Baldwin I of Jerusalem, 1100-1118 9781472433565, 9781315568645

Citation preview

Baldwin I of Jerusalem, 1100–1118

Baldwin of Boulogne was born the youngest of three sons and marked out for a clerical career, yet in turn he became a First Crusader, first Latin count of Edessa and the founder of the Latin kingdom of Jerusalem. Nevertheless, remarkably, he has never been the subject of a full-length biography. This study examines in detail the stages of Baldwin’s career, returning to the contemporary evidence to discover the qualities that enabled him not only to succeed his brother as ruler in 1100 but to maintain and expand the new kingdom of Jerusalem through the next eighteen years in the face of aggression from Muslim enemies and rivalry from fellow crusaders. Susan B. Edgington is a teaching and research fellow at Queen Mary University of London. She has written extensively on many aspects of the Crusades, but it is her close familiarity with the Latin sources for the period in question, 1095–1118, that uniquely qualifies her to write this biography. She is the editor and translator of Albert of Aachen’s Historia Ierosolimitana (Oxford, 2007), and translator (with Thomas S. Asbridge) of Walter the Chancellor’s The Antiochene Wars (Aldershot, 1999), (with Carol Sweetenham) of The Chanson d’Antioche (Farnham, 2011) and (with Steven Biddlecombe) of Baldric of Bourgueil’s History of the Jerusalemites (Woodbridge, forthcoming). Her critical comparison of the relationship of Bartolf of Nangis’s Gesta Francorum Iherusalem expugnantium with the chronicle of Fulcher of Chartres (Crusades 13, 2014) will lead to a new edition and translation of this important source.

Rulers of the Latin East Series editors by Nicholas Morton Nottingham Trent University, UK

Jonathan Phillips

Royal Holloway University of London, UK

Academics concerned with the history of the Crusades and the Latin East will be familiar with the various survey histories that have been produced for this fascinating topic. Many historians have published wide-ranging texts that either seek to make sense of the strange phenomenon that was the Crusades or shed light upon the Christian territories of the Latin East. Such panoramic works have helped to generate enormous interest in this subject, but they can only take their readers so far. Works addressing the lives of individual rulers – whether kings, queens, counts, princes or patriarchs – are less common and yet are needed if we are to achieve a more detailed understanding of this period. This series seeks to address this need by stimulating a collection of political biographies of the men and women who ruled the Latin East between 1098 and 1291 and the kingdom of Cyprus up to 1571. These focus in detail upon the evolving political and diplomatic events of this period, whilst shedding light upon more thematic issues such as: gender and marriage, intellectual life, kingship and governance, military history and inter-faith relations.

Baldwin I of Jerusalem, 1100–1118 Susan B. Edgington Godfrey of Bouillon Duke of Lower Lotharingia, Ruler of Latin Jerusalem, c.1060–1100 Simon John The Counts of Tripoli and Lebanon in the Twelfth Century Sons of Saint-Gilles Kevin James Lewis

Baldwin I of Jerusalem, 1100–1118 Susan B. Edgington

First published 2019 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2019 Susan B. Edgington The right of Susan B. Edgington to be identified as author of this work has been asserted by her in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book has been requested ISBN: 978-1-4724-3356-5 (hbk) ISBN: 978-1-315-56864-5 (ebk) Typeset in Times New Roman by Apex CoVantage, LLC

Contents

ContentsContents

List of maps List of abbreviations Preface A note on translations Maps   1 First Crusader Baldwin the younger son  1 The First Crusade  4 The journey to Constantinople  5 Contacts with Byzantium  8 Constantinople 10 The oath to Alexios  12 The evidence  13

viii ix xi xii xiii 1

  2 Nicaea to Edessa The siege of Nicaea  21 The battle of Dorylaeum  23 Cilician adventure or Cilician campaign?  24 Tarsus 25 Mamistra 30 Coming of age  32 Turbessel 34

21

  3 Count of Edessa Why Edessa?  38 The invitation from Edessa  39 The coup  42 Ruler of Edessa  45 The Armenian marriage  46

38

vi  Contents Rebellion 48 Support for the crusade  50 Pilgrimage to Jerusalem  54   4 Crisis in Jerusalem, 1100 Bohemond’s capture  59 Godfrey’s rule in Jerusalem  60 Godfrey’s title  61 Godfrey’s death  63 The succession dispute  63 Baldwin’s journey to Jerusalem  69

59

  5 King of Jerusalem A royal progress  78 Coronation 80 Tancred 82 Bohemond’s captivity and release  84 Raymond of Saint-Gilles  85 Daibert of Pisa  86 The failure of the Holy Fire, 1101  86 Six years later  88

76

  6 Ecclesiastical affairs The patriarchate of Jerusalem  93 Daibert of Pisa  95 Baldwin’s accession  98 Patriarch Evremar (1102–1108)  103 Patriarch Gibelin (1108–1112)  105 Patriarch Arnulf (1112–1118)  106

93

  7 The conquest of the littoral Introduction 111 The situation at Baldwin’s accession  112 The capture of Arsuf, 1101  114 The siege and capture of Caesarea, 1101  116 The siege and capture of Acre, 1104  119 The capture of Tripoli (1109) and Beirut (1110)  123 The capture of Sidon, 1110  124 Conclusion 125

111

Contents  vii   8 Fighting the Saracens The first battle of Ramla, 1101  129 Arrival of the ‘Crusade of 1101’  132 The second battle of Ramla, 1102  134 A Saracen ambush, 1103  139 The third battle of Ramla, 1105  141 Ambushes and raids  144

129

  9 The army, administration and allies Introduction 151 The army  151 Administration 155 The crusader states  159 Attacks by Mawdūd of Mosul, 1110–1113  161 The Turkish invasion of 1115  164 Byzantium 165 Sicily 166

151

10 Last years and legacy Introduction 173 Into Arabia  173 The last campaign  175 The funeral procession  177 Tomb and epitaph  178 Twelfth-century assessments of Baldwin  179 William of Tyre  180 Baldwin’s marriages  182 The succession  185 Conclusion 187

173

Bibliography Index

191 199

Maps

1   2  3  4 

MapsMaps

Lotharingia to Constantinople, 1096 Constantinople to Edessa, 1097 The conquest of the littoral, 1100–1118 The crusader states in 1118

xiii xiv xv xvi

Abbreviations

AbbreviationsAbbreviations

AA AK AS BN F-E FC GF GN HP IA IQ ME MGH SS MS

Albert of Aachen, Historia Ierosolimitana, ed. and trans. Susan B. Edgington (Oxford, 2007) Anna Komnene, The Alexiad, trans. E.R.A. Sewter, rev. Peter Frankopan (London, 2009) ‘The First and Second Crusades from an Anonymous Syriac Chronicle’, ed. and trans. A. S. Tritton and H.A.R. Gibb, Journal of the Royal Asiatic Society, 92 (1933), 69–102, 273–306 Bartolf of Nangis, ‘Gesta Francorum Iherusalem expugnantium’, RHC Occ, III: 491–543 Frutolfs und Ekkehards Chroniken und die Anonyme Kaiserchronik, ed. Franz-Josef Schmale and Irene Schmale-Ott (Darmstadt, 1972) Fulcher of Chartres, ‘Historia Hierosolymitana’, ed. Heinrich Hagenmeyer (Heidelberg, 1913), RHC Occ, III: 311–485 Gesta Francorum et aliorum Hierosolimitanorum, ed. and trans. Rosalind Hill (London, 1962) Guibert of Nogent, Dei gesta per Francos, ed. Robert B. C. Huygens, CCCM 127A (Turnhout, 1996) ‘Historia peregrinorum euntium Jerusolymam’, RHC Occ, III: 169–229 Ibn al-Athir: The Chronicle of Ibn al-Athir for the Crusading Period from al-Kamil fi’l-Ta’rikh AH495, ed. D. S. Richards, 3 vols (Aldershot, 2006–2008) Ibn Al-Qalānisī: The Damascus Chronicle of the Crusades: Extracted and Translated from the Chronicle of Ibn Al-Qalānisī, trans. H.A.R. Gibb (London, 1932) Matthew of Edessa, Armenia and the Crusades, Tenth to Twelfth Centuries: The Chronicle of Matthew of Edessa, trans. Ara E. Dostourian (Lanham, 1993) Monumenta Germaniae Historica, Scriptores, ed. G. H. Pertz et al., 32 vols (Hanover, Weimar, Stuttgart and Cologne, 1826–1834) Michael the Syrian, Chronique de Michel le Syrien, Patriarche Jacobite d’Antioche (1166–1199), ed. and trans. Jean-Baptiste Chabot, 5 vols (Paris, 1899–1924)

x  Abbreviations Orderic Vitalis, The Ecclesiastical History of Orderic Vitalis, ed. Marjorie Chibnall, 6 vols (Oxford, 1969–1980) RA Raymond of Aguilers, Le ‘Liber’ de Raymond d’Aguilers, ed. J. H. Hill and L. L. Hill (Paris, 1969) RC Ralph of Caen, ‘Gesta Tancredi’, RHC Occ, III: 587–716 RHC Occ Recueil des Historiens des Croisades, Historiens Occidentaux, ed. Académie des Inscriptions et Belles Lettres, 5 vols (Paris, 1844–1895) RRR Revised Regesta Regni Hierosolymitani Database, http://crusadesregesta.com UKJ Die Urkunden der lateinischen Könige von Jerusalem, ed. Hans Eberhard Mayer, 4 vols (Hanover, 2010) WC Galterii Cancellarii: Bella Antiochena, ed. Heinrich Hagenmeyer (Innsbruck, 1896) WM William of Malmesbury, Gesta Regum Anglorum, ed. and trans. R.A.B. Mynors, Rodney M. Thomson and Michael Winterbottom, 2 vols (Oxford, 1998–1999) WT William of Tyre, Chronicon, ed. Robert B. C. Huygens, CCCM 63–63A, 2 vols (Turnhout, 1986) OV

Preface

PrefacePreface

First of all I want to thank the series editors, Jonathan Phillips and Nicolas Morton, for encouraging me to write this volume and for their patience in waiting for its completion once I realised just how big a task I was taking on. As ever I have incurred enormous debts of gratitude to institutions and individuals that I acknowledge here. A number of libraries have been very helpful, including Cambridge University, the Institute of Historical Research, Reading University and the Warburg. My ‘home’ institution, Queen Mary University of London, has facilitated my research in many ways. Chris Worthington has once more prepared the maps. Thomas Asbridge read early drafts and offered advice for improvement. There is a wonderfully supportive community of people working on the Crusades and the Latin East, many of whom I meet at the Monday seminars at the Institute of Historical Research and events organised by the Society for the Study of the Crusades and the Latin East. I am grateful to all of them, but especially to the following individuals who have responded, often within minutes, to emailed SOSs: Matthew Barber; Betty Binysh; Andrew Buck; Martin Hall; Simon John; Robert Kool; Katherine Lewis; Alan Murray; Jay Rubenstein; Iris Shagrir; Carol Sweetenham; Steven Tibble. My family, as always, has provided both distraction and encouragement. Without the first I might have finished the book sooner; without the second I might not have completed it at all. Thank you all: Ben, Penny, Hannah and Rebekah Edgington; Rebecca, Mike, Emma and Annie Richardson. SBE December 2018

A note on translations

A note on translationsA note on translations

There is a wealth of primary evidence for the period 1095 to 1118, and it is used extensively in this account of Baldwin’s career. For languages other than Latin, the study is dependent on the translations of others. However, all Latin quotations have been translated by the author, and detailed references are given in the end notes to enable the original texts to be consulted. Hence the original Latin will be found in the chapters or notes only where the source is particularly hard to access or where existing English translations are contested.

Map 1  Lotharingia to Constantinople, 1096

Maps MapsMaps

Map 2  Constantinople to Edessa, 1097

Map 3  The conquest of the littoral, 1100–1118

Map 4  The crusader states in 1118

1 First Crusader

First CrusaderFirst Crusader

Baldwin of Boulogne was a crusader who became the first Latin count of Edessa and then the first Latin king of Jerusalem, a realm that he ruled for almost eighteen years. These would be outstanding achievements for any man, but for someone who was the youngest of three brothers and whose prospects at birth and as a youth were correspondingly modest, they suggest extraordinary strength of character as well as a modicum of good fortune. The balance between these assets will emerge from an investigation of his life and rule. Inevitably, however, the enquiry must begin with Baldwin’s background and early years, a period for which the documentary sources are sparse and sometimes contradictory. A fair amount of what follows, therefore, is necessarily conjectural.

Baldwin the younger son Baldwin was born in the early 1060s, the youngest of three sons of Eustace II, count of Boulogne, and Ida of Bouillon.1 Both Eustace II and Ida were prominent members of the northern European nobility. Eustace II of Boulogne (c. 1015– 1087) fought alongside William of Normandy at the battle of Hastings in 1066 and received lands in England to add to his Norman territory. When he died his eldest son became Eustace III, count of Boulogne, and his future was to be AngloNorman.2 Eustace II married Ida of Bouillon (c. 1040–1113) at a date before 1049, and it was through Ida that their second son, Godfrey, inherited lands in Lotharingia from Ida’s brother Godfrey III ‘the Hunchback’, who was childless. Ida was pious and charitable, involved in founding several religious houses, and also, evidently, an educated woman and a powerful personality whose influence over her sons’ upbringing and education was strong.3 She corresponded with a number of prominent clerics, and one of them, at least, Anselm, abbot of Bec and later archbishop of Canterbury, knew her sons personally and may even have been involved in their education.4 As was frequently the fate of a third or younger son, Baldwin was brought up in the expectation of a career in the Church. The later chronicler William of Tyre wrote of him: ‘In his adolescence he was appropriately educated in the liberal arts and became, it is said, a cleric, obtaining benefices, that are commonly called prebends, in Reims, Cambrai and Liège, thanks to the nobility that made him

2  First Crusader uniquely outstanding.’5 Baldwin’s ‘nobility’ referred to his lineage and family connections, and it is probable that his brother Godfrey influenced his preferment. Notably, we do not have any other information about Baldwin’s clerical career, and William indicated that his source was hearsay. However, Albert of Aachen, writing in the first two decades of the twelfth century, reported that Baldwin won a theological argument with the patriarch Daibert since he was ‘a man instructed in letters’.6 The witness who knew Baldwin best, Fulcher of Chartres, appears simply to have seen no need to explain the background of his patron. William of Tyre went on to say: ‘At length, for reasons obscure to us, he laid aside his clerical dress and took up military arms, becoming a knight.’ Hans Eberhard Mayer has conjectured that William’s claim was disingenuous, that Baldwin’s reasons were rooted in the Gregorian reform movement and papal condemnation of pluralism. According to this scenario, the bishop of Cambrai insisted on his priests relinquishing all but one prebend, while his opponents protested that one prebend would not suffice to support the incumbent. Baldwin therefore opted for a secular career. Mayer speculated that William suppressed his knowledge of Baldwin’s motivation because he himself held multiple prebends. There is no real evidence for this theory, but it seems unlikely that Baldwin rejected a career as a cleric, where he had inadequate means but real expectations of advancement, in favour of life as a landless knight without prospects. An alternative explanation is possible, according to which Baldwin was the chosen heir – perhaps heir presumptive – of his childless elder brother, Godfrey. Godfrey was more fortunate than many second sons and had inherited his maternal uncle’s title and substantial landholdings in Lower Lotharingia. The uncle died in 1076, but Godfrey’s title was not secure until over a decade later, during which time he had fought a long war of succession, gradually claiming his inheritance by negotiation and by force of arms. There is some evidence that Godfrey also fought for Emperor Henry IV in Italy between 1081 and 1084. According to William of Malmesbury, writing in the 1120s, Godfrey fell critically ill in the summer of 1084 when he was part of Henry IV’s army besieging Rome, and as a result he took a vow to go on the crusade.7 William was not always a reliable witness: his account of events in the East was taken mainly from Fulcher of Chartres, and his information about Godfrey’s ancestry and career before the crusade was written in light of Godfrey’s elevation to ruler of Jerusalem. Doubt has been cast on whether Godfrey was present at the siege of Rome at all: Alan Murray pointed out that ‘local sources attest to [Godfrey’s] activity in Lotharingia during this period’; however, it is not necessary to assume that Godfrey fought during the whole period from 1081 to 1084, only that he took part in the siege of Rome, for which there is some support in Albert of Aachen’s remark in relation to an epidemic that occurred outside Antioch during the First Crusade that Godfrey left the camp, ‘recalling that he had suffered from a very similar illness when he was on an expedition with King Henry IV’.8 Also, the idea that the ‘Jerusalem journey’, meaning the First Crusade, was rumoured in the 1080s must be discounted, but that is not, in fact, what William wrote. He claimed that Godfrey contracted a quartan fever (febrim quartanam iniit). Generally,

First Crusader 3 historians of medicine frown on ‘retro-diagnosis’, but in this case a quartan fever had been recognised for some 1,500 years, and it is now known to be caused by the parasite Plasmodium malaria, carried by mosquitoes and prevalent outside Rome until the marshes were drained in the twentieth century. It is, moreover, the strain of the disease least likely to prove fatal, and it can last for the lifespan of its human host, sometimes recrudescing after many years of dormancy. After discussing several possible causes, including ‘cruel mists exhaled by the river Tiber in the early morning’, William continued, ‘Nevertheless, whatever happened, it is certain that he was never free from the trouble of a continual but slow fever until, hearing a rumour of the Jerusalem journey, he vowed he would go there if God favoured him and granted him health.’ Straight away the duke recovered his health and strength, and so ‘he was the first or among the first to go to Jerusalem’. Thus, Godfrey contracted a chronic and debilitating disease in 1084, but did not make a recovery until sometime later when he made a vow to go on pilgrimage.9 How long after he started to suffer from feverish outbreaks Godfrey made his vow cannot be known, but it is unlikely to have been as late as the end of 1095, which is when reports of Urban’s speech at Clermont reached Lotharingia. William of Malmesbury may have conflated a more general vow, made in the 1080s, to make a pilgrimage to Jerusalem at some unspecified future date, with Godfrey’s response to the preaching of the crusade. Assuming that Godfrey planned to lead a major pilgrimage, it could only be undertaken once he had secured his position in Lotharingia and mustered resources. Since he was unmarried and had no heir, this could be why he summoned his younger brother to be groomed for the dukedom, to inherit if Godfrey died while on pilgrimage. Godfrey’s prayer for recovery from illness may have included a vow of chastity, at least until his pilgrimage was accomplished. His brother could stay behind and assume the dynastic responsibility of producing sons. If this was the reason behind Baldwin’s change of career, then it took place by 1086, when Baldwin was recorded as fighting alongside his brothers at the siege of Stenay to defend Godfrey’s inheritance rights. That he was fighting as a knight is a strong indication that he was at least sixteen at the time and also that his early education, like his brothers’, included training in horsemanship, weaponry and other chivalric pursuits. It is probable that Baldwin left his clerical career behind him with relief, for nothing about his later life suggests more than conventional piety, while he would prove to be an effective warrior and secular ruler. At some point between leaving the Church and departing on the First Crusade Baldwin married: ‘And then as time went on he took a wife from England, a distinguished and noble lady called Godevere.’10 Godevere, who was also known as Godehilde, was the daughter of the Norman lord Ralph II of Tosny, who had fought with William the Conqueror in 1066 and had subsequently been rewarded with extensive landholdings in England. Although his bride was not an heiress, the connection gave Baldwin an entrée into a network of powerful families within the Anglo-Norman world that included his eldest brother, Eustace, who may well have played a part in securing Baldwin’s marriage. On the surface it was a marriage more advantageous to Baldwin than to Godevere or the powerful Tosny

4  First Crusader clan: Mayer offers Baldwin’s status as literatus as the prize, arguing in circular fashion that it was an early sign that the nobility no longer considered education infra dig. It is more likely that Baldwin was Godfrey’s designated heir and thus had more concrete prospects. The date of the marriage is not known, although there is an indication in Orderic Vitalis that he was residing with his Norman inlaws early in 1090.11 There is then a gap in documentary evidence until the mid-1090s, by which time Baldwin appears to have returned to establish a presence in Lotharingia. Godfrey and their mother, Ida, were in the process of selling or giving their lands to the Church, and Baldwin appeared among the witnesses to a number of charters and in two other documents associated directly with his brother Godfrey. Alan Murray has argued from this, and from the almost complete absence of Eustace from the same documents, that on the eve of the First Crusade Baldwin was recognised as Godfrey’s heir in Lotharingia, probably because Eustace, as a vassal of the king of France, would have proved unacceptable to the German monarchy or the clergy. Baldwin’s marriage is also relevant because Godfrey had not married, and therefore his brother would be responsible for perpetuating the dynasty. There is also evidence among these documents that it was important for Baldwin’s recognition as Godfrey’s heir to be shared by key personnel who were part of Godfrey’s crusading army. A charter for the abbey of Nivelles, for example, associated Godfrey and his brother Baldwin as principals and was witnessed by lords and knights from the region of Ardennes who were to go with them on crusade: Cono, count of Montaigu; Warner, count of Grez; Henry of Esch; Henry’s brother Godfrey of Esch; Heribrand of Bouillon; Walter of Bouillon.12

The First Crusade The charter was part of the liquidation of Godfrey’s Ardennes-Bouillon inheritance that he undertook to finance his crusade. It is not known at what point Godfrey made the decision to join the crusading armies: he did not attend the papal council of Clermont in November 1095 following which Urban II had announced the departure nine months thence of an expedition for the relief of the Eastern Church and the liberation of Jerusalem. William of Malmesbury’s account of Godfrey’s illness during the siege of Rome compressed events and appeared to link Godfrey’s response to the crusade appeal directly with a vow in fact taken over a decade before. Nevertheless, it is very possible that at some time, perhaps when he was very ill, Godfrey vowed to undertake the pilgrimage to Jerusalem. He may also, then or previously, have taken a vow of celibacy and/or chastity. One of the few things about Godfrey that seems to be securely established is that – unusually for the time and especially for a substantial landholder – he never married. He was born in ‘the third quarter of the eleventh century’, and before Baldwin (born in the 1060s), so he was certainly of an age to marry by 1084.13 If he had indeed vowed to devote his life to God, then it explains why, as the evidence suggests, he had designated Baldwin as his heir in the 1080s.

First Crusader  5 It does not necessarily explain why Baldwin decided or consented to accompany him. Genuine piety should not be discounted, of course: although Baldwin had turned his back on life as a churchman, he nonetheless lived in an age of unquestioning Christian devotion. A taste for adventure was also a probable factor, in view of his later exploits on crusade. A major reason, from the later evidence, seems to have been personal ambition and the desire to carve out a territory for himself. Most significantly, though, the whole-hearted commitment of Godfrey and Ida to the forthcoming expedition, which was clearly on a much greater and more costly scale than any previous plan to travel to Jerusalem that Godfrey may have conceived, led them to dispose of all their major landholdings to finance the crusade and in return for the Church’s protection, as the charter evidence shows. This meant that Baldwin – whatever his own feelings, which, of course, we cannot know – now had no assured future in the West but was obliged to travel east as his brother’s dependant and lieutenant. Rather ironically, after supporting Godfrey in the disposition of Godfrey’s lands in Lotharingia, including the territory’s final dissolution to settle disputes and raise funds in preparation for the expedition to the East, Baldwin had rendered himself the ‘landless younger son’ once identified and widely accepted to be a significant component of the crusading armies.14 Jonathan Riley-Smith, after calculating the funds necessary to sustain a knight and his household on a long campaign, comprehensively dismissed the idea that such men were a major part of the crusading armies, but Baldwin was in fact an exception: he appears to have been funded entirely by his brother Godfrey, and this must be taken into account when considering his motivation for and activities on the crusade. A further indication that Baldwin may have seen his future in the East is that he took with him his wife, Godehilde/Godevere. By contrast, the elder brother, Eustace, who also went on the crusade, left his wife, Mary of Scotland, to administer his estates. He apparently had no need to mortgage or sell these lands, and so he probably always intended to return to them, as indeed he did after the battle of Ascalon in 1099. Eustace, who apparently self-identified as Norman rather than Lotharingian, travelled to the Holy Land separately from his brothers; the best evidence suggests that he accompanied Robert of Normandy and Robert II of Flanders.15

The journey to Constantinople Godfrey and Baldwin set out together on the overland route, following the river valleys: first the Rhine and then cutting across to the Danube. Our only detailed source for their journey is Albert of Aachen, who devoted the second book of his Historia Ierosolimitana to it. Albert was writing in the first quarter of the twelfth century and probably recording the information in this book as it arrived in Aachen: this is the only reasonable explanation for the level of detail in his account of the journey from the Ardennes to Constantinople. He himself wrote in his prologue that he ‘decided to commend to posterity at least some of the things which were made known to me by listening to those who had been there and from their reports’.16 Thus he was recording oral testimony from people who returned

6  First Crusader to the Rhineland for one reason or another, but also, and especially for this stage of the crusade, we may conjecture that letters and despatches were sent regularly from the leading crusaders to their households and to the Church figures and institutions who had effectively sponsored their journeys, who were protecting their lands and supporting them by their prayers. The carriers of such messages may have been an additional source of oral information. It should be stressed that Albert’s Historia is uniquely valuable because he did not use any of the other extant written accounts of the crusade and did not share their perspective, which was overwhelmingly French and papal. Instead he had his own bias, and, fortunately for the current project, this disposed him to focus his account on Godfrey, Baldwin and their companions. The two brothers were accompanied by a large army, led by a group of kinsmen, vassals and allies who later formed the core of the domus Godefridi, the duke’s household who assisted him in the government of the kingdom from 1099 to 1100 and were to be instrumental in securing Baldwin’s succession in 1100. Albert of Aachen provided a list of these key men. Warner of Grez was listed first after Godfrey and his brother Baldwin and described as a kinsman of the duke. The degree of kinship is not known, but there is little doubt that Warner was a Lotharingian, and he was probably a vassal of the bishop of Liège. He enjoyed a high degree of trust and would act as deputy during Godfrey’s last illness; then he led the group who supported Baldwin’s right to inherit and seized the Tower of David after Godfrey died. Warner died only four days after Godfrey (22 July 1100). Baldwin of Bourcq was likewise a kinsman of Godfrey and Baldwin, and again the degree of kinship is uncertain. He later threw in his lot with the younger brother and accompanied Baldwin to Edessa. Baldwin entrusted the county of Edessa to him when he left to take up the kingdom of Jerusalem in 1100. After Baldwin I’s death in 1118, Baldwin of Bourcq would succeed him as king of Jerusalem.17 Count Rainald III of Toul came next in the list, followed by his younger brother, Peter of Astenois (or Dampierre). Rainald and Peter were sons of Count Frederick I of Astenois, another Bouillon/Boulogne kinsman. Both brothers participated with Baldwin in the conquest of Edessa, but then they rejoined the main army besieging Antioch. They returned to Upper Lotharingia after the crusade. Dodo lord of Cons, from the Ardennes, had a relatively undistinguished crusading career; he was accompanied by his wife, Hadwida. Two more brothers completed this list of the closest circle around Godfrey: Henry and Godfrey of Esch. Henry was later described by Albert of Aachen as unus de collateralibus ducis Godefridi, which Murray interprets as another kinsman, but it is probably better translated as ‘one of Duke Godfrey’s associates’ or ‘confidants’. Both brothers were to participate in the siege of Antioch; Henry died of disease at Turbessel in late summer 1098, and Godfrey’s fate is not known.18 This was only the core of the army that left with Godfrey and Baldwin in August  1096. The names of some lesser lords and knights emerge later as the expedition progressed, and the names of others remain unknown. Each of them was accompanied by a small group, possibly some family members but certainly

First Crusader  7 servants who took care of both their master and his horses. As Riley-Smith pointed out, ‘a properly equipped knight, with armour, arms, warhorses, packhorses and servants, had to plan for a costly journey’.19 There were non-combatants, including women, children and priests. Some were probably attracted by the prospect of taking the overland route, which entailed less forward planning and less initial expense than going by sea as the other armies did. It is impossible to arrive at even a plausible estimate of the numbers of any of the individual crusading armies or of the total force that came together at the siege of Nicaea in 1097, but a best guess for the whole expeditionary force is in tens of thousands rather than the hundreds of thousands that contemporary chroniclers asserted.20 When the Lotharingian army reached Tulln, some twenty-five miles west of Vienna on the river Danube, they halted for three weeks. The reason was to find out what had befallen the ‘army of pilgrims’ that had been destroyed a short time before, the remnants of which had turned back and were coming towards them. These were not the followers of Peter the Hermit, a charismatic preacher who had set out in spring of 1096 rather than waiting for the official departure date of 15 August (the feast of the Assumption of the Blessed Virgin Mary), but less disciplined bands, inspired by Peter, who followed the same route. Naturally, they ran into supply difficulties, first because they set out before the harvest was in and second because Peter’s army had already passed the same way. They caused trouble everywhere along their route by attacking the Jews in the Rhineland and later by looting and thieving in the markets. Albert of Aachen named the leader of the first of these groups as Gottschalk and wrote that his followers (Albert said they numbered over 15,000, but this should be taken merely as an indication of a large number) came from Lotharingia, eastern France, Bavaria and Swabia and comprised ‘as many knights as common footsoldiers’. Folkmar led others, Frutolf claimed there were 12,000 of them, through Saxony and Bohemia. Emicho of Leisingen led another great rabble of men and women who became notorious for their persecution of the Rhineland Jews. They all came to grief in Hungary: Gottschalk was induced to surrender to the king at Pannonhalma; Folkmar’s band rioted and was killed and dispersed at Nitra; Emicho’s followers reached the fortress at Mosony but panicked and fled. Many were caught and killed, but a handful of leaders survived and fled towards Carinthia and Italy.21 If any of the survivors of these unofficial forces made yet another turn and accompanied Godfrey and Baldwin on crusade, we do not know, but evidently survivors of the unruly bands met Godfrey and his company and complained vociferously about their treatment by the Hungarians, whereupon the Lotharingians decided to make an enquiry into what Albert called ‘abominable murder and wickedness’. For this they sent to King Coloman of Hungary Godfrey of Esch, who was known to the king because he had been sent by Godfrey on a previous embassy, probably to pave the way for the duke’s army, and twelve other men including the duke’s seneschal, Baldric, and his chamberlain, Stabelo. The envoys carried a letter, probably reproaching the king for killing Christians and asking for safe conduct. After an exchange of letters Godfrey was invited to meet the king at Sopron, taking with him Warner of Grez, Rainald of Toul, Peter Astenois and

8  First Crusader 300 men. After the negotiations at Sopron went well all but 12 of the 300 were sent back to Tulln while Godfrey received a splendid week-long reception at Pannonhalma with further negotiations culminating in a treaty ensuring safe conduct for Godfrey’s army.22 Meanwhile Baldwin had been left in Tulln ‘to rule and look after the people’, a phrase confirming that he was seen as Godfrey’s deputy. As part of the terms of the treaty the king required hostages for the army’s good behaviour in his lands, and it was evidently agreed that these would be Baldwin, his wife and his household.23 Again this is clear confirmation that, in the absence of a son who would be expected to act as hostage in such circumstances, Baldwin was accepted and valued as Godfrey’s heir. With relief, no doubt, the duke sent messengers back to the camp at Tulln and then was reunited with his army after it had moved to Sopron. Baldwin was not at all pleased to be informed, and then reminded, that he was to be hostage for the army’s good behaviour, and he resisted and argued so vehemently that his brother told him he could take charge of the army while Godfrey became hostage in his stead. This calmed Baldwin down, and he consented to be the hostage. His resistance and his eventual acquiescence give an interesting insight into the relationship between the brothers; possibly he gave in because he was aware that he did not command the same respect and obedience his brother did, for after dire warnings to all ‘the duke and the people crossed the kingdom of Hungary, every day in peace and quiet, buying in fair and just measure’, an outcome that he might have found difficult to ensure. The king saw Godfrey and his army across the frontier, keeping back with him a strong troop of cavalry and also Baldwin and the other hostages. When Godfrey arrived at the river Sava he found his next problem on the other bank: ‘the irresistible might of the imperial army of Constantinople’ barring their way into Bulgaria. Only three ferries were available so he sent 1,000 knights across (the number is Albert’s) to establish a bridgehead, and then they built rafts to take the rest of them. Only when all were safely ashore did King Coloman appear and restore Baldwin and the other hostages. Coloman apparently parted with Godfrey and Baldwin on good terms, ‘which he showed by many gifts and the kiss of peace’.

Contacts with Byzantium The crusaders spent the night in Belgrade and then moved off into ‘the vast and strange forests’ where the Byzantine emperor’s envoys met them. Although ultimately Godfrey and Baldwin and their companions were responding to an appeal from the Byzantine emperor, Alexios I Komnenos (1081–1118), and were heading to his court in Constantinople, this was the first time they came into direct contact with the Byzantines. From now on there is another and independent account of events, that of the Greek princess, Alexios’s daughter Anna Komnene. The Alexiad, as its title indicates, was a biography of the emperor and an account of his reign. Anna was a fourteen-year-old girl in Constantinople when the armies of Godfrey and the other leaders arrived there, and although it was not until some forty years later that she composed the work, it describes vividly not only actions

First Crusader  9 but reactions, for example the impression at court that a whole continent was on the move: ‘the whole of the west and the entire people living between the Adriatic and the Straits of Gibraltar migrated in a body to Asia, marching from one end of Europe to another with their whole households in tow’.24 Anna was as disposed to praise Alexios as Albert of Aachen was to write an account favourable to Godfrey, so if we look past their prejudices to the surprising amount of common ground a quite clear picture of their contacts and confrontations emerges. According to Albert, the message conveyed by the emperor’s representatives was couched in conciliatory terms: Godfrey was not to allow his followers to devastate and plunder imperial lands; in return he would obtain a licence to buy provisions. The emperor’s guarded reception is completely understandable: in his eyes a routine appeal to the papacy for western troops to serve under him against his enemies had met with an unprecedented and overwhelming response.25 Alexios had already had to deal with Peter the Hermit’s ‘unofficial’ army of followers that had caused so much trouble that he had it shipped across the Bosphorus to Asia Minor where yet more undisciplined behaviour had led to their defeat and a massacre by the Turks. Now, in the autumn of 1096, tens of thousands of people, some of them armed knights but more unarmed pilgrims, were converging on Constantinople under different leaders. Godfrey responded to Alexios’s message by promising acquiescence, although Albert of Aachen said, ‘it was proclaimed to all that from then on they should not seize anything at all by unjust force, unless it was fodder for the horses’, which casts an interesting light on both the needs of his army and the limits to Godfrey’s readiness to obey the emperor. Under this agreement they marched on to Niš, a Byzantine stronghold, where the leaders were lavishly entertained and the rest of the army had a licence to buy and sell. After four days, they continued to Sofiya where the standard of hospitality was equally generous, then on to Philipopolis (Plovdiv) where they were plentifully supplied for a week.26 There they received the news that the king of France’s brother Hugh (often mistakenly called ‘the Great’ by chroniclers) was being detained by the emperor. With Hugh were two survivors of Emicho’s ‘crusade’, Clarembald of Vendeuil and Drogo of Nesle, who had evidently returned to France and joined the better organised army of Hugh. Both came from prominent French families and were to join Godfrey and stay with the main army until the siege of Antioch. Clarembald is last heard of at the point where Stephen of Blois deserted during the siege of Antioch. Drogo at the same time joined Baldwin in Edessa and then in Jerusalem and evidently made a career in the East where his presence is attested to 1126. Godfrey’s response to the news of the three nobles’ detention in 1096 was to send an embassy to Alexios asking him to release the prisoners, ‘otherwise he would forfeit the duke’s trust and friendship’. When his messengers returned, they reported that the emperor had not made any move to release Hugh and his companions, so Godfrey angrily ordered his army to plunder the countryside, which they duly did for the next week.27 Meanwhile, as soon as they perceived that Godfrey was likely to fall out with the emperor, two of his close circle, Henry of Esch and Count Baldwin of Hainaut,

10  First Crusader raced ahead of his envoys to arrive in Constantinople and have gifts bestowed on them by the emperor while he was still well disposed towards Godfrey and his army. Barely concealing his anger, Godfrey and his company moved on to Adrianople (Edirne), where they found the bridge and entry to the town barred to them, and then to Salabria (Silivri) on the sea of Marmara near Constantinople. It was here that Godfrey allowed his troops to devastate the countryside.28

Constantinople The emperor’s response was to send two Franks who were in his service with an offer to return the prisoners if the army stopped looting. After consulting the other leaders, Godfrey agreed and moved his whole army to encamp outside Constantinople, but they remained armed and ready for battle. The emperor kept his side of the agreement, and the former prisoners came to meet them, together with another survivor of Emicho’s crusade, William ‘the Carpenter’, the viscount of Melun (later to desert at Antioch). The emperor’s envoys invited the duke and other leaders to enter the palace while the rest of the army stayed outside the city walls. However, Godfrey was warned by some Frankish strangers that the emperor was not to be trusted, and they advised him to stay outside the walls. The situation escalated when the emperor retaliated by withholding a licence for the army to buy and sell in the markets. At this point Baldwin, who must be presumed to have approved his brother’s actions thus far, is singled out as proposing to the nobles that they should resume looting and helping themselves to the food they needed. The ensuing confrontation between Godfrey’s army and imperial forces commanded by Nikephoros Bryennios (Anna Komnene’s later husband, whom she depicted as a Homeric hero) was resolved by the emperor reinstating the licence to buy and sell. One reason for his capitulation was that it was Christmas: it is possible this was a face-saving pretext for Alexios, who would not have wanted to appear to bow to pressure. He then urged Godfrey and the other nobles with him to move into palaces along the Bosphorus ‘because of the common chills of snow and wintry weather which were a problem in the rainy season’. From then on they were quartered with the rest of the crusading armies and able to procure plentiful supplies.29 Nevertheless, the duke remained suspicious of Alexios, and when he was summoned again to the palace he refused and sent in his stead Cono count of Montaigu, Baldwin of Bourcq and Godfrey of Esch.30 In spite of the emperor’s assurances, though, another fortnight passed with messages passed to and fro between him and Godfrey. At last the emperor lost patience and reimposed the sanctions, progressively withdrawing foodstuffs from sale. Godfrey did not give way, so (according to Albert of Aachen) Alexios sent 500 Turcopoles (mercenaries) to attack Godfrey’s army. At this Godfrey ordered ‘everyone’ to arm and return to their encampment before the walls of Constantinople. They smashed up and set fire to the buildings that had housed them as they left. There was a bridge to cross to resume their former position, and, fearing that the Byzantines might seize it before they could cross, Godfrey sent Baldwin with 500 knights to occupy it. Baldwin was unable to hold his position on the bridge itself under attack from

First Crusader 11 the emperor’s Turcopoles, so he hurried across and established a bridgehead on the far side while Godfrey acted as rearguard. For the first time we get a glimpse of Baldwin’s martial qualities: But Baldwin stood fast in the agreed place, immovable and unconquerable in the face of their every attack, until from dawn until dusk the people had been taken back across the bridge in front of the city walls, and they had pitched camp in safety. He bravely attacked those same Turcopoles with his five hundred armed men as they came out of the gates and boldly fought the people, and on all sides battle was fought violently; many fell here and there; many of the Franks’ horses died of arrow-wounds. But in the end Baldwin prevailed, he sent these imperial soldiers back inside the gates, oppressed and put to flight, and he gained the plain and the victory convincingly. Truly the Turcopoles and imperial soldiers, who were angry at being defeated in war and put to flight, sallied forth from the gates again and again and in ever greater numbers to challenge and overcome the army, until the duke arrived, and because it was night he reconciled everyone in peace, reminding his brother to return to the camp with everyone else and to restrain the troops and weapons from this battle in the darkness of night.31 Albert claimed that the emperor was glad to agree; Anna Komnene described the confrontation, but although in her account it was equally deadly it ended differently: ‘In this fierce engagement many on both sides fell, and all the emperor’s men who had attacked with such recklessness were wounded. As the Romans [Byzantines] showed greater spirit the Latins gave way.’ She went on to state that not long afterwards Godfrey gave in and swore the oath that Alexios was demanding from all the western leaders, in return for which he was lavishly rewarded and entertained before crossing the Bosphorus and camping at Pelekanon. Anna glossed over thus in a short paragraph a series of events that would have exposed the emperor’s weaker negotiating position. According to Albert of Aachen’s much more detailed account, on the day after the ceasefire Godfrey ordered his followers to resume plundering around Constantinople, which they did for six days. This brought Alexios to the negotiating table, and he promised Godfrey safe conduct and hostages if he would meet him face to face. Albert reported at the same time an incident not found anywhere else, the arrival of messengers from Bohemond, still in Apulia, suggesting that Godfrey withdraw from Constantinople and they join forces to conquer the Byzantine empire when Bohemond arrived in the spring. Godfrey hesitated and spoke to his men, but if anything Bohemond’s proposal appears to have strengthened his resolve for the crusade. Godfrey was probably not unaware of Bohemond’s previous history of hostility towards Byzantium and perhaps of his tricky reputation.32 He sent a message to Bohemond: He had not left his homeland and family for the sake of profit or for the destruction of Christians, but he had embarked on the journey to Jerusalem

12  First Crusader in the name of Christ, and he wished to complete the journey and to fulfil the intentions of the emperor, if he could recover and keep his favour and goodwill.33 In view of his later actions it is doubtful that Baldwin fully shared Godfrey’s sentiments, although there is no evidence he did anything but support his brother’s decision at this stage.

The oath to Alexios Again we have only Albert’s word for it, but it is probable that the emperor learned of Bohemond’s envoys – he almost certainly had spies and informers around the Christians’ camps – and this led him to renew his overtures to Godfrey. He pledged his own son, the future emperor John II (1118–1143), as hostage to Godfrey, and the duke sent two companions who were skilled linguists, Cono of Montaigu and Baldwin of Bourcq, to fetch the boy. Once the prince was in his camp, Godfrey set out with two more of his inner circle, Warner of Grez and Peter of Dampierre, for the imperial court. His brother Baldwin ‘did not enter the palace at all at that time but stayed on the shore with the rest of the great army’.34 A reasonable inference is that Baldwin was again recognised as his brother’s deputy in his absence. Godfrey was soon made aware that he was not negotiating as Alexios’s equal. He and his retinue dressed richly in fur-trimmed robes, but the emperor remained seated on his throne, and the westerners were forced to kneel one by one to exchange the kiss of peace. They were then expected to take an oath: since this was expected of Baldwin too it is highly relevant to his future relationship with the Byzantines, but its content has been much discussed and disputed. Albert of Aachen’s report is unequivocally an account of an oath of vassalage: ‘he not only gave himself to him as a son, as is the custom of that land, but even as a vassal with hands joined, along with all the nobles who were there then, and those who followed afterwards’. Anna Komnene conveyed more concrete terms: Godfrey ‘swore on oath as he was directed that whatever towns, lands or forts he might in future subdue that had in the first place belonged to the Roman Empire would be handed over to the officer appointed by the emperor for this very purpose’. Other Latin accounts largely agree with Anna, for example Bartolf of Nangis, who was probably repeating what he found in Fulcher of Chartres’s first redaction, said of Robert of Normandy and Stephen of Blois’s meeting with Alexios, ‘they agreed a pact to this effect: that if they were able to restore to him the lands of Rūm, as many as he himself had previously possessed, he would assist them by land and sea with horses and arms and silver and all necessary supplies’. This is rather more specific than other Latin versions of the agreement in that it referred to land previously held by Alexios himself when and after he acceded to the imperial throne in 1081. This had implications for cities like Tarsus, Edessa and Antioch, which were nominally Byzantine in 1081 but over which in practice Constantinople no longer had any meaningful authority. Another point to note from the array of versions of the oath is its quasi-contractual nature. The emperor had obligations, too,

First Crusader 13 and if he failed to provide the promised support he could be considered to have broken the agreement.35 It appears that Godfrey was the first to take the oath, and as other leaders arrived they followed his lead. Bohemond proved difficult to persuade to meet the emperor at first, unsurprisingly, but ‘in the end he was convinced by the duke’s good promise and comforting words’ and took the oath.36 Bohemond’s nephew Tancred, however, managed to avoid the meeting and the oath. Robert of Flanders made no difficulty and took the oath. Albert of Aachen said that Raymond of Saint-Gilles took the same oath of vassalage, but his chaplain, Raymond of Aguilers, who presumably knew better, said he refused to swear homage but promised to respect the emperor’s life and honour.37 The last big army to arrive was that led by Robert of Normandy, Stephen of Blois and Eustace, the elder brother of Godfrey and Baldwin, and they also took the oath. Baldwin’s oath-taking is not recorded, but it may well have been on an occasion Anna Komnene related. By this time she had given up trying to list names, unfortunately, so the leaders were just an anonymous group newly arrived in Constantinople. They were invited to court and persuaded, some more easily than others, to take the oath. Godfrey had crossed back to Constantinople to witness this. One of the newcomers, a Frank, insolently sat on the emperor’s throne, and, while Alexios restrained himself, Count Baldwin made him stand and reprimanded him for failing to ‘observe the customs of the country’. How much credence we should give to the story is dubious, but it possibly shows Baldwin as an authoritative figure, and not overshadowed by his brother Godfrey.38 After each leader or group of leaders had taken the oath, they received lavish gifts and were then shipped across the Bosphorus with their armies. This was surely a deliberate policy to prevent their causing more trouble of the sort Godfrey’s forces had made. All were briefed about Turkish tactics: ‘he advised them not to pursue the enemy too far, if God gave them victory, lest falling into traps set by the Turkish leaders they should be massacred’.39 Army by army the crusaders assembled to besiege Nicaea, the emperor’s first objective across the straits in Asia Minor.

The evidence Before taking the story of Baldwin’s career any further, it is useful to survey the quality of the evidence, especially since it is frequently contradictory or suspect for other reasons. As has been seen, the documentary and literary evidence that can be used to investigate the events of Baldwin’s childhood is fragmentary. The First Crusade, on the other hand, generated a number of narrative accounts that can be compared with each other to try to get close to a complete picture.40 The eye-witness accounts are well known and much discussed. The anonymous Gesta Francorum was quite widely used and copied in the years immediately after it was written; its focus was on the deeds of the Italian Normans and the northern French participants, and it says little about Baldwin, especially after Baldwin left the main body of the crusaders in 1097. A second eye-witness, Raymond of Aguilers, focused very

14  First Crusader closely on the activities of Raymond of Saint-Gilles and fellow Provençals, and his attitude to Godfrey reflected that of Raymond, whose chaplain he was. A third participant, and much more important as a witness to Baldwin’s crusade, was Fulcher of Chartres, who wrote a long Historia Hierosolymitana. Fulcher set out from Normandy with Count Stephen of Blois, who was to become notorious for deserting the crusading armies shortly before the capture of Antioch in June 1098. Rather than escaping with Stephen of Blois at that time, Fulcher made his way to Edessa, where Baldwin was established as ruler, and became his chaplain. Fulcher’s history, which covers not only the First Crusade as a participant, but also the history of the kingdom of Jerusalem as a resident from 1100 to 1127, might, therefore, be expected to offer invaluable detail and insight for the whole of Baldwin’s crusading career, his rule in Edessa and his reign in Jerusalem. It is therefore disappointing that Fulcher generally proves rather circumspect and, moreover, that there is clear evidence he undertook a thorough revision of his earlier text when he added information for the 1120s. The reasons that he edited his memories of certain events will be explored at the appropriate points. Fulcher’s earlier, and probably much more vivid, memories of events have not survived intact. However, a copy of his first redaction, written by about 1106, evidently arrived in northern France soon afterwards. Guibert of Nogent had completed his own account of the First Crusade, the Dei gesta per Francos, when he came across Fulcher’s history, and in spite of some very dismissive remarks about the author, he incorporated quite a lot of Fulcher’s information into an added seventh book. The same redaction of Fulcher’s Historia was used by a writer, usually called Bartolf of Nangis although nothing is known of the author, to produce his Gesta Francorum Iherusalem expugnantium. This was an acknowledged adaptation of Fulcher of Chartres’s history, and its author was very open about the ways he had modified his exemplar: Now, therefore, let us move on to the beginning of our narrative, and, with God’s inspiration, let us try to elucidate that which brother Fulcher of Chartres saw with his eyes, or the deeds which were told him by those who did them and which he brought together from memory and gathered into one little book. But we who are thoroughly informed both by the content of the little book, and by the accounts of others, and by penetrating investigation, avoiding a prolix narrative, content only with those things which we feel are relevant to the matter, have taken care conscientiously to modify the text of this volume.41 Besides looking to these two early witnesses for corroboration, Fulcher’s account may also be compared with independent accounts of the same events. Foremost among these is the Historia Ierosolimitana of Albert of Aachen. This long narrative recounts events of the First Crusade, but it does not end, as do the anonymous Gesta Francorum and the account by Raymond of Aguilers, with the battle of Ascalon following the capture of Jerusalem in 1099, but carries the story on to 1119, thus covering all of Baldwin I’s career. Albert of Aachen, a Rhinelander,

First Crusader  15 was especially interested in the exploits of Godfrey of Bouillon, his brother Baldwin and other Lotharingian notables. Albert appears to have garnered his information orally from returning pilgrims and crusaders through the whole period from 1095 to 1119.42 The level of detail is therefore variable, but – as will be seen – its accuracy can often be checked against other sources, in Latin, in Arabic and in other languages, and it is very frequently found to be confirmed. Notably, William of Tyre, who used Albert of Aachen’s Historia as his main source for the First Crusade, seems to have been unaware that Albert continued to gather information and write after its end. William’s chronicle is always an important point of reference, although it has to be used with two caveats. First, William was writing some two generations later (beginning c. 1170) and with the hindsight and bias that this entailed; second, for the period of Baldwin I’s reign as king of Jerusalem, William’s account was heavily dependent on Fulcher of Chartres. There is some doubt as to whether he had languages other than Latin and French, as has sometimes been claimed, largely on the basis of his lost work on the princes of the Orient. Nevertheless, as chancellor of the kingdom of Jerusalem, William had access to royal archives and was able to include certain documents not found elsewhere.43 Other independent Latin sources include Ralph of Caen’s Gesta Tancredi, which was not apparently known by Fulcher of Chartres or William of Tyre, although it was written in the Latin East. It was written to record the deeds of the Norman Italians, Bohemond and his nephew Tancred. Ralph did not participate in the First Crusade, arriving in Outremer after accompanying Bohemond’s crusade of 1107–1108. He transferred his allegiance to Tancred in Antioch at some time before Bohemond’s death in 1111, and, according to his own account, he gained his information from both uncle and nephew. It should be noted, however, that the Gesta is extant in a single manuscript which ends abruptly in 1107, so we do not know the full scope of Ralph’s intended project, only the title hinting that it would have ended with Tancred’s death in 1112. It is thought that Ralph moved to Jerusalem after 1112 and, as he dedicated the work to the cleric Arnulf who died in 1118, that he was writing during Baldwin’s lifetime. This would explain his inclusion of lavish praise for Baldwin I: had he still been working in Antioch he might have felt freer to report the antagonism that is known to have existed between his earlier patron, Tancred, and Baldwin.44 For later events of Baldwin’s reign there are also the annals of Genoa written by Caffaro. Caffaro was part of a naval expedition that arrived in the Holy Land in 1100, and he became the official annalist of Genoa; he was therefore contemporary with the events of Baldwin’s reign, although allowance has always to be made for his (unsurprising) bias in favour of his native city.45 Ekkehard of Aura is the name usually given to a participant in the crusades of 1101 who was also, therefore, in the Holy Land within a very few years of the capture of Jerusalem. Ekkehard’s identification as the author of all the (four) continuations of Frutolf of Michelsberg’s chronicle has been challenged convincingly by T.J.H. McCarthy, but Ekkehard did write a short Hierosolimita and also the segment of the chronicles for the years 1106–1116.46

16  First Crusader Sources written in other languages than Latin sometimes add valuable details. These include the Byzantine historian Anna Komnene; the Armenian Matthew of Edessa; and Ibn Al-Qalānisī, writing in Arabic in Damascus.47 Non-written evidence includes archaeology and the landscape itself; seals, coins and other artefacts survive rather rarely.

Notes 1 The most important sources for Baldwin’s life and career are discussed at the end of this chapter. The most thorough examination of Baldwin’s career and those of his brothers and kinsmen is Alan V. Murray, The Crusader Kingdom of Jerusalem: A Dynastic History 1099–1125 (Oxford, 2000), here p.  30. See also Hans Eberhard Mayer, ‘Études sur l’histoire de Baudouin Ier roi de Jérusalem’, in Mélanges sur l’histoire du royaume latin de Jérusalem, idem (Paris, 1984), pp. 10–91. For Godfrey in particular, see Simon John, Godfrey of Bouillon, Duke of Lower Lotharingia, Ruler of Latin Jerusalem, c.1060–1100 (Abingdon, 2018). 2 Thus Eustace went on crusade in 1096 with Robert of Normandy and Robert of Flanders, almost certainly, rather than his younger brothers, and although he joined up with Godfrey after Nicaea and fought beside him at Antioch and Jerusalem, he returned to Boulogne in 1099. See Murray, Crusader Kingdom, p. 193; Heather J. Tanner, ‘In His Brothers’ Shadow: The Crusading Career and Reputation of Eustace III of Boulogne’, in The Crusades: Other Experiences, Alternate Perspectives, ed. Khalil I. Semaan (Binghampton, 2003), pp. 83–99. 3 Ida was venerated after her death as a saint, and her legend became confused with that of a mysterious ‘Swan Knight’, from whom she and therefore Godfrey of Bouillon were descended: William of Tyre referred to this in an eulogy of Ida, although he dismissed it as gossip (William of Tyre [WT], Chronicon, 10.1, ed. Robert B. C. Huygens, CCCM 63–63A, 2 vols continuously paginated (Turnhout, 1986), 9.6, p. 427), and the same story was incorporated into the Chanson d’Antioche, which achieved written form early in the thirteenth century: La Chanson d’Antioche, ed. Suzanne Duparc-Quioc, 2 vols (Paris, 1977–78), I: 372–73; The Chanson d’Antioche: An Old French Account of the First Crusade, trans. Susan B. Edgington and Carol Sweetenham (Farnham, 2011), p. 277; Simon John, ‘Godfrey of Bouillon and the Swan Knight’, 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. 129–42. 4 For Anselm’s letters to Ida, see Sally N. Vaughn, St Anselm and the Handmaidens of God: A Study of Anselm’s Correspondence with Women (Turnhout, 2002), pp. 126–59. We know little or nothing about the education of Baldwin and his brothers; John conjectures that their mother, Ida, involved Anselm in it: Godfrey, p. 42. A letter sent from Anselm to Baldwin c. 1105 is discussed in chap. 10. 5 WT 10.1, p. 453. 6 Albert of Aachen [AA], Historia Ierosolimitana, 7.61, ed. and trans. Susan B. Edgington (Oxford, 2007), p. 572. 7 William of Malmesbury [WM], Gesta Regum Anglorum, ed. R. A. B. Mynors, R. M. Thomson and M. Winterbottom (Oxford, 1998), 4.373.3–4, pp. 656–58 [NB the edition comprises 2 vols (1998–1999) but all refs are to the Latin text, which is in vol. 1]. See also Neil Wright, ‘Chapter 10: William as Historian of Crusade’, in William of Malmesbury, revd edn, ed. R. M. Thomson (Woodbridge, 2003), pp. 178–88. 8 For Godfrey’s illness at Rome, AA 5.13, p.  354; more generally, his long desire to make the pilgrimage to Jerusalem, AA 6.26, p. 436. 9 Frutolf recorded the outbreak of pestilence during the siege, but not Godfrey’s presence: Frutolfs und Ekkehards Chroniken und die Anonyme Kaiserchronik [F-E], ed.

First Crusader  17

10 11

12

13

14 15

16 17

18 19 20 21

Franz-Josef Schmale and Irene Schmale-Ott (Darmstadt, 1972), p. 96. Murray’s doubts about Godfrey’s participation in the siege of Rome (Crusader Kingdom, p.  25) are echoed by John, pp. 69–72. For ‘quartan fever’ see William E. Collins and Geoffrey M. Jeffrey, ‘Plasmodium malariae: Parasite and Disease’, Clinical Microbiology Review, 20 (2007), 579–92. WM also reported a theory that Godfrey’s death in 1100 resulted from an attack of his ‘old fever’ (febrim antiquam), although his own conclusion was that it was God’s will to take him to a ‘better kingdom’, p. 658. For Godevere/Godehilde see WT 10.1, p. 453 and Orderic Vitalis [OV], The Ecclesiastical History, ed. and trans. Marjorie Chibnall, 6 vols (Oxford, 1969–81), III: 128. Mayer, ‘Études’, pp. 14–15, 27–28, 30; Murray, Crusader Kingdom, pp. 18, 30–31. See also Alan V. Murray, ‘Baldwin I of Jerusalem’, in The Crusades: An Encyclopedia, ed. Alan V. Murray, 4 vols (Santa Barbara, 2006), I: 132–33. The date of Baldwin’s marriage is not known: there is a long discussion in Mayer, ‘Études’, pp. 32–41; Baldwin’s educated status at p. 39. Story about Baldwin having a vision at Conches that was interpreted as foretelling his glorious future: OV, IV: 216–18. Murray, Crusader Kingdom, provides full references: pp. 32–33, nn. 124–26, ‘ego . . . et frater meus’ and ‘dux Godefridus et frater eius’; p. 35, n. 130, ‘Godefridus dux et frater eius Balduinus . . . Cuno de Montacut, Arnerus de Greiz . . . Henricus de Ase, frater eius Godefirus . . . Heribrandus et Walterus de Bulon.’ We do not have a verbatim record of Urban II’s speech or any report that predates the capture of Jerusalem. Accounts of what was said at Clermont may conveniently be compared in Edward Peters, The First Crusade: The Chronicle of Fulcher of Chartres and Other Source Materials, 2nd edn (Philadelphia, 1998), pp. 25–46. For Godfrey’s age, see Murray, ‘Godfrey of Bouillon (d. 1100)’, in Crusades Encyclopedia, I: 533–35 (p. 533). For the ‘landless younger son’ topos, see, for example, Norman Cohn, The Pursuit of the Millennium, 3rd edn (London, 1970), p. 61; countered in Jonathan Riley-Smith, The First Crusade and the Idea of Crusading (London, 1986), p. 43. Eustace’s wife was the daughter of King Malcolm of Scotland and Margaret of Hungary, who was the sister of Edgar the Ætheling: Murray, Crusader Kingdom, p. 30. Eustace’s route is in AA 2.21, p. 92; Runciman considered it more likely that Eustace travelled by sea: Steven Runciman, A History of the Crusades, 3 vols (Cambridge, 1951–1954), I: 147, n. 2. AA 1.1, p. 2. Alan V. Murray, ‘Domus Godefridi’, in Crusades Encyclopedia, II: 363; Murray, ‘Daimbert of Pisa, the Domus Godefridi and the Accession of Baldwin I’, in From Clermont to Jerusalem: The Crusades and Crusader Societies 1095–1500 (Turnhout, 1998), pp. 81–102. Albert’s list is AA 2.1, p. 60. Individual biographies are in Murray, Crusader Kingdom: Warner, Count of Grez, pp.  234–35; Baldwin of Bourcq, pp. 185–86. Murray, Crusader Kingdom: Rainald III, count of Tul and his brother Peter of Dampierre, count of Astenois, pp. 49, 219, 221; Dudo, lord of Cons, pp. 191–92. The brothers Henry and Godfrey of Esch: pp. 209, 205; AA 4.35, p. 302. Riley-Smith, First Crusade, p. 43. For a thorough discussion of numbers: John France, Victory in the East: A Military History of the First Crusade (Cambridge, 1994), pp. 122–42. Godfrey’s journey from the Ardennes to Constantinople and then on to Nicaea occupies Albert of Aachen’s book 2. He and his army followed the same overland route that the ‘unofficial’ crusades of Peter the Hermit and others had already travelled with generally disastrous consequences, as described by Albert in his first book. For Gottschalk see AA 1.23, p. 44; for Folkmar, F-E, p. 108; for Emicho, AA 1.27–29, pp. 50–58 and F-E, p. 108. For the fate of the different bands, AA 1.23–25, pp. 44–48 and 1.29, p. 56; F-E, pp. 144–46.

18  First Crusader 22 Nora Berend observed that ‘The First Crusade confronted Coloman with a novel problem’. For more on Coloman (1095–1116) see N. Berend, ‘Hungary in the Eleventh and Twelfth Centuries’, in The New Cambridge Medieval History, 7 vols in 8 (1995–2006), IV: 2, ed. David Luscombe and Jonathan Riley-Smith, pp. 304–16. Albert of Aachen is the sole souce for the diplomatic exchanges with the king of Hungary: AA 2.2–4, pp. 62–64. The letters that Albert reproduced verbatim are certainly invented, even if based on a known real correspondence. Similarly, the number 300 in the legation may not be accurate. 23 AA 2.4–7, pp.  66–70. ‘Household’ translates ‘familia eius’, which used to be interpreted as meaning Baldwin’s immediate family – see for example Runciman, I: 148 – but there is no evidence that he and Godevere had children. 24 Anna Komnene [AK], The Alexiad, 10.5, trans. E.R.A. Sewter, revised edn Peter Frankopan (London, 2009), p. 275. Albert’s account is at AA 2.7, p. 70. 25 The content of Alexios’s letter received at the council of Piacenza in spring 1095 is not known. See Jonathan Harris, Byzantium and the Crusades (London, 2003), pp. 47–51, for a discussion. Peter Frankopan, The First Crusade: The Call from the East (London, 2012), as the book’s title suggests, considers the First Crusade from the Byzantine perspective; for the envoys at Piacenza, pp. 99–100. 26 Albert of Aachen gives the fullest account of Peter’s crusade: AA 1.2–23, pp. 2–44. For its reception in Constantinople see AK, 10.5–6, pp. 275–79. Albert’s description of the terms of the agreement between Godfrey and Emperor Alexios is at 2.7, p. 72. 27 Hugh of Vermandois’s cognomen ‘Magnus’ was a corruption or mistranslation of maisné (or moins né), meaning ‘the Younger’: L. Bréhier, ed., Histoire Anonyme de la Première Croisade (Paris, 1924), p. 14, n. 3. For Emicho’s crusade, see references above for Clarembald’s presence at Antioch, AA 4.13, p. 268; for Drogo’s survival, Murray, Crusader Kingdom, p. 191. The exchange of messages with Alexios is at AA 2.8, pp. 72–74. 28 See AA 2.8, p.  72. Baldwin II of Mons was killed in 1098 when on an embassy to Alexios with Hugh of Vermandois: see also Murray, Crusader Kingdom, pp. 186–87. 29 For the confrontation between Godfrey and Alexios, see AA 2.9–11, pp. 74–78. Anna Komnene, writing much later, claimed that the stand-off started in Holy Week: AK 10.9, pp. 297–98; see also n. 30, p. 514, where Frankopan points out that fighting at that time would be a violation of the Peace of God and particularly atrocious. However, Albert’s dating is now generally accepted by historians, and the details that follow of the emperor’s forecasting wintry weather support this, as does the chronology of the ensuing campaign season in Asia Minor. 30 Cono was the eldest son of Gozelo, count of Behogne. He stayed with Godfrey’s army until the battle of Ascalon, 1099, and then returned to the Ardennes. His sons Gozelo and Lambert accompanied Cono on crusade. Murray, Crusader Kingdom, pp. 189–91. The dispute with Alexios continued: AA 2.12–13, pp. 78–80. 31 AA 2.13, p. 80. For Anna’s account, see AK 10.9, pp. 288–89. The anonymous Gesta Francorum compressed the multiple confrontations into a single encounter: Alexios ordered his Turcopoles and Pechenegs to attack and kill the foragers outside the city. Baldwin lay in wait and attacked the imperial troops, taking sixty prisoners, of whom he killed some and presented the rest to his brother Godfrey. The emperor was angry; Godfrey led his men out of the city. When the emperor’s men attacked, Godfrey drove them back into the city, killing seven. Five days later the emperor agreed that if Godfrey crossed the Bosphoros he would continue to supply him: Gesta Francorum et aliorum Hierosolimitanorum [GF], ed. and trans. Rosalind Hill (London, 1962), pp. 6–7. 32 Bohemond was the eldest son of Robert Guiscard, duke of Apulia and Calabria. In the early 1080s they had waged war on Alexios along the Dalmatian coast with some success, and Alexios was therefore justifiably suspicious of Bohemond. That Bohemond expected to arrive in March confirms the likelihood that the conflict between Godfrey and Alexios should be dated to Christmas.

First Crusader  19 33 AA 2.14, p. 82. 34 AA 2.15, p. 84. 35 For contemporaries on the oath and its implications, see AA 2.16, pp. 84–86; AK 109, p. 289; and Bartolf of Nangis [BN] chap. 5, ‘Gesta Francorum Iherusalem expugnantium’, RHC Occ, III: 491–543. For two different interpretations of the evidence see John Pryor, ‘The Oath of the Leaders of the First Crusade to Emperor Alexius I Comnenus: Fealty, Homage, Pistis, Douleia’, Parergon, n.s. 2 (1984), 111–41; Thomas Asbridge, The First Crusade: A New History (London, 2004), pp.  110–11. For the ambiguous status of Antioch, Colin J. Yarnley, ‘Philaretos: Armenian Bandit or Byzantine General?’ Revue des études armeniennes, 9 (1972), 331–53. 36 For Bohemond’s oath see AA 2.18, p.  88. However, a rather different version was given in GF, pp. 11–12. This account has been disputed, however, for its interpretation gave Bohemond grounds for keeping Antioch rather than handing it over to the emperor. See the seminal essay by August C. Krey, ‘A Neglected Passage in the Gesta and Its Bearing on the Literature of the First Crusade’, in The Crusades and Other Historical Essays Presented to Dana C. Munro by His Former Students, ed. Louis J. Paetow (New York, 1928), pp. 57–78. 37 Raymond’s oath-taking was described by Raymond of Aguilers [RA], Le “Liber” de Raymond d’Aguilers, ed. J. H. Hill and L. L. Hill (Paris, 1969), pp. 41–42. Both AA and AK described a ‘special relationship’ between Alexios and Raymond of SaintGilles: AA 2.20, p. 92; AK, 10.11, p. 295. 38 The ‘insolent Frank’ is at AK, 10.10, pp. 290–91. 39 AK 10.10, p. 292. 40 For an assessment of the early histories of the First Crusade and a discussion of the relationships between them, see Susan B. Edgington, ‘The First Crusade: Reviewing the Evidence’, in The First Crusade: Origins and Impact, ed. Jonathan Phillips (Manchester, 1997), pp. 57–77; John France, ‘The Anonymous Gesta Francorum and the Historia Francorum qui ceperunt Iherusalem of Raymond of Aguilers and the Historia de Hierosolymitano itinere of Peter Tudebode: An Analysis of the Textual Relationship between Primary Sources for the First Crusade’, in The Crusades and Their Sources: Essays Presented to Bernard Hamilton, ed. John France and William G. Zajac (Aldershot, 1998), pp. 39–69. 41 The best edition of Fulcher of Chartres [FC] is: Fulcher of Chartres, Historia Hierosolymitana, ed. Heinrich Hagenmeyer (Heidelberg, 1913); see also, A History of the Expedition to Jerusalem 1095–1127, trans. Frances Rita Ryan, intro. by Harold S. Fink (Knoxville, 1969). For Fulcher’s personal history, FC 1.7, pp. 163–64; 1.14, pp. 206–208. For his own revisions of the text, FC, pp. 46–47 (intro. by Hagenmeyer); Edgington, ‘The Gesta Francorum Iherusalem expugnantium of “Bartolf of Nangis” ’, Crusades, 13 (2014), 21–35. For Bartolf’s approach to Fulcher’s text, see BN, chap. 2, p.  492. The other witnesses to early or intermediate redactions of Fulcher’s history are Guibert of Nogent [GN], Dei gesta per Francos, ed. Robert B. C. Huygens, CCCM 127A (Turnhout, 1996), book 7 and a manuscript now in Cambridge University Library: Cambridge UL MS KK.vi.15, published in the footnotes to the RHC Occ edition of Fulcher of Chartres: RHC Occ, III: 321–418. 42 AA’s independence has long been debated. For a recent view, see Jay Rubenstein, ‘Guibert of Nogent, Albert of Aachen and Fulcher of Chartres: Three Crusade Chronicles Intersect’, in Writing the Early Crusades: Text, Transmission and Memory, ed. M. Bull and D. Kempf (Woodbridge, 2014), pp. 24–37. 43 The two most important sources of information about William of Tyre, his sources and working method are Robert Huygens’s introduction to his edition (Turnhout, 1986) and Peter Edbury and John Rowe, William of Tyre: Historian of the Latin East (Cambridge, 1988). 44 Ralph of Caen [RC], ‘Gesta Tancredi in expeditione Hierosolymitana’, RHC Occ, III: 587–716. There is a more recent edition by Edoardo D’Angelo: Radulphus Cadomensis, Tancredus, CCCM 231 (Turnhout, 2011), but unfortunately it is not divided into

20  First Crusader chapters in the same way as the more widely available RHC edition or the English translation, nor is there a concordance with the RHC edition. Both editions are based on the same, unique MS  and chiefly differ only in orthographical conventions. See The Gesta Tancredi of Ralph of Caen: A History of the Normans on the First Crusade, trans. Bernard S. Bachrach and David S. Bachrach (Aldershot, 2005), intro. pp. 1–14, for biographical information. 45 Caffaro, Annali genovesi di Caffaro e de’ suoi continuatori, ed. L. Belgrano, Fonti per la storia d’Italia 11–14 bis (Rome, 1890–1901); Caffaro, Genoa and the TwelfthCentury Crusades, trans. Martin Hall and Jonathan Phillips (Farnham, 2013); see also Elena Bellomo, A servizio di Dio e del Santo Sepolcro (Padova, 2003). 46 Frutolfs und Ekkehards Chroniken und die Anonyme Kaiserchronik [F-E], ed. FranzJosef Schmale and Irene Schmale-Ott (Darmstadt, 1972); Chronicles of the Investiture Contest: Frutolf of Michelsberg and His Continuators, trans. and annotated T.J.H. McCarthy (Manchester, 2014), pp. 41–81 for a discussion of the continuators. 47 Matthew of Edessa [ME], Armenia and the Crusades, Tenth to Twelfth Centuries: The Chronicle of Matthew of Edessa, trans. Ara E. Dostourian (Lanham, 1993); Ibn Al-Qalānisī [IQ], The Damascus Chronicle of the Crusades: Extracted and Translated from the Chronicle of Ibn Al-Qalānisī, trans. H.A.R. Gibb (London, 1932).

2 Nicaea to Edessa

Nicaea to EdessaNicaea to Edessa

The siege of Nicaea The Saljūq Turks had captured Nicaea (İznik, mod. Turkey) from the Byzantines in 1092 and made it the capital city of their sultanate of Rūm. The proximity of Nicaea to Constantinople made its recapture the primary goal for Alexios Komnenos; in addition, as the place where the first council of the Christian Church met in ce 325, it had religious significance for both Byzantines and crusaders. The western crusaders and a Byzantine contingent were shipped across the Bosphoros, and one after another they converged on the city to besiege and capture it. Godfrey’s army, with Baldwin, was the first to arrive there in May 1097. There seems to have been no concerted plan for the siege, and the piecemeal arrival of the different contingents could well have led to their serial destruction. In fact, they profited from the disaster that had befallen Peter the Hermit’s ill-disciplined followers in the same region the previous year. As Albert of Aachen recounted in detail, this vast rabble had ignored the emperor’s warnings and pillaged and provoked the Turks, who defeated and massacred them with ease.1 This misled the Turkish leader Qilij Arslān into seriously underestimating the Christian threat, and thus when Godfrey’s army arrived outside Nicaea the Saljūqs were far to the east besieging Danishmendid Melitene. Nicaea was a formidable city, built on the Roman plan with a grid of roads meeting at right angles and with gates to the north, south, east and west. Raymond of Aguilers described its fortifications thus: Nicaea is a city well protected both by nature and by human ingenuity. On the western side it has a very big lake that laps against the walls and on the other three sides a ditch filled with water that is the overflow from certain streams. In addition it is surrounded by lofty walls in such a way that it does not fear attack by any man or assault by any machine.2 The lake prevented a complete blockade, so as the crusading armies arrived they encamped on the other three sides.3 As first to arrive, on 6 May 1097, Godfrey and his Lotharingians took up position to the east of the city. Albert of Aachen provided the most detailed list of crusaders and their deployment around the city.

22  Nicaea to Edessa According to him, Bohemond arrived next and set up in the neighbouring position to the north; next to him was Tancred and then Tatikios, representing the emperor with a band of auxiliaries. Robert of Flanders and Robert of Normandy were described as next in line, which implies they were also encamped towards the north. If it is accepted that Albert of Aachen was listing methodically in a counterclockwise direction up to this point, unfortunately the system appears to have broken down just as he reached Baldwin. The sentence reads: ‘Warner of the castle of Grez, a soldier irreproachable in the art of war; Eustace the brother of the aforesaid Duke Godfrey, with Baldwin their brother, a most distinguished man and unbeaten in wars, likewise took up positions in line.’ Another dozen names follow in the same chapter, not all of them likely to be associated with Godfrey’s circle, and it is by no means clear whether Baldwin was now aligned with his eldest brother, Eustace, and the Norman French rather than the Lotharingians. The following chapter listed prominent and less prominent French participants, including Bishop Adhémar of Le Puy, the papal legate; Stephen of Blois; and Hugh, the king of France’s brother. Raymond of Saint-Gilles brought up his army in mid-May and was allocated the south gate. Qilij Arslān arrived soon after Raymond with a Turkish relieving army, and he attacked in an attempt to pre-empt Raymond’s encampment. However, the Turks were successfully repulsed, and Raymond’s army took up position. Another indication that Baldwin was positioned to the south of Godfrey, and possibly dissociated from him, follows immediately after this. While Raymond’s followers were still engaged in pitching their tents, Qilij Arslān attacked again, trying to break into the city through the south gate, the one that was being blockaded by Raymond of Saint-Gilles. The Turks were driven back by Raymond, aided by Baldwin ‘the duke’s brother’ and Baldwin Calderun.4 These were then reinforced by all the other leaders, and the Turks were routed. Albert of Aachen’s vivid description of the battle owed quite a lot to his imagination: he suggested the Turks were as numerous as the sands on the seashore, that they bore golden shields and had amazingly beautiful battle standards. However, other details that Albert described appear authentic: they were mounted archers carrying composite bows of horn and bone, and their horses were swift and skilled in warfare. After the battle around the south gate the siege became more attritional, with the Christians labouring to build siege engines of various kinds to try to breach the walls and the Turks defending the walls with hurling machines mounted on the towers. This stand-off could have gone on all summer, for the lake on Nicaea’s western side that could not be blockaded by the Christians enabled the Turks to bring supplies into the city more or less at will. Decisive action was required, a situation that exposed a potential weakness in the crusading armies, namely the lack of a recognised overall leader. The siege of Nicaea marked an important step in overcoming this, for the leaders were obliged to meet and agree an effective strategy. The sources are silent on whether Baldwin attended the series of councils that was held; the likelihood is that, as before, he acted as Godfrey’s lieutenant and stayed to keep order in the Lotharingian army. The solution arrived at by the leaders’ council was to close the blockade of the city by bringing ships overland

Nicaea to Edessa  23 and floating them on the lake. Whose idea this was is disputed in the sources: for example, the Gesta Francorum claimed the initiative came from the Latins while Anna Komnene gave entire credit for the scheme to Alexios Komnenos. Albert of Aachen presented a more balanced and quite probable picture of both sides contributing to a prolonged decision-making process from which the plan evolved. According to this a number of councils were held, and the leaders sent men to Civitot where they found the boats that had been requested from the emperor and brought them seven miles over dry land to float on the lake.5 The siege of Nicaea was by no means over, and it became even more violent, but with the boats arrived another of the emperor’s generals, Manuel Boutoumites, who was detailed secretly by the emperor to open surrender negotiations with the Turkish garrison. As Alexios intended, after a siege of five weeks the Turks surrendered to the Byzantines. Eye-witnesses reported with different degrees of resentment that the Byzantines were then able to keep the crusaders outside the city walls, depriving them of the opportunity to sack and loot that they saw as their due reward. However, Alexios compensated them with lavish gifts for the leaders and a distribution of copper coin for the rest. He also crossed the straits himself and insisted on a renewal of homage from the leaders; only Tancred refused to take the oath, so Baldwin was evidently among those who acquiesced.6 At this time some of Peter the Hermit’s followers who had been taken captive by the Turks were released. With the addition of these the crusade reached full strength, at least numerically, and there may have been as many as 100,000, including non-combatants.7

The battle of Dorylaeum After leaving Nicaea the combined armies marched into Asia Minor in the heat of high summer, late June 1097. The difficulty of feeding so large a force became apparent almost immediately, and it was decided to divide and follow parallel routes to facilitate foraging. This at least was Albert of Aachen’s explanation: And so for two days they were marching as a single armed column through the heights of the mountains and the narrow passes of the way, and then they decided that the army was so big that it should be divided, so that the people could live more freely and spaciously in the camp, and if a division was thus made there would be a lot more food and fodder for the horses.8 Not unusually, the distant Rhinelander appears to have a fuller and more credible account than the eye-witnesses: the Gesta Francorum claimed they set out on the third day in the dark and simply could not see well enough to stay together; Raymond of Aguilers blamed Bohemond for a rash decision to go his own way; Fulcher of Chartres, who originally wrote only that it happened disastrously, later admitted that he did not know the reason for the division, ‘They had for some reason, I  know not what, separated from us with a large number of men at a place where the road divided. For that reason we suffered an irreparable loss;’

24  Nicaea to Edessa Ralph of Caen presented both possibilities – a strategic decision and an accidental division  – and leaned towards the latter as the probable reason. Whatever the reason for the division, contemporary sources agree that a smaller force led by the Normans Bohemond, Tancred, Robert of Normandy and Stephen of Blois led off first while the larger part of the army, led by Godfrey, Robert of Flanders, Hugh of Vermandois and Raymond of Saint-Gilles, followed with the intention of staying within a mile of the others. In the absence of Baldwin’s name from either list, it is likely that he stayed with his brother in the following group.9 If that was the case, then he received news of the Turkish ambush of the leading contingent when Godfrey did and rode to the rescue with him. Accounts of the battle of Dorylaeum stress the fear inspired by the Saljūqs’ battle tactics: the rapid charge from the mountain slopes, the hail of arrows and the ‘fiendish’ war cries. Baldwin may have been less fearful than others who had not already experienced attack by the same army on the southern gate of Nicaea, when he was in the vanguard of the defence, but there is no reliable information about the part he played in the battle. Nevertheless, thanks to the speed with which Godfrey and his allies brought up the larger army, the Turks were routed and the Christians claimed a remarkable victory. The crusaders then continued their journey eastward as a single army. They were now crossing the Anatolian plateau in punishing heat, and the different narratives contain harrowing accounts of their suffering from thirst. Undoubtedly the leaders, including Baldwin, did not share the desperation of the lower ranks, some of whom died of thirst and others of drinking too much and too rapidly when a source of water was found. They also found that the Turks had resorted to a scorched-earth policy, and this again raised the question of splitting up to alleviate the pressure on the land.10

Cilician adventure or Cilician campaign? As the expeditionary force marched east, two of the more junior leaders, Baldwin of Boulogne and Bohemond’s nephew Tancred, struck out into Cilicia, away from the main body of the army. They appear not to have been working in concert but each to have had his own purpose. Sir Steven Runciman claimed, without adducing any evidence, that ‘Baldwin had already decided upon an Armenian state’ and that he left at the same time as Tancred because ‘it would be unwise to allow any other western prince to be the first to embark on an Armenian venture’. Runciman’s belief was that both of them wanted to found lordships in the East, which may well have been the case since each was campaigning in the shadow of a senior member of his family and both had slender prospects in the West. However, the inference is that these were impulsive young men engaging in a rather frivolous and unauthorised enterprise. (It should be noted, however, that while Tancred was around twenty at the time, Baldwin was in his thirties.) John France, in contrast, has suggested that at this stage in the crusade ‘what we see is the development of an Armenian strategy’. The crusaders, according to this theory, were encouraged by Alexios to make common cause with the Armenians, who would welcome their approach and eject their Turkish garrisons. As Alexander Beihammer has pointed

Nicaea to Edessa  25 out, the population throughout Anatolia, conquered by the Turks comparatively recently, was still overwhelmingly Christian and Greek speaking, so this was not an unreasonable expectation.11 Baldwin and Tancred may either or both have undertaken their excursions into Cilicia as part of such a strategy. Thomas Asbridge also rejected the idea that their expedition ‘was simply a selfserving treasure hunt’. Instead he perceived it as a ‘carefully conceived policy’ of the leadership. The route taken by Baldwin and Tancred was shorter than the more northerly route of the main crusading army, but it traversed two narrow defiles where there might well be ambushes, the Cilician Gates and the Belen Pass. In between were the plains of Cilicia where they might hope to find support from native Christians. The probable aim of despatching the younger men by this faster route was to establish alliances with such people and to set up a supply network for the crusaders’ next military objective, which was the capture of Antioch in northern Syria, a powerful city that could not safely be bypassed, and which Alexios was anyway very keen to recapture for Byzantium.12 This provides a plausible rationale for the Cilician expedition, but if the choice of both Baldwin and Tancred was indeed made by the leadership then it is, retrospectively, open to question. They set off independently, each with a relatively small contingent of troops – Baldwin had 300 to 500 and Tancred less than half as many – and came to quarrel to the point of armed conflict. Possibly their predisposition to conflict was a factor in detaching them from the main army, where each was in the shadow of his powerful older kinsman and under-employed as a result. It is also quite likely that each of their elders, Godfrey and Bohemond, was equally loath to allow the other to establish a tactical advantage, and once one of them sent off his relative the other did the same.

Tarsus According to the anonymous author of the Gesta Francorum, the two set out at the same time, heading south towards the coast. Tancred, who had an Armenian guide, arrived first at Tarsus. As he approached the city the Turks came out to fight and were routed. Tancred then encamped before the city gate. Baldwin came up by another way and demanded that they share the city, which Tancred (not unreasonably) refused to do. Overnight the Turkish garrison fled, and the citizens shouted to the crusaders to come on in. In the morning, the chief men told Baldwin and Tancred, who were still quarrelling, that they wanted Tancred as their ruler because it was he who had defeated the Turks. Baldwin – incidentally described as ‘the extraordinary count’ at this point – continued to argue and challenge Tancred, who recognised that Baldwin commanded the stronger forces and was forced to leave. He went on to take Adana and Mamistra and many castles. The Gesta author then returned to his main narrative strand.13 Much longer and more detailed descriptions are to be found in the histories of Albert of Aachen and Ralph of Caen. Neither was a participant in the First Crusade, but each incorporated testimony by people who were: Albert from Lotharingian followers of Baldwin; Ralph from Tancred and his circle. Some embroidery can be discerned

26  Nicaea to Edessa in both accounts (it may have been by the authors or by their informants), which makes their reconciliation an interesting exercise. Albert of Aachen’s Historia agrees with the Gesta Francorum that Baldwin and Tancred went ahead while the main army lingered on the plain of Pisidia, first for rest and recreation in more pleasant countryside and then delayed by the serious wounding of Godfrey by a bear. Tancred made better progress by keeping to the main road through the mountain pass known as the Cilician Gates, while Baldwin got a bit lost in the mountains and he and his men – and more seriously their horses – began to suffer from shortage of supplies. So Tancred was the first to arrive at Tarsus, a town near the coast that retained some renown because of its biblical association with St Paul. It was a desirable objective in other ways: The town was walled on all sides, proper and suitable for its inhabitants with meadows and streams, lying in fertile plains. Its walls were so marvellously strong that it might be thought that it could never be conquered by human forces, unless God helped.14 The idea that Tarsus was a strategic objective is reinforced by the information, supplied by Albert of Aachen, that Tancred had already made a useful Armenian contact: There a certain Armenian, who had spent some time with Tancred and had become acquainted with him, promised to suggest to the townspeople, who were weighed down by the heavy yoke of the Turks, that they should give up the town into Tancred’s own hands secretly and without the Turks’ knowledge, if it so happened that they found the right time and place.15 But the Armenian inhabitants of the town were more afraid of the Turkish garrison than they were of the westerner. Tancred’s response was to forage around the town to provision his siege, and then he encamped around the walls. He proceeded to engage directly with the Turkish garrison, taunting them (according to Albert of Aachen) and threatening that if they held out till his uncle the great Bohemond arrived with the main army the town would be conquered like Nicaea. The Turks believed him and surrendered on condition that they would be protected from looting – the usual fate of captured cities – and would be allowed to retain the town when Bohemond arrived. Tancred’s flag was flown from the citadel. This was the sight that greeted Baldwin when he gained his first clear view of the town. He and his band of men, which included his comrades Peter of Astenois, Rainald of Toul and Baldwin of Bourcq, had become lost on the mountain paths and had run out of provisions. From the top of a mountain they saw Tarsus and the tents around it, but from this distance Baldwin jumped to the conclusion that it was a Turkish encampment. Tancred, in his turn, spotted the men on the mountain and thought they were part of a Turkish relieving force, as did the Turks within Tarsus. It is very likely, of course, that this comedy of errors contained elements of exaggeration, a good story reported to Albert, who was dependent on the oral

Nicaea to Edessa  27 accounts of returning crusaders. According to the Gesta Francorum Tarsus was not captured by Tancred before Baldwin’s arrival, but Fulcher of Chartres, who later travelled with Baldwin, corroborated Albert’s story in outline: ‘He took [Tarsus] away from Tancred, who had introduced his own men with the consent of the Turks.’ Ralph of Caen, who claimed to have gained his information directly from Tancred, described the same sequence of events from his patron’s point of view in highly embroidered, even fulsome, language, reinforcing the story that Tancred besieged and captured the town and was in possession when Baldwin approached. Ralph even agreed that when Baldwin’s army was spotted it was taken for a Turkish army.16 At this point it is advisable to consider Ralph of Caen’s standing as a source of evidence: as previously observed, Ralph did not arrive in the East until 1108, but he was able to speak directly with participants in the First Crusade, including his hero, Tancred, and Tancred’s uncle, Bohemond. After both these men were dead, he is thought to have resided in Jerusalem in the 1110s, when Baldwin I was king. This may well explain an unexpected interpolation in the middle of his account of the quarrel between Tancred and Baldwin over Tarsus. Baldwin’s arrival prompted a eulogy: [Baldwin], the third of Eustace’s sons after Duke Godfrey and Count Eustace, had picked out many companions for himself from the handful of thousands: among them Count Cono was the most notable. He also increased these forces with Norman knights, whose land he himself governed under Count Robert. In addition he did not lack his own youthful following, as whoever was keenest of the keen among the armed men had chosen him as leader: a man generous with money, devoted to the study of military affairs, humble in his speech, outstanding in his magnanimity; in appearance he had the figure of a nobleman from head to toe: ‘Look,’ you would say, ‘Nature has carved him into a knight with her own hand.’ No wonder that his period of life was so adorned with talents, a life he had been born into from the sceptre of Franks, and was to leave from the sceptre of Jerusalemites; and – as becomes more brilliantly clear – drawing his noble ancestry from King Charlemagne, he was drawn by divine destiny to sit upon the throne of David. Therefore by right and by merit he surpassed Alexander, for Charles distinguished his birth and David his death; and his sword ought not to become blunt, whose cradle and grave thus shone. His burning nobility, beyond torches, had separated from the great army those whom he knew to be more fervent comrades: and they numbered some five hundred knights and two thousand infantry: when they marched from the mountain onto the plain the son of the marquis [Tancred] saw our sort of arms and thought they were bringing help, not realising they were bringing harm; he showed [the new arrivals] yesterday’s stratagem, battle and victory, he told them about the spoils, showed what he had told, offered what he had shown; he divided equally with these latecomers who were uninjured the whole of what he had captured [at some cost?] before their arrival.17

28  Nicaea to Edessa On the surface this passage praised Baldwin unreservedly, but it was bracketed by hints of criticism: there was an implicit reproach in the first sentence, that Baldwin had seriously weakened the main crusading force by depriving it of its best soldiers, and there was more open censure in the last, which described Tancred’s generosity in sharing equally with Baldwin spoils he had not earned. However, as Ralph went on to explain, Baldwin was not satisfied with a half-share. He claimed that the Turks had only yielded because they were intimidated by the appearance of his own army and therefore the spoils should be his. Tancred was forced to concede, after a fierce argument with himself reported by Ralph of Caen: should he fight a fellow Christian? Should he take refuge within Tarsus and ally himself with the Armenian townsmen? Should he leave altogether? In the event Tancred stayed where he was, encamped outside the walls. Albert of Aachen’s account of the same events essentially agrees with Ralph’s, but the story was told more entertainingly.18 When Tancred spotted Baldwin’s men on the mountain he took them for Turks and prepared to fight. The Turks assembled on the town walls also thought it was a relieving force; they taunted Tancred and claimed they had only pretended to negotiate to keep him there to be destroyed by the newly arrived reinforcements (how the crusaders understood the taunts was, of course, not explained). They also set up a din of trumpets and horns as Tancred rode to meet, and fight, the newcomers. It was only a matter of time, however, before ‘the banners of Christianity were recognised on both sides, and friends and fellow countrymen were seen and they dissolved into tears of joy’. At once the forces combined and Baldwin’s men camped alongside Tancred’s ‘by common consent’, and they shared the cattle ‘they’ had foraged in the mountains – who in fact had collected this booty was not explained. It was only on the following morning that Baldwin and his men saw Tancred’s standard flying from the citadel of Tarsus  – very likely the Turks had taken it down when they thought they were about to be relieved and hastily re-erected it when they realised the advancing army was Christian. The reaction of Baldwin and his men was extreme: they insulted Tancred (and his uncle Bohemond), likening them to ‘dirt and dregs’. Fortunately, wiser men intervened, and it was decided to send a legation jointly to consult the Armenian townsmen on their choice of overlord. Unfortunately for Baldwin, the townspeople unanimously decided they would prefer Tancred. According to Albert this was because they knew Bohemond’s warlike reputation while Duke Godfrey’s was new to them. It is clear that Albert perceived the dispute as part of a larger conflict between two of the crusading armies, with Bohemond and Godfrey sparring vicariously through nephew and younger brother respectively. How widely Albert’s interpretation was shared either among and within the crusading armies or by his contemporaries in the West cannot be established; it is possible that he was rationalising from his knowledge of later events. Predictably, Baldwin reacted angrily to his rejection by the townspeople and made a vituperative speech to Tancred, the Armenians and the Turks (employing an interpreter, as Albert carefully explained). His main theme was the many ways in which Godfrey outshone Bohemond and Tancred: his rank; his ancestry; the

Nicaea to Edessa  29 esteem he was held in by the whole army. He even claimed that Godfrey had been made its ‘chief and lord’, although such an election is not corroborated by any other source, so it may have been invented by Baldwin or Albert. Baldwin moved on to make threats against the town – Godfrey would destroy it by sword and fire – and against Tancred. At the same time, he promised gifts in return for compliance in removing Tancred’s standard and replacing it with his own. The townspeople and Turks were evidently impressed, for they threw down Tancred’s standard and replaced it with Baldwin’s. Tancred recognised that his was the weaker side and withdrew. He went to try his luck at nearby Adana, but found that at first he was barred from entry by its ruler; Albert wrote it was by the Burgundian Welf who had gone before the crusading army and got there first, Ralph of Caen that it was by an Armenian Christian called Ursinus.19 Tancred was eventually admitted, and he retained Adana, suggesting that Baldwin did not consider it worth fighting for. Meanwhile Baldwin was trying to establish a hold on Tarsus, but the Turks and the townspeople were suspicious and allowed him only two towers for his use, while his men were billeted through the town. The Turks said they would stay in Tarsus until Duke Godfrey arrived. Almost as soon as this was settled, a group of Bohemond’s followers and other pilgrims – Albert said there were 300  – arrived outside the town and begged for admittance. They claimed that they had become separated from the army and were exhausted and lacked provisions, although it is likely that, as Asbridge has suggested, they were sent by Bohemond as reinforcements for Tancred. Baldwin’s men pleaded for the new arrivals, because these were Christian brothers, but Baldwin was unmoved. His excuse was that he had promised the Turks and Armenians that he would admit only his own men until Godfrey arrived. The following description is so vivid and so little to Baldwin’s credit that it surely must have been recounted by one of his men. Baldwin’s men who were billeted inside the town felt so sorry for the newcomers that they let down bread and sheep to them from the walls to alleviate their hunger. Tancred’s followers ate well and then slept deeply, as did Baldwin’s men inside the walls. While all the Christians slept the Turks made a plan, and, leaving only 200 lower ranked men to garrison the walls, 300 took their valuables and other possessions and sneaked out of the city. They fell upon the sleeping crusaders outside the walls and slaughtered them. In the morning Baldwin’s men went to the walls to see if their comrades were still there and saw the evidence of the massacre. The first that Baldwin knew of it was the uproar that followed as his men rushed to arms, denouncing the Turks for their treachery and killing those left within the town. He rode down from his tower and tried to impose order before he learned what had happened. He found that he was blamed for it: But the uproar was becoming more and more violent, and the people were very angry at the murder of Christians and shouting that Baldwin was guilty of this massacre through his fatal advice. The tumult and discharge of arrows became so fierce and so great against him that he was forced to enter the tower for refuge, driven by the necessity of saving his life.20

30  Nicaea to Edessa Baldwin’s reaction to this mutiny of his men, after panic and anger, was to repeat his previous defence, that he had promised the Turks and Armenians that he would not admit any other Christians until Godfrey arrived. He also claimed that he was ignorant of the Turks’ cruelty. He apparently persuaded his men, and they then attacked the lowly Turks who had been left in the town and slew them. Their anger was fuelled by claims that high-born women of the town had been mutilated by the Turks when they would not submit to rape. This episode, narrated with such verisimilitude by Albert of Aachen, provides more insight into Baldwin’s character and circumstances than Ralph of Caen’s eulogy, although it too is not unambiguous. On the one hand Baldwin presented himself as Godfrey’s loyal brother who was holding Tarsus on his behalf, and he regained control in a dangerous situation by putting his case convincingly to his men. On the other hand he could be seen as naive, because he trusted the Turks to keep their side of the agreement; uncharitable, because of his treatment of fellow Christians; and ruthless, if he had any idea at all of the danger he exposed them to. An absent writer’s interpretation of another’s feelings is not to be trusted, and we do not know how close Albert’s informant was to Baldwin, but Albert depicted the count as beside himself with rage, although whether at the Turkish treachery or his men’s mutiny is unclear: ‘After the fierceness of his feelings had died down and he had returned to himself, he defended himself on all charges to placate the people.’21 The events at Tarsus revealed certain character traits: the strong sense of self regard and self-control that had been hinted at over the hostage question in Hungary, but, even more emphatically, Baldwin’s undoubted ruthlessness that was to mark aspects of his rule both in Edessa and in Jerusalem. A few days later, there was a demonstration of another essential attribute of an effective commander: good luck. From the walls of the city Baldwin’s men spotted a large and miscellaneous fleet out at sea – fancifully, Albert described their goldcovered masts shining in the sun. Men put ashore and set about dividing their booty. Unsure if the sailors were friend or foe, Baldwin’s men, both knights and infantry, armed and rushed down to the shore. When questioned, the sailors identified themselves as pirates who had left Flanders and Frisia eight years before; questioning in their turn, they found out about the great pilgrimage to Jerusalem and pledged to go there too. The captain of the pirate fleet was Winemer of Boulogne, ‘from the household’ of Eustace of Boulogne. Baldwin welcomed them and, of course, their booty and their baggage, and they all feasted for several days. Baldwin’s good fortune was that he was able to deploy 300 of the ‘pirates’ to garrison Tarsus and recruit 200 more to join his own troops when he left Tarsus soon afterwards to continue his march eastward. He was accompanied by Winemer on the first stage of the journey, but the pirate captain left him at the first stop, Mamistra, and apparently returned to his fleet, for he sailed on with it to Latakia and reportedly captured the city and briefly held it, although the evidence for this is very contradictory.22

Mamistra The next stage in the increasingly fierce rivalry between Baldwin and Tancred played out at Mamistra. Once again Tancred had arrived first and captured the

Nicaea to Edessa 31 city. How he achieved this is not certain: Albert of Aachen said that the Turkish garrison resisted strongly and Tancred broke down the walls and gates and destroyed the Turks ‘by cruel slaughter’, but according to Ralph of Caen, Ursinus of Adana suggested to Tancred that the conquest of Mamistra would be easy, and indeed the Turks fled when they saw his army approaching in the evening.23 The following morning the Armenian inhabitants welcomed Tancred’s ‘sweet yoke and light burden’ and submitted to his ‘fatherly rule’. In the very next sentence Ralph contrasted the condition of Baldwin’s army: ‘Meanwhile, laden with spoils, devoid of mercy, having abused its freedom and behaved unlawfully, the army of Count Baldwin had left Tarsus.’ Baldwin found some difficulty in crossing the rivers because only the bridges near the towns were maintained, and these were guarded by Tancred’s men. Baldwin was nevertheless obliged to approach and camp outside the city and to attempt to placate Tancred, even though he was aware of the enmity his behaviour at Tarsus had aroused, because of illness in his camp: Cono of Montaigu was too sick to travel, and others were suffering from sunburn and other ailments. Moreover, the city was well fortified, with a large and wellarmed population. Therefore, Baldwin, behaving pragmatically, restrained his army from looting and pillaging and sought to negotiate and purchase supplies. Tancred, who was always magnanimous (according to Ralph), agreed to allow the Mamistran merchants to trade, and all went well for a few days. Almost inevitably, however, disputes arose between the traders and the customers, and one of these – as Ralph recounted at length – sparked off a renewal of the fighting between the leaders.24 A dispute between an innkeeper and a camp follower came to blows, and soon swords were being wielded. The response from the leaders on each side was to clamp down on the other’s men. On Tancred’s orders Baldwin’s men who were in the town were put under guard if they were ill or beaten if they were well, and then the lower ranks were expelled from the town while the nobles were put in chains. Townsmen who were outside the town were treated similarly by Baldwin’s men. Meanwhile Baldwin’s knights attacked the gates of the city and Tancred’s knights opened them so that the area became a battleground and the leaders confronted one another. Baldwin had more troops and aimed to tempt Tancred’s forces outside; Tancred had the advantage of the height of the town walls to mount hurling machines and javelins. There was a standoff while each of the leaders hoped for the other to launch the first assault and thereby forfeit the moral high ground. Nevertheless, they were unable to restrain the youths on each side who skirmished in single combat. Had one side won more fights than the other it would have claimed victory as in an informal trial by combat, but honours were fairly even, Ralph reported. The stalemate was broken when Tancred’s kinsman Richard of Salerno, who was among the skirmishers, was unhorsed and taken captive. This seems to have shocked the more prudent men on both sides into suing for peace: Cono of Montaigu appears to have been a key figure in the process. Richard of Salerno was restored to Tancred, and there was a universal exchange of prisoners so that everyone ended up in his own camp. Ralph of Caen claimed that Baldwin recognised it was time to move on and ‘went off to conquer’ while Tancred stayed for a while in Mamistra. It is likely that Baldwin’s departure was stipulated as part of the peace settlement.

32  Nicaea to Edessa Ralph of Caen’s account of the capture of Mamistra is surprisingly vivid and circumstantial, considering that he claimed to have heard it from Tancred. It seems doubtful in fact that Tancred was aware that the conflict was caused by a dispute between an innkeeper and a camp follower, but we may charitably suppose that this was the story he, or one of his circle, was telling some ten years after the event. Albert of Aachen evidently received a different report of the conflict at Mamistra.25 There is agreement insofar as Tancred is shown distributing largesse and ‘concerning himself with the security of the city’ after its capture. Baldwin arrived, accompanied by Baldwin of Bourcq and Giselbert of Clermontsur-Meuse, another Lotharingian, and he and his men pitched their tents in an orchard next to the city. Richard of Salerno took this ill and incited Tancred to attack Baldwin because he had been responsible for Tancred’s loss of Tarsus. There was a fierce pitched battle, but Tancred’s men saw that they were outnumbered and ran away. Richard of Salerno and Robert of Anzi, another prominent Norman from southern Italy, were captured, and many others were wounded or killed outright. Of Baldwin’s knights the only one taken prisoner was Giselbert. Remorse for fighting fellow Christians overtook both sides only on the following day, and then ‘on the advice of the leaders’ they made peace, and prisoners and booty were restored. Baldwin left with his 700 cavalry (and presumably his infantry too). There is sufficient agreement between the two accounts to see that they were describing the same event, but naturally each writer was keen to show his hero in the better light.

Coming of age The so-called ‘Cilician adventure’ showed Baldwin operating as his own man, and a leader of men, away from the overshadowing presence of Godfrey, although Godfrey probably gave his blessing to the campaign that would secure the territory around the Cilician Gates and to the rear of the main crusading army as it advanced on Antioch. Tancred, whose uncle Bohemond probably also looked favourably on the enterprise, initially showed himself to be the more effective commander: he took Tarsus while Baldwin was still trying to find the city. Baldwin used threats and his greater force of numbers to take the city from him and repeated the exercise at Mamistra. A ruthless streak in his character, hinted at in his dealings with Tancred, began to emerge more fully at Tarsus, where he was indirectly responsible for the deaths of exhausted stragglers whom he refused to admit to the town. It was to be out in the open at Turbessel when he tortured and threatened a horrible death to the traitor Bagrat. It should be noted that Baldwin’s bullying tactics were deployed against his Christian allies throughout the expedition. The campaign is noteworthy in the wider political context. The crusaders’ conquest of the strongholds in Cilicia was made comparatively easy by the tenuous hold of their Turkish overlords on them and the eagerness of the Armenian Christian inhabitants to dislodge the Turks. Unlike the fighting for the major cities, Nicaea before and Antioch afterwards, there appears to have been no determined

Nicaea to Edessa 33 effort by the Turkish garrisons to hold these lesser towns against the crusaders; they withdrew and presumably regrouped to fight more winnable battles. Moreover, if, as some historians claim, the Cilician campaign was conceived with a strategic purpose and undertaken with the approval, perhaps even at the suggestion, of the Byzantine emperor, then it is noteworthy that Alexios does not appear to have made any immediate claim for the restitution to Byzantine power of the lands captured. This is probably because he had his eyes on the greater prize of Antioch and believed he could tidy up such minor places once northern Syria was back in imperial hands, as indeed he would attempt to do. But it is likely he also recognised the unpopularity of Byzantine rule with the Armenians: they had no wish to be ruled from Constantinople, and a Latin Christian ruler may well have seemed to offer an expedient way of allowing them to assert and maintain a degree of independence from both Byzantines and Saljūqs. While Baldwin and Tancred were ranging through Cilicia, the main army moved towards Antioch, led jointly by Godfrey, Bohemond, Raymond, Robert of Flanders, Robert of Normandy and Adhémar of Le Puy. They stayed for a few days at Marash. Here too the crusaders found that their reputation had gone ahead of them: the Turks had fled and the Christians were prepared to trade with them, so the crusaders camped before the city walls and did not try to exert force against the town. During this brief period Baldwin returned to the main army. For what happened then we are dependent on the testimony of William of Tyre, writing much later, and as the information is not found in his known sources – Albert of Aachen and Fulcher of Chartres – it may have been an invention. This much was implied by William’s biographers, who pointed out that after Baldwin’s clashes with Tancred in Cilicia that had led to Christian deaths, and before Baldwin’s achievement in securing Edessa, ‘William had to make an abrupt about-turn and portray him as a hero rather than as a villain.’26 As William recounted the Marash interlude, Baldwin had heard of Godfrey’s serious wounding by a bear, and he was worried by this and keen to know how he was faring. This makes evident sense if, as seems almost certain, Baldwin was Godfrey’s designated heir. Baldwin found that his brother was now recovering, and because everyone was praising Tancred’s conquests in Cilicia he decided to reassemble his company and set off in search of adventure again. However, those whom he invited to join him were reluctant to do so, for they had heard of his bad behaviour towards Tancred at Tarsus. According to William, it was only respect for Duke Godfrey that prevented Bohemond from punishing Baldwin for the injustice inflicted on his nephew. Baldwin was obliged to abase himself before his brother and to accept his reprimand, promising humbly that he would suitably compensate Tancred. William claimed that this was a first transgression: Since his offence was out of character and committed rather at the suggestion of another than at his own impulse, he obtained forgiveness and won back the favour of all. For he was a man otherwise praiseworthy in all respects, and there had been no other word of this kind about him.

34  Nicaea to Edessa If indeed Baldwin did humble himself in the presence of his peers, then it may have reflected a realistic acceptance of his subordinate position, but it is more likely that he saw it pragmatically as the price to pay to gain support for a continuation of his quasi-independent campaigning. The episode probably confirmed for Baldwin that he had no role in the main army that suited his character and ambition.27 While they were at Marash Baldwin’s wife Godevere (Godehilde) died of a longstanding illness. Albert of Aachen wrote that she had been entrusted to Duke Godfrey’s care; this may have been because she was already ill or because the sort of expedition Baldwin had undertaken was unsuitable for women. We have no idea how Godevere’s death affected Baldwin emotionally, but politically it was to be an advantage for the next stage of his career in the East since it freed him to contract a second marriage. It must be remembered that Godfrey’s presumed celibacy meant that Baldwin bore the great responsibility of perpetuating their dynasty. The marriage to Godevere was some years old and was apparently childless; now that Baldwin was widowed he could undertake a second, perhaps more prolific, marriage and could seek a politically advantageous one too. After the burial of Godevere and another casualty of sickness, Udelard of Wissant, the main army moved off towards Antioch again. However, Baldwin did not go with it, for he now gave up following Tancred and bullying him into giving up his gains in favour of heading further inland, towards the river Euphrates. Asbridge claimed that he was aiming to establish ‘a new Levantine lordship’ for himself, which seems likely in view of the outcome of Baldwin’s digression, although so far as the other leaders were concerned he probably cloaked his ambition by pointing out the advantage of establishing a buffer state and possible source of supplies for the crusading armies while they were engaged in besieging Antioch.28

Turbessel Since Nicaea Baldwin had kept with him an Armenian called Bagrat (or Pakrad) who was brother of the local lord Kogh Vasil, also known as Basil the Robber. Bagrat had previously been detained by Alexios Komnenos; Albert of Aachen implied that he had broken parole in order to accompany Baldwin. However, Baldwin recognised in him a man who was warlike and multi-talented, one who could prove invaluable because of his knowledge of Armenia, Syria and Greece. Bagrat now recommended to Baldwin that he attack Turbessel. This was a strongly fortified town with an Armenian population dominated by a Turkish garrison. The citizens drove out the Turks and surrendered the town to Baldwin ‘because the Armenians wished rather to serve under a Christian duke than under gentile power’. Baldwin placed his own men to protect Turbessel and then besieged and captured Ravendel. It was reported that this was an even easier conquest because the Turks had taken fright and fled before Baldwin arrived there. The success was repeated in other local towns.29 Baldwin rewarded Bagrat by putting him in command of Ravendel, but this proved a serious miscalculation on his part. Bagrat had no intention of governing

Nicaea to Edessa  35 the town under Baldwin; he installed his son there with instructions not to admit any of Baldwin’s men, while he himself stayed with the count. The leading Armenians around Turbessel, whom Albert of Aachen named as Fer and Nicusus, knew Bagrat of old and warned Baldwin that he was not to be trusted, so Baldwin tested Bagrat by asking for the fortress back. When Bagrat refused Baldwin imprisoned him and put him to torture, finally threatening to tear him limb from limb unless he surrendered Ravendel. Bagrat then sent to his son instructing him to hand over the fortress. Baldwin freed the treacherous Armenian and cut off relations with him, transferring Ravendel to the protection of his own men. He went on to consolidate his power in the region. It appears that Baldwin had seriously misjudged his associate’s character, but when he realised this he acted promptly and ruthlessly to assert his authority. None of the other sources mentions this episode, but there is a postscript in the Armenian source Matthew of Edessa: in 1117–1118, Count Baldwin II of Edessa ‘destroyed Kogh Vasil’s principality and forced all the nobles attached to his service to take refuge in Constantinople. He caused the ruin of another Armenian pri[n]ce named Bagrat – a brave man who resided in Ravendan, not far from Cyrrhus – by pillaging his territories’. If this was the same man, he apparently enjoyed freedom to exploit Ravendel for twenty years after his treachery and before the count of Edessa had the strength to subdue him.30 If it was indeed the same Bagrat, it is a reminder of continuing instability in the region. Christopher MacEvitt has made an excellent analysis of Bagrat’s first seizure of Ravendel, concluding that ‘the conflict had little to do with Baldwin and everything to do with a complex local Christian world of political goals, struggles and squabbles into which Baldwin and his knights were drawn’. He further claimed that Baldwin was unable to assert his authority in the region outside a few strongholds like Turbessel, but that ultimately Baldwin was adept at ‘navigating the complexities of a world of competing local warlords’, helped perhaps by its similarity to the political landscape of his homeland.31

Notes 1 For Peter the Hermit’s crusade, see Albert of Aachen 1.15–22, pp. 12–44 (defeat by the Turks, pp. 32–45); the fullest modern account is Jean Flori, Pierre l’Ermite et la première croisade (Paris, 1999). ‘Rūm’ was the Turkish version of the Latin Romania and is used here to avoid confusion with the modern European state. The Saljūqs applied it to land they conquered from the Byzantine (formerly the eastern Roman) empire. 2 RA, pp. 42–43 for the topography of Nicaea. 3 AA 2.22–24, pp. 94–102 for the deployment of the crusader armies around the walls. 4 Albert’s description of the Turkish army and the battle is at AA 2.27–29, pp. 106–110. He listed Baldwin, surnamed or nicknamed (cognomine) Calderun in the same group of knights besieging Nicaea as Baldwin of Boulogne, calling him ‘a man of great renown in battle’. The only other contemporary reference is in a letter of Anselm of Ribemont to Archbishop Manasses of Reims. Anselm asked for prayers for the dead, among them Balduinus Chalderuns, qui primus in Turcos militiam fecit: Heinrich Hagenmeyer (ed.), Epistulae et chartae ad historiam primi belli sacri spectantes (Innsbruck, 1901), no. VIII, p. 145. Baldwin Calderun was killed by a stone hurled from the walls of Nicaea that broke his neck: AA 2.29, p. 110.

36  Nicaea to Edessa 5 For different versions of the decision-making process, see GF, p. 16; AK, 11.2, p. 300; RA, p. 44; FC 1.10.8, pp. 187–88; AA. 2.32, pp. 114–16. 6 AK described the surrender and oath-taking ceremony: 11.2–3, pp.  300–305. BN reported Tancred’s refusal to take the oath (chap. 7, p. 495), but AK said he was eventually persuaded to do so by Bohemond. This is also clear from RC, who recounted the episode at great length: chaps. 17–18, pp. 618–20. 7 This is the conclusion of France, Victory in the East, pp. 124–36. 8 AA 2.38, p. 128. 9 The chroniclers struggled to explain the decision of the leaders to divide into two columns; see also RM, p. 25; GF, p. 18; RA, p. 45; BN chap. 7, p. 495; FC 1.11.5, p. 194; RC chap. 20, pp. 620–21. Some later writers explicitly included Baldwin and Eustace in the same group as Godfrey and Raymond: OV, V: 60 (based on Baldric of Bourgueil, Historia Ierosolimitana, ed. Steven Biddlecombe [Woodbridge, 2014], pp. 32–33) and WT 3.16, p. 215. It is probably a fair assumption rather than an established fact. 10 The author of GF, p.  18, referred to the Turks’ ‘diabolicum sonum’ and described them as shrieking ‘demoniaca uoce’. The hardships of the march across the Anatolian plateau in high summer were described by the eye-witnesses, GF, p. 23; FC 1.12.6– 1.13.1, pp. 199–201, and most vividly by AA 3.1–3, pp. 138–40. 11 Runciman, History of the Crusades, I: 201; France, Victory in the East, p. 190; Alexander D. Beihammer, Byzantium and the Emergence of Muslim-Turkish Anatolia, 1040– 1130 (Abingdon, 2017), p. 41. 12 Asbridge, First Crusade, pp. 140–42. 13 GF, pp. 24–25. The author described Baldwin as ‘mirificus comes’. 14 Albert of Aachen’s detailed account of Baldwin’s leaving the main army and his arrival at Tarsus is AA 3.3–7, pp. 140–50; this description AA 3.7, p. 150. 15 AA 3.5, p. 146. 16 These other versions are in GF, p.  24; FC 1.14.3, pp.  207–208; RC chaps. 34–38, pp. 630–34. 17 RC chap. 37, pp. 632–33. Note that there is a lacuna in the very last sentence of the speech (there is only one manuscript of RC). The reading ‘at some cost’ is therefore conjectural. For Cono of Montaigu, in the Ardennes, see Murray, Crusader Kingdom, pp. 189–90; see also pp. 61–62 for the composition of Baldwin’s comitatus. Note that Ralph was mistaken in thinking Baldwin was a vassal of Robert of Normandy; he did not hold land from Robert, but between leaving the church and joining the crusade he seems to have been largely resident in Normandy on his wife’s (or father-in-law’s) property, and this may have given rise to the error. Tancred’s agony of indecision is in RC chap. 38, p. 633. 18 Albert’s longer version of events at Tarsus occupies AA 3.7–13, pp. 150–58. Compare Asbridge, First Crusade, pp. 143–45. 19 AA 3.10, p. 154; RC chap. 40, p. 634. Ursinus’s Armenian name was Ōshin; for more on his position in Adana, see Beihammer, Byzantium and the Emergence, p. 292. 20 AA 3.13, p. 158. 21 Ibid. 22 Albert has two accounts of Winemer’s exploits, at AA 3.14, pp.  158–60, and 3.59, pp.  230–32. There are inconsistencies between the two that are examined and discussed in Appendix 1 to Ralph-Johannes Lilie, Byzantium and the Crusader States 1096–1204 (Oxford, 1993), pp. 259–76. 23 For the competing accounts, see AA 3.15, p.  160; RC chaps. 40–42, pp.  634–37. Ralph’s description is very ornate, and although ‘Lucifer’ would normally be Venus in her guise as Morning Star, he must have intended the Evening Star as marking Tancred’s arrival. Ralph, not Albert, refers to illness among Baldwin’s followers. 24 RC chaps. 43–44, pp. 638–39. Note that ‘caupo’ can mean an innkeeper or a market stall holder, while ‘lixa’ could be a particular sort of camp follower, a sutler, who sold on provisions.

Nicaea to Edessa  37 25 AA 3.15–17, pp. 160–64. 26 Edbury and Rowe, William of Tyre, p. 74. 27 The crusaders arrived at Marash (Kahramanmaraş, mod. Turkey) in mid-October 1097: AA 3.27, pp. 180–82. For Baldwin’s concern over Godfrey’s wound, see WT 3.26 (25), p. 229. His account of Baldwin’s apology is at WT 4.1, p. 233, but note first that the ‘another’ who had encouraged Baldwin is not clearly identified: Edbury and Rowe interpret the formula as a way of portraying Baldwin as weak and easily influenced. Second, WT’s English translators wrongly made this a promise of future good behaviour on Baldwin’s part rather than saying he had a good record in the past (the Latin is ‘fuerat’): Babcock and Krey, A History of Deeds, p. 187. For William’s motive in including the apology that may have had no basis in fact, Edbury and Rowe, William of Tyre, p. 74. 28 Asbridge, First Crusade, pp. 149–50. For Udelard, a member of the domus Godefridi, see Murray, Crusader Kingdom, p. 231. RC was undoubtedly mistaken when he wrote that Baldwin overcame Artah, only twenty-two kilometres from Antioch, RC chaps. 45–47, pp. 639–41. See Asbridge, First Crusade, pp. 157–58 for Artah’s surrender to a contingent led by Robert of Flanders. 29 AA 3.17–18, pp. 164–66: For the idea that Bagrat had broken parole: Albert reported that Baldwin entrusted Ravendel to a man ‘who had broken his oath to the emperor so criminally’ (AA 3.18, p.  166). Turbessel (Tilbaşir, mod. Turkey) and Ravendel (Revanda kale, mod. Turkey) are discussed in R. Gardiner, ‘Crusader Turkey: The Fortifications of Edessa’, Fortress, 2 (1989), 23–35. 30 AA 3.18, pp. 166; for Bagrat’s later suppression by Baldwin II, ME 3.74, p. 220. 31 Christopher MacEvitt, The Crusades and the Christian World of the East: Rough Tolerance (Philadelphia, 2008), pp. 59–63. MacEvitt suggests the Armenian name P‘er for Fer.

3 Count of Edessa

Count of EdessaCount of Edessa

Why Edessa? The same historians who perceived Baldwin’s exploits in Cilicia as the product of his personal ambition tended to characterise in similar terms his leaving the main crusading army and striking out towards the river Euphrates. However, even Runciman was prepared to concede that there could be a strategic purpose as well: ‘It is doubtful if he had any planned course of action beyond a general determination to found a principality upon the Euphrates, which might be of profit to himself and to the whole crusading movement.’1 Importantly, the territory was in the hands of the Armenians, who were not only friendly, but even eager to welcome an army of western Christians who would defend them against the Turks and at the same time enable them to maintain a degree of independence from their Byzantine overlords.2 They were already in communication with the crusaders, and it is likely that their appeals had been discussed at Marash and that Baldwin’s expedition towards the Euphrates was approved by his brother and the other leaders, mainly for its strategic value, but also, quite possibly, as a good way of directing Baldwin’s disruptive energies away from the ongoing crusade. Edessa was situated about 80 kilometres east of the river Euphrates on the plateau south of the Taurus mountains. The city’s strategic value thus lay first in its position on the route between the Turkish heartlands and Antioch, which was the next objective of the crusade. Possession of the city would enable its ruler to hold up a Turkish army coming south for the relief of Antioch. The city was near impregnable, with strong walls and the citadel high above the town. Its position made it a wealthy centre of trade. Paul Gindler called Edessa ‘the key to the crossing between north Syria and Mesopotamia’ and also pointed out that it lay on the old caravan road that linked Konya to Adana, Samosata, Harran and Nikephorion (modern al-Raqqa). Edessa’s region was famously fertile, too, well watered by springs and streams, with vines growing on the nearby slopes. To the south-east there were market gardens, orchards and olive groves, while the area to the east was ‘a rich granary’. If these resources could be brought under control, they could supply life-saving provisions to the armies besieging Antioch.3

Count of Edessa  39

The invitation from Edessa While the crusading army rested at Marash in October  1097, Fulcher of Chartres, priest and chronicler, ‘left the army and turned aside into the region of the left-hand province with my lord Count Baldwin, Duke Godfrey’s brother’. Why Fulcher decided on this course of action he did not say, but there is a hint that he was keen to experience a military adventure, for he called Baldwin ‘the best possible knight’.4 His decision means that from this point in Baldwin’s career we have the testimony of an eye-witness, but in the event Fulcher tells us little about the Edessan stage of Baldwin’s career. It may be that Fulcher censored his account because his admiration for Baldwin led him to pass over actions in Edessa that reflected poorly on him, but equally probably, writing in the 1120s, he considered them no more than a sideshow to the main crusade and a preliminary to Baldwin’s reign as king of Jerusalem. So instead, Fulcher focused on the main army and the siege and capture of Antioch and Jerusalem and pieced together the history of the main expedition from the accounts of others rather than focusing on events in which he participated. However, Bartolf of Nangis explained a little more about Fulcher’s ‘left-hand province’, which he probably gleaned from Fulcher’s early account: When he heard from certain native scouts that the land towards Edessa was very good and very fertile and very easy to conquer he went off there on his own with his men; he set out towards the Euphrates, which flows between Antioch and Edessa, and captured by force and cunning fortresses situated on that same river; he brought under his yoke the very best among them, the one that is called Turbessel – the Armenians who lived in it surrendered it to him peacefully; and very many other places that were subordinate to Turbessel were obtained by this man.5 As the two versions agree, news of the capture of Turbessel (recounted earlier in chapter two) spread far and wide and led to a message from Edessa that brought Baldwin to that city. Given Fulcher’s reticence about events in Edessa, it is fortunate that Albert of Aachen continued to record with a surprising level of detail whatever he could learn of the activities of Baldwin and other Lotharingians. Albert reported that the duke of Edessa sent the city’s bishop and twelve of the city prefects to ask Baldwin to come with his men and defend Edessa against Turkish attacks. In return he would share in both the duke’s authority and his revenues. Baldwin took advice – Albert does not say from whom, but quite probably his brother Godfrey – and then set out. Now he had only 200 cavalry with him, owing to the necessity of garrisoning Turbessel, Ravendel and other conquests. They marched quickly as far as the river Euphrates, but when he attempted to cross the river he was opposed by Turks and others who had been summoned (according to Albert) by the recently released Bagrat. Vastly outnumbered, Baldwin was obliged to retreat to Turbessel

40  Count of Edessa and wait until the enemy army had dispersed before making the journey, this time unopposed.6 The entire population of Edessa turned out to escort Baldwin into the city, ‘with trumpets and all kinds of music’. Only the duke was less than delighted, apparently not so much by his presence, for he had been privy to the decision to summon Baldwin to defend the city against its enemies, as by the manner of his reception. The duke, named elsewhere as Thoros or T’oros, was in a vulnerable position because he claimed authority as a Byzantine appointee and also recognised the suzerainty of the neighbouring Turks, both unpopular positions. Furthermore, he was an Orthodox Christian, whereas most of his subjects belonged to the Separated Armenian Church. William of Tyre, writing later but with more local knowledge than Albert of Aachen, described him as a Greek sent to rule Edessa when the whole territory was under Byzantine control. He was now old, feeble and childless. He was a ‘useless appointee’ who retained his position, William implied, only by default.7 Once the crusading army appeared, seeming to offer a viable and effective alternative and the possibility of independence from both the Byzantines and the Turks under a Christian ruler, leading citizens took the opportunity. How enthusiastic Thoros’s agreement was cannot be known, but he may have envisaged some sort of quasi-retirement as co-ruler with a Latin count. Once he met Baldwin he apparently recognised him and his modest army as a threat and offered him very grudging terms: he barred him from any position of authority and from receipt of an equal share of revenues or taxes. The duke would furnish Baldwin with mules, horses and weapons, but essentially he expected him to fight the Turks as a commander of mercenaries in return for gold, silver and silks. This was not what Baldwin had in mind, and he announced that he would return to Duke Godfrey; all he asked was a safe conduct for the return journey to the main army. When this became known to the prefects and people of Edessa, they prevailed upon the duke to retain the services of ‘such a noble man and so very strong a champion’.8 It is worth emphasising that the two contemporary Latin sources for events in Edessa, Fulcher of Chartres and Albert of Aachen, wrote independently of one another, making points of agreement between them the more convincing. William of Tyre had access to both of their accounts and was writing much later. He also brought to his narrative a particular agenda to do with Baldwin’s relationship as king with the patriarchs of Jerusalem, and this must be borne in mind when evaluating his testimony. Fortunately, there is another important and independent source for this period of Baldwin’s career: the Armenian chronicler Matthew of Edessa. Matthew was hostile to Baldwin, which increases his value as an antidote to the partisan Latin chroniclers. Matthew’s modern translator divided his chronicle into three parts: the first, describing events of 952 to 1051 and based on written sources, is less relevant for us; the second, 1051–1101, drew on accounts of eye-witnesses who were alive when Matthew was; for the third, 1101–1136, he had his own experience and observation as well as reports of others. Thus we may have high expectations of the quality of Matthew’s information, and these are largely validated. Dostourian, the translator (for whose work crusades historians

Count of Edessa  41 are exceedingly grateful), made a comparison with Latin chroniclers of the First Crusade that does, however, rank them rather differently from most scholars. For him Fulcher of Chartres is not only the most dependable, but as ‘[a]n intelligent and keen observer, he was meticulous in his chronology and endowed with an unusual talent for describing events’. Albert of Aachen, in his opinion, ‘falls far short of his contemporary Fulcher’, while William of Tyre is ‘the outstanding Latin historian of the Crusades’. A  closer familiarity with the Latin historians suggests that at least for this early period when Baldwin was in Edessa, William is the least important, while Albert has far too much detailed information – much of it corroborating Matthew’s chronicle – to be dismissed as inferior to Fulcher.9 Matthew of Edessa’s account agrees with the Latin sources to the extent that it was Baldwin’s success in capturing Turbessel that came to the attention of Thoros, whom Matthew called ‘the Roman [Byzantine] commander’. Thoros called upon Baldwin for help against the neighbouring Turkish emirs. When Baldwin arrived with sixty knights the citizens rejoiced and welcomed him, and Thoros ‘acted in a friendly manner’, Matthew perhaps hinting that this was a pretence and he was concealing less amicable feelings. Thoros gave Baldwin gifts, and they formed an alliance. There is no indication in Matthew’s chronicle that Baldwin objected to a subordinate role; in fact, a few days later Thoros ‘sent’ him along with an Armenian chief called Constantine to attack Balduk, emir of Samosata.10 Albert of Aachen’s history is in complete agreement that Baldwin’s first engagement was against Balduk of Samosata but explained that the nature of the relationship with Thoros had changed significantly. Under pressure from the twelve prefects and other leading citizens Thoros reluctantly ‘made Baldwin his own adopted son according to the custom of that region and people’. The custom allegedly involved taking Baldwin under his shirt and binding him next to his chest, with an exchange of pledges.11 This might be thought fanciful on Albert’s part, especially as such a ceremony was not mentioned by Fulcher of Chartres or Matthew of Edessa, but a version also found its way into the contemporary account by Guibert of Nogent. Guibert explained that the ruler of Edessa was an old man and he and his wife were childless. Until that time he had protected Edessa by buying off the Turks, but when he heard of the Franks’ successes he conceived the idea of adopting one of their nobles, who would then defend Edessa and its environs: Meanwhile it came about that one of his household knights, who was aware of his desire, had a conversation with Baldwin himself. When this man gave him the hope of achieving a dukedom if he allowed himself to be adopted by the aforesaid old man, Baldwin believed him and went to Edessa in the knight’s company. He was received very warmly by the duke, far beyond Baldwin’s hopes, and he was adopted as son by both [the duke and his wife]. Moreover, the method of adoption according to the custom of this people is said to have been such: [the duke] had him get naked into a linen undershirt, the sort we call a nightgown, bound him to himself and confirmed it all with a kiss on the lips; his wife also did the same thing afterwards.12

42  Count of Edessa This is one of a number of striking or picturesque incidents that are shared uniquely by Albert of Aachen and Guibert of Nogent. Jay Rubenstein has discussed various possible relationships between the two chroniclers: one could have copied from the other; they could have had a shared written source; or they used oral sources. Rubenstein’s preferred hypothesis is that they ‘both heard performed one or more chansons about the crusade’, rejecting the idea that they were ‘independently debriefing several returning crusaders and coincidentally learning from them the same several points’. However, the nature of the shared information and the significant differences in detail – as in the example above – are more suggestive of anecdotes that would be just the sort of stories survivors would tell on their return to western Europe. Notably, because the adoption story is not to be found in either Fulcher of Chartres or Matthew of Edessa, it may well be no more than gossip.13 Whatever ritual Baldwin had to endure to gain power in Edessa, he clearly considered the city worth it. Edessa was an ancient city, reputed to be the Old Testament’s ‘Ur of the Chaldees’ and the birthplace of the patriarch Abraham, but also one of the earliest places to recognise the divinity of Christ and the site of an early Christian council (in ce 197). The chroniclers generally referred to the city as Rohas or Rohais, a form of its Seleucid name, Urfa, rather than Edessa, its Greek name. Politically, as the events already outlined demonstrate, Edessa paid lip service to Byzantium but paid protection money to the Turks, thereby maintaining a precarious independence. The appearance of the Latin Christians in northern Syria probably seemed to the Edessenes a practical way to resist the Turks without becoming beholden to the Byzantines. This is the story told by two Syriac chroniclers, Michael (1126–1199) and an Anonymous whose secular history ends in 1234. These two accounts, which are believed to have been independent of one another, agree that the governor (whom they called by the Greek name Theodoros) learned of the presence of the crusaders and sent a message on his own initiative to the leaders (Michael), or specifically to Godfrey and under pressure from the townsmen, asking for ‘troops to guard the land’ (Anonymous). Michael’s much shorter account relates that Theodore promised to deliver the city to Duke Godfrey, who rejoiced, saying, ‘Just as Edessa believed in Christ before Jerusalem, so she has been given to us by Our Lord Christ before Jerusalem’, and sent his brother Baldwin to rule there. In the longer, anonymous account the leaders rejoiced and sent Baldwin, ‘a pious man who feared God and was a mighty warrior’. The writer went on to explain: ‘At that time Edessa was a very great city, filled with the voices of many peoples, and famed for its clergy, monks, and much people. Its territory was full of villages, villas, and hamlets.’14

The coup Matthew of Edessa and Albert of Aachen both described Baldwin’s first campaign for Thoros and the Edessenes, against Balduk of Samosata, and although they differed in details, it was a heavy defeat for the Christians. Matthew said there was ‘a severe slaughter of as many as one thousand men’. Albert, as so often, had a more detailed narrative in which Baldwin took an army of 200 of his own men and all

Count of Edessa  43 the Edessene cavalry and infantry to Samosata, where Balduk had seized the citadel and taken leading citizens’ sons hostage. They attacked the fortress, but it was strongly defended. Albert contrasted ‘the innumerable band of the effeminate Armenian townsmen, who were fighting carelessly and slowly, [and who] fell in that place’ with ‘six of Baldwin’s excellent and vigorous soldiers [who] were shot with arrows and died’. Baldwin realised that the citadel of the fortress at Samosata was impregnable, so he garrisoned a nearby fortress and returned to Edessa with only twelve of his Franks. Baldwin identified the strength of Samosata’s defences as the main reason for his retreat, and he must have made a convincing case back in Edessa, for his own reputation appears not to have suffered. Matthew gave the information that the attack on Balduk of Samosata was made in alliance with an Armenian chief, Constantine of Gargar (Kaṙkaṙ), whom both writers then implicated in the plot that followed to depose Thoros and install Baldwin in his place. Their accounts are detailed; Fulcher of Chartres, who omitted the campaign against Balduk altogether, simply wrote: And when we had stayed there for fifteen days, the citizens of Edessa plotted wickedly to kill their prince and raise Baldwin in the palace to govern the land. For they hated [Thoros]. It was said and it was done. Baldwin and his men were very sad because they were unable to obtain mercy for him. Matthew and Albert not only recounted the plot in more detail, they differed radically in their assessment of the extent of Baldwin’s complicity.15 It should be said that Matthew was wholeheartedly on Thoros’s side, and hostile to Baldwin, because he held that the curopalates Thoros had saved Edessa from the Turks by bringing in the Franks. In his version, there were forty conspirators, ‘perfidious and evil-minded men’, who approached Baldwin, who approved their plot, and also the Armenian chief Constantine of Gargar. Then the traitors stirred up the citizens against Thoros, attacked his officers’ houses and seized the upper citadel. Thoros agreed to surrender the citadel and the city to the rebels in return for a safe conduct to go with his wife to Melitene. Matthew listed the sacred objects on which Baldwin vowed that he would not harm Thoros and wrote that Thoros was given a written safe conduct. The following day the townspeople attacked Thoros and threw him from the ramparts; they put him to the sword and dragged his corpse ignominiously through the city.16 The anonymous Syriac chronicler, writing a generation later, told the story more as a spontaneous uprising of the people, stirred up by ‘certain lewd townsmen’. The writer stressed that this was not from love of the Franks, but rather from hatred of Thoros. The citizens gathered below the upper castle and attacked Thoros as he approached, chasing him to his lower castle by the east gate. When they attacked him there he asked for a safe conduct, which they duly swore, and opened the castle gate. They broke their oath, beat him up and lowered him, naked but for a loincloth, from the city wall. Unfortunately there is a lacuna in the manuscript at this point, but the halfsentence that is extant appears to say that Thoros had brought destruction on his house. The account concluded: ‘Baldwin took all that belonged to [Thoros] and the two castles.’ There is no indication that Baldwin was involved in the plot.17

44  Count of Edessa Albert presented Thoros in a different light from Matthew. According to Albert, the citizens hated Thoros because he had taken huge quantities of gold and silver from them and, if anyone opposed him, Thoros conspired with the Turks to bring about his adversary’s death and ruin him by destroying his crops and herds. After Baldwin’s return from Samosata the senate and citizens met and decided that Baldwin offered a better prospect of defending Edessa against the Turks. They had summoned Constantine to this meeting, and it was he who proposed the plot to kill Thoros and put Baldwin in his place. After this council, one day the entire population of the city, ‘small and great’, took up arms and came to Baldwin, demanding that he join them in destroying the duke. Baldwin’s response, according to Albert, was completely honourable; in fact he appears as concerned with issues of personal honour and reputation as with moral questions: Baldwin refused with every objection to undertake such a crime, because he had been appointed the duke’s adopted son and had not yet discovered any sort of cause or evil in him which would make Baldwin agree with and take part in his destruction. He said: ‘It would be a sin beyond estimation in the sight of God for me to raise my hand without cause against this man whom I have taken to myself as father, and to whom I have also given my pledge. But I beg of you not to let me be dishonoured by his bloodshed or death, and make my name become worthless among the leaders of the Christian army. I also ask you to let me talk to him face to face in the upper room of the tower, where he has been used to remain up to now, raised up by your gift.’18 The rebels agreed to this, and Baldwin went up into the tower and broke the news to the duke, assuring him that he wanted to save him, perhaps by relieving him of his possessions. The duke may well have wondered at this, especially when the assertion gained added force from the noise of the citizens’ arrival and attack on the walls and door of the tower. He offered Baldwin enormous amounts of treasure in return for a safe conduct. Then Baldwin was ‘moved by pity’, so Albert wrote, and urged the crowd to spare the duke in return for the treasure. But the citizens refused the offer, ‘reproaching him with the insults and injustices they had often suffered under him and from the Turks at his instigation’. The duke despaired and sent Baldwin away, himself leaving by way of the window. But as he let himself down on a rope he was shot dead by ‘a thousand arrows in a moment’.19 The next day, still reluctant, Baldwin was made duke and received the rebels’ oaths of loyalty, and he was given possession of the tower with all the treasure in it. We can probably dismiss Guibert of Nogent’s touching elaboration of the same story, where he not only exonerated Baldwin from any part in the conspiracy but depicted Thoros’s ‘newly adopted son’ ready to defend the duke with his Frankish forces and the old man protesting that he himself could not be saved but Baldwin should save himself.20 The accounts we have of the conspiracy to overthrow Thoros thus agree broadly on the outline of events, which Matthew placed in Lent of 1098: the leading citizens  – the same who had invited Baldwin to Edessa  – conceived

Count of Edessa  45 a plot against Thoros, citing as justification his past collusion with the Turks at their expense. Baldwin granted a safe conduct to the duke, but he was killed by the townspeople of Edessa notwithstanding. The sources are at variance as to the extent of Baldwin’s involvement in the conspiracy. The Latin writers claimed that Baldwin tried to gain mercy for Thoros; the anonymous Syriac author did not mention him as involved in the plot, whereas Matthew of Edessa made him part of it at an early stage. Even Albert of Aachen, whose narrative is the most detailed of all, and who presented Baldwin very publicly saying and doing the honourable thing, leaves room for speculation that – like a latter-day Pontius Pilate – he was aware of the likely outcome of his actions. A picture of Baldwin begins to emerge as a ruthless, albeit effective, leader and already a master of dissimulation. It is probable that he cultivated this reputation and used it to intimidate his opponents, of whom there were still many in the region.

Ruler of Edessa After relating the murder of Thoros, Fulcher resumed his account of the crusade as it approached Antioch, ending his section on Baldwin’s detour to Edessa thus: Moreover, when Baldwin had accepted as a gift of the citizens the princely rule of that man wickedly killed, he at once embarked on a warlike struggle against the Turks who were in his [new] homeland, and very often he either conquered or killed them. But it happened that many of our men were killed by the Turks.21 It is a fair summary but scarcely does justice to more than two years’ rule. As before, for more circumstantial detail we have to turn to Albert of Aachen. He reported that Baldwin’s first priority was to deal with Balduk of Samosata, whom he had already attacked once and had failed to defeat. Balduk was now ready to parley, perhaps fearing that further resistance would attract the attention of the main crusading army, still not far away. He offered Samosata’s citadel in return for a payment of 10,000 bezants and said he would then serve Baldwin loyally as a mercenary. Baldwin saw no reason to pay for the citadel because he considered that it was being occupied unlawfully by Balduk, who had stolen it from the Edessenes, and he refused to negotiate. Balduk therefore threatened to burn down the citadel, behead his Edessene hostages and continue to harass Baldwin at every opportunity. Eventually Baldwin was persuaded by his own men to hand over sufficient treasure – gold, silver, precious purple silks, horses and mules – to buy the fortress of Samosata, and Balduk joined his household. Baldwin garrisoned Samosata with Franks and restored the Edessene hostages held there to their families. He also showed that he did not trust Balduk by demanding his wife and sons as hostages, but although Balduk agreed he kept making excuses to avoid actually handing them over.22 Meanwhile, another Turkish chief and ‘usurper’, Balak of Sororgia, approached Baldwin to ask for his aid in supplanting rebels from one of his towns. The rebels,

46  Count of Edessa who were withholding tribute from him, were ‘Saracens’, which is the word Albert used for native non-Turkish peoples. Baldwin and Balak arrived at an agreement – we are not told the terms – and Baldwin prepared to attack the city and force it to pay tribute. The rebel townsmen retaliated by inviting Balduk of Samosata to arrive with an army of Turks and defend them, and he agreed in the expectation of taking the town for himself. Baldwin became aware of this and set out to besiege the town with a strong and well-equipped army. The citizens were horrified, especially by the weaponry, and sent messages to Baldwin, promising to pay their annual tribute to him. Balduk was obliged to leave the town, and he pretended that he had never had designs on it for himself, but had arrived there to talk the townsmen into surrendering to Baldwin. Baldwin, in turn, pretended to believe him but no longer trusted him. When the town surrendered to Baldwin, Balak ceded it to him along with a fortress in the mountains. Baldwin garrisoned the two strongholds and installed Fulcher of Chartres (not the writer but a knight of the same name) in charge.23 William of Tyre made the important observation that the occupation of Sorogia opened up the route for communication between Antioch, the next goal of the main army, and Edessa.24 This was to prove valuable during the long siege of Antioch and was very probably a factor in Baldwin’s deviation to Edessa and the other leaders’ acquiescence to it. Another development worth noting is the establishment of tributary relationships with defeated adversaries. In Baldwin’s later career this was one of several devices by which Baldwin sought to alleviate the kingdom’s perpetual financial difficulties.

The Armenian marriage Baldwin was also advised by the Edessene council to consolidate his position by marrying an Armenian noblewoman. Albert stressed that this was a ‘splendid and legal marriage’. The bride was the daughter of an Armenian prince called Taphnuz, the brother of Constantine, a powerful landholder in the region of the Taurus mountains. William of Tyre, when he summarised Baldwin’s career to date at his accession, drew on Albert for both Baldwin’s adoption by Thoros and his marriage: He took in marriage the daughter of a certain noble and eminent prince of the Armenians called Taftoc, who with his brother Constantine held impregnable strongholds around mount Taurus and had many forces of brave men; on this account and because of their wealth and their immense power they were regarded as kings of that people.25 William also stressed the legality of Baldwin’s Armenian marriage when he recounted her repudiation by Baldwin once he was king. Rather strangely, neither of the writers who were in Edessa at the time mentioned the wedding. Matthew of Edessa ignored the marriage altogether; Fulcher of Chartres did not describe the wedding ceremony, although he made two passing references to the Baldwin’s wife later in his work. First he mentioned her presence at Jaffa in 1101.

Count of Edessa  47 She evidently wielded some authority, because she entrusted a letter to a mariner from the townsmen to Tancred. Second, towards the end of 1116, when Baldwin was ill, he dismissed his bigamously married third wife ‘since he had married her unlawfully because his wife whom he had previously lawfully married in the city of Edessa was still alive’.26 Guibert of Nogent also referred to the marriage and its repudiation by Baldwin, and he called the Armenian wife ‘the offspring of the most noble people of the land’.27 On balance, in spite of its omission by both of the writers in Edessa, it is likely that the wedding was as ‘splendid and legal’ as Albert recounted. The new countess may have been called Arda, but no one knows quite where this name came from: it was given currency by Edouard Dulaurier in his introduction to the nineteenth-century edition of Matthew of Edessa, but he did not provide a reference. A possible source was Paoli, 1733.28 Gérard Dédéyan has pointed out that she was probably known by the Greek name Arete (Αρετή), meaning ‘goodness’, ‘virtue’ or ‘nobility’.29 Dédéyan also constructed a plausible family tree for Arda that would explain her political importance. It has been assumed by historians that her uncle Constantine was the son of Reuben (Roupen). Matthew of Edessa linked this Constantine (1093–1100), whom he always described as ‘the son of Ruben’, with Thoros of Edessa as recipient of written notice that the Frankish forces were advancing into their area. Matthew also wrote that he and other commanders in the Taurus, along with other Christians in the region, sent provisions to the crusading army when it was suffering from famine. However, there was a second Constantine, Constantine of Gargar (Konstandin of Kaṙkaṙ), who had played a considerable part in the overthrow of Thoros the curopalates. After Baldwin was installed as Thoros’s successor, Constantine of Gargar (†1114) was closely associated in the rule of Edessa. Dédéyan has made a convincing case that it was this Constantine who promoted the marriage of the new count with his niece, the daughter of his brother Dawt‘ouk (Albert’s Taphnuz; William’s Taftoc; Michael the Syrian’s Tabtoug). Dawt‘ouk became a close ally of Baldwin: Dédéyan called it ‘une sorte de duumvirat’. The promise of a lavish dowry was no doubt another strong incentive for Baldwin to remarry: Albert of Aachen wrote that Taphnuz promised to pay 60,000 bezants, some of it to maintain mercenaries for the defence of Edessa, although he paid only 7,000 at once and put off paying the rest. Albert also confirmed some sort of working partnership between Baldwin and his father-in-law ‘because he was a man of advanced age and sensible advice’.30 The marriage probably took place soon after Baldwin seized power, that is after Easter in 1098. The alliance with Taphnuz and Constantine was relatively stable until the end of the year, during which time Baldwin was able to lend valuable support to the crusaders during the months when they were besieging Antioch and then besieged within the city. Any idea of power-sharing broke down in December  1098. To understand the reasons for this we are almost entirely dependent on Albert of Aachen. The chief cause was, he wrote, a great influx of crusaders from Antioch after the city had been captured. Discontent with the enforced inactivity of over-wintering in northern Syria was probably a factor, especially for

48  Count of Edessa those keen to press on to Jerusalem: Raymond of Aguilers, who had previously only mentioned Baldwin’s installation at Edessa to explain his absence from the council at Antioch, wrote of the unrest as a member of Raymond of Saint-Gilles’s entourage: Meanwhile the knights and the people began to complain about the journey, when it might please the princes to begin it. . . . Bohemond was saying that it would be delayed until after Easter, and it was then Christmas time. Many were also desperate because there were few horses in the army, and the duke [Godfrey] was not there, and many of the knights had gone off to Baldwin in Edessa.31 Albert of Aachen named four of the knights: Drogo of Nesle, Rainald of Toul, Gaston of Béziers and Fulcher of Chartres (not the writer). He also confirmed that there were very many others, ‘leaders and fellow soldiers in hundreds and fifties, some on horseback, some on foot’. They were there to serve for reward, after the privations and impoverishment they had suffered at Antioch, and their expectations were met: they were rewarded richly in gold and silver coin, and they assisted Baldwin in subduing the region. No wonder they flocked to Edessa, ‘until almost the whole city was besieged by Gauls and taken up by their hospitality’. And no wonder they were resented by the Edessene council: Albert reported that Baldwin was ignoring the council members in favour of the westerners.32

Rebellion The response of the Edessenes to the crusaders flooding into the city and commanding Baldwin’s attention at their expense was to conspire with the Turks to bring about Baldwin’s expulsion from the city or his assassination. Fortunately for Baldwin, one of the Armenians, whom Albert named as Enzhu, remained loyal and revealed the plot to Baldwin with names and other details. Baldwin was able to order his household knights to arrest and imprison the conspirators. Their property and ‘vast sums of money’ were confiscated, and Baldwin used them to pay his followers. The conspirators, now in captivity, pleaded with Baldwin and offered large ransoms, but he was informed by his spies that they had hidden a great deal of treasure in towns and castles outside the city, and for a while he held out for this. However, in the end he ran out of money and decided to allow the prisoners to be ransomed, with the exception of two ringleaders whom he had blinded. Some of the commoners who were in on the plot were mutilated – they had noses, hands or feet cut off – and banished from the city. Guibert of Nogent, who placed the rebellion on Easter Sunday, added lips, tongues and ears to the list of amputations and said that all the exiled conspirators were also deprived of their ‘organs of generation’.33 The savagery of the punishments may seem inordinate, but in the context of the newness and insecurity of Baldwin’s rule they were undoubtedly intended to have an exemplary and deterrent effect. It is likely that Baldwin was advised by the Edessene nobility that such mutilations were the customary

Count of Edessa  49 punishments of the Byzantines, their previous overlords. According to Albert, the sums paid in ransom were enormous: a minimum of 20,000 bezants, sometimes 30,000 or 60,000, plus mules, horses, silver vessels and other precious items. ‘From that day on Duke Baldwin became a man to be feared in the city of Edessa and his name was spread among the people right to the limits of his land.’ Baldwin’s suppression of the plot gave an early indication of his ruthless efficiency in dealing with adversaries, mitigated only by the perennial lack of funds that led to his differential treatment of prisoners who could afford to pay ransoms from others who could not and were subjected to mutilation and banishment. His fatherin-law, Taphnuz, saw this with alarm and escaped to his fortress in the mountains, refusing to return because he feared for his life as long as he still owed Baldwin the balance of Arda’s dowry.34 At this juncture Balak of Sororgia, who had importuned Baldwin for help against rebels earlier and had ceded the town to him, evidently saw an opportunity to overthrow Baldwin and re-assert his own authority. He treacherously approached Baldwin and pretended he wanted to enter into his service and was prepared to hand over his wife and sons and also a fortress – Albert called it Amacha but it has not been identified. Baldwin believed and trusted Balak and fixed a date for the handover of the fortress. Balak, however, introduced 100 armed Turks into the castle where they lay in ambush to capture Baldwin. The unsuspecting count arrived with 200 armed men and found Balak apparently ready to hand over the fortress to him. The Turk encouraged him to take a few men with him to take charge of the castle while he looked after the rest. Baldwin was deceived, but others in his company were suspicious of Balak’s ‘honeyed words’ and barred the count’s entrance, warning him against entering without any security, for example in the form of a hostage.35 This is the story as told by Albert of Aachen, but the depiction of Baldwin as so totally unaware and credulous does not ring true, especially in view of Balak’s previous treachery, so it may have been a literary device to heighten the suspense and perhaps to absolve him of two prominent knights’ deaths. Alternatively, Baldwin may have been honing his skills of dissimulation, even to the extent of deceiving his own followers. So, to continue Albert’s dramatic account, Baldwin was finally persuaded to wait outside while twelve picked men entered the castle to take its keys and shut out the Turks. The twelve were duly ambushed, and ten of them were quickly captured while the remaining two fought fiercely until they were able to make a narrow escape to an upper room where – all the while defending themselves with their swords – they put their heads out the window and shouted to Baldwin, who was still at the bottom of the castle’s steep escarpment. Finally realising the enormity of Balak’s treachery, Baldwin offered to ransom the twelve prisoners, but Balak was not interested; he wanted only to get back Sororgia. Baldwin told him he would never surrender the city, not even if Balak dismembered the captives before him. There was no concession on Balak’s part, so Baldwin withdrew to Edessa and nursed a bitter hatred (so Albert said) of the Turks. This meant that when another treacherous Turk, Balduk of Samosata, who had been delaying handing over his wife and sons as hostages, approached Baldwin ‘fawningly’,

50  Count of Edessa Baldwin had him detained and decapitated. To deal with Balak, Baldwin stationed 100 experienced soldiers in Sororgia under the command of a knight whom Albert called Folbert of Chartres. This may have been a mistake for Fulcher of Chartres, and if so it may have been the Fulcher who had arrived from Antioch or the Fulcher previously mentioned who already commanded a town and castle taken from Balak, or Albert may have confused the earlier struggle against Balak with this later one. William of Tyre had only the later episode and called the commander Fulbert of Chartres, but he may have been trying to rationalise what he read in Albert’s history. Whoever he was, his brief was to harass Amacha. He succeeded in luring the Turks out of the castle and ambushed them. A battle ensued, and Folbert captured six of Balak’s men who were then traded for six of Baldwin’s men in a prisoner exchange. Four more of Baldwin’s men took advantage of the negligence of Balak’s guards to make their escape. Balak beheaded the two who remained: Gerard, Baldwin’s ‘confidant and secretary’, and Pisellus of Wissant.36

Support for the crusade It is possible to interpret Baldwin’s activities in Cilicia and around Edessa as part of a grand ‘Armenian strategy’, as John France has done.37 According to this theory, the idea of a ‘liberated zone’ in eastern Anatolia and northern Syria came out of discussions between the crusade leaders and the Byzantine emperor. It obviously appealed to Alexios as a potential base for re-establishing Byzantine power in the region, while the crusaders could at once justify the diversion from the direct route ideologically because they were liberating Eastern Christians and strategically because gains in the area would protect their rear and provide a supply base. The ‘Armenian strategy’ is a radical revision of the ‘Cilician adventure’ theory, and neither view has clear documentary support. Nevertheless, whether Baldwin became count of Edessa as part of a Byzantine-crusader master plan or as a result of personal ambition and good fortune (or a mixture of both), his rule there was to be invaluable during the long siege of Antioch, from October 1097 to June 1098. As ever, Albert of Aachen provided the fullest account, even if it is suspected that he exaggerated somewhat: Day after day [the crusaders outside Antioch] were hard pressed by the lack not only of food but also of horses and weapons, and this severe poverty caused them more anxiety than all other concerns. While this long-term need grew greater and greater and very many were in despair because of the diminution of necessary supplies, Baldwin, who had subdued the state of Edessa, or Rohas, and had been made duke, sent very many talents of gold and silver to his brother Duke Godfrey, Robert of Flanders, Robert count of Normandy, Raymond and the other chief leaders by way of his great favourite Gerard, to make good the lack that he realised the great and noble princes were enduring. He also sent to his brother and the rest of the princes horses that were

Count of Edessa  51 remarkable runners and of excellent build, with valuable decoration on saddles and reins. He also sent weapons of wonderful value and ornament.38 Baldwin’s activities had brought him, and the greater crusading army, support from the Armenians. One of them, Nicusus, a landholder around Turbessel whom Albert had identified earlier as Baldwin’s ally and informant against Bagrat, sent a richly worked tent – a ‘pavilion’ – to Godfrey under the guard of some young men. They were ambushed by Bagrat, who then sent the tent to Bohemond as his own gift. The young escorts told Godfrey and Robert of Flanders about the ambush and theft, and the leaders asked Bohemond to hand it over. He refused. They became angry and demanded the tent again. He refused again. When Godfrey and Robert threatened to call on him with a picked band of men, Bohemond took advice and gave in. This tale might be dismissed as ‘a romantic tale for the glorification of Godfrey’, but A. A. Beaumont in his seminal study of ‘Albert of Aachen and the County of Edessa’ pointed out that the attitudes of Nicusus and Bagrat towards Baldwin – friendship and enmity respectively – were presented consistently with the earlier incident. Of more lasting advantage to Godfrey than the fancy pavilion was Baldwin’s grant to him of all the revenues of Turbessel: these included corn, barley, wine and oil and 50,000 bezants worth of gold per annum.39 Later it transpired that Godfrey had been granted lordship of Ravendel too, for Albert specified that these two cities ‘which had been subdued by his brother Baldwin before the siege of Antioch and after Baldwin’s move to Edessa were handed on to his brother the duke’.40 This raises some interesting, if unanswerable, questions: did Baldwin hand the towns over because this was always part of the strategic plan? Did he cede them to Godfrey as his overlord or from fraternal loyalty? Did he perhaps enjoy being able for once to put Godfrey in the position of beneficiary rather than donor? It is notable that he did not grant to Godfrey any title to the city of Edessa itself, so there were limits to his munificence. Nevertheless, Turbessel and Ravendel were to prove invaluable as a refuge to Godfrey when the crusader camp outside Antioch was threatened by epidemic illness. The Turkish general Karbugha, who was charged with bringing a vast army to the relief of Antioch, was well aware of the importance of Edessa as a Christian supply base. Albert of Aachen made sure his readers realised this by inventing a long speech by Qilij Arslan that summed up the progress of the crusade to date and concluded with the crusaders’ successes in Armenia: They have occupied the state of Edessa, which is very strongly fortified with ramparts and built walls and very famous indeed for its fertility, and one of their princes, Baldwin, the chief and leader of this Christian people, has even taken a daughter of a prince of the land as his wife and he was put forward in the place of the dead duke by the citizens, and he has made the whole land and region tributary to him, and thus these same Christians have invaded all places and kingdoms as far as Melitene.41

52  Count of Edessa Therefore, as the Turkish allies marched on Antioch they decided to besiege and capture Edessa and re-impose Turkish rule. Baldwin summoned his men, both Franks and Armenians, and rode out to meet the Turkish advance guard. He attacked the Turks ‘with Armenian bow and Gaulish lance’ and put them to flight. He carried off spoils of camels and mules burdened with provisions. Karbugha was incensed by the defeat of his vanguard and launched an attack on Edessa that lasted three days, but it was unsuccessful, and he lifted the siege and marched on towards Antioch. The withdrawal was hampered by the difficulty of transporting his great numbers of troops across the Euphrates that lay between, and Baldwin and the Edessenes set off in pursuit, in the hope of attacking the rearguard, but they were unsuccessful and fell back on the city. The attack on Edessa is confirmed in general terms by Matthew of Edessa, who reported that Karbugha stayed for forty days, until harvest time, and laid waste the crops. Fulcher of Chartres wrote that Karbugha besieged Edessa for three weeks.42 The discrepancies between sources about the length of Karbugha’s offensive probably relate to the difference between the total time he and his armies were in Edessene territory (forty days), the length of time he blockaded Edessa (three weeks) and the attack on the city walls (three days). The only, dubious advantage of the siege of Edessa in May 1098 was that it delayed Karbugha’s advance and allowed messengers and spies to reach the armies outside Antioch, perhaps precipitating the capture of the city that was achieved just a day before Karbugha’s arrival.43 During the month of June 1098 the crusaders were themselves besieged in Antioch and underwent extreme privation, as all the histories relate. On 28 June the desperate Latins met the Turkish army outside Antioch and won a famous victory. This was not the end of their difficulties, though, for almost immediately epidemic disease ravaged the debilitated crusaders. Its most prominent victim was Adhémar of Le Puy, the papal legate, who died on 1 August 1098, and two of the Lotharingian knights also died: Henry of Esch at Turbessel and Reinard of Hamersbach, who was buried in Antioch. Albert of Aachen stressed, though, that the disease carried off people of all ranks and conditions, ‘even women’, and although his estimate of over 100,000 deaths was undoubtedly an exaggeration, the mortality was obviously enormous. Those survivors who were able to made their way to Edessa, ‘hoping to get something from Baldwin’s hand’. One of them, whom Albert names as Folbert of Bouillon (but he may have been another Fulcher), travelling with his wife, was captured by Turks at Azaz. He was beheaded, but his wife was held in the hope of ransom and was given in marriage to a Turkish knight. Later, when the lord of Azaz was fearful of being attacked by Ridwan of Aleppo, this same knight suggested he appeal to Duke Godfrey for help, and Godfrey, after an exchange of messages assisted by one Syrian Christian and two carrier pigeons, agreed. Five days after Ridwan arrived to besiege and assault Azaz, Godfrey arrived, and Baldwin, summoned from Edessa to join him, brought 3,000 warriors – or so Albert said (medieval numbers are notoriously unreliable). Raymond and Bohemond, who had initially refused to take part in the campaign, probably each in case the other took advantage of his absence to seize Antioch,

Count of Edessa  53 now brought up their troops. Ridwan saw the numbers facing him and lifted the blockade, but his men used their knowledge of the terrain to circle around and attack those in the rear of the Christian armies, who had fallen some two miles behind their comrades. Hearing that 600 had been killed, the duke and the other leaders attacked and put to flight Ridwan’s Turks. The significance of this action is that it showed the extent of communication between Antioch and Edessa and the strength of loyalty between Godfrey and Baldwin.44 Following the conclusion of an alliance with Azaz, Godfrey took refuge with Baldwin. Albert wrote that Godfrey was fearful of the plague, which seemed to him like one he had experienced on campaign in Rome in 1084, and he withdrew into the mountains and stayed in Ravendel and Turbessel, towns he had received from Baldwin following Baldwin’s move to Edessa.45 Turbessel and Ravendel were in territory that belonged to Bagrat and his brother Kogh Vasil, and Godfrey soon came into conflict with Bagrat. Armenian monks in the two towns had suffered ‘insults’ from Bagrat’s soldiers who garrisoned another fortress nearby. The monks complained to Godfrey, who had his own reasons for bringing Bagrat down: allegedly he had intercepted other gifts sent by Baldwin to Godfrey, not only the tent, and had sent them on to Bohemond in the hope of buying his favour. Godfrey, who had evidently travelled with a considerable entourage, selected fifty men who armed and set out with Armenian archers to the fortress with the ‘guilty robbers’ in it. They attacked the castle with fire and razed it to the ground. Twenty of the soldiers from within were blinded on Godfrey’s orders. He then attacked and destroyed Kogh Vasil’s citadel as retribution for various (unspecified) insults and injuries. Since Kogh Vasil generally seems to have maintained good relations with the crusaders this may have been a case of guilt by association, or perhaps it was intended as a pre-emptive strike. In October 1098 it was judged that the disease risk had passed – it was blamed on the ‘plague-bearing month of August’ – and Godfrey was summoned back to Antioch along with the other leaders, although a number of knights stayed with Baldwin in Edessa as recounted earlier.46 The crusade was marking time in Antioch, recovering from the long siege and its aftermath until the campaigning season started in the following year. The leaders squabbled about the lordship of Antioch, a competition won by Bohemond, and they captured surrounding towns such as Albara and Ma’arra. Meanwhile Godfrey set up a meeting with his brother Baldwin. Albert of Aachen was precise about some of the details: Edessa and Antioch were seven days apart, and they met halfway between, across the river Euphrates. Unfortunately no other writer mentioned the meeting, and Albert did not know (or did not reveal, although that seems unlikely) what the brothers discussed. Perhaps it was whether Baldwin would rejoin the main expedition now that it was preparing to continue to Jerusalem, its goal, and perhaps they agreed that Edessa was too valuable a gain to jeopardise by leaving it to the rebellious Edessenes. They seem to have parted on amicable terms, and Godfrey revisited Turbessel and Ravendel on his way back to Antioch.47 The brothers would not meet again until Jerusalem was in Christian hands.

54  Count of Edessa

Pilgrimage to Jerusalem Fulcher of Chartres’s reticence about events in Edessa may well have resulted from his commitment to the crusade as an expedition to deliver Jerusalem and worship in the church of the Holy Sepulchre. From this perspective Baldwin’s activities in Cilicia and Edessa were distractions from the main action, however far they were justified in strategic terms. This interpretation is borne out by the enthusiastic way Fulcher picked up his account of Baldwin at Christmas 1099 when he and Bohemond (and Fulcher himself) made the journey to Jerusalem and the great care with which he made excuses for the two leaders: For it was necessary for the land and cities already taken with such endeavour from the Turks to be guarded, so that, the land having negligently been abandoned, if it happened that everyone went off to Jerusalem it would not be retaken in a sudden attack by the Turks who had now been driven back all the way to Persia: from this no little harm would have befallen all the Franks, both going and returning. Perhaps divine providence distracted them, judging that they would be more valuable for what needed to be done rather than what had been done.48 Fulcher went on to stress the many battles against the Turks that had been waged by Baldwin against enormous odds, and he also mentioned that the proposal to go to Jerusalem to complete the pilgrimage was initiated by Bohemond. The motives of the two leaders may have been mixed: Albert of Aachen wrote that Bohemond conceived the idea after he heard about the victory at Ascalon and Duke Godfrey’s elevation from crusaders who were returning to western Europe after completing their vows.49 It is probable that Bohemond and Baldwin, as rulers of Antioch and Edessa, wanted to find out where they stood in relation to the new ruler of Jerusalem. They may also have been concerned about the future defence of their conquests, in view of the numbers of crusaders who were leaving now that their vows were fulfilled. Baldwin’s departure with his followers – who presumably were also keen to complete their pilgrimage  – presented the Turks with an opportunity, and they made an incursion, Fulcher did not say where, that delayed Baldwin’s departure.50 He routed them with some ease and was able to embark on the journey. Marching south and west, Baldwin’s company bypassed Antioch. At Latakia, which was in Byzantine hands, they were able to buy provisions. At Valania (Bāniyās, mod. Syria) they joined Bohemond, who was encamped there with Archbishop Daibert of Pisa. Daibert had spent three months in Latakia already, having arrived with a large Pisan fleet. The pilgrimage, according to Fulcher, numbered 25,000, women as well as men, and included Baldwin’s and Bohemond’s followers and Italians of Daibert’s company. The number was assuredly an exaggeration, but nevertheless such an enormous band of people became increasingly hard to provision as they travelled through hostile territory in November of the year: Fulcher commented particularly on the hostility of the ‘Saracens’ (native Syrians) who would

Count of Edessa  55 not supply them and on the suffering of the horses and pack animals. The human pilgrims were able to help themselves to sugar cane that was growing in the fields. Fulcher also described the misery of the cold and the frequent heavy rains. The pilgrims were harassed by the Saracens on the narrow paths, but in their turn they evidently plundered goats and sheep to replace horses and pack animals that had been eaten, for he described the suffering of the poor beasts: ‘As pack animals were lacking, you would have seen she-goats and wethers stolen from the Saracens very fatigued and their backs damaged by the bulk of the baggage.’ Only twice on the journey, at Tripoli and Caesarea, were they able to buy bread and grain, albeit at a high price: these were coastal cities whose Muslim emirs were prepared to negotiate with the crusaders in the hope of preserving their independence. The main crusade had found as much, earlier in 1099.51 When the pilgrims entered Jerusalem on 21 December 1099 and worshipped in the church of the Holy Sepulchre they regarded it as the consummation of their long labour. After visiting the holy places of Jerusalem they went to Bethlehem on Christmas Eve and prayed in the church of the Nativity through the night, returning to Jerusalem on Christmas morning. When they first arrived at the city they were assailed by the stench of rotting corpses: six months after the capture of the city the bodies of slain Saracens still lay unburied. Back in Jerusalem on Christmas Day the pilgrims rested, and during this brief period Daibert of Pisa was appointed to be patriarch of Jerusalem, Fulcher said he was elected by Duke Godfrey and ‘the other chief men’, but he did not say whether Baldwin was involved in the election. Albert of Aachen, always inimical to Daibert, described the archbishop’s behaviour on the journey south as thoroughly ingratiating and claimed that he bribed his way to the patriarchate. After Daibert’s consecration, soon after New Year 1100, Baldwin and Bohemond left with him, and they met Godfrey at the river Jordan, where they all bathed and ‘enjoyed themselves’. Then they parted, Godfrey and the patriarch returning to Jerusalem and Baldwin and Bohemond setting out for Edessa and Antioch respectively. Fulcher recorded that they had replenished their supplies before the trip to the river. Their numbers were almost certainly reduced because some of their company elected to stay in Jerusalem while some of the main army decided to join one or other of the other leaders. 52 Their route took them to Jericho, where they cut palm leaves, the traditional symbol of the Jerusalem pilgrimage, then through Tiberias and along the shores of the sea of Galilee, heading north on an inland route to a town that Fulcher identified as Baalbek. This was by no means the most obvious or safest route, as was shown when they were attacked by Turks from Damascus – Fulcher numbered them at about 400  – but thanks to Bohemond in the vanguard and Baldwin in the rear, plus the rain that dissolved the glue in the Turks’ bows and arrows, the Christians were unharmed. Why they had risked travelling so close to Turkish territory, especially when the weather was evidently inhospitable, is a matter for conjecture. Very likely it was a reconnaissance mission, such as those Baldwin was later to undertake in Transjordan and Egypt. Following the skirmish with the Turks they headed for the coast, passing by Tortosa and Latakia. Raymond was

56  Count of Edessa still in Latakia, but even so the travellers were unable to buy supplies. ‘Therefore we hurried all the way to Edessa without stopping.’53 In the autumn of the same year Baldwin received the call to make the journey to Jerusalem again, this time to become its ruler.

Notes 1 Runciman, History of the Crusades, I: 202–203. He headed this section ‘Armenian Interlude’ to match his ‘Cilician Adventure’. 2 Beihammer provides several examples, p. 41. 3 Paul Gindler, Graf Balduin I. von Edessa (Halle, 1901), pp. 35–40. A more thorough and technical exploration of the topography and economy of the region is in Monique Amouroux-Mourad, Le Comté d’Edesse, 1098–1150 (Paris, 1988), pp. 19–43. The fertility of the region was noted by chroniclers and travellers including FC, 1.14.5, p. 209; F-E, p. 160. Edessa’s fortifications were remarked upon by the same authors and AA 4.12, p. 266. 4 FC 1.14.2, p. 206; writing later, he called Baldwin ‘miles quam optimus’. 5 BN chap. 9, p. 496. 6 AA 3.19, p. 168. Albert used prefecti for the duke of Edessa’s counsellors or elders of the city. Later he called them senatores. Baldwin had only eighty knights with him according to WT 4.2, p. 235. An important evaluation of Albert of Aachen as a source for events in Edessa is André Alden Beaumont, ‘Albert of Aachen and the County of Edessa’, in The Crusades and Other Historical Essays Presented to Dana C. Munro by His Former Students, ed. Louis J. Paetow (New York, 1928), pp. 101–38. 7 WT 4.2, p. 235. When WT portrayed Thoros as weak he may again, as at Marash, have been aiming subtly to undermine Baldwin, whom he considered to be weak himself and susceptible to malign influence. 8 AA 3.20, pp. 168–70. See also J. Laurent, ‘Des Grecs aux croisés: Etude sur l’histoire d’Edesse’, Byzantion, 1 (1924), 404–34. The duke of Edessa’s name, Thoros or T’oros, was an Armenian rendition of Theodoros, a Greek name. 9 See Dostourian’s Introduction, especially p. 1 and pp. 11–12. Dostourian did not list the editions of these histories he used, and it may be relevant that English translations of Fulcher and William were available at the date of publication, while there was none of Albert of Aachen. 10 ME 2.117, p. 168. He later referred to Thoros as the curopalates. 11 AA 3.21, p. 170. 12 GN 3.14, pp. 163–64. 13 Jay Rubenstein, ‘Guibert of Nogent, Albert of Aachen and Fulcher of Chartres: Three Crusade Chronicles Intersect’, in Writing the Early Crusades: Text, Transmission and Memory, ed. Marcus Bull and Damien Kempf (Woodbridge, 2014), pp. 24–37. Compare Carol Sweetenham, ‘What Really Happened to Eurvin de Créel’s Donkey? Anecdotes in the Sources for the First Crusade’, in Writing the Early Crusades, Bull and Kempf, pp. 75–88. 14 The city is now Şanlıurfa (‘Glorious Urfa’), Turkey. Chronique de Michel le Syrien, Patriarche Jacobite d’Antioche (1166–1199) [MS], ed. and trans. Jean-Baptiste Chabot, 5 vols (Paris, 1899–1924), III: 183–84; ‘The First and Second Crusades from an Anonymous Syriac Chronicle’ [AS], ed. and trans. A. S. Tritton and H. A. R. Gibb, Journal of the Royal Asiatic Society, 92 (1933), 69–102, 273–306. For the political context, Amouroux-Mourad, Le Comté d’Edesse, pp. 45–54. 15 ME 2.117, p. 169; AA 3.21, pp. 170–72; FC 1.14.13, pp. 213–14. 16 ME 2.117–18, pp. 169–70.

Count of Edessa  57 17 AS, pp.  70–71. See also Beaumont, ‘Albert of Aachen and the County of Edessa’, pp. 109–12. 18 AA 3.22–4, pp. 172–76; quoted passage, p. 174. 19 AA 3.23, pp. 174–76. 20 GN, p. 164. 21 FC 1.14.14, pp. 214–15. 22 AA 3.24, p. 176. 23 Balak ibn Bahrām belonged to the Artuqid dynasty and ruled Sororgia (Suruç, mod. Turkey). AA 3.25, pp. 176–80. 24 WT 4.6, pp. 239–40. 25 AA 3.31, p. 188; WT 10.1, p. 453; 11.1, pp. 495–96. William referred to her as ‘uxorem legitimam’. For a discussion of the repudiation, see chap. 10. 26 FC 2.14, pp. 421–22, confirming BN, chap. 53, p. 530; FC 2.59, p. 601. 27 GN 7, p. 349. 28 RHC Arm, I: LI, CXIII; Sebastiano Paoli, Codice Diplomatico del Sacro Militare Ordine Gerosolimitano Oggi di Malta (Lucca, 1733), p. 355; Mayer, ‘Études’, p. 52 and n. 104. 29 Gérard Dédéyan, Les Arméniens entre Grecs, Musulmans et Croisés: Etude sur les pouvoirs arméniens dans le Proche-Orient méditerranéen (1068–1150) 2 vols (Lisbon, 2003), II: 1039. Earlier Dédéyan made a tentative suggestion that there was an etymological link with the Ardjk‘etonk‘ dynasty: I: 213. 30 ME 2.113–14, pp. 166–67; Dédéyan, Les Arméniens entre Grecs, 2: 1036–38; for Tabtoug see MS, III: 198; AA 3.32, p. 188. The unpaid dowry was to cause some marital conflict between Baldwin and Arda, as will be recounted. 31 RA, pp. 92, 99. 32 AA 5.15–22, pp. 356–64. 33 GN, p. 165. 34 AA 5.16–18, pp. 358–60. 35 AA 5.18–19, pp. 360–62. 36 AA 5.20–22, pp. 362–64; WT 7.7, p. 351. Pisellus was nephew of Udelard: Murray, Crusader Kingdom, p. 221. Little is known about Gerard, but the extent of Baldwin’s trust in him is demonstrated in the next section where he is named as the bearer of supplies to the leaders of the army outside Antioch. Balak was probably the same as Albert’s ‘Balas of Sororgia’, who attacked the 1101 crusaders near Sinope in Asia Minor, in alliance with Qilij Arslan and Danishmend: AA 8.23, p. 616. 37 France, Victory in the East, pp. 195–96. 38 AA 4.9, pp. 260–62. 39 AA 4.9, p. 262; Beaumont, ‘Albert of Aachen and the County of Edessa’, p. 114. 40 AA 5.14, p. 354. 41 AA 4.6, pp. 256–58; 4.11–12, pp. 264–66. 42 ME 2.119, p. 170; FC 1.19.2, p. 242. 43 AA 4.12, p. 266, for the spies. Asbridge argues that ‘Kerbogha’s delay at Edessa did not save the crusade, it merely postponed the moment at which Bohemond sprang his carefully crafted plan’: Asbridge, First Crusade, p. 205. 44 AA 5.4–5, pp.  342–44. For Reinard see Murray, Crusader Kingdom, p.  224. Azaz (Albert’s ‘Hasart’) is A‘zāz (mod. Syria). The story is told in great detail by AA, 5.5– 12, pp. 344–52. Much briefer references to Ridwan’s attack on Azaz are to be found in RA, p. 88, and Kemal al-Din, RHC Or, III: 571–690 (p. 595). 45 AA 5.13, p. 354. RA, p. 84, confirmed that he set off ‘towards Edesssa’ (versus Roais). Neither FC nor ME mentioned this. 46 AA 5.14, pp. 354–56; 5.25, p. 368. 47 AA 5.26–27, p. 370.

58  Count of Edessa 48 FC 1.33.3–5, pp. 323–25. BN did not include the excuses recorded by FC, but whether this is because Fulcher added them to a later version is not certain. 49 AA, 7.6, p. 494. 50 FC’s account of the pilgrimage journey occupies 1.33, pp. 322–34; AA 7.6, p. 494. 51 See AA 5.37, p. 388, for the main crusade’s similar discovery of sugar cane. For the bad weather, cf. Hagenmeyer, Epistolae, no. X, pp. 149–52 (p. 150): Stephen of Blois had suffered from the excessive cold and rain of northern Syria the previous winter. Negotations with Muslim emirs, AA 5.37–39, pp. 386–90. 52 FC 1.33, pp. 322–34; AA 7.6–8, pp. 494–98. 53 For the journey back to Edessa: FC 1.34, pp. 335–43. ME, 2.130, p. 175, reported a severe famine throughout Mesopotamia in the year 548 (1099–1100), caused by lack of rainfall. It was blamed on God’s wrath because of the killing of Thoros. He did not report Baldwin’s absence from Edessa at Christmas 1099.

4 Crisis in Jerusalem, 1100

Crisis in Jerusalem, 1100Crisis in Jerusalem, 1100

Early in February 1100 Baldwin arrived back in Edessa after his pilgrimage to Jerusalem. He could be satisfied that he had established the Frankish county of Edessa, as Monique Amouroux-Mourad observed, ‘practically without conquest’.1 Rule in the city of Edessa itself had been ceded to him by its governing elite, and his position was consolidated first by his marriage into the Armenian nobility and second by the deposition of Thoros in a coup that probably had Baldwin’s tacit approval if not his outright collusion. He had extended his authority over surrounding fortresses by a combination of purchase, trickery, intimidation and sheer ruthlessness. All of this had been achieved with, probably, fewer than 100 knights, for when more arrived during the siege of Antioch in 1098 their presence was a provocation to the Armenians. Even though Baldwin was therefore the Frankish governor of an Armenian state, as Claude Cahen wrote, and this was a potentially precarious state of affairs, he felt secure enough to travel to Jerusalem for Christmas 1099.2 The pilgrimage also marked a rapprochement with Bohemond, now the ruler of Antioch, that led indirectly to an opportunity for Baldwin to extend his authority north of Edessa.

Bohemond’s capture The Normans of south Italy, Bohemond and his nephew Tancred, had been Baldwin’s rivals and opponents ever since the so-called ‘Cilician adventure’, if not earlier. However, Bohemond and Baldwin now had strong pragmatic reasons for co-operation in the defence of their two northern territories, and, accordingly, Baldwin came to Bohemond’s aid in July 1100. Malik Ghāzī, the Dānishmendid ruler of central and north-eastern Anatolia (1097–c. 1106), whom Matthew of Edessa as well as all the Latin chroniclers called Danishmend, attacked Melitene (Malatya, mod. Turkey), a city north-east of Edessa ruled by Gabriel, who was Greek by descent and religion and the father-in-law of the late Thoros of Edessa. Gabriel appealed to Bohemond for assistance against Danishmend, promising to cede the city of Melitene to him in return. Bohemond marched north with his cousin Richard of Salerno. The sources agree that Bohemond was careless on the march, although only Matthew of Edessa, telling the story from the Armenian point of view, went so far as to say that ‘they had put aside their weapons and

60  Crisis in Jerusalem, 1100 came dressed like women accompanying a funeral procession’.3 Ralph of Caen accused Bohemond of ‘stupid audacity’; Albert of Aachen wrote that he took only 300 knights with him, Fulcher that he was ‘ambling along unaware’.4 There is some variation between the sources as to the exact sequence of events, but the consequence of Bohemond’s carelessness was that he was ambushed and he and his cousin were captured by Danishmend, who went on to besiege Melitene. Fulcher of Chartres, reporting as an inhabitant of Edessa at the time, wrote that fugitives from Bohemond’s army brought the news to Baldwin and he immediately mustered an army of Franks from Edessa and from Antioch and marched on Melitene. Fulcher added the detail that Bohemond had sent a lock of his hair to Baldwin, which was a signal pre-arranged between them for just such an emergency. William of Tyre omitted the story of the token, perhaps dismissing it as a romantic elaboration, but Bartolf of Nangis reported it too, with the addition that it was sent by way of an Armenian, and so did Albert of Aachen, although he wrote that it was a single hair of Bohemond’s head conveyed by a Syrian.5 There seems no reason to doubt that there was some agreement for mutual assistance between the Latin rulers, probably made during their pilgrimage to Jerusalem together the previous winter: whether this amounted to an early form of confraternity is more debatable, although Fulcher’s claim that Baldwin was able to call on troops from Antioch as well as Edessa is a strong suggestion of a formal alliance.6 When Danishmend learned of Baldwin’s approach with his army he withdrew from the siege of Melitene, intent on getting his prize captives, Bohemond and Richard, beyond the reach of their allies. Fulcher and Albert agree that Baldwin pursued the retreating Turks for three days before giving up and returning to Melitene. He gave up, wrote Albert, because he feared ‘the tricks of false Christians or the ambushes of the enemy, and because he did not have many soldiers’. It may be cynical to suggest that Baldwin saw less profit in the hazardous pursuit than in returning to Melitene to earn the gratitude of the ruler. Gabriel duly handed over the city to Baldwin. According to Albert, Baldwin declined to accept the treasury of the town or lavish gifts; this was possibly because he had no intention of staying around in person and preferred to ensure a friendly alliance with Gabriel holding Melitene as a governor subordinate to him. Baldwin reinforced the agreement by garrisoning Melitene, leaving ‘fifty soldiers to remain in the city to protect and keep the town walls’. These fifty enabled Gabriel to resist a renewed attack from Danishmend, and after a long siege the Turks retreated. Baldwin returned to Edessa in mid-September, and this was when he heard that his brother Godfrey had died.7

Godfrey’s rule in Jerusalem While Baldwin had been establishing and consolidating his position in Edessa, his brother Godfrey had remained with the main crusading armies through the eightmonth-long, attritional siege of Antioch. Following its capture and cession to Bohemond, the crusaders overwintered around Antioch, and in the spring of 1099 Godfrey marched south to Jerusalem with his elder brother Eustace, Raymond of

Crisis in Jerusalem, 1100  61 Saint-Gilles, Tancred, the Roberts of Flanders and Normandy and what remained of their armies. They invested the city on 8 June 1099, and took it by assault after a five-week siege on 15 July  1099. Godfrey’s reputation was enhanced by his being one of the first to enter the city, via a siege tower against the north-eastern wall. By this time he had attracted the support of most of the other leaders, even Tancred, but not of Raymond of Saint-Gilles. During the previous winter they had adopted adversarial positions, particularly about the future of the expedition. Having failed in his attempt to oppose Bohemond’s possession of Antioch, Raymond had been intent on staying in the region and acquiring territory there. Eventually, in May  1099, he was persuaded by the strength of feeling of the other leaders and his own followers to abandon his siege of Arqa and advance with them to Jerusalem.8 Godfrey and Raymond had managed to co-operate in the capture of the city, but their opposition polarised during the subsequent weeks. When it came to electing a ruler, which happened within a week of taking Jerusalem because of the threat of an Egyptian relieving army, the two of them were the viable candidates. Raymond lost the election and ceased any co-operation with Godfrey, even to the extent that he initially refused to join the army as it went to meet the Egyptian forces near Ascalon. Following the crusaders’ victory there, the two leaders clashed over the surrender of Ascalon, which might otherwise have been achieved. Had Ascalon been taken, this would have dealt a severe blow to the Egyptians, who used it as a bridgehead for attacks on the kingdom of Jerusalem for the next half-century. After launching an abortive attack on Arsuf, Raymond joined other leaders who were assembling near Caesarea to return home to western Europe now that their pilgrimage had been accomplished. He was pursued there by Godfrey, but the two leaders were reconciled (or at least paid lip service to the idea). It was late in August 1099, and those returning home were anxious to sail before winter set in, so they marched north again and reached Latakia. At this stage Raymond was probably planning to sail too, at least as far as Constantinople, but when they arrived at Latakia they found that Bohemond was attempting to seize the city from the Byzantines, with the assistance of a large Pisan fleet that had brought Daibert, archbishop of Pisa, the legate sent by the pope to replace Adhémar of Le Puy, who had died at Antioch in August 1098. Bohemond and Daibert were persuaded to abandon their attack on fellow Christians, and Raymond took possession of Latakia’s citadel, holding it from the emperor of Byzantium, while other leaders, including Robert of Flanders, Robert of Normandy and Eustace of Boulogne, sailed for the West.9

Godfrey’s title Since Baldwin and Bohemond were preoccupied with their territories in Edessa and Antioch respectively, and Tancred remained loyal to Godfrey, capturing Bethlehem on his behalf and then striking north to Tiberias, Godfrey was able to establish and confirm his position as ruler in Jerusalem. What title he took has long been disputed by historians in a debate that had its origins in the election process as reported by contemporary writers.10

62  Crisis in Jerusalem, 1100 When a ruler was urgently required for the newly conquered city, Godfrey had – according to some sources – refused to wear a crown of gold in the city where Christ was mocked as ‘King of the Jews’ and wore a crown of thorns. However, it should be noted that the ‘witnesses’ to his refusal all depend ultimately on the same source: Fulcher of Chartres.11 It is furthermore important to be aware that our best evidence for what Fulcher wrote in his first redaction is the version attributed to Bartolf of Nangis, and according to this source, ‘They all by common consent raised Duke Godfrey to king and prince of the city of Jerusalem and of the whole region.’12 However, at the time of Godfrey’s election Fulcher was still with Baldwin in Edessa, and so his account cannot be accepted as eye-witness testimony. Those who were in Jerusalem in July 1099 present a more complicated negotiation with regard to the secular ruler of Jerusalem. In particular, Raymond of Aguilers, who was chaplain to Raymond of Saint-Gilles, narrated a process in two stages. According to this account, the clergy early took a hand in the choosing of a ruler, saying they approved the idea, but only if it were to be done properly and in due order. Their idea of the proper procedure was for a ‘spiritual vicar’ to be appointed first, but there was an unfortunate absence of good candidates for this role, and in the meantime the secular princes accelerated their election of a temporal governor. Raymond of Aguilers (one of the clerics and therefore not an unbiased reporter) put it thus: ‘Spurning our admonition and opposition they urged the count of Saint-Gilles to accept the kingdom. But he said that he shuddered at [assuming] the royal name in that city, yet he offered his consent to others if they accepted it.’13 Probably, in spite of the latter concession, Godfrey recognised a spoiler when he saw it and decided to accept the post but not the royal title. The other writer who was in a position to report first hand, the anonymous author of the Gesta Francorum, called him ‘prince of the city’. The work’s English translator rendered this as ‘ruler’ with a footnote that stated definitively that Godfrey took the title of ‘Advocate of the Holy Sepulchre’.14 This used to be a widely accepted view, but more recently it has been challenged. The probable origin of the idea was a letter ‘to the pope and all Christ’s faithful’ written from Latakia in September 1099. The author was Daibert of Pisa, and he wrote as if from himself and Count Raymond, who was alongside him in Latakia, and Duke Godfrey, who was not there but in Jerusalem (although Daibert did not mention this in the letter), as well as ‘some other bishops’ and ‘the whole of God’s army’.15 We cannot know, of course, whether Godfrey – were he present – would have approved his designation in the letter as ‘now advocate of the church of the Holy Sepulchre’; nor can we know exactly what contemporaries understood by it. One interpretation is that it can be compared to the role of advocatus as used in western Europe, where a lay ‘advocate’ might be appointed by a church or religious house to deal with temporal affairs. Godfrey had held the position for the abbey of Saint-Hubert in the Ardennes. Used in this way it would imply that the advocate accepted a role subservient to the church. There are difficulties, however, in interpreting the use of advocatus in the letter in this way. First, it was written by Daibert of Pisa (he used ego, at least according to a number of early manuscripts), and he had only recently arrived at Latakia. He had not yet met Godfrey or established any sort of relationship with him. As he was an Italian, it is possible that the role of advocatus

Crisis in Jerusalem, 1100  63 in churches in French and German territory was unfamiliar to him, and so he was either accepting Godfrey’s self-designation or using the title more loosely. In the latter case, advocatus can mean simply ‘someone called in to help’, a synonym for ‘defender’ or ‘friend’. Bartolf of Nangis, writing close to the time and possibly informed on the point by Fulcher of Chartres’s chronicle, wrote that in his lifetime Godfrey was ‘not duke or king, but servant and protector of the land’.16 In brief, it seems likely that Godfrey saw himself as both princeps and advocatus of Jerusalem and that he and contemporaries accepted that he governed a regnum, or kingdom. However, unlike a straightforward kingship, the matter of whether the position was hereditary was open to challenge.

Godfrey’s death In the event, Godfrey ruled for less than one year, dying on 18 July 1100. The cause of his death was an illness, possibly food poisoning, that gave him long enough to express his own wishes as to the succession. He was childless, and therefore he named his younger brother Baldwin as his heir in the presence of a trusted group of kinsmen and companions, but also two individuals – Tancred and Daibert – who, he probably suspected, would oppose his choice and whom, therefore, he wanted to bind by a solemn oath. The independent and often differently biased accounts of Albert of Aachen and Ralph of Caen both attest to this: Albert referred unequivocally to ‘the oath that Daibert had made with Tancred to the duke, that if he happened to die they would not confer the throne on anyone except his brothers or one of his blood’; Ralph reported that on his deathbed Godfrey, in the presence of Daibert, Arnulf and others, named Baldwin as his successor to take the highest office. Those present immediately agreed, praised his choice and took a solemn oath to obey him: ‘for they knew him to be a man generous with money, keen on military matters, humble in his manner of speech, exceedingly magnanimous’. However, Ralph continued, the decision to summon Baldwin ‘stirred up the flames of great dissension and war’.17 Understandably, having stressed the legitimacy of Baldwin’s succession, Ralph did not record Tancred’s presence at the deathbed, perhaps because he was not informed of it, or perhaps in order to avoid exposing Tancred as an oath-breaker, for Tancred was to be a central character in the opposition to Baldwin’s succession.

The succession dispute The men whom Godfrey could rely on to carry out his wishes were his personal following or household, the domus Godefridi, as it is termed by Alan Murray. Godfrey’s kinsman, the Lotharingian Warner, count of Grez, took the initiative after Godfrey died and seized the Tower of David, Jerusalem’s citadel. William of Tyre recounted this in a letter, allegedly from Patriarch Daibert, that is very hostile in tone: After Godfrey’s death Count Warner, rising against the Church like an enemy of God and valuing at nothing his pledge and pact of justice, garrisoned the

64  Crisis in Jerusalem, 1100 Tower of David against us and by way of messengers sent to Baldwin he told him that he [Warner] would seize God’s church and take violent possession of its goods until Baldwin came, the sooner the better.18 William’s attitude to the succession dispute was very much coloured by his own times and prejudices, and he tried to maintain an almost impossible balancing act between recognising Baldwin’s legitimacy as heir and enforcing the claims of the patriarchate to some sort of suzerainty. Thus he began by stressing Baldwin’s hereditary right, his nomination by Godfrey as a dying wish and the ‘common consent’ of the leaders. He went on to write a very affirmative account of Baldwin’s career to 1100, his appearance and his character. William blamed the succession crisis, and much else, on Arnulf of Chocques, saying that Baldwin emulated his brother in every way, excepting that, to his detriment, he was excessively friendly with, and ruled by, the advice of a certain Arnulf, a wicked man and the worst sort, who was archdeacon of Jerusalem and whose inclination was said to be towards every evil occupation, and of whom we made mention above because he had usurped the patriarchal see.19 According to William’s interpretation, after Godfrey was dead and buried the people to whom he had confided his dying wishes should have surrendered the Tower of David to Patriarch Daibert and handed over the city to him. Instead, a faction led by Warner of Grez had seized the citadel and sent messengers to summon Baldwin. Warner, in fact, was already ill, and he died only four days after Godfrey: William of Tyre interpreted this as a miracle wrought by the merits of the patriarch. However, the citadel remained in the dissidents’ hands, and Warner’s place as their leader was taken by the layman Geldemar Carpenel, who had abandoned support for Raymond of Saint-Gilles on crusade to join the domus Godefridi, and the cleric Arnulf of Chocques. Ecclesiastical matters will be discussed in detail in a later chapter, but here it is important to remember that Arnulf had been appointed to the patriarchate of Jerusalem as an interim measure, or provisionally, but he had been replaced by or deposed in favour of Archbishop Daibert of Pisa, the papal legate, who had forged an alliance with Bohemond during the siege of Latakia. Another Godfrey loyalist was Robert, bishop of Lydda, who set out to take the news of Godfrey’s death to Baldwin in Edessa and to invite him to come to Jerusalem as his brother’s heir. Godfrey’s attempt to bind Tancred to support Baldwin as his heir was a failure, as he might have predicted in view of the two younger men’s rivalry in Cilicia. One source appears to offer an additional excuse for Tancred’s failure to keep his oath: the Historia peregrinorum euntium Jerusolymam. This anonymous history, written at Monte Cassino around 1130, was largely based on the Gesta Francorum and the Gesta Tancredi of Ralph of Caen but also seems so have used the works of Raymond of Aguilers, Peter Tudebode, Robert the Monk and Guibert of Nogent. Occasionally it has information not found elsewhere, usually concerning the Normans of south Italy, and this is the case with events surrounding Godfrey’s

Crisis in Jerusalem, 1100  65 death. For the deathbed scene itself the Historia had an almost word-for-word copy of the Gesta Tancredi, but then the author claimed that Tancred had gone out for the purpose of foraging (or pillaging, the Latin word could mean either), and on learning that Godfrey had died, but not that he had named Baldwin as his successor, he returned to Jerusalem. There he found the gates closed against him, and he was told that he would not be allowed in until he swore allegiance to Baldwin. He refused to do so, something that was interpreted as God’s intervention when messengers arrived from Antioch with news of Bohemond’s capture, for the Antiochenes begged him to go there and advise them what they should do. Tancred set out at once and on the way met Baldwin going in the opposite direction. They affected not to see one another and did not speak. When Tancred arrived at Antioch there was a repetition: the gates were closed against him, and he was not admitted until he swore allegiance to Bohemond. The chronicler may have been showing his awareness that in spring 1099 Tancred had left Bohemond in Antioch and accompanied Godfrey to Jerusalem. Furthermore, he was almost certainly aware, since it was claimed by Ralph of Caen, that the commander of the garrison at Antioch was a kinsman of Godfrey and his brother Baldwin, namely Baldwin of Bourcq, who was loyal to his relative who ruled Edessa and always likely to oppose Tancred. According to the story, Tancred took the oath required by the garrison and became regent of Antioch during Bohemond’s imprisonment by Danishmend.20 Tancred’s active opposition to Baldwin’s succession had a powerful ally in Patriarch Daibert, who had also taken the oath to carry out Godfrey’s dying wishes. Daibert, archbishop of Pisa, had arrived in the Holy Land in the autumn of 1099 with a large fleet of Pisan ships. When he had set out from Italy the crusaders were still around Antioch, so he arrived in northern Syria, where he discovered that the crusaders had succeeded in conquering Jerusalem and Bohemond had command of Antioch. Bohemond was able to take advantage of these circumstances and Daibert’s consequent ignorance of the disputes and dissensions within the crusade’s leadership to persuade Daibert to lend his fleet for the siege of Latakia, a city that was in Byzantine (and therefore Christian) hands. However, when Raymond of Saint-Gilles arrived, as described previously, Daibert was won over to his more pro-Byzantine stance, and the siege was lifted. As also mentioned earlier, Daibert and Raymond wrote to the pope from Latakia, and Daibert was still there when Baldwin of Edessa and Bohemond of Antioch arrived on their journey to fulfil their crusading vows (at last) by worshipping in the Holy City. He travelled with them, and they all arrived in Jerusalem on 21 December 1099. It may be assumed that the journey provided each of them with ample opportunity to probe and assess his companions, which may well have influenced their actions during the succession crisis of the following summer. When Godfrey died, Daibert, acting in alliance with Tancred, appealed to Bohemond to support him in opposing the succession of Baldwin. His reasons were undoubtedly different from Tancred’s. In the case of Daibert, personal ambition and even avarice cannot be ruled out as motives for opposing the establishment of secular kingship in Jerusalem. This is an impression given by contemporary

66  Crisis in Jerusalem, 1100 sources which, of course, were all written with their own agendas, which will be discussed later in relation to Daibert’s career in Jerusalem. However, Daibert’s opposition was also more explicitly a result of his strong beliefs about the proper designation and governance of the nascent state of Jerusalem. He was an ally of Pope Urban II, the instigator of the crusade, who had probably sent him to replace Bishop Adhémar of Le Puy as his representative after the latter died (1 August 1098), but at all events expected him to bring naval and other reinforcement to the siege of Antioch, which was proving so intractable. Urban in turn was a protégé of Gregory VII, and all three clerics were prominent proponents of the Reform Papacy, which was much concerned with the proper relationship of Church and state as played out in the ‘Investiture Contest’ in western Europe, now mirrored in Jerusalem.21 Whatever his expectations, Daibert found that the situation in the East was totally transformed between the Pisan fleet’s departure from Italy and its arrival at Latakia in September 1099. First, Urban II had died (29 July 1099), and this annulled Daibert’s legatine authority, if indeed he had it. More importantly, if the Pisans were expecting to find the crusade still involved in an attritional war in northern Syria, then the capture of Jerusalem and the Christian victory over the Egyptians at the battle of Ascalon (12 August 1099) were presumably a pleasant surprise. However, they found that arrangements for the rule of the newly captured city had, necessarily, been made between these two events. Godfrey had accepted a princely title, and Daibert, who had no means of consulting quickly with his superior in Italy, had to use his own initiative and respond to the situation as he found it. A few days after he arrived in Jerusalem with Baldwin and Bohemond at Christmas 1099, Daibert was elected patriarch of Jerusalem, replacing Arnulf of Chocques. Soon afterwards there appears to have been some sort of ceremony confirming Godfrey and Bohemond as rulers of Jerusalem and Antioch respectively. The evidence for the nature of this ceremony is rather late, and therefore its reliability is in question. In the 1120s Fulcher of Chartres claimed that ‘also in Jerusalem Duke Godfrey and Lord Bohemond received their land from Patriarch Daibert for the love of God’. Even later, William of Tyre referred to the rulers’ ‘investiture’.22 No account of a formal investiture ceremony is to be found in chroniclers writing closer to the time, namely Bartolf of Nangis, Guibert of Nogent or Albert of Aachen, although this last writer was so well disposed to Godfrey and implacably opposed to Daibert that it is possible he suppressed his knowledge of any submission of the lay ruler to ecclesiastical authority. There seems to have been no question of Baldwin receiving Edessa at the patriarch’s hands, but since it is not known whether the question even arose it is difficult to know how much significance to attach to this. Bohemond, Baldwin and Daibert appear to have stayed on friendly terms, at least superficially, for Albert of Aachen recounted a trip to the river Jordan after Christmas (on 5 January 1100) during which they and Godfrey ‘bathed and enjoyed themselves’ before Bohemond and Baldwin returned to their own territories. It seems unlikely that any serious business was done.23

Crisis in Jerusalem, 1100  67 Whether a formal granting of title took place at this point or not, there is a reasonable presumption that Daibert was aiming to establish ecclesiastical dominance over the new Latin states. It used to be asserted that Daibert was trying to establish a theocracy in the Holy Land, but it is unlikely that either Bohemond or Godfrey acceded, or would have submitted, to such an arrangement or that they believed they needed ecclesiastical legitimation. There were sound pragmatic, political reasons for each of them to seek the approval of the Roman Church. Bohemond probably thought it would strengthen his position in Antioch, where the patriarch was still the Greek John of Oxeia and the Byzantine emperor still regarded the city as lawfully his. Godfrey needed a navy, and the Pisan fleet, which comprised 120 ships, could prove invaluable for conquering the coastal cities: at the time only Jaffa was in Christian hands and within easy range of Jerusalem, and not only were ports an advantage for the arrival of pilgrims and reinforcements from the West, but while they remained in Muslim hands they acted as bridgeheads for invasions from Egypt that threatened the newly established state. Therefore, even if Daibert believed he was establishing some sort of ecclesiastical suzerainty – and it is not certain that he did – Godfrey and Bohemond probably considered they had co-operated in a ritual analogous with ecclesiastical participation in the coronation ceremonies of western monarchies.24 Be that as it may, Daibert had evidently established a relationship with Bohemond shortly after his arrival in the Holy Land, and when Godfrey died, barely six months after the events described above, it was to Bohemond the patriarch appealed for support in pre-empting any claim of Baldwin to succeed his brother. Tancred was allied with them; perhaps inspired by his uncle’s example, by the summer of 1100 he had taken control of Galilee and furthered his territorial ambitions. At the news of Godfrey’s death he and Bohemond abandoned their siege of Acre and moved to take Haifa, thus extending Tancred’s lands to the coast. This was in direct contravention of Godfrey’s promise to grant the city to Geldemar Carpenel, a loyal member of his household, and thus a declaration of hostile intentions towards the domus Godefridi and Baldwin’s succession.25 While Tancred’s opposition to Baldwin is entirely explicable from the history between the two that dated from the ‘Cilician adventure’, if not before, exactly why Daibert was so opposed to Baldwin’s inheritance is more open to debate. It is possible that he knew from his acquaintance with Baldwin and from what he was told by Tancred that Baldwin would be resistant to any interference from the Church or any claim to ecclesiastical suzerainty. There is also evidence of Daibert’s having territorial ambitions, for himself or for the patriarchate, in the form of the letter addressed to Bohemond and included in William of Tyre’s chronicle.26 This has been much examined and discussed, and it remains, to say the least, ambiguous. The letter was included in William’s narrative at this critical point in events, the succession crisis following Godfrey’s death, and it purported to confirm certain grants already made in favour of the Church and to convey Godfrey’s deathbed wishes. These were, allegedly, the grant of a quarter of Jaffa made on the feast of the Purification of the Blessed Virgin Mary (2 February) and

68  Crisis in Jerusalem, 1100 ‘everything that by right belonged to the Church’ at Easter 1100, explicitly the Tower of David, Jerusalem’s citadel, and all the rest of the city plus Godfrey’s own possessions in Jaffa. Daibert also claimed that Godfrey became ‘his man and the Holy Sepulchre’s’. These grants would seem to make Daibert uncontrovertibly the ecclesiastical ruler of the new Latin state of Jerusalem. However, as William went on to say, the grants were not unconditional: Daibert had conceded in the original agreement that Godfrey needed to keep the possessions until he could increase his resources ‘by taking Cairo or other cities’.27 This had certainly not happened by July 1100. Even assuming that the letter was entirely genuine, Godfrey had therefore conceded very little in practical terms. In addition, it was only if Godfrey died without a male heir that all the grants were to go to the Church unopposed. The phrase ‘without male heir’ has been interpreted as meaning a direct heir of his body, a son, but the rules of primogeniture were established and widely accepted, and according to them a brother was entitled to inherit.28 However, the letter sent by the patriarch to Bohemond may not have made clear these flaws in his case: William himself referred to the letter he included in his chronicle as a ‘rescriptum’, which could as well be a re-write or even a revision of the original as a faithful copy of it. It is now generally agreed that William at the very least was editing a written source and that he may even have composed the letter on the basis of a documentary reference or an oral tradition that such a letter was sent: it has to be remembered that he was writing some seventy years after the events he was describing. There is, however, independent evidence that a letter with some content of this sort was sent by Daibert to Bohemond: Albert of Aachen recorded as much, with the name of the bearer and some additional information about the nature of the message and its fate: Actually, because the bearer of this message, Morellus by name, the patriarch’s secretary, was sent in deceit and against the oath which that same patriarch had made with Tancred to the duke, that if he happened to die they would not confer the throne on anyone except his brothers or one of his blood, he was opposed by the wrath of God and fell into Count Raymond’s hands in Latakia, and thus the whole sending of the letter came to nothing, and treachery was revealed on all sides.29 Not only is this independent confirmation of the existence of the letter, whatever its content, but it has added verisimilitude because Albert usually had no good word to say of Raymond of Saint-Gilles. Of course, another reason the letter ‘came to nothing’ was that Bohemond had been taken out of action when he was captured by Danishmend and carried off into captivity. The question as to how he might have responded is therefore to an extent redundant, but the probable existence of a formal alliance between Bohemond and Baldwin, as evidenced by their co-operation in the defence of Melitene, makes either outcome possible: he might have joined Daibert and Tancred in opposition to Baldwin’s inheritance, or he might have stood by his ally Baldwin, realising that Daibert was in a power bid for himself and the Church that was unlikely to benefit Bohemond.

Crisis in Jerusalem, 1100  69

Baldwin’s journey to Jerusalem While this conspiracy against Baldwin was being orchestrated, news of Godfrey’s death and Baldwin’s designation as his heir arrived in Edessa. It was brought by Robert, bishop of Lydda; a knight also called Robert; and another called Gunter. They were sent by Geldemar Carpenel; Robert FitzGerard; Ralph of Mousson; Joffrey, the duke’s chamberlain; Wiric of Flanders; Matthew, the duke’s steward; Wicher the Swabian; and the cleric Arnulf of Chocques: all of them loyal members of the domus Godefridi. Albert gave the content of the message verbatim, but he is unlikely to have been writing from knowledge, so his reported version must be treated cautiously: The knights and princes of the kingdom of Jerusalem, who were serving under the most Christian duke until now, greet you in the name of Jesus Christ, Son of the living God. We are sent here by their decision and decree, to make known to you that your brother Godfrey, duke and prince of Jerusalem, has been taken from this light. For this reason they unanimously invite you to come at once and take over the kingdom in your brother’s place and sit on his throne, for they have agreed that they will accept no other than his brother or a man of his blood, because of his inestimable goodness and very great generosity, and because of the vow which they made never to allow a foreigner to reign or to sit on the throne of Jerusalem.30 Fulcher of Chartres described Baldwin’s reaction thus: ‘When it was announced to Lord Baldwin that all the people of Jerusalem were expecting him to be the next heir and prince of the kingdom, he grieved a little for his brother’s death but rejoiced more at his inheritance.’31 Albert of Aachen reported a rather different but equally ambivalent reaction: ‘When he heard this sad message, Baldwin’s heart dissolved into very great weeping and lamentation, but nevertheless, as a man of remarkable self-control, he long pretended otherwise than was in his heart on account of his beloved brother’s death.’ Of course, neither Fulcher nor Albert could possibly know what was going on in Baldwin’s heart or mind, but of the two it is obviously Fulcher who offered the more credible account. He was, after all, Baldwin’s chaplain and with him in Edessa; Albert was offering an interpretation more acceptable to his Rhineland audience. Nevertheless, both writers portrayed a man far from grief-stricken and not even simulating shock or any of the sense of unworthiness that was a common pretence in such circumstances. This is entirely in accord with the idea that he had long been Godfrey’s designated heir, and although he can scarcely have expected his inheritance to be so soon, he was ready and keen to take it up. Baldwin was not able to set out for Jerusalem at once, however, because he had to provide for the government of Edessa. His choice was a kinsman, Baldwin of Bourcq, who at the time was in military service in Antioch ‘for money’, which might suggest mercenary status, but Ralph of Caen said that Bohemond had given him command of the Antiochene garrison.32 Baldwin of Bourcq’s employment

70  Crisis in Jerusalem, 1100 at Antioch is further evidence of some sort of formal accord between Baldwin and Bohemond, probably dating from their Christmas pilgrimage in 1099. Albert explained that the ‘old’ Count Baldwin waited until his successor was established as ruler before setting out from Edessa and that the ‘new’ count received Edessa ‘as a fief’, ruling in Baldwin I’s place.33 Once his successor was in place, Baldwin (the king designate) went to Antioch first of all. According to Albert’s account: All the soldiers and the guards of the town ran out to greet him, and they offered him the state, if he had wished to become its prince and lord. At any rate he rested there for three days in triumph and happiness, and listened graciously to all the citizens and guards on all matters, and gave wise replies, comforting them very greatly, desperate as they were about Bohemond’s loss, but he flatly refused to accept the town in Bohemond’s place.34 This reaction by the Antiochenes is another sign of good relations between the two northern states, but unfortunately it is not confirmed by Fulcher of Chartres, who was actually on this journey; he merely said that they went ‘through Antioch, and then they bypassed Latakia, Jubail, Maraclea, Tortosa and Arqa and reached Tripoli’.35 William of Tyre, however, drawing on an unknown source, recorded – plausibly – that Baldwin used his stay in Antioch prudently: He arranged for his wife and her maids who attended on her, with the heavy furnishings and the greater part of the baggage, to go down to the sea, where he had also ordered a ship to be made ready on which she could be conveyed honourably to Jaffa, for this alone of all the coastal cities had come under our authority while all the rest were still in the hands of the infidels. Baldwin’s motive, William continued, was to expedite his march and to enable him to deal more readily with any unforeseen events.36 Baldwin did indeed have to deal with enemy ambushes and attacks; he was probably also fearful of obstacles put in his way by fellow Christians who opposed his accession. His journey finally got underway on 2 October 1100. He was accompanied by what Fulcher called a ‘tiny army’ of some 200 knights and 700 infantry. The enemy threat came above all from Duqāq of Damascus (1095–1104) and his Saljūq army, whom Baldwin had already confronted as count of Edessa. Duqāq attempted to ambush Baldwin and his small force on the narrow road next to the sea north of Beirut as they travelled from Edessa. In the event, Baldwin had been forewarned by the emir of Tripoli, and Baldwin’s scouts recognised some stray Turks as they approached the ambush, judging correctly that there would be a much larger force in the vicinity. Fulcher of Chartres, who was there (although he wrote that he wished he had not been), left a long and vivid description of the ensuing battle and how the ‘tiny army’, with God’s help, eventually put the Turks to flight and took rich spoils, including horses. He stressed Baldwin’s leadership in a preliminary skirmish: Baldwin drew up battle lines and advanced in good

Crisis in Jerusalem, 1100  71 order, killing a number of Turks but losing only four of his own knights – infantry losses were not recorded, but the footsoldiers would not have been in the vanguard so were probably not involved in this engagement. Baldwin also persuaded his excessively fearful followers (this is how Fulcher portrayed them) to stand firm while being harassed by Turks from the cliffs on their left and the sea on their right, and after a sleepless night in a camp close to the enemy, he ordered a retreat, the baggage train going first and the knights following. When the rearguard was attacked from all sides by the Turks, the armed men turned and routed them, ‘just as if it had been cleverly planned’. The army then turned south again, but before they entered the narrow pass once more, Baldwin, ‘with his customary bravery’, went ahead with some of his knights to reconnoitre. Only when they found that the Turks had gone did they signal the rest of the army to follow them. When they reached Beirut, the emir sent food daily to Baldwin, ‘more by reason of fear than love’. Other town governors did the same. When they reached Haifa, which Tancred had taken (although he was not there at the time), they did not enter the city because its captor was ill disposed to Baldwin; however, the townsmen brought out bread and wine to sell to fellow Christians whom they ‘were longing to see’. Although Baldwin’s journey had been difficult, his effectiveness against the Turkish enemy had evidently enhanced his reputation very usefully.37 Albert of Aachen included a long description of the battle north of Beirut that is in essential agreement with Fulcher’s, albeit with some different details.38 He was not, for instance, aware that the narrow road was bounded on one side by cliffs and on the other by the sea and that this allowed naval support for the Turks, nor did he mention that Baldwin was employing scouts. Remarkably, both authors reported that when Baldwin’s men heard rumours of the enemy assembling to intercept them many of them deserted or feigned illness. Albert, though, wrote that Baldwin responded with a speech of encouragement; nevertheless, of the 400 cavalry and 1,000 infantry who had left Edessa with him, only 160 knights and 500 footsoldiers stayed with him. As always with medieval authors, the numbers can be taken as no more than indicative: Fulcher’s were more likely to be accurate, while Albert was using figures to exaggerate the magnitude of the Christians’ achievement in defeating a much larger Muslim force. To a similar end, Albert said that during the night the enemy lit many more fires on the hillsides than were necessary in order to intimidate the Franks. In the morning the soldiers turned back on the coast path, but whereas Fulcher seems to have thought that the ensuing battle was unplanned, Albert wrote that Baldwin was adopting a strategy of feigned flight that drew the enemy out onto the plain where they could be defeated. Overall, this seems more likely and offers evidence of Baldwin’s skill as a battle leader. The strength of the enemy in this encounter is difficult to assess because the western sources did not even give an estimate; of the Arabic sources, Ibn Al-Qalānisī (c. 1070–1160), who wrote a Damascus chronicle that was much used by later Arabic historians, is the most directly relevant, but he gave only brief and ambiguous, sometimes erroneous, details. He claimed that Godfrey was killed by an arrow while attacking Acre and that he had already rebuilt Jaffa and granted

72  Crisis in Jerusalem, 1100 it to Tancred. After Godfrey’s death, his brother Baldwin of Edessa set out with a body of 500 knights and footsoldiers. Duqāq of Damascus gathered his forces and advanced against him, accompanied by Janāh al-Dawla of Homs (also named by Albert and Fulcher). Janāh al-Dawla attacked Baldwin near Beirut and ‘he defeated him and killed some of his companions’; as Ibn Al-Qalānisī’s English translator pointed out in a footnote, ‘The pronouns in the text do not clearly specify which side was victorious.’ This allowed later Arabic writers to interpret it as a Turkish victory.39 According to the Latin writers, Fulcher and Albert, there were ‘Saracens’ fighting alongside the Turks. Although ‘Saracens’ was commonly used by Latin writers for the Egyptian Muslims, here the writers meant people from the non-Turkish population of the region, people whom Albert called ‘an innumerable multitude . . . brought together from different places and lands’. Later he clarified that they came ‘from all the towns which were on the shore of the Palestine sea, and from the mountains and different places’. These Saracens were not always hostile to the Latins. The writers had already described how the emir of Tripoli sent luxurious foods to Baldwin and his retinue and also valuable military intelligence about the planned ambush. The Saracens, then, were not a homogeneous group, nor uniform in their attitude to the Latins – or, indeed, to the Turks. The same word was applied to all native peoples, including Christians, and not only were there religious differences among them, but also religious and political rifts within peoples sharing the same religion. A number of local rulers, like the emir of Tripoli, sought to maintain their independence by assisting the Latin Christians, as on this occasion.40 Albert of Aachen, like Fulcher of Chartres, reported that after Beirut Baldwin’s and his army’s passage along the coast road to the south was relatively smooth.41 They received hospitality at Tyre but bypassed other towns without meeting any opposition ‘on account of his victory and renown, of which they had heard’. Baldwin hoped to find Tancred at Haifa, so Albert wrote, as, ‘being completely unaware of Tancred’s treachery’, he wanted to talk to him and take his advice. This seems unlikely and contradicts Fulcher’s account that they did not enter the city because of Tancred’s ill will. At all events, Baldwin soon found out that Tancred, knowing nothing about Baldwin’s approach, had set out for Jerusalem to corrupt the princes and the guards of the Tower of David so that his uncle Bohemond or he himself might receive the kingdom; he was doing all this, moreover, with the encouragement, the assistance, and the agreement of the patriarch.42 Baldwin, ‘a distinguished and farsighted man’, immediately despatched two supporters, Hugh of Saint-Omer and Robert bishop of Lydda, to Jerusalem. On the Caesarea road they met other loyal adherents of Godfrey: Ralph of Mousson, Geldemar Carpenel, Wicher the Swabian and Ralph of Montpinçon, who were in pursuit of some Saracens.43 They exchanged news and then went together to Jaffa, where they found Tancred, who had been barred from Jerusalem, laying siege to a town that was already in Christian hands. Presumably he hoped to secure it

Crisis in Jerusalem, 1100  73 for himself and hold it against Baldwin. When Baldwin’s loyal followers arrived Tancred immediately lifted his blockade and left by another route for Haifa, being careful not to meet Baldwin on the road. Tancred evidently had a fighting force with him, since he had attempted the blockade, but his failure to confront Baldwin is perhaps evidence that Baldwin too was accompanied by a significant army. Another indication of this is that he had with him, under guard, forty-five Turkish prisoners of war, as well as the booty he had brought from Beirut. After matters were settled in Jaffa Baldwin went inland to Jerusalem ‘with all his retinue’. His first action on arrival there was to place his prisoners of war under guard in the citadel, posting a clear signal of his effectiveness as a battle commander. Fulcher of Chartres reported that Baldwin was received with popular acclaim as king, first in Jaffa and then in Jerusalem. Baldwin’s arrival in Jerusalem was joyful: as Baldwin approached the city he was greeted by a great procession of both clergy and laity that included Greeks and Syrians as well as Latins.44 A notable absentee from the celebrations was Patriarch Daibert, who was exiled to Mount Zion, outside the city walls, because he was out of favour not only with the king, Fulcher said, but also the majority of the clergy. This was a situation that had to be dealt with, for it is evident that Baldwin was looking to become king and required the participation of Daibert (or conceivably another senior cleric) to make his title unassailable.45

Notes

1 2 3 4

5 6 7 8

9 10 11 12 13 14 15

Amouroux-Mourad, Le Comté d’Edesse, p. 62. Claude Cahen, La Syrie du Nord à l’époque des Croisades (Paris, 1940), p. 225. ME 2.134, pp. 176–77. RC chap. 141, pp. 704–705; AA 7.27, p. 524; FC 1.35, p. 346. Richard was the son of Robert Guiscard’s brother William. FC 1.35.4, p. 347; BN chap. 41, p. 519; AA 7.29, p. 526. Thomas Asbridge, The Creation of the Principality of Antioch, 1098–1130 (Woodbridge, 2000), pp. 104–107, raised the question whether the agreement constituted a confraternity. FC 1.35.6, pp. 348–49; AA 7.29, p. 526. The sieges of Antioch and Jerusalem were obviously the focus of the chronicles of the First Crusade, but as Baldwin was absent in Edessa through both of them they are not recounted in detail here. Narratives and source references may be found in Asbridge, First Crusade, pp. 153–327; France, Victory in the East, pp. 197–366. Since the eyewitness accounts of the First Crusade (GF, RA) ended with the battle of Ascalon, and Fulcher was absent in Edessa with Baldwin, Albert is the main source for these events: AA 6.51–60, pp. 470–84. The debate is summarised and discussed by Élisabeth Crouzet-Pavan, Le mystère des rois de Jérusalem 1099–1187 (Paris, 2013), pp. 189–92. For the biblical reference see Matt. 27: 29. The first recorded use was FC 2.6.1–4, pp. 384–87, and, derived from this, GN 7.25, pp. 317–18; Anonymous, ‘Historia peregrinorum euntium Jerusolymam’ [HP], RHC Occ, III: 169–229; WT 9.9, p. 431. BN chap. 37, p. 516. RA, p. 152. GF p. 92 and n. 2. Hagenmeyer, Epistulae, no. XVIII, pp.  167–74. The letter was brought to Bamberg by Robert of Flanders when he returned from the crusade, according to Frutolf (F-E,

74  Crisis in Jerusalem, 1100

16

17

18 19 20

21

22 23 24 25 26 27 28 29 30 31 32

p. 150); see also McCarthy’s translation, pp. 35, 154. The letter and its dissemination are further discussed in Thomas W. Smith, ‘The First Crusade Letter Written at Laodicea in 1099: Two Previously Unpublished Versions from Munich Bayerische Staatsbibliothek Clm 23390 and 28196’, Crusades, 15 (2016), 1–25; and Smith, ‘Scribal Crusading: Three New Manuscript Witnesses to the Regional Reception and Transmission of First Crusade Letters’, Traditio, 72 (2017), 133–69. BN, chap. 42, p.  520, wrote ‘dum viveret, non dux vel rex, sed servus et protector patriæ exstitit’. Definitions of ‘advocatus’ are from Lewis and Short, A Latin Dictionary, s.v. advoco, where it is also noted that in ecclesiastical Latin Christ is described as Advocatus. Murray, Crusader Kingdom, pp. 72–73, discusses the looser usage of advocatus. For the whole debate, see Jonathan Riley-Smith, ‘The Title of Godfrey of Bouillon’, Bulletin of the Institute of Historical Research, 52 (1979), 83–86, and in particular Alan V. Murray, ‘The Title of Godfrey of Bouillon as Ruler of Jerusalem’, Collegium Medievale, 3 (1990), 163–78; Murray, Crusader Kingdom, pp. 70–77. Finally, the ‘Advocate’ idea is robustly rebutted by Jay Rubenstein, Armies of Heaven: The First Crusade and the Quest for Apocalypse (New York, 2011), pp. 298–303. AA 7.27, p. 522; RC chap. 142, p. 705. It must be remembered that Ralph was writing in Jerusalem and as an adherent of Fulcher of Chartres, who was loyal to King Baldwin. This led to contradictory tendencies in his writing: his first sponsor in the East was Tancred, but his earlier mentor and later patron was Fulcher. WT 10.1–4, pp. 453–57; cf. AA 7.21, p. 516. It is important to remember that William of Tyre did not know books 7–12 of Albert’s Historia. For the part played by the domus Godefridi in the succession dispute, see Murray, Crusader Kingdom, pp. 91–93. WT 10.2, pp. 454–55. HP, pp. 227–28. John France, ‘The Use of the Anonymous Gesta Francorum in the Early Twelfth-Century Sources for the First Crusade’, in From Clermont to Jerusalem: The Crusades and Crusader Societies 1095–1500, ed. A. V. Murray (Turnhout, 1998), pp. 29–42 (NB France refers to the Monte Cassino chronicle by its alternative title, Historia Belli Sacri). Alan V. Murray, ‘Daimbert of Pisa, the Domus Godefridi and the Accession of Baldwin I  of Jerusalem’, in Clermont to Jerusalem, Murray, pp.  81–102; Murray, Crusader Kingdom, pp. 81–93. For Daibert’s career to this point see Michael Matzke, Daibert von Pisa: Zwischen Pisa, Papst und ersten Kreuzzug (Sigmaringen, 1998). For the ‘investiture’, see FC 3.34.16, pp. 741–42; WT 9.15, p. 440. Relations between Arnulf and Daibert will be discussed in a later chapter. The trip to the Jordan is AA 7.8, p. 496. For text, summary and critique of Joseph Hansen’s influential thesis, ‘Das Problem eines Kirchenstaates in Jerusalem: Ein Beitrag zur Geschichte der Kreuzzüge’ (Luxembourg, 1928), see Murray, Crusader Kingdom, pp. 84–85 and n. 95. AA 7.26, pp. 520–23; Murray, Crusader Kingdom, p. 198. WT 10.4, pp. 456–58; see also Murray, Crusader Kingdom, pp. 87–91 and n. 100, for the text and a detailed discussion. Ibid. The modern English translation of WT has ‘without male issue’ for ‘absque herede masculo’: Babcock and Krey, A History of Deeds, I: 419. AA 7.27, p. 522. For a discussion of the letter, see Edbury and Rowe, William of Tyre, p. 50. AA 7.30, p. 528. FC 2.1.1, pp. 352–54. No one has been able to establish the exact degree of blood relationship between Godfrey and Baldwin I on the one hand and Baldwin of Bourcq on the other: see Murray, Crusader Kingdom, pp. 185–86; and Appendix C, pp. 171–75; Pascal Sabourin, ‘Baudouin de Bourcq, croisé, comte d’Edesse, roi de Jérusalem: Proposition de lecture d’un

Crisis in Jerusalem, 1100  75

33 34 35 36 37

38 39

40 41 42 43

44 45

itinéraire peu ordinaire’, Revue historique ardennaise, 31 (1996), 3–15. Baldwin of Bourcq’s status in Antioch is uncertain. It is plausible that he was given command of the garrison, as reported by RC, and the writer went on to say that he was still in post when Tancred arrived to become regent: two things ‘had already aroused ill-feelings in Tancred: partly, as is usually the way, because he was the new chief, partly because Edessa had been handed to Baldwin to govern’, RC chap. 143, p. 706. AA 7.31, pp. 528–30. AA 7.31, p. 530. FC 2.1.1–4, pp. 353–56. WT 10.5, p. 458. FC 2.1.1–2.3.10, pp. 353–66; BN omits the lengthy thanks to God for the Christian victory, but otherwise tells the same story, suggesting it was written down quite close to its occurrence: BN chaps. 42–43, p. 520. After summarising the chroniclers’ accounts of this battle, Crouzet-Pavan suggested that this journey paralleled the journey of the main crusade from Antioch to Jerusalem (p. 218): ‘Sous les yeux du lecteur, un combat terrible est en œuvre entre les forces du mal rassemblées et les serviteurs de Dieu.’ AA 7.32–34, pp. 530–38. Compare IQ, p.  51 and n. 4, with The Chronicle of Ibn al-Athir for the Crusading Period from al-Kamil fi’l-Ta’rikh AH495 [IA], ed. D. S. Richards, 3 vols (Aldershot, 2006–08), I: 47. IQ’s ambiguous account of the battle’s outcome was interpreted by Ibn al-Athīr to mean a Turkish victory: ‘In the battle that followed he won a victory over the Franks.’ FC 2.1.5, pp. 355–56; BN chap. 42, p. 520; AA 7.32–33, pp. 530–34. AA and FC also used ‘Saracen’ for the Muslims of Egypt, but here the latter will be called Egyptians to keep the distinction clear. AA 7.34–36, pp. 538–40. AA 7.35, p. 538. Bishop Robert was one of those sent to summon Baldwin from Edessa. Baldwin rewarded Hugh of Saint-Omer (†1106) for his loyalty with the fief of Tiberias when Tancred took up the regency of Antioch in 1101: Murray, Crusader Kingdom, pp. 211– 12. Ralph of Mousson’s identification is conjectural: Murray, Crusader Kingdom, pp. 223–24. For Ralph of Montpinçon, see Murray, Crusader Kingdom, p. 223. FC 2.3.12–14, pp. 367–69. BN, chap. 44, p. 522, included ‘barbarians’ rather than Syrians, perhaps meaning inhabitants whose language was neither Greek nor Latin. Daibert’s fall from grace was viewed very differently by William of Tyre: see later chapters.

5 King of Jerusalem

King of JerusalemKing of Jerusalem

Despite the efforts of his Latin adversaries and his Muslim enemies Baldwin reached Jerusalem in mid-November  1100, determined to enforce his claim to succeed his brother Godfrey as its ruler. In this he had the support of the domus Godefridi, but he could have no illusions about the opposition to his prospective rule, especially from the patriarch of Jerusalem, Daibert, who had been at the centre of the conspiracy to deny him what Baldwin considered to be his legitimate inheritance. Baldwin had already shown himself to be an effective ruler in Edessa, astute and ruthless in his dealings with opponents and dissidents; he would need these qualities and more in dealing with the patriarch. No doubt he had already formulated a strategy for imposing his authority. An important step was to claim the title ‘king’ that Godfrey had refrained from adopting. Contemporary chroniclers do not seem to have questioned Baldwin’s evident intention of doing this, although they were, of course, writing after the event when Baldwin had established his kingly credentials. Also, he and they had to recognise the pragmatic advantages of the title. The political context had changed greatly in more than a year since the capture of Jerusalem and Godfrey’s election: the majority of crusaders had returned to their homes in western Europe, and those who remained were less influenced by the apocalyptic and messianic fervour of 1099.1 The alliance of Tancred with Daibert to bar Baldwin’s succession had brought into focus the need for strong secular leadership, in particular to deal with Tancred, who had been loyal to Godfrey but was a fierce opponent to Baldwin and had now seized the port of Haifa, extending his lands from Galilee to the coast. Kingship would resolve any ambiguities and enable Baldwin to try, at least, to exercise suzerainty over him. The relationship with Daibert might be more resistant to resolution, for he had the papacy and at least some of the clergy behind him, but this would not make Baldwin any less determined to assert his authority for, as Murray pointed out, he was well aware ‘of the issues in the great struggle between regnum and sacerdotium, both from his own early clerical training and his brother Godfrey’s experiences in Lotharingia, which had been one of the principal battlegrounds of the Investiture Contest’.2 Baldwin and his supporters, his opponents and the chroniclers shared a concept of kingship that had evolved in Capetian France. The hereditary principle

King of Jerusalem  77 prevailed at the expense of election, which was resorted to only in exceptional circumstances: Hugh Capet was elected in 987 following the century of disorder that followed the deposition of Charles the Simple (887), but this was the beginning of the Capetian dynasty of kings of the Franks that thereafter adhered strictly to the rule of male primogeniture. The election of Godfrey to rule the kingdom of Jerusalem could therefore, by analogy, be followed by hereditary succession. Furthermore, the brothers had a claim to royal ancestry that was stressed by certain of their historians: both of their parents were descended directly from Charlemagne, and this had much more symbolic significance than merely as a parallel to Frankish royalty, for Charlemagne had a legendary past as a warrior against the Saracens and a proto-crusader, who was thought to have made the Jerusalem pilgrimage himself. The Capetian norm was to crown the heir during his predecessor’s lifetime. This had not happened in Godfrey’s case – he probably expected to live for longer – but, as detailed in the previous chapter, he did designate his heir on his deathbed rather than leave the succession to be settled by others. Godfrey, who had achieved quasi-sainthood in his lifetime, was abiding by the idea of the realm as hereditary, and the loyalty of his household to his dying wishes has already been described.3 Baldwin’s claim to inherit the rule of the kingdom was strong, if not unassailable, but the original objection remained: could he become king in, or of, the city where Jesus was mockingly crowned the ‘King of the Jews’? Luc Ferrier suggested that the election of Godfrey as a lay ruler who would not take the title of king was a concession made by the barons to the clergy as part of a compromise that gave the laity, initially and provisionally, their choice of patriarch. Arnulf of Chocques’ subsequent deposition in favour of Daibert altered the balance of power towards the clerical side.4 This being the case, it was imperative for Baldwin to have the recognition of the Church that a formal coronation ceremony signified, and this would, preferably, mean the involvement of the patriarch. Looking again towards the Capetian monarchy as a paradigm, the ritual of consecration and coronation of a king had been fixed since the end of the ninth century.5 The first stage was a solemn vow by the future king to procure peace for the Christian people and maintain the canonical privileges and law due to the clergy of the kingdom. According to Ekkehard of Aura, soon after he entered Jerusalem Baldwin ‘bent his head over the tomb of the Lord’s sepulchre and submitted himself eternally to His service’.6 Baldwin’s joyful welcome in Jerusalem replicated stage two, the acclamation that had come to replace election: as he approached the city he was greeted by a great procession of both clergy and laity that included Greeks and Syrians as well as Latins. This may well have been orchestrated by the domus Godefridi to greet and validate the new king. Stage three, however, could not be performed without the active participation of the Church: this was the anointment of the new king with holy chrism that made him God’s elect. Only after this would the regalia be bestowed on the new king, his formal coronation. It was therefore highly desirable for a rapprochement to be reached with Patriarch Daibert, who was lodged outside the city in the abbey of Mount Zion with Baldwin’s supporters barring his entry into Jerusalem. Baldwin apparently entrusted negotiations with

78  King of Jerusalem the patriarch to others, although he may have imposed Christmas as deadline on the patriarch’s agreement. As Bartolf of Nangis wrote: Since the Saviour’s birthday was at hand, and since Baldwin himself and the patriarch Daibert himself were somewhat at odds with each other because of certain suspicions and popular rumours, the matter of his coronation and enthronement as king was put off until the feast day.7

A royal progress Although Baldwin was not formally crowned until Christmas 1100, he began to act as king as soon as he arrived in Jerusalem. Almost straightaway, according to Albert of Aachen, Baldwin was behaving as king and dispensing justice; note that Albert too mentioned acclamation and oath: Then, on the fourth day after he went up to Jerusalem, he called everyone together, great and small, from the whole assembly of Christians, and enquired about the household property of his brother Godfrey, about his equipment and money, and about the fiefs of each knight and noble. They affirmed that they had none of his brother’s things, but that these had been dispersed in alms for the poor and to pay his debts; they quoted the fiefs according as they were appointed to each person from the revenues of the towns. He patiently listened to all their replies and was silent on the subject of the possessions and weapons which had been dispersed but for which they had apologized, returning the fiefs individually to each person. Because of this he was confirmed on oath by all, and was raised powerfully onto the throne of Jerusalem where he remained gloriously. Moreover, it was mid-November, around the feast of blessed Martin bishop of Tours [11 November], when Baldwin came to Jerusalem and was appointed king and lord by all, both great and small.8 The emphasis in Fulcher of Chartres’s account is different – he included much less administrative detail, merely saying that ‘the king was rid of some business matters’ – but like Albert he reported that Baldwin had been received as king, first in Jaffa and then in Jerusalem.9 In the interval between mid-November, when he arrived in Jerusalem, and Christmas 1100 Baldwin set out to explore his new kingdom. Fulcher of Chartres went with him, and although we have to look to his later recension for a full and first-hand description of the journey, the rationale for it was preserved by Bartolf of Nangis: ‘Meanwhile, when he had refreshed himself and his men for some days peacefully in Jerusalem, Baldwin, wanting to know the land that was destined to be his and to reconnoitre his adjacent enemies, summoned his forces and set out for Ascalon.’10 Ascalon was an obvious priority: the Christians had failed to follow up their victory there in August 1099 by capturing the city, and, as the port closest to Egypt, it was the greatest threat to Jerusalem – and was to remain so throughout Baldwin’s reign and beyond. Fulcher glossed over the attempt to

King of Jerusalem  79 besiege Ascalon, but Albert of Aachen evidently also had his information from one or more of the participants on the expedition, and he explained that Baldwin had with him only 150 cavalry and 50 infantry and that when they arrived at Ascalon they found that the Egyptians had prepared for a possible attack by the new king by sending 1,000 cavalry to garrison the city. For two days the Christians remained in their camp outside the walls, then on the third day the Egyptians and Ascalonites attacked, and there was a two-day pitched battle with severe casualties on both sides. King Baldwin then called off his men; he realised (Albert reported) that they would never prevail against those numbers and the strong city walls, while any careless Christian could be killed by the enemy’s arrows. So they moved on.11 The expedition then headed inland, where it encountered a less formidable foe. Fulcher of Chartres’s more prosaic, and more plausible, account is that the inhabitants of Saracen settlements had hidden themselves and their livestock in caves. Unable to lure them out by other means, the Latins set fires and smoked them out. Those who emerged included a number of bandits who, it was believed, used to ambush and kill pilgrims on the road between Ramla and Jerusalem. There were others who were Syrian Christians, and they bought their immunity by informing on the bandits. Nearly 100 of the Saracens were killed.12 Albert of Aachen’s version of the same story is a great deal more elaborate, and, if true, it reveals Baldwin’s cunning and ruthlessness; if not, it shows the kind of reputation Baldwin was gaining and may even have been encouraging for propaganda purposes. Albert placed the caves south of Ascalon and described their occupants as an ‘Ethiopian people’ who lay hidden there to attack pilgrims on their way to Jerusalem. Rather than coming across them more or less by chance, as Fulcher implied, Baldwin set out to destroy them. First he set fires to try to smoke them out, but only two emerged, who hoped that by doing so they would be treated mercifully by Baldwin. Their hopes appeared to be justified: he treated them well, providing them with rich clothes, and asked them friendly questions about their kinsmen. This worked so well that they begged Baldwin to keep one of them in the Christian camp while the other returned to the caves and persuaded his fellows to come out. Off he went, flaunting his new clothes and other rich gifts, and ten more followed him out of the caves. Meanwhile, the one left in the camp was beheaded by Baldwin’s young men, but this was concealed from the new arrivals. His companion met the same fate, along with nine of the newcomers, but the tenth – unaware of this – was clad in rich garments and charmed by Baldwin into returning underground. This man conveyed Baldwin’s flattering promises and persuaded another thirty to come out, all but one of whom were killed, and the unsuspecting survivor was flattered by Baldwin into returning underground. Finally the 220 men remaining in the caves were deceived into coming out, and all were beheaded. Only women and children remained, and they realised what was happening when none of the men returned, so they stayed where they were. Angrily, Baldwin ordered them to be smoked out, and then, as they reluctantly emerged, they were handed over to the soldiers: ‘Some of the children were ransomed with their mothers; others were beheaded with their mothers.’13

80  King of Jerusalem The next stop on the tour was Hebron, the ‘St Abraham’s’ of the crusaders because it was the burial place of the Old Testament patriarch, and then Baldwin and his company went down to the Dead Sea via Sodom and Gomorrah, the biblical cities that Fulcher called ‘wicked’. Fulcher’s first-person description of the Dead Sea, although it drew on classical accounts, was accurate and vivid. He even tasted the water, finding it ‘bitterer than hellebore’. Tasting dates for the first time was a more pleasant experience. Still heading into the desert, the company reached Wadi Mūsa (the valley of Moses) and Mount Hor in Petra, both featured in the book of Exodus, before turning back to Jerusalem. They arrived in the city on the day of the winter solstice (21 December).14 Albert of Aachen provided some different details, for example that a breakaway group of youths suffered from hail, rain and snow when the weather turned against them in the mountains and some thirty of them died of cold.15 Both authors related that the inhabitants fled at the army’s approach and that the army found towns and villages deserted, which in view of the small numbers in Baldwin’s entourage suggests his reputation was going before him. The two writers are closely in agreement about Baldwin’s route, which took him far into territory Godfrey had not explored and proclaimed Baldwin as a proactive ruler even before his coronation.

Coronation During Baldwin’s absence certain prominent men had been working for a reconciliation between Patriarch Daibert and Baldwin that would allow a coronation to take place. The groundwork having been done, Baldwin was able to hold a meeting with the nobles and the patriarch. Albert asserted that the meeting was only consultative and that Baldwin alone decided he would be crowned in Bethlehem, but it is probable that this was a compromise already negotiated. Like the other sources, Albert explained the decision: For he was unwilling and did not presume to be exalted and wear a diadem of gold or with precious stones, and to be made king in Jerusalem, where Lord Jesus, King of kings and Lord of lords, was brought low and subject even to death for the redemption of the world, and was crowned with terrible and sharp thorns.16 A more extended version of the same rationale was included by Fulcher of Chartres. Godfrey had not been crowned, he wrote, because he did not want to be and because some people disapproved of the idea. Fulcher then included a section in direct speech representing ‘wiser and more considered advice’.17 It is tempting to believe that Fulcher himself, as Baldwin’s chaplain, was present at the meeting and took notes, or at least that he preserved the thrust of the arguments put forward, and this is possible, but Bartolf of Nangis is no help on this occasion: in a single sentence he recorded the reconciliation with Daibert and the coronation itself, the only added detail being that when Baldwin and his men arrived back in Jerusalem from their expedition they found that the golden crown and the

King of Jerusalem  81 regalia had been made ready.18 Bartolf frequently abbreviated or omitted entirely passages of theological import and may have done so here, but it is also possible that Fulcher, writing after the end of Baldwin’s successful reign, wanted to record a justification of the decision to go against Godfrey’s pious inclination. Significantly, William of Tyre omitted Fulcher’s argument; as an advocate of Daibert’s patriarchate and of ecclesiastical authority generally he may have dismissed it as naive or even dangerous. Fulcher wrote: ‘For what is the objection,’ they said, ‘if Our Lord Christ was dishonoured by jeers as a criminal and was crowned with thorns in Jerusalem, since He willingly suffered even death for us at the end? That crown, indeed, was not to their understanding one of honour and royal authority, but rather of disgrace and dishonour. But what those barbarians did as a taunt to him was by the grace of God turned to our salvation and glory. Besides, a king is not put in authority against [divine] ordinances. For when he is appointed lawfully and according to [the will of] God, he is sanctified and consecrated by an authentic blessing. He who undertakes that authority with a golden crown also undertakes the honourable burden of maintaining justice. It may appropriately be put to him, just as to a bishop concerning his episcopacy, “He who desires a kingdom, desires [to do] good work; if he does not rule lawfully then he is not a king” ’.19 This statement, if accurate, made it clear that Baldwin held that his authority to rule came directly from God and was not conferred on him by the patriarch. Baldwin may have remembered his ancestor Charlemagne, who received the imperial crown at the hands of the pope, but styled himself in official documents ‘crowned by God’.20 The statement also suggests that Baldwin and/or Fulcher viewed kingship as a contractual arrangement between ruler and ruled. The choice of date for the coronation, Christmas Day 1100, precisely 300 years after Charlemagne’s coronation, is unlikely to have been a coincidence. Of course, the date of the coronation was partly dictated by the timing of Godfrey’s death and Baldwin’s arrival in Jerusalem, but it also made Bethlehem the obvious choice of venue at Christmas, and thus it perhaps reduced the grounds for objection based on ‘wearing the crown in Jerusalem’. Additionally, both king and patriarch will have been aware of the extra significance of Bethlehem: it was the place where the prophet Samuel had anointed David as king of Israel, and therefore Baldwin was positioning himself in succession to the kings of the Old Testament. Bartolf of Nangis’s brief account of the ceremony juxtaposed it closely with the celebration of the Nativity: On Christmas Eve everyone assembled in Bethlehem, as was customary, to prepare for that most holy night in the place of the Nativity with ceremonies, prayers and vigils before that splendid crib. After their nocturnal hymns had been ritually performed with the appropriate Masses, at the third hour of the day Baldwin was crowned and exalted as king by Patriarch Daibert, and he was received with honour by all, both the clergy and the people.21

82  King of Jerusalem Contemporaries, like Albert of Aachen quoted earlier, were happy to accept the coronation in Bethlehem as a pious and sensible move to avoid the hubris of wearing the golden crown in Jerusalem.22 In any case, it is likely that Baldwin was persuaded by the ‘wiser and more considered advice’ of his supporters not to push the confrontation with Daibert at this point because he had more important tasks. The most urgent of these was to deal with the people who had opposed his succession and other potential rivals, above all with Tancred.

Tancred William of Tyre stated explicitly that Tancred still nursed the grievance against Baldwin that he had conceived at Tarsus in Cilicia.23 There is a hint in William’s account that Tancred was unwilling to take an oath of fealty to the new king. William’s main source, Fulcher of Chartres, conveyed none of this in his final redaction, but Bartolf of Nangis clearly had the same version used by William: This Tancred held Tiberias and Haifa, mighty towns that had been conquered in Duke Godfrey’s time, and in Jerusalem he had the temple of the Lord and the street adjoining the temple in his own right, but he was hostile to King Baldwin on account of certain ancient quarrels, and so he was unwilling to be subject either to the king or to his authority.24 William of Tyre offered the further explanation that Tancred was deeply religious and was urged by his conscience not to be bound by an oath of loyalty to someone whom he could not genuinely love. This is unconvincing as a motive for refusing to surrender the two towns, especially from a writer whose sympathies for Tancred and his co-conspirators had already been revealed. Albert of Aachen, whose allegiance was with Baldwin’s party, had a more credible explanation for Tancred’s uneasy conscience. Tancred, according to William, was claiming that Godfrey had granted both towns to him; Albert reported an extended dispute over Haifa. During the fortnight after Baldwin’s coronation, when he was settling such disputes and dispensing justice in Jerusalem, Geldemar Carpenel came before him and claimed that Godfrey had granted Haifa to him as a reward if he, Geldemar, succeeded in capturing it. Albert had previously described the situation, which was by no means as clear-cut as Geldemar asserted. During Godfrey’s five-week terminal illness, Tancred, with Patriarch Daibert and a Venetian fleet, had blockaded and attacked Haifa. The townsmen, many of them Jewish, had resisted stubbornly, and after two weeks the Christian attack was halted. This was, according to Albert, no wonder, since Tancred was not, as he usually did, bravely bringing faithful support with his men, on account of the envy which was gnawing at his heart, because Duke Godfrey, while he was yet alive and lying ill in bed, granted the city as a fief to Geldemar, surnamed Carpenel, a distinguished and noble soldier, if they managed to capture it.25

King of Jerusalem  83 It was Daibert who put heart into the assault, and he did this by imposing a condition: ‘that if with God’s agreement the city should be captured, it would be handed over to whomever in the opinion of the faithful worked hardest in its overthrow’. On what authority he did this (other than his own and maybe God’s) is unclear, and the likely consequences were also dubious. The Venetian forces had withdrawn, including the fleet that was now far away from the city, all except for a single young soldier manning a siege tower, who swore that he would serve until the city was conquered. The Venetian forged a pact with three of Duke Godfrey’s men: Wicher the Swabian, Wiric the duke’s chamberlain and Milo of Clermont.26 With twenty of Tancred’s followers they set about undermining one of Haifa’s towers, but the townsmen counter-attacked with fire to destroy the siege machine. In the end the Christian soldiers, who were not afraid to die for Christ, stood undaunted, suffering every torment by day and night, until their shields were burnt up by flames, shattered by slinging machines, pierced by iron pikes, and so damaged you could see right through them. Yet the following day the town was captured and the enemy fled. The Christians took possession, and, as was usual when a surrender had not been negotiated, they killed and plundered. The Venetian fleet sailed back in and took part. Geldemar then called his troops together to take possession of the town, but Tancred had the stronger forces and drove them out. At this point Daibert and Tancred heard of Godfrey’s death and decided not only to hold onto Haifa, but also to make a play for Jerusalem. To this end they sent a message (already discussed) to Bohemond in Antioch, inviting him to join them in pre-empting Baldwin’s succession. They were as yet unaware that Bohemond had been captured by Danishmend, and events unfolded very differently. This is how it stood at the beginning of 1101 when Geldemar Carpenel brought his complaint to Baldwin. Motivated, no doubt, by a desire to have Haifa under the control of an ally rather than an enemy, Baldwin, acting as king and asserting his right to administer justice, summoned Tancred to Jerusalem to answer the charges. ‘Tancred, however, answered that he was not going to make any reply concerning these things in Baldwin’s presence, because he would not recognize him as king of the city and judge of the kingdom of Jerusalem.’27 Tancred likewise ignored a second and a third message summoning him to Jerusalem, but at last he proposed a meeting on the river bank between Jaffa and Arsuf. Inconclusive talks took place there, and a further meeting in Haifa was arranged for a fortnight later. Before that, however, Tancred received a message from Bohemond’s nobles in Antioch asking him to come and take charge as Bohemond’s heir. Albert presented this as something of a face-saver for Tancred, although William of Tyre said that it was only one of several such messages.28 Tancred delayed his departure to Antioch so that he could meet the king at Haifa on the appointed day, and when it came he handed over Haifa and also Tiberias, to which he had an undisputed claim. Thus they became allies without Tancred’s having to take an oath of fealty; however, he extracted a concession from the king: if Tancred returned from Antioch within fifteen months

84  King of Jerusalem then he would get the lands back and hold them in fief from the king; if he did not return then he would no longer have any claim to them. Geldemar was awarded Haifa, and Tiberias was granted to Hugh of Fauquembergues, both subject to Tancred’s proviso. Tancred took his army and became regent of Antioch for the duration of Bohemond’s captivity.29

Bohemond’s captivity and release Tancred’s biographer, Ralph of Caen, certainly writing with hindsight and probably recording Tancred’s feelings as conveyed to him later, said that Tancred felt he was more of a ‘guest’ than a prince in Antioch because he expected his regency to be short. His first action was to despatch Baldwin of Bourcq, the captain of the guard, to Edessa. Tancred’s memory or his biographer’s record is suspect on this, for (according to Albert of Aachen) Baldwin of Bourcq had been summoned from Antioch to Edessa as soon as his kinsman decided to set out for Jerusalem, and moreover the king designate waited for Baldwin to arrive to take over the rule of Edessa before he made the journey. Ralph’s version probably preserves a larger truth, however: that Tancred regarded the two Baldwins, kinsmen and successive rulers of Edessa, with hostility and suspicion.30 This was reciprocal, at least so far as King Baldwin was concerned. The evidence to do with raising Bohemond’s ransom suggests that Tancred contributed nothing, being content to leave his uncle to languish in captivity, while Baldwin would prefer to see Bohemond in charge – they had established a working relationship already, after all, when they made their pilgrimage to Jerusalem together. Ralph of Caen stated explicitly what others argued from silence: ‘At that time Bohemond’s ransom was an anxiety for the people and especially Count Baldwin who was Tancred’s particular enemy.’31 Baldwin assailed the Antiochenes with warnings, promises and rebukes until they and Bernard, the recently appointed patriarch, came up with the money – 100,000 gold coins. Matthew of Edessa also wrote that the ransom was 100,000 dahekans, or gold coins, and that the Armenian Kogh Vasil was instrumental in raising the money, himself contributing 10,000 ‘while the count of Antioch, Tancred, gave nothing’. If Albert of Aachen is to be believed, this huge sum was a bargain, for Emperor Alexios had offered 260,000 bezants, with the idea that once Bohemond was in the Byzantine emperor’s power he would die ‘either in eternal exile or perpetual damnation, so that he could no longer harm the emperor’s realm with any of his trickery’. Bohemond was probably aware that Tancred had been less than proactive in raising the ransom, for when he returned to Antioch he demanded from Tancred not only the Antiochene lands he had assumed charge of in 1100 but also the conquests made during his regency, including Latakia: Tancred had spent eighteen months of his three-year regency capturing the town.32 Although King Baldwin was reported to be keen for Bohemond’s release, there is no record that he contributed to Bohemond’s ransom. It is likely this was partly because of the shortage of money that was to be a major concern throughout his reign and partly because he was preoccupied by multiple threats and concerns and therefore not too reluctant to see Tancred kept busy as regent of Antioch rather than causing trouble within the kingdom of Jerusalem.

King of Jerusalem  85 If the latter was Baldwin’s hope, it was granted the following year when, in the autumn of 1104, Bohemond set sail for Apulia, where he intended to begin recruiting a new crusade, not, as it turned out, to bring much-needed reinforcements to the kingdom of Jerusalem but to launch an attack on the Byzantine empire. The attack failed, but his attempt to renew his assault on the Byzantines kept him away from Antioch for the rest of his life, that is, until 1111. Meanwhile, Bohemond had entrusted ‘the care and general government of Antioch with full jurisdiction to Lord Tancred, his beloved kinsman’. Once again Tancred was able to pursue a policy of aggressive expansion aimed at defending the principality against both Byzantines and Turks. This meant that until his own death in 1112 Tancred was preoccupied with the government and defence of Antioch, and, indeed, he needed Baldwin as an ally rather than a rival.33

Raymond of Saint-Gilles The only other major leader of the First Crusade who remained in the East after the battle of Ascalon in 1099 was Raymond of Saint-Gilles. Raymond was potentially a threat to Baldwin, for he had been Bohemond’s rival for possession of Antioch and Godfrey’s rival for the rule of Jerusalem. He had acted ambiguously in the weeks following the holy city’s capture, for example by trying to retain the Tower of David, Jerusalem’s citadel, and by frustrating opportunities to take the coastal cities of Ascalon and Arsuf. However, in the autumn of 1099 he had headed north to seek out a territory of his own, preferably in the region of modern Lebanon. In the summer of 1100, however, when Godfrey died and Baldwin succeeded him, Raymond was in Constantinople with his ally Alexios I Komnenos and therefore took no part in the machinations of Daibert and Tancred around the succession in the autumn of 1100.34 He was still in Constantinople the following year when four substantial forces arrived from the West, responding to renewed appeals for reinforcements to protect the new state of Jerusalem following the departure of the First Crusaders once they had fulfilled their pilgrimage vows. The Lombard contingent, numbering some 8,000 soldiers and pilgrims, caused a disturbance in Constantinople and moved off into Anatolia, where they were joined by a smaller German force, led by the imperial constable and comprising about 2,000, and the larger northern French army, about 8,000, led by the renegade crusader Stephen of Blois and Stephen of Burgundy. Emperor Alexios was understandably nervous of the armies congregating in Constantinople, in view of his experiences of the First Crusade; to be fair, he was probably also nervous for the westerners marching through Asia Minor. He advised them to follow the same route as the earlier armies and sent Raymond of Saint-Gilles with them to ensure their safety. However, when rumours reached them of Bohemond’s captivity, the leaders of the Lombard and French armies conceived the idea of rescuing him and abandoned the recommended route. Their forces were attacked by the Turks at Mersivan and massacred. Raymond was among the survivors and managed to reach the Black Sea coast from where he took ship back to Constantinople.35 From there he made another attempt to return to Syria, but he was captured by Tancred in the winter

86  King of Jerusalem of 1101–1102 as he marched south. As a condition of his release Raymond had to agree not to compete for territories in the Antiochene region, including Latakia. Thereafter, and no doubt to Baldwin’s relief, Raymond did not return to Jerusalem, but concentrated his efforts on the region of Tripoli, where he built the castle of Mont-Pélerin. He was wounded there during an attack in the winter of 1104 and died in February 1105.

Daibert of Pisa Mainly as a result of fortunate circumstances, Baldwin’s main past and potential adversaries among the laity were thus neutralised during the early years of his reign. His chief clerical opponent, Patriarch Daibert, had conducted the allimportant consecration and coronation ceremonies, but he was nevertheless a continuing threat to Baldwin’s authority. As mentioned before, in western Europe the struggle between Church and state was decades old. It is sometimes called the ‘Investiture Contest’, but it is better understood as a clerical reform movement that affected every sector and rank of society.36 At its heart was a programme of reform that would take the appointment of bishops out of the hands of lay rulers, eliminate simony (the purchase of ecclesiastical offices) and end excessive influence by laymen who established churches and installed clergy on their own lands. These measures, if successful, would deprive lay rulers of the services of clergy who had fulfilled important roles in government, administration and even military affairs. The conflict thus had practical implications as well as ideological ones, and in the Holy Land it pitted Daibert, a staunch supporter and beneficiary of the reform papacy, against Baldwin, whose family in Lotharingia had fought for the German emperors against the reforming popes. Daibert was obviously going to continue to assert the supremacy of the Church in Jerusalem, the holiest city in Christendom. Baldwin could only rule effectively with the whole-hearted support of his patriarch, in peace and in war. He already had an ally in the previous, interim patriarch, Arnulf of Chocques, who would provide the sort of co-operation he required. Baldwin and Arnulf set about getting rid of Daibert. How they achieved this is described in the next chapter, which covers Baldwin’s relationships with all his patriarchs, but an episode of great symbolic importance took place in the church of the Holy Sepulchre at the first Easter of Baldwin’s kingship.

The failure of the Holy Fire, 1101 The Latin Christians in Jerusalem observed a liturgical year that incorporated such commemorative events as the anniversary of the capture of the city by the First Crusade on 15 July and the Exaltation of the Holy Cross on 14 September. More important events were those of the New Testament. The Nativity was naturally celebrated in Bethlehem, as happened in 1100 when Baldwin was crowned there. But undoubtedly the greatest feast in the Christian calendar was Holy Week, from Palm Sunday through the Last Supper to Good Friday and Easter Day. Since Jerusalem was the scene of Christ’s Passion there was a large element of re-enactment;

King of Jerusalem  87 for example, on Palm Sunday a procession gathered on the Mount of Olives ready to enter the city via the Golden Gate, as Jesus had done. The Latins also incorporated a traditional Orthodox ritual into their ceremonies: the miraculous appearance of the Holy Fire to light the lamps in the Holy Sepulchre on the eve of Easter Day.37 At Easter 1101, Baldwin’s first as king, Patriarch Daibert was out of favour, but he pleaded with the king to be allowed to participate in the Easter rites. On this occasion, for the first time in memory, the Holy Fire failed to appear on Holy Saturday. Besides the usual witnesses, two other chroniclers were in Jerusalem at the time: Caffaro of Genoa and Ekkehard of Aura.38 It was clearly a deeply disturbing experience: Caffaro described a sermon by Daibert claiming that God produces miracles not for the faithful but the infidel, which seems to have been a masterpiece of improvisation. Ekkehard reported a speech by a certain ‘Bishop Hermann’, a series of penitential activities undertaken and finally the spontaneous lighting of the lamps the following day and a veritable son-et-lumière as more and more lit up. Guibert of Nogent included a prominent part for Daibert of Pisa, in accordance with the writer’s sympathies, but also a role in the drama for Fulcher, whose history he was using: ‘Meanwhile Fulcher of Chartres, taking with him Patriarch Daibert’s chaplain, set out for the Mount of Olives, for it was in that place that the divine light used to arise, if ever it was not present in Jerusalem.’39 However, turning to Fulcher’s history in its surviving form, the whole event occupies just one sentence, and he certainly did not claim to have been involved.40 Contrast this with Bartolf of Nangis and you will find three chapters describing the long waiting, the grief, dread and despair: ‘They were indeed afraid that the wrath of God had fallen upon them, and they feared that He who was not sending material fire had given up the fire of His love because of their sins.’41 The Greek and Syrian Christians sang Kyrie eleison in the main church. Patriarch Daibert unlocked the door to the Sepulchre and looked in, not once but three times. Then he prostrated himself before the tomb, weeping copiously. Everyone waited, but there was no fire. As the sun went down Daibert left the Sepulchre and preached a long sermon at the end of which ‘he confessed in the presence of all that he was guilty and very culpable and implicated in many crimes from the beginning of his youth, and he had displeased God by the scantness of his worth’. He then laid aside his staff and abdicated his pastoral responsibility and threw himself down again before the altar, but there was still no result. Everyone had to leave at nightfall, although some kept vigil outside in prayer. In the morning the doors were opened and everyone went in, hoping the miracle had happened during the night. It had not. Then King Baldwin threw himself down before the doors of the Sepulchre, anxious and sorrowful, and he took the blame on himself and his iniquities for the desolation of the people. For before he was called king this misfortune had not happened in the time of any prince ruling there in Jerusalem; for this reason he almost wanted to lay aside his crown and strip himself of his kingdom, principate and dominion, unless he was soon encouraged by some words of consolation.

88  King of Jerusalem People hastened to assure the king that ‘the manner he subdued the kingdom and priesthood to Christian authority was not the reason the Holy Fire was lacking’. They then recollected that it was Easter Day and began the joyful chant. As they processed and the Eastern Christians prayed and chanted the Kyrie eleison it was announced that the Fire had arrived, and there was much rejoicing. ‘Afterwards Patriarch Daibert was restored to his see and re-elected by the common agreement and will of everyone; King Baldwin was clad in the crown and royal garments’ and the Easter Mass began. Guibert of Nogent, using the same source as Bartolf, drew out the significance of the king’s coronation: ‘That day the king himself consented to be crowned within that same city in the Lord’s Temple [i.e., the church of the Holy Sepulchre], something he had never presumed to do.’42 Bartolf’s account is so long and detailed that it cannot be doubted that he took it from his source – Fulcher – with minimal changes, as usual, to remove first-person references. It agrees closely enough with Guibert’s shorter description to reinforce this view. Fulcher later cut it out, which is not surprising in view of the ambiguous endorsement of both king and patriarch within it. Nine hundred years later, in a sceptical age, it seems evident that the whole incident was stage-managed, and to judge by the effects the purpose was to discredit Daibert, as MacEvitt observed. Rubenstein went further and concluded that ‘regnum had achieved such a stunning victory over sacerdotium in 1101’.43 In the future Jerusalem was to be a secular kingdom.

Six years later By 1106 the remaining First Crusaders, Baldwin’s chief or potential rivals for rule of Jerusalem, had been neutralised by circumstances. Daibert of Pisa had died the previous year. The Easter Fire ceremony of 1107 or 1108 was described Russian pilgrim Abbot Daniel, who also gave an impression of King Baldwin’s self-confidence and his cultivation of a regal presence.44 Daniel referred to Baldwin as ‘prince’ (kynaz) and ‘king’ or ‘emperor’ (tsar). He had learned that the king was going on a military expedition which would take him past the sea of Tiberias, and he asked if he could go with him to visit holy places in that area. Daniel particularly said that he bowed to the king when he approached him. As part of the king’s retinue and protected by the king’s soldiers he felt quite safe on the journey, although he wrote that it was impossible to travel by this route without military protection. Daniel reached Tiberias safely and stayed there, visiting the sites, until Baldwin and his soldiers returned and escorted him back to Jerusalem. Abbot Daniel took advantage of his connection with the king the following Easter when he wanted to be present at the annual miracle of the Holy Fire. At dawn on Good Friday he went to the king ‘and bowed down to the earth before him’. The abbot went on to stress the king’s friendly response, ‘for he is a kind and very humble man and not in the least proud’. Baldwin sent ‘his best servant’ with the Russian abbot to place a lamp on the tomb of the Lord, so, following instructions, Daniel went out and bought a very large lamp and filled it with oil. He described at great length the ceremony and service that followed on Holy

King of Jerusalem  89 Saturday. King Baldwin himself stood in awe and humility and wept. In the evening he led a procession, barefoot, to the church of the Holy Sepulchre, and the abbot was allowed to walk alongside the king with the abbot of St Saba. When they arrived the throng was so great that the king had to command his soldiers to force a way through. Daniel was given an excellent vantage point from which to await the manifestation of the Holy Fire. When it appeared King Baldwin’s candle was lit first from it, then from the king’s candle the ecclesiastics lit theirs and from the churchmen’s candles the laity were able to light theirs. Abbot Daniel’s lamp was one of three that stood on the altar within the tomb itself and were lit directly and miraculously by the Holy Fire. On Easter Sunday he was able to fetch it and was secretly rewarded for his devotion with a small fragment of the tomb to take back to Russia as a relic. The Easter ceremony was the natural climax and culmination of Abbot Daniel’s pilgrimage, and he returned to Russia straight afterwards. As well as demonstrating the king’s part in a successful Easter Fire service, Daniel offered a rare and sympathetic glimpse of Baldwin’s personality and authority. Notably, the patriarch of Jerusalem was not mentioned at all, a circumstance that will be explored in the next chapter.

Notes 1 Jean Flori, ‘Jérusalem terrestre, céleste et spirituelle: Trois facteurs de sacralisation de la première croisade’, in Jerusalem the Golden: The Origins and Impact of the First Crusade, ed. Susan B. Edgington and Luis García-Guijarro (Turnhout, 2014), pp.  25–50, discusses the loss of the eschatological character of the expedition: ‘la descente de la Jérusalem céleste ne s’est pas produit et Ascalon n’était donc pas Harmaguédon’ (p. 50). 2 Murray, Crusader Kingdom, pp. 94–97, for the whole question of kingship; p. 95 for the quotation. 3 See Crouzet-Pavan, Le mystère des rois de, pp.  213–22, for the Capetian parallels; Murray, Crusader Kingdom, p. 27, for the details of the Bouillon-Boulogne Carolingian ancestry. Another relevant example of evolving theories of kingship is William I of England: William M. Aird, Robert Curthose, Duke of Normandy (c. 1050–1134) (Woodbridge, 2008), pp. 63, 210. The succession crisis precipitated by Robert’s return home from crusade offers another insight into exactly contemporary attitudes to questions of succession and primogeniture: see Neil Strevett, ‘The Anglo-Norman aristocracy under divided lordship, 1087–1106: A  social and political study’ (Ph.D. thesis, University of Glasgow, 2005), pp. 182–209, especially pp. 191–94. 4 Luc Ferrier, ‘La couronne refusée de Godefroy de Bouillon: eschatologie et humiliation de la majesté aux premiers temps du royaume latin de Jérusalem’, Le Concile de Clermont de 1095 et l’appel à la croisade, Collection de l’École Française de Rome, 236 (1997), 245–65. 5 For the Capetians’ coronation rites, see Yves Sassier, Louis VII (Paris, 1991), pp. 20–21. 6 F-E, p. 162. Note that Daibert’s fall from grace was viewed very differently by William of Tyre: see chap. 6. 7 BN, chap. 44, p. 522. 8 AA 7.37, p. 540. Money-fiefs were not unknown in the West; in the East they allowed the king closer control of defence and the nobles to acquire a money-income  – something more useful in the East than in western Europe. There is more on this in chap. 9. 9 FC 2.3.12–15, pp. 367–69.

90  King of Jerusalem 10 BN, chap. 45, p. 522. 11 AA 7.38, p. 542. 12 FC 2.4.2–3, pp. 372–74. Fulcher called the settlements villas, which probably translates like the Norman vilel or English vill, small villages rather than country estates. Guibert of Nogent had a garbled version of the same story, p. 340: he described the area as the slopes of Mount Sinai and the people as barbarians who resembled Ethiopians. Baldwin spared their lives because they were so wild and ugly. 13 AA 7.39–40, pp. 542–46. The caves were probably those of Beth Guvrin (Arabic Bayt Jibrīn), about halfway between Ascalon and the Dead Sea, which were originally quarries but were occupied from the late Roman period through the Arab conquest: see A. Kloner, ‘Beth Guvrin’, in The New Encyclopedia of Archaeological Excavation in the Holy Land, ed. E. Stern, A. Lewinson-Gilboa and J. Aviram, 4 vols (Jerusalem and New York, 1993), I: 195–201. 14 FC 2.5, pp. 376–84 (Hagenmeyer cites Josephus, De bello Judaico, as a source). 15 AA 7.41–42, pp. 546–48. 16 WT 10.9, p. 463; AA 7.43, p. 550. 17 FC 2.5.12, p. 384. 18 BN chap. 45, p. 523. 19 FC 2.6.1–3, pp. 385–87 cf. WT 10.9, pp. 463–64. 20 For Charlemagne’s title, see Monumenta Germaniae Historica: Diplomata Carolinorum, I: 77. (title used from ce 801). 21 BN chap. 45, p. 523. 22 Some historians have argued that Bethlehem was a necessary relocation because Jerusalem was an ecclesiastical state headed by the patriarch. However, this claim by Jean Richard rests on the idea that Jerusalem was ‘an ecclesiastical seigneury for the benefit of the Holy Sepulcher’ placed under ‘the protection (advocatia) of a great baron already established with territories around the holy city’. Richard went on to claim that Godfrey of Bouillon, as advocatus, did not exercise temporal power in the city, but set out to create a state within the territory of Palestine. The claim was based ultimately only on William of Tyre and his version of the letter of Daibert to Bohemond sent to attempt to prevent Baldwin’s succession. Whether Daibert of Pisa believed he had a legitimate claim to suzerainty in Jerusalem, or merely wanted it to be so, cannot be known: Jean Richard, Le Royaume latin de Jérusalem (Paris, 1953), p. 32. However, Richard also claimed: ‘il était tout naturel que Jérusalem appartînt à son église puisque, comme Bethléem ou Nazareth, c’était une ville sainte’, which appears to rule out Bethlehem too. See also Jean Richard, ‘The Political and Ecclesiastical Organization of the Crusader States’, in A History of the Crusades, ed. K. Setton, 6 vols (Madison, 1969–1989), V: 193–217. For Daibert’s letter see above and WT 9.16, p. 441, and 10.4, pp. 456–58. Patriarch Daibert’s ambitions, both political and personal, are discussed in the next chapter. For Samuel and David: 1 Sam 16: 12–13. Note that several historians compared Godfrey to David, particularly when he fought with a ‘giant’, so parallels were inevitably recognised. 23 WT 10.9, pp. 463–64. 24 BN chap. 46, p. 523; cf. FC 2.7, pp. 390–93. 25 AA 7.23–27, pp. 516–24. 26 AA 7.44–45, pp. 550–54. Albert called the three named individuals Godfrey’s men. For Wiric, see Murray, Crusader Kingdom, pp. 79, 138; for Milo, p. 218. The presence of Godfrey’s men may explain the circumstantial detail of AA’s account. 27 AA 7.44–5, pp. 550–54. 28 WT 10.9, p. 464. 29 BN chap. 46, pp.  523–24, corroborates AA: the message from Antioch precipitated Tancred’s decision to hand over Haifa and Tiberias to Baldwin. RC, Tancred’s biographer, recorded none of the negotiations, still less a rapprochement between Baldwin

King of Jerusalem  91

30 31

32

33

34

35

36 37 38 39

and Tancred. As if it immediately followed Baldwin’s accession and coronation, he wrote: ‘Therefore the heirs took their places, Baldwin as ruler of Jerusalem, Tancred of Antioch, and they rivalled each other as they raced towards renown. Then – fearing that the younger would be unworthy of his elder in valour – each was on fire with envy of the other.’ RC, chap. 143, p. 706. For the history of Galilee, Aryeh Grabois, ‘Galilee’, in Crusades Encyclopedia, II: 495–98. RC, chap. 143, p. 706; cf. AA 7.32, p. 530. For the ransom: RC, chap. 147, p. 709, says decem miriadibus Michelatorum (i.e. 10 x 10,000). Coinage of Michael VII Doukas (1071–1078) that bore his bust was higher in gold content than the debased coinage of his immediate successors: AK, p. 525, n. 31. See also AA 9.34, p. 680. Orderic Vitalis, who agreed as to the emperor’s motive, said he offered 100,000 ‘philips’ (bezants): OV, V: 354–56; Tancred’s (lack of) contribution, ME 3.14, pp. 191–92. For Bohemond’s treatment of Tancred after his release: RC, chap. 147, p. 709. However, WT told a very different story, that Bohemond was welcomed back to Antioch and was very grateful for the loyalty and prudence with which Tancred had ruled: 10.24 (25), p. 483. For Latakia see RC, chaps. 144, 146, pp. 706–707, 708–709; FC 2.23.2, pp. 459–60; WT 10.22 (23), pp. 481–82. Possession of Latakia was much contested in the period of and after the First Crusade: Lilie, ‘Genoa’, Appendix 1, pp. 259–76, for the contradictory evidence. It should be pointed out that William of Tyre, writing much later, wrote as if he believed Bohemond’s motivation for his ‘crusade’ was quite different: ‘Bohemond, much wearied by the weight of borrowed money, sailed to Apulia so that he could repay his debts and in order to bring back with him greater military forces from the regions beyond the sea’ (WT 11.1, p. 495). For Tancred’s regency, see Asbridge, Creation of the Principality, pp. 59–68. For Raymond himself, see John H. and Laurita L. Hill, Raymond IV, Count of Toulouse (Syracuse, NY, 1962); Jean Richard, ‘Raymond of Saint-Gilles’, in Crusades Encyclopedia, IV: 1011. For the county of Tripoli, see Kevin James Lewis, The Counts of Tripoli and Lebanon in the Twelfth Century: Sons of Saint-Gilles (Abingdon, 2017). Collectively, this was the Crusade of 1101. The fullest discussion is Alec Mulinder, ‘The crusading expeditions of 1101–2’ (Unpublished Ph.D. thesis, University of Wales, Swansea, 1996); Mulinder, ‘Crusade of 1101’, in Crusades Encyclopedia, I: 34–37. The most complete contemporary account is AA book 8, pp.  586–636. Matthew of Edessa, always hostile to the Byzantines, asserted that Alexios ordered his officers to lead the armies into ‘desolate regions’ where the water was brackish and to feed them bread adulterated with lime as revenge for the First Crusaders’ broken vows: ME 3.5, pp. 185–86. More plausibly, see AA 8.37, p. 628, and 8.46, pp. 634–36. Ekkehard, who was part of the Bavarian army, reported that he was one of those who took ship from Constantinople to Jaffa because of rumours of Alexios’s treachery: F-E, pp. 164–70. Some of the leaders also survived and eventually reached Antioch and/or Jerusalem. Their stories, where they intersect with Baldwin’s, will be picked up at the appropriate point in the narrative. For an account of the reform papacy and the ‘Investiture Contest’, see Brian Tierney, The Crisis of Church and State 1050–1300 (Toronto, 1988; repr. 1996), pp. 55–95. For a discussion of many aspects of the Holy Fire ceremony, including the significance of its failure in 1101, see Jay Rubenstein, ‘Holy Fire and Sacral Kingship in Postconquest Jerusalem’, Journal of Medieval History, 43 (2017), 470–84. Caffaro, ‘Annales Ianuenses’, in Annali Genovesi, ed. L. T. Belgrano, pp. 3–75; F-E, pp. 176–78. GN, p. 342. BN did not include the information about Fulcher, but this would tally with his usual practice of omitting Fulcher’s references to his own experiences. WM was informed about Baldwin’s reign by FC, but he probably used an intermediate version

92  King of Jerusalem

40 41 42 43 44

(between the 1107 one used by BN and the final redaction of 1127) as he had a different account of the episode: 4.379, pp. 674–76. FC 2.8.2, pp. 395–97. BN chaps. 46–49, pp. 524–26. GN 7.42, p. 343. MacEvitt, Rough Tolerance, pp. 115–20; Rubenstein, ‘Holy Fire’, p. 483. Jerusalem Pilgrimage, 1099–1185, ed. and trans. John Wilkinson with Joyce Hill and W. F. Ryan (London, 1988), pp. 120–71 (pp. 154–55, 165–71). In the editor’s ‘Introduction to the Texts’, pp. 9–10, he suggested that the names of Daniel’s companions and the favour shown to him by King Baldwin suggest that his visit to Jerusalem was possibly as much a diplomatic mission as a pilgrimage, a supposition strengthened by the inclusion of torty-two Greek words in the text, since knowledge of Greek was probably not usual for a Russian cleric.

6 Ecclesiastical affairs

Ecclesiastical affairsEcclesiastical affairs

Unlike his brother and predecessor Godfrey, Baldwin was prepared to engage in a series of fierce and protracted disputes with the patriarch of Jerusalem in order to consolidate his royal authority. It should not be doubted that these were political confrontations, fuelled by Baldwin’s urgent need, especially in the early years of his reign, for the patriarch as an ally in the defence of the kingdom, not only with prayers and counsel but also providing soldiers and the means to support them in the king’s service. Aside from this imperative, Baldwin appears in contemporary documents and narratives as a conventionally pious man and ruler. Surviving records show that during his reign he made and confirmed several grants to the church of the Holy Sepulchre, where Godfrey had installed twenty canons in 1099; to the abbey of Mount Tabor; many to the abbey of St Mary in the valley of Jehosaphat; one each to the nunnery of St Anne in Jerusalem, the hospital of St John in Jerusalem, the abbey of St Mary of the Latins, the Templum Domini and the abbey of Mount Zion. He had a particular regard for the church in Bethlehem, the place of his coronation, and at some time in the winter of 1109–1110 he announced, after describing his coronation, that he wanted to raise Bethlehem to the status of bishopric. Archdeacon Arnulf and Deacon Aicard were sent to Rome, and Pope Paschal II ordered Archbishop Gibelin of Arles to hear their case and accompany them back to Jerusalem. Gibelin duly installed Aschetinus, who was cantor in Bethlehem, as bishop. King Baldwin confirmed the appointment and his own eleemosynary gift of the town of Bethlehem and five estates. He had settled some disputes Bethlehem had had with the church of Jerusalem through the exchange of lands and vineyards in the neighbourhood of Jerusalem. He also promised to confirm gifts to Bethlehem made by others.1 It may be significant that by this date he had found a patriarch he could work with, Gibelin of Arles. The first half of his reign had been a very different story.

The patriarchate of Jerusalem Traditionally, Christendom comprised five patriarchates: Rome, Constantinople, Antioch, Alexandria and Jerusalem. In 1054 a fissure began to open between Rome on the one hand and the four Orthodox patriarchates on the other; it was superficially about doctrinal points, but at a deeper level it was a battle for primacy

94  Ecclesiastical affairs between Rome and Constantinople. Pope Urban II’s call for an expedition to assist Constantinople and possibly to liberate Jerusalem showed that the breach was not yet viewed as irreparable. This is borne out by the co-operation of the Jerusalem patriarch with the crusade as it progressed; in fact, Patriarch Symeon II of Jerusalem may have helped initiate the crusade by making appeals for help to the Roman papacy.2 Symeon became patriarch of Jerusalem sometime after 1088. He was forced into exile by the city’s Turkish governor in October 1097 and went to Byzantine Cyprus. Around the same time, he made contact with the crusaders in northern Syria, and two letters sent from Antioch by clerical leaders to western Christendom included his name. During the siege of Jerusalem, according to Albert of Aachen, he sent costly provisions from Cyprus to Duke Godfrey in expectation of his return to the city. Albert further recorded that Symeon died before he could be restored to the patriarchate, although there is some doubt whether the crusaders were aware of this in July 1099.3 The crusaders seem to have been clear at that time that they needed both a secular ruler and a spiritual leader, and it is likely that, regardless of Symeon’s claim and even if they thought he was still living on Cyprus, they would have appointed a Latin cleric to the patriarchate.4 It is also probable that they recognised that any final appointment would rest with the pope. However, Albert of Aachen wrote explicitly that there was no one then in the Holy Land worthy of ‘so great an honour and divine governance’. Because of this, he went on, Arnulf of Chocques, ‘a cleric of wonderful wisdom and eloquence’, was appointed not as patriarch, but as chancellor, procurator of holy relics and keeper of alms.5 Bartolf of Nangis, writing before 1107, agreed that Arnulf of Chocques was advanced as ‘a sort-of patriarch’ until they had found out from the pope in Rome what they should do, while Fulcher went further when he revised his text, perhaps twenty years later: ‘Moreover, then they decided that there should not yet be a patriarch there until they had enquired of the Roman pope whom he recommended for the office.’6 Even Guibert of Nogent was not totally clear on Arnulf’s position before Daibert’s arrival. He conceded that there was a scarcity of learned men in Jerusalem, and so, ‘at a time when eloquence was valued over conduct’, Arnulf was called to become patriarch of Jerusalem. Guibert went on: ‘Therefore for a while he offered a show of being a bishop in name only, but he adorned his new appointment by his speeches.’7 The sources agree, therefore, that Arnulf was locum tenens, a caretaker patriarch. In fact, he held the post only until Christmas 1099, when Daibert of Pisa was appointed in his place. William of Tyre glossed over Arnulf’s brief and temporary tenure of the patriarchate: ‘the Jerusalem church had been as if vacant for five months until that day, not having a permanent bishop’.8 One of the reasons for Arnulf’s ready agreement to stand aside was probably his knowledge that his had been only an interim appointment. Also, he was to receive as compensation the positions of chancellor and archdeacon of the Holy Sepulchre. In addition, Arnulf remained a close ally of Godfrey and the other leaders, including Baldwin, and he was privy to their discussions in Jerusalem at Christmas 1099. There were overwhelming practical reasons for advancing the Pisan archbishop which were stated very clearly by Bartolf of Nangis. First,

Ecclesiastical affairs  95 Godfrey, with the support of the clergy and people, argued ‘that Daibert excelled Arnulf and would be a great asset to the whole realm, first because he was learned and especially erudite in letters, then because he had already learned to govern and work for the benefit of both domestic and Church affairs’. But, Bartolf added: There was also another more important reason why they [Godfrey, the clergy and people] upheld him: for Daibert had Pisans and Genoese, with whom he had come, who were as if in his power, so that whatever he wanted, they wanted and did. And so they [Godfrey etc.] brought something indispensable and very advantageous to their government if they had such a man by whose industry and ingenuity they might capture the cities situated on the sea by naval means.9 Godfrey, therefore, made a policy decision to favour Daibert in order to secure the service of the Pisan navy, which he foresaw as invaluable for the strategic priority of conquering the coastal cities and their ports. He probably did not perceive the complexities of Daibert’s personality and ambitions, or thought he could keep them in check, and of course he did not know that he would die the following July and leave an ambiguous legacy to his brother Baldwin. For Baldwin needed the Italian fleet no less than Godfrey, but Daibert had openly opposed his succession.

Daibert of Pisa Baldwin’s first acquaintance with the Italian archbishop was in December 1099 when he and Bohemond, who had also been absent from the capture of Jerusalem in July, set out for Jerusalem to complete their pilgrimage. On the surface, Daibert had impressive credentials: he had been a close ally of Urban II and had been in Urban’s entourage when he toured France in 1095–1096 to preach the crusade, which made him an obvious choice as successor to Adhémar of Le Puy when news of the papal legate’s death was received by the pope. Adhémar died at Antioch on 1 August 1098, and a surviving letter from the leaders to the pope, telling him of his legate’s death and asking him to come himself to lead the crusade to Jerusalem, was dated 11 September. The sailing season in the Mediterranean extended roughly from the spring equinox to the autumn equinox, for the avoidance of winter storms, so there could certainly be no speedy response, and it was in the summer of the following year that the new legate set out with a Pisan fleet to take to the assistance of the crusaders.10 Delays and diversions on the voyage meant that Daibert did not arrive until late September 1099, by which time circumstances had greatly changed. He probably expected to find the crusade still stalled at Antioch – hence he put in at a Syrian port rather than at Jaffa – and he now heard that Jerusalem had been captured. Furthermore, he learned that his sponsor, Urban II, had died just a fortnight after Jerusalem was taken, and therefore Daibert’s legatine authority was annulled. It would have been logical for Daibert to make his way with all speed to Jerusalem, but the first leader he came into contact with was Bohemond, who was

96  Ecclesiastical affairs besieging Latakia, a port that belonged to Byzantium. The Pisan fleet was just what Bohemond needed to complete his blockade by sea as well as land. Albert of Aachen insisted that Daibert was deceived by Bohemond, who justified the siege on the grounds that the inhabitants were heretics, but Daibert probably needed no persuasion, for the Pisan fleet had raided Byzantine islands en route to Syria. After the capture of Latakia, Daibert heard that there were crusaders nearby who had completed their pilgrimage and were now returning to their homes in western Europe, including Robert of Flanders and Robert of Normandy. Raymond of Saint-Gilles was also with them although he had probably already decided to make his future in the East. These princes revealed to the Pisans that Bohemond had tricked them into participating in the siege, and the consequent withdrawal of the Pisan fleet forced Bohemond to withdraw, ceding possession of Latakia to Raymond and the Byzantines. Finally peace was made, but Daibert did not turn against Bohemond and still did not hurry to reach Jerusalem. In the end it was at Bohemond’s instigation that Daibert accompanied him to Jerusalem as Christmas 1099.11 So at this point Baldwin met Daibert for the first time and spent some weeks in his company. According to Albert of Aachen’s hostile account, Daibert spent the journey ingratiating himself with the two leaders ‘with all his speechifying and pretended piety’. How far Baldwin was won over by Daibert’s charm offensive is debatable, but then again, he may not have foreseen how Daibert would behave once in Jerusalem. Albert claimed that the archbishop set out to achieve promotion to the patriarchate by every means, including simony, and he backed up his accusation with a surprisingly circumstantial (but uncorroborated) story that when Daibert was sent as an emissary to King Alfonso of Leon and Castile (d. 1109) he not only received lavish and precious gifts for himself, but also a golden ram for the pope which he secretly kept. Albert claimed that it was this Spanish treasure that Daibert used to bribe Bohemond, Baldwin and Duke Godfrey himself. He was duly elevated to the rank of patriarch of Jerusalem, and in this position he worked on Godfrey to secure ecclesiastical suzerainty as previously described.12 William of Tyre explained that the princes who were in the city at Christmas 1099 assembled specifically for the purpose of sorting out the situation in the Church, and Daibert was advanced to the patriarchate by common consent, ‘for as to what I said had previously been done about Arnulf: just as it had been done unwisely, so it was undone both quickly and easily’. There was no appreciation of Arnulf’s standing aside, as William of Tyre was unremittingly hostile to Arnulf and a staunch supporter of Daibert. His partiality is perhaps explained by his sense of the dignity of the patriarchal office and the proper relationship between Church and state; he was himself a prominent churchman as archbishop of Tyre (perhaps with ambitions to occupy the patriarchate of Jerusalem himself), as well as being chancellor of the kingdom of Jerusalem.13 A great deal has been made of a few lines in William’s chronicle, where he wrote that Daibert’s first action as patriarch was to confer formal investiture on Duke Godfrey of his kingdom, and to invest Bohemond with the principality of Antioch.14 He may well have been elaborating an observation by Fulcher of

Ecclesiastical affairs  97 Chartres: ‘Also in Jerusalem Duke Godfrey and Lord Bohemond received their land from Patriarch Daibert for the love of God.’15 On the surface this act seems to be an early indication of Daibert’s territorial ambitions, but that would be to view it with hindsight. Godfrey may well have regarded the patriarch’s role as similar to that played by the archbishop or senior churchman in coronation ceremonies in the West or, as Alan Murray has suggested, like the ceremony where the pope conferred full imperial authority on the elected king of Germany. William of Tyre did make the important distinction, echoing Fulcher, that the princes considered they were devoting their office to God, whose minister and vicegerent on earth Daibert was believed to be. Baldwin was almost certainly present at the ‘investiture’, although there seems to be no hint that he was invested with his county of Edessa. How he interpreted the ceremony and whether he foresaw what difficulties it would cause is not recorded.16 He had left Jerusalem for Edessa before the next act in the drama, and, again, we can do no more than speculate what part he might have played in it. William found himself in difficulty when he tried to account for the dispute that soon arose between two of his heroes, Daibert and Godfrey, about Jerusalem and Jaffa. He took the traditional course of blaming the evil counsel of troublemakers, but he did not say who was listening to the advice of these men, whether it was the duke, the patriarch or both. He presented Daibert as demanding the city of Jerusalem with its citadel and also the city of Jaffa with its appurtenances.17 Not surprisingly, his demand caused intense debate which went on for some time. In the end (William said) in February 1100, Godfrey, who was a humble and Godfearing man, handed over a quarter of Jaffa – not to Daibert personally or as patriarch, but to the church of the Holy Sepulchre. However, Daibert obviously did not give up, for at Easter when the clergy and people were assembled for the festival Godfrey granted to the patriarch Jerusalem with its citadel, reserving for himself the right to use and profit from both Jerusalem and Jaffa until he had captured one or two other cities, including Cairo. At this point he made the pledge that Daibert relied on in his opposition to Baldwin’s succession: that if Godfrey died without an heir then all the possessions would pass under the control of the patriarch. The importance of this episode for the relationship between Church and state during the whole period of Latin rule in Jerusalem is reflected in the rest of this chapter in William of Tyre’s Chronicon and the two that followed, which William inserted into his narrative, probably in the 1170s. He purported to be justifying the patriarch’s claims, but his digression dealt only with that area of Jerusalem around the church of the Holy Sepulchre that had been recognised as Christian before the capture of the city by the crusaders, and it seems to have been devised to show off his learning and historical knowledge rather than to explain how the patriarch could lay any claim to both Jerusalem and Jaffa. It is difficult to deduce anything useful from it, but Edbury and Rowe, in their study of William of Tyre, rejected the idea that the chronicler was suggesting that Daibert was trying to exert control over the kingdom of Jerusalem and maintained rather that William was intimating that Daibert aimed to establish ‘a substantial ecclesiastical lordship within the kingdom’.18 Ultimately, we cannot know what was in Daibert’s mind, but his

98  Ecclesiastical affairs reported actions around Baldwin’s succession and during the early years of his reign suggest intense ambition, either for himself or on behalf of the Church.

Baldwin’s accession Baldwin thus had his personal reasons for disliking Daibert when he arrived to rule Jerusalem, but these were less important than the necessity to resolve the ambiguities surrounding the power relationship between Church and state in the nascent kingdom. If the participants in the First Crusade had had any vision of a future state centred on Jerusalem it was probably rather vaguely and idealistically as a theocracy, where Christ ruled as king. In practical terms this was impossible to realise, and it had faded in the face of the immediate emergency of enemy attack, when it was imperative to reach compromises and temporary solutions in order to fight the battle of Ascalon. In the rather longer term, too, defending the kingdom and expanding its territories at a time when most of the pilgrims were abandoning the Holy Land induced Godfrey to prefer Daibert over the loyal Arnulf, who could not offer the strategic advantage of a naval force. Evidently, Godfrey had had to reach an accommodation with Daibert that involved receiving his lands from him – but whether Godfrey was acknowledging the overlordship of the patriarch of Jerusalem in Daibert’s own person, or as representative of the pope in Rome, or of God himself, is (and probably was at the time) open to interpretation.19 Following his coronation, Baldwin could dispense with Daibert as patriarch and was doubtless keen to do so. The king of Jerusalem needed a patriarch who would function as more than a clerical leader. In most territories in western Europe senior churchmen were ministers of state and played vital roles in secular government. Loyal ministerial support would not be enough, though: it was very clear that the new kingdom was going to have to fight for its existence. The crusaders believed that they had won victory in battle because God was on their side, and it was important for Baldwin to channel the same religious fervour in future battles against the same enemies. The visible sign of God’s presence was to be a relic of the True Cross that had been revealed to the crusaders in the days following the capture of the city in 1099 and had been borne into battle at Ascalon. The relic’s guardian and standard-bearer should be the patriarch of Jerusalem. So replacing Daibert with a loyal patriarch was a matter of urgency.20 Accordingly, Albert of Aachen recorded, Not long after this the king charged the patriarch of Jerusalem in the presence of the whole Church concerning the treachery he had practised with Tancred against him: he had claimed that Baldwin was not a worthy heir to succeed Godfrey, but that Bohemond, who was not of his blood, should possess the kingdom. He did this because the patriarch was widely accused of this crime by his nobility.21 Evidently Daibert was obdurate, for eventually Baldwin appealed to the pope, Paschal II (1099–1118), accusing the patriarch of inciting murder and discord

Ecclesiastical affairs  99 ‘between the Christians’ leaders and the new and tender Church’. Pope Paschal responded by sending a legate, Maurice the cardinal-bishop of Porto, who was charged with examining the charges against Daibert. If the patriarch was acquitted then Cardinal Maurice would act as the pope’s deputy and consecrate him; if he was condemned then Maurice had apostolic power to depose him.22 All the parties agreed to abide by his verdict, and a council of bishops and abbots was held, at which King Baldwin accused Daibert of perjury, treachery and conspiracy to murder (on the grounds that Baldwin might have been killed by Bohemond on the way from Edessa to Jerusalem). He produced the intercepted letter as evidence and called on the churchmen as witnesses. Perhaps worst of all, Daibert was accused of sacrilege because he had ‘partly diminished and dispersed’ the wood of the True Cross. Daibert was unable to mount a convincing defence, and so he was suspended from office and the council was adjourned. The council was held in Lent 1101, and Easter was approaching (21 April). This presented a crisis for Daibert because he was excluded from the traditional rituals of Maundy Thursday and Good Friday, which the cardinal performed instead. He approached the king and beseeched him in every way he could, even reminding him that he owed his coronation to Daibert. The king held firm, Albert claimed, until the patriarch whispered in his ear that he would pay a massive bribe – 300 bezants  – for Baldwin’s intervention with the cardinal.23 Baldwin, who was as ever greatly in need of money to pay for the defence of his realm, acquiesced and interceded for him. Maurice agreed to suspend the suspension until after the ceremonies around Easter, and Daibert donned the patriarchal robes and oversaw the celebration. The failure of the Easter Fire to appear has already been described, with the despair that followed until it made a belated appearance. Unforeseen consequences of the truce for Easter were, first, that Daibert and Maurice became very close friends and accomplices and, second, that Daibert realised that Baldwin’s weak point was his chronic lack of funds. Once again, we are dependent on Albert of Aachen for an account of how the situation developed, but it must be noted that Albert, as a Rhinelander, was firmly on the side of Baldwin and Arnulf. As Albert told it, later in the same year the garrison in Jaffa was importuning the king for wages owed to the soldiers for their service not only to him but also to his predecessor. So the king went to Jerusalem and asked the patriarch for (more) money for the defence of the Holy Land. The patriarch offered him 200 silver marks and said this was all he had, which the king accepted. Baldwin believed what the patriarch said, but Arnulf, as chancellor of the Holy Sepulchre, said that this money was not Daibert’s to give, and he and many others accused the disgraced patriarch of embezzling church funds. The king was angry and renewed his demands insistently. Albert then included a vivid depiction of the patriarch and the cardinal ‘feasting together on the fat of the land’ while they shared between them the offerings of the faithful that properly belonged to the church of the Holy Sepulchre. They ignored the king’s repeated demands for the patriarch to deliver and equip forty soldiers for service to the defence of the Holy Land and carried on feasting and drinking away the offerings of the faithful until it was pointed out that there would soon be none left.24

100  Ecclesiastical affairs Baldwin was presented by Albert as credulous and determined not to think ill of Daibert, but it is impossible to believe this was the case. He had already outmanœuvred Daibert’s plot to prevent his accession and stage-managed his own coronation at the time and place of his choosing; he was aware of Daibert’s venality from the bribe already offered. His demand of the patriarch now was not for another bribe; it was a legitimate claim for knight service to defend Jerusalem. Shifting the responsibility for exposing Daibert onto Arnulf distanced the king and allowed for public exposure of the patriarch’s iniquities. It was not the last time that Baldwin used Arnulf as a convenient stalking horse. Duly informed of the true state of affairs, the king made a scathing attack on the patriarch that Albert recorded as direct speech, clearly for dramatic effect, since it is unlikely he had an informant who had witnessed it. Baldwin accused Daibert of spending the offerings of the faithful on feasting (this was difficult to deny since the two churchmen were reclining at table at the time) when Christians were in danger; he reminded him that the crusaders had redeemed Jerusalem with their own blood; and he threatened to cut off the funds the patriarch was wasting. The patriarch was no less angry and accused Baldwin of trying to reduce the Church to tributary and servile status; his threat was an anathema from the pope. Cardinal Maurice meanwhile listened but merely ‘warned them about peace and concord’. The king could bear it no longer and burst out ‘harshly and intemperately’. He warned that it was better to use the income from offerings to defend Jerusalem than to allow the Saracens to carry it off by force. He threatened to strip the gold from the Sepulchre and its altar himself if he had to, in order to sustain the defence of the Holy Land, and promised that if the threat from Egypt ever passed he would restore everything even more splendidly. Albert reported that the king, ‘a man instructed in letters’, had proved the patriarch wrong, and Maurice advised Daibert to provide and maintain the soldiers, though only thirty, not the forty originally asked for. However, although Daibert agreed to do this, he soon got tired of the whole thing and realised how costly it was, so he stopped paying the soldiers, turning a deaf ear to the king’s nagging about it. The drama as written by Albert was damning to Daibert, but presented Baldwin as kingly in his wrath, as a just judge and surprisingly patient.25 However, the stand-off might have endured even longer had there not arrived an envoy from Roger I  of Sicily (1062–1101) who testified in the king’s presence and that of ‘the entire Church’ that he had brought from Apulia 1,000 gold bezants and presented them to the patriarch with instructions as to how they were to be distributed. One third was destined for the canons of the church of the Holy Sepulchre; one third was intended for the hospital for the support of the weak and the sick; one third was to go to the king ‘to maintain and reward soldiers who had lost their weapons and possessions’. The patriarch had kept the lot. It was now easy to convict Daibert of fraud and treachery, and he ran out of excuses. He was deprived of his office, and his staff and accomplices were imprisoned, although Cardinal Maurice apparently escaped censure. Daibert withdrew to Jaffa, where he spent the winter (1101–1102), and then in March he set out by sea for Antioch and his former co-conspirator Tancred. Meanwhile his chamberlains were forced

Ecclesiastical affairs 101 to reveal where Daibert kept his money buried, and 20,000 gold bezants were recovered, along with a great amount of silver. Baldwin used the money to pay his soldiers and other debts. Cardinal Maurice he kept with him and treated well, since he was a papal legate. Albert made it clear, albeit implicitly, that Baldwin’s quarrel was with Daibert as an individual and not with the clergy at large, the legate or the papal see.26 Daibert had another shot at reinstatement in 1104, when Baldwin needed support against the Egyptians. Daibert returned to Jaffa with Tancred, a large army and a party that included Baldwin II of Edessa, William of Poitiers and William the Carpenter. These leaders attempted a form of blackmail against King Baldwin, ‘asking in so many words that he reappoint the patriarch to his see, otherwise they would find themselves quite unable to go down to Ascalon to avenge his men’. The king was unwilling to agree because he was still annoyed about the hidden hoard of money, but he was persuaded by his men to reach a compromise: first they would fight the Egyptians, then he would deal with the matter of the patriarch. The proceedings would be overseen by Cardinal Robert of Paris, who had been sent as papal legate following the death of Maurice specifically ‘for the examination and correction of the unlawful events in the holy and universal Church in these eastern parts’.27 The offensive outside Ascalon duly took place but was inconclusive, and afterwards King Baldwin and the four northern allies returned to Jaffa where a church council was held. At this, bishops, abbots and other clergy unanimously decided Daibert should return to Jerusalem and be reinstated in the patriarchate. But this was not the end of it, for in Jerusalem formal court proceedings were held in the presence of Cardinal Robert. The meeting was held in the church of the Holy Sepulchre, and Arnulf, who was the church’s chancellor and archdeacon, was there. A great deal of planning had gone into it, for bishops and abbots attended from Caesarea and Ramla within the kingdom, but also Tarsus in Cilicia, Mamistra and Edessa in Armenian territory, Mount Tabor in Galilee and six from western Europe, including Laon and Piacenza. In fact, a plausible conjecture is that Baldwin had taken the dissidents away from Jerusalem to engage in a fruitless siege of Ascalon and then to waste time feasting ‘in all glory and happiness’ in Jaffa while Arnulf prepared for what was, effectively, a show trial, complete with witnesses and a comprehensive charge list. Daibert was accused of simony, murder committed on his voyage to the Holy Land, betraying King Baldwin and embezzling money intended for the faithful. The testimony against Daibert was overwhelming, and as he ‘stuck stubbornly to his crooked defence’, he was deposed and excommunicated. Only Tancred, Baldwin of Edessa and the two Williams still supported Daibert, and he returned north with them.28 So far as Albert of Aachen was concerned, this was the end of Daibert, and we have to look elsewhere for his next and final attempt to redeem himself. In 1104 he set out to take his appeal to the pope in Rome. Fulcher of Chartres, who was a cleric at King Baldwin’s court, mentioned Daibert only half a dozen times and never at any length. He wrote of Daibert’s election at the end of 1099 that he was chosen by Godfrey and the other leaders. Later

102  Ecclesiastical affairs he observed, to explain Daibert’s absence from a celebration in November 1100, that Daibert had been deprived of the patriarchate on account of accusations by some of Baldwin’s men, which had caused bad feelings. Within weeks Daibert and Baldwin were reconciled, in time for the latter’s coronation at Christmas 1100. The next time Daibert was mentioned he was on his way to Italy with Bohemond to complain to the pope about an unspecified injury done to him by the king. Another reference to Daibert’s recovery of the patriarchate and his death, and a passing mention of his presence at the battle of Harran, completed Fulcher’s treatment of the scandals surrounding Daibert, in contrast to the several chapters in Albert of Aachen’s Historia.29 Bartolf of Nangis, however, with his acknowledged access to a version of Fulcher’s history written in c. 1106, reflected the situation as it appeared then, after Daibert had been deposed and had left for Rome, but before those in Jerusalem could hear of his reinstatement. In fact, the reinstalled Daibert set out on his return journey to Jerusalem, but in the course of it he died (1105). Bartolf’s account of the dispute and trial therefore makes interesting reading: But [Satan] sent such disagreement between Patriarch Daibert and the Church entrusted to him, and sowed such hatred, that he was loved neither by the clergy, nor by the king, nor by the people; and this great individual, this great pillar of the Church, this devout shepherd and this outstanding teacher – alas – was hated by all as if he were a wolf, and he was evicted from the doors of the church, under the cloak, as it were, of Roman process, with Cardinal Robert presiding, and Arnulf the archdeacon of the Holy Sepulchre striving for the prosecution with others of his accomplices. And so this distinguished man, remarkable for every good quality, being unable to bear their ferocity, appealed to the Roman see, and soon he set out for Rome.30 The following chapter associated Daibert closely with Bohemond, with whom he travelled, and showed the two of them presenting their grievances before the pope together and each receiving a favourable response: Daibert the restoration of his patriarchate and Bohemond the right to raise an army against the Byzantine emperor under the banner of St Peter. According to this account, Daibert accused Cardinal Robert, as well as the clergy and the king of Jerusalem, of injustice against him, and in the following chapter Bartolf claimed that it was rumoured that Robert wanted the patriarchate for himself. This chapter ended with a lament about the ills which then befell the Church in Jerusalem, deprived of its shepherd, teacher and comforter.31 The difficulty with Bartolf is that when he included versions of events or opinions that are not those in the surviving final version of Fulcher’s history he could have been conveying information he found in the early redaction that he was using, or he could have been adding information from other sources.32 In the case of Daibert’s vindication by the pope, it seems most likely that this was known in northern France at the time Bartolf was writing, but not by Fulcher at the date he completed the 1106 redaction. Writing twenty years later, Fulcher had a longer

Ecclesiastical affairs 103 and perhaps more objective view of Daibert’s career in the East, but this led him to censor his previous, incomplete account of Daibert’s reinstatement.

Patriarch Evremar (1102–1108) Daibert was succeeded as patriarch by Evremar, who like Arnulf came from Chocques, near Thérouanne in the borderland between Normandy and Flanders, and had studied like him in the see of Thérouanne. The candidate had also to be approved by Cardinal Robert, so he was a less controversial choice than Arnulf himself. However, both the hostile chronicler Guibert of Nogent and the rather more neutral Bartolf of Nangis believed that it was Arnulf who was pulling the strings. Bartolf wrote that Evremar was reputed to be an uneducated but imposing and religious individual.33 Evremar was sponsored by Arnulf, according to Bartolf, to ensure that Daibert would give up any hope of regaining his see. Bartolf added another snippet of gossip: ‘All of this was done in the presence of the aforesaid Cardinal Robert who, so people say, aspired to this office himself.’ Bartolf also stressed that the king agreed to the appointment of Evremar, with all his nobles. Guibert of Nogent called Evremar ‘simple and illiterate’ and claimed that Arnulf nominated Evremar to replace him because ‘he would in no way detract from his power’.34 From the military point of view, Baldwin was able to rely on the new patriarch for support when the Egyptians invaded in 1105, first by organising prayers and penance in Jerusalem, but then by leading an auxiliary army to Ramla to join the king. He organised the hearing of confessions before the battle, too, and then carried the True Cross relic into battle. Fulcher of Chartres’s portrayal of him as pious, decisive and courageous is credible because Fulcher was an eye-witness and participant, although he himself stayed in Jerusalem rather than setting out for Ramla.35 Although Evremar was portrayed as a trusted ally of Baldwin as well as a devout colleague of Fulcher himself, albeit not by name, the chronicler touched on his patriarchal career only once again, when he left for Rome in 1106 to enquire from the pope whether he was still patriarch now that Daibert was dead.36 Albert of Aachen, as so often, included more detail about Evremar and his relationship with Baldwin and Arnulf. Albert described his election thus: Evremar, a man and cleric of good character, an excellent and cheerful distributor of alms, was appointed patriarch and succeeded to Daibert’s place and position, serving God there in the temple of the Lord’s Sepulchre with all religious devotion and good behaviour in the love of brotherly charity, and acting as faithful assistant to King Baldwin against the Saracens and unbelievers.37 This was borne out at the capture of Acre in 1104 when Evremar was associated with the king in negotiating the surrender terms. Albert’s description of what happened before the battle of Ramla in 1105 tallies very closely with Fulcher’s, even to the number of infantrymen the patriarch led to Ramla, 150. The same close

104  Ecclesiastical affairs co-operation was in place in the summer of the following year (1106) when the patriarch was one of those consulted by the king about the advantageous deployment of a pilgrim fleet that had arrived from the regions around the North Sea, and Evremar was at his side during the campaign.38 Yet in the same year the relationship soured. The reasons for this are disputed. Hamilton conjectured that ‘he lacked those skills in the arts of government and diplomacy which would enable him in moments of crisis to make decisions in the king’s name and to help in the work of administration’. Klaus-Peter Kirstein has suggested it was because Evremar quarrelled with King Baldwin over the elevation of Bethlehem to a bishopric in 1106.39 A simpler and more likely explanation hinges on the date of Daibert’s death. William of Tyre reported that Daibert had been patriarch for four years when he died; he had been in exile for three years. Some historians have added the periods together and placed his death in 1107, but if it is interpreted as ‘He was patriarch for four years, but spent three [of those years] in exile’, then he died in June 1105, three years after his deposition and replacement by Evremar in 1102.40 This meant that Baldwin and Arnulf no longer needed Evremar as a front man since there was no longer a danger that Daibert would be reinstated, so they could angle for a patriarch more to their taste. Albert of Aachen’s version of what happened next is that Evremar went to Rome ‘in order to absolve himself of all accusation and blame levelled at him by the king and Arnulf the chancellor’ and silenced Arnulf’s arguments in the presence of the pope, after which he was given permission to ‘occupy the patriarchal see from now on honourably and without impediment’ and sent back to Jerusalem with a papal letter to that effect. However, Baldwin – prompted by Arnulf – refused to accept the letter and forced Evremar to resign the patriarchal office ‘without councils and trial’. Albert concluded: ‘Although it was wrong that this dispute should take place, unless one of them had been condemned by decree and sentence of canon law, yet because the Jerusalem Church was still unformed and fragile the pope allowed it to happen.’ The episode is very revealing of the way that Baldwin and Arnulf operated, being coolly aware how far they could push the papacy.41 The letter sent by Pope Paschal to the king, dated at the Lateran on 4 December 1107, has survived.42 In it the pope summarised the brief history of the patriarchate of Jerusalem. He wrote of Cardinal Robert, who had deposed Daibert at the synod in Jerusalem in 1102 and supported the election of Evremar to the patriarchate. The pope had restored Daibert to the patriarchate, and Evremar was then promised the next vacant bishopric. Daibert was now dead, and a dispute had arisen between Evremar, who claimed to have been legitimately elected to the patriarchate, and Arnulf, the archdeacon of the church of Jerusalem. Paschal was therefore sending another legate, Gibelin of Arles, to settle the case. William of Tyre claimed that when the case was heard at a council of bishops of the kingdom, reliable witnesses attested that Daibert had been driven out ‘by the deeds of Arnulf and royal violence’. Evremar, therefore, did not lawfully occupy the patriarchate, but because of his ‘great piety and wonderful simplicity’ he was granted the see of Caesarea, which he occupied unproblematically until 1129. Amid all

Ecclesiastical affairs  105 the vituperations of the chroniclers, the true causes for the falling-out between Baldwin and Arnulf on the one hand and Evremar on the other are still unclear. The letter that Arnulf took to Rome on behalf of the king and the bishops asked the pope to remove Evremar entirely from the Jerusalem church, calling him ‘as good as useless’, but what they meant by this is open to interpretation.43 There was widespread agreement that Evremar was a simple and pious man. He co-operated with the king against the infidel while patriarch and governed his archbishopric effectively after he was deposed. One can only conjecture that he became caught up in the dispute over the patriarchate, perhaps being too keen humbly and contritely to stand aside when he heard of Daibert’s reinstatement, and was therefore seen as disloyal by the king and Arnulf.

Patriarch Gibelin (1108–1112) The election of the papal legate Gibelin of Arles as patriarch was now carried through unanimously. Even Arnulf acquiesced  – William of Tyre, who was incapable of thinking well of him, said that Arnulf maliciously brought about Gibelin’s appointment because he was a decrepit old man who would not occupy the patriarchate for long. Nevertheless, with the support of the king and the continued confidence of the pope, who allowed him to retain his full powers as legate, Gibelin was able to set about imposing some order on the Church in Jerusalem. He proved able to work with his two potential rivals, Evremar and Arnulf. Evremar’s appointment to Caesarea, the only other metropolitan see, in 1108, was a success. Gibelin was also probably responsible for the appointment of Bishop Roger to the see of Lydda-Ramla, succeeding Robert of Rouen, the First Crusader who had held it since 1099. He confirmed and oversaw the conversion of Bethlehem to a bishopric, related previously, and was responsible for bringing Galilee under the patriarch’s ecclesiastical control by installing a Latin bishop in Nazareth. He also attempted to create Latin bishoprics in the coastal cities conquered by Baldwin, but this was more difficult since they had formerly been under the authority of the patriarchs of Antioch. Despite a papal ruling, dated at the Lateran on 8 or 9 June 1111, that all the cities captured by King Baldwin were to be subject to the patriarchate of Jerusalem, the matter was not fully resolved at Gibelin’s death in 1112, although his was the solution that would be adopted.44 Meanwhile, Gibelin did not neglect his secular duties as counsellor and ally of the king. Albert of Aachen included him in the long roll of those present on campaign to Antioch in 1110, probably as keeper of the True Cross that was the army’s standard.45 However, Arnulf, who remained in the king’s confidence, may well have acted vicariously for him in secular aspects of the role during his relatively short tenure as patriarch. Shortly before Gibelin died during Lent in 1112, he petitioned the king in a document in favour of the church of the Holy Sepulchre, asking him also to ensure that his last testament was carried out and confirmed, and he referred to Arnulf as his ‘dearest friend and son’.46 Arnulf’s election to replace Gibelin as patriarch was uncontested.

106  Ecclesiastical affairs

Patriarch Arnulf (1112–1118) Like King Baldwin, Arnulf appears from documentary evidence to have been genuinely and conventionally pious. On the day of his consecration he confirmed to the brothers of the hospital of St John all their possessions within the patriarchate and specifically the tithes on their lands that had formerly been his. In the same year he granted revenue from tithes to the abbey of St Mary in Jehosaphat to rebuild their church. In 1114 he reformed the secular canons of the Holy Sepulchre: they were to observe the precept of the Augustinians and were granted sufficient benefactions to do so.47 These last arrangements were in accordance with Gibelin’s last testament but also reflected Arnulf’s having been archdeacon and chancellor of the Holy Sepulchre for the previous nine years. As secular support for King Baldwin, Arnulf had already proved his worth for the length of his reign. There was a severe test of diplomacy immediately after his installation as patriarch when he was overseeing the Holy Week ceremonies in 1112. Emperor Alexios had sent envoys to solicit help in ousting Tancred and restoring Byzantine rule in Antioch. King Baldwin was bound to refuse him, but Arnulf told him to do so on Easter Day while wearing his crown in regal solemnity. Baldwin ‘graciously commended’ the legates, bestowed gifts on them and sent them back to Constantinople.48 Anna Komnene reported the same legation rather differently: it was led by Manuel Boutoumites, a Latin-speaker who had negotiated the surrender of Nicaea in 1097 and so was familiar to the Franks. The Byzantines found Baldwin engaged in the unsuccessful siege of Tyre during Lent 1112. When it was lifted they accompanied the Latins first to Acre and then to Jerusalem for Holy Week. According to Anna they parted on ill terms: in Acre Baldwin had accused Boutoumites (accurately, as Anna conceded) of lying about the emperor’s whereabouts; in Jerusalem Baldwin demanded the money the emperor had sent to secure his alliance. Boutoumites withheld the money, and when he left, far from receiving splendid gifts, he carried ‘non-committal letters’.49 Arnulf made another foray into diplomatic affairs in the matter of Baldwin’s marriage to Adelaide of Sicily, notwithstanding Baldwin’s previous marriage to an Armenian wife, by then living in Constantinople. According to William of Tyre, this was another of Arnulf’s schemes, although there is no support for this in other sources.50 Adelaide was the widow of Roger I of Sicily, who had died in 1101. She was then regent for her son Roger II, but when he took over the government himself at the age of seventeen in 1112 she was free to remarry with her son’s approval. The attraction of marriage to Adelaide for Baldwin was her immense wealth, which was sorely needed for the defence of the kingdom. In particular, an alliance with Sicily would give the king access to the Sicilian fleet and free him from the need to grant expensive privileges to the Genoese, in particular, when they blockaded the maritime cities in his support. The attraction of the alliance for Roger was that if no

Ecclesiastical affairs  107 heir was forthcoming from the marriage then he would inherit the kingdom of Jerusalem. This was a likely outcome that alarmed the nobility of Jerusalem, although they acquiesced. Albert of Aachen included an entirely apocryphal description of the fleet’s arrival: She had two trireme dromonds, each with five hundred men very experienced in warfare, with seven ships laden with gold, silver, purple and an abundance of jewels and precious garments, besides weapons, hauberks, helmets, shields resplendent with gold, and besides all the other weaponry which powerful men are accustomed to carry for the defence of their ships. The king’s welcome to her when she landed in Acre was equally magnificent, and it was followed within days by their marriage in ‘enduring wedlock’.51 Arnulf’s involvement in the ceremony, which he knew to be bigamous, exposed him to criticism. This was only one of the serious charges against him recorded by William of Tyre. His old nickname ‘Mala Corona’ (‘Bad Crown’) was raked up, as well as his irregular birth. In addition, he had arranged for his niece to marry the wealthy and powerful lord Eustace Granarius, giving her Jericho, which formerly belonged to the Church, as her dowry. He was also accused of having mistresses. His detractors, led by Evremar of Caesarea, took their case to the pope, claiming that Arnulf had been forced on them by the king. Paschal II responded by – yet again – sending a legate to investigate the charges in 1115. Arnulf was found guilty so – yet again – he took his appeal to Rome, accompanied by senior clerics who swore that Arnulf had been freely elected and demanding his reinstatement. Arnulf was allowed to clear himself on oath and was granted a dispensation for his irregular birth (and life), and so in 1116 he returned to Jerusalem as patriarch. As ever, William accused Arnulf of using blandishments and bribery to achieve his acquittal. But, as is clear from Albert of Aachen, the one transgression the pope was unable to overlook was condoning the king’s bigamous marriage. The king, who was seriously ill by this time, acquiesced, and Arnulf presided over the formal annulment of the Sicilian marriage at Easter 1117, incurring the enmity of Sicily to the great detriment of the kingdom, as William of Tyre and modern commentators have agreed. Bernard Hamilton observed: ‘On this occasion Arnulf sacrificed the interests of the kingdom to save his own office.’52 A year later both king and patriarch were dead, Arnulf having survived long enough to receive the king’s corpse on the Mount of Olives during the Palm Sunday procession in 1118. Arnulf had been a constant presence in King Baldwin’s Jerusalem, on his side and at his side, so far as can be established from the sources.53 William of Tyre’s highly biased account depicted him as malicious and manipulative and the king as easily influenced, but in view of the efficiency Baldwin evinced in every other aspect of his rule, it is easier to believe that Baldwin was aiming to adjust to his advantage the balance of power between king and patriarch and that it was he who was using Arnulf as a convenient stalking horse and occasional scapegoat.

108  Ecclesiastical affairs

Notes 1 Digests of Baldwin’s grants to religious foundations are to be found at the Revised Regesta Regni Hierosolymitani Database [RRR], http://crusades-regesta.com [accessed 23 August  2017], at nos. 68, 70, 80, 82, 90, 93, 105‑106, 108‑10, 113, 114, 129, 145, 147, 148, 151, 155, 158, 161, 168‑69, 171‑73, 175‑78, 181‑82. Where any of the regesta are discussed in any detail, references to document collections are supplied below. For Bethlehem: Die Urkunden der lateinischen Könige von Jerusalem [UKJ], ed. Hans Eberhard Mayer, 4 vols (Hanover, 2010), I: 159–64, no. 40 [RRR, 110]. 2 See Bernard Hamilton, The Latin Church in the Crusader States: The Secular Church (London, 1980), pp. 1–12 for Patriarch Symeon and the crusade. For Symeon’s letter to the pope, carried by Peter the Hermit: Ernest O. Blake and Colin Morris, ‘A Hermit Goes to War: Peter and the Origins of the First Crusade’, Studies in Church History, 22 (1985), 79–107. The two letters sent under Symeon’s name are in Hagenmeyer, Epistulæ, no. VI, dated October 1097, pp. 141–42; no. IX, dated January 1098, pp. 146–49. 3 The provisions sent to the siege of Jerusalem included pomegranates, fat bacons and excellent wine: AA 6.39, p. 452. Pahlitzsch claimed that ‘the first version of the chronicle of Fulcher of Chartres’ had Symeon stay on pending the pope’s decision about the future of the patriarchate’, but I have not been able to locate a reference for this: Johannes Pahlitzsch, ‘Symeon II of Jerusalem’, in Crusades Encyclopedia, IV: 1135–36. Symeon may have died in exile on Cyprus rather later, at some time before 1106/1107 when a Greek Orthodox patriarch notionally succeeded him. 4 The appointment of a Latin patriarch was not a foregone conclusion. When the crusaders captured Antioch in 1098, the Orthodox patriarch John V of Oxeia remained in post. However, in 1100 the Latins suspected John of plotting with the Byzantines and forced his abdication († 1118). He was replaced by the Latin Bernard of Valence (1100–1135). For the whole question see Hamilton, The Latin Church in the Crusader States, pp. 1–12, 16–17. 5 The crusaders were, of course, unaware that Urban II would die only fifteen days after the capture of Jerusalem (29 July 1099). His successor was Pope Paschal II († 1118). Arnulf’s appointment is in AA 6.39, pp. 452–54. 6 BN chap. 37, p. 516 (quasi patriarcham); cf. FC 1.30.2, p. 308. 7 GN 7.15, p. 291 (dum vox magis quam vita curatur). 8 WT 9.15, p. 440. William’s wording was ‘proprium non habens antistitem’, and while ‘permanent’ is a legitimate translation of ‘proprius’, it could also be translated as ‘its own’ or even ‘a proper’ bishop. 9 BN chap. 40, p. 519. 10 A balanced modern biography is Michael Matzke, Daibert von Pisa (ref. earlier); see also the same author’s ‘Daibert of Pisa (d. 1105)’, in Crusades Encyclopedia, II: 339– 40. The leaders’ letter to the pope is in Hagenmeyer, Epistulae, no. XVI, pp. 161–67; Adhémar’s death at p. 164; see also Luis García Guijarro, ‘Some Considerations on the Crusaders’ Letter to Urban II (September 1098)’, in ed. Edgington and García Guijarro, pp. 151–71. 11 The Pisan fleet’s attack on some Byzantine islands in the Mediterranean was reported from the Pisan side is Bernardo Maragone, ‘Vetus chronicon Pisanum’, RIS NS, 6.2 (1936), 3–71; for the Byzantine point of view see AK 11.10, pp. 323–24. Anna dated the attacks to 1104, but they were obviously the same as those recorded in Pisa. The siege of Latakia is in AA 6.55, pp. 476–78; 6.57, pp. 480–82. AA wrote that the Genoese were also involved in the siege but was probably mistaken as their annalist Caffaro did not mention it. Pisan involvement is confirmed by the ‘Gesta Triumphalia Pisanorum in captione Jerusalem’, RHC Occ, V: 368–69. The crusaders returning home: AA 6.56, pp. 478–80; 6.60, p. 484. 12 AA 7.6–7, pp. 494–96.

Ecclesiastical affairs  109 13 WT 9.15, p.  440, hic regni, ille principatus . . . susceperunt investituram. Historians have been unduly influenced by William’s chronicle and his attitude to the clerical rivals. An early and influential example was Heinrich von Sybel, who devoted a large section of his ground-breaking history of the First Crusade to a critique of Albert of Aachen’s Historia, in the course of which he suggested that William of Tyre abandoned his use of the work as a source for events after 1099 because he rejected Albert’s account of Arnulf and Daibert: H. von Sybel, Geschichte des ersten Kreuzzugs (Leipzig, 1841); trans. by Lady Duff Gordon as The History and Literature of the Crusades (London, 1861), p. 308; S. B. Edgington, ‘The Historia Iherosolimitana of Albert of Aachen: A critical edition’ (Unpublished Ph.D. thesis, University of London, 1991), pp. 24–25, for arguments against Sybel’s theory. See Edbury and Rowe, William of Tyre, pp. 49–50 and 99–105, for an extended discussion of Daibert’s motives and actions and William’s attitudes to Daibert and Arnulf. 14 WT 9.15, p. 440, hic regni, ille principatus . . . susceperunt investituram. William’s translators, who regarded the statement as unique to him, suggested that it might represent William’s ‘conception of the ideal relationship between patriarch and king, church and state, in Jerusalem’: WT, trans. Babcock and Krey, A History of Deeds, p. 403, n. 38. 15 FC 3.34.16, pp. 741–42. 16 Hamilton, The Latin Church in the Crusader States, pp. 53–54, discusses the ceremony and dates it to Candlemas Day 1100. See also Murray, Crusader Kingdom, pp. 84–85. WT claimed Baldwin was present: 10.4, p. 457. 17 WT 9.16, pp. 441–42. WT used the verb ‘reposcere’. This means either ‘to demand back’ or ‘to demand as one’s due’. If the second of these, Daibert may not have been intent on obtaining the city for himself, because WT qualified the Holy City as ‘assigned to God’ (deo ascriptam), a point not always made clear by translators. Hamilton, The Latin Church in the Crusader States, pp. 101–103. 18 WT 9.16–18, pp. 441–45; Edbury and Rowe, William of Tyre, p. 104. 19 Runciman, A History of the Crusades, II: 310; Rubenstein linked the fading of the vision of theocracy with the failure of the ‘quest for the Apocalypse’: Armies of Heaven, pp. 293–303. 20 The hiding place of the fragment of the Cross was revealed by a Christian inhabitant of Jerusalem, and then it was kept in the church of the Holy Sepulchre: AA 6.38, pp. 450– 52; FC 1.30.4, pp. 309–10. RA, even though he disapproved strongly of Arnulf, wrote that after he was elected (provisional) patriarch in 1099 he instigated a search for the Cross, which was buried in ‘the narthex of a certain church’: p. 154. 21 AA 7.46–48, pp. 554–56. 22 The appointment of Maurice is recorded in Papsturkunden für Kirchen im Heilingen Lande, ed. Rudolf Hiestand (Göttingen, 1985), pp. 90–92, no. 4 [RRR no 27]; the pope gave thanks for the successful pilgrimage initiated by Pope Urban II and commended his legate, the bishop of Porto. 23 AA 7.49–51, pp. 556–60. 24 AA 7.58–59, pp. 568–70. 25 AA 7.60–61, pp. 570–72. Compare with WT 10.7, p. 462: Albert presented Baldwin as credulous because he trusted Daibert; William of Tyre wrote that he was gullible because he allowed himself to be deceived by Arnulf. 26 AA 7.62–64, pp. 572–74. 27 AA 9.14–17, pp. 652–58. For Cardinal Robert, Papsturkunden, ed. Hiestand, pp. 104– 107, no. 8 [RRH no. 49]. 28 AA 9.16–17, pp. 656–58. AA’s account of the proceedings should be compared with that of WT, who blamed Arnulf for misleading the king and stirring up discord; he mentioned neither the charges brought against Daibert nor his deposition (10.25(26), pp.  484–85). The accusation of murdering Byzantine subjects was echoed by Anna Komnene, who held Daibert responsible for the ravaging of Corfu, Leucas, Cephalonia and Zacynthos by the Pisan fleet AK, 11.10, pp. 323–24.

110  Ecclesiastical affairs 29 Election: FC 1. 33.20, pp. 333–34; absence November 1100: FC 2.3.14, pp. 368–69; Baldwin’s coronation: FC 2 5.12, pp. 383–84; on his way to Italy: FC 2 26.2, p. 467; recovery of patriarchate and death: FC 2.37.1, pp. 470–71; presence at battle of Harran 2.27.2, pp. 512–14. 30 BN chap. 64, pp. 537–38. 31 BN chaps. 64–66, pp. 537–38. 32 BN chap. 2, p. 492. 33 BN chap. 66, p. 538. Bartolf used the Latin term ‘idiota’ that is translated as ‘uneducated’. The Greek word had none of its modern connotations of stupidity or silliness but referred to an ordinary citizen as opposed to one who was properly engaged in civic government. 34 GN, p. 292, cf. See also Hamilton, The Latin Church in the Crusader States, pp. 56–57. 35 Evremar at the third battle of Ramla: FC 2.31, pp. 489–94; FC 2.32.1–2, p. 495; BN chap. 70, p. 540. 36 Events of 1106: FC 2.37.1, pp. 512–14. The date has been disputed among historians, some of whom have mistranslated Fulcher’s ‘Anno MoCoVIIo’ as 1107 rather than ‘In the 1107th year’, i.e. 1106. 37 AA 9.17, p. 658. 38 Evremar at Acre: AA 9.28, pp. 672–74; at Ramla: AA 9.49, pp. 706–708; 1106 campaign: AA 10.1–4, pp. 718–22; 10.17, p. 734. 39 Hamilton, The Latin Church in the Crusader States, p. 57; cf. Klaus-Peter Kirstein, ‘Evremar of Chocques (d. 1128/1129)’, in Crusades Encyclopedia, II: 417–18. 40 WT 11.4, p. 500. See also Matzke, ‘Daibert of Pisa’, p. 340. 41 AA 10.58–59, p. 772. 42 Papsturkunden, ed. Hiestand, pp. 104–107, no. 8 [RRR no. 92] (the letter includes the description of Evremar as tamquam inutilem). 43 WT 11.4, p. 501. 44 WT 11.4, p. 501. The privilegium was included by FC 3.35, pp. 742–45; WT 11.28, p. 538. See Hamilton, The Latin Church in the Crusader States, pp. 58–61 for a fuller account of Patriarch Gibelin’s achievements. 45 AA 11.40–42, pp. 816–18. 46 Gibelin’s last testament: Le Cartulaire du chapitre du Saint-Sépulcre de Jérusalem, ed. Geneviève Bresc-Bautier (Paris, 1984), pp. 85–86 [RRR 123]. 47 UKJ, I: 176–77, no. 51 [RRR 126]; UKJ, I: 179–80, no. 53 [RRR 132]; UKJ, I: 181– 83, no. 55 [RRR 146]. 48 AA 12.7–8, p. 834. 49 AK 14.2, pp. 400–406. 50 Arnulf’s culpability: WT 11.21, pp. 525–37. 51 See the concluding chapter for fuller discussion of Baldwin’s marital history. Baldwin was separated from his Armenian wife in 1108, but the marriage could not be formally dissolved. He placed her in the convent of St Anne: GN 7, p. 349. NB Baldwin’s grant to the nunnery was probably related: WT 11.1, pp. 495–96; UKJ, I: 156, no. 36 [RRR 105]. Adelaide’s arrival and the wedding: AA 12.13–14, pp. 842–46. Like all of Baldwin’s marriages, this third is dealt with at length by Mayer, ‘Études’, pp. 59–69. 52 WT 11.15, p. 519; 11.26, pp. 534–35; 11.29, pp. 542–43; Papsturkunden, ed. Hiestand, pp. 124–26, no. 19 [RRR no. 162]; AA 12.24, pp. 860–62; Hamilton, The Latin Church in the Crusader States, p. 64. Letters about the dispute are in Cartulaire de l’église du Saint-Sépulcre de Jérusalem, ed. Eugène de Rozière (Paris, 1849); to the Jerusalem Church, no. 10, pp. 8–11; to the clergy of the Jerusalem Church, no. 11, pp. 11–13; to King Baldwin, no. 12, pp. 13–14. 53 For example, he accompanied the king on campaign in 1113: FC 2.49.5, p. 569.

7 The conquest of the littoral

The conquest of the littoralThe conquest of the littoral

Introduction Baldwin’s reign, 1100–1118, was dominated by the exigencies of foreign policy. Such domestic measures as were implemented during the period were almost invariably dictated by the need to respond to external threats by fielding an effective army.1 How this affected Baldwin’s relations with the patriarchs of Jerusalem has already been seen in the previous chapter. Although King Baldwin’s interaction with his Muslim enemies has been divided into two chapters that deal with a policy of expansion in the first two thirds of his reign, while a third treats a more reactive phase in the last part, the reality was inevitably much more complex than this suggests.2 Therefore this preliminary section will set out briefly the main threats to the kingdom and some of the ways these could be countered or diminished. As formerly count of Edessa, Baldwin was already familiar with the political landscape of northern Syria. The main Turkish power base there was Mosul, whose governor, Karbughā, had mustered a coalition of forces to fight the crusaders at Antioch in 1098. After Karbughā’s death in 1102, there was a period of instability until Karbughā’s nephew Mawdūd took power in 1108. In the five years that followed, Mawdūd led three major campaigns against the Franks.3 Edessa and Antioch bore the brunt of Mawdūd’s attacks, but the kingdom of Jerusalem was inevitably drawn into the common defence of the Christian states in these years. Mawdūd’s threat to the Latin states was ended when he was assassinated at the behest of Ṭughtigin of Damascus in 1113.4 In short, there was ever-present danger from the Turks in the north, but for much of the time King Baldwin could leave the defence of Edessa and Antioch to their rulers and regents, responding only when they called upon him to do so. Baldwin’s relocation to Jerusalem brought him within the ambit of Damascus, the major city of southern Syria. Fortunately, Ṭughtigin, the effective governor of Damascus and its region throughout Baldwin’s reign, was a pragmatist who maintained his authority by way of balance-of-power politics.5 There were nevertheless potential flash points with the kingdom of Jerusalem. One was access to the Mediterranean Sea: Damascus lost Haifa to the Franks in 1100 and after that was more committed than ever to keeping open the trade route between Damascus and Tyre

112  The conquest of the littoral that was threatened by the Frankish lords of Tiberias and their expansion into the Golan Heights. Frankish designs on Tyre were frustrated in 1112 when Ṭughtigin co-operated with the Fāṭimids to relieve the city, under siege by Baldwin, thus securing the trade route until 1124 when Tyre finally fell to the Franks. Another likely area of dispute was control of the Hauran, a fertile area on the border with the kingdom of Jerusalem that was Syria’s breadbasket. Attacks launched from Tiberias were an irritant that occasionally turned into armed conflict, and in 1108 Ṭughtigin and Baldwin, both pragmatists, agreed a truce and a condominium that gave each of them one third of the region’s crops, the remaining third going to the inhabitants who produced them.6 So far as Damascus was concerned, therefore, King Baldwin was saved from any concerted attack partly by Ṭughtigin’s persistent rivalry with the Turkish powers in Mosul and Aleppo and with the Fāṭimids of Egypt that made any idea of long-term co-operation against the Franks inconceivable, but even more surely by the king’s and Ṭughtigin’s readiness to contemplate treaty agreements that ensured that each had his hands free to meet the threats of other enemies. For Baldwin in 1101, the most immediate threat was not from the Turks but from the Fāṭimids of Egypt. From 1095 to 1121 the de facto ruler of Egypt was the vizier al-Afḍal. His Shi‘ite government was implacably opposed to, and by, the Saljūq (Sunni) Turks, and in 1098 the Fāṭimid army, led by al-Afḍal, had seized Jerusalem from them. The city’s capture by the crusaders a year later was followed almost immediately by the arrival of a relieving force from Egypt, defeated on the plains near Ascalon, and a series of attacks through Ascalon in the years that followed.7 Nevertheless, as Michael Köhler has argued, it appears that the Fāṭimids preferred the Franks to the Turks as neighbours and were not averse to the Christian kingdom of Jerusalem surviving as a buffer state between them and their Sunni co-religionists.8 The sticking point, which finally induced a degree of co-operation between Damascus and Cairo, was Baldwin’s determination to conquer the sea-ports of Syria, and this is the subject of this chapter.

The situation at Baldwin’s accession When Jerusalem was captured in July 1099 the crusaders had achieved their goal, but there is no evidence that they had a plan for the future beyond the liberation of the Holy City. Thus, once they had fulfilled their pilgrim vows, many crusaders, including some of the leaders, were ready to set out for home. They participated in the election of a lay ruler, Godfrey, and a provisional patriarch, Arnulf of Chocques, and they fought the battle of Ascalon on 12 August to defeat the Fāṭimid army that was attempting to recover Jerusalem and the territory the Egyptians had lost a month before. Then, quite justifiably, the crusaders departed for the lands they had left more than three years before. Thus Godfrey’s realm consisted of little more than three isolated territories: Jerusalem and Bethlehem inland and Jaffa, the nearest port, which had been useful during the siege of Jerusalem. A small fleet including Genoese vessels had put in at Jaffa in June 1099, but they found that the town was indefensible because its fortifications were in

The conquest of the littoral 113 ruins. This led them to destroy their own ships, first to avoid their capture by the enemy and second to provide much-needed timber for the siege of Jerusalem.9 This experience illustrated for the future rulers of Jerusalem the importance of access to the coast, and it underlined the importance and urgency of securing usable ports along the littoral of Palestine. There were two reasons for this: first, ports were the key to bringing in muchneeded men and supplies from western Europe; second, so long as they were in the hands of the enemy they could be used as bridgeheads for an invasion by the Saracens of Egypt. The authorities were very aware of the manpower crisis. In December 1099 the new pope, Paschal II, had written to clerics in western Europe giving thanks for the deliverance of Jerusalem and exhorting them to urge any who had taken the cross and had not gone on the expedition, and more especially any deserters from the siege of Antioch, to go to the assistance of their brothers now.10 In April 1100 the recently elected patriarch Daibert wrote to all the Christians ‘of the German region’ and called on them in this time of crisis, explaining that even those who remained and were prepared to fight were doing it for wages: The rest, whom we were only just able to keep with us, we employ for great wages and gratuities so that they will defend Jerusalem . . . and other fortifications too . . . until God sends us assistance from your people and the Latin tongue, for without You [God] and the help of other good men we cannot supply in full the wages that we have promised to pay.11 The letter signalled the beginning of the serious fiscal difficulties that were to trouble Baldwin’s government for all his reign. Fulcher of Chartres famously numbered the forces available to Baldwin in 1101 at 300 cavalry and 300 infantry to defend all his territories, which Fulcher listed as Jerusalem, Jaffa, Ramla and Haifa. If these forces were brought together to fight an engagement, some or all of these places had to be left unprotected. The difficulty was compounded, he added, because pilgrims arrived by sea at Jaffa, running the risk of Saracen attacks; they worshipped in the holy places and then many of them returned home. Furthermore, ‘those who came across the sea to Jerusalem could not bring horses with them, while no one came to our aid by land’. Fulcher may have been exaggerating, for one of his purposes in this chapter was to enhance Baldwin I’s reputation: he claimed that the enemy was deterred from attacking by the king’s fame as a warrior.12 If by this he intended to flatter Baldwin, the chapter is likely to have been written during Baldwin’s lifetime, so before 1118. The other major source for these early years, Albert of Aachen, did not deal directly with numbers, but as has already been seen, he mentioned several times Baldwin’s chronic need for money to pay his soldiers that fuelled his dispute with Daibert. So the new kingdom needed both men and money from the West. As Fulcher pointed out, the land route taken by Godfrey and Baldwin and their men was now virtually closed; this was partly because it ran through Constantinople, and the Byzantines, who had treated the First Crusaders as allies when they arrived

114  The conquest of the littoral in 1096, had been alienated by Bohemond’s seizure of Antioch in contravention of the crusaders’ agreement with Alexios. Also, the tribulations of the journey through Anatolia, including Turkish attacks, persisted, as the 1101 expeditions were to discover. The alternative was to bring in reinforcements and supplies by sea. Fleets had arrived throughout the First Crusade, bringing much-needed men and provisions to the port of St Symeon during the siege of Antioch and later to Jaffa during the siege of Jerusalem. The crusaders had marched down the coast road from Arqa to Jerusalem in 1099, but rather than engaging the coastal cities they bypassed them, successfully negotiating truces and even the payment of tribute from their quasi-independent emirs. However, this could only be a shortterm expedient, as in the longer term, even though the crusaders had defeated the Saracens at Ascalon, the emirs would not risk making an enemy of Egypt. At the beginning of Godfrey’s rule and counting from north to south, the cities in question were Beirut, Sidon, Tyre, Acre, Haifa, Caesarea, Arsuf, Jaffa (already taken) and Ascalon.13 It would not be enough to capture one or two of them, as the others could still harbour hostile shipping and prevent the free passage of Christian ships. Furthermore, these ports could act as bridgeheads for the Egyptian navy to land an army which could threaten Jerusalem. This had happened in August 1099, culminating in the battle of Ascalon, and was to occur repeatedly in the first years of Baldwin’s reign. Ascalon, being the southernmost port, represented the greatest threat. Godfrey was aware of the opportunities and threats offered by the ports, but his death after only one year left much unfinished business. As we have already seen, Haifa was under siege in July 1100, and Godfrey promised the city to Geldemar Carpenel on his deathbed. Tancred, having prosecuted the siege, seized Haifa to extend his proto-principality of Galilee to the coast, but Baldwin forced him to surrender it to Geldemar when Tancred was summoned to Antioch to assume the regency during Bohemond’s captivity. Immediately after the Christian victory at Ascalon, Raymond of Saint-Gilles had besieged Arsuf. He was seen off by Godfrey, but Godfrey was no more able than Raymond to capture the town. There was a siege, exceptionally violent on both sides, but as winter approached and the weather turned cold Godfrey withdrew his main army, initially leaving a small force to harass the townsmen but withdrawing this too before Christmas 1099. The weather was the reason Albert gave for lifting the siege, but it is not difficult to deduce that another and more enduring difficulty was closing the blockade without naval support.14 This would explain what Bartolf reported, that Godfrey supported Daibert’s election because he needed the Pisan fleet. This was the situation inherited by Baldwin in 1100: a policy of maintaining truces with the coastal cities until they could be picked off one by one with the vital assistance of an expensively purchased Italian navy.

The capture of Arsuf, 1101 Baldwin obviously understood and was prepared to prosecute the broad policy, but he needed to modify it to fit his own priorities. At Easter in 1101, the beginning of

The conquest of the littoral  115 the campaigning season, the king held a council as a result of which ‘he decided to wait for the fleet of the Genoese and Pisans to arrive, who were over-wintering in the port of Latakia’.15 In order to neutralise the other coastal settlements while he focused his efforts on Arsuf, Baldwin concluded truces with them; this had the additional advantage of bringing in tribute from them. Thus he balanced the need for tribute and truces within a strategic plan, as Albert of Aachen explained: The king, as one who had newly arrived and was lacking much treasure for paying his soldiers, agreed to accept everything offered to him by the pagan cities of Ascalon, Caesarea, Acre and Tyre, but he refused Arsuf and its gifts.16 Clearly Baldwin was preparing to resume the siege of Arsuf that Godfrey had lifted at the end of 1099. For this Baldwin needed a fleet. After Easter King Baldwin went to Jaffa, where he made an agreement with the Italians: As long as they were willing to stay in the Holy Land for the love of God, if by God’s permission and assistance in that time they could capture any of the Saracens’ cities with the king, then they would have in common a third part of the money taken from the enemy within, [providing it were] done without any damage by the Genoese, while the king had the first and second parts; in addition they would possess by perpetual and hereditary right one neighbourhood in the city thus captured.17 This rather awkward translation of the agreement as reported by Fulcher of Chartres is included in full because in the past the agreement has been translated and interpreted as referring to the Genoese alone. This would certainly make sense in the context of Baldwin’s dispute with Daibert, who had arrived with the Pisan fleet and had been able to take advantage of Godfrey’s need for naval support to obtain the patriarchate, and it is probable that the Pisans were not operating as an official navy, but rather, as Michael Matzke has claimed, that individual ship owners and captains continued to support the king’s campaigns after the main Pisan fleet had sailed for Italy in the spring of 1100.18 This would explain why even Fulcher of Chartres, who was keen to suppress any acknowledgement of Pisan involvement, later wrote of a pact with a fleet of ‘Genoese and Italian’ vessels. It is likely that he had earlier recognised that there were Pisan vessels, for Bartolf of Nangis’s account of the agreement ran: And so when the celebration was over, the king set out for Jaffa with all his army; and calling together the senior Genoese and Pisan naval officers, he made an agreement with them that whatever cities or strongholds he might subdue through their help, he granted to them a third part of the gains, whether in plunder, or prisoners, or spoils. And when this had been confirmed on both sides by a pledge of good faith and an oath, they surrounded the nearest town, which they call Arsuf, from all sides by land and sea.19

116  The conquest of the littoral Albert of Aachen did not record any formal pact, but stated clearly that both the Pisans and the Genoese contributed fleets to the siege.20 Fulcher’s later, deliberate omission of Pisan involvement will be revisited. The sources agree that Baldwin’s closing the blockade by land and sea and launching a fierce attack straightaway were effective in bringing the inhabitants to surrender after only three days. The terms included a safe conduct for the inhabitants, who left for Ascalon with their women and children. Fulcher added that they were allowed to take all their money with them. After this the king entered the city peacefully and spent a week there, then garrisoned the city before he left. Thus by utilising the Italian ships at Arsuf to complete the blockade Baldwin had achieved the optimum outcome of a siege, which was to intimidate the enemy into a speedy surrender. This enabled him to take over Arsuf with its fortifications intact, which meant it was immediately defensible with the least number of soldiers as garrison. Furthermore, if a town was taken by assault, quite apart from any damage done to the walls, the attackers expected to kill and plunder. If Fulcher’s record of the agreement with the Italian captains was correct, the king had foreseen this danger and specifically excluded granting trading quarters to them if they violated a surrender agreement and damaged the captured city. On this occasion he was successful in preventing a sack.

The siege and capture of Caesarea, 1101 Albert of Aachen ended his account of the capture of Arsuf by writing that during his week’s rest there Baldwin consulted his nobles about the remaining towns. They agreed that Caesarea should be next, and as an opening gambit Baldwin sent a legation to the city in the hope that the Christians’ success at Arsuf would persuade the governor to surrender similarly, but he also threatened to take Caesarea by force and kill everyone in it. The citizens replied that they had already heard from the Egyptians that they were on their way, so Baldwin, presumably in the hope of capturing the city before the relieving force arrived, moved his forces to Caesarea. They found that the city was protected by a very strong wall; the besiegers lacked the timber to build the siege machines necessary either to reduce the walls or, preferably, to intimidate the enemy into surrendering as had happened at Arsuf.21 Fulcher of Chartres was present at the siege and described the siege tower built with the attackers’ limited resources: ‘So then the king ordered stone-hurlers to be made and a single, very high wooden engine constructed from the masts and oars of ships. This, as I calculate, our carpenters built higher than the wall by the length of twenty cubits.’22 However, as he went on to say, after a fortnight the Franks became impatient and stormed the city with lance and shield, using ladders they had previously made ready and not the tower, which was still unfinished. They killed nearly all the men and enslaved the women. The king kept alive for ransom the emir along with ‘the bishop who was entitled Archadius’ (that is, he was al qadi, a judge in Islamic law). Then Fulcher underlined his personal experience of the siege and sack: ‘I saw very many of the Saracens killed there piled up and

The conquest of the littoral  117 burned to ashes, the stink of whose corpses was a very great annoyance to us.’ In a well-known passage he explained that the reason for this was to extract the gold coins that men and women had variously hidden in their bodies. Bartolf condensed Fulcher’s narrative considerably and offered only one additional detail, which Fulcher may have omitted from his final version. Caesarea, according to Bartolf, contained an enormous mosque or shrine, and after exorcising ‘all its sordid demons’ it was made into cathedral for the newly consecrated archbishop.23 This was a priest named Baldwin who, William of Tyre said, had come on the expedition with Godfrey.24 He was thus probably a Lotharingian and could be expected to be loyal to Baldwin. In another detail he probably included from local knowledge, William of Tyre, who otherwise leaned heavily on Fulcher’s account of the siege, wrote that Caesarea had no harbour, but the ships tracked the army along the coast. He also mentioned running streams and well-watered gardens. Albert of Aachen described Caesarea as surrounded by densely planted orchards. Baldwin ordered that they were to be chopped down so that Saracens could not lurk among them and ambush the besieging army. It would be surprising if he did this, for he had the patriarch with him who must, like him, have been familiar with the biblical prohibition on destroying orchards.25 But Baldwin was pragmatic and ruthless enough to ignore it. Cutting down fruit trees provided timber, but of course it was green, not seasoned, and so was of limited use for building a siege tower, which is what the king’s men did. According to Albert, the assault was a much more ordered affair than Fulcher recalled. The siege tower was perfectly finished after fifteen days, when it was brought up to the walls and soldiers stationed in it. Then the order was given for attack; led by the patriarch with the True Cross relic, they used ladders to break into the city. The inhabitants retreated to the sea wall and held out from early morning to mid-afternoon before fleeing through the city. The Franks pursued them and plundered valuable spoils. Among those killed were 500 Ethiopian mercenaries sent from Egypt. The qadi (Albert called him a priest, a very old man) was captured and put in chains ‘in order to investigate a sum of silver of immeasurable size that the same priest had buried’. Whether the Christians found the hoard is not recorded, but the citizens of Acre ransomed him for 1,000 bezants. Fulcher’s account must be preferred, because he was an eye-witness, but Albert’s additional detail about the capture and ransom of al qadi is consonant with Baldwin’s constant need for money. It will be noted that the Italian sailors are nowhere in the aforementioned accounts, other than allowing masts and oars (spares, perhaps?) to be used for making the siege equipment. However, we are able to view their experience of the siege through one of them, the Genoese Caffaro, a prominent layman and annalist. He reported very briefly that Arsuf was taken ‘after three days’ fighting’ and that the Genoese then went on to Caesarea, where they beached their ships. They ravaged the cultivated land outside the city walls and began to build siege equipment. Caffaro then included a set-piece discussion between two Muslims from the city and the two clerics, Patriarch Daibert and Cardinal Maurice. After the Christian clergy had, of course, won the argument, the qadi wanted to surrender, but the

118  The conquest of the littoral emir said, ‘ “Let our swords be tested by the swords of the Genoese” ’, as if there were no other enemy to consider. The Genoese were addressed by the patriarch, who promised God’s support but also, according to Caffaro, outlined the plan for the assault: they were to take Communion and then scale the walls with ladders from the galleys and not the siege tower and other machines. He finished by promising plunder. The assault the following day was led by the Genoese consul, Embriaco, and Caffaro described in detail the part Embriaco played. Once they were in the city the Christians pursued the Saracens through the streets to the intermediate wall, where they started to resist their pursuers, but the Genoese climbed over and resumed their killing. There was a great deal of emphasis on booty in Caffaro: the Genoese chased the Saracens through the city to the mosque where 1,000 (so he said) rich merchants had taken refuge in the minaret. They promised all their wealth in return for their lives, and the Genoese allowed the patriarch to negotiate the deal. The Genoese took the city, with a great deal of treasure, by midday. Daibert and Maurice consecrated churches, the largest of them the cathedral, which had formerly been the mosque.26 It is unnecessary to labour the resemblances and minor differences between the accounts of the siege and capture of Caesarea. The most important disparity is the difference in emphasis between Caffaro, who described the whole operation as if it were conceived and led by the Genoese without mentioning King Baldwin at all, and the other chroniclers. Fulcher gave credit for the planning and implementation to Baldwin, and although he mentioned the presence of ships, or at least their masts and oars, he did not specify their nationalities. Albert wrote that the Christian king and the patriarch were there with all the cavalry and the infantry and that they attacked by land and sea with the Pisans and Genoese ‘who had idled away the whole winter at Latakia’.27 On the Muslim side, the Damascus chronicler Ibn al-Qalānisī reported briefly that in May 1101 the Franks ‘captured Qaisarīya by assault, with the assistance of the Genoese, killed its population, and plundered everything in it’.28 There can be no doubt that the Genoese fleet participated in the siege, but whether they provided the entirety of naval support is a different matter that will be addressed in relation to the siege of Acre in 1104. Even William of Tyre described the siege, the sack of the city and the massacre that followed without mentioning the forces from the ships.29 He departed from Fulcher’s account after the concealed coins and described an ‘oratory’ in the city where, because it was a place of prayer, the mass of the population took refuge. Whether this was the mosque that Bartolf and Caffaro both referred to or a Christian chapel is not clear. At all events, the fugitives were massacred within it. In the oratory there was a bright green bowl that the Genoese thought was made of emerald, so they took it as payment of a huge sum of money in kind. When the killing stopped, the spoils were gathered up and divided; the Genoese got a third ‘in accordance with the content of the agreements’ (plural), and the remaining two thirds were for the king and his men. Caffaro described how the loot was subdivided after the Genoese had sailed north to St Symeon, the port of Antioch: one fifteenth went to the galley crews, and of the rest 8,000 men each

The conquest of the littoral  119 received 48 solidi of Poitou and two pounds of peppercorns, and the consuls and sea-captains and ‘men of quality’ also had substantial rewards. They then sailed home to Genoa.30 The Genoese certainly fought at Caesarea and received rich rewards for their naval support. The siege had shown the value of such support to Baldwin, and he probably considered it worth the price in the short run. Sharing booty was one thing, however; granting areas of captured cities and associated trading privileges was likely to reduce the kingdom’s resources in the long run.

The siege and capture of Acre, 1104 Baldwin was then obliged to put his policy of conquering the littoral on hold while he confronted waves of invasion from Egypt that arrived via the southernmost port, Ascalon. In 1103 he made his first attempt to capture Acre, a particularly important goal because possession of the city brought with it a south-facing harbour, the best on the Palestinian coast.31 His first attempt was a failure, which Fulcher of Chartres (writing twenty years later) ascribed to the strength of the defences, both stone and human: But since with its wall and outer wall it was very strong, he could not capture it then, especially as the Saracens inside defended themselves with amazing prowess. So when he had destroyed their cornfields and their stands of trees, and also the gardens, he returned to Jaffa.32 Other historians identified Baldwin’s lack of a strong navy as the cause, including the thirteenth-century Arabic historian Ibn al-Athir: He almost took the place, having set up trebuchets and siege towers. He had sixteen ships at sea. The Muslims from all the coastal regions gathered together, attacked their machines and towers and burnt them. They also burnt the ships. This was a wonderful victory through which God humbled the infidels.33 Albert of Aachen also described the defeat as due an enemy fleet ‘from Tyre, and from Sidon, and from Tripoli, towns under Egyptian jurisdiction’.34 Bartolf of Nangis’s description, written like Albert’s close to the time, was more explicit: Because it was strongly fortified by a wall and an outer rampart, without a fleet he was unable to blockade it on every side, so when he had laid waste the crops and the copses and their ploughlands, he returned home, putting off the conquest of the city until he had the fleet of the Genoese and the Pisans, for which he was waiting.35 Most of the sources thus agree that the 1103 siege Acre failed for want of strong naval support.

120  The conquest of the littoral The accounts of the successful siege the following year are numerous, and in some respects they are surprisingly congruent, but in others there are striking and significant disagreements. The contemporary Damascus chronicler Ibn al-Qalānisī wrote that Baldwin had with him ‘the Genoese vessels which had captured the port of Jubail, over ninety vessels in all’.36 Like his description of the 1103 siege, Fulcher of Chartres’s account was brief, but he explained the different outcome: ‘The Genoese came to the siege with a fleet of 70 beaked ships, and when they had surrounded the city with their engines and frequent assaults for 20 days the Saracens, very fearful, had to surrender the city to the king’.37 Albert of Aachen’s account agrees with these two in outline, but it included the Pisans: Not long after the capture of Jubail the Genoese and Pisans received a legation from King Baldwin [. . .] that, for the sake of God and the saints of Jerusalem, they should attack the town of Acre, besieging it with a fleet of ships, while he himself, with God’s help and the troops of Christ’s faithful, would impose a blockade on land.38 Albert’s inclusion of the Pisans could be dismissed as an error  – as Benjamin Kedar has put it, ‘Albert repeatedly mentions Pisans and Genoese in one breath’ – were it not supported by the testimony of Bartolf of Nangis: After this, once the winter season which was then at hand had calmed down and Easter was drawing near, and a south wind was blowing, a fleet made up of Genoese and Pisans in seventy beaked ships put in at Jaffa, and they were received by the king in a seemly and welcoming fashion. Contracts were then established and agreed to the liking of both sides, and without delay they went to Ptolomaida, that is to Acre, and surrounded it by sea and land. [. . .] The Genoese and Pisans and our men were enriched with their spoils. For not even the king was able to stop them, although he strove hard to keep a watch on them lest the city should be entirely denuded of inhabitants. And when he saw that the Genoese and Pisans had gained the upper hand, he sent his men into the city to take possession of the fortifications and to hold them; and indeed, he became as rich as the others from the booty.39 The sack of the city by the Pisans and Genoese, not recounted by Fulcher, was also described by Albert of Aachen: The Pisans and the Genoese, seeing the inhabitants coming out with all their household goods, and carrying out their incredible wealth, were overcome with blind greed, and, forgetful of the trust and the truce which they had made with the king, they suddenly rushed through the middle of the city, killing the citizens, seizing gold, silver, purple of different kinds and many precious things.40

The conquest of the littoral  121 Thus, on the one side, we have substantial agreement between Albert and Bartolf; on the other side Fulcher’s account is supported by the Genoese annalist Caffaro, who confirmed the participation of his co-citizens: The Genoese [. . .] were moved by religious duty, and so they fitted out forty galleys and went to the Eastern lands. They captured by force the city of Acre along with Jubail, fighting alongside King Baldwin and Count Raymond. There King Baldwin implemented and confirmed the concession to the Genoese in the eleventh year of the indiction. He gave instructions for a copy of these concessions to be inscribed in letters of gold on a single stone in the tribune of the Sepulchre, and he swore along with twelve senior members of his court to maintain the concessions intact in perpetuity as stated in writing. This was in 1105. The Genoese made an agreement with King Baldwin that they would have one street of Acre next to the sea and one garden as representing a third of the city, and receive six hundred besants annually. In addition they would have a third of the land outside the walls up to a distance of one league. [. . .] With this all completed, the Genoese went home in triumph.41 The Golden Inscription in Caffaro’s account has been the subject of some heated debate, stretching over twenty years, between Hans Eberhard Mayer and MarieLuise Favreau on the one side, arguing that the whole thing was a Genoese invention, and Benjamin Kedar on the other, defending the Genoese position. Kedar launched another volley in 2004, but Favreau remains to be convinced.42 It does seem likely that there was an agreement between Baldwin I and the Genoese, and it was recorded within the Holy Sepulchre in Jerusalem. So what of the Pisans? There is no charter evidence for Pisa, but there is a disparity between Caffaro’s forty ships and the seventy or ninety of Fulcher and Ibn al-Qalānisī which would be explained if Pisan ships were there too. Crucially, though, Pisa had no Caffaro to compose his city’s annals. The nearest is Bernardo Marangone, a generation younger (1104–1175). He started writing in the 1130s, so did not have Caffaro’s first-hand experience, but the events of the First Crusade were still well within living memory. Marangone’s entry for 1099 has already been mentioned and quoted with regard to Daibert’s attack on Byzantine islands; it ends thus: ‘Moreover, on that same journey the Pisan people captured Maida, a very strong city, and besieged Latakia with Bohemond, and Jubail with him and Raymond, count of Saint-Gilles.’ Although the entry is under the date 1099 in the annals, it is obviously the whole story of Pisan involvement in the First Crusade, including the sieges of Latakia (1099) and Jubail (1104). In the Pisan memory, at least, the commune’s fleet was involved in all these engagements. Marangone’s entry refers to a city called ‘Maida’; since it is listed in the same sentence as Latakia and Jubail, it seems self-evident that this is (Ptolo)maida, also known as Acre, a powerful city captured in the same year as the siege of Jubail.43

122  The conquest of the littoral Given Marangone’s testimony, along with the independent accounts of Bartolf of Nangis and Albert of Aachen, it is very probable that a Pisan fleet did assist in the 1104 siege of Acre. Marangone implied that it was the official fleet sent by the commune in 1099, but Favreau is convinced that ‘at that time there was no fleet manned and sent off to Syria by Pisa’s government’, although she concedes ‘that some merchants, shipowners and sailors from Pisa who with their ships were present in the crusader kingdom at that time (1104) might have participated in the sea blockade on their own’.44 Meanwhile, if we accept the presence of Pisan ships it raises a question: why were they absent from Caffaro’s and Fulcher’s narratives? The answer in the case of both authors is political. Caffaro was writing in the 1150s to argue the city’s case when Genoa was in dispute with the Latin rulers in Syria about its trading privileges. He would not wish to make an argument for Pisan rights at the same time, especially since, as Richard Face has pointed out, the whole of the twelfth century was a period of intense conflict between Genoa and Pisa.45 So far as Fulcher of Chartres is concerned, if Bartolf of Nangis followed his exemplar and his account of the capture of Acre is an accurate record of events as originally recorded by Fulcher, then Fulcher suppressed references to the Pisans at some point in the 1110s or 1120s. His reasons for this would lie in the alliance formed no later than 1109 between the Pisans and Alexios Komnenos. On 18 April 1110, the Pisans took an oath of fealty to the Byzantine emperor, receiving trading privileges in return. Ralph-Johannes Lilie explained that ‘the sense and purpose of the treaty [was] the participation of Pisa, or at least its indirect support, in the conquest of Syria and Palestine by the Byzantine Empire’.46 The anti-crusader tendency of the treaty exacerbated the difficult relationship between the kingdom of Jerusalem and the Byzantine empire that had begun with the dispute between Baldwin I and Patriarch Daibert. Thus the earliest accounts of the siege and capture of Acre agree that Pisans were involved although they were edited out of the picture by the two chroniclers who (ironically) were in a position to present the most accurate accounts, Caffaro and Fulcher. The siege of Acre is significant because the privileges extracted by the Genoese, who contributed the majority of ships and were careful to reach an agreement before committing themselves to the siege for their naval support, reflected Baldwin’s absolute dependence on commanding a fleet to close the blockade of the maritime cities. He was setting a precedent that ultimately impoverished the kingdom of Jerusalem, but evidently he recognised that he had no choice. One thing that historians medieval and modern are agreed upon is the inestimable value of Acre as a port. Unlike Jaffa, Arsuf, Haifa and Caesarea it had a south-facing harbour that sheltered shipping from the prevalent westerly winds. As Fulcher said: ‘[Acre] was very necessary to us, because there is an extremely useful harbour there and it safely receives very many ships within its secure walls.’47 Its future importance can be gauged from the way it was fought for during the Third Crusade and then became the capital of the kingdom of Jerusalem for a century after the Holy City was lost.48

The conquest of the littoral  123

The capture of Tripoli (1109) and Beirut (1110) The capture of Tripoli was not a strategic aim of Baldwin: it was too far north to serve the interests of the kingdom of Jerusalem. However, the Genoese were prepared to trade their naval support for access and privileges anywhere along the coast, and in 1109 they were assisting Bertrand, the son of Raymond of Saint-Gilles (d. 1105), who was attempting to capture the town, as had been his father’s main purpose during his final years. There is some disagreement about the size of Bertrand’s fleet: Fulcher of Chartres said it numbered around seventy ships; Albert of Aachen wrote that Bertrand had left Provence with forty galleys, each carrying 100 soldiers plus crew, and put in at Pisa where eighty Genoese galleys joined him, having concluded a treaty; Caffaro reported that the Genoese supplied sixty galleys; Ibn al-Qalānisī wrote that Bertrand arrived with sixty vessels laden with Franks and Genoese.49 Because Bertrand was officially his vassal, Baldwin received a share of the booty when Tripoli was taken, but his main purpose in assisting at the siege was explained by Fulcher of Chartres: ‘King Baldwin came to the siege and pleaded with the Genoese to help him that year to capture Ascalon and Beirut and also Sidon.’ The king’s presence strengthened the siege. and the city governors surrendered on terms of safe conduct, but the ‘lesser folk’ among the Genoese sacked the city. Ibn al-Qalānisī did not distinguish Franks from Genoese but included a detailed description of the sack. Baldwin’s presence may have averted worse destruction: the two chroniclers agree that he allowed safe passage to those townsmen who had made the surrender agreement. Albert of Aachen added a few details: Bertrand’s swearing an oath of allegiance; the king’s arrival three weeks into the siege at Tripoli, which would not yield ‘unless King Baldwin was there in person’; the presence of Pisans along with the Genoese and the citizens’ fear that they would be attacked by the Italians in violation of the truce ‘just like the people of Acre’. Both Ibn al-Qalānisī and Caffaro reported that the Genoese were granted one third of Tripoli for their services. Baldwin did not press any claim to Tripoli, probably because he did not want to alienate Bertrand, now his vassal, or to forfeit the naval support Bertrand had brought with him. Instead he achieved his aim of persuading Bertrand and the Genoese to participate in the siege of Beirut. This siege was, unusually, through the winter of 1109–1110.50 Once again Fulcher of Chartres’s account was short. Bertrand of Tripoli was there, too, encamped a mile from the city, and although Fulcher did not mention the Genoese there were certainly ships because they were used to blockade inside the harbour some enemy vessels that had come to Beirut’s assistance. After seventy-five days, so Fulcher estimated, the Franks took the city by storm, pursued the fleeing enemy and plundered.51 Ibn al-Qalānisī told a story similar in broad terms. While the Franks were constructing siege towers nineteen Egyptian warships arrived, overcame the Frankish ships and were able to provision the city. Baldwin then sent to St Symeon ‘asking aid of the Genoese who were there with their ships’, and forty came in response. The Franks now attacked by land and sea, the commander of the Egyptian fleet was killed and the city was

124  The conquest of the littoral taken by storm. The governor fled with some of his troops, but they were captured and killed, and all their treasure was plundered. The city was sacked.52 Albert of Aachen claimed that Bernard sailed with the Pisans from Tripoli and blockaded Beirut from the seaward side. In the spring the town’s emir escaped to Cyprus, and the citizens surrendered. They were allowed to leave under safe conduct, and many did, but ‘of those who were found still in the city, who did not leave when the truce was made and had stupidly stayed, around twenty thousand were killed by Bertrand and the Pisans’.53 One of Albert’s details that is supported by Ibn al-Qalānisī is that some time into the siege Baldwin received messengers from Baldwin of Bourcq in Edessa saying that the city was under attack from a massive Turkish army and was in great need of assistance. Albert’s description of the king’s reaction to the SOS is very revealing of his character and also his belief in the imperative of securing the coastal cities: When the king heard these things he ordered the legates on pain of death to keep quiet about this unwelcome rumour, which he also pretended not to have heard, suppressing it with a wonderful silence, so that the hearts of men would not hear this arrogance and be terrified at the Turks’ boldness, and so strive less for the destruction of Beirut. The king kept quiet, so did the messengers. And he renewed his assault.54

The capture of Sidon, 1110 Baldwin’s final success was the capture of Sidon. In most respects the siege followed a familiar pattern, but there was an additional and exotic element: the participation of King Sigurd Magnusson of Norway (1103–1130).55 After a two-year journey, King Sigurd arrived with some sixty ships and a considerable army.56 He tried to tempt the Ascalonites into a fight and then sailed on to Jaffa. He went up to Jerusalem with Baldwin, who was on his way back from his march to relieve Edessa and was anxious to get to Acre, which was under attack from the Egyptians. Nevertheless, Baldwin evidently recognised the importance of flattering Sigurd and securing the naval assistance he could supply, and he arranged a magnificent reception, a guided tour of the holy places and a trip to the river Jordan. Then they formed an alliance to besiege Sidon. Sigurd moved his navy up from Jaffa, which had the result of persuading the Egyptians to withdraw from their position off Acre to Tyre, which was still in Muslim hands. It took six weeks to bring the inhabitants to the point of negotiating a surrender. The terms agreed were a safe conduct for the emir and any others who wanted to leave. Perhaps because the Italians were not involved in the siege, 5,000 Sidonites left peacefully with the possessions they could carry and the rest became subject to King Baldwin.57

The conquest of the littoral  125

Conclusion Baldwin’s successes in extending his kingdom to include the coastal cities that were so vital to the kingdom’s survival were all within the first decade of his reign. The same period was punctuated by attempts to take the two remaining ports, Tyre and Ascalon. These attempts did not stop entirely after 1110, but increasingly Baldwin was engaged in campaigning against the Turkish threat to Edessa and Antioch to the north. Tyre was captured in 1124, but Ascalon was not taken until 1153. Already the Egyptians had landed armies there to threaten the kingdom, and the coastal cities that were captured by the Christians would be under threat as long as the Egyptian navy had a harbour so close to them. A further problem for Baldwin and for his successors was that the cities had been taken at a price: as the peaceful surrender of Sidon had demonstrated, the participation of the Genoese, although vital, led to death and destruction in the short term and in the longer term to a significant loss of revenue from trade and tribute to the crown. Failure to take Ascalon and the price of Genoese naval support were emphatically not Baldwin’s only problems, however.

Notes 1 See chapter 9. 2 Köhler also set the division in 1112–1113, ‘i.e. up to the firm establishment of Frankish Rule in Edessa, Antioch, Jerusalem and Tripoli’. He went on to observe that the same years marked a change in the campaigns of Turkish armies from outside Syria and hence to a new attitude regarding alliances and treaties: Michael A. Köhler, Alliances and Treaties Between Frankish and Muslim Rulers in the Middle East: Cross-Cultural Diplomacy in the Period of the Crusades (Leiden, 2013), p. 62. 3 See chapter 9. 4 The events of Baldwin’s reign as regards the control of Aleppo in northern Syria and from Muslim perpectives are covered in chapter 2 of Nicholas Morton, The Field of Blood: The Battle for Aleppo and the Remaking of the Medieval Middle East (New York, 2018), pp. 49–82. 5 Ṭughtigin was atabeg for Duqāq of Damascus 1093–1104. After Duqāq’s death in that year Ṭughtigin manipulated the succession and soon became unchallenged ruler, nominally loyal to the Saljūq sultan Muhammad. The sultan recognised him as emir in 1115. Ṭughtigin died in 1128. See Taef El-Azhari, ‘Ṭughtigin (d. 1128)’, in Crusades Encyclopedia, IV: 1204–205. 6 Köhler, Alliances and Treaties, pp. 85–88, for this and other examples of condominia between Franks and Muslims in this period; Cobb, The Race for Paradise, p. 114. 7 See chapter 8. 8 Köhler, Alliances and Treaties, pp.  88–89. On the other hand, Cobb, working from Arabic sources, described proposals for joint offensives initiated by Cairo in 1103 and 1105 that were met with little enthusiasm from Damascus: Paul M. Cobb, The Race for Paradise: An Islamic History of the Crusades (Oxford, 2014), p. 108. For conflicting views of Egyptian policy in this period, see also Carole Hillenbrand, The Crusades: Islamic Perspectives (Edinburgh, 1999), pp. 77–78. 9 RA, p.  141. See also France, Victory in the East, pp.  336–37. France claimed that Jaffa’s fortifications were slighted by the Egyptians as they withdrew (pp. 329, 333), but I have been unable to find a reference to substantiate this. AA called it an ‘ancient

126  The conquest of the littoral

10 11 12

13

14

15 16 17 18 19 20 21 22 23 24

25

destruction’ (antiquo exterminio), although he may, of course, have been mistaken: 7.12, p. 502. Hagenmeyer, Epistulae, no. XIX, pp.  174–75. The letter was addressed to clergy ‘throughout the Gauls’. Hagenmeyer, Epistulae, no. XXI, pp. 176–77; Smith, ‘Scribal Crusading’. See chapter 6 for what Daibert did with the donations of the faithful. FC 2.6, pp. 384–90. BN did not include the same chapter on numbers, but this may have been in accordance with his usual practice of omitting information that reflected Fulcher’s personal experience. RC claimed that after the battle of Ascalon there were scarcely 200 armoured men to defend Jerusalem and about 80 of these were Tancred’s, chap. 139, pp. 703–704. This list implies an agreed boundary between the as yet unfounded county of Tripoli and the kingdom of Jerusalem, running north of Beirut and south of Jubail. See Hans Eberhard Mayer, ‘Le service militaire des vassaux a l’étranger et le financement des campagnes en Syrie du nord et en Égypte au XIIe siècle’, in Mélanges, pp. 93–161. AA 6.51–53, pp. 472–74; 7.1–6, pp. 486–94. Albert’s account included the torture of Gerard of Avesnes by the defenders of Arsuf: rather than rescue him, the Christians shot him with arrows; nevertheless, he survived and was sent from Arsuf to Ascalon. When a truce was made with Ascalon Gerard was sent back to Godfrey, cured of his wounds and well dressed and mounted. He was compensated with the wealthy fief of Hebron: AA 7.15, p.  506. Similar detail about the Arsufians’ treatment of Christian captives in this siege were included by FC as a flashback following his account of the city’s surrender in 1101: FC 2.8.7, p. 400. BN chap. 45, p. 523. AA 7.51, p. 560. FC 2.8, pp. 393–97 (the treaty is also in UKJ, I: 125–27, no. 21 [RRR 58] taken from FC). Fink and Ryan translated FC’s Ianuensibus as a dative, ‘injury done to the Genoese’ (p. 152), but my interpretation is less ambiguous and more logical. Michael Matzke, ‘Pisa’, in Crusades Encyclopedia, III: 964–66. BN chap. 50, p. 527. As before, William of Malmesbury, like Bartolf, seems to have been using a less heavily edited version of FC: He included the Pisans with the Genoese in the agreement: WM 4.380.1, p. 676. AA 7.54, p. 562. AA 7.55, pp. 562–64; FC 2.9.1–10, pp. 400–404; BN chap. 50, p. 527. FC 2.9.2, pp. 401–402. In BN (chap. 50, p. 527) I have read idoleum (or idolium) for the idolum in the printed text, since the cathedral has to be accommodated in a building and not in an image or idol. WT 10.14(15)‑15(16), pp. 469–72 for the siege and capture of Caesarea, Archbishop Baldwin at p. 472. GN claimed that Baldwin had fraudulently financed his crusade by carving a cross on his own forehead to attract alms: GN 4.17, pp. 197–98; 7.23, p. 328. See also Murray, Crusader Kingdom, pp. 184–85. AA 7.55–56, pp. 562–66. The verse reads: When thou hast besieged a city a long time, and hath compassed it with bulwarks to take it, thou shalt not cut down the trees that may be eaten of, neither shalt thou spoil the country round about with axes: for it is a tree, and not a man, neither can it increase the number of them that fight against thee. (Deut. 20: 19)

26 Caffaro, Annali genovesi, XI: 3–74; siege of Caesarea at pp. 9–13. 27 AA 7.55, p. 564. 28 IQ, p. 51, entry for ah 494. IQ used ‘Genoese’ (al-janūyūnu) to refer to Italian seafarers generally. I am grateful to Betty Binyish for advice on this. 29 WT 10.15(16), pp. 471–72.

The conquest of the littoral  127 30 Caffaro, Annali genovesi, p. 13. The ‘emerald’ chalice is still on display in the cathedral of San Lorenzo in Genoa, see WT, note on p. 471 citing W. Heyd, Histoire du commerce du Levant au moyen-âge, 2 vols (Leipzig, 1885–86), I: 137, n. 7; its continued presence there is confirmed by the website of the cathedral’s treasury: www.genova gando.it/turismo_liguria/musei_e_monumenti/tesoro_san_lorenzo/museo_tesoro.htm [accessed 6 December 2017]. The vessel is glass, although WT said the Genoese were still in his day passing it off as emerald. 31 The history and long-term importance of Acre to the kingdom of Jerusalem are explored in Acre and Its Falls: Studies in the History of a Crusader City, ed. John France (Leiden, 2018), especially Susan B. Edgington, ‘The Capture of Acre, 1104, and Baldwin I’s Conquest of the Littoral’, pp. 13–29. 32 FC 2.22.1, pp. 456–57. 33 IA, I: 61. 34 AA 9.19–20, pp. 660–62. 35 BN chap. 61, p. 536; see also WT: 10.25(26), p. 485. 36 IQ, p. 59, entry for ah 497. For the capture of Jubail see AA 9.26, p. 670 and n. 55. The Italians were acting for Raymond of Saint-Gilles, who had also appropriated Tortosa in 1102 after it was taken by survivors of the 1101 expeditions; see FC 2.17.1–2, pp. 433– 35; AA 8.42, p. 632. These cities were further north and in the region Raymond aspired to rule, the future county of Tripoli. 37 FC 2.25.1, pp. 462–63. 38 AA 9.27–29, pp. 670–74. 39 BN 63, p. 537. See also Benjamin Z. Kedar, ‘Genoa’s Golden Inscription in the Church of the Holy Sepulchre: A Case for the Defence’, in I comuni italiani nel regno crociato di Gerusalemme: Atti del Colloquio di Gerusalemme, 24–28 maggio 1984, ed. G. Airaldi and B. Z. Kedar (Genoa, 1986), pp. 317–35. 40 AA 9.29, p. 674. 41 Caffaro, ‘De liberatione civitatum Orientis liber’, Annali Genovesi, ed. L. Belgrano, I: 99–124. 42 H. E. Mayer and M.-L. Favreau, ‘Das Diplom Balduins I. für Genua und Genuas Goldene Inschrift in der Grabeskirche’, Quellen und Forschungen aus italienischen Archiven und Bibliotheken, 55/56 (1976), 22–95. Benjamin Z. Kedar, ‘Again: Genoa’s Golden Inscription and King Baldwin I’s Privilege of 1104’, in Chemins d’outre-mer: Études d’histoire sur la Méditerranée médiévale offertes à Michel Balard (Paris, 2004), pp. 495–502; Marie-Luise Favreau-Lilie, ‘Genoa’, in Crusades Encyclopedia, II: 501–506, with full bibliography of the dispute. 43 Bernard Marangone, p.  7; abstracted as ‘Gesta Triumphalia Pisanorum’, p.  368. In his more recent edition of the Gesta triumphalia per Pisanos facta (Florence, 2010), Giuseppe Scalia discussed the identity of ‘Maida’ both in his introduction (p. XXXV) and in a note to the city’s capture that is described on p. 4 (n. 16, p. 32). However, he did not consider Ptolomaida or arrive at any alternative conclusion. 44 M.-L. Favreau-Lilie, pers. comm. (email) 31 August 2010. 45 Belgrano, intro. to Caffaro, pp. XCII‑XCVIII; Richard D. Face, ‘Secular History in Twelfth-Century Genoa’, Journal of Medieval History, 6 (1980), 169–83. 46 Lilie, ‘Genoa’, p. 88. 47 FC 2.25.2, p. 463. 48 For Acre’s history before 1104 and a fuller account of the two sieges of Baldwin’s reign, see Edgington, ‘The Capture of Acre, 1104’. 49 FC 2.40–41, pp. 526–33; AA 11.3, p. 776; 11.11–14, pp. 782–86; Caffaro, ‘De liberatione’, p. 123; IQ, pp. 88–90, entry for ah 502; WT 11.10, pp. 509–10. 50 AA said the siege began in December  1109, 11.15, p.  786; FC has February  1110, 2.42.1, p. 534. WT mentioned the galleys over-wintering, implying this was unusual: 11.13, p. 515. 51 FC 2.42, pp. 534–36.

128  The conquest of the littoral 52 IQ, pp. 99–100, entry for ah 503. 53 AA 11.15–17, pp.  786–90. Caffaro did not describe the siege, but a later Genoese source said that King Baldwin captured Beirut in 1110 with twenty-two galleys: ‘Brevis historia’, in Annali Genovesi, XI: 128. For the end of the siege, FC and IQ agree 13 May 1110; AA has 27 May 1110. 54 IQ, p. 99; AA 11.16, p. 788. 55 The Nowegian king was later known as Sigurd ‘Jerusalemfarer’, and the oldest vernacular version of his journey is preserved in the Morkinskinna, which dates from early in the thirteenth century: ed. and trans. C. R. Unger (Oslo, 1867), pp. 156–98 (number of ships at p. 157; siege of Sidon at p. 187). The tale was also incorporated into the Magnússona Saga, or Saga of the Sons of Magnus, which may be read in Snorri Sturluson, Heimskringla, trans. L. M. Hollander (Austin, TX, 1964), pp.  688–714. A short account in Latin was written in the 1170s or 1180s by Theodric the Monk: see Monumenta Historica Norvegiae, ed. Gustav Storm (Oslo, 1880), pp. 3–68 (chap. 33, pp. 65–66). 56 AA 11.26, pp. 798–808 is the longest account. Cf. FC 2.44, pp. 543–48; IQ, pp. 106– 108, entry for ah 503. There were more than sixty ships according to IQ (p. 106); 55 in FC (at the siege of Sidon, p. 544); according to the Heimskringla there were sixty ships when Sigurd left Norway, but the same source recounts eight battles before they arrived in Palestine (p. 689). 57 IQ revealed that ‘the Egyptian fleet was then stationed at the port of Tyre, but it was unable to bring support to Sidon’ (p.  107), and confirmed the surrender terms, also commenting on the severity of the tribute exacted from those who remained in the city (pp. 107–108).

8 Fighting the Saracens

Fighting the SaracensFighting the Saracens

At the same time as he was pursuing an aggressive policy to achieve the capture of the coastal cities, throughout his reign Baldwin had to defend the kingdom against attack from the crusaders’ Muslim enemies. When Baldwin moved south to Jerusalem in 1100 he left Edessa and Antioch on the front line against Turkish attacks. The more immediate threat came from the Fāṭimids of Egypt who had captured Jerusalem from the Turks in 1098, only a year before the city was besieged and captured by the First Crusade. The crusaders’ blockade had been comparatively short – only five weeks – and so they were in possession of the city before the Egyptians’ relieving force, led by the vizier al-Afḍal in person, arrived at Ascalon. Thus within weeks of the conquest the Christians had to march out to meet a vast Fāṭimid army on the plains outside Ascalon. Against overwhelming odds they were successful and could have pressed on and taken Ascalon, but the garrison would surrender only to Raymond of Saint-Gilles, and Godfrey was not prepared to countenance this: Raymond had opposed his candidacy for rulership, refused to hand over Jerusalem’s citadel and joined the army at Ascalon only reluctantly and belatedly. Rather than cede possession of Ascalon to him, Godfrey forfeited the best opportunity to take the city for the coming half-century. Baldwin was to see the consequences of this in the series of battles of Ramla in the first few years of his reign.1

The first battle of Ramla, 1101 In the summer of 1101 a Fāṭimid army sent by al-Afḍal arrived at Ascalon and prepared to meet the army of Jerusalem at Ramla. The town was the former capital of Fāṭimid Palestine and was situated on the plain between Jerusalem and its port, Jaffa. The Egyptians’ goal, as Michael Brett has plausibly proposed, was to recapture Jaffa.2 Fulcher of Chartres, who was still close to Baldwin, described the Egyptians’ aim more comprehensively, if his early reaction was accurately represented by Bartolf of Nangis: ‘At that time the king of Egypt assembled a very great army and sent it to conquer Jerusalem and everyone living in it, and in the whole land of Israel, to destroy and annihilate the Franks.’3 Although, Bartolf went on, ‘all Israel’ was terrified by the news of the Egyptians’ approach, Baldwin reacted by setting out from Caesarea with his army to meet them.

130  Fighting the Saracens The Egyptians arrived at Ascalon towards the end of May, and Baldwin moved to the plains nearby. The Latin army was encamped there for twenty-four days while the Egyptians tried to lure it into advancing so as to attack and destroy it. This tactic failed, and the Latins returned to Jaffa where they stayed for seventy days, until the beginning of September. This allowed Baldwin to send to Jerusalem and Jaffa and all his possessions – still rather few – for reinforcements, although apparently he did not appeal to potential allies in Edessa or Antioch. The Latin writers emphasised the enormous disparity in numbers between the two armies. Both versions of Fulcher’s history agree that the Christian army numbered 260 cavalry and 900 infantry, but the earlier one, Bartolf, reported the enemy army as comprising 1,500,000 cavalry and 20,000 infantry, plus retainers who looked after the baggage train and were also armed. These numbers were allegedly obtained from Egyptian prisoners of war after the battle, but Fulcher apparently decided later that these were not reliable informants, for in his version of the 1120s he scaled back the number of Egyptian knights to 11,000 and adjusted the number of footsoldiers to 21,000.4 According to Albert of Aachen, Baldwin had 300 cavalry and 1,000 infantry – figures close enough to Fulcher’s – facing an army of around 200,000 cavalry and infantry combined.5 The estimates of numbers on the Christian side are probably fairly accurate, but obviously the size of the Egyptian army was grossly exaggerated by contemporary reporters to stress the overwhelming odds against a Christian victory. The same tendency to exaggerate enemy numbers is seen on the Arabic chroniclers’ side: estimates of Frankish numbers were 1,000 cavalry and 10,000 or 20,000 foot.6 Such exaggeration was routinely used to explain a defeat by the enemy or enhance a victory over it. Contemporary historians went on to describe the ensuing battle in great detail. Fulcher’s account stressed the importance of Christian belief: ‘We were putting our faith neither in weapons nor a great army, but we had placed our hope entirely in the Lord God.’7 Accordingly, the True Cross relic was carried before them, and King Baldwin addressed his troops as ‘soldiers of Christ’, exhorting them to fight for the salvation of their souls. The Christian army was divided into six divisions (Albert said five) and charged the enemy. The two front lines were quickly driven back, but Baldwin then brought up his division; Fulcher, an eye-witness, saw him kill an Arab with his lance, pull it out and continue the killing. At last the remaining enemy troops fled. Fulcher claimed that the dead were literally innumerable, but that reports said 5,000 cavalry and infantry had been killed, including the Egyptian leader. Christian casualties were eighty knights and more infantry. He concluded: ‘That day the king bore himself most valiantly; he was a very great source of encouragement, a stern slayer.’ Bartolf of Nangis’s account, intriguingly, differs both in detail and in emphasis. He reported, for example, that the king took up position between Ramla and Lydda and while he was there enemy spies suddenly appeared from a cave and sounded the signal for war. The Christians took up arms, some enthusiastically, others hesitantly, ‘as one might expect from an army of different nations and sons of many mothers brought together from all parts of the world’. This was the reason for the

Fighting the Saracens 131 king’s rousing address, which is much longer than Fulcher’s later version of it. As well as reminding the troops of God’s favour, Baldwin recalled ‘that memorable day’ when they defeated the Turks at Antioch (a battle from which he was absent) and their more recent efforts at Tripoli. He concluded by telling them to put their trust and hope in God. At this point the relic of the True Cross was displayed, confession was heard ‘and those who were previously wavering timidly and feebly and were wholeheartedly afraid were now tougher than any boxer or wrestler and had become bold’. This testimony to Baldwin’s effectiveness as a commander was followed by another comment on Baldwin’s character that did not find its way into the later Fulcher: when the two front lines were repulsed, Baldwin came to the rescue ‘like a good father helping his beloved sons at the very point of death’.8 Albert of Aachen named some of the divisional commanders, for example Geldemar Carpenel, who commanded the second division, and Hugh of Tiberias, who was in the third. In an incident that reveals the dominance Baldwin had established over his clergy, two bishops, Gerard and Baldwin, had the temerity to suggest to the king that the Christian army would not defeat the Egyptians unless Baldwin made peace with Patriarch Daibert. King Baldwin utterly refused to do this, but he was still allowed to confess and receive communion. Albert went on to describe the two divisions in the vanguard being driven back by the Egyptians so that the two in the rearguard were launched into the fray. According to Albert, the fourth division, made up of Jerusalemites, ‘men well used to war and very hardy’, was repulsed by the enemy and started to flee; the fifth, led by the king himself, came to their relief. The king’s intervention turned the battle, but writers agreed that it was the ‘power of Lord Jesus Christ and the Holy Cross’ that ensured a Christian victory. The presence of the Cross in the battle was widely affirmed, and its role (if not miraculously to strike dead the Saracens, as more than one source claimed) was evidently to be a great morale booster.9 The Christians’ losses were severe, and the battle did not end that day on the plains of Ramla. They encountered the Egyptians again outside Jaffa: these were troops who had been blockading Jaffa while the main army engaged the Franks, and so they were fresh.10 According to Albert, King Baldwin, with only forty knights and a few infantry, put 20,000 Egyptians to flight.11 When the king and a few companions returned from the pursuit, the king ‘took off his iron hauberk and purple clothing [and] it looked entirely saturated up to the neck with the blood and gore of the enemy’. They returned to Jerusalem with a great deal of plunder, and Baldwin granted a tenth of this to the hospital and the poor. According to Fulcher of Chartres, the force of Saracens encountered by the Christians outside Jaffa numbered fewer than 500, and they were survivors of the battle the previous day who had told the citizens that the king and all his army had been killed.12 When the Frankish survivors appeared they at first mistook them for fellow Arabs and approached them, only to be attacked. Fulcher wrote, though, that the Franks were too weary and wounded to pursue the enemy. As a significant postscript to the battle, the inhabitants of Jaffa had been led to believe by fugitives from the battle of Ramla that Baldwin and all his men had been killed, and this led them to write to Tancred in Antioch. Baldwin’s wife despatched the letter by ship. In view

132  Fighting the Saracens of the longstanding hostility between Baldwin and Tancred, the choice of recipient is interesting, as is Tancred’s ‘great sorrow and grief’ on reading it. However, before he could set out, another letter arrived that reported that the Saracens had in fact been ‘decisively and splendidly defeated’ and the king was safe in Jaffa. Bartolf of Nangis included a more vivid account of the encounter with Saracens coming from Jaffa who mistook the Franks for fellow Arabs and were duly slain. The Christians resumed their march to Jaffa, and Bartolf described how they were recognised from the walls and welcomed ‘like men resurrected from the dead’. Those inside the city then confessed that they had written to Tancred: But Baldwin, hearing talk of this kind, was greatly saddened that they had sent a document to Antioch about his death. He ordered a letter to be written about his being alive and his victory and well-being and he sent it to Tancred as fast as he could by a swift messenger. Tancred had already received the first letter and was ready to come to Jerusalem with all his army; the second letter-carrier met him and handed over the document signed with King Baldwin’s seal, very different from the first letter. When Tancred read it and learned of the king’s being alive and his victory and well-being, he greatly rejoiced, even though he did not love him, and gave many kinds of thanks to God, on behalf of both himself and all the Antiochenes, and so he stood down his army from that campaign.13 This earlier version of events, and especially the words I have italicised, seem a much more accurate depiction of the situation as it appeared to contemporaries than the more restrained revision Fulcher produced in the 1120s. Baldwin’s first encounter with the power of the Fāṭimid army was thus a qualified success: he routed the Egyptians, although at considerable cost in lives of his own men, and acquired spoils of war that alleviated briefly the chronic shortage of funds that made it difficult to mount an effective defence of the kingdom. The psychological effects of the Christian victory were as important: the puny forces of the infant kingdom had taken on the might of the Saracens and won. Baldwin was evidently content to attribute his victory to the power of the True Cross, and its presence on the battlefield came to have great symbolic significance.14 The campaign revealed Baldwin to be an effective and courageous general. His personal command of the rearguard had saved the day when the first divisions were repulsed and he was prepared to engage personally in hand-to-hand combat. Even after the battle he remained on the alert and dealt swiftly both with a group of Saracen survivors and with the desperate situation he found in Jaffa. As the campaigning season came to an end in 1101 he had assured, for a few months at least, a precarious peace.

Arrival of the ‘Crusade of 1101’ Eight months of peace, in fact, came to an end with the arrival of another Egyptian army at Ascalon in May 1102. In the intervening period Baldwin had an exercise

Fighting the Saracens 133 in diplomacy to deal with the fall-out from the waves of Christian reinforcements that are usually referred to collectively as the Crusade of 1101. More accurately, they were a later response to the call for a crusade in 1095. There were four separate expeditions: the first to leave comprised about 8,000 northern Italians and was led by Archbishop Anselm of Milan; another 2,000, led by Conrad the imperial constable, joined them at Constantinople in April 1101. These people all arrived overland. A second army, about 3,000 strong, was commanded by William II, count of Nevers, and made its way by sea to Constantinople, arriving by mid-June. The third contingent was from northern France, Flanders and Burgundy and included the renegade Stephen of Blois. Some 8,000 of them arrived at Constantinople by sea early in May. The fourth was a combined force from Aquitaine, led by Duke William IX, and from Bavaria, led by Welf IV. This army of maybe 16,000 also travelled overland and arrived at Constantinople in early June.15 For Alexios Komnenos it was a recurrence of the nightmare of 1096–1097.16 Both of the overland expeditions caused trouble in Byzantine lands and continued to do so outside Constantinople. The Lombard and northern French armies were shipped across to Nikomedia, and Raymond of Saint-Gilles was detailed to guide them across Asia Minor. A wrong-headed scheme to rescue Bohemond from the Dānishmendid Malik-Ghāzī took them away from the recommended route; a formidable Turkish army was waiting for them, and they were heavily defeated. Some survivors, including Stephen of Blois and Stephen of Burgundy, straggled back to Constantinople by land; Raymond of Saint-Gilles went by sea. William of Nevers was later in leaving Constantinople with his army, and, perhaps because the much larger forces that preceded him had deviated from the expected route, he failed to catch up with them. Again the Turks attacked, and this time the cavalry fled and the infantry were massacred. William himself escaped to Antioch. The last and biggest army heard rumours of these disasters before it set out from Constantinople, and the word was that Alexios was conspiring with the Turks to destroy the westerners. Some of the Germans sailed directly from Constantinople to Jaffa, including Ekkehard of Aura, possibly the author of an eye-witness account of the first battle of Ramla. The story for the rest was as before: lack of food and water compounded by Turkish raids and culminating in another Turkish attack and defeat in battle. William IX and Welf IV were among those who got away. After this series of heavy defeats the remaining forces assembled in Antioch. Under the leadership of Raymond of Saint-Gilles they left there in midFebruary 1102, and he took them to besiege Tortosa: the capture of the town was described in the previous chapter. Leaving Raymond there, the army travelled south. Baldwin met them at Beirut, and they went from there to Jaffa. Once there it was a relatively easy step to fulfil their pilgrims’ vows and to celebrate Easter. William IX then took ship for southern France; the rest of the army leaders were prevented from sailing out of Jaffa by contrary winds and were still with Baldwin when news of another Egyptian invasion arrived. The arrival of reinforcements was assuredly very welcome to Baldwin, but in the nearer future he had to try to mend diplomatic relations with Constantinople. The newcomers were insistent

134  Fighting the Saracens that Alexios had betrayed them to the Turks and generally obstructed their journey, according to Albert of Aachen: They recalled the evils and dangers befalling the pilgrims and they gave advice to King Baldwin that in the most humble way he could, and with mild entreaties, he should appeal to the emperor of Constantinople about the Christians’ miseries, that is to say, the emperor should stop ruining and betraying Christians, should assist the church in Jerusalem, should not listen to Turks and Saracens, and should not refuse to allow fully and faithfully all the trade in provisions from the islands and districts of his kingdom to Jerusalem.17 However, Baldwin listened to more temperate advice: In truth, as was reported by truthful and distinguished men, [Alexios] was not to be blamed for this crime at all, for he frequently warned and informed the army about the wastelands and shortages and Turkish ambushes in the out-ofthe-way places of Paphlagonia, and he told them that for these reasons they could not march securely and safely along that route.18 He sent an embassy and a gift of two pet lions to the emperor. Albert named three of the legates, but their identities are less certain. ‘Archbishop Gerard’ may have been the same man who bore the True Cross at the first battle of Ramla the previous year; ‘the bishop of Barzenona’ has not been identified; and there was a knight called Engelbert. The last of these took the good news back to Baldwin that the emperor wanted good relations with the king, but the mystery bishop was not convinced of the emperor’s innocence and denounced him to the pope, who (according to Albert) provided him with letters that enabled him to spread anti-Byzantine feeling ‘among all the princes of Gaul’. If this was indeed what happened, it may well have played a part in stirring up support for Bohemond’s ‘crusade’ against the Byzantines in 1106. However, Baldwin was presumably unaware of possible future damage to relations with the Byzantines when he prepared to face another Egyptian invasion in late spring 1102.

The second battle of Ramla, 1102 The Egyptian forces this time were even greater than the previous year. As before, the Latin sources generally gave exaggerated estimates, although Albert stuck with ‘countless’.19 The Arabic sources did not give numbers of Egyptian troops, but Ibn al-Qalānisī claimed that the Franks numbered only ‘about seven hundred knights and footmen, picked men’.20 The Egyptians’ first offensive on this campaign was an attack on Ramla and the nearby town of Lydda in May. They encamped before Ramla and devastated the surrounding countryside, destroying the crops that were ready for harvesting. The town was protected by a tower, garrisoned with just fifteen soldiers and defended only by them, their servants and Syrian peasants; nevertheless, they resisted the Egyptians, who therefore moved on to

Fighting the Saracens  135 nearby Lydda, which was where the bishop resided. Bartolf of Nangis described his residence, ‘the church of St George, built with wonderful workmanship, well defended by a wall and towers’, and wrote that the Ascalonites had often tried to destroy it but the bishop had always managed to protect it with his fortress. However, when the Egyptians arrived and he saw what sort of weapons they had with them and that they had settled on the plain to besiege the monastery, he summoned help from Baldwin, who was in Jaffa. Fulcher of Chartres, who was evidently in Jaffa at the time, wrote a first-hand description of what happened next. At first Baldwin thought the fire and destruction around Ramla was the work of some Egyptian scouts or outriders and he would be able to intercept them easily, so he mounted his horse and rode out with a few men. In a scene recorded only by Bartolf, he was warned by Stephen of Blois not to go, but he angrily replied ‘that even if Stephen and his allies were then in France, yet Baldwin would not let the enemy troops escape’. Guibert of Nogent told a similar story about the same incident, but Arpin of Bourges was the knight who urged caution. The king retorted: ‘If you are afraid, run off home to Bourges.’ Baldwin departed with those who were ready, and the 1101 survivors followed after with whatever knights there were in Jaffa at the time. Bartolf’s lament that follows is remarkable: Woe to that day that led out so many noble and honest men to ruin and death! That day, that bitter day, that hated day, both to be detested for all time and entirely erased from the memory of days of celebration! From the capture of the city of Jerusalem to this day there was not such disaster and misery in all the land of Israel, nor such a massacre of illustrious men.21 In a surprisingly modern simile, he likened Baldwin and his small band of men to ‘sulky kids’ (hœdi petulantes) who trusted in their own strength rather than the power of God. They even neglected to take the True Cross relic with them. The 200 faced more than 20,000 Saracens. Bartolf observed that they might just as well have thrown away their weapons and advanced on the foe with their arms tied behind their backs, they were so outclassed. Men and horses were cut down; at last, but too late, the king became fearful, and while some of his men were still alive he fled dishonourably to nearby Ramla, pursued to the very walls by the enemy. They took refuge in the town, but only the tower offered real protection, and it was too small to take even half of those who needed it.22 Albert of Aachen depicted this first stage of the campaign as an altogether more orderly foray.23 He put Baldwin’s army at 700 men, including the surviving members of the domus Godefridi, and rather than going off at half cock they moved off ‘with trumpets and bugles and purple banners’. As they descended onto the plain of Ramla they were met by an army of ‘infinite thousands’. It was too late to retreat, so instead they attacked with some slight success, penetrating the vanguard and killing some of the enemy before the Ethiopians attacked them and the Christians were either killed or took flight. Albert listed nine names of prominent crusaders among the many who were killed; he wrote that fifty escaped to Ramla

136  Fighting the Saracens and seven more named knights fled to Jaffa. As they arrived at the port they met the main army, numbered by Albert at 10,000, but hearing of the disaster the reinforcements fell back on the town. Both Albert and Fulcher/Bartolf appear to be well informed, but they present different impressions. Albert, probably informed by people who knew the survivors or the casualties, wished to show their heroism in the face of overwhelming odds; Fulcher knew that Baldwin had miscalculated and was anxious to stress that the disaster was the result of human shortcomings, not divine disfavour. The king had fortunately kept his horse, the famous Gazela, and to avoid becoming trapped in Ramla he made an escape at night, accompanied by very few companions.24 The enemy was somehow alerted to Baldwin’s escape and pursued him across country for three days until the king was able to take refuge in Arsuf. Fortunately the troop of Saracen soldiers who had been scouting there the previous day had gone away unsatisfied. While Fulcher’s account was of a heroic evasive flight through the mountains, Albert claimed that the king wandered rather aimlessly, prevented by the Saracens from taking refuge in the mountains. He eventually chanced on a road he recognised and struggled on to Arsuf despite minor arrow wounds sustained in his escape. By the time William of Tyre was writing, the story of Baldwin’s escape from Ramla had accrued the trappings of romance.25 These say more about Baldwin’s later reputation and the audience’s appetite for tales of chivalry than his real character, of course. The story started towards the end of 1101 when Baldwin took an expedition across the river Jordan and attacked an Arab encampment. He and his men chased away the men and then took women and children prisoner as well as carrying off rich spoils. One of the prisoners was the pregnant wife of an emir, and she went into labour. The king ordered that she and her new baby were to be left with food, water, a maid and two camels (for their milk). He even wrapped her in his own cloak before going on his way. Her husband was following close behind, grieving the loss of his wife, and when he found her he was overwhelmed by the king’s kindness and swore he would reciprocate when he could. The opportunity came when the king was trapped in Ramla and the emir was part of the besieging army. He slipped away, went up to the walls and persuaded the guards to admit him to the king. He warned Baldwin that the enemy was preparing to attack Ramla and personally conducted him to a safe place from where he was able to make his way to Arsuf. Casualties were very high during this first stage of the engagement. The Saracen army attacked Ramla the following day, undermining the tower where Conrad the constable, Arpin of Bourges, Stephen of Blois, Stephen of Burgundy and other knights had taken refuge. The following day the knights were smoked out, and they made a valiant stand against the enemy. But inevitably they were defeated. Conrad fought so fiercely that the Saracens kept him alive, and Arpin was able to convince them he had been the Greek emperor’s knight, and so he was also saved.26 Both of the Stephens and all the rest of the knights were killed by beheading.27 Only three out of all the knights whom the king had led out of Jaffa survived: the king’s squire; Lithard of Cambrai; and the ‘viscount of Jaffa’, all of them gravely wounded. They carried the news of the defeat to Jerusalem,

Fighting the Saracens  137 where they also reported that they did not know whether the king was alive or dead, causing an outpouring of grief. Meanwhile, Bartolf concluded, the king was in Arsuf, making up for his loss of sleep and food. Albert added another name to those who escaped Ramla: he wrote that a certain Gutmann of Brussels also arrived in Jerusalem, but rather than spreading doom and gloom he kept telling the Jerusalemites that they should not give up until they heard if King Baldwin was still alive.28 At this point in the narrative there is a significant divergence between the two main sources, although their accounts are not irreconcilable. Whereas Fulcher simply reported that the king sailed from Arsuf to Jaffa, Albert of Aachen gave a great deal more detail. Jaffa was then under siege, and the Saracens were doing their best, or worst, to spread despondency. They had the body of Gerbod of Windeke, and they cut off his head and legs and clad them in royal purple garments and displayed them to Jaffa’s defenders as the king’s remains. Evidently Gerbod bore an impressive resemblance to Baldwin, and the Egyptian ruse caused despair in the town; even the queen was considering flight. It was, though, another week before they saw the king himself. Baldwin took ship, along with a few men and also the ship’s owner or captain Godric, whom Albert called a pirate. Baldwin affixed his banner to a spear and raised it high to signal his survival to the inhabitants of Jaffa so they would not escape or surrender. Unfortunately, the Saracens also recognised his sign and were waiting for him outside Jaffa with a fleet of thirty-three warships. However, the waves were against them while the king’s vessel was running easily, and he made it into the harbour at Jaffa. He even managed to shoot six of the Saracens with his bow as his ship passed them. Of course his arrival caused great rejoicing in Jaffa, and he compounded this by immediately parading outside the city gates with half a dozen knights, encouraging the withdrawal of the enemy army. The Egyptian army duly retreated to the plains of Ascalon for the next three weeks, but when it became clear that no reinforcements were going to arrive they returned and besieged Jaffa again.29 During this renewed siege another fleet arrived from northern Europe, carrying pilgrims. Albert claimed there were 200 of these ships, although Ibn al-Qalānisī said about 40.30 The Saracens had maintained a naval blockade and prepared to attack the approaching ships, but they ran before the wind and made it into harbour. The pilgrims disembarked and joined the king and the inhabitants of Jaffa, some of them making an encampment next to the enemy. The Egyptians withdrew about a mile and took stock of the situation, deciding to renew their attacks on Jaffa. After three days the king came out of the city in strength to do battle, his cavalry and infantry augmented by the sea-borne reinforcements. They forced the Egyptian army to retreat. During the pursuit some of the enemy tried to escape by swimming and were drowned in the sea. In all, Albert claimed 3,000 Saracens were killed but few Christians. Albert’s perspective as an outsider and a northern European should not be undervalued, as he was probably informed by one or more of the pilgrims after they reached home. Not all of them made it: they set off in calm weather, but it was around the autumn equinox, and they were caught in winter storms that wrecked

138  Fighting the Saracens many of their ships when they were just two days out of port. Shipwrecked survivors fetched up in Acre, Sidon and Ascalon; these towns were in enemy hands, and the survivors were captured or killed. Many drowned; 300 ships were lost, and only some tenth of the fleet that set out was salvaged.31 Nevertheless, even if the survivors’ tales were embellished, they add an important dimension to our knowledge of the second battle of Ramla, and fortunately Fulcher’s privileged position, probably in Jerusalem at the time, provided a complementary land-based view of the battle. In May 1102, while Baldwin was still resting in Arsuf, Hugh of Tiberias arrived with eighty knights to defend the Christian cause. There was also a message from Jerusalem, perhaps carried by Hugh, that the people of Jaffa urgently needed help. The king went to Jaffa by boat, but Hugh and his men marched by land. Once the king had been joyfully welcomed in Jaffa he set about summoning reinforcements from Jerusalem and Hebron. As a messenger he recruited a Syrian Christian who was able to evade capture by the enemy and reach Jerusalem. It took him three days. Fulcher recalled that the number that could be mustered in Jerusalem was ninety, about equal numbers of knights and of other men who could obtain horses or mules.32 Avoiding the direct route, they reached Arsuf, but between there and Jaffa, as they marched along the coastal road, they were attacked. The knights were able to out-ride the pursuit on their faster horses, but some of the other men had to abandon their re-purposed pack animals and take to the water to save themselves, although their mounts were lost. The king, apparently, did not want to wait for the Egyptian armies to blockade Jaffa again and decided to attack them as they prepared their siege equipment on the plain about three miles from Jaffa. Cavalry and infantry advanced and were immediately surrounded by the far more numerous enemy. This meant that whenever the knights penetrated enemy ranks they exposed the infantry in the rear to attack. However, these footsoldiers were armed with bows and inflicted serious damage with arrows, just as the knights were doing with their lances. The enemy fled, but it was not possible to mount an effective pursuit because there were so few mounted men among the Christians. Instead, they plundered the enemy camp, finding provisions and camels and donkeys, but not the horses they badly needed, because the Egyptians had taken them with them. The key to the victory of the few over the many, Fulcher was quite clear, was that this time – unlike the first encounter that year – the Christians were fighting behind the relic of the True Cross. In Bartolf’s version of the same battle it was given even more prominence.33 There are a few significant differences in his account. The reinforcements from Jerusalem numbered 90 knights in armour and some 200 others, as many as could bear arms. When the king led out his army the following morning, the True Cross went before them. Once the Saracens realised they were under attack they took up their weapons and surrounded the Christians. In Bartolf’s account of the battle Baldwin’s army maintained strict discipline and there was no breaking of their ranks as in Fulcher’s later version; they launched a hail of arrows and then advanced cautiously behind it with drawn swords, brandishing lances, and ‘by the power of God acting through the aid of the Holy Cross’

Fighting the Saracens  139 they succeeded in forcing the enemy to flee. Bartolf described the rout vividly: the Saracens spared neither reins nor spurs and did not give the infantry a second thought. Many men on horse or foot died of heat or thirst. The Christians were unable to pursue, but they revenged themselves on any Saracens who were lagging behind. Bartolf included a much longer list of the spoils the Christians seized from the enemy camp, claiming that rich men returned to Jaffa who had left it as poor men. The whole victory was ascribed to the power of the Cross. The final version of Fulcher’s chronicle made much of the power of the Cross and the king’s failure to carry it in the earlier encounter, ‘because if that same life-giving Cross had been carried with the king in that previous battle, there is no doubt that the Lord would have favoured His people’.34 No such direct criticism is to be found in Bartolf of Nangis’s account of the battle, which suggests that Fulcher was more circumspect in his remarks while the king was alive. He concluded the chapter with the return of the king to Jaffa and observed that the land was free of war during the following autumn and winter. Albert of Aachen tied off more loose ends. After the king spent the night celebrating in Jaffa with the pilgrims, who were now in possession of a lot of plunder, he took them up to Jerusalem and ordered the church of the Holy Sepulchre to be opened so that they could fulfil their vows. We also learn that before the sea-borne reinforcements arrived Baldwin had sent urgent messages to Antioch and Edessa to ask for the rulers’ assistance. Tancred and Baldwin of Edessa gathered an army of some 500 cavalry and 1,000 infantry. They marched south and finally encamped about a mile from Jaffa. It was September by now, so the Egyptian threat had long since receded, but Baldwin took advantage of their presence to besiege Ascalon. The siege lasted only a week, during which they lay waste the crops and vines outside the walls. Finally an emir led a sortie against the Christians, in the course of which he was killed and the Ascalonites fell back on the city and barred the gates. Baldwin took advice from his nobles and withdrew. Perhaps because the siege was so short and unproductive, Fulcher either forgot about it or decided to omit it.35 However, as with the whole of Baldwin’s second campaign around Ramla, it demonstrated a combination of opportunism and realism that was to stand him in good stead. Tactically, in the absence of a large cavalry he had discovered the potential of an infantry, even one largely untrained, armed with bows and arrows. He had also had a salutary reminder of the power of the True Cross as a rallying point.

A Saracen ambush, 1103 In the period between the second and third major invasions from Egypt, there were many smaller engagements. One, reported by Fulcher of Chartres and at much greater length by Albert of Aachen, occurred between the unsuccessful and successful sieges of Acre. Fulcher merely reported that Baldwin was fighting the Saracens ‘as usual’ when an Ethiopian hiding behind a rock threw a missile that hit the king and wounded him deeply near the heart. Baldwin was close to death but was careful to get medical treatment, and after an incision of the painful scar he was restored to health. Bartolf added that his successful care took a long time,

140  Fighting the Saracens and Matthew of Edessa asserted that Baldwin’s wound was incurable.36 Albert’s much more circumstantial account is that Baldwin and ten of his knights were hunting near Caesarea when about sixty Saracens from Ascalon and Acre set out with a general intention to ambush Christians. A rumour spread that this was a much larger force than it was, and Baldwin, hearing this, instructed his knights to attack the Saracens, even though they were equipped for hunting and not for battle. He took the lead: King Baldwin, rushing more keenly than anyone into the midst of the enemy and inflicting more slaughter with his sword, through the swift pace of his speeding horse unexpectedly found himself next to some bushes of low woodland, where he was pierced through thigh and kidneys by the furtive lance of a Saracen who was lurking among the branches and thick leaves. At once streams of blood gushed ominously from this cruel wound of this powerful king, his face began to grow pale, his spirit and strength to falter, his hand to cease from fighting with his sword, until at length he fell from his horse to the ground as if dead and destroyed, and he was believed to have expired. . . . They placed him on a stretcher and took him back to Jerusalem amidst a very great weeping and lamentation of men and women, acquiring very experienced doctors for him, by whose skill and experience their king and strong champion could recover his health after this lethal wound.37 It is probable that this wound troubled Baldwin for the rest of his life, and Fulcher, at least, thought that it contributed to his death fifteen years later. The Egyptian response to news of the near-fatal wounding was to send both navy and army to join up with the Ascalonites and attack Jaffa. The Egyptian army came by land and sea, arriving at Jaffa in summer 1103, but after a few days both army and navy dispersed along the coast, perhaps with the purpose of provisioning the coastal cities that were still in Muslim hands. Albert of Aachen described a particular incident in detail.38 There were daily skirmishes between the invading forces and the citizens of Jaffa, and during the siege a large ship, over-burdened with more than 500 male pilgrims plus an unspecified number of women aboard, was shipwrecked in Jaffa’s harbour. The Saracens saw this happen and moved in quickly to attack the pilgrims and plunder their possessions, but the Christians in Jaffa had seen the wreck too and managed to drive the enemy away. A second, smaller ship that had been sailing with the larger one ploughed into the Egyptian fleet at night. The captain and seven crew abandoned ship in a tiny boat, leaving 150 men and some women, plus seven knights who were on board with their horses, to fight off Saracen attacks. In the end all the pilgrims, including the knights and all the women, were captured and killed, except for one squire who managed to swim ashore. King Baldwin, by now somewhat recovered from his wound, responded to news of the siege around Jaffa by going there himself. His very presence, which showed that he was still alive, was enough to persuade the Saracens to withdraw. It was now October (1103), and the seas were stormy, which may also have been reason enough to lift the naval siege. Ships from Jaffa

Fighting the Saracens  141 pursued the Egyptian fleet, but without success. In spite of this, there was considerable relief among the Christians, and both they and the Ascalonites settled down to a peaceful winter cultivating their crops.

The third battle of Ramla, 1105 The Egyptians had learned lessons from the 1102 campaign, and when they launched their next major invasion in 1105 they planned to besiege Jaffa simultaneously by land and sea. When King Baldwin heard, in August 1105, that the Egyptian army was at Ascalon, he gathered his forces around Jaffa. They included just about every able-bodied man: ‘Since this necessity demanded it, all who lived in the cities and were able to bear arms went into battle, excepting only those who guarded the wall at night.’39 For a while both sides waited; when the Saracens were seen to be leaving Ascalon and approaching, the king moved to Ramla and sent a messenger to Jerusalem to call up reinforcements from there. On arrival in the Christian camp the patriarch and priest received confessions and granted absolution, and then the army advanced, led by the Holy Cross, against the Saracen army encamped about four miles from Ramla. Fulcher claimed that the advance took the enemy by surprise because the Saracens’ plan had been to send a contingent against Ramla as a diversion while the main army attacked Jaffa – although he did not reveal how he came by this knowledge. The two armies joined battle. Fulcher described the Turkish mounted archers shooting their arrows and wheeling away, in a now familiar tactic, before engaging in hand-to-hand combat. The king, seeing this, seized his banner and charged in, scattering the Turks before turning his attention to the rest of the army. After telling this heroic deed, Fulcher decided to précis the rest and give the credit to God for the enemy’s retreat that followed. Fulcher reported it as a great victory for the Christians, claiming that 4,000 of the enemy were killed, including all the Ethiopians, who were on foot and could not escape, but that only sixty Christians were slain. This is at variance with the Arabic accounts that declared it a draw, with the killing about equal on both sides. However, the Egyptians had abandoned their camp, so there was considerable plunder, which was divided among the soldiers ‘according to a formula’. An emir from Acre was captured alive and was ransomed for a considerable price in gold and other things, including horses. As a postscript to the battle, Fulcher mentioned the Egyptian naval blockade of Jaffa. King Baldwin persuaded the enemy to leave by having his sailors hurl the head of Gemelmulc, emir of Ascalon, who had been decapitated in battle, onto the deck of one of the ships. This, naturally, caused alarm and despondency, and the fleet sailed off to Tyre and Sidon. Later, when the fleet was sailing home to Egypt, a storm scattered it, and the Christians captured twenty-five of the ships with their crews.40 Bartolf of Nangis followed the same narrative outline, of course, but there are some notable differences in the details he included.41 He started with the ‘king of Egypt spewing his poison against the Christians’ and sending ‘Semelimech’ at the head of the army with instructions not to return until he had obliterated the Christians from the land. Significantly, Bartolf added that the Egyptians decided on

142  Fighting the Saracens invasion then ‘because the land was empty of pilgrims and no one could come to Jerusalem from the regions across the mountains because of the quarrel between the emperor and Bohemond’. On Baldwin’s side, Bartolf claimed that he was unafraid ‘because he had so often experienced their weakness and feeble forces’. When he learned how many different components made up the Egyptians’ army, Baldwin sought to swell the ranks of his own army in the same way and sent to all corners of the kingdom. The enemy had expected flight rather than fight from the Christians, so the Egyptians hurried to occupy the plains ahead of them when they saw the Christians approaching, and when the Christians pitched camp the Egyptians spread out around them and cut off any prospect of retreat or escape. Then they waited. At this point the king sent to Jerusalem again for assistance, which indicates that the patriarch had been slow to respond to Baldwin’s first request. The messenger risked his life to get this summons through. Unlike Fulcher, who said that the Egyptian plan to attack Ramla and Jaffa simultaneously was aborted, Bartolf wrote that they had already sent a part of the army to Jaffa and a part to Ramla. Because Bartolf was using a manuscript of the redaction of Fulcher’s chronicle written no more than a year or so later, it may well be that his is the truer account and that Baldwin routed a reduced Egyptian army; perhaps Fulcher later changed the story to make the king’s triumph the greater. In the earlier account the patriarch’s arrival was the signal immediately to sound the trumpets and advance on the enemy. In another variation that has the ring of truth, Bartolf described the Turkish archers shooting from the Christian army’s left, that is, the side they bore their shields, which reduced the effectiveness of the attack, and when they were still preparing for close combat the king arrived and put them to flight. He then turned his attention to the Saracens and Ethiopians and attacked them ‘like a lion gnashing at cattle’. Thus ‘through the banner of the Holy Cross he drove them all into shameful flight’. The Egyptian cavalry got away, but the Saracens and Ethiopians were killed or taken captive. The killing of Gemelmulc is accompanied by an authentic-seeming exclamation: ‘Oh, if only he had been taken alive, how many coins he would have given as ransom!’ The Egyptians on horseback took different escape routes. Many of those who fled into the mountains, rather than making for Ascalon or more directly to Egypt, were forced to abandon their exhausted horses, and these were appropriated by the Franks. Bartolf also had a longer list of the riches found in the deserted enemy camp; the king’s men returned to Jaffa ‘stuffed and loaded’ with the spoils.42 Finally, Bartolf also turned his attention to the fleet that, he wrote, had assembled from Cairo, Alexandria, Damietta and Tripoli. Both warships and cargo vessels had been pressed into service and were laden with armed men, the idea being that once the Christian army was defeated they would be used to besiege and capture the ports. ‘But’, Bartolf wrote, ‘once they had seen Gemelmulc’s head that Baldwin had them shown from the walls of Jaffa, their hopes were dashed and they were completely despairing, and they took to flight.’ They sailed to Tripoli and Alexandria, according to this account, and thence to Cairo. Bartolf presciently called this the ‘last of the battles’ before he signed off his history.43 For the reason indicated previously, namely the closeness in date of the battle to the end of

Fighting the Saracens  143 Bartolf’s exemplar manuscript, plus his disinclination elsewhere to elaborate his source, the points of difference in this version of Fulcher from the author’s final redaction are to be respected. Albert began in Jaffa, where the king saw the Egyptian fleet arrive to blockade the port on the sea side and realised very quickly that the Egyptians’ strategy was to preoccupy the Franks with defence on that side while bringing up an army to attack from the land side.44 Baldwin immediately began to summon support, and Albert listed nine men by name; all of them were originally from Flanders or nearby and were vassals or close allies.45 They came with cavalry and infantry. Albert was also the only Latin historian to attempt to explain the presence of Turkish troops on both sides of the conflict in 1105: Also in the king’s company was a certain Turkish youth, an active soldier, Mahumeth by name, whose strength in weapons and number was a hundred Turkish archers. He had been driven out of his father’s home and from the land of Damascus because of his greed and diligence for victory for himself and he had now struck an agreement with the king that he would be loyal and prompt in all military assistance to him, and for this he would be rewarded subsequently by the king’s help against Damascus.46 The Arabic chronicler Ibn al-Athīr explained further: Several Muslims had been with the Franks, including Baktāsh ibn Tutush. Ṭughtigin had transferred sovereign rule to his nephew, Duqāq’s son, still a young child. . . . This is what motivated Baktāsh to go to the Franks and join them.47 The same author had earlier given more details of the succession dispute that resulted in Baktāsh’s leaving Damascus towards the end of 1104. He made common cause with the lord of Buṣrā, and they contacted King Baldwin to ask for military support. He went to meet them, and they concluded an agreement. It did not result in immediate action against Damascus, as Baktāsh had hoped, but evidently Baldwin renewed his promise of assistance when he needed help himself the following year. The number of troops involved seems to have been small and was outweighed by the number of Turkish mounted archers Ṭughtigin had supplied to the Egyptian armies in retaliation, but nevertheless the agreement itself was significant. His readiness to treat with the infidel illustrates Baldwin’s pragmatism in negotiating his way around the politics of his world and exploiting not only the differences between the blocs of Saracen Muslims and Turkish Muslims but also any rifts within their politics. Albert’s account of the battle largely corroborates Fulcher’s, although there are added details.48 When he left Jaffa the king had with him 6,000 men, and he left a garrison of 300 in the city under the command of Lithard of Cambrai. Once the patriarch had brought up troops from Jerusalem – Albert said only 150 infantry – the army was organised into five divisions, with Baldwin in the rearguard with

144  Fighting the Saracens only 160 knights, because of the severe shortage of horses. The two armies advanced towards one another, the 40,000 on the Egyptian side far outnumbering Baldwin’s army. They fought from early morning until noon, and then the Saracens fled, pursued by the Christians. Albert reported that 7,000 gentiles fell in the battle, including the emir of Ascalon, while the emirs of Acre and Arsuf, who had taken refuge in Ascalon after their cities were captured, were taken prisoner. Albert claimed that only 100 of Baldwin’s men were killed. Meanwhile the navy was waiting outside Jaffa for the news of a Saracen victory that would be the signal to attack, but ‘when they saw the head of the executed emir and chief of Ascalon’ they sailed away. They put in at Tripoli for one night only. William Jordan, who had succeeded Raymond of Saint-Gilles and aspired to be future lord of the city, tried to attack the fleet, but his weapons were unable to do any damage, and the ships sailed on via Ascalon back to Egypt.

Ambushes and raids After the third battle of Ramla, Albert wrote, ‘the king’s land was quiet and a great terror struck the hearts of the Ascalonites and the Egyptians, since so often they had been beaten by the king with an army of few and they had fallen or fled’. More realistically, Baldwin’s record against the Egyptians was two narrow wins and a draw, but this was enough to ensure some equilibrium in their future relations. However, as Albert also observed, there could be no permanent peace so long as Egypt held Ascalon, and the following spring, at the very end of April 1106, Baldwin marched on Ascalon and devastated the surrounding fields, setting fire to crops and trees.49 In October or November of the same year, the Ascalonites rallied support from Tyre, Sidon and Beirut for another attack on Ramla and Jaffa.50 According to Albert of Aachen they mustered an army of 7,000 cavalry and attacked a crowd of pilgrims between Arsuf and Ramla, killing 500. They then set out for Ramla, which was sparsely defended; its total garrison of eight knights was despatched to Jaffa and carried a message that enabled the army there to arm and defend the city. Rather than meet them in the field, however, the Ascalonites turned to ambush tactics, as reported by both Albert and Fulcher. The latter, who reported that the Ascalonites had some 500 horse and 1,000 footsoldiers waiting to entrap just 65 Christians, described a brief skirmish that ended with the Saracens in flight and only 3 Christians killed. Albert wrote that after this the rest of the Saracen troops emerged from the mountains and attacked. As the Christians fought back a knight called Gerard advised retreating within the defences of Jaffa, undermining the Franks’ confidence and courage so that they fled into the city. Notably, Albert said that the absence of the king was a factor in this. The Ascalonites pursued the Franks closely and caught and killed many outside the gates. Some forty were beheaded. The Ascalonites then attacked a castle (now called Yalu) west of Jerusalem and took Gunfrid of the Tower of David captive.51 The Ascalonites also launched eight galleys to attack ships in port at Jaffa. They failed to capture a big dromon, but did take a smaller, richly laden vessel. Only then did King Baldwin,

Fighting the Saracens  145 who was in Tiberias, hear what had happened. He returned to Jaffa with the patriarch, 500 cavalry and 6,000 infantry, but after consultation Baldwin decided to withdraw and wait for another occasion. Because we have only the two Latin sources, it is impossible to know just how threatening the Ascalonite campaign had been: Fulcher implied a raiding party, Albert a more concerted attack. The absence of the event from Arabic chronicles suggests a minor event. The next reported attack from Egypt was a naval one in August 1110, described fully only by Albert of Aachen.52 He reported that a great fleet of warships sailed for Beirut to try to retake the town but that during the blockade they saw three vessels approaching carrying pilgrims from Flanders and Antwerp, as well as a fourth, a Greek merchant vessel. They attacked the Christian ships, and the three pilgrim vessels were able to escape; one made it to Haifa, and the rest were wrecked in shallow waters between Haifa and Acre where the passengers could be rescued. The Greek ship was captured and plundered. The Egyptian fleet then sailed on to threaten Acre, and from there they withdrew to Tyre. The arrival of the Norwegian king and his ships to blockade Sidon persuaded them to sail back to Egypt. At the same time the Ascalonites, realising King Baldwin was fully engaged fighting the Turks and had left Jerusalem’s defences weakened, decided it was a good time to attack the city. The inhabitants summoned troops from the towns of the kingdom and were able to mount an effective defence. Albert particularly mentioned the participation of women and clerics. Three hundred soldiers, on horse and foot, went out to intercept the Ascalonites and succeeded in putting them to flight and bringing back horses, plunder and prisoners. A grand alliance of Muslim forces, both Saracens and Turks, would surely have destroyed the infant kingdom of Jerusalem, but fortunately for Baldwin this was out of the question at the time. What did happen was that the Ascalonites, sometimes with the support of the Egyptians, would take advantage of Baldwin’s preoccupation with the conquest of the maritime cities or with Turkish incursions in the north to attack the cities nearer to them when they were left comparatively defenceless. This happened again in 1115 when Baldwin had responded to an appeal to assist Antioch and was campaigning in the north. The Ascalonites besieged Jaffa by land while the Egyptians blockaded it by sea with a fleet of warships and supply vessels. An initial escalade was repulsed by the citizens, so the attackers set fire to the city gates and withdrew to Ascalon, while the ships sailed to Tyre. Ten days after their retreat the Ascalonites returned and tried an assault with hurling machines and another escalade. Once again the Franks mounted an effective defence, and the Ascalonites retreated. When the king was gravely ill in March 1117 the Egyptians set out again to launch a naval attack, but aborted the campaign when they received news of his recovery.53 The superhuman efforts and divine assistance that had been necessary in the first years of Baldwin’s reign to resist sustained and repeated attacks from Egypt were successful, therefore, and subsequent incursions were sporadic, opportunistic and smaller in scale. The defence of the kingdom would have been more straightforward if Baldwin had succeeded in capturing Ascalon; nevertheless, he

146  Fighting the Saracens established a formidable reputation and was able to rely on a loyal following to protect the kingdom from invasion from the south while he was engaged in fighting the Turks elsewhere.

Notes 1 The Fāṭimids of Egypt were usually called ‘Saracens’ in the sources, although writers also used the same word for natives of Syria and Palestine (not all of them Muslims). The word most often used for the Egyptian caliphate was ‘Babylon’, by which name they also referred to the city of Cairo. ‘Saracens’ were clearly distinguished from Turks in the early period under discussion, although later the name came to be used of Muslims (and other enemies of Christendom) indiscriminately. For a full discussion, see Nicholas Morton, Encountering Islam on the First Crusade (Cambridge, 2016), pp.  15–19, but note that Morton preferred to avoid the term ‘Saracen’ altogether, as too value-laden. For the battle of Ascalon and the battles of Ramla, see the analysis by Michael Brett, ‘The Battles of Ramla (1099–1105)’, in Egypt and Syria in the Fatimid, Ayyubid and Mamluk Eras, ed. U. Vermeulen and D. de Smet (Leuven, 1995), pp. 17–37. Brett included English translations of four Arabic accounts for comparison with the Latin ones. 2 Brett, ‘Battles of Ramla’, p. 18. 3 BN, chap. 51, p. 527; cf. FC 2.10, pp. 404–407. See also AA 7.57, pp. 566–68. 4 FC 2.11.2, p. 409. FC wrote that the king ordered every knight who could to make his squire a knight, and the number 260 was only thus achieved (not in BN). 5 AA 7.64, p.  576. Frutolf’s continuator recorded the presence of 7,000 infantry and 1,000 cavalry on the Christian side, numbers that seem wildly exaggerated, but McCarthy conjectured that he had an eye-witness account of the battle from Gerard of Schaffhausen, see F-E, p.  174; McCarthy, intro. to Chronicles of the Investiture Contest, p. 52. 6 See Brett, ‘Battles of Ramla’, pp. 31–37, for translations of the Arabic sources. 7 FC 2.11–13, pp. 407–20; BN chaps. 51–54, pp. 527–31; AA 7.65, pp. 576–84. 8 BN chaps. 51–52, pp. 528–29. 9 AA 7.66–68, pp. 578–80. Hugh of Tiberias, formerly of Fauquemberges, had accompanied Baldwin from Edessa to Jerusalem in 1100 and was sent by him to secure the citadel; this and his later history confirm that he had probably been with Baldwin when he acquired Edessa: see Murray, Crusader Kingdom, pp. 211–12. Bishop Baldwin was the recently appointed archbishop of Casarea. The identity of the cleric who bore the True Cross into battle is not entirely certain. Albert of Aachen called him Bishop Gerard, but for Fulcher he was an anonymous abbot. He was probably Gerard of Schaffhausen, treasurer of the Holy Sepulchre (‘Monachi anonymi Scaphusensis’, RHC Occ, V: 335–39; Annalista Saxo, MGH SS, VI: 736), but Hamilton suggested he was the abbot-archbishop of Tabor, Gerald (p. 61, n. 1; p. 130). Some light is cast by the 1106 continuator of Frutolf of Michelsberg’s chronicle who was a participant in the Bavarian expedition to the Holy Land in 1101 and drew on an eye-witness account of the first battle of Ramla. He numbered the Egyptian army at 40,000. It made for Jaffa first and then took up position not far from Ascalon. ‘Baldwin was not unaware of the matter and summoned his men from everywhere, that is from Jerusalem, Nicopolis, Mount Tabor, Hebron, Caesarea and Arsuf, to Jaffa where a considerable crowd of pilgrims was staying at the time.’ This source alone claimed the long pre-battle sermon was preached by ‘a certain Arnold, a venerable and well educated churchman’ (Arnulf of Chocques’s presence is not corroborated by other sources). The same writer said of the Cross: The venerable abbot Gerhard, who at that time always carried the Lord’s cross right by the king’s side, reported that he had never seen so great a density of snow or rain

Fighting the Saracens  147 as there was then of missiles let fly against the king; but after looking upon the precious Rood none of the enemy engaged with missiles or weapons, but all of them sought protection in flight. (F-E, pp. 172–76) 10 The nature of the Egyptian siege of Jaffa was described by Frutolf’s 1106 continuator: many horsemen blockaded the city from the land, and forty-two war galleys surrounded it from the sea. The inhabitants’ misery and hunger were relieved by the arrival of thirty ships carrying some 12,000 Christian pilgrims as well as plenty of grain and other foods. They too benefited from the power of the Holy Cross and broke the blockade (H-E, p. 176). 11 AA 7.70, pp. 582–84. 12 FC 2.13–14, pp. 417–24; BN chap. 53, pp. 529–30. 13 BN chap. 54, pp. 530–31. One effect of the letter, according to Mayer, was to undermine Baldwin’s trust in his Armenian wife: ‘Études’, pp. 55–56. 14 Alan V. Murray, ‘ “Mighty Against the Enemies of Christ”: The Relic of the True Cross in the Armies of the Kingdom of Jerusalem’, in The Crusades and Their Sources, ed. France and Zajac, pp. 217–38. 15 See references in chap. 5 above. 16 Nevertheless, Alexios entertained the crusaders lavishly: see Beihammer, Byzantium and the Emergence, p. 331. 17 AA 8.45, p. 634. 18 AA 8.46–48, pp. 634–36. This may be contrasted with ME and the anonymous Syriac author, who accused Alexios of sending men to lead the crusaders ‘into a desert where there was neither water nor fodder and he told the Turks of those parts so that they could surround them’ (ME 3.6, p. 186; AS, p. 74). The fullest analysis is Bunna EbelsHoving, Byzantium in westerse ogen (Assen, 1971), pp. 84–88. 19 AA 9.2, pp. 640–42; FC 2.15, pp. 425–27; BN chap. 55, p. 531. Fulcher claimed ‘there were all at once 20,000 horsemen and 10,000 foot, not counting the keepers of the pack-animals who drove camels and donkeys laden with provisions and carried cudgels and missiles in their hands for fighting’. As previously, BN’s figures were wild: ‘twenty times a hundred thousand horse [2,000,000] and thirty thousand foot’, plus attendants and servants (he may have found Roman numerals challenging). There are some difficulties in reconciling the dates in different Arabic sources, but see Brett, ‘The Battles of Ramla’, for an examination of these. 20 IQ, p. 55, entry for ah 495; Brett’s translation is less ambiguous: ‘700 chosen horse and foot’ (p. 33). 21 BN chap. 58, p. 533. 22 The account is reconstructed by comparing BN’s early version (chap. 58, pp. 533–35) with the final (FC 2.18, pp. 436–40). Two decades later, FC toned down his account considerably, permitting himself only to observe that it was very rash of the king not to wait for his men and that he should have known better. He described the king as rallying his men and attacking before most were killed and the rest fled. See also GN 7.24, p. 316, although Guibert knew little more, professing ignorance even of the fate of Stephen of Blois. 23 AA 9.3–4, pp. 640–42. 24 Gazela’s name is found in AA 7.67, p. 578 and 9.5, p. 642; BN chap. 58, p. 534; OV 10.22, V: 344–47. According to BN Baldwin was accompanied only by his squire, and three knights who followed soon after him were intercepted and killed; FC wrote that he set out with five companions but they were stopped by the enemy; the English historian William of Malmesbury also said five companions, of whom all but one were captured; AA claimed he took his squire and a knight called Hugh of Brulis; WT implied that he escaped alone with his horse, but he chanced to meet two comrades who then accompanied him. AA 9.5–6, pp.  642–44; FC 2.19–210 pp.  441–45; BN chap. 58,

148  Fighting the Saracens

25 26

27

28

29

30 31 32

33 34 35

pp. 533–35; WT 10.21(22), p. 479. WM claimed one of the captured knights was an Englishman, Robert Godwineson, who died a martyr: 3.251.2–3, p.  466; 4.384.4, p. 684. For WT’s romantic tale: 10.10(11) and 10.20–21(21–22), pp. 464–65, 478–79. AA 9.5–7, pp. 644–46; BN chap. 58, p. 534. Both Conrad and Arpin were freed from captivity in Egypt after intervention by the Byzantine emperor; for his motives, see Beihammer, Byzantium and the Emergence, p. 330. Alexios sent Conrad to Henry IV, the Holy Roman Emperor, as a gift in 1106 (AA 10.39, p. 754). Arpin returned home and became a monk (GN 7.24, p. 316; OV 10.23, V: 350–53). Fulcher wrote of the deaths of the two Stephens: ‘Alas! What noble and valiant knights we lost at that time, both in the previous battle and afterwards in the said tower. For Stephen of Blois was killed, an experienced and noble man, and the other Stephen, the count of Burgundy’ (FC 19.4, p.  443). William of Tyre transferred their deaths to the battle before the flight into Ramla and used the occasion to praise the manner of Stephen of Blois’s death because it allowed him ‘to efface by a very good end the memory of the old stigma of disgrace that he had wretchedly incurred by fleeing from the Antiochene campaign’ (WT 10.19(20), p. 477). However, BN’s account corroborates AA’s; he wrote that the two Stephens lost their heads in their last stand after the destruction of the tower of Ramla, and Hugh of Lusignan, Geoffrey of Vendôme and many others were also slain: Hugh VI of Lusignan was half-brother of Raymond of Saint-Gilles (AA 9.1, p. 638; FC 18.4, p. 438 and widely in other sources). Geoffrey of Vendôme was probably the same as AA’s ‘Geoffrey the short’ (9.4, pp. 640–42); see also FC 18.4, p. 437. For Lithard see Murray, Crusader Kingdom, p. 216. As Lithard commanded the garrison in Jaffa in 1105 he was probably also BN’s vicecomes Joppitarum and the writer made a mistake with his third fugitive. Gutmann had arrived with the 1101 crusaders (AA 9.1, p. 638) and survived at least until 1115, as attested in early charters: see Murray, Crusader Kingdom, p. 205. AA 9.8–10, pp.  646–48. Gerbod was another 1101 crusader, see Murray, Crusader Kingdom, pp. 200–201. Mayer, ‘Études’, p. 56, interpreted the queen’s preparation for flight as another betrayal of Baldwin, but she had every reason, as before, to believe the king was dead. FC (2.21.2, p. 448), called the king’s vessel a cumba; BN (chap. 59, p. 535) called it a scapha, both Latin words meaning a light skiff; for WT it was a navigium (10.21(22), p. 480), but Albert called it a ‘buss’, a type of merchant vessel common in the North Sea and English Channel that Albert, a Rhinelander, would have been familiar with. The boundary between pirates and merchants seems to have been quite permeable. Godric may have been (the future) St Godric of Finchale, who is known to have been captain of a trading vessel and to have made a pilgrimage to Jerusalem in 1101, but this claim has been disputed. See Reginald of Durham, Vita S. Godrici (Surtees Society, 1838), pp. 33–34; Christopher Tyerman, England and the Crusades (Chicago, 1988), pp. 26–27. AA 9.11–12, pp. 648–52; IQ, p. 56. AA wrote that 140,000 took ship in 330 vessels (i.e. about 425 in each): 9.18, pp. 658–60. The battle occupies FC 2.21, pp. 446–55. Realistically, we have no idea of the size of either army. If there were recently arrived pilgrims among the infantry then Baldwin’s army may have been much larger than the few hundred FC seems to suggest, and he may have exaggerated the smallness of the Christian forces to emphasise the role played by faith and the presence of the True Cross. BN chaps. 59 and 60, pp. 535–36. FC 2.21.14, pp. 453–54. AA 9.13–15, pp. 652–56. BN said the land was at peace from invasions for the next three years (chap. 60, p.  536). This was correct if he was looking towards the next major invasion from Egypt.

Fighting the Saracens  149 36 FC 2.24, pp. 460–61; BN chap. 61, p. 536; ME wrote that the wound was inflicted by an Ethiopian Muslim hiding in some bushes and that it was in Baldwin’s ribs (3.12, p. 191). 37 AA 9.22, pp. 664–66. There is a diverting addition to the information about the treatment of the wound in Guibert of Nogent’s Dei gesta (GN 7.13, pp. 287–88). According to this, the doctor (medicus) who was summoned to treat the king was concerned that if the wound were covered it might reduce scarring, but at the same time the wound would continue to fester beneath the surface. He decided to conduct an experiment and proposed to have a Saracen prisoner wounded in the same way and then have him killed so that he could dissect and examine him. To his credit, Baldwin refused to be responsible for the death of another human, even a Saracen, and so the doctor had a new suggestion. There was a bear on show in the palace, and he could conduct the same experiment on this. The king agreed, and the deed was done. The outcome was that the king’s wound was found to need draining, and once this was done he recovered. Although he does not say so, GN may have obtained the tale from FC’s first recension, while BN chose to omit it. See also Rudolf Hiestand, ‘König Balduin und sein Tanzbär’, Archiv für Kulturgeschichte, 70 (1988), 343–60; S. B. Edgington, ‘Medical Knowledge in the Crusading Armies: The Evidence of Albert of Aachen and Others’, The Military Orders: Fighting for the Faith and Caring for the Sick, ed. Malcolm Barber (Aldershot, 1994), pp. 320–26. I have discussed this with Dr Piers Mitchell, and it is possible that after this Baldwin may have suffered from a chronic and low-level septicaemia that flared up at intervals. It may have been contributed to his death in 1118. See also Mitchell, Medicine in the Crusades: Warfare, Wounds and the Medieval Surgeon (Cambridge, 2004), pp. 159–62. 38 AA 9.23–25, pp. 666–70. See also IQ, pp. 58–59, entry for ah 496 (15 October 1102–4 October 1103); Brett, ‘Battles of Ramla’, p. 33. 39 FC 2.31–33, pp. 489–503; quotation at 2.31.2, pp. 490–91. Cf. AA 9.48–50, pp. 704–10. 40 Arabic sources in Brett, ‘Battles of Ramla’, pp. 31–37. As before, chroniclers’ numbers can scarcely be trusted, but, for what it is worth, the Arabic historians Ibn al-Athīr and al-Maqrīzī wrote that the Egyptians fielded 5,000 Egyptian cavalry, reinforced by 1,300 Turkish horse from Damascus, against Baldwin’s 1,300 horse and 8,000 foot. The Turks from Damascus were over 1,000 mounted archers, Fulcher reported, and they joined an army of Arab cavalry and Ethiopian infantry that he later estimated at a total of 15,000. He claimed Baldwin had with him about 500 knights, ‘disregarding those who were mounted but did not count as knights by name’. There were also no more than 2,000 foot. Fulcher had revised his enemy figures radically from his earlier version, if that was faithfully reflected by BN, who claimed that the enemy numbered around ‘thirty times a hundred thousand’ (3,000,000) cavalry and ‘countless’ infantry, plus 1,000 Damascene Turks ‘skilled with bows and arrows’. Bartolf cast some light on the riders who were not knights, for he added the men who responded to the patriarch’s appeal in Jerusalem: ‘At once they mounted horses and left the city, and there were fifty in all, both clerics who had horses, and laymen.’ FC claimed there were 150 cavalry and infantry recruited in Jerusalem. FC regretted that the Egyptian general ‘Semelmulc’ had not been captured, although ‘Gemelmulc’ the emir of Ascalon was killed. These are identifiable as Ibn al-Athīr’s ‘Sanā’ al-Mulk Ḥusayn, one of the sons of al-Afḍal, the wazīr and effective ruler of Egypt, and Ğamāl al-Mulk, ‘who held Ascalon for the Egyptians’. The captured emir of Acre was named by al-Maqrīzī as Zahr al-Dawla Banā al-Ğuyūšī. 41 BN chaps. 69–71, pp. 539–41. 42 These are the last (human) events that BN recorded. Chapter 71 ends ‘Atque hic finis est’, although in fact a postscript describes the earthquake (Dec. 1105) and the comet (February‑March 1106) that occupy FC 2.34 and 35. 43 BN chap. 71, p. 541. His editors criticised his dating in a footnote, but they overlooked first that ‘anno millesimo centesimo sexto’ means ‘in the 1106th year’, i.e. 1105, and second that ‘ab urbe Iherusalem capta septimo’, ‘in the seventh year from the capture

150  Fighting the Saracens

44 45

46 47 48 49

50 51

52 53

of Jerusalem’, was counted in the classical and medieval way, i.e. including 1099 as year 1. AA 9.48, pp. 704–706. AA named ‘Hugh of Tiberias; Rorgius of Haifa; Gunfrid of the Tower of David; Hugh of St Abraham; Eustace Granarius; Gutmann of Brussels, a castle in Brabant; Lithard of Cambrai, a city in Gaul; Pisellus of Tuorna and Baldwin of Heestert, both from castles in Flanders’. AA 9.48, p. 706. IA pp. 93, 89. Ṭughtigin was the atabeg of Damascus 1104–1128. AA 9.49–50, pp. 706–10. AA 9.51–52, pp. 710–16. As a deer panicked at the smoke and the din, the Christian knights set off in pursuit. One of them, Arnolf of Oudenaarde, fell from his horse and was killed and decapitated. Three days after the Franks had buried his body, the Ascalonites sent back his head with a defiant message. Arnolf was another Flemish knight: Murray, Crusader Kingdom, p. 184. AA 10. 9–17, pp.  726–34 (October); FC 2.37.2–5, pp.  514–18 (November). The campaign does not feature in IQ or IA, probably because it was considered a minor Ascalonite initiative. The scared knight was probably Gerald the Chamberlain: Murray, Crusader Kingdom, pp. 199–200. For Gunfrid, see Murray, Crusader Kingdom, pp. 206–207; H.E. Mayer, ‘Die Herrschaftsbildung in Hebron’, Zeitschrift des Deutschen Palästina-Vereins, 101 (1985), 64–81. AA 11.27–30, pp. 800–804; FC 2.44.5, pp. 547–48, said only that the fleet ‘was lurking in the port of Tyre’ during the siege of Sidon and committing piratical acts but did not dare come out while the Norwegian ships were around. FC 2.53.4–7, pp. 585–86; AA 12.17, pp. 848–50, reported an Egyptian fleet at Tyre 15 August to 11 September 1115. As it sailed back to Egypt it was attacked from Acre. Two of the Saracen ships were captured. AA 12.23, pp. 858–60, for the 1117 attack.

9 The army, administration and allies The army, administration and alliesThe army, administration and allies

Introduction As the previous chapters have shown, for at least the first decade of his reign Baldwin had to maintain his kingdom on a war footing. Command of the coast was key to its security, and so at the same time as he was pursuing a policy of capturing the towns along the littoral, one by one, he had to resist repeated Egyptian attempts to retake Jaffa, using Ascalon as a bridgehead. Baldwin was not completely successful in either his offensive or his defensive campaigns, but his achievements were remarkable in view of his chronic shortage of manpower. We have already seen some of the side effects of this, particularly the need for the support of the Italian fleets to capture the coastal towns. In the short term Baldwin was unable to prevent them from sacking and looting, even when a surrender agreement was in place; in the longer term he bought their assistance at the cost of valuable trading privileges in the captured ports. Raising an army to respond quickly to attacks by the kingdom’s enemies was equally or even more problematical. This chapter will look at how Baldwin I’s army was raised and maintained; at some of the men who commanded the royal forces; at the evidence for the development of an administration to assist in the government of the kingdom; and at Baldwin’s relations with the other crusader states.

The army The mainstay of the medieval army was its knights. As previously indicated, the majority of those who had come on the First Crusade had left before Baldwin became king. Hence at the first battle of Ramla in 1101 the king was able to put no more than 300 knights in the field. The number of knights that could be mobilised increased through Baldwin’s reign, but so did the size of the kingdom, and each conquered settlement needed to be garrisoned while the army was in the field. Therefore, as a pre-requisite to the expansion and consolidation of his kingdom, Baldwin had to attract knights from western Europe, persuade them to stay and reward them for their services. The crusading expeditions of 1101 arrived during Baldwin’s reign, but they had set out in response to appeals from the crusade leaders to the pope and from the pope to the faithful made within his

152  The army, administration and allies predecessor’s time. They might have served as a model for inviting pilgrims and settlers had their crusade had a more positive outcome, but the attrition the armies suffered even before they reached Jerusalem and their losses in the second battle of Ramla (1102) ensured that no such large-scale recruitment would be repeated for decades. It is likely, anyway, that most of the participants in the expeditions intended to return home after fulfilling their pilgrimage vows by praying at the Holy Sepulchre. Mention should briefly be made of Bohemond’s plans for a crusade and his recruiting tour in France in 1106, which met with an enthusiastic response, but Bohemond’s intention was to attack the Byzantines, not to assist the king of Jerusalem, and the crusade’s failure could only add to disillusionment with the crusade idea.1 Nonetheless, Baldwin was able to call on more knights in his later campaigns, despite the losses suffered in earlier battles. Some of these will have been arrivals from the West, for the sources show a constant coming and going of ships, not only from the Italian cities but also from the regions around the North Sea. Sailors and pilgrims might be of short-term assistance, but the ships also brought prospective settlers, especially as the kingdom became more secure. There was also some movement between the states; for example, a number of Baldwin’s men accompanied him from Edessa to Jerusalem in 1100. Knights could be created from squires, and presumably this happened routinely in normal circumstances; in an emergency the process could be accelerated, as it was before the battle of Ramla in 1101. Fulcher of Chartres reported that in the following year, in even more desperate circumstances, ninety cavalry were recruited in Jerusalem, ‘so many knights and the same number of men who could obtain horses or mules’.2 However, the pack animals ridden by the non-knights were not fast enough to escape the enemy, so a number of their riders dismounted and took to the water; they saved themselves by swimming, but they lost their mounts. The knights on their swifter horses were able to escape. Hence we have no information as to whether the other mounted men would have enhanced their status had they fought and survived. Attracting knights from the West or making them in the East was only the first step in creating an army, however. Ensuring that they were properly equipped and rewarding them for their services required very substantial resources. The widely propagated idea that the western model of knight service depended entirely on providing soldiers in return for land tenure has been substantially revised, but even if there had been such a ‘feudal system’, the king of Jerusalem did not hold sufficient territory to adopt it.3 There was scope, however, for supplementing land grants by way of money-fiefs, the right to receive a part of the revenues derived from royal properties. For example, in 1107 Gerard the Chamberlain was awarded income from the revenues of Jaffa in return for military service.4 Knights might hold relatively small properties, perhaps as part of a ‘portfolio’ including moneyfiefs as well, in the countryside or in towns. Rural properties yielded crops and other agricultural produce, and in some circumstances the tenant could levy a poll tax on non-Christian inhabitants.5 Urban premises could be leased out or put to commercial use to provide income. In the cities and villages, too, a class

The army, administration and allies  153 of freemen was becoming established, burgesses who – Marwan Nader conjectured – were not only tax payers but belonged to a garrison that could be called on to defend the settlement or to augment the army.6 How far this was formalised during Baldwin’s reign cannot be established. It may safely be assumed that at times of military emergency all able-bodied men were expected to serve in the army: this explains the large numbers of infantry in the battles against the Saracens, in contrast to the smaller forces that rode on planned campaigns, for example to aid the northern states. The extent to which the infantry divisions were a formal and organised component of Baldwin’s forces is difficult to establish from any surviving evidence. An indication of one function in siege warfare is the presence of Reinold, ‘master of archers’, at the unsuccessful attack on Acre in 1103. He was said to have killed more than 150 Saracens with his crossbow before he was struck on the head by a mangonel stone and killed.7 Another source of income that could be farmed out in return for army service was the tribute exacted from the Muslim population of captured cities: Arsuf’s tribute was granted to Robert of Apulia when the city was taken in 1101.8 Tribute from Muslim-held cities had been a source of income for the crusade leaders from the time of their march south to Jerusalem in 1099, and Baldwin continued to profit from this sort of ‘protection-money’; in an example already cited, ‘The king, as one who had newly arrived and was lacking much treasure for paying his soldiers, agreed to accept everything offered to him by the pagan cities of Ascalon, Caesarea, Acre and Tyre, but he refused Arsuf and its gifts.’ It was a two-fold gain, as it allowed the king to pick off the cities one by one (in this case he intended to attack Arsuf) and provided money to pay his army.9 In a later variation, in 1106 Baldwin found it profitable to abandon the siege of Sidon in return for the sum of 15,000 bezants. In 1107 he repeated the exercise at Tyre, besieging it for a month before he was bought off with 7,000 dinars. The governors of these towns were well aware that paying tribute was the better policy. If the town was taken by assault then it could be sacked and looted, as Caesarea was in 1101.10 Even if the townsmen negotiated surrender terms to prevent this, the capture of other coastal settlements had demonstrated that the king could not control the Genoese and Pisans who participated in the blockade  – and sometimes not his own soldiers. The capture of Caesarea yielded more income in the form of ransom: the qadi of Caesarea was kept alive because of this, according to Fulcher; Albert called him a priest and wrote that he was ransomed by the citizens of Acre for 1,000 bezants.11 Victory in battle could also yield high-status captives who could bring in large sums in ransom. Albert reported that forty-eight Turks were taken for ransom in the battle near Beirut in 1100, including a ‘prince of Damascus’ who offered ‘a very great deal of treasure’. Forty-five of the captives apparently reached Jerusalem, and an enormous sum of money was paid for their release: over 50,0000 bezants.12 Plunder seized from the enemies’ camp after they had fled from the battlefield was likewise a legitimate gain, and after the first battle of Ramla in 1101 the king was unable to carry off all the spoils. He took gold and silver and horses and mules himself, and the citizens of Jaffa took in more, including tents.

154  The army, administration and allies Baldwin paid a tenth to ‘the hospital and Christ’s poor’, but the remaining riches would enable him to pay his troops and provide horses and mules to replace those lost. Even a minor engagement like the relief of Hebron in 1107 (or 1108) yielded ‘thirty-three camels, sixty-eight horses, with booty and many tents’.13 This was plunder as a by-product of success in battle; it could also be an end in itself. From first to last, Baldwin’s reign was punctuated by raids and forays into enemy territory in pursuit of spoils. On his expedition to the Dead Sea in 1100, before his coronation at Christmas, Baldwin and his men ‘seized herds and other spoils everywhere’. On later occasions Baldwin might delegate, as in August 1107 when he was engaged in the unsuccessful siege of Sidon and heard from his spies that a wealthy Arabian woman had settled on the other side of the river Jordan with great herds of camels, cattle, sheep and goats. He sent William, the illegitimate son of Robert of Normandy, who had recently arrived in the East, to collect ‘the men whom he had left to protect the city’ and to seize the woman with all her household and livestock. William collected 200 cavalry and 500 infantry from Jerusalem, crossed the river and attacked the herds, which were guarded by 500 archers. The Christians prevailed, killing men, taking women and children prisoner and carrying off some 4,000 camels and other livestock, ‘an incredible amount of booty, as price of which a great sum of gold was received that was shared among the soldiers on the king’s orders’. The episode tells us much about Baldwin’s priorities as king and the ways in which he deployed his resources. He had left a substantial garrison in Jerusalem while the main army invested Sidon from the land and Italian ships besieged it from the sea. He employed scouts, or informants, who ranged quite far afield and brought him news of an opportunity for immense gains. The king was reluctant to weaken the blockade around Sidon so he sent a trusted knight to collect troops from the garrison in Jerusalem, apparently confident both that they would follow him and that the city would be adequately defended while they were gone. The strength of the raiding party had been correctly judged and succeeded in bringing back captives and booty, which Baldwin used to reward his soldiers.14 Another raid was carried out in 1111, when one of the king’s knights guaranteed safe conduct to a treasure-laden caravan travelling from Tyre to Damascus. He had accepted a bribe of 1,000 bezants for this, but had no compunction in betraying his knowledge to the king. With 200 men Baldwin ambushed the caravan and seized its ‘wonderful and incredible treasures . . . which the king bestowed with a generous hand on his soldiers, who until now had been oppressed with long penury’. The following spring, Baldwin led a force of 200 knights and 100 footsoldiers into the valley of Moses (Wadi Mūsa in modern Jordan) for plunder ‘with which he could enrich his soldiers, who were destitute and devoid of possessions’. They encountered a caravan of Bedouin merchants, took many of them prisoner and carried off their goods, including gold, silver, silks and spices, which were shared among the soldiers. Albert recorded of a previous occasion that the division of spoils was two thirds to the soldiers and one third to the king.15 If Albert recorded only the most successful raids and those he heard about, the impression of Baldwin that is conveyed is of a king always in motion, leading

The army, administration and allies  155 his army into battle, carrying out a programme of sieges to command the coast, defending the kingdom’s hard-won possessions and carrying out raids in order to equip, pay and reward the men who were essential to these activities. Alan Murray aptly concluded: For the first five or six years of its existence, the rather grandiosely named kingdom of Jerusalem was little more than a kind of robber principality which financed itself through periodic but irregular injections of cash on a hand to mouth basis.16 Arguably, the same lack of fiscal planning characterised the rest of Baldwin’s reign too. It is remarkable that at a time when the royal mint was an instrument of government in western Europe, there was none in the kingdom of Jerusalem: not a single coin of Baldwin I has been found, nor is there any reference to a mint or moneyer in the surviving documents.17 Where coins were used for cash transactions and as money of account, Jaroslav Folda conjectured they were imitation gold dinars, indistinguishable from the genuine ones, but in the opinion of Robert Kool they were genuine gold Fāṭimid dinars and the production of imitation ones was a later development.18 Smaller value coins that changed hands in the market place, for example, were a different matter. Raymond of Aguilers listed seven types of coin in circulation among the First Crusaders from Provence: from Poitou, Chartres, Le Mans, Lucca, Valence, Melgueil and Le Puy. D. M. Metcalf considered it likely that the other armies brought more and different examples of local currency, and he pointed out that archaeological finds suggest that such coin types ‘continued to be used in the Latin East when the territorial states of the Crusaders had been established’ and as late as the 1180s. The stock of money was probably augmented by imports brought by the Italian fleets.19

Administration A picture emerges of Baldwin I as a capable warlord, but of course there was much more than this to ruling the infant kingdom of Jerusalem. As in the case of the army, he depended on nobles and knights who had elected to settle in the Holy Land to hold the major fiefs and, with clerics, to fill the offices of state and to act as administrators. In 1100 he could count on the loyalty of the men of his late brother’s household who had brought him to Jerusalem and on the men who had accompanied him from Edessa. Over the longer period of Baldwin’s reign, Alan Murray has identified by name and place of origin as many of the men involved in the government as the sources allow and extrapolated from them to reach the following conclusion: Apart from a small number of knights of southern French, German, Italian and non-Frankish origin the vast majority of the ruling classes of Jerusalem in the reign of Baldwin I can be shown to have originated from four principal areas: Flanders, Artois and Picardy; Norman Italy and the Anglo-Norman world, Lotharingia, and the Ile-de France.20

156  The army, administration and allies Many of these individuals appear only fleetingly in the sources, perhaps as someone who was killed in battle or witnessed one of the relatively few surviving charters. Two of the greatest fiefs, on the other hand, are well documented, and they illustrate well the range of origins and lack of continuity among their holders. Hebron – the crusaders’ St Abraham – was held successively by: Geldemar Carpinel, from southern France, was killed in the first battle of Ramla (1101); Gerard of Avesnes, a Lotharingian, was killed in the second battle of Ramla (1102); Hugh of Rebecques, a Fleming, held it next, but exact dates are unknown; Walter Mahomet, thought to be a converted Muslim, was lord of Hebron from 1107 or 1108; Walter was succeeded by Baldwin of St Abraham, probably in 1115, and Baldwin’s place of origin is not known, perhaps because of his long tenure of the fief to 1136. With regard to Tiberias, Tancred (a Norman from southern Italy) was obliged to recognise Baldwin’s suzerainty in 1100, and then, after he accepted the regency in Antioch in 1101, the fief was held by Hugh of Fauquembergues, a Norman who had been with Baldwin in Edessa and accompanied him south when he claimed the kingship of Jerusalem. Hugh was killed by a Turkish ambush in Syria in 1106. His successor, Gervase of Bazoches, from northern France (the modern département of Aisne), was captured and executed by Ṭughtigin of Damascus in 1108. Joscelin of Courtenay, who was lord of Tiberias 1113–1119, came from a very well-connected family in the Ile-de France. He travelled after the First Crusade, perhaps in the 1101 expeditions, and held the fief of Turbessel from Baldwin of Edessa until 1113 when, after a dispute with Count Baldwin, he went to Jerusalem and was granted the lordship of Tiberias. He relinquished the fief in 1119 to become count of Edessa as a reward for his support of the candidature of Baldwin II after Baldwin I’s death in 1118.21 Baldwin’s initial dependence on the domus Godefridi and his own loyal following from Edessa was thus unavoidably attenuated by losses in battle and to other enemy action and in other circumstances. However, he was well served by replacements from a range of disparate backgrounds. We have already seen how he delegated a perilous and profitable raid across the Jordan to William, the bastard son of Robert of Normandy, in 1107. William had arrived only very recently in the kingdom, following his father’s defeat by Henry I at Tinchebrai in 1106, but he led 700 men against a well-guarded encampment. He was made lord of Tortosa by 1111, when he accompanied the king on a campaign against Shaizar in the north.22 Another noble, one of the most powerful, was on the same expedition: Eustace Granarius had received Sidon from the king after its capture in 1110 and was probably already lord of Caesarea. He added Jericho to his fiefs when he married the niece of Patriarch Arnulf: William of Tyre added this to the scandals attaching to Arnulf, since Jericho belonged to the church of Jerusalem, and William noted that in his own day the revenues from Jericho were reported to be 5,000 gold coins. Eustace’s rise to become ‘first in the household and council of the king’ (as Albert of Aachen called him at the siege of Tyre in 1111) is tantalisingly obscure. Alan Murray has established beyond reasonable doubt that Eustace was from the county of Saint-Pol (Pas-de-Calais) and so probably travelled on

The army, administration and allies  157 crusade with Count Hugh of Saint-Pol. He was first named as a knight when summoned by the king to fight the third battle of Ramla in 1105.23 Walter Mahomet, lord of Hebron, was a third noble on the same expedition in 1111, and in the same year he reportedly earned the trust and gratitude of King Baldwin when he was left in charge of Jerusalem during the king’s absence on campaign.24 During an arranged two-year truce between Jerusalem and Ascalon some prominent Ascalonites took advantage of the opportunity to sight-see in Jerusalem, and when they returned home they wrote to Egypt proposing a scheme: the wazir would send weapons disguised as merchandise and soldiers disguised as traders. The goods were gathered, and word was sent to Jerusalem. The Jerusalemites urged Walter Mahomet to go to Ascalon in the king’s stead. As a convert, Walter had an excellent command of Arabic and so was able to understand the Ascalonites’ plotting. He feigned ignorance and granted permission for the 1,000 ‘traders’ with their 500 sacks of ‘merchandise’ to enter Jerusalem. Meanwhile he had sought out the king in Arabia and brought him back to the city. The king demanded that the sacks be opened at once, and, against some Egyptian resistance, weapons and a splendid war-trumpet were discovered. All but one of the men were put to death; the single survivor revealed that the war-trumpet was to be sounded at dawn and that at the signal a gate into the city would be opened and more men would come in, singly and unarmed, expecting to find weapons in the sacks. The Christians took over the plot, and in the morning they blew the trumpet, opened the gate, allowed 1,000 men to enter and slew them all. When the soldiers still outside heard the noise they thought it was their men slaughtering the Christians. After the sun shone and the day grew bright, the king knew it was not safe for a few to meet countless numbers, and he ordered the heads of the killed to be loaded into slings and mangonels and to be hurled into the camp, so that the Saracens would be frightened by the deaths of their own men and would be forced to flee.25 The cunning and ruthlessness shown by King Baldwin and Walter’s ability to understand Arabic are details that seems authentic among some undoubted exaggerations of numbers and wealth. There was thus an upper stratum of nobility who acted as army commanders and counsellors to the king and who held from the king and administered tracts of territory as they were acquired. The allocation of their fiefs was carefully handled by King Baldwin: urban and rural properties were retained within the royal domain and not granted as hereditary possessions. Steven Tibble has shown that this was the case for Galilee, Caesarea, Sidon, Jaffa (at least to 1110), Ramla and Arsuf. He concluded that Baldwin and his successors manipulated lordships to maintain and exercise royal control. This constitutes an important policy innovation by Baldwin I.26 Below the highest echelon on the social scale were knights who acted as castellans, commanding the garrisons of towns and cities. Some of these are known

158  The army, administration and allies from single inclusions in witness lists; others are named in the literary sources in relation to battles and sieges. It is rarely possible to date their appointments or to establish a sequence in the same post for the entire reign. The Tower of David, Jerusalem’s citadel, was commanded successively by Gunfrid (to 1106) and Anselm (to c. 1116). Both men had landholdings in the region of Jerusalem to support them. In Jaffa we have the names of two probable castellans, Lithard of Cambrai (1105) and Roger of Rozoy (1106). Baldwin of Ramla was castellan in that town from 1106, at latest, until well after Baldwin I’s death. In terms of a social hierarchy, some of these men were called viscount (vicecomes), suggesting they were sub-tenants. We also have the names of Pisellus, viscount of Jerusalem from 1108 at the latest and for the rest of Baldwin’s reign, and Ulric, viscount of Nablus from 1115. Both of their names are known from frequent inclusion in witness lists, so they were evidently prominent men who spent time at the royal court.27 The same, rather random, literary references and appearances in witness lists to charters yield a few indications of officials in the king’s service, ranging from men who might grandiosely be called officers of state to men who served in the royal household. Simon the Constable, a quite senior military official, appeared as a witness to charters between 1108 and 1115 and may also have been the representative sent by King Baldwin to Byzantine representatives at Tripoli in the spring of 1112 to invite them to meet the king outside Tyre, which he was then besieging.28 Three men were identified as steward or seneschal: the Latin dapifer was originally a ‘servant who waited at table’, but by the twelfth century he was ‘one of the great court ministers’. Matthew the Seneschal was one of the men who summoned Baldwin from Edessa in 1100; Gervase of Bazoches was also lord of Tiberias; Hugh Chostard’s name is known only from a single charter dated 1112, but he is assumed to have succeeded Gervase, who was killed in 1108.29 We can discern the development of a formal secretariat, not only through the increasing numbers of charters produced, but also by the emergence of named officials: Robert was the principal official between 1101 and 1114, and he was replaced by Pagan, the first chancellor, by 1115. Both men were priests, unsurprisingly since this was the group of people who had facility in reading and writing.30 Another important symbol of the development of a royal chancery is the seal of Baldwin I. The original does not survive, but impressions and reproductions show that it had on the obverse ‘a king seated on a backless throne holding a crosstopped orb and scepter, wearing a three-point crown with two pendants’. On the reverse was a Jerusalem gate, and behind it the Tower of David in the centre, the Temple of the Lord on the left and the church of the Holy Sepulchre on the right. It is unmistakably the Jerusalem skyline, but the dominance of the secular citadel is interesting, and the two sides of the seal may be interpreted as revealing how Baldwin perceived the nature of his kingship.31 Rather like the steward’s role, the chamberlain’s function had grown beyond recognition since its roots in the bed-chamber. Gerard appeared as camerarius or cubicularius in documents from 1108 to 1115 and was probably the same Gervase, ‘knight of King Baldwin’s household’, who was awarded part of the revenues of Jaffa for his military service in 1107. He was evidently of knightly status, and the nature of

The army, administration and allies  159 his role was not defined, but is likely to have extended to overseeing the income and outgoings of the royal household.32 Another man described as a ‘sort-of chamberlain’ (quasi cubicularius) had a much humbler function, if William of Tyre’s story reflects any truth: he was a convert from Islam who had taken his patron’s name, Baldwin, at baptism and had become a trusted domestic servant. His duties included accompanying the king to the privy. During the siege of Sidon in 1110 some nobles of the town approached the servant by way of intermediaries and offered him untold wealth and property in Sidon if he killed the king. The servant was all set to do so when word of the plot reached some Christians within the town who shot into the king’s camp an anonymous letter attached to an arrow. Predictably, the king was angry. The servant was forced to confess and was hanged. According to William of Tyre, the failure of their plot brought about the surrender of the Sidonites.33 The first decade of Baldwin I’s eighteen-year reign was dominated by battles, sieges and other military operations. These were necessary for the protection and security of the nascent kingdom of Jerusalem, but they necessitated the growth of an administrative infrastructure to keep the war machine going. It happens that the two failed plots already described indicate a turning of the tide of war by about 1110. The conspiracy of the Sidonites in 1110 failed because fellow citizens demonstrated their loyalty to a Christian king by warning him. The Ascalonites’ plot in 1111 could only have been conceived during a period of truce between the kingdom and the Egyptians. An accommodation had been reached between Jerusalem and the Saracens, and although Ascalon itself remained dangerous, no massive Saracen army was landed there after the third battle of Ramla in 1105, and no major naval attack was launched from Egypt after the assault on Beirut in 1110. The Christians’ long siege of Tyre that ended in 1112 was the last of its kind. From then on King Baldwin’s preoccupations changed. Before considering his last few years, however, we should look at the other crusader states and the dangers from the Turks in the north and east that demanded co-operation between the rulers.

The crusader states The two major ‘crusader states’ were Edessa, Baldwin’s own foundation, and Antioch. Tripoli was still disputed territory during Baldwin I’s reign, but the king’s involvement as one of the disputing parties will be discussed below. It would be natural for the established territories under Latin control to consider themselves allies, but historic rivalries persisted from the period of the First Crusade into the era of settlement, and only in times of existential threat did the three states co-operate wholeheartedly.34 Edessa and Antioch When Baldwin became the first king of Jerusalem in 1100 he left the county of Edessa ‘a patchwork of castles, cities and rural areas, some directly under Frankish authority, but most under local Armenian leaders. . . . Baldwin directly ruled a core of territory on the western bank of the Euphrates River, but in the eastern area, his authority only extended to Edessa, Saruj, and Samosata.’35

160  The army, administration and allies His successor, Baldwin of Bourcq (now Count Baldwin II), brought the county more firmly under Frankish control, but not without some necessary interventions from his kinsman Baldwin I of Jerusalem. The first crisis was in 1104 when Baldwin II, in alliance with Bohemond of Antioch, Tancred and Joscelin of Courtenay, marched out to intercept a great Turkish army advancing to besiege Edessa. They met the Turks near Harran, a city under Turkish authority less than forty kilometres to the south-east of Edessa. In the ensuing battle the Christians were defeated, and Baldwin and Joscelin were captured.36 Bohemond had escaped capture, and his response was to instal Tancred as regent of Edessa. When Bohemond left for western Europe not long afterwards Tancred returned to become regent of Antioch, appointing Richard of Salerno, another south Italian Norman, as his vice-regent in Edessa.37 King Baldwin evidently had no say in any of these arrangements, probably having his hands full in 1104 (the siege and capture of Acre) and 1105 (the third battle of Ramla). However, this did not prevent King Baldwin from interceding with Bohemond and Tancred when a question of prisoner exchange and ransom arose. According to Albert of Aachen, Tancred had captured a powerful Turkish noblewoman, and Chökürmish, the Turkish governor of Mosul, proposed either to return Count Baldwin in direct exchange or to pay a ransom of 15,000 bezants. When the king heard of this, he appealed to Bohemond and Tancred to accept the exchange, for ‘no amount of money was more important, nor should they covet it’. The Normans’ reply was disingenuous: they were prepared to obey the king’s orders, but they wanted to dissemble and pretend otherwise in the hope of extracting some money too, to pay their soldiers. These replies from them were smooth and flattering enough, but there was no good faith or truth in them at all, nor any intention of ransoming the man, on account of their ambition to have the state and its revenues . . . that added up to forty thousand bezants every single year, not counting those revenues which many castles and districts belonging to Edessa bestowed.38 The incident, which only Albert reported, may throw a little light on the relationship between the Christian princes: Bohemond and Tancred spoke of ‘obeying’ the king’s commands, perhaps paying lip service to the idea (which may have been Albert’s rather than theirs) of Jerusalem’s suzerainty; however, they had no intention of complying with the king’s wishes, and he had no way of enforcing them. In the event, Count Baldwin remained captive in Mosul for four years, and his release was secured in 1108 only by the efforts of his fellow prisoner Joscelin.39 Tripoli The following year King Baldwin summoned Tancred to Tripoli for a council of the Jerusalem Church and demanded that he restore to Baldwin II and Joscelin ‘estates unjustly stolen’. The king was also seeking redress for the recently

The army, administration and allies  161 arrived Bertrand, son of Raymond of Saint-Gilles (d. 1105), while Tancred was harbouring Tripoli’s counter-claimant, William Jordan.40 It is likely that Tancred felt more vulnerable now that Bohemond was not only absent but defeated by the Byzantines, and he obeyed, although not before assembling an army of 700 knights to accompany him and William Jordan to the meeting. Baldwin of Edessa and Joscelin followed him with another great cavalry force. Once all the parties were assembled, they were given the opportunity to air their grievances, and then a great reconciliation took place and a grand redistribution of lands. Tancred restored his possessions to Baldwin of Edessa, and it was agreed that Bertrand would receive the port of Tripoli itself (when captured), with Pilgrim’s Mount, Raymond’s castle outside the town and Jubail and all the adjacent areas. William Jordan was granted Tortosa and Arqa with their dependencies. The result, as Kevin Lewis has demonstrated, was a bipartite division of the proto-county, with Arqa and conquests to its north owing allegiance to Antioch; lands to south acknowledging the suzerainty of Jerusalem.41 As part of the same settlement, King Baldwin appeased Tancred by returning Haifa, Tiberias and Nazareth to Tancred, after receiving his fealty. The council of Tripoli was thus an exercise in pragmatism by the king, who recognised the need to make concessions in order to avoid damaging conflict between Christian states that would make all of them vulnerable to attack by the Turks. Importantly, though, despite having to make some concessions, the king had asserted his authority over Antioch. When, soon afterwards, William Jordan was killed, his possessions did not revert to Tancred, but went to the royal vassal Bertrand, who now governed territory as far north as Arqa, Tancred having seized control of Tortosa.42 Meanwhile, the council of Tripoli had brought together forces from Antioch and Edessa, who, with the Genoese fleet that had brought Bertrand and the king’s own army, concerted an assault on Tripoli which proved successful. Following Tripoli’s capture Bertrand did homage to the king for the town, and as the new count of Tripoli he served the king in the sieges of Beirut and Sidon.43 He also accompanied King Baldwin on campaigns against the Turks in the north who were becoming a major threat under the leadership of Mawdūd, governor of Mosul (1108–1113).44

Attacks by Mawdūd of Mosul, 1110–1113 Mawdūd’s first campaign against the Franks attacked the city of Edessa in May 1110.45 Count Baldwin sent messengers to inform the king, then besieging Beirut, of the attack, which he claimed the Turks had undertaken at the instigation of Tancred. The king delayed his response until he had completed the capture of Beirut and then took his army north. Albert of Aachen, whose account is the fullest in Latin, described the arrival of the king in the county of Edessa that forced the Turks to lift their siege and withdraw to Harran. He was met by Count Baldwin, who was still complaining about Tancred, so he summoned Tancred from Antioch and allowed him to state his case, which was that Edessa was rightly subject and tributary to Antioch. The king replied ‘very mildly indeed’, but he

162  The army, administration and allies dismissed Tancred’s claim and delivered a veiled threat: if Tancred preferred to be allied with infidels then he could not remain a ‘brother of Christians’. Tancred returned to the fold, according to Albert, and he and the king put the Turks to flight (Ibn Al-Qalānisī presented it as a strategic withdrawal, which seems to be nearer the truth). However, the Turks inflicted a damaging defeat on the Frankish and Armenian forces at the river Euphrates. This first battle against Mawdūd illustrates some key aspects of Baldwin’s character and policy. He showed a clear sense of priorities when he kept quiet about the appeal from Edessa and completed the capture of Beirut before revealing it. He rallied the troops before they set out, presenting the expedition as a religious imperative and stressing the duty of all Christians to lay down their lives, if necessary, for their co-religionists. He also used the situation to impose his suzerainty on both Baldwin of Edessa and Tancred of Antioch, certainly by his actions in summoning them both to be reconciled, and possibly in his speech to Tancred: Albert’s version included an explicit statement that ‘if the Christians’ affairs go so very well that we appoint a king – a head, a ruler, a defender to keep our possessions  – then we should follow him as his subjects’.46 Of course, Albert was not reporting the speech verbatim, but he was undoubtedly well informed about the campaign by someone who was on it, and if he was not conveying something actually said, then it was how the relationship between the states was perceived. In the following year, 1111, Mawdūd led another big Turkish army against the Franks, besieging Turbessel, Joscelin’s lordship. Since the Turks were unable to capture the fortress they withdrew to Aleppo in the hope of provoking Tancred to advance against them. Instead of doing so, Tancred sent messengers to King Baldwin ‘humbly’ begging him to come to aid the Christian cause and showing that he had learned something of the right approach from the king’s speech the previous year. According to Fulcher of Chartres, who was on the campaign, when the armies of Tancred and Baldwin met in the Rūj valley, they pitched camp together. The Christian and Muslim armies met in northern Syria, near Shaizar, but the encounter was inconclusive; it became a stand-off, and the Christian allies were forced to withdraw because the Turks’ activities deprived them of food and water. Once again Baldwin was able to present his defensive campaign as a holy war: as Ibn Al-Qalānisī wrote, ‘Tancred, Baldwin, and the son of St. Gilles, notwithstanding their hostility to one another, and their mutual aversion and disagreement, joined forces and formed a united front against Islām and its people.’47 By now it seems that Mawdūd had formed a habit of making annual attacks on his Frankish neighbours. Although the Latin sources did not record it, according to Ibn al-Qalānisī Mawdūd raided around Edessa and Sarūj in the spring of 1112.48 His main objective appears to have been fresh pasture for the horses, for first he grazed them outside Edessa and then the army moved on to Sarūj for the same purpose. He took no precautions against the Franks, and as his army’s mounts grazed outside Sarūj, Joscelin of Turbessel attacked and scattered them, seizing some and killing soldiers before retiring into the town. Matthew of Edessa made a great deal more of the episode.49 He agreed that it was spring 1112 (they arrived on the day after Easter) and that first Mawdūd’s ‘tremendous number of troops’

The army, administration and allies  163 moved from site to site near Edessa, then in June a detachment of 500 diverted to Sarūj. Joscelin was there with 100 cavalry and 100 infantry, and he succeeded in defeating the Turks, killing 150, taking five officers captive, and seizing all the baggage. Joscelin then withdrew to Edessa, while Mawdūd thought to find him in Sarūj. When he realised the count was no longer there, he returned to Edessa, where he was approached by some treacherous Edessenes who betrayed the city to him. The Turks garrisoned three of the city’s towers. However, Joscelin, with Count Baldwin and other Franks, mounted a strong defence, attacked the main tower and saved the city. The disparity in the accounts is very striking. It is to be expected that Ibn al-Qalānisī would make light of a foraging expedition that resulted in a reverse, while Matthew of Edessa depicted Mawdūd’s defeat as a humiliation. Matthew was also an admirer of Joscelin – although he also deplored the cruelty with which he punished the Edessenes – and was keen to contrast his bravery with the negligible role played by Count Baldwin in the defence of his own city. Stranger still is the absence of this incursion and the Turkish occupation of Edessa from the Latin narratives. Either the incident was much slighter than Matthew wrote, or it passed so quickly that there was no opportunity to send for reinforcements from Antioch and Jerusalem. By the time of Mawdūd’s last major campaign against the Christians in June 1113, both Bertrand of Tripoli and Tancred of Antioch were dead. Bertrand was succeeded by his young son Pons (1112–1137). Tancred was succeeded as regent of Antioch by Roger of Salerno (1112–1119). At the time of the Turkish invasion in 1113, Pons, as a minor, was in Antioch, presumably as ward of the regent. This time the Turkish attack was in concert with Ṭughtigin, the atabeg of Damascus (1105–1128), who asked for Mawdūd’s aid against King Baldwin, who was raiding his territory. Fulcher of Chartres, Albert of Aachen and Ibn Al-Qalānisī all described the defeat the Turks inflicted on the royal army at al-Sinnabrāh, south of Lake Tiberias, at the end of June. Although Fulcher loyally put much of the blame on ‘our men’ who rashly charged the Turks and were caught in their ambushes, there is an implicit admission in his account that the king had marched his army into an indefensible position. Matthew of Edessa’s evaluation confirms Fulcher’s: he described the Jerusalem troops as ‘puffed up with pride’, deliberately attacking before the Antiochenes could arrive and steal their glory. When the Turks counter-attacked and set about killing the Christians, Baldwin had to flee, forfeiting his banner and his royal tent with all its fine equipment. Fulcher reported that thirty of the better knights and some 1,200 infantry were lost. Albert agreed on thirty knights and estimated 1,500 infantry. Ibn Al-Qalānisī, who was ill informed about which Christian lords were present, recorded three charges by the Turkish cavalry and 2,000 Christian fatalities. It was a severe defeat, and both the Latin writers blamed it on the king’s not having waited for the armies of Antioch and Tripoli. Fulcher explained that Roger of Antioch had been summoned ‘for the love of God and affection for the king’ and that he was very aggrieved by the king’s intemperance that had caused him to attack the enemy without waiting for his advice. Following the defeat the two armies stalked one another. Fulcher included an interesting detail about ‘Saracens’ (here native Christians) deserting

164  The army, administration and allies the Franks, while Albert described the arrival of thousands of Christian reinforcements. Nevertheless, Baldwin and his allies were unable to prevent the Turks spending the rest of the summer plundering the kingdom of Jerusalem as far as the coast at Jaffa and Acre and even reaching the walls of Jerusalem itself. King Baldwin had incurred a serious dent to his reputation in this engagement, and the situation could have become even more desperate had Ṭughtigin not conceived a great suspicion and fear of Mawdūd that resulted in his hiring assassins to murder him in September 1113.50

The Turkish invasion of 1115 The death of Tancred, Baldwin I’s arch-rival since the days of the First Crusade, brought a period of easier co-operation between Jerusalem and Antioch. This enabled them to face and to respond successfully to another Turkish challenge in 1115: the invasion of Bursuq of Hamadan, commanding the sultan of Baghdad’s army. Behind this attack was a struggle between Baghdad and Damascus for the city of Aleppo.51 In early summer 1115 Ṭughtigin of Damascus, then in precarious control of Aleppo, and his son-in-law Īlghāzī ibn Artūq of Mardin (d. 1122) approached Roger of Antioch to seek a military alliance against Bursuq. Roger apparently agreed to this, for in June he led an army of some 2,000 to Apamea.52 When, in August, he heard that Bursuq had advanced and was encamped with his army near Shaizar, Roger sent an appeal for military support to Jerusalem and Tripoli. At the approach of Baldwin I  and Pons of Tripoli Bursuq’s army apparently dispersed, and the Christian allies attacked a fortress called ‘Gistrum’, probably in an attempt to provoke Bursuq to battle. When this failed, the allies each withdrew to their respective home cities, including a Damascene contingent. Bursuq’s withdrawal, however, was only a feigned retreat, as Walter the Chancellor, our best witness, made clear. In September Bursuq resumed the offensive, and the armies of Roger of Antioch and Baldwin of Edessa came to battle with them at Tell Danith on 14 September 1115, achieving a notable victory.53 King Baldwin was not involved in the battle; nevertheless, the campaign that preceded it revealed a great deal about his reputation at the time and his relationships with the other Christian rulers. When Roger of Antioch realised that Bursuq was attacking across the Euphrates with a powerful army, he appealed to the king by way of messengers bearing sealed letters.54 Baldwin’s response was immediate: he sent to his vassal Pons of Tripoli the same day and set out with an army to come to the aid of the Antiochenes. He also warned Roger not to attack the Turks before he arrived, invoking ‘their bond of brotherly affection’. Bursuq, in his turn, was aware of the approach of the king and attempted to pre-empt his arrival by attacking the Antiochenes’ camp. Mindful of the king’s warning, Roger managed to restrain his troops from responding to the assault, and when Bursuq heard of the royal army’s swift advance he feigned retreat. Both Antiochenes and Turks, therefore, viewed the king’s arrival as highly likely to change the outcome of the encounter, in spite of his defeat by Mawdūd’s army two years earlier. By 1115, therefore, the king of Jerusalem’s position as overlord of the crusader states was

The army, administration and allies  165 firmly established, and his reputation as a formidable opponent was recognised by Turks and Saracens.

Byzantium Another interested party in the politics of the crusader states was the veteran Byzantine emperor Alexios I Komnenos. The events of the First Crusade had resulted in an intensification of his enmity towards Bohemond and engendered an alliance with Raymond of Saint-Gilles. The conquest of Jerusalem in 1099 was of less concern to him than the capture of Antioch the previous year: Jerusalem, unlike Antioch, had not been a Byzantine possession within living memory, and although the Byzantine emperor might want to assert his status as protector of the Holy Places there, pragmatism dictated a policy of seeking an accommodation or alliance with Jerusalem.55 This was in Baldwin’s interest, too, because of his reliance on Byzantium and the Italian cities for the safe passage of pilgrims and settlers from western Europe to Jerusalem. Thus in 1102, following the destruction of the Lombard army in Asia Minor, Albert of Aachen recorded a conciliatory legation from Baldwin to Alexios ‘confirming their treaty and friendship’ with regard to the security of the overland route from Constantinople to Jerusalem.56 Latin authors understood the treaty of Devol made between Alexios and Bohemond in 1108 as affirming that pilgrims would in future be able to travel safely and without molestation by land or sea anywhere where the emperor’s power extended, although the Greek writer Anna Komnene did not mention this.57 Bohemond himself was no longer a threat to the Byzantine empire, but since his vow to bind Tancred to the treaty could not be enforced once Bohemond had returned to Italy rather than to Antioch, the emperor sought to isolate Antioch from the other crusader states. Anna Komnene reported the advice of his counsellors: ‘They unanimously rejected the emperor’s plan for an expedition against Tancred for now; first, they said, he must win over the other counts who controlled the places near Antioch, and in particular Baldwin the king of Jerusalem.’58 In view of the long-standing enmity between Baldwin and Tancred it was a plan that might have succeeded, but the ensuing negotiations outside Tyre during the winter 1111–1112 failed over the matter of money. Anna wrote that the emperor sent great sums of money with his envoy ‘because of Latin greed’; according to Albert, besides the king’s perennial need for money to maintain his army he was promised a Byzantine fleet to assist in the capture of Tyre, but it failed to materialise. The negotiations collapsed amidst recriminations, according to Anna, although Albert reported that the Byzantine envoys celebrated Easter in Jerusalem and then were sent on their way with splendid gifts – even though in the previous sentence he had described the king’s soldiers as destitute.59 Baldwin was probably aware of Alexios’s negotiations with the Pisans which had already resulted in a formal alliance: the Pisans took an oath of fealty to the Byzantine emperor in April 1110, and in October 1111 Alexios issued a chrysobull that set out trading privileges, annual payments to the Pisans and the protection of law. Most significantly, in return the Pisans swore to defend the emperor’s

166  The army, administration and allies interests ‘as far as Alexandria’, a vast tract of territory that would include the entire kingdom of Jerusalem, and to defend his territories against all his enemies. Assuming that Baldwin did know of the Pisan alliance, his treatment of the Byzantine legates was statesmanlike: he treated them well without ceding any diplomatic ground, despite the temptations of money and naval support, both greatly needed.60 However, having dismissed the idea of a Byzantine alliance and lost the intermittent assistance of Pisan vessels, Baldwin found himself unable to complete his plan for the conquest of the coast by capturing Tyre. The Genoese were becoming more difficult to control and were demanding greater and greater concessions in the conquered ports, but without them the king lacked the means to mount a naval blockade. This was undoubtedly a major factor behind his illadvised third marriage.

Sicily There is evidence for a cordial relationship with the Normans in southern Italy early in Baldwin I’s reign. Albert of Aachen reported that in 1101 Count Roger I of Sicily (1062–1101) sent a gift of 1,000 gold coins to be divided equally into three parts: one for the canons of the Holy Sepulchre; one for the hospital; the third for the king to maintain and equip his army. Patriarch Daibert was found to have embezzled the whole sum, and this act of treachery led directly to his deposition.61 Beyond this episode there is no evidence of co-operation between the two states until 1113, although negotiations for the marriage alliance that took place in September of that year began sooner. The bride was Adelaide of Salerno, the widow of Roger I of Sicily, who had been acting as regent for their sons since Roger I had died in 1101. The younger of these, Roger II, reached his majority at the end of 1111, and it was after this that the question of marriage to the king of Jerusalem was raised. Fulcher of Chartres devoted only two short sentences to the marriage, with no background information, but William of Tyre explained the rationale for both sides of the agreement in detail. According to him, the initiative came from King Baldwin, who sent nobles to propose marriage to the countess. She duly consulted her son, Count Roger II, and they agreed on certain conditions, subject to which the marriage would go ahead. Most importantly, if the king died without an heir by Adelaide, Roger would succeed undisputed to the throne. William further claimed that Baldwin’s envoys were instructed to accept any condition, for the king ‘thirsted to be rescued from poverty by her overflowing riches’. William imputed the whole scheme to his bête noire, Patriarch Arnulf. Baldwin’s desperation to make the alliance, no matter what conditions were imposed by Sicily, has to have come from his dire need of funds. The matter of engendering an heir and assuring the succession was surely secondary, but it cannot have seemed out of the question. By 1113, Adelaide was about thirty-eight years of age. She was probably not beyond child-bearing and had proved her fertility by producing children with her first husband, although the last was born in 1095. King Baldwin was in his forties or early fifties. He had not, so far as is known, produced children

The army, administration and allies  167 by either of his first two wives, but there is no reason to rule out his not only doing so, but surviving long enough to see a son reach his majority.62 The countess fulfilled expectations by arriving at Acre with ample supplies of corn, wine, oil and salt meat; weapons and knights; and an enormous sum of money. Albert of Aachen elaborated rather fancifully: She had two trireme dromonds, each with five hundred men very experienced in warfare; seven ships laden with gold, silver, purple and an abundance of jewels and precious garments; weapons, hauberks, helmets, shields resplendent with gold, and all the other weaponry which powerful men are accustomed to carry for the defence of their ships . . . [On the countess’s ship] the mast was clad with purest gold and put forth rays from afar like the brilliance of the sun, and both ends of the ship, covered by craftsmanship with gold and silver, were a sight to wonder at. . . . In one of the seven ships Saracen men who were very strong archers and glittered with the brilliance of their precious garments were brought as a gift for the king, and their skill at archery was considered inferior to none in the region of Jerusalem.63 Albert went on to describe the attempts of the Ascalonites to intercept the fleet and the reception of the countess by King Baldwin, which (according to him) was magnificent and accompanied by a distribution of treasure to the king’s soldiers and the conveyance of even greater wealth to the king’s treasury. Roger of Antioch was at the wedding feast and was richly rewarded too. For more than three years following the wedding there is no information about the royal couple’s conjugal relationship, or Adelaide’s activities, or even her whereabouts. Baldwin seems to have been constantly on the move, fighting for Antioch in the north or against the Ascalonites in the south or exploring into Transjordan and the Sinai desert. However, the validity of the marriage was apparently unchallenged until 1116 when two events coincided. The more important was probably Arnulf’s trip to Rome to argue for his reinstatement as patriarch of Jerusalem. He was successful, but, seemingly, his restoration was conditional on persuading the king to put away his wife. The grounds existed, for although Baldwin had separated from his Armenian wife at some time in the previous decade, there was no annulment of his second marriage and so he was not free to remarry. The following winter, the king fell critically ill and feared he was on the point of death. There was a very real possibility that Baldwin would die without children and the kingdom would pass to a Sicilian heir. Some of the nobility baulked at this, and Arnulf presided over the formal annulment at Easter 1117, incurring the enmity of Sicily to the great detriment of the kingdom, as William of Tyre and modern commentators have agreed.64 The Sicilian alliance was therefore short-lived and ultimately injurious to the kingdom of Jerusalem. It is redundant to speculate whether the outcome would have been different if Baldwin and Adelaide had produced a child. The alliance failed, and Baldwin’s deteriorating health brought the matter of the succession to the fore.

168  The army, administration and allies

Notes 1 Andrew Jotischky offered indicative figures based on the prosopographical work of Riley-Smith and Murray: Crusading and the Crusader States (Harlow, 2004), pp.  63–64. Of 659 named crusaders (definite or probable) in Jonathan Riley-Smith, The First Crusaders (Cambridge, 1997), 104 are known to have stayed, not all of them knights and not all in Jerusalem (p. 19). At the first battle of Ramla (1101), there were 250–300 knights; at the second (1102), 200; at the third (1105) 500; for the expedition to Edessa in 1110, 600; the campaign of 1113, 700: Murray, Crusader Kingdom, pp. 98–99; see also previous chapter. For letters of appeal, see Hagenmeyer, Epistulae, nos. XIX, pp. 174–75; XXI, pp. 176–77. For the 1107 crusade of Bohemond, see Riley-Smith, First Crusaders, pp. 78–79; list of named crusaders, pp. 239–42. 2 For the emergency creation of knights see FC 2.11.2, p. 409 (for 1101); FC 2.21.7–8, p. 450 (1102). 3 For a much more complex picture of ‘feudalism’, see Susan Reynolds, Fiefs and Vassals: The Medieval Evidence Re-interpreted (Oxford, 1994). 4 For money-fiefs, see Alan V. Murray, ‘The Origin of Money-fiefs in the Latin Kingdom of Jerusalem’, in Mercenaries and Paid Men: The Mercenary Identity in the Middle Ages, ed. John France (Leiden, 2008), pp. 275–86; Murray, Crusader Kingdom, pp. 113, 199–200. For Gerard: AA 10.12, p. 730. 5 Rural property: written and archaeological evidence for other revenue raising powers and the rights the king retained in the villages comes only from later periods, see Ronnie Ellenblum, Frankish Rural Settlement in the Latin Kingdom of Jerusalem (Cambridge, 1998); Heather Crowley, ‘The impact of the Franks on the Latin Kingdom of Jerusalem: Landscape, Seigneurial obligations, and rural communities in the Frankish East’ (Unpublished Ph.D. thesis, University of Cardiff, 2016). 6 Urban property: Prawer, Crusader Institutions, p. 154; Marwan Nader, Burgesses and Burgess Law in the Latin Kingdoms of Jerusalem and Cyprus (1099–1325) (Aldershot, 2006), p. 198. It should, however, be noted that Nader based his argument on western European analogies and later written evidence. 7 For Reinold: AA. 9.20, p. 662. For infantry generally: John France, Western Warfare in the Age of the Crusades, 1100–1300 (London, 1999), pp. 64–76. 8 AA 7.12, p. 502. 9 ‘Protection-money’: BN chap. 45, p.  523; AA 7.51, p.  560; the apt descriptor is Murray’s. 10 Sidon: AA 10.4, p.  722; 10.8, p.  726; Tyre: IQ, p.  82 (ah501); Caesarea: AA 7.56, pp. 564–66. 11 Ransom: the qadi of Caesarea FC 2.9, pp. 400–401 and AA 7.56, p. 566. 12 Beirut: 1100 AA 7.34, 7.53, pp. 534–38, 560–62. 13 Plunder: 1101 AA 7.70, p. 584; Hebron: AA 10.33–34, pp. 748–50. 14 Forays: 1100, AA 7.42, p. 550; 1107, AA 10.47, pp. 760–62. 15 Raids: 1111, AA 12.3–4, pp. 826–28; 1112, AA 12.8, p. 836 and see also IQ, pp. 130– 31, who wrote that the treasure amounted to over 50,000 dinars and the Christians took 300 captives. Division of spoils: AA 10.31, p. 746. It is notable that Fulcher did not mention these raids at all, probably because he viewed them as routine exercises, but also his account of the later years of Baldwin’s reign became preoccupied by celestial portents and the biblical landscape. For more on Baldwin’s marriages see the concluding chapter: the matter of the unpaid Armenian dowry is also relevant to Baldwin’s need for funds. 16 Murray, ‘Origin of Money-fiefs’, p. 281. 17 D. M. Metcalfe, in Coinage of the Crusades and the Latin East in the Ashmolean Museum Oxford (London, 1995), described a single copper coin of Jerusalem that he speculatively identified as one minted for the coronation of Baldwin I (pp.  40–42); Robert Kool of the Israel Antiquities Authority questions its authenticity: pers. com. (email dated 13 December 2017).

The army, administration and allies  169 18 On the use of Fāṭimid dinars, see Robert Kool, ‘A Hoard Twice Buried? Fatimid Gold from Thirteenth Century Crusader Arsur (Apollonia-Arsuf)’, Numismatic Chronicle, 173 (2013), 261–92; challenging the conclusions of Jaroslav Folda, The Art of the Crusaders in the Holy Land, 1098–1187 (Cambridge, 1995). 19 Use of coins imported from Western Europe during and after the First Crusade: RA, pp. 111–12; Metcalfe, Coinage, pp. 12–14. 20 For a detailed analysis of nobles by place of origin, see Murray, Crusader Kingdom, pp. 101–07. See also the statistical analysis of witnesses to royal charters by Corliss K. Slack, ‘Royal Familiares in the Latin Kingdom of Jerusalem, 1100–1187’, Viator, 22 (1991), 15–67; Murray, Crusader Kingdom, pp. 112–13 and Appendix B, pp. 166–70. 21 Murray, Crusader Kingdom, p.  107 for fief holders; p.  198 for Geldemar Carpenel; p.  199 for Gerard of Avesnes; p.  233 for Walter Mahomet; p.  187 for Baldwin of St Abraham; pp.  211–12 for Hugh of Fauqembergues; pp.  201–202 for Gervase of Bazoches; p. 214 for Joscelin of Courtenay. 22 William’s parentage was related by OV, V: 282–83. He witnessed a royal charter in 1108 as ‘Willelmus filius comitis’ [RRH no 52]. His exploits in the Holy Land are in AA 10.47, pp. 760–62; 11.40, p. 814. Although OV reported that he ‘quickly met his end’, William’s date of death is not known. See also Murray, Crusader Kingdom, p. 237, for reasons this cannot have been Robert’s legitimate son William Clito. 23 For Eustace Granarius AA 9.48, p. 706; 11.10, p. 780; 11.40, p. 816; 12.6, p. 832; WT 11.14–15, pp. 518–19. Hamilton discussed and rejected the idea that ‘niece’ (neptis) was a euphemism for an illegitimate daughter of Arnulf (p. 63). See also Murray, Crusader Kingdom, pp. 193–95. 24 The anecdote about Walter Mahomet GN: Appendix 1 at pp. 355–60. Discussion and translation: S. B. Edgington, ‘Espionage and Counter-Espionage: An Episode in the Reign of Baldwin I’, in Crusader Landscapes in the Medieval Levant: The Archaeology and History of the Latin East, ed. Micaela Sinbaldi et al. (Cardiff, 2016), pp. 157– 67. See also Murray, Crusader Kingdom, p. 233. 25 GN Appx. 1, p. 359. 26 Steven Tibble, Monarchy and Lordships in the Latin Kingdom of Jerusalem, 1099– 1291 (Oxford, 1989). 27 For source references for these minor characters, see Murray, Crusader Kingdom (in alphabetical order): Anselm, pp.  181–82; Baldwin, p.  187; Gunfrid, pp.  206–207; Lithard, p. 216; Pisellus, p. 220; Roger, p. 227; Ulric, p. 231. 28 For Simon the Constable, see Murray, Crusader Kingdom, pp. 228–29. He may have been the ‘Simon’ whom Anna Komnene mentioned in relation to the diplomatic mission, but this man was also identified by AK as the king’s nephew (‘his brother’s son’), and this seems unlikely (AK, pp. 402–406). See also Lilie, p. 86 and nn. 105–106. 29 Dictionary definitions of dapifer from C. T. Lewis and C. Short, A Latin Dictionary (Oxford, 1879) and J. F. Niermeyer, Mediae Latinitatis lexicon minus (Leiden, 1997), both s.v. See Murray, Crusader Kingdom, p. 217, for Matthew; pp. 201–202 for Gervase; p. 211 for Hugh. 30 On the chancery: Murray, Crusader Kingdom: Robert, p. 225; Pagan, p. 218, and more generally H. E. Mayer, Die Kanzlei der lateinischen Könige von Jerusalem, Schriften der MGH 40, 2 vols (Hanover, 1996). 31 For the seal: Folda, Art of the Crusaders, p. 46 and Plate 4.1; p. 559, n. 23; also Robert Kool, ‘Civitas Regis Regvm Omnium: Inventing a Royal Seal in Jerusalem, 1100– 1118’, in Crusades Subsidia: Crusading and Archaeology, ed. Vardit Shotten-Hallel and Rosie Weetch (Abingdon, 2019), forthcoming. 32 Gerard the Chamberlain: Murray, Crusader Kingdom, pp. 199–200. 33 Baldwin the valet-de-chambre: WT 11.14, pp.  518–19. Where William obtained his story is not known. 34 For an overview, see Thomas Asbridge, ‘The Principality of Antioch and the Early History of the Latin East’, in East and West in the Medieval Eastern Mediterranean

170  The army, administration and allies

35 36

37 38

39 40

41 42

43 44 45 46

47

48 49 50

II: Antioch from the Byzantine Reconquest Until the End of the Crusader Principality, ed. K. Giggaar and V. van Aalst (Leuven, 2013), pp. 1–10. Christopher MacEvitt, ‘Edessa, County of’, in Crusades Encyclopedia, pp. 379–85, at p. 382. MacEvitt covered the internal development of Edessa under the rule of Count Baldwin II in Rough Tolerance, pp. 74–92. The sources differ as to whether the Turks surrounded Edessa and then moved away to fight the battle of Harran or whether they besieged Edessa after defeating the Christians. I have adopted AA’s very detailed version: 9.39–46, pp. 690–702; see also the accounts of FC: 2.27–28, pp. 468–77; RC, chaps. 148–51, pp. 710–12; WT 10.28(29)29(30), pp. 487–91; 11.7–8, pp. 505–7; IQ, pp. 60–61; AS, pp. 78–82. Richard of the Principate’s vice-regency was reported only by ME (p. 197) and AS, p. 80. AA 9.45–46, pp.  700–702. The anonymous Syriac chronicler wrote, less plausibly, that Baldwin and Joscelin remained in prison because ‘Tancred was angry with them’ (p.  80). The same chronicler included the elaborate story of Joscelin’s negotiations with the new governor of Mosul, pp. 80–82. AA 10.37–38, pp. 752–54. For the council of Tripoli, 1109, see AA 11.10–12, pp. 780–82. Although Albert is the only authority for the council of Tripoli, details of the division of the future county of Tripoli are in WT, who wrote only that they were the result of the intervention of ‘common friends’: 11.9, pp. 507–8. See also IQ, p. 89; Beihammer, Byzantium and the Emergence, p. 362. Lewis, Counts of Tripoli, pp. 39–40. The circumstances of William Jordan’s death are rather mysterious. FC blamed a ‘furtive arrow’ (2.41.1–2, pp. 531–32); IQ wrote that the assassin was another Frank (p. 89); AA (11.15, p. 786) and Caffaro (Breuis Historia, p. 123) were in agreement that the culprit was one of William’s squires. The siege and surrender of Tripoli: AA 11.13–14, pp. 782–86; FC 2.41.3–4, p. 532; WT 11.10, pp. 509–10. For details and references for the sieges of Beirut and Sidon, see chapter 7. Taef El-Azhari, ‘Mawdūd (d. 1113)’, in Crusades Encyclopedia, III: 809. Köhler discussed whether Mawdūd’s four campaigns should be characterised as ‘counter-crusades’, pp. 96–98 (he concluded that they should not). Mawdūd’s 1st attack: AA 11.16, p.  788; 11.18–25, pp.  790–98. See also AS, p.  82; ME, 3.45–47, pp. 203–207 (Matthew wrote that it was Baldwin and Joscelin who had encouraged Mawdūd to attack Tancred); FC 2.43, pp. 537–43; IQ, pp. 101–102. AA 11.22, pp. 794–96. Asbridge interpreted Baldwin’s ‘speech’ as quite possibly indicating a formal agreement between First Crusaders that established a confraternity: Creation of the Principality of Antioch, pp. 104–107. I am inclined to see it, first, as an invention of Albert, and second, if it bears any relation to what was said, as an appeal to shared Christianity culminating in an assertion of royal suzerainty. Mawdūd’s 2nd attack: FC 2.45, pp.  549–57; IQ, pp.  114–19; ME 3.51, p.  207; AA 11.36, 38–42, pp. 810–18. AA’s much more detailed account broadly agrees with FC’s, except that he wrote that the appeal to the king came from Baldwin of Edessa, not Tancred. As a participant, Fulcher has to be preferred on the point. See also the memoirs of Usamah ibn-Munqidh: An Arab-Syrian Gentleman and Warrior in the Period of the Crusades, trans. Philip K. Hitti (New York, 1929), pp. 97–98. IQ, entry for ah 505, pp. 127–28, followed by IA (i.161), who dates it Muḥarram ah 506 (July 1112). ME 3.54–55, pp. 209–11. For Pons: Lewis, Counts of Tripoli, pp. 72–91. Roger of Salerno was the son of Richard of the Principate who had been vice-regent in Edessa 1104–1108. Tancred was regent for the absent Bohemond until the latter’s death in March 1111

The army, administration and allies  171

51

52

53 54 55 56

57

58 59 60 61

62

and then for his son, a minor, who was in Apulia. Thus Roger became regent of Antioch. For an account of his regency, see Asbridge, Creation of the Principality, pp. 69–81. The Turkish invasion of 1113: FC 2.49, pp. 565–73; AA 12.9, 11–12, pp.  836–42; IQ, pp.  132–37; WT (following FC) 11.19, pp.  523–25; ME 3.62, p. 214. See also Taef El-Azhari, ‘Ṭughtigin (d. 1128)’, in Crusades Encyclopedia, IV, pp. 1204–205. When Riḍwān ibn Tutush, Saljūq ruler of Aleppo, died in December 1113 his son was still a child, and this made the city a target for other Turkish rulers. For the ‘crisis of 1115’, see Asbridge, Creation of the Principality, pp. 70–73, 123–25. By far the most detailed account of the events of 1115 is by Walter, the chancellor of Antioch: Galterii Cancellarii: Bella Antiochena [WC], ed. Heinrich Hagenmeyer (Innsbruck, 1896); Thomas S. Asbridge and Susan B. Edgington, Walter the Chancellor’s The Antiochene Wars: A Translation and Commentary (Aldershot, 1999). For the campaigns of 1115, summarised here, see especially WC 1.2–4, pp. 65–71, whose account from the point of view of a prominent Antiochene is full and escapes the more usual Jerusalem bias of the other Latin writers. It may be compared with FC 2.53.1–3, pp. 580–84; AA 12.19, pp. 852–54; ME 3.70, pp. 218–19; IA, pp. 172–73. The date of the battle was significant because it was the Feast of the Exaltation of the Holy Cross, which was considered particularly significant in the Crusader states. WC 1.3, p. 67. For Baldwin’s response, AA 12.20, p. 854. Jonathan Harris makes the point about the protection of the Holy Places, which had been a concern in the eleventh century when the Fatimids had governed Jerusalem: Byzantium and the Crusades, p. 75. For the 1102 legation, AA 8.47–48, pp. 634–36; see also Beihammer, Byzantium and the Emergence, pp. 331–32. Baldwin’s gift to the emperor of two pet lions sent with his envoys was probably to compensate for the emperor’s tame lion killed by Lombards who broke into the imperial palace (AA 8.4, p. 590). The clause about the safe passage of pilgrims was the only one recorded by Albert and Fulcher: AA 10.45, p. 758; FC 2.39.2, p. 524. Anna Komnene set out the full text of the treaty, but the safeguarding clause is not in it: AK 13.12, pp. 387–96. See also the discussion of the treaty by R-J. Lilie, Byzantium and the Crusader, pp. 75–82. The emperor’s anger over Tancred’s intransigence, his decision to isolate him and his despatch of a legation to Baldwin: AK 14.2, pp. 400–406; discussed in Lilie, Byzantium and the Crusader, pp. 84–86. For the negotiations outside Tyre and then in Jerusalem see above and AA 12.4, 7–8, pp. 828, 834. Lilie concluded his discussion of the alliance by alluding to the Pisans’ difficult position in Jerusalem following the eviction of Daibert of Pisa from the patriarchate that had pre-disposed them to enter into it: pp. 90–91. The 1101 gift and Daibert’s crime were reported only by AA: 7.62, pp. 572–74. The endowment for the hospital is the earliest on record: see Anthony Luttrell, ‘The Earliest Hospitallers’, in Montjoie: Studies in Crusade History in Honour of Hans Eberhard Mayer, ed. Benjamin Z. Kedar, Jonathan Riley-Smith and Rudolf Hiestand (Aldershot, 1997), pp. 37–54. For the marriage, FC 2.51.3, pp. 575–77; WT 11.21, pp. 525–27. Mayer examined the circumstances in exhaustive detail: ‘Etudes’, pp. 58–69. He suggested that Adelaide was probably forty or older in 1113. Mayer also devoted a lot of space in the same study to ‘proving’ that Baldwin was homosexual as an explanation for his childlessness. The question will be revisited in the next chapter. For a brief biography of Adelaide (under her birth name of Adelaide del Vasto), see Murray, Crusader Kingdom, p. 179. It is possible another incentive for a Sicilian alliance was shared hostility to

172  The army, administration and allies Byzantium, ‘My enemy’s enemy is my friend’: see Lilie for the view that for the Byzantines, ‘An alliance with Sicily, the fundamental enemy of the Empire, was at this time unthinkable’ (p. 91). 63 Adelaide’s arrival, AA 12.13–14, pp. 842–46. 64 Hamilton, The Latin Church in the Crusader States, p. 63. FC, briefly, 2.59.3, pp. 600– 601; AA 12.23–24, pp. 858–62; WT 11.26, pp. 534–35; 11.29, pp. 542–43.

10 Last years and legacy

Last years and legacyLast years and legacy

Introduction Accounts of Baldwin’s last few years are sketchy and sometimes contradictory. It appears, however, that his usual proactive campaigning was interrupted more than once by serious bouts of illness. He died in 1118, far from Jerusalem on an expedition into Egypt. This concluding chapter will recount his final years and consider what arrangements he may have made for his succession in the absence of a son. It will attempt a character assessment and evaluation of his reign and will consider his posthumous reputation.

Into Arabia The chief sources for the events of the last three years of Baldwin’s life remain Fulcher of Chartres and Albert of Aachen, although they have to be augmented by referring to William of Tyre’s later narrative. For example, Fulcher’s account of Baldwin’s expedition east of the Dead Sea in 1115 is short: Baldwin ‘set out into Arabia and built a castle in a strong position on a small mound’, placing a garrison in it ‘to command the land for the advantage of Christianity’.1 This castle, which he named Mons regalis (Montréal), was about three days’ journey from the Red Sea and about four from Jerusalem. Albert’s account is, as usual, more detailed, but unfortunately he conflated this first expedition with a revisit the following year, 1116, when Baldwin advanced beyond Montréal and reached the Red Sea. Nevertheless, it is Albert who revealed the primary purpose of the fortification: with 200 cavalry and 400 infantry in the course of eighteen days [Baldwin] established a new fortress, so that in this way he might more powerfully subdue the land of the Arabians, and passage to and fro would no longer be available for merchants except by the king’s favour and licence.2 In other words, Baldwin was looking to control commerce and to profit from the passage of caravans through the region. The garrison would also be alert to any signs of enemy aggression. William of Tyre’s report was based closely on

174  Last years and legacy Fulcher’s, although he was able to add from his later perspective that the castle became the site of a settlement, with its fertile soil producing grain, wine and oil.3 William included a chapter immediately following this, with information not found elsewhere, that King Baldwin was concerned about the depopulation of the city of Jerusalem that rendered it vulnerable to enemy attack. His remedy was to encourage the immigration of Christians from Arabia, where they were living under harsh Muslim rule (according to William). He resettled whole families and households, with their livestock, using as inducement better conditions, liberty and the allure of living in the holy city.4 In 1116, with some 200 knights, Baldwin headed east again, revisiting Montréal, and then south through the desert to the Red Sea. This expedition is thought to have founded a castle in the valley of Moses (Wadi Mūsa) and another at a town on the coast Fulcher called ‘Helim’ (modern Aqaba, Jordan) that they had found abandoned by its inhabitants.5 On his return to Jerusalem towards the end of 1116, the king became seriously  – he feared terminally  – ill. Fulcher wrote that this was the reason Baldwin repudiated his Sicilian wife; William expanded Fulcher’s simple statement of fact to involve Baldwin’s seeking the advice of the clergy and trying to explain himself to Adelaide, who was not appeased.6 William, of course, could see the longer-term consequences of the failure of the Sicilian alliance. Albert of Aachen’s independent account of the same expedition and its aftermath added some interesting details. He did not, apparently, know about the deserted town Helim, but wrote only that the king and his men bathed in the Red Sea when they reached it, as relief from the intense heat, and ate fish. While there, Baldwin heard about St Catherine’s Monastery in the Sinai desert, and he was keen to visit it for prayer and conversation, but the monks sent messengers to dissuade him because they did not want to draw attention to the monastery lest they be expelled from it by the Saracens. Although this tale is uncorroborated, it rings true, for St Catherine’s is a Greek Orthodox foundation and has maintained its presence until today partly because of a willingness to accommodate political realities.7 Baldwin abandoned his plan to visit the monastery and returned to Jerusalem via Hebron, pausing only to raid the plains of Ascalon for camels, cattle, sheep and goats on his way. Albert gave the date of the onset of Baldwin’s serious illness as the beginning of March 1117, technically still winter but later than Fulcher implied, and Albert wrote that the king was in Acre.8 Baldwin genuinely believed he was on his deathbed, for he ordered that his worldly wealth was to be distributed: part went to the poor, along with a dole of food and wine; part to his household; and part to his soldiers, both his own and those serving for pay. However, Albert claimed that the king made a full recovery. He did not report whether the gifts were revoked but did write that the Egyptian fleets that had put in at Tyre when the Saracens heard of the king’s illness now sailed home without attacking. Another discrepancy from Fulcher’s undoubtedly better informed account is that Albert placed Arnulf’s visit to Rome to exculpate himself after the king’s recovery from his sickness. It was Arnulf who then insisted, on the pope’s orders, that Baldwin repudiate his wife because his marriage to Adelaide was adulterous and

Last years and legacy  175 unlawful. Arnulf added a charge of consanguinity between Baldwin and Adelaide, although this had passed unremarked in 1113, and he formalised the process of annulment by convoking a council in the church of the Holy Cross in Acre. ‘Sad and grieving, released by synodal law from the marriage bond, the lady sailed back to Sicily’, while Baldwin, Albert claimed, exercised ‘wonderful abstinence and chastity’ from then onward.9 Thus, although the two writers differed on the exact sequence of events, they agreed that the king’s illness was a precipitating factor in his repudiation of Adelaide. Certainly, he may have felt the prick of conscience and wanted to die absolved of his sins. Nevertheless, it is probable that there was also considerable pressure from his vassals and the senior clergy: the marriage to Adelaide had not provided the king with an heir, and according to the terms of the agreement Roger of Sicily would inherit the kingdom of Jerusalem if the king were to die while the marriage endured.10 The reminder of the king’s mortality made the matter of the succession urgent, and it is inconceivable that the matter was dropped as soon as the king made a recovery. In the short term, Baldwin appeared to regain full health, and with his accustomed energy he was mindful that two coastal cities remained unconquered. First he built a castle called ‘Scandalion’ within five miles of Tyre and garrisoned it ‘to confine the city’.11 He then embarked on a major expedition early in 1118 aimed, so Albert said, at conquering Egypt and thus removing its support for Ascalon that threatened pilgrims going to and from Jerusalem.

The last campaign Baldwin set out with a large army to punish Egypt for wrongs the Egyptians had wrought against his realm. This is William of Tyre’s very brief introduction to his account of King Baldwin’s expedition into Egypt.12 Fulcher of Chartres omitted preliminaries altogether, stating only that towards the end of March in 1118, the king attacked and plundered the city of Pharamia (al-Faramā), a place not far (William later explained) from the Nile delta.13 As so often, we have to read Albert of Aachen for fuller information, and in this case Albert’s account is so rich in circumstantial detail that it can only have been learned from someone who was with Baldwin on his last campaign.14 At the beginning of spring 1118, Baldwin set out with 216 cavalry and 400 experienced infantry. In addition there were many non-combatants because, as Albert specified, they took supply waggons with them since they could not plunder in Arabia where the natives were allies whom Baldwin did not want to alienate. After an eleven-day journey the army reached the river Nile and took the time to bathe and wash away their sweat. On Thursday, 21 March, they arrived before al-Faramā. Albert made much of the ‘tiny army’ attacking the well-fortified town, but in fact it had been deserted by its inhabitants when they heard of the king’s approach and was easily taken. The townspeople had abandoned incredible quantities of food, as well as gold and silver and other wealth, so the army was able to recover after its long, hot march. On Sunday, 24 March, some cautious men advised that there were too few of them to stay so close to Cairo, and the king responded by ordering

176  Last years and legacy the destruction of al-Faramā. They rose early the following day and slighted the town walls and burned down the buildings. It is said that while the king was working more vigorously and enthusiastically than anyone else at this destruction [. . .] he was attacked by sickness in his limbs, which were troubled beyond measure by the heat and effort, and his physical distress began to increase more and more. By sunset Baldwin was convinced that he was dying; when he told his men they were in despair because they foresaw being killed in Egypt if the king was known to have died. As so often, Albert’s account can be corroborated from a completely independent source. Ibn al-Qulzumī, a Coptic Christian who died after 1127, described events thus: Baldwin, the leader of the Franks, arrived with a great army at al-Faramā. He pillaged it and he burned it, and he determined upon a sudden attack against Cairo [Miṣr]. Then he fell ill, and on the third day his sickness became serious. He ordered his companions to carry him and to return to the Levant [al-Shām]. Then they carried him and returned, and when they reached al-ʿArīsh, he died there. They cut open his belly and they salted him, as he had commanded them. And they returned with him to Jerusalem. When news of their arrival at al-Faramā reached the noble lord al-Afḍal, he raised a great army [against] them. When Baldwin, their leader, died, and they returned, the army pursued them to the Levant and returned.15 Ibn al-Qulzumī’s knowledge of Baldwin’s crude embalming demonstrates the effectiveness of Fāṭimid intelligence, perhaps employing Coptic Christians to infiltrate Baldwin’s camp. The added information from the Egyptian author is al-Afḍal’s raising of an army in response to the attack on al-Faramā and its pursuit of the Latins as they returned to Jerusalem. This makes the fears of the king and his followers entirely reasonable. Fulcher told the story a little differently: after the destruction of al-Faramā the king was walking by the river one day with friends, and some of the knights speared fish, which they took back to the camp to eat. It was after this that the king fell sick, blaming his old wound.16 It could be inferred that food poisoning was the cause of his illness, or over-exertion in the fierce heat, but at all events, this time Baldwin was correct in believing he was dying.17 Fulcher went on to explain that the army now decided to return to Jerusalem, and as Baldwin was too weak to ride they made a litter, or stretcher, from tent poles. They reached al-‘Arīsh, about fifty miles east of al-Faramā, where Baldwin died. Before this, Albert reported, the king made a long speech, ‘although he was in terrible pain’.18 He rallied his men and urged them not to bury him in the land of the Saracens but to carry his corpse back to Jerusalem to be buried next to his brother. Quite plausibly, his audience despaired of returning ‘even empty-handed’ to Jerusalem, let

Last years and legacy  177 alone carrying a corpse in the extreme heat. The king was insistent, and his solution was to give instructions for his embalming: ‘ “Open up my belly with a knife, take out my insides, preserve my body with salt and spices and wrap it in a hide or in carpets” ’. He summoned his household cook, Addo, and gave him detailed anatomical instructions, binding him with a promise to carry out the task. According to Albert (but not Fulcher or William of Tyre) Baldwin was asked about the succession while he was still lucid: this will be discussed later. Then Baldwin was administered the last rites and breathed his last. His soldiers wailed and wept, but controlled their grief to some extent for fear that the Saracens would know the king had died. They embalmed him as instructed, sewed the body into a hide as shroud and wrapped the whole in carpets. They then bound it upright onto a horse to deceive the enemy. The men succeeded in taking the corpse back to Jerusalem in this way: cautiously as far as Hebron and then ‘with banners flying and in battle formation’ across the plains of Ascalon to Jerusalem.

The funeral procession The cortège arrived at Jerusalem on 7 April 1118, which was Palm Sunday. By what the contemporary authors seem to have regarded as a coincidence, it arrived at a key point in the ceremony for Palm Sunday when two liturgical processions met on the Mount of Olives.19 In a re-enactment of Christ’s entry into the city, the custom was for the patriarch and senior clergy to go to Bethany before dawn and to approach Jerusalem via the valley of Jehosaphat. Other church communities would come from the church of the Holy Sepulchre to meet them outside the Golden Gate of the city. Albert conveyed the drama: ‘Suddenly the dead king was borne into the midst of the people as they sang. At the sight of him their voices were hushed and their praises brought low.’20 Fulcher likewise described singing giving way to grief, happiness to lamentation. He added that Syrians and even Saracens wept along with the Franks. Nevertheless, the Palm Sunday service was completed, and then the king entered the city for the last time, carried in through the Golden Gate. By this time his body was stinking, and so it was buried hastily, in accordance with Baldwin’s wishes, next to his brother Godfrey in the Calvary chapel of the church of the Holy Sepulchre. There can be no reasonable doubt that the drama of the dead king’s arrival was contrived. News of the funeral train, as it marched with banners flying across the plains of Ascalon, had undoubtedly flown ahead of it, and messengers would have been despatched to and from the city. The most likely person to have taken charge was Patriarch Arnulf, who was preparing for the important ritual of Palm Sunday. He would have recognised the significance of the funeral train’s approach so close to this date, and he almost certainly arranged for its arrival to coincide with the Palm Sunday procession. Thus King Baldwin’s final entry into the city was his last royal adventus. As Iris Shagrir explained, ‘The ceremony of adventus, originating in classical antiquity, was the solemn celebration of the arrival of the ruler into a city.’21 Palm Sunday itself celebrated Christ’s adventus into Jerusalem and had become an annual ritual for the city. When the king was present he would

178  Last years and legacy take a prominent part: Albert recorded that in 1112 after Baldwin abandoned the siege of Tyre he entered Jerusalem through ‘the gate which looks towards the Mount of Olives, through which Lord Jesus entered riding on a donkey’. Significantly, Baldwin then spent Holy Week in the city, and on Easter Day he celebrated, ‘wearing his crown solemnly and regally in honour of the legates of the king of the Greeks and on the instruction of the lord patriarch’.22 In 1118 there was again a fusion of religious and political ceremonial, and again Arnulf appears to have been directing the spectacle. It was his final act as the king’s loyal supporter and adviser, for he fell ill and died only three weeks after Baldwin.23

Tomb and epitaph The sources agree that Baldwin was buried at once, next to Godfrey in the Calvary, or Golgotha, chapel of the church of the Holy Sepulchre. Albert described magnificent tombs for both brothers, made of white polished marble.24 Jaroslav Folda proposed, on stylistic and other grounds, that it was Baldwin I who commissioned Godfrey’s tomb early in his reign, in which case Baldwin may have stipulated, or provided for, the place and construction of his own mausoleum or had it built in advance. Godfrey’s was on the right, Baldwin’s on the left side of the entrance to the Adam chapel. Unfortunately there are no depictions of Godfrey’s and Baldwin’s tombs dating earlier than the sixteenth century, and Folda considered it likely that the royal tombs were destroyed by the Turks in 1244 and that these later drawings show them after reconstruction. However, he argued from their type and style that the renovations probably reflected the originals quite closely, in which case they were rectangular in plan and topped by a low canopy, like a roof, supported by short columns. An epitaph was carved on one side of the roof cover.25 According to the pilgrim Theoderic, who saw the epitaph of Baldwin in 1172, it comprised five lines and read as follows: HIC EST BALDUWINUS ALTER IUDAS MACHABEUS, SPES PATRIE, DECUS ECCLESIE, VIRTUS UTRIUSQUE, QUEM FORMIDABANT, CUI DONA TRIBUTA FEREBANT CEDAR ET EGYPTUS, DAN AC HOMICIDA DAMASCUS. PROCHDOLOR, IN MODICO CLAUDITUR HOC TUMULO Here is Baldwin, a second Judas Machabeus, hope of his country, ornament of the Church, glory of both. Cedar and Egypt, Dan and murdering Damascus feared him and brought him tributary gifts. Alas, he is enclosed in this little tomb.26 Fulcher of Chartres wrote his own longer tribute, which he entitled Epitaphium regis Balduini, although there is no evidence that it was ever literally inscribed on the king’s tomb: When that king fell, pious Frankish people wept, whose shield he was, whose strength and support.

Last years and legacy  179 For he was his men’s defender, a terror to the enemy and an enemy to them; a mighty leader of his country, just like Joshua. He took Acre, Caesarea, Beirut and Sidon from the wicked native enemy. Afterwards he added to his dominion and subjected to his allegiance the lands of the Arabs – or those that touched the Red Sea. He also took Tripoli, but no less threatened Arsuf, and besides he accomplished many things held in high esteem. There is no correspondence with the carved wording, so it appears that the poem was Fulcher’s personal tribute to the king’s achievements.27

Twelfth-century assessments of Baldwin Fulcher was the writer who knew Baldwin best, but his ‘epitaph’ was a verdict on him only as a militarily effective king; here and elsewhere he wrote very little about Baldwin’s appearance or his character as a man. Fulcher’s translators could cite only one criticism, from the northern campaign of 1113, when Fulcher wrote that because the king had not waited for his allies, ‘they were all very dismayed and they blamed the king’s intemperance, because he had advanced against the enemy in a disorderly and rash way without waiting for their advice and support’. Notably, the judgement was inescapable, yet even so Fulcher distanced himself by putting it in the mouths of Roger of Antioch and Pons of Tripoli.28 The other writer who is thought to have known Baldwin personally is Ralph of Caen. Ralph’s eulogy of Baldwin has already been quoted at length; it is worth repeating that although it was inserted into the narrative for 1097, it is quite clear that it was written while Baldwin was king (‘he was drawn by divine destiny to sit upon the throne of David’). Therefore, when Ralph described Baldwin thus: ‘a man generous with money, devoted to the study of military affairs, humble in his speech, outstanding in his magnanimity; in appearance he had the figure of a nobleman from head to toe’, a large helping of flattery must be suspected.29 To Albert of Aachen, writing far away in the Rhineland, Baldwin was the complete hero, even though it is from Albert’s history that we learn most about the king’s ruthlessness and ambition. It is not to be expected that the king’s deathbed would evoke a balanced judgement on his life, anyway, and so it was. Albert called Baldwin ‘a man of great wisdom’ and ‘a man who was the noblest of noble blood in his native land Lotharingia, most splendid and victorious king in the kingdom of Jerusalem, most valiant champion of God’.30 As a contrast to this emphasis on Baldwin’s military prowess, there is the testimony of the Russian pilgrim Abbot Daniel, who described the king’s regal presence, but also his kindness, humility and piety. No doubt Baldwin had put on a great show for the visiting dignitary, but it may also be suspected that Daniel wanted to give the impression of intimacy with the king.31 Apart from William of Tyre, whose characterisation of Baldwin merits its own section, there is a lack of relevant sources written after the years immediately following Baldwin’s death. Only the Historia Nicaena vel Antiochena necnon

180  Last years and legacy Jerosolymitana fills the gap. The prose body of the Historia is based closely on Robert the Monk for the period to the battle of Ascalon in 1099 and then on Fulcher of Chartres’s second book for the rules of Godfrey and Baldwin I. However, there is also a verse prologue that summarised in just thirty-four lines the First Crusade and the history of the kingdom of Jerusalem, from the capture of Nicaea to the reign of Baldwin III (1143–1163), who commissioned the Historia three years into his reign (line 29). It is, of course, designed to position Baldwin III in an illustrious royal succession and may well have suppressed some less savoury details of his forerunners’ careers, but nevertheless: Duke Godfrey held this kingdom for a single year; For eighteen years Baldwin ruled the city; These two brothers, worthy of the kingdom’s crown, Glorious in war, fearsome in combat, Fought many battles with their few men, Scattering and slaying swarms hostile to them, And through God they were mighty against the Egyptian enemy. They enlarged the kingdom that they then held. Let them now be blessed in Christ’s kingdom without end!32 The most interesting aspect of this view from one generation after Baldwin’s death is how closely he was associated with his brother Godfrey and their individual achievements allocated to both.

William of Tyre William was writing yet another generation further from Baldwin’s death, in the 1170s, but he wrote by far the most detailed description of the king. It must, however, be treated with considerable caution, not only because it was written more than half a century after the king’s lifetime, but also because William included physical descriptions and character sketches of all the rulers of Jerusalem, from Godfrey to Baldwin IV; these were clearly modelled on Einhard’s description of Charlemagne, and Einhard in turn took Suetonius as his exemplar.33 In addition, William was influenced by certain conventions and prejudices. Nonetheless, his portrayal provides a good starting point for a discussion of Baldwin. It stands at the beginning of Baldwin’s reign. The depiction begins with a physical description of Baldwin. He was very tall, much more so than his brother; his hair and beard were dark but his complexion rather white; he had an aquiline, quite prominent nose; his upper lip with the row of teeth underneath it was very slightly sunken, but not so much that it might be considered a blemish. Baldwin’s deportment was grave, and he was serious in his dress and speech, always wearing a cloak, and because of this strangers took him for a bishop rather than a layman. Credibly, William may have learned all this from oral tradition, but he may also have made some assumptions based on his knowledge of Baldwin’s clerical training and his ideas

Last years and legacy  181 about suitable regal bearing. William went on directly to contrast Baldwin’s dignified appearance with his reported behaviour: ‘he is said to have struggled with weakness of the flesh’.34 Despite his struggle, William conceded, the king managed these things so carefully that he offended or harmed no one; indeed, the matter just came to the attention of a few of his body-servants. Two things to note about this: first, it is gossip; second, the word translated here as ‘bodyservant’ is cubicularius, the same word as William used for the attendant who conspired with the Tyrians to take the king’s life. Some writers have constructed a scenario around these two passages to conjecture that Baldwin’s ‘weakness of the flesh’ was homosexual in nature. There is very little evidence to support this, and the matter is anyway of dubious relevance since contemporaries did not comment on it or apparently condemn it. A letter from Anselm of Canterbury, probably dated 1104 or 1105, that Mayer adduced as additional evidence does appear to reproach the king, but it is lacking in specifics. The letter began with the wish that Baldwin might reign over earthly Jerusalem in life as over heavenly Jerusalem after death. Anselm went on to acknowledge Baldwin as a man of education and the will to live a good life, yet in need of his ‘friendly advice’ (amica monitione). He reminded the king that he was ruling the land of the Old Testament kings and prophets as well as of the Redeemer, the King of Kings. The he came to his message: Your highness should consider, therefore, how outstanding is the grace of God that wanted you to be king in this city; and with what desire, what eagerness the king should submit to God’s will and His service, who has set him in that place. Therefore I pray you, I implore you, I admonish you as my lord and one I love, to strive diligently so to govern both your own person and all those subject to you in accordance with the law and the will of God, so that you set in your own life a shining example to all the kings of the earth.35 The letter ended with a prayer, ‘May our Lord Jesus Christ so reign in your heart and in your works that you may reign in heaven without end with King David your predecessor. Amen.’ It is evident that Anselm had heard rumours about Baldwin’s conduct that gave cause for concern, but their nature is unknown. It is possible he had heard a version of the garbled story about Baldwin’s repudiation of his Armenian wife. However, it is more likely that Anselm was concerned about Baldwin’s treatment of Daibert and the Church of Jerusalem. It should be read in conjunction with the only other letter from Anselm to Baldwin, probably sent in the spring of 1102. After Anselm expressed his joy that God had chosen Godfrey and Baldwin to rule in Jerusalem he urged Baldwin to reign according to God’s will, by which he meant: You should not think, as many bad kings do, that God’s Church has been given to you as if to a lord to be served, [but rather] that it has been entrusted to you as its advocate and defender. There is nothing in this world that God loves more than the liberty of His Church.36

182  Last years and legacy The admonition continued at some length, suggesting strongly that Anselm had already some unease about Baldwin’s intentions towards the patriarch and the Jerusalem Church. After his long but ambiguous comment on Baldwin’s private behaviour, William returned to the public persona. Baldwin was medium in build, ‘quick to take up arms, agile on horseback, tireless and diligent whenever the affairs of the kingdom called’. In fact, William continued, he excelled altogether in magnificence, spirit and military experience, an inheritance he and his brothers shared. Above all, Baldwin manifestly emulated the example of his brother Godfrey. However, William did not finish on such a positive note. He was ready to blame Arnulf of Chocques for any aberrations, and William concluded his very long sentence that had begun by praising Baldwin: . . . apart from the fact that Baldwin was reproached because he was too familiar with, and was ruled by the advice of, a certain Arnulf, a wicked man and the worst, who was archdeacon in Jerusalem, whose every deed and inclination was said to be for evil, concerning whom we made mention above because he had usurped the patriarchal chair.37 It is not unusual for biographers to blame the wayward actions of their subjects on poor advice, but William went far beyond this in his attack on Arnulf. The reason assuredly lay in Baldwin’s dealings with Daibert of Pisa and the patriarchate of Jerusalem. Arnulf, as a realist, backed Baldwin and recognised that the king had to control the resources of the nascent kingdom in order to defend and expand it. William, from his clerical stance and the later perspective of a relatively stable polity, did not accept this. This is not to argue that Arnulf had no influence with the king, only that it would be wrong to consider him some sort of éminence grise. As has been shown, Arnulf was immensely loyal from the instant of Godfrey’s death to the reception of Baldwin’s funeral train into Jerusalem seventeen years and nine months later. How far he contributed to policy-making cannot be known, but he appeared frequently, especially in the more positively disposed Albert of Aachen, in supporting roles as stage manager of public events, from the 1101 failure of the Easter Fire to the Palm Sunday processions of 1112 and 1118. William blamed Arnulf for both the Sicilian marriage and its dissolution, but the evidence is that the nobility in general favoured the match in 1113, while in 1116 and 1117 opinions had changed and Arnulf was obliged to negotiate a way out of an unpopular marriage and take the fall for it.

Baldwin’s marriages Baldwin’s serious illness in the winter of 1116 to 1117 undoubtedly focused attention on the succession. The king believed he was about to die, he had no living heir of his body and it was evident that the kingdom’s enemies would take advantage of any interregnum or confusion. Before addressing the problem as it presented at the time, a few more words about the absence of a direct heir to the throne.

Last years and legacy  183 Mayer interpreted every aspect of Baldwin’s marital history as confirmation of his homosexuality, writing at length about his ‘scandales conjugaux’.38 For example, when he recounted the death of Baldwin’s Norman wife, Godevere (Godehilde), at Marash, Mayer dismissed the idea that they had children who died there too, completely rejecting Runciman’s assumption that Baldwin was accompanied on crusade by his children as well as his wife and claiming that this is ‘d’une grande importance en ce qui concerne l’homosexualité possible de Baudouin’.39 A more considered view is that even if Baldwin was gay, this did not preclude his having sexual relations with women, especially in view of the dynastic responsibility associated with being Godfrey’s heir, probably as early as his first marriage as argued in chapter 1 above. Baldwin’s second marriage, no less than his first, was a political match. It offered an alliance with the nobility of Edessa and the prospect of a substantial dowry. At the time, Baldwin’s future seemed to lie in Edessa. Godfrey and the crusaders had not yet captured Antioch, still less Jerusalem, and Godfrey’s election to rule in Jerusalem could not be foreseen. In the absence of evidence to the contrary, it is reasonable to expect that Baldwin would want to have sons to inherit his new county. However, only some year and a half later Baldwin was summoned to Jerusalem as heir to his brother. At once he went from wanting a son to establish his comital dynasty to needing one to succeed him as king of Jerusalem. For this reason, presumably, he sent his wife by sea while he made the more hazardous overland journey to Jaffa. As previously related, she was still in Jaffa when false news arrived that the entire Christian army had been defeated at Ramla in September 1101, and she sent a letter on behalf of the townsmen to appeal for Tancred’s help, as well as in the following year when Baldwin fled after the second battle of Ramla, and she was reported as ‘considering flight’ with the rest of the citizens. Mayer reported both reactions as quasi-treasonous on the queen’s part.40 At some point in the same decade, Baldwin and Arda (to use the convenient name) separated. Guibert of Nogent, uniquely, reported the gossip about this: that the queen was ordered by her husband to come to Jerusalem; she sailed to St Symeon where she transferred to a faster ship that was driven ashore on a ‘barbarian’ island.41 The islanders seized her and detained her for a long time, after which they allowed her to depart. When she finally reached her husband, ‘the king suspected “ethnic incontinence”, not without reason, and at once abstained from her bed’. Arda was put into the convent of St Anne, and the king ‘rejoiced to live celibate’. Leaving aside for the moment the king’s alleged suspicions, there are aspects of this story that have not, to my knowledge, been examined. Arda arrived by sea at St Symeon (marina evectione), so where had she sailed from? The fact that she transferred to a faster vessel (celeriorem carinam) suggests she had made a substantial sea voyage, for example from Constantinople. This would make sense, for her father had absconded to Constantinople with the unpaid, greater part of her dowry. Baldwin, who was chronically short of money, may have sent his wife to pressure her father into paying what was due. This would suggest a level of trust between Baldwin and Arda, confirmed if she responded to his command to

184  Last years and legacy come to Jerusalem. Neither Fulcher of Chartres nor Albert of Aachen included the story of the sea voyage and kidnap. William of Tyre reported the separation thus: He set aside his lawful wife, who was not convicted, nor had she confessed, . . . without any examination of the case and without regard to the laws of matrimony, and compelled her by force to become a nun in the convent of St Anne.42 William went on to say that many different reasons had been advanced for the separation, citing just two: first, that Baldwin wanted to marry a richer and more noble wife to alleviate his burdensome poverty; second, that the queen had angered the king by disregarding the bonds of marriage. After a while the queen approached the king with ‘fabricated excuses’ and obtained his permission to visit her kin in Constantinople and seek funds from them for the convent of St Anne. Once in Constantinople she laid aside her nun’s habit and embraced an immoral life. The accounts of Guibert and William, the one separated by distance and the other by time from the scandal, share certain features, most importantly the legitimacy of Baldwin’s marriage to Arda so that the separation a thoro was not an annulment, which was the only way Baldwin would be free to take a third wife. Yet Baldwin did not seek dissolution of the marriage from the pope. Admittedly, it would be difficult because the usual convenient ground, consanguinity, was not available. Nevertheless, the curia was likely to be sympathetic, especially if adultery could be proven. (Notably, non-consummation does not seem to have been considered, at least before Mayer.) In this interpretation of Guibert’s tale and as made explicit by William, the motif of chasing the unpaid dowry in Constantinople recurs. It seems likely that failure to lay his hands on Arda’s dowry in full was the real reason behind Baldwin’s repudiation of his wife. Importantly, by consigning his wife to a nunnery, Baldwin was abandoning hope of engendering an heir. For somewhere between five and ten years he lived, apparently, a celibate life. Baldwin’s bigamous marriage to Adelaide of Sicily appears ill advised on almost every level. It is easy to interpret it as a last desperate attempt to make a political alliance that would bring wealth and other resources for the defence of the kingdom. In the short term, it was successful in this: the resources Adelaide brought with her to Jerusalem were described towards the end of the previous chapter, along with Baldwin’s rejection of her four years later. There is no information about how much time the royal couple spent together. On this and the whole subject of Baldwin’s third marriage the testimony of William of Malmesbury is interesting, although – or because – it differs so radically from William of Tyre’s.43 First, while conceding that the match was to make good Baldwin’s losses, he wrote explicitly that it was ‘for legitimate marriage’ (ad legitimum conubium). William agreed with the other chroniclers about the great riches Adelaide brought with her ‘to the king’s bedroom’, but added rather sourly ‘where the woman had amassed such infinite piles of precious goods from might seem a matter of wonder to anyone’. Baldwin ‘admitted her to his marriage bed’ (illam thoro recepit) but dismissed her soon afterwards. The reason for this, ‘they said’ (aiunt), was that

Last years and legacy  185 she was afflicted by an incurable illness and a cancer had consumed her genitals. From this William concluded: ‘One thing is certain, the king was without offspring; it was no wonder if a man for whom to be at leisure was to become unwell shrank from wifely embraces and spent his whole life in battles.’ The inescapable ambiguity is whether Baldwin dreaded all wifely embraces, or only Adelaide’s, as appears to be the implication. However that may be, and even if the story of the cancer, which William reported as hearsay, is completely untrue, William may have encapsulated a truth: throughout his reign Baldwin’s energies were absorbed by constant military campaigns, and in his last years he may have been suffering increasingly from illness whenever a pause from them allowed.

The succession It is inconceivable that following the dissolution of his marriage to Adelaide, and in view of his serious illness, Baldwin gave no thought to the succession before he left on a perilous expedition into Egypt. There was almost certainly pressure from his vassals and advisers, too, although this can only be judged from the division of opinions that followed his death. Two candidates were in the frame to succeed Baldwin as king: his brother Eustace, now back in Normandy, and his kinsman Baldwin of Bourcq, the second count of Edessa. The contemporary sources do not give an entirely clear view of the sequence of events, but Mayer has provided a plausible reconstruction which is largely followed here.44 Albert of Aachen wrote that the nobles who were with Baldwin when he was dying asked him to designate his successor, for the avoidance of dispute. Baldwin named Eustace, ‘if by any chance he would come’ (si forte uenerit). If Eustace’s age prevented his coming, then the king said that Baldwin of Edessa should be chosen. Albert then explained that on the very day of Baldwin I’s funeral the clergy and people of the Church decided that it was too dangerous for Jerusalem to be without a ruler. There was evidently a debate (‘various people said various things’) before there was universal approval for Baldwin of Edessa. Albert thereupon revealed that Baldwin was already in Jerusalem, having arrived for the celebration of Easter. Patriarch Arnulf anointed and consecrated him as ‘king and lord of Jerusalem’ on Easter Day – that is, just one week after Baldwin’s corpse had arrived in the city. The new king’s vassals did fealty for their possessions and made an oath of allegiance.45 William of Tyre’s account does not essentially disagree with Albert’s, but provides much more detail about the debate in Jerusalem. Baldwin of Edessa came to Jerusalem for Easter, and on the way he was met by a messenger with the news of King Baldwin’s death.46 He continued his journey, arriving on Palm Sunday via the Western Gate at the same time as the late king was carried in on the eastern side. After the king’s burial lay lords and clergy assembled, including Joscelin of Tiberias and Patriarch Arnulf. Sides were taken: one party held out for Eustace, in accordance with ‘the ancient law of hereditary succession’ (presumably as remembered from western Europe, since Baldwin I had been the first king of Jerusalem). Others pointed out that it would be dangerous to have an interregnum, with no one to lead out the army or bring it back or to carry out the business of

186  Last years and legacy government. According to William, it was Joscelin and Arnulf who took decisive steps to end the dispute by pointing out that Baldwin was present and an ideal candidate. William was cynical about Joscelin’s motivation, accusing him of offering Baldwin his support in return for his own succession to the county of Edessa (and he was suspicious of Arnulf because he was Arnulf). The two men persuaded the rest of the assembly to their point of view, and Baldwin was elected unanimously. The ‘unanimous’ decision may have been possible only because some of the proEustace party had already left to offer him the throne. They convinced Eustace that it was his duty to accede, and, very reluctantly, he set out and travelled as far as Apulia. There he heard that Baldwin II had already been made king, and although the envoys did their best to persuade him to carry on, he turned around and went home, no doubt with great relief. Fulcher of Chartres wrote so briefly on the election that it can be quoted in full: Therefore, after the death of King Baldwin the Jerusalemites immediately came to a decision and elected as king for themselves Baldwin – that is, the count of Edessa, kinsman of the dead king – so that they would not be thought weaker because they lacked a king. By chance he had crossed the river Euphrates and arrived in Jerusalem to speak with his predecessor. He was elected by general agreement [communiter] and consecrated on Easter Day.47 Fulcher gave no hint of a debate on the choice of heir, perhaps because he was writing in the reign of Baldwin II, but he did indicate a reason for Baldwin’s presence in Jerusalem: he was there to talk to the king, presumably at the king’s invitation and very probably to discuss arrangements for the succession.48 This would explain Matthew of Edessa’s account of the succession, assuming that Baldwin of Bourcq did not reveal in Edessa that his pilgrimage to Jerusalem had an additional purpose. Matthew wrote that the king had marched into Egypt and that on his journey home he fell sick and died, having commanded: ‘Send to Edessa and get Count Baldwin and set him up as regent of Jerusalem until my brother arrives from the Franks, in which case make him your king.’49 When the king’s men found that Baldwin was already in Jerusalem, Matthew reported, they were surprised, but also pleased. They offered him the regency, but he refused, wanting the crown for himself. The Franks then agreed that if Eustace failed to arrive within a year, Baldwin would be king. The difficulty with Matthew’s account is that the Latin sources agree that Baldwin was consecrated king at Easter 1118, and there could be no going back on this. To sum up, it seems improbable that Baldwin had given no thought to the succession. Arguably, his last expedition into Egypt when his health was evidently declining was irresponsible. However, it is unlikely that Baldwin of Bourcq’s presence in Jerusalem was coincidental and very probable that the purpose of his being there was to make arrangements for the succession. King Baldwin was aware that his brother Eustace was too elderly to want to make the journey, but he probably wanted to preserve the hereditary principle established by his own

Last years and legacy  187 succession to Godfrey by ensuring that the crown was offered to Eustace when his brother died. Since that would entail a lengthy interregnum, the best candidate for regency was Baldwin, geographically much nearer and an experienced ruler as well as a kinsman. King Baldwin’s death on campaign precipitated a crisis, and Baldwin of Edessa profited from the sense of apprehension in Jerusalem and, it must be said, Eustace’s good sense in renouncing his claim to the throne when he realised that continuing to Jerusalem could precipitate a civil war.

Conclusion King Baldwin I’s reign started as it began, with a succession dispute. Nevertheless, the intervening period, something short of eighteen years, had ensured that there was a kingdom to dispute and to inherit. Baldwin’s reputation has suffered by comparison with his brother Godfrey, who was remembered as a valiant crusader and a saintly character for centuries afterwards. Baldwin’s crusade had lacked the necessary consummation, for he was absent from the siege of Jerusalem in 1099.50 Yet despite Godfrey’s virtues, it was fortunate for Jerusalem that he died after only a year of ruling the infant kingdom of Jerusalem. His successor was singleminded in appropriating the kingship because he knew it was needed to defend and expand the realm, and he was effective in organising its limited resources to the same end. Insofar as his character emerges from the often conflicting accounts of his life and reign, he was neither a ‘nice’ man nor a good husband, but this is beside the point. Hans Eberhard Mayer observed, ‘C’était un homme politique avisé, à l’âme calculatrice, sans scrupules, violent, poursuivant froidement ses intérêts et qui dirigeait ses vassaux d’une main de fer.’51 These were precisely the ruthless qualities needed to establish and preserve the kingdom of Jerusalem. Over a century ago, W. B. Stevenson summed up Baldwin’s achievements in a paragraph that deserves to be better known: in brief, Stevenson recognised Baldwin’s effective use of his scant resources, his personal courage and military expertise, his establishing a stable government. In addition, ‘it was his determination and, indeed, his high-handed treatment of opponents that shattered the project of an ecclesiastical or papal state in Palestine’.52 It is doubtful that such a state could have survived for any length of time at all. It is impossible to say what drove Baldwin, whether personal ambition, dynastic loyalty or even crusading piety, but for the Latin kingdom of Jerusalem he was indubitably the right man at the right time.

Notes 1 FC 2.55, pp. 592–93. 2 AA 12.21, p.  856. For a description, photographs and plans of Montréal (modern ash-Shaubak, Jordan), see Hugh Kennedy, Crusader Castles (Cambridge, 1994), pp. 23–25. 3 WT 11.26, p. 535. 4 WT 11.27, pp. 535–36. 5 FC 2.56.1–3, pp.  594–95; WT 11.29, pp.  541–42; Kennedy, Crusader Castles, pp. 25–26.

188  Last years and legacy 6 FC 2.59.3, pp. 600–601; WT 11.29, pp. 542–43. 7 AA 10.21, pp. 856–58. St Catherine’s history began in 337 when St Helena ordered the monastery to be built on the believed site of Moses’ Burning Bush. It was fortified by Emperor Justinian in the sixth century, as recorded by Procopius (Buildings, V.8). See also the lavishly illustrated guide published by the monastery: St Catherine’s Monastery (Sinai, 1985). 8 AA 12.22–24, pp. 858–62. 9 AA 12.24, pp. 862. 10 William of Malmesbury claimed that Baldwin dismissed Adelaide because she had an incurable cancer, but whether he had privileged information is not known: WM 1: 688–89. The succession is discussed at some length later in this chapter. 11 The fortress at Tyre: FC 2.62, pp. 605–606; WT 11.30, p. 543. Fulcher wrote that the name meant ‘field of the lion’, but William called it Alexandrium and explained that Scandalion derived from ‘Scandar’, the Arabic name for Alexander the Great. 12 WT 11.31, pp. 543–44. 13 FC 2.64.1, pp. 609–10. Fulcher’s Pharamia is classical Pelusium, modern al-Faramā, Egypt; Albert added that the town was three days’ march from Cairo. 14 AA 12.25–28, pp. 862–70. 15 Ibn al-Qulzūmī, in Ibn al-Muqaffa’, History of the Patriarchs of the Egyptian Church, ed. and trans. ‘A. S. Atiya, Y. ‘Abd al-Musiḥ and O. H. E. Khs.-Burmester (Cairo, 1959), III 1: 35. I am very grateful to Matthew Barber for the reference and additional information about Ibn al-Qulzūmī and other historians’ treatment of this episode. He observes that later writers omitted al-Afḍal and the army he raised from the narrative (pers. com. by email 7 January 2017). For the Coptic Church as ‘a pillar of the [Fāṭimid] state’, Michael Brett, ‘Al-Karāza al-Marqusīya: The Coptic Church in the Fatimid Empire’, in Egypt and Syria in the Fatimid, Ayyubid and Mamluk Eras, Vol. IV, ed. U. Vermeulen and J. van Steenbergen (Leuven, 2005), pp. 30–60. 16 FC 2.64.2–3, pp. 610–12. 17 Frutolf of Michelsburg, like FC, blamed a longstanding illness: F-E, p.  338, as did Ibn al-Qalānisī: entry for ah 511, p. 157. Ibn al-Athīr wrote that the king swam in the Nile and a former wound reopened: IA, p. 196. The Historia Nicaena vel Antiochena necnon Jerosolymitana, written in 1146 or 1147 for King Baldwin III, linked the king’s death in 1118 explicitly with the wound Baldwin incurred from the Saracen ambush in 1103, bringing details of the injury drawn from FC into the account of Baldwin’s death from the same source: chap. 77, in RHC Occ, 5: 139–85, at 183. 18 AA 12.27–28, pp. 866–70. 19 See FC 2.64.4–6, pp.  612–14; Fulcher called the circumstances inopinabilis, i.e., ‘unthinkable’ or ‘unexpected’. 20 AA 12.29, pp. 870–72. For a detailed description of the liturgy for Palm Sunday, see Iris Shagrir, ‘Adventus in Jerusalem: The Palm Sunday Celebration in Latin Jerusalem’, Journal of Medieval History, 41 (2015), 1–20. 21 Shagrir, ‘Adventus’, pp. 12–15. 22 AA 12.7, p. 834. The word translated as ‘on the instruction’ in AA 12.7 is iussu, which might be more accurately translated as ‘at the command’ or ‘on the orders’: there will be more on Arnulf’s importance later. 23 Fulcher reported Arnulf’s death along with the deaths in 1118 of Pope Paschal (21 January), Adelaide of Sicily (16 April), Alexios Komnenos (15 August) and, of course, King Baldwin: FC 2.63.4, p. 608 and nn. 24 AA 12.29, pp. 870–72. Ekkehard of Aura, describing what he saw before he returned to Germany c. 1115, wrote that Godfrey’s tomb was built of ‘Parian stone’, i.e., marble: Hierosolymita, p. 27. 25 Folda, Art of the Crusaders, pp. 37–40; 74–75: plate 3.1a and plate 4.19 for drawing of Godfrey’s and Baldwin’s tombs from Horn, MS Vat. lat. 9233; p. 39, fig. 1 shows the

Last years and legacy  189

26

27

28 29 30 31 32 33

34

35

location of the tombs of crusader kings of Jerusalem. The general shape of the tombs is confirmed by a cutaway drawing of the seventeenth century: Vicenzo Favi, Relatione del Viaggio di Gerusalemme, London: British Library Add. 33566, fo. 90, reproduced in Chronicles of the Crusades, ed. Elizabeth Hallam (London, 1989), p. 107. There is no indication of an inscription on either tomb in Favi’s drawing. Theodericus, in Peregrinationes tres: Saewulf, John of Würzburg, Theodericus, ed. R. B. C. Huygens, CCCM 139 (Turnhout, 1995), pp. 143–97, at 154. This is the same wording as that written on the roof of the tomb in the Horn manuscript drawing; although it is at least theoretically possible that Horn in the sixteenth century took the epitaph from Theoderic, it is more likely that his drawing corroborates the pilgrim’s copy of the words. NB Cedar was a synonym for Arabia and probably made reference to Old Testament verses such as Jeremiah 49:28; Isaiah 21:17. FC 2.64.7–8, pp. 613–14; paragraph 8 is a four-line continuation that refers obscurely to the date of Baldwin’s death: ‘The king held the kingdom for eighteen years; / after this he became what he was, so that he might become what he was. / Phoebus had seen the constellation of Aries sixteen times / when the excellent king Baldwin died.’ Edbury and Rowe observed that ‘it is no doubt significant that at the end of [Baldwin I’s] career William refrained from giving him the sort of eulogy he had for Godfrey, Baldwin II and Baldwin III’: p. 74. FC 2.49.7, p. 570; Ryan and Fink, ‘Introduction’, p. 22. Fink conjectured that the criticism could only have been written after Baldwin’s death. RC chap. 37, pp. 632–33; trans. in full is in chapter 2 above; Ralph repeated the same description in the same words ‘a man . . . magnanimity’, but transferred the description to 1100 when Baldwin I succeeded Duke Godfrey. AA 12.28, p. 868. For Abbot Daniel, Jerusalem Pilgrimage, 1099–1185, ed. Wilkinson, p. 168 and chapter 5 above. Historia Nicaena, p. 139, lines 14–22. WT 10.2, pp. 454–55; Edbury and Rowe, William of Tyre, pp. 71–75; Einhard, Vita Karoli Magni, 22–27, trans. Lewis Thorpe, Einhard and Notker the Stammerer: Two Lives of Charlemagne (Harmondsworth, 1969), pp. 76–81; Suetonius, De vitis Caesarum, trans. Robert Graves (Harmondsworth, 1957); Latin texts of both online at www.thelatinlibrary.com [accessed 9 January 2017]. William explained at the end of the previous chapter (10.1, pp. 453–54) that it was unnecessary to write of the nobility of Baldwin’s birth and lineage because it was the same as Godfrey’s. William’s translators used ‘in vain’ to modify Baldwin’s struggling with weakness of the flesh; the Latin is ‘inpatienter’, which is rather less clearcut, meaning perhaps ‘reluctantly’ or ‘unwillingly’: Babcock and Krey, A History of Deeds, p. 416. Anselm’s letter: S. Anselmi cantuariensis archiepiscopi opera omnia, ed. F. S. Schmitt, 6 vols (Edinburgh, 1946–1961), IV: 255, no. 324. The passage in Latin: Consideret igitur vestra celsitudo quam eminens gratia dei sit, quod vos in hac civitate regem esse voluit; et quanto affectu, quanto studio se debeat subdere voluntati dei et eius servitio rex, quem ille ibi constituit. Precor ergo, obsecro, moneo ut dominum, ut dilectum, quatenus et vestram personam et omnes vobis subditos sic regere secundum legem et voluntatem dei studeatis, ut lucidum exemplum omnibus regibus terrae in vita vestra praebeatis.

See also The Letters of St Anselm of Canterbury, trans. Walter Fröhlich, 3 vols (Kalamazoo, 1990), 3: 36–38, n. 1 for dating (revised from Schmitt’s 1107). For Mayer’s interpretation: ‘Études’, p. 72. 36 For the earlier letter, see Schmitt, 4: 142–43, no. 235: passage quoted, Ne putetis vobis, sicut multi mali reges faciunt, ecclesiam dei quasi domino ad serviendum esse datam,

190  Last years and legacy

37 38 39

40

41

42 43 44 45

46 47 48

49 50

51 52

sed sicut advocato et defensori esse commendatam. Nihil magis diligit deus in hoc mundo quam libertatem ecclesiae suae. WT 10.2, pp. 454–55. Mayer, ‘Études’, pp. 49–72. Mayer, ‘Études’, pp. 49–50; Runciman, History of the Crusades, 1: 147, 198, 200–201. More recently, MacEvitt twice referred to the death of Baldwin’s children with his wife: pp. 57, 59. Mayer likewise dismissed Hagenmeyer’s (also unfounded) assumption that Baldwin was deeply moved by the death of his wife: FC, p. 206, n. 7. Mayer interpreted every reference in the sources pejoratively. Of the splendid wedding of Baldwin and his Armenian bride, he wrote that Albert of Aachen ‘fut, semble-t-il, si profondément frappé par les scandales conjugaux de Baudouin qu’il mit l’accent sur le fait que le mariage avec Arda était absolument légal’. GN 7.48, p. 349: his use of dicitur and traditur suggests some doubt about the story. Mayer, p. 56, assumed the king summoned her from Edessa but does not explain why she should be there when her family had fled to Constantinople and she had moved her household south to the kingdom of Jerusalem. WT 11.1, pp. 495–96. WM 4.385.2, p. 688; Mayer rejected the whole story as mere rumour possibly generated by Baldwin’s supporters, who wanted to exempt him from accusations of homosexuality, ‘Études’, p. 66. Mayer, ‘Études’, pp.  73–91. Strangely, Mayer claimed that WT had access to AA’s Historia (pp. 80, 86). For the course of events, see also Murray, Crusader Kingdom, pp. 120–23; Edbury and Rowe, William of Tyre, pp. 69–70. AA 12.28, p. 868; 12.30, pp.  872–74. There is doubt as to whether Baldwin II was crowned on Easter Day, too. AA wrote that he was but FC reported a formal coronation in Bethlehem, Christmas Day 1119 (FC 3.7.4, p. 635). However, this is a matter for Baldwin II’s biographer. WT 12.2–3, pp. 548–50. The detail about the messenger intercepting Baldwin on his way from Edessa confirms the likelihood that the Palm Sunday party on the Mount of Olives was expecting the funeral cortège. FC 3.1, pp. 615–16. Mayer cited the Chroniques des comtes d’Anjou as confirming that Baldwin came to Jerusalem to speak with his kinsman the king (gratia colloquandi cum rege cognato suo), but there is nothing in the chronicle or its dating to preclude FC as its source: Mayer, ‘Études’, p.  87 and n. 2; Chroniques des comtes d’Anjou et des seigneurs d’Amboise, ed. Louis Halphen and René Poupardin (Paris, 1913), p. 70. ME 3.75, p. 221 and nn. pp. 344–45. By the end of the twelfth century Baldwin’s absence from the siege of Jerusalem was remedied in the Old French epic, La Chanson de Jérusalem, which portrayed him fighting alongside his brother: La Chanson de Jérusalem, ed. Nigel R. Thorp; The Old French Crusade Cycle, Vol. 6 (Tuscaloosa, 1992); The Chanson des Chétifs and Chanson de Jérusalem: Completing the Central Trilogy of the Old French Crusade Cycle, trans. Carol Sweetenham (Farnham, 2016), intro. pp. 44–45. Mayer, ‘Études’, pp. 10–11. William B. Stevenson, The Crusaders in the East (Cambridge, 1907, repr. Beirut, 1958), p. 68.

Bibliography

BibliographyBibliography

Primary sources Manuscripts Cambridge UL MS KK.vi.15 Favi, Vicenzo, Relatione del Viaggio di Gerusalemme, London, British Library Add. 33566 Horn, MS Vat. lat. 9233

Editions and translations Albert of Aachen, Historia Ierosolimitana, ed. and trans. Susan B. Edgington (Oxford, 2007) Anna Komnene, The Alexiad, trans. E. R. A. Sewter, revd. Peter Frankopan (London, 2009) Annalista Saxo, in MGH SS, 6: 542–777 Baldric of Bourgueil, Historia Ierosolimitana, ed. Steven Biddlecombe (Woodbridge, 2014) Bartolf of Nangis, ‘Gesta Francorum Iherusalem expugnantium’, in RHC Occ iii: 491–543 Bernardo Maragone, Vetus chronicon Pisanum in Rerum Italicarum Scriptores, NS 6. 2, ed. Michele Lupo Gentile (1936), pp. 3–71 Caffaro, Annali genovesi di Caffaro e de’ suoi continuatori, ed. L. Belgrano, Fonti per la storia d’Italia 11-14 bis (Rome, 1890–1901); Caffaro, Genoa and the Twelfth-Century Crusades, trans. Martin Hall and Jonathan Phillips (Farnham, 2013) La Chanson d’Antioche, ed. Suzanne Duparc-Quioc, 2 vols (Paris 1977–78); The Chanson d’Antioche: An Old French Account of the First Crusade, trans. Susan B. Edgington and Carol Sweetenham (Farnham, 2011) La Chanson de Jérusalem, ed. Nigel R. Thorp. The Old French Crusade Cycle, 6 vols (Tuscaloosa, 1992); The Chanson des Chétifs and Chanson de Jérusalem: Completing the Central Trilogy of the Old French Crusade Cycle, trans. Carol Sweetenham (Farnham, 2016) Chroniques des comtes d’Anjou et des seigneurs d’Amboise, ed. Louis Halphen and René Poupardin (Paris, 1913) Daniel, ‘The Life and Journey of Daniel, Abbot of the Russian Land’, in Jerusalem Pilgrimage, 1099–1185, ed. and trans. John Wilkinson with Joyce Hill and W. F. Ryan (London, 1988), pp. 120–71 Einhard, Vita Karoli Magni, at www.thelatinlibrary.com [accessed 9 January  2017]; trans. Lewis Thorpe, Einhard and Notker the Stammerer: Two Lives of Charlemagne (Harmondsworth, 1969)

192  Bibliography Ekkehard of Aura, Hierosolymita, RHC Occ v: 1–40; see also Frutolf of Michelsberg ‘The First and Second Crusades from an Anonymous Syriac Chronicle’, ed. and trans. Arthur S. Tritton and Hamilton A. R. Gibb, Journal of the Royal Asiatic Society, 92 (1933), 69–102, 273–306 Frutolf of Michelsberg: Frutolfs und Ekkehards Chroniken und die Anonyme Kaiserchronik, ed. Franz-Josef Schmale and Irene Schmale-Ott (Darmstadt, 1972); The Chronicles of the Investiture Contest: Frutolf of Michelsberg and his Continuators, trans. Thomas J. H. McCarthy (Manchester, 2014) Fulcher of Chartres, Historia Hierosolymitana, ed. Heinrich Hagenmeyer (Heidelberg, 1913); A History of the Expedition to Jerusalem 1095–1127, trans. Frances Rita Ryan, intro. by Harold S. Fink (Knoxville, 1969) Gesta Francorum et aliorum Hierosolimitanorum, ed. and trans. Rosalind Hill (London, 1962); Gesta Francorum: Histoire anonyme de la Première Croisade, ed. and trans. Louis Bréhier (Paris 1924) Gesta triumphalia per Pisanos facta de captione Hierusalem: RIS, new series, 6.2 (Bologna, 1930); Gesta triumphalia per Pisanos facta, ed. Giuseppe Scalia (Florence, 2010) Guibert of Nogent, Dei gesta per Francos, ed. Robert B. C. Huygens, CCCM 127A (Turnhout, 1996); The Deeds of God through the Franks: A Translation of Guibert of Nogent’s Gesta Dei per Francos, trans. Robert Levine (Woodbridge, 1996) Historia Nicaena vel Antiochena necnon Jerosolymitana, RHC Occ v: 139–85 ‘Historia peregrinorum euntium Jerusolymam’, RHC Occ iii: 169–229 Ibn al-Athir: The Chronicle of Ibn al-Athir for the Crusading Period from al-Kamil fi’lTa’rikh AH495, ed. Donald S. Richards, 3 vols (Aldershot, 2006–8) Ibn Al-Qalanīsī: The Damascus Chronicle of the Crusades: Extracted and Translated from the Chronicle of Ibn Al-Qalanīsī, trans. Hamilton A. R. Gibb (London, 1932) Ibn al-Muqaffa’, History of the Patriarchs of the Egyptian Church, iii.1, ed. and trans. Yassá ʻAbd al-Masīḥ, ʻAzīz Sūryāl ʻAṭiiyah, Uswald Būrmastar (Cairo, 1959) Magnússona Saga, or Saga of the Sons of Magnus, in Snorri Sturluson, Heimskringla, trans. Lee M. Hollander (Austin, TX, 1964), pp. 688–714 Matthew of Edessa, Armenia and the Crusades, Tenth to Twelfth Centuries: The Chronicle of Matthew of Edessa, trans. Ara E. Dostourian (Lanham, 1993) Michael the Syrian, Chronique de Michel le Syrien, Patriarche Jacobite d’Antioche (1166– 1199), ed. and trans. Jean-Baptiste Chabot, 5 vols (Paris, 1899–1924) Monachi anonymi Scaphusensis, ‘De reliquiis Sanctissimae Crucis et Dominici Sepulchri Scaphusam allatis’, in RHC Occ v: 335–39 Morkinskinna, ed. C. R. Unger (Oslo, 1867) Orderic Vitalis, The Ecclesiastical History of Orderic Vitalis, ed. Marjorie Chibnall, 6 vols (Oxford, 1969–80) Procopius, On Buildings, trans. H. E. Dewing (Cambridge, MA, 1940) Ralph of Caen, Gesta Tancredi, in RHC Occ iii: 587–716; Radulphus Cadomensis, Tancredus, ed. Edoardo D’Angelo. CCCM 231 (Turnhout, 2011); The Gesta Tancredi of Ralph of Caen: A History of the Normans on the First Crusade, trans. Bernard S. Bachrach and David S. Bachrach (Aldershot, 2005) Raymond of Aguilers, Le “Liber” de Raymond d’Aguilers, ed. John H. Hill and Laurita L. Hill. Documents relatifs à l’histoire des croisades 9 (Paris, 1969); Historia Francorum qui ceperunt Iherusalem, in RHC Occ iii: 231–309 Reginald of Durham, Vita S. Godrici, Surtees Society (1838) Robert the Monk, Historia Iherosolimitana, ed. Damien Kempf and Marcus G. Bull (Woodbridge, 2013); Robert the Monk’s History of the First Crusade, trans. Carol Sweetenham (Aldershot, 2005)

Bibliography  193 S. Anselmi cantuariensis archiepiscopi opera omnia, ed. Franciscus S. Schmitt, 6 vols (Edinburgh, 1946–1961); The Letters of St Anselm of Canterbury, trans Walter Fröhlich, 3 vols (Kalamazoo, 1990) Suetonius, De vitis Caesarum, at www.thelatinlibrary.com [accessed 9 January 2017]; The Twelve Caesars, trans. Robert Graves (Harmondsworth, 1957) Theodericus, in Peregrinationes tres: Saewulf, John of Würzburg, Theodericus, ed. R. B. C. Huygens, CCCM 139 (Turnhout, 1995), pp. 143–97 Theodric the Monk, Historia de antiquitate regum Norwagiensium, in Monumenta Historica Norvegiae, ed. Gustav Storm (Oslo, 1880), pp. 1–68 Usamah ibn-Munqidh: An Arab-Syrian Gentleman and Warrior in the Period of the Crusades, trans. Philip K. Hitti (New York, 1929) Walter the Chancellor: Galterii Cancellarii: Bella Antiochena, ed. Heinrich Hagenmeyer (Innsbruck, 1896); Thomas S. Asbridge and Susan B. Edgington, Walter the Chancellor’s The Antiochene Wars: A Translation and Commentary (Aldershot, 1999) William of Malmesbury, Gesta Regum Anglorum, ed. and trans. Roger A. B. Mynors, Rodney M. Thomson and Michael Winterbottom, 2 vols (Oxford, 1998–99) William of Tyre, Chronicon, ed. Robert B. C. Huygens, CCCM 63-63A, 2 vols (Turnhout, 1986); A History of Deeds Done beyond the Sea by William Archbishop of Tyre, trans. Emily A. Babcock and August C. Krey, 2 vols (New York, 1943)

Source collections Le Cartulaire du chapitre du Saint-Sépulcre de Jérusalem, ed. Geneviève Bresc-Bautier (Paris, 1984) Chronicles of the Crusades, ed. Elizabeth Hallam (London, 1989) Codice Diplomatico del Sacro Militare Ordine Gerosolimitano Oggi di Malta, ed. Sebastiano Paoli (Lucca, 1733) Epistulae et chartae ad historiam primi belli sacri spectantes, ed. Heinrich Hagenmeyer (Innsbruck, 1901) The First Crusade: The Chronicle of Fulcher of Chartres and Other Source Materials, ed. Edward Peters, 2nd edn (Philadelphia, 1998) Monumenta Germaniae Historica, Scriptores, ed. G. H. Pertz et  al., 32 vols (Hanover, Weimar, Stuttgart and Cologne, 1826–34) Papsturkunden für Kirchen im Heilingen Lande, ed. Rudolf Hiestand (Göttingen, 1985) Recueil des Historiens des Croisades, Historiens Occidentaux, ed. Académie des Inscriptions et Belles Lettres, 5 vols (Paris, 1844–95) Die Urkunden der lateinischen Könige von Jerusalem, ed. Hans Eberhard Mayer, 4 vols (Hanover, 2010)

Secondary works Aird, William M., Robert Curthose, Duke of Normandy (c. 1050–1134) (Woodbridge, 2008) Amouroux-Mourad, Monique, Le Comté d’Edesse, 1098–1150 (Paris, 1988) Asbridge, Thomas, The Creation of the Principality of Antioch, 1098–1130 (Woodbridge, 2000) ———, The First Crusade: A New History (London, 2004) ———, ‘The Principality of Antioch and the Early History of the Latin East’, in East and West in the Medieval Eastern Mediterranean II: Antioch from the Byzantine Reconquest

194  Bibliography Until the End of the Crusader Principality, ed. K. Giggaar and V. van Aalst (Leuven, 2013), pp. 1–10 Beaumont, André Alden, ‘Albert of Aachen and the County of Edessa’, in The Crusades and Other Historical Essays Presented to Dana C. Munro by His Former Students, ed. Louis J. Paetow (New York, 1928) Beihammer, Alexander Daniel, Byzantium and the Emergence of Muslim-Turkish Anatolia, 1040–1130 (Abingdon, 2017) Bellomo, Elena, A servizio di Dio e del Santo Sepolcro: Caffaro e l’Oriente latino (Padova, 2003) Blake, Ernest O., and Colin Morris, ‘A hermit goes to war: Peter and the origins of the First Crusade’, Studies in Church History, 22 (1985), 79–107 Brett, Michael, ‘Al-Karāza al-Marqusīya: The Coptic Church in the Fatimid Empire’, in Egypt and Syria in the Fatimid, Ayyubid and Mamluk Eras, ed. U. Vermeulen and J. van Steenbergen (Leuven, 2005), pp. 30–60 ———, ‘The Battles of Ramla (1099–1105)’, in Egypt and Syria in the Fatimid, Ayyubid and Mamluk Eras, ed. U. Vermeulen and D. de Smet (Leuven, 1995), pp. 17–37 Bull, Marcus, and Damien Kempf (ed.), Writing the Early Crusades: Text, Transmission and Memory (Woodbridge, 2014) Cobb, Paul M., The Race for Paradise: An Islamic History of the Crusades (Oxford, 2014) Cohn, Norman, The Pursuit of the Millennium, 3rd edn (London, 1970) Crouzet-Pavan, Élisabeth, Le mystère des rois de Jérusalem 1099–1187 (Paris, 2013) Crowley, Heather, ‘The impact of the Franks on the Latin Kingdom of Jerusalem: Landscape, Seigneurial Obligations, and rural communities in the Frankish East’ (Unpublished Ph.D. thesis, University of Cardiff, 2016) Dédéyan, Gérard, Les Arméniens entre Grecs, Musulmans et Croisés: Etude sur les pouvoirs arméniens dans le Proche-Orient méditerranéen (1068–1150), 2 vols (Lisbon, 2003) Ebels-Hoving, Bunna, Byzantium in westerse ogen (Assen, 1971) Edbury, Peter W., and John G. Rowe, William of Tyre: Historian of the Latin East (Cambridge, 1988) Edgington, Susan B., ‘The Capture of Acre, 1104, and Baldwin I’s Conquest of the Littoral’, in Acre and Its Falls: Studies in the History of a Crusader City, ed. John France (Leiden, 2018), pp. 13–29 ———, ‘Espionage and Counter-Espionage: An Episode in the Reign of Baldwin I’, in Crusader Landscapes in the Medieval Levant: The Archaeology and History of the Latin East, ed. Kevin J. Lewis, Balazs Major, Micaela Sinibaldi and Jennifer Thomson (Cardiff, 2016), pp. 157–67 ———, ‘The Gesta Francorum Iherusalem expugnantium of “Bartolf of Nangis” ’, Crusades, 13 (2014), 21–35 ———, ‘The Historia Iherosolimitana of Albert of Aachen: A  critical edition’ (Unpublished Ph.D. thesis, University of London, 1991) Edgington, Susan B., and Luis García-Guijarro (ed.), Jerusalem the Golden: The Origins and Impact of the First Crusade (Turnhout, 2014) ———, ‘Medical Knowledge in the Crusading Armies: The Evidence of Albert of Aachen and Others’, in The Military Orders: Fighting for the Faith and Caring for the Sick, ed. M. Barber (Aldershot, 1994), pp. 320–26 El-Azhari, Taef, ‘Mawdūd (d. 1113)’, Crusades Encyclopedia, III: 809 ———, ‘Ṭughtigin (d. 1128)’, Crusades Encyclopedia, IV: 1204–205

Bibliography  195 Ellenblum, Ronnie, Frankish Rural Settlement in the Latin Kingdom of Jerusalem (Cambridge, 1998) Face, Richard D., ‘Secular history in twelfth-century Genoa’, Journal of Medieval History, 6 (1980), 169–83 Favreau-Lilie, Mari-Luise, ‘Genoa’, Crusades Encyclopedia, II: 501–506 Ferrier, Luc, ‘La couronne refusée de Godefroy de Bouillon: eschatologie et humiliation de la majesté aux premiers temps du royaume latin de Jérusalem’, in Le Concile de Clermont de 1095 et l’appel à la croisade, Collection de l’École Française de Rome, 236 (Rome, 1997), 245–65 Flori, Jean, ‘Jérusalem terrestre, céleste et spirituelle: Trois facteurs de sacralisation de la première croisade’, in Jerusalem the Golden, ed. Susan B. Edgington and Luis GarcíaGuijarro (Turnhout, 2014), pp. 25–50 ———, Pierre l’Ermite et la première croisade (Paris, 1999) Folda, Jaroslav, The Art of the Crusaders in the Holy Land, 1098–1187 (Cambridge, 1995) France, John (ed.), Acre and Its Falls (Leiden, 2018) ———, ‘The Use of the Anonymous Gesta Francorum in the Early Twelfth-Century Sources for the First Crusade’, in From Clermont to Jerusalem: The Crusades and Crusader Societies 1095–1500, ed. A. V. Murray (Turnhout, 1998), pp. 29–42 ———, Victory in the East: A Military History of the First Crusade (Cambridge, 1994) ———, Western Warfare in the Age of the Crusades, 1100–1300 (London, 1999) Frankopan, Peter, The First Crusade: The Call from the East (London, 2012) Garcia-Guijarro, Luis, ‘Some Considerations on the Crusaders’ Letter to Urban II (September 1098)’, in Jerusalem the Golden, ed. Susan B. Edgington and Luis García-Guijarro (Turnhout, 2014), pp. 151–71 Gardiner, Robert, ‘Crusader Turkey: The fortifications of Edessa’, Fortress, 2 (1989), 23–35 Gindler, Paul, Graf Balduin I. von Edessa (Halle, 1901) Grabois, Aryeh, ‘Galilee’, Crusades Encyclopedia, II: 495–98 Hamilton, Bernard, The Latin Church in the Crusader States: The Secular Church (London, 1980) Hansen, Joseph, Das Problem eines Kirchenstaates in Jerusalem: Ein Beitrag zur Geschichte der Kreuzzüge (Luxembourg, 1928) Harris, Jonathan, Byzantium and the Crusades (London, 2003) Hiestand, Rudolf, ‘König Balduin und sein Tanzbär’, Archiv für Kulturgeschichte, 70 (1988), 343–60 Hill, John H., and Laurita L. Hill, Raymond IV, Count of Toulouse (Syracuse, NY, 1962) Hillenbrand, Carole, The Crusades: Islamic Perspectives (Edinburgh, 1999) John, Simon, ‘Godfrey of Bouillon and the Swan Knight’, 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. 129–42 ———, Godfrey of Bouillon, Duke of Lower Lotharingia, Ruler of Latin Jerusalem, c.1060–1100 (Abingdon, 2018) Jotischky, Andrew, Crusading and the Crusader States (Harlow, 2004) Kedar, Benjamin Z., ‘Again: Genoa’s Golden Inscription and King Baldwin I’s Privilege of 1104’, in Chemins d’outre-mer: Études d’histoire sur la Méditerranée médiévale offertes à Michel Balard, ed. Damien Coulon, Catherine Otten, Paule Pages and Dominique Valerian (Paris, 2004), pp. 495–502 ———, ‘Genoa’s Golden Inscription in the Church of the Holy Sepulchre: A Case for the Defence’, in I comuni italiani nel regno crociato di Gerusalemme: Atti del Colloquio

196  Bibliography di Gerusalemme, 24–28 maggio 1984, ed. G. Airaldi and B. Z. Kedar (Genoa, 1986), pp. 317–35 Kennedy, Hugh, Crusader Castles (Cambridge, 1994) Kirstein, Klaus-Peter, ‘Evremar of Chocques (d. 1128/1129)’, Crusades Encyclopedia, II: 417–18 Kloner, Amos, ‘Beth Guvrin’, in The New Encyclopedia of Archaeological Excavation in the Holy Land, ed. Ephraim Stern, 4 vols (Jerusalem and New York, 1993), I: 195–201 Köhler, Michael A., Alliances and Treaties Between Frankish and Muslim Rulers in the Middle East: Cross-Cultural Diplomacy in the Period of the Crusades, trans. Peter M. Holt, revd. Konrad Hirschler (Leiden, 2013) Kool, Robert, ‘A hoard twice buried? Fatimid gold from thirteenth-century crusader Arsur (Apollonia-Arsuf)’, Numismatic Chronicle, 173 (2013), 261–92 ———, ‘Civitas Regis Regvm Omnium: Inventing a Royal Seal in Jerusalem, 1100– 1118’, in Crusades Subsidia: Crusading and Archaeology, ed. Vardit Shotten-Hallel and Rosie Weetch (Abingdon, 2019), forthcoming. Krey, August C., ‘A Neglected Passage in the Gesta and Its Bearing on the Literature of the First Crusade’, in The Crusades and Other Historical Essays Presented to Dana C. Munro by His Former Students, ed. Louis J. Paetow (New York, 1928), pp. 57–78 Laurent, Joseph, ‘Des Grecs aux croisés: Etude sur l’histoire d’Edesse’, Byzantion, 1 (1924), 404–34 Lewis, Charlton T., and Charles Short (ed.), A Latin Dictionary (Oxford, 1879) Lewis, Kevin James, The Counts of Tripoli and Lebanon in the Twelfth Century: Sons of Saint-Gilles (Abingdon, 2017) Lilie, Ralph-Johannes, Byzantium and the Crusader States 1096–1204 (Oxford, 1993) Luttrell, ‘The Earliest Hospitallers’, in Montjoie: Studies in Crusade History in Honour of Hans Eberhard Mayer, ed. Benjamin Z. Kedar, Jonathan Riley-Smith and Rudolf Hiestand (Aldershot, 1997), pp. 37–54 MacEvitt, Christopher, The Crusades and the Christian World of the East: Rough Tolerance (Philadelphia, 2008) ———, ‘Edessa, county of’, Crusades Encyclopedia, II: 379–85 Matzke, Michael, ‘Daibert of Pisa (d. 1105)’, Crusades Encyclopedia, II: 339–40 ———, Daibert von Pisa: Zwischen Pisa, Papst und ersten Kreuzzug (Sigmaringen, 1998) ———, ‘Pisa’, Crusades Encyclopedia, III: 964–66 Mayer, Hans Eberhard, ‘Die Herrschaftsbildung in Hebron’, Zeitschrift des Deutschen Palästina-Vereins, 101 (1985), 64–81 ———, ‘Die Kanzlei der lateinischen Könige von Jerusalem’, in Schriften der MGH 40, 2 vols (Hanover, 1996) ———, Mélanges sur l’histoire du Royaume latin de Jérusalem (Paris, 1984) Mayer, Hans Eberhard, and Marie-Luise Favreau, ‘Das Diplom Balduins I. für Genua und Genuas Goldene Inschrift in der Grabeskirche’, Quellen und Forschungen aus italienischen Archiven und Bibliotheken, 55/56 (1976), 22–95 Metcalf, David M., Coinage of the Crusades and the Latin East in the Ashmolean Museum Oxford (London, 1995) Mitchell, Piers, Medicine in the Crusades: Warfare, Wounds and the Medieval Surgeon (Cambridge, 2004) Morton, Nicholas, Encountering Islam on the First Crusade (Cambridge, 2016) ———, The Field of Blood: The Battle for Aleppo and the Remaking of the Medieval Middle East (New York, 2018) Mulinder, Alec, ‘Crusade of 1101’, Crusades Encyclopedia, I: 34–37

Bibliography  197 ———, ‘The crusading expeditions of 1101–2’ (Unpublished Ph.D. thesis, University of Wales, Swansea, 1996) Murray, Alan V., The Crusader Kingdom of Jerusalem: A Dynastic History 1099–1125. Original Publications of the Linacre Unit for Prosopographical Research 4 (Oxford, 2000) ——— (ed.), The Crusades: An Encyclopedia, 4 vols (Santa Barbara, 2006) ———, ‘Daimbert of Pisa, the Domus Godefridi and the Accession of Baldwin I of Jerusalem’, in From Clermont to Jerusalem: The Crusades and Crusader Societies 1095– 1500, ed. Alan V. Murray (Turnhout, 1998), pp. 81–102 ——— (ed.), From Clermont to Jerusalem: The Crusades and Crusader Societies 1095– 1500, International Medieval Research 3 (Turnhout, 1998) ———, ‘ “Mighty Against the Enemies of Christ”: The Relic of the True Cross in the Armies of the Kingdom of Jerusalem’, in The Crusades and Their Sources: Essays Presented to Bernard Hamilton, ed. John France and William G. Zajac (Aldershot, 1998), pp. 217–38 ———, ‘The Origin of Money-fiefs in the Latin Kingdom of Jerusalem’, in Mercenaries and Paid Men: The Mercenary Identity in the Middle Ages: Proceedings of a Conference held at the University of Wales, Swansea, 7th – 9th July 2005, ed. John France (Leiden, 2008), pp. 275–86 ———, ‘The title of Godfrey of Bouillon as ruler of Jerusalem’, Collegium Medievale, 3 (1990), 163–78 Nader, Marwan, Burgesses and Burgess Law in the Latin Kingdoms of Jerusalem and Cyprus (1099–1325) (Aldershot, 2006) Pahlitzsch, Johannes, ‘Symeon II of Jerusalem’, Crusades Encyclopedia, IV: 1135–36 Paliouras, Athanasios (ed.), St Catherine’s Monastery (Sinai, 1985) Prawer, Joshua, Crusader Institutions (Oxford, 1980) Pryor, John, ‘The oath of the leaders of the first crusade to Emperor Alexius I Comnenus: Fealty, homage, pistis, douleia’, Parergon, n.s. 2 (1984), 111–41 Reynolds, Susan, Fiefs and Vassals: The Medieval Evidence Re-interpreted (Oxford, 1994) Richard, Jean, Le Royaume latin de Jérusalem (Paris, 1953) ———, ‘The Political and Ecclesiastical Organization of the Crusader States’, in A History of the Crusades, ed. K. Setton, 6 vols (Madison, 1969–1989), V: 193–217 ———, ‘Raymond of Saint-Gilles’, in The Crusades: An Encyclopedia, ed. Alan V. Murray, 4 vols (Santa Barbara, 2006), IV: 1011 Riley-Smith, Jonathan, The First Crusade and the Idea of Crusading (London, 1986) ———, The First Crusaders (Cambridge, 1997) ———, ‘The title of Godfrey of Bouillon’, Bulletin of the Institute of Historical Research, 52 (1979), 83–86 Rubenstein, Jay, Armies of Heaven: The First Crusade and the Quest for the Apocalypse (New York, 2011) ———, ‘Guibert of Nogent, Albert of Aachen and Fulcher of Chartres: Three Crusade Chronicles Intersect’, Writing the Early Crusades, ed. Marcus Bull and Damien Kempf (Woodbridge, 2014), pp. 24–37 ———, ‘Holy Fire and sacral kingship in post-conquest Jerusalem’, Journal of Medieval History, 43 (2017), 470–84 Runciman, Steven, A History of the Crusades, 3 vols (Cambridge, 1951–1954) Sabourin, Pascal, ‘Baudouin de Bourcq, croisé, comte d’Edesse, roi de Jérusalem: Proposition de lecture d’un itinéraire peu ordinaire’, Revue historique ardennaise, 31 (1996), 3–15

198  Bibliography Sassier, Yves, Louis VII (Paris, 1991) Shagrir, Iris, ‘Adventus in Jerusalem: The Palm Sunday celebration in Latin Jerusalem’, Journal of Medieval History, 41 (2015), 1–20 Slack, Corliss K., ‘Royal Familiares in the Latin Kingdom of Jerusalem, 1100–1187’, Viator, 22 (1991), 15–67 Smith, Thomas W., ‘The First Crusade Letter written at Laodicea in 1099: Two previously Unpublished Versions from Munich Bayerische Staatsbibliothek Clm 23390 and 28196’, Crusades, 15 (2016), 1–25 ———, ‘Scribal crusading: Three new manuscript witnesses to the regional reception and transmission of first crusade letters’, Traditio, 72 (2017), 133–69 Stevenson, William B., The Crusaders in the East (Cambridge, 1907, repr. Beirut, 1958) Strevett, Neil, ‘The Anglo-Norman aristocracy under divided Lordship, 1087–1106: A social and political study’ (Ph.D. thesis, University of Glasgow, 2005) Sweetenham, Carol, ‘What Really Happened to Eurvin de Créel’s Donkey? Anecdotes in the Sources for the First Crusade’, in Writing the Early Crusades, ed. Marcus Bull and Damien Kempf (Woodbridge, 2014), pp. 75–88 Sybel, Heinrich von, Geschichte des ersten Kreuzzugs (Leipzig, 1841); trans. as The History and Literature of the Crusades, Lady Duff Gordon (London, 1861) Tanner, Heather J., ‘In His Brothers’ Shadow: The Crusading Career and Reputation of Eustace III of Boulogne’, in The Crusades: Other Experiences, Alternate Perspectives, ed. Khalil I. Semaan (Binghampton, 2003), pp. 83–99 Tibble, Steven, Monarchy and Lordships in the Latin Kingdom of Jerusalem, 1099–1291 (Oxford, 1989) Tierney, Brian, The Crisis of Church and State 1050–1300 (Toronto, 1988; repr. 1996) Tyerman, Christopher, England and the Crusades (Chicago, 1988) Vaughn, Sally N., St Anselm and the Handmaidens of God: A Study of Anselm’s Correspondence with Women (Turnhout, 2002) Wright, Neil, ‘Chapter 10: William as Historian of Crusade’, in William of Malmesbury, revd edn, ed. R. M. Thomson (Woodbridge, 2003), pp. 178–88 Yarnley, Colin J., ‘Philaretos: Armenian bandit or byzantine general?’, Revue des études armeniennes, 9 (1972), 331–53

Database Revised Regesta Regni Hierosolymitani Database [RRR], http://crusades-regesta.com

Index

Acre 67, 107, 114, 115, 117, 124, 138, 140, 145, 153, 160, 164, 167, 174; capture 103, 106, 119 – 22, 123, 179; emir captured 144 Adana 25, 29, 38 Addo, cook 177 Adelaide of Sicily 106, 166, 174, 184 – 5 Adhémar, bishop of Le Puy 22, 33, 52, 66, 95 Aicard, deacon 93 al-Afḍal 112, 129, 176 al-‘Arīsh 176 Albara 53 Aleppo 112, 162, 164 Alexandria 142, 166 Alexios I Komnenos 8 – 9, 10 – 13, 21, 23, 24, 25, 33, 34, 50, 61, 67, 84, 85, 106, 114, 122, 133 – 4, 142, 165 – 6 al-Faramā 175 – 6 Alfonso, king of Leon and Castile 96 alliances 25, 41, 43, 47, 53, 60, 64, 65, 68, 76, 106 – 7, 122, 124, 145, 160, 164, 165 – 7, 174, 183, 184 allies 6, 24, 32, 52, 60, 83, 101, 113 – 14, 130, 135, 143, 159, 162, 164, 175, 179 al-Sinnabrāh 163 Amacha 49 – 50 Anselm, archbishop of Milan 133 Anselm, castellan 158 Anselm of Bec, archbishop of Canterbury 1, 181 Antioch 25, 34, 38, 46, 65, 67, 70, 83, 95, 100, 105, 106, 111, 125, 130, 131, 133, 139, 145, 156, 159 – 60, 161, 163, 165, 167; battle of 52, 131; siege of 47, 50 – 3, 114 Antwerp 145 Apamea 164 Apulia 11, 85, 100, 186

Aqaba 174 Arabia 154, 157, 173 – 4, 175, 179 ‘Arda’ 46 – 7, 49, 70, 131 – 2, 137, 181, 183 – 4 Armenia/Armenians 24 – 5, 26, 32, 34 – 5, 38, 39, 52 – 3, 162 Arnulf of Chocques 15, 63, 64, 66, 69, 93, 99 – 100, 101 – 3, 104, 182; interim patriarch 77, 86, 94 – 5, 96, 98, 112; patriarch of Jerusalem 105 – 7, 156, 166, 167, 174 – 5, 177 – 8, 185 – 6 Arqa 61, 70, 114, 161 Arsuf 61, 85, 114, 122, 136 – 8, 144, 157; capture 114 – 16, 117, 153, 179; emir captured 144 Artois 155 Ascalon/Ascalonites 78 – 9, 85, 114, 115, 119, 123, 124, 125, 129, 138, 139, 140, 141, 142, 144, 145, 153, 157, 158, 167, 174, 175; battle of 54, 61, 66, 98, 101, 112, 114, 116, 129; emir killed 144 Aschetinus, bishop of Bethlehem 93 Azaz 52 Baalbek 55 Baghdad 164 Bagrat 32, 34 – 5, 39, 51, 53 Baktāsh ibn Tutush 143 Balak of Sororgia 45 – 6, 49 – 50 Baldric, seneschal 7 Balduk of Samosata 41, 42, 43, 45 – 6, 49 – 50 Baldwin, archbishop of Caesarea 117, 131 Baldwin, castellan 158 Baldwin, count of Hainaut 9 Baldwin, cubicularius 159, 181 Baldwin I: ancestry and education 1 – 4, 77; character assessment 32, 45, 49, 100, 132, 162, 179 – 82; in Cilicia 24 – 30, 64;

200 Index coronation 80 – 2; as count of Edessa 39 – 56, 70, 183; death 176 – 7, 185; First Crusade journey 4 – 13; horse Gazela 136; marriages 3, 46 – 7, 51, 70, 106 – 7, 166 – 7, 174 – 5, 181, 182 – 5; pilgrimage to Jerusalem 54 – 6; relations with patriarchs 93 – 107; at siege of Nicaea 22; succession 177, 185 – 7; succession to Godfrey 63 – 73, 97 – 8; tomb 178 – 9; wounded 139 – 42; see also coronation Baldwin Calderun 22 Baldwin of Bourcq 6, 10, 26, 32, 65, 84; as count of Edessa 35, 69 – 70, 101, 124, 139, 156, 160, 161 – 2, 163, 164, 185 – 7 Baldwin of St Abraham 156 battle of Dorylaeum 23 – 4 Bedouin 154 Beirut 70 – 1, 73, 114, 133, 144, 145, 153, 159; capture 123 – 4, 161, 179 Bernard, patriarch of Antioch 84 Bertrand of Tripoli 123 – 4, 161, 162, 163 Bethany 177 Bethlehem 55, 61, 80 – 2, 86, 93, 103, 104, 105, 112 bishop of ‘Barzenona’ 134 bishop of Liège 6 Bohemond 11, 13, 22, 23, 24, 25, 26, 28, 33, 48, 51, 52, 54, 61, 65, 66 – 7, 95 – 6, 121, 142, 160, 161, 165; captivity 59 – 60, 65, 68, 69, 83, 84 – 5, 114; recruits a new crusade 85, 102, 134, 152 booty see plunder bribery 55, 96, 99, 100, 107, 154 burgesses 152 – 3 Bursuq of Hamadan 164 Byzantine empire/Byzantines 8 – 13, 16, 21, 23, 25, 33, 38, 40 – 2, 49, 50, 54, 61, 65, 85, 94, 96, 106, 113 – 14, 121, 122, 133 – 4, 152, 158, 161, 165 – 6; see also Alexios I Komnenos Caesarea 55, 61, 101, 114, 115, 122, 129, 140, 153, 156, 157; capture 116 – 19, 179 Cairo 68, 97, 112, 142, 175 Calvary chapel 177 – 8 Capetian monarchy 76 – 7 captives 9, 10, 23, 31, 32, 35, 48 – 50, 60, 65, 68, 73, 84 – 5, 100, 114, 115, 130, 136, 142, 144, 145, 153, 154, 160, 163 castellans 157 – 8 Charlemagne 27, 77, 81, 180 Chökürmish, governor of Mosul 160

Christmas 10, 48, 54, 55, 59, 66, 70, 78, 81, 94, 96, 102, 114, 154 Cilician diversion 24 – 33 Clarembald of Vendeuil 9 Coloman, king of Hungary 7 – 8 Cono, count of Montaigu 4, 10, 12, 27, 31 Conrad, constable 133, 136 Constantine of Gargar, Armenian chief 41, 43, 44, 46 – 7 Constantinople 8 – 13, 21, 33, 35, 61, 85, 93 – 4, 106, 113, 133, 165, 183 – 4 Coptic Christians 176 coronation 67, 77 – 8, 80 – 2, 86, 88, 93, 97 – 100, 102 crown 62, 77, 80 – 2, 86 – 8, 106, 158, 178, 186 – 7 Crusade of 1101 85 – 6, 132 – 4, 151 – 2, 165 Cyprus 94, 124 Daibert, archbishop of Pisa 54, 61, 62, 65, 115, 181 – 2; deposition 101, 103; patriarch of Jerusalem 55, 63, 65 – 6, 72, 73, 76, 77, 80 – 2, 82 – 4, 86 – 8, 94 – 103, 113, 117 – 18, 122, 131 Damascus 16, 55, 111 – 12, 143, 153, 154, 164, 178 Damietta 142 Daniel, abbot and pilgrim 88 – 9, 179 Danishmend 59 – 60, 65, 68, 83, 133 Danishmendid Turks 21, 59 David, biblical king 27, 81, 179, 181 Dead Sea 80, 154, 173 diplomacy 104, 106, 132 – 4, 166; see also negotiations Dodo of Cons 6 dowry 47, 49, 107, 183, 184 Drogo of Nesle 9, 48 Duqāq of Damascus 70, 72, 143 Easter 47, 48, 68, 86 – 8, 88 – 9, 97, 99, 106, 107, 114, 115, 120, 133, 165, 167, 178, 185 – 6 Edessa/Edessenes 1, 6, 9, 12, 14, 30, 33, 38 – 56, 59 – 60, 62, 64, 65, 66, 69 – 70, 76, 84, 97, 99, 101, 111, 124, 125, 129, 130, 139, 152, 155, 156, 158, 159 – 63, 183, 186 Egypt/Egyptians 55, 61, 66, 67, 72, 78 – 9, 100, 101, 103, 112, 113, 114, 116, 117, 119, 123, 124, 125, 129 – 46, 151, 157, 158, 159, 173, 174 – 7, 178, 180, 185, 186; see also Fāṭimids Embriaco, Genoese consul 118

Index  201 Emicho of Leisingen 7 Engelbert, knight 134 Ethiopians 79, 117, 135, 139, 141, 142 Euphrates, river 34, 38, 39, 52, 53, 160, 162, 164, 186 Eustace II, count of Boulogne 1, 27 Eustace III, count of Boulogne 3, 4, 5, 22, 60, 61, 185 – 7 Eustace Granarius 107, 156 Evremar of Chocques: archbishopbishop of Caesarea 104, 105, 107; patriarch of Jerusalem 103 – 5, 143 Fāṭimids 112, 129, 132, 155, 176 First Crusade 2, 3, 4 – 13, 21 – 35, 85, 86, 98, 113 – 14, 121, 129, 151, 155, 159, 165, 180 Flanders/Flemings 30, 103, 133, 143, 145, 155, 156 Folbert of Bouillon 52 Folkmar 7 France/French 4, 6, 7, 9, 13, 14, 22, 63, 76, 85, 95, 102, 133, 135, 152, 155, 156 Franks 10, 11, 13, 27, 41, 43 – 5, 47, 52, 54, 59, 60, 71, 77, 106, 111 – 12, 116 – 18, 123, 129 – 32, 134, 142 – 5, 159 – 60, 161 – 4, 176, 177, 178, 186 Fulcher of Chartres, chaplain 14, 39, 69 Fulcher of Chartres, knight 46, 48, 50 Gabriel of Melitene 59 – 60 Galilee 67, 76, 105, 114, 157; sea of 55, 88 garrisons 23, 24, 25 – 6, 29, 30, 31, 33, 34, 39, 43, 45, 46, 53, 60, 63, 65, 69, 79, 99, 116, 129, 134, 143, 144, 151, 153, 154, 157, 163, 173, 175 Gaston of Béziers 48 Geldemar Carpenel 64, 67, 69, 72, 82 – 4, 114, 131, 156 Gemelmulc, emir of Ascalon 141, 142 Genoa/Genoese 15, 95, 106, 112 – 13, 115 – 16, 117 – 22, 123, 125, 153, 161, 166 Gerard, bishop 131, 134 Gerard, knight 144 Gerard, secretary to Baldwin 50 Gerard of Avesnes 156 Gerard the Chamberlain 152, 158 Gerbod of Windeke 137 Germany/Germans 4, 63, 85, 86, 97, 113, 133, 155 Gervase of Bazoches 156, 158

Gibelin, archbishop of Arles 93, 104; patriarch of Jerusalem 105, 106 gift-giving 8, 10, 13, 23, 29, 41, 51, 53, 60, 79, 96, 106, 134, 165, 166 – 7 Giselbert of Clermont-sur-Meuse 30 Godevere/Godehilde 3, 5, 34, 183 Godfrey III, duke of Bouillon 1, 2 Godfrey IV, duke of Bouillon 2 – 5, 21, 25, 42; at Constantinople 10 – 13; First Crusade 5 – 13, 24, 28 – 9, 33, 48, 51, 52 – 3, 60 – 1, 94, 183; ruler of Jerusalem 54, 61 – 3, 94 – 8, 112, 114, 129, 180, 181, 182, 187; tomb 178 Godfrey of Esch 4, 6, 7, 10 Godric, sea captain 137 Golan Heights 112 Gomorrah 80 Gottschalk 7 Gunfrid of the Tower of David 144, 158 Gutmann of Brussels 137 Haifa 67, 71 – 3, 76, 82 – 4, 111, 113, 114, 122, 145, 161 Harpin of Bourges 135, 136 Harran 38, 102, 160, 161 Hauran 112 Hebron 80, 138, 154, 156, 174, 177 Helim see Aqaba Henry I, king of England 156 Henry of Esch 4, 6, 9, 52 Heribrand of Bouillon 4 Holy Cross church, Acre 175 Holy Fire 86 – 9, 182 Holy Sepulchre, church 55, 87 – 9, 93, 94, 97, 99, 100, 101, 103, 105, 106, 139, 152, 158, 177 hospital see St John, hospital hostages 8, 11, 12, 30, 43, 45, 49 Hugh, count of Saint-Pol 157 Hugh Chostard 158 Hugh of Fauquembergues, lord of Tiberias 84, 131, 138 Hugh of Rebecques 156 Hugh of Saint-Omer 72 Hugh ‘the Great’ of Vermandois 9, 22, 24 Ida of Bouillon 1, 4, 5 Īlghāzī ibn Artūq of Mardin 164 infantry 7, 27, 30, 32, 43, 48, 70 – 1, 72, 79, 103, 113, 118, 130, 131, 133, 134, 137, 138 – 9, 141, 143, 144, 145, 153, 154, 163, 173, 175 Investiture Contest 66, 76, 86

202 Index Italy/Italians 2, 7, 13, 15, 32, 54, 59, 62 – 3, 64, 65, 66, 95, 102, 114 – 16, 117, 123, 124, 133, 151, 152, 154 – 6, 160, 165, 166 Jaffa 67 – 8, 72, 78, 97, 100, 101, 112 – 13, 114, 115, 119, 122, 124, 129 – 30, 131 – 2, 133, 134, 136, 137, 138, 139, 140, 141, 142 – 3, 144 – 5, 152, 153, 157, 158, 164, 183 Janāh al-Dawla of Homs 72 Jehosaphat, valley 93, 106, 177 Jericho 55, 107, 156 Jerusalem, city 53, 55, 68, 76, 78, 82, 83, 88, 106, 113, 124, 130, 137, 138, 139, 141, 145, 153, 154, 164, 173, 174, 175, 177, 183, 186; capture 61, 66, 86, 95, 97, 98, 112, 165, 187 Joffrey, Godfrey’s chamberlain 69 Jordan, river 55, 66, 124, 136, 154 Joscelin of Courtenay 156, 160, 161, 162 – 3, 185 – 6; count of Edessa 156 Jubail 120, 121, 161 Karbugha 51 – 2, 111 kingship 63, 65 – 6, 76 – 7, 81, 158 knights 2 – 10, 27, 30, 31, 32, 35, 39, 41, 46, 48 – 9, 50, 52, 53, 59, 60, 69, 70 – 2, 78, 100, 130, 131, 134, 135 – 6, 137, 138, 140, 144, 151 – 4, 155, 157 – 8, 161, 163, 167, 174, 176 Kogh Vasil 34, 35, 53, 84 Konya 38 Latakia 30, 54, 55 – 6, 61, 62, 65 – 6, 84, 86, 96, 118, 121 Lithard of Cambrai 136, 143, 158 Lombards 85, 133, 165 loot see plunder Lotharingia/Lotharingians 1 – 7, 15, 21 – 2, 25, 32, 39, 86, 117, 155, 156, 179 Lydda 130, 134 – 5; see also Ramla Ma’arra 53 Mahumeth, Turk see Baktāsh ibn Tutush Malik Ghāzī see Danishmend Mamistra 25, 30 – 2, 101 Manuel Boutoumites 23, 106 Marash 33, 38, 183 Matthew the Seneschal 69, 158 Maurice, cardinal-bishop of Porto 99 – 100, 117 – 18 Mawdūd of Mosul 111, 161 – 4

Melitene 43, 51, 59 – 60, 68 mercenaries 10, 40, 45, 47, 69, 113, 117, 174 merchants 31, 118, 122, 145, 154, 173 Milo of Clermont 83 money 27, 42, 48, 69, 78, 84, 99, 101, 106, 113, 115, 116, 117, 118, 153, 155, 160, 165 – 6, 167, 183; bezants 45, 47, 49, 51, 84, 99 – 101, 117, 153, 154; coins 16, 23, 48, 84, 117, 118, 142, 155, 155, 156, 166; dinars 153, 155; marks 99; moneyfiefs 152 Montréal castle 173, 174 Morellus, secretary 68 Mosul 111 – 12, 160 Mount of Olives 107, 177, 178 Mount Tabor, abbey 93, 101 Mount Zion, abbey 73, 77, 93 Nazareth 105, 161 negotiations 2, 8, 11, 12, 23, 28, 31, 45, 55, 62, 77 – 8, 80, 83, 103, 106, 114, 118, 124, 143, 153, 165, 166, 182 Nicaea: siege of 21 – 3, 106 Nicusus 35, 51 Nikephorion 38 Nikephoros Bryennios 10 Nikomedia 133 Nile, river 175, 176 Normandy/Normans 1, 3 – 4, 5, 13, 14, 15, 22, 24, 27, 32, 59, 64, 103, 155, 156, 160, 166, 183, 185 oaths 11, 12 – 13, 23, 43, 44, 63, 64, 65, 68, 78, 82, 83, 107, 113 – 14, 115, 122, 123, 165, 185 Pagan, chancellor 158 Palm Sunday 85 – 6, 107, 177 – 8, 182, 185 Paschal II, pope 93, 98 – 9, 101 – 5, 107, 113, 134, 174 – 5, 184 Peter of Astenois/Dampierre 6, 7, 12, 26 Peter the Hermit 7, 9, 21, 23 Petra 80 Picardy 155 pilgrimage 3, 4, 30, 54 – 6, 59 – 60, 61, 70, 77, 84, 85, 89, 95 – 6, 112, 152, 186 pilgrims 7, 9, 15, 29, 55, 67, 79, 85, 88, 98, 104, 113, 133 – 4, 137, 139, 140, 142, 144, 145, 152, 165, 175, 178, 179 Pilgrim’s Mount 161 Pisa/Pisans 61, 65, 66, 95, 114 – 16, 118, 119 – 22, 123, 153, 165 – 6

Index  203

Rainald III, count of Toul 6, 7, 26, 48 Ralph of Montpinçon 72 Ralph of Mousson 69, 72 Ramla 79, 101, 113, 141, 142, 157, 158; 1st battle 129 – 32, 133, 144, 151, 152, 153, 156, 183; 2nd battle 134 – 9, 152, 156, 183; 3rd battle 103, 141 – 4, 157, 159, 160; see also Lydda ransom 48 – 9, 52, 79, 84, 116, 117, 141, 142, 153, 160 Ravendel 34 – 5, 39, 51, 53 Raymond of Saint-Gilles 22, 24, 52, 60 – 1, 62, 64, 65, 68, 85 – 6, 96, 114, 121, 129, 133, 144, 161, 165 Red Sea 173, 174, 179 Reinard of Hamersbach 52 Reinold, master of archers 153 Richard of Salerno 32, 59 – 60, 160 Ridwan of Aleppo 52 – 3 Robert, bishop of Lydda 64, 69, 72, 105 Robert, chancery official 158 Robert, duke of Normandy 5, 12, 22, 24, 33, 61, 96, 156 Robert II, count of Flanders 5, 22, 24, 33, 51, 61, 96 Robert FitzGerard 69 Robert of Anzi 32 Robert of Apulia 153 Robert of Paris, cardinal 101 – 2, 103, 104 Roger, bishop of Lydda 105 Roger I, count of Sicily 100, 106, 166 Roger II, count of Sicily 106, 166, 175 Roger of Rozoy 158 Roger of Salerno, count of Antioch 163, 164, 167, 179 Rūj valley 162 Russia/Russian 88 – 9, 179

St George, monastery in Lydda 134 St John, hospital 93, 100, 106, 131, 154, 166 St Mary of Jehosaphat, abbey 93, 106 St Symeon, port 114, 123, 183 Saljūqs/Turks 21, 24, 33, 70 – 1, 112, 124, 125, 133, 141, 142, 143, 153, 160, 161, 165; see also Danishmendid Turks Samosata 38, 43, 45, 160 Saracens 46, 54 – 5, 72, 79, 100, 103, 118, 131, 136, 140, 142, 157, 163, 165, 167, 174, 176, 177; see also Egypt/Egyptians Sarūj 160, 162, 163 Scandalion castle 175 scouts 39, 70 – 1, 135, 136, 154 ‘Semelimech’, Egyptian army commander 141 Shaizar 156, 162, 164 ships 22 – 3, 65, 67, 70, 85, 107, 113, 114, 115 – 16, 117 – 19, 119 – 22, 123, 124, 131, 133, 137 – 8, 140, 141, 142, 144, 145, 152, 154, 167, 183 Sicily 106 – 7, 166 – 7, 174 – 5 Sidon 114, 119, 123, 125, 138, 141, 144, 145, 153, 154, 156, 157; capture 124, 158, 161, 179 siege engines 21, 22, 83, 116, 118, 119; hurling machines 22, 31, 83, 145; towers 61, 83, 116, 117 – 18, 119, 123 Sigurd Magnusson, king of Norway 124, 145 Simon the Constable 158 Sinai desert 167 Sodom 80 Sororgia 46, 49 – 50 source discussion 13 – 16, 40 – 1, 42, 64, 121 spies 12, 48, 52, 130, 154 spoils see plunder Stabelo, chamberlain 7 Stephen of Blois 9, 12, 14, 22, 24, 85, 133, 135, 136 Stephen of Burgundy 85, 133, 136 surrender 7, 23, 26, 34, 35, 39, 43, 46, 61, 82, 83, 103, 106, 114, 116 – 17, 120, 123, 124 – 5, 129, 151, 159 Symeon II, patriarch of Jerusalem 94 Syria 25, 33, 34, 38, 42, 47, 50, 65, 66, 85, 94, 95, 96, 111, 112, 122, 156, 162 Syrians 52, 54, 60, 73, 77, 79, 87, 134, 138, 177

sack 23, 116, 118, 120, 123 – 4, 151, 153 St Anne, nunnery 93, 183, 184 St Catherine’s Monastery, Sinai 174

Tancred 13, 22, 23, 24, 59, 61, 63, 71, 160, 163, 164; in Cilicia 24 – 30, 32, 33, 67; opposition to Baldwin’s succession

Pisellus, viscount 158 Pisellus of Wissant 50 plunder 7, 9, 10, 11, 23, 26, 27 – 8, 30, 31, 32, 52, 55, 70, 73, 83, 115 – 19, 120, 123 – 4, 131, 132, 136, 138 – 141, 142, 145, 151, 153 – 4, 164, 175 Pons of Tripoli 163, 164, 179 prisoners see captives qadi of Caesarea 116 – 18, 153 Qilij Arslān 21, 22, 51

204 Index 64 – 5, 67, 72 – 3, 76, 82 – 4, 114; regent of Antioch 83 – 5, 100, 101, 106, 131 – 2, 139, 156, 160, 161 – 2, 183 Taphnuz 46 – 7, 49 Tarsus 25 – 30, 32, 82, 101 Tatikios 22 Tell Danith, battle 164 Templum Domini 93, 158 Thoros, duke of Edessa 39, 40, 41, 42 – 5, 59 Tiberias 55, 61, 82, 83, 112, 144, 156, 161; lake 163 Tortosa 55, 133, 156, 161 Tower of David 6, 64, 68, 72, 85, 158 Transjordan 167 treaties 8, 112, 122, 123, 143; treaty of Devol 165 tribute 46, 114 – 15, 125, 153 Tripoli 55, 72, 86, 119, 124, 142, 144, 158, 160 – 1; capture 123, 131, 142, 179 truces 99, 112, 114 – 15, 120, 123 – 4, 157, 159 True Cross, relic 98, 99, 105, 117, 130 – 1, 132, 134, 135, 138 – 9, 141, 142 Ṭughtigin of Damascus 111 – 12, 143, 156, 163, 164 Tulln 7 – 8 Turbessel 32, 34 – 5, 39, 41, 51, 52, 53, 156, 162 Turcopoles 10 – 11

Tyre 72, 106, 111 – 12, 115, 119, 124, 125, 141, 144, 145, 153, 154, 156, 159, 165, 174, 175, 179, 181 Udelard of Wissant 34 Ulric, viscount 158 Urban II, pope 3, 4, 66, 94, 95 ‘Ursinus’ 29 Valania 54 Venetians 82 – 3 Wadi Mūsa 80, 154 Walter, count of Melun 10 Walter Mahomet 156, 157 Walter of Bouillon 4 Warner, count of Grez 4, 6, 7, 12, 22, 63 – 4 Welf IV, duke of Bavaria 133 Welf of Burgundy 29 Wicher the Swabian 69, 72, 83 William, son of Robert of Normandy 154, 156 William II, count of Nevers 133 William IX, duke of Aquitaine 133 William Jordan 143, 161 William of Poitiers 101 William the Carpenter 101 Winemer of Boulogne 30 Wiric of Flanders 69, 83