Jerusalem The Golden

Jerusalem the Golden
Outremer
Studies in the Crusades and the Latin East
Volume 3
General Editor
Dr Alan V. Murray (University of Leeds)
Editorial Board
Prof. Alfred Andrea (University of Vermont)
Prof. Simon Barton (University of Exeter)
Dr Jochen Burgtorf (California State University, Fullerton)
Prof. John France (University of Swansea)
Prof. Nikolas Jaspert (University of Bochum)
Prof. Kurt Villads Jensen (University of Southern Denmark)
Prof. Peter Lock (York St John University)
Prof. Graham Loud (University of Leeds)
Dr Christoph Maier (University of Zürich)
Prof. Helen Nicholson (University of Cardiff )
Jerusalem the Golden
The Origins and Impact
of the First Crusade
edited by
Susan B. Edgington
and
Luis García-Guijarro
F
H
Cover illustration: Photograph by Susan B. Edgington
© 2014 Brepols Publishers n.v., Turnhout, Belgium (English translation)

All rights reserved. No part of this book may be reproduced, stored
in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of the publisher.
ISBN 978-2-503-55172-2 (print)

ISBN 978-2-503-55192-0 (online)
D/2014/0095/136
Printed on acid-free paper
In memoriam
H. E. J. Cowdrey (1926–2009)
and
Marco Tangheroni (1946–2004)
Table of Contents
List of Figures ix
List of Tables ix
List of Maps ix
Abbreviationsx
Preface by Luis García-Guijarroxi
Introduction
Susan B. Edgington
The First Crusade: Expanding the Historiography 1
I. The Origins and Background of the First Crusade

†H. E. J. Cowdrey
New Dimensions of Reform. War as a Path to Salvation 11
Jean Flori
Jérusalem terrestre, céleste et spirituelle.
Trois facteurs de sacralisation de la première croisade 25
Manuel Rojas
Some Problems in the Study of the Conduct of Warfare
in the Eleventh Century 51
Mike Carr
Between Byzantium, Egypt and the Holy Land.
The Italian Maritime Republics and the First Crusade 75
Robert Somerville
The Crusade in the Councils of Urban II beyond Clermont 89
II. The Course of the Crusade

Jonathan Riley-Smith
An Army on Pilgrimage 103
Judith Bronstein
1096 and the Jews. A Historiographic Approach 117
John France
Moving to the Goal, June 1098 – July 1099 133
Some Considerations on the Crusaders’ Letter
to Urban II (September 1098) 151
Stephen J. Spencer
Constructing the Crusader.
Emotional Language in the Narratives of the First Crusade 173
vii
Table of Contents
Alan V. Murray
The Siege and Capture of Jerusalem
in Western Narrative Sources of the First Crusade 191
III. The Impact of the Crusade

Michael Brett
The Muslim Response to the First Crusade 219
Sophia Menache
Emotions in the Service of Politics.
Another Perspective on the Experience of Crusading (1095–1187) 235
Sini Kangas
Growing Up to Become a Crusader. The Next Generation 255
IV. The Afterlife of the Crusade

Elena Bellomo
Rewriting the Past.
The Conquest of the Holy Sepulchre
in the Memory of Italian Communal Cities 275
Simon T. Parsons
Making Heroes out of Crusaders.
The Literary Afterlife of Crusade Participants in the Chanson d’Antioche291
Carol Sweetenham
The Count and the Cannibals.
The Old French Crusade Cycle as a Drama of Salvation 307
Ruth Bartal
The Image of the Saracen in Romanesque Sculpture.
Literary and Visual Perceptions 329
Robert Irwin
History, Fiction and Film. Islam Faces the Crusaders 347
Maps 373
Index 377
viii
List of Figures
Fig. 1: The Cathedral of Angoulême, twelfth century (photo Bartal) 333

Fig. 2: Oloron-Sainte-Marie – West Portal (photo Bartal) 335
Fig. 3: San Julian – Santa Cruz de Mena – West Portal (photo Bartal) 338
Fig. 4: San Pelayo de Mena – West Portal (photo Bartal) 338
Fig. 5: Oloron-Sainte-Marie – West Portal (photo Bartal) 340
Fig. 6: Soto de Bureba – West Portal (photo Bartal) 342
Fig. 7: Vallejo de Mena – West Portal (photo Bartal) 342
Fig. 8: Almendres – West Portal (photo Bartal) 343
Fig. 9: Tubilla del Agua – Window of the Tower
(photo A. Ferro © Museu Frederic Marès, reproduced with permission) 343
List of Tables
Table 1: Length of Accounts of the Siege and Capture of Jerusalem

in Selected Narrative Sources of the First Crusade
(= number of pages in RHC editions) 202
Table 2: Analysis of Content in Accounts of the Siege and Capture
of Jerusalem in Selected Narrative Sources of the First Crusade 204
List of Maps
Map 1: The march to Constantinople 373

Map 2: From Constantinople to Antioch 373
Map 3: The siege of Antioch, winter 1097–98 374
Map 4: From Antioch to Jerusalem 375
Map 5: The capture of Jerusalem, 15 July 1099 376
ix
Abbreviations
MGH Monumenta Germaniae Historica
SS Scriptores (in folio)
PL Patrologia Cursus Completus. Series Latina, ed. J.P. Migne,

221 vols (Paris, 1844–64)
RHC Recueil des historiens des croisades, ed. Académie des

Inscriptions et Belles-Lettres (Paris, 1841–1906)
Arm. Documents arméniens, 2 vols (1869–1906)

Occ. Historiens occidentaux, 5 vols (1844–95)
Or. Historiens orientaux, 2 vols (1872–1906)
RIS Rerum Italicarum Scriptores, 3 series (1723-)
RS Rolls Series: Rerum Britannicarum Medii Aevi Scriptores

(London, 1858–96)
x
PREFACE
The nine-hundredth anniversary of the series of events which were much

later labelled as the ‘First Crusade’ filled the years 1995–1999 with all
types of commemorative intellectual celebrations across the world, and
logically with much more emphasis across western Europe from where
the armed contingents departed for the Levant in the summer of 1096.
Two special historical moments attracted most attention: Pope Urban II’s
preaching of the crusade at Clermont on 27 November 1095 and the
conquest of Jerusalem, the final goal of the expedition, on 15 July 1099.
Books, articles, talks and conferences were massively devoted to them.
Interest went beyond the academic world. Terry Jones, a member of the
Monty Python group, made a television series on the crusades and in
December 1995 as respected a weekly as The Economist dedicated a long
article to reinterpreting this movement with which the Middle Ages
are closely associated in modern times. At the turn of the twenty-first
century, the First Crusade and the subsequent expeditions to the eastern
Mediterranean have epitomised the mixture of fascination and rejection
with which popular minds view medieval times.
Although on a minor scale, Spain was no exception to these celebra-
tions. One of the most significant contributions in this country was a
series of three conferences which aimed to bring together – at intervals
and in different places – a select group of international historians around
themes related to the First Crusade and crusade studies. The idea was to
attract a core team of scholars representing as many intellectual traditions
as possible which would meet in Spain in 1995, 1999 and 2001 to discuss
the origin and completion of the First Crusade, and also the development
of crusade historiography in the fifty years following the publication of the
first volume of Sir Steven Runciman’s History of the Crusades. The initial
conference took place in Madrid on 16–18 November 1995. The topic was
obvious: ‘The First Crusade Nine Hundred Years After: The Council of
Clermont and the Origin of the Crusade Movement’. Fifteen historians
gathered: eight Spanish, four British, one American, one Italian and one
Portuguese. The next meeting was in Huesca between 7 and 11 September
1999. The main point of discussion was of course the conquest of ‘Jerusalem
the Golden’, while there was a subtheme related to events in Iberia. Huesca
had been taken from the Muslims in 1096 and that fact favoured com-
parisons between the eastern and western ends of the Mediterranean. The
number of historians swelled to twenty-four. The Madrid group was made
xi
Preface
richer by very valuable additions. Finally, the conference on historiogra-

phy was held at the end of July 2001 in Teruel, on the southern frontier
of the medieval kingdom of Aragon. The number of active participants
doubled that at Huesca. The project had ended successfully, achieving its
main goals of interrelating intellectual traditions in debates on the origins
of the crusade and on modern historiography of that movement. Most of
the historians who met in Madrid did so later in Huesca and Teruel. This
fact, not so usual in conferences, gave a sense of continuity which helped
to make discussions more fruitful and conviviality more enjoyable.
But, as Ovid wrote and crusaders recollected in at least one of their
letters, ‘sometimes happiness is clouded by sadness’. Conferences are two-
act plays. Discussions in situ are only preliminaries which vanish without
the careful afterthoughts of a published text. In this respect the project
did not fully fulfil its goals. The Madrid conference came out swiftly as a
book in 1997. Huesca and Teruel were not so lucky. The latter has faced
intractable problems due to the massive number of papers. The proceed-
ings of Huesca were about to be issued on several occasions, but all sorts of
circumstances prevented that end at its due time or even later. When I had
almost despaired, Brepols has come to the rescue of highly valuable mate-
rial. But life moves on, almost fifteen years have passed since the golden
September days in 1999, and consequently the book cannot comprise
all the texts presented at Huesca. Some historians have understandably
published their contributions elsewhere or think that the content has
grown old without repair. Fourteen papers have unfortunately been lost
this way. Ten contributions remain and finally see the light of day (Ruth
Bartal, Michael Brett, H. E. J. Cowdrey, Jean Flori, John France, Luis
García-Guijarro, Robert Irwin, Jonathan Riley-Smith, Manuel Rojas,
and Robert Somerville). To fill the gap nine scholars have volunteered
to write papers related to the original theme of the conference and now
of the book (Elena Bellomo, Judith Bronstein, Mike Carr, Sini Kangas,
Sophia Menache, Alan V. Murray, Simon Parsons, Stephen Spencer and
Carol Sweetenham). To the former I wish to express my gratitude for
their biblical patience and understanding. I give my warmest thanks to
the latter not only for their willingness to participate and to keep afloat a
foundering ship, but mainly for the quality of their contributions which
makes a fine original product even better. This book can no longer be the
proceedings of the Huesca conference, but the breadth and quality has
suffered in no way from it. Quite the contrary.
In this long process to achieve publication the project has lost two of its
most valuable original participants. In February 2004 Marco Tangheroni
passed away without being able to prepare his Huesca contribution for
xii
Preface
publication. H. E. J. Cowdrey sent his paper in due time, but most sadly
death caught him in December 2009 before he could see it printed. I
wish to offer both of them a tribute of appreciation and gratitude for
their brilliant scholarship and personal warmth.
I would not be writing these lines if Alan V. Murray and Susan
Edgington had not offered to publish Huesca with Brepols and worked
most diligently and effectively to achieve that end and complete the exist-
ing material to produce a sound book. I am very grateful to both of them.
Madrid, February 2014
xiii
Introduction
The First Crusade: Expanding the Historiography
Susan B. Edgington
Queen Mary University of London
The two elements in this collection’s title reflect the period of historio-
graphical ferment to which my co-editor has referred in his preface. The
Huesca conference in 1999, commemorating the capture of Jerusalem
900 years before, was entitled ‘Jerusalem the Golden: The Conquest of
the Dream’, an elegant reference to the aspirations of the First Crusade via
John M. Neale’s 1858 translation of the twelfth-century poem by Bernard
of Morlaix:
Urbs Sion aurea, patria lactea, cive decora,
Omne cor obruis, omnibus obstruis et cor et ora.
Nescio, nescio quae iubilatio, lux tibi qualis,
Quam socialia gaudia, gloria quam specialis.1
The other liminal date, 1095, was marked by conferences in Madrid, ‘La
Primera Cruzada’; in Clermont, ‘Autour de la Première Croisade’; and
in London, ‘Deus Vult: The Origins and Impact of the First Crusade’.2
1
Bernardus Morlanensis, ‘De contemptu mundi’ (‘The Better Land’), lines
269–7, The Oxford Book of Medieval Latin Verse, ed. Stephen Gaselee (Oxford,
1928), pp. 9–91. Neale’s hymn begins: ‘Jerusalem the golden, with milk and
honey blest, / Beneath thy contemplation sink heart and voice oppressed. / I
know not, O I know not, what joys await us there, / What radiancy of glory,
what bliss beyond compare’: Hymns Ancient and Modern, New Standard
Edition (1983), no. 184, accessed at http://www.oremus.org/hymnal/amns.
html [16 February 2014].
2
La Primera Cruzada, Novecientos Años Después: El Concilio de Clermont y los
Orígines del Movimento Cruzado, ed. Luis García-Guijarro Ramos (Castellón,
1997); Autour de la Première Croisade, ed. Michel Balard (Paris, 1996) [In the
same city at the same time another international conference resulted in the
collection, Le Concile de Clermont de 1095 et l’appel à la croisade (Rome, 1997)];
The First Crusade: Origins and Impact, ed. Jonathan Phillips (Manchester, 1997).
Jerusalem the Golden. The Origins and Impact of the First Crusade, ed. by Susan B. Edgington and
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. 1–7.
FHG DOI: 1.1/M.OUTREMER-EB.1.11
Susan B. Edgington
It was the subtitle of the latter that we felt could not be improved upon
to describe the scope of the current volume.
Fifteen years have passed since this surge of commemorative events and
it might be thought that enthusiasm for the study of the First Crusade
would subside in its aftermath. That it did not is only partly owing to the
interest engendered in a new generation of crusades students by these same
commemorative events. The terrorist attacks of 11 September 2001 had
their impact on the small world of crusades studies as well as causing global
shock. There was a period of introspection and reassessment, perhaps most
clearly evinced in Jonathan Riley-Smith’s Bampton Lectures in America,3
and Christopher Tyerman’s The Debate on the Crusades 1099–2010.4
Focusing, as we must here, on the First Crusade, reverberations of 9/11
and subsequent events are seen, for example, in the concluding sentences
of Thomas Asbridge’s monograph, written soon after:
By 1300, the memory of the crusade as a war engendered by fanatical
hatred had become embedded in the collective consciousness of western
and eastern society. The lines of religious discord hardened; Christendom
and Islam had been set on the path to enduring conflict.5
Notably, the US paperback edition of the same book added a second subti-
tle: The Roots of Conflict between Christianity and Islam.6 This may now be
seen as a misguided attempt by the publishers to make the book ‘relevant’,
and it obscures the real merits of this book, which is a very readable, if
conventional, narrative of the events of the First Crusade, amply supported
by evidence from the primary sources. By 2011, Jay Rubenstein made a
clear distinction between popular reactions and scholarly responsibility:
Some historians, and no doubt some readers, will be inclined to see in the
crusade […] the birth of the modern world, as we find ourselves living in
a time again marred by religious strife and characterised by an instinctive
division between East and West.
But he continued,
Caution is necessary on this point. The word ‘crusade’ has been used to
distressing effect by all sides in recent global conflicts and never with
anything resembling thoughtfulness or precision.7
3
Published as The Crusades, Christianity, and Islam (New York, 2008).
4
Christopher Tyerman, The Debate on the Crusades 1099–2010 (Manchester, 2011).
5
Thomas Asbridge, The First Crusade: A New History (London, 2004), p. 339.
6
Thomas Asbridge, The First Crusade: A New History: The Roots of Conflict
between Christianity and Islam (New York, 2005).
7
Jay Rubenstein, Armies of Heaven: The First Crusade and the Quest for Apocalypse
(New York, 2011), p. 324.
2
Introduction
Leaving aside this gesture towards modern geo-political concerns,

Rubenstein’s interpretation of the events, and especially the motivation, of
the First Crusade is a millenarian one: ‘Woven into this history, then, more
thoroughly than in any previous telling of the First Crusade saga are all the
dreams, visions, and miracles that occurred during the expedition’.8 It is not
a radical approach – Norman Cohn had pointed the way in his seminal
work of 19579 – but it is thoroughly worked out and provides an interesting
perspective. The same may be said in all respects of the monograph by Peter
Frankopan subtitled The Call from the East.10 This is an interpretation cen-
tred on the Byzantines and the role of Alexios I Komnenos, in the tradition
of Steven Runciman’s very influential 1951 history of the First Crusade.11
A much more original treatment is Conor Kostick’s The Social Structure
of the First Crusade.12 Kostick tackles head-on the most eminent of con-
temporary historians of the crusades, specifically those who have doubted
the validity, or even the possibility, of a Marxist or materialist analysis.13
His valiant attempt to provide this depends on a close analysis of the lan-
guage used by the early historians of the crusade. This source material has
its limitations, as Kostick fully acknowledges, which prevent an entirely
successful reconstruction of the social structure of the movement, but
there are some percipient observations nonetheless, and a methodology
that could usefully be applied to other crusades and other areas of study.
Kostick also produced a more conventional narrative history in 2009,
entitled The Siege of Jerusalem.14
The anglophone world, then, has produced a handful of book-length
histories of the First Crusade. In France Jean Flori has performed almost the
same feat single-handed – alongside more general studies, including one on
apocalyptic prophecy in the medieval period – publishing biographies of
Peter the Hermit and Bohemund and an excellent study of the chroniclers
8
Rubenstein, Armies of Heaven, p. xiv.
9
Norman Cohn, The Pursuit of the Millennium: Revolutionary Millenarians and
Mystical Anarchists of the Middle Ages (London, 1957).
10
Peter Frankopan, The First Crusade: The Call from the East (London, 2012).
11
Steven Runciman, A History of the Crusades, 3 vols (Cambridge, 1951–54),
1: The First Crusade and the Foundation of the Kingdom of Jerusalem.
12
Conor Kostick, The Social Structure of the First Crusade (Leiden, 2008).
13
Kostick, Social Structure, pp. 299–300, citing Marcus Bull, ‘Muslims and
Jerusalem in Miracle Stories’, in The Experience of Crusading 1: Western
Approaches, ed. Marcus Bull and Norman Housley (Cambridge, 2003), pp. 13–38
(here p. 18); Norman Housley, Contesting the Crusades (Oxford, 2006), p. 79.
14
Conor Kostick, The Siege of Jerusalem: Crusade and Conquest in 1099 (London,
2009).
3
Susan B. Edgington
of the crusade in the last fifteen years.15 The twenty-first century has, in fact,
seen a remarkable number of monographs on the First Crusade, more than
any comparable period in the past. However, monographs are not the cutting
edge of historical research, since journals and essay collections offer shorter
publication times for smaller scale studies in depth. It would be impossible to
list here all the articles and essays focused on the First Crusade and published
since 1999, although some of them will be commented on below when sur-
veying current historiographical trends, and many more will be found in the
notes to the chapters in this volume. Nevertheless, it is important to mention
the journal Crusades, published annually for the Society for the Study of the
Crusades and the Latin East from 2002. Because Crusades incorporates the
society’s bulletin as well as articles and book reviews it contains important
advance information on forthcoming publications and work in progress,
which has undoubtedly contributed to the increase in quantity and qual-
ity of published research on all aspects of crusades studies. For an excellent
example, see the complementary papers published in Crusades by Benjamin
Kedar and Konrad Hirschler, using a detailed examination of Western and
Eastern sources respectively, on the conquest of Jerusalem in July 1099.16
The first years of the third millennium have also seen the publication
of new editions of some of the most important Latin sources for the First
Crusade, not only establishing sound texts but also reassessing matters
of authorship, dating, bias, intended audience, and so on. These are – in
date of publication order – Albert of Aachen, the ‘Historia belli sacri’,
Ralph of Caen, Robert the Monk, and Baldric of Bourgueil.17 Again, the
15
Jean Flori, Pierre l’Ermite et la Première Croisade (Paris, 1999); Flori,
Bohémond d’Antioche: Chevalier d’aventure (Paris, 2007); Flori, Chroniqueurs
et propagandistes: Introduction critique aux sources de la première croisade
(Genève, 2010); Flori, L’Islam et la fin des temps: L’interprétation prophétique
des invasions musulmanes dans la chrétienté médiévale (Paris, 2007).
16
Benjamin Z. Kedar, ‘The Jerusalem Massacre of July 1099 in the Western
Historiography of the Crusades’, Crusades 3 (2004), 15–75; Konrad Hirschler,
‘The Jerusalem Conquest of 492/1099 in the Medieval Arabic Historiography of
the Crusades: From Regional Plurality to Islamic Narrative’, Crusades 13 (2014),
37–76.
17
Albert of Aachen, Historia Ierosolimitana: History of the Journey to Jerusalem, ed.
and trans. Susan B. Edgington (Oxford, 2007); Historia de via et recuperatione
Antiochae atque Ierusolymarum, ed. Edoardo D’Angelo (Firenze, 2009); Radulphi
Cadomensis Tancredus, ed. Edoardo D’Angelo (Turnhout, 2011); The Historia
Iherosolimitana of Robert the Monk, ed. Damien Kempf and Marcus G. Bull
(Woodbridge, 2013); The Historia Ierosolimitana of Baldric of Bourgueil, ed. Steven
J. Biddlecombe (Woodbridge, 2014).
4
Introduction
appearance of so many key publications of First Crusade texts within a

decade is remarkable. At the same time, the internet has made old edi-
tions of texts freely available, including all sixteen folio volumes of the
Recueil des historiens des croisades which are accessible online courtesy
of the Bibliothèque nationale de France.18 While good editions of texts
are proliferating, in some countries, including the United Kingdom, the
numbers who are able to read them are lessening. This is clearly regrettable
from the point of view of scholarship, but it has led to the production of a
number of translations of primary sources into modern languages. These
are not, of course, only into English and French (though my own ability
to benefit from them stops there): I am aware of a translation of the Gesta
Francorum into Russian, and a scholar is working on a Chinese translation
of Albert of Aachen’s Historia.19 Monolingual anglophone students are
particularly well provided for. They have not only the greatly expanded
second edition of Edward Peters’ primary source collection,20 but also
Ashgate’s Crusade Texts in Translation series, which notably includes
the chronicle of Ibn al-Athīr as well as Latin and Old French sources.21 It
may thus be said that in terms of published scholarship the study of the
First Crusade is in a healthy condition.
So, it may be asked, what need of another collection of essays on the
subject? In 1999 Luis García-Guijarro brought together an international
panel of historians, as he has explained in his preface. Their papers were
largely ‘mainstream’, in that they presented the participants’ recent
research into events leading up to and during the First Crusade. Notable
outliers were Ruth Bartal on the visual representation of Saracens and
Robert Irwin on modern portrayals of the crusades in fiction and film.
Both of these papers have been updated for inclusion in this collection.
Delays in publication allowed the selection of essays to be refined in 2007:
Judith Bronstein and Alan Murray filled gaps by writing on the Jewish
experience in the Rhineland and the narrative sources for the capture of
Jerusalem (respectively); Sophia Menache broke new ground by writing
18
At www.gallica.bnf.fr
19
Valentin Portnykh, The Deeds of the Franks and the Other Pilgrims to Jerusalem:
Commented Russian Translation (Novosibirsk, 2010) and Wang Xiang-Peng,
a PhD candidate at China’s Northeast Normal University, projected date of
publication November 2013 (communicated by email, 26 March 2012).
20
The First Crusade: The Chronicle of Fulcher of Chartres and Other Source
Materials, ed. Edward Peters, 2nd edn (Philadelphia, 1998).
21
The Chronicle of Ibn al-Athīr for the Crusading Period from al-Kāmil fī’l-ta’rīkh.
Part 1: The Years 491–541/1097–1146. The Coming of the Franks and the Muslim
Response, trans. D. S. Richards (Aldershot, 2006).
5
Susan B. Edgington
on the emotional impact of crusading. All three essays have been revised
for inclusion here. Further vicissitudes delayed publication again, so that
in 2013, when it was decided to go ahead as part of the Outremer series,
further gaps had opened up, but also an opportunity to recruit histori-
ans, some of them in the early stages of their careers, working in new or
neglected fields.
It has to be admitted, however, that one topic that remains overlooked
here – although it is an historiographical strand that remains popular
(for example in universities) – is the experience of women. Various
attempts have been made to look at this for the crusades in general, for
example, by Régine Pernoud.22 However, the ‘time of ’ the crusades in
the title identifies the evidential problem, which also makes two studies
in German, both published in 2003, less than satisfactory.23 In English,
Natasha Hodgson grounded her study of women and the crusades more
soundly on ‘historical narrative’, but again it takes the longer time period.24
The most satisfactory treatment of women in general on the First Crusade
is probably by Conor Kostick, who followed his chapters looking at the
social stratification of the crusading armies with one on ‘Women and the
First Crusade’.25 More recently, Alan Murray has considered the plight
of women who were single, or became so as a result of the deaths of their
husbands or protectors.26 Wider issues of gender also merit a more closely
focused examination than has hitherto been accorded.27 This is a gap in
the historiography that remains; others we have attempted to fill for this
publication.
22
Régine Pernoud, La Femme au temps des croisades (Paris, 1990).
23
Christine Dernbecher, ‘Deus et virum suum diligens’: Zur Rolle und Bedeutung
der Frau im Umfeld der Kreuzzüge (St. Ingbert, 2003); Sabine Geldsetzer,
Frauen auf Kreuzzügen 1096–1291 (Darmstadt, 2003); see also the review of
Dernbecher and Geldsetzer by Christoph Maier, Crusades 5 (2006), 213–16;
Maier, ‘The Roles of Women in the Crusade Movement: A Survey’, Journal of
Medieval History 30 (2004), 61–82.
24
Natasha R. Hodgson, Women, Crusading and the Holy Land in Historical
Narrative (Woodbridge, 2007).
25
Kostick, Social Structure, pp. 271–85.
26
‘Sex, Death and the Problem of Single Women in the Armies of the First
Crusade’, in Shipping, Trade and Crusade in the Medieval Mediterranean:
Studies in Honour of John Pryor, ed. Ruthy Gertwagen and Elizabeth Jeffreys
(Farnham, 2012), pp. 255–70.
27
See essays in Gendering the Crusades, ed. Susan B. Edgington and Sarah
Lambert (Cardiff, 2001) for the scope of such an approach, although again
individual essays take a longer time frame.
6
Introduction
A current area of interest is the memorialisation of the crusades, and

we have been able to include essays that reflect this.28 Mike Carr’s chapter
on the involvement of the Italian maritime republics in the First Crusade
is complemented by Elena Bellomo’s on the crusade in the memory of
Italian communal cities. Both Simon Parsons and Carol Sweetenham
have examined the way the Chanson d’Antioche and other Old French
epics acted to enhance the heroism (Parsons) or villainy (Sweetenham)
of participants in medieval memory of the First Crusade. This use of
literary texts to evidence medieval attitudes and ideas is important after
a long period when the only question asked of the Chanson d’Antioche
seemed to be whether it contained any authentic eye-witness material or
not. Sini Kangas has also used literary evidence alongside narrative texts
to examine reception of the First Crusade in western Europe by the two
generations following the crusade; this not only takes her into memori-
alisation, but also into the history of childhood (an area more neglected
than women’s history so far as the crusades are concerned). Finally, of our
‘new’ contributors, Stephen Spencer is working in the fast growing field
of the history of emotions, and his work on emotional language in the
narratives of the First Crusade provides a counterpoint to Menache’s on
emotions in Outremer.
Thus we hope we have identified some new directions in crusades histo-
riography as well as presenting the best of established scholarship. I wish
to conclude, however, with some words about the original ‘Huesca ten’.
The late John Cowdrey’s essay is included in its original version (as
accepted for publication in 2007). Professors Jean Flori and Jonathan
Riley-Smith are both, at the time of writing, contending with serious
illness. Both have been amazingly generous and supportive of our efforts
to bring their essays to publication. Their texts have remained virtually
unchanged, but the notes have been updated, in particular to take account
of recent editions of primary sources. The other contributors, all of them
eminent and busy people, have updated their contributions and responded
graciously to queries and comments. I present their collective contribu-
tion to the historiography of the First Crusade with gratitude and pride.
28
For the theoretical underpinning of this trend, see Nicholas Paul and Suzanne
Yeager, ‘Introduction: Crusading and the Work of Memory, Past and Present’,
in Remembering the Crusades: Myth, Image, and Identity (Baltimore, MD,
2012), pp. 1–25 (here 2–11).
7
I
The Origins and Background

of the First Crusade
New Dimensions of Reform
War as a Path to Salvation
†H. E. J. Cowdrey
University of Oxford
The subject of this chapter is the epoch-making expansion in the notion

of war as offering a path to salvation which took place under popes
Gregory VII (1073–85), Victor III (1086–87), and Urban II (1088–99);
the most notable result of this expansion was the First Crusade, which
culminated in the capture of Jerusalem in 1099. This chapter is based upon
an assumption about the crusade which must be declared at the outset. It is
that in its origins as in its course from beginning to end, the First Crusade
had, in roughly equal proportions, the character of an armed pilgrimage
and that of a holy war. Such an assumption may the more readily be made
in view of the cogent and fully documented case presented by Jean Flori.1
Granted such prominence of holy war in the genesis of the crusade, it
may first be suggested that death in battle, when the cause is a righteous
one, as martyrdom carrying immediate access to the glories of heaven,
was the principal tradition that the popes and other makers of ecclesi-
astical opinion drew upon and developed. It is widely recognised that,
by Gregory VII’s time, there was a history of popes who promised that
death on military campaigns, especially against the Saracens, Hungarians,
and Northmen, would be rewarded by everlasting life; such promises
were notably made by Popes Leo IV (847–55) and John VIII (871–82).2
Most famously, after an army, mainly of Germans, led by the first of the
major eleventh-century reform popes, Leo IX (1049–54), suffered defeat
at the hands of the Normans in 1053, a strong tradition had it that Leo
1
Jean Flori, Pierre l᾿ermite et la première croisade (Paris, 1999), esp. pp. 132–51,
208–25.
2
See, for example, Carl Erdmann, Die Entstehung des Kreuzzugsgedankens
(Stuttgart, 1935), pp. 21–24, trans. as The Origin of the Idea of Crusade, Marshall
W. Baldwin and Walter Goffart (Princeton, 1997), pp. 25–28; Flori, Pierre
l’ermite, pp. 123–24, 134–35, 220.
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. 11–.
FHG DOI: 1.1/M.OUTREMER-EB.1.11
H. E. J. Cowdrey
proclaimed the fallen to have died in defence of righteousness and to have

taken their place in heaven among the martyrs.3
Under Gregory VII, the theme of death in properly directed warfare
as a path to eternal salvation in the company of the martyrs underwent
intensification and development. It did so in connection with Gregory’s
concern to win Christians for a militia sancti Petri or a militia Christi – for
a service or warfare (the double significance of the Latin noun militia must
be kept in mind) of St Peter and, through him, of Christ. In Gregory’s let-
ters such a militia is illustrated in his plans of 1074 for a widely recruited
expedition to rescue Eastern Christians who were suffering slaughter and
persecution at the hands of the Saljūq Turks – a Muslim enemy. Those
who gave their lives for their afflicted brothers could thereafter expect to
be crowned as their reward. Both the example of Christ the Redeemer
who gave his life for men and the duty of fraternal charity in imitation
of Christ who thus died (cf. 1 John 3:16) should move men to offer their
own mortal flesh on behalf of Christ who is life eternal.4
If one views Gregory’s teaching overall, a distinctive feature of it was
that he insistently presented the challenge of martyrdom as thus conceived
to all sorts and conditions of men. First and foremost, it was a challenge
addressed to himself. He did so in connection with his eastern plans of
1074: the suffering of Eastern Christians moved him to a desire to die
himself, since he would rather lay down his life for them than neglect
them and command the whole world according to the will of the flesh.5
He was to write in similar terms of the papal office in general: thus, in
1083, he insisted that his succession in the line of popes imposed upon
him an inevitable necessity to fight against the enemies of God with the
sword of the Spirit even to death, if that should be necessary.6 Archbishops
had a like duty and even privilege. In Muslim North Africa, Archbishop
Cyriacus of Carthage had suffered beatings and mutilation at the hands
of the emir. Gregory praised the archbishop’s fortitude: by such witness
3
See esp. ‘Historia mortis et miraculorum S. Leonis IX’, cap. 1.2, in PL 143, col.
527; Bruno of Segni, ‘Vita S. Leonis papae IX’, in PL 165, cols 1157–58; Bonizo of
Sutri, Liber ad amicum, ed. Ernst Dümmler, Lothar von Heinemann, Friedrich
Thaner et al., MGH Libelli de lite imperatorum et pontificum saeculis XI et
XII conscripti, 1 (Hannover, 1891), nos 5, 9, pp. 589, 620.
4
Das Register Gregors VII (Gregorii VII Registrum), 1.49, 2.37, ed. Erich Caspar,
2 vols, MGH Epistolae selectae (Berlin, 1920–23), 2: 75, 173; The Epistolae
vagantes of Pope Gregory VII, ed. and trans. H. E. John Cowdrey (Oxford,
1972), no. 5, pp. 12–13.
5
Register Gregors VII, 2.31, 2: 166.
6
Register Gregors VII, 9.35, 2: 622–23.
12
the Church of old had grown; by such means it was now being renewed.
But his culminating thought was that it would have been better still if
the witness had been to the point of death: ‘how much more precious
would have been the confession of your faith if, after the stripes that you
suffered, you had by showing the Muslims their error and by preaching
the Christian faith come to the pouring out of life itself !’7 Bishops, too,
must be prepared for the supreme sacrifice; thus, for example, in 1076,
Bishop Simeon of Oca-Burgos was exhorted to strive and to labour to the
shedding of his blood, if that should be necessary.8 In 1082, in a pastoral
letter addressed to all clergy and laity who were faithful to the apostolic
see, Gregory ended by exhorting them:
If we wish with God’s help quickly and bravely to rout the ancient enemy
[…] let us strive not only not to avoid the persecution that he inflicts and
death for righteousness, but from love of God and for the defence of the
Christian religion even to desire them.9
Amongst the laity, Gregory addressed such calls to kings,10 great princes,11
and above all to knights. It was to them especially that his calls in 1074
to an eastern expedition were addressed. In his last encyclical letter from
Salerno in 1084, he had above all in mind faithful knights when he spoke
of the few who for love of Christ’s law were determined to stand firm to
death in the face of the ungodly.12
If Gregory thus addressed the challenge of martyrdom extensively
through Latin Christendom, he also focused it intensively upon individu-
als. The familiar examples are the lay knight Erlembald and the priest
Liprand who were leaders of the savagely militant Patarene party at Milan
until its débâcle in 1075. Erlembald was Gregory’s archetypal strenuissi-
mus miles Christi.13 He died violently in the vicious fighting that he had
done much to provoke. The German chronicler Berthold of Reichenau
recorded the death and the miracles of Erlembald and also of the Roman
city prefect, Gregory’s supporter Cencius, who had likewise died violently.
At his Lent synod of 1078, Gregory virtually canonised them, and Berthold
held Erlembald up to all knights as an example to follow:
7
Register Gregors VII, 1.23, 2: 39–40.
8
9
10
11
12
Epistolae vagantes, no. 54, pp. 132–33.
13
Register Gregors VII, 1.27–28, 2: 44–46.
13
H. E. J. Cowdrey
Let every knight of the highest King and defender of the holy faith not
be slow to press ahead strongly and exultantly in Christ with Erlembald’s
industry and purpose, if, having accomplished his lawful strife, he
would rejoice to be crowned eternally with so honourable a triumph of
heavenly glory.14
As Erdmann pointed out, these canonisations were a turning-point.
Soldier-saints of the past, like St Martin of Tours, had been saints despite
their military status or else in reaction against it; Erlembald and Cencius
were adjudged to be martyrs because of their earthly arms and warfare.15
The Patarene priest Liprand is significant for a different reason. Like
Archbishop Cyriacus of Carthage, he suffered mutilation – in his case,
the cutting off of his nose and ears – but not death. He was nevertheless
a martyr who by his scars was indelibly stamped as such, and he would
and should persist to the end of his life in active sanctity:
Wherefore, martyr of Christ, be strong in the Lord […] For we know that
you are always thus assailed and afflicted by the enemies of holy church;
but you should not fear them or be dismayed, for with great charity we are
keeping both you and all that is yours under our own and under apostolic
protection.16
To be a martyr, it was not necessary actually to die; it was enough to show
and to persist in heroic witness: being prepared to die but, if providence
should so dispose, continuing to live a courageous life.
Gregory thus extended the dimensions of martyrdom in earthly strife
both extensively, by his energetically setting it forth to all grades of clerical
and lay society, and intensively, by holding up for praise and imitation
individual figures who had recently deserved the martyr’s palm.
The example of Liprand, with its moral that the quality of life no less
than the contingency of a violent death was a touchstone of martyrdom,
suggests a further reflection upon Gregory VII’s extending the dimen-
sions of warfare as a potentially saving and sanctifying activity. It was
not necessary (as the evidence about late eleventh-century warfare at
first sight seems to suggest) that a knight should be killed in order to
win entry into heaven; in truth, there could also be a sanctifying effect
upon the individual in battles well fought with purity of motive. Such a
14
Berthold, Annales, a. 1077, in MGH SS, 5: 304–6.
15
See the discussions in Erdmann, Origin of the Idea of Crusade, pp. 13–15, and
Flori, Pierre l’ermite, p. 205.
16
Quellen und Forschungen zum Urkunden- und Kanzleiwesen Papst Gregors VII,
1: Quellen: Urkunden. Regesten. Facsimilia, ed. Leo Santifaller (Vatikanstadt,
1957), no. 106, pp. 94–95.
14
thought was, indeed, well established by Gregory’s time. In the context

of the holy wars of Ottonian Germany, the Romano-Germanic Pontifical
of c. 960 contains a prayer for the army which includes a petition that
soldiers might have a right motive in fighting (proeliandi recta voluntas).
Then they might fight effectively, ascribing their victory not to their own
strength but to Christ the true victor who by the humility of his passion
triumphed on the cross over death and the prince of death.17
Gregory’s letters show traces of at least four interrelated ideas about
how the warfare of living warriors should be conducted and how, if so
conducted, it might be salutary. First, men should practise the skills of
arms well in military terms. They should fight viriliter – manfully;18 estote
ad pugnandum fortissimi – ‘be most strong to do battle’, Gregory urged
the warriors, mostly from north of the Alps, whom he summoned to his
eastern campaign of 1074.19 Warriors in a good cause should do a profes-
sional job. They should fight with confidence, for in the service of Peter
and with his patrocinium they could count upon the help and prosperity,
the victory and the ensuing peace, that Peter brought in the affairs of this
world as of the next.20 Second, men should fight morally. Gregory had
little or no conception of a just war in the Augustinian sense. But his
counsel to the Normans who were about to fight the Muslims of Sicily
was that they should abstain from capital crimes – that they should rise
superior to the lust for killing and for plunder that was the besetting sin
of fighting men.21 More positively, those who fought should, as beseemed
Christians, keep true faith in their behaviour towards those for whom
and also against whom they fought.22 Third, in connection with warfare
against the Sicilian Muslims, Gregory therefore alluded to a theme that
recurs in the complexity of his pre-crusading attitude to Islam. In the
spirit of Christ’s bidding his followers to let their light so shine before men
that other men should glorify their Father in heaven (Matt. 5:16), Count
Roger of Sicily with his army was to study so to fight that regard for the
name of Christian would be magnified among the heathen (quatinus […]
christiani nominis culturam inter paganos amplificare studeat).23 Fourth
17
Le Pontifical romano-germanique du dixième siècle, ed. Cyrille Vogel and
Reinhard Elze, 3 vols (Cité du Vatican, 1963–72), 2: 380.
18
e.g. Register Gregors VII, 1.28, 2: 46; cf. 7.27, 2: 508.
19
20
e.g. Register Gregors VII, 1.7, 2: 12; 3.15, 2: 277; 6.1, 2: 390; 6.14, 2: 419.
21
22
23
Register Gregors VII, 3.11, 2: 272. For Gregory and Islam, see further H. E. J.
Cowdrey, Pope Gregory VII, 1073–1085 (Oxford, 1998), pp. 489–94.
15
H. E. J. Cowdrey
and most important, in a way that blends with and enhances Gregory’s
perception of martyrdom as a way of life as well as of death, warfare rightly
undertaken was a supreme act of love for God and for one’s fellow men. It
was the leading theme of his attempted recruiting of 1074 for his eastern
expedition. The prayer with which his latest appeal ended was:
May Almighty God, who has summed up his whole law in the precept
of charity, grant that you may so love him with your whole heart, your
whole mind, and your whole strength that, by loving your neighbours as
yourselves, you may attain, if it shall prove necessary, to lay down your
lives for them.24
It should be remembered that, in accordance with the custom of the
times, Gregory expected that warriors embarking upon a campaign or
battle would make full confession of their sins and receive absolution.25
His letters go far to show and to publicise how they might form a proeli-
andi recta voluntas and how, in life as well as by death, warfare might be
a means of grace. Increasingly in the world of Gregory VII, war was seen
as a possible path to salvation. If warriors might not yet win remission of
sins by fighting – that came with the crusade – they nevertheless might
do so through it.
A further element in the papal handling of warfare as it concerned the
individuals who took part at first sight may seem to have contradicted
this development, though in fact it probably served to accentuate it. A
long tradition required those who committed homicides or woundings
in battle, even when their warfare served an acceptable purpose and was
directed by a legitimate ruler, to do penance. In canon law, the relevant
passages of the Decretum of Bishop Burchard of Worms may serve as an
example;26 in practice, the Normans who fought in 1066 at the battle of
Hastings faced severe penances imposed under the supervision of papal
legates.27 Especially in the decrees of his Roman synods between 1078 and
1080, Gregory took account of the virtually inescapable sinfulness of the
avocation of knights as of those of merchants and stewards or other such
secular administrators. His most negative assessment was that expressed
in the November synod, 1078. He referred to the hatred that knights
harboured in their heart and to the plunder that they desired: one recalls
the ‘capital crimes’ against which he had cautioned the South Italian
24
See above p. 12; for the citation, see Register Gregors VII, p. 173.
25
Register Gregors VII, 3.11, 2: 272; 8.6, 2: 514.
26
Burchard of Worms, Decretum, in PL 140, nos 6.23, 19.5, cols 770–71, 952.
27
Councils and Synods with Other Documents Relating to the English Church, ed.
Dorothy Whitelock et al, 2 vols in 4 (Oxford, 1981), no. 88, 1/2, 581–84.
16
Normans.28 Gregory returned to the subject in March 1080, when he not

only spoke of the necessary sinfulness of the three avocations but also
sought to redress the anomalies of current penitential practice by insist-
ing on the confession of all sins and, more important, by requiring inner
contrition and conversion of life.29 This forms an important complement
to the four points that he had already made about warfare in the service
of St Peter. It began to suggest how knights could fight in such service,
under the direction of religious men, with an altogether clear conscience
and purity of intention.
Gregory’s initial synodal decree of 1078, rather than his more radical
decree of 1080, was prominent among the surprisingly few of his rulings
that were taken up in the twelfth-century canonical tradition.30 Against the
run of his thinking, it gave his name to a perpetuation of the established
view that military activity necessarily involved sin that must be purged by
specific penance. Yet by so doing it also heightened the contrast that he
had drawn between sinful and unprofitable secular warfare and warfare
in the service of St Peter through which, if his intention were pure and
single-minded, a knight might find salvation and the remission of sins.
Here was a source of the opposition between the malitia of secular warfare
and the militia of its Christian counterpart which was already formulated
by Ivo of Chartres and which found classic expression in St Bernard’s
‘praise of the new warfare’ of the Knights Templar, that quintessential
crusading institution.31 According to Fulcher of Chartres’s account, at
Clermont Pope Urban II called upon knights to abandon the self-seeking
plunder of domestic violence and as soldiers of Christ to aid their eastern
brethren, thus deserving an eternal reward.32 The continuing influence of
Gregory’s decree of 1078 tended by reaction to add urgency to such a call.
So far, the subject of this paper has been the opening up in the decades
that preceded the First Crusade of warfare as a path to the salvation of the
28
Register Gregors VII, 6.5b, 2: 404, cf. 3.11, 2: 272; also 7.10, 2: 472; 8.21, 2: 559.
29
Register Gregors VII, 7.24a, 2: 481–82; cf. 8.21, 2: 559.
30
For documentation and further discussion, see Cowdrey, ‘Pope Gregory VII
and the Bearing of Arms’, 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. 21–35; repr. in Cowdrey, The Crusades
and Latin Monasticism, 11th-12th Centuries (Aldershot, 1999), Essay III.
31
Yves de Chartres, Correspondance, 1: 1090–1098, ed. Jean Leclercq (Paris, 1949),
no. 44 (1095), pp. 178–79; ‘De laude novae militiae’, caps. 1–4, in Sancti
Bernardi Opera, ed. Jean Leclercq et al., 8 vols (Roma, 1957–77), 3: 214–21.
32
Fulcher of Chartres, Historia Hierosolymitana, I.3.2–8, ed. Heinrich
Hagenmeyer (Heidelberg, 1913), pp. 132–38.
17
H. E. J. Cowdrey
individual. No less momentous in the adding of new dimensions to reform

was the extension and consolidation of the causes in pursuit of which such
salvation might be found. Gregory VII’s pontificate represents only a stage
in this process, for which Urban II’s was perhaps more critical. It is worth
comparing them. Among the causes that Gregory proposed to Christian
fighting men, three stand out. First, there was his call of 1074 to resist an
external, pagan enemy in the Muslim Saljūq Turks who, as Gregory put it,
were slaughtering Eastern Christians like cattle and ravaging their lands;33
but, in Gregory’s lifetime, this call came to nothing. Second, and of far
greater duration and importance, were his sponsorings of militant action
against simoniacs and schismatics – the enemies of Christendom from
within. In his early years as pope, the Patarenes of Milan are an example
of his allies against simoniacs.34 King Henry IV of Germany’s naming of
Archbishop Guibert of Ravenna at Brixen in 1080 to be anti-pope ranged
Gregory against schismatics for the rest of his pontificate. One may espe-
cially notice his immediate reaction in 1080 of seeking to call forth and
himself to lead a military expedition against Ravenna.35
Third, throughout his pontificate Gregory was increasingly concerned
with the problems of peace and social harmony within the kingdoms
and regions of Latin Christendom, especially France and Germany, and
with the means by which they might be fostered, upheld, and enforced.36
Like all, great and small, clerical and lay, who since the crumbling of
the Carolingian order had sought to be instrumental in the creation of
peace, Gregory was appalled by the slaughter and the arson, the rapine
and the perjuries, the sacrilege against churches and the violence against
the unprotected – monks, clergy, pilgrims, merchants, women, minors –
that were caused by civil wars and by the oppressions of the powerful.
His concern was the greater because in his view the breakdown of peace
in one region, especially when occasioned by the shortcomings of kings,
threatened to spread and to engulf all Christendom.37 His preferred
defence against social anarchy was, indeed, strong monarchy when acting
both effectively (utiliter) and obediently to papal admonitions. But when
33
See above p. 12.
34
See Cowdrey, Pope Gregory VII, pp. 68–70, 280–86.
35
See esp. Register Gregors VII, 8.5, 8.7, 8.12–14, 2: 521–25, 531–35; discussion by
Cowdrey, Pope Gregory VII, pp. 204–5, 310–11.
36
For fuller discussion, see Cowdrey, ‘From the Peace of God to the First Crusade’,
in La Primera cruzada novecientos años después: El concilio de Clermont y los
orígines del movimiento cruzado, ed. Luis García-Guijarro Ramos (Castellón,
1997), pp. 51–61; repr. in Cowdrey, The Crusades and Latin Monasticism, Essay II.
37
Register Gregors VII, 2.5, 2: 131; 4.24, 2: 337; 5.7, 2: 358.
18
such kingship failed, the duty to take counsel and to act devolved upon
the pope. And he must engage the support of bishops, princes, and above
all faithful laymen at large in a service of St Peter and of the apostolic see
that would promote order by all means (modis omnibus), including the
rightful use of armed force.
In 1074, during a short-lived but bitter contretemps with King Philip I
of France whom he deemed to be ruling inutiliter so far as peace and
order were concerned, Gregory announced the duty which rested upon
clergy and laity alike of opposing him if he were impenitent. Even if Philip
disposed of the power that pagan emperors had brought to bear upon
Christian martyrs of old, Gregory could leave his many and great iniqui-
ties in no wise unpunished, and he called upon the feudatories of France
to back him up.38 In Germany, his involvement was to become far more
protracted and fundamental.39 Between 1077 and 1080, his concern was to
be an arbiter between the rivals for the kingship, Henry IV and Rudolf of
Swabia. In the devastating civil conflicts that were destroying the peace of
the kingdom and empire, good Christians should rally to whichever king
righteousness showed to be the true ruler, so that he might duly obtain
the royal dignity and help a church that was almost perishing.40 In 1080,
it should be noted, the charge that Henry had proved to be a destroyer
of German peace figured among the reasons why Gregory again declared
him excommunicate and deposed: he had handed over to death a great
multitude of Christians; he had caused churches to be laid waste; and he
had devoted to desolation practically the whole of the German kingdom.41
With this condemnation, followed later in the year by the death in battle
of Rudolf of Swabia, and with the opening up of the Guibertine schism,
Gregory became no longer the arbiter in the struggle with Henry IV but
the rival protagonist. Significant highlights were Gregory’s call in 1083
for a general synod of faithful clergy and laity to assemble in some safe
place to deal radically with the issues of peace and of the disorders by
which Christendom was beset.42 Abortive though it was, the proposal
was a foreshadowing of Urban II’s council of Clermont in 1095. And
Gregory’s last encyclical from Salerno in 1084 was a resounding call to all
Christians, not least the military classes, to help and succour their father
38
Register Gregors VII, 1.74–75, 2.5, 2.18, 2.32, 2: 105–7, 129–33, 150–51, 168–69.
39
For an excellent discussion of the topics of this paragraph, see Monika Suchan,
Königsherrschaft im Streit. Konfliktaustragung in der Regierungszeit Heinrichs
IV. zwischen Gewalt, Gespräch und Schriftlichkeit (Stuttgart, 1997), pp. 121–58.
40
e.g. Register Gregors VII, 4.23–24, 2: 334–38.
41
Register Gregors VII, 7.14a, 2: 486.
42
Register Gregors VII, 9.29, 2: 612–13; cf. 9.29, 2: 613.
19
H. E. J. Cowdrey
(St Peter) and mother (the Roman Church) if through them they would
have the absolution of all their sins, and blessing and grace in this world
and in the world to come.43 To such horizons Gregory’s militant quest
for peace and order could expand.
And yet, despite some overlapping, the causes that Gregory proposed
to fighting men – his eastern campaign, action against simoniacs and
schismatics, and the establishment or defence of internal peace within
Christendom – remained largely discrete and dissociated so far as the
historical plane is concerned. This was despite the fact that cosmically,
all the enemies of Christendom, without and within, were servants and
members of the devil and the Antichrist whom Gregory called upon
the faithful to resist by all means as servants and members of Christ.44
Urban II, by contrast, during the early years of his pontificate developed
a much more comprehensive overview of past Christian history and of
current Christian objectives. He thereby set war as a path to salvation in a
fuller, more unified, and deeper setting, which issued in the call to crusade.
Urban developed an overview of the whole of past Christian history
which has been well explained by Alfons Becker in what he calls the schema
that Urban propounded.45 There were four chronologically consecutive
stages. (1) Initially, in the earliest days of the Church, there was a flower-
ing of the Christian life in communities that embodied true Christian
libertas and were rich in holy martyrs and confessors. (2) With the lapse
of time, and especially in the seventh and eighth centuries, owing to the
sins of Christians (peccatis exigentibus) many of these communities became
subject to Saracen supremacy and tyranny by which the libertas of the
Christian religion was reduced to nothing. (3) Now, however, nostris
temporibus, the God who, according to the scriptures, changes times and
seasons, removing kingdoms and setting them up (Dan. 2:20–1, Ecclus.
43
Epistolae vagantes, no. 54, pp. 128–35.
44
e.g. Saljūq Turks: Register Gregors VII, 2.37, 2: 173; simoniacs: Register Gregors
VII, 1.11, 2: 18; 1.15, 2: 23–24; 4.1, 2: 289–92; schismatics: Register Gregors VII,
8.5, 2: 522–23; Epistolae vagantes, no. 54, pp. 132–35; peace-breakers: Register
Gregors VII, 9.21, 2: 601–3, Epistolae vagantes, no. 25, pp. 66–67.
45
Alfons Becker, Papst Urban II. (1088–1099), 3 vols (Stuttgart, 1964–2012),
2: 352–62, 374–76, 398–403. For examples of documents which reflect this
schema, see Urban II, Epistola 5 (1088), in PL 151, cols 288–89; Epistola 50
(1091), in PL 151, cols 329–30; Epistola 52 (1091), in PL 151, cols 331–33; Epistola
59 (1092), in PL 152, cols 339–41; Epistola 93 (1093), in PL 151, cols 370–72;
Epistola 135 (1095), in PL 151, cols 407–8. Gregory VII had formulated such a
schema in the separate cases of Spain and the see of Palermo: Register Gregors
VII, 63.4, 1: 91–94; Quellen und Forschungen, no. 212, p. 253.
20
10:8), was at last showing mercy upon his people. Under the leadership of
godly kings and princes, and by the toil of the entire people, the Saracens
were being expelled. (4) Thus, a Christian restoration was imminent
through the outpouring of God’s grace and mercy. In it, the pope was
God’s fellow-worker (cooperator). But God’s purpose in a critical juncture
of history would not be achieved without the collaboration of the military
classes in the design of his providence. Thus, for example, in 1091 Urban
assured Count Ermengol IV of Urgel in anticipation of a campaign to
recover Tarragona that it would be a sacrifice acceptable to Almighty God
and a means to secure the remission of his sins, if by his aid Tarragona
was secured for Christendom.46
Progressively as Urban’s pontificate developed, such an overview of
previous Christian history was complemented by a unified understanding
of the renewal of Christendom in the present time. Especially throughout
the Mediterranean region – Spain, Sicily, Byzantium, Jerusalem – areas
of Christian recovery were considered together.47 As early as 1089, Urban
was telling the lay and ecclesiastical nobiles et potentes of Catalonia that
the spiritual benefits attaching to a Jerusalem pilgrimage were available
to those who would help to restore the church of Tarragona as a wall
and bulwark of Christendom against the Saracens.48 After the council
of Clermont, Urban renewed the call. If knights from other lands had
flooded to aid the churches of Asia and to free its peoples from Saracen
tyranny, so the Catalans should strive in regions adjacent to them. Those
who fell on a campaign undertaken there for love of God and their neigh-
bour might also count upon forgiveness of their sins and God’s merciful
reward of eternal life.49 Again in 1098, with reference to the restoration
of the church of Huesca, Urban rejoiced that in his own day, through
the strength of Christians, God had subdued the Moors in Europe as the
Turks in Asia, and he had restored cities once famous to the worship of
46
Paul Kehr, ‘Papsturkunden in Spanien, 1: Katalanien’, Abhandlungen der
Gesellschaft der Wissenschaften zu Göttingen, philologisch-historische Klasse, n.s.
18/2 (Berlin, 1926), pp. 286–87, n. 22; cited in Becker, Papst Urban II, 2: 398, n.
386.
47
See esp. Erdmann, Entstehung des Kreuzzugsgedankens, pp. 292–96; Erdmann,
Origin, pp. 314–19; Becker, Papst Urban II, 1: 227–30; 2: 299–307; Michael
Matzke, Daibert von Pisa. Zwischen Pisa, Papst und erstem Kreuzzug
(Sigmaringen, 1998), pp. 79–84, 139–40.
48
Epistola 20, in PL 151, cols 302–3.
49
Kehr, ‘Papsturkunden’, no. 23, pp. 287–88. Kehr’s date-range of 1089/91 is too
early in view of the manifest reference to the First Crusade.
21
H. E. J. Cowdrey
his religion.50 Besides his plans for Tarragona, Urban’s raising of the see
of Pisa to metropolitan status (1092) testifies to his purpose of building
up ecclesiastical organisation in the western Mediterranean.51 Renewed
papal interest under Urban in the Constitutum Constantini (‘Donation of
Constantine’) may well arise from its proclamation of a papal principatus
over the four patriarchal sees of Antioch, Alexandria, Constantinople,
and Jerusalem as well as over all churches in the world and from the vast
if imprecise endowments that Constantine bestowed upon Pope Silvester
and his successors in the four quarters of the world;52 that is to say, from
the comprehensiveness of its ecclesiastical and topographical perspectives.
Into this overarching view of Christian restoration and renewal, Urban
seems, not least by way of the diplomacy of which he was a master, to
have built the three causes that Gregory VII had proposed to knights,
but that in his day remained largely discrete. The taking up of Gregory’s
eastern plans of 1074 goes without saying. Indeed, the Liber pontificalis
commented upon how, in his crusade preaching, Urban had heard how
Gregory had called upon the ultramontani to go to Jerusalem for the
defence of the Christian faith and to free the Lord’s sepulchre from its
enemies’ hands.53 As regards simoniacs and schismatics, in the west Urban
was concerned to win offenders away from their errors and false loyalties
to the worthier and more satisfying cause of current papal objectives. His
winning over and promotion of Daibert of Pisa, at first made deacon by
the Henrician Archbishop Wezilo of Mainz, and the consolidating of
the service of the Pisans, may serve as an example.54 In the east, a prime
objective of Urban in and through the crusade was to bring to fruition
Gregory VII’s hope of claiming the churches of the east, and especially
that of Constantinople, for a proper filial obedience to the holy Roman
Church.55 As regards the domestic peace of Latin Christendom, Urban’s
call to knights to abandon their warfare and plunder at home in order
50
Epistola 237, in PL 151, col. 504.
51
See esp. Matzke, Daibert von Pisa, pp. 75–85.
52
‘Constitutum Constantini’, ed. Horst Fuhrmann, Fontes iuris Germanici
antiqui 10 (Hannover, 1968), lines 156–76, 203–8, pp. 80–83, 85–86. For the
revival of interest in this text under Urban II, see Cowdrey, ‘Eleventh-Century
Reformers’ Views of Constantine’, Byzantinische Forschungen 24 (1997),
63–92, repr. in Cowdrey, Popes and Reform in the 11th Century (Aldershot,
2000), Essay I, esp. pp. 86–89.
53
Le Liber pontificalis. Texte, introduction et commentaire, ed. Louis Duchesne
and Cyrille Vogel, 3 vols (Paris, 1886–1957), 2: 293.
54
Matzke, Daibert von Pisa, esp. pp. 26–101.
55
See esp. Becker, Papst Urban II, 2: 1–205, 414–34.
22
to find an external foe in the Saljūq was far from a cynically calculated
attempt to divert the roughnecks of the West to violence outside their
own regions. Rather, it was intended as an implementation of the pope’s
duty to provide for the peace of both West and East by enlisting clergy
and laity, according to the respective vocations of their social orders, in a
service that had as its ruling motive a willingness in charity to give one’s
life as a martyr for the cause of God and of one’s fellow men.56
To such parameters the later eleventh century witnessed the expansion
of papal concepts of Christian renewal and expansion, and within them
of military service as a path to salvation.
In conclusion, it may be worth adding a brief word about how modern
historians understand this expansion. Is it to be epitomised as a shift from
just to meritorious violence? It may be that this is a question mal posée, at
least if it envisages as a starting-point a concept of the just war which was
consciously and substantially based upon the teachings of St Augustine of
Hippo.57 For the subject of this paper has been the development already,
in the later eleventh century, of an idea of meritorious violence, at least in
the sense of divinely approved warfare as a way through which individuals
and societies might find salvation. One may, indeed, detect diffused influ-
ences of classical and patristic discussions of war; fathers like Augustine
and Gregory the Great were sometimes cited – especially in Augustine’s
case – in texts falsely ascribed which distorted their true teaching, such as
the letter Gravi de pugna, with its message that, in battle, God vindicates
the cause of the righteous.58 But deliberate and informed reference to
Augustine’s authentic teaching about the just war was slow to develop. It
is hard to find any trace of it in the letters of Pope Gregory VII. His dis-
ciple, Bishop Anselm II of Lucca, began to explore Augustine more fully
apropos military action against the domestic heretics and schismatics of
the 1080s.59 But only with Gratian’s Decretum of c.1140 did the material
relating to the just war begin to be comprehensively assembled, tested
56
See Cowdrey, ‘From the Peace of God to the First Crusade’.
57
For the just war, see Frederick H. Russell, The Just War in the Middle Ages
(Cambridge, 1975), with a discussion of St Augustine’s teaching at pp. 16–26;
see also Flori, Pierre l’ermite, pp. 119–28.
58
‘Gravi de pugna conquereris: dubites nolo, utile tibi tuisque dabo consilium;
arripe manibus arma, oratio aures pulset Auctoris; quia quando pugnatur, Deus
apertis coelis prospectat, et partem quam inspicit iustam, ibi dat palmam’:
Pseudo-Augustine, Epistola 13, in PL 33, col. 1098.
59
For comment on Anselm of Lucca, see Cowdrey, ‘Pope Gregory VII and the
Bearing of Arms’, pp. 33–34.
23
H. E. J. Cowdrey
for authenticity, and systematically discussed.60 And Gratian did not in

any way deliberately address the problems raised by the crusade. That
remained for popes and canonists yet to come.
The view that there was a shift from just to meritorious violence should
be regarded as an over-simplification, if not a distortion. It is, perhaps,
truer to say that both concepts grew together in a long and inter-acting
relationship. But ideas of meritorious violence on the whole developed
earlier, with more sophisticated ideas of just war emerging in the twelfth
century as Christian thought gradually moved into the scholastic age. The
result of this interaction was a view of the just war which carried a legal and
political, if not a moral, obligation to wage it. Like the crusade, men could
find salvation not only through it but also by it. Hence the disturbing, if
seldom commented upon, final sentence of the fullest English-language
monograph on the just war in the Middle Ages: ‘It remains an open ques-
tion whether just-war theories have limited more wars than they have
encouraged.’61 For the ideas of crusade and of the just war came in the
twelfth century to complement and to interact with each other, leaving
a long and ambivalent legacy to Christian Europe and to a wider world.
60
See Gratian’s long discussion in his Decretum, Pars 2, Causa 23: Corpus juris
canonici, ed. Emil Friedberg, 2 vols (Leipzig, 1879–81), 2: 889–965. Gratian is
fully discussed by Russell, Just War, pp. 55–85.
61
Russell, Just War, p. 308.
24
Jérusalem terrestre, céleste et spirituelle
Trois facteurs de sacralisation de la première croisade
Jean Flori
CNRS et Centre d’études supérieures de civilisation médiévale
L’étude des motivations et des buts des premiers croisés a connu depuis
Carl Erdmann des rebondissements nombreux si bien connus de tous
qu’il est inutile d’y insister ici.1 Depuis les travaux de mon ancien maître
et ami Paul Rousset et de H. E. J. Cowdrey, l’accent a été mis, à juste titre,
sur Jérusalem comme but et comme incitation, et en particulier sur son
impact psychologique auprès des croisés potentiels. Cette démarche a
remis au premier plan leurs motivations spirituelles, rapprochées de celles
des pèlerins se rendant aux lieux saints, même si l’on s’accorde à admettre
que croisade et pèlerinage sont deux institutions distinctes.2
Pourtant, dans quelques travaux, John France et Sylvia Schein, par
exemple, ont contesté l’ampleur du caractère attractif de Jérusalem à la
1
Carl Erdmann, Die Entstehung des Kreuzzugsgedankens (Stuttgart, 1935) ;
Erdmann, The Origin of the Idea of Crusade, trad. angl. Marshall W. Baldwin
et Walter Goffart (Oxford, 1977) ; la thèse de C. Erdmann a été très (trop)
vigoureusement attaquée par plusieurs historiens contemporains. Voir par
exemple John Gilchrist, ‘The Erdmann Thesis and the Canon Law’, dans
Crusade and Settlement, éd. Peter W. Edbury (Cardiff, 1985), pp. 37–45, et
plus encore Jonathan Riley-Smith, ‘Erdmann and the Historiography of the
Crusades, 1935–1995’, dans La primera cruzada novecientos años después : El
concilio de Clermont y los orígenes del movimiento cruzado, éd. Luis García-
Guijarro Ramos (Castellón, 1997), pp. 17–29.
2
Paul Rousset, Les origines et les caractères de la première croisade (Neuchâtel,
1945) ; Rousset, ‘L’idée de croisade chez les chroniqueurs d’occident’, dans X°
congresso internazionale di scienze storische (Roma, 1955), pp. 546–65 ; H.E.
John Cowdrey, ‘Pope Urban II’s Preaching of the First Crusade’, History 55
(1970), 177–88 (repris dans Cowdrey, Popes, Monks and Crusaders (Aldershot,
1984), Essay XVI) ; Cowdrey, ‘Pope Urban II and the Idea of Crusade’, Studi
Medievali 36 (1995), 721–42.
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. –.
FHG DOI : 1.1/M.OUTREMER-EB.1.11
Jean Flori
fin du xie siècle et minimisé quelque peu son rôle dans les motivations
de départ des premiers croisés.3
L’accent mis sur la ville sainte conforte évidemment les thèses des
historiens qui, sans aller jusqu’à assimiler totalement la croisade à un
pèlerinage, privilégient pourtant, dans les motivations de départ des
croisés, les notions de pénitence et de rémission des péchés acquises par
l’accomplissement de la route (via), du ‘chemin’ (iter) conduisant aux lieux
saints de Jérusalem. N’étant pas de ceux-là, je n’en suis que plus à l’aise
pour souligner cependant l’omniprésence de Jérusalem dans la plupart
des documents relatifs à la première croisade, aussi bien dans les discours
mobilisateurs plus ou moins reconstruits que dans les chartes des croisés,
dans les lettres ou dans les chroniques qui mentionnent leur départ ou
leur passage. Certes, ce n’est pas la ville en elle-même qui les attire, mais
uniquement la résonance, spirituelle et émotionnelle, du mot Jérusalem
dans les mentalités contemporaines. Ce mot, avec toutes ses acceptions,
loin d’être marginal ou secondaire, me semble être au contraire au centre
des thèmes de prédication de la croisade, au coeur des aspirations de ceux
qui ont répondu à cet appel. Sans aucun doute, les croisés partaient bien
pour Jérusalem.
Est-ce pour autant une preuve de l’assimilation, dans leur esprit, de la
croisade à un pèlerinage (armé) aux lieux saints ? Les croisés répondant
à l’appel étaient-ils seulement, ou même surtout, motivés par le désir
d’accomplir un pèlerinage, fût-ce les armes à la main, en tant que pénitents,
pour y gagner l’absolution de leurs péchés confessés ? Je ne le pense pas, et
cette communication a précisément pour but de mettre en évidence trois
autres facteurs de sacralisation de la croisade, trois motivations d’ordre
spirituel, liés à Jérusalem, mais qui n’en sont pas moins totalement indé-
pendants de la notion de pèlerinage. On peut les rattacher aux trois buts
suivants : reconquérir la Jérusalem terrestre, gagner la Jérusalem céleste,
bâtir la Jérusalem spirituelle.
3
John France, ‘The Destruction of Jerusalem and the First Crusade’, Journal
of Ecclesiastical History 47 (1996), 1–17 ; France, ‘Les origines de la première
croisade. Un nouvel examen’, dans Autour de la première croisade, éd. Michel
Balard (Paris, 1996), pp. 43–56 ; France, ‘Le rôle de Jérusalem dans la piété du xie
siècle’, dans Le partage du monde : Echanges et colonisation dans la Méditerranée
médiévale, éd. Michel Balard et Alain Ducellier (Paris, 1998), pp. 151–61 ; Sylvia
Schein, ‘Jérusalem, objectif originel de la première croisade ?’, dans Autour de la
première croisade, pp. 119–26.
26
1. Reconquérir la Jérusalem terrestre

Ce premier aspect est lié au but à atteindre : Jérusalem et ses lieux saints.
Pour assimiler la première croisade à un pèlerinage, ou du moins pour
rapprocher leurs motivations, on a souvent insisté sur la similitude des
expressions employées à leur propos : via, iter, peregrini, etc, désignent
indifféremment l’une et l’autre. Toutefois, cette similitude, indéniable
au demeurant, ne me paraît pas très significative, car la destination même
de l’entreprise – à savoir la libération par les armes des lieux saints trans-
formait nécessairement l’expédition en pèlerinage dès lors qu’elle tendait
vers le même but, Jérusalem.4 Rien n’indique, en revanche, que cet aspect
de pèlerinage ait occupé la première place dans les motivations initiales
des croisés. De nombreux autres mobiles à caractère religieux ou spirituel,
voire matériels, pouvaient tout aussi bien contribuer à susciter leur départ,
même si les documents ecclésiastiques ne mentionnent généralement que
l’aspect pénitentiel mis en avant par l’Église pour l’expédition comme
pour les pèlerinages qui la précédèrent.
Pour tenter de percevoir directement les véritables sentiments des futurs
croisés, nous ne disposons guère que de leurs lettres et de leurs chartes de
départ. Mais l’exposé de leurs mobiles, exprimés principalement dans les
chartes rédigées à l’occasion des actes juridiques accomplis par eux avant
de se mettre en route, suscite quelques doutes ou pour le moins quelques
remarques. Les études de Marcus Bull et de Jonathan Riley-Smith ont
souligné la profonde influence exercée sur les familles des laïcs croisés par
les communautés monastiques des environs, leur commune religiosité et
les motivations spirituelles avancées par les chartes.5 L’ouvrage de Jonathan
4
On peut même dire que la première croisade ne devenait pèlerinage que si elle
atteignait son but, à savoir la libération des lieux saints. En toute rigueur, les
‘privilèges spirituels’ du pèlerin n’étaient imputés qu’au terme du pèlerinage,
liés et ainsi subordonnés à la réussite de l’expédition militaire. Un siècle
plus tard, Richard Coeur de Lion exprime cette différence de conception en
refusant, dit-on, de faire le pèlerinage à Jérusalem (dont il n’est séparé que par
quelques lieues) parce qu’il n’a pas réussi à atteindre le but de sa croisade, à
savoir la libération de la Ville Sainte. Voir sur ce point Jean Flori, Richard Coeur
de Lion, le roi-chevalier (Paris, 1999), p. 175.
5
Marcus Bull, ‘The Roots of Lay Enthusiasm for the First Crusade’, History
78 (1993), 353–72 ; Bull, Knightly Piety and the Lay Response to the First
Crusade : The Limousin and Gascony, c. 970–c. 1130 (Oxford, 1993) ; Bull,
‘The Diplomatic of the First Crusade’, dans The First Crusade : Origins and
Impact, éd. Jonathan Phillips (Manchester, 1997), pp. 35–54 ; Jonathan
Riley-Smith, The First Crusade and the Idea of Crusading (London, 1986) ;
27
Jean Flori
Riley-Smith (The First Crusaders, 1997), tout en marquant un salutaire

retour à l’idée de guerre sainte trop souvent minimisée dans ses travaux
antérieurs, souligne à nouveau l’importance des motivations religieuses
révélées par ces chartes, en particulier les notions de conscience du péché,
de désir de pardon et d’acceptation de pénitence. Ces éléments sont
incontestables, et je ne cherche en rien à les nier ni à les minimiser. Tout
le monde s’accorde aujourd’hui à voir dans les motivations religieuses des
croisés les principaux mobiles de leur départ.
Mais de quelles motivations religieuses s’agit-il ? Nous ne connaissons
guère la religiosité des laïcs qu’à travers ce que nous en disent les moines
ou les clercs. On peut légitimement se demander si l’influence des moines
sur les laïcs, en particulier sur les chevaliers, a nécessairement conduit à
une commune forme de spiritualité. Certes, les uns et les autres pratiquent
bien la même religion formelle, voire formaliste, dont ils observent les
rites essentiels, mais ils n’en ont pas nécessairement pour autant la même
perception intime. On peut se demander, dans cette perspective, si les
chartes des croisés n’expriment pas davantage la religiosité des moines
que celle des laïcs concernés dans ces actes.
La question se pose avec plus d’acuité encore à propos des mobiles
de départ des chevaliers. Sommes-nous bien sûrs que les rédacteurs des
chartes traduisent vraiment leurs motivations réelles ? Entendons-nous
bien : il n’est pas question, dans les lignes qui suivent, de mettre en doute
la sincérité des chevaliers, de sonder leur psychisme profond, ni même
de souligner l’inévitable distorsion qui existe toujours, nécessairement,
à toutes les époques, entre les mobiles avoués et les raisons profondes et
mystérieuses qui conduisent un homme à l’action. C’est là un champ de
recherche réservé au psychanalyste, ce que je ne suis en aucune manière. Il
s’agit seulement de rappeler que les mobiles des croisés, que nous trouvons
parfois exprimés par les rédacteurs des chartes, ont nécessairement subi,
en passant du croisé au moine rédacteur, une coloration, une formulation,
voire une altération de la part des ces rédacteurs. Qu’elle soit volontaire
ou non, cette transcription confère aux mobiles des croisés une tournure
ecclésiastique qui rapproche ces mobiles de ceux que l’Église estimait
jusqu’ici devoir être à l’origine de telles décisions de départ. Elle les rend
conforme à la norme admise, au ‘politically correct’ de l’époque. D’où,
Riley-Smith, ‘L’idée de croisade dans les chartes de la première croisade’, dans

Les croisades. L’Orient et l’Occident d’Urbain II à saint Louis, 1096–1270, éd.
Monique Rey-Delqué (Milan, 1997), pp. 130–33 ; Riley-Smith, ‘The Idea of
Crusading in the Charters of Early Crusaders, 1095–1102’, dans Le Concile de
Clermont de 1095 et la Croisade (Rome, 1997), pp. 155–66 ; Riley-Smith, The
First Crusaders, 1095–1131 (Cambridge, 1997).
28
nécessairement, un accent mis sur la conscience du péché, sur le besoin de

repentance, sur le pèlerinage aux lieux saints destiné précisément à servir de
pénitence. La coloration cléricale du langage des chartes ne me paraît guère
pouvoir être niée. Il ne faut donc pas en minimiser l’impact. La distorsion
n’est pas seulement quantitative, exagérant l’intensité des sentiments
religieux ; elle peut être aussi qualitative, transformant une motivation
religieuse peu orthodoxe en expression plus ‘religieusement correcte’. Les
quelques exemples qui suivent suffiront à illustrer ce processus.
Dans une charte datée d’octobre 1097, la comtesse de Flandre fait réfé-
rence au départ pour la croisade de son mari. Elle l’exprime en ces termes : ‘la
grâce du saint Esprit a illuminé le coeur de mon seigneur et époux Robert,
comte de Flandre, pour lever une grande armée et faire une expédition
militaire pour réprimer la perfidie des Perses.’6 La comtesse, personnage de
la haute aristocratie laïque, n’hésite pas à décrire la croisade pour ce qu’elle
est aux yeux de la plupart des chevaliers : une expédition militaire suscitée
par le Saint-Esprit. Elle ne dit rien d’un pèlerinage ni d’une pénitence ni
d’une rémission des péchés. La chose est plus rare dans les chartes ordinaires.
Pourtant, dans une charte de La Réole, Amanieu de Loubens dit qu’il a été
inspiré par l’amour du Saint-Esprit pour aller ‘combattre et tuer les ennemis
de la religion chrétienne, et surtout purifier le lieu ou Notre Seigneur Jésus
Christ a daigné souffir la mort pour la restauration de la race humaine.’7 Une
autre charte évoque le fait que quelques-uns des personnages cités sont ‘sur
le point de prendre la route de Jérusalem pour combattre les barbares’,8 etc.
Dans ces chartes, c’est l’action militaire elle-même, la guerre contre l’ennemi
de la foi, qui est mise en avant et se trouve en soi valorisée.
Certes, j’en suis le premier convaincu et en ai maintes fois tenté la
démonstration, la notion de ‘guerre sacralisée’ était déjà présente bien
avant la première croisade. Elle pouvait cependant, à cette date, choquer
encore les mentalités monastiques ; son expression vigoureuse et sans
nuance par les chevaliers a sans aucun doute pu être atténuée en de très
nombreux cas par les rédacteurs des chartes. Tous n’étaient sans doute pas
prêts à affirmer sans ambage, comme c’est le cas ici, que le Saint Esprit
poussait des hommes à aller tuer d’autres hommes, fussent-ils des ennemis
6
Charte de Clémence de Bourgogne, comtesse de Flandre, oct. 1097, Epistulae
et chartae ad historiam primi belli sacri spectantes, éd. Heinrich Hagenmeyer
(Innsbruck, 1901), no. VII, pp. 142–43.
7
Cartulaire du prieuré de Saint-Pierre de La Réole, éd. Charles Grellet-Balguerrie
(1863), no. 100, p. 140.
8
Cartulaire de l’abbaye de Saint-Chaffre du Monastier, éd. Ulysse Chevalier (Paris,
1884) : ‘Nunc autem quidam eorum, Jerosolimitanum iter ad expugnandos
barbaros arripientes […]’, Charte 398, (déc. 1096).
29
Jean Flori
de la foi. En d’autres termes, l’absence dans la plupart des chartes d’une

formulation aussi claire et brutale de la ‘guerre sainte’ ne signifierait nul-
lement qu’une telle notion n’existait pas chez les chevaliers. Sa présence,
même rare, en revanche, en est la preuve manifeste. Alors que les cheva-
liers mettaient l’accent sur leur désir de combattre pour Dieu dans une
guerre sacralisée ayant pour but la reconquête de l’héritage du Christ, les
moines ont très vraisemblablement préféré traduire ce désir en termes plus
conformes à la spiritualité admise, en termes de pèlerinage ou de ‘voyage’
accompli en rémission de leurs péchés, masquant ou atténuant ainsi l’as-
pect militaire même lorsqu’ils osent employer des expressions suggérant
pour le moins une action guerrière, comme expeditio ou exercitio.9 Ainsi
Achard de Montmerle se décrit comme ‘témoin de ce grand mouvement
ou expédition du peuple chrétien s’apprêtant à marcher sur Jérusalem
pour combattre en faveur de Dieu contre les païens et les Sarrasins’, et se
dit lui-même animé des mêmes intentions.10 Il évoque clairement ainsi
une opération guerrière de reconquête. On a d’ailleurs un exemple du
même genre antérieur à la première croisade, à propos d’une expédi-
tion militaire contre les Maures d’Outre Pyrénées, puisque Hugues de
Lusignan fait une donation avant de partir combattre contre les Sarrasins
en Espagne, pour le salut de son âme.11 Ces quelques cas suffisent, me
semble-t-il, à établir l’atténuation ‘doctrinale’ des motivations guerrières
des croisés par les moines rédacteurs.
D’autre part, même si ces rédacteurs n’ont pas eu volontairement
l’intention de remodeler l’expression des motivations des croisés, ils ont
probablement été amenés à le faire inconsciemment par simple confor-
mité aux modèles lexicaux précédents que leur fournissait l’histoire
passée. La croisade était, ne l’oublions pas, un phénomène nouveau qui
n’avait encore aucun vocabulaire propre. Pour le décrire, il fallait user du
langage utilisé antérieurement pour d’autres phénomènes auxquels ont
pouvait la comparer, même si on savait l’en distinguer mentalement. Ne
sommes-nous pas alors en présence, dans la plupart des cas, d’un langage
9
Sur l’emploi de ce terme expeditio à propos de la première croisade, voir par
exemple Cartulaire de l’abbaye de Saint-Chaffre du Monastier, charte no. 245
(25 juillet 1096) ; Cartulaire de l’abbaye du Ronceray d’Angers (1028–1184), éd.
Paul Marchegay (Paris, 1900), charte no. 294, p. 186. L’entreprise est parfois
plus clairement encore nommée exercitio : voir par exemple Recueil des chartes
des l’abbaye de Cluny, éd. Auguste Bernard et Alexandre Bruel, 6 vols (Paris,
1876–1903), charte no. 3712, 5 : 59.
10
Recueil des chartes de Cluny, charte no. 3703, 5 : 51 (12 avril 1096).
11
Charte de Hugues de Lusignan, no. 157, Chartes de l’abbaye de Nouaillé, de 678
à 1200, éd. Dom P. de Monsabert (Poitiers, 1936), p. 248.
30
monastique stéréotypé, calqué sur celui que l’on employait auparavant

à propos des pèlerinages antérieurs ? L’objectif même des croisés (c’est-
à-dire Jérusalem et plus précisément la délivrance du saint Sépulcre)
conduisait tout naturellement les scribes à rédiger ces chartes de départ
en des termes identiques à celles des pèlerins, jusqu’ici seuls concernés.
Comme ces pèlerins, et davantage encore, les croisés venaient chercher
auprès des établissements ecclésiastiques les subsides nécessaires à leur
voyage pour la Terre sainte ; les chartes furent rédigées précisément pour
enregistrer ces transactions, ventes ou mises en gage déguisées. Quoi de
plus naturel, dès lors, pour les rédacteurs, que d’attribuer une même cause
à des phénomènes produisant apparemment les mêmes effets ?
Le langage des moines rédacteurs des chartes déforme d’ailleurs triple-
ment la réalité : une première fois en assimilant ces croisés aux pèlerins et
pénitents antérieurs ; une seconde fois en assimilant leurs ventes, renon-
ciations et mises en gage à des donations pieuses ; une troisième fois en
attribuant très souvent à ces concessions dûment enregistrées (et non à
l’expédition en elle-même) le mérite d’avoir été faites ‘pour la rémission
de leurs péchés’ ou ‘pour le salut de leur âme’.
L’accent bien réel mis sur les péchés antérieurs des donateurs et sur le
désir de pénitence manifesté par ces donations ou renonciations me semble
donc devoir être surtout mis au compte de la formulation monastique
traditionnelle des actes bien plus qu’à un réel désir de repentance des
chevaliers partant pour la croisade. Le besoin d’argent ou de soutien de
toute nature (y compris, d’ailleurs, celui des prières des moines) condui-
sait probablement les croisés à accepter de se voir ainsi désignés comme
repentants et pénitents, en des termes parfois très humiliants, comme
on l’a récemment observé.12 Par ailleurs, soulignons-le à nouveau, les
chartes confèrent valeur méritoire et rédemptrice à la donation, qui est à
l’origine de l’acte écrit, tout autant qu’à l’expédition qui en est pourtant
la raison d’être par les besoins financiers qu’elle nécessite, mais qui peut
avoir été suscitée par des mobiles très divers. En d’autres termes, l’aspect
pénitentiel mis en avant par les chartes pour toutes les raisons déjà évo-
quées a certes pu être à l’origine du départ de nombreux chevaliers ; il
serait en revanche très périlleux de généraliser et d’en faire l’unique ou la
principale motivation des chevaliers. Ils étaient certes décrits en pèlerins,
pénitents, allant à Jérusalem pour la rémission de leurs péchés. Eux-mêmes
se percevaient probablement en combattants de la foi, en milites Christi,
mettant leur vaillance et leur épée au service de leur Seigneur, par amour,
certes, mais attendant de lui des récompenses spirituelles pour prix de leur
12
Riley-Smith, First Crusaders, en particulier p. 106.
31
Jean Flori
service, comme les chevaliers du siècle servant par les armes leur seigneur
temporel.13
Malgré la brillante démonstration de Riley-Smith concernant le finan-
cement d’un croisé, on ne peut pas non plus exclure l’existence de motiva-
tions matérielles, même si elles sont évidemment masquées et demeurent
probablement mineures : certes, le coût du départ pour Jérusalem était
pour les familles une charge très lourde ; certes, la plupart des croisés
n’avaient pas l’intention de rester en terre sainte et presque tous sont reve-
nus bien plus pauvres qu’ils n’étaient partis. Mais ces faits indubitables ne
peuvent être pris en compte qu’a posteriori et il n’est pas sage de mesurer
les motivations de départ des croisés à l’aune des résultats obtenus après
coup. En d’autre termes, et de façon plus triviale : nous savons, nous, que
l’expédition ne serait pas ‘rentable’ ; les premiers croisés, eux, ne le savaient
pas. Ils pouvaient se bercer d’illusions. De plus, sur un plan strictement
méthodologique, rejeter la possibilité d’existence de motivations maté-
rielles pour ces raisons-là équivaudrait à nier le succès remporté, de nos
jours, par les jeux de hasard, sous prétexte que les chances de gagner sont
insignifiantes. C’est aussi faire peu de cas de l’espérance du pillage, du butin
pris sur l’ennemi, si fréquemment évoqué dans les récits de la croisade et
qui suscita entre les croisés tant de jalousie, de discorde et de rancoeur.
Au demeurant, la mention expresse du concile de Clermont restreignant
l’octroi des ‘privilèges spirituels’ (ici commutation de pénitence) à ceux-qui
partiraient ‘sans désir d’acquérir gloire ou richesse’ prouve à l’évidence que
ces motivations étaient bien présentes chez ceux-là mêmes qui prenaient
la route en rémission de leurs péchés. À plus forte raison chez ceux qui
y étaient poussés par d’autres mobiles, et qui pouvaient être nombreux.
En effet, pour indéniable qu’il soit, cet aspect de pèlerinage pénitentiel
ne me semble ni premier, ni prépondérant. Dans la prédication d’Urbain
II comme dans récits de la croisade ou même, on l’a vu, les chartes de
départ des croisés, c’est bien plus l’expédition guerrière de libération des
lieux saints, de l’héritage du Christ, qui se trouve valorisée. L’étude de
la notion de reconquête chrétienne dans la correspondance pontificale
13
Les lettres de Grégoire VII et le discours d’Urbain II à Clermont n’hésitent
pas à faire cette comparaison entre les deux services assortis de récompenses,
et soulignent sans vergogne la supériorité du ‘salaire’ spirituel sur les gages
matériels. Voir sur ce point Jean Flori, ‘Réforme-reconquista-croisade. L’idée
de reconquête dans la correspondance pontificale d’Alexandre II à Urbain II’,
Cahiers de civilisation médiévale 40 (1997), pp. 317–35 ; Flori, ‘Le vocabulaire
de la reconquête chrétienne dans les lettres de Grégoire VII’, dans De Toledo a
Huesca. Sociedades medievales en transicion a finales del siglo XI (1080–1100), éd.
Carlos Laliena Corbera et Juan F. Utrilla Utrilla (Zaragoza, 1998), pp. 247–67.
32
de Grégoire VII à Urbain II montre bien qu’une telle notion n’était pas
neuve et répondait à une forte préoccupation de la papauté. Urbain II
a su, avec le succès que l’on sait, donner à cet appel à la reconquête une
forme répondant également aux aspirations et aux mentalités des che-
valiers en associant le pèlerinage à la guerre dite ‘sainte’ que je préfère
pour ma part nommer ‘guerre sacralisée’. J’ai assez insisté, dans plusieurs
travaux antérieurs, sur ces aspects de ‘guerre sainte’ pour ne pas y reve-
nir ici.14 Dès 1095 l’idée d’une guerre sacralisée méritoire au point de
constituer un moyen de salut était, rappelons-le, admise par la doctrine
ecclésiastique et adoptée par les chevaliers, non pas tant par une évo-
lution de la guerre juste vers la guerre sainte, comme on le dit souvent,
que par une affirmation croissante de la sacralité du combat armé pour
Dieu, pour l’Église, pour Saint-Pierre.15 Les promesses de récompenses
spirituelles accordées par Grégoire VII et Urbain II à ceux qui combat-
traient pour l’Église et principalement pour le Saint-Siège expriment
clairement cette dimension de combat sacralisé méritoire et l’on voit
mal comment la guerre menée contre les ‘infidèles’ pour le Christ et la
délivrance de son tombeau et de son héritage pourrait avoir été moins
valorisante aux yeux des chevaliers que les appels à combattre pour les
intérêts de la papauté en Italie, en Espagne et en Sicile. Par ailleurs, il
était à coup sûr plus ‘attractif ’, pour les chevaliers, de risquer la mort
pour obtenir le paradis que pour accomplir une pénitence.16
14
Outre les travaux mentionnés à la note précédente, voir par exemple Jean Flori,
‘Guerre sainte et rétributions spirituelles dans la seconde moitié du xie siècle :
lutte contre l’islam ou pour la papauté ?’, Revue d’Histoire Ecclésiastique 85 (1990),
617–49 ; Flori, ‘L’Eglise et la guerre sainte, de la paix de Dieu à la croisade’,
Annales : Economies, Sociétés, Civilisations (1992), 88–99 ; Flori, ‘De la chevalerie
féodale à la chevalerie chrétienne ?’, dans Militia Christi et crociata nei secoli XI-XII
(Milano, 1992), pp. 67–101 ; Flori, ‘“Pur eshalcier sainte crestïenté”, croisade,
guerre sainte et guerre juste dans les anciennes chansons de geste françaises’, Le
Moyen Age 97 (1991), 171–87 ; Flori, ‘La préparation spirituelle de la croisade :
L’arrière-plan éthique de la notion de miles Christi’, dans Il Concilio di Piacenza
e le crociate (Piacenza, 1996), pp. 179–92 ; Flori, ‘L’idée de croisade dans quelques
chansons de geste du cycle de Guillaume d’Orange’, Medioevo Romanzo 21
(1997), 476–95 ; Flori, Croisade et chevalerie, xie-xiie siècles (Paris, 1998).
15
Notion fort opportunément rappelée ici même par John Cowdrey, dont je
partage pleinement l’analyse.
16
Sur l’existence de la notion de martyre des guerriers combattant pour le Christ
à l’époque de la première croisade, voir Jonathan Riley-Smith, ‘Death on the
First Crusade’, dans The End of Strife, éd. David Loades (Edinburgh, 1984), pp.
14–31, qui la nie ; Cowdrey, ‘Martyrdom and the First Crusade’, dans Crusade
33
Jean Flori
Toutes les motivations jusqu’ici mentionnées ont pu exister, parfois

même mêlées et réunies chez un seul homme. Elles sont d’ailleurs le plus
souvent parfaitement compatibles (sauf, on l’a vu, les motivations maté-
rielles pour ceux des croisés qui partaient en pénitents). À plus forte raison
devait-on les retrouver présentes dans un ensemble aussi vaste que l’armée
des croisés. Pour les croisés, les motivations spirituelles (pénitence, guerre
sainte méritoire et perspective du martyre) ne s’excluaient aucunement.
On ne peut donc pas affirmer que les guerriers ont tous pris leur décision
de départ pour des raisons de pénitence, ni qu’ils se considéraient comme
des pénitents. Tous, en revanche, se considéraient comme des combattants
pour le Christ, des milites Christi. C’est bien ainsi, d’ailleurs, que le voulait
Urbain II qui déconseillait aux non combattants d’entreprendre cette
marche, soulignant par là-même le caractère essentiellement guerrier de
l’expédition méritoire.
En bref, Jérusalem et les lieux saints sacralisent à la fois le pèlerinage
d’une part, la croisade d’autre part. La sacralité de la croisade se trouve
ainsi, pour ainsi dire, doublée, car elle est à la fois pèlerinage par sa des-
tination et guerre sainte de reconquête par ses intentions. Elle est même
triple, comme nous allons le voir, indépendamment du pèlerinage qui
n’en est peut-être que la retombée mineure, valorisée après coup à cause
de l’échec relatif des autres objectifs visés.
2. Gagner la Jérusalem Céleste

Le second facteur de sacralisation tient à la dimension eschatologique de
Jérusalem. Les médiévistes ne croient plus depuis longtemps aux préten-
dues ‘terreurs de l’an mil’,17 mais la réalité d’une attente eschatologique ne
doit pas pour autant être négligée comme elle l’a trop été depuis plusieurs
and Settlement, pp. 47–56, qui en accepte éventuellement la possibilité, et

Jean Flori, ‘Mort et martyre des guerriers vers 1100 ; l’exemple de la première
croisade’, Cahiers de civilisation médiévale 34 (1991), pp. 121–39, qui en affirme
l’existence. Je ne suis pas convaincu par les arguments réaffirmés par Riley-
Smith, First Crusaders, pp. 72–74. Voir à ce propos Flori, Pierre l’ermite et la
première croisade (Paris, 1999), pp. 135–38 et 216–21.
17
L’année 1999 a vu réapparaître, au sein même de notre société pourtant si
laïcisée, un certain climat d’attente de la fin du monde, voire de ‘peur de l’an
2000’, reposant sur des fondements infiniment plus fragiles (pour ne pas dire
plus stupides !) qu’il y a mille ans, et assorti d’un ensemble encore plus navrant
de superstitions diverses et de naïve crédulité. Ceci devrait inciter les historiens
à se montrer moins catégoriques dans leur rejet d’une attente eschatologique au
34
années sous l’influence de Bernard McGinn.18 Sans adopter totalement ici

les conclusions parfois excessives de Richard Landes,19 force nous est de
constater que cette attente, mêlant inquiétude et espérance, était vivace
au xie siècle.20 Elle est alors associée à l’idée de vengeance des chrétiens et
du Christ sur leurs ennemis, les infidèles, Juifs, hérétiques et musulmans
confondus. Les multiples variantes apportées aux traités sur l’Antichrist
témoignent assez de l’intérêt qu’on leur portait.
Ces écrits annoncent, on le sait, l’avènement du dernier millénaire qui
serait marqué par la fin de l’empire perse et la suprématie temporaire des
‘fils d’Ismaél’, à savoir les Arabes, qui devaient étendre leur domination
sur les terres jusqu’ici chrétiennes de Palestine, de Syrie, et d’Égypte. Ils
domineraient et souilleraient les lieux saints jusqu’à ce qu’un ‘roi des
Grecs et des Romains’ se lève et les asservisse, vengeant ainsi les chrétiens
avec l’aide de la milice divine. Ce roi devrait règner à Jérusalem pendant
un an et demi : alors paraîtrait l’Antichrist, qui serait à son tour vaincu
Moyen Age. Mieux vaut d’ailleurs parler, pour cette époque, d’une attente des
‘Temps de la fin’ mêlant angoisse et espérance.
18
Bernard McGinn, Visions of the End : Apocalyptic Traditions in the Middle
Ages (New York, 1979) ; McGinn, ‘Apocalypticism in the Middle Ages : An
Historical Sketch’, dans Apocalypticism in the Western Tradition, éd. McGinn
(Aldershot, 1994), pp. 252–86 ; McGinn, ‘Portraying Antichrist in the Middle
Ages’, dans The Use and Abuse of Eschatology in the Middle Ages, éd. Werner
Verbeke, Daniel Verhelst et Andries Welkenhuysen (Leuven, 1988), pp. 1–47.
19
Voir en particulier, parmi ses très convaincantes études, Richard Landes, ‘Lest
the Millennium be fulfilled : Apocalyptic Expectations and the Pattern of
Western Chronography, 100–800 ce’, dans The Use and Abuse of Eschatology,
pp. 137–211 ; Richard Landes, Relics, Apocalypse and the Deceits of History :
Adémar of Chabannes, 989–1034 (Cambridge, MA, 1995) ; Richard Landes,
‘Sur les traces du Millennium : la “via Negativa”’, Le Moyen Age 98 (1992), 356–
77, et 99 (1993), 5–26.
20
Paul Alphandéry et Alphonse Dupront, La chrétienté et l’idée de croisade (Paris,
1954), a bien mis en lumière ce climat d’attente, même s’il en a peut-être exagéré
la portée, ce qui n’est pas sûr, en tout cas pour la croisade dite ‘populaire’. Il en a
posé les fondements dans Alphandéry, ‘Notes sur le messianisme médiéval latin,
xie-xiiie s.’, dans Rapport annuel de l’Ecole des Hautes Etudes, section des Sciences
Religieuses (Paris, 1912) ; ce climat a été depuis trop minimisé dans le sillage de
Bernard McGinn, ‘Iter sancti sepulchri : the piety of the first crusaders’, dans
Essays on Medieval Civilization, éd. Bede K. Lackner et Kenneth R. Philp
(London, 1978), pp. 33–71. Voir sur ce débat Sylvia Schein, ‘Die Kreuzzüge als
volkstümlich-messianische Bewegungen’, Deutsches Archiv für Erforschung des
Mittelalters 47 (1991), pp. 119–37, et dans Flori, Pierre l’ermite, pp. 175–77.
35
Jean Flori
par le Christ mettant fin à l’histoire de ce monde.21 La Sybille Tiburtine

précise que ce roi des Romains et des Grecs appellera tous ses sujets à se
faire baptiser et punira du glaive ceux qui n’adoreront pas la croix ; même
les Juifs, à la fin de son règne, se convertiront et le sépulcre du Seigneur
sera vénéré par tous. Alors viendra l’Antichrist, finalement vaincu par
l’archange saint Michel sur le Mont des Oliviers.22 Cette tradition d’ori-
gine orientale s’occidentalise chez Adson de Montier-en-Der, vers 968 ;
‘certains de nos docteurs’, affirme-t-il, disent que ce roi des Romains sera
un roi franc qui viendra à Jérusalem pour y déposer sa couronne et son
sceptre sur le Mont des Oliviers. Alors paraîtra l’Antichrist qui rassem-
blera derrière lui les siens, à savoir les Juifs, mais sera finalement vaincu
par le chef des milices céleste et par les fidèles du Christ.23
On répète trop souvent que ces spéculations intellectuelles n’intéres-
saient guère qu’une petite minorité de moines lettrés. Rien n’est plus
faux, comme en témoignent à la fois les divers remaniements subis par
ces textes jusqu’à la fin du xie siècle et les très ombreux manuscrits qui
nous les transmettent.24
L’époque qui précède la croisade était-elle favorable à l’éclosion d’un
mouvement de masse reposant sur de telles espérances apocalyptiques
ou millénaristes ? C’est vraisemblable. Au milieu du xie siècle, le Pseudo-
Alcuin en témoigne. Son texte s’inspire de traditions diverses, dans un
milieu faisant de Charlemagne le héros du pèlerinage et le défenseur
des lieux saints. Il récupère du même coup la légende de l’empereur
des derniers temps et fait de celui-ci un roi franc qui ne peut être que
Charlemagne. Cette adaptation est intentionnelle et lie ainsi la fin des
21
On trouve ces éléments dans plusieurs textes anciens, glosant sur les écrits de
Méthode, composés en Syrie dès la fin du viie siècle et traduits en latin au viiie
siècle, annonçant ces temps de la fin et décrivant les modalités de l’apparition de
l’Antichrist, en relation avec la récente invasion arabe. Voir Pseudo-Méthode,
‘Sermo de regnum cantium et in novissimis temporibus certa demonstratio’,
texte dans Sibyllinische Texte und Forschungen, éd. Ernst Sackur (Halle, 1898),
pp. 60–96 ; éd. par Otto Prinz, ‘Eine frühe abendländische Aktualisierung
der lateinischen Übersetzung des Pseudo-Methodios’, Deutsches Archiv für
Erforschung des Mittelalters 41 (1985), 5–17 ; trad. française dans Claude
Carozzi et Huguette Carozzi-Taviani, La fin des temps (Paris, 1982), p. 20.
22
Texte dans Sackur, Sibyllinische Texte, pp. 177–87.
23
Adson de Montierender, Adso Dervensis, De ortu et tempore Antichristi, éd.
Daniël Verhelst, (Turnhout, 1976) ; trad. partielle dans Carozzi et Carozzi-
Taviani, La fin des temps, pp. 20–34.
24
Verhelst, Adso Dervensis, p. 139, relève 190 manuscrits pour la seule version du
Pseudo-Méthode.
36
temps, suivie de la manifestation de la gloire divine au Saint Sépulcre,

au pèlerinage du ‘dernier empereur’ à Jérusalem.25 La croisade répondait
particulièrement bien à ces prédictions.
Nous avons un autre indice de l’impact de ces spéculations sur la pen-
sée politique à l’époque précédant de peu la croisade. Vers 1085, Benzo
d’Albe se réfère à la prophétie sybilline pour présenter Henri IV comme
le héros attendu. Il cite pour cela un texte plus modifié encore que celui
du Pseudo-Alcuin dans le sens d’une interprétation politique contempo-
raine : le roi Franc est pour lui Henri IV, qu’il veut pousser à l’action ; il
rétablira l’ordre en Pouille et en Calabre, sera couronné roi des Grecs à
Constantinople, puis marchera sur la Palestine où, après avoir vénéré le
sépulcre et les lieux saints, il sera couronné à Jérusalem ; alors s’accomplira
la glorification du saint Sépulcre.26 Cette interprétation de Benzo montre
bien que quelques années avant la croisade, la légende de Charlemagne et
la croyance concernant le dernier empereur romain s’étaient déjà combi-
nées avec l’idée de pèlerinage, ou plutôt de reconquête de la Terre sainte.27
De telles interprétations valorisant le pouvoir franc étaient particu-
lièrement bien reçues dans les régions germaniques, longtemps restées
hostiles au pape et attachées à la mystique impériale. Elles ont fort bien
pu être utilisées, fût-ce à demi-mot, par les prédicateurs de la croisade en
Allemagne, région où le message d’Urbain II ne fut guère prêché directe-
ment. La (fausse ?) lettre de l’empereur Alexis à Robert de Flandre appelle
les chevaliers (bellatores Christi) à venir secourir l’empire ‘pour le salut
de leur âme’ et ‘pour mériter dans le ciel une glorieuse récompense’, et les
presse d’agir vite ‘avant la survenue de l’Antichrist’ afin que le royaume
des chrétiens et le tombeau du Christ ne leur échappe.28 L’idée d’une pro-
chaine survenue de l’Antichrist était donc bien présente dans les esprits à
l’époque de la première croisade, et donnait lieu à diverses spéculations.
Godefroy de Bouillon fut-il influencé par de telles spéculations ? C’est
loin d’être impossible. Selon Albert d’Aix, le duc de Lorraine aurait été
luimême jadis gratifié d’une vision relative à son rôle futur à Jérusalem.29
Albert rapporte aussi que, dix ans avant l’expédition, une autre vision
25
L’empereur des Romains, désigné par ‘Constans’ par la Sybille Tiburtine, est
changé ici en ‘C’, pour s’appliquer à Carolus : Verhelst, Adso Dervensis, p. 107.
26
Benzo d’Albe, Ad Henricum IV Imperatorem, dans MGH SS 11 : 605.
27
Verhelst, Adso Dervensis, p. 109.
28
Paul Riant, Alexii Comneni Romanorum imperatoris ad Robertum 1 Flandriae
comitem epistola spuria (Genève, 1879) ; Epistulae et chartae, no. I, p. 134.
29
Albert d’Aix, VI.26–27 : Albert of Aachen, Historia Ierosolimitana : History
of the Journey to Jerusalem, éd. et trad. Susan B. Edgington (Oxford, 2007),
pp. 436–39.
37
Jean Flori
accordée à un chevalier digne de foi, avait désigné Godefroy comme

devant être institué chef et prince de tout le peuple chrétien.30 Plus précise
encore, une autre vision, accordée à un probable compagnon d’Albert,
aurait révélé, sept mois après le départ de Godefroy, que celui-ci avait été
choisi par Dieu pour être le chef de l’armée chrétienne et devenir le pre-
mier à Jérusalem.31 Le prestige de Charlemagne et la force des traditions
attribuant à l’empereur la défense du saint Sépulcre contribuaient aussi
à désigner Godefroy. Celui-ci descendait, on le sait, de Charlemagne par
sa grand’mère ; de plus, une tradition, recueillie par Richard le Pèlerin
et Matthieu d’Édesse, fait de lui le possesseur de la couronne de l’empe-
reur romain Vespasien et surtout de son épée mythique, dont Vespasien
(comme tous les conquérants de Jérusalem qui l’avaient avant lui possédée)
aurait usé, après sa conversion supposée au christianisme, pour venger sur
les Juifs la mort du Christ, à l’instigation de Dieu.32 Il y avait là de quoi
nourrir, dans l’esprit de Godefroy et de son entourage, des spéculations
mystico-politiques.
Une autre tradition, inspirée du Pseudo-Méthode, met plus encore
l’accent sur les Juifs, et sur le rôle du dernier empereur à Jérusalem. Cette
fois ce roi franc, qui devait règner dans les derniers temps, ne viendrait pas
à Jérusalem pour y déposer sceptre et couronne, mais pour les y recevoir au
contraire sur le mont des Oliviers. La date de ce texte est certes controver-
sée, car ce dernier point évoque si nettement la conquête de 1099 et les ori-
gines du royaume de Jérusalem qu’il incline beaucoup d’érudits à le situer
après l’élection de Godefroy de Bouillon.33 La démarche méthodologique
30
Albert d’Aix, VI.33 (éd. Edgington, pp. 444–47). Albert prend soin par la
suite de montrer que cette prophétie annonçant un ‘élu de Dieu’ ne fut pas
accomplie par les chefs des armées précédentes, exterminées en cours de route ;
ils n’étaient donc pas ceux par qui devait advenir le salut d’israël.
31
Albert d’Aix, VI.36 (éd. Edgington, pp. 448–49).
32
Selon La Chanson d’Antioche, vers 4169–75 (éd. Suzanne Duparc-Quioc,
2 vols (Paris, 1977), 1 : 223) cette épée mythique appartint successivement
à Alexandre, puis à Judas Macchabée, à Vespasien, puis à Cornumaran, tous
furent conquérants de Jérusalem. Voir aussi Matthieu d’Édesse, ‘Chronique’,
dans RHC Arm. 1 : 25 ; voir aussi La Venjance Nostre Seigneur, éd. Loyal A .T.
Gryting (Ann Arbor, 1952).
33
Verhelst, Adso Dervensis, p. 141 ; voir aussi Daniël Verhelst, ‘Les textes
eschatologiques dans le liber floridus’, dans The Use and Abuse of Eschatology,
pp. 299–305. Ce texte est nécessairement antérieur à 1120 puisqu’il est contenu
dans le Liber Floridus de Lambert de Saint-Omer. Lambert, témoin privilégié
des préparatifs matériels et psychologiques à la croisade de Robert II de
Flandre, en serait lui-même l’auteur peu après la croisade ; il aurait adapté les
38
est contestable ; mais quoi qu’il en soit de sa date précise, ce texte prouve
en tout cas l’intérêt considérable porté, au moins dans ces régions, à plu-
sieurs traits prophétiques définitivement liés : la légende du pèlerinage
de Charlemagne associée à celle du roi des derniers jours.34 Les récits de
croisade, malgré leur date de rédaction, portent encore quelques traces de
ce climat : la marche des croisés d’Antioche à Jérusalem et le siège de la ville
sainte s’accomplissent clairement, on l’a récemment souligné, dans une
perspective eschatologique que confirme encore l’élection de Godefroy
de Bouillon.35 Pierre l’ermite a-t-il utilisé ces traditions prophétiques, si
en faveur à la fin du xie siècle, dans les régions traversées ? On l’ignore,
écrits antérieurs aux informations qu’il avait peutêtre reçues concernant la

prise de Jérusalem et la fondation du nouveau royaume. Soulignons ici une
certaine forme de pétition de principe dans ce refus d’attribuer à la période
antérieure à la croisade un texte qui s’y insère pourtant fort bien, pour la seule
raison qu’il serait trop précis. Cette précision est d’ailleurs toute relative
puisque ni Godefroy ni ses successeurs n’ont reçu les attributs royaux au
mont des Oliviers. En revanche, la cérémonie de pénitence, d’humilité et de
réconciliation des chefs au mont des Oliviers, précédant la procession autour
des murailles de Jérusalem, doit peut-être quelque chose à cette tradition : au
Mont des Oliviers, les princes oublient leurs querelles de préséance politique
pour ‘remettre’ le pouvoir à Dieu seul. Cf. Albert d’Aix, VI.8 (éd. Edgington,
pp. 412–15). Rien n’empêche donc de situer ce texte, comme les précédents,
avant le déclenchement de la croisade.
34
Sur la légende de Charlemagne, prototype du pèlerin-croisé à Jérusalem et sur la
légende de l’empereur des derniers jours, voir Carl Erdmann, ‘Endkaiserglaube
und Kreuzzugsgedanke im 11. Jhdt’, Zeitschrift für Kirchengeschichte 51 (1932),
394–414 ; Steven Runciman, ‘Charlemagne and Palestine’, English Historical
Review 50 (1935), 609–19 ; Robert Folz, Le souvenir et la légende de Charlemagne
dans l’empire germanique médiéval (Paris, 1950), pp. 134–42 surtout ; Kurt
Kloocke, ‘Kreuzzugsideologie und chansons de geste’, dans Festschrift für
K. Wais, éd. Johannes Hösle et Wolfgang Eitel (Tübingen, 1972), pp. 1–18 ;
Aryeh Grabois, ‘Charlemagne, Rome and Jerusalem’, Revue Belge de Philologie
et d’Histoire 59 (1981), 792–809 ; Riley-Smith, First Crusade, p. 112.
35
Voir sur ce point la trop timide (mais significative) concession récente de
André Vauchez, ‘Les composantes eschatologiques de l’idée de croisade’,
dans Le Concile de Clermont, pp. 233–43, rejoignant en cela Flori, La première
croisade. L’occident chrétien contre l’islam (Bruxelles,1992), pp. 40–43. Sur le
sens eschatologique de l’élection de Godefroy, voir Luc Ferrier, ‘La couronne
refusée de Godefroy de Bouillon : eschatologie et humiliation de la majesté aux
premiers temps du royaume latin de Jérusalem’, dans Le Concile de Clermont,
pp. 245–65 ; sur l’atmosphère prophético-eschatologique de la prise de
Jérusalem, voir Flori, Pierre l’ermite, pp. 415–17.
39
Jean Flori
mais le contraire serait surprenant. Pierre s’est attardé à prêcher à Pâques,

puis à la Pentecôte, périodes qui se prêtaient particulièrement bien à des
sermons de ce genre. Beaucoup de pogroms débutèrent d’ailleurs peu de
temps après ses prédications. Quant aux persécutions des bandes d’Emich
ou de Volkmar, elles résultent manifestement d’un antisémitisme profond
lié à une espérance apocalyptique dont la réalisation était conditionnée
par la conversion massive des Juifs ou par leur extermination, comme je
crois l’avoir montré ailleurs, avec d’autres.36 Selon Ekkehard, Emich se
disait suscité par des révélations divines pour répandre la foi chrétienne
et assurer définitivement son triomphe avant les derniers jours.37 On com-
prend mieux, dans ce contexte, l’acharnement de ses adeptes à convertir les
Juifs plus encore qu’à les tuer, afin que se réalise pleinement la prophétie
s’appliquant à leur chef. Celui-ci, après la conversion finale des Juifs, devait
être couronné par le Christ et triompher à Jérusalem, vaincre l’Antichrist
et préparer l’instauration du royaume de Dieu.
L’espérance eschatologique ne se cantonnait probablement pas exclu-
sivement aux seuls partisans d’Emich et ne donnait pas nécessairement
lieu à des croyances et à des déviances aussi funestes. Cette espérance
n’apparaît pourtant que de manière très discrète et très voilée dans les
chroniques de croisade, particulièrement dans les sources dites ‘fran-
çaises’. Pourquoi ? Deux explications sont possibles. Ou bien il s’agirait
là d’une croyance limitée aux régions traversées par Pierre l’ermite et ses
émules ; ou bien les chroniqueurs occidentaux auraient volontairement
choisi d’occulter cet aspect devenu triplement gênant après les massacres
36
Voir sur ce point Jonathan Riley-Smith, ‘The First Crusade and the Persecution
of the Jews’, dans Persecution and Toleration, éd. William J. Sheils, Studies in
Church History, 21 (1984), pp. 51–72 ; Robert Chazan, European Jewry and
the First Crusade (London, 1987), pp. 250–51 ; Jean Flori, ‘Une ou plusieurs
“première croisade” ? Le message d’Urbain II et les plus anciens pogroms
d’occident’, Revue historique 285 (1991), pp. 3–27 ; Flori, La première croisade,
pp. 53–54 ; Benjamin Z. Kedar, ‘The Forcible Baptisms of 1096 : History and
Historiography’, dans Forschungen zur Reichs-, Papst- und Landesgeschichte, éd.
Karl. Borchardt et Enno Bünz, 2 vols (Stuttgart, 1998), 1 : pp. 187–200.
37
Ekkehard, ‘Hierosolymitana’, 12, dans RHC Occ. 5 : 20. Tel un gourou de notre
époque, il semble avoir parfaitement su utiliser la crédulité populaire, et se
serait identifié, pour accomplir sa mission et recruter des adeptes, à ce roi des
derniers jours annoncé prophétiquement par Dieu. On lui attribue aussi l’usage
de signes prétendument célestes pour attirer à lui les foules ; voir Chronique
de Solomon bar Simson, dans Chazan, European Jewry, p. 250, et Kedar, ‘The
Forcible Baptisms of 1096’, p. 195.
40
des Juifs, celui des troupes de Pierre l’ermite et la non-réalisation de la

prophétie.
De tous les chroniqueurs ‘français’, en effet, seul Guibert de Nogent se
fait clairement l’écho de ces croyances. Pour lui, Urbain II y aurait fait
allusion en établissant un lien entre l’expédition des croisés et l’ultime
combat du Christ et des siens contre l’Antichrist et ses suppôts. Lien
logique à ses yeux dès l’instant où il prêche une guerre destinée à restaurer
le christianisme en Palestine : car, aurait dit le pontife, ‘il est évident que ce
n’est pas contre les Juifs ni contre les païens que l’Antichrist fera la guerre ;
mais contre les chrétiens, selon l’étymologie même de son nom.’ Si donc
le christianisme n’est pas rétabli en Terre sainte où aujourd’hui règne le
paganisme, ces prophéties ne seront jamais réalisées :
Il est nécessaire, avant la venue de l’Antichrist, que l’empire du christianisme
soit, par vous ou par ceux que Dieu aura choisis, rétabli en ces régions, afin
que ce chef de tous les maux, qui y établira le trône de son règne, y trouve
quelque foi contre laquelle il puisse livrer combat. Réfléchissez donc à cela :
le Tout-Puissant vous a peut-être prédestinés afin que, par vous, Jérusalem
cesse d’être ainsi foulée aux pieds.38
Par ces mots, s’il les a bien prononcés, Urbain II n’annonçait d’ailleurs pas
explicitement la venue imminente de l’Antichrist et de la fin des temps,
mais seulement la nécessité, pour l’accomplissement ultérieur des pro-
phéties, de rétablir préalablement l’autorité des chrétiens sur Jérusalem.39
L’historien français Alphandéry a jadis mis l’accent, avec (peut-être un
peu trop de) force, sur l’impact de cette espérance eschatologique dans
laquelle il voyait l’élément mobilisateur principal des premiers croisés.
Mais, ajoutait-il à propos de ce texte, ‘à ce stade d’élaboration, il n’y a à
peu près sûrement presque plus rien des paroles d’Urbain II à Clermont.’40
J’ai d’abord partagé sur ce point l’avis de la plupart des historiens
récents de la première croisade, dissociant radicalement le message des
prédicateurs populaires de celui d’Urbain II à Clermont.41 Je n’en suis
plus aussi certain aujourd’hui. Certes, Guibert de Nogent a très proba-
blement développé ce thème eschatologique pour lui donner l’ampleur
38
Guibert de Nogent, Dei gesta per Francos, éd. Robert B.C. Huygens (Turnhout,
1996), p. 115.
39
Peut-être est-ce là une ‘atténuation’ – dont Guibert serait l’auteur – destinée
à concilier le thème abordé par le pape et la non-réalisation de la venue de
l’Antichrist ?
40
Alphandéry et Dupront, La chrétienté et l’idée de croisade, pp. 41–42 et passim.
41
Voir en particulier Flori, ‘Une ou plusieurs “première croisade” ?’, p. 327, et
Flori, La première croisade, pp. 35–44.
41
Jean Flori
qu’il a dans son récit. Mais peut-être ne l’a-t-il pas totalement inventé.
L’allusion d’Urbain II à une nécessaire reconquête chrétienne des lieux
saints dans une perspective prophétique était dans la nature des choses,
logique, surtout sous cette forme relativement anodine. Une telle allusion
n’est donc pas totalement irrecevable.
Il resterait alors à expliquer pourquoi Guibert, qui n’était pas présent
à Clermont, est le seul à y faire allusion. Peut-être le fait-il d’après des
rapports de témoins ? Dans ce cas, la rigoureuse honnêteté intellectuelle
qu’on lui reconnaît le conduisait à en faire état. On voit mal, en revanche,
pourquoi il aurait de son propre chef introduit ce thème lorsqu’il rédige
son récit entre 1104 et 1108, à une époque où, de toute évidence, les temps
de la fin ne s’étaient aucunement manifestés. Au contraire, si Urbain II
a bien fait allusion, même de façon très voilée, à ces aspects dans ses pré-
dications, on peut comprendre les raisons qui ont pu conduire les autres
chroniqueurs (en particulier Baudri de Bourgueil et Robert le Moine)
à en faire, après coup, l’économie. Cette suppression présentait en effet
deux avantages : elle exonérait le pape de toute allusion eschatologique
non réalisée, puisqu’en Terre sainte ni l’Antichrist ni le Christ n’étaient
apparus, avant comme après la prise de Jérusalem ; elle l’exonérait aussi de
toute responsabilité dans les massacres des Juifs dont les premiers croisés,
plus ou moins liés à Pierre l’ermite, s’étaient rendus coupables. On ne
doit pas s’étonner de telles ‘coupures’ à intentions idéologiques : Foucher
de Chartres se livre à une ‘épuration’ du discours d’Urbain II bien plus
radicale encore puisqu’il en supprime toute allusion à Jérusalem, pourtant
au centre de la prédication du pape à Clermont, dans ses lettres et dans sa
tournée de propagande.42
Dans la lettre, extraordinaire à bien des égards, que Bohémond et les
princes croisés adressent en septembre 1098 à Urbain II, les chefs résu-
ment les mobiles de leur action, font le bilan des victoires accomplies et
exposent leurs craintes et leurs espérances pour les jours à venir. Les croisés,
‘Hierosolymitani Iesu Christi’, ont vengé sur les Sarrasins les injures infli-
gées à Dieu et à Jésus Christ. Grâce à Dieu, au miracle de la sainte Lance, ils
ont vaincu les païens, pris Antioche, et soumis toute la région à la religion
chrétienne, ou plus précisément à la foi ‘romaine’. Mais ces guerriers, qui
42
J’ai tenté ailleurs de fournir une explication à cet ‘oubli’ : Baudouin, après sa
conquête d’Edesse, n’a pas pris part à la marche sur Jérusalem et à sa conquête.
Foucher de Chartres, passé au service de Baudouin, aurait ainsi voulu éviter de
suggérer que son maître n’aurait pas pleinement accompli la mission prêchée
par Urbain II. Son éloge des mérites de Baudouin et de Bohémond dans la
réussite de l’opération va dans le même sens. Voir sur ce point Flori, La première
croisade, p. 31.
42
savent fort bien vaincre et tuer les infidèles, se disent démunis devant les
hérétiques, Grecs, Arméniens, Syriens, Jacobites, qui pullulent en Orient.
Ils ont besoin du pape pour leur désigner l’adversaire à combattre et à
détruire. De plus, avant de poursuivre la conquête vers Jérusalem, ils ont
besoin d’un guide et font appel à Urbain II pour qu’il vienne en personne
achever ‘sa guerre’. L’appel se termine par cette exortation superbe :
Ainsi, avec nous, achève la voie de Jésus Christ que nous avons commen-
cée et que tu as prêchée ; ouvre pour nous les portes des deux Jérusalem,
libère le Sépulcre du Seigneur et fais que le nom des chrétiens soit exalté
au dessus de tout nom.43
La première Jérusalem, sans aucun doute, est bien la ville sainte que les
croisés ont pour tâche de conquérir. Mais quelle est ici la seconde ? On peut
hésiter : dans une perspective eschatologique, il s’agirait de la Jérusalem
céleste qui doit descendre aux temps de la fin : le pape, alors, y mènerait
ses élus ; dans une perspective sotériologique, il peut s’agir de la Maison
de Dieu, le royaume des élus, de ceux qui, convertis, ont décidé de suivre
le Christ. À cette date, il est permis de croire que les deux dernières inter-
prétations pouvaient être mêlées. Il est clair, en tout cas, que les croisés
entendaient bien, par la conquête armée de la Jérusalem terrestre, s’ouvrir
aussi les portes d’une autre Jérusalem, céleste ou spirituelle, qui n’avait
rien à voir avec les lieux saints du pèlerinage ni avec la satisfaction d’une
pénitence.
3. Bâtir la Jérusalem spirituelle

Troisième facteur possible de sacralisation : la conversion des infidèles.
On a souvent affirmé que la première croisade n’avait aucunement la
dimension d’une guerre ‘missionnaire’ et j’ai moi-même abondé dans ce
sens.44 J’en suis, là encore, moins sûr aujourd’hui. Cette dimension est
évidente, on vient de le rappeler, dans les intentions des croisés ‘populaires’
à l’égard des Juifs. Mais on peut aussi en trouver quelques traces dans la
croisade générale à l’égard des musulmans.
Certes, la conversion des infidèles ne pouvait pas être prêchée comme
telle et les documents connus n’en font pas état. Par ailleurs, il ne s’agit
43
Lettre de Bohémond et des princes croisés à Urbain II (sept. 1098), dans
Epistulae et chartae, no. XVI, 161–65.
44
Voir par exemple, entre autres, Jean Richard, L’esprit de la croisade (Paris, 1969),
p. 28 ; Riley-Smith, The First Crusade, pp. 109–11 ; voir toutefois Flori, La
première croisade, pp. 229–30, esquissant déjà, en 1992, la position que j’adopte
pleinement aujourd’hui.
43
Jean Flori
évidemment pas d’une conversion au sens actuel du terme, mais au sens ou

l’entendaient alors les chevaliers :45 leur rôle ne consistait pas à convaincre
les musulmans, mais à les vaincre ; à ramener à la chrétienté d’abord des
territoires, et non des hommes, dans une perspective un peu comparable,
sur ce point précis, à celle des guerriers musulmans dans le djihâd.46 En
ce sens, la croisade était – et restera encore longtemps par la suite dans la
pensée populaire – non pas un obstacle à la conversion des peuples, mais
un moyen d’y parvenir en rendant manifeste, par la victoire, la supériorité
du Dieu des chrétiens sur les faux dieux supposés des infidèles, en une
sorte d’ordalie ou de jugement de Dieu.47 C’est peut-être dans ce sens
qu’Albert d’Aix signale la conversion du seigneur sarrasin de Ramlah, dont
le baptême aurait eu lieu, selon certains, avant la bataille d’Ascalon mais
que d’autres situent plutôt après la victoire des croisés.48 Les chevaliers ont
donc pu concevoir leur action guerrrière comme destinée à eshalcier sainte
crestïenté dans les trois sens de cette expression, comme le disent aussi les
chansons de geste et, dans une certaine mesure, la lettre de Bohémond à
Urbain II déjà citée,49 et par làmême d’amener à Dieu au moins les infidèles
‘de bonne volonté’.50 L’étude du motif de la conversion des ‘païens’ dans la
45
Voir sur ce point Flori, Chevaliers et chevalerie au Moyen Age (Paris, 1998),
pp. 203–34.
46
Voir sur ce point Flori, ‘Croisade et djihâd : Le problème de la guerre dans le
christianisme et l’islam’, dans Les croisades. L’Orient et l’Occident d’Urbain II
à saint Louis, 1096–1270, éd. Monique Rey-Delqué (Milan, 1997), pp. 49–62 ;
Flori, ‘Croisade et Gihad’, dans Le Concile de Clermont , pp. 267–85 et, plus
récemment, Flori, Guerre sainte, jihâd, croisade : Violence et religion dans le
christianisme et l’islam (Paris, 2001).
47
Elizabeth Siberry, ‘Missionaries and Crusaders (1095–1274) : Opponents or
Allies ?’, dans The Church and War, éd. William J. Sheils, Studies in Church
History 20 (1983), pp. 103–110. Voir aussi sur ce point, dans une perspective
plus évangélique, Paul Rousset, ‘La croisade obstacle à la mission’, Nova et
vetera 57 (1982), 133–41.
48
Albert d’Aix, VI.44 (éd. Edgington, pp. 460–61).
49
Voir sur ce point David A. Trotter, ‘“Por eshalcier sainte crestïenté”’, French
Studies Bulletin 3/7 (1983), 1–3 ; Jean Flori, ‘Pur eshalcier sainte crestïenté ;
croisade, guerre sainte et guerre juste dans les anciennes chansons de geste
françaises’, Le Moyen Age 97 (1991), 171–87.
50
Le Moniage Guillaume traduit assez bien ce point de vue en faisant exprimer par
Guillaume, son héros, chevalier devenu moine, son regret d’avoir choisi l’ordre
monastique : ‘Mieus vaut li ordenes de la cevalerie, Qu’il se combatent vers la
gent sarrasine, Prendent lor terres et conquierent lor viles, Et les païens a no
loi convertissent’ : Le Moniage Guillaume, éd. Wilhelm Cloetta (Paris, 1906),
44
Chanson de Roland, par exemple, illustre bien la mentalité des chevaliers

à cet égard, et le rôle qu’ils pouvaient s’attribuer.51
On en trouve aussi quelques exemples dans les récits de croisade.52
Albert d’Aix raconte qu’à Nicée un espion sarrasin fut capturé dans le
camp ; menacé de torture par Godefroy pour lui faire révéler les intentions
des musulmans, il les avoua et demanda le baptême. Albert, toutefois, émet
quelques doutes sur la sincérité d’une telle demande, due bien plus, selon
lui, à la peur qu’à une réelle attirance pour la foi catholique.53 Il raconte
pourtant une nouvelle tentative du même genre, omise par les autres
sources, lors du siège de Jérusalem par les chrétiens. Ceux-ci avaient réussi
à capturer l’un des sarrasins qui avaient tendu une embuscade où avaient
péri plusieurs chrétiens, parmi lesquels Gilbert de Trèves et Achard de
Montmerle ; ils leur avaient coupé la tête. Le captif était un personnage
important et Albert précise que les croisés firent tous leurs efforts pour
le convertir à la foi chrétienne. N’y parvenant pas, ils lui tranchèrent la
tête devant les murailles, pour effrayer les défenseurs.54
D’autres tentatives de conversion furent sans doute mieux conduites,
car Albert d’Aix signale la présence de Turcs convertis parmi les armées
chrétiennes. L’un d’entre eux le fut par l’intermédiaire de Bohémond,
dont il adopta le prénom à son baptême ; choisi pour aller espionner le
camp des renforts musulmans près d’Antioche, il aurait servi par la suite
vers 640–43. Bien que plus tardif et désignant sans doute les ordres religieux
militaires, ce texte n’en traduit pas moins une ‘mentalité’ commune chez les
chevaliers.
51
Voir sur ce point Jean Flori, ‘La croix, la crosse et l’épée. La conversion des
infidèles dans La Chanson de Roland et les chroniques de croisade’, dans ‘Plaist
vos oïr bone cançon vallant ?’ Mélanges de Langue et de Littérature Médiévales
offerts à François Suard, éd. Dominique Boutet, Marie-Madeleine Castellani,
Françoise Ferrand et Aimé Petit, 2 vols (Lille, 1999), 1 : pp. 261–72 ; voir aussi
Jean Flori, ‘Première croisade et conversion des “païens”’, dans Migrations et
diasporas méditerranéennes (X-XVIs.), éd. Michel Balard et Alain Ducellier
(Paris, 2002), pp. 449–57.
52
Riley-Smith, The First Crusade, n’accepte que deux cas de conversions forcées
massives, dont l’une (celle d’al-Barah en septembre 1098, p. 110), résulterait selon lui
d’une lecture fautive de Robert le Moine lisant crediderant au lieu de reddiderant. La
chose est vraisemblable, et je passe sous silence cette occurrence. Mais il y en a d’autres.
53
Albert d’Aix, II.26 (éd. Edgington, pp. 104–5).
54
Albert d’Aix, VI.5 (éd. Edgington, pp. 410–11). Voir sur ce point Kedar, Crusade
and Mission, p. 63.
45
Jean Flori
de négociateur entre Bohémond et ‘Firouz’ dans les tractations qui allaient

livrer la ville au Normand.55
Avant ce succès, un combat sous Antioche entraîna aussi, selon Albert,
un grand nombre de ‘Gentils’ à adopter le christianisme.56 Lors du siège
d’Antioche, après les victoires chrétiennes des 5 et 6 mars 1098, de nom-
breux habitants de la ville perdirent courage et sortirent pendant la nuit
pour se rendre, se disant prêts à embrasser la foi chrétienne : ils furent ‘admis
au nombre des fidèles’.57 Après la prise de Jérusalem, les croisés tentèrent
aussi d’obtenir la conversion de leurs captifs juifs, comme le signale une
lettre contemporaine louant la résistance d’un jeune juif de 9 ans.58
Les gesta de l’Anonyme présentent aussi plusieurs cas de conversions col-
lectives offertes ou suggérées. Ainsi, à Antioche, le chef turc de la citadelle
conclut avec Bohémond un traité selon lequel les ‘païens’ qui voudraient
recevoir le christianisme pourraient rester dans la ville tandis que les autres
seraient autorisés à la quitter sains et saufs. De fait, quelques jours plus tard,
les Sarrasins voulant rester musulmans furent conduits dans ‘leur pays’ tan-
dis que le chef turc fut baptisé ‘avec tous ceux qui préférèrent reconnaître le
Christ’, à la plus grande joie du camp chrétien.59 La lettre de Bohémond à
Urbain II, on l’a vu, suggère non seulement une soumission des habitants
de la région aux croisés, mais à la religion romaine, ce qui pourrait bien
55
Albert d’Aix, III.61 (éd. Edgington, pp. 234–35) et IV.15 (pp. 270–73).
56
Albert d’Aix, III.66 (éd. Edgington, pp. 246–47). Peut-être s’agissait-il
d’Arméniens ou de ‘convertis’ à l’islam ? Rien, toutefois, ne permet de
l’affirmer ; quoi qu’il en soit, l’épisode, comme les précédents, témoigne de la
mentalité des croisés plus que de la réalité des faits.
57
Albert d’Aix, III.66 (éd. Edgington, pp. 246–47). S’agissait-il de Turcs de la
garnison ? d’habitants musulmans de la ville ? On l’ignore.
58
Lettres juives contemporaines de la capture de Jérusalem : ‘Contemporary
Letters on the Capture of Jerusalem’, éd. Shelomo D. Goitein, Journal of Jewish
Studies 3 (1952), 162–77 ; trad. fr. dans Claude Cahen, Orient et Occident au
temps des croisades (Paris, 1983), pp. 224–25.
59
Anonyme, Gesta Francorum et aliorum Hierosolimitanorum, éd. Louis Bréhier
(Paris, 1924), pp. 158 et 190 ; Chanson d’Antioche, éd. Duparc-Quioc, vers
9540–69 ; Robert le Moine, VII.18, RHC Occ. 3 : 836 ; Voir aussi Tudebode,
Historia de Hierosolymitano itinere, éd. John H. Hill et Laurita L. Hill (Paris,
1977), pp. 108–9, 79–80, cf. La chanson de Roland, éd. Jean Dufournet (Paris,
1993), pp. 83, 113 ; Raymond d’Aguilers, ‘Liber’, éd. John H. Hill et Laurita
L. Hill (Paris, 1969), pp. 79, 115 ; Robert le Moine, IX.18, V.17, VII.22, VII.9 ;
Baudri de Bourgueil, ‘Historia Jherosolimitana’, dans RHC Occ. 4 : 74 ; Guibert
de Nogent, Dei gesta, II.1 et VII.17 ; Raoul de Caen, ‘Gesta Tancredi’, cap.43,
dans RHC Occ. 4 : 638.
46
impliquer des conversions. Un peu plus tard, Raymond Pilet et ses hommes
s’emparèrent ‘avec l’aide du Christ’ d’une place forte et firent un tri rapide
parmi les habitants qui s’y étaient réfugiés : ‘ils mirent à mort tous ceux
qui ne voulaient par recevoir le christianisme ; quant à ceux qui préfé-
rèrent reconnaître le Christ, ils les laissèrent vivants.’60 Lors de la marche
des croisés d’Antioche vers Jérusalem, plusieurs garnisons furent passées
au fil de l’épée, et il en résulta des redditions et des demandes d’alliance.
Ainsi le ‘roi de Tripoli’ offrit-il son ‘amitié’ à Raymond de Toulouse qui la
repoussa, déclarant qu’il ne ferait jamais la paix, sauf s’il se convertissait.
Un traité fut conclu par la suite avec lui, entraînant la délivrance de 300
prisonniers croisés ; il stipulait que si les Francs parvenaient à l’emporter
sur les armées fatimides du Caire et à prendre Jérusalem, le ‘roi de Tripoli’
se ferait chrétien et tiendrait d’eux sa terre.61
Autre indice allant dans le même sens : les princes chrétiens, en envoyant
des ambassades aux Fatimides d’Égypte, ennemis des Turcs, espéraient
obtenir leur alliance, mais aussi parfois leur conversion ; ils pensaient
qu’ils y seraient inclinés en voyant les miracles accomplis pour eux par
Dieu, qui leur donnait sous leurs yeux la victoire.62 Les chroniqueurs rap-
portent que les croisés, lors du siège d’Antioche, reçurent des Fatimides
d’Égypte une proposition d’alliance contre les Turcs. La réalité de cette
ambassade est indiscutable, ainsi que les objectifs d’alliance poursuivis.
Mais Albert d’Aix va plus loin : en reconstituant à sa manière, selon les
coutumes du temps, la teneur du message adressé aux chefs croisés, il fait
état d’une possible adoption de la foi chrétienne par le pouvoir fatimide.
Cette reconstruction est certes fantaisiste, mais elle témoigne cependant
de la croyance ambiante selon laquelle la conversion massive des ‘païens’
était possible par le truchement de combats victorieux.63 Selon Guibert
de Nogent, le calife fatimide avait proposé de se faire chrétien si les croi-
sés l’aidaient à reconquérir les territoires récemment conquis sur lui par
les Turcs, y compris Jérusalem.64 L’Historia belli sacri, qui puise à une
60
Anonyme, Gesta, éd. Bréhier, p. 165.
61
Anonyme, Gesta, éd. Bréhier, p. 185 ; voir aussi le récit de la conversion du fils
du prince d’Azaz, malgré la mort de son fils otage, Albert d’Aix, V.24 (éd.
Edgington, pp. 366–67) et V.32 (pp. 378–79).
62
Sur les relations diplomatiques entre croisés et musulmans dès cette époque, voir
Hadia Dajani-Shakeel, ‘Diplomatic Relations between Muslim and Frankish
Rulers, 1097–1153 ad’, dans Crusaders and Muslims in Twelfth-Century Syria,
éd. Maya Shatzmiller (Leiden, 1993), pp. 190–215.
63
Albert d’Aix, III.59 (éd. Edgington, pp. 230–33).
64
Guibert de Nogent, Dei gesta, IV.13, éd. Huygens, p. 189. Le pouvoir était alors
entre les mains d’al-Afdal Shahanshah, vizir de 1094 à 1121.
47
Jean Flori
tradition indépendante, signale aussi l’envoi, de Nicée, d’une ambassade

occidentale auprès de la cour fatimide pour l’informer de l’intention des
croisés d’aller délivrer Jérusalem, offrant au calife l’alternative suivante :
ou bien se convertir au christianisme et devenir ainsi leur allié, ou bien
s’apprêter à les affronter.65
Foucher de Chartres exprime, à propos du siège d’Antioche, une concep-
tion compatible de la guerre sainte et de la conversion qui, par bien des aspects,
rejoint celle qu’exprimera quelques années plus tard Bernard de Clairvaux :
Lorsque (Dieu) permet que, par les Turcs, des chrétiens soient tués, c’est
pour leur salut ; quant aux Turcs, c’est pour la perte de leur âme. Cependant,
il plut à Dieu que quelques-uns d’entre eux, qui avaient été prédestinés au
salut, fussent baptisés par nos prêtres. Car ceux qu’il a prédestinés, il les a
aussi appelés et glorifiés.66
Le cas le plus significatif de cette mentalité relative à la conversion des
Sarrasins s’exprime dans le récit de l’affrontement décisif qui, à Antioche,
eut lieu entre les troupes chrétiennes et musulmanes, en une bataille assi-
milée dès l’abord à un véritable jugement de Dieu. Toutes les sources font
état, avant cette bataille, d’une ambassade des croisés auprès de Karbuqa
afin d’obtenir de lui un tel combat qui déciderait du sort de la ville. Selon
plusieurs d’entre-elles, les chrétiens firent état de leurs droits sur Antioche
et offrirent à Karbuqa et à ses hommes le choix de se faire chrétiens et de
conserver ainsi leurs possessions, ou bien d’avoir à affronter les troupes
du vrai Dieu. Karbuqa, bien entendu, aurait rejeté cette offre avec mépris
et aurait fait à son tour une proposition symétrique de conversion à
l’islam. L’épisode est d’autant plus significatif de la mentalité religieuse
des croisés qu’il a été forcément recomposé, selon les poncifs du temps,
par les chroniqueurs dont aucun n’était évidemment présent, peut-être
sur les indications des rares membres de cette ambassade. Or, le principal
d’entre eux était précisément Pierre l’ermite, dont la prédication initiale,
en France du Nord et en Allemagne, pouvait fort bien déjà refléter une
telle conception. Toutefois, l’étude du contexte historique et littéraire de
cette ambassade suggère que Pierre l’ermite a fort bien pu être réellement
chargé par les chefs croisés d’une telle proposition de conversion ou de
combat singulier. En effet, leur sort était, ‘à vues humaines’, totalement
désespéré, mais les visions antérieures, cautionnées quelques heures plus
tôt par la découverte de la Sainte Lance, annonçaient le salut et la victoire
65
‘Historia belli sacri’, cap. 22, RHC Occ. 3 : 181 ; voir aussi caps. 99–103, 212–15, le
récit de l’ambassade du calife et la libération des prisonniers suite au miracle du
feu sacré.
66
Foucher de Chartres, ‘Historia Iherosolymitana’, I :16, RHC Occ. 3 : 341.
48
des chrétiens, sans toutefois en indiquer les modalités. La moins irréaliste

de toutes était peut-être à leurs yeux la conversion de Karbuqa.67
Une lettre des princes croisés aux fidèles d’Occident va tout à fait
dans le même sens. Sa date est discutée, mais se situe presque à coup sûr
entre mai et juillet 1098, c’est-à-dire très peu de temps avant (ou après)
l’affrontement des croisés et des armées de Karbuqa. Elle raconte elle aussi
les victoires remportées jusqu’alors par les croisés, mais ne minimise pas
leurs souffrances et leurs pertes ; elle annonce que bientôt aura lieu une
bataille décisive entre les chrétiens et l’armée du ‘sultan de Perse’ qui,
ayant rassemblé toutes ses forces, leur a fait savoir qu’il se disposait à les
affronter en une sorte de jugement de Dieu. La bataille devait avoir fin
octobre 1098. Le fait qu’aucun affrontement majeur n’eut lieu vers cette
date m’incite à croire que cette bataille prévue vers la Toussaint eut lieu
en réalité bien avant cette date et qu’il s’agit probablement du combat
livré à Antioche contre Karbuqa le 28 juin de cette même année.68 Cette
bataille a pu en effet être annoncée, sur un ton de défi, par une ambassade
musulmane alors que les croisés assiégeaient encore Antioche. Or, dans
cette lettre, les croisés affirment sans ambages que le souverain musulman
les avait avertis de ce combat futur et lui avait attribué la fonction d’une
ordalie, promettant même de se convertir à la foi chrétienne s’il était
vaincu, ce qui pousse les croisés à réclamer les prières de tous les chrétiens
avant la date choisie pour cette bataille décisive, preuve qu’ils croyaient
à la réalité de cette perspective. Le texte mérite attention tant il traduit
bien l’état d’esprit des croisés à ce moment, tout à fait conforme à ce que
l’on a pu constater à propos de l’ambassade de Pierre l’ermite :
Il faut encore que vous sachiez que le roi des Perses nous a fait savoir
qu’il nous livrerait bataille le jour de la fête de tous les Saints, assurant
que s’il en sort vainqueur, il ne cessera de faire la guerre aux chrétiens
de concert avec le roi de Babylone et la plupart des autres rois païens.
Mais s’il perd la bataille, il se fera chrétien avec tous ceux qu’il pourra
entraîner à sa suite.69 En conséquence, nous vous supplions de vous livrer
à cette intention au jeûne et aux aumônes et de célébrer la sainte messe
avec dévotion et assiduité. Et en particulier, observez dévotement, par les
aumônes et les prières, le troisième jour précédant la fête, qui se trouve
67
Voir sur ce point Flori, Pierre l’ermite, pp. 376–83.
68
Dans ce cas, la lettre aurait été écrite probablement en mai, avant la prise
d’Antioche par les croisés, prise que la lettre, d’ailleurs, ne mentionne pas, ce
qui confirme encore cette datation.
69
‘[P]orro si perdiderit, se omnesque quos adiungere poterit, christianos futuros
esse spopondit’.
49
Jean Flori
être un vendredi, jour du triomphe du Christ, que nous choisissons pour

livrer cette mémorable bataille.70
De toute évidence, la conversion des musulmans par l’exemple de la force
des armes paraissait aux croisés une éventualité plausible et souhaitable,
espérée sinon attendue. On ne peut donc pas éliminer totalement des
objectifs des croisés la conversion des musulmans ainsi comprise. La croi-
sade est tout à la fois, dans la mentalité des chevaliers, guerre de reconquête
chrétienne, guerre sainte à connotation eschatologique et guerre de religion
destinée à favoriser la conversion de ceux que Dieu avait prédestinés au salut.
Conclusion
Outre son aspect de pèlerinage et d’entreprise pénitentielle, soulignée
jusqu’à l’excès par les textes ecclésiastiques et cependant bien réelle pour
certains au moins des croisés, la croisade offre ainsi une triple dimension
de sacralisation qui subsiste jusqu’à la bataille d’Ascalon, ultime jugement
de Dieu. Mais deux de ces aspects se sont alors affaiblis, évanouis, et n’ont
guère pu subsister après cette date. En effet, la conversion massive des
‘infidèles’ n’a pas eu lieu et il faudra désormais concevoir autrement leur
conversion, leur entrée dans la Jérusalem spirituelle qu’est l’Église, ce ‘Tout
Israél’ qui doit être sauvé. De même, la descente de la Jérusalem céleste ne
s’est pas produite et Ascalon n’était donc pas Harmaguédon. La fin des
temps est encore à venir et l’Histoire humaine continue. Reste, seule, la
Jérusalem terrestre reconquise, le Saint Sépulcre qu’il faudra désormais
défendre. Ces lieux saints, sources de grâce dans l’Histoire ordinaire,
prennent alors une importance suréminente. La croisade, dès lors, perd
ses caractères eschatologiques et prophétiques pour devenir à son tour, en
elle-même (et non plus dans son but), accomplissement des prophéties.
C’est ainsi que la présentent désormais les chroniqueurs qui, après coup,
en rédigent l’histoire. Pour tous, désormais, les croisés ne sont pas ceux qui
ont ouvert le chemin et la porte du royaume de Dieu, mais seulement ceux
qui ont réouvert la route des pèlerinages à Jérusalem. On pourrait presque,
à leur propos, inverser la formule de Riley-Smith : loin de prendre peu à
peu conscience en cours de route d’être des milites Christi, les croisés qui,
jusqu’ici et dès leur départ se percevaient avant tout comme des milites
Christi, prennent conscience, après leur conquête de la ville sainte et en
l’absence de toute manifestation tangibles d’une autre Jérusalem, d’être
aussi des pèlerins ; il accomplissent alors leur pèlerinage avant de s’en
retourner chez eux. Parce que, décidément, l’Histoire terrestre continue.
70
Lettre des princes croisés reçue par l’évêque Hugues de Grenoble, dans
Epistulae et chartae, no. XII, 154–55.
50
S ome Problems in the Study of the Conduct
of Warfare in the Eleventh Century
Manuel Rojas
University of Extremadura
This essay is not intended as a bilans et perspectives approach of the kind

which is so dear to modern French historiography, but a survey which
briefly outlines some basic questions relating to the conduct of warfare
in western Europe before the First Crusade. Some important aspects
will need to be left out for reasons of space: for example, grand strategy
and operational strategy, technology in combat and during sieges, the
routine of warfare, logistical support, generalship and leadership, and
the composition and size of armies. These are the most relevant, but
not the only facets which have to be omitted, although some aspects of
them will necessarily impinge upon the topics discussed here, because
all of them were inextricable parts of warfare through the ages. These
limitations mean that it is not possible to consider some important works
in the field of medieval war, even in the period we are concerned with.
Similarly, we cannot examine in detail what has come to be known as
the ‘Gillingham Paradigm’,1 although – ironically – the British historian
himself thinks that it would be better renamed as the ‘Smail Paradigm’.2
1
This designation has been given in publications with revisionist character, e.g.
Clifford J. Rogers, ‘The Vegetian “Science of Warfare” in the Middle Ages’,
Journal of Medieval Military History 1 (2003), 1–19, and Stephen Morillo, ‘Battle
Seeking: The Contexts and Limits of Vegetian Strategy’, Journal of Medieval
Military History 1 (2003), 21–41, as well as other writings: João Gouveia
Monteiro, ‘Estratégia e risco em Aljubarrota: A decisão de dar batalha à luz
do “Paradigma Gillingham”’, reprinted in Gouveia Monteiro, Entre Romanos,
Cruzados e Ordens Militaires: Ensaios de História Militar Antiga e Medieval
(Coimbra, 2010), pp. 137–68; L. J. Andrew Villalon, ‘Battle-Seeking, Battle-
Avoiding or Perhaps Just Battle-Willing? Applying the Gillingham Paradigm
to Enrique II of Castile’, Journal of Medieval Military History 8 (2010), 131–54.
2
John Gillingham, ‘“Up with Orthodoxy!” In Defense of Vegetian Warfare’,
Journal of Medieval Military History 2 (2004), 149–58 (here p. 153), replying
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. 1–.
FHG DOI: 1.1/M.OUTREMER-EB.1.116
Manuel Rojas
This thesis focuses on how war was actually fought, assuming that pitched
battles were rare events, fraught with physical, but above all political risks,
which the majority of commanders attempted to avoid whenever possible.
Consequently, the most characteristic features of warfare were sieges and
ravaging of an enemy’s territory.3
Any study of the conduct of warfare in the eleventh century (as for
other periods) has important geographical restrictions. While it would
be desirable to cover all of Latin Christendom, it is very difficult if not
impossible to make an exhaustive and balanced study of how war was
conducted in every part of that geopolitical area. Guy Halsall feared that
his book on warfare in the second half of the first millennium would be
criticised for being ‘Frankocentric’, so that it might have been called The
Franks and their Enemies.4 Similarly, the present essay could easily have
been ‘Normancentric’ or ‘Anglo-Normancentric’; the number and quality
of sources, the areas where Normans were active (Francia Occidentalis,
Italy, the Iberian Peninsula, England, the Celtic lands, Byzantium and
Outremer), the many opponents they faced (adversaries within their own
cultural milieu, as well as Muslims, Anglo-Saxons, Celts and Greeks)
and the attention historians have devoted to all these questions are all so
remarkable that it is advisable to choose Norman warfare as one of the
main threads in this article.5 During the eleventh century the Normans
fought decisive battles such as Civitate (1053) and Hastings (1066) as
to Rogers and Morillo, and with reference to R. C. Smail, Crusading Warfare

(1097–1193) (Cambridge, 1956).
3
Gillingham’s ‘paradigm’ was set up in three seminal articles: ‘Richard I and
the Science of War in the Middle Ages’, in War and Government in the Middle
Ages: Essays in Honour of J. O. Prestwich, ed. John Gillingham and J. C. Holt,
(Woodbridge, 1984), pp. 78–91; ‘War and Chivalry in the History of William
Marshal’, in Thirteenth Century England, II, ed. P. R. Coss and Simon D. Lloyd
(Woodbridge, 1988), 1–14; ‘William the Bastard at War’, in Studies in Medieval
History presented to R. Allen Brown, ed. Christopher Harper-Bill, Christopher
Holdsworth and Janet L. Nelson (Woodbridge, 1989), pp. 141–58 [all reprinted
in Anglo-Norman Warfare: Studies in Late Anglo-Saxon and Anglo-Norman
Military Organization and Warfare, ed. Matthew Strickland (Woodbridge,
1993), pp. 194–207, 251–63, and 143–60].
4
Guy Halsall, Warfare and Society in the Barbarian West, 450–900 (London,
2003), p. 12.
5
Among the vast literature on this topic, see David C. Douglas, The Norman
Achievement, 1050–1100 (Berkeley, 1969); Douglas, The Norman Fate, 1100–1154
(Berkeley, 1976); R. Allen Brown, The Normans (Woodbridge, 1984); Marjorie
Chibnall, The Normans (Oxford, 2000); A Companion to the Anglo-Norman
52
well as other important engagements. They were adventurers and mer-

cenaries who knew how to plan and implement raids; they were eager to
erect castles and skilful in siege techniques. They were led by formidable
leaders, who generally were up to date in tactics and military technology
and achieved astounding victories. In short, they were conquerors who
built themselves a positive image as fierce and untameable warriors. This
brilliant curriculum more than justifies following a Norman course when
dealing with war and its conduct in the eleventh century. However, I have
tried not to confine discussion to these limits, but to include discussion
of other geographical areas.
The amount of information offered by sources on politics and warfare
in Latin Christendom is far from homogenous, differing widely in space
and time in the course of the long tenth century, that is from the end of
the ninth century to around 1030, which historians have regarded as a
crucial epoch in European history.6 However, knowledge of ‘what really
happened’ is far from complete. It is widely believed that this long period
lacks reliable sources, and indeed only the seventh century matches this
deficiency of information. Yet the writing of history is not only based on
the sequences of facts transmitted in narrative sources. The tenth century
was a golden age for hagiography, which rigid nineteenth-century clas-
sifications considered as fiction rather than as a reservoir of information.7
Orality, which suffers from the passage of time and has left few written
traces, is far more difficult to interpret than texts, and it is certainly a
more treacherous source for historians.8 Yet this in no way means that
World, ed. Christopher Harper-Bill and Elisabeth van Houts (Woodbridge,

2003); David Bates, The Normans and Empire (Oxford, 2013).
6
Joseph Calmette, L’enfondrement d’un empire et la naissance d’une Europe,
ix e-xi e siècles (Paris, 1941); Geoffrey Barraclough, The Crucible of Europe:
The Ninth and Tenth Centuries in European History (London, 1976); Robert
Fossier, Enfance de l’Europe, xe-xiie Siècles: Aspects économiques et sociaux, 2 vols
(Paris, 1982); Johannes Fried, Die Formierung Europas, 840–1046 (München,
1991). No author has expresssed these ideas more clearly than Robert Sabatino
López, ‘Still Another Renaissance?’, American Historical Review 57 (1951), 1–21.
7
There is no clear-cut distinction between ‘intention’ and ‘practice’; many
‘historians’ were also ‘hagiographers’, as Felice Lifshitz has pointed out: ‘Beyond
Positivism and Genre: “Hagiographical” Texts as Historical Narrative’, Viator
25 (1994), 95–113. On the utility of hagiography for this subject, see John
France, ‘War and Sanctity: Saints’ Lives as Sources for Early Medieval Warfare’,
Journal of Medieval Military History 3 (2005), 14–22.
8
Jean Flori, ‘La Notion de chevalerie dans les chansons de geste du xiie siècle:
Etude historique de vocabulaire’, Le Moyen Age 81 (1975), 211–44 (here 211).
53
Manuel Rojas
it is not an important tool to study the mentality of warriors.9 Orally

transmitted sources open up perspectives for a tentative understanding
of the values of bellatores or for cautious deductions regarding combat
techniques of the time.10 Despite negative assessments in historiography, it
is not fair to conclude that the tenth century was short of great historians.
Contemporary authors did not always present a full and precise narrative
of events; they are often wrong when dealing with dates and places and
they are thus spurned by the histoire événementielle for not providing
undisputed accuracy. Yet they are a mine of information if their works
are read in a constructive and unprejudiced way.11
Earlier authors such as Sir Charles Oman and Ferdinand Lot presented a
negative view of the material available for this period. Oman was a prolific
military historian, but a generalist with a shallow training as a medieval-
ist.12 Lot’s work is more surprising, because he was one of the great French
medieval historians of pre-Annales times and an expert on the early Middle
Ages, to which he devoted most of his research. Yet his treatment of the
‘long tenth century’ is confined to seven pages. He considered that there
was a historiographical void in the period and that sources could not be
fully relied on because of their deficient information; his severe criticism
of the narrative of Richer of Saint-Remi on the decisive battle of Soissons
9
Almost all historians of literature admit that epic poetry depicts the mentality
of warriors better than the roman courtois: Matthias Waltz, ‘Spontanéité
et responsabilité: De la Chanson de Roland à Raoul de Cambrai’, Société
Roncesvals: IVe Congrès International (Heidelberg, 1964), pp. 194–200 (here
194); William Calin, The Old French Epic of Revolt: Raoul de Cambrai, Renaut
de Montauban, Gormond et Isembart (Genève, 1962), pp. 113, 126, 130; Erich
Köhler, ‘Quelques observations d’ordre historico-sociologique sur les rapports
entre la chanson de geste et le roman courtais’, in Chanson de Geste und höfischer
Roman (Heidelberg, 1963), pp. 21–36. Köhler indicates that epic poetry shows
an ideological identity between author and public while there was a great rift
in the romans between the ideal of adventure and popular mentality.
10
John F. Benton, ‘Nostre Franceis n’unt talent de fuir: The Song of Roland and
the Enculturation of a Warrior Class’, Olifant 6 (1979), 237–49; B. Gitton, ‘De
l’emploi des chansons de geste pour entraîner les guerriers au combat’, in Le
Chanson de geste et le mythe carolingien: Mélanges René Louis, 2 vols (Saint-
Père-sur-Vézelay, 1982), 1: 3–19.
11
Timothy Reuter, ‘Introduction: Reading the Tenth Century’, in The New
Cambridge Medieval History, 3, ed. Reuter (Cambridge, 1999), pp. 1–24 (here
1–9).
12
Charles Oman, A History of the Art of War in the Middle Ages, 2nd edn, 2 vols
(London, 1924), 1: 101–48.
54
in 923 is clear evidence of his opinion on this point.13 Another renowned

specialist, J. F. Verbruggen, also covers the period very superficially because
of the lack of ‘good sources in greater numbers’.14
There are other problems related to written sources. On occasion, rea-
sonable doubts arise about whether they are entirely original when describ-
ing events or are influenced by classical authors or the Bible, especially the
Old Testament.15 Some events, whether great or small, may receive great
attention in the sources while others are ignored.16 Exceptional circum-
stances are thus important aids for the historian. In the eleventh century,
as in later periods, the most complex military operations were sieges or
blockades designed to capture important strongholds, and the unusual
character of these events attracted the attention of chroniclers to such
an extent that they could inspire entire works or concentrate the focus
of narratives on them.17 The blockade and capture of Antioch (1097–98)
during the First Crusade is probably the most relevant, but not the only
13
Ferdinand Lot, L’art militaire et les armées au Moyen Age en Europe et dans le
Proche Orient, 2 vols (Paris, 1946), 1: 119–23.
14
J. F. Verbruggen, The Art of Warfare in Western Europe during the Middle Ages:
From the Eighth Century to 1340, trans. Sumner Willard and R. W. Southern,
2nd edn (Woodbridge, 1997), p. 11.
15
See Richard Abels and Stephen Morillo, ‘A Lying Legacy? A Preliminary
Discussion of Images of Antiquity and Altered Reality in Medieval Military
History’, Journal of Medieval Military History 3 (2005), 1–13, and the reply
by Bernard S. Bachrach, ‘“A Lying Legacy” Revisited: The Abels-Morillo
Defense of Discontinuity’, Journal of Medieval Military History 5 (2007), 153–
93. Peter Heather’s very graphic comment on the historiography of the Late
Roman empire may be fully applied to the case under discussion: ‘This type of
interpretation requires historians to treat authors, not as sources of fact, but
rather like second-hand-car salesmen whom they would do well to approach
with a healthy caution’, Heather, The Fall of the Roman Empire: A New History
of Rome and the Barbarians, 2nd edn (Oxford, 2007), p. xiii.
16
Halsall has discussed the invisibility of most campaigns in a generalisation
about a wider period in the Middle Ages: ‘Most early medieval campaigning
was probably small scale and aimed at the acquisition of booty’, but ‘the
majority of such warfare escapes notice’ (Warfare and Society in the Barbarian
West, p. 156).
17
That was the case with the Viking attack on Paris: Abbo of Saint-Germain-des-
Prés, Viking Attacks on Paris: The Bella Parisiacae urbis, ed. and trans. Nirmal
Dass (Leuven, 2007). And in the twelfth century, The Conquest of Lisbon: De
expugnatione Lyxbonensi, ed. and trans. Charles W. David (New York, 1936),
and ‘The Poem of Almería’, in The World of El Cid: Chronicles of the Spanish
55
Manuel Rojas
case; siege operations in the Norman conquest in Italy and the capture
by the Muslims of the Castilian advanced position at Aledo in south-
eastern Iberia (1088) are evidence of this type of military activity in the
western Mediterranean.18 For similar reasons, an important pitched battle
could bring forth ample and diverse information, but the more frequent
smaller-scale engagements rarely did so.19 The most famous case in the
period under discussion is the clash at Hastings: ‘More is known about
Hastings than any battle fought in the West since the end of the Roman
empire, and any event in English history prior to 1066 and for another
century thereafter’.20
It was common for one side to produce a detailed but subjective nar-
rative of an engagement while its opponents remained largely or com-
pletely silent, as was the case with the battle of Sagrajas or Zalaca (1086).
Contemporary and later Christian sources barely referred to the severe
defeat of King Alfonso VI of León-Castile at the hands of the Almoravids
and their allies in al-Andalus, while Muslim sources on the event are
abundant and precise.21 Sources reflect different military aspects in a
more or less detailed way, as can be seen with early castles. Despite their
proliferation, there are very few iconographic representations of them –
although some are of great quality, the Bayeux Tapestry being the obvious
example –,22 while descriptions in annals, chronicles and other written
sources are normally very poor.23 This scarcity means that particular
attention has been given to passages related to this theme in the works of
Lambert of Ardres, Walter the Archdeacon and Lawrence of Durham, the
last one written in florid and obscure Latin.24 However, earlier historians
Reconquest, ed. and trans. Simon Barton and Richard Fletcher (Manchester,
2000), pp. 250–63.
18
Randall Rogers, Latin Siege Warfare in the Twelfth Century (Oxford, 1992),
pp. 91–105, and Ambrosio Huici Miranda, Las grandes batallas de la Reconquista
durante las invasiones africanas (almorávides, almohades y benimerines) (Madrid,
1956), pp. 85–99, and sources cited in both works.
19
Georges Duby, Le Dimanche de Bouvines, 27 juillet 1214, 2nd edn (Paris, 2005),
pp. 8–9.
20
M. K. Lawson, The Battle of Hastings, 1066 (Stroud, 2002), p. 45.
21
Huici, Las grandes batallas de la Reconquista, pp. 48–75, 77–80.
22
Robert Higham and Philip Barker, Timber Castles (London, 1992), pp. 147–71.
23
Higham and Barker, Timber Castles, pp. 114–46.
24
Lambert of Ardres, The History of the Counts of Guines and Lords of Ardres, trans.
Leah Shopkow (Philadephia, 2001), pp. 141, 160–61; Walter the Archdeacon,
‘Vita Iohannis episcopi Teruanensis auct. Waltero’, ed. O. Holder-Egger, in
MGH SS 15/2 (Hannover, 1888), xcviii, pp. 1136–50 (here xii, pp. 1146–47);
56
were not always keen to deal with the material that greatly interests
present-day scholars. Authors in the tenth and eleventh centuries did
not match the liveliness of later vernacular sources when touching on
‘the thoughts and actions of soldiers’.25 They were normally concise and
aseptic to an extreme when explaining and describing the circumstances
of military engagements. There is a generalised assumption that literati
(basically clerics) were totally or partially ignorant in matters related to
war.26 There were exceptions, such as William of Poitiers and Fulk IV
le Réchin. The former had been a warrior, as Orderic Vitalis relates:
‘before he entered the church he had himself been keenly involved in the
business of war. He had borne arms in service of his prince and, having
himself lived through the dire perils of war, was all the better placed to
give an accurate description of the conflicts he had seen’.27 The latter, a
lay aristocrat, planned to write the history of his bellicose predecessors.28
For all these reasons, the study of warfare based on Latin texts which are
not always considered as fully historical is a real tour de force.29 However,
the aristocratic origins of many members of the higher clergy tends to blur
distinctions between ecclesiastics and knighthood.30 A close examination
of sources allows a perception of intersections between the secular and
the sacred, as ‘the intimate links between the martial and spiritual have
often been obscured by modern historical perspectives’.31 The monks who
wrote annals and chronicles were no timid, cloistered clerics unaware of
trans. in R. Allen Brown, English Castles (London, 1976), pp. 60–61; Lawrence
of Durham, Dialogi Laurentii Dunelmensis Monachi ac Prioris, ed. James Raine
(London, 1880), pp. 8–15.
25
Gillingham, ‘Richard I’, p. 79.
26
John H. Beeler, Warfare in Feudal Europe, 730–1200 (Ithaca, 1971), p. xi;
Bernard S. Bachrach, ‘Medieval Siege Warfare: A Reconnaissance’, Journal of
Military History 58 (1994), 119–33, (here 121).
27
Orderic Vitalis, The Ecclesiastical History, ed. and trans. Marjorie Chibnall, 6
vols (Oxford, 1969–80), 2: 258–59. Translation from Gillingham, ‘William the
Bastard at War’, p. 143.
28
‘Fragmentum historiae Andagavensis’, in Chroniques des comtes d’Anjou et des
seigneurs d’Amboise, ed. Louis Halphen and René Poupardin (Paris, 1913), pp.
232–38.
29
See for example, Bernard S. Bachrach, ‘Dudo of Saint Quentin as an Historian
of Military Organisation’, Haskins Society Journal 12 (2002), 165–85.
30
Matthew Strickland, War and Chivalry: The Conduct and Perception of War in
England and Normandy, 1066–1217 (Cambridge, 1996), p. 73.
31
Abigail Wheatley, The Idea of the Castle in Medieval England (York, 2004),
p. 89.
57
Manuel Rojas
the realities of the outside world, and war was a pursuit which involved
some who had professed in monasteries.32 Well known prelates such as
Odo of Bayeux, Geoffrey of Coutances, Henry of Blois and Philip of
Beauvais bore arms, in contravention of canon law.33 There were many oth-
ers who acted in the same way. The Cantar de Mio Cid provides a stirring
account of Jerome of Périgord’s arrival in Valencia ‘longing to encounter
Moors on the field of battle’.34 Many more, as a result of landholdings or
political status, were builders or custodians of key castles. Bishop Gundulf
of Rochester, who was William the Conqueror’s master of works at the
White Tower, received from William Rufus the commission to rebuild
Rochester castle in stone.35
Any inexperience of literati in relation to military affairs should not
be stressed as much as the way they chose to describe activities in war,
strategies and tactics, as well as all the details that give an approximate
idea of how a battle, siege or punitive raid was conducted. Men who con-
trolled the art of writing were not always willing, or were even reluctant,
to describe in detail the eventus bellorum.36 Sometimes there is a sharper
concentration on events from the perspective of an opponent. Without
information from Muslim sources, specifically in the Tibyān, the ‘mem-
oirs’ of the last Zīrid king of Granada, it would be difficult to reconstruct
accurately the expansionist strategies of Fernando I of León-Castile and
his son Alfonso VI against the taifas (petty kingdoms), which combined a
32
Katherine Allen Smith, War and the Making of Medieval Monastic Culture
(Woodbridge, 2011), p. 4.
33
Strickland, War and Chivalry, pp. 47, 73 n. 88.
34
The Poem of My Cid, trans. Peter Such and John Hodgkinson (Warminster,
1987), line 1292. See Simon Barton, ‘El Cid, Cluny and the Medieval Spanish
Reconquista’, English Historical Review 126 (2011), 517–43.
35
Marylou Ruud, ‘Monks in the World: The Case of Gundulf of Rochester’,
Anglo-Norman Studies XI, ed. R. Allen Brown (Woodbridge, 1989), 245–60
(here 248).
36
John Gillingham, ‘Fontenoy and After: Pursuing Enemies to Death in France
between the Ninth and Eleventh Centuries’, in Frankland: The Franks and the
World of the Early Middle Ages. Essays in Honour of Dame Jinty Nelson, ed. Paul
Fouracre and David Ganz (Manchester, 2008), pp. 242–65, (here 245), lists
twenty-eight engagements that occurred in West Francia and the Rhineland
between 841 and 1068, ‘a few so obscure that they can be neither precisely dated
nor located’. Between 994 and 1047, however, Gillingham points out that kings
‘were directly involved in only two out of fifteen [battles]’. Could the absence
of kings be one the reasons for the slight attention paid to such combats by
authors writing what they believed their lords would like to hear?
58
systematic policy of exhaustion of their resources through parias (tribute

payments) with the continuous threat of military intervention.37 One pas-
sage in the Tibyān deserves to be included in any anthology on strategy,
in connection with the strategy of indirect approach:38
‘This is a business,’ – [Alfonso VI of León-Castile] said to himself – ‘out
of which I must make the most, even if the city [Granada] is not taken, for
what shall I get out of taking it away from the hands of these ones to give
it to those other ones but giving the latter reinforcements against myself ?
The more agitators there are and the more rivalry arises among them, all
the better it is for me’. Therefore he decided to obtain money from both
sides [the taifa kingdoms of Granada and Seville] and make one of the
adversaries clash against the other without any intention of obtaining ter-
ritories for himself. ‘I do not believe in their religion’ – he said to himself
making his calculations – ‘and they all hate me. What reason do I have to
take Granada? Its subjugation without fighting is impossible and, if I have
to conquer it by war, considering the number of men that would die and
the amount of money I would spend, the losses would be greater than the
benefits derived from the military action, and that in the case I succeeded
in winning it. On the other hand, should I win it, I would not be able to
keep it unless I could count on the loyalty of the inhabitants, who would
not give it to me, and it would also be out of the question that I should kill
all the inhabitants to repopulate the city with people of my own religion.
Therefore, there is no other possible line of conduct than to sow discord
37
Angus MacKay, Spain in the Middle Ages: From Frontier to Empire, 1000–1500
(London, 1977), p. 15, refers to this system as a ‘protection racket’. The best
guides to the tribute regime imposed by the Christian principalities on the
Muslim petty kingdoms are still José María Lacarra, ‘Aspectos económicos
de la sumisión de los reinos de taifas (1010–1102)’, in Lacarra, Obra Dispersa,
1961–1971 (Pamplona, 2010), pp. 219–49, and Hilda Grassotti, ‘Para la historia
del botín y de las parias en Castilla y León’, Cuadernos de Historia de España
39–40 (1964), 43–132.
38
This coinage by Sir Basil Liddell Hart may be applied to much of medieval
warfare: The Strategy of Indirect Approach, 2nd edn (London, 1946), pp. 184–
216. It was a strategy of progressive weakening of the enemy as a means to
guarantee its final defeat by the repeated gaining of small victories, but without
fighting direct combats such as pitched battles. At the level of operational
categories this method is called the relational manoeuvre; see Edward N.
Luttwak, The Logic of War and Peace, 2nd edn (Cambridge, MA, 2001), p. 115.
For the Iberian context, see Manuel Rojas, ‘A Decisive Century in the Struggle
against Islam in Iberia, ca. 1031–1157: Grand Strategy Perspectives’, in The
Medieval Way of War: Studies in Medieval Military History in Honor of Bernard
S. Bachrach (forthcoming) and works cited therein.
59
Manuel Rojas
among the Muslim princes and get money from them continuously, so
that they exhaust their resources and become weaker. When that happens,
Granada, unable to resist, will spontaneously surrender to me and will be
willingly subjugated; in the same way it is happening with Toledo which,
because of the misery and the fragmentation of the population and the
fleeing of its king, is coming into my hands without any effort’.39
The specific characteristics of the practice of war in the Celtic lands
and late Anglo-Saxon England, the volatile territorial principalities of
north-western France or the Christian polities of the Iberian Peninsula
(to give only four examples) were diverse. Differences in the conduct
of war reflected the idiosyncrasies of distinct societies and the diverse
circumstances they faced, as well as the institutional and administrative
frameworks they had developed to win their armed conflicts or at least to
restrict the ambitions of their rivals. The question at issue is to ascertain
whether, in all Western armed conflicts within the same period, the con-
duct of warfare aimed to achieve similar goals through equivalent opera-
tional strategies and methods, although technological levels may have
been different and tactics more or less refined. A circumstance common
to all situations was low-intensity warfare, an activity which dated from
long before the eleventh century.40 Large or small punitive raids, whose
goal was the seizure of livestock and destruction of material resources,
had different military and legal characteristics in Ireland, Iberia or France,
and the technology used in them was also diverse. Yet, despite such vari-
ation, they amounted to the same hit and run tactics when viewed from
the perspective of ‘the normal business of war’.41
39
The Tibyān: Memoirs of ‘Abd Allāh B. Buluggīn, Last Zīrid Amīr of Granada,
trans. Amin T. Tibi (Leiden, 1986) is a generally accurate translation of this
remarkable source, but it does not achieve the precision and quality of the
Spanish version by Évariste Lévi-Provençal and Emilio García Gómez, El siglo
XI en 1ª persona: Las ‘Memorias’ de ‘Abd ‘Allāh, último rey Ziri de Granada,
destronado por los Almorávides (1090), 2nd edn (Madrid, 2010), pp. 182–83. The
translation given here is my own, based on the Spanish version.
40
Although some of their proposals have been criticised by authors with
‘normalist’ inclinations (see below); see Georges Duby, The Early Growth of the
European Economy: War and Peasants from the Seventh to the Twelfth Century,
trans. Howard B. Clarke (Ithaca, 1974), pp. 48–50; Timothy Reuter, ‘Plunder
and Tribute in the Carolingian Empire’, and ‘The End of Carolingian Military
Expansion’, reprinted in Medieval Polities and Modern Mentalities, ed. Janet L.
Nelson (Cambridge, 2006), pp. 231–50 and 251–67.
41
See, for Ireland, Katherine Simms, ‘Images of Warfare in Bardic Poetry’, Celtica
21 (1990), 608–19 and T. M. Charles-Edwards, ‘Irish Warfare before 1100’, in
60
In connection with conquest and ‘colonisation’ we must ask whether,

sooner or later, the conduct of warfare became similar everywhere due to
the fact that superior military techniques spread from the centre to the
peripheries of western Europe.42 This extension was caused by a process of
osmosis or the introduction into conquered territories of external military
and social factors which had already shown their efficacy as instruments of
aggression and expansion.43 It can be argued that if conquerors finally won,
that was not a matter of luck but of their superior war technology which
allowed them to develop more efficient operational strategies and tactics
than those whom they attacked. One might also consider whether some
less sophisticated Western societies were able to find similar operational
A Military History of Ireland, ed. T. Bartlett and K. Jeffery (Cambridge, 1996),

pp. 26–51; for Iberia, Francisco García Fitz, Castilla y León frente al Islam:
Estrategias de expansión y tácticas militares (siglos XI-XIII) (Sevilla, 1998), pp.
59–170; for France, Dominique Barthélemy, ‘Feudal War in Tenth-Century
France’, in Vengeance in the Middle Ages: Emotion, Religion and Feud, ed. S. S.
Throop and P. R. Hyams (Farnham, 2010), pp. 105–14, where he synthesised
some of the arguments already developed in Caballeros y milagros: Violencia y
sacralidad en la sociedad feudal, trans. F. Miranda García (Granada, 2005), pp.
19–31. But see John France, ‘La Guerre dans la France féodale à la fin du ixe
et au xe siècle’, Revue Belge d’Histoire Militaire 23 (1979), 177–98, for a more
military and less feud-type view.
42
‘Colonisation’ is a concept difficult to transfer to medieval times. A brief
discussion is given in Brendan Smith, Colonisation and Conquest in Medieval
Ireland: The English in Louth, 1170–1330 (Cambridge, 1999), pp. 1–9. For the
Latin East, see ‘The Crusading Kingdom of Jerusalem –The First European
Colonial Society? A Symposium’, in The Horns of Hattin, ed. Benjamin Z.
Kedar ( Jerusalem, 1992), pp. 341–66. Centre and periphery are geographical
but also historical and mental constructs which depend on the criteria
determining which place is regarded as ‘centre’. If a citizen of Córdoba, Baghdad
or Constantinople at the beginning of the eleventh century were told that he
lived in the ‘periphery’ of Europe, he would be deeply surprised. See Robert
Bartlett, ‘Heartland and Border: The Mental and Physical Geography of
Medieval Europe’, in Power and Identity in the Middle Ages: Essays in Memory
of Rees Davies, ed. H. Pryce and J. Watts (Oxford, 2007), pp. 23–36.
43
The victories of the Normans in southern Italy were to a great extent due to
their superior technological and tactical abilities. See Errico Cuozzo, ‘Quei
Maledetti Normanni’: Cavalieri e organizzazione militare nel Mezzogiorno
normanno (Napoli, 1989); Charles D. Stanton, ‘The Battle of Civitate: A
Plausible Account’, Journal of Medieval Military History 11 (2013), 25–56. I
would like to thank Dr Stanton for sending me a copy of his article.
61
Manuel Rojas
solutions of their own and employ them in their internal conflicts or

against attempts at conquest from outside.44 These could be regarded as
the ‘fearful symmetries’ which Ehrenreich referred to following William
Blake’s well known poem, or conversely, as asymmetries between neigh-
bouring but distinct societies in conflict.45
In relation to these questions, Robert Bartlett has proposed an interest-
ing explicative approach. He considers that an advanced military technol-
ogy was disseminated in conjunction with an aristocratic diaspora to the
Celtic and Slavic peripheries of Europe from a ‘central area’ comprising
the German and Frankish heartlands, and from England after 1066. The
three main technological components of this process were heavy cavalry,
missile weapons (in particular, the crossbow), and castles with associated
siege techniques. Superior equipment, fighting techniques and fortifications
were assimilated by peripheral cultures not only as a result of conquest but
also by imitation and cultural emulation.46 Matthew Strickland considers
that Bartlett’s thesis may be applicable with modifications to the Celtic
lands and those beyond the German Ostmark, but less so to England before
44
Irish crannogs (artificial islands constructed on lakes by private hands) and
other strongholds were not strictly castles but they could serve similar purposes.
This appreciation can be extended to the Gaelic worlds of medieval Scotland
or Wales. See, for example, Tom E. McNeill, Castles in Ireland: Feudal Power
in a Gaelic World (London, 2000), pp. 9–10; Nancy Edwards, The Archaeology
of Early Medieval Ireland (Abingdon, 1990), pp. 34–41. Although this is a
different question altogether, neither the husun nor other Muslim strongholds
in Iberia were castles in a strict sense. See Valerie Dalliere-Benelhadj, ‘Le
“château” en al-Andalus: Un problème de terminologie’, in Habitats fortifiés et
organisation de l’espace en Méditerranée médiévale, ed. André Bazzana, Pierre
Guichard and J. M. Poisson (Lyon, 1983), pp. 63–67; André Bazzana, Patrice
Cressier and Pierre Guichard, Les Châteaux ruraux d’al-Andalus: Histoire et
archéologie des husun du sud-est de l’Espagne (Madrid, 1988).
45
Barbara Ehrenreich, Blood Rites: Origins and History of Passions of War (New
York, 1997), pp. 132–43. An excellent analysis of the socio-economic conditions
which allowed northern Christian societies to impose their rule over al-Andalus
before and during the African ‘dictatorships’ (Almoravids and Almohads) may
be found in José María Mínguez, ‘Sociedad feudal, guerra feudal’, in La guerra
en la Edad Media, ed. José Ignacio de la Iglesia Duarte (Logroño, 2007), pp.
17–48. See also Felipe Maíllo Salgado, De la desaparición de al-Andalus, 2nd
edn (Madrid, 2011), pp. 67–95.
46
Robert Bartlett, The Making of Europe: Conquest, Colonization and Cultural
Change, 950–1350, 2nd edn (London, 1994), pp. 60–84.
62
1066.47 The same conclusion may be applied to Christian polities in the

Iberian Peninsula.48 Bartlett’s brilliant thesis works for some geo-political
zones, but not completely for all of them. I am fully aware that this is a
controversial issue or even an obsolete question, especially for many anglo-
phone scholars who have abandoned the concept of feudalism.49 It can be
argued that this conceptual iconoclasm leads nowhere. If it is accepted
that medieval societies had economic foundations based on agriculture
and livestock, then control of lands and men were substantial elements of
power, and competition between polities to increase these or prevent others
achieving the same end was the foundation of effective social, economic and
political power. If there is agreement on these points, it is not easy to explain
according to medieval historical frameworks what incited certain societies
to attack others, or why there was a continuous struggle between kingdoms
or polities within the same socio-cultural milieu if explanations based on
the outdated ‘feudal aggression’ are done away with. Tentative responses
of an anthropological and ethnic, mental and doctrinal (not ideological)
or institutional and juridical character may be useful to understand why
disputes, aggression and expansive thrusts emerged, but they do not fully
explain by themselves the deepest reasons for hostilities and conquests.
An important and scarcely discussed question which has to be taken
into account is what Georges Duby called temporalité différentielle.50 This
refers to the changes in the conduct of warfare over time caused by the
appearance of new factors or by the modification of existing elements.
One evident case is the emergence, consolidation and proliferation of
castles as military tools and the high level of innovation and experi-
mentation in different forms of defence-works – most obviously walled
towns – which changed the nature of conflict on both defensive and
offensive levels. There is no longer support for the ‘romantic’ view which
claimed that the origin of castles was connected with the attacks carried
out by Norsemen, Saracens and Hungarians, at a time when individual
47
Matthew Strickland, ‘Military Technology and Conquest: The Anomaly
of Anglo-Saxon England’, in Anglo-Norman Studies, XIX, ed. Christopher
Harper-Bill (Woodbridge, 1997), 353–82.
48
Rojas, ‘A Decisive Century’.
49
The intellectual position of these scholars causes great perplexity in non-
Anglo-Saxon academic milieus where the concept of ‘feudalism’ causes neither
rejection nor problems. Susan Reynolds has published a compilation of her
articles which has a revealing overall title: The Middle Ages without Feudalism:
Essays in Criticism and Comparison on the Medieval West (Aldershot, 2012).
50
Review of Arno Borst, Lebensformen im Mittelalter, in Francia 3 (1975), 742–
45 (here p. 745).
63
Manuel Rojas
heroes full of good intentions speckled lands with strongholds with the
purpose of safeguarding the unarmed majority of the population, amid
general tumult and disorder.51 The same applies to ideas which related
the emergence of castles to the rise of aristocracies and social disorder
– what earlier institutionalist authors called the continuous disturbing
presence of ‘feudal anarchy’.52 In general terms, the most probable option
is that the rise and proliferation of castles were a symptom of a new order
which meant ‘the emergence of a “new aristocracy” with the means and
motivation to defend their homes and estates, compete with their peers
and intimidate communities’.53
Although it is true that the military dimension was only one of the many
functions of castles, the encastellation of western Europe during the tenth
and especially the eleventh century meant a notable alteration of political
strategies, organisation and routine of campaigns and military operations.
In spite of trenchant attacks against ‘military fundamentalism’,54 revisionist
authors cannot shut their eyes to ‘the military role of castles’, even though
51
As early as 1938, Roger Aubenas doubted that the rapid increase in the number
of strongholds had such a spontaneous cause, ‘Les Châteaux forts des xe et xie
siècles: Contribution à l’étude des origines de la féodalité’, Revue historique de
droit français et étranger ser. 4, 17 (1938), 548–86.
52
From an archaeological perspective, private defence had a long ancestry. For
example, private fortification was absent in Merovingian times, a period of
endemic violence between competing nobles. The few sources available show
that Merovingian aristocrats lived in villa complexes enclosed by walls, but
this related more to symbolic definition of property rights than defence. The
typical fortifications of the period were not proto-castles, but former Roman
walled towns, military camps and prehistoric-style forts in natural defensive
positions. See Ross Samson, ‘The Merovingian Nobleman’s Home: Castle or
Villa?’, Journal of Medieval History 13 (1987), 287–315.
53
Oliver Creighton, Early European Castles: Aristocracy and Authority, AD 800–
1200 (Bristol, 2012), p. 49. See also André Debord, Aristocratie et pouvoir: Le
rôle du château dans la France médiévale (Paris, 2000). For a lavish catalogue
of castle architecture of a specific area, see Marie-Pierre Baudry, Châteaux
‘Romans’ en Poitou-Charentes, x e-xiii e siècles (La Chèche, 2011).
54
See Charles Coulson, Castles in Medieval Society: Fortresses in England, France,
and Ireland in the Central Middle Ages (Oxford, 2003). For conflicting positions
which synthetically interpret castles as either fortresses or residences, see
Colin Platt, ‘Revisionism in Castle Studies: A Caution’, Medieval Archaeology
51 (2007), 83–102, and the reply by Oliver Creighton and Robert Liddiard,
‘Fighting Yesterday’s Battle: Beyond War or Status in Castles Studies’, Medieval
Archaeology 52 (2008), 161–68. A summary view which is more favourable to
64
‘the majority of castles never saw an arrow shot in anger’.55 This is to a certain
extent a tendentious and biased conclusion, based almost solely on castles
as military architecture and taking only England as the area under study.
It is true that the majority of medieval castles and fortresses were never
involved in direct military operations and that they did not withstand
sieges. But ‘revisionist’ scholars perhaps forget a couple of important func-
tions which fortifications also carried out: offensive roles and deterrence.
Many strongholds were considered difficult to capture with the resources
available at the time or only at great cost in manpower and resources after
enduring indirect pressure. This included the building of counter-castles
(Gegenburgen) and systematic devastation of the neighbourhood, actions
which were usually made by garrisons from other enemy strongholds.56
The offensive role of a stronghold was particularly significant when it was
located in enemy territory. Aledo fell into Christian hands in 1086; it was
situated hundreds of kilometres away from any friendly stronghold, but
very close to Lorca and Murcia, two Muslim-held walled towns. Its func-
tion was not defensive; it was a base for launching predatory expeditions
into neighbouring Andalusi lands and also served as a refuge for Christian
raiders.57 However, the most perfect strategy for conquering enemy enclaves
and the lands subject to them followed a well-known method summed
up in three Ps: presence, patience and persistence. That is to say, to have
enough military power to erode the economic and material resources of
the enemy without the need to combat it directly, except at specific times
and as a simple demonstration of superiority.58
There was no linear evolution which discarded earlier responses when
new architectural alternatives were adopted. ‘Old’ and ‘new’ castles
‘revisionism’ is Sarah Speight, ‘British Castle Studies in the Late 20th and 21st
Centuries’, History Compass 2 (2004), 1–32.
55
Speight, ‘British Castle’, p. 22; Robert Liddiard, Castles in Context: Power,
Symbolism and Landscape, 1066 to 1500 (Macclesfield, 2005), pp. 70–96.
56
For some of many examples, see The Gesta Guillelmi of William of Poitiers, ed.
and trans. R. H. C. Davis and Marjorie Chibnall (Oxford, 1998), pp. 60–63;
Matthew Bennett, ‘Wace and Warfare’, Anglo-Norman Studies, XI, 37–57: ‘The
point is, that to Wace warfare was inconceivable without the castle’ (p. 41).
57
Al-Hulal al-Mawsiyya Crónica árabe de las dinastías almorávide, almohade y
benimerín, trans. Ambrosio Huici Miranda (Tetuán, 1951), p. 82; ‘Abd ‘Allāh, p.
239; Rawd al-qirtas, trans. Ambrosio Huici Miranda, 2 vols (Valencia, 1964), 1:
294–95.
58
Abd ‘Allāh, pp. 229–30. Also see Emillio García Gómez and Ramón Menéndez
Pidal, ‘El conde mozárabe Sisnando Davídiz y la política de Alfonso VI con las
taifas’, Al-Andalus 12 (1947), 27–41.
65
Manuel Rojas
coexisted in space and time while fulfilling their functions, but this does
not contradict the fact that, as time passed, this type of fortification
became more complex in architectural structure and from a poliorcetic
perspective. A great step forward came when important castles and urban
fortifications began to be made of stone and old defensive devices and
architectural practices were reemployed or new ones were introduced.
In the middle or at the end of the tenth century, with some very specific
exceptions, timber and earthwork strongholds stood at the highest level
of castle building techniques,59 and in many places cave dwellings and the
use of relief advantages of a specific site were common.60 At the end of
the eleventh century the keep, the tour maitresse and the Bergfried had
become the most characteristic signs of military architecture, although
timber constructions were not abandoned.61 Reusing fortifications from
antiquity and employing communal enclosures that had previously func-
tioned well as defensive points,62 gave way to a new type of fortress, the
59
Higham and Barker, Timber Castles; André Débord, ‘Les fortifications de terre
en Europe occidentale du xe au xiie siècles’, Archéologie Médiévale 11 (1981),
5–123; Walter Herrmann, Le Château préfabriqué au Moyen Age: Construction
et protection du pan de bois (Strasbourg, 1989).
60
As late as 924, the inhabitants of Pamplona preferred the protection of nearby
mountains and forests to the walls of the city in the event of Muslim attack:
Ibn Hayyan, Crónica del califa ‘Abdarrahman III an-Nasir entre los años 912
y 942 (al-Muqtabis V), trans. María Jesús Viguera and Federico Corriente
(Zaragoza, 1981), p. 151. For a general view see Bernabé Cabañero Subiza, ‘De
las cuevas a los primeros castillos de piedra: Algunos problemas del origen de
la castellología altomedieval en el norte peninsular’, Turiaso 4 (1985), 167–88;
Cabañero Subiza Los castillos catalanes del siglo X: Circunstancias históricas y
cuestiones arquitectónicas (Zaragoza, 1996), pp. 126–31.
61
See above, n. 60. In addition, Hermann Hinz, Motte und Donjon: Zur
Frühgeschichte der mittelalterlichen Adelsburg (Köln, 1981); André Chatelain,
Donjons Romans des Pays d’Ouest: Etude comparative sur les donjons romans
quadrangulaires de la France de l’Ouest (Paris, 1973); Derek Renn, Norman
Castles in Britain, 2nd edn (London, 1973); Jean Yver, ‘Les châteaux forts en
Normandie jusq’au milieu du XIIe siècle’, Bulletin de la Société des Antiquaires
de Normandie 53 (1955–56), 28–115; Eric Fernie, The Architecture of Norman
England (Oxford, 2000), pp. 49–82.
62
The Anglo-Saxon burhs were one of the most notable cases. See The Defence
of Wessex: The Burghal Hidage and Anglo-Saxon Fortifications, ed. David Hill
and Alexander R. Rumble (Manchester, 1996); John Baker and Stuart Brookes,
Beyond the Burghal Hidage: Anglo-Saxon Civil Defence in the Viking Age
(Leiden, 2013).
66
castle, which was less costly when limited in size, easier and quicker to
erect and smaller and taller than its predecessors.63 These and other fac-
tors related to wider social changes at the time produced an astounding
proliferation of castles,64 a new factor which was so significant that it
changed the character of hostilities: ‘it was a warfare of position’.65 Even
the most humble fortification could turn into a considerable obstacle,
difficult to take by storm or neutralise unless by compromise or accord.66
From this time, all those who attained power – many of them new men
from lineages of obscure origins –67 would, in the well-known words of
Otto of Freising on Frederick II of Swabia, ‘always haul a fortress with
him by the tail of his horse’.68 Making the greatest use of all resources at
their disposal – or using their own talents – nobles such as Fulk Nerra or
Arnau Mir de Tost,69 peoples such as the Normans,70 or emerging polities
63
Bartlett, The Making of Europe, pp. 65–67. Bernard S. Bachrach, ‘The Cost of
Castle Building: The Case of the Tower at Langeais, 992–994’, in Medieval
Castle: Romance and Reality, ed. K. Reyerson and F. Powe (Dubuque, 1984),
pp. 47–62.
64
In Cinglais (Lower Normandy), in an area of 12 × 20 kilometres, there were
three or four stone castles and around thirty earthwork redoubts. Michel Fixot,
Les Fortifications de terre et les origines féodales dans le Cinglais (Caen, 1968), pp.
30–31.
65
John France, Victory in the East: A Military History of the First Crusade
(Cambridge, 1994), p. 27.
66
Strickland, War and Chivalry, pp. 204–29.
67
Georges Duby, ‘French Genealogical Literature’, in The Chivalrous Society
(London, 1977), pp. 149–57; Régine Le Jan, ‘Continuity and Change in the
Tenth-Century Nobility’, in Nobles and Nobility in Medieval Europe: Concepts,
Origins, Transformations, ed. A. J. Duggan (Woodbridge, 2000), pp. 53–68;
Thomas N. Bisson, ‘Princely Nobility in an Age of Ambition (c. 1050–1150)’, in
Nobles and Nobility in Medieval Europe, pp. 101–13.
68
The Deeds of Frederick Barbarossa, ed. and trans. Charles Christopher Mierow,
2nd edn (New York, 2004), p. 45.
69
Bernard S. Bachrach, Fulk Nerra, the Neo-Roman Consul, 987–1040: A Political
Biography of the Angevin Count (Berkeley, 1993). Francesc Fité i Llevot and
Eduard González i Montardit, Arnau Mir de Tost: Un senyor de frontera al
segle XI (Lérida, 2010); Philippe Araguas, ‘Les châteaux d’Arnau Mir de Tost:
Formation d’un grand domaine féodal en Catalogne au milieu du xie siècle’, in
106e Congrés National des Sociétés Savantes (Perpignan, 1981), pp. 61–76.
70
Orderic Vitalis, 2: 218–19: ‘To meet the danger the king [William I] rode all the
remote parts of his kingdom and fortified strategic sites (opportuna loca) against
enemy attacks. For the fortification called castles (castella) by the Normans
67
Manuel Rojas
such as the northern kingdoms and counties of the Iberian peninsula,71

based much of their success in expansion and conquest on the efficient
use of castles and other fortifications. All of them were eager to make such
constructions because they became important assets in consolidating their
power. They learned to maximise the efficiency of strongholds in offensive
and defensive warfare, but also searched for the most productive ways to
(Galli) were scarcely known in the English provinces, and so the English – in
spite of their courage and love of fighting – could put up only a weak resistance
to their enemies […]’. See Marjorie Chibnall, ‘Orderic Vitalis on Castles’, in
Studies in Medieval History Presented to R. Allen Brown, pp. 43–56; reprinted
in Anglo-Norman Castles, ed. Robert Liddiard (Woodbridge, 2003), pp. 119–
32. But see Wheatley, The Idea of the Castle, pp. 25–26, for a revision of the
traditional interpretation of this passage. Relevant literature is vast, but see, for
example, Stuart Prior, A Few Well-Positioned Castles: The Norman Art of War
(Stroud, 2006) and Anglo-Norman Castles.
71
For surveys of the Iberian peninsula, see André Bazzana, ‘Le début du
château dans l’Espagne septentrionale’, in Château Gaillard: Etudes de
Castellogie Médiévale 16 (1994), 33–44; Philippe Araguas, ‘Le château dans
l’Espagne chrétienne autour de l’an mil’, in Guerre, pouvoir et idéologies dans
l’Espagne chrétienne aux alentours de l’an mil, ed. Th. Deswarte and Ph. Sénac
(Turnhout, 2005), pp. 67–80. For Catalonia, Bernabé Cabañero Subiza, Los
castillos catalanes; Philippe Araguas, ‘Les châteaux des marches de Catalogne
et Ribagorce (950–1100)’, Bulletin Monumental 137 (1979), 205–24; Francesc
Fité, ‘Arquitectura militar y repoblación en Catalunya (siglos VIII al XI), in III
Congreso de Arqueología Medieval Española (Oviedo, 1989), pp. 193–236; Jordi
Bolòs Masclans, ‘Fortificaciones y organización del territorio en la “Marca”
o frontera catalana durante los siglos IX-XII’, in La fortificación medieval en
la Península Ibérica (Aguilar de Campoo, 2001), pp. 101–24. On Aragón,
Francisco Íñiguez Almech, ‘Las empresas constructivas de Sancho el Mayor:
El castillo de Loarre’, Archivo Español del Arte 43 (1970), 363–73; Philippe
Araguas, ‘Le château de Loarre et les châteaux de la frontière aragonese au XIe
siècle: Leur place dans l’architecture militaire de l’Occident chrétien’, in La
Marche supérieure d’al-Andalus et l’Occident chrétien, ed. Ph. Sénac (Madrid,
1991), pp. 165–75. On Navarre: Bernabé Cabañero Subiza, ‘La defensa del reino
de Pamplona-Nájera en el siglo X: Materiales para el estudio de la evolución
de sus castillos’, in La Marche supérieure d’al-Andalus, pp. 99–114. On the
origins of the name of Castile from the word castellum: Gonzalo Martínez
Díez, El condado de Castilla (711–1038): La historia frente a la leyenda, 2 vols
(Valladolid, 2005), 1: 132–37. On Leonese fortifications: José Avelino Gutiérrez
González, Fortificaciones y feudalismo en el origen y formación del reino leonés
(siglos IX-XIII) (Valladolid, 1995).
68
take them by storm.72 The military careers of the most competent and
successful commanders among rulers of the period show that they spent
most of their time organising captures of opponents’ strongpoints in wars
of attrition in which raids and counter-raids were the norm and battles
only a much avoided final recourse. As stated earlier, the appearance of
castles must be related to a new order which produced a transformation
in the conduct of warfare. Within major polities, factional castellans were
thorns in the sides of their lords, who found it difficult to control them
fully. This was the case of Hugh of Abbeville (956–1000); according to
Hariulf of Saint-Riquier, ‘defended by his castle, he did what he wanted
without fear’.73 What castellans wanted was to live as best they could
from those who were in no position to refuse their demands, and as far
as possible to hold operational bases for their activities.
At this point it is relevant to raise an issue which reflects antagonistic
positions about the relationship between written sources and the every-
day practice of warfare. Discussion has focused on whether it is right to
assume that the commanders of armies, who came from the upper classes
of medieval society, were more or less literate and thus could have had
access to treatises with information applicable to the conduct of warfare.
Apart from some minor authors, the two main works in this context
were Frontinus’s Strategemata and, above all, Vegetius’s De Re Militari.
Both treatises were repeatedly copied and in some cases annotated and
augmented by those who consulted them.74 From the time when Henri
Delpech saw Vegetian influence everywhere – even to the point that one
might assume from his arguments that no medieval general could have
planned any strategy or applied any tactical manoeuvre without having a
deep knowledge of De Re Militari –,75 scholarly opinion has been divided
on the practical application of Vegetian principles to military practice in
72
Besides Rogers, Latin Siege Warfare, the standard works for the period are Jim
Bradbury, The Medieval Siege (Woodbridge, 1992), and Peter Purton, A History
of the Early Medieval Siege, c. 450–1200 (Woodbridge, 2009).
73
Chronique de l’abbaye de Saint-Riquier (v e siècle-1104), ed. Ferdinand Lot
(Paris, 1894), p. 163.
74
Christopher Allmand, ‘A Roman Text of War: The Strategemata of Frontinus
in the Middle Ages’, in Soldiers, Nobles and Gentlemen: Essays in Honour
of Maurice Keen, ed. Peter Coss and Christopher Tyerman (Woodbridge,
2009), pp. 153–68. Philippe Richardot, Végèce et la culture militaire au Moyen
Age (ve-xve siècles) (Paris, 1998); Allmand, The De Re Militari of Vegetius:
The Reception, Transmission and Legacy of a Roman Text in the Middle Ages
(Cambridge, 2011).
75
Henri Delpech, La tactique au xiiie siècle, 2 vols (Paris, 1886).
69
Manuel Rojas
western Europe. The numerous manuscripts which have survived, many of

them translated into the vernacular, support the view that De Re Militari
was a hidden but very influential presence in the conduct of warfare.
Other scholars, while accepting that Vegetius was the military auctoritas
par excellence, consider that his influence is difficult to determine or
even that his popularity was ‘mainly a matter of antiquarianism and the
desire to acquire classical learning’,76 or that correspondences between
this ancient work and the realities of war actually resulted from a series
of ‘eternal common-sense principles’.77
The use of De Re Militari by Geoffrey Plantagenet during the siege of
the stronghold of Montreuil-Bellay in 1150 has often been advanced as
solid proof of Vegetius’s theory being put in practice. However, perhaps
insufficient attention has been paid to the fact that the count, although
he ‘was literate [and] took counsel by Vegetius’, relied on monks of the
nearby abbey of Marmoutier who were at the siege to understand the
text he was trying to read.78 I stress this point because Bernard Bachrach
has been at pains to prove that there was a profitable Angevin military
tradition, especially on the part of Fulk Nerra, which would explain
why these nobles, due to Vegetian influence, were successful in warfare
at that time: ‘Fulk’s strategy and indeed also his tactics were informed
by Roman military science, most likely Vegetian in inspiration, which
he modified according to his needs’.79 The main issue here is once again
whether the work of Vegetius could exert real influence on the conduct of
warfare because military commanders had enough education to consult
it, learn from its contents and put its ideas into practice on the field of
battle.80 If this had been the case, it would have affected the highest strata
76
Halsall, Warfare and Society in the Barbarian West, p. 145.
77
Smail, Crusading Warfare, p. 15 n. 2.
78
‘Historia Gaufredi ducis Normannorum et comitis Andegavorum’, in
Chroniques des comtes d’Anjou, p. 218.
79
Bachrach, Fulk Nerra, p. xi. See also Bachrach, ‘The Practical Use of Vegetius’
De Re Militari during the Early Middle Ages”, The Historian 47 (1985), 239–
55, who presents an alternative reading of sources: either the authors drew
attention to the influence of Vegetian principles on the behaviour of military
commanders when recording events, or they stressed coincidences between
the conduct of war and late Roman military science or the work of Vegetius
which did not imply any formal knowledge of those historical precedents by
war leaders.
80
Bernard S. Bachrach, ‘Writing Latin History for a Lay Audience c. 1000: Dudo
of Saint Quentin at the Norman Court’, Haskins Society Journal 20 (2008),
58–77, and works cited therein.
70
of secular society and not the lower or middle nobility. The maxim of
Kekaumenos, the Byzantine military strategist of the eleventh c entury,
is difficult to apply to the Western nobility: ‘When you have free time,
and are not occupied with military affairs, read strategic works and
books, histories and the books of the Church’.81 At this time, the high-
est offices were sometimes occupied by people who were unable to read
Latin: Roger le Poer, one of those new men whom King Henry I of
England had ‘raised from the dust’,82 first as chancellor and then as chief
justiciar, was described as ‘almost illiterate (fere illiteratus)’ by William
of Newburgh.83 There is a need for research on this topic to remove once
and for all the unacceptable situation which has been expressed thus:
‘there are still historians who sneer at medieval commanders as if they
were fools and idiots, but in all areas of war commanders, then as since,
were quite capable of carefully weighing up the position and coming to
sensible and practical decisions’.84
Many of the points touched on so far show two almost irreconcilable
historiographical stances on how the practices of war reflected the s ocieties
which employed them:
We can broadly distinguish two camps in the debate, simplified as
minimalist-exceptionalists and maximalist-continuists. Minimalists and
exceptionalists believe in little continuity from Late Antiquity or cor-
relation with Byzantine and Islamic military organisation. Maximalists
81
Consejos de un aristócrata bizantino, trans. J. Signes Codoñer (Madrid, 2000),
p. 53. The genre of military treatises continued its development in Byzantium,
and it is probable that emperors and generals transferred their precepts to real
combat: see John Haldon, Warfare, State and Society in the Byzantine World,
565–1204 (London, 1999), pp. 200–1; Eric McGeer, Sowing the Dragon’s Teeth:
Byzantine Warfare in the Tenth Century, 2nd edn (Washington, DC, 2008),
pp. 253–328; Márton Tösér, ‘Byzantine Expansion in the 10th Century: Siege
Techniques in Theory and Practice’, Hadtörténelmi Közlemények 119 (2006),
459–82. I have had no access to Taxiarchis Kolias, ‘Byzantine Military Tactics:
Theory and Practice’, in Byzantium at War, ed. N. Oikonomidès (Athens,
1997), pp. 153–64.
82
Richard W. Southern, Medieval Humanism and Other Studies (Oxford, 1970),
p. 212.
83
‘Historia rerum Anglicarum’, in Chronicles and Memorials of the Reigns of
Stephen, Henry II and Richard I, ed. R. Howlett, 4 vols (London, 1884), 1: 36.
84
Bradbury, Medieval Siege, p. 80.
71
Manuel Rojas
and continuity-proponents believe in a high degree of continuity and

large-scale operations.85
What is most germane here is the fact that both positions normally
relate to historians who can be labelled as ‘substantivists’ and ‘normalists’
respectively.86 The latter have been usually guided by the idea of a series
of active guidelines related to human nature which consequently were
timeless. Thus there was a set of general principles in warfare, tactics
and strategy which have been constant and unchanging throughout
history. Signs of this standpoint – only two examples are needed – are
searches for ‘systems’, ‘networks’ or ‘nets’ of castles which were con-
sciously designed,87 or sophisticated administrative and logistic abilities
for making war. ‘Normalist’ scholars feel a great desire to reconstruct
the course of a campaign or military action according to those eternal
principles of strategic and tactical conduct, even subscribing to the risky
method which Lieutenant-Colonel Alfred H. Burne called ‘Inherent
Military Probability’ (IMP).88 This view has received many criticisms,89
85
Leif Inge Ree Petersen, Siege Warfare and Military Organization in the Successor
States (400–800 AD): Byzantium, the West and Islam (Leiden, 2013), pp. 235–45
for a complete discussion of both proposals, although the author clearly sides
with the ‘continuists’. See also David Bachrach, Warfare in Tenth-Century
Germany (Woodbridge, 2012), pp. 1–5. The debate has become bitter on
some occasions. See the review by Guy Halsall of Bernard S. Bachrach, Early
Carolingian Warfare: Prelude to Empire (Philadelphia, 2001), in Peritia 16
(2002), pp. 483–88; and the review by Bernard S. Bachrach of Halsall, Warfare
and Society in the Barbarian West in American Historical Review 109 (2004), 959.
86
This is the distinction made by Halsall in Warfare and Society in the Barbarian
West, pp. 6–8.
87
Manuel Rojas, ‘Labrando fortalezas: Planteamientos sobre el castillo como
núcleo de estrategias en el escenario bélico anglo-francés (c. 1066–c. 1216)’, in
Os Reinos Ibéricos na Idade Média: Livro de Homenagem ao Professor Doutor
Humberto Carlos Baquero Moreno, ed. L. Adão da Fonseca, L. C. Amaral and
M. F. Ferreira Santos, 2 vols (Lisboa, 2003), 2: 867–80 (here 878–79); María
Dolores García Oliva, ‘Los sistemas defensivos cristianos y musulmanes (siglos
XII-XV) en la historiografía hispana de las últimas décadas’, in The Conduct of
Warfare in the Middle Ages, 1: Historiography, ed. Manuel Rojas (forthcoming).
88
Alfred H. Burne, The Crécy War: A Military History of the Hundred Years
War from 1337 to the Peace of Bretigny, 1360 (London, 1955), p. 12; Burne, The
Agincourt War: A Military History of the Latter Part of the Hundred Years War
from 1369 to 1453 (London, 1956), pp. 12–13.
89
Ferdinand Lot expressed negative opinions avant la lettre: ‘Etudes sur la
bataille de Poitiers de 732’, Revue Belge de Philologie et d’Histoire 26 (1948),
72
but it has been adopted by many military historians. The basic idea
underlying IMP is the use of a kind of deductive test when doubts or
difficulties arise when determining what actually happened. According
to this test, an adequate knowledge of the field, of the varied elements
which may have been present in the military action (including types
of armament, technology, and typologies of castles) and of the written
sources may offer a reliable basis to infer tactics, to reconcile conflicting
contemporary evidence, and finally to establish a coherent narrative of
events. However, it is widely accepted that the further one goes back in
time the more difficult it is to apply IMP.90
Medieval generals were obviously neither naive nor inept and they knew
how to search for and profit from all types of advantages to outmanoeuvre
their opponents. This is perfectly compatible with the notion that war
can only be perceived in its own historical period, which means that a
fair number of aspects have to be taken into account: the chronological
context, as well as the entirety of norms, motives, beliefs, reasoning and
aims of men under arms, and of the social and mental background which
urged them to fight or made them capable of explaining the significance
of war. Perception and experience of emotions and the ways in which they
appear do not always need to be identical in different historical periods,91
but some human emotional reactions to combat or its immediacy may
be atemporal up to a point: anxiety, fear, anger, courage or shame among
33–59. See also Halsall, Warfare and Society in the Barbarian West, pp. 6–7; The
Battle of Agincourt: Sources and Interpretations, ed. Ann Curry (Woodbridge,
2000), pp. 400–1; Michael K. Jones, ‘The Battle of Verneuil (17 August 1424):
Towards a History of Courage’, War in History 9 (2002), 375–411.
90
As David Howarth points in his book on Hastings: ‘Strictly speaking, every
sentence in a story nine centuries old should include the word perhaps: nothing
is perfectly certain. But that would be boring, and I have left out the qualification
whenever things seem reasonably certain, either from the early sources or from
deduction and inference’: 1066: The Year of the Conquest (London, 1981), p. 8
(Howarth’s italics).
91
Rolf Petri, ‘The Idea of Culture and the History of Emotions’, Historien: A
Review of the Past and Other Stories 12 (2012), 21–37. Anger’s Past: The Social
Uses of an Emotion in the Middle Ages, ed. Barbara H. Rosenwein (Ithaca,
1998); Catherine Cubitt, Barbara Rosenwein, Stuart Airlie, Mary Garrison
and Carolyne Larrington, ‘The History of Emotion: A Debate’, Early Medieval
Europe 10 (2001), 225–56; Jilana Ordman, ‘Feeling like a Holy Warrior:
Western Authors’ Attribution of Emotions as Proof of Motives for Violence
among Christians Actors in Military Conflicts, Tenth through Early Twelfth
Centuries’ (unpublished Ph. D. thesis, Loyola University, Chicago, 2013).
73
Manuel Rojas
others. But these basic emotions should be understood in the context

of the value system of which they formed a part. Death in battle, for
example, was perhaps the greatest physical fear for a warrior, but for a
king or a high nobleman it had even greater consequences because of the
succession problems that this could cause for their domains.92 But it was
an anxiety tempered by the fact that battles were regarded as ordeals;
death in them was considered a martyrdom which was actively sought
and which facilitated salvation.93
92
For Alfonso VI’s reaction to the death of his son and heir Sancho and a
considerable number of Leonese-Castilian nobles at Uclés in 1108, see Primera
Crónica General, ed. Ramón Menéndez Pidal (Madrid, 1906), p. 555.
93
H. E. John Cowdrey, ‘Martyrdom and the First Crusade’, in Crusade and
Settlement: Papers Read at the First Conference of the Society for the Study of the
Crusades and the Latin East and Presented to R. C. Smail, ed. Peter Edbury
(Cardiff, 1985), pp. 46–56. For an example, see the sermon given by Adhemar,
bishop of Le Puy, before the battle of Antioch (1098): Robert the Monk’s History
of the First Crusade: Historia Iherosolimitana, ed. and trans. Carol Sweetenham
(Aldershot, 2005), p. 169.
74
Between Byzantium, Egypt and the Holy Land
The Italian Maritime Republics and the First Crusade
Mike Carr
University of Edinburgh
The question of what drove the citizens of the Italian maritime republics
to participate in the First Crusade is as complex and difficult to answer as
that regarding the participants of the crusader armies. The diverse mixture
of sailors, captains and other crewmembers who travelled by sea from Italy
to the Levant shared as many contrasting motives as the clerics, pilgrims
and knights who made the passage overland to Jerusalem. In these melt-
ing pots of society, pious devotion and religious zeal stood alongside and
were blended with desires for material gain and opportunity overseas.1
However, with the exception of the Normans, the Italian crusaders stood
apart from their European counterparts in one major way: they already
had pre-existing links with the territories of the Mediterranean with
which the crusade would come into contact. Consequently, while some
of the men who made up the Italian crusader fleets had visited foreign
ports in Egypt, Syria and Byzantium as merchants, others had fought for
their mother city against Muslims in the waters of the Mediterranean
and in North Africa.
This chapter aims to explore the different commercial, political and
religious concerns of the republics against this Mediterranean context.2
1
This has been the focus of the excellent article by Christopher Marshall, ‘The
Crusading Motivations of the Italian City Republics in the Latin East, 1095–
1104’, in The Experience of Crusading, ed. Marcus Bull and Norman Housley, 2
vols (Cambridge, 2003), 1: 60–79.
2
On the Mediterranean perspective of the Genoese crusaders, see Giovanna Petti
Balbi, ‘Lotte antisaracene e militia Christi in ambito iberico’, in ‘Militia Christi’
e Crociata nei secoli XI-XIII (Milano, 1992), pp. 519–45 (here pp. 519–25, 533–37,
539–45); Petti Balbi, ‘Genova e il Mediterraneo occidentale nei secoli XI-XII’, in
Comuni e memoria storica. Alle origini del comune di Genova (Genova, 2002), pp.
503–26 (here pp. 506–19); Michel Balard, ‘Genova e il Levante (secc. XI-XII)’,
in Comuni e memoria storica, pp. 527–49; Elena Bellomo, A servizio di Dio e del
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. –.
FHG DOI: 1.1/M.OUTREMER-EB.1.11
Mike Carr
The crusade was a military endeavour within the trading networks of the
eastern Mediterranean where commercial rivalries as well as devotional
impulses affected the republics’ different levels of participation, which
has, in turn, polarised the historiographical debate regarding their con-
tribution. This study, by building on current scholarship and by exploring
economic data regarding Italian trade in the Levant and Byzantium aims to
highlight the influence which pre-existing trade networks, relations with
Islam, and economic rivalries had on the contribution of the maritime
republics to the crusade.
On the eve of the First Crusade, the principal Italian cities trading in
the eastern Mediterranean were the Ligurian and Tuscan communes of
Genoa and Pisa, the city of Amalfi on the western coast of southern Italy,
and the Adriatic republic of Venice. Of these, merchants from Amalfi were
the most actively engaged in trade with Egypt and also firmly established
in Byzantium, specialising in the import of luxury goods, such as silks,
spices and high-value artisanal works. The presence of the Amalfitans
is attested to in Constantinople from the tenth century and they cul-
tivated close trading links with the Fātimids of Tunisia, after providing
naval assistance to them in 969 for the conquest of Egypt.3 A number of
Amalfitans were attacked and killed in Cairo for burning Egyptian ships
in the late tenth century, but before long they had returned to their former
prominence in the city.4 In the 1090s the Norman chronicler William
of Apulia wrote that merchants from Amalfi were actively importing
Santo Sepolcro: Caffaro e l’Oriente latimo (Padova, 2003), pp. 78–198; Jonathan
Phillips, ‘Caffaro of Genoa and the Motives of Early Crusaders’, in Religion as
an Agent of Change, ed. Per Ingesman (forthcoming). I thank Professor Phillips
for sending me a draft of this article in advance of publication.
3
For a background to Amalfitan trade at this time, see Armand O. Citarella,
‘The Relations of Amalfi with the Arab World before the Crusades’, Speculum
42 (1967), 299–312; Citarella, ‘Patterns in Medieval Trade: The Commerce of
Amalfi before the Crusades’, Journal of Economic History 28 (1968), 531–55; Vera
von Falkenhausen, ‘Il commercio di Amalfi con Constantinopli e il Levante nel
secolo XII’, in Amalfi, Genova, Pisa e Venezia: Il commercio con Costantinopoli e
il vicino Oriente nel secolo XII, ed. Ottavio Banti (Pisa, 1998), pp. 19–38, esp. pp.
19–26; Yaacov Lev, ‘A Mediterranean Encounter: The Fatimids and Europe,
Tenth to Twelfth Centuries’, in Shipping, Trade and Crusade in the Medieval
Mediterranean: Studies in Honour of John Pryor, ed. Ruthy Gertwagen and
Elizabeth Jeffreys (Farnham, 2012), pp. 131–56 (here pp. 140–43).
4
Samuel M. Stern, ‘An Original Document from the Fatimid Chancery
Concerning Italian Merchants’, in Studi orientalistici in onore di Giorgio
Levi Della Vida, 2 vols (Roma, 1956), 2: 529–38 (here pp. 533–34), reprinted
76
goods from Alexandria and Antioch, and that they had trading partners
amongst the Arabs, Libyans, Sicilians and Africans.5 Like Amalfi, the
northern communes of Genoa and Pisa were also engaged in trade with
Egypt during the eleventh century, although probably to a lesser degree
than their southern Italian counterparts. Documents from the Cairo
Geniza attest to Genoese and Pisan ships visiting Egypt on the eve of the
crusade.6 Moreover, the presence of bacini, Islamic pots inserted into the
fabric of Pisan churches, implies a high volume of trade linking northern
Italy with Egypt, Tunisia and Sicily.7
in Samuel M. Stern, Coins and Documents from the Medieval Middle East
(London, 1986), Essay V.
5
William of Apulia, La geste de Robert Guiscard, ed. Marguerite Mathieu
(Palermo, 1961), p. 190. William’s actual words are slightly cryptic, but he
certainly implies strong Amalfitan trading relations with these regions and
peoples: ‘Huc et Alexandri diversa feruntur ab urbe, Regis et Antiochi; gens
haec freta plurima transit; his Arabes, Libi, Siculi noscuntur et Afri’.
6
See, for example, Stern, ‘An Original Document’, pp. 529–38; Shelomo
D. Goitein, A Mediterranean Society: The Jewish Communities of the Arab World
as Portrayed in the Documents of the Cairo Geniza, 6 vols (Berkeley, 1967–93),
1: 32, 40–46, 59, 211. For a background to Genoese and Pisan trade in these years,
see Benjamin Z. Kedar, ‘Mercanti genovesi in Alessandria d’Egitto negli anni
Sessanta del secolo XI’, in Miscellanea di studi storici 2 (Genova, 1983), 21–30,
repr. in Kedar, The Franks in the Levant, 11th to 14th Centuries (Aldershot,
1993), Essay I; Gabriella Airaldi, ‘Groping in the Dark: The Emergence of
Genoa in the Early Middle Ages’, in Miscellanea di studi storici 2 (Genoa, 1983),
9–17; Steven A. Epstein, Genoa and the Genoese, 958–1528 (Chapel Hill, NC,
1996), pp. 24–27; Enrica Salvatori, ‘Pisa in the Middle Ages: The Dream and
the Reality of an Empire’, in Empires and States in European Perspective, ed.
Steven Ellis (Pisa, 2002), pp. 13–32, esp. pp. 15–20; Marco Tangheroni, ‘Pisa,
l’Islam, il Mediterraneo, la Prima Crociata: alcune considerazioni’, in Toscana
e Terrasanta nel Medioevo, ed. Franco Cardini (Firenze, 1982), pp. 31–56 (here
pp. 36–41).
7
See David Abulafia, ‘The Pisan Bacini and the Medieval Mediterranean
Economy: A Historian’s Viewpoint’, in Papers in Italian Archaeology 4,
ed. Caroline Malone and Simon Stoddart (Oxford, 1985), 285–91, repr. in
Abulafia, Italy, Sicily and the Mediterranean, 1100–1400 (London, 1987), Essay
XIII; Benjamin Z. Kedar, ‘Una nuova fonte per l’incursione musulmana del
934–935 e le sue implicazioni per la storia genovese’, in Oriente e Occidente tra
Medioevo ed Età moderna. Studi in onore di Geo Pistarino, ed. Laura Balletto, 2
vols (Genova, 1997), 2: 605–16.
77
Mike Carr
The Venetians were also engaged in trade with Egypt during the eleventh
century. Their presence in Alexandria is attested to in the Geniza docu-
ments of the 1060s and 70s and Venetian records of specific shipments
also survive, such as in 1071, when the merchant Giovanni Marinacio
imported nine sportae of alum from Alexandria to the Rialto.8 However,
Venice, situated in the Adriatic and with a long history of collaboration
with nearby Byzantium, had closer trading links with the empire than with
the Muslim states of North Africa and the Levant. Like the Amalfitans a
little earlier, towards the end of the eleventh century Venetian merchants
had begun to integrate themselves into the internal trade and maritime
networks of the Byzantine Empire, where they were able to take advantage
of economic opportunities in ports of call in the triangular trade system
linking Constantinople, Egypt and Syria.9 Here they frequently conveyed
agricultural, pastoral and industrial commodities, as well as silk, between
them.10 By the late eleventh century Venetian merchants were travelling
from Constantinople to the Levant, as is confirmed by a document of
1095, in which a merchant travelled from Venice to Constantinople and
then on to Saljūq-controlled Antioch.11
Of these Italian maritime republics, Amalfi was the only one not to
participate in the First Crusade. Indeed, according to legend, Amalfitan
merchants were present in Jerusalem during the siege of the city by the
crusaders, where they were ordered by the Fātimid authorities to bombard
the Christian armies with stones. However, by miracle these stones were
transformed mid-air into bread which instead had the opposite effect of
8
Documenti del commercio veneziano nei secoli XI-XIII, ed. Raimondo Morozzo
della Rocca and Agostino Lombardo, 2 vols (Roma, 1940), 1: 10–11 (doc. 11).
For references to the Geniza documents, see David Jacoby, ‘Byzantine Trade
with Egypt from the Mid-Tenth Century to the Fourth Crusade’, Thesaurismata
30 (2000), 25–77 (here p. 43).
9
David Jacoby, ‘Byzantine Crete in the Navigation and Trade Networks of
Venice and Genoa’, in Oriente e Occidente tra Medioevo ed Età moderna, 1:
517–40 (here pp. 518–19); Jacoby, ‘Italian Privileges and Trade with Byzantium
before the Fourth Crusade’, Anuario de estudios medievales 24 (1994), 349–68
(here p. 365); Jacoby, ‘Mercanti genovesi e veneziani e le loro merci nel Levante
crociato’, in Genova, Venezia, il Levante nei secoli XII-XIV, ed. Gherardo Ortalli
and Dino Puncuh (Genova, 2001), 213–56 (here pp. 216–17).
10
Jacoby, ‘Silk in Western Byzantium before the Fourth Crusade’, Byzantinische
Zeitschrift 84/85 (1991/1992), 452–500 (here pp. 493–94), reprinted in
Jacoby, Trade, Commodities and Shipping in the Medieval Mediterranean
(Aldershot, 1997), Essay VII.
11
Documenti del commercio veneziano, 1: 27–28 (doc. 24).
78
relieving the starving crusaders.12 In contrast to the Amalfitans, both the

north Italian communes were enthusiastic participants in the crusade.
The Genoese, in particular, made the greatest and earliest contribution,
dispatching a flotilla to the East in 1097, a second fleet which reached the
Holy Land in 1099, and another four in the years 1100–1104. The first
force, numbering twelve galleys and one small ship, supplied the crusad-
ers during the siege of Antioch from the port of Saint Symeon, while the
sailors of the second fleet famously helped in the capture of Jerusalem
by building siege machines from the timber of their dismantled vessels.13
The other fleets then assisted the Latins in consolidating their hold on the
Levantine seaboard, such as in the capture of Arsuf and Caesarea in 1101,
12
This legend is mentioned in David Abulafia, ‘Trade and Crusade, 1050–1250’,
in Cross Cultural Convergences in the Crusader Period, ed. Michael Goodich,
Sophia Menache and Sylvia Schein (New York, 1995), pp. 1–20 (here p. 4);
Abulafia, The Great Sea: A Human History of the Mediterranean (London,
2011), ch. 2, pp. 269–70. This may be a confused version of a similar legend
relating to a monk from the hospital of Jerusalem, originally founded by the
Amalfitans. This monk threw loaves to the crusaders during the siege, which
miraculously turned to stones after his arrest by the Muslim authorities,
see Josephi Historiographi, ‘Tractatus de exordio sacrae domus Hospitalis
Jerosolimitani’, in RHC Occ. 5: 405–10 (here p. 409); ‘Le commencement de la
fondacion de la sainte maison de l’ospital de S. Johan de Jerusalem’, in RHC Occ.
5: 411–21 (here pp. 419–20); Anthony T. 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 (here pp. 38–39).
13
The main source for this is the account of the Genoese annalist Caffaro, see
‘De liberatione civitatum orientis’, in Annali Genovesi di Caffaro e de’ suoi
continuatori dal MXCIX al MCCXCIII, ed. Luigi Tommaso Belgrano, 5
vols (Genoa, 1890–1929), 1: 99–124 (here pp. 99–110); Caffaro, Genoa and
the Twelfth-Century Crusades, trans. Martin Hall and Jonathan Phillips
(Farnham, 2013), pp. 107–25 (here pp. 107–16). See also Le ‘Liber’ de Raymond
d’Aguilers, ed. John H. Hill and Laurita L. Hill (Paris, 1969), pp. 49, 141, 147;
Gesta Francorum et aliorum Hierosolymitanorum, ed. and trans. Rosalind Hill
(London, 1962), p. 88. A detailed account of these events is given by Marie-
Luise Favreau-Lilie, Die Italiener im Heiligen Land vom ersten Kreuzzug bis
zum Tode Heinrichs von Champagne (1098–1197) (Amsterdam, 1989), pp. 43–51.
See also Elena Bellomo, ‘Tra Bizantini e Normanni: I Genovesi in Oltremare
agli esordi del XII secolo’, in Miscellanea in memoria di Giorgio Costamagna,
ed. Dino Puncuh, 2 vols (2003), 1: 143–66; Bellomo, A servizio di Dio, passim;
Epstein, Genoa and the Genoese, pp. 28–30.
79
Mike Carr
Tortosa in 1102, and the successful sieges of Jubail and Acre in 1104.14 As
a result of this, Genoa received extensive trading privileges in the newly
conquered ports, such as in Antioch where they were granted their own
church and thirty houses rent free by Bohemund.15 Pisa was a little slower
to respond, but nevertheless dispatched a large fleet (given the inflated
number of 120 vessels by the city’s annalists) under the command of
Archbishop Daibert to the Levant in the late winter of 1098 or spring
1099. This fleet participated in the siege of Laodicea, before contributing
to the rebuilding of Jaffa. This was followed by another fleet which later
assisted in the sieges of Caesarea and Acre.16 Like the Genoese, the Pisans
were also awarded privileges in the crusader states, but mostly at a later date
than their Ligurian counterparts.17 Finally the Venetians, who probably
14
For these events, see Caffaro, ‘De liberatione civitatum orientis’, pp. 111–24,
Caffaro, Genoa and the Twelfth-Century Crusades, pp. 117–25; Caffaro, ‘Annales
Ianuenses’, in Annali Genovesi di Caffaro, pp. 3–75 (here pp. 5–14), Caffaro,
Genoa and the Twelfth-Century Crusades, pp. 49–101 (here pp. 50–57); Albert
of Aachen, Historia Ierosolimitana: History of the Journey to Jerusalem, ed. and
trans. Susan B. Edgington (Oxford, 2007), pp. 671–75; Favreau-Lilie, Die
Italiener im Heiligen Land, pp. 79–110; Bellomo, A servizio di Dio, pp. 89–140.
15
The privileges were expanded and renewed by subsequent rulers of Antioch,
see the relevant documents in: Codice diplomatico della repubblica di Genova,
ed. Cesare Imperiale di Sant’Angelo, 3 vols (Roma, 1936–42), 1: 11–13 (docs.
7–8), 16–18 (doc. 12); I Libri Iurium della Repubblica di Genova, ed. Eleonora
Pallavicino, Dino Puncuh and Antonella Rovere, 8 vols (Genova, 1992–2002),
1/2: 152–55 (doc. 337), 157–60 (doc. 340). An overview is given by Geo Pistarino,
‘Genova e il Vicino Oriente nell’epoca del Regno Latino di Gerusalemme’, in
I Comuni Italiani nel Regno Crociato di Gerusalemme / The Italian Communes
in the Crusading Kingdom of Jerusalem, ed. Gabriella Airaldi and Benjamin Z.
Kedar (Genova, 1986), pp. 57–139, esp. pp. 65, 77–86.
16
The main sources for this are: ‘Gesta triumphalia per Pisanos facta’, in Gli
Annales Pisani di Bernardo Maragone, ed. Michele Lupo Gentile, RIS n.s. 6/2
(Bologna, 1930), pp. 87–96 (here pp. 89–90); Bernardo Maragone, ‘Annales
Pisani’, in Gli Annales Pisani, pp. 1–74 (here p. 7); ‘Chronicon Pisanum seu
Fragmentum auctoris incerti’, in Gli Annates Pisani, pp. 97–103 (here p. 102);
Albert of Aachen, Historia Ierosolimitana, ed. Edgington, pp. 671–75. See also
Favreau-Lilie, Die Italiener im Heiligen Land, pp. 51–61; Michael Matzke,
Daibert von Pisa: Zwischen Pisa, Papst und erstem Kreuzzug (Sigmaringen,
1998), pp. 135–52; Maria Luisa Ceccarelli Lemut, Pisa e l’Oriente latino dalla I
alla III Crociata (Pisa, 2010), pp. 9–13.
17
Favreau-Lilie, Die Italiener im Heiligen Land, pp. 382–495 and passim;
Abulafia, ‘Trade and Crusade’, pp. 12–13.
80
began preparing their forces in 1096–97, only managed to dispatch a fleet

in the second half of 1099. This flotilla wintered at Rhodes, where it came
into conflict with a Pisan force, before sailing to Palestine, helping capture
Haifa in 1100.18 The republic was also granted trading exemptions in the
crusader states, including at Jaffa, Haifa and Acre, as well as elsewhere.19
In the past it has been claimed that commercial concerns were the most
important, or sometimes the only, factors that motivated the Italian mari-
time republics to participate in the First Crusade and in the subsequent
expeditions to the eastern Mediterranean. The mercantile privileges that
the republics gained by assisting the crusaders in capturing Muslim ports
on the Levantine seaboard, in particular, have been used as evidence to
support this thesis, along with the seemingly self-centred behaviour of
the Italian merchants in the Levant during the later twelfth and thir-
teenth centuries, especially their continued willingness to trade with
Muslim Egypt and to pursue their factional rivalries in crusader ports,
not to mention the role of the Venetians in the diversion of the Fourth
Crusade.20 Eugene Byrne and William Heywood were two of the most
vociferous propagators of this in the early twentieth century, with Byrne
writing that the Genoese only helped the crusaders ‘in return for rewards
and privileges of deep import’, while Heywood went as far as to refer to
the history of Pisan expansion as ‘a sordid chronicle of commercialisation
18
The principal Venetian source for this is the ‘Monk of Lido’, ‘Historia de
translatione sanctorum magni Nicolai terra marique miraculis gloriosi, ejusdem
avunculi, alterius Nicolai, Theodorique, martyris pretiosi de civitate Mirea in
monasterium S. Nicolai de Littore Venetiarum’, in RHC Occ. 5: 253–92 (here
pp. 253–81). See also Favreau-Lilie, Die Italiener im Heiligen Land, pp. 62–79;
Elena Bellomo, ‘The First Crusade and the Latin East as Seen from Venice:
the Account of the Translatio sancti Nicolai’, Early Medieval Europe 17 (2009),
420–43 (here pp. 425–31).
19
Joshua Prawer, ‘The Venetians in Crusader Acre: A Reconsideration’, in Cross
Cultural Convergences in the Crusader Period, pp. 216–23; David Abulafia, ‘Gli
italiani fuori d’Italia’, in Storia dell’economia italiana, ed. R. Romano, 2 vols
(Turin, 1990), 1: 261–86 (here pp. 272–75), repr. in Abulafia, Commerce and
Conquest in the Mediterranean, 1100–1500 (Aldershot, 1993), Essay X.
20
For an analysis of the changing contemporary perceptions of the Italian
republics in the crusades, see Sylvia Schein, ‘From milites Christi to mali
Christiani: The Italian Communes in Western Historical Literature’, in I
Comuni Italiani, pp. 680–89, esp. pp. 681–85, and Marshall, ‘The Crusading
Motivations of the Italian City Republics’, pp. 60–61.
81
Mike Carr
and greed’.21 In 1985, Philip Hitti, rehashing the language used by Steven
Runciman thirty years earlier, even stated that ‘[t]rade – at least in the
case of the Genoese, Venetians and Pisans, the shrewdest money-makers
of the age – was a primary motivation in this venture [crusading]’.22 This
view was further expanded in two economic histories of the First Crusade,
published in the 1990s, although unfortunately in both accounts with
very flimsy references to the available economic data.23
However, as many recent studies have shown, it would be foolish to
describe Italian participation on the crusade as being driven solely by
a desire to obtain commercial privileges in the newly established Latin
territories overseas.24 This is especially the case when we consider that
21
Eugene H. Byrne, ‘Genoese Trade with Syria in the Twelfth Century’, American
Historical Review 25 (1919–20), 191–219 (here p. 193); William Heywood, A
History of Pisa: Eleventh and Twelfth Centuries (Cambridge, 1921), pp. 115–16.
For comments about the lack of Venetian religious sentiment in crusading, see
Richard A. Newhall, The Crusades (London, 1930), p. 47 and Aziz S. Atiya, The
Crusade in the Later Middle Ages (London, 1938), pp. 114–16.
22
Philip K. Hitti, ‘The Impact of the Crusades on Eastern Christianity’, in A
History of the Crusades, ed. Kenneth M. Setton, 6 vols (Madison, WI, 1969–
89), 5: 33–58 (here p. 38); Steven Runciman, A History of the Crusades, 3 vols
(Cambridge, 1951–54), 1: 351–66.
23
Gary M. Anderson, Robert B. Ekelund, Jr., Robert F. Hébert and Robert D.
Tollison, ‘An Economic Interpretation of the Medieval Crusades’, Journal of
European Economic History 21 (1992), 339–63, esp. 353–55, 360–61; and the
book by the same authors with Audrey B. Davidson, Sacred Trust: The Medieval
Church as an Economic Firm (Oxford, 1996), pp. 131–51. These works also make
the flimsy assertion that Saljūq hostility to Italian trading interests was one of
the main reasons for the participation of the maritime republics on the crusade:
Anderson et al., ‘An Economic Interpretation of the Medieval Crusades’, pp. 343–
46, 353–55; Anderson et al., Sacred Trust, pp. 134–35, 144. A similarly dubious
argument is propagated by Andrew S. Ehrenkreutz, ‘The Fatimids in Palestine:
The Unwitting Promoters of the Crusades’, in Egypt and Palestine: A Millennium
of Association (868–1948), ed. Amnon Cohen and Gabriel Baer (New York,
1984), pp. 66–72. Ehrenkreutz’s argument has been convincingly refuted by
Yaacov Lev, State and Society in Fatimid Egypt (Leiden, 1990), pp. 49–52.
24
See, in particular, Jonathan Riley-Smith, ‘Government in Latin Syria and the
Commercial Privileges of Foreign Merchants’, in Relations Between East and West
in the Middle Ages, ed. Derek Baker (Edinburgh, 1973), pp. 109–32; Marshall,
‘The Crusading Motivations of the Italian City Republics in the Latin East’, pp.
60–77; Phillips, ‘Caffaro of Genoa and the Motives of Early Crusaders’; Abulafia,
‘Trade and Crusade’, pp. 7–14. For similar arguments, but in different crusading
82
many of the Italian cities had already established secure trading links
in Egypt and the Levant before 1095, and that in the early years of the
twelfth century the long-term survival of the Latin coastal cities was
far from being guaranteed. Indeed, when considering the varying levels
of Italian contribution to the First Crusade, it is clear that commercial
concerns instead provide some explanation for a lack of participation on
the part of some cities, rather than for the commitment to the venture
of others. Amalfi, in particular, was tied up in the complex economic
and political milieu of southern Italy and the Levant during the late
eleventh century. Considering the city’s prominence in eastern trade,
it seems plausible that the Amalfitans decided not to join the crusade
in order to safeguard their strong trading position in Egypt and the
Levant. In addition to this, they had matters of defence to deal with
closer to home, having rebelled against Norman rule for the second time
in 1096.25 As Citarella has shown, in the tenth and eleventh centuries
the Amalfitans had rarely sided with their fellow Christians against
Muslims in the Mediterranean, preferring instead to cultivate com-
mercial links with Islamic rulers.26 Even after the capture of Jerusalem,
the Amalfitans were actively selling timber to the Fātimids in Cairo,
presumably for ship construction.27 Ironically, the decline of Amalfi’s
commercial importance during the twelfth century was probably a result
of its reluctance to participate in the crusade, a fact that is reflected in
the failure of the city to gain commercial privileges in the kingdom of
Jerusalem comparable to those obtained by the northern communes.28
contexts, see Rasa Mažeika, ‘Of Cabbages and Knights: Trade and Trade Treaties
with the Infidel on the Northern Frontier, 1200–1390’, Journal of Medieval History
20 (1994), 63–76; Mike Carr, ‘Trade or Crusade? The Zaccaria of Chios and
Crusades against the Turks’, in Contact and Conflict in Frankish Greece and the
Aegean, 1204–1453: Crusade, Religion and Trade between Latins, Greeks and Turks,
ed. Mike Carr and Nikolaos Chrissis (Farnham, 2014), pp. 115–34.
25
Robert Guiscard had laid siege to Amalfi in 1096, but was forced to abandon
the campaign after the greater part of his army left to join Bohemund on the
crusade: Geoffrey of Malaterra, The Deeds of Count Roger of Calabria and
Sicily and of His Brother Duke Robert Guiscard, trans. Kenneth Baxter Wolf
(Ann Arbor, 2005), pp. 203–5. For a background to this rebellion, see Ulrich
Schwarz, Amalfi im frühen Mittelalter (9.-11. Jahrhundert): Untersuchungen zur
Amalfitaner Überlieferung (Tübingen, 1978), pp. 65–68.
26
Citarella, ‘The Relations of Amalfi with the Arab World before the Crusades’,
pp. 299–312; Citarella, ‘Patterns in Medieval Trade’, pp. 531–55.
27
Stern, ‘An Original Document from the Fatimid Chancery’, pp. 533–34.
28
Jacoby, ‘Byzantine Trade with Egypt’, p. 48; Abulafia, ‘Trade and Crusade’, pp. 4–5.
83
Mike Carr
In Genoa there was also fear that involvement in the crusade could
endanger the city’s commercial relations with Muslim countries; a concern
which proved to be justified in the early years of the twelfth century after
the arrest of a number of Genoese merchants in Cairo for supporting the
crusaders.29 But in contrast to Amalfi, Genoa and neighbouring Pisa both
had a long-standing tradition of holy warfare against Muslims.30 The two
cities, fearful of Islamic domination of the western Mediterranean, had
collaborated in expelling the Moors from Sardinia in the early eleventh
century and, in 1087, launched a combined naval attack on the Muslim
port of Mahdia in North Africa.31 This campaign had strong religious
connotations; it was sponsored by Pope Victor III and many of the
participants made a pilgrimage to St Peter’s in Rome on their outward
voyage, after which they wore the pilgrim’s badge during the fighting.32 As
historians have argued, the Mahdia campaign encapsulated a number of
factors that would later characterise the participation of the Italian cities
on the First Crusade: it represented a growing engagement with Islam
through trade and warfare; it provided an early link between pilgrimage
and holy war, as seen by the badge of St Peter worn by the combatants;
and it showed a closer connection between the papacy and the republics
of Genoa and Pisa.33 This is not to say that these two cities were motivated
29
Bellomo, A servizio di Dio, pp. 141–54. For the arrest of the merchants, see
Goitein, A Mediterranean Society, 1: 45; the letter from the Cairo Geniza reads:
‘The sultan [al-Malik al-Afdal], imprisoned the Genoese, which caused great
consternation amongst the Rūm [Europeans] and no goods can be sold because
of this. It looks as if this recession will last long, so that everyone’s business has
come to a halt.’
30
See Citarella, ‘Patterns in Medieval Trade’, p. 550. A prominent Amalfitan
merchant, known as Pantaleone, did partiticipate in the Mahdia campaign of
1087, but his presence does not suggest that Amalfi played any official part:
Citarella, ‘The Relations of Amalfi with the Arab World before the Crusades’,
p. 311; von Falkenhausen, ‘Il commercio di Amalfi’, esp. p. 26 and pp. 20–22;
H. E. John Cowdrey, ‘The Mahdia Campaign of 1087’, English Historical
Review 92 (1977), 1–29 (here pp. 15–16).
31
For a detailed treatment of the Mahdia campaign, see Giuseppe Scalia, ‘Il carme
pisano sull’impresa contro i saraceni del 1087’, in Studi di filologia romanza
offerti a Silvio Pellegrini (Padova 1971), pp. 565–627; Cowdrey, ‘The Mahdia
Campaign of 1087’, pp. 1–29.
32
Cowdrey, ‘The Mahdia Campaign of 1087’, pp. 17–18, 22.
33
The characteristics of the Mahdia campaign have been subject to a long scholarly
debate. Recent studies which argue that the motivations of the participants
shared similarities with those of the crusaders include Balbi, ‘Lotte antisaracene
84
to participate in the crusade purely by their common history of confronta-

tion with Islam, or that Amalfi was deterred solely for commercial reasons,
but the contrasting relations with Muslims prior to 1095, in terms of both
commerce and warfare, do provide one explanation for their contrasting
response to the crusade.
Venice, on the other hand, stood somewhere between Amalfi and the
north-western communes. It had a history of confrontation with Islam
dating back to the tenth and early-eleventh centuries, when the city had
fought alongside Byzantium against the Saracens in the Adriatic, but in
the last decades of the eleventh century the struggle against Islam was
not a prominent feature of Venetian Mediterranean policy.34 In fact, as
Bellomo has suggested, conflicts with Muslims had a limited impact on
the construction of a public identity in the city, in contrast to Pisa and
Genoa.35 Instead, Venice was primarily concerned with the defence of the
Adriatic from the Normans, who had begun to threaten Venetian mastery
of the sea through a series of successful campaigns against Byzantium
in Albania. In order to prevent Robert Guiscard from capturing Corfu,
Durazzo and Valona, and thus cutting off the lower Adriatic, the repub-
lic had allied with Emperor Alexios I Komnenos in the war against the
Normans of 1081–85.36 In return for this, Venice received a chrysobull
from the emperor in 1082 which granted the city’s merchants extensive
trading rights throughout the empire. These amounted to freedom to
trade in all commodities and a total exemption from commercial dues,
toll payments and port duties, including a full exemption from the kom-
merkion.37 As we saw earlier, by the time of the First Crusade the Venetians
had integrated themselves into the local Byzantine trade networks, an
e ‘militia Christi’ in ambito iberico’, pp. 533–36; Bellomo, A servizio di Dio, pp.
11–13; Cowdrey, ‘The Mahdia Campaign of 1087’, pp. 22–23; Phillips, ‘Caffaro
of Genoa and the Motives of Early Crusaders’. On the contrary, Franco Cardini,
Le crociate tra il mito e la storia (Roma, 1971), pp. 342–43, n. 4, has argued that a
religious value was only ascribed to this expedition at a later date.
34
See Frederick C. Lane, Venice: A Maritime Republic (London, 1973), p. 32;
Abulafia, ‘Trade and Crusade’, pp. 23–28.
35
Bellomo, ‘The First Crusade and the Latin East as Seen from Venice’, p. 422.
36
See Anna Komnene, The Alexiad, trans. E. R. A. Sewter, revised Peter Frankopan
(London, 2009), esp. pp. 109–29, 160–64; William of Apulia, La geste de Robert
Guiscard, pp. 214–31, 236–49; Geoffrey of Malaterra, The Deeds of Count Roger
of Calabria, pp. 153–60; Donald M. Nicol, Byzantium and Venice: A Study in
Diplomatic and Cultural Relations (Cambridge, 1988), pp. 55–57.
37
For full details of the terms of the chrysobull, see Ralph-Johannes Lilie,
Handel und Politik zwischen dem byzantinischen Reich und den italienischen
85
Mike Carr
act which was facilitated by the preferential trading arrangements which

they had received from Emperor Alexios.38 In contrast, the Genoese and
Pisans were not regular traders in Byzantium until after the First Crusade
and even then they were unable to obtain privileges as generous as those
granted to Venice in 1082.39
When this economic background is considered, an explanation for
Venice’s delayed response to the crusade can be found; like Amalfi, the
Venetians did not want to compromise their position in the eastern mar-
kets, in particular that of Byzantium whose emperor was lukewarm in his
enthusiasm for the crusade.40 This is why we see the Venetians defending
the Aegean on their way to the Holy Land, possibly in order to help the
Byzantines guard it from Saljūq incursions, but also to prevent commercial
rivals, namely the Pisans, from encroaching on their markets.41 Evidence
for this is given by the main Venetian source for the crusade, the Translatio
sancti Nicolai, written by an anonymous monk from the Lido. In reporting
the passage of the Venetian fleet, the author states that it arrived in Aegean
waters in October 1099, where it wintered at Rhodes. At this time, the
Venetians were attacked by a Pisan fleet of fifty vessels flying Byzantine
standards and masquerading as a squadron from the imperial navy. The
Venetians sent envoys to the Pisans, urging them not to insult the Greeks
and Venetians, but to continue their voyage as righteous Christians and
peaceful pilgrims. The warnings were to no avail and the Venetians were
compelled to send out thirty vessels, which subsequently routed the Pisan
fleet, bringing back twenty-eight ships and 4000 prisoners in triumph
to Rhodes.42 Here the Venetians, in an act of great clemency, decided
to release the captives on the condition that they were never to trade in
Kommunen Venedig, Pisa und Genua in der Epoche der Komnenen und der
Angeloi (1081–1204) (Amsterdam, 1984), pp. 8–16.
38
See Jacoby, ‘Italian Privileges and Trade with Byzantium’, pp. 365–66.
39
Jacoby, ‘Italian Privileges and Trade with Byzantium’, pp. 357–62; Jacoby,
‘Byzantine Trade with Egypt’, pp. 55–58.
40
According to the Translatio sancti Nicolai, when the Venetian fleet arrived in
the Aegean in 1099 Emperor Alexios tried to persuade them to abandon their
mission to the Holy Land by promising gifts and threatening them: Monk of
Lido, ‘Historia de translatione’, p. 257.
41
The Turks may have been launching raids into the Aegean during these years,
see Anna Komnene, The Alexiad, pp. 202–3, 309.
42
Monk of Lido, ‘Historia de translatione’, pp. 257–58; the account of the battle
is discussed in detail by Favreau-Lilie, Die Italiener im Heiligen Land, pp. 646–
48; Bellomo, ‘The First Crusade and the Latin East as Seen from Venice’, esp.
pp. 425–27.
86
Byzantine lands nor to make war on fellow Christians.43 As Bellomo has

shown, the account of this battle (although embellished by the author)
is highly suggestive of the commercial tensions which existed during the
crusade.44 The rivalry between Pisa and Venice can therefore be regarded
as one reason for the republic’s delayed participation in the crusade; a fact
reflected by the author of the Translatio who clearly wished to emphasise
that the priorities of the Venetians were to protect their interests and
those of their Byzantine allies against the hostile Pisans. Moreover, if we
presume that the Pisan fleet had been sent from Palestine to intercept the
Venetians on their way to the Holy Land, as Favreau-Lilie has suggested,
then this also provides an indication that the Pisans too were wary of a
rival fleet sailing to the eastern Mediterranean.45
An analysis of the maritime republics in this wider Mediterranean
context shows that the crusade could not be easily separated from the
political concerns and commercial rivalries of the Italian communes in
the latter decades of the eleventh century. Consequently the republics’
pre-existing trade links and relations with Byzantium and the Islamic
powers of Egypt and the Levant provide some interesting explanations for
their varying levels of commitment to the venture. This is not, however, to
say that the Italians were motivated primarily by the prospect of material
gain, as has been claimed in the past; recent scholarship has shown this
not to be the case.46 Instead, this background helps to demonstrate that
concerns over the preservation of good relations – be they commercial or
political – in Egypt, Byzantium and the Levant largely acted to temper
commitment to the crusade, as seen in the case of Amalfi and to an extent
Venice, which was tardy in sending a fleet to the East. Pisa and Genoa, on
the other hand, with their common history of confrontation with Islam
and few discernible links with Byzantium, were enthusiastic contributors
to the crusade. The contrasting commitment of the maritime republics is
thus evidence of the challenges they faced in balancing the crusade with
the wider concerns of the mother city in the Mediterranean.
43
Monk of Lido, ‘Historia de translatione’, pp. 258–59 (‘se nunquam scilicet
deinceps Romaniam causa mercimonii intraturos, vel praelium in christianos
ullo modo excitaturos’).
44
Bellomo, ‘The First Crusade and the Latin East as Seen from Venice’,
pp. 439, 442.
45
Favreau-Lilie, Die Italiener im Heiligen Land, pp. 66–67.
46
See, in particular, Marshall, ‘The Crusading Motivations of the Italian City
Republics’, pp. 64–65; Bellomo, ‘The First Crusade and the Latin East as Seen
from Venice’, pp. 420–43; Phillips, ‘Caffaro of Genoa and the Motives of Early
Crusaders’.
87
The Crusade in the Councils
of Urban II beyond Clermont
Robert Somerville
Columbia University
The importance of papal councils in the Latin Church during the late
eleventh- and early twelfth-century reform has often been noted, and
need not be reiterated here. At the Lateran in Rome, elsewhere in Italy
especially in the south, and also north of the Alps especially in France,
the popes of the time convened dozens of synods. The extent to which
modern historians can reconstruct what happened in those assemblies
varies greatly from event to event. Some are little more than a line or two
in a chronicle; for others sources abound. Pope Urban II’s Council of
Clermont is in the latter category, and the welter of information available
about Clermont shows that launching the First Crusade was only one
piece of business among many treated. Clermont was the fifth council
which Urban celebrated during his eleven-year pontificate, and from
1096 to 1099 he presided at five more. Did they deal with the crusade?
The answer is that surely they did, as will be delineated in the pages to fol-
low, but it is important to remember that among the popes of this period
only for Gregory VII is a chancery register known, so access to official
records for conciliar activity is rare.1 The history that is pursued must be
reconstructed from a variety of unofficial sources.
1
For the crusade and Clermont, including references to works treating papal
councils of this period, see Robert Somerville, ‘Clermont 1095: Crusade
and Canons’, in La primera cruzada, novecientos años después: El concilio de
Clermont y los orígenes del movimiento cruzado, ed. Luis García-Guijarro
Ramos (Castellón, 1997), pp. 63–77. The present article is an edited version of
a lecture given at Huesca during the Segundas Jornadas Internacionales sobre La
Primera Cruzada in September 1999, and it retains aspects of that presentation.
Despite its age the discussion is not obsolete, but the bibliography on the First
Crusade published between 1999 and 2013 is large and it has been impossible
to add references to much that is important. Two works require mention: Ian
S. Robinson, The Papacy 1073–1198 (Cambridge, 1990), especially Chapter 3
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. –1.
FHG DOI: 1.1/M.OUTREMER-EB.1.11
Robert Somerville
In searching for the crusade in Urban II’s synods after Clermont it is

useful to recall the tendency of the Reforming popes to repeat the same
enactments from council to council. Furthermore, from the time of
Clermont until Pope Urban’s death in the summer of 1099 the develop-
ment of the crusading enterprise offered new issues for papal attention.
It would be expected, therefore, to find evidence about them in the synods
celebrated after Clermont, be it in repetition of Clermont’s decisions or
the formulation of new regulations.
A very brief summary of conciliar activity throughout Urban’s reign
(1088–99) is a useful point of departure. Early in his pontificate, while
in residence mainly in southern Italy, the pontiff held synods at Melfi in
1089, at Benevento in 1091, and at Troia in 1093. Legislation survives from
all three, as well as records of various conciliar acts.2 In early 1095, gain-
ing the upper hand over his rival Pope Clement III and having returned
to Rome, Urban set out on a long journey northward which would take
him across the Alps into France, to Clermont, and back to Italy only in
late summer of 1096. In March of 1095 he held an important council in
Lombardy, at Piacenza, followed in November by the famous gathering
at Clermont, and much is known about both of these assemblies. Other
synods in France followed Clermont, in 1096 at Tours and at Nîmes;
and after the pope’s return to Italy he convened a councils in 1097 at the
Lateran, in 1098 at Bari in Apulia, and in 1099 at St Peter’s in Rome.3
That makes ten councils, from 1089 to 1099. It can be pointed out
immediately, and is a disappointment to historians of the First Crusade,
that the first five, that is, Melfi through Clermont, are much better
documented than the last five, that is, Tours through St Peter’s. As will
be discussed below, lists of canons about whose authenticity no ques-
tion is possible survive from three of these last five, that is from Tours,
Nîmes, and St Peter’s, yet none of this legislation deals with the crusade.
on papal councils; and Somerville, Pope Urban II’s Council of Piacenza, March
1–7, 1095 (Oxford, 2011), Chapter 6, for a general treatment of Urban’s synods
after Clermont, repeating some but not all of what is given here.
2
Robert Somerville (with the collaboration of Stephan Kuttner), Pope Urban
II, the Collectio Britannica, and the Council of Melfi (1089) (Oxford, 1996); see
Appendix II for canons of Benevento and Troia.
3
In additional to the works of the present author listed above in nn. 1–2,
mention must now be made of the monumental achievement by Alfons Becker,
Papst Urban II, 3 vols, Schriften der MGH, 19/1–3 (Stuttgart, 1964–2012); see
Somerville, Piacenza, p. vii. See also the summary in Somerville, ‘Urban II,
Pope’, in Dictionary of the Middle Ages 12 (1988), 302–4, especially for the papal
councils.
90
The Crusade in the Councils of Urban II beyond Clermont
Historians have disagreed about whether or not canons survive from the
Lateran Council of 1097, an issue with ramifications for the crusade. No
known decrees are attributed specifically either to this synod or to Bari,
although Urban himself, in a letter to the bishop of Maguelonne, refers
to legislation enacted in the former, and a letter of Pope Paschal II refers
to a decision from the latter.4
The loss of the acts from the Council of Bari is a severe blow to under-
standing Pope Urban’s eastern policy, for the council met in southern Italy
on the frontier between Latin and Greek Christianity. Bari witnessed the
striking episode in which Anselm of Canterbury, in exile from England,
was asked by the pontiff to lecture the assembled churchmen, both Latins
and Greeks from southern Italy. The issue was the Latin doctrine of the
Double Procession of the Holy Spirit, that is, Procession from both the
Father and the Son, as opposed to the Greek teaching that the Spirit
proceeds only from the Father.5 Anselm’s biographer, the monk Eadmer,
went on to report that an anathema then was pronounced against those
who rejected the Latin teaching on the Procession. Eadmer, and follow-
ing him the twelfth-century English historian William of Malmesbury,
are the only sources for that condemnation.6 If this sentence actually
was promulgated it seems a curious gesture on Urban II’s part, for it no
doubt would have caused the Greeks to stomp out of the assembly. It can
be wondered whether Eadmer is describing a condemnation which was
officially issued at Bari, a synod in which the pope must have hoped for
consensus between Latins and Greeks.
Directly relevant to the crusade is a general letter written by the clergy
and people of Lucca virtually contemporary with the synod at Bari. In a
fascinating yet ambiguous conclusion the letter recounts that Pope Urban
‘is holding’ (tenet) a council at Bari. He is deliberating and planning
4
JL 5775, and JL 5929: papal letters are cited by number according to the calendar
in Regesta pontificum Romanorum ab condita ecclesia ad annum post Christum
natum MCXCVIII, ed. Philipp Jaffé, Wilhelm Wattenbach et al., 2nd edn
(Leipzig, 1885–88). The section covering Urban and Paschal’s pontificates was
revised by Samuel Loewenfeld, hence the preface ‘JL’.
5
For Bari see the assembled articles in Salvatore Palese and Giancarlo Locatelli,
Il Concilio di Bari del 1098 (Bari, 1999), several of which treat the issue of the
Procession of the Spirit. For Anselm at Bari see Eadmer, Historia novorum in
Anglia, ed. Martin Rule (London, 1884), pp. 104–5, and Becker, Urban II, 2: 193.
6
Eadmer, Historia novorum, ed. Rule, p. 106; see also William of Malmesbury,
Gesta Pontificum Anglicorum, ed. and trans. Michael Winterbottom and
Rodney M. Thomson, 2 vols (Oxford, 2007), 1: 156 (i. 53), an account derived
from Eadmer. See Becker, Urban II, 2: 193.
91
Robert Somerville
along with ‘many senators of the region’ (multis terrae senatoribus) to go

without question to Jerusalem.7 Setting aside questions about the origin
and transmission of this letter, the suggestion that Urban II planned to
travel to Jerusalem is remarkable.8 Is that possible? Did the pope consider
going east to join the crusaders, as, in fact, they had asked him to do in
their famous letter from Antioch dated September 11 of 1098?9
It does indeed seem possible and is maybe even likely that Urban II
professed such an intention, for early in the twelfth century Bohemund
of Antioch wrote to Pope Paschal II that Urban made such a promise in
a council – a council where Paschal himself had been present, Bohemund
reminded him – but death prevented fulfillment of that plan. It is, unfor-
tunately, unclear to which of Urban’s councils Bohemund is referring,
and both Bari and the assembly at St Peter’s in 1099 are possibilities.10
Cardinal Rainerius, the future Paschal II, attended both gatherings,
and Bohemund’s reminder to the pope that he was on hand when his
predecessor made that promise strongly suggests that it happened.11 But,
planned or not, Urban made no such trip, and little more can be said on
the matter except to point out that Bohemund’s letter was discovered in
a manuscript in Lincoln Cathedral Library in the 1930s, thus pointing
7
Heinrich Hagenmeyer, Epistulae et chartae ad historiam primi belli spectantes
(Innsbruck, 1901), no. XVII, p. 167: ‘notum quoque uobis facimus, quod
dominus papa Urbanus apud Barum tenet concilium, tractans et disponens
cum multis terrae senatoribus ad Ierusalem profecto tendere.’
8
The letter is found in MS Paris, Bibliothèque Mazarine, 1710, ff. 72r-73r, in a
book composed of pieces of various eleventh- and twelfth-century manuscripts:
see Auguste Molinier, Catalogue général de la Bibliothèque Mazarine, 4 vols
(Paris, 1885–92), 2: 179–80, and the online catalogue of the manuscripts at the
Mazarine, http://www.calames.abes.fr/Pub/ms/MAZB11576 . MS 1710 comes
from the abbey of Saint-Martin-des-Champs in Paris although it is not clear
that all pieces therein were copied at that house.
9
Hagenmeyer, Epistulae, no. XVI, p. 164. See Paul Kehr, Italia pontificia 9, ed.
Walther Holtzmann (Berlin, 1962), 442, no.4.
10
Walther Holtzmann, ‘Zur Geschichte des Investiturstreits (Englische
Analekten II)’, Neues Archiv 50 (1935), 280–82. See Paul Kehr, Italia pontificia
8 (Berlin, 1935), 27–28, no.91; Kehr-Holtzmann, Italia pontificia 9, no. 7; and
Becker, Urban, 2: 198, nn. 379 and 428 (where a date of 1106 or 1108 is offered
for Bohemund’s letter). For St Peter’s, and a date of 1108, Carl Erdmann, Die
Entstehung des Kreuzzugsgedankens (Stuttgart, 1935), p. 303, n. 68.
11
See JL 5929 for Bari, and Kehr-Holtzmann, Italia pontificia, 9, p. 205, no.*10,
for St Peter’s.
92
out that modern research still can uncover important sources pertaining
to the crusades.12
Bari has been treated first because, from the perspective of the crusade,
the synod’s location in Apulia, its involvement with Latin-Greek issues,
and the hint that Urban discussed therein a trip to the East, make it espe-
cially tantalising. Yet in the end there is a let-down, and what is true for Bari
is also the case in three of Urban’s other post-Clermont councils, that is,
Tours and Nîmes in 1096, and St Peters in 1099. In each case the evidence
points to the fact that the crusade was treated, but details are scarce.
Council of Tours
It once was thought that Urban II celebrated a synod at Limoges at
Christmastime, 1095, a month after Clermont. The pontiff and his entou-
rage spent Christmas at Limoges, and Urban preached the crusade, but
he did not formally convoke a council.13 Furthermore and along similar
lines, a recently discovered thirteenth-century text from a manuscript at
Seo de Urgel had Pope Urban convening a council at Chartres, but this
too is a mistake.14 Three months after Clermont, in late March during
the third week of Lent, Urban assembled at Tours the second council
held during his trip to France. Two sources nearly contemporary with
the synod provide vignettes about what happened. Bernold of Constance
reported that Urban reaffirmed the statutes of his earlier synods; and the
Gesta Andegavensium more specifically noted that in the Council of Tours
Urban ‘confirmed those things that he had dealt with at Clermont’, which
thus included whatever Clermont enacted about the crusade. One canon
ascribed to Tours survives, repeating legislation from Clermont but not
12
MS Lincoln, Cathedral Library, 233, fol. 91v: see Rodney M. Thomson,
Catalogue of the Manuscripts of Lincoln Cathedral Chapter Library (Cambridge,
1989), p. 193, at the end of a collection of the letters of Bishop Ivo of Chartres.
13
Robert Somerville, ‘The French Councils of Pope Urban II: Some Basic
Considerations’, Annuarium Historiae Conciliorum 2 (1970), 56–65 (repr. in
Papacy, Councils and Canon Law in the 11th-12th Centuries (Aldershot, 1990),
Essay V; see the ‘Additions and Corrections’ at the end of the volume). See
also Becker, Urban II, 2: 442–43, and Heinrich Hagenmeyer, Chronologie de la
première croisade (1094–1100) (Paris, 1902), pp. 11–12.
14
Robert Somerville, ‘Pope Urban II, a Pseudo-Council of Chartres, and
Congregato (C.16, q.7, canon 2 ‘palea’)’, in Reform and Renewal in the Middle
Ages and the Renaissance: Studies in Honor of Louis Pascoe, S.J., ed. Thomas M.
Izbicki and Christopher M. Bellitto (Leiden, 2000), pp. 18–34.
93
Robert Somerville
the crusading regulations.15 On the Sunday before the council opened

the pope delivered a public sermon from a great wooden pulpit which
was erected on the banks of the Loire. The source said nothing about the
crusade, but why would Urban have been silent about it on this occasion,
given his repeated advocacy of the venture throughout the spring and
summer of 1096?16
Council of Nîmes
The situation at the Council of Nîmes, which Urban assembled in July,
1096, is a bit clearer, although the evidence is hardly plentiful. Perhaps
the decision to send the bishops of Grenoble and Orange to seek
Genoese help for the expedition to the Holy Land was made in this
synod, although exactly when Urban took that action is unclear.17 It also
is possible, although this has not been confirmed as far as this author is
aware, that from Nîmes the pontiff dispatched a remarkable letter to the
counts of Besalú, Empurias, Roussillon, and Cerdaña and their followers.
He urged, with the promise of remission of sins, the restoration of the
church and city of Tarragona in place of going to liberate the churches
in Asia.18 Notwithstanding these uncertainties, without question the
crusade appeared on the conciliar agenda. Urban preached a recruit-
ing sermon for the expedition, and a dispute was resolved between the
abbey of Saint-Gilles and Count Raymond of Toulouse, ‘about to go
on the expedition to Jerusalem’.19 Furthermore, the canonical tradition
15
Bernold of Constance, ‘Chronicon’, in Die Chroniken Bertholds von Reichenau
und Bernolds von Konstanz 1054–1100, ed. Ian S. Robinson, MGH Scriptores
rerum Germanicarum, n.s. 14 (Hannover, 2003), 527; ‘Gesta Andegavensium’,
in RHC Occ. 5: 345 (p. lxxiii for the date); see. Hagenmeyer, Chronologie, p. 15.
For the canon of Tours see Somerville, Piacenza, pp. 119–20.
16
La Chronique de Saint-Maixent, 751–1140, ed. Jean Verdon (Paris, 1979),
p. 154: ‘Ubicumque fuit, precepit cruces facere hominibus et pergere Jerusalem
et liberare eam a Turcis et aliis gentibus.’
17
Paul Kehr, Italia pontificia 6/2 (Berlin, 1914), 323, no.*5 (= JL *5651).
18
Paul Kehr, ‘Papsturkunden in Spanien, 1: Katalanien’, Abhandlungen der
Gesellschaft der Wissenschaften zu Göttingen, philologisch-historische Klasse, n.s.
18/2 (Berlin, 1926), pp. 287–88, no. 23. For the date see Becker, Urban II, 2:
347, n. 169, and also Jonathan Riley-Smith, The First Crusade and the Idea of
Crusading (London, 1986), pp. 19–20.
19
Hagenmeyer, Chronologie, p. 30; Becker, Urban II, vol. 2, pp. 452–54; JL 5659.
See also Michael Matzke, ‘Die origine Hospitalariorum Hierosolymitanorum:
94
from Nîmes is more plentiful than from the Council of Tours, but as
with Tours, Nîmes’s surviving decrees repeat provisions from Clermont
although nothing on the crusade.20 Yet crusading measures probably were
issued, and an early twelfth-century text from the monastery of Fleury
implies that at Nîmes Urban ordered bishops to preach the crusade in
their dioceses, and commanded that those who were to participate in the
venture should wear crosses.21
Council at St Peter’s, Rome

In late April of 1099, less than three months before the capture of
Jerusalem by his armies and slightly more than three months before his
death, Urban II held his last council, at St Peter’s. Bernold of Constance
wrote that in this assembly the pontiff urged that reinforcements be
sent to bolster the Latin armies in the East.22 It is impossible to say how
much the pope knew of events in the East, but Becker surely was correct
in suggesting that he was receiving considerably more information than
surviving sources reveal.23 At the time of the gathering in St Peter’s did
Urban have an idea of how close to their goal the Latin armies were? Was
it, furthermore, at the Council of Rome rather than at Bari that the pontiff
promised to join those warriors? In 1099 Urban was urging Archbishop
Anselm of Milan to help recruit forces for the crusade, and was encour-
aging the citizens of Pisa to send a flotilla to the East at about the same
time.24 Whether either of these actions was connected specifically to the
gathering at St Peter’s toward the end of April in 1099 is unknown.
The legislative tradition from the Roman assembly is fuller than for
any of Urban’s other synods after Clermont, but the nineteen surviving
Vom klösterlichen Pilgerhospital zur internationalen Organisation’, Journal of

Medieval History 22 (1996), 1–23 (here p. 9).
20
For authentic canons from Nîmes see Stephan Kuttner and Robert Somerville,
‘The So-called Canons of Nîmes (1096)’, Tijdschrift voor Rechtsgeschiedenis 38
(1970), 175–89 (repr. in Kuttner, Medieval Councils, Decretals, and Collections of
Canon Law (London, 1980, 2nd edn 1992), Essay III, with ‘Retractationes’ for both
printings; repr. also in Somerville, Papacy, Councils and Canon Law, Essay IX).
21
‘Narratio Floriacensis de captis Antiochia et Hierosolyma et obsesso Dyrrachio’,
in RHC Occ 5: 356. See also p. xci of the volume’s Introduction.
22
Bernold of Constance, Chronicon, ed. Robinson, p. 537.
23
Becker, Urban II, 2: 424–25.
24
Paul Kehr, Italia pontificia, 6/1 (Berlin, 1913), 54, no.*129; 3 (Berlin, 1908), 359,
no.*23.
95
Robert Somerville
decrees, repeating some canons from the earlier councils at Melfi in 1089,
and Piacenza in 1095, deal mainly with pressing issues of the Reform
movement such as simony and clerical appointments.25 None concerns the
crusade, yet an eye-witness at the council offered a little more information
which suggests such a connection. Bishop Lambert of Arras attended the
synod in St Peter’s, and had, moreover, a close relationship with Pope
Urban II.26 He wrote that this assembly was convened to deal with ‘the
error and heresies of the Greeks’ (pro errore et haeresibus Graecorum), and
that it enacted, together with reforming measures for the Latin Church,
‘decrees concerning the Latin and the Greek Church’ (decreta de Latina
et Graeca ecclesia).27 Whatever that actually means, it points eastward, at
least as far as Constantinople and maybe further. Lambert continued,
and wrote that a diligent reader could find these canons and read them;
but he did not bother to say where, and quickly moved on to speak of
the death of the pope and his burial in St Peter’s next to Leo the Great.
If what Lambert said was accurate, however, an early twelfth-century
investigator could have put his hands on a series of canons from the 1099
Council of Rome treating Greek-Latin issues, perhaps including matters
relating to the crusade. And if they ever existed, it is not impossible that
these provisions will someday re-emerge.
That the legislation from Rome was more extensive than the decrees
which have survived is also revealed by an entry in the Chronicle of the
abbey of Saint-Maixent. Following a notice which relates that in this
synod Pope Urban and the others on hand ‘reaffirmed the journey to the
holy sepulchre of our Lord Jesus Christ’ (Urbanus papa […] confirmavit
et ceteri viam Sancti Sepulchri Domini Nostri Jhesu Christi), the Chronicle
continues and says that the pope also decreed therein that all Christians
should fast every Friday on behalf of their sins, especially those which
they forgot to confess.28 Friday abstinence was an ancient Christian prac-
tice, although no other source known to this investigator ascribes such a
regulation to an eleventh-century papal decree. Gregory VII and Urban II
25
Giovanni Domenico Mansi, Sacrorum concilorum nova et amplissima collectio
20 (Venezia, 1775), 961–64. There are 18 numbered items in the list, but canon
17 seems to embody two decrees. For Melfi and Piacenza see nn. 1 and 2 above.
26
Becker, Urban II, vol. 2, p. 200.
27
Gesta Atrebatensium, in Recueil des Historiens des Gaules et de la France, nouv.
éd., 14 (Paris, 1877), 756. For Lambert and this text see now Lotte Kéry, Die
Errichtung des Bistums Arras, 1093/1094 (Sigmaringen, 1994), 71, n. 22.
28
Chronique de Saint-Maixent, ed. Verdon, p. 168: ‘Urbanus papa […] et decrevit
in ipsa synodo, omni sexta feria, jejunare pro peccatis suis omnibus christianis
et maxime pro illis quibus non confessi sunt inmemores.’
96
early in his pontificate both promulgated canons differentiating true from

false penance, including a provision from the Council of Melfi in 1089
which stated that penance for one sin only was in vain and was needed
for all sins committed if forgiveness was to be gained.29 John Cowdrey
has linked this concern about full confession to Urban’s plan at Clermont
that the crusade indulgence be applicable only for those who had made
an honest and complete confession, and undertook the journey out of
devotion only, and not for honour or money.30 Was this specific link made
an issue in the Council of Rome? The juxtapositioning in the notice from
Saint-Maixent of confession and penance for forgotten sins could suggest
as much, but again, as repeatedly is the case in these investigations, the
evidence to pursue the question is not at hand.
Thus far attention has been devoted to four of Urban’s five post-
Clermont councils, that is Bari, Tours, Nîmes, and Rome. Virtually
nothing other than its existence is known about the remaining synod,
that is, the Lateran Council of 1097, yet that gathering requires a place
in discussion of the legislation from Urban’s later synods.31 In 1884 Julius
von Pflugk-Harttung published a list of 10 canons which carry no attri-
bution, but to which he added the editorial conjecture that they were
the decrees from a synod of Urban II held between 1097 and 1099.32
Pflugk-Harttung found these texts in an early-modern manuscript: MS
C.24 at the Biblioteca Vallicelliana in Rome. They are accompanied by
a note from the great sixteenth-century Spanish scholar Miguel Tomás
Taxaquét (Michael Thomasius) indicating (1) that these decrees were
transcribed from a codex owned by his contemporary, the Italian humanist
and canonist Hieronymus Parisetti, and (2) that perhaps they derive from
a synod of Urban II.33 Parisetti’s manuscript has not come to light, so it
29
For Melfi canon 16 and the Gregorian provision see Somerville, Council of
Melfi (1089), p. 257, and also pp. 293–94.
30
H. E. John Cowdrey, ‘Pope Gregory VII and the Bearing of Arms’, in Montjoie:
Studies in Crusade History in Honour of Hans Eberhard Mayer, ed. Benjamin
Z. Kedar, Jonathan Riley-Smith and Rudolf Hiestand (Aldershot, 1977), pp.
24–25, with full references to the relevant texts.
31
See Somerville, Piacenza, pp. 123–25. The ‘Excursus’ on pp. 124–25 about the
1097 councils duplicates some but not all of the analysis given below.
32
Julius von Pflugk-Harttung, Acta pontificum Romanorum inedita 2 (Stuttgart,
1884), 167–68.
33
Pflugk-Harttung, Acta, 2: 168; see also Laura Gasparri, ‘Osservazioni sul Codice
Vallicelliana C.24’, Studi Gregoriani 9 (1972), 467–513 (here pp. 483, 485), and
for this well-known manuscript see Somerville-Kuttner, Pope Urban, pp. 193
and 215–23. For Parisetti, a member of the post-Tridentine papal commission
97
Robert Somerville
is impossible to know whether or not the context in which these canons

occur there provided the basis for attributing them to Urban.
A few of the texts relate to crusading issues: canon 5 speaks about pen-
ance for an arsonist being that he ‘remains in God’s service for a year in
Jerusalem or in Spain’, and canon 7 offers protection for the possessions
of those who go to Jerusalem. The verbal formulae used in this list of
decrees is fully consistent with formulations to be found in papal synods
at the time of Urban II, for example the use of the first person plural ‘we
command’ (praecipimus, canons 1–2), ‘we forbid’ (interdicimus, canon 3),
‘we prohibit’ (prohibemus, canon 6), and also the invocation in canon 1
of the authority and command of the holy fathers and the entire holy
Roman Church’ (Ex auctoritate et praecepto sanctorum patrum totiusque
sanctae Romanae ecclesiae). Furthermore, as a point of reference a ‘Lateran
Council’ is cited in canons 1–2. Pflugk-Harttung speculated that he was
dealing with regulations issued in the Lateran Council which Urban II
convened in 1097, and which survived in this copy in a form for distribu-
tion, perhaps as part of the acts of the 1099 Council at St Peter’s.
The origin of these canons has been debated on and off since Pflugk-
Harttung first printed them.34 There is no question that some of the
decrees repeat, at times verbatim, provisions from Calixtus II’s Lateran
Council of 1123, and especially Innocent II’s Lateran Council of 1139.
Furthermore, this set of decrees lacks items to be expected in legislation
from one of Urban II’s late synods, especially a prohibition against lay
investiture, which was promulgated both at Bari and at the Council of
Rome, and presumably was issued at the Lateran Synod too.35 Pflugk-
Harttung’s canons could represent only a segment of legislation from a
late synod of Urban II, but it is more reasonable to conclude that they
belong to a gathering which met after 1139, although until light is shed on
why Tomás Taxaquét thought that these texts might belong to an assembly
of Pope Urban the matter remains slightly open.
for producing a new edition of the Corpus iuris canonici (the ‘Correctores
Romani’), see Stephan Kuttner, ‘Some Roman Manuscripts of Canonical
Collections’, Bulletin of Medieval Canon Law n.s. 1 (1971), 13–23 (repr. in
Medieval Councils, Decretals, and Collections Law of Canon), Essay II, and in
general for the ‘Correctors’ see Mary E. Sommar, The Correctores Romani:
Gratian’s Decretum and the Counter-Reformation Humanists (Berlin, 2009).
34
See, for example, Francis J. Gossman, Pope Urban II and Canon Law
(Washington, DC, 1960), p. 1, and Hartmut Hoffmann, Gottesfriede und
Treuga Dei, Schriften der MGH, 20 (Stuttgart, 1964), 222, 225 for a summary
of the discussion.
35
For Bari see JL 5929; for Rome see Mansi, Amplissima collectio, 20: 964.
98
The foregoing pages yield, by way of a summary, the following results.

The crusade was not forgotten in Urban’s councils after Clermont, but
exactly what was done is uncertain, sometimes tantalisingly so. Pope
Urban would have recruited for the expedition at all of his later synods,
and various actions concerned with the crusade emerge in the episodic
conciliar sources at hand. Clermont’s decrees were reiterated, but just what
the rulings from November 1095 stipulated is a well-known conundrum.
But they certainly included the crusading indulgence, placement of cru-
saders’ goods under the Peace of God, instructions to bishops to recruit
participants in local churches, and other regulations, for example about
the wearing of crosses, and maybe even about the disposition of territory
which might be captured in the East.36 Urban’s councils at Bari and Rome
of 1098 and 1099 offer other possibilities. Each seems concerned with
recruitment for the crusade, and broadly speaking with ‘Latin-Greek’
issues, which could have included matters pertaining to the crusade. At
one or both of these gatherings, furthermore, Urban may have promised to
go east himself; and at the Council of Rome he appears to have enjoined
Friday fasting as penance for incomplete confession of sins, which perhaps
was linked to the crusading indulgence.
Some thought might be given, in conclusion, to the question of why,
notwithstanding these scattered pieces of evidence, Urban’s crusade is not
presented in the later councils with the level of clarity found, for example,
in the decrees on the Wibertine Schism promulgated in the Council of
Piacenza. The answer must be sought within the general question of the
survival of the acts of Urban’s synods after 1095. Despite the problems in
attempting to reconstruct the canons of Clermont, Clermont shows that
provisions about the crusade survive, when they survive at all, in extended
lists of decrees preserved by local churches. The crusade was one issue
among many treated at Clermont, and regulations about it stand side by
side with decrees on many other issues. If fuller accounts of the decrees
from the post-Clermont assembles were available the First Crusade would
no doubt be found therein as one concern among many. Urban II’s pon-
tificate was a time of papal schism and a period of turmoil in the Latin
Church. For churchmen in Europe, engulfed by these ecclesiastical prob-
lems in the 1090s, the crusade would not rank as more noteworthy than
regulations about canonical appointments to ecclesiastical offices, or the
place of laity in Church affairs, to name two pressing issues of the time.37
36
See Somerville, ‘Pope Urban’, passim.
37
Along these lines see the discussion of priorities for the Reform papacy by
Johannes Laudage, ‘Ad exemplar primitivae ecclesiae: Kurie, Reich und
Klerusreform von Urban II. bis Calixt II.’, in Reformidee und Reformpolitik im
99
Robert Somerville
Without comprehensive accounts of Urban II’s later councils h istorians

cannot assess how much emphasis was given to the crusade in these
assemblies. Yet this is true of other matters as well. Were it not for a refer-
ence in a letter of Pope Paschal II, for example, no evidence would reveal
that the Council of Bari enacted a decree against lay investiture.38 Why
these synods are so poorly documented in comparison to Urban’s earlier
assemblies is a difficult question for which no satisfactory answer exists.
There is no reason to doubt, however, that new discoveries about the acts
of Urban’s last five assembles would include new information about how
those councils treated the crusade, or about how Latin-Greek relations
were handled late in the pontificate. A reader in the early twenty-first
century thus can be left to ponder Lambert of Arras’ statement that a
reader who wished to do so could find the decrees from the Council of
Rome on the Latin-Greek question, although the studious bishop did
not bother to indicate where.
spätsalisch-frühstaufischen Reich, ed. Stefan Weinfurter (unter Mitarbeit von

Hubertus Seibert) (Mainz, 1992), pp. 58–59.
38
JL 5929.
100
II
The Course of the Crusade

An Army on Pilgrimage
University of Cambridge
In this paper I will argue that although crusading drew on the traditions
of eleventh-century pilgrimage, there was no gradual development of one
institution into the other, as has sometimes been suggested. The crusade
was a new departure. It was distinctive from the start because each volun-
teer had to make a vow, signified by the wearing of a cross, and because he
was performing a penance in arms. Once it was over, the customary rule
that pilgrims, as penitents, should be unarmed again applied, but some
twelfth-century pilgrims to the east can be found serving in the armies of
the kingdom of Jerusalem and I will try to explain in what circumstances
they felt able to do so.
The influence of pilgrimage upon crusading and the parallels between
them are, of course, easy to find. Much of language of the First Crusade
drew directly on pilgrim traditions, although it was also shot through
with military terminology. In writing of it Pope Urban II used the terms
of pilgrimage – iter, via, labor – although he also employed the military
phrase Jherosolimitana expeditio. The pilgrim terms peregrinatio, via, iter,
iter beatum, iter Domini and sanctum iter were used in letters written by
the crusaders on the march and in these they occasionally referred to
themselves as peregrini, but they also wrote of the exercitus in which they
were serving. In two Burgundian charters the crusade was referred to as
both a peregrinatio and an expeditio. A Provençal one referred to two
brothers taking the cross ‘on the one hand for the grace of the pilgrimage
and on the other, under the protection of God, to wipe out the defile-
ment of the pagans’.1 In their preparations crusaders employed on a larger
scale measures which had already been put to use by pilgrims, who had
arranged for the management of their lands while they were away and had
pledged their estates to raise funds; one surviving eleventh-century charter
included a clause, which would also appear in those of some crusaders
1
Jonathan S.C. Riley-Smith, The First Crusaders 1095–1131 (Cambridge, 1997),
pp. 67–68.
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. 11–11.
FHG DOI: 1.1/M.OUTREMER-EB.1.11
and must have already been a standard formula, covering the eventual-
ity that the pilgrim might decide to stay in the Holy Land. Pilgrims had
settled disputes with churches over rights to which they had doubtful
claims, although only one of these agreements seems to have been made
in exchange for money, whereas renunciations in exchange for cash were
to be a feature of the crusaders’ preparations, suggesting, when taken
together with the small number of surviving pledge-charters drawn up for
pilgrims, that, unlike a crusade, a pilgrimage to Jerusalem was just within
the financial capabilities of most landowners. On departure pilgrims and
crusaders alike made endowments in return for the promise of prayer; and
the three most travelled pilgrim roads to Constantinople – down Italy to
Bari and across to Greece; by way of Hungary and Bulgaria; and through
Dalmatia – were also to be taken by the armies of the First Crusade.2 On
the march crusaders who scrupulously fulfilled the liturgical and peni-
tential requirements of pilgrimage won praise.3
More striking, however, than the similarities between eleventh-century
pilgrimages and early crusades are the differences. In the first place, all
crusaders were expected to make a vow, a promise made in the present
tense in front of witnesses, and assemblies of local men and women,
marked by emotional preaching and near hysteria, were the most com-
mon environments in which they committed themselves to an expedi-
tion. Their vows were signified by the wearing of crosses. As each man
made his vow he was presented with a cloth cross and he was supposed
to have it attached to his clothes at once. This aspect of the proceedings
needed careful preparation and as late as 1463 Cardinal Bessarion issued
the f ollowing instruction to preachers:
The manner of fixing on the sign (of the cross) shall be identical in all […]
places and shall be set in motion as rapidly as possible. When a sign has
been made from red silk or cloth, they shall attach it to the breast with a
needle. Those receiving it may afterwards sew it firmly in place.4
Crusaders were expected to go on wearing their crosses, which were very
distinctive from the first and got bigger as the twelfth century progressed,
until they came home with their vows fulfilled: in 1123 the bishops at the
2
Riley-Smith, First Crusaders, pp. 33–36.
3
See, for example, Guibert of Nogent, Dei gesta per Francos et cinq autres textes,
ed. Robert B. C. Huygens (Turnhout, 1996), pp. 198–99.
4
See Norman Housley, Documents on the Later Crusades, 1274–1580 (Basingstoke,
1996), p. 151.
104
First Lateran Council decreed against those ‘who had taken their crosses
off ’ without departing on crusade.5
Pilgrims to Jerusalem, on the other hand, had never been under the
general obligation to make vows and although some did so6 one must sup-
pose that most never did. They were, broadly speaking, of three types. The
first, and perhaps the most numerous, were those performing severe pen-
ances imposed on them by their confessors. By their very nature penances
needed no vow; they required contrition, demonstrated by obedience to
one’s confessor, as a necessary element in absolution. The second, often
hard to distinguish from the first because there was a penitential element in
their journeys as well, were those engaged in what was called a peregrinatio
religiosa, an act of devotion undertaken voluntarily and perhaps vowed,
but not enjoined by a confessor. The third were those who were going to
Jerusalem to live there until they died; the special location of the city in
the geography of providence meant that it was a place in which devout
Christians wanted to be buried.7 The popular misconception that the
crusader’s vow originated in that of the pilgrim probably stems from the
fact that in the surviving pontificals and manuals, dating from the third
quarter of the twelfth century onwards, actions, which earlier had been
separate, had become conflated, so that the cross, the sign of the crusader’s
vow, and the symbols of pilgrimage were received in a single ceremony.
Reading these led to the belief that there was at first little distinction made
between pilgrims and crusaders and that the rite for the taking of the cross
developed out of the ceremony for blessing the insignia of pilgrims.8 But
it can be demonstrated that in the earliest period of crusading there was
not one action, but two: the taking of the cross, which signified the mak-
ing of the vow,9 and the granting of the pilgrim’s purse and staff, which
5
Conciliorum Oecumenicorum Decreta, ed. Giuseppe Alberigo et al. (Freiburg,
1962), p. 168.
6
James A. Brundage, Medieval Canon Law and the Crusader (Madison, WI,
1969), p. 17.
7
Cyrille Vogel, ‘Le pèlerinage pénitentiel’, in Pellegrinaggi e culto dei santi in
Europa fino alla I crociata (Todi, 1963), pp. 37–94. For Jerusalem as a shrine
in which to die, see also Benedicta Ward, Miracles and the Medieval Mind
(London, 1982), pp. 124–25.
8
James A. Brundage, ‘Cruce Signari: the Rite for Taking the Cross in England’,
Traditio 22 (1966), 289–310; Brundage, Medieval Canon Law, pp. 119–20.
9
In the accounts of Pope Urban II’s sermon at Clermont by Robert the Monk and
Baldric of Bourgueil, the taking of the cross and the vow are closely related: The
Historia Iherosolimitana of Robert the Monk, ed. Damien Kempf and Marcus G. Bull
(Woodbridge, 2013), pp. 6–8; ‘Baldrici episcopi Dolensis Historia Jerosolimitana’,
105
would follow, perhaps some time later. The distinction between them
is illustrated in Eudes of Deuil’s description of the preparations of King
Louis VII of France for the Second Crusade. The king went through two
ceremonies, separated in time and space. He took the cross on 31 March
1146 at Vézelay, where a large gathering had assembled. It looks as though
there were two cross-takings on this occasion, a semi-private one for the
king and the greater nobles, at which Louis was given a cross sent by
the pope, and another – public and emotional – for the rest. Both were
marked by the attachment of cloth crosses. Over a year later, on 11 June
1147 at St Denis, Louis received the pilgrimage insignia, the purse and the
oriflamme (in place of a staff ), together with a blessing, from the pope.10
This sequence of events appears to have been typical. Although it
was possible to take the cross, like Louis, semi-privately,11 it was usually
assumed at large assemblies. The earliest of the public cross-takings, in
a field outside Clermont on 27 November 1095, immediately followed
Pope Urban’s proclamation of war12 and during his preaching tour there
were others presided over by him, including one possibly at Le Mans13
in RHC Occ. 4: 1–111 (here pp. 15–16). This is confirmed in the appeals from the
crusaders on the march to get laggards to fulfil their vows: Epistulae et chartae ad
historiam primi belli sacri spectantes, ed. Heinrich Hagenmeyer (Innsbruck, 1901),
pp. 142, 160, 165; and see 176. There can be absolutely no doubt from these that
the taking of the cross marked the vow; and it is borne out in the words put into
the mouth of Helias of La Flèche by Orderic Vitalis: The Ecclesiastical History of
Orderic Vitalis, ed. and trans. Marjorie Chibnall, 6 vols (Oxford, 1969–80), 5:
228–30, and those of Philip of Le Perche-Goet in 1140: Cartulaire des abbayes de
Saint-Pierre de la Couture et Saint-Pierre de Solesmes (Le Mans, 1881), p. 56. The
wording of an entry of 1146, in which the public nature of the wish is expressed but
there is only reference to the staff and purse and not to a cross, suggests that we are
here faced by a pilgrim, not a crusader: Cartulaires et chartes de l’abbaye d’Absie, ed.
Bélisaire Ledain (1895), p. 109.
10
Eudes of Deuil, De profectione Ludovici VII in orientem, ed. Virginia G. Berry
(New York, 1948), pp. 8–10, 16–18. Note the conjunction of the blessing and
the reception of the purse.
11
See Reginald of Durham, Libellus de Vita et Miraculis S. Godrici, Heremitae
de Finchale, ed. Joseph Stevenson (London, 1847), p. 33; Recueil des chartes
des l’abbaye de Cluny, ed. Auguste Bernard and Alexandre Bruel, 6 vols (Paris,
1876–1903), 5: 89.
12
Robert the Monk, Historia, ed. Kempf and Bull, pp. 6–8; Baldric of Bourgueil,
in RHC Occ. 4: 15–16; Fulcheri Carnotensis Historia Hierosolymitana (1095–
1127), ed. Heinrich Hagenmeyer (Heidelberg, 1913), pp. 138–43.
13
See Orderic Vitalis, Ecclesiastical History, 5: 228.
106
and certainly another at Tours.14 Many obviously had to take place with-
out him: at Rouen, the archbishop of which had been on pilgrimage to
Jerusalem thirty-eight years before – it was followed by a riot15 – and at
Etampes,16 during the siege of Amalfi17 and possibly at Verdun.18 They
also continued to occur throughout the century: for example at Tours
and Le Mans in the summer of 1128,19 during Archbishop Baldwin of
Canterbury’s preaching journey through Wales in 118820 and at Mainz
on Laetare Sunday of the same year; the Mainz cross-taking was another
which generated public disorder.21 In none of the early descriptions of
cross-taking are the pilgrimage insignia mentioned.
It seems that it was only after nobles and knights had taken the cross
that they would make private arrangements to receive the purse and staff,
and perhaps a blessing which certainly appears in the rites for pilgrim
symbols,22 from a local bishop, abbot or prior, sometimes associating this
with a financial arrangement with, or an endowment for, the religious
community concerned. Fulk Doon of Châteaurenard was given a napkin
and staff by the abbot of Lérins on 22 May 1096; the abbot enjoined the
14
‘Gesta Ambaziensium dominorum’ in Chroniques des comtes d’Anjou et des
seigneurs d’Amboise, ed. Louis Halphen and René Poupardin (Paris, 1913), p. 101.
Note the closeness of the language to that of Baldric of Bourgueil, RHC Occ. 4: 16.
15
Guibert of Nogent, De vita sua, ed. Edmond-René Labande (Paris, 1981), pp.
246–48. For the archbishop’s pilgrimage, see Orderic Vitalis, Ecclesiastical
History, 2: 68–72, 254.
16
La Chronique de Morigny (1095–1152), ed. Léon Mirot (Paris, 1909), p. 40.
17
Gesta Francorum et aliorum Hierosolimitanorum, ed. and trans. Rosalind Hill
(London, 1962), p. 7.
18
In 1100: Laurence of Liège, ‘Gesta episcoporum Virdunensium’, ed. Georg H.
Pertz, MGH SS 10: 497.
19
‘Chronica de gestis consulum Andegavorum: additamenta’, in Chroniques des
comtes d’Anjou, p. 161; Guigues A.M.J.A. d’Albon, Cartulaire général de l’ordre
du Temple (Paris, 1913–22), p. 9.
20
Gerald of Wales, ‘Itinerarium Kambriae’, in Giraldi Cambrensis Opera, ed. John
S. Brewer and James F. Dimock, 7 vols (London, 1861–91), 6: 13–16, 20, 48–49,
55, 61–62, 67, 72–74, 80–83, 104–5, 112–13, 119, 122–27, 136–37, 142–44,
146–47.
21
Historia de expeditione Friderici imperatoris, ed. Anton Chroust, in MGH
Scriptores rerum Germanicarum, n.s. 5 (Berlin, 1928), 14–15; Robert Chazan,
‘Emperor Frederick I, the Third Crusade and the Jews’, Viator 8 (1977), 83–93.
22
Brundage, ‘Cruce Signari’, pp. 297–310; Cyrille Vogel and Reinhard Elze, Le
pontifical romano-germanique du dixième siècle, 3 vols (Città del Vaticano,
1963–72), 2 : 227–29, 362.
107
journey to Jerusalem on him as a penance, which suggests that Fulk had

made confession and had had imposed on him as a formal penance the
obligation he had already vowed.23 Herbert II of Thouars wished to receive
the habitum peregrinationis from the bishop of Poitiers in 1100.24 Miles
of Vignory was given the purse at Vignory on 26 May 1100.25 It was at
times like these that a control envisaged at Clermont by the pope, who
had decreed that those who had taken the cross should seek a blessing
from their parish priests,26 should have taken effect. One example is that
of the female crusader Emerias of Altejas, who was obviously unsuited
for the expedition of 1096. She went to her bishop for his blessing and
was persuaded to commute her vow to the foundation of a hospice for
the poor.27 For obvious reasons there was no reference to cross-taking on
these occasions and the distinction between the two ceremonies may have
been preserved in some regions long after the two rites had been joined
together elsewhere: John of Joinville, for instance, described his recep-
tion of pilgrimage insignia, and apparently them alone, from the abbot
of Cheminon in 1248.28
The early distinction between the taking of the cross and the giving of
the symbols of pilgrimage has led me to conclude that the formal ceremony
for taking the cross did not grow out of existing pilgrimage rites, but was
a distinct, and obviously deliberate, accretion. There is no evidence, as
far as I am aware, that cross-wearing had been regularly associated with
pilgrims before the First Crusade,29 while the language that the pope was
23
Cartulaire de l’abbaye de Lérins, ed. Henri de Flamare (Nice, 1885), p. 312.
24
‘Cartae et Chronica prioratus de Casa-Vicecomitis’, in Chroniques des églises
d’Anjou, ed. Paul A. Marchegay and Emile Mabille (Paris, 1869), p. 341. For
Herbert’s crusade, see Riley-Smith, The First Crusade and the Idea of Crusading
(London, 1986), pp. 127–29, 131–32.
25
Chartes et documents de Saint-Bénigne de Dijon, ed. Georges Chevrier and
Maurice Chaume, 2 vols (Dijon, 1943–86), 2: 172. It is possible, from the
opening words of his charter (p. 171), that Miles of Vignory was a pilgrim
rather than a crusader.
26
See Robert the Monk, Historia, ed. Kempf and Bull, p. 7.
27
Histoire générale de Languedoc, ed. Joseph Vaissète, Claude Devic and Auguste
Molinier, 3rd edn, 16 vols (Toulouse, 1872–1904), 5: 756–58. Archbishop
Hugh of Lyons was to seek permission formally from Pope Paschal II: Hugh of
Flavigny, ‘Chronicon’, ed. Georg H. Pertz, in MGH SS 8: 487.
28
John of Joinville, Vie de Saint Louis, ed. Jacques Monfrin (Paris, 1998), p. 60.
29
Carl Erdmann was convinced that ‘this symbol was unquestionably an
innovation’: The Origin of the Idea of Crusade, trans. Marshall W. Baldwin and
Walter Goffart (Princeton, 1977), p. 345. But note the mos peregrini of Helias
108
reported using in 1095 suggests that he was doing something new,30 as

does the way the blessing and handing over of the cross was incorporated
into the later, conflated, rites.31 If the introduction of the cross was new
and the ceremony for taking the cross was at first generally distinct from
the reception of the symbols of pilgrimage then crusades and pilgrimages
were to be clearly distinguished from the start. This is certainly implied in
the demands of the first crusaders on the march that those who had taken
the cross should be forced to join them, because they were assuming that
other volunteers were clearly identifiable.32
Pilgrims and crusaders were penitents in a period when great stress
was being laid on the need for adequate rules for penance, but a second
important distinction between them was that crusaders were armed. This
was a radical departure from the rule that penitents were not normally
to carry weapons. Its strict observance had been somewhat mitigated by
Pope Gregory VII, who had believed that knights when penitent must
not pursue their calling, but had allowed for exceptions – in defence of
the penitent’s rights and those of his lord, friend, the poor and the church
– although only on the advice of an experienced religious mentor. This
mitigation was very limited, however, and it had been expressed in the
context of a much more rigorous enforcement of the general regulations
relating to penance.33 It is clear that right up to the First Crusade pilgrims
to Jerusalem were supposed to travel weaponless.34
of La Flèche, according to Orderic Vitalis (5: 230), although this would make
more sense if treated as the mos tante peregrinationis in the description of Fulk
of Anjou’s cross-taking in 1128: ‘Chronica de gestis consulum Andegavorum:
Additamenta’, p. 161.
30
Robert the Monk, Historia, ed. Kempf and Bull, pp. 6–8; ‘Baldrici episcopi
Dolensis Historia Jerosolimitana’, in RHC Occ. 4: 16.
31
Brundage, ‘Cruce Signari’, pp. 299–310, see especially p. 307 (the rite in the
Coventry Pontifical): ‘Benedictio crucis dande ituris in sanctam terram vel
ituris in aliam terram ad expugnandum inimicos crucis’.
32
See above, note 9.
33
H. E. John Cowdrey, ‘Pope Gregory VII and the Bearing of Arms’, in Montjoie.
Studies in Crusade History in Honour of Hans Eberhard Mayer, ed. Benjamin
Z. Kedar, Jonathan S.C. Riley-Smith and Rudolf Hiestand (Aldershot, 1997),
pp. 21–25.
34
Some historians, however, have appeared to believe that this was still an open
question. See Paul Alphandéry, La chrétienté et l’idée de croisade, ed. Alphonse
Dupront, 2 vols (Paris, 1954–59), 1: 26; Edmond R. Labande, ‘Pellegrini
o crociati? Mentalità e comportamenti a Gerusalemme nel secolo XII’, Aevum
109
This was at a time when the journey to Jerusalem was perceived to

be dangerous. Although the sea route was open, the primitive nature of
shipping, the expenses of the passage and the terror most westerners felt
when confronted by open water meant that very few would have taken that
option. But the land route was difficult enough. Once Constantinople had
been left behind everyone seems to have taken the old imperial highway
to the east, which ran to Antioch, from where the pilgrims travelled south
down the Syrian and Palestinian coasts, but that final stretch – by way
of Laodicea, Tripoli, Caesarea and Ramla to Jerusalem – was considered
to be a perilous one and in 1057 Norman pilgrims, who had reached
Antioch, debated whether to avoid it by taking a ship. This was, of course,
Muslim territory. An English hermit, who was persuaded by his bishop
not to make the pilgrimage, had envisaged his death at the hands of the
Muslims. Dhimma law tolerated Christianity, but only provided it was not
ostentatious and its rites were not performed publicly: in 1027 Richard of
Saint-Vanne was stoned when he said Mass openly and in c. 1040 Ulrich of
Breisgau, whose pilgrimage was eccentric – he took with him one servant
and one horse, which he never rode until he had recited the whole psalter
– was stoned by a hostile crowd near the river Jordan. In 1022 Gerald of
Thouars, the abbot of Saint-Florent-lès-Saumur, was taken prisoner and,
it was believed, martyred by Muslims before he reached Jerusalem. Forty-
three years later Bishop Gunther of Bamberg, who brought trouble from
Palestinian robbers on himself by the ostentatious richness of his train,
was warned in Laodicea that there were Muslim bandits ahead and his
large party was severely mauled near Caesarea; it was reported later that
two-thirds of his following did not return to the west. After 1071 condi-
tions must have deteriorated as Asia Minor was swept by marauding Turks
and Palestine was disputed by Saljūqs and Fātimids.
The stay in Jerusalem itself could also be dangerous. The atmosphere in
the city seems to have been tense in 1027. Stones were thrown by Muslims
into the Holy Sepulchre compound during Holy Week and the pilgrims
were worried by the number of armed men in the streets at the time of
the miracle of the Holy Fire. In 1055 the authorities expelled Christians
from the Sepulchre compound and closed the pilgrim roads: Lietbert of
Cambrai and his companions were held up for three months at Laodicea
and for a further two in Cyprus; they never reached their destination.35
Although one supposes that pilgrims could have hired escorts of mer-
cenaries to protect them I am not aware of them doing so and a feature
54 (1980), 217–30 (here p. 218); Giles Constable, ‘The Place of the Crusader in
Medieval Society’, Viator 29 (1998), 377–403 (here pp. 384–85).
35
Riley-Smith, The First Crusaders, pp. 36–38.
110
of the early twelfth century was to be the measures taken to afford them
security; the Templars, for example, were founded to protect the pilgrim
roads in Palestine.36 Nor is there any evidence that eleventh-century
pilgrims themselves carried arms. When in 1065 Bishop Gunther of
Bamberg and his followers were attacked, they agonised whether they
could resist their assailants at all, although in the end some fought back
with whatever they could get their hands on.37 Pilgrims continued to travel
without weapons even after conditions worsened from the 1070s onwards:
Robert I of Flanders travelled unarmed in 1086.38 That an unarmed pil-
grimage to Jerusalem was still the norm ten years later is demonstrated
in the chronicle of Monte Cassino in which one of the authors, who may
himself have been in Urban II’s entourage during the pope’s preaching
journey through France in 1095–96, gave as a reason for the crusade the
facts that ‘the princes […] could not do penance at home for their innu-
merable crimes and as laymen were very embarrassed to be seen keeping
company without weapons’.39 In other words the penances incurred by
west European magnates were likely to be so severe that they would have
to pilgrimage abroad and this entailed travelling without arms.
So in this respect, as in others, the First Crusade really was something
new. Even if the occasional eleventh-century pilgrim had carried arms
36
Guillaume de Tyr, Chronique, ed. Robert B.C. Huygens (Turnhout, 1986),
p. 554.
37
Lampert of Hersfeld, Opera, ed. Oswald Holder-Egger, MGH Scriptores
rerum Germanicarum 38 (Hannover, 1894), 94–98; Annales Altahenses
maiores, ed. Edmund L. B. von Oefele, MGH Scriptores rerum Germanicarum
4 (Hannover, 1891), 68–70; Marianus Scottus, ‘Chronicon’, ed. Georg Waitz,
in MGH SS 5: 559.
38
There seems to be no evidence whatever for Robert I of Flanders engaging in
military activities while in Constantinople on his return journey, although he
took some sort of oath, doubtless connected with his promise to send military
aid to the Byzantine Empire. See Charles Verlinden, Robert Ier le Frison, comte
de Flandre (Anvers, 1935), pp. 158–59. For a forged letter purporting to be from
the Emperor Alexios to him, which suggested that the Greeks had become so
desperate that they had been prepared to envisage Latin rule in Constantinople,
see Epistulae et chartae, no. I, pp. 129–36. Two scholars – Einar Joranson, ‘The
Problem of the Spurious Letter of Emperor Alexius to the Count of Flanders’,
American Historical Review 55 (1949–50), 811–32, and Michel de Waha, ‘La
lettre d’Alexis Comnène à Robert I le Frison’, Byzantion 47 (1977), 113–25 –
have argued that the letter is a genuine one which has been retouched, rather
than a total forgery.
39
‘Chronica monasterii Casinensis’, ed. Hartmut Hoffmann, MGH SS 34: 475.
111
he would have done so illegitimately. Now the church was positively

urging knights to undertake an armed pilgrimage. Pope Urban ‘imposed
on (the lords and subjects of Gaul) the obligation to undertake such a
military enterprise for the remission of all their sins’.40 And he claimed to
be ‘stimulating the minds of knights to go on this expedition, since they
might be able to restrain the savagery of the Muslims by their arms and
restore the Christians to their former freedom’.41 The crusade, however,
was not establishing any norm and once Jerusalem had been secured in
the battle of Ascalon the ordinary rules of penitential pilgrimage again
prevailed. This is demonstrated by Albert of Aachen’s report that those
crusaders who were going home rather than staying to continue the fight
carried few weapons with them. Most bore nothing but the palm fronds,
which were evidence that they had completed their pilgrimage,42 not-
withstanding the fact that the return journey, at least in its initial stages,
must have been as dangerous as the crusade itself had been, if not more
so now that Islam had been aroused. A graphic description of its perils
was provided by the Norman crusader Richard, son of Fulk of Aunou,
who was one of more than 1400 passengers crammed into a large ship
at Jaffa. Cruising north it ran into a storm and was wrecked off Tortosa,
which was in ruins and empty. The travellers ransacked it for what they
could find, but, concerned that they might be attacked by the Muslims,
nearly 100 of them boarded another boat which was Armenian and was
bound for Cyprus.43
This reversion to the normal rules for penitents is underlined in a
famous letter written in 1102 by Bishop Ivo of Chartres, the canonist, to
Pope Paschal II. Raimbold Croton, a hero of the First Crusade, had had
a monk in charge of an abbey estate close to his seized and castrated for
permitting, or probably ordering, the abbey servants to mow hay from
a field which Raimbold claimed was his own. Like so many returning
crusaders Raimbold had been caught up once more in the violence of
French rural life, in which disputes over rights and property often ended
in acts of insouciant force. Ivo described his decision and its consequences.
40
Epistulae et chartae, p. 136.
41
‘Papsturkunden in Florenz’, ed. Wilhelm Wiederhold, Nachrichten von der
Gesellschaft der Wissenschaften zu Göttingen. Phil.-hist. Kl. (1901), pp. 306–25
(here p. 313).
42
Albert of Aachen, Historia Ierosolimitana: History of the Journey to Jerusalem,
ed. and trans. Susan B. Edgington (Oxford, 2007), VI.54, p. 474.
43
‘Miracula S.Nicolai conscripta a monacho Beccensi’, in Catalogus Codicum
Hagiographicorum Latinorum antiquiorum saeculo XVI qui asservantur in
Bibliotheca Nationali Parisiensi, ed. Bollandists, 2 (Bruxelles, 1890), 427–29.
112
We took away Raimbold’s weapons and imposed a fourteen-year penance

on him, in such a way that he should abstain from the more luxurious
foods during the prescribed period […] He accepted this obediently, but
afterwards, having sent for many and great men to intercede for him, he
exhausted us with much earnest pleading that he be allowed the use of
arms on account of the way his enemies disturbed him.
Since Raimbold had lost a hand in the first assault on Jerusalem on 13 June
1099 it is highly unlikely that he could have used a sword effectively; the
point, of course, was that his status would have been affected if for four-
teen years he had to go about unarmed. Ivo, obviously pestered beyond
endurance, sent Raimbold on pilgrimage to Rome with this letter; he
wrote that the penance of a pilgrimage to Rome might induce the pope
to be merciful in this case.44
All penitents, therefore, were still barred from using weapons. An
interesting and late case is the great pilgrimage of Henry the Lion in
1172. Henry was perhaps accompanied by as many as 1500 persons: senior
churchmen, nobles, knights, ministeriales and servi. Arnold of Lübeck,
who may well have been one of the pilgrims, described the journey over-
land to Constantinope, from where Henry was to take a boat to Acre.
Finding their way barred by a Serbian force in Bulgaria, the German
pilgrims resorted to arms, but only reluctantly and somewhat inconclu-
sively, since the Serbs soon fled. At this point in his narrative Arnold put
a speech into Henry’s mouth in which he drew attention to the wholly
exceptional circumstances in which the pilgrims found themselves and
justified their action. ‘Since we are on pilgrimage it is our duty to journey
in peace and gentleness and for this reason we ought not to proceed under
military banners.’ On the other hand, resistance would be justified in this
case because of the threat they faced.45
But how can a general prohibition on the bearing of arms be reconciled
with the fact that after 1100 pilgrims to Palestine were quite often engaged
in military actions? One should first carefully distinguish from the rest
those individuals who must have been crusaders. The men who arrived
in a large fleet in May 1102 ut adorarent in Iherusalem, but took part in
44
Ivo of Chartres, ‘Epistolae’, in PL 162, cols 144–45. For the loss of Raimbold’s
hand, see Radulphi Cadomensis Tancredus, ed. Edoardo D’Angelo (Turnhout,
2011), p. 99.
45
Arnold of Lübeck, Chronica Slavorum, ed. Johann M. Lappenberg, MGH
Scriptores rerum Germanicarum 14 (Hannover, 1868), 10–31, esp. p. 16; Einar
Joransson, ‘The Palestine Pilgrimage of Henry the Lion’, in Medieval and
Historical Essays in Honor of James Westfall Thompson, ed. James L. Cate and
Eugene N. Anderson (Chicago, 1938), pp. 146–225.
113
the capture of Jaffa before going on to ‘fulfil their vows in Jerusalem’, were
almost certainly crusaders arriving late for the expedition of 1101.46 So
were the Pisan and Genoese sailors who arrived in 1104 causa adorandi
Iherusalem, but helped take Jubail in Lebanon and attack Acre.47 The
nature of the military operations in Lebanon is described in a charter of
January 1103 in which Raymond IV of Saint-Gilles, ‘prince, with God’s
aid, of the Christian knighthood on the Jerusalem journey’, gave the abbey
of St Victor of Marseilles half of Jubail in advance of its occupation; the
wording makes it clear that the crusade was not yet considered to be over
and would not be – however many vows had been fulfilled at the Holy
Sepulchre – until the coast was conquered and a line of communications
from Jerusalem north through Antioch to Asia Minor and beyond had
been secured.48 In 1108 Raymond’s son Bertrand and the Genoese, who
were finishing off the reduction of the Lebanese coast and had vowed to
serve ‘God and the Holy Sepulchre’, seem to have believed themselves to
be on crusade.49 The peregrini in the ships of the Venetian crusade which
arrived in 1123 were also crusaders.50
But the historians Fulcher of Chartres, who had settled in the East as
chaplain to Baldwin of Boulogne, the king of Jerusalem from 1100, and
Albert of Aachen, who was never in the East but had wanted to join the
First Crusade and had exceptionally good eye-witness sources for the
early years of the settlement, also referred to men who were not crusaders,
46
Albert of Aachen, Historia Ierosolimitana, IX.11, IX.18, ed. and trans.
Edgington, pp. 648, 658.
47
Albert of Aachen, Historia Ierosolimitana, IX.26, IX.28, ed. and trans.
Edgington, pp. 670, 672–74.
48
Cartulaire de l’abbaye de Saint-Victor de Marseille, ed. Benjamin Guérard, 2
vols (Paris, 1857), 2 : 151–53. See Jonathan S. C. Riley-Smith, ‘Raymond IV
of St Gilles, Achard of Arles and the Conquest of Lebanon’, in The Crusades
and their Sources. Essays presented to Bernard Hamilton, ed. John France and
William G. Zajac (Aldershot, 1998), pp. 1–8.
49
Albert of Aachen, Historia Ierosolimitana, XI.4–7, XI.17, XI.19, XI.32, ed.
and trans. Edgington, pp. 776–88, 790, 792, 806; Caffaro, ‘De liberatione
civitatum orientis’, ed. Luigi T. Belgrano, Annali Genovesi 1 (Genoa, 1890),
122–24; Guillaume de Tyr, Chronique, ed. Huygens, pp. 507–9.
50
Fulcheri Carnotensis Historia, ed. Hagenmeyer, p. 657. See Riley-Smith, ‘The
Venetian Crusade of 1122–1124’, in I Comuni italiani nel regno crociato di
Gerusalemme, ed. Gabriella Airaldi and Benjamin Z. Kedar (Genova, 1986),
pp. 337–50.
114
but were obviously pilgrims prepared to fight.51 Fulcher of Chartres stated

that in 1105 the Egyptians were planning an invasion ‘because we were so
few and were without the help of the usual pilgrims’.52 And he wrote with
respect to 1113 that ‘it is usual in these overseas parts for our army to grow
daily at this time, because the pilgrims have arrived’.53 It is possible that a
party of English, Flemings and Danes, which arrived causa adorandi in
Iherusalem in 1106 and after visiting the holy places helped prepare an
attack on Sidon, was made up of pilgrims.54 So were probably the follow-
ers of the king of Norway in 1110. In the past I have assumed that these
comprised a delayed element of Bohemund of Taranto’s crusade of 1107.55
I have now come to conclusion that they were not. They arrived desiderio
adorandi in Iherusalem, according to Albert of Aachen, and ut Hierusalem
peregrinarentur, according to Fulcher of Chartres.56 They agreed to assist
the kingdom militarily only after they had been to Jerusalem, because
Christ had ordered ‘his faithful followers first to seek the kingdom of God
and afterwards to find all the beneficial things they sought’.57 Pilgrims,
therefore, seem to have had access to weapons, but their military service
conformed to a particular pattern. They do not seem to have volunteered
for military action until after they had fulfilled their religious obligations
by visiting the holy places in and around Jerusalem. Fulcher of Chartres’s
reference to pilgrim military assistance ‘at this time’ seems to refer to July
1113, by which date pilgrims arriving on the Easter passage would have
already seen the holy places. In 1153, when the kingdom’s army was laying
siege to Ascalon, pilgrims were forbidden to return home before they
51
Compare what follows with the references to entirely pacific examples of
pilgrimage after 1102 in Fulcheri Carnotensis Historia, ed. Hagenmeyer,
pp. 435, 670–71, 761, 799, 805; Albert of Aachen, Historia Ierosolimitana,
XII.10, XII.25, XII.33, ed. and trans. Edgington, pp. 838, 862, 880.
52
Fulcheri Carnotensis Historia, ed. Hagenmeyer, p. 490. See also Guillaume de
Tyr, Chronique, ed. Huygens, p. 498.
53
Fulcheri Carnotensis Historia, ed. Hagenmeyer, p. 575.
54
Albert of Aachen, Historia Ierosolimitana, X.1, X.7, ed. and trans. Edgington,
pp. 718, 724.
55
Riley-Smith, The First Crusaders, p. 241.
56
Albert of Aachen, Historia Ierosolimitana, XI.27, ed. and trans. Edgington,
p. 800; Fulcheri Carnotensis Historia, ed. Hagenmeyer, pp. 543–48. Guillaume
de Tyr, Chronique, ed. Huygens (p. 517), writing much later, seems to have
believed that they were crusaders.
57
Albert of Aachen, Historia Ierosolimitana, XI.30, ed. and trans. Edgington,
p. 802.
115
had supplemented the besieging force58 and in October 1183, when the
departure of the passage for home was near, pilgrims, who must, again,
have already been to Jerusalem, were called upon to serve as footsoldiers
in an army facing Saladin in the valley of Jezreel.59
It may well have been the case, therefore, that from 1100 onwards pil-
grims felt free to take part in warfare only after they had visited the holy
places; in this they behaved in precisely the opposite way to so many of the
first crusaders, who in 1099 had self-consciously reverted to an unarmed
state once they had liberated the Holy Sepulchre. Arnold of Lübeck’s
account of Henry the Lion’s pilgrimage throws light on a related question.
It is not always clear whether twelfth-century pilgrims brought weapons
with them or were armed once they had arrived; with Palestine now in
Christian hands there must have been stockpiles at hand. In Henry the
Lion’s case it is certain that his company carried arms, even if it did not
intend to use them.
At any rate, while Jerusalem was in Christian hands, there appear to
have been, broadly speaking, two patterns of behaviour. Those on crusade,
such as the soldiers in 1101 and 1147–48, fought on the march to the East
and again, if need be, once they had visited the holy places. Those who
came as pilgrims pure and simple only felt free to take up arms after they
had achieved their goal by venerating the Holy Sepulchre.
58
Guillaume de Tyr, Chronique, ed. Huygens, p. 793.
59
Guillaume de Tyr, Chronique, ed. Huygens, p. 1053.
116
1096 and the Jews
A Historiographic Approach
Judith Bronstein
Universiy of Haifa
In the spring of 1096 bands of crusaders – poor people as well as experienced

knights – attacked and injured the Jewish communities of Speyer, Mainz,
Worms, Cologne, Metz, Trier, Regensburg, and Prague. Jews were massa-
cred, their property was despoiled and destroyed. Faced with the choice of
baptism or death, many Jews converted to Christianity; but many others
chose to take their own lives and those of their families and friends in the
sanctification of God’s name. These terrible events, set forth in twelfth-
century chronicles, have attracted much attention in modern historiography.
This article seeks to study the most recent historiographic trends in the
subject while probing the different methods and the approaches behind
them. To what degree does contemporary research narrow the gap between
crusades historiography and Jewish history? To what degree can knowledge
of the tragic fate of the Jews contribute to our understanding of the First
Crusade and the ideology of its participants? A brief perusal of the most
recent studies on the subject may add new perspectives on these matters.
The 1096 massacres are described in three unique Hebrew narrative
sources: the so-called Mainz Anonymus, or The Narrative of the Old
Persecutions, which seems to have been written close to the events; the
Solomon bar Samson Chronicle, an edited work dating to the 1140s and
composed of fragments of diverse contemporary accounts; and the so-
called Eliezer bar Nathan Chronicle, which, as recently argued, was based
on the Solomon bar Samson Chronicle and written by a well known mid-
to late twelfth-century poet and legalist of that name after the Second
Crusade. The last two chronicles concentrate on the attacks and the
eventual fate of the Jewish communities in Mainz, Cologne, Xanten and
Trier; whereas the Mainz Anonymus, focuses mainly on the attacks on
the communities in Speyer, Mainz, and Worms. Robert Chazan dated the
Mainz Anonymus as the oldest of the three narratives, composed close to
the time of the events and as such highly valuable. It describes the spec-
trum of crusaders’ behaviour towards the Jews and Jewish responses, as
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. 11–11.
FHG DOI: 10.1484/M.OUTREMER-EB.1.102320
Judith Bronstein
well as the reactions of the local Christian communities to the attacks.1

These three chronicles are considered the earliest extant examples of local
Jewish historical writing in medieval Europe.2 Besides narrative sources,
the Jews massacred in 1096 were memorialised in lists of victims and in
piyyutim, namely liturgical poetry.3
1
The chronicles were first published in Hebräische Berichte über die
Judenverfolgungen während der Kreüzzuge, ed. Adolf Neubauer and Moritz
Stern (Berlin, 1892) and re-edited by Sefer Gezerot Ashkenaz ve-Zarfat, ed.
Abraham Habermann ( Jerusalem, 1945). For the latest edition and translation
of the chronicles see ‘Hebräische Berichte über die Judenverfolgungen während
des Ersten Kreuzzugs’, ed. Eva Haverkamp, in Hebräische Berichte über die
Judenverfolgungen während des Ersten Kreuzzugs. Hebräische Texte aus
dem mittelalterlichen Deutschland, MGH and Israel Academy of Sciences
1 (Hannover, 2005). For another English translation see The Jews and The
Crusades: The Hebrew Chronicles of the First and Second Crusades, ed. and trans.
Shlomo Eidelberg (Madison, WI, 1977). For English translations of the Mainz
Anonymous and the Chronicle of Solomon bar Samson see also Robert Chazan,
European Jewry and the First Crusade (Berkeley, 1987), pp. 223–97. For a
partial listing of Jewish and Christian sources and short quotations from them,
Heinrich Hagenmeyer, Chronologie de la première croisade, 1094–1100 (Paris,
1902), pp. 13–28 (first published in the Revue de l’Orient Latin 6 (1898), 228–
41). For the authorship, dating and interrelationship of the chronicles see Anna
Sapir Abulafia, ‘The Interrelations between the Hebrew Chronicles of the First
Crusade’, Journal of Semitic Studies 27 (1982), 221–39; Robert Chazan, God,
Humanity and History: The Hebrew First Crusade Narrative (Berkeley, 2000),
pp. 28–111; Chazan, ‘Latin and Hebrew Chronicles: Some Shared Themes’, in
The Medieval Crusade, ed. Susan J. Ridyard (Woodbridge, 2004), pp. 19–20;
Hebräische Berichte, ed. Haverkamp, pp. 231–40.
2
Jeremy Cohen, ‘A 1096 Complex? Constructing the First Crusade in Jewish
Historical Memory, Medieval and Modern’, in Jews and Christians in Twelfth
Century Europe, ed. Michael A. Singer and John Van Engen (Notre Dame, IN,
2001), p. 10. An earlier example of an extant Jewish narrative may have been
the so-called ‘1007 Anonymous’, but it has been dated by Kenneth Stow to
the thirteenth century: The ‘1007 Anonymous’ and Papal Sovereignity: Jewish
Perceptions of the Papacy and Papal Policy in the High Middle Ages (Cincinnati,
1984), pp. 26–33.
3
Abraham Grossman, ‘Shorashav shel Qiddush ha-Shem be-Ashkenaz
ha-Qedumah’, in Sanctity in Life and Martyrdom: Studies in Memory of Amir
Yekutiel, ed. Isaiah M. Gafni and Aviezer Ravitzky ( Jerusalem, 1992), pp. 101–3
(in Hebrew); Robert Chazan, In the Year 1096. The First Crusade and the
Jews (Philadelphia, 1996), pp. 107–8. For references to the 1096 massacres in
non-Ashkenazi and in early modern Jewish sources see Chazan, In the Year
118
1096 and the Jews: A Historiographic Approach
Most Jewish sources offer detailed descriptions of the events of 1096, and
the Jewish responses to it. In recent years, historians have also paid more
attention to the attitudes expressed in Christian sources, and found that,
although not as rich and detailed as the Jewish sources, numerous local
Christians annalists and chroniclers also include passages on the waves of
violence throughout 1096 and the forcible conversion of the Jews.4
The most complete Latin account can be found in Albert of Aachen’s
Historia Hierosolymitana. Albert gives a vivid and terrible account of the
slaughter of the Jews in Cologne and Mainz, perpetrated by crusaders
and the citizens of Cologne and by the followers of Emich of Flonheim.
Although Albert’s main interest focuses on the crusaders’ behaviour, he
also portrays the diversity of the Jews’ responses: their flight, their attempts
to find refuge at the palace of Rothard, bishop of Mainz, and their efforts
to resist the attacks (about which he gives no details). Albert was appalled
by the decision of some of the Jews to take their own lives:
The Jews, indeed, seeing how the Christian enemy were rising up against
them and their little children and were sparing none of any age, even turned
1096, pp. 116–20; Simon Schwarzfuchs, ‘The Place of the Crusades in Jewish
History’, in Culture and Society in Medieval Jewry: Studies Dedicated to the
Memory of Haim Hillel Ben-Sasson, ed. Menahem Ben-Sasson, Robert Bonfil
and Joseph R. Hacker ( Jerusalem, 1989), pp. 251–67 (in Hebrew).On piyyutim
see Susan L. Einbinder, Beautiful Death: Jewish Poetry and Martyrdom in
Medieval France (Princeton, 2002). This important book examines the
evolution and roles of a large corpus of Jewish martyrological poetry –
liturgical laments and penitential hymns – from northern France, mainly from
the laments for the martyrs of Blois (1171) onwards.
4
For a description of the Latin sources see Benjamin Z. Kedar, ‘The Forcible
Baptisms of 1096: History and Historiography’, in Forschungen zur Reichs-, Papst-
und Landesgeschichte: Peter Herde zum 65. Geburtstag von Freuden, Schülern
und Kollegen dargebracht, ed. Karl Borchardt and Enno Bünz, 2 vols (Stuttgart,
1998), 1: 187–200. For additional listings of Christian local sources with short
references to the 1096 persecution see Hagenmeyer, Chronologie, pp. 17, 19, 23–25;
Jonathan Riley-Smith, The First Crusade and the Idea of Crusading (London,
1986), p. 176, nn. 88, 89; Chazan, God, Humanity and History, pp. 132–33. For
a positive assessment of the value of the Christian sources in describing these
events see Rudolf Hiestand, ‘Juden und Christen in der Kreuzzugspropaganda
und bei den Kreuzzugspredigern’, in Juden und Christen zur Zeit der Kreuzzüge,
ed. Alfred Haverkamp (Sigmaringen, 1999), pp. 153–208. For a categorisation
of the Christian sources according to extent and content see David Malkiel,
Reconstructing Ashkenaz: The Human Face of Franco-German Jewry, 1000–1250
(Stanford, 2009), pp. 80–86. For more on Malkiel’s approach see below.
119
Judith Bronstein
upon themselves and their companions, on children, women, mothers and

sisters, and they all killed each other. Mothers with children at the breast –
how horrible to relate – would cut their throats with knives, would stab
others, preferring that they should die thus at their hands, rather than be
killed by the weapons of the uncircumcised.
Albert concluded that very few Jews survived these attacks, and many
of those who did had converted, ‘rather through fear of death than for
love of the Christian religion’.5 The attackers took great plunder from the
Jews. But Albert saw in their greed one of the reasons for their downfall.
He explained the defeat of Peter the Hermit’s followers at Niš as God’s
punishment against the pilgrims (crusaders) ‘who had sinned in His eyes
by excessive impurities and fornicating unions, and had punished the
exiled Jews (who are admittedly hostile to Christ) with a great massacre,
rather from greed for their money than for divine justice, since God is
a just judge and commands no one to come to the yoke of the Catholic
faith against his will or under compulsion’.6 Although not as rich as the
Jewish chronicles, Albert of Aachen’s is the most detailed Latin account
of the events. Another twelfth-century crusader chronicler, Ekkehard of
Aura, also highlights the role of Emich of Flonheim and his followers in
the assaults. He notes in passing that when Emich’s followers
were led through the cities of the Rhine, the Main and the Danube, they
either utterly destroyed the execrable race of the Jews wherever they found
them […] or endeavoured to force them into the bosom of the Church.7
Other Latin sources include annals, chronicles and gesta. Eva Haverkamp
has shown that most of these are monastic writings from the Regnum
Teutonicum, and that they are highly informative sources. The fact that
the information was incorporated in sources from geographically diverse
areas, some of them remote from the locations where the events took place,
attests to the interest in the topic; an interest which started immediately
after 1096 but continued well into the thirteenth century. Using careful
analysis of the context of these local Latin sources, and their interrelation-
ship, Haverkamp is able to reconstruct the way that information relating
5
ed. and trans. Susan B. Edgington (Oxford, 2007), 1.26–27, pp. 50–53.
6
Albert of Aachen, Historia Ierosolimitana, I.29, ed. and trans. Edgington, pp. 56–59.
7
Ekkehard of Aura, ‘Hierosolymita’, in RHC Occ. 5: 20. On Ekkehard of Aura
and especially Albert of Aachen’s writing see Susan Edgington, ‘The First
Crusade: Reviewing the Evidence’, in The First Crusade: Origins and Impact,
ed. Jonathan Phillips (Manchester, 1997), pp. 60–73. On Emich of Flonheim
see also Stow’s article below.
120
to the 1096 persecution was disseminated and copied. She focuses on the
role of the existing monastic networks in this process and examines the
influence of local circumstances, and agendas. These local sources are not
only important with regard to the events of 1096, but they also shed light
on the attitudes of Christian writers to the Crusade and its participants.8
Modern historiography of the crusades has tended to refer to the 1096
massacres as just one of many aspects of the First Crusade. Contemporary
writers and modern scholars who concentrated on the crusades were
generally more interested in the perpetrators and their motivation than
the victims’ fate. Taking motivation as the criterion for categorisation,
Benjamin Kedar has identified six major approaches in crusader histo-
riography with regard to the 1096 massacres. The first two approaches,
attributed mainly to historians of the seventeenth to the nineteenth
century, make moralistic evaluations: the first criticises the crusader
movement as such, and sees the massacres as an expression of the fanati-
cism of its members; the second glorifies the crusades but condemns the
attackers. The other four approaches reflect the views of professional
historians, who in their attempts to reconstruct the past differ in their
identification of the motives: the perpetrators were motivated by greed,
by the Jews’ economic activities, by the Investiture Controversy, and/or
by their own ideology.9
Ideology as a motive for action has become predominant among
crusade historians, especially owing to the prolific writings of Jonathan
Riley-Smith, who has dealt specifically with the relationship between
the crusades and the persecution of 1096, combining Latin and Hebrew
sources. In an article published in 1984 Riley-Smith contends that the
assaults were conducted not by gangs of undisciplined peasants but by
armies made up of poor people as well as crusaders, all led by experienced
knights. Their motives were material and spiritual. Though looting and
material extortion from the Jews were not the crusaders’ goal, they were
a customary way to supply an ill-equipped army. Riley-Smith sees as the
main reason for the anti-Jewish violence the crusaders’ desire to avenge
the crucifixion of Christ. In their wish for vengeance against the enemies
of the faith, they failed to make a clear distinction between the Jews and
8
Eva Haverkamp, ‘What Did the Christians Know? Latin Reports on the
Persecutions of Jews in 1096’, Crusades 7 (2008), 59–86.
9
Benjamin Z. Kedar, ‘Crusade Historians and the Massacre of 1096’, Jewish
History 12 (1998), 11–31. For a bibliographical survey of the First Crusade
see Alan V. Murray, ‘Bibliography of the First Crusade’, in From Clermont to
Jerusalem: The Crusades and Crusader Societies, 1095–1500, ed. Alan V. Murray
(Turnhout, 1998), pp. 267–310.
121
Judith Bronstein
the Muslims, against whom the crusaders were called to fight in order to
avenge the sacrilegious occupation of Christ’s lands. Revenge against the
Jews was not a part of crusading ideology, as propounded by educated
crusader preachers. It was the idea of vengeance manipulated by popular
preachers, who might also call for a missionary war. This could have led to
a misconception of the crusade as a war for the expansion of Christianity
and incited some crusaders, such as Emich, driven by the general hysteria
and possibly also by eschatological expectations, to force conversion on
the Jews. Yet Riley-Smith is not convinced that conversion was a main
aim. Forced baptism was against Canon Law and contrary to the teach-
ing of the Church. If the crusaders misinterpreted the message and set
out to fight for the expansion of Christianity, more incidents of forced
mass conversion in the course of the First Crusade would be expected,
but Riley-Smith found none on such a scale.10
In a study dedicated to forced baptism, Benjamin Kedar challenges
these views. He asserts that in Latin and Hebrew sources alike there is
ample evidence that in 1096 the Jews were given the choice of baptism
or death; accordingly, their Christianisation was one of the crusaders’
major goals. Kedar concludes that most historians of the crusades tend to
ignore or underestimate this goal. Kedar ascribes this contrast between the
centrality of forced baptism in Latin and Hebrew sources and its under-
estimation by most historians to the dependence of early historians on
inadequate sources such as William of Tyre and Albert of Aachen, from
whose writings it is difficult to grasp the importance of conversion. He
also notes the tendency of later historians to consider the perpetrators
as participants in the ‘Peasants’ Crusades’, interested only in plunder and
murder, hence devoid of spiritual motives. Against Riley-Smith’s view
that forced conversion contradicted Canon Law, Kedar argues that Jews
10
Jonathan Riley-Smith, ‘The First Crusade and the Persecution of the Jews’,
Studies in Church History 21 (1984), 51–72; Riley-Smith, The First Crusade
and the Idea of Crusading, pp. 49–57. The view that greed was the real motive
behind the attacks is expressed, for example, by Hans E. Mayer, The Crusades,
2nd edn (Oxford, 1988), pp. 40–41. Examining the motivation of the crusaders
in general, Riley-Smith argues that owing to the enormous expenses incurred
by the crusaders and their families the material explanation is not satisfactory:
The First Crusaders, 1095–1131 (Cambridge, 1997), pp. 109–35. On the cost of
crusading in its early stages see also Riley-Smith, ‘Early Crusaders to the East
and the Cost of Crusading, 1095–1130’, in Cross Cultural Convergences in the
Crusader Period: Essays Presented to Aryeh Grabois on His Sixty-Fifth Birthday,
ed. Michael Goodich, Sophia Menache, and Sylvia Schein (New York, 1995),
pp. 237–59.
122
were coerced into Christianity as early as in the seventh century, when

ecclesiastical and secular authorities presented the Jews with the alterna-
tive of baptism or expulsion. Attempts at Christianisation were principally
made on Jewish children, and in the decade before 1096 they may have
been prompted by eschatological expectations. Kedar sees in the baptism-
expulsion motive a precedent to the baptism-death choice which the Jews
faced in 1096. Their conversion was probably seen by some contemporaries
as the goal of the crusade. Kedar claims that in the embryonic state of the
crusade movement in 1096 there was no single normative concept of the
crusade; there were many, and propagation of the Christian faith was one
of them. The fact that this conception did not lead to similar massacres
in Asia Minor and Syria is explained by him mainly by the annihilation
in Hungary of the armies which upheld these views.11
Forced baptism is a central topic in modern historiography both of
the crusades and of Jewish history in recent years. David Malkiel claims
that the option of baptism was presented to the Jews only after the first
deadly attacks. After analysing Christian sources, he concludes that the
attackers’ aim was slaughter. Conversion was only a by-product of the
assaults, and availed of primarily on an individual basis. The Hebrew
sources also mentioned the choice of baptism, but they too considered
slaughter to be the perpetrators’ main aim. Baptism was offered as an
alternative to certain individuals: those who survived a massacre ‘in
cases like this, the enemy offers conversion both because vengeance has
already been achieved and because of the humanising effect of the close
encounter’.12 Other individuals to whom baptism was offered were Jews
with whom local Christians had personal relations. Children were an
important exception to the perpetrators aim. Both Latin and Hebrew
sources indicate that they were expected to be baptised. This and the
motive of murder rather than baptism, increase, according to Malkiel, the
credibility of the Hebrew narratives ‘because they would have highlighted
the heroism of the martyrs by depicting the Jews as rejecting an offer of
baptism and submitting to martyrdom voluntarily’.13 Kenneth Stow has
argued that forced baptism was a central topic in medieval thought, as it
sparked great concern and fear in ecclesiastical circles. It challenged the
equilibrium which papal and canonical circles wished to establish in the
relations between Christians and Jews and could result in apostasy, namely
11
Kedar, ‘The Forcible Baptisms of 1096’, pp. 87–88.
12
Malkiel, Reconstructing Ashkenaz, p. 92; Malkiel, ‘Destruction or Conversion:
Intention and Reaction, Crusaders and Jews, in 1096’, Jewish History 15
(2001), 257–81.
13
Malkiel, Reconstructing Ashkenaz, p. 86.
123
Judith Bronstein
a return to Judaism. ‘This ‘backsliding’, as it was called, was opprobrious.

It led to the violation not only of the canons forbidding apostasy but
also of the canons protecting the Jews and their right to preserve their
religion’.14 Count Emich of Flonheim is portrayed in both Christian and
Jewish chronicles as the chief perpetrator of anti-Jewish violence and
forced baptism in 1096. Nonetheless, Stow shows Emich to be a symbol
of contemporary society, carried away by ecstasy, while the chroniclers
who criticise him point to a Church in search of discipline and control
and fearful of a body of relapsed converts.15
The modern historical memory of the 1096 persecution is David Malkiel’s
starting point in his ‘Jews and Apostates in Medieval Europe: Boundaries
Real and Imagined’, in which he challenged prevailing views among
modern scholars regarding apostasy in medieval Ashkenaz (Franco-
German Jewry). The memory of the First Crusade, according to Malkiel,
is crucial for modern Jewish historiography. The latter, by referring on
the one hand to the scarce mention of apostasy in the Hebrew chroni-
cles and on the other hand to the same chronicles’ emphasis on mass
voluntary martyrdom, emphasises the heroic behavior of the Jews. This
historical memory of Ashkenazi Jews contrasts with that of the Iberian
Jews in the fifteenth century, many of whom, faced with similar threats
and dilemmas, opted for baptism. This contrast has been explained in
modern historiography by the high level of cultural, economic, social
and political integration of the Iberian Jews into Christian society, and
the insularity of the Ashkenazi Jews. By showing that the number of cases
of apostasy in medieval Ashkenaz were much higher than previously
accepted, Malkiel is part of a current historiographic debate which ques-
tions the reliability of the Hebrew chronicles of the First Crusade, includ-
ing their portrayal of apostasy as rare. This, as is emphasised by Malkiel, is
of extreme importance, ‘because it undermines the Ashkenaz-Sepharad
dichotomy: one can no longer reduce the medieval experience to a binary
structure of cultural engagement versus insularity’. Malkiel shows that
apostasy and apostates were a common feature of Jewish communities
in Ashkenaz. Conversion was not only the result of gruesome coercion
and religious tension, as it is portrayed in modern historiography, but
14
Kenneth Stow, ‘Conversion, Apostasy and Apprehensiveness: Emicho of
Flonheim and the Fear of Jews in the Twelfth Century’, Speculum 76 (2001),
911–33. For the view that forcible conversion was the perpetrators’ major
objective see also Jean Flori, ‘Une ou plusieurs “première croisade”? Le message
d’Urbain II et les plus anciens pogroms d’Occident’, Revue historique 285
(1991), 3–27 (here pp. 9–12).
15
Stow, ‘Conversion, Apostasy and Apprehensiveness’, pp. 911–33.
124
was due mostly to ‘a high degree of social and cultural intimacy with
their Christian neighbours’.16
The central issues of ideological motive and forcible conversion have
been reconsidered by Riley-Smith, who, rethinking his past positions,
proposes a different approach. Anti-Jewish violence on the scale mani-
fested in Europe was the result of the internalisation of the concept of
holy war in an attempt to purify and unify Christian society. The per-
secution of the Jews had no theoretical basis. Nothing in the traditions
of Christian violence or crusade theory could have justified it. Yet holy
war – proclaimed against an external enemy – tends to turn inwards, ‘to
be directed against the members of the very society that generated it’.17
Riley-Smith gives several examples of religious movements that demanded
purification of their society and unification of their members around the
same religious ideals as a prerequisite for a successful external war. As
often declared by Church leaders, the crusade movement also required
the reform of the Church and Christendom together with religious uni-
formity. Forced conversion of the Jews was not, as Riley-Smith had been
inclined to believe, a manifestation of millenarianism but the result of
pressure to create a uniform Christian society by eliminating the alien –
by baptism or by death. Although the use of force against the Jews was
forbidden by the leaders of the Church, charismatic preachers pointed
to the Jews as an alien group against whom Christian violence should
be directed. Moreover, by careful analysis of the process of recruitment
and the techniques used by crusader preachers, Riley-Smith shows that
in order to induce their audiences to take the cross, the preachers had to
whip them into such an emotional state that once unleashed they were
out of control.18
While historians of the crusades tend to focus on the motivation of
the attackers, most scholars of Jewish history deal with the victims. The
1096 massacres have attracted a huge scholarly interest among historian
specialising in Jewish history. Because of the deep significance attached
to them for past and contemporary Jewish memory,19 these massacres
16
David Malkiel, ‘Jews and Apostates in Medieval European – Boundaries Real
and Imagined’, Past and Present 194 (2007), 3–34; on the Ashkenazi-Sepharadi
dichotomy see also Malkiel, Reconstructing Ashkenaz, pp. 254–61.
17
Jonathan Riley-Smith, ‘Christian Violence and the Crusades’, in Religious
Violence between Christians and Jews: Medieval Roots, Modern Perspectives, ed.
Anna Sapir Abulafia (New York, 2002), pp. 3–21 (here p. 8).
18
Riley-Smith, ‘Christian Violence and the Crusades’, pp. 16–17.
19
For the centrality of the 1096 in contemporary Jewish memory see David
Nirenberg, ‘The Rhineland Massacres of Jews in the First Crusade: Memories
125
Judith Bronstein
aroused a wide-ranging discussion, in some cases highly emotional. Jewish

historiography on 1096 has been categorised by three major tendencies.
The first focuses on periodisation, that is, on a chronological division of
research influenced by major historical events and by political and philo-
sophical trends: in the nineteenth century by the Jewish enlightenment
(Haskalah), and later by Zionism; during the twentieth century by the
Holocaust and the creation of the Jewish State. The second tendency is
defined by the debate between most of the above scholars who, regardless
of the events that shaped their writing or their aims, consider the massacres
of 1096 as a turning point in Jewish history, and a number of historians
who argue against the magnitude and impact attributed to them. The third
and most recent tendency discusses the factual accuracy of the Hebrew
chronicles of the First Crusade. While some historians consider them
factual descriptions of historical events, the latest trend examines them as
literary texts. As such, the historical, emotional and cultural circumstances
that influenced these writings are examined together with the ideological
and educational messages they wished to convey, mainly that death in the
sanctification of God’s name (Mavet al kidush ha-Shem) should serve as
a model of behaviour for future generations.20
Medieval and Modern’, in Medieval Concepts of the Past: Ritual, Memory,

Historiography, ed. Gerd Althoff, Johannes Fried and Patrick J. Geary
(Washington, DC, 2002), pp. 279–309.
20
For categorisations and historiographic evaluations see Jeremy Cohen,
Sanctifying the Name of God: Jewish Martyrs and Jewish Memories of the First
Crusade (Philadelphia, 2004), pp. 31–55; Jeremy Cohen, ‘Between History
and Historiography: On the Study of the Persecution and the Determination
of Their Significance’, in Facing the Cross: The Persecutions of 1096 in History
and Historiography, ed. Yom Tov Assis et al. ( Jerusalem, 2000), pp. 16–31 (in
Hebrew); Chazan, European Jewry, pp. 192–222; Chazan, In the Year 1096, pp.
120–70; Schwarzfuchs, ‘The Place of the Crusades in Jewish History’, pp. 251–
55. On the historians of the Wissenschaft des Judenthums school see also Nils
Roemer, ‘Wissenschaft des Judenthums and the Martyrs of 1096’, Jewish History
13 (1999), 65–81. For additional historiographic studies which refer indirectly
to the 1096 massacres see Kenneth Stow, ‘Ha-Knesia ve-Historiographia
Neutralit’, in Studies in Historiography: Collected Essays, ed. Joseph Salmon
et al. ( Jerusalem, 1987), pp. 101–15 (in Hebrew); Jeremy Cohen, ‘Recent
Historiography on the Medieval Church and the Decline of European Jewry’,
in Popes, Teachers and Canon Law in the Middle Ages: Essays in Honor of Brian
Tierney, ed. James R. Sweeney and Stanley Chodorow (Ithaca, NY, 1989),
pp. 251–62.
126
Mavet al kidush ha-Shem is the focal point of Jewish historiography on

1096. Most nineteenth- and twentieth-century historians highlight the
heroic nature of Jewish martyrdom as a manifestation of the Jews’ devo-
tion to their faith.21 Nevertheless, while some historians find its roots in
a set of Jewish customs and traditions, others consider it to be the result of
cross-cultural interaction.22 In 1961 Jacob Katz saw Jewish martyrdom as
influenced by Christian religious awakening, such as the Cluniac move-
ment. And yet, by sacrificing their lives, the Jews also expressed their
exclusiveness and their rejection of Christian values.23 In order to reject
21
Yitzhak Baer, ‘Gezerat Tatn’u’, in Sefer Asaf, ed. Moses D. Cassuto et al.
( Jerusalem, 1953), pp. 136–37 (in Hebrew); Hayim H. Ben-Sasson, ‘Effects of
Religious Animosity on the Jews’, in A History of the Jewish People, ed. Hayim
H. Ben-Sasson (Cambridge, MA, 1976), pp. 414–18.
22
See, for example, Grossman, ‘Shorashav shel Qiddush ha-Shem be-Ashkenaz
ha-Qedumah’, pp. 105–30; Salo W. Baron, A Social and Religious History of
the Jews, 2nd edn, 18 vols (New York, 1952–83), 4: 95–96; Yitzhak Baer, ‘The
Religious-Social Tendency of “Sefer Hasidim”‘, Zion 3 (1937), 1–50; Baer,
Studies in the History of the Jewish People, 2 vols ( Jerusalem, 1985), 2: 157–59;
Baer, ‘Gezerat Tatn’u’, pp. 136–37 (all in Hebrew); Hayim H. Ben-Sasson,
‘Effects of Religious Animosity on the Jews’, in A History of the Jewish People,
ed. Ben-Sasson pp. 414–18.
23
Katz, Exclusiveness and Tolerance, pp. 90–93. This view of martyrdom as a
manifestation of rejection was also expressed by Kenneth Stow, Alienated
Minority: The Jews of Medieval Latin Europe (Cambridge, MA, 1992), p. 116.
Yisrael Yuval, however, suggested that martyrdom derived from messianic
beliefs, whereby the martyrs hope that their sacrifice will hasten not only
God’s redemption but also the destruction of the enemies of the Jewish people.
These eschatological beliefs, Yuval maintains, were known to contemporary
Christians and perceived by them as an act aimed at destroying their faith,
which led, subsequently, to Christian blood libels against the Jews. Israel Yuval,
‘Vengeance and Damnation, Blood and Defamation: From Jewish Martyrdom
to Blood Libel Accusations’, Zion 58 (1993), 33–90 (in Hebrew). Yuval further
developed his concept of the messianic nature of Jewish martyrdom and the
centrality of the idea of vengeful redemption in Two Nations in Your Womb:
Perceptions of Jews and Christians in Late Antiquity and the Middle Ages, trans.
B. Harshav and J. Chipman (Berkeley, 2006) (first published in Hebrew in
2000). For responses to Yuval’s arguments in ‘Vengeance and Damnation’, see
Jeremy Cohen, ‘Gezerot Tatnu: Martyrdom and Martyrology in the Hebrew
Chronicles of 1096’, Zion 59 (1994), 169–208 (in Hebrew); Robert Chazan,
Medieval Stereotypes and Modern Antisemitism (Berkeley, 1997), pp. 75–76.
This debate has been summarised by Elsa Marmursztejn, ‘Reason in the History
of Persecution. Observations on the Historiography of Jewish-Christian
127
Judith Bronstein
Christian values, Jews had to be familiar with Christian culture, as Ivan

Marcus has shown. This knowledge was the result of a daily interaction
with Christians of the Jewish minority in Ashkenaz, which did not live
in social and political isolation from their Christian neighbours. Jews
resisted the pressures of the majority culture and retained a strong col-
lective identity by internalising Christian motifs, symbols and rituals
and merged them, often in polemic, parodic and neutralised ways , with
ancient Jewish cultures and traditions, a process which Marcus had named
‘inward acculturation’. The sacrificial and self-sacrificial acts of the Jews
in 1096, exemplify how they incorporated elements of Christian culture
into their own, through a denial of their Christian meaning. ‘The ritu-
alisation of the specific metaphor that a Jewish community is collectively
a sacrificed Isaac takes on special significance as an acted-out polemical
rispote to Christian claims that Jesus’s death was an atoning sacrifice.’24
As exposed by Eva Haverkamp in ‘Martyrs in rivalry: the 1096 Jewish
martyrs and the Thebean Legion’, scholars increasingly portrayed Jews
and Christians as sharing a common conceptual language of religious
images and symbols, even though they also developed quite detailed,
albeit disfigured, concepts of each other. Haverkamp demonstrates that
Shlomo bar Shimshon had a good knowledge of Christian martyrology,
and made use of it in his description of the Xanten martyrdom and in
his response to the cult of Thebean martyrs.25 In Jewish Martyrs in the
Pagan and Christian World Shmuel Shepkaru further examines Marcus’
concept of ‘Inward acculturation’. He shows that together with traditional
Jewish symbols of martyrdom, a set of central crusader motifs, such as
the concept of absolute love and devotion, the ideology of chivalry, and
the belief in a reward in heaven for the martyr have been incorporated
into the Hebrew narratives. Thus, Shepkaru identified a new Jewish
martyrology, which addressed the crusaders in their own theological
language, refuted and mocked their conducts and aims, and also spoke
to the Jewish audience. This new martyrology portrays Jewish martyrs
as ‘Jewish Crusaders’, and exhibits the superiority of their sacrifices over
the Christians’. ‘A new martyrology that, like Christian martyrology and
Relations from the Perspective of Forced Baptism’, Annales. Histoire, Sciences

sociales 6 (2012), 29–32.
24
Ivan G. Marcus, Rituals of Chilhood: Jewish Acculturation in Medieval Europe
(New Haven, 1996), p. 7; Marcus, ‘A Jewish-Christian Symbiosis: The Culture
of Early Ashenaz’, Cultures of the Jews: A New History, ed. David Biale (New
York, 2002), pp. 449–516
25
Eva Haverkamp, ‘Martyrs in Rivalry: The 1096 Jewish Martyrs and the Thebean
Legion’, Jewish History 23 (2009), 319–42.
128
hagiography, presented martyrs as voluntarily dying in public spectacles

to deliver taunting proclamations, with the new assurance that it would
earn them a place next to the Divine’.26 Anna Sapir Abulafia has shown
that although the Hebrew Chronicles were very much aware of Christian
spirituality, they described every element of the Christian faith in negative
terms and in anti-Christian invectives; this was a deliberate strategy to
inculcate the negation of Christianity into every member of the Jewish
community. But this vituperative denial of Christianity could also have
been, Sapir Abulafia suggests, the result of concern over attraction to
certain concepts of Christian theology: ‘Just as Christian anti-Judaism
was in part fuelled by Christian ambivalence towards the tradition from
which it had developed, Jewish anti-Christianity was in part fuelled by
the continuing need for Jews to keep themselves from being absorbed
into an increasing Christianising environment.’27
Indeed, cross-cultural influences seem to stand nowadays at the centre
of historical research with regard to 1096, as further exemplified by two
different works of recent years. Robert Chazan, one of the most prolific
writer on the 1096 massacres, examines the objectives of the Jewish chroni-
cles. He identifies the ‘time-bound’ message, namely the description of
the historical events, and the ‘timeless’ message, which sets out to guide
Jewish behaviour, memorialises Jewish martyrs, and places them within the
context of Jewish history. The combination of the two objectives consti-
tutes a new style of Jewish historical writing and reflects new views on the
relationship between God and humanity, hence a new approach to history.
The focal point of these views is human capacity to understand God’s will
and to carry out God’s dictates: ‘The Jews who died in the assaults of 1096
and those who survived were part of a vibrant and creative young society
that was forging a new understanding of the divine, the world, humanity
and history. The First Crusade was grounded in this new view of human
beings, their capacities, and their potential […] The Jewish narrators of the
events of 1096 shared with contemporary Christian writers assumptions
about the limited immediate role played by God during the First Crusade
and the correspondingly grand role of human heroes and villains.’28 By
comparing Hebrew and Christian contemporary sources, Chazan wishes
26
Shmuel Shepkaru, Jewish Martyrs in the Pagan and Christian World
(Cambridge, 2006), p. 210.
27
Anna Sapir Abulafia, Christian-Jewish Relations, 1000–1300: Jews in the
Service of Medieval Christendom (Harlow, 2011), p. 145; Anna Sapir Abulafia,
‘Invectives against Christianity in the Hebrew Chronicles of the First Crusade’,
in Crusade and Settlement, ed. Peter W. Edbury (Cardiff, 1985), pp. 66–72.
28
Chazan, God, Humanity and History, pp. 211–12.
129
Judith Bronstein
to demonstrate the historical accuracy of the Hebrew narrative and, no

less crucially for his argument, the Jews’ absorption of Christians values
and attitudes: ‘Jewish behaviours in 1096 show that the radical spiritual-
ity manifested in the crusade had considerable influence on the Jews.’29
Jeremy Cohen deals with the ‘timeless’ objective, to borrow Chazan’s
terminology: ‘How did Jews in twelfth-century Germany remember and
memorialise those who preferred death as a Jew to life as a Christian?
What historical circumstances gave rise to these memories of martyrdom,
as opposed to the events that they narrate?’30 Though Jewish crusader
chronicles do not necessarily convey true historical facts, they reflect the
inner world, experiences, and wishes of those who wrote them. Cohen
introduces a new approach to the study of the 1096 narrative, which tells
us mainly about what he calls the martyrologists, those survivors who tell
the martyrs’ stories. These survivors, many of whom survived because they
chose baptism over death, convey in their writing guilt or doubts about
their own or the martyrs’ behaviour, and are propelled by their impulse
to resolve this ambivalence. Cohen evaluates their function ‘as therapy for
traumatised communities plagued by guilt over the very fact and the means
of their survival’.31 Like Chazan, and other historians before him, Cohen
believes that Ashkenazi Jews were deeply involved and influenced by the
Christian world that surrounded them; he therefore examines concepts
of Christian martyrdom and crusading to provide an insight into Jewish
martyrdom and the chronicles that memorialised it.32
Chazan’s and Cohen’s works exemplify the most recent historiographic
debate, which discusses the factual accuracy of the Hebrew chronicles.
They reflect a growing and welcome tendency in modern historiography
29
Chazan, God, Humanity and History, p. 192.
30
Cohen, Sanctifying the Name of God, p. viii. One of the main promoters of
this debate is Ivan Marcus, who argues against the factual accuracy of the
Jewish chronicles in a number of works; see, for example, ‘From Politics to
Martyrdom: Shifting Paradigms in the Hebrew Narratives of the 1096 Crusade
Riots’, Prooftext 2 (1982), 40–52.
31
Cohen, Sanctifying the Name of God, pp. 68–69. In this context see also
Einbinder, Beautiful Death, which examines the cross-cultural facets of Jewish
martyrological poetry, and claims that the role of this poetry was not only to
commemorate the victims but also to shape the Jewish identity of the living,
who faced persecution and conversionary pressures.
32
Cohen, Sanctifying the Name of God, passim. See also Jeremy Cohen, ‘The
Hebrew Crusade Chronicles in their Christian Cultural Context’, in Juden und
Christen zur Zeit der Kreuzzüge, ed. Alfred Haverkamp (Sigmaringen, 1999),
pp. 17–34.
130
which, unconstrained by artificial divisions, studies these narratives as a

product not of an isolated Jewish society but of a society influenced by
Christian culture and values. Such an approach may foster new attitudes
not only to the Jews, as a persecuted minority in Christian society, but
to the different components of Christendom, whether distinguished by
gender, socio-economic status, or ideological differences. Moreover, the
Jewish narratives of 1096, as research on the factual nature of the Jewish
chronicles shows, should be analysed not as texts by which necessarily true
historical facts were conveyed, but as channels through which collective
memory and religious or/and national consciousness were shaped.
131
Moving to the Goal, June 1098 – July 1099
John France
Swansea University
The great army of Christians only just defended itself from the countless
nations and weapons of the gentiles when it was still both whole and
undivided at Antioch. Now, however, part is left at Antioch, part in this
siege of Jabala, part had been taken away and has moved to Arqa to con-
quer the enemies’ fortresses and cities, and thus the strength of our men
was diminished and they could not now manage to stand against so many
thousands of gentiles.1
Thus Albert of Aachen summarised the state of the army of the First
Crusade as it was in early March 1099, divided, dispersed, and in a hostile
land. Yet in June 1098, only ten months before, the army had enjoyed a
miraculous victory over the army of Karbugha to secure possession of
the great city of Antioch.2 So why was it still besieging Arqa in Syria in
a somewhat desultory fashion, bickering and divided? Albert of Aachen
tells us that Bohemund, Godfrey of Bouillon and Robert of Flanders had
not joined Raymond of Toulouse’s army when in January 1099 it began its
march through north Syria towards Arqa, near Tripoli. As a result, their
men became discontented and forced the leaders to promise to proceed
south on 1 March. They duly did so, but Bohemund left them to return
to Antioch while Godfrey and Robert, far from joining with Count
Raymond, besieged the coastal city of Jabala.3 While there they received
a message from Count Raymond of Toulouse saying that he feared an
enemy army was coming against him at Arqa and asking for their support.
Godfrey and Robert, according to Albert, promptly marched to Arqa,
only to discover there was no threat and that the emir of Jabala had prom-
ised Raymond great tribute to lure them away from the siege of his city:
1
V.34; ed. and trans. Susan B. Edgington (Oxford, 2007), pp. 382–83.
2
(Cambridge, 1994), pp. 269–96.
3
Albert of Aachen, Historia Ierosolimitana, V.33–34, ed. and trans. Edgington,
pp. 380–83.
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. 1–1.
John France
‘Count Raymond had falsely claimed this assembly of enemies and had
called them up now to his assistance for no other reason than to receive
the money which the inhabitants of Jabala promised for their freedom,
to ensure that the Christians would move away from the blockade of
their walls.’4 We should not assume Albert’s story is true, because Albert
is generally hostile to Count Raymond and he never went to the East,
relying on others who had participated in the journey. It is interesting
that some believed this story and repeated it after their return to the west.
Other sources say that Godfrey and Robert received the tribute of Jabala.
For we are not lacking in sources for this period. Important among them
is the Gesta Francorum, the work of an anonymous author who was clearly
closely connected to Bohemund and seems to have written his account of
the crusade very shortly it ended. Despite his admiration for Bohemund,
the Anonymous seems to have left him and become attached to the army
of the count of Toulouse by January 1099, which perhaps suggests that
he was a pious man eager to fulfil his oath to liberate Jerusalem. Perhaps
for this reason he is discreet about the divisions in the army during this
period. His work was used extensively by a series of other writers in the
early twelfth century, all of whom agree in contradicting Albert by saying
that Godfrey and Robert of Flanders were given the tribute of Jabala.5
4
Albert of Aachen, Historia Ierosolimitana, V.33–34, ed. and trans. Edgington,
pp. 380–83.
5
Anonymous, Gesta Francorum et aliorum Hierosolimitanorum, ed. Rosalind
Hill (Edinburgh, 1962), p. 84. Amongst writers who relied heavily on the work
of the Anonymous were: Baudri of Dol, ‘Historia Jerosolimitana’, in RHC Occ.
3: 3–111 (here p. 92); Robert the Monk, ‘Historia Iherosolimitana’, in RHC Occ.
3: 717–882 (here pp. 855–56); Robert the Monk’s History of the First Crusade,
trans. Carol Sweetenham (Aldershot, 2005), p. 191; Peter Tudebode, Historia
de Hierosolymitano Itinere ed. John H. and Laurita L. Hill (Paris, 1977),
p. 104; Historia de Hierosolymitano Itinere, trans. John H.and Laurita L. Hill
(Philadelphia, 1974), p. 108; Guibert de Nogent, Dei gesta per Francos et cinq
autres textes, ed. Robert B.C. Huygens (Turnhout, 1996), p. 260; The Deeds of
God through the Franks, trans. Robert Levine (Woodbridge, 1997), p. 120. For
the interrelations of all these works see 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. Alan
V. Murray (Turnhout, 1998), pp. 29–42 and Marcus Bull, ‘The Relationship
between the Gesta Francorum and Peter Tudebode’s Hierosolymitana Itinere’,
Crusades 11 (2012), 1–18. There is a substantial literature on the Gesta, most
notably Colin Morris, ‘The Gesta Francorum as Narrative History’, Reading
Medieval Studies 19 (1993), 55–71.
134
However, from time to time they reveal something of the quarrels divid-
ing the army.
Two other chroniclers went on the crusade and made limited use of the
Gesta Francorum. Raymond of Aguilers was a Provençal and chaplain to
the count of Toulouse and his account appears to date from very soon after
the capture of Jerusalem. It is markedly partisan because he was very closely
associated with the visionary Peter Bartholemew, who played such a major
role at Antioch, but perhaps because of this is sometimes implicitly critical
of the count. According to him the ruler of Jabala sent 5000 gold pieces,
together with horses and mules, to all the princes in order to buy off any
resumption of the siege of his city.6 Fulcher of Chartres was a cleric who
went on the First Crusade, but in 1097 he left the main body of the army
and proceeded to Edessa with his master Baldwin of Boulogne who later
became king of Jerusalem (1100–18). As a servant of King Baldwin he had
his own reasons for discretion which explains why his account of this period
is even more bland than that of the Gesta.7 Shortly after 1102 Albert of
Aachen wrote an immensely important account of the First Crusade which
does not rely on any other.8 He never went to the East, but evidently spoke
to many people who did, and produced a very significant source for this
period. The most important Muslim sources for the First Crusade are the
chronicles of Damascus and Aleppo. The route of Count Raymond’s army
took him through the lands of both cities. Both chronicles take consider-
able note of the siege of Antioch and the capture of Ma’arrat al-Nu‘man in
December 1098 which involved a great slaughter. But they have nothing
to say about the internal affairs of the crusade during the journey to Arqa,
6
Raymond of Aguilers, Le ‘Liber’ de Raymond d’Aguilers, ed. John H. and
Laurita L. Hill (Paris, 1969), p. 112; Raymond d’Aguilers, Historia Francorum
qui ceperunt Iherusalem, trans. John H. and Laurita L. Hill (Philadelphia, 1968)
p. 91. References to Raymond’s work will be given both to the Latin text and
(in parentheses) to the English translation.
7
Fulcher of Chartres makes no mention of tribute and simply says that
Godfrey and Robert came to the summons, but ‘the expected battle did not
materialise’: Fulcheri Carnotensis Historia Hierosolymitana (1095–1127),
ed. Heinrich Hagenmeyer (Heidelberg, 1913), pp. 269–70; A History of the
Expedition to Jerusalem 1095–1127, trans. Frances R. Ryan and Harold S. Fink
(Knoxville, 1969), p. 113.
8
Albert’s account of the First Crusade was written in 6 books, but he later added
6 more on the establishment of the Latin States in Syria: Albert of Aachen,
Historia Ierosolimitana, ed. and trans. Edgington, pp. xiv-xv.
135
John France
and while the Damascus account describes the siege and fall of Jerusalem
in July 1099, that of Aleppo barely mentions it.9
Some of these sources contribute to our knowledge of the events of this
period, but if we wish to understand the discord in the army it cannot be
too often emphasised that the crusade was not one, but a collection of
armies, each loosely made up of retinues. To refer to them as ‘The First
Crusade’ is useful shorthand, but the reality was a collection of individual
and group outlooks and attitudes which were only loosely united by
ideological imperatives. The long sojourn of the army in north Syria after
the capture of Antioch can only be understood in terms of the interplay
between the demands of the ideological objectives of the crusade and the
ambitions and needs of particular groups and individuals.
After the victory over Karbugha on 28 June 1098, the leaders decided
to delay the journey to Jerusalem until 1 November. There were a number
of substantial reasons why the army as a whole needed such a long delay:
(a) The army was tired and, as the Anonymous commented, it was sum-
mer and ‘very dry and waterless’.10 Yet this was hardly a decisive reason,
for they would ultimately besiege Jerusalem in the height of summer, and
a four-month rest seems excessive.
(b) The length of the delay was almost certainly the consequence of politi-
cal factors. In May 1098, after seven months of siege, the army of the First
Crusade was still outside Antioch. Bohemund, leader of the south Italian
Normans, had suborned Firuz, the commander of a sector of Antioch’s
wall, and without revealing this he approached the other princes with
the suggestion that whoever could find entry should receive control of
the city. This was sharply rejected by the other leaders as unfair, and con-
trary to the oath they had all sworn to Alexios I Komnenos (1081–1118)
to return to him all former imperial lands. Shortly thereafter it became
clear that Karbugha of Mosul’s enormous army intended to march to
the relief of Antioch, and in this situation Bohemund’s request had to
be reconsidered despite the fact that to concede it would risk breaking
the treaty with Byzantium. The Byzantine alliance was so important that
even in this dire situation, when they risked being trapped before the
walls of Antioch by the army of Karbugha, the princes were not willing
to concede Bohemund’s demands unconditionally. They would make
him only a conditional promise of Antioch; if Alexios came to their aid
in fulfilment of what they regarded as his obligations as an ally, he would
9
The Damascus Chronicle of the Crusades, ed. Hamilton A. R. Gibb (London,
1932); Kemal ad-Din, ‘Chronicle of Aleppo’, in RHC Or. 3: 577–665 (in French).
10
Gesta Francorum, ed. Hill, p. 72.
136
receive the city; if not it would be granted to Bohemund: ‘If Bohemond

can take this city, either by himself or by others, we will thereafter give
it to him gladly, on condition that if the emperor come to our aid and
fulfil all his obligations which he promised and vowed, we will return the
city to him as it is right to do. Otherwise Bohemond shall take it into
his power.’11 This obligation to Alexios was evidently taken very seriously
indeed. Immediately after the victory over Karbugha, they sent envoys, led
by Hugh of Vermandois and Baldwin of Hainaut, to Constantinople, and
it was probably to give Alexios time to reach Antioch with an army that
they subsequently arranged the long delay until 1 November. It should be
noted that the crusader army itself had left Nicaea in late June 1097 and
arrived at Antioch on 20 October 1097: they were giving Alexios precisely
the same period to come to Antioch.12 But this was not the only reason
for a prolonged delay.
(c) After the siege of Nicaea, an anonymous history of the First Crusade
probably produced at Monte Cassino in the early twelfth century, gen-
erally known as the Historia Belli Sacri, reports that Alexios advised the
crusaders to send a legation to the Fātimids of Egypt who were potential
allies against the Saljūq Turks.13 As a result of this, Egyptian ambassadors
appeared at Antioch in early February, at the time of the Lake Battle, and
stayed for about a month. According to Stephen of Blois, the Egyptians
at this time ‘established peace and concord with us’, while a number of
sources mentioned their comings and goings, including Albert of Aachen
who later accused the Egyptians of breaking an agreement made with the
army at Antioch.14 We do not know what passed between the crusaders
11
12
Gesta Francorum, ed. Hill, p. 72; John France, ‘La stratégie arménienne de la
première croisade’, in Les Lusignans et l’Outre Mer: Actes du Colloque Poitiers-
Lusignan, 20–24 octobre,1993, ed. Claude Mutafian (Poitiers, 1995), pp. 141–49,
France, Victory in the East, pp. 190–96, 219–20, 169–92, 297–98; France, ‘The
Crisis of the First Crusade from the Defeat of Kerbogah to the Departure from
Arqa’, Byzantion 40 (1970), 276–308.
13
New edition: Historia de via et recuperatione Antiochiae atque Ierusolymarum,
ed. Edoardo d’Angelo (Firenze, 2009), pp. 28, 47–48, 104–9. This work is
usually regarded as being based on the Gesta Francorum, but d’Angelo claims
(pp. xliv-lv) that its main source was material predating the composition of
the Gesta. References are also given to the old edition: ‘Historia Belli Sacri’, in
RHC Occ. 3: 181, 189–90, 212.
14
Gesta Francorum, ed. Hill, pp. 37, 43; Albert of Aachen, Historia Ierosolimitana,
III.59, III.62, V.46, VI.31, ed. and trans. Edgington, pp. 230–31, 236–39, 402–3,
442–45. Albert shows an impressive knowledge of the hatred between the
137
John France
and the Fātimid embassy at Antioch, but presumably mutual hostility to

the Saljūq formed the basis of an understanding, and later events suggest
that this was taken very seriously by the Egyptians who probably regarded
the westerners as agents of Alexios. The prospect of getting Jerusalem with-
out a fight was a powerful reason for delay in the summer and autumn of
1098. Once the crusaders realised that in July 1098 the Fātimids had seized
Jerusalem, it must have been apparent that the balance of advantage in
any negotiations had swung firmly in the direction of the Fātimids. Even
so, the die was not cast against an agreement with the Fātimids until May
1099 when they finally made unacceptable proposals which the crusader
leaders rejected.15 It is hardly surprising that, in the wake of the capture
of Jerusalem and the defeat of the Egyptians, crusader sources chose to
make very little of this alliance with the Fātimids.
But if these were sound reasons for delay, there were other and very
substantial factors at work. At the instant of victory over Karbugha a
sharp dispute had erupted between Bohemund and the count of Toulouse
about possession of the citadel of Antioch. Subsequently there was bad
blood between them to the extent that each occupied different parts of
the city and there was disorder.16 On 14 July 1098 Bohemund made a grant
of privileges to the Genoese, in which they promised to aid him against
any who threatened his possession of the city, with the significant excep-
tion of the count of Toulouse. This charter clearly envisaged the possibil-
ity of active hostilities between Bohemund and the count of Toulouse.
In November, when the leaders conferred about the resumption of the
march to Jerusalem, the count emerged as the champion of the rights of
the Byzantine emperor and therefore contested Bohemund’s claims to
Antioch. We can assume that this was already the issue in the summer
of 1098, especially as Bohemund’s charter made no mention of Alexios’s
Fātimids and the Turks, and although he first says that the Fātimids had
reclaimed Jerusalem from the Turks before the arrival of the crusaders, he
afterwards makes clear that they actually captured it at the end of the crusader
siege of Antioch, and his account of a surrender on terms by its ruler Soqman
is confirmed by The Damascus Chronicle, ed. Gibb, p. 45. For the whole subject
of crusader relations with the Fātimids see France, Victory in the East, pp. 211,
251–52.
15
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, p. 110 (pp. 89–90).
16
Gesta Francorum, ed. Hill, says that after the victory over Karbugha, Bohemund
took possession of the citadel of Antioch (pp. 70–71) while Albert of Aachen,
Historia Ierosolimitana, V.2, ed. and trans. Edgington, pp. 340–41, says that
Count Raymond took possession of the Bridge Gate.
138
rights.17 The sharpness of this dispute suggests enormous potential for

delay and clearly endangered the success of the whole crusade. Could they
continue without Byzantine support, which had been so valuable? If the
argument between Bohemund and Raymond was not settled, could the
crusade continue without one of their armies?
The peculiar nature of authority on the crusade made finding a solu-
tion problematic. People had joined the crusade as individuals seeking
salvation. The great leaders had direct authority only over those who
were their vassals. Others might attach themselves to one of the princes:
the men of Gascony, the Velay, the Limousin and Poitou worked with
the count of Toulouse for most of the journey, but they were free to make
other arrangements if they so wished. At the siege of Jerusalem, Gaston
of Béarn, who came from the Pyrenees, fought not with the count but
with the north French.18 As a result, the crusade was certainly run by a
committee and we tend to assume that the leading princes dominated it,
but they deliberated with the bishops and with other secular leaders who
are usually not specified, but amongst whom was sometimes Tancred,
a man of slender means. This was a rather ramshackle machinery for
decision-making, but it was inevitable in the circumstances. It was only
in the direst emergency that they were prepared to countenance one
of their number being raised over them. The appointment of Stephen
of Blois as leader in the spring of 1098 suggests that the leaders were
becoming aware of the risks of this situation. But Stephen deserted the
crusade during Karbugha’s siege of Antioch, so the experiment was not
a notable success.19
17
Gesta Francorum, ed. Hill, pp. 70–71, 83–84, 87; Raymond of Aguilers, ‘Liber’,
ed. Hill and Hill, p. 93 (pp. 74–75); Epistulae et chartae ad historiam primi
belli sacri spectantes, ed. Heinrich Hagenmeyer (Innsbruck, 1902), no. XIII,
pp. 155–56.
18
John H. and Laurita L. Hill, Raymond IV de Saint Gilles 1041 (ou 1042)-1105
(Toulouse, 1959), p. 29; Raymond of Aguilers, ‘Liber’, ed. Hill and Hill,
pp. 145–46 (pp. 123–24); Gesta Francorum, ed. Hill, pp. 91–92.
19
Bohemund was accepted as the commander at the Lake Battle in February
1098 and again against Karbugha, though with Adhemar, in June 1098: France,
Victory in the East, pp. 245–46, 279. Stephen of Blois was chosen to lead the
army as he reports his appointment in a letter to his wife: Epistulae et chartae,
no. X, pp. 149–52; this is supported by Raymond of Aguilers, ‘Liber’, ed. Hill
and Hill, p. 77 (p. 59), and Gesta Francorum, ed. Hill, p. 63. He fell ill and
when he saw Karbugha besiege Antioch fled westwards. He never seems to
have wielded real authority. For a discussion, see France, Victory in the East,
pp. 255–56.
139
John France
There was a papal legate, Adhemar bishop of Le Puy, and Urban spoke of
him as undertaking the journey in his own place. Raymond of Aguilers has
the priest Stephen say that although there was no single leader, the army
trusted greatly in him.20 Adhemar was not the only legate on the crusade,
but he enjoyed special influence because he knew the mind of Urban II
and because he led his own army, but he was not a supreme commander.
His death at Antioch on 1 August during an outbreak of plague removed
a man who would surely have been a moderating force in the disputes of
the leaders. But Raymond of Aguilers is simply highlighting his influence
when he says that the leaders went out to different places as a result of
his death, for the Anonymous says they were already dispersed, and it is
unlikely that if the leaders had been together at the moment of the legate’s
death they would have delayed informing the pope until September.21
In these circumstances the dispersal of the army in the summer of
1098 was not unnatural: there was no threat from any Muslim power and
continued concentration made supply difficult. But the dispersal took a
particular form: ‘each went off into his own territory’.22 This is not quite the
first we have heard of such territories: the count of Toulouse, at this time
ill at Antioch, seems to have controlled Rugia in the Orontes valley since
early in the siege of Antioch, and his expedition in September to Albara
represented an extension of that base. Bohemund went to Cilicia, which
Tancred was already holding for him, and Robert of Normandy probably
returned to Laodicea where he had spent much of the winter. Godfrey
of Bouillon, probably with Robert of Flanders, built up a dominion in
the Afrin valley in the approaches to his brother Baldwin’s principality
of Edessa. A minor Provençal leader, Raymond Pilet, led an expedition
which after initial success, was defeated before Ma’arrat al-Nu‘man. It
may seem surprising that Bohemund left Antioch, given the dispute with
Raymond, but presumably the decision to send Hugh of Vermandois had,
for the moment, inhibited overt action by either party, which is possibly
why Albert speaks of Bohemund acting secretly at this stage. Moreover
the outbreak of plague at Antioch was an excellent reason for leaving.23
20
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, p. 73 (p. 56).
21
Epistulae et chartae, no. II, pp. 136–37; Raymond of Aguilers, ‘Liber’, ed. Hill
and Hill, pp. 73, 84 (pp. 66, 74); Jean Richard, ‘La Papauté et la direction de la
première croisade’, Journal des Savants (1960), pp. 49–58; Gesta Francorum, ed.
Hill, pp. 72–74, and see below p. 141.
22
23
Gesta Francorum, ed. Hill, pp. 72–73, 26–27; Raymond of Aguilers, ‘Liber’, ed.
Hill and Hill, p. 84 (p. 66); Robert L. Nicholson, Tancred (Chicago, 1940),
pp. 72–73; John France, ‘The Departure of Tatikios from the Army of the
140
But this precarious stability was shattered at some time during that sum-
mer. On 11 September the crusader leaders wrote to Urban II informing
him of the death of Adhemar and asking him to come to the East to lead
them to Jerusalem. This extraordinary letter seems to have been consider-
ably influenced by Bohemund, for in it he speaks of his capture of Antioch
in the first person, while in a postscript he begs Urban to break off all
relations with Alexios Komnenos. But even if we discount this last part,
the letter is evidence of an extraordinary change in attitudes. The Greeks
are classified with Armenians, Syrians and Jacobites as heretics whom
the Holy Father is called upon to expel from Antioch. The contrast with
two earlier letters, of October 1097 and January/February 1098, in which
Adhemar and Symeon the Greek patriarch of Jerusalem jointly wrote to
the west appealing for support and thereby underscoring the unity of the
churches, is patent. In the sequence of letters sent by crusaders during
the course of their journey, this letter of the princes to the pope dated 11
September is the first to show any hostility to Byzantium. The bitterness
which it reveals is all the more stunning because of this and because of
the official nature of the document. Only one event could have brought
all the princes together and precipitated such depth of feeling: news of
the decision of the emperor, made at Philomelium in June 1098, to turn
back from an expedition to aid the crusade at the very moment of their
deepest peril, during the second siege of Antioch by Karbugha.24 Once
this was known it became very evident that Alexios would not come to
take possession of Antioch, and it could clearly be argued that he was in
breach of his obligations to the crusaders.
The request that Urban II should come to the East may have been
meant seriously, or it may have been a device to put off an imminent
quarrel over the fate of Antioch until 1 November. If the latter, it was
partially successful. Godfrey mounted an expedition to aid the emir of
Azaz against Ridwān of Aleppo, with the reluctant aid of Bohemund
First Crusade’, Bulletin of the Institute of Historical Research 44 (1971), 131–47;

Charles W. David, Robert Curthose Duke of Normandy (Cambridge, MA,
1920), p. 239; Albert of Aachen, Historia Ierosolimitana, IV.40, ed. and trans.
Edgington, pp. 310–13.
24
Epistulae et chartae, no. VI, pp. 141–42, no. IX, pp. 146–49, no. XVI, pp. 161–
65. Hagenmeyer suggested that the postscript was written years later, but the
present writer disagrees. As late as the second letter of Anselm of Ribemont,
written in early July 1098, there is no hint of anti-Byzantine feeling: Epistulae
et chartae, no. XV, pp. 157–60. For events at Philomelium, where the fleeing
Stephen of Blois met the emperor Alexios, see Gesta Francorum, ed. Hill,
pp. 63–65.
141
John France
and Raymond of Toulouse. The count of Toulouse then went on to seize

Albara on 25 September, while Bohemund returned to consolidate his
position in Cilicia. Shortly after, Godfrey was back in Edessa. The sense
here is of men looking to their own positions in north Syria, as a hedge
against the imminent storm over Antioch. The certainties of possessions
and lands here in the north, as against the deep uncertainties about the
journey to Jerusalem now evident, must have been a powerful reason for
further delay.25
When the leaders finally met on 1 November 1098 to consider resump-
tion of the march to Jerusalem it is hardly surprising that there was violent
disagreement over the question of Antioch. Bohemund claimed the city
under the terms of the conditional promise made in May 1098 and this
was contested by Raymond of Toulouse who stood for the promise made
to the emperor to return former Byzantine territories. It may seem sur-
prising, in view of the bitterness against Alexios expressed by the leaders
in their letter of 11 September, that there was not an immediate decision
in favour of Bohemund. However, when it came to it the other princes
were reluctant to act openly against the Byzantines, perhaps because their
aid had been so important that to risk its withdrawal must have seemed a
terrible threat. Moreover, Raymond of Toulouse commanded the single
largest army and it was unthinkable to take any action that might provoke
him to desert the expedition. But a new factor now entered the situation.
According to Raymond of Aguilers, grave discontent within the army
forced Bohemund and Raymond to a ‘discordant peace’ which, according
to the Anonymous was no more than the status quo: each of the two men
held his possessions in Antioch but resolved to press on to Jerusalem.26
This serious popular discontent was not new. Raymond’s support in
September 1098 for Godfrey’s effort to assist the rebellious emir of Azaz,
and his slightly later expedition against Albara were undertaken in part
to satisfy discontent within his army.27 Agitation was hardly surprising.
The crusaders had come to the east to deliver Jerusalem. To all of them,
25
‘Kemal ad-Din’, in RHC Or. 3: 586–87; Albert of Aachen, Historia
Ierosolimitana, V.13–14, V.25, ed. and trans. Edgington, pp. 354–57, 368–69;
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, p. 89 (p. 66); Gesta Francorum,
ed. Hill, pp. 74–75; Historia de via, ed. D’Angelo, pp. 92–95 (‘Historia Belli
Sacri’, in RHC Occ. 3: 207).
26
Gesta Francorum, ed. Hill, p. 76; Raymond of Aguilers, ‘Liber’, ed. Hill and
Hill, pp. 94–95 (p. 75).
27
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, pp. 88–89, 91 (pp. 73–75);
Albert of Aachen, Historia Ierosolimitana, V. 7–12, ed. and trans. Edgington,
pp. 346–55.
142
though inevitably some more than others, the wrangling over Antioch
must have seemed a distraction and, indeed, a threat to the spiritual
reward which they had been promised and for which they had suffered
so much. Moreover, long delay in friendly territory had a material effect
on the well-being not merely of the poor, but of the broad mass of people
in the army who could not hope to profit from the manoeuvrings of the
princes. The quarrels among the leaders had created a vacuum of power
into which there stepped the visionary, Peter Bartholomew, who was well
able to give expression to popular anxiety.
Peter was one of a series of visionaries who announced revelations which
inspired the army during the second siege of Antioch, when the crusaders
were trapped in the city by Karbugha’s army. He was pre-eminent because
through a series of visions he had discovered the Holy Lance, a tangible
manifestation of the divine will which was carried as a standard in the
great battle against Karbugha. Subsequently it became a token of victory
accepted by all the leaders as such.28 This gave Peter enormous prestige
and by way of reported visions he expressed the anger of the mass of the
army at the delay and its hardships, and urged the army on to Jerusalem.29
In Peter’s visions the count of Toulouse was explicitly revealed as God’s
instrument charged with the care of the Holy Lance and leadership of the
whole expedition, and it was on him that enormous popular pressure was
directed. After the ‘discordant peace’ reached at Antioch in November
1098 Raymond of Toulouse led a part of the army, not to Jerusalem,
but to Ma’arrat al-Nu‘man which was besieged from 29 November to 12
December. Bohemund accompanied him, and Robert of Flanders was
definitely present, but Godfrey seems not to have been there and the
location of Robert of Normandy is uncertain. This was no more than an
extended raid into Syria to satisfy the rank-and-file. In this it failed because
the siege dragged on in an area which had already been systematically
pillaged and the army experienced difficulties of supply. After a difficult
siege the city was stormed on 11 December. Acquisition of Ma’arrat al-
Nu‘man strengthened the Provençal hold on Rugia and the cities of the
Belus massif like Albara, and this is perhaps why, on its fall, Bohemund
tried to trade off a part of the city against the count’s holdings in Antioch.
This bargaining wrecked the ‘discordant peace’ and Bohemund returned
28
John France, ‘Two Types of Vision on the First Crusade: Stephen of Valence
and Peter Bartholomew’, Crusades 5 (2006), 1–20.
29
On the importance of Jerusalem to the crusaders see Jonathan Riley-Smith,
The First Crusade and the Idea of Crusading (London, 1986), esp. pp. 31–57;
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, pp. 84–88, 89–92, 95–97
(pp. 66–69, 70–72, 76–78).
143
John France
to Antioch. As the army delayed around Ma’arrat al-Nu‘man the crisis of

supply became ever more acute and some of the army resorted to eating
the bodies of their fallen enemies. In these circumstances discontent grew
and Peter Bartholomew’s visions urging the crusade on to Jerusalem fell
upon more and more willing ears.30
For a month the army remained dispersed and rudderless. But after vari-
ous tentatives at reconciliation, on 4 January 1099 Raymond of Toulouse
made an overt bid for the leadership of the crusade. In a council at Rugia
he offered 10,000 solidi each to Godfrey and Robert of Normandy,
6000 to Robert of Flanders and 5000 to Tancred, with smaller sums
to lesser people. Moreover, at Arqa in March he would pay substantial,
though unspecified, sums to Robert of Flanders and Godfrey of Bouillon.
Raymond of Aguilers tells us about the offers made at Rugia, and he later
reveals the terms upon which they were made. He says that Tancred, with
40 knights, entered the count’s service until Jerusalem should be captured,
though Tancred broke the agreement, alleging that Raymond was not
paying ‘in proportion to the effort and effectiveness of the soldiers who
were maintained and led by that same Tancred’.31
These sums disbursed by Raymond, well over 30,000 solidi, were enor-
mous. By way of comparison, before departure Robert of Normandy had
pawned his duchy for 10,000 marks (roughly 5000 solidi). The English
king’s income from the county farms was likewise 10,000 marks.32 Besides
Tancred, only Robert of Normandy accepted the offer. It would be inter-
esting to know the sources of Count Raymond’s wealth. Alec Mulinder,
who has written a thesis on the ‘Crusade of 1101’, thinks that Raymond
30
France, Victory in the East, pp. 311–15; see also Colin Morris, ‘Policy and Visions:
The Case of the Holy Lance Found at Antioch’, in War and Government in the
Middle Ages: Essays in Honour of J. O. Prestwich, ed. John Gillingham and James
C. Holt (Woodbridge, 1984), pp. 33–45; Lewis A. M. Sumberg, ‘The “Tafurs”
and the First Crusade’, Medieval Studies 21 (1959), 224–46; my view is that this
cannibalism was simply a result of necessity, and I am sceptical of the ideas
advanced by Michel Rouche, ‘Cannibalisme sacré chez les croisés populaires’,
in La Religion populaire: Aspects du Christianisme populaires à travers l’histoire,
ed. Yves-Marie Hilaire (Lille, 1981), pp. 29–41.
31
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, pp. 99–100, 112 (pp. 80, 92);
Albert of Aachen, Historia Ierosolimitana, V.35, ed. and trans. Edgington,
pp. 384–85.
32
Austin Lane Poole, Domesday Book to Magna Carta, 1087–1216 (Oxford, 1944),
p. 417.
144
may well have received money from the Emperor Alexios, and certainly
the scale of his wealth is suggestive.33
But the most interesting thing about this passage is the light it casts
on the size of the army, something which has of late become a conten-
tious subject. There can be little doubt, from the evidence of Raymond of
Aguilers and Albert, that Count Raymond based his offer on the size of
military followings. I would guess that the primary factor was the cavalry
force of each leader of which the number was probably easier to tally. On
the basis of Tancred’s 40 cavalry and the supposition that there were other
unnamed contingents, we can suppose that Robert of Flanders had 50,
while Godfrey and Robert of Normandy had 80–100 each. Since we know
that Count Raymond had 300, this means that the force which he gathered
for the march south from Ma’arrat al-Nu‘man that began on 13 January
1099 and on which he was joined by Robert of Normandy and Tancred,
comprised some 450–500 cavalry. Godfrey and Robert of Flanders, with
a cavalry force of some 150–200, stayed at Antioch. This roughly bears
out the figure of 1000 cavalry which Tancred said there were in the whole
crusader army in February 1099. Godfrey and Robert of Flanders were
driven south by discontent within their forces, though even so they refused
to join the southern French at Arqa and instead besieged Jabala. When all
these armies gathered at Jerusalem their cavalry, according to Raymond of
Aguilers, numbered 1200–1300, but the increase can be explained by the
prosperity of the army in this period which enabled them to buy horses.
The same estimate suggested 12,000 foot whom we can divide between the
south French and the north French as 8000 to 4000.34 I cannot pretend
that these figures are precise – we must allow for a drift southwards of
garrison forces, which is clearly attested, in the wake of the Provençals,
33
Alec Mulinder, ‘The Crusading Expeditions of 1101–2’ (unpublished Ph.D.
thesis, University of Wales, 1996), pp. 215–17.
34
The whole subject of numbers on the First Crusade is highly controversial as a
glance at some recent works shows: Jean Flori, ‘Un problème de méthodologie:
La valeur des nombres chez les chroniqueurs du Moyen Age. A propos des
effectifs de la Première Croisade’, Le Moyen Age 99 (1993), 399–422, and Flori,
Pierre l’Ermite et la première croisade (Paris, 1999), p. 453; Bernard S. Bachrach,
‘The Siege of Antioch: A Study in Military Demography’, War in History 6
(1999), 126–46; Jonathan Riley-Smith, ‘Casualties and the Number of Knights
on the First Crusade’, Crusades 1 (2002), 13–18. Especially useful sources are
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, pp. 103–4, 148 (pp. 84, 125),
and Albert of Aachen, Historia Ierosomitana, V.35, ed. and trans. Edgington,
pp. 384–85. The numbers cited here are from France, Victory in the East,
pp. 122–42.
145
John France
and the presence of non-combatants, though I suspect that by this time

they were relatively few. However, I am convinced that these figures are
of the right order. They explain much about the hesitancy of the leaders
of the army in general, and the count of Toulouse in particular, in chal-
lenging the power of the Fātimids who held Palestine south of Tripoli
including Jerusalem. This sense of uncertainty is made very clear in the
work of Raymond of Aguilers who reports that during the march south
from Ma’arrat al-Nu‘man Count Raymond seriously considered leaving
inland north Syria, crossing the Nosairi mountains, and laying siege to
Jabala.35 There were, therefore, very good reasons why Count Raymond
brought his extended raid south to a halt at Arqa.
But why was there such bitterness in the army? It was so great by March
1099 that the story that Count Raymond had been bribed by the Muslims
to lure Godfrey and Robert of Flanders to Arqa was believed. Nor was that
the end of disputes, for despite the count’s diplomacy, relations remained
strained and the whole question of besieging Arqa remained controversial,
with only the count dedicated to its capture. At Easter 1099 a Byzantine
embassy arrived complaining of Bohemund’s seizure of Antioch and
saying that Alexios would come to join the crusaders and lead them to
Jerusalem. This merely opened old wounds, for only the count favoured
this course of action; the rest feared they would be drawn into Byzantine
plots against Bohemund, and the whole business brought to a head sim-
mering discontents and anxiety to press on to Jerusalem.36 On the face of
it Count Raymond had behaved well, for of all the leaders he had, since
the fall of Antioch, been most consistently in contact with the enemy
and he had led the march south, but clearly there was enormous rancour
against him by January 1099. Albert of Aachen is generally fair to all the
princes, and yet he treats Raymond and the Provençals with real contempt.
A number of factors explain this. Whatever the doubts of the leaders
about attacking Jerusalem and challenging the Fātimid hold on Palestine
with a relatively small army, in the end it had to be done and the siege
of Arqa became, increasingly, an irrelevance. Raymond’s friendship with
Byzantium was bound to arouse bad feeling after Philomelium, the more
particularly, perhaps, because the other leaders may have felt uneasy
about their break with Byzantium, and the possible personal benefits he
received in terms of subsidy may have created envy. But probably far more
important was the question of leadership. From the very first Raymond
of Toulouse seems to have regarded himself as having a special position
35
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, pp. 103–4 (p. 84).
36
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, pp. 125–26 (p. 105).
146
amongst the leaders, perhaps because he alone had met Urban II.37 Alone
of the princes, at Constantinople in April 1097 he refused to become the
man of Alexios.38 It seems as if he was claiming a different and superior
position. Now such an eminence could only be achieved by the agree-
ment of the other leaders and Raymond never gained that. Indeed, in
military affairs he was quite overshadowed by Bohemund, and Raymond
of Aguilers says that he was ill in 1097 and there was murmuring against
him even in the ranks of his own army. In consequence he took charge
of a fortress outside the Bridge Gate of Antioch in March 1098 in order
to re-establish his reputation.39
Moreover, the situation arose whereby the visionary, Peter Bartholomew,
exerted influence by magnifying the count’s role as the chosen of the Lord
and leader of the army, and the count accepted this by using the Holy
Lance as his standard. Peter Bartholomew expressed the anxiety of the
crusaders to press on to Jerusalem, but there is also a note of Provençal
patriotism and pride in the count of Toulouse running through his visions.
The pretensions of this upstart visionary to influence political decisions
must have been bitterly resented by the other leaders, while many in the
clerical hierarchy must have shared Adhemar’s initial doubts about him.
On the night of 5 April 1099 Peter had a vision in which he spoke of
the army divided into 5 ranks. The first of these ranks was compared to
Christian martyrs and the last of them to betrayers like Judas and enemies
like Pontius Pilate. These ranks, he suggested, would be made manifest if
the count drew up the whole army for an attack on Arqa. As agitation to
press on to Jerusalem was then rising in the army, and as some of the leaders
doubted the value of attacking Arqa anyway, this was highly controver-
sial. Moreover, there were five armies present: those of the Provençals,
the Normans of Normandy, the Lotharingians, the Flemings and the
followers of Tancred. This vision, therefore, was blatant exploitation of
spiritual prestige for a clearly partisan end and provoked the doubters
into an open challenge the end result of which was the trial of the Lance
through an ordeal by fire and the death of Peter Bartholomew. The trial
did not produce a decisive outcome, and it is interesting that, according
to Raymond of Aguilers, Adhemar’s cross was then used to validate the
37
René Crozet, ‘Le voyage d’Urban II et ses arrangements avec le clergé de
France’, Revue Historique 179 (1937), 270–310.
38
Gesta Francorum, ed. Hill, p. 13; Raymond of Aguilers, ‘Liber’, ed. Hill and
Hill, pp. 41–42 (pp. 23–24).
39
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, p. 62 (pp. 44–45).
147
John France
Lance, and a new visionary, Peter Desiderius, assumed the mantle of Peter
Bartholomew and exerted some influence on events.40
The trial of the Lance certainly weakened the count’s position, and there
are other indications of his unpopularity. Tancred had received 5000 solidi
and two horses to enter the service of Count Raymond until Jerusalem
had been captured. However, shortly before the trial of the Lance he took
advantage of tension between the count and the other leaders to attach
himself to Duke Godfrey.41 Of course this may not have been Raymond’s
fault, but it is clear evidence of his isolation as a result of the futile siege
of Arqa.42 All this seems to have steeled his colleagues, especially Godfrey
of Bouillon, who emerges as the leader of the drive to Jerusalem, to grasp
the nettle and march on to Jerusalem. This was an enormous risk for a
relatively small army, and probably explains why the count exerted such
influence for so long; his case was one that any responsible leader was
bound to take into account. In the end, agitation, fanned by the rivalries of
the leaders, grew in the army for a resumption of the journey to Jerusalem.
It is notable that even then Count Raymond had doubts. At Ramla it was
almost certainly he who shrank from the risks of an assault on Jerusalem
and suggested an attack on Egypt.43
A Fātimid embassy arrived while the army was encamped at Tripoli in
early May, escorting back the ambassadors whom the crusaders had earlier
sent to Egypt. The Fātimids may well have known by then of the breach
between Alexios and the crusaders and, of course, they held Jerusalem,
and the concession they offered, free access to the Holy City for small
groups of pilgrims, were regarded with contempt. Raymond of Aguilers
thought they were also negotiating with the Saljūqs, and he reveals a good
deal of knowledge of Islamic tenets in discussing this.44 The attitude of
40
Gesta Francorum, ed. Hill, pp. 13–14; Raymond of Aguilers, ‘Liber’, ed. Hill
and Hill, pp. 112–24, 128–30, 131–34 (pp. 93–103, 110–12, 122–23).
41
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, p. 112 (p. 92).
42
Shortly after Tancred’s death in 1112 one of his close servants, the cleric Ralph
of Caen, composed an account of his life. Ralph clearly knew of the troubles
in the army at Arqa but is at pains to emphasise the religious dispute over
the Lance: ‘Gesta Tancredi’, in RHC Occ. 3: 587–716 (here p. 682); 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), pp. 126–27.
43
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, pp. 136–37 (p. 115). It is the
fact that he does not name the leader who proposed this that suggests it was
Raymond of Toulouse.
44
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, pp. 109–10 (p. 90); Albert of
Aachen also shows considerable knowledge of the politics of the Muslim world
148
the Fātimids seems to have ended all inhibitions, and on 13 May the army
marched into Egyptian territory.
The crusaders could hardly have expected that their march to Jerusalem
would be unopposed, but it was, probably because the Fātimids were confi-
dent of coming to an agreement with the leaders of the expedition and had
not reinforced their garrisons. The cities of the coast, like those of inland
Syria, were prepared to come to terms as the army sped south. Its purpose-
fulness is evident in its speed. When the army left Ma’arrat al-Nu‘man on 13
January and travelled the 160 km to Arqa in 32 days, the crude average daily
rate of march was only 5 km and there were numerous rests. In fact, when the
army was marching it made about 13 km per day which is on the slow side
of what a big western army could achieve. But it left Tripoli on 16 May and
covered 360 km to Jerusalem by 7 June, a crude average daily rate of 15 km.
Allowing for only eight days of rest, the army actually made 24 km per day,
rising, between Tripoli and Beirut, to 40 km. Moreover, while their march
to Arqa was marked by discussions about route and diversions, this was not
the case as they approached Jerusalem.45 Clearly the doubts, the hesitations
and the personal interests of the leaders had been put aside in the drive to
reach Jerusalem which had flagged since the summer of 1098. They must
have been expecting a Fātimid relieving force at any time and so were aware
that the siege of the Holy City would be a race against time.
But the army which approached Jerusalem was terribly divided. The
divisions which had emerged were too fundamental to be cured. Their
immediate root was the quarrel over the fate of Antioch, but underlying
this was a rather older tension arising from the attitudes and personality
of the count of Toulouse which sharpened the differences over policy.
The period June 1098 to July 1099 saw the breakup of the unity of the
First Crusade, which had always been precarious. The ideological unity
of the crusade was not strong enough to hold all its diverse components
together. The Saljūq powers of north Syria were too weak and divided to
profit from this while the Fātimids were unduly complacent. Even so it is
symbolic of the conflicts amongst the crusaders that the count of Toulouse
besieged the south gate (Zion Gate) of Jerusalem, while all the other lead-
ers attacked along its northern wall. And even after the capture of the city
bitterness between the leaders continued over arrangements for its rule.46
(Historia Ierosolimitana, VI.32, ed. and trans. Edgington, pp. 444–45), as does
the Historia de via, ed. D’Angelo, pp. 104–9 (‘Historia Belli Sacri’, in RHC Occ.
3: 214–15).
45
France, Victory in the East, pp. 326–27.
46
France, Victory in the East, pp. 356–66.
149
Some Considerations on the Crusaders’ Letter
to Urban II (September 1098)
University of Zaragoza
The origins of the crusading movement have been much debated, and in
particular the role of the reformist papacy in the First Crusade. In the
1930s Carl Erdmann proposed an interpretation of the leading role of
the pope which nowadays seems too unequivocal and unidirectional.1
Erdmann’s work has been submitted to severe criticism in the last decades,
mainly in the anglophone academic world.2 Disagreements on points of
detail have given way to almost wholesale rejection of the idea that this
new meritorious warfare was connected with the papacy in a straightfor-
ward way, and to an undisguised distaste for the ‘over-arching theories’
which sustained such ideas.3 These critics traced the crusade rather from
the long and well-established tradition of pilgrimage, a genesis which
had been explicitly dismissed by Erdmann.4 More recent research has
tended to view the crusade as an expedition autonomous in all respects,
which did not depart from western Europe with a fully formed concept
1
Carl Erdmann, Die Entstehung des Kreuzzugsgedankens (Stuttgart, 1935);
Erdmann, The Origin of the Idea of Crusade, trans. Marshall W. Baldwin and
Walter Goffart (Princeton, 1977).
2
For an overall criticism of Erdmann’s thesis, see Jonathan Riley-Smith,
‘Erdmann and the Historiography of the Crusades’, in La Primera Cruzada
novecientos años después: El Concilio de Clermont y los orígenes del movimiento
cruzado, ed. Luis García-Guijarro Ramos (Castellón, 1997), pp. 17–29. An
appreciative view of Erdmann’s work is given in Luis García-Guijarro, ‘Los
orígenes del movimiento cruzado. La Tesis de Erdmann y sus críticos en la
segunda mitad del siglo XX’, in As Ordens Militares e as Ordens de Cavalaria na
Construção do Mundo Ocidental. Actas do IV Encontro sobre Ordens Militares,
ed. Isabel Cristina F. Fernandes (Lisboa, 2005), pp. 87–107.
3
For a description of central European historiographical trends in the 1920s and
1930s, see Riley-Smith, ‘Erdmann and the historiography of the crusades’, p. 17.
4
Erdmann, The Origin of the Idea of Crusade, p. xxxiii.
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. 11–11.
of the innovative character of the enterprise, but developed the crusading

ethos en route, as knights, infantry and non-combatants moved through
Anatolia, Syria and along the coast down to Jerusalem. Contemporary
chroniclers offered a first depiction of this new style of pilgrimage based
on their own experiences, while a second generation of writers revised the
narrative to conform with ecclesiastical correctness.5 Crusade historio
graphy at the turn of the twenty-first century has put forward additional
perspectives. In some of them, the focus of research has moved again to
the West, not to the papacy, but to sectors of the lay nobility which are
conceived as potent elements in shaping crusade ideology. The lay nobil-
ity is thus regarded as a significant, though hitherto neglected, factor in
understanding the march to Jerusalem.6
The historiography of the First Crusade has thus moved from the
perception of a clear imprint of the Roman Church to the image of tra-
ditional practices reshaped by piecemeal ideological constructions during
the expedition and in the years after. According to this interpretation, the
new form of pilgrimage had no single or undisputed institutional origin,
and a concept of crusade was created by the participants themselves. This
presents one basic problem, however: the movement now appears to have
slowly advanced to its goal propelled by its own impulses alone, lacking
any kind of institutional grip which might provide coherence, and only
building itself as it went along. It is difficult to visualise such a potent,
diverse, and altogether innovative expedition without any sort of overall
control, even a merely theoretical one, to back and justify its great nov-
elty and provide a consistent ideology. There is documentary evidence to
show that such a lack of institutional leadership was indeed not the case.
The letter sent from the crusade leaders in Antioch to Pope Urban II (11
September 1098) is an example.
The letter is well known. Joseph Michaud and the Recueil des Historiens
des Croisades reproduced it.7 But Comte Paul Riant was the first scholar
to make specific comments on it in his ‘Inventaire’, pointing out two
formal aspects: first, the small number of contemporary copies of the
written message; second, the possibility of earlier, unknown letters
from the leaders of the expedition informing Pope Urban II of previous
5
See Jonathan Riley-Smith, The First Crusade and the Idea of Crusading
(London, 1986).
6
See, for example, Marcus Bull, Knightly Piety and the Lay Response to the First
Crusade: The Limousin and Gascony, c. 970–c.1130 (Oxford, 1993).
7
Joseph Michaud, Histoire des Croisades, 7th edn, 7 vols (Paris, 1867), 1: 508–10
(French trans. of Fulcher of Chartres’s Historia as ed. by Bongars); Fulcher of
Chartres, ‘Historia Iherosolymitana’, in RHC Occ. 3: 350–51.
152
Some Considerations on the Crusaders’ Letter to Urban II
successful campaigns, such as the conquest of Nicaea, with as much

precision as the September 1098 text did in relation to the capture of
Antioch.8 Heinrich Hagenmeyer mentioned the letter in his chronology
of the First Crusade and included it in his selection of crusaders’ corre-
spondence written 1088–1100.9 The twenty-three items he chose, with
their introduction, transcription and authoritative notes, now form the
standard version of this basic source for the study of the First Crusade.
There is a recent English translation of ten of these letters, including the
one discussed here.10 The first appeared in 1921 among the First Crusade
sources compiled by August C. Krey.11 Six years earlier, Frances Rita Ryan
had concluded an English version of Heinrich Hagenmeyer’s recently
published edition of the Historia Hierosolimitana, which contained
Fulcher of Chartres’ version of the letter (I.24). Her work remained
unpublished until 1969, and it is now the standard rendering into English
of the shorter version of the letter.12
It has reached us in a shorter and a longer version. The basic difference
is the omission of the final part of the text (sections 15 and 16) by Fulcher
of Chartres and other editions based on his Historia Hierosolimitana.13
Possible explanations of the discrepancy between the two versions are
that the author of the chronicle either cut the ending as inappropriate
or saw a manuscript which contained only the original wording of the
8
Paul Riant, ‘Inventaire critique des lettres historiques des croisades’, Archives de
l’Orient Latin 1 (Paris, 1881), 1–224 (here no. 114, pp. 181–83).
9
Heinrich Hagenmeyer, ‘Chronologie de la Première Croisade (1094–1100)’,
Revue de l’Orient Latin 7 (1899), 275–339 (here pp. 323–24); Epistulae et chartae
ad historiam primi belli sacri expectantes, ed. Heinrich Hagenmeyer (Innsbruck,
1901), no. XVI, pp. 93–100 (introduction); pp. 161–65 (transcription);
pp. 341–58 (notes).
10
Letters from the East: Crusaders, Pilgrims and Settlers in the 12th-13th Centuries,
trans. Malcolm Barber and Keith Bate (Farnham, 2010), no. 8, pp. 30–33; used
below for English translations of the letter of 11 September 1098.
11
August C. Krey, The First Crusade: The Accounts of Eye Witnesses and
Participants (Princeton, 1921), pp. 160–61, 192–93 and 194–95.
12
Fulcheri Carnotensis: Historia Hierosolimitana (1095–1127), ed. Heinrich
Hagenmeyer (Heidelberg, 1913); Fulcher of Chartres, A History of the
Expedition to Jerusalem 1095–1127, trans. Frances S. Ryan, ed. Harold S. Fink
(Knoxville, 1969).
13
For the manuscripts and printed editions which followed the two versions,
see Epistulae et chartae, pp. 97–100 and 161; for Fulcher’s version, Historia
Hierosolimitana, ed. Hagenmeyer, pp. 258–64; History of the Expedition to
Jerusalem, pp. 107–12.
153
letter (sections 1–14) as agreed by the leaders who remained in Antioch

in September 1098: Bohemund, Raymond of Saint-Gilles, Godfrey
of Bouillon, Robert of Normandy, Robert of Flanders and Eustace of
Boulogne (Stephen of Blois had fled in June and Hugh of Vermandois
had been sent to Constantinople).14 A postscript, which bluntly reflected
Bohemund’s views without the restraints imposed by common policy
with the rest of the commanders, must have been added by the Norman
prince immediately afterwards, although some authors suggest that it was
done much later, at the time of Bohemund’s visit to western Europe in
1105–6.15 Rudolf Hiestand, for example, has suggested that the original
text was altered not only by the addition of the postscript but also by
the sentence at the beginning of section 14 which declared that Eastern
Christians heretics were to be annihilated.16 According to Hiestand, this
aggressive attitude cannot be reconciled with the crusaders’ and particular
Bohemund’s favourable behaviour to non-Latin Christians between 1097
and 1100. It would fit better with the Norman prince’s anti-byzantine
policies of 1105–7. But a series of circumstances in the summer of 1098 –
the absence of and lack of communication from Alexios I Komnenos, the
papal legate’s death and internal disputes among crusaders being the most
relevant – put such a stress on the leaders, and especially on Bohemund
who wanted to be reckoned as lord of Antioch at any cost, that it prob-
ably hardened to extremes the attitude to Eastern Christians. When
Bohemund reinstated John IV Oxeites as Greek patriarch of Antioch
in June 1098 after the conquest of the city, the relationship of crusaders
with non-Latins was relaxed. Three months later there was high suspicion
towards easterners that was channelled through religious intolerance.
This situation lasted while the deep sense of uncertainty which had been
building up after the victory over Karbugha prevailed among Western
leaders. When the impasse ended, a policy of religious toleration towards
non-Latin Christians was resumed and Bohemund gave back to Greek,
Armenian and Syriac monks their monasteries in Cilicia. Nothing thus
contradicts the assumption that the main corpus of the 11 September 1098
14
Hagenmeyer suggested that Fulcher of Chartres may well have known the
postscript and decided to cut it due to its vividness: Epistulae et chartae, p. 356,
n. 83.
15
Fink in Fulcher of Chartres, History of the Expedition to Jerusalem, p. 112, n.18.
16
Rudolf Hiestand, ‘Boemondo I e la prima Crociata’, in Il Mezzogiorno
normanno-svevo e le Crociate (Atti delle quattordicesime giornate normanno-
sveve, Bari, 17–20 ottobre 2000), ed. Giosuè Musca (Bari, 2002), pp. 65–94
(here pp. 68 and 84–85).
154
letter, the reference to Eastern Christians as heretics, and the postscript

were written at roughly the same time.
The imperious content and tone of the addition at the end of the let-
ter contrast with the moderate tenor of most of the rest of it. The pope
was still recognised in the postscript as promoter of the crusade, ‘you are
the originator of this holy expedition’, but he was criticised for leniency
towards people who had taken the cross but not travelled east.17 The
attitude to the Greek emperor, who had not been mentioned previously,
was openly hostile: ‘the unjust emperor who has never fulfilled the many
promises he has made us. In fact, he has hindered and harmed us in every
way at his disposal’.18 Bohemund’s addition showed his most pressing
preoccupations at a time when he was close to procuring de facto the
rulership of Antioch because he controlled the fortress of the city.19 He
needed to secure enough warriors for Antioch’s defence once the con-
tingents headed for Jerusalem, and to check any possible intervention by
Alexios I Komnenos which would deprive him of his coveted lordship.
He presented his concerns to the pope, whom he considered at that point
the authority on everything related to the expedition. Intention, themes
and style were different from other sections of the letter. It is difficult to
imagine such sharp wording addressed to Urban II by other leaders, prob-
ably because they were less concerned by shortage of manpower at that
time. Also such hostility towards the Greek emperor would surely have
enraged Raymond of Saint-Gilles. But, in spite of divergences, there is a
common thread between the corpus of the letter and the postscript: the
acknowledgement of the pope as the supreme authority who should look
after the crusade and promote certain policies at a time when the appeasing
figure of Adhemar of Le Puy, Urban II’s legate, had died and the leaders
deeply disagreed on essential aspects of the expedition. There was no king
among the commanders. No monarch would concede such authority and
power to the pope, but the higher nobility did so as part of the spiritual
vassalage that bound them when they decided to take the cross.
The centrality of Bohemund in both parts of the letter is also evident.
The postscript was an addition to press his interests more clearly. The main
part of the letter was far more conciliatory, but it bore Bohemund’s stamp
as well. He was the first of the leaders to be mentioned at the beginning,
17
‘[…] tu inceptor sancti itineris cum sis’, Epistulae et Chartae, no.XVI, p. 165;
Letters from the East, p. 33.
18
‘[…] ab iniusto imperatore, qui multa bona promisit nobis, sed minime fecit.
Omnia enim mala et impedimenta quaecumque facere potuit, nobis fecit’:
Epistulae et Chartae, no. XVI, p. 165; Letters from the East, p. 33.
19
Rudolf Hiestand, ‘Boemondo I e la prima Crociata’, pp. 78–79.
155
although he had no official entitlement to that priority while Godfrey,

Raymond, both Roberts, and Eustace were given their titles of dukes and
counts.20 Furthermore, the narrative of the conquest of Antioch centres
on him. It is then reasonable to think, as did Hagenmeyer, that the entire
letter was written by someone in close contact with the Norman prince,
perhaps by the author of the Gesta Francorum or by a person who knew
this chronicle well.21
Most of the letter (sections 2–13) is devoted to informing the pope
of events after the battle of Dorylaeum, of which Urban II already had
knowledge. Although no records have survived, we may assume from
the words ‘as you have already heard’ that the legate Adhemar of Le
Puy sent oral or written reports to Rome, and that the papacy was well
acquainted with the course of the crusade until at least the summer
1097.22 Comte Riant and Hagenmeyer thought so.23 Both mentioned a
letter of the pope to Peter, bishop of Huesca, dated 11 May 1098, in which
Urban II rejoiced in the Christian victories over the Turks in Asia and
the Muslims in Europe, and in the restoration of Christian worship in
cities which had been famous in former times.24 The pope probably had
in mind Nicaea and Tarragona, which had been taken in the early 1090s
and was still in Christian hands when Urban II wrote to the bishop of
Huesca, though soon to be stormed by the Almoravids.25 In any case, the
20
As Hiestand points out, Bohemund’s rank was without doubt inferior to that
of any of the other great commanders of the crusade: ‘Boemondo I e la prima
Crociata’, p. 70.
21
Epistulae et Chartae, pp. 96–97; Gesta Francorum et aliorum Hierosolimi
tanorum: The Deeds of the Franks and the other Pilgrims to Jerusalem, ed. and
trans. Rosalind Hill (London, 1962).
22
‘cum igitur capta Nicaea, illam maximam multitudinem Turcorum, sicut
audistis, in Kalendas Iulii nobis obviam in valle Dorotillae devicessemus
et illum magnum Solimannum fugavissemus suisque omnibus et terra et
rebus exspoliassemus, adquisita et pacificata tota Romania, ad obsidendam
Antiochiam venimus’, Epistulae et Chartae, no. XVI, p. 161.
23
Riant, ‘Inventaire critique’, p. 183; Epistulae et Chartae, pp. 93 and 343, n. 15.
24
‘Nostris si quidem diebus Dominus in Asia Turcos, in Europa Mauros
Christianorum viribus debellavit, et urbes, quondam famosas, religionis suae
cultui gratia propensiore restituit’, Epistulae et Chartae, p. 343, n. 15; Riant,
‘Inventaire critique’, p. 183, n. 8. Riant omits the decisive sentence on the
Christian conquest of cities.
25
By mid-1092 at latest Tarragona was in Christian hands: Lawrence J. McCrank,
‘Restauración canónica e intento de reconquista de la sede tarraconense’,
Cuadernos de Historia de España 61–62 (1977), 145–245 (here p. 224).
156
reference to victories over the Turks and the conquest of cities is far from
precise. Nonetheless, it was not merely rhetoric but real knowledge of
what was happening in the East, though it is impossible to deduce what
information the pope had or how recent it was; the expression ‘in our
own days’ cannot be taken too literally. The letter seems to suggest only
that news from the engagement at Dorylaeum on 1 July 1097 had reached
Urban II and that he was aware in the spring of 1098 that Nicaea had
been conquered the previous year.
The description of the siege and Christian conquest of Antioch on 3
June 1098 (sections 3–6), Karbugha’s investment of the city during twenty-
five days (section 7), the discovery of the Holy Lance (section 8), and the
final battle that freed the Latins from Turkish pressure on 28 June 1098
is concise but detailed. Bohemund’s prime role is clearly marked at every
stage and specifically at two crucial moments: the entry into the city early
in the morning of 3 June 1098 (‘I, Bohemund, persuaded a certain Turk
to betray the city to me, and just before daybreak on the third day before
the nones of June with many Christian fighters I placed scaling ladders
against the wall, and took possession of the city that had previously resisted
Christ’) and the surrender of the citadel after the Muslim defeat of 28
June with the conversion to Christianity of the emir who held it (‘There,
the emir who was in the fortress mentioned above with a thousand men
surrendered to Bohemund and through him totally subjected himself to
the Christian faith’).26
A reference to the papal legate’s death on 1 August 1098 (section 12),
after military operations had concluded and peace was established at
Antioch, ends the description of events between July 1097 and September
1098, and connects with the second part of the letter, much shorter, but
far more relevant for understanding the character of the crusading move-
ment, especially its origins. Adhemar of Le Puy’s death was the necessary
link between the past of the crusade and what its leaders thought should
be its future at a time of internal crisis of leadership and policies, now that
the widely accepted authority and moderating influence of the bishop of
Le Puy were no longer the key to solve problems. The leaders referred to
him as the pope’s delegate (‘whom you had appointed as your vicar’).27 It is
important to estimate what were his functions in that motley contingent
of combatants. His activities did not focus on a restricted (and modern)
view of spiritual welfare for the members of the expedition. At that time,
and especially for the reformist Church, ‘spiritual’ meant global care and,
26
Letters from the East, pp. 31 and 32.
27
‘quem tuum vicarium nobis commiseras’, Epistulae et Chartae, no. XVI, p. 164;
Letters from the East, p. 32.
157
in this case, supreme authority over the various contingents and leaders.
Urban II expressed this clearly shortly after the council of Clermont in a
letter he addressed to the faithful of Flanders, December 1095: ‘And we
have constituted our most beloved son, Adhemar, bishop of Puy, leader of
this expedition and undertaking in our stead, so that those who, perchance,
may wish to undertake this journey should comply with his commands, as
if they were our own, and submit fully to his loosings or bindings, as far
as shall seem to belong to such an office.’28 The overall authority over the
contingents invested in Adhemar of Le Puy by Urban II implied a papal
will to control the crusade. That would have been utterly impractical if a
united front of warriors with clear and agreed policies had existed or if
kings had been in command as they were in later crusades. But that was
not the case before Antioch was taken, and was even less so when the city
was freed from Muslim menace and disagreements unleashed. Silent or
open disputes among the leaders reinforced the role of the legate, who
managed to preserve a sense of unity among such diversity of attitudes
and policies. When Adhemar of Le Puy died the leaders were conscious
that the unifying thread had gone, and decided to urge Urban II into
direct action to prevent collapse. No new legate was requested; only the
pope’s presence would do. This cry for help indicates that internal and
external problems had shattered the group so grievously that it no longer
relied on itself for the success and even for the survival of the project. It is
significant that there is no mention in the written message of the leaders’
previous decision to separate forces in the summer and rearrange them on
1 November, as if their confidence in their own capacity for conducting the
expedition had been steadily decreasing. A remarkable military triumph
had given way to despair. The few lines in the letter which announced
the legate’s death were thus the necessary link between the successes
of the crusade, of which the conquest of Antioch was the last proof, and
the suggestions they put forward to Urban II to ensure that the victories
would last and not vanish in disagreements and ill-feelings.
In sections 13 and 14 of the letter, the leaders begged Urban II to
travel east. They based their petition on three points: the pope, whose
idea it was to march to Jerusalem and who had pressingly preached the
journey, should take the lead in the expedition ‘and personally complete
28
‘et carissimum filium Ademarum, Podiensem episcopum, huius itineris ac
laboris ducem, vice nostra constituimus, ut quibus hanc viam forte suscipere
placuerit, eius iussionibus tamquam nostris pareant atque eius solutionibus seu
ligationibus, quantum ad hoc negotio pertinere videbitur, omnino subiaceant’,
Epistulae et Chartae, no. II, pp. 136–37; Krey, The First Crusade, p. 43.
158
a war which is your own’.29 This is perhaps the most revealing sentence
in the whole letter. The leaders, paradoxically hard-pressed by a deep
sense of lack of leadership, took up the arguments which were implicit in
Urban II’s letter of December 1095: the pope was the spiritual father of all
crusaders, but he should also make his authority visible in situ at the time
when his representative had died. ‘If you do come to us to complete with
us the expedition you began, the whole world will obey you.’30 The Iter
Hierosolymitanum was a spiritual pursuit, yet it could only be achieved
through material means, an army. The pope should embody this dual
character by being physically present in the East and by leading the people
to both the heavenly and earthly Jerusalem (‘in this way you will complete
the expedition of Jesus Christ which we began and you preached. Thus
you will open the gates of both Jerusalems, liberate the Sepulchre of the
Lord and exalt the Christian name over every other one’).31 The goals of
the so-called Gregorian reform and the meaning of the crusade could not
have been expressed more clearly and vividly. The sharp split between the
spiritual and the secular worlds meant for reformist clerics the superiority
of the former and control of the latter. War was spiritualised and made
meritorious by warriors committed by taking a vow. Urban II saw himself
as leader of the expedition when he delegated his power to the bishop
of Le Puy in December 1095. Three years later princes, dukes and counts
required him to guide them in the East. It was a special moment which
soon vanished when leaders reaffirmed their autonomy and the papacy
betrayed the contradictions which made unfeasible the dual power it
coveted. But that brief moment showed the fabric and consistency of
the crusade, which was soon engulfed by the secular practices of power
and by the incapacity of the papacy to transform authority into tangible
potestas when dealing with the material world.
The second argument was deliberately put forward to convince the
pope to journey east by stressing the relevance of Antioch to Christian
history and papal power (section 13). It was in that city that followers of
the new faith began to be called Christians. St Peter, the papacy’s ultimate
29
‘et bellum, quod tuum proprium est, ex tua parte conficias’, Epistulae et Chartae,
no. XVI, p. 164; Letters from the East, p. 32.
30
‘si enim ad nos veneris et viam per te inceptam nobiscum perfeceris, totus
mundus tibi oboediens erit’, Epistulae et Chartae, no. XVI, p. 165; Letters from
the East, p. 33.
31
‘et sic nobiscum viam Iesu Christi a nobis inceptam et a te praedicatam perficias
et portas etiam utriusque Hierusalem nobis aperias et Sepulcrum Domini
liberum atque Christianum nomen super omne nomen exaltatum facias’,
Epistulae et Chartae, no. XVI, pp. 164–65; Letters from the East, p. 33.
159
justification for its doctrine of universal jurisdiction, was enthroned as

bishop in the very church the crusaders watched daily. By visiting Antioch,
the pope would travel back to the origins of Christianity and to the roots
of papal authority which ecclesiastical reform aimed at invigorating
through visible worldwide power. ‘So what on earth could appear more
appropriate than that you, the father and leader of the Christian religion,
should come to the first leading city of the Christian name? […] We ask
you again and again […] as father and leader to come to the place of your
fatherhood, and as vicar of St Peter to sit on his throne.’32 Bohemund’s
probable influence on the raising of Antioch, of which he aspired to be
lord, to a status higher than that of Rome may also have been intended to
impress the pope by suggesting that the idea and practice of universal papal
authority could be reinforced if he visited the city which gave Christianity
its name and had such potent early connections with St Peter. There was
also the bonus of latinising Antioch and extending the universal character
of the Roman Church to Christian lands that were schismatic since 1054,
or heretic much earlier and jealous of their autonomy long before the rift
between Constantinople and Rome. The aspirations of Bohemund and the
pope may have seemed to the Norman prince closer than they were in fact.
Urban II was firm but respectful towards Greek Christianity and wished
for deeper cooperation and not confrontation with Alexios Komnenos.
The goal of Christian unity (unitatis redintegratio) was in his mind at a
time when ideas and feelings of schism had not hardened and both sides
were still flexible enough to reach compromises.33 The Norman prince
may or may not have been aware of this difference of attitude when he
(probably) advocated the inclusion in the letter of references to Antioch
which were worded in a way that sharply contradicted papal policies on
this matter.34
32
‘quid igitur in orbe rectius esse videtur, quam ut tu, qui pater et caput
Christianae religionis exsistis, ad urbem principalem et capitalem Christiani
nominis venias? […] Mandamus igitur et remandamus tibi,[…], ut tu pater et
caput ad tuae paternitatis locum venias, et qui beati Petri es vicarius, in cathedra
eius sedeas’, Epistulae et Chartae, p. 164; Letters from the East, pp. 32–33.
33
Alfons Becker, ‘Urbain II et l’Orient’, in Il Concilio di Bari del 1098 (Atti del
Convegno Storico Internazionale e celebrazioni del IX Centenario del Concilio),
ed. Salvatore Palese and Giancarlo Locatelli (Bari, 1999), pp. 123–44 (here
pp. 123–34).
34
Hagenmeyer devoted a brief note to this question, but he did not press the
matter further than positivist conclusions might allow: Epistulae et Chartae,
p. 354, n. 75.
160
There was yet a third reason in favour of the pope travelling east. The
Latin army had defeated the pagans, that is to say the Muslims, but it could
not do likewise with heretics, specifically Greeks, Armenians, Syrians
and Jacobites (section 14). Official Church policy had been to keep on
good terms with them during the march, a line of action that was strictly
followed by the legate Adhemar of Le Puy. In accordance with growing
anti-Greek feeling among combatants, crusade leaders imagined that papal
authority exercised on the spot could and should uproot these deviations.
The words ‘eradicating and destroying all types of heresy’ show intense
emotion derived not so much from uncompromising orthodoxy, but
from the animosity towards Greeks and other local populations which
slowly accumulated among Latins, as argued earlier.35 The Byzantines
are named first. This primacy shows Bohemund’s hostility to the eastern
empire, which he saw as an obstacle to his own ambitions. Without the
restraining force of the legate and with the general conviction, fuelled by
Bohemund, that Alexios Komnenos had not fulfilled his agreements and
had withdrawn any help to the crusaders at their worst moments in the
late spring and early summer of 1098, most of the leaders were ready to
accept a subtle but determined pressure on the emperor and his subjects
at the Roman Church’s most sensitive point: heresies. The postscript,
added to the main letter shortly after it was composed, bluntly expressed
Bohemund’s feelings without the need for temperance to comply with
the other leaders’ more moderate attitudes. According to the Norman
prince, the pope ought to free the crusaders from any obligations to the
Greek emperor, to whom all the suffering and difficulties they had so
far experienced were ascribed. This aggressive attitude went far beyond
the tact and care with which the pope approached Greek matters. It was
certainly more Norman than papal, more political than religious. This is
why, according to Alfons Becker, the letter may have caused Urban II’s
disappointment with developments that undermined one of the basic
aims envisaged by the pope: political and religious understanding with
the Greek empire.36
Leaving aside the postscript, the meaning of this short second part of
the letter is extremely rich and rather complex. It is difficult to deny that
crusaders regarded the pope as the promoter of and highest authority over
the expedition. However, the control he might exert over the march once
the legate had died depended entirely on the acquiescence of a group of
35
‘et omnes haereses, cuiuscumque generis sint […] eradices et destruas’, Epistulae
et Chartae, no. XVI, p. 164; Letters from the East, p. 33. See above, pp. 154–155.
36
Alfons Becker, Papst Urban II (1088–1099), 3 vols (1964–2012), vol. 2, Der
Papst, die griechische Christenheit und der Kreuzzug (Stuttgart, 1988), p. 199.
161
leaders often at odds with each other. The date of the letter, 11 September
1098, tells us much about the feelings of the princes and makes their
submission to Rome easy to understand. Just over a month previously,
the only authority in situ that they all accepted had been taken away by
death. Deep mutual misunderstanding prevented the Byzantine emperor
from playing a significant role in the expedition. Once the enemy had
been defeated, and without any dominant force on the spot to impose its
will and check disarray, the leaders’ conflicting ambitions emerged and
clashed, compromising the future and even the present of the march to
Jerusalem. The first important conquest over which the crusaders had
a complete say, the city and territory of Antioch, was bitterly disputed
between Bohemund and Raymond of Saint-Gilles. The crusade was thus
afflicted by a paralysing crisis that could only be overcome by recourse to
the authority which had promoted the enterprise and which was recog-
nised by all. The urgent call to the pope was really an appeal to unify and
thoroughly latinise the expedition. It sprang not so much from conviction
or desire as from sheer necessity. Bohemund’s addition to the text shared
with its main corpus a vivid sense of need which was expressed rather
more forcefully than in the rest of the letter. The whole message sprang
from a temporary sense of weakness when the crusade was still seen as
a common enterprise. Confidence was soon regained, but the frailty in
the letter of 11 September 1098 exposed basic features of the expedition
which would not otherwise have been so clearly disclosed.
There is no reason to doubt that Urban II received the letter, but it
is not clear when he then did. He may already have known its content
when he presided over a council at Bari in October 1098. The precise date
of this gathering of Latin and Greek clerics has only been conjectured
and this uncertainty raises the question whether discussions at Bari were
prompted by the letter of 11 September 1098 or were independent from it.
Comte Riant and Hagenmeyer were not clear about the dates and more
recently Alfons Becker, the modern authority on Urban II, has reckoned
cautiously that the council ‘lasted apparently a week, from October 3
to 10’.37 Becker has stated, also tentatively, that the pope ‘may well have
37
Riant mentioned two different dates, 3–11 October and 21 October (‘Inventaire
critique’, no. 116, p. 185; no. 117, p. 186). The former seems to relate to the
sessions of the council, but he did not provide any reason or source for the
latter. Hagenmeyer spotted the contradiction and, although he suggested that
the date of 21 October might be a misprint, he clung to it to leave more time for
the reception of the crusaders’ letter, suggesting that five weeks allowed time
to sail from St Simeon to Bari: Epistulae et Chartae, p. 102. Both Riant and
Hagenmeyer took for granted that the letter was known at the council, having
162
received [the crusaders’ letter] at the beginning or towards the end of

his council in Bari’ or simply ‘during his stay in Bari in October 1098’.38
A major discussion at Bari turned on divisive theological points with
Greek bishops of southern Italy, specially the filioque controversy.39 But,
as with most regional councils at that time, it also dealt with the usual
reformist points, and lay investiture of ecclesiastics was condemned once
again.40 This dense and specific agenda and not the crusade alone led to a
high participation of clerics: 183 bishops.41 The location of the city where
the assembly was summoned favoured the attendance of a considerable
number of Greek Church dignitaries. Anselm of Canterbury, who was pre-
sent, probably on account of his investiture controversy with the English
king, became the main exponent of orthodoxy which was reaffirmed
in the conclusions, while Greek deviations were again condemned. But
crusade affairs, perhaps the crusaders’ proposal, must also have been
treated at the council. In October 1098 the clerics and people of Lucca
previously reached Urban II, but neither pressed his argument to allow for time
inconsistencies. In fact they glossed over the obvious difficulties. For Becker’s
dating of the council, Papst Urban II , 2: 190.
38
Becker, Papst Urban II, 2: 197; 427. Becker’s estimate of 3–4 weeks or 20–25
days for the letter to reach the pope and the council may be too optimistic
and certainly fits too tightly the dates of the council sessions if they were at
the beginning of October 1098: 2: 197, n. 378. Other historians have endorsed
Becker’s position: Francesco Panarelli, ‘Il Concilio di Bari: Boemondo e la
Prima Crociata’, in Il Concilio di Bari del 1098, pp. 145–67 (here p. 163). Despite
consensus of classic and modern scholars, early October makes reception almost
impossible and even late October would be difficult. It has been assumed that
the message was sent off on the day it was written, but this may be unwarranted.
A short delay for logistical or other reasons would affect the overall calculation.
39
Riant, ‘Inventaire critique’, no. 117, p. 186; Epistulae et Chartae, pp. 102 and
370. Hagenmeyer highlighted the crusade discussions in the council of Bari.
He assumed that they centred on whether the pope should go to Jerusalem
in person. Jean Flori has followed Hagenmeyer: Pierre l’ermite et la Première
Croisade (Paris, 1999), p. 583, n. 71.
40
For a summary of discussions at the council, whose canons have been lost, see
Becker, Papst Urban II, 2: 190–200. He argues that the theological debate was
not on Urban II’s agenda, but was fostered by Greek attendants at the council
once the sessions started, and that the influence on such discussions of the
crusaders’ letter with its anti-Greek flavour can be imagined though it is not
attested in sources (Becker, Papst Urban II, 2: 191–92 and 197).
41
Epistulae et Chartae, p. 369. Other sources mention 185 or a much lesser
number, 130: Becker, Papst Urban II, 2: 192, n. 365.
163
wrote a general letter to ecclesiastics and faithful of Latin Christendom

to expound the news of the siege and capture of Antioch and of the final
victory over Karbugha as reported by a citizen from Lucca, a certain
Bruno, who had arrived to the crusaders’ camp in early March 1098 and
had returned to his city in the summer of that year.42 At the end of this
letter there was an additional information: ‘We also make it known to you
that Lord Pope Urban holds a council at Bari, considering and arranging
(matters) with many senators of the land (who are) soon to take the road
to Jerusalem.’43 This valuable item of news does not connect smoothly
with the rest of the letter and leads the reader to wonder what its purpose
was and precisely what information was being delivered. The general view
among historians is that these lines show that the letter of 11 September
1098 may have been discussed at the council. Its writers were naturally
more concerned with that part of the assembly at Bari relating to the
crusade than with theological disputations with the Greeks, although it
is a mystery how they knew that the points raised by the crusade leaders
would be dealt with at Bari when these had not been part of the agenda
known beforehand by attendants to the council.44 Krey’s translation of
the last sentence of the letter suggests that those to journey east were ‘the
senators of the land’. But there is an alternative reading of the Latin text
which centres on the pope’s plan to travel to Syria himself: ‘[…] consider-
ing and arranging with many senators of the land (how) to take soon the
road to Jerusalem’.45 This reading supports the conjecture that talks with
notables were about the organisation of Urban II’s travel to Antioch, as
42
Epistulae et Chartae, p. 100. His suggestion that Bruno may have had some
knowledge of the crusaders’ letter and made comments about it which were
misinterpreted locally in Lucca to suppose that the pope would surely travel east
does not take into account chronology. Bruno reported in the letter of October
1098 that he returned home three weeks after the victory over Karbugha, that
is at the end of July 1098, one and a half months before the crusaders wrote the
letter in Antioch: Epistulae et Chartae, p. 370.
43
‘Notum quoque vobis facimus, quod Dominus papa Urbanus apud Barum
tenet concilium, tractans et disponens cum multis terrae senatoribus ad
Ierusalem profecto tendere,’ Epistulae et Chartae, no. XVII, p. 167; Krey, The
First Crusade, p. 192.
44
Hagenmeyer denied that this question could have been included in the letter
of invitation to the council, a possibility that Riant had kept open: Epistulae et
Chartae, pp. 370–71; Riant, ‘Inventaire critique’, no. 117, p. 187.
45
The translation keeps as many of Krey’s words as possible.
164
Hagenmeyer thought and most historians have done since, and not about
their own participation in the expedition as crusaders.46
An additional source supports the view that Urban II was seriously
considering travelling east at the council of Bari. Bohemund urged Pope
Paschal II in 1106 or 1108 to carry out the journey that his predecessor
had promised to undertake at the ecclesiastical assembly and that death
had prevented him from accomplishing. Bohemund’s desperate search
at this time for support in his struggle with Alexios Komnenos may cast
doubts on this assertion, but the fact that he argued that Paschal II knew
Urban II’s intentions well because he had taken part in the council seems
to disperse suspicion.47 However, not all the information provided by this
letter of Bohemund is reliable. Urban II died on 29 July 1099, and long
before that date, even as early as the last months of 1098, he seems to have
dropped any idea of travelling east.
Whatever was discussed at Bari in relation to the crusade, connected
or not with the 11 September 1098 letter, and whatever Urban II’s initial
reaction may have been to Bohemund’s and the other leaders’ proposal,
if it arrived before the council, it is certain that the pope did not cross
the sea and so he did not continue the legate’s work of strengthening the
46
Fink, editor of Ryan’s translation of the Historia Hierosolimitana, added a
note: ‘What Urban thought of this strange letter is unknown; he died before
he could take any action’: History of the Expedition to Jerusalem, p. 107, n. 1.
If my reading of the last sentence of the Lucca letter is correct, according to
that text Urban II took an intense interest in the crusaders’ proposals, whatever
happened at or after the council of Bari. In any case, discussion there of matters
related to the crusade may have been stirred by the September letter. It is true,
as I have already stressed, that we do not know for certain the precise date of
the council and so the question whether there was enough time for the pope to
receive the letter and react to it before the council began remains open. Urban
II seems to have reached Bari on 2 or 3 October (Becker, ‘Urbain II et l’Orient’,
p. 141, n. 47; Epistulae et Chartae, p. 369), but this information is of limited use,
because he could have learned about the crusaders’ proposals while in the city.
My whole argument collapses if Urban II was unaware of the letter when the
council met, but nearly all historians think he was aware of it. Chronological
problems remain in spite of this almost unanimous accord.
47
‘quod papa Urbanus beate memorie, sicut promiserat in concilio apud beatum
(Nicolaum) celebrato vobis presentibus, morte superveniente minime complere
potuit, vos […] perficere studeretis […]’, quoted in Becker, Papst Urban II, 2:
198, n. 379.
165
papal grip over the crusade.48 He appointed instead a new legate, Daibert,
archbishop of Pisa, who reached Outremer in the late summer of 1099 after
Jerusalem had been conquered and Urban II had died, but who had been
designated much earlier.49 Consequently, the control the pope exercised
over the whole enterprise once Adhemar of Le Puy was no longer there
remained that of distant authority, but it did exist at that level and it was
certainly the only basis for crusaders’ claims in Outremer until polities
emerged and consolidated around their rulers’ own legitimacy not the
pope’s. They were some of the dukes, counts and princes who had begged
Urban II to assist them.
The crusaders’ letter of 11 September 1098 raises two linked questions.
On the one hand, the fact that the papacy did not exercise real control over
the expedition (the legate’s great influence should be neither under- nor
overestimated) does not mean that it had no power whatsoever over it,
remote though it might be. It is inconceivable that such a huge movement
of Latins to the East lacked a unifying force behind it. The crusade was
not just a massive pilgrimage and even less was it an armed pilgrimage,
which is a contradiction in terms and a periphrasis much used by historians
to reconcile novelty with tradition so as to stress continuity. According
to Guibert of Nogent, at Christmas 1096 the inhabitants of Kastoria in
Macedonia refused to supply Bohemund’s contingent with food on its way
to Constantinople because they looked on them as warriors not pilgrims.50
The crusade was a radically new movement that needed a coherence and
48
Becker suggests that Urban II’s first spontaneous reaction was to plan a journey
east, the destination of which, however, is not clear: Papst Urban II, 2: 427–28;
Becker, ‘Urbain II et l’Orient’, p. 142. For an opposite view note Panarelli’s
personal scepticism that the journey had been imagined or projected by Urban
II: ‘Il Concilio di Bari’, p. 166.
49
Riant thought that the appointment had taken place between October and
December 1098 and that the archbishop of Pisa may have been at the council
of Bari: ‘Inventaire critique’, no. 119, p. 188. The exact date of the designation
is unknown: Klaus-Peter Kirstein, Die lateinische Patriarchen von Jerusalem:
Von der Eroberung der Heiligen Stadt durch die Kreuzfahrer 1099 bis zum Ende
der Kreuzfahrerstaaten 1291 (Berlin, 2002), pp. 147–48. What is certain is that
if Urban II ever envisaged a journey east, he must have changed his mind very
soon. Bohemund twisted information in his favour in his letter to Paschal II.
50
‘militares illos estimantes, non peregrinos’, Dei gesta per Francos et cinq autres
texts, ed. Robert B. C. Huygens (Turnhout, 1996), III.2.88–89, p. 139. The
reaction of Macedonian villagers was probably stirred by remembrances of the
Norman campaigns in the region over the years 1081–1085 in which Bohemund
had played an active part.
166
backing that only the papacy could then offer. In that sense, Erdmann’s
ideas about the central role of the papacy in the crusade’s origins, when
trimmed of possible oversimplifications, are still valid for understanding
this novel type of military campaign.
The second question derives from the aspects mentioned above. The
absence of practical control over a movement which needed basic ideo-
logical guidelines to give the journey significance, guidelines that were
not built up as the expedition went along, brings out the contradictions
embedded in papal aspirations to universal jurisdiction. The internal
conflict affecting papal policies shows the lack of means to move effec-
tively and swiftly from the consistency of solid religious principles to the
practicalities of power, and also the difficulties of transferring spiritual
into earthly material supremacy. At a moment when ecclesiastical reform
was beginning to be firmly established, more in principle than in practice,
the crusade, one of its finest products, slipped away from the hands of
the papacy as soon as the basic spiritual guidelines of meritorious warfare
had to be combined with settlement and territorial organisation, which
were the natural product of the type of warfare that had been preached
at Clermont and of its astounding success.
There are other well known and controversial texts that may help
historians to understand the contradictions between papal principles
and the practice of power during the First Crusade in the wider context
of ecclesiastical reform. This perspective may also throw additional
light on the meaning of the letter of 11 September 1098. The decree of
the council of Clermont concerning lay and ecclesiastical organisation
of territories in the East, its confirmation by Adhemar of Le Puy at a
council in Antioch, as well as the information provided by the cartulary
of the Holy Sepulchre on the drawing of a frontier line between the
principality of Antioch and the kingdom of Jerusalem have all been
sources thoroughly studied by Rudolf Hiestand.51 Fulcher of Chartres
dealt with the Clermont decree and its confirmation when touching
on ecclesiastical disputes between the principality of Antioch and the
kingdom of Jerusalem over the recently conquered Tyre: ‘For in the
Council of Auvergne, so authoritative and justly renowned, it was decreed
by unanimous consent that whatever city across the Great Sea could be
wrested from the yoke of the pagans should be held forever without
contradiction. Moreover this was reaffirmed and conceded by all at the
51
Hiestand, ‘Les canons de Clermont et d’Antioche sur l’organisation
ecclésiastique des Etats croisés: autentiques ou faux?’, in Autour de la Première
Croisade, ed. Michel Balard (Paris, 1996), pp. 29–37.
167
Council of Antioch, over which the bishop of Le Puy presided.’52 A text

without date included in the cartulary of the Holy Sepulchre reaffirmed
and added precision to Fulcher’s information: ‘It is well remembered
and known that, when venerable Pope Urban of holy memory presided
over a well attended assembly at Clermont and urged to take the way to
Jerusalem, he decreed that in the case that any prince conquered regions
or cities from the pagans, the recovered churches would belong to their
territories, once they had been cleansed of the old rites.’53
Hiestand reached the conclusion that the texts concerning Urban’s rul-
ing at Clermont, and about the fixing of boundaries between Antioch and
Jerusalem by the legate Adhemar of Le Puy were alterations of authentic
records, and in consequence false.54 The Clermont decree was composed
for the purpose of settling disputes in the Iberian kingdoms and counties,
and later used for solving ecclesiastical conflicts in the east after substituting
Outremer for the original geographical zone.55 The council of Antioch was
rather a secular assembly of the leaders and the legate to fix the southern bor-
der of what had been Byzantine territory until the Turks stormed the region
in the last decades of the eleventh century and which, according to previous
agreements between crusaders and Alexios Komnenos, belonged to the Greek
emperor after the fall of Antioch. The meeting of July 1098, of which there
were no records, was later transformed into a council which renewed Urban’s
decree and applied it to the border disputes between Antioch and Jerusalem.56
Some of the arguments advanced by Hiestand are open to debate,
especially those referring to the decree of Clermont which ratified the
52
‘nam in concilio Alvernensi tam authentico et nominatissimo constitutum
unanimi adsensu fuit, ut quaecumque civitas, mari magno transito, a
paganorum posset excuti iugo, sine contradictione perenniter obtineretur.
hoc etiam in Antiocheno concilio, episcope Podiense magistrante, replicatum
et concessum ab omnibus est’, Fulcher of Chartres, Historia Hierosolimitana,
III.34.15, ed. Hagenmeyer, pp. 740–41; History of the Expedition to Jerusalem,
trans. Ryan, p. 268.
53
‘Sancte memorie venerabilis Urbanus papa, quando concilium populosissime
congregationi(s) in Monte Claro celebravit viamque Jerosolimitanam
suscitavit, decrevisse memoratur et scitur quod quicumque principes provincias
vel civitates super gentiles conquirerunt, eliminatis gentium ritibus, eorum
principatibus ecclesie restitute pertinerent’, Le Cartulaire du Chapitre du
Saint-Sépulcre de Jérusalem, ed. Geneviéve Bresc-Bautier (Paris, 1984), no. 89,
pp. 203–4.
54
Hiestand, ‘Les canons de Clermont et d’Antioche’, p. 37.
55
Hiestand, ‘Les canons de Clermont et d’Antioche’, pp. 35–36.
56
168
possession by conquerors of lands taken from the infidels and rearranged

ecclesiastical provinces to fit into Christian polities. Hiestand thinks that
Pope Paschal II, who received a request at the council of Benevento in
February/March 1113 to solve disagreements about ecclesiastical bounda-
ries between the principality of Antioch and the kingdom of Jerusalem,
used an old text from Clermont that was written to clarify the muddled
diocesan geography in Iberia.57 By dropping precise references to time
and place, the pope turned a specific ruling into a general principle valid
for Outremer, thus giving at that assembly an ad hoc answer to problems
which faced the newly settled Latin Christians in the east.58 This type of
forgery had been stimulated earlier in Outremer, Hiestand claimed, in
1110 at the latest, as soon as the conquest of more northerly cities by the
king of Jerusalem raised the question of the limits of the old ecclesiasti-
cal provinces and of the need to adapt their borders to those of the new
kingdom. It is significant that Fulcher of Chartres brought this question
into his chronicle after mentioning the conquest of the last of the coastal
cities, Tyre, on 7 July 1124.
The main problem, however, does not arise from the multiplicity of
forgeries and alterations at both ends of the Mediterranean, but from the
assumption that the Clermont decree could not have been written with
the East in mind. Hiestand questioned at least three basic ideas: first,
that Jerusalem was a political and military objective at Clermont; second,
that the penitential tone of the crusade indulgence was compatible with
such down-to-earth issues as territorial allotment or ecclesiastical limits;
third, that there could be such a precise forecast of future events and
disputes.59 It is difficult to ignore Jerusalem as the goal of the expedition.
57
‘Sancte memorie venerabilis Urbanus papa, quando concilium populosissime
congregationis in Monte Claro celebravit viamque Ierosolimitanam suscitavit,
decrevisse memoratur et scitur quod quicumque principes provincias vel
civitates supra gentiles conquirerent, eorum principatibus eliminatis gentium
ritibus ecclesie restitute pertinerent. Quod discretissimi patris decretum
rescindere non audemus’, Rudolf Hiestand, Papsturkunden für Kirchen im
Heiligen Land (Göttingen, 1985), no. 15, p. 121.
58
Hiestand, ‘Les canons de Clermont et d’Antioche’, pp. 33–34 and 35–36. It is
difficult to assume that Pope Paschal II used a decree destined for Iberia to
solve problems in Outremer without mentioning its original objective. Had
he done so, he would have changed the geographical scope of the decree and
altered it accordingly. This does not fit with his will to abide by Urban II’s
Clermont ruling. Besides, a reference to the decree’s previous destination had
been necessary when dealing with affairs of Outremer at Benevento.
59
169
The Clermont indulgence was offered to those who went to that city to
free the Church, so it was assumed that they would reach the goal.60 If
the liberation of the Church by conquering Jerusalem and other urban
centres was meant to be lasting, Latin Christians would have to settle
in Palestine and Syria to defend what had cost so much to achieve; this
required some guidelines touching secular and ecclesiastical affairs.61
The crusade was not a raid and some idea of how settlement in the East
should develop must have been clear from the start. If that were so and
if there were more than one leader ready to take up residence across the
sea, disputes must have been foreseeable.
The only alternative which could have made such vague plans superflu-
ous would be to conceive the expedition as a Latin campaign under Greek
banners. If that had been the case, there would have been no need for
thoughts about the organisation of conquered territories, because once in
Christian hands these would have been controlled by the Greek emperor
directly or through some of the crusade leaders.62 The pope and the leaders
of the crusade were surely willing at the start to respect Alexios Komnenos’
rights over former Byzantine lands, and the meeting at Antioch in July
1098 to fix the southern frontiers reinforces the view that territories as far
as the region of Tortosa and Tripoli were regarded as Byzantine, but it is
very doubtful that Urban II and the crusaders envisaged Greek control
of Jerusalem, even in the rosy days of understanding between the leaders
and the emperor, and despite the goodwill of the pope towards the latter.
Therefore it is difficult to assume that no attention was paid in Clermont
to an earthly Jerusalem. The decree under discussion was probably the
vague answer to these questions. There is no need for an Iberian setting
to explain it. But this Clermont precept in no way assured effective papal
direction of the movement. The words of the decree do not justify by
themselves an image of the First Crusade as a progressive unfolding of a
predetermined papal plan.
60
‘Quicumque pro sola devotione, non pro honoris vel pecunie adeptione, ad
liberandam ecclesiam Dei Hierusalem profectus fuerit, iter illud pro omni
penitentia ei reputetur’, Robert Somerville, The Councils of Urban II, 1, Decreta
Claramontensia (Amsterdam, 1972), canon 2, p. 74.
61
Becker thought this way in Papst Urban II (1088–1099), 2: 384. He has followed
Hiestand’s arguments in later works: ‘Urbain II et l’Orient’, pp. 138–39.
62
Becker’s suggestion is not wholly convincing: ‘On peut donc supposer que,
dans la pensée d’Urbain II, une future reorganisation en Orient était plutôt une
question de politique impériale et d’éventuels arrangements entre l’empereur
byzantine et les croisés […]’: ‘Urbain II et l’Orient’, p. 139.
170
The meaning of the Clermont decree connects with the letter dis-
patched from Antioch almost three years later. Both show the papacy at
the forefront of the movement, shaping its spiritual lines and advancing
general ideas on the future organisation of the Latin establishments over-
seas. This leadership always remained at the theoretical level of author-
ity and progressively vanished until it was reduced to just a distant and
somewhat annoying moral reference. But this switch only began to take
place clearly after the First Crusade, when Baldwin I (1100–1118), like any
other contemporary king, started to build up a kingdom around Jerusalem
which was based on a legitimacy of his own.
The first expedition to Outremer and the origins of the concept of
crusade are still the basis of most debates among historians who study
this movement. There are two linked levels in these discussions which
perhaps have not been duly considered. The first of them touches upon
the difference between the principles and the practice of papal power at
a time when Latin monarchies of all sorts began their long struggle to
consolidate their standing. The second deals with the contrast between the
vague initial sketch and the full completion of the first crusading project
in which secular powers had most of the say. At the level of principles and
initial sketch the papacy was dominant. This control vanished when ideas
mixed with and took shape from the practicalities of warfare, settlement
or government, and the final ideological shape of the movement was
greatly influenced by its practicalities. But in spite of the fact that after
the first expedition crusades became affairs of kings, popes remained the
only authority which could proclaim a crusade or convert existing warfare
into one. In this way, one of the principles of Church reform persisted: the
extension of a spirituality controlled by the Roman Church to activities
and social classes which had previously been alienated from it, namely
warfare and noble warriors.
171
Constructing the Crusader
Emotional Language in the Narratives of the First Crusade*
Stephen J. Spencer
Queen Mary University of London
In recent years the history of medieval emotions has been established as

a valid and worthwhile field of enquiry. This is owed in large part to the
efforts of scholars, most notably Barbara Rosenwein, who have begun
to overturn traditional opinions which dismissed the emotional lives of
people in the Middle Ages as ‘childlike’ and volatile in comparison to
those of the modern period.1 Unsurprisingly, this new enthusiasm for the
historical study of emotions has, at times, extended into the popular field
of crusader studies, although most considerations to date have focused on
a single sentiment. For example, Jonathan Riley-Smith has determined that
crusading was preached and interpreted as two types of love: the love of
God and love of neighbour.2 By far the most detailed survey of an emotion
in the context of the crusades is Susanna Throop’s recent monograph, in
*
I would like to thank Thomas Asbridge, Susan Edgington and Andrew Buck, who
read earlier versions of this chapter and offered suggestions for its improvement.
1
Barbara H. Rosenwein, ‘Worrying About Emotions in History’, American
Historical Review 107 (2002), 821–45; Rosenwein, Emotional Communities
in the Early Middle Ages (Ithaca, NY, 2006). For traditional attitudes, see
Johan Huizinga, The Waning of the Middle Ages, trans. Frederick Hopman
(Harmondsworth, 1965), pp. 13–14; Norbert Elias, The Civilizing Process, trans.
Edmund Jephcott (Oxford, 1994).
2
Jonathan Riley-Smith, ‘Crusading as an Act of Love’, History 65 (1980),
177–92. A broader perspective is, however, adopted by Sophia Menache in her
contribution to this volume and in a recent article: ‘Love of God or Hatred
of Your Enemy? The Emotional Voices of the Crusades’, Mirabilia 10 (2010),
1–20. See also Miikka Tamminen, ‘A Test of Friendship: Amicitia in the
Crusade Ideology of the Thirteenth Century’, in De Amicitia: Friendship and
Social Networks in Antiquity and the Middle Ages, ed. Katariina Mustakallio
and Christian Krötzl (Rome, 2009), pp. 213–29; Stephen Bennett, ‘Fear and its
Representation in the First Crusade’, Ex Historia 4 (2012), 29–54.
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. 1–1.
Stephen J. Spencer
which she persuasively argued that during the twelfth century crusading
became increasingly conceived as an act of vengeance.3 Another i mportant
contribution is her analysis of zelus, ‘a composite of passionate love, jeal-
ous protectiveness, and angry hostility’ which has no equivalent in the
English language.4 Yet, as Throop demonstrated, vindicta, ultio and zelus
rarely appeared in the earliest crusade texts.
To my knowledge, nobody has yet conducted a broader investigation
into the emotions and affective displays that contemporaries expected
crusaders to possess or, for that matter, to reject. It is this gap in the his-
toriography which the present study seeks to start to address, through
an examination of the emotional language contained within the Latin
narratives of the First Crusade.
These include three works which have traditionally been classified as
‘eyewitness’ testimonies: the anonymous Gesta Francorum et aliorum
Hierosolimitanorum, Raymond of Aguilers’ Historia Francorum qui
ceperunt Iherusalem, and the first book of the Historia Hierosolymitana
by Fulcher of Chartres.5 The First Crusade also generated a vast cor-
pus of literature by authors who did not participate in the expedition.
Chief among these are the trio of histories composed within the first
decade of the twelfth century by the Benedictine monks Guibert of
Nogent, Robert the Monk and Baldric of Bourgueil, who took the
3
Susanna A. Throop, Crusading as an Act of Vengeance 1095–1216 (Aldershot,
2011). See also Phillipe Buc, ‘La vengeance de Dieu: de l’exégèse patristique à la
réforme ecclésiastique de la première croisade’, in La Vengeance 400–1200, ed.
Dominique Barthélemy, François Bougard and Régine Le Jan (Rome, 2006),
pp. 451–86.
4
Throop, Crusading as an Act of Vengeance, p. 170.
5
(London, 1962); Raymond of Aguilers, Le ‘Liber’ de Raymond d’Aguilers, ed.
John H. Hill and Laurita L. Hill (Paris, 1969); Fulcher of Chartres, Fulcheri
Carnotensis Historia Hierosolymitana (1095–1127), ed. Heinrich Hagenmeyer
(Heidelberg, 1913). Strictly speaking, Peter Tudebode’s Historia de
Hierosolymitano Itinere, ed. John H. Hill and Laurita L. Hill (Paris, 1977), was
also a first-hand account, but his text is so similar to the Gesta Francorum that
it has been excluded from this study. See Jay Rubenstein, ‘What is the Gesta
Francorum, and Who was Peter Tudebode?’, Revue Mabillon 16 (2005), 179–
204; Marcus Bull, ‘The Relationship Between the Gesta Francorum and Peter
Tudebode’s Historia de Hierosolymitano Itinere: The Evidence of a Hitherto
Unexamined Manuscript (St Catherine’s College, Cambridge, 3)’, Crusades 11
(2012), 1–17.
174
Emotional Language in the Narratives of the First Crusade
Gesta Francorum (or versions of it) as their foundation text.6 A u seful

Lotharingian c ounter-weight is Albert of Aachen’s emotionally rich
Historia Ierosolimitana, the first six books of which may have been finished
as early as 1103.7 Alongside these works, I have also examined the Latin
verse account by Gilo of Paris; Ralph of Caen’s biography of Tancred
of Hauteville; and the Gesta Francorum expugnantium Iherusalem – a
text largely derived from Fulcher of Chartres’ narrative and commonly
attributed to Bartolf of Nangis.8
It must be made clear that I have no intention of attempting to ascertain
the crusaders’ actual emotions, which remain well and truly lost. Instead,
this study treats the emotional language of the narratives in terms of tex-
tual representation and function. There are several reasons for this. For
a start, the very notion of an ‘eyewitness’ has come under close scrutiny
in recent years, and scholars have increasingly propagated the view that
these texts were as much cultural and literary artefacts, and even propa-
gandist entities, as they were faithful reports of precisely ‘what happened’
on crusade.9 Moreover, recent research into the emotional descriptions
6
Guibert of Nogent, Dei gesta per Francos et cinq autres textes, ed. Robert B. C.
Huygens (Turnhout, 1996); Robert the Monk, The Historia Iherosolimitana of
Robert the Monk, ed. Damien Kempf and Marcus Bull (Woodbridge, 2013);
Baldric of Bourgueil, ‘The Historia Ierosolimitana of Baldric of Bourgueil – A
New Edition in Latin and an Analysis’, ed. Steven J. Biddlecombe (unpublished
Ph.D. thesis, University of Bristol, 2010). References to the currently more
accessible editions of Robert’s and Baldric’s histories are provided in
parentheses (Robert: RHC Occ. 3: 717–882; Baldric: RHC Occ. 4: 1–111).
7
Albert of Aachen, Historia Ierosolimitana, ed. and trans. Susan B. Edgington
(Oxford, 2007).
8
Gilo of Paris, The Historia Vie Hierosolimitane of Gilo of Paris and a Second,
Anonymous Author, ed. and trans. Christopher W. Grocock and J. Elizabeth
Siberry (Oxford, 1997); Ralph of Caen, Radulphi Cadomensis Tancredus, ed.
Edoardo D’Angelo (Turnhout, 2011); Bartolf of Nangis, ‘Gesta Francorum
Expugnantium Iherusalem’, in RHC Occ. 3: 489–543. Bartolf used an early
recension of Fulcher’s Historia, to which he probably had access in northern
France. On this, and the uncertainty surrounding the author’s identity, see
Susan B. Edgington, ‘The Gesta Francorum Iherusalem Expugnantium of
“Bartolf of Nangis”’, Crusades 13 (2014), 21–35.
9
Elizabeth Lapina, ‘“Nec signis nec testis creditur […]”: The Problem of
Eyewitnesses in the Chronicles of the First Crusade’, Viator 38 (2007), 117–39;
Yuval Hariri, ‘Eyewitnessing the First Crusade: The Gesta Francorum and Other
Contemporary Narratives’, Crusades 3 (2004), 77–99; Marcus Bull, ‘Views of
Muslims and of Jerusalem in Miracle Stories, c.1000–c.1200: Reflections on the
175
Stephen J. Spencer
found in twelfth-century historical narratives has elucidated that, while

they may allow us to gauge contemporary evaluations of emotions, they
do not automatically reflect the lived feelings of protagonists.10
This chapter explores the emotions contemporary writers considered
appropriate or inappropriate for crusaders to display, using the construc-
tion of the ‘idealised crusader’ as a diagnostic tool to consider, in turn,
representations of the fear of death, weeping and anger. Although the
‘idealised crusader’ is my own construct, it nevertheless has firm ground-
ing in the sources: following the expedition’s remarkable achievements,
writers set about articulating and reconfiguring the enterprise’s history
and, in so doing, seemingly started to formulate ideas about the desired
characteristics of Christian combatants. The theological embellishments
of monastic authors and the transformation of the expedition into a
distinctly ‘Frankish’ venture in certain texts are just two by-products of
that process.11
It is the primary contention of this study that emotions and emotional
expressions played an important part in the configuration of the ideal miles
Christi, above all because they functioned as effective textual indicators
of the spirituality and motives participants were thought to have pos-
sessed. Furthermore, by highlighting differences between the first-hand
testimonies and the accounts of non-participants, it will also be suggested
Study of the First Crusaders’ Motivations’, in The Experience of Crusading, 1:

Western Approaches, ed. Marcus Bull and Norman Housley (Cambridge, 2003),
pp. 13–38; Bull, ‘The Eyewitness Accounts of the First Crusade as Political
Scripts’, Reading Medieval Studies 36 (2010), 23–37; Jean Flori, Chroniqueurs et
propagandistes: introduction critique aux sources de la première croisade (Genève,
2010).
10
See, for example, Stephen D. White, ‘The Politics of Anger’, in Anger’s Past:
The Social Uses of an Emotion in the Middles Ages, ed. Barbara H. Rosenwein
(Ithaca, NY, 1998), pp. 127–52 (here p. 137).
11
Jonathan Riley-Smith, The First Crusade and the Idea of Crusading (London,
1986), pp. 135–52; Susan B. Edgington, ‘The First Crusade: Reviewing the
Evidence’, in The First Crusade: Origins and Impact, ed. Jonathan P. Phillips
(Manchester, 1997), pp. 55–77 (here pp. 59–60). On the reconfiguration
of the expedition’s history and its significance in providing ‘avatars’ for
contemporaries to emulate, see Barbara Packard, ‘Remembering the First
Crusade: Latin Narrative Histories 1099–c.1300’ (unpublished Ph.D. thesis,
Royal Holloway, University of London, 2011); Nicholas Paul, To Follow in their
Footsteps: The Crusades and Family Memory in the High Middle Ages (Ithaca,
NY, 2012), pp. 21–53.
176
that the emotional rhetoric of crusading substantially developed during

the twelfth century.
Fear of Death
As Jan Verbruggen, Jonathan Riley-Smith and, more recently, Conor
Kostick have all acknowledged, fear of death was regularly imputed to
Christian protagonists in the expedition’s narratives without eliciting a
great deal of criticism.12 Despite this, it will be argued here that the ideal-
ised crusader discernible in many of the texts confronted death with an
undaunted spirit, chiefly because he held an unbreakable trust in God.
The chroniclers repeatedly claimed that participants journeyed with
humility, rather than pride, and it was their unswerving hope in God
that rendered them intrepid in the face of death.13 For example, com-
menting on the mindset of the Christians as they left Jerusalem to engage
the army of the Egyptian vizier al-Afdal on 12 August 1099, Robert the
Monk stated that they were ‘bearing God, the victor of war, in body and
mind, and for this reason they were not frightened by any great number
of men, because they were trusting not in themselves, but in the strength
of Him’.14 Indeed, even their Muslim opponents were said to have appre-
ciated this characteristic.15 Linked to this theme was the perception that,
12
Jan F. Verbruggen, The Art of Warfare in Western Europe During the Middle
Ages: From the Eighth Century to 1340, trans. Sumner Willard and Richard
W. Southern, 2nd edn (Woodbridge, 1997), pp. 37–44; Riley-Smith, The
First Crusade and the Idea of Crusading, p. 71; Conor Kostick, ‘Courage and
Cowardice on the First Crusade, 1096–1099’, War in History 20 (2013), 32–49
(here p. 41).
13
It is worth noting that the relationship between intrepidity and trust in God
was not unique to a crusading context; indeed, it has been identified as a central
theme of battlefield orations. See John R. E. Bliese, ‘Rhetoric and Morale: A
Study of Battle Orations from the Central Middle Ages’, Journal of Medieval
History 15 (1989), 201–26 (here pp. 206–7); David S. Bachrach, Religion and the
Conduct of War, c.300–c.1215 (Woodbridge, 2003), pp. 85–90. More generally,
consult Fear and its Representations in the Middle Ages and Renaissance, ed.
Anne Scott and Cynthia Kosso (Turnhout, 2002).
14
Robert the Monk, Historia Iherosolimitana, ed. Kempf and Bull, p. 103 (873):
‘Deum, victorem belli, corpore et mente gestantes, et propterea non terrentur
ulla hominum multitudine, quia non confidunt in sua sed in ipsius virtute.’
15
See Albert of Aachen, Historia Ierosolimitana, III.65, IV.6, ed. and trans.
Edgington, pp. 242–43, 254–57; Robert the Monk, Historia Iherosolimitana,
ed. Kempf and Bull, pp. 29, 71, 106 (765, 827, 877).
177
Stephen J. Spencer
in light of their startling victories, the crusaders became convinced that

they had received celestial assistance. This found particular expression
in Raymond of Aguilers’ narrative. Upon arrival at Antioch, Count
Raymond of Toulouse supposedly reminded his comrades that there was
no need to be afraid since the Lord had helped them in their previous
ordeals. Later, assured that God would come to their aid, the crusaders
possessed ‘security’ (securitas) before the battle of Ascalon.16
A good example of the relationship between fearlessness and confidence
in divine intervention is the emotional characterisation of combatants
as they marched out to fight Karbugha of Mosul’s forces in the battle
of Antioch on 28 June 1098. Most writers attested to participants’ lack
of dread, which some linked to the discovery of the relic of the Holy
Lance on 14 June 1098 and other divine revelations confirming God’s
support. According to the Gesta Francorum, the peasant visionary Peter
Bartholomew was instructed by St Andrew to ‘tell the people of God to
have no fear, but that they should trust steadfastly with their whole heart
in the One True God’; and Fulcher of Chartres reported two additional
visions conveying the same message.17 Despite rejecting the Lance’s authen-
ticity, Ralph of Caen had Peter the Hermit announce to Karbugha that the
Christians were unafraid because the acquisition of Antioch confirmed
that they were under St Peter’s protection, whilst Bartolf of Nangis wrote
that, ‘fearing nothing’ (nihil […] formidantes), they went to fight boldly
for Christ and ‘would not have hesitated to meet and undergo death in
a cheerful spirit’.18
Bartolf ’s account also hints at another theme frequently connected to
fear, present in earlier texts but especially in works by non-participants:
the crusaders were regularly cast as intrepidly and joyfully accepting
16
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, pp. 47, 157.
17
Gesta Francorum, ed. and trans. Hill, p. 60: ‘dic populo Dei ne timeat, sed
firmiter toto corde credat in unum uerum Deum’; Fulcher of Chartres, Historia
Hierosolymitana, ed. Hagenmeyer, pp. 245–47.
18
Ralph of Caen, Tancredus, ed. D’Angelo, p. 72; Bartolf of Nangis, ‘Gesta
Francorum Expugnantium Iherusalem’, p. 503: ‘occumbere et mortem
subire hilari animo non dubitarent’. See also Fulcher of Chartres, Historia
Hierosolymitana, ed. Hagenmeyer, p. 249; Raymond of Aguilers, ‘Liber’, ed.
Hill and Hill, pp. 81–82; Guibert of Nogent, Dei gesta, ed. Huygens, p. 236;
Baldric of Bourgueil, Historia Ierosolimitana, ed. Biddlecombe, p. 260 (76);
Robert the Monk, Historia Iherosolimitana, ed. Kempf and Bull, pp. 75–76
(832); Albert of Aachen, Historia Ierosolimitana, IV.49, IV.52, ed. and trans.
Edgington, pp. 326–27, 330–31.
178
martyrdom for Christ. The same emphasis on the fearlessness and jubila-
tion of participants as they contemplated death marked Albert of Aachen’s
account of their preparations for the battle of Ascalon. They proceeded,
so Albert repeatedly claimed, ‘as happy as if they were going to a feast’
(tamquam ad conuiuium pergentes letati), and Godfrey of Bouillon
apparently explained to a perplexed Muslim ambassador that ‘we do not
fear death or the charge of the enemy, since we are certain of His eternal
reward after death in this world’.19 Similarly, just as Robert the Monk
insisted that the crusaders besieged Nicaea ‘not at all fearing death in
return for life’ (pro vita mori minime formidantes), Gilo of Paris told of
a certain Apulian who, considering it glorious to die in battle for Christ,
attacked Antioch’s Muslim defenders ‘without any fear’ (haud timide).20
There were, of course, contexts in which it was entirely suitable for
participants to experience dread, such as timor Dei; yet, when it came to
fear of death, writers appear to have conceived the idealised crusader as
being utterly fearless in combat either due to his trust in God or, as later
commentators attested, his willingness to embrace martyrdom.
Religious Weeping
On the rare occasions that historians have drawn attention to instances
of weeping in the narratives of the First Crusade, they have tended to
focus on the tears of women. Those interested in gender distinctions
have, for example, pointed to Fulcher of Chartres’ emotive account of an
unidentified wife, overcome with tears, bidding farewell to her husband as
he departed on crusade.21 The gendered implications of crying aside, it is
important to recognise that the crusaders were also frequently represented
as having wept, and not just over their dead. In fact, writers assigned a
variety of roles to their tears. They could, for instance, express their love
19
Albert of Aachen, Historia Ierosolimitana, VI.43, ed. and trans. Edgington, pp.
458–61: ‘non timemus mortem aut impetum inimicorum, cum certi simus post
temporalem mortem de eterna illius remuneratione.’
20
Robert the Monk, Historia Iherosolimitana, ed. Kempf and Bull, p. 23 (756);
Gilo of Paris, Historia Vie Hierosolimitane, ed. and trans. Grocock and Siberry,
pp. 172–75.
21
Fulcher of Chartres, Historia Hierosolymitana, ed. Hagenmeyer, p. 163; Sarah
Lambert, ‘Crusading or Spinning?’, in Gendering the Crusades, ed. Susan B.
Edgington and Sarah Lambert (Cardiff, 2001), pp. 1–15 (here pp. 7–8); Natasha
Hodgson, Women, Crusading and the Holy Land in Historical Narrative
(Woodbridge, 2007), pp. 113–14.
179
Stephen J. Spencer
for a coreligionist or serve as devices for petitioning other individuals. To

give just one example, Raymond of Aguilers recorded that members of
the crusader host at Ma’arrat al-Nu‘man in 1098 knelt before Raymond of
Toulouse, tearfully begging him to continue the journey to Jerusalem.22
Of greater importance is the fact that a number of authors seem to have
conceived participants’ tears as performing a series of religious functions.23
Weeping was often represented as a legitimate mode of imploring divine
assistance. For instance, Raymond of Aguilers explained that the ‘flowing
of tears’ (affluentia lacrimarum) in the crusader camp during the siege
of Antioch made one think God’s compassion was at stake.24 Some even
credited Christian triumphs to the weeping of participants: for Bartolf of
Nangis, their tears and prayers during the investment of Antioch were, at
least in part, responsible for the city’s eventual betrayal; and Robert the
Monk implied that the tears shed upon arrival at Jerusalem in 1099 were
better weapons than the material arms with which the city was besieged,
since the former ascended into heaven and procured Christ’s support.25
It is worth noting that, in most of the narratives, religious weeping was
not the preserve of ecclesiastics, but was also attributed to combatants
or the crusader army collectively; for example, in one text the ‘leaders
and clergy’ (principes et clerus) beseeched the Lord’s mercy ‘with many
prayers and tears’ (cum multis orationibus et lacrimis) before the battle of
Ascalon.26 The downplaying of traditional social distinctions in accounts
of the crusaders’ weeping appears to result from the clerical authorship of
most, if not all, of the texts, and the broader process whereby monastic
22
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, p. 99. The communicative
function of weeping receives attention in Gerd Althoff, ‘Empörung, Tränen,
Zerknirschung: “Emotionen” in der öffentlichen Kommunikation des
Mittelalters’, Frühmittelalterliche Studien 30 (1996), 60–79.
23
The seminal work on religious weeping remains Piroska Nagy, Le don des
larmes au Moyen Age: un instrument spirituel en quête d’institution (ve-xiiie
siècle) (Paris, 2000). See also Pierre Adnès, ‘Larmes’, Dictionnaire de Spiritualité
9 (1976), 287–303; Nagy, ‘Religious Weeping as Ritual in the Medieval West’,
Social Analysis 48 (2004), 119–37; Holy Tears: Weeping in the Religious
Imagination, ed. Kimberley C. Patton and John S. Hawley (Princeton, 2005);
Peter Dinzelbacher, Warum weint der König? Eine Kritik des mediävistischen
Panritualismus (Badenweiler, 2009), pp. 62–65; Crying in the Middle Ages:
Tears of History, ed. Elina Gertsman (London, 2011).
24
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, p. 60.
25
Bartolf of Nangis, ‘Gesta Francorum Expugnantium Iherusalem’, p. 499;
Robert the Monk, Historia Iherosolimitana, ed. Kempf and Bull, p. 96 (863).
26
Raymond of Aguilers, ‘Liber’, ed. Hill and Hill, p. 155.
180
ideals became assimilated into crusading histories.27 The crusaders, as

warrior-monk hybrids, effectively inherited the kind of tears which were
traditionally associated with professed religious.28
Tears could also symbolise participants’ dedication to Jerusalem. Three
writers – Albert of Aachen, Robert the Monk and Baldric of Bourgueil –
envisioned the Latins weeping copiously when they first sighted Jerusalem
on 7 June 1099.29 In all three accounts, the lachrymose performances
attributed to the Christians encapsulated their desire to reach the Holy
City; thus, in the versions of Baldric and Robert, weeping appeared
alongside other somatic gestures – the bending of their bodies and kissing
of the ground – which expressed humility. Uniquely, Baldric even went
so far as to suggest that they were actually imitating the tears of Christ,
remarking that ‘they wept over [ Jerusalem], over which their Christ had
also wept’.30 Furthermore, the crusaders’ tearful worshipping at the Holy
Sepulchre on 15 July was described by the majority of commentators;
indeed, their innumerable tears allegedly flooded the Sepulchre’s floor.31
Accounts of this episode highlight a further spiritual function of tears:
they could express participants’ gratitude to God. According to Albert
of Aachen, Godfrey of Bouillon offered thanks to God at Christ’s tomb
by ‘persisting in tears, prayers, and divine praises’ (in lacrimis, orationibus
et diuinis laudibus persistens).32 Other writers depicted the entire army
as tearfully thanking God for the completion of their pilgrimage, and
Guibert of Nogent claimed that the Christians again poured out ‘endless
27
See William Purkis, Crusading Spirituality in the Holy Land and Iberia
c.1095–c.1187 (Woodbridge, 2008), pp. 30–58.
28
On the weeping of professed religious, see Giles Constable, The Reformation
of the Twelfth Century (Cambridge, 1996), p. 273; William M. Aird, ‘The Tears
of Bishop Gundulf: Gender, Religion, and Emotion in the Late Eleventh
Century’, in Intersections of Gender, Religion and Ethnicity in the Middle Ages,
ed. Cordelia Beattie and Kirsten A. Fenton (Basingstoke, 2011), pp. 62–84.
29
Albert of Aachen, Historia Ierosolimitana, V.45, ed. and trans. Edgington, pp.
402–3; Robert the Monk, Historia Iherosolimitana, ed. Kempf and Bull, p. 96
(863); Baldric of Bourgueil, Historia Ierosolimitana, ed. Biddlecombe, p. 298
(97).
30
Baldric of Bourgueil, Historia Ierosolimitana, ed. Biddlecombe, p. 298 (97):
‘Fleuerunt igitur super illam, super quam et Christus illorum fleuerat.’ See Luke
19:41.
31
Robert the Monk, Historia Iherosolimitana, ed. Kempf and Bull, p. 100
(869).
32
Albert of Aachen, Historia Ierosolimitana, VI.25, ed. and trans. Edgington,
pp. 436–37.
181
Stephen J. Spencer
tears of thanks’ (infinitis gratiarum lacrimarumque) when they returned

to Jerusalem after the battle of Ascalon.33
This all suggests that the crusaders’ tears were understood to be a visual
manifestation of their piety. At times, however, chroniclers attempted to
communicate the internal feelings behind the crusaders’ weeping. One
of the terms used in a small but significant number of cases to relate such
hidden feelings was compunctio. Inspired by God, compunctio was a sort
of piercing of the heart which symbolised penitents’ awareness of, and
resolution to rectify, their previous sins, and was usually accompanied by
the external symptom of sorrowful tears.34 This phenomenon, known as
the donum lacrimarum (the ‘gift of tears’), can be detected in Guibert of
Nogent’s narrative of the battle of Dorylaeum on 1 July 1097. During the
conflict the crusaders were weeping and grieving: ‘the piety of compunc-
tion and confession’ (piae compunctionis ac confessionis) rose from their
minds and, in this way, their sins called out to Christ.35
According to the medieval doctrine of compunction, however, tears of
joy stemmed from a longing for salvation and, therefore, surpassed those of
sorrow in their spiritual efficacy.36 There are signs that ‘second-generation’
monastic writers interpreted the weeping of participants before the Holy
Sepulchre in light of this. Baldric of Bourgueil theologically improved the
Gesta Francorum’s comment that they visited the Sepulchre ‘rejoicing and
weeping from excessive joy’ (gaudentes et prae nimio gaudio plorantes) by
adding that each vowed to spend the day ‘with the sacrifice of compunc-
tion’ (cum incenso compunctionis).37 Bartolf of Nangis also referred to the
crusaders’ ‘tears of compunction’ (compunctionis lacrymas) at this moment,
and a similar appreciation likely accounts for William of Tyre’s decision
to clarify that their tears and sighs were of the sort inspired, not by sorrow
or worry, but by ‘burning devotion and the perfect joy of the inner man’
33
Gilo of Paris, Historia Vie Hierosolimitane, ed. and trans. Grocock and Siberry,
pp. 248–49; William of Tyre, Chronique, ed. Robert B. C. Huygens (Turnhout,
1986), p. 414; Guibert of Nogent, Dei gesta, ed. Huygens, p. 300.
34
Nagy, Le don des larmes, pp. 88–94; Sandra J. McEntire, The Doctrine of
Compunction in Medieval England: Holy Tears (New York, 1990), pp. 32–80.
On the etymology of compunctio, see Nagy, Le don des larmes, pp. 92, 425–30;
Columba Stewart, Cassian the Monk (Oxford, 1998), pp. 122–25.
35
Guibert of Nogent, Dei gesta, ed. Huygens, pp. 156–57.
36
Nagy, Le don des larmes, pp. 127–29; McEntire, The Doctrine of Compunction in
Medieval England, pp. 22–23, 50–57.
37
Gesta Francorum, ed. and trans. Hill, p. 92; Baldric of Bourgueil, Historia
Ierosolimitana, ed. Biddlecombe, pp. 311–12 (103).
182
(fervens devotio et interioris hominis consummata leticia).38 William’s con-

cern for the ‘spiritual joy’ (spirituali gaudio) of the homo interior, externally
signalled by tears, appears to have been shared by Guibert of Nogent, who
portrayed the crusaders’ experience as tantamount to a spiritual rebirth.
Having endured unparalleled torments, like those of childbirth, they now
resembled ‘newborn sons’ (nati filii) and attained ‘the fresh joys of the
longed-for vision’ (nova visionis […] desideratae gaudia). There is a strong
case for arguing that Guibert was alluding to participants’ spiritual joy,
not least because, in accordance with the doctrine of compunction, the
crusaders then tearfully ‘embraced’ (complectuntur) Christ, still suspended
from the Cross.39
It is clear that chroniclers struggled to articulate the emotions of par-
ticipants before Christ’s tomb; indeed, some maintained that the blessed
nature of their tears could not be truly comprehended.40 Nevertheless,
later writers, seemingly influenced by the doctrine of compunction,
offered theological commentaries in which the crusaders acquired the
most sought-after tears – those of spiritual joy, which symbolised their
attainment of God’s realm.
Anger and Rage

At the turn of the twelfth century anger came to be perceived in at least
two ways: on the one hand, as a vice; and, on the other, as an acceptable
or even, within certain criteria, praiseworthy sentiment.41 With this in
mind, it is surprising to find that not only is anger, like vengeance, one of
38
Bartolf of Nangis, ‘Gesta Francorum Expugnantium Iherusalem’, p. 515;
William of Tyre, Chronique, ed. Huygens, pp. 413–14.
39
Guibert of Nogent, Dei gesta, ed. Huygens, p. 282.
40
Guibert of Nogent, Dei gesta, ed. Huygens, p. 282; William of Tyre, Chronique,
ed. Huygens, p. 414.
41
The literature is vast. See especially: Anger’s Past, ed. Rosenwein; Richard E.
Barton, ‘Gendering Anger: Ira, furor, and Discourses of Power and Masculinity
in the Eleventh and Twelfth Centuries’, in In the Garden of Evil: The Vices and
Culture of the Middle Ages, ed. Richard Newhauser (Toronto, 2005), pp. 371–92;
Kate McGrath, ‘The Politics of Chivalry: The Function of Anger and Shame
in Eleventh- and Twelfth-Century Anglo-Norman Historical Narratives’, in
Feud, Violence and Practice: Essays in Medieval Studies in Honor of Stephen D.
White, ed. Belle Tuten and Tracey Billado (Aldershot, 2010), pp. 55–69; David
Bates, ‘Anger, Emotion and a Biography of William the Conqueror’, in Gender
and Historiography: Studies in the History of the Earlier Middle Ages in Honour
183
Stephen J. Spencer
the more poorly represented emotions in the earliest crusade texts, but
also that the wrath of participants was rarely portrayed as possessing a
religious dimension.42 Sophia Menache has pointed to the slaughter at
Jerusalem in 1099 as an episode which was ‘depicted in the category of a
reasonable, just, useful anger that was directed at extirpating sin or punish-
ing evil doers’.43 It is important to note, however, that anger terminology
is curiously absent from the earliest accounts of the massacre, and it was
only in the later testimonies of non-participants that such terminology
was used, with a noticeable religious edge. The Norman historian Ralph
of Caen suggested the crusaders seethed with ‘equal anger’ (par ira) and
he affirmed that this was ‘holy fury’ (sancte furor).44 Albert of Aachen
referred to the Christians as ‘raving and venting their rage’ (bachantes
ac seuientes) on a third day of slaughter, but we cannot be sure that he
approved of their fury, for this appears in a passage in which the author
seemed to sympathise with the victims.45
In fact, in most cases, it was neither the religious errors nor the sup-
posed polluting presence of the Muslims which was presented as arousing
crusader indignation, but rather the physical injuries they inflicted on
the Christian army. As a straightforward example, consider the descrip-
tions of the Latins’ feelings when they learned of the Turkish slaughter
of their compatriots who had been escorting craftsmen to the siege of
Antioch in March 1098. The version in the Gesta Francorum reads: ‘Then
we, inflamed by the murder of our men […] at once came together to
fight against them.’46 Several writers adapted this account significantly.
Guibert of Nogent claimed that they were inspired by a combination
of anger (efferatis) and sorrow (dolore), thereby magnifying the wrongs
of Pauline Stafford, ed. Janet L. Nelson, Susan Reynolds and Susan M. Johns
(London, 2013), pp. 21–33.
42
There are, however, a few instances in later texts: Gilo of Paris, Historia Vie
Hierosolimitane, ed. and trans. Grocock and Siberry, pp. 6–7, 74–75; William
of Tyre, Chronique, ed. Huygens, pp. 354, 401.
43
Menache, ‘Love of God or Hatred of Your Enemy?’, pp. 10–11. See also Menache,
‘Emotions in the Service of Politics: Another Perspective on the Experience of
Crusading (1095–1187)’, in this volume pp. 235–54 (here p. 245).
44
Ralph of Caen, Tancredus, ed. D’Angelo, pp. 108–10.
45
pp. 442–43; Benjamin Z. Kedar, ‘The Jerusalem Massacre of July 1099 in
the Western Historiography of the Crusades’, Crusades 3 (2004), 15–75 (here
pp. 22–23).
46
Gesta Francorum, ed. and trans. Hill, p. 40: ‘Tunc nos accensi occisione
nostrorum […] iuncti simul peruenimus contra eos ad bellum.’
184
committed by the Turks, whereas in Baldric of Bourgueil’s Historia this

incident took the form of a more developed emotional sequence, with
the introduction of vengeance; he not only characterised the Christians
as being ‘more inflamed with wrath than they were frightened’ (magis in
iram excitati quam exterriti), but also as seeking to be avengers (ultores)
of the blood of their brethren.47
Episodes like this, few and far between in the ‘eyewitness’ accounts, are
more characteristic of later works, especially Albert of Aachen’s Historia,
in which participants were regularly described as displaying righteous
anger in reaction to injuries inflicted on the Christian host.48 Nevertheless,
susceptibility to anger was usually derided, for it was an emotion which
needed to be controlled. Thus, Baldric presented his central character,
Bohemund of Taranto, as an individual in control of his wrath. Despite
being indignant at the emperor of Constantinople, Alexios I Komnenos,
for impeding the journey of the southern Italian Norman contingent,
Bohemund ‘silently contained his feeling of anger’ (iram animi tacitus
continuit).49 In similar vein, for Ralph of Caen Tancred of Hauteville’s
judgement was never clouded by wrath: he reported how, in a quarrel
with Raymond of Toulouse, ‘Tancred was hardly able to curb his feelings
and restrain himself from assuaging his anger with the slaughter of the
Provençals; but reason came to the man, preventing him from shedding
Christian blood.’50 Instead, he took revenge (ultus) in a suitable manner
– by ejecting the count’s men from Antioch.
47
Guibert of Nogent, Dei gesta, ed. Huygens, p. 191; Baldric of Bourgueil,
Historia Ierosolimitana, ed. Biddlecombe, pp. 208, 209 (50). Baldric’s version
reflects the widely attested ‘script’ whereby anger in response to injuries led to
vengeance. See White, ‘The Politics of Anger’, pp. 142–45; Throop, Crusading
as an Act of Vengeance, pp. 21–22, 158–71.
48
Ira, the anger word most frequently used, appears twelve times in the Gesta
Francorum, seven times in Raymond of Aguilers’ Historia, and is absent from
the first book of Fulcher of Chartres’ history. In contrast, the same term was
used on twenty-two occasions in the first six books of Albert of Aachen’s text.
For examples, see Albert of Aachen, Historia Ierosolimitana, II.31, III.55, ed. and
trans. Edgington, pp. 114–15, 224–25.
49
Baldric of Bourgueil, Historia Ierosolimitana, ed. Biddlecombe, p. 156 (23–24);
Steven J. Biddlecombe, ‘Baldric of Bourgueil and the Flawed Hero’, Anglo-
Norman Studies 35 (2013), 79–93 (here pp. 87–88).
50
Ralph of Caen, Tancredus, ed. D’Angelo, p. 84: ‘uix compescit Tancredus
animos, quin Prouincialium strage iram leniat. Sed occurrit uiro ratio, quae
sanguinem uetet fundi Christianum.’
185
Stephen J. Spencer
Ralph’s comment raises another issue which chroniclers appear to have

appreciated: that, unless properly harnessed, anger was a potentially dan-
gerous emotion, one which could sow discord and breed violence between
fellow Christians. Accounts of the unruly conduct of the earliest crusaders
as they passed through Hungarian territory in 1096 are a useful case in
point. Several authors depicted these crusaders as possessing a boundless,
unmediated rage, which verged on insanity.51 Albert of Aachen recorded
an incident whereby, following ‘a most vile dispute with a certain Bulgar’,
a hundred Swabians satisfied their ‘rage’ (furoris) by destroying seven mills
and numerous houses, resulting in the slaughter of many pilgrims in retali-
ation.52 Albert’s attitude towards this incident is indicated by his decision
to distance Peter the Hermit, whom he promoted as instigator of the First
Crusade, from the Swabians’ behaviour. Entirely unaware of their actions,
Peter ‘called together the more prudent and intelligent men from the army,
and spoke to them thus, saying: “A serious and severe misfortune threatens
us, arising from the rage of the senseless Germans.”’53 Several facets of this
passage underline the illegitimacy of the Swabians’ fury, not least the use
of furor – a term with strong connotations of irrationality – to describe
their rage, but also the contrast established between Peter and ‘the more
prudent and intelligent men’ on the one hand, and the ‘senseless Germans’
on the other.54 A few lines later Albert made a similar juxtaposition,
noting that an assault on Niš by a thousand ‘frivolous youths’ (leuitatis
iuuentus) – again motivated by furore – was contrary to the instructions
of Peter and ‘all sensible men’ (omnibus sensatis).55
Albert’s presentation of the early crusaders’ rage ought to be considered
a reflection of his broader concern for the potentially destructive conse-
quences of anger between Christians, especially members of the crusader
army. This is undoubtedly indicated by his representation of wise counsel
as, in modern parlance, a kind of ‘anger management’. Throughout Albert’s
text, when an individual’s wrath was depicted as threatening to provoke
51
Guibert of Nogent, Dei gesta, ed. Huygens, pp. 121–22; Gilo of Paris, Historia
Vie Hierosolimitane, ed. and trans. Grocock and Siberry, pp. 36–37, 52–53.
52
Albert of Aachen, Historia Ierosolimitana, I.9, ed. and trans. Edgington,
pp. 20–21.
53
Albert of Aachen, Historia Ierosolimitana, I.10–11, ed. and trans. Edgington,
pp. 22–23: ‘conuocat sapientiores et magis sensatos de exercitu quibus sic
loquitur dicens: “Graue et durum nobis infortunium ex furore insipientium
Theutonicorum ortum imminet”’.
54
On furor, see Barton, ‘Gendering Anger’, pp. 383–87.
55
Albert of Aachen, Historia Ierosolimitana, I.11, ed. and trans. Edgington,
pp. 22–23.
186
inter-Christian violence, prudent men interceded to mitigate their fury.56

Take, for instance, his account of Tancred of Hauteville’s reaction to the
execution of Muslims who had been granted his banner as a sign of pro-
tection during the massacre at Jerusalem: ‘Tancred, the glorious knight,
was fired with violent anger about this insult to him, and his rage would
not have quietened down without discord and great vengeance, except
for the advice and opinion of greater and wiser men, who tempered his
mind.’57 It is clear that Albert felt Tancred had legitimate cause for anger:
it was in response to an iniuria and would have manifested vengeance.58
Nevertheless, despite being justly motivated, the implication is that his
anger – characterised as ira and then, tellingly, as furor – was excessive and
out of control, to the point that it would have resulted in Christian blood-
shed. As such, ‘greater and wiser men’ were required to calm his rancour.
Conclusion
In recent assessments of crusader motivation, historians have prioritised
documentary evidence – above all, charters – over the narrative accounts
of the First Crusade, all of which were contaminated by hindsight.59
While the narratives may not allow us to gauge what it actually felt like
to be a crusader, they can still offer an insight into the passions twelfth-
century contemporaries believed participants ought to possess. Emotions
and emotional expressions, it has been argued, constituted a relatively
modest, but nonetheless significant, part of the overall make-up of the
56
On the wise men topos, see Miriam R. Tessera, ‘“Prudentes homines […] qui
sensus habebant magis exercitatos”: A Preliminary Inquiry into William of
Tyre’s Vocabulary of Power’, Crusades 1 (2002), 63–71.
57
pp. 440–41: ‘Tancradus uero miles gloriosus super hac sibi illata iniuria
ira uehementi accensus est, nec sine discordia et grandi ultione furor illius
quieuisset, nisi consilium et sententia maiorum ac prudentium illius animum
in hiis uerbis temperasset.’
58
Throop, Crusading as an Act of Vengeance, pp. 36–37.
59
Giles Constable, ‘Medieval Charters as a Source for the History of the
Crusades’, in Crusade and Settlement, ed. Peter W. Edbury (Cardiff, 1985), pp.
73–89; Marcus Bull, Knightly Piety and the Lay Response to the First Crusade:
The Limousin and Gascony, c.970–c.1130 (Oxford, 1993); Bull, ‘The Diplomatic
of the First Crusade’, in The First Crusade: Origins and Impact, ed. Phillips,
pp. 35–54; Jonathan Riley-Smith, The First Crusaders, 1095–1131 (Cambridge,
1997).
187
Stephen J. Spencer
ideal miles Christi represented in the narratives. Herein lies the significance
of emotions in the texts: they helped to communicate the spirituality and
motives expected of a crusader.
Although admissions of participants experiencing dread in battle do
feature in the narratives, it appears that the idealised crusader did not
fear death, but instead placed his hope completely in God. Furthermore,
in at least some of the texts, martyrdom was accepted in a cheerful and
undaunted spirit. In both of these ways, the vocabulary of fear and fear-
lessness exemplified the devotion of protagonists. In similar fashion,
weeping was recognised as an outward sign of the internal feelings and
pious convictions participants were believed to hold. Writers assigned a
range of functions to the Christians’ tears, both secular and religious in
orientation. They might, for instance, invite the mercy of another pro-
tagonist or express sorrow over the dead. But they could also symbolise
the importance of Jerusalem or serve as instruments for petitioning and
thanking God, upon whose mercy, for many narrators, the expedition’s
fortunes rested.
Conversely, anger featured as a violent passion which rarely possessed a
distinctly religious character. Instances of the crusaders’ anger over Muslim
sacrilege or their supposed contamination of Jerusalem’s holy places are, on
the whole, quite rare; more frequently they were depicted as demonstrat-
ing righteous wrath in response to physical injuries, which had the effect
of justifying acts of brutality prosecuted by combatants. At the same time,
chroniclers seem to have regarded anger as a socially dangerous sentiment
which, if unrestrained, could lead to inter-Christian violence and which,
therefore, needed to be managed – either by the protagonists themselves
or by wiser men in the army. In both regards, the First Crusade does not
appear to have engendered a distinct or particularly novel set of approaches
to anger, for it operated in precisely the same ways in contemporaneous
historical narratives outside a crusading context: as the legitimate reaction
to injuries and as a passion which needed to be controlled.60
Close reading of emotional language in the narratives also reveals
differences between the ‘eyewitness’ testimonies and texts composed by
non-participants. Although discernible in the former, the notion that cru-
saders ought to fearlessly accept martyrdom gained greater currency in the
60
See Gerd Althoff, ‘Ira Regis: Prolegomena to a History of Royal Anger’,
in Anger’s Past, pp. 59–74; Richard E. Barton, ‘“Zealous Anger” and the
Renegotiation of Aristocratic Relationships in Eleventh- and Twelfth-Century
France’, in Anger’s Past, pp. 153–70; Albrecht Classen, ‘Anger and Anger
Management in the Middle Ages: Mental-Historical Perspectives’, Mediaevistik
19 (2006), 21–50.
188
accounts of non-participants, and there are strong indications that several

of these ‘second-generation’ commentators interpreted participants’ tears,
most notably those shed before the Holy Sepulchre in 1099, in light of
the doctrine of compunction. Moreover, anger was poorly represented
in the earliest narratives but became more prominent in certain second-
hand accounts, particularly Albert of Aachen’s Historia, although it was
seemingly still conceived as an inherently dangerous passion.
Thus, taken together with Throop’s observation that non-participant
chroniclers of the expedition emphasised the theme of vengeance more
than the ‘eyewitnesses’, the three case-studies considered here testify that
even within the first decade of the twelfth century the emotional rhetoric
of crusading underwent significant developments.61
61
Throop, Crusading as an Act of Vengeance, pp. 49–52.
189
The Siege and Capture of Jerusalem
in Western Narrative Sources of the First Crusade
Alan V. Murray
University of Leeds
One of the most tangible signs of the impact of the First Crusade on
the consciousness of its time was the way in which it was reflected in the
history writing of western Christendom. Few annals or local histories
written around the beginning of the twelfth century fail to mention the
council of Clermont and the capture of Antioch and Jerusalem. In many
cases, these bare narrative bones are fleshed out with other details, such
as the names of departing crusaders from the region where the source was
written. Yet it is only at the other end of the historiographical scale that
the impression made on contemporaries can be fully appreciated. One
might question whether any other event in the Middle Ages brought
forth so many dedicated narratives as did the First Crusade: more than
a dozen accounts in prose, verse, or mixed format composed during the
following sixty years which were devoted primarily to the course of the
expedition. Between these two generic extremes some longer histories
contain accounts of the crusades which are themselves comparable in
length with some of the dedicated narratives. Three of the most significant
of these originated in the Anglo-Norman realm: the Historia Anglorum
of Henry of Huntingdon,1 the Gesta Regum Anglorum of William of
Malmesbury,2 and the Historia Ecclesiastica of Orderic Vitalis,3 but the
1
Henry of Huntingdon, Historia Anglorum: The History of the English People,
ed. and trans. Diana Greenway (Oxford, 1996).
2
William of Malmesbury, Gesta Regum Anglorum, ed. and trans. R. A. B. Mynors,
Rodney M. Thomson and Michael Winterbottom, 2 vols (Oxford, 1998–99);
Rodney M. Thomson, William of Malmesbury, 2nd edn (Woodbridge, 2003);
Thomson, ‘William of Malmesbury, Historian of the Crusade’, Reading
Medieval Studies 23 (1997), 121–34.
3
The Ecclesiastical History of Orderic Vitalis, ed. and trans. Marjorie Chibnall,
6 vols (Oxford, 1969–80); Marjorie Chibnall, The World of Orderic Vitalis
(Oxford, 1984).
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. 11–1.
Alan V. Murray
trend that they exemplified continued for hundreds of years. Well into the
fifteenth century, vernacular accounts of the crusade were still being woven
into elaborate universal histories produced by anonymous compilers for
French patrons, some of them in the form of rolls displaying multiple
parallel narratives.4 Such histories tended to derive primarily from Old
French translations of the chronicle of William of Tyre, whose lengthy
Latin history of Outremer was written in the 1180s.5 Although its account
of the crusade was largely derivative of earlier sources, William’s chronicle
was undoubtedly the single most influential history of the expedition
from the later Middle Ages to the second half of the nineteenth century,
when a new range of sources became known as a result of the work of
scholars such as Heinrich von Sybel and Heinrich Hagenmeyer, and
above all through the collaborative publishing programme of the Recueil
des Historiens des Croisades (RHC), produced under the auspices of the
Académie des Inscriptions et Belles-Lettres of Paris.
1. The Dedicated Narrative Sources

The main ‘dedicated’ accounts of the crusade are (in approximate order
of composition):6
(1) the anonymous Gesta Francorum et aliorum Hierosolymitanorum
(c. 1100) written by a follower of Bohemund, later prince of Antioch,
and generally considered to have been a layman;7
4
Good examples of this genre are MSS. Leeds, University Library, Brotherton
Collection 100, and Manchester, John Rylands University Library, Fr.99. See
Oliver Pickering, ‘The Crusades in Leeds University Library’s Genealogical
History Roll’, in From Clermont to Jerusalem: The Crusades and Crusader
Societies, 1095–1500, ed. Alan V. Murray (Turnhout, 1998), pp. 251–66.
5
Guillaume de Tyr, Chronique, ed. Robert B.C. Huygens, 2 vols (Turnhout,
1986).
6
The precise datings of the works are in many cases debateable or uncertain,
particularly because of the sometimes considerable gap between the
commencement of writing and the dissemination of a completed version.
Dates given here are therefore simply meant to provide an approximate relative
chronology.
7
Anonymi Gesta Francorum et aliorum Hierosolimitanorum, ed. Heinrich
Hagenmeyer (Heidelberg, 1890); La Geste des Francs: Chronique anonyme de la
Première Croisade, trans. Aude Matignon (Paris, 1992); Histoire anonyme de la
Première Croisade, ed. and trans. Louis Bréhier (Paris, 1924); Gesta Francorum
et aliorum Hierosolimitanorum, ed. and trans. Rosalind Hill (London, 1962);
La Gesta dei Franchi e degli altri pellegrini gerosolimitani, ed. and trans. Luigi
192
The Siege of Jerusalem in Narrative Sources
(2) the Historia Francorum qui ceperunt Iherusalem or simply Liber

(c. 1101) by Raymond of Aguilers, chaplain to Raymond of Saint-Gilles,
count of Toulouse;8
Rosso (Alessandria, 2003); 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; Conrad Witzel, ‘Le problème de
l’auteur des Gesta Francorum’, Le Moyen Age 61 (1955), 319–28; Hans Oehler,
‘Studien zu den Gesta Francorum’, Mittellateinisches Jahrbuch 6 (1970), 58–97;
Rosalind Hill, ‘Crusading Warfare: A Camp-Follower’s View, 1097–1120’, in
Proceedings of the Battle Conference on Anglo-Norman Studies, I, ed. R. Allen
Brown (Ipswich, 1979), pp. 75–83, 209–11; B. Skoulatos, ‘L’auteur anonyme
des Gesta Francorum et aliorum Hierosolimitanorum et le monde byzantine’,
Byzantion 50 (1980), 504–32; Kenneth Baxter Wolf, ‘Crusade and Narrative:
Bohemond and the Gesta Francorum’, Journal of Medieval History 17 (1991),
207–16; Colin Morris, ‘The Gesta Francorum as Narrative History’, Reading
Medieval Studies 19 (1993), 55–71; John France, ‘The Use of the Anonymous
Gesta Francorum in the Early Twelfth-Century Sources for the First Crusade’,
in From Clermont to Jerusalem, pp. 29–42; France, ‘The Anonymous Gesta
Francorum and the Historia Francorum qui ceperunt Iherusalem of Raymond
of Aguilers and the Historia de Hierosolimitano 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. 39–70;
Jeanette Beer, ‘Heroic Language and the Eyewitness: The Gesta Francorum
and La Chanson d’Antioche’, in Echoes of the Epic: Studies in Honor of Gerard F.
Brault, ed. David P. Schenck and Mary Jane Schenck (Birmingham, AL, 1998),
pp. 1–16; Emily Albu, ‘Probing the Passions of a Norman on Crusade: The Gesta
Francorum et aliorum Hierosolimitanorum’, in Anglo-Norman Studies, XXVII,
ed. John Gillingham (Woodbridge, 2005), pp. 1–15; Conor Kostick, ‘A Further
Discussion of the Authorship of the Gesta Francorum’, Reading Medieval
Studies 35 (2009), 1–14; Léan Ní Chléirigh, ‘The Treatment of the Byzantines
in the Anonymous Gesta Francorum et aliorum Hierosolymitanorum’, Bulletin
of International Medieval Research 15–16 for 2009–10 (2011), 77–101.
8
‘Raimundi de Aguilers canonici Podiensis historia Francorum qui ceperunt
Iherusalem’, in RHC Occ. 3: 231–309; Raymond of Aguilers, Liber, ed. John H.
Hill and Laurita L. Hill (Paris, 1969); Raymond of Aguilers, Historia Francorum
qui ceperunt Iherusalem, trans. John H. Hill and Laurita L. Hill (Philadelphia,
1968); Jean Richard, ‘Raymond d’Aguilers, historien de la première croisade’,
Journal des Savants 3 (1971), 206–12; Christoph Auffarth, ‘“Ritter” und
“Arme” auf dem Ersten Kreuzzug, ausgehend von Raimund von Aguilers’,
Saeculum 40 (1989), 39–55. There is a short account written as a continuation
193
Alan V. Murray
(3) the Historia de Hierosolimitano itinere (1101/1111) by the southern

French priest Peter Tudebode;9
(4) the Historia Hierosolimitana by Fulcher, a priest from Chartres who
became chaplain to Baldwin of Boulogne, later count of Edessa and future
king of Jerusalem, written in a first redaction by 1106 and considerably
revised and augmented by 1127;10
(5) the Hierosolymita (1103/1125), a repeatedly reworked chronicle by
Ekkehard of Aura, a monk of St Michael at Bamberg and a participant
in the Crusade of 1101;11
(6) the Dei Gesta per Francos (1106/1109) of Guibert, abbot of
Nogent-sous-Coucy;12
of an incomplete version of the history of Raymond of Aguilers, extant in MS.

Paris, Bibliothèque nationale de France, lat.5132, ff. 15v-19v. Evidently by an
anonymous writer from southern France, it may actually be an independent
source, although it drew primarily on Raymond for its information. See John
France, ‘An Unknown Account of the Capture of Jerusalem’, English Historical
Review 87 (1972), 771–83.
9
Historia de Hierosolymitano itinere, ed. John H. Hill and Laurita L. Hill
(Paris, 1977); ‘Petri Tudebodi seu Tudebovis sacerdotis Sivracensis historia de
Hierosolymitano itinere’, in RHC Occ. 3: 1–117; Historia de Hierosolymitano
itinere, trans. John H. Hill and Laurita L. Hill (Philadelphia, 1974); France,
‘The Use of the Anonymous Gesta Francorum’.
10
Fulcheri Carnotensis Historia Hierosolymitana (1095–1127), ed. Heinrich
Hagenmeyer (Heidelberg, 1913); Fulcher of Chartres, A History of the Expedition
to Jerusalem 1095–1127, trans. Frances S. Ryan, ed. Harold S. Fink (Knoxville,
1960); Wolfgang Giese, ‘Untersuchungen zur Historia Hierosolymitana des
Fulcher von Chartres’, Archiv für Kulturgeschichte 69 (1987), 62–115; Verena
Epp, Fulcher von Chartres: Studien zur Geschichtsschreibung des ersten
Kreuzzuges (Düsseldorf, 1990); Epp, ‘Miles und militia bei Fulcher von
Chartres und seinen Bearbeitern’, in Militia Christi e Crociata nei secoli XI-XIII
(Milano, 1992), pp. 769–84; Edward Peters, The First Crusade: The Chronicle of
Fulcher of Chartres and Other Source Materials, 2nd edn (Philadelphia, 1998).
11
‘Ekkehardi abbatis Uraugiensis Hierosolymita’, in RHC Occ. 5: 1–40; Frutolfs
und Ekkehards Chroniken und die Anonyme Kaiserchronik, ed. Franz-Josef
Schmale and Irene Schmale-Ott (Darmstadt, 1972); Franz-Josef Schmale,
‘Überlieferungskritik und Editionsprinzipien der Chronik Ekkehards von
Aura’, Deutsches Archiv für Erforschung des Mittelalters 27 (1971), 110–24.
12
‘Historia quae dicitur Gesta Dei per Francos’, in RHC Occ. 4: 113–263;
Guitbertus abbas S. Mariae Nogenti, Dei Gesta per Francos et cinq autres textes,
ed. Robert B. C. Huygens (Turnhout, 1996); Huygens, La tradition manuscrite
de Guibert de Nogent (Steenbrugge, 1991); The Deeds of God Through the Franks:
194
(7) the Gesta Francorum Iherusalem expugnantium (c. 1108) by Bartolf

of Nangis;13
(8) the Historia Hierosolymitana (1107/1100) of Robert, a monk of the
abbey of Saint-Rémi at Rheims;14
(9) The Historia Jerosolimitana (c. 1108) of Baldric of Bourgueil, later
archbishop of Dol-de-Bretagne;15
(10) the prosimetric Gesta Tancredi (1112/1118) by the Norman priest
Radulph of Caen;16
A Translation of Guibert de Nogent’s Gesta Dei per Francos, trans. Robert

Levine (Woodbridge, 1997); Geste de Dieu par les Francs, trans. M. C. Garand
(Turnhout, 1998); Jay Rubenstein, Guibert of Nogent: Portrait of a Medieval
Mind (New York, 2002); Maria Lodovica Arduini, ‘Il problema christianitas
in Guiberto di Nogent’, Aevum 78 (2004), 379–410; Léan Ní Chléirigh, ‘Anti-
Byzantine Polemic in the Dei gesta per Francos of Guibert, Abbot of Nogent-
sous-Coucy’, in Sailing to Byzantium, ed. Savvas Neocleous (Newcastle upon
Tyne, 2009), pp. 53–74.
13
‘Gesta Francorum Iherusalem expugnantium’, in RHC Occ. 3: 491–543; Susan
B. Edgington, ‘The Gesta Francorum Iherusalem expugnantium of “Bartolf of
Nangis”’, Crusades 13 (2014), 21–35.
14
‘Roberti Monachi historia Iherosolimitana’, in RHC Occ. 3: 717–882; Robert
the Monk’s History of the First Crusade: The Historia Iherosolimitana, trans.
Carol Sweeetenham (Aldershot, 2005); The Historia Iherosolimitana of Robert
the Monk, ed. Damien Kempf and Marcus G. Bull (Woodbridge, 2013);
Georg Marquardt, Die Historia Hierosolymitana des Robertus Monachus: Ein
quellenkritischer Beitrag zur Geschichte des 1. Kreuzzugs (Königsberg, 1892);
Thomas Martin Buck, ‘Von der Kreuzzugsgeschichte zum Reisebuch: Zur
Historia Hierosolymitana des Robertus Monachus’, Deutsche Vierteljahres
schrift für Literaturwissenschaft und Geistesgeschichte 76 (2002), 321–55; Luigi
Russo, ‘Ricerche sull’Historia Iherosolimitana’, Studi medievali ser.3, 43 (2002),
651–91.
15
‘Baldrici episcopi Dolensis Historia Jerosolimitana’, in RHC Occ. 4: 1–111;
The Historia Ierosolimitana of Baldric of Bourgueil, ed. Steven J. Biddlecombe
(Woodbridge, 2014); Steven J. Biddlecombe, ‘Baldric of Bourgueil and the
Flawed Hero’, in Anglo-Norman Studies, XXXV, ed. David Bates (Woodbridge,
2013), pp. 79–93.
16
‘Gesta Tancredi in expeditione Hierosolymitana […] auctore Radulfo
Cadomensi’, in RHC Occ. 3: 587–716; Radulphi Cadomensis Tancredus, ed.
Edoardo D’Angelo (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); H. Glaesener, ‘Raoul de Caen, historien
et écrivain’, Revue d’histoire ecclésiastique 46 (1951), 5–21; Laetitia Boehm, ‘Die
Gesta Tancredi des Radulph von Caen: Ein Beitrag zur Geschichtsschreibung
195
Alan V. Murray
(11) the Historia vie Hierosolimitane (before 1120), an epic poem writ-
ten by a monk of Cluny, Gilo of Paris, and subsequently augmented by
an anonymous author from Champagne or Lotharingia, now generally
known in scholarship as the Charleville Poet;17
(12) the Historia Ierosolimitana (1102/1130) by Albert, a cleric of
Aachen;18
der Normannen um 1100’, Historisches Jahrbuch 75 (1956), 47–72; Jean-

Charles Payen, ‘Une légende épique en gestation: Les Gesta Tancredi de Raoul
de Caen’, in La Chanson de geste et le mythe carolingien: Mélanges René Louis,
ed. Emmanuèle Baumgartner et al., 2 vols (Saint-Père-sous-Vézelay, 1982),
2: 1051–62; Kaspar Elm, ‘O beatas idus ac prae ceteris gloriosas! Darstellung
und Deutung der Eroberung Jerusalems 1099 in den Gesta Tancredi des
Raoul von Caen’, in Es hat sich viel ereignet, Gutes wie Böses: Lateinische
Geschichtsschreibung der Spät- und Nachantike, ed. Gabriele Thome and Jens
Holzhausen (München, 2001), pp. 152–78.
17
The Historia vie Hierosolimitane of Gilo of Paris, and a Second, Anonymous Author,
ed. and trans. Chris W. Grocock and Elizabeth Siberry (Oxford, 1997); Chris
W. Grocock, ‘L’aventure épique: le traitment poétique de la Prèmiere Croisade
par Gilon de Paris et son continuateur’, in Autour de la Première Croisade:
Actes du Colloque de la Society for the Study of the Crusades and the Latin East
(Clermont-Ferrand, 22–25 juin 1995), ed. Michel Balard (Paris, 1996), pp. 17–28;
Peter Christian Jacobsen, ‘Die Admonter Versifikation der Kreuzzugsgeschichte
Roberts von St-Remi’, in Literatur und Sprache im europäischen Mittelalter:
Festschrift für Karl Langosch zum 70. Geburtstag, ed. Alf Önnerfors, Johannes
Rathofer and Fritz Wagner (Darmstadt, 1973), pp. 142–72.
18
‘Alberti Aquensis Historia Hierosolimitana’, in RHC Occ. 4: 265–713; Albert
of Aachen, Historia Ierosolimitana: History of the Journey to Jerusalem, ed. and
trans. Susan B. Edgington (Oxford, 2007); Geschichte des ersten Kreuzzuges,
trans. Hermann Hefele, 2 vols ( Jena, 1923); André Alden Beaumont, ‘Albert
of Aachen and the County of Edessa’, in The Crusades and Other Historical
Essays presented to Dana C. Munro, pp. 101–38; Peter Knoch, Studien zu
Albert von Aachen: Der 1. Kreuzzug in der deutschen Chronistik (Stuttgart,
1966); Dietrich Lohrmann, ‘Albert von Aachen und die Judenpogrome des
Jahres 1096’, Zeitschrift des Aachener Geschichtsvereins 100 for 1995–1996
(1996), 129–51; Colin Morris, ‘The Aims and Spirituality of the Crusade as
Seen Through the Eyes of Albert of Aix’, Reading Medieval Studies 16 (1990),
99–117; Susan B. Edgington, ‘The First Crusade: Reviewing the Evidence’,
in The First Crusade: Origins and Impact, ed. Jonathan Phillips (Manchester,
1997), 57–77; Edgington, ‘Albert of Aachen Reappraised’, in From Clermont to
Jerusalem, pp. 55–67; Edgington, ‘Albert of Aachen and the chansons de geste’,
in The Crusades and Their Sources: Essays Presented to Bernard Hamilton,
ed. John France and William G. Zajac (Aldershot, 1998), pp. 23–37; Filippo
196
(13) a work now generally known as the Historia Belli Sacri (c. 1130) by
an anonymous monk of Montecassino;19
(14) the Expeditio Ierosolimitana (1146/1165), a verse epic by the
Bavarian monk Metellus of Tegernsee;20
(15) the Historia Nicaena vel Antiochena (c. 1146) composed by an
anonymous author in Outremer at the behest of King Baldwin III of
Jerusalem.21
The authorship of these fifteen sources was overwhelmingly clerical,
as we might expect for this period in history. More significant is the
geographical distribution of the areas where their authors wrote their
accounts: five were situated in northern France (Guibert of Nogent,
Baldric of Dol, Robert the Monk, Bartolf of Nangis, Gilo of Paris and
his continuator), three in the kingdom of Germany (Ekkehard of Aura,
Albert of Aachen, and Metellus of Tegernsee), two in Occitan-speaking
southern France (Raymond of Aguilers and Peter Tudebode), two in
the kingdom of Jerusalem (Fulcher of Chartres and the author of the
Historia Nicaena vel Antiochena), and two in the Norman principali-
ties of Antioch and southern Italy (the Anonymous of the Gesta and
Radulph of Caen), to which grouping we could add the Historia Belli
Sacri of Montecassino. Local or regional interests meant that many of
the narratives view the crusade from the perspectives of particular con-
tingents and leaders: Raymond of Saint-Gilles and the southern French
in the accounts of Raymond of Aguilers and Peter Tudebode; Godfrey
of Bouillon, Baldwin I of Jerusalem and the Lotharingians and Germans
in those of Albert, Metellus, Gilo of Paris and the Charleville Poet; and
Bohemund, Tancred and the Normans of southern Italy and Normandy
Andrei, ‘Alberto di Aachen e la Chanson de Jérusalem’, Romance Philology 63

(2009), 1–69.
19
‘Tudebodus imitatus et continuatus […] Historia peregrinorum euntium
Jerusolymam ad liberandam Sancti Sepulcri de potestate ethnicorum’, in RHC
Occ. 3: 167–229; Historia de via et recuperatione Antiochae atque Ierusolymarum,
ed. Edoardo D’Angelo (Firenze, 2009); John France, ‘Note sur le manuscrit
6041A du fonds latin de la Bibliothèque nationale: Un nouveau fragment d’un
manuscript de l’Historia Belli Sacri’, Bibliothèque de l’Ecole des Chartes 126
(1968), 413–16.
20
Metellus von Tegernsee, Expeditio Ierosolimitana, ed. Peter Christian Jacobsen
(Stuttgart, 1982); Jacobsen, ‘Die Admonter Versifikation der Kreuzzugsgedichte
Roberts von St-Remi’.
21
‘Historia Nicaena vel Antiochena’, in RHC Occ. 5: 136–85.
197
Alan V. Murray
in those of the Anonymous and Radulph of Caen.22 The northern French

authors offer a noticeably less particularist perspective, although Guibert
does betray an explicit contempt for the Germans and their part in the
expedition.23
Titles of medieval works were often unstable as a result of the vagaries
of transmission, but the list given above reveals something about how
the authors conceived their themes. Several titles relate to the ‘deeds of
the Franks’, achieved implicitly, or – in the case of Guibert of Nogent –
explicitly, by divine direction and favour.24 More of them, however, stress
the centrality of Jerusalem in the crusade: the ‘journey to Jerusalem’, the
‘Jerusalemite history’ and so on. The majority of the narratives treat the
capture of Jerusalem on 15 July 1099 as the culmination of the expedi-
tion and end their accounts either with the occupation of the holy city
or with the battle of Ascalon on 15 August 1099, which secured the
Christian possession of southern Palestine from the Fātimids and which
was followed by the return of the majority of pilgrims to their homes in
the West. Fulcher of Chartres, Albert of Aachen, and to a lesser extent,
Guibert of Nogent, Baldric of Dol, Bartolf of Nangis and Radulph of
Caen continue their narratives to take in the early history of Outremer,
but it is the crusade itself which is at the centre of their works. Since an
exhaustive treatment of all of the sources is not possible here, the fol-
lowing pages will attempt to highlight some of the differences in those
narratives which are generally considered to be most important by
22
One aspect of these perspectives is the prosopographical information that can
be obtained about the different contingents. On this matter, see especially
Evelyn M. Jamison, ‘Some Notes on the Anonymi Gesta Francorum, with
Special Reference to the Norman Contingent from South Italy and Sicily
in the First Crusade’, in Studies in French Language and Medieval Literature
presented to Professor Mildred K. Pope (Manchester, 1939), pp. 195–204 and
Ernesto Pontieri, ‘I Normanni dell’Italia meridionale e la Prima Crociata’,
Archivio storico italiano 114 (1956), 3–17 (for the southern Italian Normans);
Charles W. David, Robert Curthose, Duke of Normandy (Cambridge, MA,
1920) (for the Normans of Normandy); and Alan V. Murray, ‘The Army of
Godfrey of Bouillon, 1096–1099: Structure and Dynamics of a Contingent on
the First Crusade’, Revue belge de philologie et d’histoire 70 (1992), 301–29 (for
the Lotharingians and Flemings).
23
‘Historia quae dicitur Gesta Dei per Francos’, in RHC Occ. 4: 136.
24
On this idea in particular, see Antonius Hendrikus van Erp, Gesta Francorum:
Gesta Dei? Motiviering en rechtvaardiging van de eerste kruistochten door
tijdgenoten en moslimse reactie (Amsterdam, 1982).
198
examining their portrayal of the siege and capture of the city of Jerusalem
in June–July 1099.
2. Participants or Eyewitnesses?
Since the late nineteenth century historical scholarship has repeatedly
stressed the primacy of those accounts written by participants in the
crusade: Peter Tudebode, Raymond of Aguilers, Fulcher of Chartres, and
the anonymous author of the Gesta Francorum. Even among the narra-
tives written by non-participants there is a clear hierarchy. Largely thanks
to the efforts of its most recent editor, Susan Edgington, the Historia of
Albert of Aachen has been re-established as the most important of these,
even though it differs significantly in content from the narratives by par-
ticipants, while the histories of Robert the Monk, Guibert of Nogent,
Radulph of Caen and Baldric of Dol are also frequently cited by modern
historians. By contrast, one will rarely find references to the likes of the
Historia Nicaena, Gilo of Paris or Metellus of Tegernsee in studies of the
crusade, even though they contain details not known from other sources;
the neglect of the latter two may have to do with modern historians’
suspicions of poetry as a vehicle for the writing of history.
Of the four accounts composed by participants, the Gesta Francorum
has acquired an almost unique position in scholarship, which is reflected
in the high number of studies devoted to it relative to the other sources.
It had a noticeable influence on the account of Peter Tudebode, and to a
lesser extent, on that of Raymond of Aguilers, as well as on many of the
other works. John France has argued that the influence of the Anonymous
on an entire ‘Gesta family’, together with the wide availability of relatively
cheap modern editions, has led to a situation in which the Gesta has come
to be regarded as the ‘normal’ account of the crusade, a default setting
by which other sources are judged.25 The identification of this norma-
tive tendency should serve as a warning against overvaluing the Gesta
over the accounts of the other participants, but France also makes the
point that even the other branches of the Gesta family need to viewed
as works in their own right, whose testimony should not be dismissed
as merely derivative. It is because of their status as eyewitnesses that the
Anonymous and the other participants in the crusade have tradition-
ally been accorded a greater importance than authors whose works were
removed in space or time from the events they described. However, it
25
France, ‘The Use of the Anonymous Gesta Francorum in the Early Twelfth-
Century Sources for the First Crusade’, pp. 29–30.
199
Alan V. Murray
is important to remember that few medieval sources can be considered

as eyewitnesses in an absolute sense. Thus, while Fulcher of Chartres
was a participant in the expedition, according to his own testimony
he was not present at one of the key events he describes, the capture of
Jerusalem, since he had remained with Baldwin of Boulogne after the
latter established the county of Edessa.26 As Jean Flori has argued, this
circumstance is probably the most convincing explanation for Fulcher’s
failure to mention Jerusalem in his account of Urban II’s speech at the
council of Clermont.27 By contrast, Robert the Monk, who was not a
participant in the expedition itself, was an eyewitness at Clermont, a
circumstance which probably brought him his commission to write his
history.28
The distinction made between participants and eyewitnesses in the above
remarks is perhaps a helpful one to maintain in view of recent research,
which has called the whole idea of an ‘eyewitness account’ into question.
Yuval Harari has argued that ‘eyewitness accounts’ are a distinct type of
historical writing ‘defined by their purposes as much as by their sources’,
and possessing particular values and characteristics; according to such
criteria, ‘a text lacking these values and characteristics can hardly be consid-
ered an eyewitness account, even if it was written by an eyewitness’.29 His
arguments enable a useful distinction to be made among writers who were
participants in the expedition, specifically between Fulcher of Chartres
and Raymond of Aguilers on the one hand and the Gesta Francorum on
the other. Thus Fulcher repeatedly draws attention to the question of truth
and truthfulness, and explicitly underlines his own status as a protagonist
in events as a means of establishing the veracity of his statements. He also
makes a consistent formal distinction through the use of the first person
plural to document his own presence as an eyewitness, in contrast to the
third person in connection with events where he was absent, such as the
capture of Jerusalem. By Harari’s criteria, the Gesta fails the test of an
eyewitness account, despite being written by a participant. Its author does
not try to establish an eyewitness status, nor is it all clear at which events
he was himself present. Indeed, the Anonymous often gives the impres-
26
Fulcheri Carnotensis Historia Hierosolymitana, ed. Hagenmeyer, I.xxxiii,
pp. 322–34.
27
Jean Flori, La première croisade: L’Occident chrétien contre l’Islam (Bruxelles,
1992), pp. 31–32.
28
Robert the Monk’s History of the First Crusade, ed. Sweetenham, p. 2.
29
Yuval Noah Harari, ‘Eyewitnessing in Accounts of the First Crusade: The Gesta
Francorum and Other Contemporary Narratives’, Crusades 3 (2004), 77–99.
200
sion of an omniscient narrator, for example, in telling of secret councils

and plans, or in the famous constructed dialogue between the Turkish
emir Karbugha and his mother during the siege of Antioch.30 This scene,
at which the Anonymous could not possibly have been present, is not
only fictitious but quite fanciful in its depiction of the Muslim world.
Harari concludes that the author of the Gesta ‘was not attempting to write
a factually true eyewitness account of the crusade’, and sees it as closer
in spirit to an epic account.31 Together with the arguments advanced by
France, Harari’s findings undoubtedly caution us against setting too great
store by the testimony of the Gesta Francorum.
3. Narrative Structure and Content

A comparison of the relative lengths of the accounts of the siege and
capture of Jerusalem and subsequent events up to the battle of Ascalon
in ten of the most important narratives is given at Table 1.
The most striking result of this comparison is that the two accounts by
participants which are generally considered to be the more important,
those of Fulcher of Chartres and the Gesta Francorum, are the shortest.
The brevity of Fulcher’s account can be explained by his absence during
the siege, yet as a resident of Jerusalem from 1100 to 1127 he had every
opportunity to question people who had been present and remained in
the East after 1099. A more plausible reason is that his hero, Baldwin
I, was still in Edessa while the siege was going on, and that he did not
wish the capture of the city in 1099 to overshadow Baldwin’s subsequent
deeds as king of Jerusalem. Two of the accounts which are considered
to be heavily dependent on the Gesta, Raymond of Aguilers and Peter
Tudebode, are over twice as long as its treatment, while the longest nar-
rative, that of Albert of Aachen, is almost ten times the length of that
given by the Gesta. To some extent these differences can be explained by
the effects of a more prolix Latin style on the part of some of the later
authors compared with the simple and laconic language of the Gesta.
However, the variations in length are primarily the result of significant
differences in content.32
30
Gesta Francorum et aliorum Hierosolimitanorum, ed. and trans. Hill, pp. 52–54.
31
Harari, ‘Eyewitnessing in Accounts of the First Crusade’, p. 91.
32
It must be admitted that the relative shortness of Albert’s chapters means that
the layout of the RHC edition contains much blank space, but this only partly
explains the significant difference in length.
201
Alan V. Murray
Table 1: Length of Accounts of the Siege and Capture of Jerusalem in Selected

Narrative Sources of the First Crusade (= number of pages in RHC editions)
Gesta Francorum
Raymond of Aguilers
Peter Tudebode
Fulcher of Chartres
Guibert of Nogent
Bartolf of Nangis
Baldric of Dol
Robert the Monk
Ralph of Caen
Albert of Aachen
A comparison of the content of the various accounts is given in Table 2.

This does not claim to be a totally comprehensive analysis, but is rather
an attempt to highlight those sections which seem to be key elements of
each account.
The basic structure of the narrative of the siege and capture of Jerusalem
common to most of the sources consists of the following elements (in
varying order):
(1) the arrival of the crusade army on 7 June and its dispositions for
the siege;
(2) the first assault on the city, on 13 June;
(3) the expedition sent under Raymond Pilet and others to meet ships
which had put in at Jaffa on 17 June;
(4) privations suffered by the crusaders because of lack of water and
food;
(5) the search for wood and construction of siege engines;
(6) a procession around the walls, culminating in a sermon on the
Mount of Olives, on 8 July;
(7) the series of assaults with siege machinery, culminating in the cap-
ture of the city on 15 July;
(8) the role in the breakthrough played by Godfrey of Bouillon, his
brother Eustace of Boulogne, and the knight Letold of Tournai (and
his brother Engelbert);
202
(9) the plundering and occupation of houses and seizure of captives;

(10) the massacre of the Muslim and Jewish inhabitants of the city;
(11) the surrender of the Tower of David by its Fātimid garrison to
Raymond of Saint-Gilles;
(12) the seizure of the Temple by Tancred;
(14) worship at the church of the Holy Sepulchre;
(15) the election of Godfrey of Bouillon as ruler on 22 July;33
(16) the election of Arnulf of Chocques as patriarch of Jerusalem;
(17) the defeat of a Fātimid army at the battle of Ascalon on 12 August.
As far as the basic narrative is concerned, there is considerable congruence
between the four accounts by participants, as one might expect, bearing
in mind the influence of the Gesta Francorum. Does this point to the
primacy of the Gesta and its derivatives, as was long accepted by crusade
scholarship, or do the different depictions provide support for the doubts
raised about it by France, Harari and others? The Gesta is undoubtedly
the most rudimentary and terse of all the narratives, describing the march
from Ramla to Jerusalem in less than a sentence.34 The other sources add
considerably more detail to the basic outline made by the Gesta, as well as
several points of independent testimony. Peter Tudebode and Raymond of
Aguilers, for example, tell how the Turks on the walls of the city mocked
the besiegers.35 Raymond gives an excursus on the capture of Bethlehem
which leads on to the digest of a debate among the crusaders about the
future form of government for Jerusalem once it was in Christian hands.36
The sources other than the Gesta Francorum, both participants and
non-participants, provide context for the events of the siege which is often
absent from the Anonymous’s account. This can be seen in the accounts
of the construction of siege machines, which proved to be the key to cru-
sader success after the first, fruitless assault in early June. The Anonymous
introduces this important tactical development with the unmotivated
statement that ‘Our leaders then decided to attack the city with engines,
so that we might enter it and worship at our Saviour’s Sepulchre’. The only
33
Many of the sources mistakenly refer to Godfrey as ‘king’. On the problem of
his title as ruler of Jerusalem, see Alan V. Murray, ‘The Title of Godfrey of
Bouillon as Ruler of Jerusalem’, Collegium Medievale 3 (1990), 163–78.
34
Gesta Francorum et aliorum Hierosolimitanorum, ed. and trans. Hill, p. 87.
35
Historia de Hierosolymitano itinere, ed. Hill and Hill, p. 137; ‘Raimundi de
Aguilers canonici Podiensis historia Francorum qui ceperunt Iherusalem’, in
RHC Occ. 3: 297.
36
Iherusalem’, in RHC Occ. 3: 295–97.
203
204
Table 2: Analysis of Content in Accounts of the Siege and Capture of Jerusalem in Selected Narrative Sources of the First Crusade
GF RA PT FC GN BN BD RM RC AA
Alan V. Murray
Arrival of armies at Jerusalem (7 June) • • • • • • • • •

Disposition of besiegers • • • • • • • • •
Tancred reconnoitres and meets hermit •
Description of Jerusalem • • •
First assaults on city (13 June) • • • • • • • • • •
Turks on walls mock Christians • • • •
Arrival of fleet at Jaffa and crusader force sent there • • • • • • •
Privation of thirst • • • • • • • • •
Privation of hunger • • • • •
Capture of Bethlehem • • •
Debate about future government •
Vision of Peter Desiderius •
Construction of siege engines • • • • • • • • • •
Spies and communications of garrison with Egypt • •
Sermon and procession around the city • • • • • •
Assaults with siege engines (9–15 July) • • • • • • • • • •
Breakthrough by Godfrey/Eustace/Letold • • • • • • • •

GF RA PT FC GN BN BD RM RC AA
Biblical history of Jerusalem • • • • •
Massacre of inhabitants • • • • • • • • • •
Surrender of Tower of David to Raymond • • • • • • • • • •
Seizure of Temple by Tancred • • • • • • • •
Dispute between Tancred and Arnulf •
Visions of Stabelo, Hecelo and Giselbert •
Worship at Holy Sepulchre • • • • • • • • •
Plundering and seizure of prisoners • • • • • • • • •
Rulership offered to Raymond • • • •
Election of Godfrey as ruler • • • • • • • • •
Election of Arnulf as patriarch • • • • • • • • •
Discovery of True Cross • • • •
Disposal of corpses • • • •
Surrender of Nablus • • • • •
Previous occupation of city by Turks • •
Fatimid reaction to capture • • • • •
Battle of Ascalon • • • • • • • • • •
205
Alan V. Murray
difficulty that he recognises is that timber for the siege towers had to
be brought from far afield.37 Peter Tudebode recasts this into the more
logical statement that ‘our lords studied means by which they could take
Jerusalem and enter the Sepulchre for the purpose of adoring their Lord
and Saviour’, adding the important detail that Saracen prisoners were
employed to haul timber from distant locations.38 This is confirmed by
Raymond of Aguilers, who adds the information that the collection of
materials and construction of machinery were entrusted to Gaston of
Béarn, William Embriaco and Peter, bishop of Albara.39 Albert of Aachen
gives a far fuller account of motivation and context, stating that ‘when
this tumult of battle had subsided, since the duke and the foremost in the
army saw that the city was unconquerable by force of arms and attack,
they returned from the assault to the camp and reached a common con-
clusion, that unless the city was taken by siege engines and mangonels
it could never be conquered by any other strength of weapons’. He also
gives considerable detail about the collection of timber and the process
of construction (but omitting the Saracen forced labour).40 Guibert of
Nogent gives an even more sophisticated account of the motivation and
purpose of the construction of machinery.41 Yet it is Albert who provides
the fullest depiction of the deployment and effects of the siege engines
as well as the response of the Fātimid garrison to them up to the success-
ful final assault, in marked contrast to the Gesta, which telescopes the
account of the last event into a couple of sentences.42 This example – to
which one could add numerous others – suggests that what the reading
and listening public wanted from the later narratives of the crusade was
primarily a greater degree of detail and elucidation of events than was
given by the Gesta Francorum and its immediate derivatives.
37
Gesta Francorum et aliorum Hierosolimitanorum, ed. and trans. Hill, p. 89:
Tunc seniores nostri ordinauerunt quomodo ingeniare possent ciuitatem, ut ad
adorandum nostri Salvatoris intrarent Sepulchrum.
38
Historia de Hierosolymitano itinere, ed. Hill and Hill, p. 138: Statim nostri
seniores ordinaverunt quomodo ingeniare possent hanc civitatem, et adorando ad
Domini et Salvatoris intrarent civitatem.
39
Iherusalem’, in RHC Occ. 3: 297–98.
40
Albert of Aachen, Historia Ierosolimitana, VI.2–3, ed. Edgington, pp. 406–7.
41
‘Historia quae dicitur Gesta Dei per Francos’, in RHC Occ. 4: 225–27.
42
Albert of Aachen, Historia Ierosolimitana, VI.9–11, ed. Edgington, pp. 414–19;
Gesta Francorum et aliorum Hierosolimitanorum, ed. and trans. Hill, p. 87.
206
4. The Public Demand for Crusade Historiography:

Style, Substance, or Elucidation?
The privileged status of the Gesta Francorum in modern historiography
stands in stark contrast to the medieval reception of the events of the
crusade. Despite being promoted by Bohemund, in its original form it
evidently met little favour with contemporary consumers of literature on
the crusade, who demanded a fuller and more sophisticated treatment.
This process started with redactors of the Gesta itself; Rosalind Hill points
out that an unknown reviser, whom she calls X, made some 150 changes
in word order and 260 in vocabulary simply to improve the quality of
the Anonymous’s Latin.43 It was probably the Gesta Francorum that was
criticised for its simplicity of style by both Baldric of Dol and Guibert of
Nogent, although they claimed to know nothing about its authorship.44
It is also probable that the Gesta was the exemplar that Robert identi-
fies as having been procured for him by his abbot; even though Robert
claims a plain style for his own work, he criticised the Gesta for its lack of
sophistication, but equally for its incompleteness, notably the lack of an
account of the council of Clermont.45 It was Robert’s history that went
on to become the most popular and widely disseminated contemporary
narrative of the crusade, surviving in almost 100 manuscripts, and being
translated into German, Dutch and Italian.46 Similarly, Bartolf of Nangis
and Guibert criticised the style of the first redaction of Fulcher’s chroni-
cle, and both sought to improve on its information in their own works.
Fulcher’s second redaction shows considerable revision, which was done
not only for stylistic reasons, but also to modify his depiction of several
episodes in keeping with his perception of the subsequent history of the
kingdom of Jerusalem.47
43
Gesta Francorum et aliorum Hierosolimitanorum, ed. and trans. Hill, p. xli.
44
‘Baldrici episcopi Dolensis Historia Jerosolimitana’, in RHC Occ. 4: 10;
‘Historia quae dicitur Gesta Dei per Francos’, in RHC Occ. 4: 120.
45
‘Roberti Monachi historia Iherosolimitana’, in RHC Occ. 3: 721.
46
Robert the Monk’s History of the First Crusade, ed. Sweetenham, pp. 8–10;
Friedrich Kraft, Heinrich Steinhöwels Verdeutschung der Historia
Hierosolymitana des Robertus Monachus: Eine literarhistorische Untersuchung
(Straßburg, 1905). Historia Hierosolymitana von Robertus Monachus in
deutscher Übersetzung, ed. Barbara Haupt (Wiesbaden, 1972).
47
Edgington, ‘The Gesta Francorum Iherusalem expugnantium of “Bartolf of
Nangis”’.
207
Alan V. Murray
A good style was desired by the reading and listening public, but it was
not the only demand that fuelled the continued production of histories of
the crusade. One of the most noticeable features of the sources produced
by non-participants is the amount of space devoted to broader contextual
elucidation. This is done in several ways. Radulph of Caen and (uniquely
among the participants) Fulcher of Chartres provide descriptions of the
city of Jerusalem. Robert, Guibert, Baldric and Albert present more
sophisticated accounts of the city’s biblical history, with an emphasis on
the place of the Temple. The clearest expositions of contemporary political
circumstances are given by Guibert and Albert: they explain how, when
the crusade appeared in Syria, Jerusalem was under the control of the
Artūqid emir Sūqman, a Turkish commander in the service of the Great
Saljūq sultan, but that the city had subsequently been seized by the forces
of the Fātimid caliph of Egypt.48
A further evident demand was for a greater degree of detail. This
can be seen above all in the narratives of Radulph of Caen and Albert
of Aachen, whose depictions of the capture of Jerusalem are twice as
long as those in any of the other sources discussed here. The Gesta
Tancredi is the single source that diverges most from the common nar-
rative structure, since much of the depiction of the siege and capture
is focused on its hero, Tancred. Radulph states that Tancred arrived at
Jerusalem before the other commanders, and the four chapters imme-
diately preceding the description of the city are told entirely from his
perspective.49 This tendency becomes even more marked after the cap-
ture of the city. The election of Godfrey as ruler is not mentioned at the
appropriate point in the narrative; rather, after describing the battle of
Ascalon, Radulph mentions incorrectly that the kingdom had fallen to
Godfrey by lot. He fails to mention the election of Arnulf of Chocques
as patriarch, despite giving a lengthy account of a dispute between
Arnulf and Tancred as an illustration of the other leaders’ supposed
jealousy and victimisation of his hero.50 There is a general agreement
that the first crusader to mount the walls in the successful assault on
15 July was a knight called Letold of Tournai, fighting in a siege tower
commanded by Godfrey of Bouillon, but Radulph relegates him to a
secondary position, giving the place of honour to a Norman, Bernard
48
Albert of Aachen, Historia Ierosolimitana, VI.32, ed. Edgington, pp. 444–45;
‘Historia quae dicitur Gesta Dei per Francos’, in RHC Occ. 4: 223–24.
49
‘Gesta Tancredi in expeditione Hierosolymitana’, in RHC Occ. 3: 683–86.
50
208
of Saint-Valery-sur-Somme.51 Radulph’s tendency to laud Tancred and

the Normans to the exclusion of other leaders and contingents makes
it difficult to find agreement between the Gesta Tancredi and the other
sources. Despite its apparent relative length compared to the other
accounts, much of this is achieved through florid, prolix and often
allegorical language which frequently adds little to the detail or the
development of the narrative, and by the depiction of two extended
speeches by Arnulf and Tancred himself.52
While the account of Albert of Aachen shows a far greater corre-
spondence with those of the other sources, it also has several unique
passages, some of which serve a similar function to the digressions in the
Gesta Tancredi in that they glorify Godfrey of Bouillon, duke of Lower
Lotharingia. One relates to a vision experienced by Stabelo, the duke’s
chamberlain, before the crusade, which is interpreted by Albert as foretell-
ing Godfrey’s long desired journey to the Holy City.53 A similar episode
concerns a knight called Hezelo nomine de Kenzwilare villa, quae est in
Ribuario, who while hunting with Godfrey in a wood known as Ketena,
falls asleep and is transported to Mount Sinai where he sees a vision of
Moses, which is interpreted as a prefiguration of Godfrey’s elevation to
the rulership of Jerusalem.54 The third vision is a premonition revealed
to Giselbert, a canon of St Mary’s at Aachen, in the seventh month of
Godfrey’s pilgrimage, that the duke would be raised to be ‘chief of all and
prince in Jerusalem, preordained and appointed by God’.55 At first sight
these reports of visions seem to be fanciful elaborations, but they may also
be indicative of Albert’s working methods. The canon of Aachen would
undoubtedly have been regarded as a reliable witness and may even have
been a colleague of Albert’s. Of the knight Hezelo we know nothing,
but the geographical names associated with him can be pinpointed with
some certainty. Kenzwilare is the modern Kinzweiler, a hamlet now part
of Eschweiler about twelve kilometres north-east of Aachen, and thus in
the Ripuarian region indicated by Albert.56 The silva […] dicitur Ketena can
most readily be connected with the name of Kettenis, a village just north
51
‘Gesta Tancredi in expeditione Hierosolymitana’, in RHC Occ. 3: 693.
52
53
Albert of Aachen, Historia Ierosolimitana, VI.26, ed. Edgington, pp. 436–39.
54
Albert of Aachen, Historia Ierosolimitana, VI.34–35, ed. Edgington,
pp. 446–49.
55
Albert of Aachen, Historia Ierosolimitana, VI.36–37, ed. Edgington, pp. 448–51.
56
Maurits Gysseling, Toponymisch woordenboek van België, Nederland,
Luxemburg, Noord-Frankrijk en West-Duitsland (vóór 1226), 2 vols (Brussel,
209
Alan V. Murray
of Eupen; this area is part of the Cantons de l’Est, the German-speaking

community of eastern Belgium, and situated in the hilly, wooded
country known as the Hautes Fagnes (Ger. Hohes Venn).57 Stabelo the
Chamberlain was killed at the Second Battle of Ramla in May 1102.58 Any
information about his vision, which can only have acquired significance
after Godfrey’s election, is thus most likely to have been brought back to
the Rhineland by returning crusaders.
The inclusion of these three visions may be an indirect testimony to
a fairly meticulous process by which Albert collected information in
Aachen and beyond, from participants in the crusade as well as those who
had remained in the West but nevertheless had information about Duke
Godfrey. Such an approach would also explain the degree of detail that is
evident in Albert’s descriptions of events belonging to the crusade proper.
As we have seen, most of the sources examined agree that the first cru-
sader to mount the walls of Jerusalem on 15 July was Letold of Tournai.59
Albert gives the precise information that Letold was accompanied by
his brother Engelbert, and that the two had taken up position on the
middle story of the siege tower, while Godfrey and Eustace of Boulogne
were fighting from the upper story.60 Albert also gives details which are
not essential to the narrative, such as that Godfrey’s tower had a crucifix
mounted on it, or that a soldier standing next to the duke was killed by a
stone thrown from a mangonel on the city walls. These and other details
of the crusader assault suggest that Albert undertook a careful process of
questioning of eyewitnesses to come up with a depiction of events that
was more rounded and detailed than those produced by participants in
the expedition.
1960), p. 562. The modern form Kinzweiler shows the diphthong <ei> which
developed from the Middle High German long vowel <î>.
57
Neither geographical name is identified in the RHC edition of Albert’s
history. The author’s identifications of these two places were accepted by Dr
Susan Edgington in her edition and translation (personal communication, 27
October 2006).
58
Albert of Aachen, Historia Ierosolimitana, IX.4, ed. Edgington, pp. 640–43.
59
Gesta Francorum et aliorum Hierosolimitanorum, ed. and trans. Hill, pp. 90–91;
Historia de Hierosolymitano itinere, ed. Hill and Hill, pp. 139–40; ‘Baldrici
episcopi Dolensis Historia Jerosolimitana’, in RHC Occ. 4: 102; ‘Historia quae
dicitur Gesta Dei per Francos’, in RHC Occ. 4: 226–27; ‘Roberti Monachi
historia Iherosolimitana’, in RHC Occ. 3: 866–67.
60
Albert of Aachen, Historia Ierosolimitana, VI.11, 19, ed. Edgington, pp. 416–19,
426–29.
210
5. The Massacres of 15–17 July:

Divine Will or Military Necessity?
Albert of Aachen may well be the only source to provide a rational
explanation for one of the most controversial and bloody episodes of
the entire crusade. All of the primary accounts agree that, as soon as they
had broken into Jerusalem on 15 July, the crusaders began a massacre of
the city’s Muslim and Jewish inhabitants (its Christian population hav-
ing been expelled by the Fātimid garrison), which continued on the next
day. Albert alone relates that the crusade leaders ordered a continuation
or resumption of the massacres for a third day, that is on 17 July.61 It is
noticeable that most sources present these events in terms of Christian
theology. Raymond of Aguilers tells how the crusaders ‘rode in blood [up]
to the knees and bridles of their horses’. John and Laurita Hill identified
this description as deriving from a passage in the Book of Revelation,
which describes the wrath of God using the image of a winepress flow-
ing with blood up to the bridles of horses; Raymond’s description thus
primarily serves to underline his point that the effusion of pagan blood
in the Temple was a just punishment for the pagan blasphemies that had
taken place there.62 Guibert of Nogent and Robert the Monk state that
the massacres were carried out in retribution for Muslim crimes against
the Christians, and go on to place these events in a theological context by
citing biblical prefigurations, particularly from Isaiah and Zechariah, of a
liberated Jerusalem.63 These are merely a few strands in the rich seam of
contemporary theological interpretation of the crusade and the conquest
of Palestine as a whole.64 However, the essential point is that all these
sources give theological justifications of the massacre, rather than rational
explanations for why it took place.
61
62
Iherusalem’, in RHC Occ. 3: 300: equitabatur in sanguine usque ad genua, et
usque ad frenos equorum; Raymond of Aguilers, Liber, ed. Hill and Hill, p. 150
n. 2; Raymond of Aguilers, Historia Francorum qui ceperunt Iherusalem, trans.
Hill and Hill, p. 128 n. 22; Biblia Sacra iuxta Vulgatam versionem, ed. Robert
Weber et al., 3rd edn (Stuttgart, 1969), p. 1896: et calcatus est lacus extra civitatem
et exivit sanguis de lacu usque ad frenos equorum per stadia mille sescenta.
63
‘Historia quae dicitur Gesta Dei per Francos’, in RHC Occ. 4: 227–29, 237–38;
‘Roberti Monachi historia Iherosolimitana’, in RHC Occ. 3: 868–82.
64
Sylvia Schein, Gateway to the Heavenly City: Crusader Jerusalem and the
Catholic West (1099–1187) (Aldershot, 2005), pp. 21–33.
211
Alan V. Murray
The most thorough examination of these events has been carried out by
Benjamin Kedar, who in an exhaustive, diachronic study has analysed its
depiction in both primary and secondary narratives and documents. He
makes the crucial point that the recognition of the use of extensive biblical
imagery of blood and vengeance by the chroniclers does not mean that
carnage did not occur. He concludes that after the entry into the city on 15
July the crusaders started a slaughter of its Muslim and Jewish inhabitants,
which continued into the next day, even engulfing Muslims taking refuge
in the Temple area who had been promised protection by Tancred and
Gaston of Béarn; the Fātimid garrison which still occupied the Tower of
David surrendered and was allowed to leave on 17 July, accompanied by
some Muslim and Jewish civilians. Kedar also shows that two letters sent
by the Jewish elders of Ascalon to the Jews of Alexandria corroborate the
report by Albert of Aachen that on 17 July the crusader leaders ordered
the final annihilation of all inhabitants who still remained in the city.65
Some historians have argued that according to medieval conventions
of war, besieged populations could only secure their lives by surrendering
and could not expect mercy in the event of a town being taken by assault.66
However, Kedar demonstrates that the massacres which extended over
three days were clearly far more extensive than in other cases of capture
by storm. After their many privations, the crusaders may well have been
thirsting for revenge, and in the frenzy of the first irruption into the city
probably made little effort to distinguish between armed opponents and
unarmed civilians; the gory descriptions of the Western narratives are
eloquent testimony to an orgy of violence. Yet it is difficult to believe that
frenzied killing in hot blood could have gone on beyond the day of the
assault, given that the crusaders hoped to secure plunder, and certainly
needed to find food, and rest from their exertions.
Alone among the Western sources, Albert of Aachen provides reflec-
tions on the strategic considerations that led to the final slaughter of the
third day (17 July). He describes Tancred’s fury at hearing of the killing
of a group of Saracens on the roof of the Temple of Solomon (al-Aqsā),
65
Benjamin Z. Kedar, ‘The Jerusalem Massacre of July 1099 in the Western
Historiography of the Crusades’, Crusades 3 (2004), 15–75.
66
(Cambridge, 1994), pp. 355–56; Kaspar Elm, ‘Die Eroberung Jerusalems im
Jahre 1099. Ihre Darstellung, Beurteilung und Deutung in den Quellen zur
Geschichte des Ersten Kreuzzugs’, in Jerusalem im Hoch- und Spätmittelalter.
Konflikte und Konfliktbewältigung – Vorstellungen und Vergegenwärtigungen,
ed. Dieter Bauer, Klaus Herbers and Nikolas Jaspert (Frankfurt am Main,
2001), pp. 31–54.
212
to whom he had given his banner as a sign of his protection. By way of

reply to Tancred, Albert constructs a speech which he attributes not to a
named individual but to ‘the greater and wiser men’:
Jerusalem, city of God on high, has been recovered, as you all know, with
great difficulty and not without harm to our men, and today she has been
restored to her own sons and delivered from the hands of the king of
Egypt and the yoke of the Turks. But now we must be careful lest we lose
it through avarice or sloth or the pity we have for our enemies, sparing
prisoners and gentiles still left in the city. For if we were to be attacked in
great strength by the king of Egypt we should be suddenly overcome from
inside and outside the city, and in this way carried away into eternal exile.
And so the most important and trustworthy advice seems to us that all the
Saracens and gentiles who are held prisoner for ransoming with money, or
already redeemed, should be put to the sword without delay, so that we
shall not meet with any problem from their trickery or machinations.67
Here Albert encapsulates the strategic predicament of the now victori-
ous crusaders. Once the crusaders entered Palestine, al-Afdal, the vizier
of the Fātimid caliph (the rex Babylonie in Albert’s account) had started
to send troops to Palestine to intercept them. In the event the Fātimid
army had not completed its concentration by the time that Jerusalem fell,
but it was only a matter of time before it would attempt to retake the city
before the crusaders had the chance to organise a proper defence; it was
this army that they eventually defeated outside Ascalon on 12 August. This
strategic situation explains the actions of the crusade leadership on 17 July.
Most of the 400-strong Fātimid garrison (and possibly some civilians)
had retreated into the Tower of David, a strongpoint which dominated
the western walls, and continued to hold out under the command of the
governor, Iftikhār al-Dawlah.68 The crusaders had already experienced a
similar predicament after capturing Antioch, when the citadel remained
67
Albert of Aachen, Historia Ierosolimitana, VI.29, ed. Edgington, pp. 440–41:
Ierusalem, ciuitas Dei excelsi, ut uniuersi nostis, magna difficultate et non sine
dampno nostrorum recuperata, propriis filiis hodie restituta est, et liberata de
manu regis Babylonie et iugo Turcorum. Sed nunc cauendum est ne auaritia
aut pigricia uel misericordia habita erga inimicos hanc amittamus, captiuis
et adhuc residuis in urbe gentilibus parcentes. Nam si forte a rege Babylonie in
fortitudine graui occupati fuerimus, subito ab intus et extra expugnabimur, et sic
in perpetuum exilium transportabimur. Vnde primum et fidele consilium nobis
uidetur quatenus uniuersi Sarraceni et gentiles qui captiui tenentur pecunia
redimendi aut redempti sine dilatione in gladio corruant, ne fraude aut ingeniis
illorum nobis aliqua aduersa occurrant.
68
213
Alan V. Murray
in the hands of the Turks at a time when a relieving army was a pproaching.
Nevertheless, it was not the only potential threat.
Although large numbers of Muslims and Jews had been killed on 15–16
July, many had taken refuge on the Temple platform, or were hiding in
houses or cellars, while Albert’s account indicates that many others were
being held captive with a view to ransom. France estimates that by the
time of the battle of Ascalon the number of crusaders had been reduced
to about 10,000, a figure that gives us an approximate strength for the
crusaders after the capture of Jerusalem.69 These exhausted Westerners
were confronted with the need to mount a defence while also trying to
subdue and guard a large number of surviving inhabitants, in a city whose
topography was unfamiliar. Albert’s comment that the crusade leaders
feared being ‘overcome from inside and outside the city’ reveals the danger
that the inhabitants might be tempted to mount an active resistance at
the moment that the Fātimid army arrived outside the walls.
The actions of the crusaders after the initial massacres can be inter-
preted as an attempt to deal with these potential dangers. They did not
waste their efforts in trying to capture the Tower of David, but removed
the threat by granting the garrison free passage to Ascalon after it had
surrendered its weapons. They believed that the remaining Muslim and
Jewish inhabitants also had to be removed, but allowing them to leave
would have placed a large potential labour force at the disposal of the
advancing Fātimids. It is thus most likely that the crusade leaders decided
that the most efficient way of securing their own safety would be to anni-
hilate the remaining Muslims and Jews, even if this meant a great loss of
ransom money. Other chroniclers provide support for the systematic and
calculated manner of this final slaughter, in which the captives were split
into groups to be slaughtered, with some being spared only long enough
to dispose of the corpses.70
69
France, Victory in the East, p. 131.
70
Historia de Hierosolymitano itinere, ed. Hill and Hill, p. 142; ‘Historia quae
dicitur Gesta Dei per Francos’, in RHC Occ. 4: 228; ‘Roberti Monachi historia
Iherosolimitana’, in RHC Occ. 3: 869; ‘Historia quae dicitur Gesta Dei per
Francos’, in RHC Occ. 4: 228. For a more detailed analysis of the strategic
situation and its consequences, see Alan V. Murray, ‘The Demographics of
Urban Space in Crusade-Period Jerusalem, 1099–1187’, in Urban Space in the
Middle Ages and the Early Modern Period, ed. Albrecht Classen (Berlin, 2009),
pp. 205–24, and Murray, ‘La construcción de Jerusalén como capital cristiana:
Topografía y población de la Cuidad Santa bajo el dominio franco en el siglo
XII’, in Construir la ciudad en la Edad Media: Encuentros internacionales
214
Albert’s analysis of the strategic situation demonstrates his unique value

as a source. The frenzied slaughter on the day of the crusader conquest
and the organised large-scale executions that followed it were undeniable
facts. The majority of those writers who produced dedicated accounts of
the crusade chose to interpret and justify this massacre in terms of divinely
approved purification of the holy sites and revenge on their Muslim
occupiers. It is remarkable that unlike the others, both participants and
non-participants alike, Albert, writing in the distant Rhineland, managed
to research the history of the First Crusade so thoroughly that he was able
to produce a rational if chilling explanation that enables us to glimpse
the true reasons that impelled the leaders of the First Crusade to carry
out such a slaughter, so ‘that the streets of the whole city of Jerusalem
are reported to have been so strewn and covered with the dead bodies of
men and women and the mangled limbs of infants, that not only in the
streets, houses, and palaces, but even in places of desert solitude numbers
of slain were to be found’.71
del Medievo 2009, Nájera, ed. Beatriz Arizaga Bolumburu and Jesús Angel
Solórzano Telechea (Logroño, 2010), pp. 91–110.
71
Albert of Aachen, Historia Ierosolimitana, VI.30, ed. Edgington, pp. 442–43:
Vnde platea totius urbis Ierusalem corporibus exstinctis uirorum ac mulierum
ac laceris menbris infantium adeo strate et operte fuisse referuntur, ut non solum
in uicis, soliis et palaciis, sed etiam in locis deserte solitudinis copia occisorum
reperiretur.
215
III
The Impact of the Crusade

The Muslim Response to the First Crusade
Michael Brett
School of Oriental and African Studies, University of London
The Near East on the Eve of the Crusades1

For the conference in Huesca at which this paper was first presented,
the proposed title was ‘Islam defeated’. That begged too many questions,
even when rendered problematic with a question mark: ‘Islam defeated?’
Controversy over the role of the faith in the Muslim response to the First
Crusade has nevertheless characterised the subject, not only at the time
but in modern scholarship. For this revision of the original, therefore,
the attempt has been made to distinguish between the reaction of an
Islamic society to invasion and conquest and its response in religious
terms, beginning with the situation in the Near East on the eve of the
crusades. By the end of the fifth/eleventh century, the realm of Islam as
a religion, a way of life and a civilisation formed what Hugh Kennedy
has usefully called a Muslim commonwealth of successors to the original
Arab empire, which had finally ceased to exist in the middle of the tenth
century CE.2 Internally complicated by a whole series of state- and empire-
builders, externally its realm had remained largely intact, despite the loss
of Cilicia, Antioch and the thughūr or marches of Syria to Byzantium in
the tenth century. In the second half of the eleventh century, however, it
was thrown onto the defensive in the western Mediterranean by Christian
aggression in Spain and Sicily. In 1085 Toledo had been lost to Castile,
and by 1095 Sicily had been conquered by the Normans. Disturbingly,
the papacy under Gregory VII and his successor Urban II had begun to
1
The title of my contribution to the previous conference at Madrid in 1995, in
La Primera Cruzada, novecientos años después: El concilio de Clermont y los
orígenes del movimiento cruzado, ed. Luis García-Guijarro Ramos (Castellón,
1997), pp. 119–36.
2
Hugh Kennedy, The Prophet and the Age of the Caliphates: The Islamic Near
East from the Sixth to the Eleventh Century (London, 1986), pp. 200–11.
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. 1–.
Michael Brett
think in terms of a holy war upon Islam to east and west, giving its bless-
ing to attacks across the northern, Pyrenean frontier of al-Andalus. In the
eastern Mediterranean the arrival of the First Crusade nevertheless took
the Muslim world by surprise. In hindsight we can see it as an outcome,
not simply of developments in western Europe, but of a chain of events at
the centre of the Islamic world that began in the tenth century with the
rise of the Fātimids to challenge the ‘Abbāsid caliphs of Baghdad. Ruled
from Cairo, their empire reached no further than Egypt and Syria; but
the challenge of its Shī‘ite propaganda to the Sunnī world of which the
‘Abbāsids were the nominal rulers provoked a reaction which from the end
of the tenth century onwards provided the Turkish lords of eastern Iran
and central Asia with an ideology of conquest and empire in the name of
the ‘Abbāsid caliph. Where the Ghaznāvids had posed as champions of
Baghdad and threatened to overthrow the Fātimids, the Saljūq Turks who
took their place in the 1040s carried their purpose into effect, going on
to occupy Baghdad in the 1050s and Fātimid Syria in the 1060s and 70s.3
Fātimid Egypt itself, which from 1066 to 1074 was crippled by a mutinous
soldiery, was nevertheless spared, partly because in 1071 the Saljūq sultan
Alp Arslan turned aside into Asia Minor to defeat the Byzantines at
Manzikert, and was himself murdered the following year. It was the defeat
of the Byzantines, on the other hand, which eventually led to the appeal of
Constantinople to the papacy for aid to recover Asia Minor from the Turks
who had overrun it. By that time, however, not only had Egypt made a full
recovery at the hands of the Armenian military vizier Badr al-Jamālī and
his son al-Afdal, but Syria had relapsed into its component parts. While the
Fātimids at Acre held the cities along the coast of Palestine and Lebanon
as far as Beirut, and the inland plain of Ramla in southern Palestine, the
departure from Syria of the Saljūq prince Tutush, followed by his death
in the struggle to succeed his brother the sultan Malik Shah in 1092–93,
had left Antioch to its governor Yaghisiyan, Aleppo and Damascus to his
young sons Ridwān and Duqāq, and Jerusalem to the Turcoman dynasty
of Artūq. With the evaporation of the Saljūq overlordship, and al-Afdal
in Egypt safely in power after installing his protégé on the Fātimid throne
in 1094–96, there was a clear opportunity to bring the country back into
the Fātimid fold. As the new champion of the Fātimid caliphate and its
imperial destiny, al-Afdal took the first step in 1097 with the recovery of
Tyre, independent since the loss of Fātimid Syria. At Aleppo, he was briefly
recognised by the Saljūq Ridwān, an earnest of the political calculations
3
See Michael Brett, The Rise of the Fatimids: The World of the Mediterranean and
the Middle East in the Tenth Century CE (Leiden, 2001), pp. 429–30, and The
New Cambridge History of Islam, 6 vols (Cambridge, 2010–12), 4: 691–93.
220
which were to determine Syrian relations with the Latin states. Most
important in this context, however, was the capture of Jerusalem from
the Artūqids in 1098, the year in which the armies of the First Crusade
took Antioch, and turned south.
Up to this point, both Saljūq and Fātimids had reacted to an invasion
of the realm of Islam rather than responded to a challenge to its faith.
With the armies of Qilij Arslān, the Saljūq prince in Anatolia, swept aside
by the crusaders, Karbugha, the Saljūq atabeg of Mosul, had come to the
relief of Antioch, only to be routed after the fall of the city. Al-Afdal,
on the other hand, seeing a horde dispatched by the Byzantine emperor
to reconquer his lost lands from the Saljūq enemy, sought to turn its
arrival to his own advantage, all the more perhaps as an Armenian from
a Christian background, at the head of an Armenian army with strong
Anatolian connections.4 Thus a Fātimid embassy was sent to the crusaders
outside Antioch, which returned to Cairo with presents and a Frankish
delegation. Its purpose appears to have been an understanding, if not an
alliance, by which Aleppo, say, the capital of northern Syria, would be left
to the crusaders, and Damascus to the Egyptians as the former capital of
their Syrian province. From his reaction to the taking of Jerusalem, that
certainly appears to have been al-Afdal’s understanding, though not of
course that of the crusaders. Their purpose went unrecognised; the fall of
Jerusalem was all the more unexpected given the exceptional determina-
tion which drove the victors on; and the foundation of the kingdom of
Jerusalem all the more unforeseen. How long it then took for that pur-
pose to be recognised for what it was is not clear. Despite its association
with the Prophet, the hill city of Jerusalem had remained a backwater
in the Fātimid empire, only brought into prominence by the Turcoman
Artūqids; it was for the crusades themselves to turn its recapture for Islam
into a sacred mission, and that not to any great extent before the time of
Nūr al-Dīn and Saladin over fifty years later. There was indeed a precedent
for this, the First Crusade, in the campaigns of the Byzantine emperor
John Tzimisces, which in 975 had reached as far south as Caesarea with
the apparent objective of recovering Jerusalem for the Christian empire.
Only aborted by his death, these campaigns had moreover come at a time
when the Fātimids, newly installed in Egypt, were making a public stand
as champions of Islam against the infidel.5 But since the beginning of the
eleventh century, the two empires had been on treaty terms which at times
4
See Michael Brett, ‘Badr al-Ğamālī and the Fatimid Renascence’, in Egypt and
Syria in the Fatimid, Ayyubid and Mamluk Eras, IV, ed. Urbain Vermeulen and
Jo van Steenbergen (Leuven, 2005), pp. 61–78.
5
Brett, Rise of the Fatimids, pp. 331, 346.
221
Michael Brett
had amounted to friendship and co-operation; and al-Afdal’s immedi-

ate reaction to the fall of Jerusalem is reported in the Arabic tradition
to have been an embassy rebuking the crusaders for taking the city after
what he considered to be their agreement. The sijill-s or formal letters to
the crusaders at Antioch and again at Jerusalem have not, unfortunately,
survived to reveal the vizier’s appreciation of their intentions. Written to
a prescribed formula, they would certainly have been addressed de haut
en bas in the name of the Imām Caliph, Commander of the Faithful, the
Lord of Islam, to Christians considered to be in some form of treaty rela-
tionship, actual or presumed, with himself and the religion he represented.
But how this evident breach of contract was interpreted, as a relatively
minor or as a major affront, not only to the majesty of the sovereign but
to the faith itself, cannot be known. What al-Afdal did was to bring his
army up to Ascalon with the apparent intention of recovering the city. But
like the army of Karbugha at Antioch, his own was routed, in this case by
a surprise attack. Al-Afdal fled back to Cairo, retaining only Ascalon as
the Fātimid foothold in southern Palestine, and obliged to contemplate
the installation of a quite different foe from the Saljūqs in the former
dominion of the dynasty he served.
The Response of the Muslim World

The wider response of the Muslim world to the creation of the kingdom of
Jerusalem and the Latin states of Edessa, Antioch and Tripoli has attracted
a literature which considers the impact of the crusades down to the time
of Saladin and beyond to the fall of Acre in 1291. Its most substantial
treatment is by Carole Hillenbrand in The Crusades: Islamic Perspectives;6
in the Madrid conference preceding that at Huesca where this paper was
first given, it was elegantly summarised by Robert Irwin.7 His piece drew
attention to the literature of holy war as exemplified in a treatise, ‘The Sea
of Precious Virtues’, which c.1160 explained the intrusion of the Franks
as a punishment for the moral decadence of the Islamic world. Decadent
or not, what the anonymous author denounced as a shocking failure to
unite against the infidel, to the extent of allying with him against fellow
Muslims, was indeed characteristic of the response of the Saljūq world to
the First Crusade, with the possible exception of the Danishmendids in
north-eastern Anatolia. The crusaders themselves proved only too willing
6
Carole Hillenbrand, The Crusades: Islamic Perspectives (Edinburgh, 1999).
7
Robert Irwin, ‘The Impact of the Early Crusades on the Muslim World’, in La
Primera Cruzada, pp. 137–51.
222
to enter into the politics of that world, most dramatically in 1115, when
an army dispatched from Mosul by the Great Saljūq sultan Muhammad
was defeated by the forces of Aleppo and Damascus in alliance with the
crusader Roger of Antioch. Hillenbrand suggests that what the Turkish
princes of these cities were most afraid of was an attempt by the sultan to
reincorporate them into his empire under the pretext of holy war upon
the infidel. Given such behaviour, the cause of Islam had been left to the
poets and jurists of Syria. Of the jurists, the qādī of Damascus, al-Harawī,
had gone to Baghdad in 1099 to appeal in vain to the sultan Barkyārūq
for his leadership. In 1111 Ibn al-Khashshab, Shī‘ite qādī of Aleppo, led a
second delegation which made such a fuss that it persuaded the Sultan
Muhammad to dispatch Mawdūd the governor of Mosul to Syria; there,
however, he was not only shut out of Aleppo, but murdered in Damascus.
Meanwhile in 1105–6 the Damascene jurist al-Sulamī composed a treatise
on the conduct of jihād or holy war, specifically reproaching the princes of
Islam for their failure to unite against the infidel, and calling upon them
to do so. Selectively quoted by Hillenbrand, the poets in their turn had
reached for the familiar tropes of death and destruction in hyperbolic
lament for the victories of the polytheists over Islam. What both jurists
and poets reveal is a popular indignation at odds with the politics of
their rulers, which was rooted in the character of Damascus and Aleppo
as self-governing city-states under the military protection of more or
less unpopular Turkish warriors. Damascus had resisted the Fātimids for
twenty years after their arrival in Egypt, and had readily broken away from
them in the crisis c.1070; the failure of the Fātimids to take Aleppo had
halted any ambition to invade Iraq and advance on Baghdad.8 Twenty
years after the fall of Jerusalem, Aleppo was instrumental in mounting
the campaigns which culminated in the counter-crusade of Nūr al-Dīn
and Saladin. In the aftermath of the foundation of the Latin kingdom,
however, it was the coastal cities of Arsuf, Caesarea, Acre, Tyre, Sidon,
Beirut and Tripoli which were most immediately threatened, and whose
response came down to a choice between capitulation and resistance.
Sustained by the Fātimid fleet, their resistance delayed their fall over a
period of more than twenty years, from that of Arsuf and Caesarea in
1101 to Acre in 1104, Tripoli in 1109, Beirut and Sidon in 1110, and finally
Tyre in 1124. While there is no doubt about the determination of their
8
The discussion of the city and the city-state in Islam has a long history, but
see Michael Brett, ‘The City-state in Mediaeval Ifriqiya’, in Brett, Ibn Khaldun
and the Mediaeval Maghrib (Aldershot, 1999), Essay XIV. For Damascus and
Aleppo under the Fātimids, see Brett, Rise of the Fatimids, pp. 309–14, 331, 349
(Damascus), 310–11 (Aleppo).
223
Michael Brett
resistance, however, the commitment of the Fātimids to their survival

has been variously interpreted. What has not been generally proposed is
that it was part of a commitment to jihād, to holy war, the first specifi-
cally Islamic response of the rulers of the Muslim commonwealth to the
Christian challenge to their faith.
This is surprising given the nature of the Fātimid caliphate, for which,
since we are dealing with a conflict between Islam and western Christendom
in which the papacy played a crucial role, a comparison between Rome
and Cairo is not inappropriate. A hundred years after the First Crusade,
the biblical figure of Melchizedek became central to Pope Innocent III’s
conception of himself as the vicar of Christ. Seen as both king and priest,
Melchizedek was taken to be not only a prefiguration of Christ, but of
the pope who exercised the spiritual and temporal power and author-
ity of the Messiah. At the same time as the papacy of Pope Innocent III,
Melchizedek makes an appearance in the Persian literature of the Ismā‘īlīs,
the successors of the Fātimids, shortly after the end of the Fātimid caliphate
in 1171. The coincidence is entirely accidental, although the source of the
reference was evidently Christian, and the borrowing may go back to the
time of the Fātimids’ greatest philosopher/theologian al-Kirmānī in the
heyday of the caliphate in the first half of the eleventh century. Despite
the changes made in this Ismā‘īlī literature to accommodate his figure in
the sacred history of Islam, Melchizedek nevertheless fulfils the same role
as a prefiguration of the imām caliph, the Fātimid monarch who claimed
for himself religious authority and political power over the entire Muslim
world on behalf of God and in succession to Muhammad. The difference
in principle was that the papacy, while reserving to itself the overall direc-
tion and supervision of the government of Christendom, had entrusted the
execution to its princes, beginning with the emperor. In practice, however, a
similar delegation had taken place in Islam, at the expense of direction and
supervision. The ‘Abbāsid caliph, whose claims by this time were restricted
to the government of the community rather than the authority for its faith,
had entrusted the government of the East and the West to the Saljūq sultan
in the middle of the eleventh century, while since 1074 the Fatimid caliph
had been represented by a plenipotentiary vizier. It was that vizier, with
the grandiloquent title of al-Sayyid al-Ajall al-Afdal, Sayf al-Imām Jalāl
al-Islām Sharaf al-A’nām Nāsir al-Dīn Khalīl Amīr al-Mu’minīn, sc. The
Mighty and Gracious Lord, Sword of the Imām, Glory of Islam, Honour of
Mankind, Protector of Religion, Friend of the Commander of the Faithful,
to whom fell the caliphal duty of opposition to the crusaders as intruders
into the lands of Islam.9
9
For this comparison, see Brett, Rise of the Fatimids, pp. 417–34.
224
That duty was very specific. In his capacity as imām and caliph, ‘king
and priest’ as it were ‘after the order of Melchizedek’, the Fātimid monarch
ruled on behalf of God over a miscellaneous collection of Muslim and non-
Muslim subjects, fī dhimmatihi wa fī nusratihi, ‘under his protection and
with his support’. The dhimma in question was not the narrow protection
offered to non-Muslims under Muslim rule, but the far wider dhimma or
protection of God which lies behind the original Constitution of Medina,
and which had been proclaimed at the entry of the Fātimids into Egypt in
969 as the essence of their rule. In return for that protection, the commu-
nity gave its bay‘a, its oath of acceptance and obedience.10 But at the heart of
this compact lay a much simpler undertaking that dated back to the Arab
conquests, that the caliph, whether Umayyad, ‘Abbāsid or Fātimid, would
lead the faithful in war upon the infidel, justly and equitably, sharing out
the spoils. That leadership and its concomitant, the defence of the lands of
Islam against attack, was one of the very few specific obligations laid upon
him in his capacity as their commander.11 His performance of that duty
correspondingly served as a test of his ability, of his fitness for office. That
also was true of al-Afdal, the vizier who ruled in the name of the caliph,
and like the notorious al-Mansūr in Spain a hundred years earlier, required
the vindication that only the prosecution of the holy war could provide.
The campaigns that al-Afdal set on foot against the crusaders in the early
years of the century were clearly aimed at his glorification in the holy war,
as the remarkable text culled by Geoffrey Khan from the Geniza mate-
rial amply demonstrates. It is a fragment of a report to al-Afdal from the
commander in the field of the capture of a Frankish fortress, conceivably
the tower at Ramla which was central to the battle of 1102, which hails the
vizier’s ‘impetuousness for the defence of Islam and the Muslims […] the
territory and the religion. His steadfastness and his sacrifice of money and
treasures, arms and equipment, men and children […]’.12
The Fātimid Campaigns against the Kingdom of Jerusalem

The evidence, if not for that impetuousness and steadfastness, at least for the
vizier’s commitment, is provided by the campaigns themselves, despite
the differences in the sources and in their interpretation. After the rout
10
See Brett, Rise of the Fatimids, pp. 300–3.
11
‘Abd al-Wahīd al-Marrākushī, History of the Almohades, ed. Reinhard P. A.
Dozy, 2nd edn (Leiden, 1881), p. 239.
12
Geoffrey Khan, Arabic Legal and Administrative Documents in the Cambridge
Genizah Collections (Cambridge, 1993), no. 111, p. 428.
225
Michael Brett
at Ascalon in 1099, three of these were undertaken in 1101, 1102, and 1105,
giving rise to three major battles around Ramla, the old Fātimid capital
of Palestine. They were moreover followed by a period of twenty years of
intermittent frontier excursions, and more importantly as far as the crusad-
ers were concerned, the dispatch of the Egyptian fleet to the assistance of
Sidon, Tyre, Beirut and Tripoli when these came under attack. None of
this succeeded in halting, let alone reversing, the consolidation of the Latin
kingdom and the county of Tripoli, although at Ascalon the approach to
Egypt was forbidden to the Franks until the Fātimid caliphate entered its
final phase in the middle of the century. In view of this persistent failure, the
question is why the Egyptian effort was sustained over so long a period. The
original aspiration to universal empire was now clearly out of the question,
as was any attempt to recreate the Fātimid empire in greater Syria. A more
modest attempt to restore and maintain the Fātimid position in southern
Palestine and along the coast in the face of crusader expansion was surely a
factor, especially since al-Afdal remained bent on upholding what remained
of the Fatimid empire, witness his concern at this time with enforcing the
terms of the Baqt, the treaty which established the tributary status of the
Christian kingdom of Muqurra in the Sudan.13 Modern opinion has varied.
Insofar as he was concerned with Egyptian motive, Runciman agreed that
al-Afdal was attempting to reconquer Palestine, while following Grousset
in supposing that ‘the resources of Egypt were enormous’.14 For Grousset,
that meant an equally large army, but one that was similarly ill-trained, its
numbers no match for the Frankish knights. Hamblin, on the other hand,
has argued that the Fatimid army was not only a well-constructed force of
different cavalry and infantry units, but for that reason relatively small.15 Lev
broadly agrees, but despite Hamblin’s instances of hard fighting and near
victory, concludes from the long tale of failure to make headway against
the Franks that ‘their army […] lacked any reasonable fighting capability’.16
Turning to the navy, Lev sees the same economic resources and organisa-
tional skills devoted to the fleet as to the army, but finds that once again
13
Heinz Halm, ‘Der nubische baqt’, in Egypt and Syria in the Fatimid, Ayyubid
and Mamluk Eras, II , ed. Urbain Vermeulen and Daniel De Smet (Leuven,
1998), pp. 63–103.
14
Steven Runciman, A History of the Crusades, 3 vols (Cambridge, 1951–54), 2:
12–13, 76. See René Grousset, Histoire des Croisades et du royaume franc de
Jérusalem, 3 vols (Paris, 1935), 1: 225, 229–30.
15
William J. Hamblin, ‘The Fatimid Army during the Early Crusades’
(Ph.D. thesis, University of Michigan, 1985), published in facsimile (Ann
Arbor, 1987), pp. 294–301.
16
Yaacov Lev, State and Society in Fatimid Egypt (Leiden, 1991), pp. 93–103.
226
it was relatively small; crucially, moreover, it was sent out in response to

crusader attacks along the Lebanese coast, which, given the time it took to
arrive, meant it was likely to be ineffective. This argument is essentially that
of Hamblin: given that the Fātimids were on the defensive, the distance of
Palestine-Syria from Egypt made each Fātimid response to each crusader
attack too little, or more importantly, too late.17 Meanwhile Köhler, quoted
by Hillenbrand, has suggested that such defensiveness was deliberate; far
from wishing to reconquer their lost lands, the Fātimids regarded the king-
dom of Jerusalem as a buffer between themselves and the Saljūqs of Syria.18
Hillenbrand herself simply refuses to be drawn. But of these explanations
of Fātimid motives behind Fātimid failures, only Lev’s account of the
Fātimid navy supports the conclusion. Hamblin’s argument is apologetic,
an attempt to explain away the embarrassing failures in battle of a fine army.
Köhler’s speculation has to accommodate the awkward fact that al-Afdal
sought the alliance of Damascus in the third battle of Ramla. There is in
the end something to be said for Lev’s conclusion that the Fātimid army,
however constituted and however led, was found wanting; but equally it is
difficult to think that the effort which went into these expeditions by land
and sea over so long a period had no more positive purpose in the eyes of
al-Afdal and his audience at home and abroad than the rearguard defence
of Egypt and the Palestinian-Lebanese littoral. The Fātimids were at war
in a way that the Saljūqs of Syria were not.
The course of that war may be broadly clear, but for the battles of Ramla,
the sources vary between an Egyptian, a Syrian and a Latin tradition.19
The Egyptian tradition, as might be expected, is the more favourable; Ibn
al-Qalānisī, the principal Syrian source, is more in accord with the Latin
sources, which are chronologically more accurate in dating the second
battle to 1102 rather than 1103. The Egyptian objective in all cases appears
to have been Jaffa, the port whose capture by the crusaders in 1099 had
been vital to the taking of Jerusalem, and whose recapture would clearly
have dealt them a major blow. In the first battle, the Egyptian commander
was killed, at some cost to the Franks; but whereas the Latin sources have
the Egyptians driven back to Ascalon, the Arabic ones have the crusad-
ers driven back to Jaffa. What is important is not so much the accuracy
17
Lev, State and Society, pp. 104–21.
18
Hillenbrand, The Crusades, pp. 77–78. See Michael A. Köhler, Alliances and
Treaties between Frankish and Muslim Rulers in the Middle East, trans. P. M.
Holt, rev. Konrad Hirschler (Leiden, 2013), pp. 88–89.
19
Both the sources and the battles are discussed in Michael Brett, ‘The Battles of
Ramla (1099–1105)’, in Egypt and Syria in the Fatimid, Ayyubid and Mamluk
Eras, I, ed. Urbain Vermeulen and Daniel De Smet (Leuven, 1995), pp. 17–37.
227
Michael Brett
or otherwise of their accounts, but the language of holy war employed

by the Syrian Ibn al-Qalānisī.20 In one sense this is retrospective: Ibn
al-Qalānisī was writing in Damascus about 1160, at the time when the
Zangid Nūr al-Dīn had become the champion of jihād, and moreover at
a time when the Fātimid vizier Tala’i‘ ibn Ruzzīk was soliciting his alli-
ance in the holy war. On the other hand it betrays a sense of precedent,
of historical memory on the part of the Egyptians and the Damascans
which was capable of recalling the battle as a victory over the infidel in the
way in which it may have been celebrated at the time. This is all the more
possible in the light of the second battle in 1102, a more substantial affair
in which Ramla was taken along with many prisoners, although Baldwin
narrowly escaped to Jaffa, and the arrival of a Frankish fleet forced the
Egyptians under the command of al-Afdal’s son to retreat to Ascalon.
For Ibn al-Qalānisī it is once again a question of God’s having given the
victory; but more importantly, it is surely this battle, which turned on
the capture of a tower at Ramla, which is celebrated in the fragment of
a report to al-Afdal from the commander in the field of the capture of a
Frankish fortress, hailing the vizier’s
impetuousness for the defence of Islam and the Muslims […] the territory
and the religion. His steadfastness and his sacrifice of money and treasures,
arms and equipment, men and children […] in the strengthening of His
religion. The humiliation of His enemy is His heart’s desire.21
The fragment not only attests to al-Afdal’s engagement in holy war with
the crusaders, as befitted the Sword of the Imām. It is equally an example
of the way in which this engagement was proclaimed and recorded, and a
clue to the diplomatic means by which it was developed in the aftermath
of the battle into a fresh approach with important implications for the
caliphate and the holy war. From the time of their coming to power in
Ifrīqiya at the beginning of the tenth century, the Fātimid caliphs had
been great letter-writers; from the time of their arrival in Egypt, their more
personal correspondence had been overlaid by their public correspond-
ence for administrative, diplomatic and propaganda purposes, conducted
through their chancery according to fixed and elaborate scribal formulae.
The instrument was the sijill, from the Latin sigillum, although the place
of the seal had been taken by the ‘alāma, the formulaic signature of the
caliph, and by this time in the life of the dynasty the signature of his vizier.
The composition of what was not merely a letter but an authoritative proc-
lamation of majesty, has been fully described by John Wansbrough in his
20
Brett, ‘Battles of Ramla’, pp. 33–34.
21
See Khan, Legal and Administrative Documents, p. 428.
228
Lingua Franca in the Mediterranean.22 Few of the originals have survived,

but their formula is well illustrated for the present purpose by a copy of such
a document from the year 1053–54, the proclamation of a victory which
typically recapitulated the history of the event before the final instruction to
publish the triumph from the pulpit.23 What is especially important is that
this particular sijill then served as a source for the Egyptian tradition of the
episode. In the present instance, all the indications point to a similar con-
clusion, that the adulatory letter addressed to al-Afdal from the battlefield
formed the basis for a similar proclamation of victory by al-Afdal himself;
and moreover, that the story it told entered into the correspondence with
Damascus that led up to the third battle in 1105. In 1103, in the year, that
is, after the second battle, a letter was sent by al-Afdal to Damascus, which
according to Ibn al-Qalānisī requested ‘the help of its army in the holy war
for Muslim land and Muslim folk’.24 In the manner of the Fātimid sijill, the
request itself would have been preceded by an appropriate introduction,
a rehearsal of the glorious efforts of the Egyptians in the previous battles,
to encourage the Syrians to join in a final successful campaign. Its content
may in fact be inferred from the date of 1103 which is given for the second
battle in the majority of the Arabic sources: if the letter, in Fātimid fashion,
had served, as is probable, as a source for the Egyptian tradition, then its
date, inscribed in the letter itself, will have furnished the chroniclers in
that tradition, and those acquainted with it, with their date for the battle
instead of what is certainly the true date of 1102 given in the Latin sources
and by Ibn al-Qalānisī. The invitation itself was to join the Egyptians in
their campaign of that year; but in spite of the assured magnificence of this
product of the Fātimid chancery, written in all probability by its chief scribe,
Ibn al-Sayrafī, it was nevertheless declined. Accepted in principle, says Ibn
al-Qalānisī, the request was refused in practice by the Saljūq prince Duqāq
for various good reasons. But it was not the end of the matter.
The campaign in question was a fiasco. In the spring of 1103, the Fātimid
fleet relieved a siege of Acre, but the subsequent expedition by land and
sea to recapture Jaffa came to nothing with the failure of the army to
advance beyond Ascalon; and in 1104 Acre was finally taken after the
Fātimid fleet failed to arrive. In 1105 al-Afdal nevertheless renewed his
invitation to Damascus, now ruled after the death of Duqāq by his atabeg
22
John Wansbrough, Lingua Franca in the Mediterranean (Richmond, 1996),
pp. 76–145.
23
See Michael Brett, ‘The Ifriqīyan Sijill of al-Mustansir, 445/1053–4’, in Egypt
and Syria in the Fatimid, Ayyubid and Mamluk Eras, VI, ed. Urbain Vermeulen
and Kristof D’Hulster (Leuven, 2010), pp. 9–16.
24
Brett, ‘Battles of Ramla’, pp. 28–29.
229
Michael Brett
Tughtakīn. In the complicated politics of Syria, Tughtakīn had problems

of his own with the succession to Duqāq, not least since Duqāq’s brother
Irtash was negotiating with Jerusalem; and it was presumably this threat
that prompted him to send a squadron to join the Egyptians under the
command of al-Afdal’s son in the third and final battle of Ramla. This
time it was evidently a Frankish victory. Though glossed in the Arabic
sources as something of a draw, with losses on both sides about equal, these
recognise its finality with the rueful reflection that the wheel of fortune
had spun for the Franks against the Muslims. What is important as far as
the Muslim response to the First Crusade is concerned, however, has less
to do with any failure on the part of the Fātimids to dislodge the infidel
from Jerusalem, let alone drive him into the sea, than with the implica-
tions of their commitment to holy war in defence of the faith. Down to
the second battle of Ramla, this can be construed as the straightforward
fulfilment of the caliphal duty to wage such a war, undertaken by the
vizier for his own glorification as the champion of the dynasty. From
1103 onwards, however, his attempt to win over his erstwhile enemies,
the Saljūqs of Syria, as allies in this holy war was a revolutionary change,
one which set aside the exclusive demand of the Fātimid imamate and
caliphate for recognition as the sole source of authority and power in
the Muslim world, in favour of a coalition of all Muslims, irrespective of
their religious and political allegiance, under its auspices in defence of
the land of Islam. Ineffective as it may have been in driving the crusaders
from Jerusalem, or preventing the fall of the coastal cities, it was the first
attempt to orchestrate a united Muslim resistance to the crusades.
As an initiative it came to nothing; but as a formula for the prosecu-
tion of the holy war by a dedicated leader it came into its own with Nūr
al-Dīn at Damascus some fifty years later. It may indeed have served as a
precedent. Not only do we find Ibn al-Qalānisī in the 1150s remember-
ing the battles of Ramla, and al-Afdal’s approach to Damascus, as early
episodes in the holy war of which Nūr al-Dīn was now the champion. At
the same time the Fātimids renewed their call for an alliance in remark-
ably similar terms.25 In 1150 the Fātimid vizier Ibn al-Sallār wrote to Nūr
al-Dīn, then at Aleppo, requesting such an alliance, to no avail. He did
so nevertheless through the good offices of Usāma ibn Munqidh, that
Arab-Syrian warrior and gentleman whose memoirs add a human touch
to the literature of the crusades; and it was to Usāma that Ibn al-Sallār’s
successor, the vizier Tala’i‘ ibn Ruzzīk, turned to plead on his behalf in
25
Michael Brett, ‘The Fatimids and the Counter-Crusade, 1099–1171’, in Egypt
and Syria in the Fatimid, Ayyubid and Mamluk Eras, V, ed. Urbain Vermeulen
and Kristof D’Hulster (Leuven, 2007), pp. 9–16.
230
support of his own embassy and letter to Nūr al-Dīn in 1158. Neither of
these letters is extant, but their content may be gauged from the vizier’s
poetic exchanges with Usāma, in which in the manner of a sijill he glori-
fies Egyptian victories before exhorting Nūr al-Dīn to rise up against the
enemy in the assurance of Egyptian support.26 Ibn Ruzzīk’s overriding
concern is declared to be the holy war upon unbelief, which is prescribed
for all Muslims in word and deed; what is different is that in the proposed
alliance the senior partner is no longer the Fātimid caliph and his vizier
but Nūr al-Dīn, the man of destiny who is urged to go to Jerusalem and
claim his reward from God. It is ironic that when the alliance eventually
came about in the 1160s, it should have ended in the suppression of the
Fātimid caliphate by Saladin; but this should not obscure the way in
which the Fātimids had, in virtue of their caliphate, responded to the
First Crusade with the proclamation and practice of a holy war which
anticipated and may have contributed to the subsequent development of
the counter-crusade under Nūr al-Dīn and Saladin. The role of Fātimid
Egypt in the Muslim response to the First Crusade, which is generally
considered to be defensive as well as ineffective, is in need of reappraisal.
Reaching a Modus Vivendi

Although the third battle of Ramla may have been the last in the series, its
status as a closure in the course of the holy war only became apparent in
retrospect. Fātimid Egypt, apart from a brief interlude in 1130–31, remained
at war with the Christian kingdom of Jerusalem down to the final crisis of
the 1160s. Coveted by the crusaders until its final capture in 1153, Ascalon
served the caliphate as a ribāt or frontier fortress for attack and defence
against the infidel. The frequent sorties from the city were not always simple
raids. In 1113 and 1115, al-Afdal was in a position to launch more substantial
expeditions, which provoked Baldwin I to invade Egypt itself in 1118, only to
fall ill and die. His death, moreover, prompted a reactivation of the Fātimid
alliance with Damascus, and a joint invasion of the Latin kingdom which
in the event came to nothing. When after the assassination of al-Afdal in
1121, the Fātimid caliph al-Āmir regained control of the state in conjunc-
tion with the new vizier al-Batā’ihī, the profile was raised again. In what
looks like a return to splendid isolation, Tughtakīn’s governor at Tyre, in
place since the atabeg had relieved the siege of the city in 1111, was ousted
in 1122 by a Fātimid naval expedition and replaced by a Fātimid nominee.
In the following year, al-Āmir showed his hand with a major campaign by
26
Brett, ‘The Fatimids and the Counter-Crusade’, pp. 9–16.
231
Michael Brett
land and sea, prepared with all the administrative resources of the state
and dispatched with due ceremony. The objective once again was Jaffa; but
the grand army, furnished with muezzins and Qur’an readers to sustain its
morale, was soundly defeated and its baggage train plundered, while the
fleet lifted the blockade of Jaffa only to be set upon and severely damaged by
the Venetians.27 When therefore Tyre was once again besieged in 1124, the
attempt to coordinate its relief by the Egyptians and Damascenes came to
nothing. What was conspicuous then was the determination of the defence,
which had saved the city in 1111 and now, after five months of siege, secured
a safe-conduct for the citizens, either to go or stay. Many did indeed go,
marking the completion of the crusaders’ conquest of their territories in
Palestine-Syria, and an end to the Fātimid response that began with the
battles of Ramla. But what their long-drawn-out resistance demonstrates is
that below the reactions of the rulers of Islam to this invasion of its territory
were those of the Muslim populace. In this case, that reaction depended
upon the character of the Muslim city, as represented by the cities of the
Palestinian-Syrian littoral from Ascalon as far as Laodicea.
In the language of the Egyptian sources as well as Ibn al-Qalānisī, Acre,
followed by Tyre and Tripoli, is described as a thaghr, pl. thughūr, a frontier
against the infidel whose inhabitants were committed to the holy war.28 As
a description, it corresponded to the expectation of g overnment and p eople
that the community would participate in its own defence. Thus on the
occasion of Baldwin’s invasion of Egypt in 1118, when Farama was sacked,
we are told that the authorities mustered the people and the muqta’ūn or
military fiefholders of the district, together with the Bedouin, to encounter
the Crusaders in advance of the main Egyptian army. The term markaziyya,
‘locals’, ‘people of the district’, is obscure, but at the battle of Ascalon in 1099,
al-Afdal’s army was swollen by volunteers and the folk of the town with
their own militia, the ahdāth.29 In 1111 its citizens rose up against a traitorous
governor who had gone over to Jerusalem and installed himself in the citadel
with a Frankish garrison, murdering him and massacring them. At the great
siege of Tyre in 1111–12, graphically described by Ibn al-Qalānisī, the citizens
and their militia were enthusiastically involved in the defence of the city
along with the ajnād or garrison troops, assisted by volunteers from the
surrounding countryside and Turkish troops sent from Damascus. All these
episodes are comparable to one that I have discussed under the heading of
‘The armies of Ifriqiya, 1052–1160’, namely the successful resistance of the
27
Lev, ‘The Fatimid Army and Navy’, in State and Society in Fatimid Egypt,
pp. 102–3.
28
Ibn Muyassar and Ibn al-Qalānisī in Brett, ‘Battles of Ramla’, pp. 32, 33.
29
Ibn al-Qalānisī in Brett, ‘Battles of Ramla’, p. 32.
232
people of Mahdia in modern Tunisia to the attack on the city by Roger II

of Sicily in 1123.30 On the one hand were the regular troops of the Zīrid
sultan, on the other the Bedouin tribes of the interior, but between them a
levy of the citizens, whose commitment to the defence of the city was quite
clearly crucial. The strength of their opinions, and their willingness to fight,
went on to become a major factor in the middle of the twelfth century,
when the coastal cities of Ifrīqiya, with the exception of Tunis, finally fell
under Norman rule, and their inhabitants were faced with a critical choice
between emigration, submission or resistance. This whole episode, which
I have discussed elsewhere,31 has a strong bearing on the fate of the coastal
cities of Palestine-Syria in the first quarter of the century.
The main difference between the two cases was that the cities of Ifrīqiya
were peripheral to the Norman kingdom of Sicily, not central as those of
Palestine-Syria were to the Latin kingdom of Jerusalem and the county
of Tripoli. They were on the whole treated far more leniently; with the
exception of Mahdia, they were left to govern themselves; they were able
in consequence to rebel; and they were finally delivered by a Muslim force,
the army of the Almohad caliph ‘Abd al-Mu’min. The emotions, however,
the issues and the solutions, were essentially the same. As at Ascalon in
1111, the upstart ruler of Gabes who had turned to Roger to establish his
independence was put to death in 1147 in a popular uprising, while at
Mahdia in 1148, the Zīrid sultan Hasan abandoned the city rather than
submit. But in accordance with the practice of Islamic Law, the Normans
offered first amān, ‘safe-conduct’, and secondly ‘ahd, ‘binding agreement’
to rule with equity in return for obedience. Applied not only to Mahdia
but to the other cities of the coast, from Sousse through Sfax and Gabes to
Tarābulus al-Gharb, the Tripoli in modern Libya, such agreements were
the foundation of the new regime. Along the coast of Palestine-Syria the
same applied at Arsuf in 1101, Acre in 1104, Tripoli in 1109, Sidon in 1110,
and finally at Tyre in 1124. Only Caesarea in 1101 and Beirut in 1110 were,
like Jerusalem, taken by assault and massacre. The citizens were left free to
emigrate, or to remain under the dhimma or protection of the new rulers.
The decision to go or to remain was vital. At stake was the fundamental
need of the Muslim community to live under the law of Islam. That required
the presence of a qādī or judge, duly appointed by a Muslim ruler with the
power to enforce his judgements. It required, in other words, residence
30
Michael Brett, ‘The Armies of Ifriqiya, 1052–1160’, in Brett, Ibn Khaldun and
the Medieval Maghrib, Essay XII.
31
Michael Brett, ‘Muslim Justice under Infidel Rule: The Normans in Ifriqiya,
517–555H/1123–1160AD’, in Brett, Ibn Khaldun and the Medieval Maghrib,
Essay XIII.
233
Michael Brett
within the Dar al-Islam, in a land where Islam was politically as well as
religiously supreme. The problem arose when this was not so. It had long
been debated in Ifrīqiya, where Muslim merchants who crossed the Sahara
found themselves in pagan territory, under pagan rule. The c ommonsense
solution of the jurists was that in the absence of the qadi, they should choose
for themselves a nāzir or ‘watchman’, an episcopus, no less, who would
judge them by the law in accordance with his knowledge and ability, with
the consent of the pagan monarch.32 The question became more difficult
when it was raised at home, first in Sicily and then in Ifrīqiya itself, when
the Muslim state was replaced by the government of the infidel, and the
qādī himself was appointed by non-Islamic authority. Both in Sicily and
in Ifrīqiya, however, the consensus of the jurists was the same. The need to
ensure that the community lived by the Law, without which it would perish,
overrode the illegality of the appointment of the judge responsible for its
administration. The situation was to some extent regularised by the ‘ahd, the
pact which the community had made with the infidel ruler, and which had
the force of law. The moment that it was broken by the ruler, however, the
‘ahd was dissolved, and the community was obliged to strive for the resto-
ration of legitimate Islamic rule. In the case of Ifrīqiya, this led to the swift
expulsion of the Normans after a mere decade. In Sicily, on the other hand,
the occasion never arose: Muslim rebellions against increasing oppression
simply resulted in the eventual eradication of the community in the island.
Whatever the choice of the inhabitants of the coastal cities of Palestine-
Syria, it was evidently made by Muslims as Muslims who were viewed and
treated as such by their conquerors. It was thus a decision based upon
religion, if not necessarily at the behest of the Law. Emigration was not
to be undertaken lightly, and in the absence of compulsion, including
fear, Hillenbrand supposes that as in Ifrīqiya only the relatively wealthy
could afford to leave, and even then may have looked to return.33 If such
emigrés took with them the ardour shown in defence of their cities, it may
be supposed that they contributed to the groundswell of popular opinion
in Syria and in Egypt in favour of holy war. On the other hand, those like
the Muslims of Sicily who remained, may always have preferred a peace-
ful outcome, and like the Sicilians, settled for existence under an infidel
regime. In either case, the mentality of the thaghr, the frontier against
the infidel, if indeed it ever existed save in the minds of the chroniclers
and their sources, is likely to have faded with the fall of the cities and the
dissolution of their communities.
32
Michael Brett, ‘Islam and Trade in the Bilad al-Sudan, Tenth-Eleventh Century
AD’, in Brett, Ibn Khaldun and the Medieval Maghrib, Essay V.
33
Hillenbrand, The Crusades, pp. 75–76, 358–60, 362–63.
234
Emotions in the Service of Politics
Another Perspective on the Experience
of Crusading (1095–1187)
Sophia Menache
University of Haifa
Jonathan Riley-Smith, in an inaugural lecture delivered at Royal Holloway

College, elaborated on the concept of crusade as ‘a blood-feud waged
against those who had harmed members of Christ’s family’. He further
discussed the problematic relationship between the love of God and
‘love your neighbour’ in Christian thought, on the one hand, and the
crusades, with all their acts of violence perpetrated against the Muslims,
on the other.1 One may add, in this regard, that not only Muslims but
also Jews and Eastern Christians were victimised by Christian zeal from
the end of the eleventh century onwards. Although this early ‘Clash of
Civilisations’ has been satisfactorily studied from the political, socio-
economic, and cultural perspectives, the wide spectrum of emotions
provoked by the crusades remains in many aspects terra incognita.2 Such
disregard of emotions, moreover, characterises common historiographical
trends in medieval history at large.3 Riley-Smith elucidated one aspect of
the multifarious emotional baggage that the crusades carried in medieval
Christendom: love or, more specifically, love of God. He further clari-
fied in later studies the emotional manipulation of medieval audiences
1
Jonathan Riley-Smith, ‘Crusading as an Act of Love’, History 65 (1980), 177–92
(here p. 191).
2
Though a growing interest in emotions, especially as manifested in the Early
Middle Ages, is clearly detectable in recent years, much work is still to be done.
See, for instance, Mary Garrison, ‘The Study of Emotions in Early Medieval
History: Some Starting Points’, Early Medieval Europe 10 (2001), 243–50;
Catherine Cubitt, ‘The History of the Emotions: A Debate’, Early Medieval
Europe 10 (2001), 225–27; Barbara H. Rosenwein, ‘Writing without Fear about
Early Medieval Emotions’, Early Medieval Europe 10 (2001), 229–30.
3
Barbara H. Rosenwein, ‘Worrying about Emotions in History’, American
Historical Review 107 (2002), 821–45.
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. –.
Sophia Menache
by crusade preachers, who fostered a variety of feelings, such as horror,

sorrow, cruelty, and vengeance.4
In the general framework of medieval emotions, this paper focuses
on the gamut and essence of feelings raised at three cornerstones of the
history of the crusades and the Latin East, namely: the papal call to the
crusade and the resulting conquest of Jerusalem;5 the fall of Edessa and
the Second Crusade, with its outcomes; and the Christian defeat at the
Horns of Hattin. We strongly believe that research into the different
sources may expose the degree of reception of the papal message from
Clermont onwards, first and foremost among contemporary chroniclers
but perhaps also among medieval audiences at large.
Due to the scarcity of historical research on emotions in medieval
Christendom, some methodological clarifications are necessary. The
social-constructionist view, according to which emotions and their display
are formed and shaped by the society in which they are expressed, offers
a suitable framework.6 One may further note that emotions are managed
by individuals in a definite place and time. What may be seen as official
representations of emotions – as they appear in contemporary narrative
sources – were in fact effective shapers of individual representations.7
Some emotions, moreover, functioned both as collective and individual
shapers and as intermingled expressions of social needs/desires and, in
parallel, of individual feelings. Thus, violence, which in the crusader con-
text acquires crucial importance, could be evaluated as a medium through
which power was not only manifested but also understood and manipu-
lated.8 Violence might be employed not only as a political and a legal act,
a focus of fantasy, but also as an ultimate guarantor of honour, status, and
4
Jonathan Riley-Smith, ‘Christian Violence and the Crusades’, in Religious
Violence between Christians and Jews: Medieval Roots, Modern Perspectives, ed.
Anna Sapir Abulafia (New York, 2002), pp. 16–17.
5
Reference to liberation or conquest is quite indiscriminate and tries to follow
the medieval use of these terms. See Penny J. Cole, ‘Christian, Muslims, and the
“Liberation” of the Holy Land’, Catholic Historical Review 84 (1998), 1–10.
6
Ron Mallon and Stephen P. Stich, ‘The Odd Couple: The Compatibility of
Social Construction and Evolutionary Psychology’, Philosophy of Science 67
(2000), 133–54.
7
William M. Reddy, ‘Emotional Liberty: Politics and History in the
Anthropology of Emotions’, Cultural Anthropology 14 (1999), 256–88.
8
Gerd Althoff, ‘Empörung, Tränen, Zerknirschung: “Emotionen” in der
öffentlichen Kommunikation des Mittelalters’, Frühmittelalterliche Studien 30
(1996), 60–79.
236
Another Perspective on the Experience of Crusading
hierarchy, and even as an enticement to justice.9 Other medieval terms, as

well, permeated the crusader thesaurus and were commonly used by con-
temporary chroniclers to the point that sometimes it becomes difficult,
if not impossible, to differentiate between ‘real’ situations and literary
topoi. Some examples may clarify this state of affairs: ‘hatred’ (odium) was
a conventional term used in medieval secular jurisprudence to depict a
lasting public relationship between two adversaries. Conversely, ‘anger’
(ira), often described a short-term, thus reparable rage between members
of a kin group. In moral literature, the opposite of hatred was usually
love, while anger was balanced with patience.10 Such antonym, however,
was disregarded by Thomas Aquinas, who approached anger as the only
passion that had no contrary.11 Lonely or not, anger appears very often
in medieval sources as an integral part of the war, as both weapon and
armour.12 A fourteenth-century preachers’ handbook tried to explain
such contingency: ‘for many people today cannot take their revenge with
material weapons and therefore retain hatred through hardened anger
(iram induratam) in their hearts’.13 Different emotions often appear in
close association, their dissociation becoming difficult and sometimes
artificial, even with regard to clear antonyms.14 Love (amor) and hatred
(odium); desire (desiderium) and aversion (fuga); sadness (dolor) and joy
(delectatio); hope (spes) and despair (desperatio); fear (timor) and daring
(audacia): all reflect some associations commonly used by medieval
authors. The question still remains to what degree and, more important
still, for what purposes such terms permeated crusader sources as well.
The almost obligatory starting point for any research into the emotional
aspects of the crusades are the declarations pronounced by or attributed
to Pope Urban II, the main designer of the Christian enterprise overseas.
In his letter to the faithful in Flanders (December 1095), written shortly
9
Richard W. Kaeuper, ‘Introduction’, in Violence in Medieval Society, ed. Kaeuper
(Woodbridge, 2000), p. xiii.
10
Daniel Lord Smail, ‘Hatred as a Social Institution in Late Medieval Society’,
Speculum 76 (2001), 90–126 (here pp. 90–91).
11
Thomas Aquinas, Summa Theologiae, ed. and trans. Dominican Fathers, 5 vols
(London, 1964–80), quaestiones 1a, 2ae, 23.
12
Nancy Sherman, Stoic Warriors: The Ancient Philosophy behind the Military
Mind (Oxford, 2005), p. 65.
13
Fasciculus morum: A Fourteenth-Century Preacher’s Handbook, ed. and trans.
Siegfried Wenzel (Philadelphia, 1989), no. 2.3, pp. 118–21.
14
On the classical approaches, see Susan James, Passion and Action: The Emotions
in Seventeenth-Century Philosophy (Oxford, 1997), pp. 5–6.
237
Sophia Menache
after the council at Clermont, the pope mentioned the main factors that
legitimised, even dictated, the massive departure of the faithful eastwards:
We believe that you, brethren, learned long ago from many reports the
deplorable news that the barbarians in their frenzy have invaded and
ravaged the churches of God in the eastern regions. Worse still, they have
seized the Holy City of Christ, embellished by his passion and resurrec-
tion, and – it is blasphemy to say it – they have sold her and her churches
into abominable slavery.15
The atrocities committed by the Muslims – here referred to as ‘barbar-
ians’ – to the most precious shrines of the Christian faith were emphasised
time and again in papal correspondence, thereby provoking and justify-
ing feelings of anger, injury, and insult among contemporary audiences.
In the papal letter to the religious congregation of Vallombrosa, Urban
shared the means that would resolve such an abominable state of affairs:
We were stimulating the minds of knights to go on this expedition, since
they might be able to restrain the savagery of the Saracens by their arms
and restore the Christians to their former freedom.16
Papal rhetoric thus bestowed on the crusades all the attributes of a just
war, one that was intended to repair the harm done not only to the most
holy shrines of the Christian religion but also to the faithful themselves.
One may note, in this regard, that according to classical jurisprudence,
justice was ‘a steady and enduring will to render unto everyone his right’.
Urban took further care, in this context, to remind his contemporaries
of the crucial damage that the Christian enterprise would inflict on the
enemy’s pride, not only in the Holy Land but far and wide, in all Christian
lands that with God’s help would be released from Muslim rule. Referring
to the capture of Tarragona in the Iberian Peninsula, the pope argued:
‘For you know what a great defence it would be for Christ’s people and
what a terrible blow it would be to the Saracens if, by the goodness of
God, the position of that famous city were restored.’17 The search for
vengeance, then, appears to have played a most crucial role, offering
15
Epistulae et chartae ad historiam primi belli sacri expectantes, ed. Heinrich
Hagenmeyer (Innsbruck, 1901), pp. 136–37; The Crusades: Idea and Reality
1095–1271, ed. and trans. Louise Riley-Smith and Jonathan Riley-Smith
(London, 1981), p. 38.
16
‘Papsturkunden in Florenz’, ed. Wilhelm Wiederhold, Nachrichten von der
Gesellschaft der Wissenschaften zu Göttingen. Phil.-hist. Kl. (1901), pp. 306–25
(here pp. 313–14); The Crusades: Idea and Reality, p. 39.
17
Paul Kehr, Papsturkunden in Spanien. I Katalonien (Berlin, 1926), pp. 287–88;
The Crusades: Idea and Reality, p. 40.
238
an appealing component in the papal message, one that was especially

attractive to medieval knights. The Christian attack on the Muslims was
furthermore presented as a just reprisal for their desecration of the holy
shrines of the Christian faith. Indeed, the Muslims’ acts of sacrilege, their
despoliation of and vandalism against the Christians’ ancestral rights
to the Holy Land – which, by that time, had became the Patrimonium
Christi – actually rendered the crusades all the power and connotations
of the biblical just war.18
Urban’s correspondence thus offers an instructive perspective on the
emotional papal message as it was presented by the pope to contemporary
audiences, if not on the political goals pursued by the vicar of God on
earth in practice. Rather clearly, the pope appealed to the basic feelings
of the faithful without making any clear differentiation between ecclesi-
astical and secular groups. One should further note that the feelings to
which the pope pleaded did not pertain to the evangelical message of love,
either of God or your neighbour; they returned, instead, to the German
heritage of the cult of war and the bravery expected from the soldier on
the battlefield.19 By the end of the eleventh century such values/virtues
received renewed impetus from the consolidation of knighthood as a
well-defined social class.20 The question still remains, however, as to the
degree to which the anger, vengeance, and/or retaliation that the pope
attempted to foster did in fact permeate the narrative sources. In other
words, to what degree was the papal call indeed echoed in contemporary
sources, thus further justifying the atrocities that the crusaders, in their
18
There is a rich bibliography on the idea of just war and holy war. See, for
example, James A. Brundage, ‘Holy War and the Medieval Lawyers’, in The
Holy War, ed. Thomas F. Murphy (Columbus, 1976), pp. 99–140; H. E. John
Cowdrey, ‘The Genesis of the Crusades: The Spring of Western Ideas of Holy
War’, in The Holy War, pp. 9–32; Esther Cohen and Sophia Menache, ‘Holy
Wars and Sainted Warriors: Christian War Propaganda in the Middle Ages’,
Journal of Communication 36 (1986), 52–62.
19
The basis of the stereotyped image of medieval knights can be found in the
first-century Roman historian, Tacitus, in his appealing description of German
values and behaviour in De Germania, ch. 14, ed. Henry Furneaux (Oxford,
1894), pp. 64–65.
20
Jean Dunbabin, ‘From Clerk to Knight: Changing Orders’, in The Ideals and
Practice of Medieval Knighthood II: Papers from the Third Strawberry Hill
Conference 1986, ed. Christopher Harper-Bill and Ruth Harvey (Woodbridge,
1988), pp. 26–39; Richard Mortimer, ‘Knights and Knighthood in Germany
in the Central Middle Ages’, in Ideals and Practice of Medieval Knighthood II,
pp. 86–103.
239
Sophia Menache
turn, perpetrated against all those ‘neighbours’ whom they met in their
pilgrimage to earthly Jerusalem? The different versions of Urban’s speech
at Clermont (as they appear in various chronicles written about ten years
later; that is, after the First Crusade had materialised) may clarify the
breadth of the transmission of the papal message and the kind of emo-
tions it aroused.21 The very fact that the original version of the speech has
not reached us – a fact that Joshua Prawer accurately called ‘one of the
paradoxes of history’22 – allowed a rather free transfer of the ideas attrib-
uted to the pope into contemporary chronicles, their respective authors
recording in their own way and according to their own experiences what
they remembered or what they wanted to assign to Urban II.
Fulcher of Chartres, who probably was at Clermont and participated
in the First Crusade as chaplain of Stephen of Blois,23 emphasised the
search for vengeance implied in the words that he attributed to Urban II:
For the Turks, a Persian people, have attacked [the Eastern Christians].
[…] They have seized more and more of the land of the Christians, have
already defeated them in seven times as many battles, killed or captured
many people, have destroyed churches, and have devastated the kingdom
of God […] I, not I, but God exhorts you as heralds of Christ […] to has-
ten to exterminate this vile race from our lands and to aid the Christian
inhabitants in time.24
This touching empathy with the suffering of the Eastern Christians should,
in other words, justify the crusaders’ retaliation against their infamous
enemies – presented here as Turks and Persians25 – who had slaughtered
21
On the various versions of the papal speech, see Dana Carleton Munro, ‘The
Speech of Pope Urban II at Clermont, 1095’, American Historical Review 11
(1906), 231–42; H. E. John Cowdrey, ‘Pope Urban II’s Preaching of the First
Crusade’, History 55 (1970), 177–88. On the appeal to the crowds in the Middle
Ages, especially in the crusader context, see Gary Dickson, ‘Medieval Christian
Crowds and the Origins of Crowd Psychology’, Revue d’histoire ecclésiastique
95 (2000), 61–67.
22
Joshua Prawer, A History of the Latin Kingdom of Jerusalem, 2 vols ( Jerusalem,
1963), 1: 82.
23
On Fulcher and his writing, see Fulcher of Chartres, A History of the
Expedition to Jerusalem (1095–1127), trans. Frances Rita Ryan, ed. Harold S.
Fink (Knoxville, 1969), pp. 3–56.
24
Fulcherii Carnotensis Historia Iherosolymitana, I.3.3–4, ed. Heinrich
Hagenmeyer (Heidelberg, 1913), pp. 132–33; History of the Expedition to
Jerusalem, trans. Ryan, ed. Fink, p. 66.
25
About the crusaders’ lack of knowledge of the Muslims and Islam at this early
stage, see Svetlana Loutchitskaja, ‘Barbarae nations: Les peuples musulmans
240
and captured so many of the faithful while destroying the houses of God.
A vindictive God looking for vengeance and the recovery of Christian
lands by the force of arms thus replaced the evangelical loving God. As
noted earlier, anger often appears in medieval literature as a temporary
emotion that may be satisfied, and hence annulled, when vengeance is
achieved. Only when the last stage did not crystallise could anger then turn
into hatred of a more stable nature. Contemporary theoretical manuals
further differentiate between hatred as an emotion, and revenge, which
was depicted as an action achieved by material weapons.26 Medieval his-
toriography of the crusades, however, does not corroborate theoretical
differentiations of this kind: though vengeance was achieved and – at least
on the first stages of the crusades – the Muslims were indeed defeated on
the battlefield, anger and hatred joined to leave their mark on the Christian
approach to Muslims and Islam throughout the whole crusader period.
Robert the Monk, who was also at Clermont,27 conjured up the satanic
image of the Muslims in a message he ascribed to Pope Urban:
The race of Persians, a foreign people and a people rejected by God […]
has invaded the land of those Christians, depopulated them by slaughter
and plunder and arson, kidnapped some of the Christians and carried
them off to their own lands and put others to a wretched death, and has
either overthrown the churches of God or turned them over to the rituals
of their own religion. They throw down the altars after soiling them with
their own filth, circumcise Christians, and pour the resulting blood either
on the altars or into the baptismal vessels. When they feel like inflicting
a truly painful death on some they pierce their navels, pull out the end
of their intestines, tie them to a pole and whip them around it until, all
their bowels pulled out, they fall lifeless to the ground […] When Pope
Urban had eloquently spoken these words and many other things of the
same kind, all present were so moved that they united as one and shouted,
‘God wills it! God wills it!28
dans les chroniques de la première croisade’, in Autour de la Première Croisade,

ed. Michel Balard (Paris, 1996), pp. 99–108; Matthew Bennett, ‘The First
Crusaders’ Image of Muslims: The Influence of Vernacular Poetry’, Forum for
Modern Language Studies 22 (1986), 101–22.
26
Guy Halsall, ‘Violence and Society in the Early Medieval West: An
Introductory Essay’, in Violence and Society in the Early Medieval West, ed.
Halsall (Woodbridge, 1998), pp. 7–18.
27
On the few biographical facts known about Robert and his writing, see The
Historia Iherosolimitana of Robert the Monk, ed. Damien Kempf and Marcus
G. Bull (Woodbridge, 2013), pp. xvii-xxxiv.
28
Robert the Monk, ed. Kempf and Bull, pp. 5–8; Robert the Monk’s History of the
First Crusade, trans. Carol Sweetenham (Aldershot, 2005), pp. 79–81.
241
Sophia Menache
By ascribing to the Muslims the forced conversion of Christians, coupled

with what contemporaries saw as the barbarous practice of circumcision,
Urban II, in the version of Robert the Monk, actually appealed to the most
primitive and powerful fear that could grip contemporary audiences, that
of castration.29 In addition, given the detailed description of the painful
torture of Christians condemned to death,30 together with the charge of
desecration of churches, the papal message brought about the desired goal:
the immediate emotive response of the congregation, who saw in the papal
call the voice and will of an almighty God.31 Robert’s version is crucially
important from a communication perspective, since it reflects complete
synchronism between the pope and his audience, with the mobilisation
of the latter manifesting a very positive response to the papal call.
Guibert of Nogent further elaborated the satanic image of the Muslims,
while adding macabre details of his own in a letter allegedly written by
the emperor of Constantinople to Robert I, Count of Flanders. The lurid
detail, on the one hand, and the fact that it was attributed to the highest
Christian ruler of the time on the other, secured the desired impact on
contemporary audiences:
After Christianity was driven out, the churches which the pagans held
had been turned into stables for horses, mules, and other animals […] they
carried out all kinds of filthy activity in them, so that they had become
not cathedrals, but brothels and theatres […] They took virgins and made
them public prostitutes […] Mothers were violated in the presence of
their daughters, raped over and over again by different men, while their
daughters were compelled, not only to watch, but to sing obscene songs
29
On circumcision and the Christian approach to it, see, Ivan G. Marcus, Rituals
of Childhood: Jewish Acculturation in Medieval Europe (New Haven, 1996), pp.
76–78; David Nirenberg, Communities of Violence: Persecution of Minorities in
the Middle Ages (Princeton, 1996), pp. 147–48; Shaye J. D. Cohen, ‘Between
Judaism and Christianity: The Semicircumcision of Christians according to
Bernard Gui, His Sources and R. Eliezer of Metz’, Harvard Theological Review
94 (2001), 285–321; Richard Schenk, ‘Covenant Initiation: Thomas Aquinas
and Robert Kilwardby on the Sacrament of Circumcision’, in Ordo sapientiae
et amoris: Image et message de Saint Thomas d’Aquin, ed. Carlos-Josaphat Pinto
de Oliveira (Fribourg, 1993), pp. 555–93.
30
The martyrdom of Christians, including priests, at the time of mass became
a very common topos in Robert’s writing: see Historia Iherosolimitana, ed.
Kempf and Bull, p. 11; trans. Sweetenham, pp. 86–87.
31
Rachel Dressler, ‘Deus hoc vult: Ideology, Identity, and Sculptural Rhetoric at
the Time of the Crusades’, Medieval Encounters: Jewish, Christian and Muslim
Culture in Confluence and Dialogue 1 (1995), 188–218.
242
and to dance […] Finally reverence for all that was called Christian was
handed over to the brothel […] Their lust overflowed to the point that
the execrable and profoundly intolerable crime of sodomy, which they
committed against men of middle or low station, they also committed
against a certain bishop, killing him […]32
Johan Huizinga has emphasised the fascination that manifestations of
violence and macabre spectacles exerted on medieval audiences, a tendency
clearly discernible from the Early Middle Ages onwards.33 Still, what may
appear to modern eyes as extravagance and/or salacious exhibitionism
was not due to any primitive lack of restraint inherent in medieval rep-
resentations but to the public and theatrical nature of communication
at the time.34 The testimonies quoted above portrayed for the ordinary
eleventh-century believer a realistic image of the enemy, while translating
abstract terms – such as ‘infidel’ – into everyday practice, one that covered
the most abominable crimes perpetrated not only against Christians and
their faith but against humanity as a whole. Emphasis on sexual abuses,
covering a wide range from rape to sodomy, must have had a very dramatic
and traumatic effect on medieval audiences, who were educated by the
castrating catechism of Pauline dogmas and monastic ideals. The papal
call to the crusade and the rich range of emotions it evoked thus received
a propitious welcome in contemporary narrative sources. Indeed, they
reflect a profound identification with the papal call: hatred, anger, the
pursuit of revenge, the justification of the war; all these notes accompanied
the two-hundred-year symphony whose overture was heard in Clermont.
The conquest of Jerusalem by the crusaders on 15 July 1099 provides
a strange mixture of holy and profane emotions, with praying and fast-
ing, litanies and processions accompanying the massive slaughtering of
Muslims. The many sources reporting the Christians’ tremendous victory
offer a clear picture of the turmoil of emotions that seized the faithful as a
whole, without distinction of gender, age, or social status. After referring
32
Guibert de Nogent, Dei gesta per Francos et cinq autres textes, ed. Robert B. C.
Huygens (Turnhout, 1996); The Deeds of God through the Franks, trans. Robert
Levine (Woodbridge, 1997), pp. 36–37.
33
Johan Huizinga, The Waning of the Middle Ages: A Study in the Forms of Life,
Thought and Art in France and the Netherlands in the XIVth and XVth Centuries,
trans. Frederik Hopman (London, 1976), pp. 1–21. See also Philippe Ariès,
‘Huizinga et les thèmes macabres’, in Bijdragen en Mededelingen betreffende de
geschiedenis der Nederlanden (‘s-Gravenhage, 1973), pp. 246–57.
34
Gerd Althoff, ‘Ira Regis: Prolegomena to a History of Royal Anger’, in Anger’s
Past: The Social Uses of an Emotion in the Middle Ages, ed. Barbara H. Rosenwein
(Ithaca, NY, 1998), pp. 59–74 (here p. 74).
243
Sophia Menache
to the fears (timor) among the crusaders who had besieged Jerusalem, the
anonymous author of the Gesta Francorum – who probably participated
in the First Crusade – noted the extreme change of mood that followed
the conquest of the city:
Before we made this assault on the city, the bishops and priests persuaded
all, by exhorting and preaching, to honor the Lord by marching around
Jerusalem in a great procession, and to prepare for battle by prayer, fasting,
and almsgiving. [After the Christians entered the Holy City,] our men
followed, killing and slaying even to the Temple of Solomon, where the
slaughter was so great that our men waded in blood up to their ankles […]
But this time the pilgrims entered the city, where the enemy gathered in
force. The battle raged throughout the day, so that the Temple was covered
with their blood. When the pagans had been overcome, our men seized
great numbers, both men and women, either killing them or keeping them
captive, as they wished […] Later, all of our people went to the Sepulchre
of our Lord, rejoicing and weeping for joy, and they rendered up the offer-
ing that they owed […] No one has ever seen or heard of such a slaughter
of pagans, for they were burned on pyres like pyramids, and no one save
God alone knows how many there were.35
Albert of Aachen, as well, describes the cruel slaughtering of thousands
of Muslims, with the crusaders (whom he refers to as peregrini) not dif-
ferentiating among their victims.36
One may note, in this regard, the success of contemporary ecclesiastics
in giving a new meaning to the Christian love of God, while the divine
command to love your enemy (Matthew 5.43–48) disappears as though
it had never been. In the crusader context, the love of God justifies, and
even legitimises, the massacre of Muslims that followed the conquest of
Jerusalem. Prayers and abstention, processions and liturgies, before and
after the slaughtering of the infidel, appear as manifestations of gratitude to
a vindictive God, Who had bestowed upon the Christians an irrefutable,
35
Gesta Francorum et aliorum Hierosolimitanorum, 39, ed. and trans. Rosalind
Hill (London, 1962), pp. 89–92. See also Peter Partner, ‘Holy War, Crusade
and Jihad: An Attempt to Define Some Problems’, in Autour de la Première
Croisade, pp. 333–43; Benjamin Z. Kedar, ‘Croisade et Jihad vus par l’ennemi:
Une étude des perceptions mutuelles des motivations’, in Autour de la Première
Croisade, pp. 345–55.
36
Historia Ierosolimitana: History of the Journey to Jerusalem, ed. Susan B.
Edgington (Oxford, 2007), VI.22, pp. 430–31. In a letter to Pope Paschal II,
Godfrey of Bouillon, Raymond of Saint-Gilles, and Daibert estimated that
about 100,000 Saracens had been killed during the conquest of Jerusalem. See
Epistulae et chartae, no. XVIII, pp. 167–74.
244
just victory. Moreover, the different ways of approaching the infidel were
not accompanied by any signs of remorse.37 Contemporary reports of the
conquest of Jerusalem clearly reflect the twofold approach to anger found
in medieval sources. Medieval theologians criticised anger that had got
out of control; as such, it was unreasonable, sinful, and more suitable to
animals than to human beings. According to crusader chroniclers, how-
ever, the Christian liberation of Jerusalem was not such a case. The release
of Jerusalem from Muslim sacrilege, with all the atrocities this implied
in practice, was categorised as a reasonable, just, useful anger, for it was
directed at extirpating sin or punishing evil doers. It pertained, therefore,
to that kind of anger possessing a moral dimension.38
Peter Tudebode, who was also an eyewitness to the conquest of Jerusalem,
contrasted the Christi milites to illis paganis, placing emphasis on the cru-
saders’ joy (exultantes, gaudenter) as they approached the holy city. Peter
further refers to the exchange of visual and oral insults between the Muslim
inhabitants of Jerusalem and the crusaders assaulting the city, an exchange
that from the Muslim side was accompanied by invectives against the most
precious values of the Christian faith.39 William of Tyre, who gives a simi-
lar report, emphasises the feelings of ira that engulfed the faithful as the
Muslims committed acts of sacrilege right in their face.40 He also depicts
in most touching tones the devotion and piety that the crusaders expressed
in the church of the Holy Sepulchre following the conquest of the city.41
The crusaders’ conquest of Jerusalem thus served as a fruitful arena to
express conflicting feelings that had been fermenting for years, at least
37
On parallel phenomena in Muslim literature, see Zouhair Ghazzal, ‘From
Anger on Behalf of God to “Forbearance” in Islamic Medieval Literature’, in
Anger’s Past, pp. 203–30.
38
On both kinds of anger, see Lester K. Little, ‘Anger in Monastic Curses’, in
Anger’s Past, pp. 9–35 (here pp. 12–13); Richard E. Barton, ‘“Zealous Anger”
and the Renegotiation of Aristocratic Relationships in Eleventh- and Twelfth-
Century France’, in Anger’s Past, pp. 153–70 (here p. 155).
39
Petrus Tudebodus, Historia de Hierosolymitano itinere, ed. John H. and Laurita
L. Hill (Paris, 1977), p. 134.
40
Willelmi Tyrensis Archiepiscopi Chronicon, 8.11, ed. Robert B.C. Huygens, 2 vols
(Turnhout, 1986), p. 401.
41
‘Intueri erat amenissimum et spirituali plenum iocunditate quanta devotione,
quanto pii fervore desiderii ad loca sancta fidelis accederet populus, quanta
mentis exultatione et spirituali gaudio dominice dispensationis deosculabantur
memoriam: ubique lacrime, ubique suspiria, non qualia meror et anxietas
solet extorquere sed qualia fervens devotio.’ Willelmi Tyrensis Archiepiscopi
Chronicon, 8.11, ed. Huygens, pp. 413–14.
245
Sophia Menache
since the day of the call of Clermont. At a first stage, hatred, aversion,
despair, fears, and, above all, anger, were integral to the Christian enter-
prise. The conquest of Jerusalem, however, brought about unlimited joy
and hope. These positive emotions appeared as just compensation for
those who had suffered the vicissitudes of the long road to the Holy Land
and finally, with the help of God, had victoriously entered the Holy City.
The emotive notes that accompanied the liberation of terrestrial Jerusalem
were faithfully expressed by Fulcher of Chartres:
Oh day so ardently desired! Oh time of times the most memorable! Oh
deed before all other deeds! Desired indeed because in the inner longing
of the heart it had always been hoped by all believers in the Catholic faith
that the place in which the Creator of all creatures, God made man, in
His manifold pity for mankind, had by His birth, death, and resurrection,
conferred the gift of redemption would be restored to its pristine dignity
by those believing and trusting in Him. They desired that this place, so
long contaminated by the superstition of the pagan inhabitants, should
be cleansed from their contagion […] And this same work which the Lord
chose to accomplish through His people, His dearly beloved children and
family, chose, I believe, for this task, shall resound and continue memorable
in the tongues of all nations until the end of time.42
The mass euphoria that followed the conquest of Jerusalem was not a
short-lived phenomenon, and Fulcher provides a faithful reflection of
the sense of mission — or mission accomplished — that characterised the
first settlers in the crusader kingdom about twenty years later:
Consider, I pray, and reflect how in our time God has transformed the
Occident into the Orient. For we who were Occidentals have now become
Orientals. He who was a Roman or a Frank has in this land been made into
a Galilean, or a Palestinian […] We have already forgotten the places of our
birth […] Words of different languages have become common property
known to each nationality, and mutual faith unites those who are ignorant
of their descent […] He who was born a stranger is now as one born here;
he who was born an alien has become a native. Our relatives and parents
join us from time to time, sacrificing, even though reluctantly, all that they
formerly possessed […] You see, therefore, that this is a great miracle, and
one which the whole world ought to admire.43
Divine intervention did not, therefore, come to an end with the lib-
eration of Jerusalem. It was part and parcel of the crusader enterprise
42
Historia Hierosolymitana, ed. Hagenmeyer (Heidelberg, 1913), I. 29.3–4,
pp. 305–6; History of the Expedition to Jerusalem, p. 123.
43
Historia Iherosolymitana, III.37.2–8, pp. 748–49; History of the Expedition to
Jerusalem, pp. 271–72.
246
thereafter and, as such, it created a new kind of solidarity and sense of

mission among all Christians in the Holy Land. The words of the apostle
thus received new meaning: ‘Where there is neither Greek nor Jew, cir-
cumcision nor uncircumcision, Barbarian, Scythían, bond nor free: but
Christ is all, and in all’ (Colossians 3.11). The Christian faith provided the
strongest relationship among those who fixed their residence overseas.
The holiness of the land further strengthened the bonds among its new
inhabitants, who became Galileans and Palestinians thus forgetting their
European roots. But this sense of mission, which was a direct outcome
of the first victories, began to decay, even to collapse, with the continual
reversals on the battlefield. New feelings, most of them of a negative, pes-
simistic nature, thus came to the fore forty-five years later.
Disappointment, frustration, and sometimes despair were, indeed, the
outcomes of the fall of Edessa (25 December 1144) and further Christian
defeats during the Second Crusade and its aftermath (1147–1148).44
William of Tyre describes in detail the turmoil of emotions in the papal
curia that brought about the call for a new crusade. He lays emphasis
on the enlistment of Bernard of Clairvaux, who brought about a posi-
tive response not only among the political elite of the time but among
the masses as well.45 On the other hand, the essay written by Bernard of
Clairvaux for his former pupil, Pope Eugene III (1148–1149), reflects
the negative emotions and the sense of frustration engendered by the
continuous Christian defeats overseas:
The sons of the Church and those who are counted as Christians have been
overthrown in the desert, slain by the sword or consumed by hunger […]
Fear and sorrow and confusion are in the inner chambers of their kings.
How confused are the feet of those who preach peace, of those who bring
glad tidings of good things! We have said ‘Peace’ and there is no peace; we
have promised good things and you see there is disorder, so that it looks as
though we have gone into this business rashly without stopping to think
[…] But perhaps our contemporaries say, ‘How can we know that what
you say is truly inspired by the Lord? What proof can you give us to make
44
On the historical context, see, John G. Rowe, ‘The Origins of the Second
Crusade: Pope Eugenius III, Bernard of Clairvaux and Louis VII of France’, in
The Second Crusade and the Cistercians, ed. Michael Gervers (New York, 1992),
pp. 79–90; Jonathan Phillips, ‘Papacy, Empire and the Second Crusade’, in
The Second Crusade: Scope and Consequences, ed. Jonathan Phillips and Martin
Hoch (Manchester, 2001), pp. 15–31; Martin Hoch, ‘The Price of Failure: The
Second Crusade as a Turning-Point in the History of the Latin East?’, in The
Second Crusade: Scope and Consequences, pp. 180–200.
45
Willelmi Tyrensis Archiepiscopi Chronicon , 16.18, ed. Huygens, pp. 739–41.
247
Sophia Menache
us believe in you?’ I have no answer to their questions; they must spare

my embarrassment.46
The confusion and mortification alluded to by St Bernard was a direct
result of the disappointment of contemporaries following the Second
Crusade, whose undertaking had been so vehemently preached by the
Cistercian monk a few years earlier. We find here the first seeds of the
doubts and even suspicions that became a crucial element in the criticism
of the crusade throughout the thirteenth century; namely, whether the
movement as a whole was inspired by God or, rather, by the greed and
ambitions of human beings, the pope at their head, who were not immune
to error and miscalculation.47
As the situation on the battlefield did not improve and the fall of Edessa
turned into an incontrovertible fact, the pursuit of revenge became more
acute and pressing. The search for vengeance turned into a pattern shared
not just by the crusaders but allegedly also by Jesus Christ Himself, Who
at the zenith of His passion was said to have anticipated the coming of the
crusaders to the Holy Land. In a scene added by Graindor to La Chanson
d’Antioche (c. 1180),48 Jesus tries to comfort the thief at his side by declaring:
Friend […] the people are not yet born
Who will come to avenge me with their steel lances.
So they will come to kill the faithless pagans
Who have always refused my commandments.
Holy Christianity will be honoured by them
And my land conquered and my country freed.
A thousand years from today they will be baptised and raised
And will cause the Holy Sepulchre to be regained and adored.
And they will serve me as though they were my offspring;
They will all be my sons, I promise them that
In heavenly paradise shall their heritage be.49
46
Bernard de Clairvaux, De consideratione, in Sancti Bernardi Opera, ed. Jean Leclercq
et al., 8 vols (Rome, 1963), 3: 410–13; The Crusades: Idea and Reality, pp. 62–63.
47
On the growing criticism of the crusades, see Elizabeth Siberry, Criticism of
Crusading 1095–1274 (Oxford, 1985), passim, and the bibliography included.
48
One should note, however, that there is considerable debate about the
authorship of the Chanson. See Susan B. Edgington and Carol Sweetenham,
The Chanson d’Antioche: An Old French Account of the First Crusade (Farnham,
2011), pp. 3–8.
49
La chanson d’Antioche, ed. Suzanne Duparc-Quioc, 2 vols (Paris, 1976), 1:
25–28; The Crusades: Idea and Reality, p. 72.
248
The prophetic tone of these words was aimed at strengthening the confi-
dence of the crusaders, whose real reward was not in this world but in the
next. The motifs of divine retribution in the world to come were hardly
new: they were at the core of the Sermon of the Mountain (Matthew 5) and
were very popular among the early Christians.50 Reinforced by Bernard of
Clairvaux in his preaching for the Second Crusade, the hope for divine
retribution provided a kind of psychological compensation for the many
dilemmas and difficulties that faced the crusaders in the Holy Land from
the 1140s onward.
The Christian setback at the Horns of Hattin (3 July 1187),51 moreover,
served as a catalyst for renewed emotional turmoil, either caused or justi-
fied by the unprecedented scope of the defeat and its awful consequences.
Furthermore, it brought about an extraordinary flow of guilt, which the
enormous calamity transformed into an unquestionable truth: the sins
of the crusaders and of the inhabitants in the Latin East as a whole had
brought about God’s anger and were responsible for their collapse and the
victory of the infidel. This interpretation is fully developed by Patriarch
Eraclius in a letter written to Pope Urban about two months after Hattin
(5–20 September 1187):
The enormity of our lamentation and sorrow, Reverend Father, we are
scarcely able to convey to your piety’s ears. It has fallen to us to see in our
days the oppression of our people, the doleful and lamentable desolation
of the holy church of Jerusalem and that which is holy given unto dogs
(Matthew 7: 6). Truly, Holy Father, the anger of the Lord has come upon
us and His terrors have put us to confusion (Psalm 88: 16) […] He has
given over our king and the whole Christian army into the hand of pagans
[…] Nor are these things enough to satiate the barbarity of the enemies of
the cross of Christ. Indeed, striving to blot out the Christian name from
under heaven […] Alas, alas, O Reverend Father, that the Holy Land, the
inheritance of the Crucified, should be given into the hands of pagans.
Alas, alas that the Lord has thrown away His inheritance and has not
spared it, withholding His mercy behind his anger.52
50
Robert Bultot, Valeurs chrétiennes: La doctrine du mépris du monde au xie siècle
(Louvain, 1964), pp. 11–57; Melvin Storm, ‘A culpa et a poena: Christ’s Pardon
and the Pardoner’s’, Neuphilologische Mitteilungen 83 (1982), 439–42.
51
On the battle, its course, and consequences, see Benjamin Z. Kedar, ‘The Battle
of Hattin Revisited’, in The Horns of Hattin, ed. Benjamin Z. Kedar ( Jerusalem,
1992), pp. 190–207; William J. Hamblin, ‘Saladin and Muslim Military Theory’,
in Horns of Hattin, pp. 228–38.
52
Trans. in The Conquest of Jerusalem and the Third Crusade: Sources in
Translation, ed. Peter W. Edbury (Aldershot, 1998), pp. 162–63.
249
Sophia Menache
The disaster on the battlefield thus became an outcome and faithful

reflection of the fury of the vindictive God, Who had accompanied and
supported the crusades from their very beginnings, though by the end of
the twelfth century His victims were no longer Muslims but Christians.
Hence the imperative to find the reason that might justify such rigorous
punishment. It was easily found in a well-known topos in medieval litera-
ture, peccatis nostris exigentibus; that is, the many sins of the faithful were
the cause and justification of divine punishment. A letter written to the
prior of the Hospitallers in Italy refers to the fact that ‘as a consequence of
our sins many of our men were killed, and the Christians were defeated’.53
An Old French continuation to the chronicle of William of Tyre returns
to the same topos, while also mentioning one of its many victims, the
vicar of God, Urban III, who had died of grief:
After this division [of the count of Tripoli] had been defeated, the anger
of God was so great against the Christian host because of their sins that
Saladin vanquished them quickly […] The disaster befell Christendom at
a place called the Horns of Hattin […] The news of it struck the hearts
of those faithful to Jesus Christ. Pope Urban who was at Ferrara died of
grief when he heard the news.54
Pope Urban’s distress did not remain an isolated phenomenon. On the
contrary, his successor at the Apostolic See, Gregory VIII, depicted in
ominous colours the emotional reactions in the papal curia to the defeat
at Hattin (Audita tremendi, November 1187):
On hearing with what severe and terrible judgment the land of Jerusalem
has been smitten by the divine hand, we and our brothers have been con-
founded by such great horror and affected by such great sorrow that we
could not easily decide what to do or say; over this situation the psalmist
laments and says, ‘Oh God, the heathens are come into thy inheritance.’
[…] We ought not to be so downhearted that we fall into want of faith and
believe that God, angered by his people in such a way as to allow himself
to become infuriated by the manifold actions of a host of common sinners,
will not through his mercy be quickly placated by penance, that he will
not console us and that after weeping and tears he will not bring rejoicing.
For anyone of sane mind who does not weep at such a cause for weeping
[…] would seem to have forgotten not only his Christian faith […] but
even his very humanity, since every sensible man can surmise the details
which we have left out, from the very magnitude of the peril, with those
53
The Conquest of Jerusalem and the Third Crusade, p. 161.
54
La continuation de Guillaume de Tyr (1184–1197), ed. Margareth R. Morgan
(Paris, 1982), pp. 54–55; The Conquest of Jerusalem and the Third Crusade, p. 47.
250
savage barbarians thirsting after Christian blood and using all their force
to profane the Holy Places and banish the worship of God from the land.55
The emotional turmoil affected not only the pope and the cardinals but
the whole of Christendom, which was further stirred by the atrocities
committed anew by the infidel to the most holy shrines of the Christian
faith. Hattin became in the papal document the most faithful reflection
of God’s anger against Christians: their sins and malevolence brought
about not only the loss of the Holy Land but also the capture of the Cross,
which now lay in the contaminated hands of the infidel.56
Acknowledgment of God’s infinite justice vis-à-vis the deliberate
attempt to foster a widespread sense of mea culpa was not an ephemeral
strategy of the papal curia alone. An appeal to similar emotions appears
in a letter written by Conrad of Montferrat to Baldwin, Archbishop of
Canterbury, almost two years after Hattin (1189):
The holy city of Jerusalem, despoiled of its worshippers, is to be mourned
and lamented. As a consequence of their sins, its inhabitants have been
placed under tribute to Saladin, and, having paid the capitation tax, are
driven far from the kingdom. The walls of Jerusalem are bereft of their
hermit occupants. God has stood back as if from the defilement of our evil,
and Mohammed has taken over; where Christ was prayed to day and night
at the appointed hours, now Mohammed is praised with uplifted voice.57
The widespread use of ‘because of our sins’ may raise questions as to the
actual value of this expression to express genuine feelings. As well pointed
out by Mary Garrison, the very use of what she calls ‘prefabricated units
of expression’ – quotations, allusions, proverbs, or clichés – may have
been deliberate precisely because of their communicative power.58 Peter
Dronke has further demonstrated that a distinctive use of a topos can
itself constitute individuality within a tradition.59 The extensive use of the
55
Historia de expeditione Friderici imperatoris, ed. Anton Chroust, MGH
Scriptores rerum Germanicarum, n.s. 5 (Berlin, 1928), 6–10; The Crusades: Idea
and Reality, pp. 64–65.
56
On Gregory’s crusading policy and the preaching of the crusade at this stage,
see Penny J. Cole, The Preaching of the Crusades to the Holy Land, 1095–1270
(Cambridge, 1991), pp. 62–79.
57
Ralph of Diceto, Opera Historica, ed. William Stubbs in Chronicles and
Memorials of the Reign of Richard I, RS, 2 vols (London, 1865), 2: 60–62; The
Conquest of Jerusalem and the Third Crusade, pp. 168–69.
58
Garrison, ‘The Study of Emotions’, pp. 245–46.
59
Peter Dronke, Poetic Individuality in the Middle Ages: New Departures in Poetry
1000–1150 (Oxford, 1970), pp. 11–12.
251
Sophia Menache
cliché, ‘because of our sins,’ cannot by itself deny or diminish the sense
of anguish and despair that embraced Christians in the twelfth century.
Faced with the difficult question of how the Christ had refrained from
bestowing victory on the Cross and the Christian armies and, instead, had
chosen the infidel – whose terrible crimes and perceived barbarous essence
were by now well known – contemporaries could only reply with a sense
of mea culpa, a form of expiation that had been preached continuously by
the Church’s representatives for generations.60 Moreover, the ubiquity of
the topos in ecclesiastical correspondence and in narrative sources as well,
may testify to its roots in the climate of opinion prevailing in medieval
Christendom.
This short review hints at the broad spectrum of emotions that were
closely connected with the Christian epos in the Latin East and reveals
some of the ways by which contemporaries tried to understand and explain
the historical process. No clear differences in approach are discernible
among the sources, a situation that allows the conclusion that the papal
message of Clermont – at least in its core – was widely accepted by
contemporary authors. Medieval chroniclers used all the rich spectrum
of words at their disposal without making clear differentiations among
anger, hatred, violence, and revenge – terms that were distinctively
employed in contemporary jurisprudence and theological treatises alike.
Thus, reference only to the words themselves limits the field of research,
while historical contextualisation allows a better analytical and theoreti-
cal perspective of the intricate emotional spectrum that was inherent in
medieval Christendom. A process of continuous emotional adjustment
to changing circumstances is also clearly discernible, while the thesaurus
of emotional expressions changed and revitalised itself according to new
exigencies. Still, the hatred of Muslims and Islam remained constant
throughout the crusader period, notwithstanding the meeting with the
enemy and the complete victory over it in the early stages. The attempt to
deny any legitimacy to the Muslim rule over Jerusalem did not diminish
over the years. Although a more moderate image of the ‘pagans’ emerged
toward the end of the twelfth century,61 Peter of Blois still referred to the
60
Elisabeth F. Vodola, ‘Fides et culpa: The Use of Roman Law in Ecclesiastical
Ideology’, in Authority and Power: Studies on Medieval Law and Government
Presented to Walter Ullman on his Seventieth Birthday, ed. Brian Tierney and
Peter Linehan (Cambridge, 1980), pp. 83–97; Jean C. Payen, Le motif du
repentir dans la literature française médiévale: Des origines à 1230 (Genève,
1968), passim.
61
Rainer C. Schwinges, Kreuzzugsideologie und Toleranz (Stuttgart, 1977),
pp. 102–8.
252
Muslims as those ‘who blaspheme against Christ, pollute the sanctuary

of the Lord and in their pride and miscreance abase the glory of our
Redeemer’.62 This hatred, moreover, was actually a sine qua non of the
crusades and, therefore, was continuously fostered by ecclesiastical
propaganda. On the other hand, the historical process left its mark on
the changing emotions among the crusaders, whose anger and hatred
towards the Muslims were strengthened, if not replaced, by the increas-
ing search for revenge and the subsequent search for their own sins as
justification for their repeated defeats.
Contextualisation of the negative emotional vocabulary that was used
to delineate and describe Christian-Muslim relations proves, though, to
be misleading. In fact, medieval sources described the Muslims with the
same spectrum of negative emotions that were used to reproach different
groups within Christendom itself and the heretical sects provide a clear
example of this approach. The crusades, moreover, took place shortly
after Christian society had allegedly settled its internal differences in
the form of the Treuga Dei; the ideal of knighthood crystallised across
the same spectrum of values and emotions that delineated the discourse
in which the Muslims were discussed. Less than a century before the
crusades, different groups in Christian society had been the target of
the same pejorative emotions that were used to denounce and reproach
the Muslims. These terms should, therefore, be seen and analysed, not to
produce a superficial moral reading of the vilification of the Muslims but
as an essential part of the thesaurus by which Christian society analysed
itself. In fact, the use of the same Augustinian emotional index transforms
negative attitudes toward the Muslims and Islam into an act of inverted
inclusion of the Muslims within the Christian sphere;63 in other words,
using illusionary inclusion in order to exclude. This inverted inclusion
means that within its inner discourse, Christian society had defeated the
Muslims symbolically, independent of the real outcome on the battlefield.
The transformation of the crusaders from ‘Westerners’ into ‘Easterners’ in
Fulcher’s sense of mission (see above, notes 44–45) is a conscious practice
of erasing the ‘other’ by expropriating its identity. This was not, how-
ever, an act of including the Easterner in the crusaders’ Weltanschauung,
but a symbolic denial that served to exclude the Easterners altogether.
62
Peter of Blois, Epistolae, PL 207, col. 1064; Riley-Smith, ‘Crusading as an Act
of Love’, p. 177.
63
See, for instance, the use of Augustine’s arguments by Jacques de Vitry, bishop
of Acre, to justify violence against infidels (c. 1216–25): Jacques de Vitry,
‘Sermones Vulgares’, in Analecta novissima Spicilegii Solesmensis, ed. Cardinal
Jean Baptiste Pitra, 2 vols (Paris, 1888), 2: 419–20.
253
Sophia Menache
The inverted inclusion of the Muslims thus became the last step on the
long march of both inclusion and, at the same time, erasing the infidel,
for it was Christianity that defined the cultural boundaries of the West.
In this way, in almost an embryonic ‘Orientalist’ manner, crusader soci-
ety had subordinated the East; indeed, had turned it into a part of the
West. The emotional discourse thus proves to be a political narrative, and
in this fashion it should be read.
254
Growing Up to Become a Crusader
The Next Generation*
Sini Kangas
University of Helsinki
Who can tell of the boys, the old men, who are stirred to go to war? Who
can count the bands of virgins […]? They all sing of battle, but do not say
that they fight.1
Around of the turn of the twelfth century, Westerners were deeply con-
cerned with the fate of Jerusalem. Guibert of Nogent, who concluded his
chronicle in about 1108, depicted European roads filling with groups of
crusaders, setting forth for Jerusalem with their entire households: adults,
boys, and girls, of various ages. In European memory the First Crusade
has remained a religious mass movement, involving the whole community
from the newborn to the elderly.
Among the Western nobility, crusading evolved into a prestige-
enhancing family institution entwined with social networks formed by
marriage, landholding, vassalage and religious sponsorship. Fighting the
holy war and visiting shrines in the Holy Land was a characteristic start-
ing point in creating a crusader profile for the family, but the impact of
the phenomenon in medieval communities extended far beyond this.
The longevity of crusading in western societies depended on commit-
ment over generations throughout the twelfth and thirteenth centuries.
Like the great majority of European adults, most children did not leave
their homes to wander eastwards, but they grew up in societies highly
*
I wish to thank Niilo Helander Foundation for generous support in the writing
of this article, Dr Susan Edgington for her insightful comments, as well as
Dr Philip Line, who has not only revised the English language of my texts, but
also improved the content by accurate questions.
1
‘Quis pueros, quis dicat anus ad bella moveri, quis quit virgineas annumerare
manus, quis referat […]? Bella canunt omnes; nec se pugnare fatentur’: Guibert
of Nogent, Dei gesta per Francos et cinq autres textes, ed. Robert B.C. Huygens
(Turnhout, 1996), pp. 119–20.
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. –.
FHG DOI: 10.1/M.OUTREMER-EB.1.1
Sini Kangas
appreciative of crusading and pilgrimage. Children provided the most

important reserve of future crusaders.
This chapter concerns underage participation in twelfth-century cru-
sades, beginning with the definitions of a child in the sources and going on
to discuss the children who travelled east with their families. It concludes
by looking at the broader frameworks that supported the commitment of
future generations to crusading: roles in the family, education, religiosity,
and lay culture.
Under-Age Crusaders in the Sources

Twelfth-century sources include sporadic information about child
ren.2 What is clear is that European (as well as Armenian, Syrian and
Palestinian) children were present on the crusades from the outset,
and that the sources usually associate children with the adults of their
sub-group without much ado. The children we learn most about are
likely to be noble boys older than seven years. Noble girls appear in the
sources in more limited circumstances – inheritance of remarkable areas
of land or marriage – whereas references to children of lower orders are
infrequent. When they occur they usually mention children as part of
a larger group which comprises either mothers and their children, or
only children.
In spiritual terms, adult crusaders understood under-age participation
to be equal to their own. Children shared the privileged status of adult
pilgrims, including plenary indulgence. Perceptions of martyrdom seem
to have been relatively loose, encompassing victims of famine or sickness.
After the crusader victory at Nicaea in 1097, children were accounted
martyrs along with the adults. It was popularly believed that God would
permit salvation for anyone who perished with ‘right intention’.3
The sources describe children playing, hurting themselves, and asking
naive questions from the adults. The adults’ response was sympathy for
their vulnerability. Young crusaders played war games when they had an
opportunity to do so, and the adults are reported to have kept an eye on
them, groaning at the sight of children on both sides being wounded,
2
Chronicles, letters, charters, legal agreements, canon law, secular law codes,
chansons de geste, esp. the first and second crusade cycles.
3
(London, 1962), p. 17; Guibert of Nogent, Dei gesta, ed. Huygens, p. 120;
Baldric of Dol, ‘Historia Jerosolimitana’, in RHC Occ. 4: 30.
256
Growing Up to Become a Crusader: The Next Generation
and removing the children from their battlefield.4 Guibert of Nogent

mentions children bored by prolonged travelling, and asking if they had
reached Jerusalem before every castle or city, or if there was still a long
way to go.5
What is important in these examples is not that children are described
as behaving in a childlike manner, but that they represent model behav-
iour: they attack enemies and long for the holy city. The basic virtues of
crusaders were unshakeable faith, determination to die for God, loyalty to
fellow Christians, and, especially in the case of the nobility, prowess, skill
and endurance in fighting. The first crusaders had been elected among the
nations6 to win back the Christian inheritance, the Holy Land, to avenge
the unjust attack on Christians and, as noble crusaders themselves under-
stood the agenda, to wipe away the shame of the Christian nation caused
by the hanging of Christ on the cross.7 In crusade propaganda these char-
acteristics were attributed to exemplary crusaders regardless of their age.
In medieval society the proportion of children and young adults was
larger than in our own, and it is logical to assume that the audiences for
preaching, histories and stories likewise had a strong underage represen-
tation. The authors of both ecclesiastical and more secular sources were
propagating holy war, and they wanted to make their ideas persuasive.
Allusions to children were included in order to inspire and influence the
next generation to follow the example of the first crusaders, and therefore
these sources may tell more about the target audience and their expecta-
tions than the actual events, especially in the case of texts not written by
participants. The triumphant news of the conquest of Jerusalem in 1099
and its reception in the West (re)constructed the narrative tradition of
the First Crusade.
Realistic stories are rare, especially in the case of children of common
birth. Fulcher and Lisiard include an unusual story of an Armenian
rusticus (peasant) and his infantula (little daughter), who helped Count
Joscelin I of Edessa (d. 1131) to get back safely to Turbessel when he was
escaping from captivity. The peasant gave the desperate count bread and
a ride on his donkey, and he made his little girl sit in front of the riding
4
Guibert of Nogent, Dei gesta, ed. Huygens, pp. 309–10.
5
6
Ps. 32:12. ‘Gens Francorum […] a Deo electa […] quibus prae ceteris gentibus
contulit Deus’: Robert the Monk, ‘Historia Iherosolimitana’, in RHC Occ. 3:
727–28, 723, 882; ‘Christi gente, Christi militibus’: Gesta Francorum, ed. Hill,
pp. 6, 53–54, 56; Guibert of Nogent, Dei gesta, ed. Huygens, pp. 123, 192–93.
7
La Chanson d’Antioche, ed. Suzanne Duparc-Quioc, 2 vols (Paris, 1976), 1: 27,
lines 205–9.
257
Sini Kangas
count, so that passers-by would think that he was the father of the child.
She, however, pestered the count, cried aloud and could not be pacified
until they reached their destination. Both sources underline the great
relief that followed the count’s safe return, as well as the remuneration
he paid to the peasant.8
Twelfth-century chronicles form the main group of sources. Some of
them, including the Gesta Francorum, hardly mention children at all,
whereas some writers include several references to the participation of
children. Among these are the works of Guibert of Nogent (not later
than 1108), Albert of Aachen (c. 1118), Fulcher of Chartres (by 1126), and
William of Tyre (until 1184), whose Chronicon – despite its late date – is
the richest source among the genre.
Other sources can be used to complement the inadequacies of historio
graphy. The chansons de geste, believed to have originated from early in
the twelfth century, and later thirteenth-century romances include details
about childhood and youth.9 Their contents are notorious in mixing fic-
tion with contextualised fact, and it is wise to limit their use as a histori-
cal source to the social norms, expectations, and material culture of the
medieval nobility they were written for.
Among the works of the so-called crusade cycles the Chanson d’Antioche
makes perhaps a partial exception. The poem survives in a revision
from the 1180s or later, but it was very probably based on earlier ver-
nacular accounts perhaps originating with the first crusaders themselves.
Although it is impossible to state for certain how much was changed
and added to an earlier version, the extant poem is strongly linked to
the Latin chronicles of the First Crusade. It also is a rare example of the
secular response to the First Crusade otherwise under-represented in
the source material.10
As difficult as fictional texts are for a historian, they cannot be com-
pletely omitted from the study. The world of narrative history, chansons
and epics was readily accessible to noble children and the children of serv-
ants of noble households, forming their idea of the past. From the second
decade of the twelfth century onward, storytelling tradition increasingly
8
Fulcheri Carnotensis Historia Hierosolymitana (1095–1127), III.25, ed. Heinrich
Hagenmeyer (Heidelberg, 1913), pp. 683–86; Lisiardus Turonensis, Historia
hierosolymitana, PL 174, cols 1630–31.
9
Phyllis Gaffney, Constructions of Childhood and Youth in Old French Narrative
(Farnham, 2011), p. 59.
10
The Chanson d’Antioche: An Old French Account of the First Crusade, trans.
Susan B. Edgington and Carol Sweetenham (Farnham, 2011), pp. 34–35,
49–57, 60–61.
258
included descriptions of the main characters’ births, upbringing and

training, an indication, again, that the audience favoured this kind of
information. In addition to the first and second crusade cycles, the
Arthurian cycle and other poetry and romances loosely related to crusad-
ing all have children among their central figures.
Who is a Child?
The obvious difficulty is to decide how medieval writers actually defined
a child. In principle, they divided iuvenis (youth) into three (or four)
phases, including infantia (0–7 years), pueritia (c. 7–12) and adolescen-
tia (c. 14–20). However, in the sources puer can extend to males aged
about 30 and adolescens to persons even older than that. In the plural,
pueri may also include girls.11 The same applies to French enfes/enfant,
bacheler, escuier, damoisele, pucele etc., whose usage expands their modern
meaning.12 Furthermore, the words pater (father) and filius (son), bore
biblical implications. Men with full capacity could be spoken of as filii
to indicate their spiritual subordination to religious authorities. In their
letter to the pope, the leaders of the First Crusade thus called Urban II
(c. 1042–1099) their father and themselves his sons.13 The medieval ter-
minology is ambiguous, because it indicates not only the actual age of
human beings but also their social standing. Accordingly, an adult can
only be someone with legal capacity as well as the ability to sustain himself
financially. This ruled out not only under-age persons, but women (with
few exceptions), and a significant proportion of grown-up males including
serfs and landless household knights.
If puer, adolescens, iuvenis and filius remain terms to be interpreted with
caution, the sources also use words and concepts explicitly related to early
childhood. Infans is such a term for children under six or seven, parvulus
usually being limited to even smaller children. Sometimes puer is defined
with participles clearly indicating breastfeeding, such as lactans or sugens,
in which case we can be certain that the subject is indeed very young. As
a good example for the variety of vocabulary used for babies and small
children, Albert of Aachen mentions pueros sugentes, pueris lactentibus,
11
Shulamith Shahar, Childhood in the Middle Ages (London, 1990), pp. 21–28;
Nicholas Orme, Medieval Children (New Haven, 2001), pp. 6–8.
12
Gaffney, Constructions of Childhood, pp. 25–27.
13
Epistulae et chartae ad historiam primi belli sacri spectantes, ed. Heinrich
Hagenmeyer (Innsbruck, 1901), no. XVI, pp. 161–65.
259
Sini Kangas
tenellis filiis, infantulis, paruulos, fetus, infantes, infantes adhuc sugentes,

sugens masculus (for a male infant), and infans unius anni.14
Children on the First Crusade (1096–99)

Children participated in crusading right from the outset, but unlike those
in the legend of the Children’s Crusade,15 the great majority of crusader
children never had a genuine ideological choice to make. They marched
east because their parents, relatives or guardians decided to, either assum-
ing that the children’s participation would earn them eternal reward, or
because there was no other place to leave them, or, by contrast, because
mutual bonds of affection made it difficult to leave a child behind. Some
children were born during the crusade.16
The First Crusade was a grim experience for children and adolescents,
involving pestilence, injury and other severe hardships. The winter of
1097–1098 was extremely cold, the lack of food escalated into famine,
and infectious diseases reaped a grim harvest among the ranks. When
Anselm of Ribemont wrote back home in November 1097 begging for
prayers for his deceased companions, he mentioned that six out of thirteen
had died of disease – and he was referring to adult males of high social
standing, better equipped for the trying circumstances than a child.17
In contemporary sources for the First Crusade, even members of the
leading military elite are sometimes described as starving and living on
charity and, irrespective of their social standing, children caused further
logistical difficulties. Family status affected physical circumstances on the
march when it came to the availability of food, medication, and modes
14
Albert of Aachen, Historia Ierosolimitana: The History of the Journey to
Jerusalem, ed. Susan B. Edgington (Oxford, 2007), pp. 34, 52, 336, 124, 130, 138,
140, 432, 442.
15
Peter Raedts and Gary Dickson have shown that the Children’s Crusade was
a later fictional interpretation of Alberic of Troisfontaines († c. 1252). It was
not a crusade in either the theological or the legal sense, nor particularly child-
oriented, the majority of participants not being children, and thus the myth
is irrelevant here. Peter Raedts, ‘The Children’s Crusade of 1212’, Journal of
Medieval History 3 (1977), 279–323; Gary Dickson, The Children’s Crusade:
Medieval History, Modern Mythistory (London, 2008), pp. 106–11.
16
Thus far I have not been able to find children who left home independently
to join a crusade. In many cases the situation changed during the march when
members of their families died and children became independent actors.
17
Epistulae et chartae, no. VIII, pp. 144–46.
260
of transportation, but wealth and connections did not protect children

from infectious diseases or the consequences of military attack.
Generally speaking, crusader children, despite their social background,
were in a far more vulnerable position than adults. The number of unsup-
ported children is likely to have been substantial, as many were orphaned
in the atrocities of the campaign. Infant mortality, especially, must have
been high; mothers are reported miscarrying and abandoning their child
ren. Albert of Aachen tells how babies were born to famished and forlorn
mothers on the way, and how these women left their newborn behind by
the route, where others found them either dead or in a pitiful condition.18
What became of these abandoned children? Back home, they would
have been taken care of by relatives or possibly by the Church, but during
a military campaign conventional societal factors facilitating the survival
of the weakest ceased to exist. When deprived of familial support, children
who were old enough continued to follow the army and tried to earn their
living as best as they could by begging for alms or by making themselves
useful to the adults. According to Guibert of Nogent, orphaned children
asked for victuals from the princes, who supported them as well as they
could out of pity and religious duty.19 The sources reveal auxiliary tasks
which women and children performed during military encounters, basi-
cally logistical support such as bringing water, missiles, and other neces-
sities to soldiers, and filling ditches to prepare the way for siege towers
and ladders.20 For instance, during the siege of Jerusalem they helped to
protect siege engines from fire by gathering material for the mantlets.21
Unlike in the modern era, effective use of many medieval weapons
required strength and skill beyond the ability of a child. In twelfth-century
noble German families, the education in arms was only completed during
18
In western cultures abandonment was shameful but not criminal, whereas
attempted infanticide was a crime that was severely punished. Albert of
Aachen, Historia Ierosolimitana, III.2, ed. Edgington, pp. 138–41.
19
20
La Chanson d’Antioche, ed. Duparc-Quioc, 1: 408–9, lines 8296–349; David
Nicolle, Crusader Warfare. Byzantium, Europe and the Struggle for the Holy
Land 1050–1300 AD (London, 2007), p. 13; Walter Porges, ‘The Clergy, the
Poor and the Non-Combatants on the First Crusade’, Speculum 21 (1946),
1–23 (here pp. 1–4, 9–12); Randall Rogers, Latin Siege Warfare in the Twelfth
Century (Oxford, 1992), pp. 8, 61. In twelfth-century chansons, children
and young people run errands, prepare horses for battle, serve knights and
accompany them: Gaffney, Constructions of Childhood, p. 62.
21
261
Sini Kangas
the late teens or early twenties,22 and this seems to have been the case also
in other areas. This means that, unlike modern child soldiers, the major-
ity of crusader children were not effective in combat, which limited the
functions that they could perform in the army. In general, ancient and
medieval weaponry was more effective when wielded by stronger people.
Crusader military actions – like warfare of all times and regardless of
time and space – involved children, but minors’ participation is depicted
in a different light from that of modern western treatises. In the High
Middle Ages, the presence of children on the battlefield was understood
to be normal, even if not an ideal situation. In descriptions of the great
crusader battles, their wide fields were bathed in the blood of crusaders
of all ages amid scenes of wholesale slaughter.23 Many children got lost
or died during the battles:
The Turks, rejoicing […] slaughtered the wretched band of pilgrims […]
Going into the pilgrim camp they found those who were there, the feeble
and crippled, clerics, monks, aged women, boys at the breast, and put
them all to the sword, regardless of age. They took away only young girls
and nuns, whose faces and figures seemed to be pleasing to their eyes, and
beardless and attractive young men.24
Children were among those who suffered when there was fighting, a group
routinely including the old and sick, women and maidens, children and
babies,25 even if depictions of deliberate cruelty inflicted on children by
the crusaders are rarer.26 The exception is the capture of Jerusalem, during
which the sources stress the totality of slaughter, listing deliberate killing,
22
Jonathan R. Lyon, ‘Fathers and Sons: Preparing Noble Youths to be Lords in
Twelfth-Century Germany’, Journal of Medieval History 34 (2008), 291–310.
23
Until the latter part of the fourteenth century there was no commonly approved
idea of the immunity of non-combatants. Christopher Allmand, ‘War and the
Non-Combatant in the Middle Ages’, in Medieval Warfare: A History, ed.
Maurice Keen (Oxford, 1999), pp. 253–72.
24
‘Turci itaque gaudentes […] detruncabant miseram manum peregrinorum
[…] Tentoria uero illorum intrantes quosquos repererunt languidos ac debiles,
clericos, monachos, mulieres grandeuas, pueros sugentes, omnem uero etatem
gladio extinxerunt. Solummodo puellas teneras et moniales, quarum facies et
forma oculis eorum placere uidebatur, iuuenesque inberbes et uultu uenustos
abduxerunt’: Albert of Aachen, Historia Ierosolimitana, I.21, ed. Edgington, pp.
42–43; see also pp. 52, 130, 318, 336.
25
William of Tyre, Chronicon, ed. Robert B. C. Huygens, 2 vols (Turnhout,
1986), 16.16, p. 737.
26
For example, Ralph of Caen mentions crusaders roasting Saracen pueri (boys)
on skewers. Ralph of Caen, ‘Gesta Tancredi’, RHC Occ. 3: 675; The Gesta
262
both with bare hands and with weapons, of Saracen babies and infants
who were found in the city with their mothers.27
In the aftermath of raiding or battles small children usually remained
with their mothers, whether in death, captivity or slavery.28 The Anonymous
mentions that the Turks used to loot the churches and houses and take
Christian children into captivity.29 The sources mention Khorasan as the
place where Christian captives were taken, but as the whole concept of
Khorasan, the Saracen homeland, was quasi-mythical among medieval
westerners, it is far from clear what place was actually meant. Captured
crusaders and Muslims unable to pay a ransom were sold into slavery in
the markets of Antioch, Aleppo, Damascus and other cities, after which
we do not hear of them. Raymond of Aguilers paints a pessimistic picture
of the destiny of the Christian children forcibly seized from their mothers
to be circumcised and converted to Islam or placed in brothels,30 and it is
likely that among the captives there were young children who were raised
as Muslims, but if so, there is little evidence of them in Muslim sources.
Bringing up an Exemplary Crusader

To become attached to his or her community, the child is required to adopt
its collective beliefs, norms, myths and rituals, and most children became
familiar with the concept through their interaction with members of their
family and local community. In the medieval West positive indications of
religious conviction included crusading and related practices. Thus some
children grew up in families deeply committed to the crusading ideology
and participation in crusading began at home, a long time before tak-
ing up arms and the pilgrim’s staff. Influential crusader families formed
networks connected by marriage, landowning, and military campaigns
Tancredi of Ralph of Caen, trans. Bernard S. Bachrach and David S. Bachrach

(Aldershot, 2005), p. 116.
27
Albert of Aachen, Historia Ierosolimitana, VI.23, 30, ed. Edgington, pp. 432–33,
442–43; Fulcher of Chartres, Historia Hierosolymitana, ed. Hagenmeyer, I.27,
p. 301.
28
Albert of Aachen, Historia Ierosolimitana, VII.40, ed. Edgington, pp. 546–47;
William of Tyre, Chronicon, 13.20, ed. Huygens, p. 612.
29
30
Raymond of Aguilers, ‘Historia Francorum qui ceperunt Iherusalem’, in RHC
Occ. 3: 288; trans. John H. Hill and Laurita L. Hill (Philadelphia, 1968), p. 109.
Guibert of Nogent similarly mentions Christian girls being prostituted by
Saracens: Dei gesta, ed. Huygens, p. 102.
263
Sini Kangas
both in the Latin East and the West. Among the upper nobility, relations
between east and west remained close through continuous intermarriage
and various forms of mutual support.31
From the beginning, papal preaching referred to crusading as a
longstanding privilege of true Christians chosen by God from genera-
tion to generation. While preaching the First Crusade at the council of
Clermont 1095, Pope Urban II placed the listeners at the end of a long
line of Christian champions who were descended from the Maccabees,
the founders of Christian kingdoms and from Charlemagne, and whose
campaign represented the concluding phase of universal warfare before
the last battle and the second coming of Christ.
A generation and a half later, similar attitudes prevailed. In Pope
Eugene’s crusade bull Quantum praedecessores from 1145, the honour of
predecessors became the most important reason to take the cross, and the
recipients were seen as part of a continuous line of respectable defenders
of holy Christendom. The blood spilt by their ancestors was mentioned
repeatedly.32 At this point, we find prominent western noble houses within
which crusading has been established as a family tradition, as well as a
source of prestige and pride.
Parents were primarily responsible for bringing up their children
as morally upright members of the religious society. William of Tyre
describes a child as a flexible creature made of wax, a metaphor popular in
his time.33 The wisdom behind the image was that children were morally
weak and could not yet tell right from wrong. It was the obligation of the
parents to mould him into shape, so that he would learn to be pious and
honest, as well as adapting to models of behaviour proper for their rank.
If the parents neglected their duty, the youth would be badly formed and
harden irredeemably into a creature disgusting to God and men.
According to the medieval idea, mothers were responsible for the
primary care of infants and young children. Serious training could begin
only later, with the exception of basic religious education, prayers and the
basics of the Christian faith, which belonged to maternal duties. Ralph
of Caen tells us that Godfrey of Bouillon’s warlike abilities came from his
father, whereas his cultivation of God was because of the example given
31
This is especially the case with the princely houses of the Latin East, the
offspring of Baldwin of Bourcq (Baldwin II, d. 1131), Bohemund of Taranto
(c. 1058–1111), and Raymond of Toulouse (1052–1105), the founders of the
dynasties of Jerusalem, Antioch and Tripoli.
32
‘Quantum praedecessores’, PL 180, cols 1064–66.
33
William of Tyre, Chronicon, 16.3, ed. Huygens, p. 717; Shahar, Childhood in the
Middle Ages, pp. 88–89.
264
by his mother.34 In Guibert of Nogent’s case the path of his whole life was
set before his birth by the religiosity of his parents, especially the mother.35
The seventh year was the traditional point for noble boys to turn from
female care to tuition by males, whereas girls continued to be taught by
women.36 In knightly families, fathers trained their sons to become lords
with the help of masters and tutors. In crusader families sons were likely
to be taught in military skills by crusaders, and to socialise with people
from crusader backgrounds. Young sons are reported to have accompanied
their fathers when they were making property agreements, visiting courts
of other magnates, and conducting various other family businesses. This
was an essential part of the education of the child to take his father’s place
successfully one day, should the occasion arise.37
The sources give disproportionate space to affectionate relations
between fathers and eldest sons, but fatherly care was not limited to the
firstborn male child. Younger sons had to be able to replace their elder
brothers in case they died. If they did not share the family’s hopes for them
or if they lacked the proper education to live up to these hopes, problems
would arise, as in case of Louis VII (1120–1180), whose path to kingship
was long and rocky after having spent his tender years among the monks
of Saint-Denis. Female heirs were also important to their parents. The
Chanson d’Antioche mentions noble girls joining the crusade with their
fathers, en vont qui les ont engenrees (‘who had given them life’), and also
describes how daughters remaining at home mourned alongside their
mothers the loss of their fathers.38
The chronicles include several descriptions of fathers losing an heir.
Gosson’s heartbroken father kisses and embraces his dead (adult) son and
laments that he will never enter the Holy Sepulchre. In a conventional
phrase related to the death of a hero, one thousand knights weep with him
to show respect.39 Fatherly feelings transgress the religious boundaries in
34
Ralph of Caen, ‘Gesta Tancredi’, RHC Occ. 3: 615; The Gesta Tancredi of
Ralph of Caen, p. 37.
35
Guibert de Nogent, Autobiographie, ed. Edmond Réne Labande (Paris, 1981),
I.3; A Monk’s Confession: The Memoirs of Guibert of Nogent, trans. Paul J.
Archambault, 3rd edn (University Park, PA, 2004), pp. 9–10.
36
Nicholas Orme, From Childhood to Chivalry: The Education of the English
Kings and Aristocracy 1066–1530 (London, 1984), p. 7.
37
Lyon, ‘Fathers and Sons’, pp. 291–310.
38
La Chanson d’Antioche, ed. Duparc-Quioc, 1: 55, lines 845–46; The Chanson
d’Antioche: An Old French Account, trans. Edgington and Sweetenham,
p. 124.
39
La Chanson d’Antioche, ed. Duparc-Quioc, 1: 145–46, lines 2527–40.
265
Sini Kangas
the Chanson d’Antioche; Garsion and Soliman mourn for their slain sons
as bitterly as Gosson.40
Mothers are far less present in the chronicles and chansons, but should
the father die prematurely, a mother’s duty was to take over his role of
ensuring the children’s rights and abilities to inherit. Like fathers, mothers
require respectful behaviour from both underage and grown-up children.
In the Gesta Francorum and the Chanson d’Antioche Corbaran (Karbugha)
behaves exceptionally by disobeying his mother’s wise advice to avoid
attacking the Franks. In the Gesta’s version Corbaran treats her with
care and respect as a dutiful son, but cannot refuse battle honoris causa,41
whereas in the Chanson d’Antioche he snaps at his mother: ‘Laisiés vos
sermoner. Toute estes redotee, L’en vos devroit tuer’ (‘Stop preaching.
Everyone is terrified, you [just] have to kill [the Franks]’)42 and leaves her
rashly to die because of his foolishness. If the roles of mother and wife
conflicted, it was appropriate for the wife to take precedence. For instance,
when Lady Mamilia of Roucy, the wife of Hugh II du Puiset, gave birth
to their son in Apulia in 1107, she left the baby in the care of Bohemund
of Taranto, a kinsman, and continued her journey with her husband.43
In addition to parents and relatives, networks of peers were important
for the later career of noble boys. Wardship was traditional among noble
Normans: the children left home, usually in their early teens, to be brought
up in their lord’s court.44 It was also fairly common practice to send boys
to ecclesiastical households to be educated.45 Since the youths were raised
in small groups, they spent a lot of time together, often forming strong
bonds of friendship and support.46
Children preferred each other’s company, as always, but most of them
also spent a lot more time with adults than would later be the norm. It
was not usual for children to be educated in age groups, although it may
to some extent have applied to aristocratic boys being trained in military
40
La Chanson d’Antioche, ed. Duparc-Quioc, 1: 126–28, lines 2213–51; 1: 154–55,
lines 2702–19.
41
Gesta Francorum, ed. and trans. Hill, pp. 53–56; La Chanson d’Antioche, ed.
Duparc-Quioc, 1: 268–69, lines 5252–68; 1: 274–75, lines 5376–98.
42
La Chanson d’Antioche, ed. Duparc-Quioc, 1: 275, lines 5394–95, trans. Kangas.
43
44
Orme, From Childhood to Chivalry, pp. 46, 50.
45
In this case their military training was taken care of by masters: Orme, From
Childhood to Chivalry, p. 56.
46
Matthew Bennett, ‘Military Masculinity in England and Northern France,
c. 1050–1225’, in Masculinity in Medieval Europe, ed. Dawn M. Hadley
(London, 1999), pp. 71–88; Orme, From Childhood to Chivalry, p. 29.
266
skills away from home. Other children spent their days with adults and
elder siblings. Away from the nobility, infants would often be taken to
work with their mothers or nurses, and apprentices and children educated
at home spent the majority of daytime with their masters and parents.
Spare time activities were similarly shared. Medieval children and adults
played the same games, listened together to the same stories, sang the
same songs, and went to church with their families. Even toys were not
exclusively restricted to children, as grown-ups might play with them
as well. Orme concludes a discussion of children’s culture in medieval
England: ‘How distinctive then was [it]? There were few striking dif-
ferences of content from the culture of adult life.’47 The same applies to
crusade-related religious and cultural traditions.
The chronicles are disappointingly sparse when it comes to the syl-
labus of twelfth-century crusaders or of their children born in the Latin
East.48 William of Tyre states that Baldwin III was fairly well educated,
far more so than his younger brother Amalric, but does not include any
details of Baldwin’s education. What we learn is that the king liked both
reading to himself and listening to others read aloud, and that he was
particularly fond of history, that is, information about past local events as
well as broader depictions of war, peace and the heroic exploits of other
princes. Baldwin was also very well acquainted with the customary law
of the Latin East, and enjoyed conversations with the men of letters, as
well as any acquaintances who might possess interesting information
about matters of the realm.49 Since Amalric seems to have been similarly
knowledgeable about customary law and history, as well as interested in
reading about the affairs in his kingdom in his spare time,50 it is probable
that the royal princes had been tutored together at least some of the time,
and that their syllabus had included studies in history, secular law, read-
ing and conversational skills. William of Tyre was himself paid tutor to
the future king Baldwin IV,51 but again he does not reveal any more than
that he had been given instructions to train the prince, then about nine
years old, in liberal studies and knowledge of letters (reading and writing
Latin), and that the prince was allowed to participate in rough games
with other boys under his supervision. Of the prince’s mother Sibylla
47
Nicholas Orme, ‘The Culture of Children in Medieval England’, Past and
Present 148 (1995), 48–88 (here p. 86).
48
Most of the information comes from William of Tyre, and it explicitly concerns
royal children.
49
William of Tyre, Chronicon, 16.1–2, ed. Huygens, pp. 714–16.
50
William of Tyre, Chronicon, 19.2, ed. Huygens, pp. 864–65.
51
267
Sini Kangas
we learn that she had been raised by her Aunt Iveta at the Convent of St
Lazarus in Bethany.52
What was the result of the noble upbringing likely to be? The protago-
nists of the chansons are described as impulsive, fearless, hot-blooded and
passionate, as well as self-confident and arrogant, unrestrained and lacking
in reason in their youth.53 William, brother of Tancred, sounds exactly
this kind of young man, as indeed does the young Tancred himself.54 The
Chanson d’Antioche mentions that William was still very young and had
only recently been knighted when taking up the cross. He was very keen
to show his valour in battle, but though brave he was impatient to the
point of recklessness, and completely unable to assess the potential risks
of his actions.55 Whether due to these characteristics or not, William died
during the First Crusade.
A model example of noble youth is without doubt Baldwin III, pre
eminent among others in beauty and form. He was an attentive listener,
had a good memory, and learned quickly. William of Tyre adds to this list
brilliancy of speech, honesty and a good heart, lack of greed and a sincere
fear for God. Unlike Tancred’s brother William, Baldwin was also patient.56
The Church, Children and Crusading

Children in the crusading societies of the West were raised in an atmos-
phere of strong support for religious warfare. Their socialisation with
crusading ideals was boosted by collective religious acts including masses,
liturgy, prayers, processions, church architecture and art, as well as familial
networks – many children had in the family someone who had been on
pilgrimage, intended to do so, or who had known someone involved in
the tradition. They were also familiar with the very rich storytelling tradi-
tion celebrating the champions of the holy war. According to Guibert of
Nogent, it was not necessary for any ecclesiastical person to make ora-
tions in churches to stir up the people, since one man told another, both
by word and by example, at home and elsewhere. A great rumour spread
into every part of France, and whoever heard the papal call urged his
52
53
Gaffney, Constructions of Childhood, pp. 38–39, 73.
54
For example Ralph of Caen, ‘Gesta Tancredi’, in RHC Occ. 3: 607; Robert the
Monk, ‘Historia Iherosolimitana’, in RHC Occ. 3: 746.
55
La Chanson d’Antioche, ed. Duparc-Quioc, 1: 117–18, lines 2068–80.
56
William of Tyre, Chronicon, 16.1–2, ed. Huygens, pp. 714–16.
268
neighbours and family about undertaking the Way of God.57 In medieval

castles, towns and villages, people lived in close proximity and local news
travelled rapidly.
The increasing importance of pilgrimage to Jerusalem from the eleventh
century occurred throughout western Christendom. The city was seen
as the centre of the world; no other shrine was as important. Crusaders
founded churches and other ecclesiastical buildings and donated money
towards the purchase of religious items viewable by rich and poor alike.
Sometimes the monuments were built as replicas of the Holy Sepulchre
(in England, Spain, Lombardy, France, Germany), or housed a miniature
or model. Returning pilgrims also brought relics with them: pieces of
the cross, holy oil, water from the river Jordan, dust from the Holy City,
etc.58 When they passed by, everybody could see the symbol of the cross
on their garments and banners.
It is important to understand that in the twelfth and thirteenth cen-
turies the ideology of the crusades was accessible irrespective of level of
literacy. Amnon Linder found that after the loss of Jerusalem in 1187 there
were at least five types of Holy Land liturgies circulating all over Europe:
the Holy Land Clamour (either appended to office services or inserted
in the Mass), the Holy Land Mass (extant and new triple sets of Mass
prayers), the dedicated War Mass, and Holy Land segments in bidding
prayers. The spoken word and crusade-related music – liturgy, psalms and
hymns, prayers and metaphors as a part of a mass – were complemented
by shared physical actions with fellow worshippers such as prostration,
genuflection, and the kiss of peace, and by the physical surroundings, the
sounds, sights and smells in the church.59 All this was available to children
whatever their social standing.
The importance of ceremonial liturgy was that it brought sacred history
alive in an illiterate society, and enhanced and maintained the collec-
tive memory. Though written down in Latin, sermons for the laity were
57
‘Non erat aecclesiasticae cuiquam personae necessarium ut ad excitandos pro
hoc ipso populos in aecclesiis declamaret, cum alter alteri non minus monitis
quam exemplo domi forisque profectionis vota clamaret […] Magnus per
universas Franciae partes rumor emanate et quisque, ad que primo pontificis
preceptum prevolans fama detulerat, de proponenda “via dei” – sic enim
antonomasice vocabatur – contiguos sibi ac familiares quosque sollicitat.’
Guibert of Nogent, Dei gesta, ed. Huygens, pp. 88, 117–18.
58
Colin Morris, The Sepulchre of Christ and Medieval West (Oxford, 2005), pp.
202, 220–23, 230–42.
59
Amnon Linder, Raising Arms: Liturgy in the Struggle to Liberate Jerusalem in
the Late Middle Ages (Turnhout, 2003), pp. 364–65.
269
Sini Kangas
preached and prayers were held in the vernacular,60 so that everybody

would be able to understand what was going on. The common people
were active propagators, willing and able to contribute to the develop-
ment of the tradition.
Crusade liturgy evolved as a mixture of pilgrimage liturgy, that had
existed in Europe since the early Middle Ages, and religious traditions
connected to warfare, such as blessing the arms, mass and confession before
battle, common prayers for military expeditions, and the papal banner.61
Collective religious practices supporting crusading included the rite of
taking the cross in public, visits and prayers at shrines before and after the
crusade, daily masses and prayers on the road, public acts of penance, fasts,
processions, almsgiving, and tolling of the church bells, both preceding
the expedition and during it. Special departure rituals were taking place
by the thirteenth century, including confirmation of vows, blessing of the
pilgrim staff and wallet, and the singing and chanting of songs, hymns
and prayers for a safe return.62
I have not found any specific descriptions in twelfth-century material
of children’s reception of the preaching of the crusades or stories circulat-
ing around the subject. The chronicles stress the unanimity of popular
reaction to crusade propaganda, and the ideal was that everybody would
participate in crusading either physically, financially or spiritually. The
aim, developed further by Pope Innocent III (1198–1216), was to oblige
everyone to become involved by paying taxes and giving donations and
participating in penance and intercessory exercises, thereby extending
the responsibility for the campaign’s outcome to the whole of society.63
Urban II was the first pope to initiate and organise an extensive preaching
60
Carolyn Muessig, ‘Sermon, Preacher and Society in the Middle Ages’, Journal
of Medieval History 28 (2002), 73–91 (here p. 79); see also Roberto Rusconi,
‘La predicazione. Parole in chiesa, parole in piazza’, in Lo spazio letterario del
Medioevo. I. Il Medioevo latino, ed. Guglielmo Cavallo, Claudio Leonardi and
Enrico Menestò (Salerno, 1992), pp. 571–603.
61
Christoph T. Maier, ‘Crisis, Liturgy and the Crusade in the Twelfth and
Thirteenth Centuries’, Journal of Ecclesiastical History 48 (1997), 628–57.
62
Anne E. Lester, ‘A Shared Imitation: Cistercian Convents and Crusader
Families in Thirteenth-Century Champagne’, Journal of Medieval History 35
(2009), 353–70 (here p. 365); William Chester Jordan, ‘The Rituals of War:
Departure for Crusade in Thirteenth-Century France’, in The Book of Kings:
Art, War and the Morgan Library’s Medieval Picture Bible, ed. William Noel
and Daniel H. Weiss (London, 2002), pp. 99–105; Jonathan Riley-Smith, The
First Crusaders and the Idea of Crusading, 3rd edn (London, 2009), pp. 139–43.
63
Maier, ‘Crisis, Liturgy and the Crusade’, p. 656.
270
tour, the sole purpose of which was to provide support for a forthcom-
ing crusade. Thereafter, regularly before major expeditions, the church
authorised preachers and legates to propagate crusading ideology.
New religious orders, military brotherhoods and the Cistercian order
shaped western society’s positive perception of crusading from the sec-
ond quarter of the twelfth century onward. The connection between
crusader families and religious and military orders developed intensely
over time. Larger groups related by kin and friendship, consisting typically
of knightly iuvenes, entered Cistercian monasteries together.64 Convents
also provided a safe repository for the female members, daughters, sisters
and wives, when their male relatives travelled east.65
Conclusions
In terms of the medieval West, crusading was present in the daily life of
children irrespective of their social status. Crusading ideology was always
strongly based on the popular support of the laity and it represented a
shared stock of common values appealing to rich and poor alike. As impor-
tant as ecclesiastically authorised preaching and history-writing were the
stories and memories circulated in the local community, the heart of which
was the parish church. Collective participation is the key to understand-
ing why and by what means western societies remained keenly involved
in crusading through the twelfth and well into the thirteenth century.
At church children came into contact with crusading in various ways:
together with their parents, they listened at masses, sermons, litanies
and prayers propagating crusading. They were brought up to think that
by donating money for a good and pious purpose, they would not only
contribute to the common Christian cause but to their personal spiritual
health and eventual redemption as well. Some of them later took monastic
vows to support relatives leaving for the Holy Land, and some of them
ended up as clerical chroniclers boosting the ideology with their learning.
Religious education was an important part of the process of committing
forthcoming crusaders to the cause.
Many children participated in the marches east with their families and,
if the family around them ceased to exist, on their own. The information in
the sources is scattered, and in the majority of cases we do not learn who
64
Jochen Schenk, Templar Families: Landowning Families and the Order of the
Temple in France, c. 1120–1307 (Cambridge, 2012), pp. 253, 259.
65
The problem here is that the charters do not list the ages of the females who
entered convents. Some of them were certainly children.
271
Sini Kangas
these children were and what became of them. Even noble children from
the highest social stratum like the younger son of the Count of Toulouse,
who participated in the First Crusade with his parents66 and died in the
East, may not be mentioned by name. Most of the time authors do not
treat children as a group separate from adults. Military actions evidently
involved children. In the Middle Ages, the appearance of children on the
battlefield was understood to be a normal practice, even if not an ideal
one. Even if the adults took pity on children, they are not described as a
distinct group under immunity.67
Crusader children imitated the adults around them, and shared their
fates. Early involvement in family networks, feuds and alliances, and the
specific religious and cultural traditions practiced by the family circle all
prepared them for adulthood. According to the sources, in most cases
children participated willingly. In the medieval west, crusading often
developed into a family tradition transmitted from generation to
generation.
66
67
272
IV
The Afterlife of the Crusade

Rewriting the Past
The Conquest of the Holy Sepulchre
in the Memory of Italian Communal Cities*
Elena Bellomo
Cardiff University – University of Verona
Italy is centrally located in the Mediterranean and has always served as

a natural bridge linking east and west, the Mediterranean area and con-
tinental Europe. Over the centuries the Italian peninsula and its islands
performed a delicate and essential role of connection and mediation
between different cultures. In the middle ages this crucial role endowed
Italy with a primary importance in international events and organisations
which connected both sides of the Mediterranean, such as the crusades
and the Military Orders. However, the participation of Italian communal
cities in the expedition to conquer Jerusalem in 1099 was much less influ-
ential than their contribution to later crusader contingents. The political
fragmentation of central and northern Italy and the internal conflicts
which preceded the creation of communes made it impossible for central
and northern Italian cities to orchestrate major expeditions. Thus refer-
ences to Italian forces originating from these areas in the chronicles of
the First Crusade are sporadic and limited. However, the conquest of the
Holy Sepulchre was an event of unique importance to Latin Christians
and the First Crusade came to play a role in the public memory and the
identity of several communal towns of central and northern Italy.1 This
chapter aims at outlining how the First Crusade was perceived in some
*
I would like to thank Susan Edgington, Luis García-Gujarro Ramos, Alexia
Grossjean, and Valeria Polonio for their constructive comments. I thank
Martin Hall and Jonathan Phillips who kindly provided me with their English
translation of Caffaro’s crusading chronicles. Jonathan Phillips has also kindly
sent me the text of his forthcoming article on the Genoese annalist Caffaro.
1
On the role of the crusades in the culture of Italian cities, see Franco Cardini,
‘Crociata e religione civica nell’Italia medievale’, in La religion civique à l’époque
médiévale et moderne (Chrétienté et Islam), ed. André Vauchez (Rome, 1995),
pp. 155–64 (here pp. 159–64).
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. –.
Elena Bellomo
of these cities between the twelfth and fifteenth centuries, focusing only
on some emblematic examples which can offer original insights into the
complex relationship between the crusade and communal Italy. They offer
the opportunity to investigate, in a comparative approach, civic environ-
ments which had diverse links with the east and had varying involvement
in crusading expeditions. The long period covered by this investigation
also allows for an evaluation of the evolution, meaning, and use of the
memory of the First Crusade from the birth of the communal government
to the transition to the Signoria.
The Conquest of Jerusalem: a Marginal Memory?

The medium for the intense and fruitful contacts between Italy and
the Middle East was the Mediterranean, a sea which had already been
crossed before the crusades by experienced Italian sailors acutely aware
of its dangers as well as its promises. Thus it is not surprising that Italian
maritime cities such as Genoa, Pisa and Venice played a central part in a
Mediterranean enterprise such as the First Crusade. Genoa was the first
Italian maritime city to join the crusader army bound for Jerusalem and
was involved in this undertaking by Pope Urban II himself who sent two
high prelates to preach the cross in the Ligurian port. Two Genoese naval
contingents supported the Christian fighters in their conquests of Antioch
and Jerusalem respectively. According to local tradition, on the way back
from Antioch the Genoese claimed to have found the relics of St John
the Baptist in Myra.2 The second Genoese contingent was composed of
only two galleys commanded by the Embriaci brothers who, dismantling
their ships, provided the crusaders with the timber needed to build siege
machines. After the battle of Ascalon the Embriaci left Palestine and
brought to Genoa the exciting news of the conquest of the Holy City.3
While these first expeditions were private in nature and left a town deeply
divided by civil and religious factionalism, the third crusader fleet which
sailed from Genoa was armed by the newly established communal gov-
ernment: the Compagna comunis. A young Genoese crusader embarked
2
Caffaro, ‘De liberatione civitatum Orientis Liber’, in Annali genovesi di Caffaro e
de’ suoi continuatori, ed. Luigi Tommaso Belgrano, 5 vols (Roma, 1890–1919), 1:
101–9; Caffaro, Genoa and the Twelfth-Century Crusades, trans. Martin Hall and
Jonathan Phillips (Farnham, 2013), pp. 109–15; Iacopo da Varagine, ‘Legenda
translationis beatissimi Joannis Baptistae Genuam’, in RHC Occ. 5: 229–35.
3
Caffaro, ‘De liberatione’, 1: 110–11; Caffaro, Genoa and the Twelfth-Century
Crusades, pp. 115–17.
276
The Conquest of the Holy Sepulchre in the Memory of Italian Communal Cities
with this fleet. His name was Caffaro and he decided to write a history of
his city. Caffaro’s initiative was a revolutionary one: his chronicle is the
first town chronicle written in the middle ages and the first annals ever
composed by a layman. The work begins with a detailed account of the
crusading expedition in which the author participated. In 1152 Caffaro’s
annals became the official history of the commune of Genoa, a chronicle
which was unprecedented in merging historical narrative, crusading and
patriotic ideals, and notarial authenticity.4
Genoa was not new to anti-Muslim warfare. It had already participated
in maritime expeditions against Muslim African and Spanish ports,
namely Mahdia (1087) and Tortosa (1093).5 However, the real turning
point for the city’s Mediterranean adventures was the crusade. That we
are today aware of the crucial importance of Genoese involvement in the
Latin East is directly attributable to the testimony of Caffaro. Moreover,
the annalist wrote a short chronicle on the Genoese crusading expedi-
tion which conquered the Muslim cities of Almeria and Tortosa in Spain
(1146–48) and a history of the First Crusade and the first decade in the
life of the kingdom of Jerusalem.6 Thus crusading played a prominent
role in Caffaro’s historical works. The close link between the author, the
Latin East and Genoese anti-Muslim warfare was further supported by the
belief that he returned to the Holy Land during the first half of the twelfth
century. Caffaro also led the Genoese fleet which attacked Minorca and
Almeria in 1146 and went to Tortosa in 1149 to sign the carta divisionis
of the newly conquered town.7
Given the crucial importance of the 1100–1101 Genoese expedition
to both the Genoese public and Caffaro’s life it is not surprising that
4
Caffaro, ‘Annales’, in Annali genovesi, 1: 3–75; Giovanna Petti Balbi, Caffaro e la
cronachistica genovese (Genova, 1982); Antonio Placanica, ‘L’opera storiografica
di Caffaro’, Studi Medievali, ser.3, 36 (1995), 1–62; Elena Bellomo, A servizio di
Dio e del Santo Sepolcro. Caffaro e l’Oriente Latino (Padova, 2003).
5
Giuseppe Scalia, ‘Il carme pisano sull’impresa contro i saraceni del 1087’, in
Studi di filologia romanza offerti a S. Pellegrini (Padova, 1971), pp. 565–625; H.
E. John Cowdrey, ‘The Mahdia Campaign of 1087’, English Historical Review
92 (1977), 1–29; Giovanna Petti Balbi, ‘Lotte antisaracene e ‘militia Christi’ in
ambito iberico’, in ‘Militia Christi’ e Crociata nei secoli XI-XII (Milano, 1992),
pp. 519–45 (here p. 537).
6
Caffaro, ‘Ystoria captonis Almarie et Turtuose’, in Annali genovesi, 1: 79–89;
Caffaro, ‘De liberatione’, pp. 98–124.
7
Diplomatari de la Catedral de Tortosa (1062–1193), ed. Antoni Rovira i Virgili
(Barcelona, 1997), no 18.
277
Elena Bellomo
he decided to inaugurate his annals with that campaign.8 The author’s

crusading experience took place after the success of the First Crusade
and thus in his annals he decided to focus on later expeditions. In this
work the only mention of the conquest of Jerusalem is a short reference
in a list of Genoese anti-Muslim expeditions which includes the fleets
that attacked Mahdia, Tortosa, and Antioch. Several decades after he had
begun to write the annals, Caffaro decided to compose a second chroni-
cle exclusively focusing on the Latin East, the De liberatione civitatum
Orientis liber or Liberatio. It was probably intended to defend Genoese
trade rights contested overseas as well as to offer an example of courage,
union and sacrifice to Caffaro’s fellow-citizens involved in deep factional-
ism at that time. This chronicle was not included in the official history
of the Compagna which already comprised the aforementioned Caffaro
chronicles. It was only accidentally found at the end of the thirteenth
century by the Genoese annalist Iacopo Doria. The text of the Liberatio
found by Iacopo, extant in a single codex, was a copy of the original draft
and the chronicle did not attain a definitive form.9 Therefore it is hard to
assess how Caffaro actually wanted to portray the Genoese contribution
to the First Crusade in his last work. Despite this, the Genoese perspec-
tive and the use of poetic and narrative sources make the Liberatio a very
interesting and challenging source. I have extensively discussed elsewhere
Caffaro’s description of the sieges of Antioch and Jerusalem in this
work by examining the writing stages, primary sources and reliability
of these passages. Here I approach the subject from a different perspec-
tive. Caffaro’s account of the siege of Antioch mixes limited Genoese
information, probably orally transmitted, and information drawn from
poetic and narrative sources dealing with the First Crusade. Thanks to the
use of primary sources not originating from the Genoese environment,
the annalist penned a vivid portrayal of the Antiochene campaign.10 In
contrast, the siege of Jerusalem, the crusaders’ final victory, and the con-
tribution made by the Genoese to this success are very briefly mentioned
in the Liberatio, and Caffaro’s reconstruction is almost exclusively drawn
from Genoese information.11 Unfortunately the incompleteness of the
Liberatio makes it impossible to ascertain whether Caffaro intended this
section of the chronicle to be revised and completed by inserting further
information from other primary sources.
8
Caffaro, ‘Annales’, pp. 3–13; Caffaro, Genoa and the Twelfth-Century Crusades,
pp. 49–56.
9
Bellomo, A servizio di Dio, pp. 39–156.
10
Bellomo, A servizio di Dio, pp. 62–78, 94–101.
11
278
By virtue of the presence of its combatants at Antioch and Jerusalem,

Genoa gained an undisputed preeminence among the other Italian
maritime cities. However, this accomplishment is neither claimed nor
emphasised in Caffaro’s works. In fact, the author’s main aim was to
underline the success of the newly established political form of the com-
mune. Thus, the Genoese anti-Muslim efforts which preceded the first
expedition officially sponsored by the Compagna are only marginally
mentioned in the annals. Moreover, personal observation is of primary
importance to Caffaro’s historical perspective and he pays much more
attention to events in which he had been involved and with which he had
direct acquaintance. Caffaro also focuses on the privileges granted to the
Genoese in the east. The local contribution to the capture of Jerusalem is
mentioned in the text of the inscription praising the Genoese crusaders in
the Holy Sepulchre.12 However, at the time the Embriaci did not receive
any special concessions. Finally, not only were the first two Genoese
crusader contingents private expeditions, but they left a city afflicted
by an acute conflict which also led to the interruption of the election of
the consuls. These tensions originated from the fear that involvement
in the crusade could compromise the existing economic ties with Egypt
as well as from acute religious conflict, which jeopardised the establish-
ment of the new communal form of government. In his annals Caffaro
highlighted the positive effect that the crusade eventually had in settling
these internal conflicts.13 However, the presence of such acute friction
when the Embriaci left Genoa very probably influenced the annalist’s
understanding of their success.14
Not only was Caffaro the first official annalist of Genoa but he was
also elected consul of the commune and consul of justice several times.
He led the Genoese fleets in battle and was a skilled ambassador. His
consulates coincided with crucial periods for the fortunes of Genoa and
the development of its institutions.15 It is worth noting that Caffaro’s
approach to the crusade matches the propagandistic vision of anti-Muslim
warfare developed by the Genoese authorities in this period. Genoa did
not have a prestigious past to uphold. Its international fortune started
with the crusade and the crusade provided the frame and foundation for
the image of power and distinction that the Genoese ruling class wanted
12
I Libri iurium della Repubblica di Genova, 1/1, ed. Antonella Rovere (Roma,
1992), no. 59.
13
Caffaro, ‘De liberatione’, p. 111; Caffaro, Genoa and the Twelfth-Century
Crusades, p. 117.
14
15
279
Elena Bellomo
to convey at international level.16 However, in this ambitious plan such

a prestigious achievement as contributing the conquest of Jerusalem
was ignored: the ambassadors (Caffaro was among them) who talked to
Barbarossa in Roncaglia in 1158 pragmatically claimed that Genoa had
effectively defended the European coasts from Rome to Barcelona, while
the inscription on Porta Soprana in Genoa focused on the Spanish suc-
cesses of Almeria and Tortosa with only a general reference to victories
in Asia.17 Again the fact that only a limited and unofficial contingent had
joined the crusaders in 1099 seems to be a crucial factor and to have made
other expeditions, jointly planned and conducted with the patronage of
the Genoese commune, more significant for both propaganda and public
memory. Given Caffaro’s significant role in Genoese political activities of
this period, it is permissible to wonder whether he influenced the attitude
of the Genoese authorities to the First Crusade.
It is now worth investigating whether the Genoese contribution to the
conquest of the Holy City left a specific mark on ecclesiastical records
in Genoa. Despite heavy losses of material which already occurred in
the early medieval period, Valeria Polonio has been able to outline the
general features of Genoese ecclesiastical historical writings, which
included hagiographic texts dealing with specific saints, chronotaxes and
prosopographical notes on bishops and archbishops, accounts of major
events and translations of relics.18 Unfortunately, most of these sources
are lost. However, Polonio has convincingly argued that a later work, the
chronicle written at the end of the thirteenth century by the Genoese
archbishop Iacopo da Varagine, also drew information from sources now
lost but kept at that time in the archbishop’s palace.19 When describing
the Genoese crusader expeditions, Iacopo clearly made use of Caffaro’s
annals and, most probably, the work of William of Tyre. Iacopo did not
mention the presence of the Embriaci brothers in the Holy Land in 1099
and ascribed the final success of the crusade to a Genoese contingent of
forty ships. However, such a large fleet only left Genoa in 1104. The author
16
John Brian Williams, ‘The Making of a Crusade: The Genoese Anti-Muslim
Attacks in Spain 1146–1148’, Journal of Medieval History 23 (1997), 29–53;
17
Caffaro, ‘Annales’, p. 50; Caffaro, Genoa and the Twelfth Century Crusades,
p. 81; Corpus Inscriptionum medii aevi Liguriae III. Genova centro storico, ed.
Augusta Silva (Genova, 1987), no. 216.
18
Valeria Polonio, ‘Identità ecclesiastica, identità comunale: la memoria a
Genova’, in Comuni e memoria storica. Alle origini del Comune di Genova
(Genova, 2002), pp. 449–82 (here pp. 449–59).
19
Polonio, ‘Identità ecclesiastica’, pp. 458–59 and n. 17.
280
also antedated to 1099 events connected to the Genoese expedition of

1100–1101, such as the impressive miracle of the Holy Fire described by
Caffaro which is, however, only briefly evoked in Iacopo’s chronicle.20
Despite this clear attempt to amplify the Genoese role in the liberation
of the Holy Sepulchre, another crusader success, the capture of Antioch,
is more extensively described in the archbishop’s work. This distinction
is not a reference to the Liberatio, but rather to the local tradition that
after the conquest of Antioch the Genoese stole the relics of St John the
Baptist in Myra and brought them to Genoa. According to Iacopo, the
aforementioned Genoese fleet of forty galleys supported the crusaders in
both Antioch and Jerusalem, returning to Genoa in 1099 with the Baptist’s
relics. The author recalls the two sieges of Antioch, including an account
of the Invention of the Holy Lance most probably derived from William
of Tyre and certainly not from the Liberatio. He also mentions an earlier
text which narrated the translation of St John’s relics to Genoa and states
that he had written another account of these events with some hymns.21
Iacopo’s Ystoria translationis illustrates the Antiochene campaign in a very
different way from the chronicle of Genoa, probably following the earlier
text on the relics’ translation.22
Iacopo da Varagine’s treatment of Genoese involvement in the First
Crusade highlights the archbishop’s obviously propagandistic approach.
In all likelihood he consulted a manuscript of the annals which did not
include the Liberatio and deliberately chose to manipulate the informa-
tion in Caffaro’s work in order to magnify the glory of Genoa. He merged
events relevant to several Genoese expeditions to the east to suggest
a prompt and massive response to Urban II’s appeal and a substantial
Genoese contribution to the conquest of the Holy Sepulchre. However,
from his point of view the acquisition of the relics of St John gave a fur-
ther significance to the Antiochene campaign. Iacopo’s work is important
because it attests the existence of an earlier account of the translation of
the Baptist’s relics, while in the archbishop’s chronicle there is not the
20
Iacopo da Varagine e la sua cronaca di Genova dalle origini al MCCXCVII, ed.
Giovanni Monleone, 3 vols (Roma, 1941), 2: 84–86; Caffaro, ‘Annales’, pp. 7–9;
Caffaro, Genoa and the Twelfth-Century Crusades, pp. 51–53.
21
Iacopo da Varagine, 2: 88–89, 300–4; Caffaro, ‘De liberatione’, p. 120; Caffaro,
Genoa and the Twelfth-Century Crusades, p. 122.
22
Iacopo da Varagine, Legenda translationis, pp. 230–31; Valeria Polonio, ‘L’arrivo
delle Ceneri del Precursore e la diffusione del culto al Santo a Genova e nel
Genovesato’, in San Giovanni Battista nella vita sociale e religiosa a Genova e
in Liguria tra medioevo ed età contemporanea (Genova, 2000), pp. 35–65 (here
pp. 35–38, 44–48).
281
Elena Bellomo
slightest trace of an ecclesiastical memory of Genoese involvement in the

conquest of Jerusalem. Even in ecclesiastical circles this event seems not
to have been the subject of a specific record. It is entirely possible that
this was due to the aforementioned internal frictions that involved the
local Church, led initially by schismatic bishops and then leaderless until
1099. Recent research has emphasised that the Genoese were not strangers
to the religious motivations which inspired the crusade.23 However, the
local memory of these events was deeply affected on both the civic and
the ecclesiastical sides by contingent factors such as propagandistic and
political issues and the fact that no relics were gained by the Genoese
crusaders who contributed to the conquest of Jerusalem. Thus, paradoxi-
cally, the only Italian city that could actually claim to have helped the
crusaders significantly in 1099 marginalised the memory of this success
in favour of other crusader achievements.
Turning Defeat into Victory:

The Case of the Milanese Crusaders
The presence of combatants from Milan in the army of the First Crusade
is disputed.24 Albert of Aachen reports that during the second siege of
Antioch a Langobardus cleric encouraged his companions, telling them
23
Elena Bellomo, ‘La componente spirituale negli scritti di Caffaro sulla prima
crociata’, Atti della Società Ligure di Storia Patria n.s. 37 (1998), 63–92; Bellomo,
‘Un’identità composita: la percezione dell’oltremare negli scritti dell’annalista
genovese Caffaro’, in Studi sull’Europa medievale. L’Europa di fronte all’Oriente
cristiano tra alto e pieno Medioevo, ed. Annamaria Ambrosioni (Alessandria,
2001), pp. 77–94; Bellomo, A servizio di Dio, pp. 155–87; Jonathan Phillips,
‘Caffaro of Genoa and the Motives of Early Crusaders’, in Religion as an Agent
of Change, ed. Per Ingesman (forthcoming). For a first comparative view see
Christopher Marshall, ‘The Crusading Motivation of the Italian City Republics
in the Latin East. 1096–1104’, in The Experience of Crusading, ed. Marcus Bull,
Norman Housley, Peter W. Edbury, Jonathan Phillips, 2 vols (Cambridge,
2003) 1: 60–79, which, however, ignores any Italian secondary works on these
maritime cities.
24
Giancarlo Andenna, ‘I conti di Biandrate e le loro clientele vassallatiche alla
prima crociata’, in Deus non voluit. I Lombardi alla prima crociata (1100–1101).
Dal mito alla ricostruzione della realtà, ed. Giancarlo Andenna and Renata
Salvarani (Milano, 2003), pp. 233–62 (here p. 235); Alfredo Lucioni, Anselmo
IV da Bovisio, arcivescovo di Milano (1097–1101). Episcopato e società urbana sul
finire dell’XI secolo (Milano, 2011), p. 189.
282
that a priest ‘who lived in Italian parts’, known to him from boyhood,
had been approached by a pilgrim who foretold him the success of the
crusade. Then the pilgrim was revealed to be St Ambrose.25 The vision
of the patron saint of Milan and the clerics’ direct acquaintance since
boyhood seems to be plausible evidence that here the word Langobardus
indicates a man from the area of modern Lombardy. Moreover, his story
would be even more effective if addressed to fellow citizens devoted to
St Ambrose. Thus this passage suggests that crusaders from Lombardy,
and perhaps Milan in particular, were present in the Christian army.26 The
Milanese chronicler Galvano Fiamma claims in his Manipulus Florum,
dating from the 1330s or 1340s, that the Milanese Giovanni de Raude and
Pietro Salvatici were the first crusaders to enter Jerusalem. The Lombard
contingent also included a number of Milanese nobles such as Ottone
Visconti and Benedetto, also known as Rozino de Cortesella.27 While
local traditions ascribe a significant role to Milanese crusaders in the final
attack on Jerusalem, primary sources on the First Crusade, with the sole
exception of the episode recorded by Albert of Aachen, do not contain
any further information that can be linked to the area of Milan.28 Any
Milanese participation in this undertaking must actually have been very
limited and in fact Milan and north Italy made a much more incisive con-
tribution to the crusader expedition which left in 1100 and was defeated
by the Turks in 1101.29 This Lombard campaign had as its spiritual leader
the Milanese archbishop Anselmo IV da Bovisio (1097–1101). Alfredo
Lucioni has recently shed new light on this figure, penning a portrait
of a determined and skilled prelate. Lucioni has carefully reconstructed
25
Albert of Aachen, Historia Ierosolimitana. History of the Journey to Jerusalem, ed.
and trans. Susan B. Edgington (Oxford, 2007), IV.38, pp. 306–9; Jean Flori, Pierre
l’ermite et la Première Croisade (Paris, 1999), p. 367; Aldo A. Settia, ‘L’esercito
lombardo alla prima crociata’, in Deus non voluit, pp. 11–29 (here p. 23).
26
27
Galvano Fiamma, Manipulus Florum, ed. Ludovico A. Muratori, in RIS 11
(Milano, 1727), ch. 141, cols 617–18.
28
Aldo A. Settia has convincingly argued that the Lombard siege master mentioned
in Albert of Aachen’s chronicle (II.35, ed. Edgington pp. 120–23) was from
southern Italy: Aldo A. Settia, ‘Un “Lombardo” alla prima crociata: Tecnologie
militari fra occidente e oriente’, in Società, istituzioni e spiritualità. Studi in onore
di Cinzio Violante, 2 vols (Spoleto, 1994), 2: 843–55; Settia, Comuni in guerra.
Armi ed eserciti nell’Italia delle città (Bologna, 1993), pp. 249–60.
29
Settia, ‘L’esercito lombardo’, pp. 11–29; Giuseppe Ligato, ‘Le vicende della
crociata lombarda. Gerusalemme o ‘regnum Babilonicum’?’, in Deus non voluit,
pp. 31–103.
283
Elena Bellomo
Anselmo’s ascendancy, proving his membership of the de Raude family,

a line of capitanei and notable vassals of the archbishop of Milan well
integrated in the town elite.30 Probably following their prestigious relative,
several members of the de Raude family took the cross and left in 1100.
In 1102 Berlinda, mother of Vifredo de Raude, left some property to the
monastery of Sant’Ambrogio, but only if her son did not return de (most
probably Hierosolymitano but here the text is lacking) itinere.31 As Lucioni
hypothesises, at that time news of the Lombard defeat in the east had
probably reached Milan but Berlinda still hoped that her son was alive.32
A charter from 1104 also informs us that Ariprando de Raude was asked
by the late Alberico da Settala to settle Alberico’s last will. This happened
while both were in Hierosolimitano itinere, either on the Lombard crusade
or an ensuing pilgrimage to Jerusalem by some of the surviving crusaders.33
The only mention of a Giovanni de Raude in existing primary sources dates
to 1103 when, according to the chronicle of Landolfo Iuniore or di San
Paolo, his horse stepped on the foot of the priest Liprando after he had suc-
cessfully undergone an ordeal. Carlo Castiglione has identified him with
the eponymous consul of Milan attested to in 1135 but this supposition is
not confirmed by any primary sources.34 According to Galvano Fiamma,
when Benedetto Rozino returned to Milan, he founded the church of
the Holy Sepulchre in order to celebrate the Milanese contribution to the
liberation of the holy places.35 However, Galvano’s account is the result of
an apparent manipulation of the real facts recorded in primary sources. In
the euphoria which preceded the departure of the crusader army, on 15 July
1100, Anselmo IV reconsecrated the Milanese church of the Holy Trinity
and added a new dedication to the Holy Sepulchre. He also established an
annual festival on 15 July to commemorate the conquest of the Holy City
30
Lucioni, Anselmo IV da Bovisio, pp. 65–66. On the de Raude, see Hagen Keller,
Signori e vassalli nell’Italia delle città. Secoli IX-XII (Torino, 1995), p. 85, n. 58;
Andrea Castagnetti, ‘Feudalità e società comunale. II. Capitanei a Milano
e a Ravenna fra XI e XII secolo’, in La signoria rurale in Italia nel Medioevo
organizzato da C. Violante e M. L. Ceccarelli Lemut (Pisa, 2006), pp. 117–215
(here pp. 122, 132, 144–51).
31
The iudicatum can be consulted at http://cdlm.unipv.it/edizioni/mi/milano-
sambrogio-mon3–1/carte/ambrogiomon1102–04–23 .
32
Lucioni, Anselmo IV da Bovisio, pp. 189–90.
33
Lucioni, Anselmo IV da Bovisio, pp. 190–91.
34
Landolfo Iuniore o di San Paolo, Historia Mediolanensis, ed. Carlo Castiglione,
RIS, ser.2, 5/3 (Bologna, 1934), ch. 18, p. 13; ch. 59, p. 36. For the date of the
ordeal see Lucioni, Anselmo IV da Bovisio, p. 191, n. 17.
35
Galvano Fiamma, Manipulus Florum, ch. 153, col. 627.
284
and to be celebrated in the church itself. According to Lucioni, this was

the final and most spectacular act of Anselmo’s episcopate. The festival
was solemnised by an indulgence granting remission of a third of their
sins to those faithful who visited the Holy Sepulchre of Milan. In order
to protect the pilgrims an eight-day truce before and after the festival
was established as well. The celebration was rounded by the holding of
market.36 It is certain that the church already existed when Anselmo IV
reconsecrated it and, most probably, also restored the building.37 In
thirteenth-century sources the church is only referred to by the new
dedication to the Holy Sepulchre. In all likelihood it had been gradually
changing its dedication, while keeping an apparent link with the Rozoni
family, the founders of the church in 1030.38 The possible participation
of a Rozone, the aforementioned Benedetto, in the Lombard crusade
may have created a further link between this edifice and the expedition.
Galvano Fiamma’s account clearly shows a well calculated shift in the
memory of the Milanese crusade: the bitter defeat suffered by this expedi-
tion was turned into triumphant participation in the contingent which
had conquered Jerusalem. The church of the Holy Sepulchre, probably
restored for its new consecration, became the tangible proof of the Milanese
contribution to the liberation of the Sepulchre in Jerusalem. At the same
time, the foundation of the church was postdated. In the course of the
centuries a disaster which had ended with the defeat of the Christian army
and the death of the Archbishop Anselmo IV in Constantinople has been
turned into a glorious episode in the history of Milan. The memory of the
Lombard who actually reached Antioch and encouraged his companions
by telling them a prophecy of victory ascribed to St Ambrose was lost, while
in contrast, the Milanese participation in the catastrophic expedition of
1100–1101 was framed by a new splendid scenario with the impressive image
of Giovanni de Raude heroically climbing the walls of Jerusalem in 1099.39
36
Ambrosianae Mediolani basilicae ac monasterii hodie cisterciensis monumenta,
ed. Giovanni Pietro Puricelli (Milano, 1645), no. 298; pp. 481–85; Lucioni,
Anselmo IV da Bovisio, pp. 193–95; Renata Salvarani, ‘San Sepolcro a Milano
nella storia delle crociate’, in Deus non voluit, pp. 263–82 (here pp. 272–73).
37
Salvarani, San Sepolcro a Milano, p. 271; Roberto Ottolini, ‘La crociata milanese
del 1100: vicenda storica e riflessi icnografici’, Studi medievali 48 (2007),
761–66.
38
Salvarani, San Sepolcro a Milano, pp. 274–75.
39
Lucioni, Anselmo IV da Bovisio, p. 192; I am currently working on a
comprehensive analysis of the perception of the Lombard crusade of 1100–1101
in the Milanese primary sources from the twelfth to the sixteenth century.
285
Elena Bellomo
Crusade as Family Pride

Galvano Fiamma’s accounts of Milanese participation in the First Crusade
open up another promising line of research: the use of the crusade in the
development of family traditions among the communal aristocracy. In
1288, writing his De magnalibus civitatis Mediolani, the Milanese chroni-
cler Bonvesin da la Riva stated that the commune of Milan used to offer
the Visconti family the most important standard of the town, which
portrayed an indigo snake eating a red man. According to Bonvesin, this
standard preceded the Milanese army and had to be hung in a visible
position before it encamped. Bonvesin concludes saying: ‘Hanc autem
dignitatem propter excellentem cuiusdam Ottonis Vicecomitis, viri
strenuissime indolis, probitatem et victoriam quam contra Sarracenos
ultra mare in bello exercuit, dicitur h[abuisse] Vicecom[itum] nobilissima
parentela.’40 In his Manipulus Florum, Galvano Fiamma reports that dur-
ing the First Crusade Ottone Visconti duelled with a Muslim knight who
bore the image of a viper eating a man on his shield and helmet. Ottone
took the symbol from the defeated enemy and put it on his banner. In
order to celebrate this victory, the Milanese decided that their army could
only set up camp after the standard with the viper was hung on a tree.41
The origin of the Viscontis’ coat of arms with the viper eating a man is
still unknown and has been given various explanations by scholarship.42
This debate is not relevant to the subject of this article. What is interesting
is that Bonvesin da la Riva and Galvano Fiamma present two accounts
which are at the same time similar but very different. While Bonvesin
does not mention the First Crusade and refers to an unspecified fight
overseas against the Turks, Galvano links the coat of arms of the Visconti
family to Ottone’s participation in the First Crusade. Bonvesin claims
that the snake was a communal emblem adopted by the Milanese army;
Galvano attributes Muslim origins to it, and its importance is only linked
to Ottone’s bravery. It is apparent that Galvano’s account, written several
decades after Bonvesin’s one, retains a more sophisticated version of the
origin of the Viscontis’ crest, and it cannot be ruled out that Galvano’s
narration was also partially influenced by the eleventh-century Milanese
chronicler Landolfo Seniore who recorded a duel between Eriprando,
40
Bonvesin da la Riva, Le meraviglie di Milano (De magnalibus Mediolani), ed.
Paolo Chiesa (Roma, 2009), V.22, pp. 126–29.
41
Galvano Fiamma, Manipulus Florum, ch. 154, cols 617–18.
42
The theories explaining the origin of the Visconti’s viper are summarised in
Bonvesin da la Riva, Le meraviglie, pp. 237–38. On this issue also see Andenna,
‘I conti di Biandrate’, p. 234.
286
father of Ottone, and a relative of Conrad III that occurred in 1037.43

Ottone Visconti is mentioned in primary sources from 1075 to 1111.44
Unfortunately, no contemporary evidence attests his presence amongst
the Milanese who left in 1100. However, at that time the Viscontis already
played a prominent role in the Milanese aristocracy, and it is very likely
that a member of such an important family, closely linked to the arch-
bishops of Milan, had accompanied Anselmo IV on his way east.45
The antedating of the Lombard crusade to 1099 allowed the integration
of this fiasco into the history of the Viscontis, turning it into a glorious
episode connected to the family coat of arms. Moreover, the manipula-
tion of the past which had turned the Lombard crusade into a successful
expedition was exploited by the Viscontis precisely when Ottone Visconti,
archbishop of Milan, had established his dominance in the town after
the final defeat of his rivals, the Torrianis (1277).46 Bonvesin da la Riva’s
account presents several credible elements, such as the description of a
Milanese banner and its attribution to the Viscontis thanks to the heroic
deeds of the crusader Ottone. In the following century, the legend was
fully developed, reinterpreting the data totally in favour of the Viscontis:
Ottone, just like his father, defeats an enemy in a duel, and the insignia with
the viper becomes a war trophy. It is worth noting that the first example of
the Viscontis’ coat of arms is on the archbishop’s palace in Legnano built
43
Landolfo Seniore, ‘Historia Mediolanensis’, ed. Ludwig Bethmann and Philipp
Jaffé, in MGH SS 20 (Hannover, 1868), III.25, p. 62.
44
Gli atti privati milanesi e comaschi del secolo XI, ed. Cesare Manaresi and
Caterina Santoro, IV, c. 1075–1100 (Milano, 1969), nos 557, 558; I placiti del
‘regnum Italiae’, ed. Cesare Manaresi, III/2, 1085–1100 (Roma, 1960), no. 467;
Le pergamene del secolo XII della Chiesa maggiore di Milano (Capitolo maggiore
– Capitolo minore – Decumani) conservate presso L’Archivio di Stato di Milano,
ed. Maria Franca Baroni (Milano, 2003), no. 1; Landolfo Iuniore o di San
Paolo, Historia Mediolanensis, ch. 26, p. 17; Girolamo Biscaro, ‘I maggiori dei
Visconti, signori di Milano’, Archivio Storico Lombardo 16 (1911), 5–76 (here p.
31). Keller (Signori e vassalli, p. 85, note 58) suggests that he can be identified
with the Otto Mediolanensis mentioned in a placitus from 1086: I placiti del
‘regnum Italiae’, ed. Cesare Manaresi, III/1, 1285–1084 (Roma, 1960), no. 459.
45
Keller, Signori e vassalli, pp. 175–88; Castagnetti, Feudalità e società comunale.
II, pp. 122, 132–34, 144, 145, 150.
46
On Archbishop Ottone Visconti see Grado G. Merlo, ‘Ottone Visconti e la
curia arcivescovile di Milano. Prime ricerche su un corpo documentario’, in
Gli atti dell’arcivescovo e della curia arcivescovile di Milano nel sec. XIII. Ottone
Visconti (1262–1295), ed. Maria Franca Baroni (Milano, 2000), pp. ix-xxxiv.
287
Elena Bellomo
by Archbishop Ottone Visconti (1262–85).47 Thus the establishment of

the Viscontis’ supremacy in Milan seems to be a turning point in a further
modification of the past to the advantage of the ruling family.
The Milanese manipulation of the local crusading experience displays
several elements which later became topoi in the legends connecting Italian
towns to the First Crusade. In the case of Milan the falsification was sup-
ported by actual participation in a crusading expedition very soon after the
conquest of Jerusalem. In the course of the following centuries the appeal
of the First Crusade did not diminish. On the contrary, it provided the
perfect frame for fake family glories, especially developed in the fifteenth
and sixteenth centuries when the Italian aristocracy was becoming a closed
class with its own symbols, traditions, and codes of conduct. For instance,
in the early seventeenth century the Pisan scholar Raffaello Roncioni in
his Delle Istorie pisane claimed that the Pisans contributed to the con-
quests of both Antioch and Jerusalem. In this version Cucco Ricucchi
was the standard bearer of the Pisans during the siege of Jerusalem. The
crucifix that was on the top of his banner inspired the crusaders to attack.
According to Roncioni, on a partigiana (a kind of halberd) found in the
Ricucchi palace there was an inscription in Italian in which Cucco himself
claimed to have been the first crusader to climb the walls of Jerusalem. An
identical inscription was also found on the gate of the Fortezza Vecchia
of Livorno, but in this case it was the Pisan Coscetto del Colle who, also
in Italian, claimed the same honour for himself.48 It is apparent that the
two inscriptions are forgeries but they effectively contributed to enhance
the glories of the relevant families and of Pisa, whose naval contingent
first arrived in the east after the conquest of Jerusalem.49
Another interesting case is that of the Pazzis and the Scoppio del carro in
Florence, investigated by Sergio Raveggi. Even today the Scoppio del carro
is a popular festival in Florence. At Easter an antique cart, loaded with
fireworks, crosses the city centre while a fire, struck with flints claimed to
be from Jerusalem, is carried in procession to the Duomo. With this Easter
fire the archbishop of Florence lights a mechanical dove (the colombina)
47
Giorgio D’Ilario, Egidio Gianazza, Augusto Marinoni and Marco Turri, Profilo
storico della città di Legnano (Legnano, 1984), p. 215.
48
Raffaello Roncioni, ‘Delle Istorie pisane libri XVI’, ed. Francesco Bonaini, in
Archivio Storico Italiano 6/1 (1844), 138–45; Sergio Raveggi, ‘Storia di una
leggenda: Pazzo dei Pazzi e le pietre del Santo Sepolcro’, in Toscana e Terrasanta
nel Medioevo, ed. Franco Cardini (Firenze, 1982), pp. 299–315 (here p. 315,
n. 42); Maria Luisa Ceccarelli Lemut, Pisa e l’Oriente latino dalla I alla III
Crociata (Pisa, 2010), pp. 22–24.
49
See Mike Carr’s chapter in this volume pp. 75–87.
288
which ignites the cart outside the church. According to local tradition
this ceremony commemorates the participation of a Florentine contingent
in the First Crusade. Among the Florentine crusaders Pazzino de’ Pazzi
particularly stood out as the first crusader to reach the top of the Jerusalem
walls carrying the crusader banner (on 15 July 1099 the walls of Jerusalem
must have been crowded with Italian crusaders). Thanks to his bravery
Pazzino received from Godfrey of Bouillon a coat of arms which resem-
bled that of the duke, and three small stones from the Holy Sepulchre. In
Florence Pazzino was triumphantly welcomed by his fellow citizens who
prepared a chariot for his parade in the city. The stones from the Holy
Sepulchre were kept first in S. Maria sopra Porta and later in S. Biagio. Every
Easter they were used to light a fire which was then distributed in town.50
This legend clearly merges several elements with different origins: the
involvement of Florentines in a crusade; the praise of an ancestor of the
Pazzis, one of the most notable Florentine families; and the Easter cer-
emony of the Holy Fire, a well-known ceremony in medieval Jerusalem
where celestial fire miraculously appeared in the Holy Sepulchre every
Easter and was then distributed to the faithful.51
The first mention of a Holy Fire rite in Florence dates to the early
fourteenth century when Giovanni Villani described it in his chronicle.
According to this account, since ancient times, just as in Jerusalem, a new
fire was blessed on Easter Saturday. Torches were lit with this fire and it was
brought by the Florentines to their houses. Giovanni Villani explains that
the Pazzis gained a prominent role in this ceremony because one of their
ancestors, named Pazzo, was strong enough to carry the biggest torch.52 In
Villani’s account the link between this celebration and the Holy Sepulchre
is very clear but there is no reference to the crusade. It is in the sixteenth
century that the first evidence of the legend about the crusader Pazzino
de’ Pazzi is found. It is recalled in a chronicle written in 1535 by Ghinozzo
Uguccione de’ Pazzi. This passage claimed to be a copy of an older account
and was later included in further works such as, for instance, the Notizie
istoriche delle Chiese fiorentine by Giuseppe Richa (1755).53
Thus it is likely that in the course of the fifteenth century, or at the lat-
est in the sixteenth century, the Pazzis developed a story which explained
50
Raveggi, ‘Storia di una leggenda’, pp. 299–300, 303.
51
See above, note 20.
52
Giovanni Villani, Nuova Cronica, II.23, ed. Giuseppe Porta, 2 vols (Parma,
1990–1991), 1: 89–90.
53
Giuseppe Richa, Notizie historiche delle chiese fiorentine divise ne’ suoi quartieri,
3 vols (Firenze, 1755), 3: 233 (and on this ceremony, 3: 232–36); Raveggi, ‘Storia
di una leggenda’, p. 303.
289
Elena Bellomo
their link with the rite of the Holy Fire in a more heroic and satisfactory
way. As Raveggi has noted, they tried to elaborate a falso verosimile. In the
legend of Pazzino de’ Pazzi the First Crusade is antedated to 1088 because
precisely one century later, in 1188, a Florentine contingent joined a crusad-
ing expedition.54 In the battle of Montaperti, in 1260, between Florentine
Guelphs and Guibellines, Iacopo del Nacca Pazzi was the standard bearer
of the Florentine commune (just as Pazzino was the crusader standard
bearer). He intrepidly defended the banner even after losing both arms.
Another notable member of the family, Pazzino de’ Pazzi, eponymous
with the legendary crusader, was exiled from Florence after the defeat of
Montaperti, and it has been claimed that he was the first member of the
family to use the coat of arms described in the crusader legend.55
A Florentine Easter celebration resembling the Jerusalem rite of the
Holy Fire, the actual role of the Pazzis in this ceremony, the Guelph mili-
tancy of the family, and the bravery of one of its members were merged in
an appealing legend that magnified and glorified the origins of a family
which had actually gained its power though trade and banking activities.
The binding agent in the development of this myth was the First Crusade
which provided the perfect setting for an antique legend of valour.
Conclusions
Memory and history are affected by several factors. A complex process
shapes personal and collective, family and town memories according to
both enduring values and changeable ambitions. The key to the success
of written memory is the ability to match the plans of the ruling class, the
prevalent mentality, and authors’ and readers’ expectations. In this respect
defeat can easily be turned into victory, and even victory can be overlooked
in cases where its protagonists do not correspond to the desired version of
the story. From 1099 the crusade proved to be a very powerful concept,
the perfect foundation for international political propaganda as well as for
more local and limited purposes. Later on it became the ideal context to
provide many legends, objects of devotion, and symbols with prestigious
explanations and origins. At this stage the crusade was losing its historical
character and becoming a cultural reference. Despite its apparent failures
and alterations, it still exerts a formidable draw and remains a concept deeply
rooted in Italian and European civilisation, able to survive the era which
gave it birth and to influence the European culture for centuries to come.
54
Raveggi, ‘Storia di una leggenda’, p. 308.
55
Raveggi, ‘Storia di una leggenda’, p. 313.
290
Making Heroes out of Crusaders
The Literary Afterlife of Crusade Participants
in the Chanson d’Antioche
Simon T. Parsons
Royal Holloway, University of London
The First Crusade spawned a great number of literary and historical works
concerned with explaining, disseminating, glorifying, and appropriating
the success of the expedition. While academic study has naturally focused
on the closest, and presumably best informed, of these accounts – the Latin
‘crusade chronicles’ – in an attempt to ascertain the historical truth of
the expedition, an important group of texts have been less studied: those
which make up the vernacular Old French Crusade Cycle.
Despite concerns over the validity of such labels, and questions over the
impartiality of sources, our current understanding of the First Crusade
suggests that the expeditionary force consisted of, and was certainly led,
predominately by those whom we should now consider ‘French’, and who
may have considered themselves Franci, namely, speakers of the langue d’oïl
and langue d’oc. The twelfth century, so abundant with writing about the
crusades, was also the period of genesis for vernacular French literature,
where contemporary concerns were reflected in assonanced or rhymed
epic poems known as the chansons de geste. It is not surprising to discover,
therefore, that the events of the crusade eventually found their way into
the burgeoning genre of the chansons. This relevance to contemporary
historical events is a unique case in the genre: the traditional chanson de
geste may exemplify twelfth-century concerns, but is normally set in the
pseudo-historical, fictive world of the eighth and ninth centuries.
The text known to us as the Chanson d’Antioche represents the earliest
extant form of these langue d’oïl vernacular accounts of the First Crusade.
This is not to say that there were not earlier versions, either spoken or
written: the chansons de geste were oral poems, intended for recital and
performance. They were probably composed, reimagined, and transmitted
without written texts being used.1 To try and understand a predominantly
1
There is much discussion over orality in the chansons. For a seminal starting
point, see Joseph J. Duggan, The Song of Roland: Formulaic Style and Poetic
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. 1–.
Simon T. Parsons
oral form of ‘literature’ with reference to existing written texts poses its
own challenges to scholars.
The Chanson d’Antioche is found in nine medieval manuscripts (and two
fragments),2 each including only part of the work known as the Old French
Crusade Cycle. This cycle occupies an interesting position, somewhere
between historical account and fantastic literature. There are arguably twelve
poems contained within this cycle,3 and not one of the cyclical manuscripts
preserves them all: the manuscripts are compilations of material, dating
from the middle of the thirteenth century onwards. The common conven-
tion, with which there is no reason to disagree, is that the central trilogy of
Chanson d’Antioche – Chanson des Chétifs – Chanson de Jérusalem predates
the other poems in the cycle.4 These three are very loosely based on the
‘historical’ events of the First Crusade, respectively: on the progress of the
crusade up until the aftermath of the battle outside Antioch in the summer
of 1098; on the adventures of certain captives in the aftermath of the People’s
Crusade; and on the eventual capture of the Holy City.
There is a strong argument for dissociating the Chanson d’Antioche
further. Both the Chanson des Chétifs and the Chanson de Jérusalem have
little correlation with known Latin sources for the First Crusade, and
contain much material which we would now consider to be legendary.
The Chanson d’Antioche, however, is very similar in terms of its structure
and content to the accounts of Albert of Aachen and Robert the Monk,
and further contains parallels with other Latin First Crusade texts. There
is limited fantastical material, certainly not significantly more than occurs
in Latin crusade sources. The Chanson de Jérusalem appears to be based
on the Chanson d’Antioche, comprising what Anouar Hatem described
memorably as ‘une imitation, parfois servile, de la Chanson d’Antioche’.5
Craft (Berkeley, 1973), esp. pp. 16–62.

2
MS Paris, Bibliothèque nationale de France, fr.12558, dateable to before 1250,
is considered to represent the oldest tradition. See La Chanson d’Antioche, ed.
Suzanne Duparc-Quioc, 2 vols (Paris, 1978), 2: 43.
3
La Naissance de Chevalier au Cygne, Le Chevalier au Cygne, La Fin d’Elias, Les
Enfances Godefroi, La Retour de Cornumarant, Antioche, Les Chétifs, Jérusalem,
La Chrétienté Corberan, La Prise d’Acre, La Mort Godefroi, La Chanson des Rois
Baudouin. There are variant versions of many of these texts, so the exact number
in the cycle depends on interpretation of what comprises a separate poem.
4
Suzanne Duparc-Quioc, Le Cycle de la Croisade (Paris, 1955), pp. 70–74.
5
Anouar Hatem, Les Poèmes épiques des croisades: genèse – historicité –
localisation. Essai sur l’activité littéraire dans les colonies franques de Syrie au
Moyen-Age (Paris, 1932), p. 269.
292
It is here that the ‘text’, as we possess it in the earliest manuscripts,

and the ‘chanson’ divide. The text was composed or redacted at the same
time as the Chanson des Chétifs and the Chanson de Jérusalem, as internal
narrative coherence demonstrates. Estimates for when this took place
have varied from c.1177 to c.1204. The earlier dating of the text is based
upon Suzanne Duparc-Quioc’s study of the poem, where she convinc-
ingly argued that many of the cultural references in the extant Chanson
d’Antioche, particularly to Prester John and the Seven Sleepers of Ephesus,6
were particularly apposite to the 1170s-1180s.7 The later dating, resting
predominately upon a perceived propagandistic link with the Fourth
Crusade, has been proposed by Carol Sweetenham and Susan Edgington.8
But the chanson itself, as opposed to the extant text, existed before that.
Indeed, the Chanson d’Antioche itself announces that it is a reworking of
an earlier poem, which was ostensibly composed by a figure called Richars
li pelérins (Richard the Pilgrim).9 It also states that the instigator of this
reworking, and probably the same reworking which aligned the Chanson
d’Antioche with the Chanson des Chétifs and the Chanson de Jérusalem,
was a figure known as Grainsdor of Douai. Despite attempts to establish
the historicity of either of these figures,10 there has been no convincing
independent evidence of their identities.
There is strong evidence for an earlier written version of the Chanson
d’Antioche, whether or not a Grainsdor or Richars was involved in its produc-
tion. Close textual similarities with early to mid twelfth-century crusade
accounts, particularly Albert of Aachen, suggest more than simple copying
on the part of the Chanson d’Antioche.11 In particular, the relationship to the
heroic verse account of Gilo of Paris has been understudied, and suggests
6
See Chanson d’Antioche, ed. Duparc-Quioc, 1: 345, line 6953; 1: 37, line 413; 1:
169, line 2974.
7
Duparc-Quioc, Chanson d’Antioche, 2: 137–39.
8
Carol Sweetenham, ‘Antioch and Flanders: Some Reflections on the Writing
of the Chanson d’Antioche’, in Epic and Crusade, ed. Philip E. Bennett, Anne
E. Cobby, Jane E. Everson (Edinburgh, 2006), pp 131–53; and The Chanson
d’Antioche: An Old French Account of the First Crusade, trans. Susan B.
Edgington and Carol Sweetenham (Farnham, 2011), p. 47.
9
Chanson d’Antioche, ed. Duparc-Quioc, 1: 443, line 9014.
10
See esp. Hermann Kleber, ‘Graindor de Douai: remanieur-auteur-mécène?’, in
Les Épopées de la Croisade – Premier colloque international (Trèves, 6–11 août
1984), ed. Karl-Heinz Bender (Stuttgart, 1987), pp. 66–75; Robert Francis
Cook, Chanson d’Antioche, chanson de geste: Le cycle de la croisade est-il épique?
(Amsterdam, 1980), pp. 23–30.
11
Discussed in The Chanson d’Antioche, ed. Edgington and Sweetenham, pp. 3–48.
293
Simon T. Parsons
common sources, but not direct influence.12 A possibly related Occitan

tradition of vernacular poems is attested by Geoffroy of Viegeois, writing
in the late twelfth century, who tells us of an ingens volume […] materna
lingua ritmo vulgari (‘a great work […] in his mother tongue and the com-
mon metre’) written by a Gregory Bechada, drawing on information from
a figure called Gaubert the Norman.13 There is evidence from the end of the
twelfth century that troubadours were expected to know, or did know, chan-
sons de geste on the First Crusade, and in particular the siege of Antioch.14
Historians and literary scholars can only study the extant text, and
any attempts to establish which elements existed in earlier versions of the
Chanson d’Antioche are at best tendentious.15 But the Chanson d’Antioche
is not just a text, more importantly it is a tradition. Its attempts to
glorify and exalt the leaders of the First Crusade to hero status cannot
be assigned entirely to the early thirteenth century: these same processes
were probably under way during the twelfth century as well. With this
understanding of the process by which the extant text of the Chanson
d’Antioche was formed, it is possible to consider the ways in which the
Chanson d’Antioche glorified the participants of the First Crusade and
constructed heroic identities for them.
The Chanson d’Antioche is, despite its ‘historical’ subject matter, a
chanson de geste, and, as such, takes on the conventions of its genre. This
is evident from its stylistic principles, formulaic passages, and general
martial ethos. The inclusion of the story of the First Crusade in this sort
of text ‘heroicises’ the material in much the same way that, for example,
portraying material in a comic book might do today. Portraying the events
of the First Crusade in a chanson de geste has obvious implications for
the way the material is understood by its audience. As in the canonical
chansons, the successes of the expedition are a result of acts of personal
heroism, of individual agency supported and ordained by God. This is
not in conflict with the crusading ideology of the late twelfth and early
thirteenth centuries which, like the chansons, is predominately driven
12
See Suzanne Duparc-Quioc, ‘Un Poème Latin du xiie siècle sur la première
croisade par Gilon de Toucy, augmenté par Fulco’, in Les Épopées de la Croisade,
pp. 35–49.
13
See The Canso d’Antioca: An Occitan Epic Chronicle of the First Crusade, ed.
Carol Sweetenham and Linda M. Paterson (Farnham, 2003), pp. 5–9.
14
Lambert of Ardres, The History of the Counts of Guines and Lords of Ardres,
trans. Leah Shopkow (Philadelphia, 2000), p. 130; Stefano M. Cingolani,
‘The sirventes-ensenhamen of Guerau de Cabrera: A Proposal for a New
Interpretation’, Journal of Hispanic Research 1 (1992), 191–201.
15
See, for example, Chanson d’Antioche, ed. Duparc-Quioc, 2: 99–242.
294
not by kings and institutions, but by the martial power of nobles, bound
together, under God, by complex bonds of kinship and fealty.
From the earliest stanzas, the Chanson d’Antioche portrays the First
Crusade as an heroic enterprise:
Segnor, oés cançon de grant cevalerie
Des bons barons de France drois est que jo vos die
Ki par force conquisent la celestiel vie.16
Barons, a term which means not only ‘lords’ but also ‘heroes’, or ‘valiant
nobles’, are the protagonists of the expedition. The song is explicitly a tale
of Christian knighthood: how it avenges Christ’s death on the cross; how
it raises Christianity; and casts down the pagans. The concept is elucidated
further elsewhere in the text:
Cele ost a Nostre Sire meïsmes commandée
Des bons barons de France fu faite et porpensée.17
There is nothing unique about this concept: it is a neat analogue to the
title of Guibert of Nogent’s work Dei gesta per Francos. But the Chanson
d’Antioche goes further than any other text in following through with this
promise, displaying many examples of personal valour which are signifi-
cant in turning the tide of battle and allowing God’s will to be carried out.
The author or redactor of the Chanson d’Antioche is careful to position
the text within the same narrative world as the conventional chansons de
geste, namely the mythological Carolingian past. During a violent assault in
the final battle of Antioch, the climactic set-piece of the text, the Christians
are explicitly compared to the heroes of the Carolingian chansons:
Cil qui a pié remest molt fu grains et dolans
Son escu triers son dos totes voies soufrans.
Les grans paines que ot Oliviers ne Rollans
Ne celes que soufri Iaumons ne Agolans
Ne li bers Vivïens quant fu en Aliscans
Ne valut a cestui le pris de .III. bezans.18
16
‘My lords, hear a song of great knighthood./ It is right that I should speak of
how the noble lords of France/ gained heavenly life through force of arms’:
Chanson d’Antioche, ed. Duparc-Quioc, 1: 23, lines 105–7.
17
‘This army was commanded by Our Lord himself,/ conceived of and carried
through by the noble lords of France’: Chanson d’Antioche, ed. Duparc-Quioc,
1: 25, lines 144–45.
18
‘He who remained on his feet was harassed and distraught,/ his shield thrown
behind him on his back and attacked from all sides./ Compared with this,
the great sufferings of Oliver or Roland,/ or those which Iaumont or Agolant
295
Simon T. Parsons
The storylines of three popular chansons are invoked to aggrandise

the suffering which the crusaders endured, and to link them with
these heroes of the mythic past. Oliver and Roland are the Christian
leaders who die heroically on the battlefield of Roncesvals after being
ambushed in the Chanson de Roland; Iaumont and Agolant are two
Saracen kings who are struck down in the field by a younger Roland
in the narrative of the Chanson d’Aspremont; and Vivien is the young
nephew of William of Orange who is heroically slaughtered after a
last stand on the shoreline of Aliscans, an episode occurring in both
the Chanson de Guillaume and the slightly later Aliscans. Later a ref-
erence is made to there being no need to recall the deeds of ‘Bertran
ne Aïmer’ when one could see the Saracens being dismembered and
sent somersaulting over each other by the mighty blows of Godfrey, a
reference to two Christian knights in Aliscans, ferocious nephews of
the hero William of Orange.19
The leaders of the First Crusade are brought into the same world as the
heroes of the mainstream chansons de geste not solely by referential means,
but also genealogical. When trapped in Antioch, the crusaders decide to
send a mission to Corbaran to propose a duel between warriors in place
of a full battle. This episode occurs in many Latin sources, but is exag-
gerated and lengthened significantly in the Chanson d’Antioche.20 While
the mission is under way, the nobles discuss who will get the honour of
being the army’s champion:
Et al seul ont eslit Godefroi de Buillon
K’il est praus et delivres, del linage Carlon.
[Robert of Normandy becomes so enraged that he considers leaving]
Tous plains de maltalent vint a son pavellon
Et fait se siele metre, il et si compaignon.
‘Que voles faire, sire?’ dist Fouques d’Alençon
endured,/ or that which the hero Vivien suffered when he was at Aliscans,/
were not worth the price of three bezants’. Chanson d’Antioche, ed. Duparc-
Quioc, 1: 424, lines 8611–16.
19
I agree with Duparc-Quioc, Chanson d’Antioche, 1: 453, over Edgington and
Sweetenham, The Chanson d’Antioche, p. 316, in identifying Aïmer with Aïmer
li Chaitif, William’s nephew whom we see fighting at Aliscans alongside
Bertrand, rather than William’s father (more often called Aïmeri). See Aliscans,
ed. Claude Régnier, 2 vols (Paris, 1990), 2: 357.
20
See. the mission of Balant in the Aspremont, or the trial by combat of Ganelon
in the Roland: Aspremont: Chanson de geste du xiie siècle, ed. and trans. François
Suard (Paris, 2008), pp. 80–112, lines 194–753; La Chanson de Roland, ed. and
trans. Ian Short (Paris, 1990), pp. 250–254, lines 3858–3933.
296
‘Par foi jo m’en irai en nostre region.

Dont ne sui del linage Ricart le fil Doon?
Ainc por un chevalier ne wida son arçon.21
Here, the two nobles are positioned as rivals in heroic pedigree: a rivalry
based upon their patrilineal heritage, with links to the ‘literary’ world of
the chansons de geste. Godfrey is described as a descendant of Charlemagne
(there is no convincing historical argument to suggest that this was
the case). Robert’s ancestor is Doon of Mayence, the progenitor of a line-
age of barons rebellious towards Charlemagne and a figure who gives his
name to a whole cycle of chansons de geste. By virtue of their ancestors, the
two are immediately set up as natural rivals for an audience conversant
with the storylines of the chansons. Crucially, in the text the status of
Godfrey and Robert as leaders relies upon their links with the popular
heroes of the literary past.
This kind of alignment of the worlds of crusade and epic is by no means
the only way in which heroes are constructed in the Chanson d’Antioche.
Frequently, the Chanson d’Antioche augments, expands, and perhaps
invents new episodes to demonstrate the heroic attributes of particular
figures. Two of the plentiful occurrences in the text will be examined here.
Estatins
In the Chanson d’Antioche, Estatins (or Statins) is a general of the
Byzantine emperor Alexios and his main representative on the crusade.
This character is certainly to be associated with Tatikios, historical general
of Alexios during the late eleventh and early twelfth centuries, mentioned
frequently by Anna Komnene.22 Nearly all the Latin sources, including the
Gesta Francorum, Raymond of Aguilers, Guibert of Nogent, and Albert of
Aachen attest his presence, agreeing on his position, his prowess, and the
fact that he was deprived somehow of his nose – Guibert of Nogent tells
21
‘They chose Godfrey of Bouillon for the single combat, / who was valiant and
worthy, and of the line of Charlemagne […] Filled with fury, he [Robert] went
to his tent / And put his saddle on his horse, along with his companion. /
“What do you want to do, Lord?” said Fulk of Alençon,/ “By my faith, I want
to go back home. / Am I not of the lineage of Richard, the son of Doon? / He
was never unhorsed by any knight.”’: Chanson d’Antioche, ed. Duparc-Quioc, 1:
371, lines 7437–45.
22
Basile Skoulatos, Les personnages byzantins de l’Alexiade (Louvain, 1980),
pp. 287–92.
297
Simon T. Parsons
us that he had it replaced with a gold prosthesis.23 This attribute is unique

to Latin crusade sources: Byzantine accounts do not report a missing
nose. Where the Chanson d’Antioche diverges from the majority of Latin
crusade sources however, is in portraying him relatively positively. In the
Latin sources, he is not a crusader, but a sort of Byzantine ‘chaperone’,
keeping an eye on the crusaders and ready to flee at a moment’s notice
if things turn sour. In the Chanson d’Antioche, he becomes a defender of
Christianity, and a crusader in his own right.24
According to the account of the Chanson d’Antioche, while the cru-
saders were still in Christian lands, before crossing over into Anatolia,
the emperor Alexios allegedly came up with a plan to crush the crusade
utterly, since he was worried about its intentions. Estatins, despite being
a nephew of Alexios (unique information to the Chanson d’Antioche, and
reminiscent of nephew-uncle relationships in vernacular literature),25 leapt
to the defence of the crusaders, agreeing with their mission and asserting
that ‘harm done to them is harm done to me’.26 When Alexios persisted:
Estatins s’escria: ‘M’espee m’aportés!’
Illuec li fu li brans en se main delivrés.
Estatins le traist fors com s’il fust forsenés.
Adont fust li estors sus el palais levés
Jan i eüst de mors, d’ocis et d’afolés,
Quant Guis li senescaus en monta les degrés,
Dist a l’empereor: ‘Malvais consel creés.
Por coi avés esté en cest palais mellés?’27
23
Gesta Francorum et aliorum Hierosolimitanorum, ed. and trans. Rosamund Hill
(London, 1962); Le ‘Liber’ de Raymond d’Aguilers, ed. John H. Hill and Laurita
L. Hill (Paris, 1969), p. 54; Dei gesta per Francos, ed. Robert B.C. Huygens
(Turnhout, 1996), p. 182; Historia Ierosolimitana: History of the Journey to
Jerusalem, ed. and trans. Susan B. Edgington (Oxford, 2007), pp. 94–95.
24
Although he does not strictly take the cross, the Chanson d’Antioche does not
emphasise this aspect of crusading. Interestingly, Estatins is introduced only
one stanza after the council of Clermont, and so is not narratively disconnected
from the canonical leaders of the crusade.
25
See William O. Farnsworth, Uncle and Nephew in the Old French Chansons de
geste: A Study in the Survival of Matriarchy (New York, 1913).
26
27
‘Estatins cried out: “Bring me my sword!” / His blade was put into his hand,
/ and he drew it from its sheath as if he had gone mad. / Battle commenced
within the palace, / there were so many dead, so many people cut up, so many
thrown down ,/ when Guy the seneschal came up the palace steps/ and said to
298
This episode transforms Estatins into a defender of Christianity against

a corrupted and misled king, unworthy of his faith. Two elements are
particularly striking about this scene. Firstly, it takes place within the
palace, the inner sanctum of the ruler’s power and a demilitarised zone,
and is thus a socially transgressive occurrence. Guy’s objection is not
based on Christians fighting Christians, or vassals of Alexios fighting
vassals of Alexios, but rather on the location of the battle. Second is the
intriguing emphasis on Alexios having received bad counsel. This scene
has notable parallels with scenes in the chanson de geste, Aliscans, a work
with which we know the author was familiar because of the reference to
its heroes, quoted above. In Aliscans, the hero William of Orange travelled
to the king’s court looking for help to fight against the pagans, but was
mocked because of his lack of courtly accoutrements and royal favour. In
a number of scenes, he threatened the king, whom he accused of receiving
bad counsel, with the ultimate threat of resorting to violence within the
boundaries of the palace.28
The argument that the figure of Estatins carries oblique references to
William of Orange is strengthened by the recurring epithet esnasé, mean-
ing ‘with his nose cut off or effaced’, used nine times out of twenty men-
tions of the character in the entire Chanson d’Antioche. For an audience
familiar with the chansons de geste, a figure with a cut-off nose would be
reminiscent of William of Orange, whose nose was cut off in a duel with
the pagan leader Corsolt in the action of the Couronnement de Louis, and
who was thenceforth known by the epithet al curb niés (with the short,
hooked, or cut-off nose).29
These elements of the disfigured nose, his nephew status to the cor-
rupted Alexios, and the battle scene in the palace are not truly enough to
confirm that Estatins was explicitly modelled on or aligned with William
of Orange. Yet reminiscences of the epic hero can be seen in Estatins’
character, which would have coloured the audience’s perception of him,
and transformed him into an heroic figure through which to reconcile
perceived Byzantine antagonism to the crusade with the active Byzantine
military and logistical support of the expedition.
the emperor: “You have taken bad counsel./ Why is there such a battle inside
the palace?”’: Chanson d’Antioche, ed. Duparc-Quioc, 1: 57, lines 896–903.
28
Aliscans, ed. Régnier, 1: 129–38, lines 2962–3278.
29
It has also been argued that Tatikios’ portrayal in Gilo of Paris may contain
references to William of Orange. See The Historia Vie Hierosolimitane of Gilo of
Paris and a Second, Anonymous Author, ed. and trans. Christopher W. Grocock
and Elizabeth Siberry (Oxford, 1997), pp. 116–17.
299
Simon T. Parsons
The Chanson d’Antioche continues to portray Estatins positively, where

other accounts of the First Crusade are more negative. In the earliest Latin
accounts of the First Crusade, the handing over of the city of Nicaea to
Byzantine troops rather than Frankish ones is a source of much discon-
tent.30 In the Chanson d’Antioche, the city is not directly handed over to
the emperor, instead being claimed by Estatins. As soon as this transaction
takes place, the Chanson d’Antioche tells us: ‘si ovrirent les portes, li Franc
i sont entré’ (‘[The Turks] opened the gates, and the Franks poured in’).31
Estatins is thus not demarcated as an imperial vassal: in practice, he is one
of the Franks and his dominance over Nicaea is the Franks’ dominance
as well. There is no hint of the censure found elsewhere that the empire
benefited from the work of the crusaders. When the Christian leaders are
described setting up position to besiege the city of Antioch, ‘Estatins qui
ot cuer de lion’ (‘Estatins who had the heart of a lion’) is included along
with the other major leaders of the crusade.32 In this section, each of the
major leaders, referred to later as ‘nos barons’ (‘our leaders / heroes’), is
devoted a stanza describing their position and their sworn intentions to
starve the city into submission, and Estatins is included among them.33
Estatins’ ‘heroicisation’ and adoption by the westerners as one of their own,
and as a crusader, has the function of removing the awkward factionalism
between Byzantine and Latin Christian components on the expedition
of the Crusade after it departed Constantinople.
Robert of Normandy
The sixth most mentioned figure in the Chanson d’Antioche,34 Robert of
Normandy was one of the high-profile leaders of the First Crusade. A well
attested historical figure in his own right, Robert was duke of Normandy
from the death of his father, William I of England, in 1087, until his defeat,
capture, and imprisonment after the battle of Tinchebrai in 1106. While
the Chanson d’Antioche inserts many heroic episodes about all of the leaders
30
Fulcheri Carnotensis Historia Hierosolymitana, ed. Heinrich Hagenmeyer
(Heidelberg, 1913), I.10.10, pp. 188–89; Fulcher of Chartres: A History of
the Expedition to Jerusalem, trans. Frances R. Ryan, ed. Harold S. Fink
(Knoxville, 1969), p. 83.
31
32
33
34
Lewis A. M. Sumberg, ‘Au confluent de l’histoire et du mythe: La Chanson
d’Antioche’, in Les Epopées de la Croisade, p. 64.
300
of the crusade,35 and Robert is in many ways a typical case study, there are
a few intriguing elements about his portrayal which elicit in-depth con-
sideration. In the earliest, ‘eyewitness’ accounts, he plays a relatively small
role, and although clearly one of the leaders, performs no notable exploits
before the capture of Jerusalem itself: for example, in the Gesta Francorum
and the account of Peter Tudebode he is always mentioned as part of a list
of leaders, never acting alone, until the battle of Ascalon after Jerusalem
has fallen.36 Raymond of Aguilers, apart from introducing him separately
at Nicaea, never mentions him as an independent agent.37 The only action
on crusade ascribed to Robert uniquely in Fulcher of Chartres’ account
is his leaving the town of Ma’arrat al-Nu‘man to join Raymond’s army.38
The earliest sources do not make Robert an active ‘hero’; the Chanson
d’Antioche makes him one of the most significant.
Corbaran, or Karbugha, is the most important adversary in the Chanson
d’Antioche – he is in command, ultimately, of all of the pagan enemies
and organises the fight against the crusaders. In the final set-piece bat-
tle of Antioch, Corbaran is supported by his two primary lieutenants:
Amedelis, and Rouges Lion. Of these three principal pagans, Robert of
Normandy is responsible for unhorsing Corbaran and killing Rouges Lion,
while Amedelis escapes without being defeated in combat. In defeating
Corbaran, Robert is casting down the most powerful of the pagans and,
in a genre so heavily based on single combat between heroes, this is tan-
tamount to turning the tide of the battle. These single combats are both
carried out in traditionally ‘heroic’ epic style:
Li quens de Normendie fu molt de fier regart
Par la bataille vait, plus i ot de soi quart
[…]
Corbarans fu armés, ne sanbla pas Lonbart
L’escu ot a fin or, ne crient lance ne dart,
Ses elmes fu forgiés en la cit da Baudart,
El nazel par devant uns escarboucles art
Hanste ot et fort et roide et si porte falsart,
En l’escu de son col ot paint un papelart.
35
For the ‘heroicisation’ of Godfrey, see Simon John, ‘The Creation of a First
Crusade Hero: Godfrey of Bouillon in History, Literature and Memory,
c.1100–c.1300’ (unpublished PhD thesis, Swansea University, 2012).
36
Gesta Francorum, ed. Hill, pp. 94–97, 112. Petrus Tudebodus: Historia de
Hierosolymitano itinere, ed. John H. Hill and Laurita L. Hill (Paris, 1977), p. 155.
37
‘Liber’, ed. Hill and Hill, p. 164.
38
Historia Hierosolymitana, I.25.5, ed. Hagenmeyer, p. 268. Robert’s journey and
wintering through Italy, before the crusade proper, are also described in detail.
301
Simon T. Parsons
Quant li quens l’aperçoit, vers lui va cele part

Tel coup li a doné desor son touenart
Ke gambes reversees le trebuche el begart.39
It is worth noting that this progression – description of sumptuous arms
and armour before an overthrowing of one’s opponent – is traditional
for heroic encounters in epic, and associates Robert with figures such
as Roland, William of Orange, and Charlemagne.40 But this is to be
expected: the Chanson d’Antioche is, after all, a chanson de geste, and it
utilises the traditional structure and clichés of the genre. The slaying of
Rouges Lion is even more formulaic and epic in tone:
Li quens de Normendie a le ciere menbree
Broce le bon destrier s’a ensegne escriee.
Fiert le Rouge Lion sor se targe roee
Que par desos la boucle li a fraite et troee
Le clavain desrompu et le car entamee,
Parmi le gros del pis li est l’anste passee.
Del ceval l’abat mort, l’arme s’en est alee
En infer le hisdeus illuec est ostelee.41
These epic combats are only one way in which the Chanson d’Antioche
aggrandises the role Robert played on crusade when compared with the
Latin sources. His supposed descent from Doon of Mayence has already
been discussed. He continues to be portrayed as the equal of Godfrey in
39
‘The Count of Normandy was terrible to behold, / through the battle he rode,
with just a small group of men […] Corbaran was armed, and didn’t look at
all like a Lombard. / His shield was made of pure gold, and feared not lance or
dart, / his helm was forged in the city of Baghdad, and on the nose guard was
a burning red gemstone, / his spear was strong and sturdy, with a curved blade,
/ and on the shield that hung from his neck was painted a parrot. / When
the count saw Corbaran, he rode straight towards him, / and gave him such a
blow upon his shield/ that he sent the fool tumbling off his horse, legs in the
air’: Chanson d’Antioche, ed. Duparc-Quioc, 1: 431, lines 8749–64.
40
For discussion and examples of the typical progression of such combats, see
Genette Ashby-Beach, The Song of Roland: A Generative Study of the Formulaic
Language in the Single Combat (Amsterdam, 1985); Jean Rychner, La Chanson
de geste: Essai sur l’art épique des jongleurs (Genève, 1955).
41
‘The count of Normandy, wonderful to behold, / spurred on his fine warhorse,
and shouted his battle cry. / He struck Rouge Lion on his round shield, / which
shattered and was pierced beneath the boss, / his mail coat was penetrated and
his flesh torn apart:/ right through his chest did the lance slice. / He flung him,
dead, from his horse, and his soul departed, / to be housed in the horrific realms
of hell’: Chanson d’Antioche, ed. Duparc-Quioc, 1: 443–44, lines 9029–9036.
302
valour, a surprising depiction given the centrality of the latter to the cru-
sade cycle as a whole, with the duke of Bouillon at one point admitting to
him: ‘Hé! Robert, jentius quens, frans hom, ciere hardie, | Miux valés vos
de moi, ce ne renoi jo mie’ (‘Oh, Robert, noble count, valiant man, and
tough to look at, | you are more worthy than me, I tell no lie.’).42 Before
the final battle of Antioch is joined, Robert, and not Godfrey, gives the
pre-battle speech to the crusaders, emphasising the opportunities ahead
for martyrdom and encouraging the hesitant.43
The reasons for this weight placed upon Robert’s role in the narrative
remain unclear. If we accept the hypothesis that the Chanson d’Antioche
was essentially a Picard undertaking (and this is supported by the language
of the current recension)44 what were the motivations for emphasising
Robert’s feats-of-arms, and not, for example, those of Robert of Flanders,
or indeed Godfrey of Bouillon, thereby cementing the latter’s position as
the main hero of the text and cycle?
Although unrepresented in other crusade sources, this primary role
on crusade attributed to Robert occurs in ‘mainstream’ histories from
the Anglo-Norman sphere of influence. The Gesta Regum Anglorum of
William of Malmesbury (written approximately between 1124 and 1127)
records a story reminiscent of the Chanson d’Antioche where Robert kills
Corbaran in single combat.45 This is continued in the vernacular history
of Geoffrey Gaimar (1136–1137) , which claims ‘pur Curbarant k’il out
oscis | entrat li duc en si halt pris’ (‘For Corbaran, whom he had killed, the
duke got so much respect’),46 and Wace’s Roman de Rou (1160–1174).47
The unedited twelfth- or thirteenth-century Old French crusade poem
contained within the manuscripts Hatton 77 and BM Add. 34114, writ-
ten in an Anglo-Norman dialect, also seem to assign a dominant role
to the heroics of Robert on crusade, implicating him in duels with the
most respected of the pagan leaders. At some point in the mid-twelfth
century, although it is unclear exactly when, a feat whereby ‘R. Dux
42
43
44
La Chanson d’Antioche, ed. Jan A. Nelson, in The Old French Crusade Cycle,
ed. Jan A. Nelson and Emanuel J. Mickel, 10 vols (Tuscaloosa, 1977–2003), 4:
34.
45
William of Malmesbury: Gesta Regum Anglorum, ed. Roger A.B. Mynors,
Rodney M. Thomson, and Michael Winterbottom, 2 vols (Oxford, 1998), 1:
702. For dating see 2: xvii-xxxv.
46
Geffrei Gaimar: Estoire des Engleis, ed. Ian Short (Oxford, 2009), pp. xii, 312.
47
The History of the Norman People: Wace’s Roman de Rou, trans. Glyn S. Burgess
(Woodbridge, 2004), p. 200. For dating see pp. xxiii-xxiv.
303
Simon T. Parsons
Normannorum Partum prosternit’ (‘Robert, duke of the Normans, cast

down the Parthian’) was commemorated in the now lost ‘crusading win-
dows’ at Saint-Denis, which were thankfully preserved in engravings by
Bernard of Monfaucon in the eighteenth century.48
It is not unreasonable to assume that there is a link between the two
traditions: either the Chanson d’Antioche was influenced by the stories
found in twelfth-century historiography in the Angevin sphere, or these
accounts were influenced by a proto-Antioche. Ultimately, this exalta-
tion of Robert’s status remains confounding, but we should consider the
possibility of Anglo-Norman influence in the composition of the extant
Chanson d’Antioche or its preceding tradition.
In summary, the Chanson d’Antioche constructs heroes out of the
participants of the First Crusade in a variety of ways and for a variety
of reasons. Partially due to the shrouded development of the Chanson
d’Antioche tradition, many of these reasons are now obscure. It casts the
crusaders as lords, fighting under God against a monstrous and ‘othered’
enemy for glory and salvation. Allusions, both explicit and implicit,
structural and narratorial, regarding characters and regarding situations,
to the mainstream chansons de geste, the dominant epic literature of the
period, translated the story of the First Crusade from the historic to
the heroic (a distinction less evident to a medieval audience than it is to
us), allowing subsequent poems of the crusade cycle to build a bona fide
mythology upon it. Lewis Sumberg described this process succinctly:
‘C’est ainsi que l’histoire vraie engendre l’histoire légendaire laquelle, a
son tour, fait naître le mythe.’49
Yet this is not unique. Latin sources for the First Crusade carry out the
same process, and frequently contain allusions to the chansons, both gen-
eral and specific. For example, the Gesta Tancredi describes how those who
saw Robert of Flanders and Hugh drive off the Turks at Dorylaeum would
think Roland and Oliver had been reborn.50 Godfrey’s famed bisection of
48
Bernard de Montfaucon, Les Monumens de la monarchie françoise, 5 vols
(Paris, 1729), 1: 277, 384–97. See also Elizabeth A.R. Brown and Michael W.
Cothren, ‘The Twelfth-Century Crusading Window of the Abbey of Saint-
Denis: Praeteritorum enim recordatio futurorum est exhibitio’, Journal of the
Warburg and Courtauld Institutes 49 (1986), 1–40, although the conclusion
that Robert’s victory was the capture of the standard in the battle of Ascalon is
unproven.
49
Sumberg, ‘Au confluent de l’histoire et du mythe’, pp. 64–65.
50
Ralph of Caen, Radulphi Cadomensis Gesta Tancredi, ed. Edoardo d’Angelo
(Turnhout, 2011), p. 30, line 923.
304
a Turk in Albert of Aachen;51 the discussion between Karbugha and his

mother in the Gesta Francorum;52 Karbugha playing chess before battle in
Raymond of Aguilers53 – episodes with echoes of vernacular literature are
plentiful in mainstream crusade accounts. Even those examples of aggran-
disation discussed above have parallels in Latin literature. We have already
discussed the treatment of Robert of Normandy in Angevin literature and
the Saint-Denis windows. Gilo of Paris says of Estatins (Statinus), ‘dum
vivebat nase non laude carebat’ (‘while he lived, he lacked a nose, but he
did not lack praise’).54 It is clear that the Chanson d’Antioche belongs to
the same tradition as these chronicles.
In this, the Chanson d’Antioche should be viewed not as the beginning
of an independent tradition of mythologising the events of the First
Crusade, but rather in continuity with the Latin accounts, which carried
out the same processes: those of aggrandisation, of making the stories
and exploits more fantastic, and of constructing heroic identities for the
crusade participants. The Latin texts have their own emphases on heroic
deeds by certain participants and these emphases should not be accepted
as valid and representative of historical truth while those of the Chanson
d’Antioche are rejected as untrustworthy unless the source material of the
Chanson d’Antioche can be established with absolute certainty – and this
is patently a long way from being the case.
51
Albert of Aachen, Historia Ierosolimitana, III.65, ed. Edgington, pp. 244–45.
See also John, ‘The Creation of a First Crusade Hero’, pp. 70–72, for discussion
of this feat of arms.
52
Gesta Francorum, ed. Hill, pp. 53–56.
53
‘Liber’, ed. Hill and Hill, p. 80.
54
Gilo of Paris, Historia Vie Hierosolimitane, ed. Grocock, p. 119.
305
The Count and the Cannibals
The Old French Crusade Cycle as a Drama of Salvation*
Carol Sweetenham
Independent Scholar
What more miraculous undertaking has there been (other than the mys-
tery of the redeeming Cross) than what was achieved in our own time by
the journey of our own people to Jerusalem? […] this was not the work of
men; it was the work of God.
The First Crusade was seen as literally miraculous by contemporaries,
as these words of Robert the Monk suggest.1 Over successive crusades it
became increasingly clear that it was the only crusade to have achieved its
goal, the liberation of Jerusalem. As such it gradually made the transition
from history to legend: the events of the crusade formed the basis of the
early thirteenth-century Old French Crusade Cycle, a cycle of chansons de
geste standing alongside the cycles describing Roland and William, while
Godfrey of Bouillon was elevated to one of the Nine Worthies alongside
Alexander, David and Charlemagne.2
*
Earlier versions of this paper were given at the 2011 conference of the British
branch of the Société Rencesvals, and the 2011 Pécs conference of the Medieval
Chronicle Society. It has benefited considerably from discussion at both, and
in particular from insights by Philip Bennett, David Pattison, Patricia Gillies
and Nicholas Coureas: they of course bear no responsibility of any kind for the
use I have made of their comments.
1
‘Post creationem mundi quid mirabilius factum est praeter salutiferae crucis
mysterium, quam quod modernis temporibus actum est in hoc itinere
nostrorum Iherosolimitanorum?[…] Hoc enim non fuit humanum opus, sed
divinum’. The Historia Iherosolimitana of Robert the Monk, ed. Damien Kempf
and Marcus Bull (Woodbridge, 2013), p. 4; Robert the Monk’s History of the
First Crusade, trans. Carol Sweetenham (Aldershot, 2005), p. 77.
2
The Old French Crusade Cycle, ed. Jan A. Nelson and Emanuel J. Mickel, 10 vols
(Tuscaloosa, 1977–2003). On the Nine Worthies, see Maurice Keen, Chivalry
(New Haven, 1984), pp. 121–24.
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. –.
Carol Sweetenham
But no human enterprise, even a divinely inspired one, can be perfect.

And the First Crusade certainly had its darker side: ‘the luminosity of
the story of triumph over adversity in the cause of God … casts shade as
well as light’.3 Greed for land and possessions, indiscriminate slaughter,
fornication, cannibalism, cowardice and treachery are all evidenced if
not emphasised in the sources.4 As a rule this less edifying behaviour is
passed over, explained away or used to settle scores.5 However the early
thirteenth-century Old French Crusade Cycle takes a different approach.
It uses and indeed emphasises that darker side to underpin the central
message of the crusade: its power as an instrument of redemption. This
in turn underlines the central exhortatory message of the text.
The Crusade as a Means of Redemption

The exact nature of what Urban II offered potential crusaders in 1095
has been discussed at length. Briefly, it seems clear that crusaders were
offered remission of sins provided they fulfilled their vows, and that this
covered the fighting and killing which would inevitably form part of the
crusade.6 Whilst it is not clear that martyrdom was officially offered by
3
Christopher Tyerman, God’s War: A New History of the Crusades (London,
2006), p. 58.
4
For plunder at the fall of Antioch and capture of Jerusalem see Fulcher of
Chartres I.17.7 and I.28.1–2: Fulcheri Carnotensis Historia Hierosolymitana
(1095–1127), ed. Heinrich Hagenmeyer (Heidelberg, 1913), pp. 234–35,
301–3. For massacres see Fulcher, Historia Hierosolymitana, I.27.12–13, ed.
Hagenmeyer, pp. 300–1. For fornication see Gesta Francorum et aliorum
Hierosolymitanorum, ed. Rosalind Hill (London, 1962), p. 58. For cowardice
see flight of William the Carpenter, Gesta Francorum, ed. Hill, pp. 33–34. For
treachery see Raymond of Aguilers, Historia Francorum qui ceperunt Jerusalem,
trans. John H and Laurita L. Hill (Philadelphia, 1969), p. 59, where Christians
tell the Turks about conditions in Antioch.
5
For passing over embarrassing truths, see Fulcher, Historia Hierosolymitana,
I.14, who carefully avoids giving details on how Baldwin gained control of
Edessa. For explaining the awkward see Gesta Francorum, ed. Hill, p. 12, which
argues that the leaders swore an oath to Emperor Alexios Komnenos through
force of circumstance. For settling scores see the waspish pen portraits of the
leaders of the crusade by Ralph of Caen, Radulphi Cadomensis Gesta Tancredi,
ed. Eduardo D’Angelo (Turnhout, 2011), pp. 18–19.
6
There is extensive literature on these issues and this is the briefest of summaries.
See Jonathan Riley-Smith, The First Crusade and the Idea of Crusading
(London, 1986), pp. 27–29; Tyerman, God’s War, pp. 66–67.
308
The Old French Crusade Cycle as a Drama of Salvation
the pope as a reward for those who died, the connection was made early
and is found in contemporary sources.7 Those who failed to fulfil their
vows were threatened with excommunication and shame.8 So crusade
was a route to redemption, and those who failed to fulfil their vows were
beyond redemption until and unless they completed them.
By the end of the twelfth century the rhetoric had evolved. The loss of
Jerusalem in 1187 had sent a shock wave through the west, and the need
to recover it gave a new impetus to crusade preaching. Against this back-
drop the memory of the First Crusade a century before acquired a new
significance: not only had it been the only successful crusade, but that
success provided a precedent for what was now seen to be needed, the
reconquest of Jerusalem.9 Innocent III stressed crusade as a repayment for
the forgiveness of sin which brought spiritual rewards; conversely those
who shirked crusading were roundly condemned.10 Allied to this were
concerns about the number of lapsed crusaders which both Celestine III
and Innocent III tried to tackle at the end of the twelfth century.11
The Old French Crusade Cycle: Bad Behaviour on

Crusade and How it can be Redeemed (or not)
The Old French Crusade Cycle dates from the first half of the thirteenth
century. The three texts which lie at its heart portray the First Crusade:
the Chanson d’Antioche, the Chanson des Chétifs and the Chanson de
Jérusalem.12 These offer a fictionalised account of the crusade written a
7
H. E. John Cowdrey, ‘Martyrdom and the First Crusade’, in Crusade and
Settlement, ed. Peter Edbury (Cardiff, 1985), pp. 46–56; Riley-Smith, The First
Crusade, pp. 114–18; Penny Cole, The Preaching of the Crusades to the Holy
Land, 1095–1270 (Cambridge, MA, 1991), pp. 30–31.
8
Riley-Smith, The First Crusade, pp. 124–25.
9
James M. Powell, ‘Myth, Legend, Propaganda, History: The First Crusade,
1140-ca.1300’, in Autour de la Première Croisade, ed. Michel Balard (Paris,
1996), pp. 127–41 (here pp. 136–37).
10
Cole, Preaching, pp. 80–83.
11
Tyerman, God’s War, pp. 483–84.
12
La Chanson d’Antioche, ed. Suzanne Duparc-Quioc, 2 vols (Paris, 1976–78);
The Chanson d’Antioche , ed. Jan Nelson, The Old French Crusade Cycle, 4
(Tuscaloosa, 2003); ed. Bernard Guidot, La Chanson d’Antioche: Chanson de
geste du dernier quart du xiie siècle (Paris, 2011); The Chanson d’Antioche: An
Old French Account of the First Crusade, trans. Susan Edgington and Carol
Sweetenham (Farnham, 2011). La Chanson de Jérusalem, ed. Nigel Thorp,
309
Carol Sweetenham
century later, albeit drawing on earlier source material.13 But there is a seri-
ous purpose behind the entertainment. It is very clear from the prologue
of the Chanson d’Antioche that there is an exemplary function: the text
exhorts its listeners to emulate their forebears and go to the Holy Land
to fight the infidel. If they do, they can expect forgiveness of sins, and
salvation if they die in the attempt:
Nostre Sire vos rueve en Jherusalem aler,
Le deffaee gent ocire et afoler,
Ki Deu ne voelent croire ne ses fais aorer
Ne ses commandements volentiers escouter […]
[…] Li bon baron de France ne vaurent arrester,
En estranges pais s’alerent deserter,
La devinrent salvage por lor armes salver,
Ki la puet par bon cuer son cors por Deu pener,
Jhesus li rois de gloire nel vaura oblier,
Ains le fera en glorie hautement coroner.14
The message in the Old French Crusade Cycle is clear and simple and
expressed repeatedly in what is in effect a prologue of some 200 lines.
Crusade redeems sin, and the First Crusade is an example:
Segnor, oés cancon de grant cevalerie’
Des bons barons de France drois est que jo vos die
Ki par force conquisent la celestiel vie.
Il n’orent cure entr’els ne engine ne envie […]
[…] Ki moru par boin cuer Diex ne l’oblia mie,
Les armes en a mis en molt grant segnorie.15
The Old French Crusade Cycle, 6 (Tuscaloosa, 1992); La Chanson des Chétifs,
ed. Geoffrey Myers, The Old French Crusade Cycle, 5 (Tuscaloosa, 1980).
13
A case powerfully made by Robert Cook in Chanson d’Antioche, chanson de
geste: Le cycle de la croisade est-il épique? (Amsterdam, 1980). For summary of
recent scholarship and discussion see The Chanson d’Antioche, trans. Edgington
and Sweetenham, Introduction, ch. 1.
14
‘Our Lord asks you to go to Jerusalem to kill and confound the wicked
pagans who refuse to believe in God and adore His works or pay heed to His
commandments […] The noble lords of France did not hesitate. They went to
wander through the desert in foreign lands and saved their souls by becoming
savages. If anyone willingly allows his body to be broken for God, Jesus King of
Glory will not forget him: on the contrary, He will crown him in glory on high’.
Chanson d’Antioche, all references are to the Duparc-Quioc edition, here 1: 24,
lines 128–31; 1: 24–25, lines 138–43.
15
‘My lords, listen to a song of true knightly achievements: it is only fitting that
I should tell you about the noble French lords who by force of arms gained
310
Conventional and Unconventional Heroism in the Old French

Crusade Cycle
By and large the main characters in the trilogy are distinguished by a
somewhat bland heroism befitting crusade heroes. Heroic single com-
bats are pretty evenly distributed, and each leader is credited with some
achievement.16 More interestingly, lower rank knights such as Gontier of
Aire receive time in the spotlight.17 In particular Rainalt Porcet’s capture
and refusal to apostasise are described in some detail.18 Some knights
whom we know from other sources to have been bywords for violence are
portrayed as straight heroes. Raimbaut Creton, for example, had castrated
a monk.19 Most famously, Thomas of Marle, a byword for wickedness
who was anathematised and ultimately attacked and killed by Louis VI,
is portrayed as the hero of the Chanson de Jérusalem, being flung over the
city wall on the spears of the Christian army.20 The question of redemption
through crusade does not even arise. Presence and heroism are enough to
ensure salvation and posthumous fame; and there is a subliminal message
that going on crusade is a good way for those with relatively little power
and influence to improve their standing.
celestial life […] God did not forget anyone who died there willingly: He raised
their souls on high.’ Chanson d’Antioche, 1: 23, lines 105–8, 1: 24, lines 118–19.
16
See for example the start of the battle of Antioch, where Bohemund, Hugh,
Godfrey, Robert of Flanders, Robert of Normandy and Tancred all receive
honourable mention in less than 200 lines: Chanson d’Antioche, 1: 423–31, lines
8593–767.
17
Chanson d’Antioche, 1: 173–75, lines 3058–108: Gontier captures the prestigious
warhorse Faburs and is rewarded by being made a knight and seneschal.
18
Chanson d’Antioche, 1: 205–6, lines 3766–79; 1: 212–17, lines 3899–4038, 1:
224–32, lines 4209–425.
19
Chanson d’Antioche, 1: 208–12, lines 3815–93 for Raimbaut’s exploit at the bridge
in Antioch: line 3893: ‘Molt fu bons chevaliers et de tos fu amés’. For the monk see
PL 162, cols 144–45, letter 135 from Ivo of Chartres to Pope Paschal: Raimbaut
castrated the monk in revenge for his servants’ stealing hay, and a penance of
fourteen years involving removal of arms, fasting and alms was imposed.
20
Suger, Vie de Louis VI le Gros, ed. Henri Waquet (Paris, 1964), ch. 7, pp. 31–33 for
his eviction from his castle of Montaigu; ch. 24, pp. 173–79 for his condemnation
by the Council of Beauvais; ch. 31, pp. 251–53 for his death and inability to receive
Communion. Thomas was ‘virum omnium quos novimus hac aetate nequissimum’
(‘the most infamous man of his time we have come across’) according to Guibert
of Nogent, Autobiographie, ed. and trans. Edmond-René Labande (Paris, 1981),
pp. 328–29; Chanson de Jérusalem, ed. Thorp, lines 4743–54, 4807–8.
311
Carol Sweetenham
However, there are a number of episodes where discreditable behaviour

is deliberately highlighted. There is the repeated cowardice of Stephen of
Blois. There is the gory emphasis on cannibalism by the Tafurs at Antioch.
And there is a humorous counterpoint in the story of the theft of Pierre
Postel’s donkey. At first sight this seems surprising in a work which puts such
a firm emphasis on heroism. Why stress so strongly anti-heroic behaviour?
The Count: Stephen of Blois …

Stephen of Blois is vilified throughout the Chanson d’Antioche: indeed his
cowardice at each new challenge becomes something of a topos.
Quant l’entendi Estievenes, s’en ot al cuer tencon,
Il vausist a cele eure ester a Blois se maison.
Tos tramble dusqu’en terre quant prist le gonfanon.21
In his final appearance he declares that he is too ill to fight and must retreat
to Alexandretta. He is carried out of Antioch on a litter by twelve paupers;
the moment he is out of sight of the army he leaps from the litter and
literally runs away, forcing his bearers to go with him in case they go back
to the army and tell the truth.22 The army, unable to count on terrestrial
help from the Byzantine emperor, receives instead something much more
potent: help from the warrior saints on the battlefield at Antioch in an
unequivocal demonstration of divine support for the crusaders’ cause.
Up to a point, this is accurate. Stephen’s reputation in the sources
for the First Crusade is that of the failed crusader. Although he appears
to have held some kind of leadership role, he seems to have made little
impact. Faced with what looked like the overwhelming certainty of defeat
at Antioch, he left the crusade to return to Byzantium. Here he appears
to have told Emperor Alexios that the crusaders were beyond help. His
subsequent return to the Holy Land with the crusade of 1101 and heroic
death pass with little comment.23
21
‘When Stephen heard this his heart was torn in two: how he wanted at that
moment to be back in his house in Blois! As he took the pennon he trembled
all over.’ Chanson d’Antioche, 1: 86, lines 1438–40; for his cowardice at Nicaea 1:
86–87, 92–93, 94–95, lines 1428–62, 1557–80, 1610–22.
22
Chanson d’Antioche, 1: 285–86, lines 5608–44.
23
James L. Cate, ‘A Gay Crusader’, Byzantion 16 (1942–43), 503–26; James
Brundage, ‘An Errant Crusader: Stephen of Blois’, Traditio 16 (1960), 380–95;
Paul Rousset, ‘Etienne de Blois: Croisé, fuyard et martyre’, Geneva 11 (1963),
183–95. The sources’ views of Stephen are clear: Gesta Francorum asserts that
his illness was feigned: ‘turpiter recessit’ and ‘vehementique captus timore
312
Quite how Stephen’s actions led to his reputation being so compre-

hensively blackened is an interesting question. The link is Bohemund
and his determination to legitimise his possession of Antioch. Stephen’s
relationship with Byzantium was in sharp contrast to that of Bohemund.
Bohemund initially tried to gain a position of trust at the Byzantine
court. His later claim to Antioch hinged crucially on the emperor’s
failure to supply help during the sieges first outside and then inside the
city: this, Bohemund argued, freed him from his obligation to hand
Antioch to the emperor since the latter had refused to assist him while
he was effectively acting as a vassal.24 Stephen, meanwhile, had built
up a good relationship with Alexios, and was clearly believed when he
reported to the emperor that the crusade was on the verge of failure.
How far the two were effectively rivals on the crusade is hard to say.
But it is interesting that the capture of Antioch was orchestrated by
Bohemund on the evening of 3 June 1098, a few hours after Stephen
had finally left the siege.
A portrayal of Stephen, therefore, which not only depicted him as
the only leader to fail to complete the crusade but also as the cause of
the emperor’s failure to send help underlined by contrast Bohemund’s
heroism and right to Antioch.25 In this respect it is striking that Hugh of
recessit’, ed. Hill, p. 63. Fulcher of Chartres, Historia Iherosolimtana, I.16.7, ed.
Hagenmeyer, p. 228, describes him as a disgrace. Somewhat later William of
Malmesbury is critical: ‘clam effugit, mendatiis suis advertantes retro agens’:
Gesta Regum Anglorum, ed. and trans. Roger A.B. Mynors, Rodney Thomas and
Michael Winterbottom, 2 vols (Oxford, 1998–99), 1: 634–35. Orderic Vitalis
comments ‘magna detentus aegritudine ut asserebat’, IX.9: The Ecclesiastical
History of Orderic Vitalis, ed. and trans. Marjorie Chibnall, 6 vols (Oxford,
1969–80), 5: 90–91; Orderic later gives a more detailed and critical description
of Stephen climbing a mountain, fleeing in terror and dissuading the Emperor
from sending help (IX.10, 5: 106–7). For Stephen’s further departure and death
on crusade in 1101 see William of Malmesbury, Gesta Regum, ch. 383–84, ed.
Mynors et al., 1: 680–85 and Orderic Vitalis, Ecclesiastical History, X.22, ed.
Chibnall, 5: 342–51. William of Tyre, 10.9–44 is equally clear that the illness
was feigned and Stephen’s actions damaging: Willelmi Tyrensis Archiepiscopi
Chronicon, ed. Robert B. C. Huygens, 2 vols (Turnhout, 1986), pp. 284–85.
24
William B. McQueen, ‘Relations between the Normans and Byzantium 1071–
1112’, Byzantion 56 (1986), 427–76.
25
A point made explicitly by Baldric of Bourgueil, Historia Jerosolymitana, III.13,
in RHC Occ. 4: 1–111 (at p. 73): ‘Ex Dei tamen dispositione factum nequaquam
dubitamus, qui disposit omnia suaviter […] Si enim Alexius imperator advenisset
Turcosque superasset, triumpuis genti suae non exercitio Dei ascriberetur et
313
Carol Sweetenham
Vermandois, who also failed to complete the crusade, attracts no criticism:

indeed he is portrayed as one of its main heroes in Robert the Monk’s
Historia Iherosolimitana.26 Stephen’s reputation became increasingly
negative: the crusader who ran away.27
However both the emphasis on Stephen and the vitriol directed at him
in the Chanson d’Antioche outstrip anything in any other source. To some
extent this may reflect the passing of time: it is easier to attack someone
who had died a century before, and there had been ample time for fic-
tion to mould and change actual fact. But in the theology of redemp-
tion through crusade, Stephen is guilty of the ultimate sin: he ran away.
Our final view of him is, quite literally, running away in the company
of paupers. His departure could not be more ignominious. He did not
accomplish his crusading vow and therefore cannot achieve redemption.
This explains the viciously critical depiction in the text. Stephen is the
anti-exemplar of redemption. His role is to serve as the counterweight
to the depiction of redemption and salvation, the example not to follow.
Indeed his cowardice is a convenient device for the author: if it had not
existed it might have had to be invented.
… the Cannibals: the Tafurs …

The Tafurs are an intriguing and shadowy group in the sources for the
crusade.28 They are part of the Christian army but only just, a ragged
and vicious force of 10,000 led by their own king. They are monstrously
Graecorum multitudini deputaretur, non Francorum fortitudini.’ Ralph is

very clear that Stephen bore responsibility for dissuading the emperor: Gesta
Tancredi, ed. D’Angelo, pp. 66–67.
26
Marcus Bull, ‘The Capetian Monarchy and the Early Crusade Movement:
Hugh of Vermandois and Louis VII’, Nottingham Medieval Studies 40 (1996),
25–46; Robert the Monk, trans. Sweetenham, p. 19.
27
For general criticism of crusaders who failed to fulfil their vows see Elizabeth
Siberry, Criticism of Crusading 1095–1274 (Oxford, 1985), pp. 47–70.
28
See Lewis Sumberg, ‘The ‘Tafurs’ and the First Crusade’, Medieval Studies 21
(1959), 224–46; Michel Rouche, ‘Cannibalisme sacré chez les croisés populaires’,
in La Réligion Populaire: Aspects du christianisme populaire à travers l’histoire,
ed. Yves-Marie Hilaire (Lille, 1981), pp. 29–41; Jill Tattersall, ‘Anthropophagi
and Eaters of Raw Flesh in French Literature of the Crusade Period’, Medium
Aevum 57 (1988), 240–53; Jay Rubenstein, ‘Cannibals and Crusaders’, French
Historical Studies 31 (2008), 525–52; Magali Janet, ‘Les scènes de cannibalisme
aux abords d’Antioche dans les récits de la première croisade: Des chroniques à
la chanson de croisade’, Bien dire et bien aprandre 22 (2004), 179–91. On poverty
314
emaciated, diseased, in rags and wielding unknightly weapons; indeed

they are barely human.
La peüssiés veïr tant vies dras depanés
Et tante longe barbe et tans ciés hurepés,
Tans magres et tans ses et tans descolorés,
Et tante torte eskine et tans ventres enfl’es,
Et tante jambe torte et tans pies bestornés,
Et tans mustiaus rostis et tans caukains crevés.
Portent haces danoises et coutels acerés,
Gisarmes et macues et pels en son arsés.
Li rois porte une fauc dont l’aciers fu tenprés;
Qui il en ataindra molt ert mal assenés.29
In the Chanson d’Antioche they appear several times and are guilty of the
worst excesses. While they are accused of raping Saracen women, what
they are notorious for is cannibalism. Encouraged by Godfrey and Peter
the Hermit, they respond to the desperate famine by digging up and eat-
ing the corpses of Turks outside the walls of Antioch (not, as historically
attested, Ma’arrat al-Nu‘man). They comment pointedly that the Saracens
taste better than pork or bacon, which no Muslim would ever touch. In
a grotesque parody of the Eucharist Godfrey offers them a bottle of wine
to wash the meat down, though the text is careful to point out that they
had no bread.30
and the social structure of the crusade see Conor Kostick, The Social Structure of
the First Crusade (Leiden, 2008).
29
‘Had you been there you would have seen countless tattered rags, countless
long beards and matted heads of hair, countless thin and desiccated and faded
[bodies], countless twisted spines and swollen stomachs, countless crippled legs
and feet sticking out at all angles, countless burnt calves and worn-out shoes.
They carried Danish axes and sharp knives, thrusting weapons and maces and
stakes. The King carried a scythe of tempered steel: anyone he attacked with it
would be very badly injured’. Chanson d’Antioche, 1: 407, lines 8256–65.
30
For rape Chanson d’Antioche, 1: 317, line 6413; cannibalism 1: 217–21, lines
4039–118; references to pork and bacon 1: 219, line 4074. There was a
persistent medieval legend that Muhammad was killed and eaten by pigs
whilst in a drunken stupor: see Guibert of Nogent, Dei gesta per Francos et
cinq autres texts, ed. Robert B. C. Huygens (Turnhout, 1996), I.4, p. 100; Le
Couronnement de Louis, ed. Ernest Langlois (Paris, 1984), lines 851–52. For
parody of the Eucharist, see Chanson d’Antioche, ed. Duparc-Quioc, 1: 218, 220,
lines 4054, 4102–6. Rouche argues that there is a deliberate parallel with the
Eucharist, with the cannibalism taking place at the start of Lent and the flesh
being like the Israelites’ manna: ‘Cannibalisme sacré’, pp. 33, 36.
315
Carol Sweetenham
The portrayal of the Tafurs in the Chanson de Jérusalem is puzzlingly

different. They are certainly the same people as in the Chanson d’Antioche.
The physical description is familiar: they are skinny, dressed in rags and
use unknightly weapons.31 While there is no actual incident of cannibal-
ism in the Chanson de Jérusalem, they are referred to in passing as can-
nibals (cist manguent no gen) and are portrayed deliberately acting the
part to frighten the enemy.32 Their poverty is, if anything, more heavily
emphasised: they are referred to as ribals and l’autre gent menue, and the
contrast is commented on when Godfrey dresses them in rich clothes to
fool the enemy.33 In fact more than once there seems to be a pointed con-
trast between them and the better off: when the Tafur king is wounded
twenty times in the assault on Jerusalem, two princes lean down from
their chargers to ask how he is and the bishop of Mautran is pointedly
described as molt bien noris by contrast.34
However their role is very different and much more developed com-
pared to their activities in the Chanson d’Antioche. They operate as a
quasi-autonomous part of the army: thus at the siege of Jerusalem their
encampment is described separately and in more detail than that of
Godfrey.35 It is striking how often the Tafur king is seriously wounded,
something which does not happen to the other leaders on the crusade.36
They serve as a workforce for the army: they gather wood for revet-
ments, and a couple of times are used to play a variant of the old school
photograph trick on the Saracens.37 Despite this apparently subordinate
role they command respect. The Tafur king is treated as an equal by the
leaders of the crusade, addressed by Robert of Flanders for example as
bels cosin.38 More than that, they are accorded two signal honours. They
launch the first attack on Jerusalem. They are second in after Thomas of
31
Chanson de Jérusalem, ed. Thorp, lines 1814–36, 3010–25. Interestingly the text
says they had burnt and leathery skin (line 1820): this is a symptom of scurvy.
I am indebted to Dr Piers Mitchell for this information.
32
Chanson de Jérusalem, ed. Thorp, lines 6424; 6417–24; 7302–3.
33
Chanson de Jérusalem, ed. Thorp, lines 5888; 6378–411.
34
Chanson de Jérusalem, ed. Thorp, lines 2250–53; line 3033.
35
Chanson de Jérusalem, ed. Thorp, lines 1974–85.
36
Chanson de Jérusalem, ed. Thorp, lines 2250; 3019–20; 3426–30.
37
Chanson de Jérusalem, ed. Thorp, lines 6387–93; 7290–97. They run from the
beginning of the line of soldiers parading for the Saracens to the end so that the
numbers look larger. On the association of trickery and outlaws see Timothy S.
Jones, Outlawry in Medieval Literature (New York, 2010), pp. 89–90.
38
Chanson de Jérusalem, ed. Thorp, lines 1858.
316
Marle, who has meanwhile declared himself liegeman to the Tafur king.39
This in turn buys the latter the privilege of crowning Godfrey king on the
Temple Mount at the instance of Raimbaut Creton and Godfrey himself.40
The Tafurs also play a leading role in other encounters. The Tafur king
is the first to volunteer to stay in Jerusalem to support Godfrey. When
Raymond of Toulouse is captured the Tafur king and Peter the Hermit
ride out to rescue him. They are often seen alongside Godfrey: thus it is
Godfrey who gives the signal for them to launch an attack on Jerusalem
and has the horn sounded for them to retreat. They are also found in
company with Peter the Hermit, for example fighting alongside him at
the Gate of David.41
It is striking that they have their own king, who is accorded particular
prominence in the Chanson de Jérusalem. In part this marks their status
as a contingent of the army with its own leader on a par with the other
contingents. But it also acts as a subversion of the established order.
The Tafur king is a kind of mock king acting in counterpoint to the
main army: ‘misrule was thought to be allied to rule as an interrelated
opposite’.42
The Old French Crusade Cycle deliberately plays up the contrast
between their behaviour at Antioch and their behaviour at Jerusalem. In
the Chanson d’Antioche the cannibalism is underlined by heavy humour
and lip-smackingly exaggerated relish. The Tafurs make jokes about it.
The text deliberately emphasises the episode, describing the detail of
how the bodies were cooked and dwelling on the horrified reactions of
the Saracens.43 It uses laisses similaires to amplify the episode, something
reserved for significant points.44 The episode is referred back to later at
39
Chanson de Jérusalem, ed. Thorp, lines 3351–57; 4636–812.
40
Chanson de Jérusalem, ed. Thorp, lines 5324–29; 5335–40.
41
Chanson de Jérusalem, ed. Thorp, lines 5418–30; 5760–74; 3405–30; 5912–30.
42
Sandra Billington, Mock Kings in Medieval Society and Renaissance Drama (Oxford,
1991), p.1. Interestingly an official in Paris was designated as the Roi des Ribauts:
Frédéric Godefroy, Dictionnaire de l’ancienne langue francaise, 10 vols (Paris,
1880–1902), 7: 183c; Arne Terroine, ‘Le roi des ribauds de l’Hôtel du roi et les
prostituées parisiennes’, Revue historique de droit francais et étranger 56 (1978), 253–67.
43
Chanson d’Antioche, 1: 219, lines 4070–72: ‘Voiant paiens, les ont par pieces
decolpés, / En l’eve et es carbons les ont bein quisinés. / Volontiers les manjuent,
sans pain, tos dessalés’.
44
See for example Chanson d’Antioche, 1: 117–20, laisses 95–97 for the death of
William Fitzmarquis; 1: 126–28, laisses 103–5 for Soliman’s lament; 1: 161–71,
laisses 127–39 for establishing siege of Antioch; 1: 391–409, laisses 315–27 for
forces marching out for battle of Antioch.
317
Carol Sweetenham
Antioch.45 By contrast in the Chanson de Jérusalem the Tafurs are referred

to repeatedly as the mainstay of the army.
We cannot explain their presence in the text as a part of the story which
could not be omitted because it was so well known. Several sources for
the crusade do refer to cannibalism, but do not emphasise it and with one
exception do not link it to a specific group.46 Cannibalism was regarded as
a rare event to be treated with extreme repugnance.47 But clearly something
of the kind did happen on crusade under conditions of extreme famine
at Ma’arrat al-Nu‘man; and it was known to have happened to an extent
where it could not be ignored.48 The sources tackle this unpalatable truth
in three ways. Firstly, they use avoidance strategies, asserting that it was
rumour rather than fact and that only a few people did it.49 Secondly,
they dissociate by describing it in terms of extreme revulsion.50 Thirdly,
and increasingly evident in later sources, it is described as a ceremonial
parade of apparent cannibalism at Antioch to demoralise the Saracens.
According to William of Tyre, at Antioch Bohemund staged an elabo-
rate pantomime of cooking Turks to frighten spies he suspected to be in
the camp.51 This particular topos was to have a long life: it reappears in
glorious detail in the Middle English romance about Richard I, where
45
Chanson d’Antioche, 1: 408, lines 8286–87 before the battle of Antioch: 8286,
‘plus desirent car d’ome que cisnes enpeuvrés’; lines 8921–31 at the battle: 8930,
‘A celui qui le voit vis est que le manjue’.
46
Rubenstein, ‘Cannibals and Crusaders’, pp. 527–40, gives a useful summary.
47
Cannibalism was ascribed to Cain: see John Block Friedman, The Monstrous
Races in Medieval Art and Thought (Cambridge, MA, 1981), pp. 95–96. Raoul
Glaber IV.10–11 describes with horror the extremes of cannibalism caused by
famine in the 1030s: Raoul Glaber, Histoires, ed. Mathieu Arnoux (Turnhout,
1996), p. 242: ‘horret denique referre que tunc generi humano corruptiones
acciderunt. Heu proh dolor! Quod olim raro audiri contigerit, tunc rabia edies
compulit vorari ab hominibus humanas carnes.’
48
Gesta Francorum, ed. Hill, p. 80; Fulcher of Chartres, Historia Iherosolimitana,
I.25, ed. Hagenmeyer, pp. 265–81.
49
Guibert of Nogent, VII.23, Dei gesta, ed. Huygens, p. 311, ‘furtim et quam
rarissime factum constat, atrox apud gentiles fama percrebuit’. William of
Malmesbury, Gesta regum, ch. 362, ed. Mynors et al., 1: 634–35 describes the
cannibalism taking place in the mountains.
50
William of Tyre, Chronicon, VII.11, ed. Huygens, p. 357, ‘tam immundis et
pestilentibus cibis’.
51
William of Tyre, Chronicon, IV.23, ed. Huygens, p. 266. In the work of Adhemar
of Chabannes Roger de Tosny – another Norman – cuts up and boils a Saracen
then pretends to feast on him quasi porcum: Ademari Cabannensis Chronicon,
318
he has the heads of Saracen prisoners cooked and served to Saracen

ambassadors complete with nametags, pretends extreme offence at their
queasy refusal of his hospitality, and finally allows them to leave the table
completely demoralised.52 However none of these texts is specific about
the perpetrators. The only source to describe a group analogous to the
Tafurs is Guibert of Nogent. They are poor and ragged, led by an ex-knight
and used for heavy manual labour in the army. Their leader ‘began to be
called the Tafur king in their barbarous tongue’.53 They are not cannibals
themselves; they observe cannibalism by others and use it to demoralise
the enemy.54 This description contains much of what is in the Old French
Crusade Cycle in embryo: their poverty, a clearly identified leader, their
use for manual tasks and their pretence of cannibalism. We do not know
whether the Old French Crusade Cycle borrows directly from Guibert
(there is little other evidence that it did so) or whether it reflects some
kind of oral tradition which Guibert also picked up. What it does do is
to combine the memory of the Tafurs as a group with the cannibalism
attested at Ma’arrat al-Nu‘man and the parade of cannibalism at Antioch.
The author goes out of his way to include them, and indeed to give them
far more emphasis in the Old French Crusade Cycle than in any other text.
There are some obvious superficial explanations for this portrayal.
The Tafurs are picaresque, a foil to the rest of the army and bringers of
(rather dubious) light relief: ‘played as scenes of high comedy’.55 Clearly as
such they caught the imagination: in the later adaptation of the text, the
Chevalier au Cygne et Godefroid de Bouillon, they appear nine times and
their cannibalism is heavily emphasised.56 It could be argued that their
cannibalism underlines the sub-human status of the Saracens, although
III.36–56, ed. Pascale Bourgain (Turnhout, 1999), p. 174. This suggests that the
episode may owe as much to fictional topos as to reality.
52
Der Mittelenglische Versroman über Richard Löwenherz, ed. Karl Brunner
(Leipzig, 1913), lines 3410–520. On the importance of ritual in political
discourse see Timothy Reuter, ‘Velle sibi fieri in forma hac: Symbolic Acts in
the Becket Dispute’, in Medieval Politics and Modern Mentalities, ed. Janet L.
Nelson (Cambridge, 2006), pp. 167–90.
53
Guibert of Nogent, Dei gesta, VII.23, ed. Huygens, p. 310.
54
Guibert of Nogent, Dei gesta, VII.23, ed. Huygens, pp. 310–12.
55
Tattersall, ‘Anthropophagi’, p. 249.
56
Le Chevalier au Cygne et Godefroid de Bouillon, ed. Baron Reiffenberg, 3 vols
(Bruxelles, 1846–59). See for example 2: 469–72, lines 16665–752, where
Godfrey explains their approach in excruciating detail to Cornumarant; 3: 134,
lines 22968–91 where they roast Saracens and leave their bodies lying about to
scare their opponents.
319
Carol Sweetenham
this is not a strong theme in the Old French Crusade Cycle. They also
provide a useful way of diverting blame. Crimes committed on the cru-
sade – rape, cannibalism and indiscriminate slaughter – are firmly laid at
their door. This is classic displacement technique: the Tafurs are singled
out as a group very different from the rest of the army and can safely be
inculpated. A similar technique is used in the Canso de la Crozada, where
the ribautz in the army are blamed for the sack of Béziers.57 The Tafurs
are described frequently as ribaut.58 This is a pejorative term used of a
general bad lot, sometimes following the army in hopes of pillage and
on occasion contrasted with the main army.59
None of this, however, explains their sudden metamorphosis into
heroes in the Chanson de Jérusalem. The answer goes straight back to
the central ideology of the crusade: that to go on crusade buys redemp-
tion from sin, and that death on crusade counts as martyrdom. This
is precisely what we see with the Tafurs. They start in the Chanson
d’Antioche as barely human and guilty of the worst excesses. By the
time they reach Jerusalem they are transformed. They take the brunt
of hard work and wounds in battle. Their reward is to crown Godfrey
himself. Through the power of crusade they move from outcast sinners
to redeemed quasi-saints.
Looked at from this angle, the significance of the grotesque descrip-
tion of the Tafurs becomes apparent. They are desperately poor and
hence marginalised: they are twice disguised in rich clothes to trick the
enemy which points up the contrast.60 Physically they are deformed, a
marker of exclusion. Their hair stands on end, they have gaping jaws and
their skin is dark.61 This type of description is not only typical of mon-
57
La Chanson de la Croisade albigeoise, ed. Eugène-Martin Chabot, trans. Henri
Gougaud, intro. Michel Zink (Paris, 1989), pp. 60–65, laisses 19–22 for the
sack of Béziers.
58
See, for example, Chanson d’Antioche, 1: 169, 218, 220, lines 2987, 4057, 4089;
Chanson de Jérusalem, ed. Thorp, lines 1983, 3011; Le Chevalier au Cygne et
Godefroid de Bouillon, ed. Reiffenberg, 2: 469, lines 16665–71, clearly portrays
them as well armed equivalents of the free companies.
59
Adolf Tobler and Ernst Lommatzsch, Altfranzösisches Worterbuch, 10 vols
(Berlin, 1925–2002), 8: 1253–57; Robert, Dictionnaire historique de la langue
francaise, 2 vols (Paris, 1992), 2: 805. For contrast with the army see L’entrée
d’Espagne: Chanson de geste franco-italienne, ed. Antoine Thomas, 2nd edn, 2
vols (Firenze, 2007), 2: 76, line 10039: ‘ces chevalier montent n’i a un seul rebaut’.
60
Chanson de Jérusalem, ed. Thorp, lines 6387–93, 7290–97
61
Ruth Mellinkoff, Outcasts: Signs of Otherness in Northern European Art of
the Late Middle Ages, 2 vols (Berkeley, 1993), 1: 125 for deformity, 1: 122 for
320
sters but of devils: and the text explicitly describes them as such.62 They
have the gaping jaws familiar from depictions of Hell: devils consume
human souls and so can be shown as cannibals.63 Cannibalism is itself
something practiced by outsiders remote in time and place like Gog and
Magog, and indeed occasionally by Saracens themselves.64 Rather like
the grotesques round the outermost arch of a portal or the Blemmies and
Cynocephaloi on the edge of the Hereford mappa mundi, the Tafurs are
liminal grotesques, part of the crusade but on the very edge.65 As such
their difference from the rest of the army is emphasised. They look dif-
ferent, dress differently and fight differently. They have their own leader
who is at best loosely under the control of the crusade leaders: there is
no suggestion they have sworn allegiance to anyone. The Saracens rec-
ognise this difference and describe them as barely human.66 But they are
not actually monsters: they remain just sufficiently human to be able to
experience salvation. The redemption of even the most monstrous and
marginal through Christianity is central to its mission: ‘[T]he exile of
the Plinian races to the edges of the world was countered by an opposing
impulse that sought to draw them back, figuratively at least, to the center
big teeth, 1: 123–24 for staring eyes, 1: 126–27 for dark skin. For devils with
glaring eyes and hair standing on end see the portal at Sainte-Foy de Conques.
The Winchester Psalter shows gleeful devils with large eyes, hair on end,
grotesque features, hairy, dark and in rags: see Kristine Edmondson Haney,
The Winchester Psalter: An Iconographic Study (Leicester, 1986), folios 18 (the
Temptation of Christ) and 38 (the Torment of the Damned).
62
Chanson d’Antioche, 1: 220, line 4086, ‘diable’, line 4116 ‘une gens moult averse’.
63
See for example the mosaic of the Last Judgement at Torcello, Venice for a devil
swallowing a human.
64
Tattersall, ‘Anthropophagi’, pp. 242–43. For cannibalism by Saracens see
Floovant: Chanson de geste du xiie siècle, ed. Sven Andolf (Uppsala, 1941), p. 58,
lines 1838–43: a Saracen threatens to grind his Christian opponent into powder
and feed him to the Saracen forces to make them brave.
65
Michael Camille, Image on the Edge: The Margins of Medieval Art (London,
1992), p. 14: ‘the further one moves away from the centre-point of Jerusalem,
the more deformed and alien things become.’ In the same way grotesques are
banished to the outermost voussoirs of arches as at Aulnay (Camille, Image
on the Edge, pp. 65–75), or the chancel arch of St John’s at Adel, Leeds, which
depicts cannibals among other grotesques. See Friedman, Monstrous Races, pp.
37–58 for monsters on the margins of maps; Paul D. A. Harvey, Mappa Mundi:
The Hereford World Map (Hereford, 2010).
66
Chanson d’Antioche, 1: 219, line 4063: ‘Co ne sont pas Francois, ancois sont vif
malfé’.
321
Carol Sweetenham
of the world in Jerusalem.’67 The subtext is that even the most marginal
and desperate can be redeemed through the power of crusade.
The need to emphasise the cannibalism thus becomes clear: even those
guilty of the most monstrous crimes can be redeemed. But there is a
further dimension: not only redemption, but their function as a part of
the overall divine plan. The text brings forward the cannibalism to the
siege of Antioch rather than placing it at Ma’arrat al-Nu‘man. The ritual
public nature of the cannibalism and heavy stress on the Saracen reaction
underlines the motivation: to demoralise the Saracens so they can be more
easily defeated.68 Nowhere was victory more important than at Antioch:
had the city not fallen just before Karbugha’s arrival the army would have
been wiped out.69 The timing and extent of the cannibalism suggest that
it was crucial to success at Antioch. As such it is no coincidence that
Peter the Hermit and Godfrey, respectively the leading preacher of the
crusade and the first ruler of Jerusalem, are shown actively encouraging
the Tafurs in their cannibalism. The Devil and monsters form part of the
overall divine plan for salvation.70 In the same way the Tafurs – diabolical,
grotesque outsiders – and their monstrous crime become essential to the
working out of the divine plan for the crusade at Antioch which begins
at the start of the Chanson d’Antioche with Christ on the Cross. And they
go on to be crucial to the taking of Jerusalem.
… and a Donkey Supper: The Episode of Eurvin

de Creel and Pierre Postel
This episode is unlike any other in the Chanson d’Antioche, more reminis-
cent of a fabliau than a chanson de geste.71 It describes how two relatively
67
Friedman, Monstrous Races, p. 59.
68
Chanson d’Antioche, 1: 219–20, lines 4065, 4077–86.
69
(Cambridge, 1994), pp. 282–83: ‘it was no less than a miraculous delivery, the
very climax of the crusade’.
70
Friedman, Monstrous Races, p. 85 on the Ebstorf and Hereford maps cataloguing
monsters ‘as part of the symmetry and order of the cosmos’; p. 109: ‘a sign of
God’s power over nature and his use of it for didactic ends’. For the same theme
in sculpture see Peter Scott Brown, ‘Scoundrels and Scurrilitas at St Pierre
de Sévignac’, in Difference and Identity in Francia and Medieval France, ed.
Meredith Cohen and Justine Firnhaber-Baker (Farnham, 2010), pp. 197–226
(here p. 198), the grotesques on the portal are ‘separate but inseparable from the
sacred sculpture in the […] tympanum’.
71
See Fabliaux du Moyen-Age, ed. and trans. Jean Dufournet (Paris, 1998).
322
low-ranking soldiers fall out just before the battle of Antioch. While
Pierre Postel is at Mass Eurvin de Créel steals Pierre’s donkey to feed his
starving household. Pierre, not surprisingly, is furious when he returns
to find his donkey made into kebabs and casserole. There is a furious row
with Eurvin, which ends, however, with the two being reconciled in the
knowledge that by the next day they will either be dead or have as many
donkeys as they could possibly want, and sitting down to gorge themselves
on a pre-battle donkey supper.72
It would be tempting to dismiss this as a light-hearted interpolation.
But it is found right through the various manuscript traditions. And there
is more to the episode than meets the eye. Firstly, while the crime may
sound trivial, it is not. The text lays heavy stress on the famine ravaging
the army. To steal food at this point is therefore a major crime; and it
is exacerbated by the fact that Pierre is at Mass. Secondly, it is the only
episode in the text to focus on people of such low rank. Thirdly, it is the
very last episode before the start of the climactic battle of Antioch, on
which the trilogy and indeed the success or failure of the crusade hang.
The message is clear. The redemption offered by possible death on
crusade extends all the way down to even relatively humble crusaders.
They may be non-nobles guilty of unglamorously bad behaviour rather
than major sins: but they are redeemed just the same.
The Old French Crusade Cycle as a Drama of Redemption

and Salvation
The Old French Crusade Cycle thus shows how the offer of redemption
by going on crusade plays out in practice. The Tafurs, barely recognisable
as human, end up in the place of honour at Jerusalem. Petty vavasors are
forgiven. The one person who cannot be forgiven and redeemed is the
one who was notorious, rightly or wrongly, for deserting the crusade,
Stephen of Blois.
This drama of salvation is built into the structure of the cycle itself
through the treatment of the chétifs, those taken prisoner at Civetot at
the start of the crusade. Their desperate battle with Corbaran occupies
some fifteen laisses at the start of the Chanson d’Antioche.73 There is no
suggestion that they have not fought bravely; on the contrary the text
72
Chanson d’Antioche, 1: 377–79, lines 7576–642.
73
Chanson d’Antioche, 1: 38–53, laisses 20–34.
323
Carol Sweetenham
emphasises their heroism against overwhelming odds.74 But nevertheless

captivity is shameful.75 This is clear from the detailed portrayal: they are
chained up in a way faintly reminiscent of sinners in a Last Judgement,
and put to work doing menial labour in a quarry under the orders of
Saracens.76 Worse, defeat cuts across the theology of ‘Deus vult!’, the belief
that because the crusade is a divinely inspired and approved enterprise
then victory is inevitable.
This left the author with a problem. How could an enterprise enjoy
divine favour when it began with such a crushing defeat? The dilemma
could have been avoided by simply ignoring the fact and starting the text
elsewhere. Instead the author makes it the first episode of the Chanson
d’Antioche, devoting some 600 lines to it; and he insists that he is giving
the ‘vrai commencement’: it would be hard to emphasise the point much
more strongly.77
The chétifs then disappear from sight for the remainder of the Chanson
d’Antioche. They reappear in the next instalment, the Chanson des Chétifs,
inserted between the Chanson d’Antioche and the Chanson de Jérusalem.
This is entirely devoted to the salvation of the captives. Through their
military prowess they regain their freedom, they return to the crusade
in time to take part in the siege of Jerusalem, and for good measure leave
the Saracen arch-enemy Corbaran on the verge of baptism.
The disgrace of defeat and captivity is thus redeemed. The erstwhile
captives achieve salvation through achieving their crusading vow, thus
accomplishing their own redemption. They contribute to the success of
the crusade by returning in time to help at the siege of Jerusalem, and by
undermining their most dangerous opponent through bringing him to the
point of redemption by baptism too. The defeat with which the crusade
began is turned into a triumph. The entire structure of the trilogy is designed
to make this a reality within the fictional depiction of the crusade.
74
Thus Chanson d’Antioche, 1: 46, line 634: ‘Fierement se deffendent mais rien ne
lor vaura’.
75
Yvonne Friedman, Encounter between Enemies: Captivity and Ransom in the
Latin Kingdom of Jerusalem (Leiden, 2002), p. 1: ‘the captive […] was the classic
anti-hero, the failed warrior.’
76
For sinners chained up and led away see the portal of Saint-Trophime at Arles;
Chanson d’Antioche, 1: 52, lines 785–89; their forced labour is described in
detail in the first part of the Chanson des Chétifs. Compare Fulcher of Chartres,
Historia Hierosolymitana, II.27.9, ed. Hagenmeyer, p. 475, where the archbishop
of Edessa is used like a pack animal by Saracens; Friedman, Encounter between
Enemies, pp. 113–16.
77
Chanson d’Antioche, 1: 19, lines 12–15.
324
Conclusion
The crusade thus becomes a stage for a drama of salvation. On the one
hand, this is shown through divine reward for heroism and ultimately
death, most strikingly in the episode of Rainalt Porcet. Exactly as prom-
ised in the preaching of the crusade, those who die are rewarded with
immediate salvation and a place in Heaven.
But participation on crusade can lead to redemption for all. The menue
gent can be saved. The Tafurs, marginalised to the very edge of the crusade
and guilty of one of the worst crimes of all, cannibalism, can not only be
redeemed but given the honour of crowning the first ruler of Jerusalem.
Those defeated at Civetot are redeemed through their bravery and
rewarded with the chance to help take Jerusalem. The one exception who
does not achieve redemption is Stephen, the crusader who abandons the
crusade, fails to fulfil his vows and remains forever outside divine grace.
The very worst elements of the crusade are thus not only redeemed but
shown to be integral to its success. This in turn reflects the medieval belief
of salvation history: everything was part of the divine plan, even the most
liminal. The success of the crusade was pre-ordained by God. Without the
cannibalism of the Tafurs the morale of the enemy would not have been
broken. Without the triumphant arrival of the chétifs at Jerusalem the
army would have struggled to take the city and might have been attacked
again by Corbaran. And because heroic actions need counterweights and
texts need light and shade, even Stephen’s cowardice finds its place as the
antithesis of heroism and redemption.
Thus all who succeeded in achieving their crusade vows could find
salvation, no matter how poor or marginal. The relevance of this to a
text exhorting the audience to go on crusade is obvious. No matter how
menial their status, they will achieve salvation through crusade provided
they stay the course. And no matter how small their contribution, it will
still be part of the divine plan: as Jesus prophesied on the Cross:
[…] dela oltremer venront novele gent,
De la mort de lor pere pranderont venjement.
Ne remanra paiens desci qu’en Orient;
Li Franc aront le terre tot a delivrement,
Et ki pris et fines iert en cel errement
L’arme del cors ira en nostre salvament
Et li tiue i voist hui par mon commandement.78
78
‘A new race will come from over the sea to avenge the death of their Father.
Not a single pagan will remain between here and the East: the Franks will
liberate the whole land. The soul of every man who is taken and killed on this
325
Carol Sweetenham
Afterword: The Etymology of the Name Tafur and

What it Might Tell Us
The term ‘Tafur’ was clearly current at or shortly after the time of the
crusade: it is used by Guibert of Nogent. Guibert gives a detailed explana-
tion of the etymology: ‘“tafur” autem apud gentiles dicuntur quos nos,
ut nimis litteraliter loquar ‘trudennes’ vocamus, qui ex eo sic appellantur
quia trudunt, id est leviter transigunt quaquaversum peragrantes, annos’.79
There is no particular reason to disbelieve Guibert’s statement that this
group called itself the ‘tafurs’.
There is no entirely clear etymology for the term Tafur, however. It
has been sourced to thakavor, Armenian for ‘king’.80 However, Schmitz
argues that it is not attested in this meaning until the end of the twelfth
century and should instead be derived from Armenian thaphar (hard up)
or indeed thapharh, a vagrant.81 Cahen suggested Arabic tâfoûr meaning
miserably poor.82 Raynouard suggests a derivation from another Arabic
word, dahur, meaning a trickster or deceiver.83 The only common thread in
these theories is an oriental origin for a word with no western equivalent.
It is clear that the word had wide currency as a general term of abuse
although it is not attested before the late twelfth century. In Old French
it is found, for example, in the late twelfth century Tristran by Béroul: ‘ne
dot pas que je n’alle au plet / Atapiné comme tafurs’.84 It is found in the
Anglo-Norman romance of Horn: ‘joe sui cunréé cumme tafor tapin’.85
journey will receive My salvation. And your soul will find salvation today at My
command’. Chanson d’Antioche, 1: 27–28, lines 206–12.
79
Guibert of Nogent, Dei gesta, VII.23, ed. Huygens, p. 310.
80
Anouar Hatem, Les poèmes épiques des croisades: génèse-historicité-localisation
(Paris, 1932), p. 226; Rouche, citing Gérard Dédéyan, ‘Cannibalisme sacré’,
p. 37, n. 17.
81
Michael Schmitz, ‘Herkunft des altfranzösischen Wortes tafur’, Romanische
Forschungen 32 (1913), 608–12.
82
Claude Cahen, La Syrie du Nord à l époque des croisades (Paris, 1940), p. 15, n. 1.
83
Francois Raynouard, Lexique Roman, 6 vols (Paris, 1838–44), 5: 294. This
is also the view of Daniel Scheludko, ‘Über die arabischen Lehnwörter im
Altprovenzalisch’, Zeitschrift für Romanische Philologie 47 (1927), 418–42
(here p. 439).
84
‘have no doubt that I shall go to the trial in the disguise of a Tafur’. The Romance
of Tristran by Beroul, ed. Alfred Ewert, 2 vols (Oxford, 1939–70), 1: 100, lines
3344–45.
85
‘I am disguised in the costume of a Tafur’. The Romance of Horn by Thomas, ed.
Mildred K. Pope, completed T. B. W. Reid, 2 vols (Oxford, 1955–64), 1: 126,
326
There are various occurrences in Occitan. Bertran de Born for example

refers to ‘rassa vilana, tafura’ insulting those who criticise nobles in ‘Mout
mi plai’ in the late twelfth century; and Gaucelm Faidit to ‘avols gens
tafura’ as those who criticise his love for his lady in ‘Ab cossirier plaing’.86
As a description of a specific group on the crusade the term is not found
outside Guibert. However there is some evidence of the Tafurs having a
recognised identity. Bertran de Born in ‘Pois lo gens’ refers to the Tafur
king and his court; he comments in the context of the king of Aragon
betraying him in his war on the Young King ‘Del rei tafur / Pretz mais sa
cort e son atur’: in other words the king of Aragon is lower in his esteem
than the king of the Tafurs.87 The Canso de la Crozada ascribes the sack
of Béziers to a group loosely attached to the main army and variously
described as truans, ribautz, arlotz and tafurs with their own king. The text
explicitly models itself on an Occitan Canso d’Antioca describing the First
Crusade, and the description may well deliberately evoke its predecessor.88
The term continued to be used after the twelfth century. Wartburg
cites taffeur as Paris argot for a coward.89 Mistral attests tafur as a general
word for a scoundrel and rascal.90 In Spanish Corominas and Pascual
argue that the term passed into general use in the thirteenth century as a
result of the popularity of the Spanish account of the crusade, the Gran
Conquista de Ultramar.91 It survives in modern Spanish as a colloquial
term for a gambler or cardsharp.92
We can draw a number of conclusions. The first is that there is no agreed
or identified European etymology of the word tafur, which suggests (but
does not prove) that it did indeed originate on the First Crusade. The
second is that it had wide currency at least from the late twelfth century
line 3690; see also lines 3970, 4290.

86
L’Amour et la Guerre: L’oeuvre de Bertran de Born, ed. Gérard Gouiran, 2 vols
(Aix-en-Provence, 1985), vol.2: 845–56 (line 33); Les Poèmes de Gaucelm Faidit,
ed. Jean Mouzat (Paris, 1965), pp. 142–48 (lines 49–52).
87
‘I rate the court and retinue of the King of the Tafurs more highly’. Bertran de
Born, ed. Gouiran, 1: 451–74 (lines 33–34).
88
La Chanson de la Croisade albigeoise, ed. Chabot, pp. 19–22, laisses 19–22 for
sack of Beziers, pp. 64–65, laisse 22: line 4 for ‘li gartz tafur pudnais’; pp. 40–41,
laisse 2: lines 1–3 for reference to Canso d’Antioca.
89
Walther von Wartburg, Französisches etymologisches Worterbuch, 18 vols (Bonn,
1928–67), 13: 31.
90
Frédéric Mistral, Lou Trésor du Félibrige, 2 vols (Paris, 1932–79), 2: 944.
91
Joan Corominas and José A. Pascual, Diccionario Crítico Etimologico castellano
e hispaníco, 6 vols (Madrid, 1980–91), 5: 377–78.
92
Oxford Spanish Dictionary (Oxford, 1998), p. 706.
327
Carol Sweetenham
as a general term of abuse; this shows a persistence of memory, perhaps

in oral tradition, and seems to suggest a renewed interest at around the
time of the defeat at Hattin. The third is that in at least two Occitan
texts around the turn of the century the Tafurs are depicted with a king,
which suggests a more deliberate evocation of the First Crusade history.
All of this suggests that the Tafurs were well known and indeed a byword
for appalling behaviour by the end of the twelfth century, which in turn
accounts for the prominence given to them by the author of the Chanson
d’Antioche. If even these notorious reprobates could be saved on crusade,
there was hope for any potential crusader.
328
The Image of the Saracen
in Romanesque Sculpture
Literary and Visual Perceptions*
Ruth Bartal
Tel-Aviv University
Christian perceptions regarding the Saracens in chronicles and epics dur-

ing the crusades have been studied by historians and literary critics from
various perspectives. Scholars concur that despite the physical encoun-
ter between Christians and Muslims during the period of the crusades,
Christian characterisation of the Saracens was not based on empirical
observation. The early epics and chronicles structured the image of the
Saracen as antithetical to that of the fair noble and virtuous Christian
knight. The Saracens’ conduct, physical appearance, and moral qualities
were contrasted to those of the Christians.1
*
I am grateful to Prof. Nurith Kenaan-Kedar for calling my attention to the
sculpture of Tubilla del Agua, and for bringing me the photograph from the
Museum Marés in Barcelona. I would like to thank also Dr. Jordi Camps Soria
for sending me the slides of Soto de Bureba and Bercedo. A version of this
chapter previously appeared in Assaph: Studies in Art History 8 (2003), 85-102.
1
On the image of the Saracens in medieval literature see: Caroline Cazanave,
‘Sarrasins térrifiants, Sarrasines attrayantes: Double effet de la vision de l’autre
dans les Chansons de Geste’, in L’Exotisme: Actes du colloque de Saint-Denis de
la Réunion 1988, ed. Alain Buisine, Norbert Dodille and Claude Duchet (Paris,
1988), pp. 61–71; Paul Brancourt, ‘De l’imagerie au réel: l’exotisme oriental
d’Ambroise’, in Images et signes de l’Orient dans l’Occident médiéval, Aix-en-
Provence, ed. Jeanne Laffitte (Marseille, 1982), pp. 29–39; Danielle Buschinger,
‘L’image du Musulman dans le Rolandslied’, in Images et signes, pp. 61–73;
Kazimierz Kupisz, ‘Autour de l’exotisme du roman de Tristan et Yseult’, in
Exotisme et creation: Actes du colloque international, ed. Jean Moulin (Lyon,
1983), pp. 31–40; Claude Lachet, ‘L’exotisme dans Sone de Nansy, fantaisie
et réalisme’, in Exotisme et creation, pp. 43–53; Sylvia Schein, ‘From “Holy
Geography” to “Ethnography”: “Otherness” in the Descriptions of the Holy
Land in the Middle-Ages’, in Miroirs de l’altérité et voyages au Proche-Orient, ed.
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. –.
Ruth Bartal
The Saracens were demonised: they were cruel, deceitful and possessed
bestial sexual appetites. They were black and hideous sometimes appeared as
giants. In Aliscans Baudus was fifteen feet tall, black, with eyes like burning
coals, a huge face, carrying the mast of a ship as his weapon. The Saracens’
supernatural, monstrous dimensions were designed to emphasise their
ferocity, in order to magnify the Christian victory or alternatively to ridicule
them in order to reduce fear.2 Their dark complexion also functioned as a
metaphor for their morally negative character. This metaphorical function
of ostensibly realistic physical features is apparent through the contrasting
descriptions of the converted Muslim or Muslim princess as fair-skinned.3
The profound contrast between Christians and Saracens in the epics, and
the analogies made between the wild man and the Moor in medieval
romances defined the Saracens as belonging to an inferior civilisation. Their
worst fault, however, was embodied in their pagan faith which associated
them with Satan.4 The battle against the Saracens was thus perceived as
a just, moral and spiritual war against demonic aggressors and idolators.
These well-known stereotypical depictions of the Saracens in literary
texts form the basis for comparison with the Romanesque images to be
analysed here. Can one draw direct analogies between these two different
forms of artistic expression? To what extent did the visual presentation fol-
low the literary descriptions, and what were its other sources of inspiration?
Romanesque art depicted the Saracen-Christian conflict in numerous
ways. The iconography of these representations relied on different textual
and visual sources, and direct analogy with contemporary literary sources
is not always evident. In order to examine representations of Saracens in
Romanesque sculpture and study the affinities and differences between the
textual and visual presentations, I will briefly survey several Romanesque
programmes in which the Christian-Muslim conflict is portrayed. These
will be classified by their source of inspiration, whether visual or textual,
Ilana Zinguer (Genève, 1991), pp. 115–23; Michel Balard, ‘Concept et images de
l’Orient au Moyen Age’, in L’Orient, concept et images. Actes du colloque organisé
par l’IRCOM (Paris, 1988), pp. 17–26; Norman Daniel, Heroes and Saracens:
An Interpretation of the Chansons de Geste (Edinburgh, 1990); Daniel, Islam
and the West: The Making of an Image (Edinburgh, 1960).
2
Susan Stewart, On Longing: Narrative of the Miniature, the Gigantic, the
Souvenir, the Collection (Durham, NC, 1993), pp. 70–103.
3
Jacqueline de Weever, Sheba’s Daughters: Whitening and Demonizing the
Saracen Woman in Medieval French Epic (New York, 1998), pp. 53–71.
4
Ron Barkai, Cristianos y musulmanes en la España medieval: El enemigo en el
espejo (Madrid, 1984), pp. 19–58, 156–58; C. Meredith Jones, ‘The Conventional
Saracen of the Song of Geste’, Speculum 17 (1942), 201–25.
330
The Image of the Saracen in Romanesque Sculpture
and by their compositional formulation. Some of these programmes fea-

ture the confrontation between Christians and Saracens, others present
the notion of Christian triumph in concise symbolic form. Alongside
the better known programmes a less familiar group will be introduced in
which the Saracens are depicted as captives. It is in this latter group that
the image of the Saracen bears the closest affinities to his stereotypical
image in the early epics.
The first group presents the notion of Christian domination through
biblical themes: The portal of Ripoll, which is structured as a triumphal
arch, incorporates scenes from the Apocalypse, the Book of Daniel, and
scenes from the Pentateuch. The battle at Rephidim (Exodus 17.8–13)
presents Moses standing between Aaron and Hur and raising his hands
during Joshua’s battle against the Amalekites.5 Urban, according to Baldric
of Dol in his Historia Jerosolimitana, used this example to encourage
the crusaders: ‘It is our duty to pray; let it be yours to fight against the
Amalekites, with Moses we extend our tireless hand in prayer to heaven;
you fearless warriors, take out and brandish the sword against Amalek.’6
Humbert of Romans maintained that God himself had authorised the
crusades when he not only directed Moses to make war against the
Amalekites, but added , ‘God will be at war with Amalek from age to
age.’7 The defeated enemy in the battle scene, however, does not bear
distinguishing characteristics; he is depicted through the classical image
of fallen enemies under the feet of their conquerors, as for example on
Roman and Byzantine monuments, where the defeated enemy lay on the
ground or was trampled by the horse’s hooves.8
Another biblical theme illustrating Christian-Saracen polemic is the
sacrifice of Isaac. Prevalent in medieval art in various contexts, this sub-
ject on the portal of the Lamb in San Isidoro in Léon is depicted in a
5
Manuel Castiñeiras, ‘The Portal at Ripoll Revisited: An Honorary Arch for the
Ancestors’, in Romanesque and the Past, ed. John McNeill and Richard Plant
(Leeds, 2013), pp. 121–41; Marisa Melero Moneo ‘La propagande politico-
religieuse du programme iconographique de la façade de Sainte-Marie de
Ripoll’, Cahiers de Civilisation Médiévale 46 (2003), 135–57.
6
Quoted in Penny J. Cole, The Preaching of the Crusades to the Holy Land
(Cambridge, MA, 1991), p. 18.
7
James A. Brundage, ‘Humbert of the Romans and the Legitimacy of Crusader
Conquests’, in The Horns of Hattin, ed. Benjamin Z. Kedar ( Jerusalem, 1992),
pp. 302–13.
8
See for example the defeated enemy on the Column of Trajan: André Grabar,
Christian Iconography: A Study of its Origins (Princeton, 1980), fig. 136, or on
the Column of Arcadius fig. 126.
331
Ruth Bartal
very unusual manner: The scene is read from right to left: portrayed on
the right hand side are Sarah standing at the entrance of her tent and
Isaac on his way to the land of Moriah; at the centre one can identify
the sacrifice of Isaac. The iconography of the central sacrifice of Isaac is
more or less conventional; however, the presence of Sarah goes against
the Genesis story, and moreover on the left hand side another woman is
depicted raising her garment in an immodest gesture, and there is a rider
with a bow and arrow aimed at the Agnus Dei which is presented in the
upper zone. These figures have been convincingly identified by Williams
as Ishmael and his mother Hagar. Williams links the scene to the victory
of Alfonso VII in the battle at Baeza in 1147.9 The identification of Hagar
and Ishmael with the Saracens relies on many descriptions of the geneal-
ogy of Muhammad as the offspring of Ishmael. Already in the Chronicle
of Albelda the Saracens were identified as the offspring of the servant
Hagar and her son Ishmael.10 The unusual scheme in Léon conveys the
idea that Abraham’s illegitimate progeny was at war with the legitimate,
and the promise: ‘Your descendants shall possess the gate of their enemies’
(Genesis 22:17). The composition is structured so that Isaac is seen march-
ing onto the stage of history while Ishmael is seen departing from it, thus
furnishing the viewer with a very clear image of Christian conquest.
Allusions to Old Testament exempla can be found in all contemporary
chronicles. Humbert of Romans in his manual for preachers encouraged
the use of Old Testament heroes as model protocrusaders.11 The presen-
tation of the conflict through biblical exempla elevated the conflict to a
spiritual level. However, the audience could easily link Hagar’s immod-
est gesture with prevalent popular convictions concerning the immoral
Saracen woman, and identify the aggressive Ishmael with his bow and
arrow as the Saracen warrior, the adversary of God.
The second group, comprising battle scenes between Christians and
Saracens, has been identified as relying directly on the Song of Roland. The
battle may be rendered either as being between two armies, such as the
battle scene on the lintel of Angoulême from the beginning of the twelfth
century, or as a battle between two warriors, such as the scene depicted on
one of the capitals in the Palacio de los Reyes de Navarra in Estella, which
9
John Williams, ‘Generationes Abrahae: Reconquest Iconography in Leon’,
Gesta 16 (1966), 4–11.
10
‘Chronique d’Albelda’, in Chroniques asturiennes ( fin ixe siècle), ed. and trans.
Yves Bonnaz (Paris 1987), pp. 3–4.
11
On the Old Testament prototypes see Brundage, ‘Humbert of Romans’, pp. 302–13.
332
Fig. 1: The Cathedral of Angoulême, twelfth century (photo Bartal)
has been identified as Roland fighting the Saracen giant Ferragut.12 The
scene in Angoulême is identified as Marsile fainting before the city of
Saragossa. On the right, Marsile faints before the captured city while his
wife, Queen Bramimonde, is looking down from the tower despairingly at
the sight: ‘King Marsile, flees to Saragossa […] He lies down, a grisly sight,
on the green grass. He has lost his right hand completely […] Before him
his wife Bramimonde weeps and wails, and sets up a loud lament’ (fig. 1).
It has been suggested that the patron of this program was Bishop Girard
of Angoulême, who was compared to Archbishop Turpin in the poem.13
These representations rely on written texts and resemble literary descrip-
tions in form. As in the epics, the scenes are depicted through the mode
of confrontation. They represent an ongoing struggle in a narrative form.
In Angoulême, the Saracens can be identified by their different shields
and headgear; there is, however, no distinction between the Saracen and
the Christian warriors, whether in their dimensions or in other features.
Their falling positions and the figure lying on the ground are once again
12
Pedro de Palol and Max Hirmer, Early Medieval Art in Spain (London, 1967),
fig. 239.
13
Michael Camille, The Gothic Idol (Cambridge, 1989), pp. 146–151.
333
Ruth Bartal
based on classical models. On the capital dating to the end of the twelfth
century, the image of Feraggut seems to rely to a greater extent on literary
perceptions of the Saracen. He is portrayed as a monstrous creature: he
has a big head with curly hair, he is bearded, and his mouth and nostrils
are wide open; he uses different weaponry, some kind of a mace, and is
further disgraced by being shown falling off his horse.
The third group differs considerably from the earlier depictions both
in form and in content. It does not show the encounter, but rather the
moment of defeat. The idea of triumph is embodied in two themes: the
rider trampling under his feet the defeated enemy, and the conquering
lion. Both themes rely on well-known classical models, and their formu-
lation is based on visual rather than textual sources. The use of a vertical
axis depicting superior versus inferior, instead of the previously employed
horizontal axis, reinforces the notion of defeat. The focus in these images
is on the conquering rider or lion. The defeated figure of the enemy varies:
at times it is the image of the classical barbarian, as in the cathedral of
Oloron-Sainte-Marie (fig. 2).14 Like the barbarians, the figure in Oloron
is highly expressive in its appearance and gestures. The realistic expression
on his face is that of fear and pain. We can see the same expression on the
faces of the defeated barbarians on many Roman monuments. Wearing a
short tunic and unkempt hair, the figure tries to protect himself from the
horse’s hooves, raising his hand above his head. In other representations, as
in Parthenay-le-Vieux, the enemy is depicted in a more symbolic manner,
personified as a prostrated figure at the rider’s feet.15 The scene is based on
a tradition of classical ‘adventus’ iconography, the emperor’s entry into
a captured city, as for example the ‘adventus’ of Constantius Chlorus on
the Arras Medallion, where he is welcomed by the personification of the
city of London.16 In all these representations both the barbaric figure and
the prostrated personification, possibly representing idolatry, are antique
14
Ruth Bartal, ‘Le programme iconographique du portail occidental de Sainte
Marie d’Oloron et son contexte historique’, Les Cahiers de Saint-Michel-de
Cuxa (1987), pp. 95–124 (here pp. 102–5). On the history of Gaston IV see
Pierre Tucoo-Chala, Quand l’Islam etait aux portes des Pyrenées, de Gaston IV
le croisé à la croisade des Albigeois (Biarritz, 1994), pp. 85–142.
15
Linda Seidel, Songs of Glory: The Romanesque Façades of Aquitaine (Chicago,
1987), fig. 53. The image also recalls depictions of the idol prostrated at the feet
of the Holy Family in the scene of the Flight into Egypt, as can be seen in the
eleventh-century manuscript of the Annals of Saint-Germain-des-Prés in Paris:
Camille, The Gothic Idol, fig. 5.
16
The Arras Medallion: see Sabine G. MacCormack, Art and Ceremony in Late
Antiquity (Berkeley, 1990), pl. 9.
334
Fig. 2: Oloron-Sainte-Marie – West Portal (photo Bartal)
motifs of the subjugated enemy and could easily be identified by medieval

audiences as the contemporary Saracen. The debate concerning the iden-
tity of the triumphant riders in Romanesque art is still unresolved. Do they
represent the moral struggle, the general concept of the conquest of good
over evil, or rather the conquering ‘milites Christi’? Can we identify the
riders as historical crusaders? Or alternatively as literary characters? Linda
Seidel argues that the ‘Song of Roland can be read as an armed fight of
good and evil in which the Franks are personifications of Christian virtue
and the Muslims are the embodiments of pride, greed and treachery’.17
These analogies indicate the difficulty in making clear-cut distinctions
between the symbolic and the mundane natures of Christian victory. It
would seem that in the eyes of the medieval beholder the spiritual and the
temporal victories were intertwined and thus perceived as a single entity.
Christian victory may also be represented by the conquering lion. The
conquering lion is likewise drawn from classical models; yet in classical
art the lion is usually attacking a bull or a gazelle. In Romanesque art the
17
Seidel, Songs of Glory, pp. 55–56, 70–80.
335
Ruth Bartal
lion often devours or tramples a human being. In Oloron-Sainte-Marie a

lion devouring a man is placed opposite the rider.18 As in the case of the
rider, the symbolic meaning of the conquering lion has been extensively
studied both for its Christological and its demonic symbolism. Another
aspect of its presentation was very convincingly demonstrated by Hartner
and Ettinghausen, who showed its evolution as a symbol of political
triumph. The Muslims adopted it to represent their victories over the
Christians. The Christians, aware of the impact of this emblem and its
political connotations, adopted it as a symbol of their domination.19
In Spanish Romanesque art the conquering lion is depicted in various
contexts. The lion may symbolise Christ, the Lion of Judah, as suggested
by the inscription in Jaca,20 or it may symbolise the Christian warrior, a
metaphor well-rooted in the contemporary literary sources. In the Poema
Almería the Christian warriors are compared to lions: ‘William Pellis,
without waiting for his commanders’ order, rushed ahead of the others,
and first killed one Saracen in the front line with his lance, then, like a
lion among cattle that tears at their bodies with his claws, killed many
others.’21 In the Historia Silense, Ordoño is called the Lion of Libya.22 In
the Primera Crónica General de España, Count Fernán González exhorts
his soldiers to fight the army of Al-Mansūr; he compares his soldiers to
lions and the enemy to sheep.23 The Spanish version of the Chrismon
flanked by two lions, replacing the lambs of the early Christian formula,
may serve as an example of the ideological import of this metaphor in
Spanish Romanesque art.24
18
Ruth Bartal, ‘Le programme iconographique’.
19
Willy Hartner and Richard Ettinghausen, ‘The Conquering Lion, the Life of a
Symbol’, Oriens 17 (1964), 161–71.
20
Susan Harens Caldwell, ‘Penance, Baptism, Apocalypse: the Easter Context of
Jaca Cathedral’s West Tympanum’, Art History 3 (1980), 25–40.
21
Quoted in Colin Smith, Christians and Moors in Spain, 2 vols (Warminster,
1988–89), 1: 176–77.
22
Historia Silense, ed. Justo Pérez de Urbel and Atilano González Ruiz Zorrilla
(Madrid, 1959), p. 157.
23
Primera crónica general de España, ed. Ramón Menéndez Pidal (Madrid, 1977),
p. 393, cap. 689. The same metaphor appears in the French epics: Chrétien de
Troyes describes Erec as a lion; see Chrétien de Troyes, Erec et Enide, ed. Mario
Roques (Paris, 1955), p. 68. Referring to Richard the Lionheart, Bertran de
Born says: ‘the lion’s custom appeals to me’; see Anthology of Troubadour Lyric
Poetry, ed. and trans. Alan R. Press (Austin, 1971), p. 167.
24
Jose Perez Carmona, Arquitectura y escultura romanicas en la provincia de Burgos
(Burgos, 1974), p. 145, explains the preference for the lion over the sheep in
336
In the monastery of San Julian in Santa Cruz de Mena, the tympanum

presents a dramatic scene: on the left a lion is trampling under his feet a
human victim, whereas on the right a horse is standing without its rider,
with the sword of the fallen rider alone depicted above the horse. The lion
is roaring, his head raised in a proud gesture; Huidobró mentions traces
of an inscription on the lion’s tail reading Vicit Leo.25 The horse inclines
his head submissively, while the fallen enemy lies under the lion’s paws.
The well-known Constantinian emblem of triumph, the Chrismon, is
placed between the horse and the lion as a central emblem. In Santa
Cruz the image of the defeated enemy is not given any individual traits
or specific characteristics, only its inferior place and reduced size indicate
his symbolic defeat (fig. 3).
The conquering lion appears in the same context on the tympanum
of San Pelayo de Mena, in the vicinity of Santa Cruz, but the program
in this case is more complex: on the right-hand side a lion is attacking a
man, on the left a man is attacking a lion, while a group of four captives
is standing in between them. Above, a choir of seven angels forming an
arch seems to glorify the event. The inscription on the lintel reads: ‘Ego
sum Pelagius Corduba’ (fig. 4). The interpretations of the tympanum
link the scene to the inscription, identifying it as the martyrdom of St
Pelagius, the famous martyr of Córdoba. According to these interpreta-
tions the man being devoured is St Pelagius, and the captives are martyrs
awaiting their execution. The image on the left, which was erroneously
identified as a rider, is either Caliph ‘Abd al-Rahmān II, or his envoy sent
to carry out the execution; the angels represent the triumph of Christian
faith over their Muslim persecutors. This interpretation is unacceptable
because the figure on the left is not a rider: the raised tail with its button-
like end can only be that of a lion and not that of a horse. The man is
sitting astride the lion, his arm outstretched and his hand touching the
mouth of the animal. The position astride the animal is characteristic of
depictions of Samson killing the lion. Hence the program of San Pelayo
does not depict the martyrdom of St Pelagius, nor does the inscription
refer to the scene on the tympanum, but rather to the church which
is consecrated to the martyr. The tympanum presents two emblems of
triumph: the conquering lion and Samson. The identity of the captives
is more difficult to establish. If they are Christian prisoners as identified
by Huidobró, then their position in between the two emblems of power
Spanish Romanesque art; see also Ruth Bartal, ‘The Survival of Early Christian
Symbols in 12th century Spain’, Principe de Viana 181 (1987), 299–315.
25
L. Huidobró Serna, ‘The Art of the Reconquest in Castille, Valle de Mena’, Art
Bulletin 14 (1931), 172–76.
337
Ruth Bartal
Fig. 3: San Julian – Santa Cruz de Mena – West Portal (photo Bartal)
Fig. 4: San Pelayo de Mena – West Portal (photo Bartal)
338
would imply that they are awaiting their salvation, hence they would not
be connected with the devoured man who symbolises the defeated evil.
However, their appearance suggests that they are not Christians: their
beards, their headgear, the edge of their trousers visible under their long
tunics, and their bound hands indicate that they are Saracen captives,
standing in between the two emblems of Christian victory. In contrast
to other presentations, here the defeated enemy is given a central place,
and he is depicted in a more realistic mode.26
The presentation of Samson and the conquering lion in San Pelayo can
be compared to other programs where the image of Samson is depicted
opposite the conquering rider, as in the church of Santa Maria in Carrión
de los Condes or in Partheney-le-Vieux. The image of Samson is, in fact,
synonymous with the lion. Like the lion, he serves as a metaphor for
Christ, an allegory of Fortitude; like the lion he functions in contem-
porary secular writings as an emblem of the daring and heroism of the
Christian knight. In the Poema Almería the Duke Pedro Alfonso is said to
be beautiful like Absalom and strong like Samson.27 These compositions
which include variants of the same images – equestrian, lion, Samson – all
present what Sauerländer defined as ‘sacred heraldry’.28
The last category comprises the image of the Saracen captives. In this
group the focus is placed not on the conqueror, but rather on the defeated.
This group can be divided into two types of chained figures: one consist-
ing of bearing figures (Atlantes), the other of standing chained figures.
In Oloron-Sainte-Marie, two gigantic figures support the trumeau:
they are bound with their backs to each other by heavy chains. They are
bearded, wearing headgear, their physiognomy has some negroid traits,
and their expression reflects fear and suffering (fig. 5). This image evokes
descriptions in the Gesta Francorum and in the Chanson d’Aspremont
where the Saracen leaders upon their defeat recall their desire to capture
crusaders and bind them together in pairs.29 Here the situation is, of
course, reversed. The trumeau is placed on a vertical axis under the cross
26
Ruth Bartal, ‘Interpretacion iconografica del tympano de San Pelayo de Mena’,
Goya: Revista de Arte 192 (1986), 322–29.
27
El Poema de Almeria y la épica románica, ed. H. Salvador Martinez (Madrid
1975), p. 32, line 119.
28
Willibald Sauerländer, ‘Romanesque Sculpture in its Architectural Context’,
in The Romanesque Frieze and its Spectator, ed. Deborah Kahn (London, 1992),
pp. 17–44.
29
Matthew Bennett, ‘First Crusaders’ Images of Muslims: The Influence of
Vernacular Poetry?’ Forum for Modern Language Studies 22 (1986), 101–22
(esp. 109, n. 67); Gesta Francorum et aliorum Hierosolimitanorum, ed. Rosalind
339
Ruth Bartal
Fig. 5: Oloron-Sainte-Marie – West Portal (photo Bartal)
and the Christogram, emphasising the divine nature of the victory. The
bearing figure in art has a long metaphorical history. In his chapter of On
Architecture dealing with the training of architects, Vitruvius explains the
meaning of the bearing figures, maintaining that the caryatides symbolise
the victory over Caria: ‘the town was captured, the men were killed, their
matrons were led away into slavery […] And so the architects of that time
designed for public buildings figures of matrons to carry burdens; in order
that the punishment of the sin of the Caryatid women might be known
to posterity and historically recorded’. The victory of the Spartans over
the Persian army was glorified in the same manner : ‘they placed statues of
their captives in barbaric dress, punishing their pride with deserved insult
to support the roof ’.30 The two supporting figures of Oloron are restored
Hill (London, 1962), p. 22, n. 2. La Chanson d’Aspremont, chanson de geste du

xiie siècle, ed. Louis Brandin, 2 vols (Paris, 1923), 1: 95.
30
Vitruvius, On Architecture, 2 vols (Cambridge, MA, 1962), 1: 8–10.
340
in Morlaas, and recur in Saint-Gilles-du-Gard.31 The cathedral of Sainte

Marie in Oloron was erected under the patronage of Gaston IV of Béarn
who participated in the First Crusade; he was also an ally of Alfonso I
the Battler, and participated in the reconquest of Saragossa. The façade of
Oloron-Sainte-Marie features all the triumphant images: the conquering
rider, the conquering lion, and the defeated Saracens.
The standing type recurs several times in the province of Burgos: in
Bercedo, Soto de Bureba (fig. 6), Vallejo de Mena (fig. 7), Almendres
(fig. 8), and Tubilla del Agua (fig. 9).32 All the churches are from the
northern region, not far from San Pelayo and Santa Cruz de Mena. All
date to the last quarter of the twelfth century. Artistic expression in this
region, as we have already noticed in Santa Cruz and San Pelayo, differs
considerably from the intellectual discourse which we have encountered
in Léon or Ripoll. The image of the captives introduces a local mode of
presenting the spirit of the Reconquest.
All the figures of the captives, except the one in Tubilla del Agua, are
located at the bottom of the archivolt. In Tubilla del Agua, he supports
the window of the tower. In Vallejo de Mena, the figure is crouching and
flanked by a warrior. In Bercedo, Vallejo de Mena and Almendres, the
figure is placed under an angel. In Tubilla del Agua and Soto de Bureba
the figures are clothed, while in other representations they are shown
in the nude. All the figures are giants, with big heads, long untidy hair,
large eyes and noses. They are all chained in the same manner by their
feet and neck. This mode of binding the enemy appears in the Cantigas
de Santa Maria, which describes the cruelty of the Moors who bound the
Christian captives by their hands or by their legs and necks.33 The initial
clue leading to my identification of these figures as Saracen captives was
the figure of Soto de Bureba clothed in an Oriental garb. However, it is not
31
Jacque Lacoste, Sainte Foy de Morlaas (Pau, 1976); Marcel Durliat ‘Le portail
de Morlaas’, Bulletin Monumental 136 (1978), 55–61; Whitney S. Stoddard, The
Façade of Saint-Gilles-du-Gard: Its Influence on French Sculpture (Middletown,
CT, 1973).
32
Carmona, Arquitectura y escultura romanicas: for Bercedo, pp. 233–34; San
Pelayo de Mena, 236–37; Santa Cruz de Mena, 237–38; Soto de Bureba, 114–
15, figs 109–110. Frederic Marés, Cataleg del Museu Frederic Marés I Deuloval
(Barcelona, 1979): for Tubilla del Agua, p. 28, nos 980–81. Paloma Rodríguez-
Escudero Sánchez, Arquitectura y escultura románicas en el Valle de Mena
(Salamanca, 1987): for San Lorenzo de Vallejo, pp. 53–70; Santa Cruz de Mena,
pp. 99–101.
33
Joseph F. O’Callaghan, Alfonso X and the Cantigas de Santa Maria: A Poetic
Biography (Leiden, 1998), p. 92, n. 36.
341
Ruth Bartal
Fig. 6: Soto de Bureba – West Portal (photo Bartal)
Fig. 7: Vallejo de Mena – West Portal (photo Bartal)
342
Fig. 8: Almendres – West Portal (photo Bartal)
Fig. 9: Tubilla del Agua – Window of the Tower (photo A. Ferro © Museu
Frederic Marès, reproduced with permission)
343
Ruth Bartal
only the Oriental garb that supports my identification. The giant captives,
hideous in their appearance and demonic attributes, are conceptually
reminiscent of the literary descriptions of the Saracen. Their gigantic,
frightening image and the chains signify an infernal existence. The giants
present a physical world of disorder and disproportion; as metaphors of
physical force they allegorise lawlessness. Thus the victory over the giant
is the triumph of culture over chaotic primordial forces.34 While the
chains may define these figures as captives, we must also bear in mind their
metaphorical level: chains in medieval art serve as signs and attributes
of Satan. In depictions of the Last Judgment Satan and the doomed are
chained. Thus the chained captives are associated with Satan and belong
in hell. Although this group is a medieval creation and does not derive
from classical models, the angels above the figures in Bercedo, Vallejo de
Mena and Almendres play the part of the classical Victories placed above
the barbaric captives, emphasising that the victory is by Divine Providence.
Thus in twelfth-century Romanesque art, the portrayal of Christian-
Saracen conflict often alluded to textual references, whether in the Old
Testament or contemporary chronicles and epics. However, the literary
descriptions were too subtle to serve as pictorial propaganda aimed at a
large audience; images had to be more explicit in order to reach a wider
public. Thus artists preferred the concise symbolic forms. In the same man-
ner we can understand the advice of Guibert of Nogent to preachers to
present their examples through contrasts, in order to be more intelligible.35
In most cases the artists employed visual models prevalent since antiquity
as signifiers of triumph. These models formed an accessible pictorial lan-
guage through which symbolic presentations could be readily transformed
into mundane temporal situations. The image of the defeated enemy also
derives from the same vocabulary: its identification as the Saracen can be
deduced only from the context. In most representations the emphasis is
on the conqueror rather than on the defeated. Only in the last group is
the enemy highlighted, capturing the attention of the beholder. In these
images of captives the visual depictions converge with the contemporary
literary descriptions, both aiming at the same goal, namely to induce an
emotional response of hatred and contempt from the audience.
The Saracen in Romanesque sculpture in its various manifestations is
not represented as a human being; he is dehumanised and depicted as a
symbol of defeat. This symbolic figure is substituted by a more realistic
image in later manuscripts.
34
Stewart, On Longing, pp. 70–103.
35
Cole, The Preaching of the Crusades, pp. 20–23; Guibert de Nogent, Liber quo
ordine sermo fieri debeat, PL 156, col. 26.
344
An extended analysis of the image of the Saracen in thirteenth-century

manuscripts is beyond the scope of this paper. However, contrary to the
symbolic image of the Saracen in Romanesque sculpture, in thirteenth-
century manuscripts he is portrayed as a real human being, at times even
dignified. In an illustration to Augustine’s City of God, now in Paris, the
pagans, who may be identified as Saracens due to their headgear, point at
a mosque with an idol, while the Christians point at a church. Although
the miniature depicts the polemic between the Christians and the Saracen
idolators, the two groups are not dissimilar.36 In the Cantigas the Moors
are depicted in various contexts; although their negative nature emerges in
numerous episodes, they are portrayed in a dignified manner.37 This new
image of the Saracen does not imply, however, a real change in Christian
attitudes toward the Saracens. We may ask whether the change is due to
a closer acquaintance with the Muslims, or possibly to the new Gothic
style, or to the different media. The answer to these questions is not a
simple one. The change is probably due to all the above reasons. I believe,
however, that the different approaches stem mainly from the different
media. The manuscripts were designed for an educated audience and thus
could be more sophisticated and exhibit more nuances in their presenta-
tion. Romanesque sculpture, on the other hand, was a public art form in
which the image must appeal to its audience with a powerful immediacy
and could not risk being ambivalent.
36
Camille, The Gothic Idol, fig. 75.
37
On the representation of the Moor in the Cantigas de Santa Maria see
O’Callaghan, Alfonso X and the Cantigas de Santa Maria, n. 38; J. Esten
Keller and A. Grant Cash, Daily Life Depicted in the Cantigas de Santa Maria
(Lexington, 1998), p. 58, pls. 60, 68; Amparo Garcia Cuadrado, Las Cantigas:
El Codice de Florencia (Murcia, 1993), pp. 113–18; Robert I. Burns, Islam under
The Crusaders: Colonial Survival in the Thirteenth-Century Kingdom of Valencia
(Princeton, 1973), pp. 318–19.
345
History, Fiction and Film
Islam Faces the Crusaders
Robert Irwin
School of Oriental and African Studies, University of London
What follows is, in a sense, a sequel to my essay ‘Saladin and the Third
Crusade: A Case Study in Historiography and the Historical Novel’.1 That
essay not only examined the ways in which novelists who wrote about
the crusading past were influenced by the historians they read, but also
considered the possibility that the historians’ view of their subject may
have been unconsciously influenced by images and stereotypes derived
from historical novels. In the course of that essay it was noted that the
‘boundary between historical fact and fiction is by no means as clear cut
as one might have expected’ and the essay concluded that not only do ‘the
historical novels show a close dependence on the histories, but also […]
the histories make use of the tricks of the novelist’.2 Here however, I shall
concentrate on the First, rather than the Third Crusade, and, moreover,
the focus is somewhat different. Only a select handful of novelists will
be considered in some detail, but Western and Arab film-treatments will
also be discussed.
The First Crusade has proved to be a less popular setting for romantic
fictions and films than the Third Crusade. Arab writers have naturally
preferred to deal with the triumphs of Saladin, rather than with the
earlier period of Islamic disunity and defeats. Even European novelists
have tended to consider the crusading enterprise as an unedifying episode
1
Robert Irwin, ‘Saladin and the Third Crusade: A Case Study in Historiography
and the Historical Novel’, in Companion to Historiography, ed. Michael Bentley
(London, 1997), pp. 139–52. Since ‘History, Fiction and Film: Islam Faces the
Crusaders’ was delivered at the Huesca conference, Susan Edgington has
published a wide-ranging and penetrating survey of historical novels which
feature the First Crusade, ‘The First Crusade in Post-War Fiction’, in The
Experience of Crusading, 1, ed. Marcus Bull and Norman Housley (Cambridge,
2003), pp. 255–80. Our two surveys are, I think, complementary.
2
Irwin, ‘Saladin and the Third Crusade’, pp. 148, 150.
Luis García-Guijarro, Turnhout, 1 (Outremer, ), pp. –1.
Robert Irwin
in medieval history. They have been influenced in this by the decidedly

cynical treatment of the crusades by such early historians as Edward
Gibbon in The Decline and Fall of the Roman Empire (1776–88) and by
Sir Charles Mills (1788–1826) in his History of the Crusades (1820). The
input of both Gibbon and Mills is obvious in the pages of Sir Walter
Scott’s romance, Count Robert of Paris (1831). This novel presents a series
of implausible romances and valiant encounters in Constantinople at the
time of the arrival of the First Crusade. Sir Paul Harvey, a Scott devotee,
gave a dutiful summary of what is by any standards a rather poor novel
in The Oxford Companion to English Literature:
The scene is Constantinople in the days of Emperor Alexius Comnenus
(1081–1118), and the story centres on the arrival there of the first crusaders,
and in a plot of Nicephorus Briennius, the husband of Anna Comnena,
to dethrone his father-in-law. Anna Comnena herself figures largely in
the novel and provides some of its best pages. Count Robert of Paris, a
proud and valiant Frankish knight, and his Amazonian wife Brenhilda,
are among the crusaders. When they pay homage to the emperor, Count
Robert grossly insults him by seating himself on his throne. He thereby
arouses the wrath of Hereward, an English soldier of the emperor’s
Varangian guard. The count and his wife, by a device of the emperor’s, are
detained as hostages for the crusaders when these cross to Asia; the count
is thrown into prison and rescued by the chivalrous Hereward. Meanwhile
his wife Brenhilda is exposed to the unwelcome attentions of Briennius and
challenges him to a duel, agreeing to surrender herself to him if defeated.
When the time for the duel comes, Count Robert presents himself in her
stead, and as Briennius fails to appear, Hereward fights on his behalf. He
is defeated, but his life is spared by the count in consideration of his past
services. Hereward attaches himself to the count, having discovered his
old Saxon love, Bertha, in the countess’s waiting-woman.3
Harvey’s efficient summary omits some of the more colourful highlights,
such as the fights, first Robert’s with a tiger and then the strangling of the
wicked philosopher Agelastes by an orang-utan, as well as the portrait of
Emperor Alexios Komnenos enjoying a glass of English beer on a hot day.
Count Robert of Paris is almost the last of the Waverley novels and
Scott wrote it at a time when he was unwell. He was a stroke victim and
had only a year to live. He was also under great financial pressure, having
shouldered all the debts of a bankrupt publishing company with which he
had been involved, albeit only marginally. Abysmally plotted and slackly
3
Paul Harvey, ‘Count Robert of Paris’, in The Oxford Companion to English
Literature, ed. Margaret Drabble, 5th edn (Oxford, 1985), p. 233.
348
History, Fiction and Film: Islam Faces the Crusaders
written, Count Robert of Paris is a pot-boiler of curiosity value only.4 The

prose is sententious, didactic and overloaded with parentheses. As an
example of florid and pompous writing which is hard to stomach today,
consider a couple of sentences of dialogue uttered by the commander of
the Varangian guard to Hereward:
All I mean to make thee understand, my dear Hereward, is that there are,
though perhaps such do not exist in thy dark and gloomy climate, a race
of insects which are born in the first rays of the morning and expire with
those of sunset (thence called by us ephemerae, as enduring one day only),
such is the case of a favourite at court, while enjoying the smiles of the most
sacred emperor. And happy is he whose favour, rising as the person of the
sovereign emerges from the level space which extends around the throne,
displays itself in the first imperial blaze of glory, and who keeping his post
during the meridian splendour of the crown, has only fate to disappear
and die with the last beam of imperial brightness.5
And here is the heroine of the romance:
The Countess Brenhilda was now above six-and-twenty years old, with as
much beauty as can well fall to the share of an Amazon. A figure, of the
largest feminine size, was surmounted by a noble countenance, to which
even repeated warlike toils had not given more than a sunny hue, relieved
by the dazzling whiteness of such parts of her face as were not usually
displayed.6
But Scott seems to have expected his farrago to be taken seriously, as he
provided it with footnotes, a glossary and three short appendices. As
already noted, Scott’s sources included Gibbon’s Decline and Fall of the
Roman Empire and Mills’s History of the Crusades. Gibbon’s attitude to
crusaders and crusading was on the whole contemptuous and he presented
the heroism of the crusaders as a function of their ‘savage fanaticism’. Their
barbarian energy was what distinguished them from the Byzantines and
Gibbon’s contempt for crusader fanaticism was balanced by his contempt
for Byzantine decadence and deviousness (a deviousness which contrasted
so badly with, for example, Hunnish honesty). Byzantine politics had
taken on the trappings of ‘the indolent luxury of Asia’. Scott seems to have
gone along with the view that the eleventh-century empire was wealthy,
4
For a more positive (minority) view, see Graham McMaster, Scott and Society
(Cambridge, 1981), pp. 207–15; on Scott and the crusades more generally, see
Elizabeth Siberry, The New Crusaders: Images of the Crusades in the 19th and
Early 20th Centuries (Aldershot, 2000), pp. 112–31.
5
Walter Scott, Count Robert of Paris (Edinburgh, 1871), p. 46.
6
Scott, Count Robert, p. 153.
349
Robert Irwin
degenerate and vulgar. But, unlike Gibbon, he thought that, even so, the
adoption of Christianity had been for the benefit of the East Roman
Empire. Count Robert of Paris seems to have been inspired (if ‘inspired’
really is the right word) by a brief reference in The Alexiad (which was
relayed by Gibbon) to an otherwise not terribly well-known French lord,
Robert of Paris, who, during the oath-taking of the crusader leaders to
Alexios, was bold enough to seat himself on Alexios’s throne.7
As for Mills’s attitude to the crusaders, he was divided between awe and
revulsion. He was awed by their chivalry and prowess, while yet he was
horrified by their fanaticism that was, as far as he was concerned, a product
of the popery which was such a repugnant leading feature of the Middle
Ages. Repugnance for popery mingled with revulsion for ‘the violent
excesses and barbarities of the European mob’ and, most particularly, for
the ‘wretched fanatics’ of the People’s Crusade. Even so, Mills, strongly
influenced by the chivalric revival of the nineteenth century, was also keen
to stress the English role in the crusades: ‘[T]he shores of Palestine may
not be improperly regarded merely as the theatre of English chivalry.’8
Mills made use of such sober sources as Albert of Aachen, Fulcher of
Chartres and William of Tyre. However, he was also fond of referring
to a much more fanciful work: La Gerusalemme Liberata, a romantic
epic published in 1581 by Torquato Tasso.9 In Tasso’s epic, the historic
achievements of Godfrey of Bouillon during the First Crusade serve as
the background to an intertwined series of love stories, most notably the
love of the Christian warrior, Rinaldo, for Armida, the sorceress daugh-
ter of the Saracen king of Damascus. Although Tasso had set out to give
Italy its own epic, such characters as the Amazon warrior, Clorinda, and
such incidents as Armida’s and Rinaldo’s dalliance in the bower of bliss
became popular with painters, poets, and musicians throughout Europe
– and even with some historians who cited Tasso’s fantasy as if it were
real history. Gerusalemme Liberata is about destiny and the fatality of
love and its tone is therefore appropriately idyllic, sensuous and mysteri-
ous. As Christopher Tyerman has summed it up, Gerusalemme Liberata
7
Edward Gibbon, Decline and Fall of the Roman Empire, ed. John B. Bury
(London, 1896), pp. 291–92.
8
Charles Mills, History of the Crusades, for the Recovery and Possession of the Holy
Land, 3rd edn, 2 vols (London, 1822), 1: 80–81. On Mills, see Siberry, New
Crusaders, pp. 10–14
9
For Tasso as an inspiration see, for example, Mills, History, pp. 84, 109, 120.
On Tasso’s influence on crusades historians, novelists, musicians, etc., see also
Siberry, ‘Tasso and the Crusades: History of a Legacy, Journal of Medieval
History 19 (1993), 163–69; Siberry, New Crusaders, esp. pp. 4–5 and 35–37
350
‘reinvented the First Crusade as a romantic tale of chivalry, love and magic
within the more familiar story of Godfrey of Bouillon […] His approach,
appealing to rose-tinted images of valour, loyalty and love was to have a
long history’.10 Tasso was translated into English in the sixteenth century
and was very widely read.
Although it never quite passed out of fashion in Britain, the epic’s
reputation benefited from the Gothick revival of the late eighteenth
century. A new translation was produced in 1824, significantly only a few
years before the publication of Count Robert of Paris. Scott had definitely
read and been impressed by Tasso (he had already read the work in John
Hoole’s translation of 1763) and it is noteworthy that in Scott’s novel,
Waverley, a young lady asks Waverley to help with her translation of
Tasso. So Tasso’s lightly fanciful view of the First Crusade came to Scott
directly, as well as mediated through Mills’s history. A critical account
of the 1824 translation of Tasso in The Eclectic Review should serve to
remind us that hostile approaches to the history of the crusades did not
start with Terry Gilliam, or even with Sir Steven Runciman: ‘The grand
objection however to Tasso’s poem is the false view which it gives of the
achievement which it celebrates […] we must forget that the crimes and
cruelties of the Croisés as well as their fanaticism sank them below the
Moslems […] and we must strive to believe that the delivery of Jerusalem
was an object worthy of the interposition of the highest intelligences.’11
To return to Mills and to Scott’s reading of him, Mills had described the
Emperor Alexios Komnenos as unprincipled, vacillating and cowardly.12
Scott mostly went along with this unsympathetic view, though in Scott’s
novel this was tempered by what he had taken from Anna Komnene’s
Alexiad. He appears to have read Anna in an edition of Du Cange. He
also made use of Du Cange’s account of the Fourth Crusade.13
10
Christopher Tyerman, The Invention of the Crusades (London, 1998), p. 108.
11
On the translation of and impact of Tasso on English literature, see Charles P.
Brand, Torquato Tasso: A Study of the Poet and of His Contribution to English
Literature (Cambridge, 1965), ch. 9. The anonymous reviewer in The Eclectic
Review is cited by Brand, Torquato Tasso, p. 274. For the impact of Tasso
on another belated literary enthusiast for the crusades, Chateaubriand, see
Elizabeth Siberry, ‘Images of the Crusades in the Nineteenth and Twentieth
Centuries’, in The Oxford Illustrated History of the Crusades, ed. Jonathan Riley-
Smith (Oxford, 1995), pp. 365–85 (here p. 366).
12
Mills, History, 1: 95–96.
13
Charles Dufresne Seigneur du Cange, ‘Corpus byzantinae histoire’, in
Nicephori Bryennii […] Annae Comnenae et J. Cinami (Paris, 1670); also author
of Histoire de Constantinople sous les empereurs français (1657).
351
Robert Irwin
As in the earlier crusading novel, The Talisman, Scott also made play
with the imagery of The Thousand and One Nights and similar oriental
fantasies, such as James Ridley’s pseudo-oriental story collection, Tales
of the Genie (1764).14 Although the most obvious source for the sinister
Paradise Garden of Agelastes in Count Robert of Paris can be found in
Tasso’s account of the enchanted garden of Armida, Agelastes’s garden
may also owe something to the medieval stories about the Old Man of
the Mountains and the Paradise Garden of the Hashishins.15
Again, as in another of his earlier novels, Ivanhoe (1820), so also in
Count Robert of Paris: racial and cultural distinctions preoccupied Scott
– between European and Asian, between Greek and Frank, between
Frenchman and Norman, and between Norman and Anglo-Saxon. This
preoccupation with the antagonism between Saxon and Norman, with
the liberties of the Saxons and their dispossession by the Normans was
very much of its time. As Christopher Hill has observed of these notions,
‘the years 1820–80, the years of the Gothic revival, were in a sense their
heyday’ and it ‘became a real stimulus to historical research’.16 In Scott’s
novel, the Saxon Hereward incarnates the natural virtues of plain speaking
and plain dealing. Count Robert is a more sophisticated character, but
to a lesser extent, when compared to the decadent and politic Greeks he
too embodies the same virtues.
Scott’s view of Asians and Africans might be seen as an expression of
nineteenth-century imperialist attitudes. On the Asians, for example:
A party of heathen Scythians whom they beheld, presented the deformed
features of the demons they were said to worship – flat noses with expanded
nostrils, which seemed to admit the sight to their very brain; faces which
extended rather in breadth than length, with strange unintellectual eyes
placed in the extremity; figures short and dwarfish, yet garnished with legs
and arms of astonishing sinewy strength, disproportioned to their bodies.17
As for the Africans, Scott describes ‘six deformed Nubian slaves, whose
writhen and withered countenances formed a hideous contrast with
14
See, for examples of references to the Nights, Scott, Count Robert, pp. 37 and
338 (Douban is the name of the physician in one of early tales in the Nights).
For an allusion to Ridley’s story collection: Count Robert, p. 22.
15
On such stories, see Farhad Daftary, The Assassin Legends: Myths of the Isma‘ilis
(London, 1994).
16
Christopher Hill, ‘The Norman Yoke’, in Puritanism and Revolution (London,
1958), p. 111; cf. Léon Poliakov, Le Mythe Aryen: Essai sur les sources du racisme
et des nationalismes (Paris, 1971), pp. 63–85.
17
352
their snow-white dresses and splendid equipment’.18 Yet it is likely that

Scott’s racism owed little to the contemporary and barely formed ethos
of empire. Rather, he unthinkingly reproduced the racial attitudes of his
sources, most notably Anna Komnene.
Scott, the ‘father of the historical novel’, had difficulty in engaging
seriously with the religious motivations of the crusaders. Most historical
novelists have experienced a similar difficulty. Alfred Duggan, in a brief
guide which he wrote for the National Book League in 1957, was to criticise
Scott’s Talisman on those grounds:
If he [Scott] had stopped to think, he would have seen that a Scotsman
would not ride all the way to Palestine unless he felt pretty strongly about
the fall of the Holy Places; but his crusaders in The Talisman feel no more
hatred of the infidel than if they were engaged in opposite sides in a test
match.
Scott, according Duggan, had no real feel for or understanding of the
Middle Ages.19
Yet, as we shall see, when Duggan came to write his novel about the
First Crusade, he tended to treat it as primarily as a socio-economic
phenomenon. This is all the more surprising in Duggan’s case, as he was a
devout Catholic. Knight with Armour, published in 1950, tells the rather
gloomy story of Roger fitzRalph, a young knight from Bodeham in Sussex,
who having enlisted in the service of Robert of Normandy takes part in
the First Crusade and finds and loses a wife in the course of the expedi-
tion, before finally losing his life in the assault on the walls of Jerusalem
in 1099. Ralph is one of those land-hungry younger sons who, in older
histories of the crusades, used to form the rank and file of the crusader
armies. Generally, land-hunger plays a larger part in the deliberations of
Roger and his companions than does spiritual fervour. Duggan’s novel
is a Bildungsroman in which Ralph learns the disillusioning realities of
life as he moves from youth to manhood. It is a story of the doings of the
great – of Godfrey, Bohemund, Tancred and so on – as seen as through
the eyes of a poor knight, a knight, indeed, who, as he loses servants and
pack animals, has difficulty in retaining even the somewhat lowly rank
he laid claim to at the beginning of the crusade. Roger is something of
a sad-sack figure, dull and shy, the follower and victim of men who are
smarter and grander than he is. He is neglected by the lord to whom he
had sworn fealty, and betrayed by his wife and his cousin, and betrayed
18
19
Alfred Duggan, Readers’ Guides, National Book League, ser. 2, 11 (Cambridge,
1957), 6–7.
353
Robert Irwin
also, one might add, by the vision of the (almost certainly fraudulent)
Holy Lance.
Duggan was forty-five years old when he made his debut with Knight
with Armour. There is a sense in which Ralph’s long arduous journey is in
a disguised form the chronicle of his creator’s return from dissipation to
playing a proper role in society – a story of redemption. Alfred Duggan
was born in 1903 in Argentina, the son of a rich Argentinian settler of Irish
descent and an American mother. As a result of his widowed mother’s
remarriage in 1917, he became the stepson of Lord Curzon and resettled
in England; he studied at Eton and Balliol. Prior to the eventual publi-
cation of Knight with Armour, Alfred’s chief claim to fame had been his
reputation as a rake and a hell-raiser. He spent the days hunting and the
evenings in night-clubs. Evelyn Waugh called him ‘a full-blooded rake
of the Restoration’.20
After his negligently conducted studies, Alfred travelled widely in the
Levant and elsewhere. I should say that he was made to travel in those
parts, for his stepfather was so concerned with Alfred’s drinking that he
kept ‘planning for him to go on archaeological or geographical expeditions
where he would find nothing to drink’.21 Since Curzon was passionately
interested in the Near East, it was there that the young Alfred mostly
travelled. When war broke out in 1939, Alfred joined the Irish Guards.
As a lowly but upper-class private he spent all his wages paying his ser-
geant to clean and press his uniform for him. After fighting in Norway,
he was invalided out and worked first in an aircraft factory and then as
a dairyman. By now the youthful arrogance had quite vanished and it
seems that Alfred would have liked to have spent the rest of his life in
humble employment as a shop assistant or a porter. However, since his
mother would not let him descend so low, he turned to novel writing as
the next best thing.
Before turning to the background of the novel itself, I should like to
note that Alfred Duggan was not merely a fabricator of fictions; his own
life provided the material for yet other fictions. At Oxford he had been the
contemporary of Anthony Powell and Evelyn Waugh. It seems evident to
me that Alfred’s spectacular drinking and his subsequent lowly wartime
assignments, which amounted to a via humilitatis, were used by Powell to
20
On the life and work of Alfred Duggan, see Evelyn Waugh, ‘Alfred Duggan’,
Spectator, 10 July 1964, pp. 38–39; Christopher Sykes, Evelyn Waugh (London,
1975), pp. 290–91; The Reader’s Companion to Twentieth-Century Writers,
ed. Peter Parker (London, 1995), p. 207. Anthony Powell described the young
Alfred as ‘rackety’ in Powell, To Keep the Ball Rolling (London, 1983), p. 100.
21
David Gilmour, Curzon (London, 1994), p. 577.
354
furnish some of the career and character of Charles Stringham in Powell’s

cycle of novels, A Dance to the Music of Time.22 Equally, the attempt by
Alfred’s stepfather, Lord Curzon, to send Alfred under chaperonage on
culturally improving trips to foreign parts where he would find it hard
to get hold of drink is strongly reminiscent of the way the Marchmain
family sought to deal with Sebastian Flyte’s drinking problem in Evelyn
Waugh’s novel Brideshead Revisited from 1945 (in neither the novel nor
in real life did this remedy work).
The environment created for the young Alfred by his stepfather must
have had a considerable role in shaping the future historical novelist.
Alfred Duggan’s step-father was a rather grand figure, that ‘most superior
person’, George Nathaniel Curzon, Marquess Curzon of Keddlestone
(1859–1925), viceroy of India, earl, marquis and Foreign Secretary
(1919–24), a statesman and major player on the international stage – none
of which concerns us here. What is relevant is Curzon’s prominent role
in the chivalric revival of the late nineteenth and early twentieth centu-
ries. This revival was fuelled by the paintings of the pre-Raphaelites, the
writings of William Morris and the rediscovery of Malory. In the 1880s
and 90s the members of the secretive upper-class coterie, the Souls,
entertained a cult of Burne-Jones and the Morte d’Arthur.23 In its revived
Edwardian version the chivalric ethos fed into the British imperial ethos.
Gentlemen and ladies played at managing castles and at holding feasts
and archery contests. Curzon was a particular enthusiast for the Middle
Ages. In 1912 he was one of the select band of aristocrats who jousted at the
Empress Hall in (appropriately) Earls Court, and Viscountess Curzon, his
first wife, presided as the Queen of Beauty.24 Curzon also collected castles
and restored them. In the first instance he restored Tattershall Castle in
Lincolnshire.25 Subsequently, in 1917, he bought the castle and estate of
Bodiam in Sussex. Bodiam Castle, which is late medieval, was restored by
Curzon in the 1920s. Curzon’s imaginative passion for the medieval was
well developed and, as he contemplated Bodiam, he wrote that ‘it could
hardly surprise anyone, were a train of richly clad knights, falcons on their
wrists, were to emerge from the Barbican Gate, for the enjoyment of the
chase, or even were the flash of spearheads and the clatter of iron-shod
22
Anthony Powell, A Dance to the Music of Time, 12 vols (London, 1951–75); see
Hilary Spurling, Handbook to Anthony Powell’s Music of Time (London, 1977),
pp. 156–59.
23
Mark Girouard, The Return to Camelot: Chivalry and the English Gentleman
(New Haven, 1981), p. 209.
24
Girouard, Return to Camelot, pp. 6–7.
25
Girouard, Return to Camelot, p. 138.
355
Robert Irwin
hooves to indicate the exit of a party with more serious intent’.26 As a

youth, Alfred Duggan must surely have participated in such reveries of
the past. As an adult, he researched Knight with Armour while living in
Bodiam Manor on the castle’s estate. It is therefore not surprising that
Ralph, the protagonist of that novel, is the son of Osbert fitzRalph, who
‘held the manor of Bodeham in Sussex’. More generally, throughout the
novel one finds a detailed interest in the way in which castles were man-
aged, defended and attacked.
Like a medieval squire, Alfred Duggan grew up in a world of castles
and horses. In fact as a young blood, he had kept a string of hunters.
His detailed familiarity with horses, their capacities and distinguishing
characteristics, as well as the techniques of girthing and loading features
again and again in the course of the narrative of Knight with Armour:
‘Jack, the warhorse, was ten years old, and considerably past his prime;
but he was perfectly trained, ran straight at the mark without guidance
[…]’; ‘All his life he had been terrified of the way the ground comes up
to hit a rider whose horse trips at full gallop […]’; ‘Blackbird was a big,
powerful horse, not very fast, which made him safe in the charge, but
sound in legs and wind, and with the deep ribs that mean staying-power
[…]’27 Duggan grew up in a world in which horses were routinely used
not just for hunting, but also for warfare, haulage and deliveries. In that
important sense, Duggan was actually closer to his beloved Middle Ages
than he was to our own age. Zoé Oldenbourg (on whom more shortly)
also grew up with horses and was passionately devoted to them. They both
belonged to the last generation of historical novelists to be thoroughly
familiar with this crucial aspect of pre-modern life.
More generally, Duggan interested himself in the mundane practicali-
ties of going on crusade. When he eventually turned to novel writing, he
wrote, of course, as a man who had experienced the realities of war. He
knew how boring the soldier’s life was and how much time one had to
spend arranging transport and supplies. Duggan’s First Crusade is a hard
slog. It is also something of an educational experience, as most of Roger’s
companions on the march are prone to give lectures – on Byzantium, on
Greek Orthodoxy, on crusader goals and strategy – rather as if they were
school-teachers specialising in history who just happened to be going on
crusade.
‘Men resemble their times more than they do their fathers’: although
Duggan’s fascination with the Middle Ages has its roots in his personal
background, it is also the case that it was something in the air. The
26
Gilmour, Curzon, p. 463.
27
Alfred Duggan, Knight with Armour (London, 1950), pp. 25, 103, 112.
356
medievalist enthusiasms of Britain in the 1940s and 50s were in part a

legacy of the Second World War and a stress on the insignia of war –
cap-badges, flags and so forth – and the interest in bright colours, the
pageantry and heraldry were further fanned by the coronation of 1953.
This new version of the Middle Ages appears in films like Henry V (1944)
and The Story of Robin Hood (1952), in paintings by John Piper and other
members of the neo-Romantic school and in novels by Hilda Prescott,
Cynthia Harnett, T. H. White and others.
Finally the single most important thing to be said about Knight with
Armour is that both its contents as well as the simple act of writing it
testify to Duggan’s contrition for his misspent youth. The Duggan fam-
ily was Catholic and Alfred, after flirting with Marxism and atheism as
a young man, was to return to his faith in the last part of his life. By the
time he came to write, his dissipated alcoholic past was behind him. Yet
it is noticeable just how many references there are to wine in the course of
his first novel. The crusaders breakfast on wine, just as Duggan had once
breakfasted on whisky and brandy. One senses that the dreadful tempta-
tion was always in his mind as he wrote. Roger’s and Alfred’s paths to
redemption are both bleak ones. Knight with Armour is a kind of votive
offering which testifies to its author’s interior crusade.
In an obituary of his friend for The Spectator, Evelyn Waugh drew atten-
tion to the fact that Duggan’s literary and moral trajectory was opposite
to that of so many literary artists. While people like Scott Fitzgerald and
Dylan Thomas started out with vast literary talents and ambitions, which
they then proceeded to destroy in devastating drinking bouts, Duggan
‘emerged from his years of dissipation with his mind acute, his remarkable
memory unimpaired and a prose style perfectly fitted for use’. Duggan
was to write fourteen more historical novels, most of which are set in
the Middle Ages. He does not seem to have been ashamed of his craft.
Oldenbourg, by contrast, actually denied that she was a historical novelist.
Although Zoé Oldenbourg wrote several novels set in the Middle Ages,
Joie des pauvres (1970), translated as Heirs of the Kingdom (1972), is, I
think, the only one to deal with the First Crusade. The (sarcastic) refer-
ence in the title is of course to the New Testament: ‘Blessed are the poor,
for they shall inherit the Kingdom of Heaven’. Oldenbourg, like Duggan,
shuns the war-tents of the great and tells the story of the First Crusade
from the point of view of the humble, though her chosen protagonists
are much lowlier than Roger fitzRalph. The core of what amounts to a
kind of group protagonist, or millennial chorus, is provided by a group
of weavers from the suburbs of Arras: Jacques the weaver and his wife,
the glamorous blonde Marie, the murderous Elie le Grêlé, and Alix, the
repentant prostitute. This core group picks up other figures on the march
357
Robert Irwin
including Brother Barnabé, an itinerant monk; Evrard, a knight who

prefers to associate with the poor and humble; Saint-John, a crazy man
who thinks he is the reincarnation of John the Baptist; and Philotheus,
a eunuch who does not seem to have lost his ability to experience desire.
Craziness and eunuchry apart, Oldenbourg’s characters are mostly ordi-
nary people swept up in extraordinary events. The viewpoint shifts from
one protagonist to another and then at times Oldenbourg takes it upon
herself to speak for them as a group; she addresses the reader directly and
her prose acquires a burning, incantatory quality – the First Crusade
envisioned as a passion play.
On the way the weavers and their travelling companions argue, make
love, sire children and learn new crafts. There are biblical resonances to
their years in the wilderness. However, the fierce Christianity of these
pilgrims draws as much upon folklore as it does upon any canonical text
and there is a fairy-tale quality to their beliefs: Peter the Hermit receives
from God a parchment made from snow crystals; in the East there is no
winter and golden apples grow on the trees and silk costs less than linen
does in France; the capture of Jerusalem will inaugurate a thousand years
of peace; and there is a more earthy dream of Jerusalem as ‘a city paved
with barley cakes, where every leaf on the trees was a cooked bean and
the wood fires had fagots of toasted bread you could crunch between
your teeth’.28 The crusaders’ fantasies are regularly contrasted with the
gruelling hardships they endure, first as weavers and then as pilgrims on
the long road to salvation.
The capture of Jerusalem is both the fulfilment of a dream and the gate-
way to yet deeper disillusion. In the immediate aftermath of the conquest,
the poor are put to work dealing with the corpses of those massacred by
the crusaders. Oldenbourg’s prose here is revoltingly vivid and detailed
and there are surely resonances here of the mass graves and death camps
of World War Two. In an autobiographical memoir by Oldenbourg we
find the remark: ‘The Jerusalem of Joie des Pauvres is so beautiful that
thousands of hearts burn with love for it and they are consumed like moths
in the flame; they arrive there only to discover a horror which exceeds all
understanding. And yet that is Joy even so’.29
Discouragingly (from my point of view at least), Oldenbourg prefaces
her novel with an Author’s Note which announces that, ‘[T]his is not a
historical novel.’ She goes on to elaborate: ‘The author wishes to declare
her firm and considered conviction that: (1) To praise an author for his or
her scholarship, whether real or presumed, amounts to calling that author
28
Zoé Oldenbourg, The Heirs of the Kingdom (London, 1972), p. 225.
29
Oldenbourg, Visages d’un Autoportrait (Paris 1977), p. 354.
358
a pedant. (2) Source material is freely available to anyone and the writer
is at liberty to make use of it if he so wishes.’30 Oldenbourg then argues
that whatever the surface content of a novel may be, all novels are in the
end about their authors. If the reader wants the conventional historical
novel ‘combining entertainment with a modicum of historical instruc-
tion’, Joie des pauvres is not it. Rather, Oldenbourg has produced a book
about ‘the human condition’.
It is likely that Oldenbourg used René Grousset’s Histoire des croisades
(1934–36) for some of the humdrum chronology. It is even more prob-
able that the plot and spirit of Joie des pauvres owes an immense amount
to La Chrétienté et l’idée de Croisade by Paul Alphandéry and Alphonse
Dupront.31 This was a pioneering work on the history of popular men-
talités which sought to understand the movements of people swept by
eschatological panics and millennial enthusiasms and the mass psychology
of the crowd on pilgrimage. The subheadings of La Chrétienté et l’idée
de Croisade can be used as a skeleton guide to the story told in Joie des
pauvres: ‘From the western lands to Antioch: the roads, the first trials.
The crusade army at Antioch: the poor, the Tafurs, the eschatological
crisis. Visions and prophecies: the Holy Lance, the victory at Antioch,
the march to Jerusalem. The power of the poor: the capture of Jerusalem
and the triumph of poverty, the eschatological fulfilment.’ Only the
fulfilment is queried in the concluding chapter of Oldenbourg’s novel.
Alphandéry’s preoccupation with biblical quotation as giving the First
Crusade chronicles their imagery, vocabulary and texture is faithfully
echoed in Oldenbourg’s fiction.
It is also possible that Oldenbourg had read Norman Cohn’s enor-
mously influential The Pursuit of the Millennium. Cohn’s study of millenar-
ian expectations ran from the early Middle Ages up to the messianic reign
of John of Leiden in the sixteenth century. Cohn included a section on
‘The poor in the first crusades’, which dealt with such topics as the appeal
of Peter the Hermit to the poor, the reverence of Raymond of Aguilers
for the poor, the belief that it was the special destiny of the pauperes to
take Jerusalem, and the activities of the Tafurs32 (Oldenbourg’s knight,
Evrard, may be inspired by le roi Tafur, who ‘is said to have been a Norman
knight who discarded horse, arms and armor in favour of sackcloth and a
30
Oldenbourg, Heirs of the Kingdom, p. i.
31
1st edn, 2 vols (Paris, 1954–59); 2nd edn with postface by Michel Balard (Paris,
1995).
32
Norman Cohn, The Pursuit of the Millennium: Revolutionary Messianism in
the Middle Ages and its Bearing on Modern Totalitarian Movements (London,
1957), pp. 40–52.
359
Robert Irwin
scythe’). Unlike Pope Urban and many of the great lords, the poor became
obsessed with capturing the Heavenly City: ‘In their eyes the crusade
was an armed and militant pilgrimage, the greatest and most sublime of
pilgrimages’.33 It was also ‘a collective imitatio Christi’.34 In a concluding
chapter, Cohn underlined the relevance of the study medieval millenarian-
ism for the understanding of such contemporary phenomena as Nazism
and Communism – both movements which sought to celebrate the epic
exploits of a chosen people, the fulfilment of eschatological expectations,
and the killing of the Jews in preparation for a (secular) millennium.35
Zoé Oldenbourg was born in 1916 in St Petersburg, the daughter of a
historian and journalist. Though she spent the earliest years of her life in
Russia, when she was nine her family fled to France and Oldenbourg was
educated at a French lycée. She studied theology briefly in England. She
subsequently became a painter and it was only after the Second World War
that she turned to literature. Although she is best known for her novels
about the Middle Ages, especially about Cathars and crusaders, she also
wrote novels about Russian emigré life, as well as historical non-fiction
and two autobiographical memoirs. I do not think it entirely fanciful
to speculate that Oldenbourg’s story of the disillusionments of the First
Crusade projects back into the past, her own perceptions of twentieth-
century millennial dreams and disillusionments. She was, after all, an
eyewitness of Russia’s turmoil during and after the Bolshevik revolution.
As she wrote in the first of her autobiographical memoirs, ‘I have been
the witness of great events, albeit not a witness from the grand box. Just
like everyone else.’36
In declaring that she had written about ‘the human condition’, she
was probably intending to refer to André Malraux’s novel, La Condition
humaine (1933), in which Malraux, speaking through one of his protago-
nists, appears to suggest that an individual’s life only acquires meaning in
death: ‘Qu’eût valu une vie pour laquelle il n’eût pas accepté de mourir?’
Malraux’s novel, set against the background of an abortive Communist
insurrection in Shanghai in 1927, is a chronicle of defeated idealism, disil-
lusionment and death on a massive scale. For Malraux, individual exist-
ence is anguishing and absurd. Human dignity and meaning are found
in fraternity, revolution and sacrificing oneself for a transcendent cause.
It is easy to see how such notions can be reapplied to the First Crusade.
However, it would not do to overstress the parallels between Oldenbourg’s
33
Cohn, Pursuit of the Millenium, p. 43.
34
Cohn, Pursuit of the Millenium, p. 44.
35
Cohn, Pursuit of the Millenium, pp. 307–19.
36
Oldenbourg, Visages, p. 11.
360
medieval romance and modern realities, for she claims to like history for
its beauty rather than for its relevance to modern concerns.37
As Paul Bailey noted, in a review of Russell Hoban’s Pilgermann, it ‘is
not a historical novel about the dark ages of the kind one associates with,
say, Zoe Oldenbourg […]’ 38 Not, then, a conventional historical novel and
Hoban is not in any sense a conventional novelist (apart from anything
else, he likes to discuss the developing drafts of his novels with his analyst,
‘which has helped him be “good friends” with his head’).39 Russell Hoban
was born in 1925 in Pennsylvania. At first he earned a living as a writer
and illustrator of children’s books. The Mouse and His Child (1967) is
an enduring classic of children’s fiction. Subsequently he turned to the
writing of adult fiction. Hoban’s early idiosyncratic adult novels, such as
Kleinzeit and Turtle Diary, have their fans, but Hoban’s chief claim to
literary fame was as the author of Riddley Walker, a fantastic literary vision
of post-holocaust Britain, narrated in a specially devised post-holocaust
English that is both barbarous and poetic.40 Russell Hoban died in 2011.
I met Russell Hoban at a Christmas party in South London in 1981. He
was looking a bit bored, but he brightened up when I told that I worked on
crusader history, for he was at that time still at the planning and research
stage of his novel, Pilgermann. Among other things, I suggested that he
consult the works of S. D. Goitein both for what they had to say specifically
about Jewish life in the Levant in the eleventh century and more generally
for details of material life in the same period.41 Hoban not only did so, but
wrote to Goitein (then at Princeton) asking for advice. Goitein’s reply was
characteristically both helpful and eloquently phrased. Although Hoban
did not allow himself to be dictated to or restricted by his research, he
read widely in both primary sources, like the Gesta Francorum, and the
secondary sources including the Wisconsin History of the Crusades, as well
as Prawer, Erdmann and Smail. As for Runciman, Hoban told me that his
prose is ‘not the juiciest kind of writing, very conservative writing. His
37
Oldenbourg, Visages, pp. 11–12.
38
Pilgermann, reviewed by Paul Bailey in The Standard, 19 March 1983, p. 199.
39
The Reader’s Companion, ed. Parker, p. 335.
40
On the life and work of Hoban, see Christine Wilkins, Through the Narrow
Gate: The Mythological Consciousness of Russell Hoban (Cranbury, NJ, 1989).
41
In addition to his magnum opus, A Mediterranean Society: The Jewish
Communities of the Arab World as Portrayed in the Documents of the Cairo
Geniza, 5 vols (Berkeley, 1967–88), Shelomo Goitein had published several
short articles on the evidence concerning the immediate prelude and aftermath
of the First Crusade from Jewish letters.
361
Robert Irwin
take on the crusades is a little less passionate than mine’. Pilgermann cost
its author £3000 in reference books.42
The dust-jacket of the first edition of Pilgermann (London, 1983) con-
tains a statement by the author concerning the novel’s origins:
It was on May 15th and 16th that I saw, first in the evening, then in the
night, then in the dawn, the ruined stronghold of Montfort in Galilee,
built in the twelfth century by the knights of the Teutonic Order of Saint
Mary and enlarged by them in the thirteenth century. The look of the
stars burning and flickering over the ruin, those three stars between the
Virgin and the Lion with their upward swing like the curve of a scythe,
the stare into the darkness, the hooded eagleness of the stronghold high
over the gorge, the paling into dawn of its gathered flaunt and power pre-
cipitated Pilgermann into his time and place and me into a place I hadn’t
even known was there. Later on, within the body of the novel, this kind
of transtemporal intuition is echoed in a passage which begins: ‘With
the ear of the mind I hear the army of the Franks on the march, I hear
the massed clinking of the tread, I hear the horses snort and whinny, the
rattling of leaves of iron […]’43
The eponymous Pilgermann is a German Jew who, after a passionate
night with Sophia, the wife of a gentile tax-collector, is castrated by an
anti-Semitic mob. Then he has a vision of Jesus Christ and engages in
wise-cracking dialogue with Him. Inspired by the vision, he decides to
make his way to Jerusalem, even as the hordes of the First Crusade are
setting out. Pilgermann’s pilgrimage is phantasmagoric and the people he
meets on the way are strange – the decapitated tax-collector, a purveyor
of fraudulent relics, a bear which is thought by its owner to be God, a
talking pig, an ill-fated pilgrimage of children and Death himself. There
are echoes of such medieval tropes as the Ship of Fools and the Dance of
Death. Pilgermann enters Muslim Antioch in advance of the coming of
the armies of the First Crusade and is befriended by a Persian merchant,
Bembel Rudzuk. Under Rudzuk’s patronage, Pilgermann designs and
makes a geometrically patterned pavement of tiles. Its pattern, which in
the mind’s eye can be seen to extend beyond its actual boundaries to infin-
ity, hints at a mysterious sense in which consciousness may transcend the
human lifespan. The novel ends with the crusaders forcing their way into
Antioch and slaughtering the inhabitants, including Pilgermann – but
by then the reader has understood that the individual consciousness of
42
The information here and much of what follows is based on an interview that I
conducted with the author on 5 July 1999.
43
Hoban, Pilgermann, p. 99.
362
Pilgermann is part of a much larger pattern and not really bounded by

time and space.
Hoban’s mystical sense of the past suchness of things is unusual but
not unique. The subject has been touched on in Marghanita Laski’s
Ecstasy in which she briefly discussed the sensation in which ‘a unitive
symbol triggers an experience of which it is believed that the ecstatic has
been translated into time past’. Among her witnesses here were William
Wordsworth of The Preludes; Richard Jefferies, author of The Story of My
Heart; and the historians Edward Gibbon and Arnold Toynbee. Jefferies,
for example, as he gazed at a Roman wall at Pevensey recorded that,
‘[S]eeing it thus clearly, and lifted out of the moment by the force of
seventeen centuries, I recognised the full mystery and depth of things
in the roots of the grey grass on the wall, in the green sea flowing near.’44
Pilgermann is a ‘historical novel’ which attacks history and the chrono-
logical arrangement of things. When Hoban’s narrator comes to meditate
on the chronology of historical guilt – guilt for the crucifixion of Christ,
or for the massacres perpetrated by crusaders, or for the Holocaust – he
comments that ‘a purist may argue that God, being everywhere in time
at once, would not have written one thing “before” and another “after”
but that argument is well answered when we point out that the Creator
characteristically employs a sequential mode of presentation, even going
so far as to work six days one after the other and rest on the seventh’.45 Yet,
from Hoban’s mystical perspective, chronological sequence is a convention
and the violence of the sacks of Jerusalem and Constantinople is constantly
re-enacted in a kind of eternal nunc stans. ‘The fall of Constantinople that
begins in 1204 with the French and the Flemings is consummated by the
Turks in 1453; what is required is not that a particular enemy shall attack
the dome, only that by sword and fire beauty shall be brought low, only
that the holy books shall be trampled. Echah!’46 Pilgermann (and Hoban
presumably) are appalled by history and ‘the rushing forward, the rushing
back, that so much of history is made of ’.47
At one point in the novel, Pilgermann finds himself weeping from
‘an attack of history’. Pilgermann dead has escaped the constrictions of
chronology and therefore he observes that: ‘I can’t tell this as a story
because it isn’t a story; a story is what remains when you leave out most
44
Quoted in Marghanita Laski, Ecstasy: A Study of Some Secular and Religious
Experiences (London, 1961), pp. 111–15.
45
46
Hoban, Pilgermann, pp. 90–91.
47
363
Robert Irwin
of the action: a story is a sequence of picture cards’. 48 So, although the

crusades and the adventures of a castrated Jewish pilgrim with a propensity
for puns are at the foreground of Hoban’s novel, ultimately the book’s
concerns transcend the mundanely historical, as Hoban meditates on the
nature of the universal mind, a consciousness which transcends time and
individuals. ‘It seems to be looking out through the eyeholes of my face
for as long as I have a face to look through.’49
While at Hoban’s style at times carries resonant echoes of the Bible and
Qur’an, at other times it is slangy and pun-laden. Some of the dialogue is
reminiscent of the repartee of Hollywood screwball comedy. Hoban told
me that the voice is a direct development from Kleinzeit, the eponymous
hero of his first novel. Pilgermann bristles with anachronistic references
to Hieronymus Bosch, Vermeer, the Tarot pack, the Holocaust and so
forth. The first edition also contains at least one unintentional anachro-
nism, for, when I congratulated the author on teasing his readership by
including the deliberate anachronism of having people drink coffee in
the eleventh century, he looked rather pained (in a subsequent edition,
substitute drinks have been provided). Nevertheless, despite its inten-
tional and unintentional anachronisms, Pilgermann is exceptional – and
really rather medieval – in taking the religious dimension seriously. Its
language, imagery and use of typology draws heavily on the Old and
New Testaments, as well as on the Qur’an (he bought six copies of the
Qur’an in order to discover the best translation). Even before he started
to write the novel, he had been steeping himself in the book of Isaiah.
Pilgermann, like a medieval treatise, is dense with allegory and typology
and Hoban’s transtemporal and moralistic approach to history means
that his superficially anachronistic novel is actually more ‘medieval’ in
its concerns than any of its competitors.
Hoban, who is Jewish, meditates on, among other things, the histori-
cal significance of the crusaders’ attacks on Jews on the way to Jerusalem.
Hoban did not like crusaders. In a conversation with me, he characterised
himself as ‘a twentieth-century Jew, a guy who sits at a desk, thinking of
these eleventh- to twelfth-century predators – real hard men these Jew-
killers. These stories really talked to me’. For Hoban, the crusaders were
like the Lord George Gordon rioters, as described in Dickens’s Barnaby
Rudge. The crusaders ‘were men with armour and energy and a liking for
violence’. The novel contraposes Pilgermann, the Jewish victim-type, with
Bohemund, the triumphant warrior. Bohemund is ‘a kind of centaur, who
has come into existence at the same as the stirrup was invented’. Bohemund
48
49
Hoban quoted in Wilkins, Through the Narrow Gate, p. 101.
364
is ‘questing on the death-track that is Christ’. ‘So whoever asks, “Are you
a Jew?” is saying at the same time, “Are you one of those who has not
so far been slaughtered?”’ Bohemund is explicitly everything that the
narrator is not. Pilgermann’s (and Hoban’s) cosmic consciousness stops
short of empathy for Bohemund. He is a horrid, fascinating barbarous
mystery. Towards the end of the novel, as the Franks are storming into
Antioch, Pilgermann pours a bucket of excrement on Bohemund and, a
few pages later, Bohemund takes his bloody vengeance on Pilgermann.
However, though Hoban confirmed to me that Bohemund is everything
that Pilgermann is not, he then cryptically added, ‘Everything contains
its opposite.’
Scott, Duggan, Oldenbourg and Hoban, in their very different ways,
attempted exercises in empathy. Through their novels, they sought to
understand what it must have been like to have been alive in the eleventh
century and to have been a witness of or participant in one of the greatest
events of that century. As Sir George Trevelyan once famously declared,
‘The poetry of history lies in the quasi-miraculous fact that once, on
this earth, once on this familiar spot of ground, walked other men and
women, as actual as we are today, thinking their own thoughts, swayed
by their own passions, but now all gone, one generation vanishing into
another, gone as utterly as we ourselves shall shortly be gone, like ghosts at
cockcrow’.50 The ‘poetic sense of transience and the tragedy of the human
condition, of men and women trapped by the frailty of flesh in time and
circumstance […]’ is a leading feature both of the non-fiction of historians
like Trevelyan and of the better class of historical novelist.51 But when
we turn to consider film-makers, it is hard, almost impossible, to detect
any traces of imaginative empathy. As we shall see, modern concerns and
attitudes are imported wholesale into the Middle Ages.
Apart from al-Hariri’s The Splendid Story of the Crusading Wars (1899),
there was no modern Arab historiography of the crusades to speak of
until after the Second World War. Modern Arab historians have tended
to present the crusading movement as the precursor of imperialism, or
Zionism, or both. Similarly they have tended to present Saladin as the pre-
cursor of one or other vaunted great leader and unifier of the Arab nation:
perhaps Nasser, or Saddam Hussein, or Hafez Assad.52 Arab historians
50
George M. Trevelyan, An Autobiography and Other Essays (London, 1949),
p. 13.
51
This characterisation of Trevelyan’s poetico-historical sensibility comes from
David Cannadine, G.M. Trevelyan: A Life in History (London, 1992), p. 190.
52
Emmanuel Sivan, ‘Modern Arab Historiography of the Crusades’, in
Interpretations of Islam. Past and Present (Princeton, 1985), pp. 3–45; see Carole
365
Robert Irwin
have drawn heavily on western historians like Runciman. The novels of

Sir Walter Scott have influenced the perceptions of both Arab historians
and novelists. Scott’s Tales of the Crusaders was translated under the title
Salah al-Din by Najib Haddad (1867–99). Haddad also dramatised The
Talisman. There have been lots of dramatisations and literary versions of
the career of Saladin in the Arab world, but I have found no novels about
the First Crusade. Presumably the latter subject is doubly unattractive,
because not only were the Muslims defeated in the 1090s, but Arabs did
not even play a leading role in the armies facing the crusaders.
Arab novels and films on the subject of the crusades preached the need
for strong leaders (like Saladin and Gamal Abdul Nasser) and stressed the
need for Arab unity – most specifically the unity of Muslim and Christian
Arabs in the face of Western imperialist aggression. Since Egypt is and
always has been the centre of the Arab film industry, it is not surprising to
find Egypt’s leading role in repelling the crusades stressed in films. Ibrahim
Lama’s Saladin (1941) stole spectacular sequences from DeMille’s The
Crusades (on which more shortly) and inserted them in his low-budget,
small-cast film in which they did not fit at all. After some criticism of this
procedure, the filched scenes were removed but this of course did nothing
for narrative coherence. So one sees a vast battlefield and then what should
be a might cavalry charge is executed by a dozen local extras.53
Yusuf Chahine’s Al-Nasr Salah al-Din (‘Saladin the Victorious’, 1963)
has a somewhat higher reputation amongst connoisseurs of Arab cinema.
Chahine, who comes from a Christian Arab family in Alexandria, is one
of the most famous Arab film-makers. His Saladin film was based on an
idea by Yusuf Siba‘i (1917–78) a leading romantic novelist and short-story
writer. ‘Abd al-Rahman al-Sharqawi, a poet and novelist famous for his
writings on propagandistic socialist themes, wrote the script and even
the Nobel-prize-winning Naguib Mahfouz took a hand in developing
the script.
The story is as follows: Saladin defeats the crusaders at Alexandria.
Then, provoked by Renaud de Châtillon’s massacre of Mecca pilgrims,
Saladin fights crusaders at Hattin and kills Renaud de Châtillon in
single combat. Then the Third Crusade arrives and takes Acre thanks
to the treachery of its governor. The crusaders win again at Ascalon, but
then they are defeated on the desert road to Jerusalem. The crusaders are
divided, for Richard wants a truce, whereas Philip and Conrad do not.
Hillenbrand, The Crusades: Islamic Perspectives (Edinburgh, 1999), pp. 4–5.

53
Ahmed Raafat Baghat, ‘Cinéma et Histoire: de “un baiser dans le désert”
á “l’Emigré”’, in Egypte, 100 ans de cinéma, ed. Magda Wassaf (Paris, 1995),
pp. 179–80.
366
Arthur tries to murder Richard with a poisoned arrow, but Saladin heals
him. There is a final battle and Issa, the romantic young hero, smuggles
some new kind of war-winning Greek fire into Jerusalem. Richard defeated
proposes a truce and the crusaders agree to leave Holy Land. Only Louise
de Lusignan remains in the arms of Issa.54
Chahine’s film shows the local Muslims and Christians combining to
drive out the alien invader. Islam and Christianity can and should coexist.
In the course of the film Saladin repeats again and again that all religions
are tolerated and that Jerusalem is open to all. There is more than a pass-
ing resemblance between al-Nasr Saladin and Gamal Abdel Nasser. Arab
problems in facing the crusaders are shown to be in large part due to trai-
tors within their own ranks, notably the wicked general, Omar, and the
treacherous commander of the Arab garrison at Acre. The film takes it
for granted that the crusaders are all malignant, greedy warmongers – all
except Louise de Lusignan and she marries Saladin’s Christian general,
Issa al-Awam.55
Despite the large budget and the distinguished list of people who
worked on this film, the result is not entirely satisfactory. As Chris
Peachment put it in Time Out: ‘After years of Anthony Quinn playing
Arabs, the Egyptian film industry finally wreaks dreadful revenge. With
a nomadic cast of zillions, Saladin sets out to boil your brains with
three hours of unrelieved apology for hanging on to Jerusalem. Caught
up in endless exposition Richard I is no Coeur-de-Lion, but, like all the
Europeans a red-wigged mongoloid given to lines like “We can take
Acre by lunchtime”. Saladin smoulders […]’56 Chahine has returned to
the themes of tolerance and East-West relations in the Middle Ages in a
more recent big-budget film, Destiny (1997). This focuses on the conflict
between al-Ghazālī and Ibn Rushd (Averroes). The treatment of the
philosophical issues raised by Ghazālī’s The Incoherence of the Philosophers
(which is treated as an attack on liberal thinking) is broken up by some
lively song-and-dance numbers, as well as by various fights and murders.
However, when it comes to feature films about the crusades, the west-
ern contribution is not so very distinguished either. Western films on the
crusades were also strongly influenced by the literary legacy of Tasso and
Scott. La Gerusalemme Liberata (1911), directed by Guazzoni, was one
of the earliest four-reel films and was, as its title suggests, based, however
loosely, on Tasso’s epic. Its huge sets and cast set the precedent for other
54
Youssef Chahine l’Alexandrin, ed. Christian Bosséno (Paris, 1985), pp. 26–27,
142–43.
55
Baghat, Cinéma, p. 180.
56
Chris Peachment, Time Out Film Guide, 6th edn (London, 1998), p. 751.
367
Robert Irwin
better-known silent historical epics such as Cabiria (1914) and Ben Hur
(1925) (incidentally Oldenbourg attributes the development of her
historical consciousness in part to her having gone as a child to see such
early silent epics as Last Days of Pompei and William Tell).57 There was a
remake of Gerusalemme Liberata (also known as The Mighty Crusaders)
in 1957, directed by Carlo Bragaglia. I have not seen it, but The Motion
Picture Guide describes it as a ‘costume spectacular with poor dubbing
and scantily dressed women’.58
I come now to Cecil B. DeMille, of whom Nicholas Clerihew Bentley
declaimed:
Cecil B. DeMille,
Rather against his will,
Was persuaded to leave Moses
Out of The Wars of the Roses.
In his film The Crusades (1935), DeMille set out to teach the American
public the real story of the crusades. He was explicit about his commit-
ment to history: ‘The duty of an historian is to give an accurate report
of the known and proven facts. The duty of an historical dramatist,
however, is to fill the crevasse between them’.59 DeMille hired ‘one of the
top historians in the field’ to do the job. This was the popular historian
Harold Lamb and DeMille actually bought the film rights to Lamb’s The
Crusades: Iron Men and Saints. Lamb’s book was the stirring story of the
men who ‘started out on what they called the Voyage of God’. The epic
tale starts in ‘Dark-Age’ Europe:
The damp forests were there as before, and the grey ruins where owls
glided from the vines. The wolves hunted in packs as usual. Only small
patches of land were cultivated, in stony ground, near the hamlets. Clay
and stone huts, roofed with thatch, clustered below the hewn logs of a
lord’s hall and a stone tower.60
The story culminates in the siege of Jerusalem:
Tortured wood creaked and crashed. Ropes whined and hissed, and the
long arms of the stone casters shot towards the sky. Heavy stones soared
and fell splintering against the wall. Crossbows snapped, and the short
57
Oldenbourg, Le Procès du Rêve (Paris, 1982), p. 47, see also p. 109.
58
The Motion Picture Guide 5 (1986), 1950.
59
Albert Johnson, ‘The Tenth Muse in San Francisco’, Sight and Sound,
Jan-March, 1955, p. 152.
60
Harold Lamb, The Crusades: Iron Men and Saints (London, 1930), p. 20.
368
black bolts whirred up, towards the ramparts. Beneath the tumult of sound
echoed the steady thud-thud of the rams pounding against the foundations.
The crusaders force their way into the city and a terrible massacre ensues:
Old men stood with the calm of fatalists, and saw the heads of their sons
roll to musty carpets. They in turn, were struck down, their skulls shattered
by axes or their bowels cut open by a sword’s edge. Women clung to their
bending knees, screaming and panting until steel wrenched open their soft
bodies and they cried out no longer.61
Once the crusaders have finished massacring everyone they can find,
‘the anger left them, and they were filled with happiness. They stared at
the small white churches and smiled. They walked with clasped hands,
saying little’.62
DeMille was a great admirer of this sort of narrative and he thought
he knew fine writing when he saw it. In particular Lamb was noted for
his sympathetic understanding of Middle Eastern people (Lamb spread
himself quite widely as a historian and also wrote on such subjects as
Genghis Khan, Tamerlane and Babur). Lamb’s film-treatment of the
crusades is somewhat compressed. What happens is that Saladin and ‘the
Saracens invade the Holy Land, desecrating crosses, burning books and
selling Anne Sheridan into slavery’.63 Anne Sheridan (then just starting
on her career in film) as she is about to be sold into slavery cries out ‘The
Cross! The Cross! Give me the Cross!’ Foolishly amidst all the massacring,
Saladin spares Peter the Hermit, despite the Hermit’s warning that, if his
life is spared, he will go to Europe to stir up a thing called a crusade. But
Saladin is too arrogant to heed the warning. Over in England, Richard is
keen to go on crusade because this will give him an excuse not to marry
a French princess. Unaccountably, while travelling out East, he agrees to
marry Berengaria (Loretta Young ‘in an exceptionally attractive wig’), but,
not being very keen on women or sex, he sends his sword to deputise for
him at the marriage. Once in Palestine all the other kings disagree with
Richard’s plan to try and take Jerusalem. While Richard is fighting in
Palestine, Berengaria is kidnapped. Peter the Hermit is also captured and
martyred. But Saladin has fallen in love with Berengaria. Finally, Richard
agrees to a treaty with Saladin under which all Christians except him are
allowed to enter Jerusalem. Arm-in-arm with Berengaria, he watches some
Christians entering Jerusalem. Love has replaced the will to conquer in
61
Lamb, Crusades, p. 200.
62
Lamb, Crusades, p. 204.
63
George MacDonald Fraser, The Hollywood History of the World, 2nd edn
(London, 1996), p. 56.
369
Robert Irwin
his heart. Berengaria lays Richard’s sword on a pillow. Somewhere in the

background the crusaders are singing.
Ernst Lubitsch, a regular visitor to the set, declared, ‘I’m hypnotised.
There isn’t a cocktail-shaker or tuxedo in sight’. The mild and bespectacled
Harold Lamb was also a regular visitor. When asked by DeMille what
he thought of one of the battle sequences, he replied that the extras were
not showing enough hostility: ‘What you need is a thousand battle axes
red with blood’, he said. ‘That would make it a shambles,’ DeMille said.
‘Exactly. And also history,’ Lamb replied.
Graham Greene, who in the 1930s did a lot of film-reviewing, enjoyed
the typical DeMille set-pieces with the vast crowds of extras moving to
fulfil the director’s vision of history. However, he commented that:
Mr de Mille’s evangelical films are the nearest equivalent today to the glossy
German prints which decorated mid-Victorian Bibles. There is the same
lack of period sense, the same stuffy horsehair atmosphere of beards and
whiskers, and, their best quality, a childlike eye for detail’.64
DeMille had been influenced by the Italian silent films and his manage-
ment of the crowd scenes did show a definite flair, as Greene acknow
ledged. Poor though it is as an authentic recreation of the past, The
Crusaders compares quite well with a later film, King Richard and the
Crusaders.
King Richard and the Crusaders (1954), which starred George Sanders
(as Richard the Lionheart), Rex Harrison (as the Emir Ilderim) and
Virginia Mayo (as Lady Edith), was clearly but loosely based on Scott’s The
Talisman. Ilderim saves Richard’s life so that the Saracens will have some-
one worth fighting, Ilderim sings and the rest of the plot is profoundly
silly. Virginia Mayo gets to mouth the immortal line, ‘War! War! That’s
all you think about, Dick Plantagenet!’65 According to Time magazine,
the film ‘shows why the Crusades never really amounted to much’.66
Kingdom of Heaven, directed by Ridley Scott, was released in 2005.
Though visually splendid, in other respects it was not an improvement on
its predecessors. A severely critical review in the Times Literary Supplement
had this (among other things) to say:
The villains that Balian [the hero] is up against are very villainous indeed.
One wonders what possible fun they get out of being so very evil and so
brainless. Kingdom of Heaven seems to be telling us that medieval people
64
The Graham Greene Film Reader: Mornings in the Dark, ed. David Parkinson
(Manchester, 1993), p. 22.
65
Fraser, Hollywood History, p. 57.
66
Time, quoted in Halliwell’s Film & Video Guide 2000 (London, 1999), p. 456.
370
were just like us, only much stupider. One person who would certainly
have enjoyed this film, if only he were alive, is William Archbishop of
Tyre. William died in 1184, but he would have been delighted to see
the polemical and malicious portraits of Guy de Lusignan, Reginald of
Chatillon and the Patriarch Heraclius that he presented in his History of
Deeds Done Beyond the Sea being given renewed currency in a film in the
twenty-first century.67
Evidently readers of this survey cannot have learned much about the
crusades from Chahine’s, DeMille’s or Scott’s interpretations. It is sad
that the crusading legacy should have come to such stuff, but, as Graham
Greene observed, with reference to DeMille’s The Crusaders, ‘[F]or the
rest of the time, we must be content with a little quiet fun at the expense
of Clio’.68
67
Robert Irwin, ‘It’s God Guignol’, Times Literary Supplement, 27 May 2005,
p. 17.
68
‘Cinema’, The Spectator, 29 August 1935, p. 14.
371
MAPS
Map 1: The march to Constantinople
Map 2: From Constantinople to Antioch
373
Maps
Map 3: The siege of Antioch, winter 1097–98
374
Maps
Map 4: From Antioch to Jerusalem
375
Maps
Map 5: The capture of Jerusalem, 15 July 1099
376
Index
‘Abbāsids 220, 224, 225 al-Andalus 56, 65, 220

absolution 16, 20, 26, 105 Andrew, St 178
Achard of Montmerle 30, 45 Anselm, archbishop of Canterbury
Acre 80, 81, 113, 114, 220, 222, 223, 91, 163
229, 232, 233, 366, 367 Anselm IV, archbishop of Milan 95,
Adhemar, bishop of Le Puy 74n, 283-84, 285, 287
140, 141, 147, 155-58, 161, 166, Anselm II, bishop of Lucca 23
167, 168 Anselm of Ribemont 14n, 260
al-Afdal, Fātimid vizier 47n, 84n, Antichrist 20, 35, 36, 37, 40, 41 42
177, 213, 220-22, 224-29, 230, Antioch 263, 359
231, 232 before the crusade 77-78, 110, 219,
Albania 85 220, 362
Albara 140, 142, 143 Christian siege and capture 42,
Albert of Aachen, chronicler 114, 46-49, 55, 79, 113, 178, 180, 184,
119-20, 134, 135, 146, 175, 189, 201, 221, 276, 278-79, 281, 288,
197, 199, 209-10, 212-15 294, 300, 312, 365
Aledo 56, 65 Turkish siege 139, 143, 213, 281,
Aleppo 220, 221, 223, 230, 263 282-83, 285, 296, 319-323
Alexandretta 312 battle (1098) 48, 49, 74n, 157, 178,
Alexandria 77, 78, 212, 366 221-22, 292, 295, 301, 303, 311n,
patriarchate 22 312
Alexios I Komnenos, Byzantine from capture to march south
emperor 160, 297-99, 348, 350-51 135-47, 149, 151-171, 185, 313-14
before the crusade 37, 85, 86, 111n after 1099 80, 114, 222, 281
oath of crusaders 136-37, 168, 170 patriarchate 22, 154, 160
relations with crusaders 138, 141, Apocalypse 36, 40, 331
145-48, 154, 155, 161, 165, 185 Apulia 37, 90, 93, 266
at Philomelium 142, 313-14 Arabs 35, 77, 219, 225, 365-67
Alfonso VI, king of Léon and Castile Armenia 43, 112, 141, 154, 161, 220,
56, 58-59, 74n 221, 256, 257, 326
Alfonso VII, king of Léon and Arnulf of Chocques, patriarch of
Castile 332 Jerusalem 203, 205, 208-9
Almeria 277, 280 Arqa 133, 136, 144-49
Almohads 233 Arsuf 79, 223, 233
Almoravids 56, 156 Artūqids 208, 220, 221
Alp Arslan, Saljūq sultan 220 Ascalon 115, 212, 214, 222, 227, 228,
Amalekites 331 229, 231, 232, 233, 366
Amalfi 76-77, 78, 83-85, 107 battle (1099) 44, 50, 112, 178-79,
Amalric, king of Jerusalem 267 180, 182, 198, 201, 203, 205, 208,
Ambrose, St 283, 285 213, 214, 225-26, 232, 276, 301
Amedelis 301 Augustine of Hippo, St 15, 23,
Anatolia 152, 221, 222, 298 253, 345
377
Index
Baghdad 220, 223, 302n Caesarea 79, 80, 110, 221, 223, 233
baptism 36, 46, 48, 117, 122-25, 130, Caffaro, chronicler 277-81
248, 325 Cairo 47, 76, 83, 84, 220-22, 224
see also conversion Geniza 77, 78, 84n, 121, 126, 128,
Baldwin I, count of Edessa and king 129-31, 225, 361
of Jerusalem 42n, 114, 135, 140, Calabria 37
171, 194, 197, 200, 201, 228, 231, Calixtus II, pope 98
232, 308n cannibalism 144, 308, 312, 315-22
Baldwin II, count of Edesssa and captives 45, 46, 47, 48n, 86, 110, 203,
king of Jerusalem 264n 205, 206, 213, 214, 228, 244, 257,
Baldwin III, king of Jerusalem 197, 263, 292, 300, 319, 324, 325, 331,
267, 268 337, 339, 340, 341, 344, 348
Baldwin IV, king of Jerusalem 267 castles 56, 62, 63-69, 72
Baldwin, archbishop of Canterbury Catalonia 21, 68n
107, 251 Celestine III, pope 309
Baldwin, count of Hainaut 137 Cencius, Roman city prefect 13-14
Bari 104 Chahine, Yusuf, film-maker 366-67,
council (1098) 90, 91-92, 95, 97, 98, 371
99, 100, 162-66 chansons de geste 44, 258, 291, 294-95,
Barkyārūq, Saljūq sultan 223 297, 302, 304, 307-28
Beirut 149, 220, 223, 226, 233 Aliscans 295-96, 299, 330
Benevento council (1091) 90 Chanson d’Antioche 258-59,
council (1113) 169 291-305, 307-8
Bernard of Clairvaux, St 17, 48, 247-49 Chanson d’Aspremont 295-96, 339
Bernard of Saint-Valery 207 Chanson des Chétifs 292-93,
Bertrand, count of Tripoli 114 324-25
Bethlehem 203, 204 Chanson de Guillaume 295-96
Bohemund I, prince of Antioch 80, Chanson de Jérusalem 292, 311-12,
133, 147, 161, 185, 207, 264n, 266, 316-18, 320-21
319, 353, 364-65 Chanson de Roland 45, 295-96,
role in capture of Antioch 45-46, 302, 307, 332-33, 335
136-37, 146, 313-14 Charlemagne, Holy Roman emperor
dispute over rule in Antioch 36-39, 264, 297, 302, 307
138-39, 140, 143-44, 162 charter evidence 26-32, 103-4, 108n,
letter dated Sept. 1098 42, 44, 46, 114, 138, 187, 256n, 271n, 284
92, 141-42, 154-56 children 255-72
crusade of 1107 115, 154, 165 Cilicia 140, 142, 154, 219
booty, see plunder Cistercians 248, 271
Bulgaria 104, 113, 186 Civetot 323, 324, 325
Burchard, bishop of Worms 16 Civitate, battle (1053) 52
Byzantium 71, 331, 349, 356 Clement III, pope 90
before the crusade 52, 75-76, 85, Clermont, council (1095) 17, 19, 21,
111n, 168, 219, 220, 221 32, 41-42, 89, 90, 93, 95, 97, 99,
and the First Crusade 136, 141, 142, 106, 108, 167-170, 171, 191, 207,
146, 161-62, 170, 297-300, 312-13 236, 240-41, 252
economic activity 75-76, 78, 86-87 Cologne 117, 119
378
Index
Constantinople 37, 76, 78, 96, 104, Embriaci, Genoese family 276, 279,
137, 147, 154, 220, 285, 348, 363 280
patriarchate 22, 160 see also William Embriaco
conversion 36, 38, 40, 43-50, 117, 119, Emich, count of Flonheim 40, 119,
120, 122-25, 157, 242, 263, 330 120, 122, 124
Corbaran 266, 296, 301-2, 303, 323, emotions 173-89, 235-54
324, 325 anger 73, 143, 183-87, 188, 237-38,
see also Karbugha 239, 241, 243, 245-46, 249-51,
Corfu 85 253, 369
Crusades confidence 15, 158, 162, 178-79, 18,8
Second 106, 117, 236, 247-49 249, 268
Third 250-52, 309 fear 73-74, 120, 177-79, 188, 237,
Children’s 260 242, 244, 246, 330, 334, 339
Cyprus 110, 112 hatred 16, 137n, 237, 241, 243, 246,
Cyriacus, archbishop of Carthage 252-53, 344, 353
12-13, 14 joy 178, 182-83, 237, 244, 245, 246,
358
Daibert, archbishop of Pisa 22, 80, 166 love 13, 16, 21, 128, 173, 174, 179-80,
Dalmatia 104 235, 237, 239, 244
Damascus 135-36, 220, 221, 223, 227, weeping 179-83, 188
228, 229, 230, 231, 232, 263 see also vengeance
Danes 115 Engelbert of Tournai 202, 210
Danishmendids 222 English 56, 110, 115, 144, 163, 348, 350
DeMille, Cecil B., film-maker 368-70, Eraclius, patriarch of Jerusalem 249,
371 371
Dorylaeum, battle (1097) 156, 157, Erlembald, knight 13-14
182, 304 Ermengol IV, count of Urgel 21
Duggan, Alfred, novelist 353-57, 365 eschatology 34, 43, 50, 122, 123, 127n,
Duqāq 220, 229-30 359-60
Durazzo 85 Estatins (Tatikios) 297-300, 305
Eugenius III, pope 247, 264
Eastern Christians 12, 18, 43, 141, Eurvin de Créel 323
154-55, 160-61, 235, 240 Eustace III, count of Boulogne 154,
Edessa 135, 140, 142, 200, 201, 222 156, 202, 204, 210
conquest (1144) 236, 247, 248, 308n
Egypt 177, 204, 366-67 family influences 27, 255-56, 261-65,
economic activity 75-78, 81, 83, 87, 268, 271, 272, 284, 286-90
279 Fātimids 83
before the crusade 75-78, 81, 83, 87, before the crusade 76, 110, 219-22
220, 223, 225 negotiations with the crusaders
embassy (1097) 47, 137-38, 148-49 47-48, 137-38, 148-49
occupy Jerusalem (1098) 208, 213, occupation of Jerusalem (1098)
221 78, 146, 198, 203, 205, 206, 208,
after the crusade 115, 226-31, 232, 234 211-14
see also Fātimids response to the crusade 222-34
379
Index
Fernando I, king of Léon and Castile Gregory I, pope 23

58 Gregory VII, pope 11-20, 22, 23, 32n,
finance 31-32, 65, 67, 104, 107, 110, 33, 89, 96, 109, 219
122n, 144-45, 259, 270 Gregory VIII, pope 250-51
Firuz, betrayer of Antioch 4-6, 136, Guibert, archbishop of Ravenna 18
157 Guibert of Nogent, chronicler 41-42,
Flemings 115, 147, 198n, 363 198, 206-8, 319
Florence 288-90
France 18, 19, 48, 60, 89, 90, 111, 197, hagiography 53, 128-29, 280
268, 269, 295, 310-11, 358, 360 Haifa 81
Frederick I Barbarossa, Holy Roman Hastings, battle (1066) 16, 52, 56
emperor 280 Hattin, battle (1187) 236, 249-51, 328,
Fulcher of Chartres, chronicler 135, 366
153-54, 194, 200, 201, 253, 308n Hebrew sources 117-18
see also Cairo Geniza
Gaston IV of Béarn 139, 206, 212, Henry IV, Holy Roman emperor 18,
334n, 341 19, 37
Genoa 76, 77, 80, 84, 85, 87, 276-82 Henry ‘the Lion’, duke of Bavaria
Genoese 79, 80, 81, 82, 84, 86, 94, and Saxony 113, 116
114, 138, 276-82 Hoban, Russell, novelist 361-65
Germans 11, 113, 186, 197, 198, 261-62, Holy Fire 110, 281, 289-90
362 Holy Lance 42, 48, 143, 147-48, 157,
Germany 15, 18, 19, 37, 62, 130, 197, 178, 281, 354, 359
239, 269 holy places 25, 26, 27, 29, 32, 34, 35,
Gesta Francorum et aliorum 36, 37, 42, 43, 50, 115, 116, 188,
Hierosolimitanorum 134, 192, 251, 284, 353
199-201, 207, 308n holy war 11, 15, 28, 29, 3,0 33, 34, 48,
Ghaznāvids 220 50, 84, 125, 220, 222-32, 234, 239,
Gilbert of Traves 45 244, 255, 257, 268
Girard, bishop of Angoulême 333 see also just war
Godfrey of Bouillon, duke of Hospitallers 79n, 250
Lower Lotharingia and ruler of Huesca 21, 156
Jerusalem 296, 311n, 315, 316 Hugh, count of Vermandois 137, 140,
ancestry 264, 297 154, 304, 314
on crusade 45, 133-34, 140-46, 148, Hungarians 11, 63
154, 156, 353 Hungary 104, 123, 186
role in capture of Jerusalem 179,
181, 202-3, 204, 208, 210, 289, Iacopo da Varagine 280-82
election as ruler 202-3, 205, 208, ideology 39n, 42, 117, 121-22, 125, 126,
317, 321 128, 136, 149, 152, 167, 171, 173n,
visions relating to 37-39, 209-10 220, 242n, 252nn, 263, 269, 271,
posthumous reputation 301n, 303, 294, 320-21, 336
305, 307, 322, 350-51 Ifrīqiya 228, 232-34
Granada 58-60 Iftikhār al-Dawlah, governor of
Greece 104 Jerusalem 213
380
Index
indulgence 97, 99, 169, 170, 256, 285 John IV Oxeites, patriarch of
Innocent II, pope 98 Antioch 154
Innocent III, pope 224, 270, 309 John Tzimiskes, Byzantine emperor
Italy 33, 52, 56, 75-87, 89, 90, 91, 104, 221
163, 250, 275-90, 301n, 350, 370 Joscelin I, count of Edessa 257
Ivo of Chartres, St 17, 93n, 112-113, 311n Jubail 80, 114
just war 15, 23, 24, 238, 239
Jabala 133-35, 145-46
Jaffa 80, 81, 112, 114, 202, 204, 227, Karbugha, atabeg of Mosul 48-49,
228, 229, 232 133, 136, 137, 138, 139, 141, 143,
James of Voragine, see Iacopo da 154, 157, 164, 178, 201, 221, 222,
Varagine 266, 296, 301, 303, 305, 322, 324,
Jerusalem 92, 141 et passim 325
as goal of the crusade 21, 22, 25-50, see also Corbaran
94, 98, 134, 142, 146-48, 169-70, Khorasan 263
307, 351 knights 13-14, 16-17, 21, 22, 28-31, 33,
as pilgrimage goal 21, 22, 25-27, 31, 37, 44-45, 50, 57, 75, 107, 109,
34, 104-5, 110-11, 112, 114-16 112, 113-14, 117, 121, 144, 152, 226,
eschatological significance 34-43, 238, 239, 253, 259, 265, 268, 271,
159, 358-59 295, 311, 315-16, 329, 339, 348, 353,
Fātimid occupation (1098) 138, 355, 358-59
146, 220-21
crusader siege 78, 145, 149, 180-81, Lambert, bishop of Arras 96, 100
281-82, 288, 316-17, 325 Laodicea 80, 110, 14,0 232
capture (1099) 79, 95, 113, 136, 139, Lateran, council (1097) 90, 91, 97,
184, 187, 188, 191-215, 222, 227, 98, 105
243-46, 257, 261, 262, 278, 279- Leo IV, pope 11
80, 282, 283, 285, 288, 289, 353, Leo IX, pope 11-12
363 Letold of Tournai 202, 204, 208, 210
Fātimid reaction to capture 222-23 letter evidence 42-43, 46n, 49, 91-93,
capture by Saladin (1187) 309, 367 103, 111n, 141, 151-71, 212, 222,
church of the Holy Sepulchre 22, 228-29, 231, 237-38, 242, 251,
31, 36, 37, 38, 43, 50, 96, 110, 114, 361n
116, 139, 181, 182, 189, 203, 205, Liprand, priest 13-14, 284
206, 244, 245, 248, 265, 269, liturgy 104, 118, 244, 268-70
275, 279, 281, 285, 289 Lotharingians 147, 175, 197
cartulary 167, 168 Louis VI, king of France 311-12
mount of Olives 36-38, 202 Louis VII, king of France 106, 265
patriarchate 22 Lucca 91, 163-64
Jews 35, 36, 38, 40-41, 42, 43, 46, 117-
31, 203, 211, 212, 214, 235, 247, Ma’arrat al-Nu‘man 135, 140, 143-44,
360, 365 145, 146, 149, 180, 301, 316, 318,
jihād 44, 223, 224, 228 320, 322
John the Baptist, St 276, 281, 358 Mahdia 84, 233, 277, 278
John VIII, pope 11 Mainz 107, 117, 119
381
Index
al-Mansūr 225, 336 penance 16, 17, 26, 28, 29, 31, 32, 33,
Manzikert, battle (1071) 220 34, 43, 97, 98, 99, 103, 105, 108,
martyrdom 11-14, 16, 19, 20, 23, 33-34, 109, 111, 113, 250, 270, 311n
74, 110, 118-19, 123-24, 126-30, Peter Bartholomew 135, 143-44, 147, 178
147, 178-79, 188, 242, 256, 303, Peter, bishop of Albara 206
309, 321, 337, 369 Peter, bishop of Huesca 156
massacres 40, 42, 117-18, 120-21, 123, Peter Desiderius 148, 204
125-26, 129, 184, 187, 203, 205, 211- Peter the Hermit 39-41, 42, 48, 49,
15, 233, 244, 308, 358, 363, 366, 369 120, 178, 186, 315, 317, 322, 358,
Melfi, council (1089) 90, 96, 97 359, 369
Metz 117 Peter, St 15, 33, 84, 159-60, 178
Milan 13, 18, 282-85, 286-88 militia 12, 15, 17, 19
millenarianism 34-37, 125, 264, 359-60 Philip I, king of France 19
see also Apocalypse, eschatology Philomelium 141, 146
Military Orders 271, 275 Piacenza, council (1095) 90, 96, 99
see also Hospitallers, Templars Pierre Postel 312, 323
Minorca 277 pilgrimage 11, 18, 21, 25, 26, 27, 29, 30, 31,
Moors 21, 30, 58, 84, 330, 341, 345 32, 33, 34, 36-37, 39, 43, 50, 75, 84,
Mosul 223 86, 103-16, 120, 148, 151-52, 166,
181, 186, 198, 209, 240, 244, 256,
Nablus 205 262, 263, 268, 269, 270, 283, 284-
narrative sources 175-76, 188-89, 85, 358, 359, 360, 362, 364, 366
191-215 Pisa 22, 76, 77, 80, 81, 82, 84-87, 9,5
Nicaea 45, 48, 137, 153, 156, 157, 179, 114, 276, 288
256, 300, 301, 312n plunder 15, 16, 17, 22, 32, 55n, 120, 122,
Nîmes, council (1096) 90, 93, 94-95, 97 203, 205, 212, 232, 241, 308
Niš 120, 186 Prague 117
non-combatants 146, 152, 255-72 preaching 13, 22, 37, 40, 41, 43, 44,
Normans 11, 15, 16-17, 52-53, 56, 61n, 93-95, 104, 106-7, 111, 122, 125,
67, 75, 83, 85, 110, 112, 136, 147, 158, 159, 167, 173, 236, 237, 244,
161, 185, 197, 198n, 208-9, 219, 247-49, 252, 257, 264, 266, 269-
233, 234, 266, 352 71, 276, 309, 322, 325, 332, 344
Northmen 11, 63 prisoners see captives
Norway 115, 354 Provençals 103, 135, 140, 143, 146,
numbers 145-46 147, 185
Nūr al-Dīn 221, 223, 228, 230-31
Qilij Arslān 221
Oldenbourg, Zoé, novelist 357-61, 365
Raimbold Croton 112-13
Palestine 35, 37, 41, 81, 87, 110, 111, 113, as Raimbaut Creton in the
116, 146, 170, 198, 211, 213, 220, Chanson d’Antioche 311, 317
222, 226-27, 232, 233, 234, 276, Rainalt Porcet 311, 325
350, 353, 369 Ralph of Caen, chronicler 148n, 195,
Paschal II, pope 91, 92, 100, 112, 165, 208-9, 308n
166n, 169, 244n Ramla 44, 110, 148, 203, 210, 220,
Patarenes 13, 14, 18 225, 226, 227, 228, 230, 231, 232
382
Index
Raymond of Aguilers, chronicler 135, council (1099) 90, 92, 93, 95-96, 98
193 St Symeon, port 79
Raymond Pilet 47, 140, 202 Saljūq Turks 12, 18, 20n, 23, 78, 82n,
Raymond of Saint-Gilles, count of 86, 110, 137, 138, 148, 149, 208,
Toulouse 220-23, 224, 227, 229, 230
before the crusade 94, see also Turks
on crusade 47, 133-34, 135, 155, 178, Saracens 11, 20, 21, 30, 42, 44, 45,
180, 185 46, 48, 63, 85, 206, 212, 213, 238,
dispute with Bohemund 138-48, 162, 244n, 262-63, 296, 315-16, 317-
role in capture of Jerusalem 203-5, 317 22, 324, 329-45, 369, 370
after the crusade 114, 264n Saragossa 333, 341
redemption 127n, 246, 271, 307-28, Sardinia 84
354, 357 Satan 241-42, 330, 344
Reform movement 11-24, 89-90, 96, Scott, Ridley, film-maker 370-71
99n, 151, 157, 159-60, 163, 167, Scott, Walter, novelist, 348-53, 365,
171, 174n 366, 370
Regensburg 117 Seville 59
remission of sins 16, 17, 21, 26, 29, 30, Sicily 15, 21, 33, 77, 219, 233, 234
31, 32, 94, 112, 285, 308 Sidon 115, 223, 226, 233
Rhodes 81, 86 Silvester, pope 22
Richard I the Lionheart, king of Simeon, bishop of Ova-Burgos 13, 15
England 27n, 319, 336n, 366-67, Soissons, battle (923) 54-55
369-70 Spain 21, 30, 33, 52, 56, 58-60, 63, 68, 98,
Richard the Pilgrim 38, 293 124, 168-69, 219, 225, 238, 269, 277
Ridwān of Aleppo 141, 220 Speyer 117
Robert I, count of Flanders 37, 38n, spies 45, 204, 319
111, 242 Stabelo, chamberlain 205, 209-10
Robert II, count of Flanders 29, Stephen, count of Blois 137, 139, 154,
133-34, 140, 143, 144, 145, 146, 240, 312-14, 324, 325
154, 156, 311n, 316 Stephen, priest 140
Robert Guiscard 83n, 85 Sūqman, emir 208
Robert Curthose, duke of Normandy Sibylla, queen of Jerusalem 267
140, 143, 144, 145, 154, 156, 296, symbolism 64n, 105-9, 124, 128, 149,
297, 300-5, 311n, 353 181, 183, 188, 253, 269, 286, 288,
Roger, prince of Antioch 223 290, 329-45
Roger I, count of Sicily 15 Symeon, Greek patriarch of
Roger II, king of Sicily 233 Jerusalem 141
Rothard, archbishop of Mainz 119 Syria 75, 78, 123, 136, 142, 143, 149,
Rouge Lion 301, 302 208, 219-21, 223, 226, 227, 230,
Rudolf of Swabia 19 232, 233, 234
Rugia 140, 144
Tafurs 144n, 312, 314-23, 324, 325,
Sagrajas, battle (1086) 56 326-28, 359
Saladin 116, 221, 222, 223, 231, 250, Tancred of Hauteville 139, 140, 144,
251, 347, 365, 366-67, 369 145, 147, 148, 175, 185, 187, 197,
St Peter’s, Rome 84, 96 203-5, 208-9, 212-13, 268, 353
383
Index
Tarragona 21, 94, 156, 238 letter from the princes 42- 44, 46,
Tasso, Torquato 350-51, 367-68 141, 151-71
Tatikios 297 Urban III, pope 249, 250
Templars 17, 111, 271n
Thomas of Marle 311, 317 Valencia 58
Toledo 60, 219 Valona 85
Tortosa (Spain) 277, 278, 280 Vegetius 51n, 69-70
Tortosa (Syria) 80, 112, 170 vengeance 35, 38, 42, 121-22, 123,
Tours 107 173-74, 183, 185, 187, 189, 212,
council (1096) 90, 93-94, 95, 97, 107 215, 236-41, 243, 248, 252-53,
Trier 117 257, 295, 311n, 326n, 365
Tripoli (Lebanon) 47, 110, 133, 146, 148, Venice 76, 78, 80-81, 82, 85-87, 114,
149, 170, 222, 223, 226, 232, 233 232, 276
Tripoli (Libya) 233 Vespasian, Roman emperor 38
Troia, council (1093) 90 Victor III, pope 11, 84
Tughtakīn 230, 231 violence 17, 18, 23, 24, 64n, 112, 119,
Tunisia 76, 77, 233 121, 124, 125, 186-87, 188, 212, 235,
Turbessel 257 236, 243, 252, 299, 311, 363, 364
Turks 12, 18, 21, 45, 46, 47, 48, 110, Visconti, Milanese family 283, 286-88
137, 156-57, 168, 184-85, 203, 304, visions 37-38, 48-49, 135, 143-44,
205, 213-14, 220, 223, 232, 240, 147-48, 178, 204, 205, 209-10,
262, 263, 283, 286, 300, 304, 305, 282-83, 354, 359, 362
316, 319, 363 Volkmar 40
see also Saljūq Turks
Tutush 220 Wezilo, archbishop of Mainz 22
Tyre 167, 169, 220, 223, 226, 231-33 William I, king of England 58, 67n,
300
Urban II, pope 89-100, 219-20, 259, 360 William II, king of England 58
and Christian history 20-22 William fitzMarquis 268, 318n
and Reform movement 18 William Embriaco 206
concept of crusade 32-34, 37, 103, William of Orange 296, 299, 302, 307
237-42, 308 Worms 117
at Clermont 17, 19, 41-42, 106, 140,
200, 240-42, 264 Xanten 117, 128
preaching the crusade 111-12, 270-71,
276, 281, 331 Yaghisiyan, governor of Antioch 220
384

Jerusalem The Golden

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Jerusalem The Golden

Uploaded by

Copyright:

Available Formats

Jerusalem the Golden

© 2014 Brepols Publishers n.v., Turnhout, Belgium (English translation)

ISBN 978-2-503-55172-2 (print)

I. The Origins and Background of the First Crusade

II. The Course of the Crusade

III. The Impact of the Crusade

IV. The Afterlife of the Crusade

Fig. 1: The Cathedral of Angoulême, twelfth century (photo Bartal)  333

Table 1: Length of Accounts of the Siege and Capture of Jerusalem

Map 1: The march to Constantinople 373

MGH Monumenta Germaniae Historica

SS Scriptores (in folio)

PL Patrologia Cursus Completus. Series Latina, ed. J.P. Migne,

RHC  Recueil des historiens des croisades, ed. Académie des

Arm. Documents arméniens, 2 vols (1869–1906)

RS Rolls Series: Rerum Britannicarum Medii Aevi Scriptores

The nine-hundredth anniversary of the series of events which were much

richer by very valuable additions. Finally, the conference on historiogra-

Leaving aside this gesture towards modern geo-political concerns,

appearance of so many key publications of First Crusade texts within a

A current area of interest is the memorialisation of the crusades, and

The Origins and Background

The subject of this chapter is the epoch-making expansion in the notion

proclaimed the fallen to have died in defence of righteousness and to have

thought was, indeed, well established by Gregory’s time. In the context

Normans.28 Gregory returned to the subject in March 1080, when he not

individual. No less momentous in the adding of new dimensions to reform

for authenticity, and systematically discussed.60 And Gratian did not in

1. Reconquérir la Jérusalem terrestre

Riley-Smith (The First Crusaders, 1997), tout en marquant un salutaire

Riley-Smith, ‘L’idée de croisade dans les chartes de la première croisade’, dans

nécessairement, un accent mis sur la conscience du péché, sur le besoin de

de la foi. En d’autres termes, l’absence dans la plupart des chartes d’une

monastique stéréotypé, calqué sur celui que l’on employait auparavant

Toutes les motivations jusqu’ici mentionnées ont pu exister, parfois

2. Gagner la Jérusalem Céleste

and Settlement, pp. 47–56, qui en accepte éventuellement la possibilité, et

années sous l’influence de Bernard McGinn.18 Sans adopter totalement ici

par le Christ mettant fin à l’histoire de ce monde.21 La Sybille Tiburtine

temps, suivie de la manifestation de la gloire divine au Saint Sépulcre,

accordée à un chevalier digne de foi, avait désigné Godefroy comme

écrits antérieurs aux informations qu’il avait peutêtre reçues concernant la

mais le contraire serait surprenant. Pierre s’est attardé à prêcher à Pâques,

des Juifs, celui des troupes de Pierre l’ermite et la non-réalisation de la

3. Bâtir la Jérusalem spirituelle

évidemment pas d’une conversion au sens actuel du terme, mais au sens ou

Chanson de Roland, par exemple, illustre bien la mentalité des chevaliers

de négociateur entre Bohémond et ‘Firouz’ dans les tractations qui allaient

tradition indépendante, signale aussi l’envoi, de Nicée, d’une ambassade

des chrétiens, sans toutefois en indiquer les modalités. La moins irréaliste

être un vendredi, jour du triomphe du Christ, que nous choisissons pour

This essay is not intended as a bilans et perspectives approach of the kind

to Rogers and Morillo, and with reference to R. C. Smail, Crusading Warfare

well as other important engagements. They were adventurers and mer-

World, ed. Christopher Harper-Bill and Elisabeth van Houts (Woodbridge,

it is not an important tool to study the mentality of warriors.9 Orally

in 923 is clear evidence of his opinion on this point.13 Another renowned

systematic policy of exhaustion of their resources through parias (tribute

In connection with conquest and ‘colonisation’ we must ask whether,

A Military History of Ireland, ed. T. Bartlett and K. Jeffery (Cambridge, 1996),

solutions of their own and employ them in their internal conflicts or

1066.47 The same conclusion may be applied to Christian polities in the

I. The Origins and Background of the First Crusade

II. The Course of the Crusade

III. The Impact of the Crusade

IV. The Afterlife of the Crusade

Fig. 1: The Cathedral of Angoulême, twelfth century (photo Bartal) 333

Table 1: Length of Accounts of the Siege and Capture of Jerusalem

Map 1: The march to Constantinople 373

PL Patrologia Cursus Completus. Series Latina, ed. J.P. Migne,

RHC Recueil des historiens des croisades, ed. Académie des

RS Rolls Series: Rerum Britannicarum Medii Aevi Scriptores

Without comprehensive accounts of Urban II’s later councils h istorians

Gesta Francorum (or versions of it) as their foundation text.6 A u seful