100% found this document useful (1 vote)
2K views435 pages

Kelly DeVries - Infantry Warfare in The Early Fourteenth Century - Discipline, Tactics, and Technology (Warfare in History) (2000, Boydell Press)

Uploaded by

fig8fm
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (1 vote)
2K views435 pages

Kelly DeVries - Infantry Warfare in The Early Fourteenth Century - Discipline, Tactics, and Technology (Warfare in History) (2000, Boydell Press)

Uploaded by

fig8fm
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 435

Infantry Warfare in the Early Fourteenth Century :

title: Discipline, Tactics, and Technology Warfare in


History, 1358-779X
author: DeVries, Kelly.
publisher: Boydell & Brewer Ltd.
isbn10 | asin: 0851155715
print isbn13: 9780851155715
ebook isbn13: 9780585202143
language: English
Infantry--History, Military art and science--History-
-Medieval, 500-1500, Infantry drill and tactics--
History, Military history, Medieval, Battles--
subject
England--History, Battles--Europe--History, Great
Britain--History--14th century, Europe--History-
-476-1492
publication date: 1996
lcc: UD59.D48 1996eb
ddc: 356/.1/09409023
Infantry--History, Military art and science--History-
-Medieval, 500-1500, Infantry drill and tactics--
History, Military history, Medieval, Battles--
subject:
England--History, Battles--Europe--History, Great
Britain--History--14th century, Europe--History-
-476-1492
Page i
Warfare in History

Infantry Warfare in the Early Fourteenth Century


Discipline, Tactics, and Technology
This study departs from the conventional view of the dominance of cavalry in
medieval warfare: its objective is to establish the often decisive importance of infantry.
In pursuit of evidence, Kelly DeVries examines the role of the infantry, and the nature
of infantry tactics, in nineteen battles fought in England and Europe between 1302 and
1347. In most of these battles it was the infantry which secured victory. Evidence from
first-hand accounts of the battlesa major feature of this studyis employed to argue that
victory came not because of superior technology, even when the longbow was used,
but due to a solid and disciplined infantry line making a defensive stand able to
withstand the attacks of opposing soldiers, whether cavalry or infantry.
KELLY DEVRIES teaches medieval history and the history of technology at Loyola
College in Maryland.
Page ii
Warfare in History
General Editor: Matthew Bennett
ISSN 1358-779X
The Battle of Hastings: Sources and Interpretations
edited and translated by Stephen Morillo
Infantry Warfare in the Early Fourteenth Century:
Discipline, Tactics, and Technology
Kelly DeVries
The Art of Warfare in Western Europe during the Middle Ages from the Eighth
Century to 1340
J.F. Verbruggen
Knights and Peasants:
The Hundred Years War in the French Countryside
Nicholas Wright
Page iii

Infantry Warfare in the Early Fourteenth Century


Discipline, Tactics, and Technology
Kelly DeVries

THE BOYDELL PRESS


Page iv
© Kelly DeVries 1996
All Rights Reserved. Except as permitted under current legislation no part of this work
may be photocopied, stored in a retrieval system, published, performed in public,
adapted, broadcast, transmitted, recorded or reproduced in any form or by any means,
without the prior permission of the copyright owner
First published 1996
The Boydell Press, Woodbridge
Reprinted in paperback 1998
ISBN 0 85115 567 7 hardback
ISBN 0 85115 571 5 paperback
The Boydell Press is an imprint of Boydell & Brewer Ltd PO Box 9, Woodbridge,
Suffolk IP12 3DF, UK and of Boydell & Brewer Inc. PO Box 41026, Rochester, NY
14604-4126, USA
A catalogue record for this book is available from the British Library
Library of Congress Catalog Card Number: 9612439
This publication is printed on acid-free paper
Printed in Great Britain by St Edmundsbury Press Ltd, Bury St Edmunds, Suffolk
Page v

Contents
Acknowledgements vi
Abbreviations vii
Introduction 1
I
9
The Battle of Courtrai, 1302
II
23
The Battle of Arques, 1303
III
32
The Battle of Mons-en-Pévèle, 1304
IV
49
The Battle of Loudon Hill, 1307
V
58
The Battle of Kephissos, 1311
VI
66
The Battle of Bannockburn, 1314
VII
86
The Battle of Boroughbridge, 1322
VIII
100
The Battle of Cassel, 1328
IX
112
The Battles of Dupplin Moor, 1332, and Halidon Hill, 1333
X
129
The Battle of Laupen, 1339
XI
137
The Battle of Morlaix, 1342
XII
The Battles of Staveren, 1345, and Vottem, 1346 145

XIII
155
The Battle of Crécy, 1346
XIV
176
The Battle of Neville's Cross, 1346
Appendix
Three Infantry Ambushes:
188
The Battles of Morgarten, 1315, Auberoche, 1345, and La
Roche-Derrien, 1347
Conclusions 191
Bibliography 199
Index 211
Page vi

Acknowledgements
Upon completing my dissertation, Perceptions of Victory and Defeat in the Southern
Low Countries during the Fourteenth Century: A Historiographical Comparison, at
the University of Toronto's Centre for Medieval Studies in 1987, it was suggested that
I should divide what I had written into two studies: one which discussed what was
happening on the battlefield and another which analyzed how medieval writers
rationalized military defeat. I will be forever grateful to my advisors, Professors Bert
S. Hall and Michael R. Powicke, for their constant stewardship and friendship in the
initial writing of my dissertation. Also owed thanks are Professors Joseph Goering,
John Munro, and John Gilchrist who read and commented on that dissertation.
This book is the first result of that suggestion, discussing what occurred on the
battlefields of Europe from 1302 to 1347. Three of the chapters, on the battles of
Courtrai, Mons-en-Pévèle, and Cassel, are drawn directly from the research which I
completed in that dissertation. The other chapters grew out of a desire to discover
whether what was occurring in battles fought in the southern Low Countries during
the early fourteenth century was also happening elsewhere throughout Europe. I
quickly realized that it was, and the book from that time on simply 'wrote itself'.
I have been assisted in this endeavor primarily by a Junior Faculty Sabbatical awarded
to me by Loyola College and funded by the Center for the Humanities there. Other
funding came from a Loyola College Summer Research Grant and a Wilfrid Laurier
University Research Initiation Grant. Assistance has also come from the Interlibrary
Loan Departments at Loyola College and Wilfrid Laurier University Libraries.
I again must thank my wife, Barbara Middleton, for reading and editing this
manuscript, and for many others who constantly urged me to complete it. I wish also
to express my love and thanks to my children. Finally, I wish to thank the Little
League Baseball teams which I have coached during the past three years. They have
taught me that every team has sufficient offense to win, but that it is defense that
ultimately decides victory.
Page vii

Abbreviations
AHDG Archives historiques du département de la Gironde
CCF Corpus chronicorum Flandriae
CL Chroniques Liégeoises
CRHB Commission royale d'histoire de Belgique
CS Camden Society
EETS Early English Text Society
EHS English Historical Society
FRG Fontes rerum Germanicarum
HS Historians of Scotland
MGH Monumenta Germaniae Historica
MGH, SS Monumenta Germaniae Historica, Scriptores
PG Patrologia Graeca
RHF Recueil des historiens de la Gaule et de la France
RS Rolls Series
SHF Société de l'histoire de France
SRI Scriptores rerum Italicarum
Werken uitgegeven door het Historisch Genootschap te
WHGU
Utrecht
WUVG Werken uitgegeven door de Vereenigen Gelre
Page 1

Introduction
This book is about medieval battles. Until only recently, up to the last forty years or
so, the study of medieval battles had been an important scholarly endeavor. Historians
writing large surveys of medieval military history in the late nineteenth and early
twentieth centuries, such as G. Kohler, Sir Charles Oman, Hans Delbrück, E. Daniels,
and W. Erben, all based their works on battle narrative, or what John Keegan would
later call the 'battle piece.'1 One need only take a cursory look at the table of contents
in these works to see that chapters on other aspects of medieval warfare, most notably
military technology, are almost completely ignored among those devoted to medieval
battles. Also omitted is nearly all discussion of siege warfare, campaign strategy,
logistics, military obligation and recruitment, booty, the laws of war, ethics and
religious aspects of military conflict, military intelligence, the psychology of the
medieval soldier, and the relationship between war, government, and society.
This tradition survived into the middle of this century. Ferdinand Lot's two-volume
L'art militaire et les armées au Moyen Age, en Europe et dans le Proche-Orient,
published in 1946, is nearly identical to Oman's or Delbrück's previous works in its
approach to the subject, except that Lot dwells far more on the numbers of soldiers
participating in medieval battles than do his predecessors.2 A decade later, Lieutenant-
Colonel Alfred H. Burne devoted two volumes to the study of the Hundred Years War
which, although also addressing the siege warfare and campaign strategies of the
conflict, used a 'battle piece' methodology to link it all together.3
Finally, perhaps the most important and certainly one of the most enduring studies of
medieval battles appeared in 1954, although to a large extent its popularity was not
realized until after a flawed English translation was published in 1977.4 J.F.
1 G. Kohler, Die Entwicklung des Kriegswesens und der Kriegfuhrung in der Ritterzeit von
Mitte des 11. Jahrhunderts bis zu den Hussitenkriegen, 3 vols. (Breslau, 1886); Sir Charles
Oman, A History of the Art of War. The Middle Ages from the Fourth to the Fourteenth
Century (London, 1898) and A History of the Art of War in the Middle Ages, 2 vols. (London,
1905); Hans Delbrück, History of the Art of War Within the Framework Political History,
vol. III: Medieval Warfare, trans. W.J. Renfroe, Jr. (Westport, 1984), originally Geschichte der
Kriegskunst im Rahmen des Politischen Geschichte, vol. III: Mittelalter, 2nd ed. (Berlin,
1923); E. Daniels, Geschichte des Kriegswesens, vol. II: Das mittelalterliche Kriegswesen,
2nd ed. (Berlin, 1927); and W. Erben, Kriegsgeschichte des Mittelalters (Berlin, 1929). See
also John Keegan, The Face of Battle (Harmondsworth, 1978).
2 Ferdinand Lot, L'art militaire et les armées au moyen âge en Europe et dans le proche orient,
2 vols. (Paris, 1946).
3 Alfred H. Burne, The Crecy War: A Military History of the Hundred Years War from 1337 to
the Peace of Bretigny, 1360 (London, 1955) and The Agincourt War: A Military History of the
Latter Part of the Hundred Years War from 1369 to 1453 (London, 1956).
4 J.F. Verbruggen, The Art of Warfare in Western Europe during the Middle Ages: From the
Eighth Century to 1340, trans. S. Willard and S.C.M. Southern (Amsterdam, 1977). This
publication lacked notes or bibliography, abridges the first section by eliminating many of
Verbruggen's examples, and
(footnote continued on next page)
Page 2
Verbruggen's De krijgkunst in west-europa in de middeleeuwen (IXe tot XIVe eeuw)
uses several battle narratives to explain his theses on medieval strategy and tactics.5 In
presenting these theses Verbruggen disagrees with most other medieval military
historians, including all mentioned abovealthough Lot, Verbruggen's mentor, receives
his softest criticisms; these other historians, Verbruggen complains, frequently misread
and thus misused original sources describing the battles on which they were writing.
In particular, they disregarded vernacular sources, preferring to use Latin ones. They
also often ignored battle eyewitnesses, despite the obvious superiority of such
sources. Finally, and most importantly for Verbruggen, those authors he criticizes
considered medieval warfare to be without an 'art': medieval soldiers did not fight
with a strategy or tactic in mind, and generalship, by strict definition, did not exist. To
Verbruggen, there definitely was an 'art' of medieval warfare. Both cavalry and
infantry warriors fought with a tactical skill for which modern scholars have generally
not given them credit. They fought as units of soldiers and not as individuals; these
units fought with order and cohesion.
Cavalry soldiers, knights, were trained to fight in this way, as skilled in battlefield
tactics as they were in the use of warhorses and weaponry. It was, as Verbruggen calls
it, a 'psychology of fighting', which certainly did not disappear once they entered
battle. Knights knew the tactical meaning of the standard, the importance of order in
formations, the protection of the flanks, the place of the suppliers on the battlefield,
and the regrouping of formations after contact. The 'psychology of fighting' also was
aided by a strict obedience to commands and the use of banners and trumpets to
increase communication. It resulted in difficult and learned battlefield maneuvers such
as the feigned retreat.
But cavalry alone rarely won battles. Only when infantry was used to support the
knights, and when archers were used to soften the enemy in preparation for cavalry
charges, were great victories had in medieval battles. Verbruggen describes in detail
seven battles in which the cavalry/infantry tactics show a planned order and an
understanding of good military methods: Antioch (February 9 and June 28, 1098),
Axpoel (June 21, 1128), Arsoef (September 7, 1191), the Steppes (October 13, 1213),
Bouvines (July 27, 1214), and Woeringen (June 5, 1288).
While cavalry alone rarely won battles, infantry alone sometimes did. In what he
describes as 'the miracle of the fourteenth century', Verbruggen cites the success of
peasant infantry forcesthe Swiss, Welsh, and Scotsand urban infantry forcesthe Italian
militias, the Flemings, and the Liégeoisas examples of infantry-based armies which
were highly skilled in battlefield tactics. These led to several 'victories', although
Verbruggen only recounts three: Courtrai (July 11, 1302), Arques (April 4, 1303), and
Mons-en-Pévèle (August 18, 1304). (Even though Verbruggen does not characterize
Mons-en-Pévèle as a victory, he does claim that the Flemish
(footnote continued from previous page)
excludes, inexplicably, the author's narrative on the battle of the Steppes. A new translation
restoring these omissions is currently being prepared.
5 J.F. Verbruggen, De krijgkunst in west-europa in de middeleeuwen (IXe tot XIVe eeuw)
(Brussels, 1954). See also J.F. Verbruggen, 'La tactique militaire des armées de chevaliers,' Revue
du nord 29 (1947), 161-80.
Page 3
infantry's adept knowledge of tactical maneuvers used there caused great slaughter
among the much larger French cavalry-based force.)
Although challenging some of his infantry warfare theses, and also his battle
narratives of Courtrai, Arques, and Mons-en-Pévèle, this study must consider
Verbruggen's work as its scholarly grandfather.
It has been more than forty years since Verbruggen's work originally appeared and
more than eighteen years since its English translation was published. Most historians
have continued to praise it,6 but few have followed its lead or tested its theses. The
most popular surveys on medieval military history written after Verbruggen, Philippe
Contamine's War in the Middle Ages and John Beeler's Warfare in Feudal Europe,
have very little to say about medieval tactics.7 The same holds true for works which
survey shorter chronological periods.8 Justifiably, medieval military historians have
looked elsewhere to increase the knowledge of military topics. Good, lengthy works
have recently been written on siege warfare, military organization and obligation, Just
War and the Peace of God, military orders, chivalry, and military technology.9
6 See, for example, Claude Gaier, 'Relire Verbruggen,' Moyen Age 85 (1979), 105-12 and
Bryce Lyon, 'The Role of Cavalry in Medieval Warfare: Horses, Horses All Around and Not a
One to Use,' Mededelingen van de Koninklijke Academie voor Wetenschappen, Letteren en
Schone Kunsten van België 49 (1987), 77-90. Another article which draws upon Verbruggen
for inspiration is Dennis E. Showalter, 'Caste, Skill, and Training: The Evolution of Cohesion
in European Armies from the Middle Ages to the Sixteenth Century,' Journal of Military
History 57 (1993), 407-30.
7 Philippe Contamine, War in the Middle Ages, trans. M. Jones (London, 1984) and John Beeler,
Warfare in Feudal Europe, 730-1200 (Ithaca, 1971).
8 See, for example, John Beeler, Warfare in England, 1066-1189 (Ithaca, 1966); Stephen Morillo,
Warfare under the Anglo-Norman Kings, 1066-1135 (Woodbridge, 1994); R.C. Smail, Crusading
Warfare, 1097-1193 (Cambridge, 1956); Christopher Marshall, Warfare in the Latin East, 1192-
1291 (Cambridge, 1992); Christopher Allmand, The Hundred Years War: England and France at
War, c.1300-c.1450 (Cambridge, 1988); and Anne Curry, The Hundred Years War (Houndmills,
1993).
9 For an excellent bibliography of medieval military topics (to 1983) see Philippe Contamine,
War in the Middle Ages, pp. 309-60. The following include some of the book titles which have
appeared since then. On siege warfare: Jim Bradbury, The Medieval Siege (Woodbridge, 1992);
R. Rogers, Latin Siege Warfare in the Twelfth Century (Oxford, 1992); and Ivy A. Corfis and
Michael Wolfe, eds., The Medieval City Under Siege (Woodbridge, 1995). On military obligation
and organization: Richard P. Abels, Lordship and Military Obligation in Anglo-Saxon England
(Berkeley, 1988); James F. Powers, A Society Organized for War: The Iberian Municipal
Militias in the Central Middle Ages, 1000-1284 (Berkeley, 1988); Mark C. Bartusis, The Late
Byzantine Army: Arms and Society, 1204-1453 (Philadelphia, 1992); and Andrew Ayton, Knights
and Warhorses: Military Service and the English Aristocracy under Edward III (Woodbridge,
1994); On Just War and Peace of God: Thomas Head and Richard Landes, eds., The Peace of
God: Social Violence and Religious Response in France around the Year 1000 (Ithaca, 1992).
On military orders: Alan Forey, The Military Orders: From the Twelfth to the Fourteenth
Century (Toronto, 1992); Helen Nicholson, Templars, Hospitallers and Teutonic Knights:
Images of the Military Orders, 1128-1291 (Leicester, 1995); Peter Partner, The Knights Templar
and their Myth (Oxford, 1981); and Malcolm Barber, The New Knighthood: A History of the
Order of the Temple (Cambridge, 1994). On chivalry: Maurice Keen, Chivalry (New Haven,
1984); Juliet R.V. Barker, The Tournament in England, 1100-1400 (Woodbridge, 1986); and
Howell Chickering and Thomas H. Seiler, eds., The Study of Chivalry: Resources and
Approaches (Kalamazoo, 1988). On military technology: Kelly DeVries, Medieval Military
Technology (Peterborough, 1992); John R. Kenyon, Medieval Fortifications (New York, 1990);
N.J.G. Pounds, The Medieval Castle in England and Wales: A Social and Political History
(Cambridge, 1990); M.W. Thompson, The Rise of the Castle (Cambridge, 1991); M.W.
Thompson, The Decline of the Castle (Cambridge, 1987); R.H.C.
(footnote continued on next page)
Page 4
A few good biographies on military leaders have also been published.10
However, battle narrative has all but disappeared, and despite Verbruggen's continued
popularity, there still endures a nearly universal belief that medieval tactics were non-
existent. Sir Charles Oman's characterization of medieval fighting skill, 'when mere
courage takes place of skill and experience, tactics and strategy alike disappear' and ' .
. . it was impossible to combine the movements of many small bodies when the troops
were neither disciplined nor accustomed to act together', still holds for most writing
about medieval warfare.11 For example, one of the most popular textbooks of military
history, Men in Arms, describes medieval tactics:
A feudal army in the field was an indescribably undisciplined force. Many tenants-in-chief
would take orders only from their immediate overlord, the king; therefore an effective chain of
command was impossible. There was a superabundance of courage, which tended to aggravate
rather than to relieve the normal disorder. Long centuries of control of the art of war by one
class, the exaggerated concentration upon cavalry warfare alone, and the absence of any
provision for group training except in a restricted fashion in the reformed tournament meant that
the study and practice of organized tactics had all but vanished . . . once the battle was joined,
all semblance of order disappeared, and the struggle became nothing more than a confused
melee of hundreds of individual encounters.12
Medieval military scholars are no different in their criticism of medieval tactics and
tactical leadership. As Philippe Contamine writes:
(footnote continued from previous page)
Davis, The Medieval Warhorse (London, 1989); and John H. Pryor, Geography, Technology,
and War: Studies in the Maritime History of the Mediterranean, 649-1571 (Cambridge,
1988).
10 On Fulk Nerra see: Bernard S. Bachrach, Fulk Nerra, the Neo-Roman Consul, 987-1040
(Berkeley, 1993). On William the Conqueror see: David C. Douglas, William the Conqueror: The
Norman Impact upon England (Berkeley, 1964) and Frank Barlow, William I and the Norman
Conquest (New York, 1965). On the Cid see: Richard Fletcher, The Quest for El Cid (Oxford,
1989). On Richard the Lionheart see: John Gillingham, Richard the Lionheart (London, 1978). On
Saladin see: Malcolm Cameron Lyons and D.E.P. Jackson, Saladin: The Politics of the Holy War
(Cambridge, 1982). On William Marshal see: Georges Duby, William Marshal: The Flower of
Chivalry, trans. R. Howard (New York, 1985) and David Crouch, William Marshal (London,
1994). On Edward I see: Michael Prestwich, Edward I (Berkeley, 1988). On Robert Bruce see
G.W.S. Barrow, Robert Bruce and the Community of the Realm of Scotland, 3rd ed. (Edinburgh,
1988) and Ronald McNair Scott, Robert the Bruce: King of Scots (New York, 1982). On Edward
the Black Prince see: Richard Barber, Edward, Prince of Wales and Aquitaine: A Biography of
the Black Prince (London, 1978). On Henry V see: Christopher Allmand, Henry V (Berkeley,
1992). On Charles VII see: M.G.A. Vale, Charles VII (Berkeley, 1974). On John Talbot see: Hugh
Talbot, The English Achilles: The Life and Campaigns of John Talbot, 1st Earl of Shrewsbury
(London, 1981) and A.J. Pollard, John Talbot and the War in France, 1427-1453 (London,
1983). And on Joan of Arc, among many others, see: Marina Warner, Joan of Arc: The Image of
Female Heroism (Harmondsworth, 1981) and Reginé Pernoud, Joan of Arc: By Herself and Her
Witnesses, trans. E. Hyams (New York, 1964).
11 Sir Charles Oman, The Art of War in the Middle Ages A.D. 378-1515, ed. J.H. Beeler (Ithaca,
1953), pp. 58, 60. This is Oman's undergraduate essay, published in 1885, and should not be
confused either with his 1898 version or his 1905 two volume revision.
12 Richard A. Preston, Alex Roland, and Sydney F. Wise, Men at Arms, 5th ed. (Fort Worth,
1991), pp. 69-70. See also John Keegan, Face of Battle, p. 336; Martin van Creveld, Command
in War (Cambridge, 1985); and Archer Jones, The Art of War in the Western World(Oxford,
1989), pp. 144-45.
Page 5
It should be agreed that medieval military history includes many battles which were nothing but
hasty, instinctive and confused confrontations in which captains played the role of simple
leaders of men, incorporated almost anonymously into the first line of battle, and where the
warriors' chief concern was to find an adversary worthy of rank or valour, without any
preoccupation for their companions in arms. They grappled on the battlefield with a sort of holy
fury, free to flee precipitately as soon as things seemed to be going against them, and the
individual search for booty and ransom was all-important.13
Even those who do give some plausibility to medieval tactical efficiency seem to
'damn with faint praise', as B.H. Liddell Hart illustrates:
In the West during the Middle Ages the spirit of feudal chivalry was inimical to military art,
though the drab stupidity of its military course is lightened by a few bright gleamsno fewer
perhaps, in proportion, than at any other period in history.14
This has led to an attempted determinism of battlefield victory that removes the
soldier or his general as the cause of victory. Greater numbers, the presence of the
mounted shock combat, and superior technology are more frequent determinants for
victory in medieval battles. One example of this is the enduring belief that knightly
cavalry continued to dominate the battlefield. These troops after all excelled in the
tactics of mounted shock combat. Couching lances under their arms and using the
force of their charging warhorses, they were able to deliver a blow with such an
impact that few could defend against it. Those few infantry victories accorded by
defenders of 'The Age of the Horse' are portrayed as temporary accidents.15 Such a
portrayal completely ignores the numerous infantry victories of the early fourteenth
century, nineteen of which are recounted here.
Another good example of this, and one that is particularly pertinent to this work, is the
'tenacious myth' of the longbow. Since at least the sixteenth century many military
historians have believed that the longbow significantly changed English strategy and
tactics in the later Middle Ages, so much so that England was to gain many victories
solely because of its use in warfare. In 1298, for example, Edward I took a troop of
over 10,000 archers with him on his conquest of Scotland (a ratio of three archers to
one mounted man-at-arms), an extremely large number in
13 Contamine, War in the Middle Ages, p. 229.
14 B.H. Liddell Hart, Strategy: The Indirect Approach (New York, 1954), p. 75. See also John
Beeler, 'Towards a Re-Evaluation of Medieval English Leadership,' Journal of British Studies 3
(1963), 1-10.
15 See Contamine, War in the Middle Ages, pp. 126-32; Bryce Lyon; Preston, Roland, and Wise,
pp. 62-66; Michael Howard, War in European History (Oxford, 1976), pp. 1-19; Claude Gaier,
'La cavalerie lourde en Europe occidentale du XIIe au XVIe siècle,' Revue internationale
d'histoire militaire 31 (1971), 385-96; and Robert Bartlett, 'Military Technology and Political
Power,' in The Making of Europe: Conquest, Colonization and Cultural Change, 950-1350
(Princeton, 1993), pp. 60-63. See also J.F. Verbruggen, 'La tactique de la chevalerie Française de
1340 à 1415,' Publications de l'université de l'état à Elisabethville I (1961), 39-48 and J.F.
Verbruggen, 'De rol van de ruiterij in de middeleeuwse oorlogvoering,' Revue Belge d'histoire
militaire 30 (1994), 389-418. Verbruggen insists in these articles that, although cavalry armies
remained dominant until the end of the Middle Ages, they did adapt their tactics to counter the new
tactics of infantry armies.
Page 6
comparison with the numbers of archers included in English armies previous to this
time. And he was victorious. Also victorious was the English army, again including a
large contingent of archers, which faced the Scots at Dupplin Moor in 1332 and at
Halidon Hill in 1333. Finally, English archers participated in the decisive victories over
the French army at the battles of Sluys (1340), Morlaix (1342), Crécy (1346), Poitiers
(1356) and Agincourt (1415).16 Crediting the longbow with all of these victories,
however, removes any recognition of the non-archery infantry soldiers which were
also present among the English forces. Furthermore, it also devalues English
generalship, determining that Edward III, the Black Prince, or Henry V did little more
than take advantage of this 'invincible' technology. Finally, it also fails to explain the
victories of non-English armies during the same period.

This book is about medieval battles. In particular, it is about nineteen battles which
were fought in Europe between 1302 and 1347. This is not to say that medieval battles
were the most important aspect of warfare during this period, or in fact during any
period of medieval history, nor that any of the battles discussed here could be
considered 'decisive' in the way that Sir Edward Creasy, Major General J.F.C. Fuller,
or Joseph Dahmus would define the term.17 Indeed, only in the battle of Kephissos,
fought between the Catalan Company and the duke of Athens in 1311, was one side so
defeated, the Athenians, that the other side immediately took over their previous land
holdings and began to govern their previous subjects. Yet, this government lasted only
for seventy years, hardly establishing the battle as 'decisive'.
In most other instances, the battles discussed here did not even decide the immediate
future of the two sides, necessitating that further battles be fought. Thus the Scots
fought with the English at Loudon Hill in 1307, at Bannockburn in 1314, at Dupplin
Moor in 1332, at Halidon Hill in 1333, and at Neville's Cross in 1346; the English
fought the French at Morlaix in 1342, at Auberoche in 1345, at Crécy in 1346, and at
Le Roche-Derrien in 1347; and the Flemings fought the French at Courtrai in 1302, at
Arques in 1303, at Mons-en-Pévèle in 1304, and at Cassel in 1328. (They also assisted
the English against the French at Sluys and Tournai in 1340 and at Calais in 1346-47,
conflicts not discussed in this work.) The threat of Flemish assistance to the English
may even have been the determining factor in rushing the French attack at Crécy in
1346.
16 T.H. McGuffie, 'The Long-bow as a Decisive Weapon,' History Today 5 (1955), 737-41;
Jim Bradbury, The Medieval Archer (New York, 1985), 71-138; Robert Hardy, Longbow: A
Social and Military History, 3rd ed. (London, 1992); Robert Hardy, 'The Longbow,' in Arms,
Armies and Fortifications in the Hundred Years War, ed. A. Curry and M. Hughes
(Woodbridge, 1994), pp. 161-82; Clifford J. Rogers, 'The Military Revolutions of the Hundred
Years War,' Journal of Military History 57 (1993), 249-51; and Gareth Rees, 'The Longbow's
Deadly Secrets,' New Scientist 138 (June 5, 1993), 24-25. Against the decisiveness of the
longbow, see John Keegan, Face of Battle, pp. 78-116 and Claude Gaier, 'L'invincibilité
anglaise et le grande arc après la guerre de cent ans: un mythe tenace,' Tijdschrift voor
gescheidenis 91 (1978), 378-85. See also my discussion in Medieval Military Technology, pp.
37-39.
17 Sir Edward Creasy, The Fifteen Decisive Battles of the World, 36th ed. (London, 1894); J.F.C.
Fuller, The Decisive Battles of the Western World and their Influence upon History, 2nd ed., 2
vols. (London, 1970); and Joseph Dahmus, Seven Decisive Battles of the Middle Ages (Chicago,
1983).
Page 7
Other battles may stand out as far as opponents are concerned, but they too cannot be
considered 'decisive' in that the conflicts had very limited results. This is especially
true in the battles which were fought against rebels, where independence, if gained at
all by the battle, was generally fairly short in duration. This includes not only all of the
Flemish/French battles, but also the battle of Morgarten, fought in 1315 between the
duke of Austria and Swiss rebels; the battle of Boroughbridge, fought in 1322 between
the army of King Edward II of England and the rebellious force of the duke of
Lancaster; the battle of Staveren, fought in 1345 between the William, the count of
Hainault, and his Frisian rebels; and the battle of Vottem, fought in 1346 between the
prince-bishop of Liège and his rebellious subjects. Of these, only the battle of
Morgarten brought some lasting independence for its victors. Finally, almost as an
anomaly, the battle of Laupen was fought in 1339 between two neighboring Swiss
towns; it also decided very little.
What all of these battles have in common, beyond the chronological similarity, is that
large infantry forces participated in all of them; and in most instances they were
victoriousthe battles of Mons-en-Pévèle and Cassel being the only exceptions.
Moreover, in each instance victory came not because of superior technology, even
when the longbow was present, but because a solid, disciplined, and tactically
proficient infantry line making a defensive stand was able to withstand the attacks of
opposing soldiers, whether cavalry or infantry.

Although nineteen battles are detailed here, these certainly are not all of the battles of
the early fourteenth century. The selection of these battles was made simply because
original sources detailing the tactics on the battlefield were available. In some cases,
notably the battles of Courtrai, Bannockburn, and Crécy, there were several sources
which reported the battlefield action, while other battles had only one or two sources
which discussed the tactics, namely Loudon Hill, Laupen, and Staveren. Even then,
the number of early fourteenth-century battles which are mentioned in the sources
without a discussion of tactics are numerous. This undoubtedly makes the early
fourteenth century the most prolific period for battles in the entire Middle Ages, and
certainly one of the reasons for the narrow focus of this study.
The chronological boundaries of this study are, hopefully, justified. The Black Death,
which begins in Europe in 1347, changed the nature of warfare for the remainder of
the Hundred Years War. Thus while there is surely some comparison between the
battle of Poitiers, fought in 1356, and those battles which are discussed here, the
difference in sizes of English and French forces at Poitiers and those at Crécy must be
ascribed to the loss of population. The decision of 1302 as the beginning point of the
study is perhaps more difficult to explain. In choosing to begin there, I have decided
to follow Verbruggen's lead in recognizing the originality of the battle of Courtrai
rather than moving the chronological time-frame back to the battles of Woeringen,
Stirling Bridge, or Falkirk, despite some similarities between those battles and the
ones which are included in this book.
Finally, in order to understand what this book is trying to say there is a need to define
the word 'battle'. While most dictionaries may define 'battle' in a rather
Page 8
general wayMerriam-Webster defines it as 'a general military engagement'only battles
which saw two armies entirely assembled and ordered against each other are included
for study here. That comprises all but three of the nineteen conflicts listed above,
excluding the battles of Morgarten, Auberoche, and La Roche-Derrien, all of which
are for the purposes here defined as 'ambushes'. (Details of these conflicts have been
included in an appendix.)
Page 9

I
The Battle of Courtrai, 1302
In 1302, Philip the Fair ruled France. Four years earlier, he had signed a truce with his
enemy, Edward I of England, after a small war fought entirely on Flemish soil which
resulted in Edward's loss and his expulsion from the continent. For his participation in
the war, Guy of Dampierre, the count of Flanders and ally to Edward I, was
imprisoned by Philip.1 James of St. Pol became the royal governor of Flanders who
enforced the king's laws with a large French army. (Indeed, it was the only standing
army in Philip's France.) The independent towns of Flanders considered any intrusion
by France into their affairs, especially occupation by a large fighting force, to be a
breach of their political power. James of St. Pol recognized the hatred of the Flemish
towns, especially Bruges, and on May 17, accompanied by Pierre Flote, the king's
chief advisor and negotiator, he took his army into Bruges to negotiate a peace with
the towns.
When they arrived at Bruges, most of the rebellious faction of the city had fled, and
those who were left refused to negotiate with James. The French army harassed the
townspeople, especially the families of those who had fled, and stories of
mistreatment were conveyed to the rebels hiding outside the town. At sunrise on May
18, while the French army for the most part was asleep, the Brugeois rebels attacked,
killing over 300 soldiers. James of St. Pol and Pierre Flote fled the city, James to the
castle of Courtrai and Pierre to the city of Lille, where they informed the king of the
rebellion and massacre.2
Knowing that this massacre would bring certain violent repercussions, the people of
Bruges amassed their forces and sent representatives to the other Flemish towns
asking for support. All but Ghent responded favorably. This army then moved to
Oudenaarde, the nearest stronghold of the French. Guy of Namur and William of
Jülich, the son and grandson of the imprisoned count, met the army at Oudenaarde
and joined their small forces to it, becoming the leaders of the entire armed
contingent. Oudenaarde fell within a few days, and the Flemish soldiers marched to
1 An excellent study of Philip's relations with Flanders is Frantz Funck-Brentano, Philippe le
Bel en Flandre: Les origines de la guerre de cent ans (Paris, 1896).
2 Several sources detail the 'Matins' massacre of May 18, 1302 in Bruges. The most notable
original source is the anonymous Annales Gandenses, ed. Hilda Johnstone (London, 1951), which
is, however, partisan to the Brugeois rebels. The best secondary narratives are Funck-Brentano,
Philippe le Bel, pp. 400-04, and J.F. Verbruggen, De slag der guldensporen, bijdrage tot de
gescheidenis van Vlaanderens vrijheidsoorlog, 1297-1305 (Antwerp, 1952), pp. 262-75.
Page 10
the next fortress held by French troops, the castle of Courtrai, reaching it on June 26.3
Philip the Fair was determined to avenge the Brugeois massacre, and he responded by
sending a large French army, composed mainly of mounted knights and led by Robert,
the count of Artois, against the rebels.4 On July 11, the two armies met in battle
outside the town and castle of Courtrai.
By evening, the Flemings had achieved the greatest victory in the county's history. The
French had fled from the battlefield with the Flemings chasing them for as many as
seven miles. The bodies of the dead were stripped, and from five to seven hundred
golden spurs, spurs given for victory in tournaments, were collected by the Flemings
and hung in the Cathedral of the Virgin at Courtrai. (This in turn gives the battle its
modern name: the Battle of the Golden Spurs.) The list of French dead was long and
contained a number of important French nobles. Dead were Robert of Artois, the
leader of the French force, the count of Eu, the count of Aubermarle, Godfrey of
Brabant, Jean of Hainault, Raoul de Neele (the constable of France), James of St. Pol,
Henry of Luxembourg, Pierre Flote and many more.
News of the victory spread quickly throughout Europe. The pope, Boniface VIII, who
had had previous difficulties with Philip, received by letter a detailed account of the
battle and rejoiced in the news.5 The victory was also reported in England, Holland,
Austria and Switzerland.
The Flemings were exuberant; they paved the streets of Bruges with flowers, dragging
the fleur-de-lys through the mud and shouting, 'On with the lion, and down with the
lily.' Lodewijk van Velthem, a contemporary Flemish chronicler, writes: 'Never did we
hear such sounds from the trumpet. The victors paraded up and down the town.'6
Another contemporary writer, the Florentine Giovanni Villiani, adds: 'So proud and
bold the Flemings have become through their victory at Courtrai, that one Fleming
with his goededag would dare invite a battle against two French knights on horse.'7
To these and other contemporary chroniclers, the battle of Courtrai was the most
important battle of their age, and indeed more fourteenth-century writers comment on
it than on any other battle of the century including those fought at Crécy and at
Poitiers. To many commentators the Flemish victory compared with the greatest
victories in history: with the Greeks over the Trojans,8 with the Israelites under David
3 See Verbruggen, Slag, pp. 262-75, for a discussion of the Flemish army movements between
May 18 and July 11, 1302.
4 For Philip's reaction to the Matins massacre and his efforts to mobilize an army see Funck-
Brentano, Philippe le Bel, p. 405. For a discussion of the size of this force see nos. 23-24 below.
5 See Gilles li Muisit, Chronicon, in Corpus chrvnicorum Flandriae, ii, ed. J.J. de Smet
(Brussels, 1841), p. 196.
6 See Lodewijk van Velthem, Voortzetting van de Spiegel historiael (1284-1316), ed. H. Vander
Linden et al, Commission royale d'histoire de Belgique (Brussels, 1922), II:338.
7 Giovanni Villani, Istorie Fiorentine, in Scriptores rerum Italicarum, xiii, ed. L. Muratori
(Rome, 1728), c.338: ' . . . per queste vittorie salirono in tanta superbia & ardire, che uno
Fiamingo a pie con uno Godendac in mano harebbe atteso due cavaglieri Franceschi a cavallo.'
8 See Jan de Klerk, Brabantse yeesten of rijmkroniek van Braband, ed. J. F. Willems and J. H.
Bormans (Brussels, 1839), I:426.
Page 11
at Gilboa9 and with the Romans under Scipio Africanus against the Carthaginians.10
To others the French defeat echoed Roland's loss at Roncesvalles.11 In fact, the story
of the battle of Courtrai became so well known that by the time the Dutchman Melis
Stoke wrote (c.1305) he refused to give the details of the battle itself saying: 'I will not
tell what happened there, for it is well known.'12
Although none of these contemporary or near contemporary sources claims to be an
eyewitness to the battle, several contain extensive accounts of what occurred there. As
well, there exists a contemporary historiography which presents both a French and a
Flemish version of what occurred at the battle. For the Flemings there is the Annales
Gandenses, written c.1310 by an anonymous Franciscan monk living in Ghent,
Lodewijk van Velthem's Spiegel historiael, written c.1316 by a Flemish ecclesiastic,
the Chronicon comitum Flandriae, the first part of which was written in 1329 by a
monk living in the abbey of Clairmarais in St. Omer, the Ancienne chronique de
Flandre and the Chronique de Flandre, which may be two redactions of the same
mid-fourteenth century chronicle written in St. Omer, and the early fifteenth century
Rijmkroniek van Vlaenderen. There is also the famous Oxford Chest, carved by a
Flemish sculptor shortly after the battle, which depicts the events of the conflict. For
the French there is the Chronique Artésienne, written c.1304 by an anonymous writer
who was probably from Arras, Guillaume Guiart's historical poem, the Branche des
royaux lignages, which was written by a French soldier who served in the Franco-
Flemish war but probably did not fight at Courtrai, Geoffroi de Paris' Chronique
rimée, written between 1314 and 1317 by a bourgeois writer living in Paris, Gilles le
Muisit's Chronicon, written by the abbot of St. Martin's in Tournai between 1346 and
1348, the Récits d'un bourgeois de Valenciennes, written c.1366, two continuators of
Guillaume de Nangis' Chronicon, the first who wrote in 1303 and the second in 1316,
the Grandes chroniques de France, written continually throughout the fourteenth
century at the Abbey of St. Denis in Paris and intended to be the official chronicles of
France, and the Chronographia regum Francorum, written in the early fifteenth
century by an anonymous author possibly also writing at St. Denis in Paris. As well,
there are four significant accounts of the battle written by 'foreign' authors, men who
undoubtedly spent much time in the wealthy southern Low Countries and there heard
the story of the battle: Ottokar von Stiermarken's Oesterreichische reimkroniek,
written by an official, perhaps a diplomat, of the court
9 See the poem appended to the manuscript of Johannes Thilrode, Chronicon, in Monumenta
Germanica historiae, scriptores (hereafter MGH, SS), xxv, ed. G.H. Pertz (Hannover, 1880),
p. 583 and Otto von Stiermarken, Oesterreichische reimkroniek, in MGH, Deutschen
Chroniken, v, ed. J. Seemuller (Hannover, 1893), II:858.
10 See Ottokar von Stiermarken, p. 858.
11 See Geoffroi de Paris, Chronique rimée, in Recueil des historiens de la Gaule et de la
France (hereafter RHF), xxii, ed. Guignant and de Wailly (Paris, n.d.), ll. 1364-66; Jean de
Winterthur, Chronicon, in MGH, SS (nova series), iii, ed. F. Baethegen (Hannover, 1924), p. 65;
Willem Procurator, Chronicon, ed. Pijnacker Hardwijk, Werken uitgegeven door bet Historisch
Genootschap te Utrecht, 3rd ser., 20 (Amsterdam, 1904), p. 65; and Jan de Klerk, Brabantse
yeesten, I:421.
12 Melis Stoke, Rijmkroniek van Holland, ed. W.G. Brill, Werken uitgegeven door het Historisch
Genootshcap te Utrecht, n.s. xl and xlii (Utrecht, 1885), p. 183: 'Te telne, wat hem daer na
gheval:/Wat si daden, weet men wel.'
Page 12
of Otto von Lichtenstein c.1309, Jean de Winterthur's Chronicon, written in the mid-
fourteenth century by a Swiss Franciscan, Willem Procurator's Chronicon, written by
the abbot of the Benedictine monastery of Egmond in the county of Holland c.1332,
and Giovanni Villani's Istorie Fiorentine, written c.1340 by an ambassador from
Florence who frequently visited Flanders during this period.13 From these sources a
good picture of what happened at the battle of Courtrai emerges.
The Flemish army arrived at Courtrai first, on June 26, planning to besiege the town's
castle which held a small garrison of French troops. The Flemish army was quite
large, having been drawn from many towns and villages in the county. Lodewijk van
Velthem records their numbers at 13,000, while the Annales Gandenses claims a total
of 60,000.14 Modern historians are not as generous in their calculations, although the
number they give also indicates a sizable force, between 7,378 and 11,000.15 Most of
these were infantry soldiers, clothed with very little armor and equipped only with a
popular local weapon known as a 'goededag.'16 There were
13Annales Gandenses, pp. 27-33; Lodewijk van Velthem, II:285-342; Chronicon comitum
Flandriae, in Corpus chronicorum Flandriae, i, ed. J.J. de Smet (Brussels, 1837), pp. 168-71;
Chronique de Flandre, in Istore et croniques de Flandres, ed. Kervyn de Lettenhove
(Brussels, 1879), I:476-79; Ancienne chronique de Flandre, in RHF, xxii, ed. Guignant and de
Wailly (Paris, n.d.), pp. 378-79; Rijmkroniek van Vlaenderen, in Corpus chronicorum
Flandriae, iv, ed. J.J. de Smet (Brussels, 1865), 797-801; Chronique Artésienne et chrunique
Tournaisienne, ed. F. Funck-Brentano (Paris, 1898), pp. 44-52; Guillaume Guiart, Branche des
royaux lignages, in RHF, xxii, ed. Guignant and de Wailly (Paris, n.d.), pp. 232-240; Geoffroi
de Paris, pp. 92-105; Gilles le Muisit, pp. 194-97; Récits d'un bourgeois de Valenciennes, ed.
Kervyn de Lettenhove (Brussels, 1877), pp. 113-16; Guillaume de Nangis, Chronicon et
continuationes, ed. H. Geraud, Société d'histoire de France (Paris, 1843), pp. 317-24; Les
grandes chroniques de France, ed. J. Viard, Société d'histoire de France (Paris, 1934),
IX:203-09; Chronographia regum Francorum, ed. H. Moranville (Paris, 1891), I:104-12:
Ottokar von Stiermarken, pp. 846-54; Jean de Winterthur, pp. 31-32; Willem Procurator, pp.
63-65; and Giovanni Villani (ed. Muratori), cols. 387-88. A good description of the Oxford
Chest is Charles ffoulkes, 'A Carved Flemish Chest at New College, Oxford,' Archaeologia,
2nd ser., 15 (1914), 113-28.
Henri Pirenne is the earliest historian to remark on the existence of a varying Flemish and French
version of the events at the battle of Courtrai. In an article entitled 'La version flamande et la
version française de la bataille de Courtrai, note historiographie de XIVe sièle' (Bulletin de la
commission royale d'histoire 4th ser., 17 (1890), 11-50), and a later 'Note supplémentaire'
(Bulletin de la commission royale d'histoire 5th ser., 2 (1892), 85-123), which was written in
response to questions which arose in Franz Funck-Brentano's work, Mémoire sur la bataille de
Courtrai (1302, 11 juillet) et les chroniqueurs qui en ont traité, pour servir à l'historiographie
du règne de Philippe le Bel (Paris, 1891), Pirenne argued that by studying the earliest and most
complete sources commenting on the battle of Courtrai it was possible to see two divergent
versions of the battle based on national origin of the commentator. For a comment on Pirenne's
thesis and a discussion of all the sources on Courtrai see Verbruggen, Slag and Verbruggen, 'De
historiografie van de guldensporenslag,' De leiegouw (1977), 245-72.
14 Lodewijk van Velthem, II: 289 and Annales Gandenses, pp. 28-29.
15 For an estimation of the number of Flemish soldiers at Courtrai see J.F. Verbruggen, Slag, pp.
199-237; Joseph de Smet, 'Les effectifs Brugeois à la bataille de Courtrai en 1302,' Revue Beige
de philologie et d'histoire 12 (1933), 631-36; and Ferdinand Lot, L'art militaire et les armées au
moyen âge en Europe et dans le proche orient (Paris, 1946), I:263-64.
16 That the Flemings were mostly, if not all, infantry soldiers is remarked on by several of the
sources. See, for example, the Annales Gandenses, p. 29; the Chronique de Flandre, I:477; and
Gilles le Muisit, p. 195. The use of the 'goededag' is also of interest to many, among them the
Grandes chroniques (VIII: 204). The 'goededag' seems to have been Flemish in origin and to have
been used there almost
(footnote continued on next page)
Page 13
also a few crossbowmen and a few knights, although the numbers of these troops
were quite small.17 Of the knights, few had military experience, and theseamong them
Guy of Namur, William of Jülich, John of Renesce, Henry of Loncin, Gossuin of
Godenshoven, Dietrich of Hondeschoote, Robert of Leewergem and Baldwin of
Popperordetook up positions of leadership on the field. Pieter de Coninck, the leader
of the massacre at Bruges, was also present and may have had some leadership
responsibilities.18
The Flemings surrounded the castle and began the siege. At the same time, perhaps
recognizing the probability of a battle there, they began to dig a number of ditches in
the fields beyond their besieging forces. These ditches would play an important role in
the battle which was to follow. Many of them were connected to the Lys river and thus
filled with water; others were hidden by dirt and branches.19 A later account, that of
the Kronyk van Vlaenderen, written c. 1477, claims that fog on the battlefield further
hid the ditches.20
While the Flemings were besieging the castle at Courtrai, Philip the Fair had begun to
gather an army. The response to the massacre of French troops at Bruges was
impressive. The large size of the French army is noted in almost every account of the
battle. Some of these sources, such as the Grandes Chroniques report only that the
army contained 'many great French knights and a large multitude of infantry,'21 while
the Chronicon comitum Flandriae compares the force with its 'multitude' to the
Flemish army 'with few men.'22 Several sources give specific numbers for these
troops ranging from a low number of 7,024 given by Lodewijk van Velthem to a high
of 20,000 mentioned in the Chronique Artésienne and in the Chronicon comitum
(footnote continued from previous page)
exclusively. Its chief function was to bring down a knight from his horse. The best
contemporary illustration of this weapon appears on a side panel of the famous Oxford Chest,
carved by a Flemish sculptor shortly after the battle to celebrate the victory. (For a discussion
of the Oxford Chest see Charles ffoulkes, 'A Carved Flemish Chest at New College, Oxford,'
Archaeologia 2nd ser. 15 (1914), 113-28.) J.F. Verbruggen has written an excellent article on
this weapon entitled 'De goededag,' Militaria Belgica 1977, 65-70.
17 Both Lodewijk van Velthem and the anonymous author of the Annales Gandenses mention
crossbowmen among the Flemish troops at Courtrai. They are also shown on the Oxford Chest.
Their number, however, is not mentioned. Verbruggen has determined that there were fewer than
500 crossbowmen present in the Flemish army at Courtrai (Slag, pp. 210-11). The number of
knights present offers even a more difficult calculation. While several authors note their presence,
and the fact that they dismounted to fight with the infantry, no firm estimate has been given. Even
Verbruggen seems hesitant to guess, again giving a total of under 500 (Slag, p. 211).
18 Few sources record the names of the Flemish leaders, except for Guy of Namur, William of
Jülich and John of Renesse (who may have had the chief military command). It is only the Annales
Gandenses which provides the list above (p. 29).
19 Almost all of the contemporary sources mention the ditches. Those who mention their
construction and concealment include Otto von Stiermarken, p. 852; Geoffroi de Paris, p. 148;
Jean de Winterthur, p. 22; and the Chronographia, 1: 105. The strategy of digging ditches to
disrupt the charge of an opponent was not new, having been used since the sixth century, if not
earlier. See R.H.C. Davis, The Medieval Warhorse (London, 1989), p. 14.
20Kronyk van Vlaenderen van 580 tot 1467, ed. P. Blommaert and C.P. Serrière (Ghent, 1839),
I:159. Although this is not mentioned by more contemporary sources, it is a possible occurrence as
this part of modern Belgium suffers frequent patches of quite thick fog during the summer months.
21Grandes chroniques, VIII:204.
22Chronicon comitum Flandriae, p. 168.
Page 14
Flandriae.23 Modern historians have reduced the latter number significantly,
estimating a total of 3,000 knights and 4,000-5,000 infantry in the French army.24 Still,
by medieval military standards, the French force was quite large.
Moreover, while the Flemish soldiers fought on foot, the main and most impressive
part of the French army was the cavalry. All the great and proud knights of France
were present at Courtrai, clothed impressively with expensive armor and equipped
with lances and swords. Many had won golden spurs at tournaments, and these
symbols, together with the large number of heraldic banners, represented a force of
soldiers considered by some to be the 'flower of French chivalry.'25 At their head was
the militarily experienced and distinguished Robert of Artois.
The French arrived at Courtrai on July 8. For three days the French army surveyed the
situation and planned for their attack. The Flemish army had chosen the site and had
prepared the field for battle. The French in turn jockeyed for a strong position from
which to attack them and raise the siege of the castle.26
Little is known about the military activity which occurred during this period. Both the
Annales Gandenses and Lodewijk van Velthem claim that some pre-battle skirmishing
took place between the two armies, although this seems to have resulted in little actual
combat.27 Some fighting may have taken place around the site of a destroyed bridge
across a 'river' running in front of the castle. Two French sources, the first continuator
of Guillaume de Nangis' Chronicon and the Grandes chroniques, claim that during the
days preceding the battle, the French troops attempted to repair the bridge which had
been destroyed by the Flemings during their preparation of the battlefield. However,
every attempt was thwarted, with the Flemings 'always attacking the French and
disturbing whatever work they attempted.'28 This may be an incorrect report,
however, as it is not mentioned by any of the Flemish sources, despite the fact that
such an occurrence should have elicited some comment. If it did occur, and the river
referred to is the Lys River, this could represent an attempt by the French army to
surround the Flemings and to attack them from the rear.
The Annales Gandenses also claims that Robert of Artois allowed his army to pillage
the Flemish countryside around Courtrai. According to this Flemish
23 Lodewijk van Velthem, II:302, 311; Chronique Artésienne, p. 40; and Chronicon comitum
Flandriae, p. 168.
24 For modern historical estimates of the French army size see Verbruggen, Slag, 237-50 and Lot,
I:261-62.
25 Many sources note the large number of knights in the French army, especially in comparison to
the few knights who fought alongside the Flemings. Some, like the Chronique Artésienne, even list
the names and titles of the most prominent French knights (pp. 44-45). See also Annales
Gandenses, pp. 27-28; Willem Procurator, p. 63; Ancienne chronique, p. 378; and La chronique
Liégeoise de 1402, ed. E. Bacha (Brussels, 1900), p. 246. The pride of these troops is
acknowledged in the Chronographia, I: 110.
26 See the Chronique de Flandre, I:476.
27Annales Gandenses, p. 29 and Lodewijk van Velthem, II:302.
28Continuatio chronici of Guillaume de Nangis, I:318: 'Ultra enim transire non poterant propter
pontem quem Brugenses fregerant super aquam fluminis prope Corteriacum decurrentis. Ubi dum
reparationi pontis gens Francorum intenderet, Brugenses saepius aciebus dispositis occurrentes, et
opus quantum poterant disturbantes, [quotidie] ad bellum Francigenas provocabant.' See also
Grandes chroniques, VIII:143.
Page 15
chronicle, the French killed women, children and the sick and decapitated and defaced
church statues in order 'to show their ferocity and terrorize the Flemings.' That this did
not have the desired effect for the French is attested to by the anonymous author's
further comment: 'However, such doings did not terrorize the Flemings, but stimulated
and provoked them to still greater indignation and rage and violent fighting.'29
Robert of Artois was also busy gathering intelligence about the Flemish troops and
their preparation of the battlefield. In particular, according to his register of expenses,
he purchased a map of the ditches made on the battlefield. This was bought from a
man known only as Pierre l'Orrible (perhaps a pseudonym) for the exorbitant amount
of xiii lb. x s. x d. par.30
Because they anticipated an attack at any time after the arrival of the French, the
Flemish army must have ordered their position and held that order until the battle was
fought. They knew that they held the best ground, with their backs to the river and the
ditches in front of them. With the exception of Gilles le Muisit, who writes that the
Flemings initially did not exhibit a strong spirit,31 almost every account of the battle
describes their high morale.32 They certainly were frightened, as is attested to by
Lodewijk van Velthem, not knowing whether they would live or die; but they believed
that God was on their side, and they believed that He would lead them to victory.33
Ultimately they were, in the words of Jean de Brusthem, 'rejoicing and excited, roaring
in the manner of lions.'34
The Flemings were also well led. It was these leaders who ordered their troops in a
single line as a shield wall against the Lys river, in a win-or-die formation.35 It was
they who also ordered the dismounting of knights and nobles to fight as infantry
alongside the non-noble soldiers. And it was they who kept the morale of the Flemish
army high. Many contemporary sources report that before going into battle the leaders
of the Flemings addressed their troops. John of Renesse took the lead, telling the
soldiers of the oncoming attack and asking them neither to flee nor to retreat. He
asked the troops to attack the horses with their weapons and to push them into the
ditches. He requested that they 'fight strongly for their wives and their children, for
29Annales Gandenses, p. 28: 'Franci autem Flandriam intrantes Flamingantem, ut ostenderent
ferocitatem animi sui, et Flandrenses terrere volentes, non parcebant mulieribus, nec infantibus,
nec decrepitis, quin eos occiderent quos invenire poterant; imo et imagines sanctorum in
ecclesiis, ac si homines fuissent vivi, decapitaverunt, alia eis etiam membra amputantes. Hoc
autem factum Flandrenses non terruit, sed magis ad iram et furorem et ad crudeliter pugnandum
animavit et provocavit.'
30 Robert of Artois' register of expenses is found in Funck-Brentano, Mémoire, pp. 79-85. The
purchase of the map is found on p. 80.
31 Gilles li Muisit, p. 195.
32 See, for example, the description of the Flemings in the Grandes chroniques, VIII:204 and in
the Rijmkroniek van Vlaenderen, p. 800.
33 Lodewijk van Velthem, II:303. See also the Récits d'un bourgeois de Valenciennes, pp. 114-
15.
34 Jean de Brusthem, Chronique, in Chroniques Liégeoises, ii, ed. C.S. Balau (Liège, 1931), p.
60: 'Quo gaudentes et animati Flandrenses, more leonum rugientes et viriliter decertantes, tandem
Francos bello vicerunt.' This chronicle, essentially an abridgement of Jean de Hocsem's
Chronicle, was written in the early sixteenth century. This passage is not found in Hocsem's work.
35 A number of sources characterize the Flemish formation as a shield wall. These include:
Lodewijk van Velthem, p. 303; Chronographia, I: 106; and Chronique de Flandre, I:477.
Page 16

Battle of Courtrai
(First Phase)
the laws and liberty of their homeland.' He promised the troops that if they resisted
pride, putting their faith in God, that 'He would be merciful to those who were
humble, giving consolation and victory to them.'36 It was only the leaders, according
to the Chronicon comitum Flandriae, who kept the Flemings from fleeing at the sight
of the French army.37
On the French side a different picture is given. Most sources report that the French
36 There are two different versions of this oration. The Chronicon comitum Flandriae (p. 168)
does not mention who gave the speech, but records it as:
Estote, inquit, viri fortes et viriliter pugnate pro uxoribus et liberis vestris ac pro legibus ac
libertatis patriae vestrae, quia nisi praestiteritis, non remanebit de vobis neque de genere
vestro mingens ad parietem, quin omnes gladio perfodiantur. Quicumque autem fugerit vel
retrocesserit, vel ad spolia se inclinaverit usque post bellum perfossus pereat. Equos invadite
primo cum lanceis et baculis, et equo perfosso, miles subito superatur.
The second version of the battlefield oration, found in Willem Procurator's Chronicon (p. 64), is
given by John of Renesse and is entirely different:
O fortissimi et divites, patriam et filios defendentes! Non turbetur cor vestrum neque formidet;
nam vos animas vestras pro justitia ponitis et Deus justus suam pro vobis semel positam adhuc
minus subtrahit, si ejus pietati et misercordie spem et fiduciam tribuatis. Ecce, vos humiliter
proponitis vestrum defendere et ipsi in superbia et abusione alienum tollere. Unde qui superbis
resistit, ipsis deficiet, et qui humilibus dat gratiam, ipse vos consolatione et victoria decorabit.
See also Lodewijk van Velthem, II:305-06 and Kronyk van Vlaenderen, I:158. On the praise for
the Flemish leaders see Lodewijk van Velthem, II:305-06, 313.
37Chronicon comitum Flandriae, p. 169.
Page 17
were haughty, believing that they could defeat the Flemings at will.38 Certainly Robert
of Artois held this opinion. When he called a council on the day of battle, the
Chronographia regum Francorum notes, he ignored the advice of many of his nobles
who did not want to attack the Flemings,39 and shortly before noon, with trumpets
blaring he ordered his troops in three lines and began to attack the Flemish army.40
The battle began with an exchange of crossbow fire. Because there were so few
archers on either side this attack was largely ineffectual.41 The French infantry then
began their approach, making their way across the ditch-filled field quite easily it
seems, until they began to attack the Flemings. So effective was this attack that, in the
words of Gilles le Muisit, 'they were almost at the point of victory.'42 But the infantry
attack was halted by Robert of Artois who insisted that the knights be granted the
victory. As the Ancienne chronique de Flandre reports the incident, a French knight
seeing that the infantry was about to defeat the Flemings approached Robert of Artois
and asked him: 'Lord, why do you wait any longer? Our infantry . . . advances so that
they will have victory here, and we will obtain no honor.' Robert responded by
recalling the infantry and charging his knights across the field.43
The French cavalry charge was disastrous. The knights charged 'pompously and
without order' passing over the ditches and hitting a solid, impenetrable Flemish line.
The French charge was halted; unable to pierce the infantry shield wall, the French
horses stopped and the knights were pushed into ditches behind them.44 Ensuing
cavalry charges did not help, as these knights and their horses became further
enmeshed in the ditches and the confusion. Nor did a desperate charge from the castle
provide relief for the fallen French cavalry, for it was easily turned back by a Flemish
contingent placed there to prevent such a charge.45 In the words of a Middle English
poem written to celebrate the victory, the French knights were like a 'hare' caught
38 See the Chronique Artésienne, p. 48.
39Chronographia, I:107. On the battle council see also the Chronique de Flandre, I:477 and the
Récits d'un bourgeois de Valenciennes, p. 114.
40 On the order of the French lines see the Annales Gandenses, pp. 29-30; the Chronicon
comitum Flandriae, p. 169 and the Oxford Chest (ffoulkes, pl. I). On the time of the battle see the
Annales Gandenses, p. 30.
41 See the Chronographia, I:107; the Oxford Chest (ffoulkes, pl. I; and Jean Desnouelles,
Chronique, in RHF, xxi, ed. Guignant and de Wailly (Paris, 1855), p. 90.
42 Gilles li Muisit, p. 195: ' . . . et pedites dicti comitis Flandrenses superare, et esse quasi sub
puncto habendi victoriam.' See also the Chronographia, I:107.
43Ancienne chronique de Flandre, p. 378: ' ''Sire, que attendez vous plus? Noz gens de pié, quy
de pres nous sièuent, s'avanceront tellement, que ilz en auront la victoire, et nous ne y acquerrons
point d'honneur." ' See also the Chronographia, I:107 and the Chronique de Flandre, I:476-77.
The Chronique de Flandre insists, however, that the French knights charged onto the battlefield
only because they thought that the Flemings were fleeing from it (I:476-77).
44 On the disastrous French charge see the Annales Gandenses, p. 30; Lodewijk van Velthem,
II:307-19; the Chronicon comitum Flandriae, p. 169; the Grandes chroniques, VIII:204-05;
Giovanni Villani (ed. Muratori), col. 387; the Kronyk van Vlaenderen, I:161; Thomas Gray,
Scalachronicon, ed. J. Stevenson (Edinburgh, 1836), pp. 127-28; and Ives, Pars ultima
chronicon, in RHF, xxi, ed. Guignant and de Wailly (Paris, 1855), p. 204.
45 The castle attack is mentioned only in the Annales Gandenses, p. 30, but it is also shown on the
Oxford Chest (ffoulkes, pl. I).
Page 18

Battle of courtrai
(Second Phase)
in a 'trap.'46 Jean de Hocsem uses another metaphor describing the knights falling into
the ditches as 'oxen slaughtered as a sacrifice without defense.'47
With victory at hand, the Flemish infantry marched forward across the field. They
took no prisoners, killing any French knight not yet slain or drowned.48 Among those
yet to die was Robert of Artois. According to several sources he had lost his horse in a
ditch, but had not fallen in himself. Unhorsed and defeated, in a scene reminiscent of
Shakespeare's Richard III, he tried to save himself by pleading for his life. But his
pleas went unheeded as he was killed.49
46 There are several editions of this poem. The one I have used is edited by Rossell Hope
Robbins in his collection, Historical Poems of the XIVth and XVth Centuries (New York,
1959), pp. 9-13. Other editions can be found in Political Songs of England, Camden Society,
vi, ed. T. Wright (London, 1839), pp. 187-195 and Chants historique de la Flandre, 400-1650,
ed. L. de Baecker (Lille, 1855), pp. 161-72. The above reference can be found in Robbins, pp.
192-93.
47 Jean de Hocsem, La chronique de Jean de Hocsem, ed. G. Kurth (Brussels, 1927), p. 105: ' . .
. et apud Curtracum commisso prelio Francigenis ruentibus in fossatis sicut boves ad victimam
sine defensione.' The Annales St. Jacobi Leodensis (in MGH, SS, xvi, ed. G.H. Pertz (Hannover,
1850), p. 653), in copying Hocsem's account, dropped the 'b' from boves to describe the knights as
sheep, giving the phrase its more accustomed proverbial nature but losing the bulky image which
Hocsem had probably intended initially: 'Fuit bellum apud Curtracum, ubi Francigenis in fossatum
ruentibus, sicut oves sine defensione mactantur.'
48 See the Annales Gandenses, p. 31 and the third continuation of the Gesta abbatum St.
Trudonien-sum, in MGH, SS, x, ed. G.H. Pertz (Hannover, 1851), p. 409.
49 On the death of Robert of Artois see Lodewijk van Velthem, II:321-22; the Récits d'un
bourgeois de Valenciennes, p. 115; and the Middle English poem (Robbins, pp. 192-93). All
portray Robert
(footnote continued on next page)
Page 19
There were many French dead strewn across the field and filling the ditches. The
Flemings gathered their booty, among which were the golden spurs to be hung in the
Courtrai church. Then they retired from the field, leaving the bodies of their dead foes
to putrefy.50
All of the chroniclers of the battle are impressed by the heavy casualties. Many French
sources even record the names of the knights killed in lists which often surpass in
length the account of the battle itself.51 Modern historians too are impressed by the
large numbers of knights killed, estimating that between forty and fifty percent of the
French cavalry were lost.52
Why did the Flemings defeat the French knights? Modern historians disagree on their
answer to this question. Sir Charles Oman claims that the victory was a 'mere accident'
and that the French nobility 'comforted themselves with the reflection that it was the
morass and not the Flemish infantry which won the battle.'53 Henri Nowé blames the
French loss on the battlefield ditches hidden deceptively by the Flemings.54 Several
other historians point to the ineptitude of Robert of Artois as a military leader. Indeed,
no less a historical figure than Napoleon III remarks on the inane charge of the French
cavalry over their own infantry, an event which does not seem to have occurred, the
infantry having left the battlefield without incident. Yet, others echo his complaint.
Georges Digard blames Robert for sending his knights 'blind' over a ditch-filled field,
and Joseph Strayer sees a certain victory if the infantry had been allowed to
continue.55 But, it is Henri Pirenne who levels the
(footnote continued from previous page)
pleading unsuccessfully for his life. Only Gilles li Muisit differs with this picture, claiming
instead that Robert chose to die rather than live in humiliation after the defeat of his troops (p.
319).
50 For a reference to the rotting and putrification of the French corpses see the continuator of
Guillaume de Nangis' Chronicon, pp. 319-20 and the Oxford Chest (ffoulkes, pl. I). This image is
also a central theme in the ribald Flemish 'Gospel,' the Passio Francorum secundum Flemingos,
written after the battle and edited in Paul Lehmann, Die Parodie im Mittelalter (Stuttgart, 1963),
pp. 30-31. (Another edition and extended study of this source is J.M. de Smet, 'Passio francorum
secundum flemyngos,' De leiegouw (1977), 289-319. See also Verbruggen, 'Historiografie,' pp.
246-47.) Contemporary sources report that none of the fallen soldiers was buried on the
battlefield; however, the Chronographia (I:112-14) does note that angels saved the body of
Robert of Artois from mutilation by sneaking it off to the safety of a nearby monastery.
51 See, for example, the lengthy death lists found in the Chronique Artésienne, pp. 49-51; the
Chronicon comitum Flandriae, p. 170; the Chronographia, I:110-12; the Chronique de Flandre,
I:478-79; the Grandes chroniques, VIII:207-08; the third continuation of the Gesta abbatum St.
Trudoniensum, p. 409; the Chronique Normande de xiv siècle, ed. A. and E. Molinier (Paris,
1882), pp. 18-19; and the Chronique des Pays-Bas, de France, d'Angleterre et de Tournai, in
Corpus chronicorum Flandriae, iii, ed. J.J. de Smet (Brussels, 1856), pp. 123-24.
52 Philippe Contamine in his War in the Middle Ages (trans. M. Jones (London, 1984), p. 258)
puts the loss at forty percent of the French knights while both Verbruggen (De krijgkunst in west-
europa in de middeleeuwen (IXe tot XIVe eeuw) (Brussels, 1954), p. 318) and Funck-Brentano
(Philippe le Bel, p. 415) select the larger, fifty percent figure.
53 Sir Charles Oman, A History of the Art of War: The Middle Ages from the Fourth to the
Fourteenth Century (London, 1898), pp. 592-93. It should be noted that Oman himself changed
his mind on the importance of the battle of Courtrai in his two volume revision of this work.
However, on the issue of whether the defeat of the knights was an 'accident' or not, Oman does not
alter his conclusions. See A History of the Art of War in the Middle Ages (London, 1905), II:118.
54 Henri Nowé, La bataille des éperons d'or (Brussels, 1945), p. 80. See a similar argument
used by R.H.C. Davis, p. 27.
55 Napoleon III and I. Favé, Du passé et l'avenir de l'artillerie (Paris, 1856), I:33-34; Georges
Digard,
(footnote continued on next page)
Page 20
harshest accusations against Robert of Artois calling him 'imprudent' and claiming that
'a charge of cavalry against a numerous infantry, formed in compact mass, solid and
immobile as a wall, on a marsh plain and intersected by ditches is unable to end in
anything but disaster.'56
Three historians, Frantz Funck-Brentano, Ferdinand Lot and J.F. Verbruggen, all
credit the infantry tactics for providing the victory at Courtrai; it was the first victory
of infantry over cavalry in a line of many later successes. Funck-Brentano is almost
poetic in his discourse on the subject:
Jacques de Châtillon and Robert of Artois were both representatives of the old world, of
chivalry, of a feudal society. They had power, simple character and grandeur. But in Flanders
they found themselves to be in the presence of a new world, where thought was stronger than
force, a modern world by which they were conquered. Robert of Artois was the hero of
chivalric wars where grand strokes of the sword decided the victory on an equal terrain
between adversaries of equal arms. To strike a horse on the head was a felony at least if you
were not fighting heretics. A whole other war awaited them, all the strategy and ambushes
wherein the chivalry of France was to fall in one glorious catastrophe; what a fateful day for the
cause they were defending.57
In looking at the contemporary accounts of the battle, one cannot help but agree with
this trio of distinguished modern historians. Surely the battle of Courtrai shows that
infantry was capable of defeating knights if they thoughtfully chose and prepared their
terrain and if they stood their ground fighting a defensive rather than an offensive
battle. It is too easy simply to blame the defeat on the ineptitude of Robert of Artois.
Certainly he was a proud man, as were his knights who considered the Flemings to be
'poor and unarmed rusticos' in comparison to their expertise in warfare.58 Perhaps he
was also incautious and should have heeded the advice of his councilors who warned
him not to attack the Flemish army that day. But we are not
(footnote continued from previous page)
Philippe le Bel et le Saint Siège de 1285 à 1304 (Paris, 1936), II:118; and Joseph Strayer, The
Reign of Philip the Fair (Princeton, 1980), p. 334.
56 Henri Pirenne, 'Note supplementaire,' pp. 89-90: 'Une charge de cavalerie contre une infantrie
nombreuse, formée en masse compacte, solide et immobile comme un mur, dans une plaine
marécageuse et coupée de fosses, ne pouvait qu' aboutir à un désastre.'
57 Funck-Brentano, Philippe le Bel, pp. 405-06: 'Jacq. de Châtillon et Rob. d'Artois étaient tous
deux les représentants d'un monde vielli, de la chevalerie, de la société féodale; ils en avient la
force, le caractère simple et la grandeur: mais en Flandre ils devaient se trouver en présence d'un
monde nouveau, où la réflexion avait plus d'action que la force, le monde moderne, par lequel ils
devaient être vaincus. Rob. d'Artois était le héros des guerres chevalereques où les grands coups
d'épée décidaient de la victoire sur un terrain égal, entre adversaires également armés. Frapper
les chevaux à la tête était félonie, à moins que l'on ne combattit contre les hérétiques. Une toute
autre guerre l'attenfait, toute de strategie et d'embûches, où la chevalerie française allait s'écrouler
en une catastrophe glorieuse, mais néfaste à la cause qu'elle défendait. See also Lot, I:252 and
Verbruggen Krijgkunst, p. 245.
58 The description of the Flemings as 'rusticos' is found in Jean de Paris, Memoriale temporum,
in RHF, xxi, ed. Guignant and de Wailly (Paris, 1855), p. 638 and the Continuatio chronici of
Guillaume de Franchet, in RHF, xx, ed. Guignant and de Wailly (Paris, 1840), p. 20. For the pride
of the French army see also the Chronographia, I:110 and the Continuatio chronici of Guillaume
de Nangis, I:331. The Chronographia also claims that the French army was infelicitous and
corrupt (I:110).
Page 21
told why these nobles did not want to fight: whether they actually perceived the
danger in fighting an infantry army on a battlefield filled with ditches, or whether they
had some other reason, perhaps not so prophetic, for desiring not to fight. In any case,
a general's incaution in fighting against medieval infantry forces before this time had
rarely meant defeat.
Nor can we attribute the Flemish victory only to the presence of the ditches on the
battlefield. Despite the fact that these ditches are almost always mentioned by
contemporary chroniclers, with some calling them 'treasonous,' 'evil' or 'mischievous,'
and others claiming that the French 'trembled' when recounting them, there are many
reasons why they cannot be recognized as the single cause of the French cavalry's
defeat.59 For one thing, few Flemish sources accord them any importance in
determining the outcome of the battle. They are not portrayed on the Oxford Chest
nor are they mentioned in the Annales Gandenses, and while they are discussed
briefly in Lodewijk van Velthem's Spiegel historiael, he is not impressed by their
destructive presence.
Second, those Flemish sources which do mention the ditches and the role they played
in the defeat see their presence only as a secondary cause of defeat. For example, the
Chronicon comitum Flandriae claims that the Flemings did not expect what
eventually occurred; they were surprised that the ditches were so effective.60 As well,
the Ancienne chronique de Flandre and the third continuation of the Gesta abbatum
St. Trudoniensum insist that the French knights did not encounter the ditches until
after they had passed them. Only after being pushed back by the Flemish infantry into
the ditches were they a factor in the defeat.61
Third, it was not the ditches themselves which caused the deaths of so many of the
French knights. As Geffroi de Paris points out, it was the fact that they were filled
with water and mud which seemed to have marked the end for so many heavily
armored soldiers, especially if in being pushed back by the infantry line the knight's
horse fell on top of him trapping him in the mud and under the water.62
Finally, despite the fervent belief of Giovanni Villani and others that the French
knights did not know about the ditches on the battlefield,63 other sources attest that
they were indeed aware of their presence. For not only does the Ancienne chronique
de Flandre report this knowledge, especially concerning a large ditch nearest the
Flemish lines which the anonymous writer of this chronicle describes as 'very large
and very deep,'64 but the mere fact that the French infantry, visible to their cavalry
59 These descriptions of the ditches can be found in the Chronique anonymé Française
finissant en MCCCLVI, in RHF, xxi, ed. Guignant and de Wailly (Paris, 1855), p. 139; the
Chronique Artésienne, p. 48; Guillaume Guiart, 1. 6478; and Geoffroi de Paris, II, 1488-90.
60Chronicon comitum Flandriae, p. 169.
61Ancienne chronique de Flandre, p. 378 and the third continuation of the Gesta abbatum St.
Trudoniensum, p. 409.
62 Geoffroi de Paris, II. 6086-89.
63 Giovanni Villani (ed. Muratori), col. 387.
64Ancienne chronique de Flandre, p. 378. See also the Récits d'un bourgeois de Valenciennes,
p. 115; the third continuation of the Gesta abbatum St. Trudoniensum, p. 409; and the Chronique
de Flandre, I:476.
Page 22
cohorts, would also have needed to cross over them is evidence enough to support the
knowledge of their presence. Ultimately, perhaps the best evidence to support the
French knowledge of the Courtrai ditches is that Robert of Artois, as shown in his
expense ledger, had purchased a map from a spy showing him where the ditches were
on the battlefield.
The Flemings had dug the ditches there only to disrupt the charge of the French
knights so that it might be weakened by the time it came into contact with the infantry
line. It was the infantry line then which was the cause of victory at Courtrai. As was
promised by their leaders, and because they had no place to flee with their backs to
the river, the Flemish infantry did not break from their line. They stood bravely
awaiting the charge of the French knights who rode with lance couched, clothed in
expensive and impressive armor, and atop large warhorses. Their bravery should not
be understated, for they could not have anticipated what would happen next. Unable
to find a path through the Flemish line, and unwilling by nature to run over a man, the
horses stopped.65 The knights were unable to fight as accustomed, and the Flemings
were able to advance without hindrance, pushing their horsed opponents into the
ditches which they had just ridden over. A reinforcing assault by another French
cavalry line would not only meet with the same result, but it would also run into its
own first line, a situation which occurred at Courtrai.
An infantry force could and did defeat an experienced cavalry army. At Courtrai the
Flemish infantry chose their field of battle, selecting a good defensive position from
which to fight. They prepared the field for battle by digging ditches in an attempt to
hinder and perhaps also to narrow any orderly attack against them, especially a
cavalry charge. Finally, they chose to fight a defensive battle, ordered in a solid line,
unbroken by flight or retreat. And in the end the cavalry opposing them was unable to
defeat them, suffering humiliation and a heavy loss of life. It was, in the words of the
anonymous author of the Chronicon Rotomagensi, 'as if the whole flower of French
knighthood had disappeared.'66
There is an interesting postscript to the battle of Courtrai. Writing nearly one hundred
years after the battle and chronicling a French victory at the same town in 1382, Jean
Froissart reports that the French remembered that it was there that 'the count of Artois
and all the flower of French nobility were slain.' To avenge this loss, the French
burned the city and confiscated the golden spurs which still hung in the Cathedral of
the Virgin.67
65 For a convincing discussion of this phenonemon at the battle of Agincourt see John Keegan,
The Face of Battle (Harmondsworth, 1978), pp. 94-96.
66Chronicon Rotomagensi, in RHF, xxiii, ed. de Wailly, Delisle and Jourdain (Paris, n.d.), p.
236: ' . . . quasi totus flos militiae galicanae decedit, non quidem virtute hostium, sed incautela
pugnantium, cadentium in fossata quae adversarii antea fecerant et cooperverant fradulenter.' See
also Bernard Gui, Flos chronicorum necnon e chronico regum Francorum, in RHF, xxi, ed.
Guignant and de Wailly (Paris, 1855), p. 713.
67 Jean Froissart, Chroniques, in Oeuvres de Froissart, ed. Kervyn de Lettenhove (Brussels,
1870), XIII: 177-78: 'Là y ot de rechief grant ochission et persécution faite, aval la ville, des
Flamens qui y estoient repus, ne on n'en prendoit nul à merchy; car li François haioient la ville
durement pour une bataille qui fu devant Courtray, où li contes d'Artois et toute la fleur de France
fu jadis morte.'
Page 23

II
The Battle of Arques, 1303
All of the county of Flanders joined in the celebrations which followed the important,
and unexpected, victory over the French at the battle of Courtrai. Towns, like Ghent,
which had neglected to join the rebellion before the battle quickly federated
themselves with the victorious rebels; support for the rebellion also came from the
rural areas of Flanders. Guy of Namur and William of Jülich were heralded as heroes
and assigned the task of governing the newly independent county. For administrative
and military assistance in this task, Guy sent for his brothers, John and Henry of
Namur and Philip of Chieti.
There was little doubt that the king of France, Philip IV, would attempt to retake his
northern county. Flanders was far too wealthy a possession for Philip to allow it to be
independent from the French realm.1 Moreover, a secession of one of France's fiefs
might encourage other unstable provinces, such as Bordeaux, Picardy, or Gascony, to
seek their own independence.2 So the sons and grandsons of Guy of Dampierre, still
imprisoned by Philip IV, prepared for another French attack by strengthening the
defenses of the major towns and building new fortresses which would hinder a
French assault of their county.
The Flemings did not concern themselves solely with defensive matters, however, as
they took advantage of the French confusion following Courtrai to secure their
territory and to make new conquests. Within days of this victory, the castles of
Courtrai and Cassel fell to the rebels, and their defeat was followed closely by the
surrender of the towns of Lille, Douai, and Termonde; only Termonde resisted the
Flemish attacks.3
There was no celebration held in France after the loss at Courtrai. A stunned Philip the
Fair was forced to gather new funds and a new army to try to regain his lost
territories. In addition to this, on November 18, perhaps instigated by the Flemish
1 The two best discussions, despite their age, concerning the events between the battles of
Courtrai and Arques can be found in Pirenne, Histoire de Belgique, I:392-95 and Funck-
Brentano, Philip le Bel et Flandre, pp. 430-71. As for Philip the Fair's realization of the
economic consequences of losing Flanders, see the Chronicon comitum Flandriae, p. 173.
2 Both the first and second continuations of Guillaume de Nangis's Chronicon (I:324, 334)
describe a rebellion which erupted in Bordeaux inspired by the success of the Flemish rebellion.
As well, the Chronographia regum Francorum (I: 119) reports that in 1303 some of the French
nobles approached Philip IV asking him to raise an army to put down the Flemish rebellion as they
feared that his inaction might encourage their own lands to rebel against them.
3 See the Annales Gandenses, pp. 34-35.
Page 24
victory, Philip's old foe, Pope Boniface VIII, issued his famous bull Unam sanctam.
Unam sanctam reinforced the pope's earlier decretal, Clericos laicos, forbidding the
use of church subsidies by lay rulers unless consented to by the Pope himself. There
is no doubt that this bull was directed at Philip's situation.4
The French king, it seemed, had been defeated on two fronts, and yet he held his
church and people together.5 Moreover, Philip ably kept the English from seizing the
opportunity to attack a weak and militarily unprepared French realm by negotiating an
alliance with his potential enemy based on the cession of Gascony to England.6 Philip
could now concentrate on the recovery of the county of Flanders. However, it would
be nearly two years before France was capable of attacking the rebels. Philip did
gather an army before the end of 1302 and marched as far north as Vitry, two miles
south of the Flemish town of Douai. But after a stay of five to six weeks, this army
ran out of food and was forced to return to France, a return described as 'inglorious'
by at least one contemporary French author.7
The Flemings did not wait for the French army's recovery, and before 1302 had
ended, rebel armies attacked the counties of Holland and Hainault, lands to the north
and southeast of the county of Flanders but controlled by the same lord, Count Jan I.
Essentially the Flemish army had split into two parts, with William of Jülich leading a
force south to Tournai and Guy of Namur leading another army north into Zeeland.8
William's goal initially was much larger than Guy's; he besieged the French-partisan
town of Tournai.
Tournai did not fall to the Flemings. Indeed, the impatient rebels only besieged the
town for three days before they became fatigued and moved on to 'easier' targets. The
anonymous author of the Annales Gandenses writes:
The Flemings . . . besieged Tournai for three days, and did much damage to the city and the
neighbourbood. They could not have taken it, however, unless by a long siege. So, being weary
of the tumult of war and already burdened with heavy expense all through the summer, they left
it and returned to their tents.9
William of Jülich's Flemings moved on to the supposedly easier siege of Lessines.
4 On the effect of Unam sanctam on Philip the Fair see Strayer, pp. 272-76.
5 It seems somewhat amazing that Philip, in the wake of military and ecclesiastical defeat, was
able to keep his church together in support of him. But by a brilliant use of propaganda, through
letters sent to his bishops and read aloud as 'sermones' before his people, Philip was indeed
successful in uniting his kingdom in the wake of this defeat. See, for example, the original sources
contained in Digard, II: 122; Funck-Brentano, Mémoire sur la bataille de Courtrai, pp. 85-88;
and Jean Leclerq, 'Un sermon prononcé pendant la guerre de Flandre sous Philippe le Bel,' Revue
du moyen âge latin 1 (1945), 165-72.
6 See EM. Powicke, pp. 653-54.
7 See, for example, the second continuator of Guillaume de Nangis, I:337.
8 See Chronographia regum Francorum, I:118, for a description of the Flemish attack of
Zeeland.
9Annales Gandenses, p. 37: 'Flandrenses . . . Tornacum tribus diebus obsederunt, damna plurima
dicte civitati circumquaque inferentes. Ipsam tamen, nisi longiori tempore sedissent circa eam,
non poterant obtinere; unde bellico tumultu lassati ac expensis maximis per totam estatem
pregravati, eam reliquerunt, in tabernacula sua revertentes.'
Page 25
By 1303, William of Hainault, the son of Jan I, decided that he could no longer wait
until a French army would save his two counties from destruction at the hands of the
Flemish rebels. He gathered an army from his counties and proceeded against the
Flemings at Lessines.10 The onslaught was successful; the rebels, unaccustomed to
military opposition, were forced to give up their siege and retreat north. Concurrent
with this minor defeat, and maybe because of it, Guy of Namur increased his attack
against Holland, assaulting Zeeland from the land and sea in a concerted effort to take
the town of Middleburg. William of Hainault was forced to march his small army
north to defend the county of Holland.11
This maneuver lessened the pressure on William of Jülich's force, and he again moved
south. First, he attacked and captured Lessines, now made vulnerable with the
Hainaulter army in Holland. His next target was the French-garrisoned town of St.
Omer. The French soldiers inside the town, among them the constable of France,
Gaucher of Châtillon, responded by mounting a small attack on the rebel armies at the
village of Arques, a short distance from St. Omer. On April 4, 1303, the battle of
Arques was fought; by the end of the day, the French had retreated to the town,
leaving the battlefield to the Flemings. But it was a costly 'victory' for the Flemings, as
they had lost many, and on the following day, after burying their fallen troops and
those of the French in a large mass grave, they too retreated from the site of the
battle.12
The original sources for the battle of Arques are the same as those for Courtrai and
Mons-en-Pévèle, with the Annales Gandenses, the Chronique de Flandre, and
Guillaume Guiart, a combatant on the French side, as the most complete narratives.13
The battle of Arques is really a play in three acts. The first act begins with the travel of
William of Jülich with his rebellious Flemish troops from the town of Lessines to St.
Omer. Lessines, held by a garrison of German mercenaries according to the Annales
Gandenses, had been attacked forcefully, 'with machines of war and daily assaults,' by
the Flemings. They had taken it, but in the end their victory was somewhat hollow; the
whole town was burned, with the gates, towers, and walls destroyed.14
Why William of Jülich chose to go to St. Omer after his victory at Lessines is a
question left unanswered by the original sources. Only a few months prior to
William's journey, an army of Flemings had tried to raid the county of Artois by
traveling past St. Omer. Just outside the town they met Eudes of Burgundy whose
large force put an end to the invasion by massacring most of the raiders.15 Perhaps
10 See Annales Gandenses, pp. 37, 42.
11 For a description of the Flemish military activities in Zeeland see Annales Gandenses, pp. 42-
48.
12 There are few modern descriptions of the battle of Arques. The longest is Verbruggen, De
krijgkunst, pp. 319-25. See also Funck-Brentano, Philippe le Bel et Flandre, pp. 437-41 and
Nowé, pp. 85-86.
13 In their narratives of the battle of Arques, unlike their accounts of Courtrai and Mons-en-
Pévèle, the Chronographia regum Francorum is a Latin translation of the Chronique de Flandre.
14Annales Gandenses, p. 39.
15 See Chronique Artésienne, pp. 58-59; Chronique de Flandre, I:259-60; Chronographia
regum Francorum, I:123-24; Récits d'un bourgeois de Valenciennes, pp. 117-18; Chronique
Normande, p. 23; Ancienne chronique de Flandre, p. 384; and Rijmkroniek van Vlaenderen, p.
806. Jean Desnouelles also mentions the defeat of the Flemings in Artois, but interchanges it with
the 1303 battle of Arques (p. 193).
Page 26

Battle of Arques
(according to the Annales Gandenses)
in response to this victory, or perhaps to guard against further invasions, Philip IV
placed a large garrison of troops in St. Omer'mercenaries' the Annales Gandenses
calls themunder the command of Jacques of Bayonne and Gaucher of Châtillon.16
Thus such a target seems to have been an unwise selection by the young Flemish
military leader. Why attack it then? The Chronique de Flandre claims that William
merely wanted to retrieve the body of an unnamed brother killed in Eudes'massacre,
while the Annales Gandenses, not a fan of this Flemish warrior, uses the words
'foolish' and 'indiscreet' to describe William's decision to go against St. Omer.17
But the explanation may also be one of more rationale military strategy. William of
Jülich, brash of character and already many times victorious against the French,
wished merely to use his large force to defeat the only sizeable army which Philip IV
could employ to oppose the complete independence of Flanders. Victory would also
in all likelihood cause a withdrawal by William of Hainault from his fight against the
Flemings in Zeeland. However, to accept this explanation for William's move, we must
agree with the claim of most original sources that the Flemish army was far larger than
the numbers of troops garrisoned in the town. Numbers ranging from 10,000 to
50,000 may well be exaggerations, but they are always recorded in comparison to
much smaller numbers of French soldiers.18 Also in acceptance of this
16Annales Gandenses, p. 48. Gaucher succeeded Raoul de Nesle as constable after the latter's
death at the battle of Courtrai.
17Chronique de Flandre, I:261; Chronographia regum Francorum, I:128; and Annales
Gandenses, p. 39. See also the Chronique des Pays-Bas, p. 126.
18 The Flemish numbers are recorded as: 10,000 (Chronique Artésienne, p. 59); 30,000 (Jean
Desnouelles, p. 193 and Chronique des Pays-Bas, p. 126); and 50,000 (Chronique Normande, p.
23). Other sources which describe the large size of the Flemish force without indicating a
numerical figure include: Chronique de Flandre, I:261; Chronographia regum Francorum, I:128;
Chronicon comitum Flandriae, p. 172; Rijmkroniek van Vlaenderen, p. 808; and Giovanni
Villani (ed. Muratori), col. 409.
(footnote continued on next page)
Page 27
explanation we must note the numerous accounts of Flemish destruction around St.
Omer in an effort to draw out the French troops from the town.19
Immediately on reaching the outskirts of St. Omer, William of Jülich divided his force
into three parts. The first division, made up of troops from Ypres and recognizable
with their red insignia, was ordered closest to the town. The second, men from St.
Winoksbergen, was ordered behind the Yprois, 'a good league distant' according to the
Annales Gandenses. It also contained the supply train. And the third contingent, the
largest Flemish force, composed primarily of troops from Furnes and Cassel and led
by William himself, was camped in the rear.20 Why William ordered his troops in this
manner rather than keeping everyone together cannot be determined from the sources.
The Yprois contingent met the French first at the small town of Arques. There they
encountered sixty French soldiers, called bidauts in the Chronique Artésienne. The
Chronique Artésienne claims that these were killed, the Annales Gandenses that they
were put to flight.21 At least some of them must have escaped to report the news to St.
Omer, however, as in consequence of this attack, Jacques of Bayonne called his
troops together and prepared to march out of the town to confront the Flemings.22
The approach of the Flemings frightened their French counterparts. Recognizing their
much smaller numbers and undoubtedly remembering the Flemish victory at Courtrai,
the Chronique de Flandre remarks that the French soldiers all 'expected certain death,'
preparing for this by taking absolution from priests inside the town.23 Still, there were
many lords and knights among the French troops in St. Omer, and also many
Flemings who had stayed loyal to the French, known as Leliaerts; they were not
willing to hide behind the walls of a town while relief could be achieved by defeating
the 'lesser' individuals outside.24 They were also encouraged by a short speech given
by Jacques of Bayonne promising them honor and victory.25 The French,
(footnote continued from previous page)
French numbers are recorded as: 1,300 (Guillaume Guiart, 1. 15461); 4,000 (Giovanni Villani
(ed. Muratori), col. 409); and 20,000 (Chronique des Pays-Bas, p. 126).
19 Arques and the suburbs of St. Omer were burned. See Annales Gandenses, p. 40; Chronicon
comitum Flandriae, p. 172; Récits d'un bourgeois de Valenciennes, p. 119; Chronique des Pays-
Bas, p. 126; and Giovanni Villani (ed. Muratori), col. 409. The Annales Gandenses claims,
however, that Arques was burned by French troops fleeing from the Flemings into St. Omer.
20Annales Gandenses, p. 40; Chronique de Flandre, I:262; and Chronographia regum
Francorum, I:129-30. J.F. Verbruggen counts five divisions of Flemings, one from Ypres, two
from St. Winoksbergen, and one each from Cassel and Furnes (De krijgkunst, p. 320). How he
arrived at this number is a mystery, as it appears nowhere in the original sources.
21Chronique Artésienne, p. 60; Annales Gandenses, p. 40; and Guillaume Guiart, p. 249.
22 Guillaume Guiart, p. 243; Chronique de Flandre, I:262; and Chronographia regum
Francorum, I:131.
23Chronique de Flandre, I:262; Chronographia regum Francorum, I:129; and Guillaume Guiart,
P. 243.
24 On the named French knights see Guillaume Guiart, p. 244; Chronique de Flandre, I:262;
Chronographia regum Francorum, I:129; Gilles li Muisit, p. 199; Rijmkroniek van Vlaenderen,
p. 805; and 'Une chronique Valenciennoise inedité,' p. 53.
25 Guillaume Guiart, p. 243.
Page 28
with banners flying and trumpets sounding, marched from St. Omer to do battle with
their enemy.26
The first force they confronted was the same Yprois who had been successful at
Arques. The Yprois seem not to have anticipated this fight, but they had time to form
a solid line, identified as a 'grosse bataille' by the Chronique de Flandre, and this
seems to have kept the French from attacking them. Jacques of Bayonne left his
infantry behind to keep the Yprois from advancing on the town, and his cavalry,
numbering some 800, rode toward the remaining Flemish force.27 The curtain had
fallen, with relatively little violence, on act one.
Act two was very short and much more violent. Only a brief time after their encounter
with the Yprois, the French cavalry rode into the St. Winoksbergen soldiers. This
Flemish contingent was 'unprepared for battle and in loose order,' according to the
Annales Gandenses; nevertheless they fought valiantly, 'like a whole group of
Rolands,' wrote the author of the Chronique de Flandre. Eventually, however, they
were defeated and scattered. The supply train was also attacked and plundered. The
number of Yprois killed was large and included many 'grooms and drivers.'28
Flushed with victory, the French cavalry moved on, and the third act began. By some
unknown means, William of Jülich had learned of the attacks on his forward
contingents, and he now took a defensive posture, dismounting all of his troops and
ordering them to stand against a French cavalry charge. Goededags and pikes were put
forward. This formation was not a line, as the Flemings had formed at Courtrai, but
was, as the Annales Gandenses describes it, 'a circle like a bowl or a crown with
[William] in the center.'29 It was an impenetrable defensive formation which did not
necessitate the use of terrain to be effective.
On this all of the original sources are in agreement. But what happened after William
ordered his formation is in dispute. The Annales Gandenses claims that the French
cavalry, recognizing the futility of attacking such a formation, rode 'cunningly' around
it, looking for a weak spot. Deaths occurred only when a French soldier moved too
closely to the Flemings, or when a Fleming broke from his protective formation. The
annalist describes the scene:
And thus they remained for nearly two hours, the king's mercenaries doing some little harm to
the Flemings and vice versa. For after the battle of Courtrai the French, whether horse or foot,
never ventured to launch their whole army against the Flemings, who always fought on foot, but
always, riding round them if possible,
26 Guillaume Guiart, p. 243.
27Chronique de Flandre, I:262; Chronographia regum Francorum, I:130; and Annales
Gandenses, p. 40. The number 800 comes from the Annales Gandenses.
28Annales Gandenses, p. 40; Chronique de Flandre, I:262-63; Guillaume Guiart, pp. 243-44; and
Chronique Artésienne, p. 60.
29Annales Gandenses, p. 40: 'Hoc audiens Wilhelmus velocius quo potuit suis in adjutorium
venit, vidensque se paucos habere equites et hostes multos, appropinquans eis equos suos reliquit,
et omnem exercitum suum pedes pugnare volens, aciem magnam rotundam ad modum cupe et
corone ordinans, ipseque in medio ejus existens, hostes ad pugnam provocavit.' See also
Guillaume Guiart, p. 246 and Chronique Artésienne, p. 60.
Page 29

Battle of Arques
(according to Other Sources)
attacked them by threes and fours at a time, swiftly withdrawing with their horses, to entice
them away from their own line, so that the others might then by a charge overthrow them.30
The Flemings never did break from their formation, and with the arrival of
reinforcements from the Yprois and St. Winoksbergen divisions, the French fled from
the field to St. Omer. Several met their death in the flight, but most reached St. Omer
in safety.31
The other original sources, including that of eyewitness Guillaume Guiart, differ
considerably from this account. They claim that in coming onto the Flemish
formation, Jacques of Bayonne met with his troops and set his tactics. He divided his
army into four lines: two lines would attack the front of the formation (at least that
part of the formation directly in front of the French) with one line attacking the left
flank and one the right. A small number of soldiers were left in reserve to prevent a
reinforcing assault from the other Flemish contingents.32 It was a 'bad plan' reports
the Chronique Artésienne; despite pushing the Flemings back some distance, the
French cavalry continually failed to penetrate the defensive formation. Charge upon
charge proved futile.33
Finally, after a long battle with many killed on both sides and with night falling,
Jacques of Bayonne sounded the call to retreat, and the fatigued French attempted to
make their way back to St. Omer. Their retreat was orderly, with the Flemings in
fierce pursuit. Several times the French turned as if to renew their attacks against the
30Annales Gandenses, pp. 40-41: 'Sicque stabant fere per duas horas stipendiarii regis
Flandrensibus in modico nocentes et e contrario. Post bellum enim Curtracense numquam
Franci ausi sunt, nec eques nec pedes, Flamingos semper pedes pugnantes insimul cum tota acie
impugnare, sed semper eos circumequitantes quando poterant, per tres vel quatuor eorum
aliquantulum invadebant, cito cum equis ab eis resilientes, ut ipsos de acie sua extraherent, et
tunc alii ipsos cum impetu ad terram prosternerent.'
31Annales Gandenses, p. 41.
32 Guillaume Guiart, p. 245.
33Chronique Artésienne, p. 60. See also Guillaume Guiart, pp. 245-47.
Page 30
Flemings, and on each occasion the Flemings stopped and prepared for the charge.
Even though some in the French leadership wanted to attack their pursuers, no charge
came, but the pattern was repeated five or six times. Ultimately, the French reached the
protective walls of St. Omer.34
The French had lost the battle of Arques, at least they had been forced to retreat from
the battlefield. But to give the Flemings the victory is perhaps too generous, for they
could not attack St. Omer, perhaps because of the losses they had taken in the battle,
nor did they occupy the battlefield for long. Indeed, according to the Chronique de
Flandre, on the day following the battle Jacques of Bayonne sent Aury the German to
the Flemings in an attempt to bury French dead. Aury found no Flemings on the
battlefield; the dead of both sides, some 15,000 according to this author, had been
buried together in one large burial pit.35 The play had ended.
Arques is a difficult battle to analyze, for although there are more than sufficient
numbers of original sources, the fact that they differ on what occurred when the two
main armies met each other has given rise to confusion and misunderstanding. J.F.
Verbruggen, for example, has tried to accomodate both the account of what happened
as recorded in the Annales Gandenses and that reported by the other original
sources.36 This simply cannot be done and begs the question of original source
credibility. The Annales Gandenses has generally been seen as accurate in its account
of the Flemish rebellion of 1302-1305,37 but its details of the battles of Arques and
Mons-en-Pévèle (seen in the following chapter) are often unreliable.
Still, even without discounting the Annales Gandenses narrative, the battle of Arques
ably represents the tactics of infantry warfare that were prevalent everywhere in
Europe during the early fourteenth century. William of Jülich may well have been
presumptious in seeking to fight against the French garrison at St. Omer. At least that
is what the Chronique des Pays-Bas reports the other Flemish leaders accused him of
after the battle.38 Moreover, the division of his force certainly displays a lack of
wisdom that defies the unwritten strategy of 'not dividing one's forces.' But the fact
remains that William's army was large enough to fight against the French even when
so divided.
The Flemish leader also seems to have made no effort to select and prepare the site of
battle, a tactic which had led to success at Courtrai and which would be repeated by
most victorious infantry armies during this half century. Perhaps he was seeking to
find a suitable spot for battle, and that is why he separated his troops into three
divisions. It might also explain why his supply train was not with the largest
34Chronique de Flandre, I:263-64; Chronographia regum Francorum, I:132; and Guillaume
Guiart, p. 247. See also Annales Gandenses, p. 41 on the retreat.
35Chronique de Flandre, I:264 and Chronographia regum Francorum, I:132. Other death tallies
include: 2,000 (Chronique Normande, p. 24); 12,000 (Jean Desnouelles, p. 193 and Récits d'un
bourgeois de Valenciennes, p. 119); 15,000 (Guillaume de Franchet, p. 21); 16,000 Flemings
killed (Chronique Artésienne, p. 60); and 24,000 Flemings killed (Chronique des Pays-Bas, p.
126).
36 Verbruggen, De krijgkunst, pp. 319-25.
37 Verbruggen, Funck-Brentano, and Pirenne all accept the Annales Gandenses without question.
38Chronique des Pays-Bas, p. 126. See also Annales Gandenses, pp. 39-41 and Rijmkroniek van
Vlaenderen, p. 805.
Page 31
contigent of troops at the rear, but instead was with the smallest force, from St.
Winoksbergen, in the middle. However, without more information from the original
sources concerning this, such a conclusion would be weak conjecture at best.
Yet, once in battle and despite being somewhat surprised at encountering their foe, the
Flemings performed the tactics that were necessary to provide victory over an
experienced and noble cavalry army. With the exception of the second contingent,
described as being 'in unprepared and loose order,' which was easily defeated by a
cavalry charge, neither the Yprois division nor the main Flemish force fell victim to
the French attacks. Both of these contingents, the Yprois seemingly containing no
cavalry and the main force dismounting, ordered themselves in a defensive formation
and prepared for a French cavalry assault. This alone seems to have discouraged the
enemy. The French chose not to deliver a charge against the solid Yprois line, norif we
trust the Annales Gandensesagainst the main army's 'bowl or crown.' Even when the
more credible version of the third act of the battle is accepted, that the French cavalry
did indeed charge against William of Jülich's circle, perhaps many times, the defensive
formation remains the significant feature leading to victory. For it could not be
penetrated, and those manning it did not break under the pressure of constant cavalry
onslaught. Ultimately, whether because of fatigue or because of approaching night,
and undoubtedly having suffered numerous casualties from Flemish goededags and
pikes, the French were forced to retreat and leave the battlefield to the Flemings.
Even then, with the French marching away from them, William of Jülich's army did
not veer from their tactical plan. Each time the French turned around as if to mount
another charge against them, some five or six times according to the Chronique de
Flandre, the Flemings reordered their defensive formation and stood ready to fight.
The French could do nothing except continue their retreat.
William may have won the battle of Arques, but he could not take advantage of the
victory. He had lost many men, if only from the huge number of casualties suffered by
the St. Winoksbergen contingent. More importantly, he had not been able to defeat the
French decisively enough to prompt the surrender of St. Omer. Further military
efforts against the town would necessitate a siege, and his army had already shown
that they had no patience for such a military endeavor. So William of Jülich
abandoned the field on which he had achieved such a hard-fought victory and also
marched away. This provoked the harsh but perhaps justified comment by the annalist
from Ghent following his description of the battle of Arques: 'Wherefore it seems that
it was by the just judgment of God that he [William of Jülich] never, after the battle of
Courtrai, was fortunate in any battle or other business.'39
39Annales Gandenses, p. 41: 'unde, ut videtur justo Dei judicio, numquam post bellum
Curtracense in aliquo bello vel negotio bene prosperabatur.' To justify this statement the
Annales Gandenses claims that William had extorted and lavishly spent money, cavorted with
magicians and enchanters, and consulted evil spirits.
Page 32

III
The Battle of Mons-En-Pevele, 1304
After the victory at the battle of Arques, William of Jülich's army again moved against
Tournai. Still the town did not fall, although the Flemings were able to destroy its
suburbs, and even at times to breach the gates.1 From there the army traveled to the
borders of Hainault and began undertaking raids and chevauchées into this
neighboring county.
Philip the Fair could no longer remain inactive. It was clear that his loyal noble, Count
William of Hainault and Holland, needed royal assistance if he were to keep his lands
from the rebels. However, Philip was not yet prepared to send a large army against the
Flemings so he resorted to other, more diplomatic means to stop the Flemish conquest
of Hainault and Holland. Truces were negotiated and peace was sought. The king even
permitted the imprisoned count of Flanders, Guy of Dampierre, to be released so that
he might aid in this peace process.2
Ultimately, these measures failed. The truces were not kept by the rebels and peace
was not attained by Guy of Dampierre who, bound by his chivalric code of honor,
returned to prison. However, the time spent negotiating these peace measures did
allow Philip's taxes to be collected and his army to regroup. By the beginning of 1304,
the French king was ready to attack the Flemish rebels. While the French army, led by
the king himself, marched north to attack William of Jülich's force, the French navy
sailed to Zeeland to unite with the army of Hainault and Holland. It was this combined
northern force in Zeeland which struck the first blow on May 11, 1304 when it
soundly defeated Guy of Namur's army and navy at Zierkzee; Guy was captured and
the Flemish conquest of Holland was halted.3
However, the main Flemish force was still active in the south. Philip the Fair had
moved into the county and was waiting at Mons-en-Pévèle for word on the outcome
1 For contemporary accounts of this siege see Annales Gandenses, p. 49, Gilles li Muisit, p.
198; and Chronographia regum Francorum, I:125-26.
2 See Pirenne, Histoire de Belgique, I:396.
3 The best contemporary French accounts of the battle of Zierkzee are found in Chronographia
regum Francorum, I:148-49; Continuatio chronici of Guillaume de Nangis, I:340; and Etienne
Delcambre, ed., 'Une chronique Valenciennoise inedité,' Bulletin de la commission royale
d'histoire de Belgique 94 (1930), 57. The best Dutch sources are Melis Stoke, pp. 184-89; Jan
Beke, Chronographia Johannis de Beke, ed. H. Bruch ('s Gravenhage, 1973), pp. 257-67; and
Willem Procurator, pp. 70-74. And the best Flemish source for this battle is Annales Gandenses,
pp. 58-63. This battle was fought almost entirely on shipboard; it has been wholly neglected by
modern military or naval historians despite the large number of contemporary sources for it.
Page 33
of the battle of Zierkzee. William of Jülich, desiring a battle with the French king,
advanced on his position. Here he was joined by his uncles, John of Namur and Philip
of Chieti, with their smaller levies. Another uncle, Henry of Namur, would also arrive
at Mons-en-Pévèle with his force, but not until the day of the battle.4 On August 18
the long and bloody battle of Mons-en-Pévèle took place. By the end of the day, the
Flemings had fled from the battlefield, leaving Philip in possession of it. Casualties
were high on both sides; modern historians estimate that the Flemings lost between
seven and eight thousand men including their capable young general, William of
Jülich, and that French losses were nearly as high.5
The defeat at Mons-en-Pévèle marked the end of this Flemish independence
movement, for although Philip the Fair still encountered a large rebel army when he
traveled to Lille, the Flemish morale had been broken, and they feared the destruction
of their own homes and towns.6 The remaining Flemish leaders, led by Philip of
Chieti, readily agreed to a truce and signed the Treaty of Marquette at Lille in
September 1304.7 This time the truce would not be broken, and finally, in June 1305,
after months of continual negotiations, a conclusive treaty of peace was signed at
Athis-sur-Orge.8 The conditions of this treaty were exceptionally harsh and not
entirely conclusive but, for the moment at least, the Flemish independence movement
had been halted.
The battle of Mons-en-Pévèle is one of the most difficult medieval battles to describe.
This is not because of the paucity of contemporary sources on the battle. On the
contrary, there is quite a large number of sources which recount the battlefield events,
including most of the authors who wrote on the battle of Courtrai. Moreover, one of
these sources, Guillaume Guiart, describes the battle from memory, having served
with the French army there. His description thus ought to be the most reliable
contemporary account, and in most cases it can be used with little scrutiny. However,
Guiart's narration does not report one important battlefield event which every other
source recountsa Flemish attack which reached the French king nearly killing him.
This might indicate an account which should be questioned more carefully, but it
probably suggests only that as a relatively unimportant member of the French army,
Guiart was not near the king when the Flemish attack against him was made.
But this is not the only problem that exists among the contemporary sources
recounting the battle. The generally reliable Annales Gandenses, which provides the
most descriptive account of the Flemish rebellion of 1302-05, including perhaps the
most detailed narration of the battle of Mons-en-Pévèle from either the Flemish or
French side, accords the victory not to the French but to the Flemings. This naturally
presents many problems to the modern historian attempting to reconstruct what
4 See Chronographia regum Francorum, I:156.
5 See Verbruggen, Krijgkunst, p. 335 and Funck-Brentano, Philippe le Bel, pp. 476-77.
6 For a description of the siege of Lille see Chronographia regum Francorum, I:162-65.
7 On the treaty of Marquette see Franz Funck-Brentano, 'Le traité de Marquette (Septembre 1304),'
in Melanges Julien Havet (1895; rpt. Geneva, 1972), pp. 749-58.
8 On the treaty of Athis-sur-Orge see Funck-Brentano, Philippe le Bel, pp. 485-560; Strayer, pp.
336-37; and Hans van Werveke, 'Les charges financières de traité d'Athis (1305),' Revue du Nord
22 (1950), 81-93.
Page 34
occurred in the battle; it may also be the cause of the doubt of several prominent
historians, including Henri Pirenne, Henry Stephen Lucas, Ferdinand Lot, and Hans
Delbrück, as to the conclusivity of the French victory there.9 Still there seems to be
little doubt that Philip the Fair's army did defeat the Flemings at Mons-en-Pévèle,
although at a fairly high cost. And, while the report of the battle found in the Annales
Gandenses should be used, the problems which it presents by claiming a Flemish
victory should not be discounted.

All sources agree that it was the French army which first arrived at Mons-en-Pévèle
and set up camp there.10 This appears to have been a conscious decision of Philip the
Fair, and most accounts insist that he chose this place to fight the Flemings. Only the
Annales Gandenses claims otherwise, contending that Philip chose this place not as
one of battle but as one of defence until he learned of the outcome of the northern
battle of Zierkzee.11
All sources also agree on the large size of the French force. Most of these are
impressed with the size of this army using simple phrases such as 'copiosa multitudo'
and 'grand puissance' to describe it, while others actually attempt to tally the numbers
usually counting a total in excess of 100,000 men.12 Modern historians picture a much
smaller force, although one quite sizeable for the time, totaling no more than 3,000
knights and 10,000 infantry.13 It was this large number of knights which was most
impressive to contemporary authors, and also to the Flemings who would face them.
Indeed, Guillaume Guiart describes the French knights as a 'very noble race' which
God had given primarily to defeat the rebellious Flemings, a foreboding which was
undoubtedly felt by the smaller numbered rebels themselves.14
Also present with the French army were a number of artillery pieces, some trebuchets
and mounted crossbows. It is, however, difficult to determine the number, size or
capabilities of these machines from the sources reporting their presence. The
Chronique Artésienne, for example, reports that five 'engiens' plus some 'espringales'
were at Mons-en-Pévèle, while Guillaume Guiart mentions the presence of
9 Pirenne, Histoire de Belgique, I:396; Lucas, The Low Countries, p. 42; Lot, L'art militaire,
I:268; and Hans Delbrück, History of the Art of War Within the Framework of Political
History, trans. W.J. Renfroe, Jr. (Westport, Conn., 1984), III:540. While these authors never
indicate the problems presented by the Annales Gandenses in studying what occurred at Mons-
en-Pévèle, all use this source in their narrative of the battle. Other historians, namely
Verbruggen (Krijgkunst, pp. 323-35), Funck-Brentano (Philippe le Bel, pp. 474-75), and Henri
Nowé (p. 91), all have no difficulty in recognizing that this was a French victory; each also
uses the Annales Gandenses as a source.
10 See Chronographia regum Francorum, I:154; Chronicon comitum Francorum, p. 173; Gilles
li Muisit, p. 201; Chronique de Flandre, I:495; Continuatio chronici of Guillaume de Nangis,
I:343; and Jean de Paris, p. 643.
11Annales Gandenses, p. 57.
12 For the estimate of 100,000 men see Chronique Normande, p. 25; Rijmkroniek van
Vlaenderen, p. 809; and Jean Desnouelles, p. 194. For other comments on the size of the French
force see Chronicon comitum Flandriae, p. 173; Giovanni Villani (ed. Muratori), col. 413;
Ottokar von Stiermarken, p. 1001; Grandes chroniques, VIII:239; Ives, p. 204; Jean de Paris, p.
643; and Continuatio chronici of Girard de Franchet, p. 24.
13 See Verbruggen, Krijgkunst, p. 324.
14 Guillaume Guiart, pp. 290-91.
Page 35
only three 'perdriaus' and two 'espringales.' These sources agree, however, that despite
some initial damage caused by these weapons, in the long run they proved to be an
inconclusive factor in the battle, the Flemings having destroyed them long before the
actual fighting began.15
The French army was led by Philip himself who handled his military leadership as if
he were leading a Crusade. He had prepared for the campaign by offering prayers for
victory in the most sacred of French churches, the church at Saint-Denis, where he
had also obtained the oriflamme, a banner of almost relic reputation which was only
to be used in fighting against heretics or heathens.16 Assisting Philip in his command
were his constable and two marshals, the most important military leaders in the
kingdom. Philip had also requested the aid and leadership of William of Hainault and
Holland, but the count had declined because of ill health.17
The Flemish army arrived at Mons-en-Pévèle later than the French and set up their
camp opposite to them, but within their sight.18 Their numbers are reported to be
larger than the French, with calculations ranging as high as 200,000;19 modern
historians claim a smaller, although equally impressive, 12,500 to 15,000 tally.20 All
the Flemish soldiers were on foot, armed with godedags and lances.21
The morale of the Flemings was high. They desired to fight the French king, and they
believed that they could defeat him. They were, according to the bourgeois of
Valenciennes, inspired by their need to 'defend themselves and their heritage.'22
Moreover, the Annales Gandenses reports that when they heard of the defeat of their
companions at Zierkzee they were 'for the most part . . . stimulated rather than
dejected' by the news, 'since they believed they would get their own back for Guy [of
Namur]'s loss from the king and his army.'23
The Flemings set up their line opposite the French. This made the opposing lines,
again according to the Annales Gandenses, 'so close that the crossbowmen of one
could have shot at the other.'24 The Flemings were ordered in a defensive formation
15Chronique Artésienne, pp. 85-86 and Guillaume Guiart, pp. 288, 291-93. See also
Chronicon comitum Flandriae, pp. 173-74.
16 See Chronographia regum Francorum, I:150 and Chronique de Flandre, I:493. On the
oriflamme see Philippe Contamine, L'oriflamme de Saint-Denis aux XIVe et XVe siècles (Nancy,
1975).
17 The constable and marshals are only mentioned before the battle (see Chronographia regum
Francorum, I:152-53); they are not mentioned at Mons-en-Pévèle although they surely would have
been there. On the request for aid from William of Hainault and Holland and his illness keeping
him away from the battle see Chronographia regum Francorum, I:151-52.
18 On the Flemish arrival at Mons-en-Pévèle see Chronographia regum Francorum, I:154;
Chronique Artésienne, p. 84; Récits d'un bourgeois de Valenciennes, p. 120; and Chronique de
Flandre, I:495.
19 On the estimate of 200,000 see Jean Desnouelles, p. 194 and Chronique Normande, p. 25.
Rijmkroniek van Vlaenderen estimates a smaller, although still outrageous, 160,000 size force.
20 See Verbruggen, Krijgkunst, p. 324.
21 See Guillaume Guiart, pp. 291-92.
22Récits d'un bourgeois de Valenciennes, p. 121. See also Chronographia regum Francorum,
I:154; Chronique de Flandre, I:276; and Ancienne chronique de Flandre, p. 394.
23Annales Gandenses, p. 65: 'Sed Flandrenses, rumore predicto non concussi pro majori parte,
sed magis exacerbati, credentes se damnum Guidonis et sperantes a rege et suo exercitu
recuperaturos.'
24Annales Gandenses, pp. 63-64: 'Cumque duo hostiles exercitus sic ordinati in tantum essent
propinqui, quod balistarii unius exercitus in oppositum sua spicula poterant jaculari . . . '
Page 36

Battle of Mons-en-Pévèle
(Initial Battle Formation)
similar to that which had been successful at Courtrai, but with the addition of a quasi-
fortification made with their wagons and carts at their rear and on their flanks. The
Annales Gandenses describes this fortification:
. . . some Flemings detailed for the purpose sent away all the drivers with the horses and mares
drawing the wagons, and made a sort of fortification out of the wagons in their rear of their own
line, connecting one with another, but taking off one wheel from each, so that the French, if they
wished to encircle them, should not be able to attack them from the rear.25
Jean de Paris provides a more 'wild west' description of this defensive formation
describing a 'fortress' of carts and wagons which were circled 'so that they [the
Flemings], in necessity, might be able to make a stand and fight inside such a circle.'26
There also seems to have been some marshes and ditches separating the two lines.
These may have been simply natural hindrances as there is no mention of their
construction for the battle. However, that the Flemings knew of their existence and
planned their formation accordingly is attested to by Guillaume Guiart who reports
25Annales Gandenses, p. 66: par ' . . . aliqui Flandrenses ad hoc ordinati, dimissis omnibus
aurigis cum equis et equabus trahentibus currus, de ipsis curribus a tergo aciei sue quasi
quamdam munitionem unum alteri connectendo composuerunt, a quolibet una rota ablata, ne
Franci, si eos voluissent circumequitare, ipsos possent a tergo invadere.'
26 Jean de Paris, p. 643: 'Flandrenses autem astute fecerant quasi munitionem de suis curribus et
quadrigis, magnum terrae spatium hujus modi vehiculis circumdantes, ut infra talem circuitum
possent in necessitate consistere et pugnare.' See also Chronographia regum Francorum, I:155-
56 and Chronique Artésienne, p. 84.
Page 37
that the Flemish leaders told their soldiers that the ditches would save them as they
had at Courtrai. In fact, the marshes and ditches had little effect on the outcome of the
battle.27
On August 13, with a flurry of crossbow fire, the Flemish troops tried to provoke the
French to do battle.28 The French armed themselves and assembled for a charge on
the Flemish line.29 Whether a charge actually occurred at this time however is
questionable, as only the Annales Gandenses claims that it took place. Even then, the
anonymous author of this chronicle contends that it was stopped before actually
encountering the Flemish infantry.30 Other writers report only skirmishingcross-bow
fire, artillery fire and small raidsbetween the two sides at this time. According to the
Chronicon comitum Flandriae, the French hoped that this skirmishing would fatigue
the Flemings and cause them to withdraw from the field.31
This was not to be the result; nor did battle ensue at this time. Instead, fighting was
halted so that a peace parliament could be called to discuss the rebellion without
further bloodshed. This is a difficult scenario to investigate because there are several
differing stories about this peace council. It does seem certain that such a parliament
took place on the battlefield and that the battle was stopped until the question of peace
was resolved as so many sources report it, but these sources differ on who asked for
peace and on whom the blame should rest for peace not being declared.
On the one hand, there is a tradition among contemporary writers that it was the
Flemings who desired peace at Mons-en-Pévèle. Both the author of the Chronique
Normande and Jean Desnouelles relate a story that has the Flemings beginning to lose
the battle and asking the count of Savoy, whose friendship they had recognized in the
past, for assistance in obtaining a peace accord to stop further fighting. This is a
request which the count was more than willing to grant if the Flemings promised 'to
give and ordain one hundred chapels where masses might be directed for the French
who were slain at Bruges and elsewhere.'32 The Flemings readily agreed to this offer,
but peace was thwarted when Philip refused to accept these conditions and chose
instead to pursue the engagement.
Other sources claim that it was the French who desired peace on this day and that this
was denied them by the Flemings. Guillaume Guiart, for example, claims that the
hunger, thirst and filth of the battlefield began to demoralize the French troops who
thus were encouraged by the talk of peace.33 Some authors also comment on the
intense heat of the noonday sun which fatigued the French soldiers causing them to
disarm and disorganize. Indeed, Jean de Paris claims that the heat and thirst endured
27 Guillaume Guiart, p. 291. See also Annales Gandenses, p. 70.
28 See Annales Gandenses, pp. 63-64.
29 See Ancienne chronique de Flandre, p. 394.
30Annales Gandenses, pp. 66-67.
31Chronicon comitum Flandriae, pp. 173-74. See also Guillaume Guiart, p. 286 and Chronique
Artésienne, pp. 85-86.
32Chronique Normande, p. 25: ' . . . il leur voulsist pardonner leur meffait pour donner et
ordonner cent chappelles, ou on diroit messes pour les Francois, qui avoient este occis a Bruges
et allieurs.' See Jean Desnouelles, p. 194 and Chronographia regum Francorum, I:155.
33 Guillaume Guiart, p. 294.
Page 38
by the French was so bad that 'many died in camp from their great thirst, including the
count of Auxerre and others.'34 All of these authors generally blame the Flemings for
refusing to grant peace to the French at Mons-en-Pévèle, claiming instead that the
rebels used this opportunity to treacherously attack the demoralized enemy.35
A final variant on this story is found in the Annales Gandenses. Here it is reported that
it was the French king who pushed for peace, but that the Flemings 'who were always
desirous of peace' also welcomed a conference which might end the battle. However,
the call to peace was merely a ruse used by Philip IV to reform his army. The
Ghentenaar annalist writes:
But while negotiations as to peace were going on between the parties, and orders had been
given in both hosts that all should remain inactive and do each other no injury, the king, who
had large numbers of horse, sent one company of horse, with many foot, against the left flank of
the Flemish army, and another against the right, as though to attack the Flemings from the rear.
The Flemings, 'perceiving the deception,' thus withdrew from the peace negotiations
and returned to their line. In the process, a knight, whom the Annales Gandenses
reports bore the 'cognizances' of the count of Savoy, was calling for a truce; but, he
was killed by the Flemings, 'not wanting to hear any more about peace.'36
On August 18, after the failure of the peace parliament, the battle resumed. Unlike the
skirmishing five days earlier, contemporary writers do not report who initiated this
new phase of fighting. It may have been a French initiative, however, as they opened
the battle by charging into the Flemish line in a fierce cavalry charge imitative of that
attempted at Courtrai. Fighting became so vicious during this phase of the battle that
Guillaume Guiart, who may have participated during these French attacks of the
Flemish line, sums it up simply: 'death and blood dwelt there.'37
The 'horrible and cruel' battle continued into the midday.38 The French fought well,
but the Flemings held their position; both sides took heavy casualties. Soon it became
apparent that the Flemings were beginning to win. Panic ran through the
34 Jean de Paris, p. 643: 'Franci namque ex calore et siti plurimum fuerunt tunc gravati, ita
quod plures ex maxima siti in campo obierunt, ut comes Antissidorensis aliique plures.' See
also Jean Desnouelles, p. 194.
35 See Ives, p. 204; Continuatio chronici of Guillaume de Nangis, I:343; and Jean de Hocsem, p.
121.
36Annales Gandenses, pp. 67-68: ' . . . ex parte regis petitum est cum maximo dolo, ut pace
tractatus haberetur. Flandrenses igitur, qui semper pacem desiderabant, in tractatum consenserunt.
Sed dum de pace inter partes tractaretur, et in utraque acie esset proclamatum, quod omnes starent
in quiete, et unus alteri non noceret, rex, qui in equitibus abundabat, unam aciem equitum cum
multis peditibus misit versus sinistrum latus exercitus Flandrensis et aliam versus dextram, quasi
per eos Flandrensibus a tergo noceret. Qui dolum cognoscentes, ulterius in quiete stare nolentes,
sicut prius pugnare ceperunt. Venit autem quidam eques de exercitu regis cum fortissimo equo et
optime armatus, signis bellicis comitis ornatus Sabaudie, clamans dolose, ''Pax! Pax!" quem
Flandrenses, dictum comitem esse credentes, occiderunt, nihil audire volentes ulterius de pace.'
37 Guillaume Guiart, p. 293: 'Mort et sanglent ileuc demeure.' (Guiart's entire description of this
phase of the battle appears on pp. 291-93.) See also Chronographia regum Francorum, I:156 and
Ives, p. 204.
38 See Ancienne chronique de Flandre, p. 394.
Page 39

Battle of Mons-en-Pévèle
(French Attack)
French army, and several began to flee from the battlefield. The Chronographia
regum Francorum describes the scene:
A marvelous battle was fought on both sides. And when they had fought most fervently for a
long time, the Flemings prevailed so strongly and enthusiastically that the French retreated and
left the field, as if they were semi-conquered.39
Something had to be done to prevent a Flemish victory. The French leadership chose
to attack the flanks of the Flemish line, hoping that such a maneuver might divert
attention from the frontal assaults and perhaps break the Flemish defensive formation.
As such, the attack on the flanks proved somewhat successful reaching the Flemish
wagon fortification and sacking their camp. An attempt was also made to disassemble
the wagons to allow for a rear assault on the Flemish line. But, the Flemings countered
this by sending some troops against the French ultimately chasing them from the camp
and preventing further damage to their fortification.40 The infantry line continued to
hold its position.
The length of the battle and the August heat began to take its toll on both armies.
39Chronographia regum Francorum, I:157: 'Factaque est pugna mirabilis ab utraque parte;
que cum ferventissima jamdiu durasset, prevaluerunt Flamingi ita fortiter et animose quod
Gallici jam se retrahebant et per campos dividebantur, quasi semivicti.' See also Annales
Gandenses, p. 67; Ottokar von Stiermarken, p. 1001; Gilles li Muisit, p. 201; Chronique de
Flandre, I:495-96; and Chronique anonymé Française, p. 136. Jean de Paris (p. 643) does not
agree with these other accounts, claiming instead that the Flemings lost this phase of the battle.
40Annales Gandenses, pp. 68-69 and Ottokar von Stiermarken, p. 1002. The Annales Gandenses
contends that this attack was by both French cavalry and infantry.
Page 40
Fatigue and thirst plagued all involved in the fighting, but the French seem to have
been especially affected by it. They wore heavier armor than did their opponents, and
they had been faced with the demoralization of constant defeat. Jean Desnouelles
reports that this led to disorder among the French army, many of whom were trying to
remove their armor to prevent heat prostration.41 This did not work, however, as the
Annales Gandenses notes that many died without being wounded, suffocating
'through the weight of their armor and the summer heat.'42
As the battle had turned to their side, some of the Flemish leaders, including John and
Henry of Namur, retreated to Lille with their tired troops, leaving William of Jülich,
Robert of Namur and Philip of Chieti to continue to fight the French.43 The remaining
Flemish leaders then met in council to plan how to end the battle; they decided to
make a conclusive attack with their infantry line on the French camp hoping to drive
their enemy from the field.44
The Flemish attack was brutal and initially successful. The discouraged French troops
still facing the infantry line were taken by surprise and fled to the rear. Soon the
Flemings were among the French camp, and they began to threaten Philip the Fair.
More contemporary chronicles comment on this attack, and the subsequent French
recovery, than any other part of the baffle at Mons-en-Pévèle. It seems to both the
French and Flemish authors that this was the deciding point of the battle, a point when
the success of the French army against the Flemings was determined.45 The Flemings
must have felt the same, for the sole purpose of this battlefield venture, according to
the Chronique Artésienne, was to try and reach the French king and possibly to kill
him, thereby gaining victory.46
The contemporary chronicles contend that the Flemish attack on the French camp
nearly succeeded in its goal of killing Philip. Ives reports that the Flemings slew all of
the king's bodyguard save two,47 and the Chronicon anonyme regum Franciae claims
that only one man separated Philip from the Flemings and that this knight as
decapitated assisting the king onto his horse.48 Several chroniclers also mention the
destruction of the oriflamme in this attack.49
Most descriptive in its narrative of this scene is the Annales Gandenses which
41 Jean Desnouelles, p. 194. See also Guillaume Guiart, p. 294 and Jean de Paris, p. 643.
42Annales Gandenses, pp. 69-70.
43 See Annales Gandenses, p. 70.
44 See Annales Gandenses, p. 70 and Chronique Artésienne, p. 86.
45 See Guillaume Guiart, p. 296; Annales Gandenses, pp. 71-72; Chronographia regum
Francorum, I:157-58; Chronique Artésienne, pp. 86-87; Chronicon comitum Flandriae, p. 174;
Jean de Paris, p. 643; Jean Desnouelles, p. 194; Récits d'un bourgeois de Valenciennes, p. 121;
Ives, pp. 204-05; Gilles li Muisit, p. 201; Ancienne chronique de Flandre, p. 394; Chronique
anonymé Française, p. 136; Continuatio chronici of Guillaume de Nangis, I:344; Chronique
Normande, p. 26; Grandes Chroniques, VIII:241; Chronique de Saint-Denis, p. 678; Geoffroi de
Paris, p. 117; Rijmkroniek van Vlaenderen, p. 809; Giovanni Villani (ed. Muratori), col. 415;
Ottokar von Stiermarken, pp. 1001-03; Jean de Winterthur, p. 32; and Chronicon anonyme regum
Franciae, p. 18.
46Chronique Artésienne, p. 86.
47 Ives, p. 204.
48Chronicon anonyme regum Franciae, p. 18.
49Chronographia regum Francorum, I:161; Ancienne chronique de Flandre, p. 395; and
Chronique anonymé Française, p. 136.
Page 41

Battle of Mons-en-Pévèle
(Flemish Counter-attack)
asserts that Philip was fiercely attacked, but defended himself nobly and bravely until
his horse was killed under him. However, the Flemings were unable to recognize the
fallen king and thus his life was spared. The anonymous Ghentenaar annalist writes:
[The Flemings] did not know how to distinguish him from the others lying on the ground, for his
bodyguard, as a precaution in the peril of the conflict, had removed his surcoat, adorned with
the sign of the lilies, so that he should not be recognized by the enemy, who would have more
gladly killed him than anyone else.50
The Annales Gandenses continues the story by maintaining that finally a group of
Philip's fleeing knights came upon him lying stunned on the ground and helped him to
mount a horse. But Philip had difficulty guiding his steed which was attacked and
wounded by a Flemish soldier. Fortunately, the horse led the stunned king back
towards his own troops and he was saved.51
Philip the Fair soon recovered, and provoked by the attack which had come close to
killing him, he rallied his army and led a counterattack against the Flemings. The
Chronographia regum Francorum has left a compelling description of Philip at this
time:
50Annales Gandenses, p. 71: 'Nesciens ipsum inter alios prostratos discernere, eo quod
custodes predicti in periculo conflictus pre timore superumicale suum bellicum, suo signo,
scilicet liliorum, decoratum, ne ab hostibus, qui libentius ipsum quam aliquem alium
occidissent, agnosceretur, abruperant.'
51Annales Gandenses, p. 71.
Page 42
The king was nobly armed with his regal arms, and he sat upon his war horse holding a metal
mace in his hands . . . and finally that most ferocious lion entered into the middle of the battle
killing here and there and slaying all whom he touched with his mace. For he held himself so
powerfully then that by his own bodily strength and no other were the Flemings suppressed.52

The Chronique de St. Denis adds that by doing this he was able to demonstrate how
'very confident and brave' he was to his enemies, striking fear among them and
causing them to flee.53
It was Philip's ability to rally his fleeing army against the Flemish attackers which led
to the French victory. Jean de Paris writes:
And when most of the king's army had fled terrified from the sudden invasion and confusion of
the Flemish army, the king whom we ought to praise very highly in this, taking on an air of
constancy and bravery, remained with few men, and he gave heart and courage to those fleeing
and trembling with fear, that returning they obtained a pleasant victory from the Flemings that
very evening and night.54
This counterattack fought by both French knights and infantry caught a disordered
Flemish army unprepared for such an assault. After only a small amount of defense
they turned and fled in rout;55 the French troops pursued the fleeing Flemings well
into the night.56
It is this last part of the battle with which the Annales Gandenses differs in its story. It
contends that rather than returning to the battle and rallying his troops to victory,
Philip was 'forced and compelled' by his soldiers to take flight, despite not wishing to
leave the battlefield. His army fled with him. The Flemings then plundered the French
camp before returning to the battlefield to celebrate their victory. Finally, these
Flemings joined their comrades at Lille. But, this was not because they had lost the
battle, simply because they had run out of food. The annalist writes:
From a high and elevated position, where they [the Flemings] united to rest for a short time,
they watched with joy the French returning with their troops and companies, their torches and
great wax tapers lighted, seeking in great sorrow their noble dead, slain, crushed, or choked in
the pits and ditches, and showing no
52Chronographia regum Francorum, I:157-58: 'Rex veto nobiliter indutus suis armis
regalibus et sedens super dextrarium suum tenesque machiam ferream in manibus suis . . . et
tanquam leo ferocissimus intravit medium belli, percutiens hinc et inde et prosterens
quoscumque attingebat de machia. Nam ita valide tunc se habuit, quod propria strenuitate
corporis sui et non alterius, Flamingi refutati sunt.'
53Chronique de Saint-Denis, p. 678.
54 Jean de Paris, p. 643: 'Cumque plurimi de regis exercitu ex invasione subita et tumultu
Flandrensis exercitus perterriti fugerunt, rex, in hoc laudandus quamplurimum, constantiam et
audiciam assumens, remansit cum paucis, deditque cor et audiciam tantam fugientibus ac
trepidantibus prae timore, quod revertentes satis pulchram victoriam eadem vespere et nocte de
Flandrensibus habuerunt.'
55 See Chronographia regum Francorum, I:158.
56 On the pursuit of the Flemings see Chronographia regum Francorum, I:159-60; Chronique
Artésienne, p. 87; Ancienne chronique de Flandre, p. 395; Chronique de Flandre, I:496; and
Ottokar von Stiermarken, p. 1003.
Page 43
desire for further battle, although the moon rose in clear splendor. On the said hill, however,
they could not find victuals sufficient for their number, or coverings, or tents to shelter them. So
after consultation they went away towards Lille, not as vanquished or in flight, for there was
none to pursue them, but driven by hunger and necessity.57

But the Annales Gandenses is alone in its assertion of Flemish victory; all other
contemporary sources report a French victory.
Philip the Fair greeted the victory with much thanksgiving. He had subdued the
Flemings, restoring peace to his torn kingdom, and he responded, according to the
Chronicon anonyme regum Franciae, by giving a gift of one hundred l.p. to those
knights who stayed with him when it seemed that the Flemings would defeat them; to
the son of the knight who was slain helping him mount his horse, the king gave five
hundred l.p.58 Philip also celebrated the victory by giving gifts to several important
churches in his kingdom.59
But Philip also grieved over those men lost on the battlefield. Both sides had suffered
heavy losses at the battle of Mons-en-Pévèle. Estimates of Flemish dead range from a
low of 4,000 noted in the Annales Gandenses to a high of 70,000 found in the
Chronicon anonyme regum Franciae. The French may have lost as many or perhaps
more, although the only tally given by contemporary sources is that of 9,000 found in
the Annales Gandenses.60
Some of the most interesting anecdotes of the battle concern the deaths of two of the
prominent Flemish leaders: Pieter de Coninck, the man responsible for the Bruges
Matins massacre which had prompted the rebellion in 1302, and William of Jülich, the
leader of the Flemish army and the man chiefly credited with the strategy which
57Annales Gandenses, p. 73: 'Ex quo loco, cum in eo aliquantulum more quiescentes
contraherent, alto et eminenti viderunt cum gaudio Francos cum turmis et cuneis reversos,
accensis faculis et cereis suis magnis torcis, cum maxima mestitudine mortuos suos nobiles
occisos vel oppressos, suffocatos per puteos et fossata requirere, nec eos, quamvis luna
ascenderet splendendo clare, amplius velle de bello curare. In dicto itaque monte, quasi nihil
victualium quod multitudini eorum sufficeret, nec cooperimenta, nec tentoria aliqua, quibus
tegerentur, invenientes, non ut victi vel fugati, quia nullus eos insequebatur, versus Insulam inito
consilio abierunt, sed inedia et necessitate coacti.'
58Chronicon anonyme regum Franciae, p. 18.
59 While besieging Lille, a few days after his victory at Mons-en-Pévèle, Philip gave an annuity
of one hundred l.p. to the church of St. Denis in commemoration of the victory in Flanders; he also
promised an equestrian statue of himself for the Notre Dame Cathedral in Paris. These gifts are
enumerated and pledged by Philip in a letter written at the siege of Lille in September 1304.
Copies of this letter can be found in Gallia christiana (Paris, 1744), VIII:374-75; Michel Felbien,
Histoire de l'abbaye royale de Saint-Denis en France (1707; rpt. Paris, 1973), p. cxxx; and
Caesar Egassio Du Boulay, Historia universitatis Parisiensis, (1668; rpt. Frankfurt, 1966), IV:71.
Philip made similar offerings in commemoration of the victory to two churches, Mont-Saint-
Michel and the Notre Dame de Boulogne, which he visited on his return trip home to Paris. See
Chronique anonymé Française, p. 136.
60Annales Gandenses, p. 75; and Chronicon anonyme regum Franciae, p. 18. See also
Chronique Artésienne, p. 87; Jean de Paris, p. 643; Chronicon comitum Flandriae, p. 174;
Ancienne chronique de Flandre, p. 394; Récits d'un bourgeois de Valenciennes, p. 122; Jean
Desnouelles, p. 195; Continuatio chronici of Guillaume de Nangis, I:344-45; Chronique de
Saint-Denis, pp. 678-79; Chronique anonymé Française, p. 136; Rijmkroniek van Vlaenderen, p.
810; and Chronique Tournaisienne, in Chronique Artésienne et Tournaisienne, ed. F. Funck-
Brentano (Paris, 1898), p. 87 n1.
Page 44
resulted in the Flemish victory at Courtrai. Concerning the death of Pieter de Coninck,
the Chronographia regum Francorum records that, after the battle was won and the
Flemings had left the field, a riderless Flemish horse was found 'adorned by the
wooden image of St. George.' Said to be the horse of Pieter de Coninck, it was
concluded that he had made this image in an effort to gain the intercession of God.
That this horse was without its rider, who was presumed to be slain, was a message to
the French that God had been with them and not with the Flemings.61
More colorful stories accompany the death of William of Jülich. Almost all
contemporary sources report that William was killed during the last stages of this
battle, trying to rally his fleeing troops, and that it was important ultimately in the
defeat of the Flemings; but by whom or how William was killed is contested by the
contemporary authors. The Récits d'un bourgeois de Valenciennes claims that the
credit for William's death ought to be given to Charles of Valois, the king's brother,62
while three other chronicles, Jean Desnouelles' Chronicon, the Chronique Normande,
and the Chronique Tournaisienne, claim that William was killed by Renaud of
Dammartin, the count of Boulogne, out of vengeance for the death of his father at
Courtrai.63 The Chronographia regum Francorum agrees with these latter
conclusions on who was responsible for the death of the Flemish leader, but its
account is very different from the other sources:
Meanwhile, William of Jülich . . . seeing that he was unable to flee, constricted not a little by
the heat and thirst, unshod his feet, and also those with him taking up their sword points placed
the pommels into their mouths so that they might extinguish their thirst, thus awaiting death.
Immediately the count of Boulogne . . . attacked them and cut off the head of the aforementioned
William, count of Jülich, and the others perished by the sword.64
As well, three sources, the Chronographia regum Francorum, the Récits d'un
bourgeois de Valenciennes and the Ancienne chronique de Flandre, note that after the
battle a man, described as a 'client' by the Chronographia, delivered the head of
William of Jülich to Philip's tent, but that the French king 'did not care to receive' this
prize.65
Again it is the Annales Gandenses which differs from the other contemporary sources
on this story. For one thing, the annalist is certain that William's death came honorably
on the battlefield: 'when pursuing the enemy [William of Jülich] either perished by
suffocation being crushed after falling (for he was of slender build,
61Chronographia regum Francorum, I:161.
62Récits d'un bourgeois de Valenciennes, p. 122.
63 Jean Desnouelles, p. 194; Chronique Normande, p. 26; and Chronique Tournaisienne, p.
87n1.
64Chronographia regum Francorum, I:158-59: 'Interea Guillelmus . . . videns quod evadere non
valebat, nimium constrictus calore et siti, discaltiavit pedes suos; qui quidem et qui cum eo erant,
sumptis mucronibus suis, ponebant pomalia in oribus suis ad extinguendam sitim, sic mortem
expectantes. Porro comes Bolonie, qui prius perceperat quod evadere vellent et ob hoc viam, qua
possent fugere, clauserat, percipiens eorum dolorem, aggressus est eos abscinditque caput prefati
Guillelmi, comitis Juliocensis: ceteri vero gladiis intelierunt.'
65Chronographia regum Francorum, I:161; Récits d'un bourgeois de Valenciennes, p. 122; and
Ancienne chrunique de Flandre, p. 395.
Page 45
courageous though he was), or, as the French assert, when in pursuit of them with a
company of about eighty men.'66 At the same time, he dispels some of the folklore
which had arisen around the death of this Flemish leader:
And because afterwards no certain trace of his body or his arms could be found, either by the
French or the Flemings, the common Flemish folk for a long time afterwards said that he had
been spirited away by the magic to which he was devoted, and later, at a fitting time, when they
were in great military peril, would return. But these are frivolous and fabulous tales. For it is
certain that he perished that day, even though no trace of his body or his arms could be found
among the slain or smothered. The same happened with regard to many nobles at Courtrai. It is
said that a certain most wicked enchanter of his household, who was with him, promising a
certain magic incantation which he had learned for himself that he could become invisible to the
enemy and others whenever he wished or found it necessary. But the said incantation availed
him nothing, so far as saving his life went. I do not know if it was of use in concealing his body;
for it is easy for demons to hide some dead body.67
The Annales Gandenses also dispels the notion that William's head had been delivered
to the French king after the battle, contending instead that the head was that of 'a
certain chaplain of Ghent, rather like him.'68
As can be seen from a narrative of the battle of Mons-en-Pévèle, this was not, as
Charles Oman insists, the vindication of fourteenth-century French cavalry warfare.69
Indeed, it was not until the end of the battle, after the failure of the Flemish attack on
the French camp, that the French knights played any role in realizing the victory. Even
then, at least according to the author of the Chronique de Saint-Denis, it was not the
cavalry alone which defeated the Flemings, but a combination of cavalry and infantry,
fighting together, which crushed the attack on the camp.70
All else points to a distinct infantry superiority on the battlefield of Mons-en-Pévèle.
Despite not arriving first on the field, the Flemings ordered themselves precisely as
they had at Courtrai: in a single line of infantry formed as a defensive unit prepared to
meet the onslaught of French knights. And while they were unable
66Annales Gandenses, pp. 73-74: 'Inter quos fuit precipuus Wilhelmus Juliacensis, qui
insequens hostes vel oppressionis et casus suffocatione interiit (erat enim tenuis complexionis,
licet esset animosus), vel, ut Franci asserunt, eos insequens, cum parvo cuneo circiter lxxx
virorum.'
67Annales Gandenses, p. 74: 'Et quia postea nullum indicium certum corporis ejus vel armorum a
Francis nec a Flandrensibus poterat inveniri, vulgares Flamingi longo postea tempore arte magica,
cui inserviebat, ipsum affirmabant ablatum, et in posterum, tempore suo competenti, quando ipsi in
majori essent belli periculo, rediturum. Sed hec frivola eunt et fabulosa. Certum est enim quod
isto die periit, quamvis indicia corporis vel armorum ejus inter multos occisos vel oppressos non
poterant inveniri; hoc enim de multis nobilibus accidit in Curtraco. Dicitur quidam pessimus
incantator de familia sua mala (et tunc cum eo erat) ipsum decepisse, quia promiserat sibi, quod
quadam magica incantatione, quam sibi didicerat, quandoque vellet et indigeret, hostibus et
quibuscunque aliis invisibilis fieret. Sed dicta incantatio nihil sibi hic profuit, quantum ad vite
conservationem; facile enim est demonibus corpus aliquod admortuum occultare.'
68Annales Gandenses, pp. 74-75: ' . . . numquam fuisse Wilhelmi, sed cujusdam capellani de
Gandavo sibi aliquantulum similis, qui in bello corruit et quem ego bene novi.'
69 Sir Charles Oman (1905), II:118.
70Chronique de Saint-Denis, p. 678.
Page 46
to prepare the battlefield by digging ditches as they had done at Courtrai, they were
well aware of the natural hindrances which lay between themselves and the
Frenchmarshes and perhaps irrigation ditchesand, according to Guillaume Guiart, they
planned their formation in order to take advantage of them. Furthermore, with the
frequently mentioned construction of the wagon-fortress, they prepared the field
behind and to the sides of their line in order to protect themselves from attacks in their
rear and on their flanks. This fortification also prevented easy flight if the Flemings
began to lose.
The French, too, may have been aware of the vulnerability of their knights in a charge
against this well prepared Flemish infantry line. Perhaps this is why the Chronicon
comitum Flandriae contends that the French initially only skirmished, hoping that by
this the Flemings would grow fatigued and give up the field without necessitating a
charge by the French knights. This also could be the reason behind Philip's use of
artillery on the battlefield, as this was uncommon in the early fourteenth century. If he
could have broken the Flemish line with the fire from these weapons, he could then
have countered with an effective cavalry charge. But, the life of the artillery at the
battle was short-lived, for little is mentioned further about it except that it was easily
destroyed by the Flemings in the skirmishes which preceeded the August 18 fighting.
Finally, if the Annales Gandenses is to be believed, the French did try a 'feigned'
cavalry charge in the first phase of the battle which stopped short of hitting the
Flemish line. Perhaps this was also an attempt by Philip to see if the infantry
formation would break; however, when it did not, another tactic was trieda peace
parliament.
If this was the case, it gives credence to the story told by several of the contemporary
writers, including the eyewitness Guiart, that the peace parliament was summoned by
the French. Perhaps it also gives credence to the rather cynical statement of the
anonymous author of the Annales Gandenses that the peace conference was only a
'crafty' artifice used by the French king to hide the reformation of his army.
If it occurred, the reformation must not have worked, for when battle began again,
after the failure of the peace parliament, the French returned to the strategy which had
met with failure at both Courtrai and Arques: they charged on horseback into the
solidly formed Flemish infantry line. And, as at both of those battles, the charge was
entirely ineffective. The infantry line held, the horses stopped, and the cavalry began
to lose. The French attempted cavalry charge after cavalry charge, but they neither
broke the Flemish line nor disrupted it sufficiently to gain even the slightest victory.
Thus, the French were forced to alter their battle plans. A contingent of horse and foot
was even sent against the flanks of the Flemings which pierced their wagon
fortification. But, they too were incapable of disrupting the Flemish line and were
easily put to flight by a counterattack.
Furthermore, although there were heavy casualties on both sides, it was not the
Flemings who tired first. The French knights, dressed in heavy armor and having to
wage a constant offensive battle, were the ones who became more quickly fatigued
and thirsty, and several of them deserted the battlefield.
Thus, all hinged on the success of the Flemish attack on the French. While the
Page 47
Flemings had not been defeated by endless French cavalry charges, they had also not
gained victory over their opponents. The French kept charging, despite their frequent
setbacks, and they showed no signs of surrender. Moreover, unlike at Courtrai, the
French leader himself, in this case the king, did not take part in the cavalry charges,
and therefore was still alive. As well, fatigue began to affect the Flemish army, and
some had already retired from the battlefield. Consequently, the Flemish leaders
recognized the need to attack the French and drive them from the field.
The charge began in an ordered fashion; without breaking ranks, the Flemish line
advanced against the French who were immediately facing them. This brought instant
success, with the French troops fleeing to the rear. But, the French camp did not join
in the rout, and the Flemish line found itself among the tents of the French battling all
the way to the king.
At this point it becomes interesting to ponder what might have occurred had the
Flemings succeeded in killing or capturing the king. The battle undoubtedly would
have had a different result. But, saved by his bodyguard's dramatic defense, Philip IV
was not killed, and ultimately he was able to rally his troops and counter the Flemish
attack. Finally, the Flemish line had broken, although not by any action of the French.
The tents and other camp equipment had so disrupted the line that their defensive
formation was impossible to hold. This then allowed the French to charge easily
through the Flemings, killing many and sending the others fleeing from the field. A
few Flemish soldiers, led by William of Jülich, did try to regroup, but their numbers
were too few to withstand a French assault, and, as asserted in the Chronographia
regum Francorum, they were also very fatigued and dehydrated so that their defense
was very weak, and all were killed. In the end then, it was the early success of the
Flemish charge which eventually led to their defeat. They had disordered themselves
with their attack on the French camp, and this, coupled with the very quick French
recovery inspired by the king himself, led to the defeat of the Flemish infantry at
Mons-en-Pévèle.
The impact of the French victory at Mons-en-Pévèle may never be entirely
understood. While it seems that some lessons learned at Courtrai and Arques about
sending knights against a solid infantry line altered the maneuvers of the French army
at Mons-en-Pévèle, on the whole it appears that the French were unwilling to change
their battlefield tactics. They were destined later, at Crécy and elsewhere, to repeat the
same tactical mistakescontinually charging their knights into infantryand this brought
frequent defeats.
Therefore, the impact of Mons-en-Pévèle may have been to hide the problems knights
faced when encountering defensive infantry formations. The defeat at Courtrai could
be blamed on the ditches and other battlefield hindrances prepared by the Flemings,
on the poor leadership of French generals, on the pride of their soldiers, or simply on
the swing of the wheel of fortune. As well, the result of these battles could also be laid
at the feet of the Flemish rebel leaders, especially William of Jülich and Pieter de
Coninck, and perhaps this is the purpose behind the rather lengthy and anecdotal
stories of their deaths. The death of a losing military leader is always of interest to
contemporary chroniclers, but in the case of the battle of Mons-en-Pévèle the extent to
which most of the major accounts go to explain how
Page 48
William of Jülich died, and what happened to his body, must have other underlying
significance. Perhaps it was because of the success William had at Courtrai and
Arques, not only as a rebel leader, but also as a leader of non-chivalric infantry forces
that necessitated a description of his death. This in turn would assure the French and
Flemings that he was dead and incapable of rising again, while at the same time it
would show that the infantry victories at Courtrai and Arques were accidents, led by a
general who eventually met his fate at the hand of a knight, either the king's brother,
Charles of Valois, or the vengeful Renaud of Dammartin, the count of Boulogne.
Page 49

IV
The Battle of Loudon Hill, 1307
On March 25, 1306, Robert Bruce crowned himself king of Scotland at the castle of
Scone, the traditional site of Scottish royal coronations. In doing so he confirmed his
intention to open a second phase of the Scottish rebellion against the English king,
Edward I.
Robert Bruce was an unusual candidate to lead this phase of the rebellion. Prior to
1297, the Bruce family was loyal to Edward. Hoping eventually to gain the Scottish
throne left vacant by the deaths of Alexander III (1286) and his granddaughter,
Margaret, (1290), both Robert and his father, also named Robert, had done fealty and
homage to Edward as late as 1296. Still hoping for coronation, the Bruces neither
participated in nor condoned the rebellions of other Scottish nobles against England in
the previous decade.1 However, by May 1297 when the Scottish rebellion began again,
Robert Bruce had become impatient with Edward's reluctance to fill the vacant
northern throne. As earl of Carrick he joined the other nobles in a rebellion led by
William Wallace.2
William Wallace proved to be a disappointing military leader. Despite having a rather
large and enthusiastic Scottish army, as well as the blessings of both King Philip IV of
France and Pope Boniface VIII, and despite initially defeating the English army at the
battle of Stirling Bridge, fought on September 11, 1297, Wallace suffered a critical loss
to the English knights at the battle of Falkirk, fought on July 22, 1298.3 The Scottish
rebellion seemed at an end. Robert Bruce was not involved in the defeat at Falkirk,
but was named as one of the rebel leaders, and as such was
1 The best modern history of Robert Bruce is undoubtedly G.W.S. Barrow's Robert Bruce and
the Community of the Realm of Scotland, 3rd ed. (Edinburgh, 1988). A less satisfactory, but
also recent biography of Bruce is Ronald McNair Scott, Robert the Bruce: King of Scots (New
York, 1982). For Robert Bruce's early life see Barrow, pp. 20-38 and Scott, pp. 3-37. For his
and his father's allegiance to Edward I before 1297 see Barrow, pp. 39-68. On the latter point
see also Ranald Nicholson, Scotland: The Later Middle Ages (Edinburgh, 1974), p. 51; F.M.
Powicke, The Thirteenth Century, 1216-1307 (Oxford, 1953), pp. 606-14; Michael Prestwich,
Edward I (London, 1988), pp. 366-69; and The Three Edwards: War and State in England,
1272-1377 (London, 1980), pp. 42-47.
2 On the origins of the 1297 rebellion see Barrow, pp. 79-83; Scott, pp. 38-42; and Prestwich,
Edward I, pp. 476-79. The standard work on the Scottish rebellion is Evan M. Barron, The
Scottish War of Independence: A Critical Study, 2nd ed. (Inverness, 1934), but it is dated and has
been revised almost completely by Barrow. For Barron's discussion of the 1297 rebellion see pp.
18-67. On the relationship between Bruce and Wallace see Andrew Fisher, 'Wallace and Bruce:
Scotland's Uneasy Heroes,' History Today 39 (February 1989), 18-23.
3 The standard description of the battles of Stirling Bridge and Falkirk is found in Oman (1905),
II:75-81. Ferdinand Lot also discusses Falkirk (I:323-24). See also Barron, pp. 68-78; John E.
Morris,
(footnote continued on next page)
Page 50
pursued by Edward I after the battle. He was forced to go into hiding along with many
of the rebellious Scottish nobles.4
Robert Bruce reappeared in February 1304 as he and most other Scottish nobles
surrendered to the English and renewed their oaths of fealty and homage. The nobles
lost no land or titles, and there was no exchange of hostages or judicial action taken
against them. Moreover, a new government was established in Scotland which,
although being led by English nobles in the dominant positions of lieutenant and
treasurer, did contain some Scottish representation on the advising council. Among
this leadership was John Comyn, who was chiefly responsible for the reconciliation of
the Scots and the English, and Robert Bruce.5
But peace was short-lived. Two years after renewing his oaths of fealty and homage to
Edward, Robert Bruce rebelled against the English king, murdered John Comyn, and
crowned himself king of Scotland.6 Although some historians contend that this was a
rash act precipitated by an argument with Comyn, it seems more likely that Bruce had
carefully planned his coup-d'état. Bruce knew that Edward I was ill, and that he was
unlikely to lead an invading army into Scotland. Furthermore, Comyn was a symbol
of Scottish submission to England. His death would be a symbolic gesture which
would unite the Scottish people in a rebellious fervor.7 At the same time, Robert
Bruce's coronation as king returned the rule of Scotland to its rightful lords, the Bruce
family able to trace its claim to the throne through Alexander II.8
News of Robert Bruce's rebellion reached Edward at the end of February. On May 22,
1306, in a colorful oration delivered after the knighting of his son, the future Edward
II, at an Arthurian feast now known as the Feast of Swans, Edward I insisted that he
would avenge the death of Comyn. Swearing on the roast swans which made up the
chivalric centerpiece on the banquet table, he promised that he would not sleep more
than one night in the same place until he reached Scotland with an army to put down
Bruce's rebellion. This campaign, the king promised, would be the catalyst behind his
greatest military adventure, one which would conclude in the Holy Land.9
Edward summoned his army to meet him at Carlisle in July 1307. In the meantime,
however, he appointed Aymer of Valence, his half-cousin and brother-in-law to
Comyn, to be his special military lieutenant in Scotland. The king commissioned
(footnote continued from previous page)
The Welsh Wars of Edward I (Oxford, 1901), pp. 282-94; Barrow, pp. 83-104; Scott, pp. 42-
52; and Prestwich, Edward I, pp. 478-83.
4 See Barrow, pp. 104-31; Scott, pp. 53-64; F.M. Powicke, pp. 692-94; and M. Keen, England in
the Later Middle Ages: A Political History (London, 1973), pp. 35-37.
5 See Barrow, pp. 132-44; Scott, pp. 64-73; F.M. Powicke, p. 713; and Keen, p. 37.
6 See Barrow, pp. 145-64; Scott, pp. 73-76; Nicholson, Scotland, p. 71; F.M. Powicke, pp. 713-
14; Prestwich, Edward I, pp. 505; and T.M. Smallwood, 'An Unpublished Early Account of
Bruce's Murder of Comyn,' Scottish Historical Review 54 (1975), 1-10. For this murder Bruce
was excommunicated.
7 For a discussion on the different historical interpretations of this coup d'état see Barrow, pp.
145-53 and F.M. Powicke, pp. 713-14.
8 On Bruce's royal lineage see F.M. Powicke, p. 611 n.1 and Barrow, pp. 39-40.
9 Several sources mention the Feast of the Swans. The most colorful narrative is that of Prestwich,
The Three Edwards, pp. 34-35. See also F.M. Powicke, p. 515; Barrow, pp. 153-54; and Scott,
pp. 79-80.
Page 51
Valence to harass Bruce and to try to demoralize the Scots so that the rebellion might
be contained until the larger army arrived the following summer. Edward also sent his
son to aid Valence in this task.
Initially, Valence was very successful against the Scottish leader. By the end of
summer, Bruce's rebellion had nearly collapsed. Valence had met and defeated Bruce's
army at Methven on June 20, and this was followed by a second defeat at Dalry fought
two months later.10 Meanwhile, Edward, the prince of Wales, had also captured
Bruce's castle at Lochmaben. Many rebels had been captured and executed, including
a number of Bruce's own family members. But Robert Bruce himself had escaped
capture, and by the end of the campaigning season he, and with him the rebellion,
remained alive.11
After this successful beginning, however, nothing went right for the English cause.
Edward I became increasingly ill on his journey north and was unable to proceed
further than Carlisle. His efforts to arrange a parliament in January to discuss the
Scottish situation also were unsuccessful. Furthermore, in February, Bruce once again
eluded capture, despite being encircled by Aymer of Valence's army in the
neighborhood of Glen Trool.12 Bruce went north followed by Valence, and finally the
two armies met on May 10 at the battle of Loudon Hill. This time the outcome did not
favor the English. Robert Bruce won a certain victory, handily defeating the English
cavalry.
The victory was a stunning upset of the English strategic plans. Aymer and his army
fled to Carlisle where he met the ill Edward and resigned his commission in disgrace.
The remaining English armies too were recalled, so that a new offensive could be
undertaken. However, before that happened Edward I died (on July 7), turning the
responsibilities of English rule, including the handling of the newly motivated Scottish
rebellion, over to his son who became Edward II.13
Burdened by his new responsibilities, Edward was unable to respond promptly to
Robert Bruce's victory. The rebellion grew in numbers and morale. A now famous
letter, dated May 15, 1307 and attributed to Alexander Abernethy, recounts the spirit of
the Scottish people following Bruce's victory at Loudon Hill:
I hear that Bruce never had the good will of his followers or of the people generally so much
with him as now. It appears that God is with him, for he has destroyed
10 The tactical action of both of these battles is difficult to determine from the original sources;
as such, I am unable to write more at length on them in this work.
11 For a description of the events of 1306 see Barrow, pp. 154-61; Scott, pp. 80-100; Nicholson,
Scotland, pp. 72-75; F.M. Powicke, pp. 715-17; Barron, pp. 236-59; Prestwich, Edward I, pp.
507-09; and Keen, p. 38. Among the relatives of Robert Bruce captured during this year were his
second wife, Elizabeth de Burgh, his daughter, Marjorie, three sisters, Elizabeth Siward, Christina
Seton, and Mary Bruce, and three brothers, Neil, Thomas and Alexander. Only the three brothers
were executed although Mary was confined in a wooden cage at Berwick.
12 On Edward's attempt to call a parliament in January 1307 see Barrow, pp. 161-64 and Keen, p.
38. On Bruce's escape at Glen Trool see Barrow, p. 172 and Scott, pp. 100-02.
13 On the transformation of English power after the death of Edward I see Prestwich, Edward I,
pp. 556-58. On the impact of the victory on Scotland see Barrow, pp. 172-74 and Scott, pp. 104-
06.
Page 52
King Edward's power both among the English and Scots. The people believe that Bruce will
carry all before him . . . 14
It would not be until 1310 that Edward II again would return to Scotland to fight
Robert Bruce, the Scottish king.
Unlike most of the battles fought in the early fourteenth century, the battle of Loudon
Hill is recorded in only one source, John Barbour's historical poem The Bruce.15
Despite its historical significance, only one English chronicler, Thomas Gray, mentions
the battle, and his chronicle contains no details beyond simply recording it as a
Scottish victory. No other English source chronicling the Scottish rebellion even
mentions this battle, and this reticence is echoed by the Scottish chronicles which
discuss the life of Robert Bruce.16 This silence in itself seems odd, especially when
compared to the large number of contemporary records on the battle of Bannockburn
fought seven years later. Coupled with the fact that John Barbour's poem was not
written until 1376, the absence of a large number of sources has led most medieval
military historians to neglect or to downplay the battle's significance.17 Still, Barbour's
poem, composed in Middle English, is considered by most scholars to be of great
historical accuracy.18 Moreover, its narration of the battle of Loudon Hill shows
Bruce's successful use of infantry against cavalry, a feat which he would duplicate
later at the battle of Bannockburn.

Barbour begins book VIII of his poem after Robert Bruce's escape from Valence at
Glen Trool. Bruce traveled north followed by an angered Valence who was harassed
in his pursuit by Bruce's lieutenant, James, lord of Douglas.19 Finally, Bruce's army,
all infantry, arrived at Loudon Hill where he was sent a message from Valence
challenging him to battle on the plains near the bill. Bruce received Valence's
challenge, and, recognizing the need to eventually fight the English army despite
having a much smaller army than his English counterpart, he accepted the
14 This letter is found in Calendar of Documents Relating to Scotland, ed. J. Bain (Edinburgh,
1883), II:#1926. The translation I have used is found in Barrow, pp. 172-73.
15 While there are several editions of Barbour's Bruce, the most accessible and widely used is
that edited by Walter W. Skeat (The Bruce, or the Book of Robert de Broyss, King of Scots (1286-
1332), Early English Text Series, 2 vols. (1870-77; rpt. Oxford, 1968)). Barbour's description of
the battle of Loudon Hill is found on I:177-92.
16 Thomas Gray, pp. 34-35. Later Scottish chronicles mention the battles of Methven and Dalry
but omit mention of Loudon Hill. See John of Fordun, Chronica gentis Scotorum, ed. W.F. Skene
(Edinburgh, 1871), I:340-42 and the Liber Pluscardensis, ed. F.J.H. Skene (Edinburgh, 1877),
I:231-33.
17 The only military historian who discusses the battle is Oman (1905), II:83-84, but his narrative
is quite short. Barron's description of the battle is also quite short (pp. 262-63). Even Aymer of
Valence's biographer, J.R.S. Phillips, does not discuss the battle at any length, despite Valence's
defeat there being the reason for his retirement from military leadership. See Aymer de Valence,
Earl of Pembroke, 1307-1324: Baronial Politics in the Reign of Edward II (Oxford, 1972).
18 On the historical veracity of John Barbour's poem see George Neilson, John Barbour, Poet
and Translator (London, 1900) and Barrow, pp. 312-13.
19 Barbour, I:177-81.
Page 53
challenge.20 Valence was pleased with Bruce's desire to fight him for he felt that the
English cavalry would easily triumph over the Scottish infantry:
For he thought, through his great might,
If the king appeared to fight,
That, through the great chivalry
That was in his company,
He would so overwhelm the king,
That there would be no recovery.21
But, the Scottish army was already at Loudon Hill, while the English force had not yet
arrived. So, Bruce was able to reconnoiter the battlefield, establish his position, and
prepare the field to disrupt the charges of the English cavalry. He discovered that
while the plain itself was flat and dry, on one side there was a morass, 'deep and wide,'
which could be used as a border for fighting.22 However, the field was still too wide
for an infantry army to fight cavalry, and thus Bruce decided that alterations were
needed to aid him in fighting the English. He therefore cut three large and deep
trenches across the field, leaving small gaps in them which would decrease the
number of cavalry who could charge the infantry line. Robert Bruce also hoped that
these trenches would slow down the progress of the English cavalry charges so that
their shock might be lessened.23 Bruce then gathered his army, numbering only 600
and camped beyond the trenches.24
On the day of battle, Aymer of Valence assembled his army, numbering 3,000, and
marched to Loudon Hill. The sun was shining and the cavalry, ordered in two lines,
were impressive to their Scottish opponents. Barbour describes the chivalric array:
The sun rose, shining and bright,
So that it glimmered on the large shields.
...
Their helmets were brightly polished,
Again, on them the sun glistened;
Their spears, their pennons, and their shields
Illuminated the entire field with bright light.
Their best, brightly colored armor,
And horse the shade of such color,
20 Barbour, I: 181-82.
21 Barbour, I: 182-83:
For he thoucht, throu his mekill mycht,
Gif the king durst apeir to ficht,
That, throu the gret chevelry
That suld be in his cumpany,
He suld swa ourcum the king,
That thar suld be na recouering.
22 Barbour, I: 183,
23 Barbour, I: 183-84. Barbour claims that these gaps were so narrow that they would only allow
the passage of 500 knights charging side by side. This still seems to be rather high number and
may indeed be an exaggeration.
24 Barbour, I: 184.
Page 54

Battle of Loudon Hill


And coat-of-arms of such color,
And hauberks, that were as white as flour,
Made them glow, so that they appeared
to be angels of a heavenly kingdom.25
To offset this impressive and imposing display, Bruce addressed his soldiers,
reminding them that they would be killed for rebelling against the English king if they
failed to win the battle. They should therefore fight 'hardely,' meeting the first line's
charge bravely and solidly, so that the second line would be terrified of repeating the
same unsuccessful result. Bruce also promised his troops that victory would bring
happiness and valor. The Scots responded: 'Sir, if God wishes, we will do it / so that
no disgrace shall be placed on this deed.'26 Bruce then led his men in
25 Barbour, I: 184-85:
The sone wes rysyn schynand bricht,
That blenknyt on the scheldis braid.
...
Thair basnetis burnyst var brycht,
Agane the sone [glemand] of licht;
Thair speris, thair pennownys, & thar scheldis
Of licht Illumynit all the feldis.
Thair best [&] browdyn bricht abneris,
And hors hewit on seir colour,
And cot-armouris off seir colour,
And hawbrekis, that war quhit as flour,
Maid thame glitterand, as thai war lik
Till angellis he, of [hewinis] rik!
26 Barbour, I: 186: 'Schir, gif god will, we sall sa do / That no repruf sall ly thar-to.'
Page 55
a battlefield prayer, and they prepared for the English attack by moving to the first
trench and ordering their defensive line formation.27
Aymer of Valence saw the formation of the Scots among the trenches, but he still did
not feel that there would be any difficulty in achieving a quick victory given his larger
numbers and the fact that his cavalry faced only infantry. He too addressed his troops,
promising them that they would gain rewards and renown by defeating the rebels.28
The trumpets then announced the beginning of battle, and the English cavalry charged
into the Scots. The Scottish infantry in turn met the charge ' . . . with such vigor / that
the best and the most valorous / were dragged to the ground at the meeting.'29 Before
long the tide of battle began to favor the Scots. The English cavalry charge was
powerful and impressive, but by the time it hit the line of Scottish infantry its impact
had deteriorated, disrupted by the trenches placed on the battlefield. Their Scottish
opponents, on the other hand, held their position, using their line first to stop the
charge and then to push the cavalry back. Scottish spears killed both men and horses,
while other cavalry, trying to escape the infantry, were thrown from their mounts into
the trenches and morass:
The king's men, who were worthy,
With their spears so sharp,
Stabbed both men and horses,
Until red blood flowed freely from the wounds.
The wounded horses tried to flee,
And charged their own men in their flight,
So that those who were in the vanguard,
Were thrown here and there into the trenches.30
Among the Scots fought Robert Bruce himself, and his military exploits boosted
morale within his own force while inspiring fear among the English.31
Soon the battle was over. The field was covered with dead and dying English cavalry
and their horses. The Scots too had suffered losses, but they were small in
comparison to their chivalric opponents'.32 Still in line, and still led by Robert Bruce,
27 Barbour, I: 186-87.
28 Barbour, I: 187.
29 Barbour, I: 188: ' . . . vith sa gret vigour / That the best and of mast valour / Var laid at erd at
thair metyng.'
30 Barbour, I: 188-89:
The kingis men, that worthy war,
With thair speris that scharply schar,
Stekit men and stedis bath,
Till red blude ran of voundis rath.
The hors that [woundyt] war can fling,
And ruschit the folk in thair flynging,
Swa that thai that than formast war
War stekit in soppis heir and thar.
31 Barbour, I: 188. Also singled out for their bravery in battle were James of Douglas and
Edward Bruce, Robert's brother.
32 Barbour, I: 189. No estimates of casualties on either side are given by Barbour although he
does indicate that when Bruce advanced after the fleeing English troops only 500 soldiers went
with him
(footnote continued on next page)
Page 56
the Scottish infantry began to advance across the field in pursuit of the fleeing
soldiers. Seeing this rout, the English rearguard, which had yet to enter the fray,
turned and fled as well. Aymer of Valence tried to rally his troops, but finally he too
was forced to flee from the field; the English had lost the battle of Loudon Hill.33
The Scots left the battlefield with their prisoners. They were obviously elated with the
victory, and they praised Bruce for leading them against the English. Barbour ends his
narration of the battle by writing:
They were without a doubt happy;
Because they knew that they had won,
As the king had helped them so.
From that they gave him more homage.
Then his power grew more and more.34
There are several similarities between this battle and that of Courtrai fought five years
previously. Despite being challenged to battle by the English, a challenge which
included the site of the battlefield, Robert Bruce was given the opportunity to choose
on what part of the field the fighting was to take place. Arriving at the site first, he
was able to survey the area, noting its natural hindrances to fighting, and to prepare it
to serve his advantage in battle. He knew that he was fighting with a small infantry
army against a larger cavalry force, and he used this to formulate his battle plan.
Although he found a morass bordering the battlefield, he judged the plains to be too
large to serve his purposes. Therefore, he directed that the field be narrowed by
digging three large trenches across it. He then ordered his troops in a defensive line
with the intention of standing against the charge of English cavalry.
On the English side, as with the French at Courtrai, there was no concern for the pre-
battle activities of the Scots. Despite seeing the trenches, and perhaps knowing that
they would decrease the number of cavalry who could charge at one time, the English
army proudly formed their mounted shock formation and surged forward. Their
display was impressive and their charge seemed powerful, and yet they lost. The
Scottish infantry simply held their line, and their opponents on horseback could not
pierce their formation. Again as at Courtrai, the warhorses must have stopped, and
this allowed the infantry to use their weapons advantageously against the heavily
armored cavalry. Horses became a target, as did the fallen cavalry with the weak joints
of their armor exposed.
(footnote continued from previous page)
indicating a possible loss of one hundred men. Oman (1905, II:85) reports that one hundred
English knights were killed there, but he gives no source for his tally, and I have yet to find any
mention of this elsewhere.
33 Barbour, I: 189-90. The English rearguard probably contained infantry, but Barbour does not
report who was in that formation, only that they fled with the routed cavalry.
34 Barbour, I: 191:
So war thai blith forouten dout;
For feill that wonnyt thaim about,
[Fra] thai the king saw help him swa,
Till him thar homage can thai ma.
Than vox his power mair and mair.
Page 57
Robert Bruce rejoiced in the defeat of the English. Although we are not told what he
thought about the battle, or whether he believed that it was his tactics which achieved
victory, the next time a serious threat was made against his claim to the throne, at
Bannockburn in 1314, he repeated many of his Loudon Hill tactics and again achieved
victory. At Bannockburn, however, Bruce did not face and defeat a lieutenant of the
king but the king himself, Edward II.
Page 58

V
The Battle of Kephissos, 13111
There are perhaps few more interesting medieval military organizations than the
Catalan Grand Company. A band of mercenaries by definition, they are more
appropriately described as a loose-knit group of late medieval warriors, largely
infantry troops, who travelled widely throughout the eastern Mediterranean region.
Effectively leaderless for most of their early history, during that period the Catalan
Company still faced and defeated armies made up of Turkish, Caucasian, Genoese,
Thracian, Macedonian, Athenian, Byzantine, Burgundian, and French soldiers. In
doing so they captured large amounts of land, dominating and ruling most of Greece
throughout much of the fourteenth century Kenneth M. Setton's dates are 1311-1388.2
The Catalan Company was organized late in 1302 by Roger de Flor of Brindisi. Once a
Knight Templar and very much an experienced warrior, both on land and sea, Roger
de Flor gathered together a group of experienced Spanish soldiers who had taken the
side of Frederick II against the Napolese Angevins in the War of the Sicilian Vespers
which had lasted for more than twenty years. Having fought for so long, most of those
who joined his company had little else they could do for employment.3 The Company
was not small; most historians estimate a number of 6,500, 4,000 of whom were 'the
famed and formidable almugávares', infantry troops of outstanding
1 There are several spellings of this battle, which was named after the river near where it was
fought. I have chosen to use the Greek. Other spellings include Cephissus and Céfu. Jep Pascot,
Les almugavares: Mercenaires catalans du moyen âge (1302-1388) (Brussels, 1971), pp.
149-55 calls the battle Copaïs after a nearby lake.
2 Kenneth M. Setton, Catalan Domination of Athens, 1311-1388 (Cambridge, 1948) is the
seminal work on the Catalan Company. Other works of importance include: Pascot; Kenneth M.
Setton, 'The Catalans in Greece, 1311-1380,' in A History of The Crusades, ed. K.M. Setton, vol.
III: The Fourteenth and Fifteenth Centuries, ed. H.W. Hazard (Madison, 1975), pp. 167-224; R.
Ignatius Burns, 'The Catalan Company and the European Powers, 1305-1311,' Speculum 29
(1954), 751-71; David Jacoby, 'La ''compagnie catalane" et l'état catalan de Grèce. Quelques
aspects de leur histoire,' Journal des savants (1966), 78-103; Nicolas Cheetham, Mediaeval
Greece (New Haven, 1981), pp. 130-51; and J. Longnon, 'The Frankish States in Greece, 1204-
1311,' in A History of The Crusades, ed. K.M. Setton, vol. II: The Later Crusades, ed. R. L. Wolff
and H.W. Hazard (Madison, 1969), pp. 270-74.
3 Setton, Catalan Domination, pp. 2-3; Pascot, pp. 32-44; Setton, 'Catalans,' pp. 168-69; Burns,
p. 752; Jacoby, pp. 79-81; Cheetham, pp. 134-35; Longnon, p. 270; and Norman Housley, The
Later Crusades, 1274-1580: From Lyons to Alcazar (Oxford, 1992), p. 161. The Catalan
Company initially was made up of Catalan, Aragonese, Majorcan, and Navarrese soldiers. Later,
Turks, Byzantines, Thracians, and Macedonians joined the mercenary group.
Page 59
skill and ability.4 The Byzantine Emperor, Andronicus II Palaeologus, was looking for
experienced soldiers, hoping to import an army to face the growing power of the
Ottoman Turks in the eastern Mediterranean. He could pay well, and the Catalan
Company jumped at the opportunity to put their skills to use again.5
By August 1303, the Company had their first success, by sacking the island of Ceos.
The following month they arrived in Constantinople. Quickly they were put up against
the Ottomans in Asia Minor, and just as quickly they began to push them back from
the Byzantine capital. Soon they had won so much favor and were so feared in
Byzantium that Roger de Flor, now with the title 'Caesar,' had even married into the
Palaeologi family. But this turned out to be a bad decision, for he was murdered by
the same family in April 1305.6
Devoid of their founder and leader, but respected and feared by all in the east, the
Catalan Company withdrew from Constantinople but refused to return to the western
Mediterranean. They went instead to the Gallipoli peninsula where they established
their first state.7 This state did not last long. In June 1307, the Company left Asia
Minor and moved into the west. They traveled swiftly. With little opposition, the
Company, now bolstered by Turkish and Greek recruits, quickly conquered Thrace
and Macedonia, and by spring 1309 they entered Thessaly, threatening Athens,
Thebes, and the lower Greek peninsula.8 Walter (Gautier) I of Brienne, the Frankish
duke of Athens, did what little he could to stop their advancement: rather than fighting
against them, he hired them to fight for him. For six months, until the end of 1310,
they waged war for their new employer, taking more than thirty villages, towns, and
strongholds, conquering other rulers of Greece, in Thessaly and Epirus, and even
attacking Byzantine forces in the southern peninsula.9
But the wisdom Walter had shown in hiring instead of firing the Catalan Company,
now disappeared. After so many victories he sought to dismiss them still owing four
months wages. Instead, he selected 500 of the mercenaries200 knights and 300
almugávarespaid them their wages, gave them lands and titles, and then requested that
they keep their comrades out of Athenian territories. The rest of the Catalan Company
moved into fortifications in southern Thessaly, but they refused to be so easily
dismissed, and during the end of 1310 and the beginning of 1311 both they and the
duke of Athens prepared for battle. A blow to the Athenian military strength came
with the defection of the 500 Company soldiers whom the duke of Athens had
4 Setton, Catalan Domination, p. 3; Pascot, p. 44; and Burns, p. 752.
5 Setton, Catalan Domination, p. 2; Burns, p. 752; and Jacoby, p. 79.
6 Setton, Catalan Domination, pp. 3-4; Pascot, pp. 47-85; Setton, 'Catalans,' p. 169; Burns, p.
752; Jacoby, pp. 80-81; Cheetham, pp. 134-37; and Housley, p. 161.
7 Setton, Catalan Domination, p. 4; Pascot, pp. 87-123; Setton, 'Catalans,' p. 169; Jacoby, pp. 81-
86; and Longnon, p. 270. At Gallipoli, the Catalan Company was forced to defend themselves not
only against the Byzantines, but also against the Genoese and Caucasians.
8 Setton, Catalan Domination, pp. 4-5 and Pascot, pp. 125-40. Pascot claims that the Catalan
Company vacated Gallipoli because of discord among the Company. Setton believes that it was
because the Catalans had devastated the peninsula and were unable to continue to feed themselves.
9 Setton, Catalan Domination, pp. 7-8; Pascot, pp. 140-49; Setton, 'Catalans,' pp. 169-70;
Cheetham, pp. 132, 137-40; Longnon, pp. 270-71; and Housley, pp. 161-62.
Page 60
separated from the rest of the mercenary band; they rejoined their comrades.10 A more
major blow came on March 15, 1311, when the still outnumbered Catalan Company
met and defeated the duke of Athens and his Frankish knights at the battle of
Kephissos. Athenian losses were numerous; Walter of Brienne was among the dead on
the field. Greece was now the land of the Catalan Company.11
There are only three sources for the battle of Kephissos and all are written from the
Catalan perspective. The longest and most detailed is Ramón Muntaner's Crònica,
written sometime before his death in 1336. The second is the Aragonese version of the
Chronique de Morée known as the Libro de los fechos et conquistas delprincipado de
la Morea which was compiled during the first half of the fourteenth century under the
direction of the Don Fray Johan Ferrandez de Heredia, Master of the Hospital of St.
John of Jerusalem.12 And the third is found in Nikephoros Gregoras's Byzantina
historia written in Constantinople c. 1359. Although composed a great distance from
and much later than the battle of Kephissos, Gregoras'narrative seems to have been
taken from Catalan or Turkish participants.13 There is no source on Kephissos written
from the Athenian position.

Probably because there are no original sources written on the battle of Kephissos from
the Athenian perspective, there is little disagreement in the story given by Ramón
Muntaner, the Aragonese version of the Chronique de Morée, and Nikephoros
Gregoras.
In 1311 there was undoubtedly no military force more experienced in warfare than the
Catalan Company. Although there is little record of them fighting full-scale battles
before Kephissos, they were involved in the raids, chevauchées, and sieges, both
against armed forces and unarmed civilians, that generally characterized late medieval
warfare. They had also fought several small engagements at which their battle tactics
would have been strengthened. As well, their military training had been developed by
fighting several different foes, from the peasant factions quickly put together to
defend the lands of Thrace and Macedonia to well-organized and armed professional
Byzantine Imperial contingents.
The Catalan Company was also quite numerous for a mercenary band. The Aragonese
Chronique de Morée indicates that at the time of the battle of Kephissos the Company
numbered 6,000, 2,000 of whom were cavalry and 4,000 infantry.14 This number was
not much smaller than the initial recruitment count given by Ramón
10 Setton, Catalan Domination, pp. 8-9; Setton, 'Catalans,' p. 170; Cheetham, p. 140; Longnon,
p. 271; and Housley, p. 162.
11 Setton, Catalan Domination, pp. 9-13; Pascot, pp. 149-54; Setton, 'Catalans,' pp. 167, 170-71;
Jacoby, p. 87; Cheetham, pp. 140-42; Longnon, pp. 271-72; Housley, p. 162.
12 Ramón Muntaner, Crònica, trans. H.M. Goodenough (London, 1921), II:575-78 and Libro de
los fechos et conquistas del principado de la Morea compilado por comandamiento de Don
Johan Ferrandez de Heredia, ed. A. Morel-Fatio (Geneva, 1885), pp. 119-21.
13 Nikephoros Gregoras, Byzantina historia, ed. J.-P. Migne, Patrologia Graeca, 148 (Paris,
1865), 411-22.
14Libro de los fechos, p. 120.
Page 61
Muntaner,15 although Nikephoros Gregoras indicates that more than 1,100 Turks had
been added to the Company total before their invasion of Thessaly.16
The Athenian army was much larger, with Nikephoros Gregoras counting 6,400
knights and 8,000 foot, and Ramón Muntaner enlarging the latter figure to 24,000.17
These numbers are probably exaggerations, at least that is the conclusion of most
modern historians, yet there is little doubt that the Athenians outnumbered the Catalan
Company. Indeed, the Aragonese version of the Chronique de Morée reports that
when the Company saw the number of their enemy at Kephissos, 'that great multitude
of men,' they tried to make peace.18 Many of these troops were also mercenaries
themselves, with Muntaner estimating 700 'Frankish' knights fighting for the duke of
Athens.19
But while outnumbering the Catalan Company, the Athenians certainly did not have
the same level of military experience as did their counterparts. After all, it was because
of their experience that the Catalans had been hired as warriors by Walter of Brienne
in 1309. This relative lack of experience may have been a significant factor in the
Athenian loss at Kephissos.
The story of the battle of Kephissos opens with the hiring and then firing six months
later of the Catalan Company by the duke of Athens, Walter of Brienne. This was an
odd occurrence, claims Ramón Muntaner, as Walter had 'made himself beloved by the
Catalans and spoke Catalan.' In turn, the Company had responded by fighting
valiantly for the duke, recovering much territory which had been taken from him by
the Byzantines and others.20
Still, there was the question of money. The Catalan Company had been in the employ
of Athens for six months, but had only been paid for two. This is when Muntaner
claims that Walter of Brienne decided, 'when he saw that he was at peace with all his
neighbours,' on 'a very wicked plan': he would choose 500 of the Catalans and bribe
them to stay with him, paying them for their past service and giving them land in the
duchy. The others he commanded to be driven from his duchy without paying what
was owed to them. In devising this plan, Walter was certain that the mercenaries who
remained would defend him against those whom he had driven away.21
There were two grave miscalculations made by the duke in this plan. First, instead of
leaving Greece as ordered, the Catalan Company retreated into some of the
strongholds which they had recently won. While the Aragonese Chronique de Morée
insists that the Company only wanted to remain in peace there, these were technically
the duke's holdings, and thus it appears that the Catalans were staying close enough to
Athens to monitor the situation and to wait for an opportunity to gain vengeance
15 Ramón Muntaner, pp. 485-86.
16 Nikephoros Gregoras, pp. 414-15.
17 Nikephoros Gregoras, pp. 423-24 and Ramón Muntaner, p. 576.
18Libro de los fechos, p. 120. See also Setton, Catalan Domination, p. 9; Pascot, pp. 151-52;
and Cheetham, p. 140.
19 Ramón Muntaner, p. 576.
20 Ramón Muntaner, pp. 575-76 and Libro de los fechos, p. 119.
21 Ramón Muntaner, p. 576. See also Libro de los fechos, pp. 118-19.
Page 62
for their mistreatment by the duke.22 Second, although mercenaries by occupation,
money and lands alone could not force the 500 Catalans who had been paid to Stay
behind to fight against their comrades. When placed in a battle situation at Kephissos,
they refused to contend against the Company and instead joined with them in fighting
the Athenians. However, at the time the duke did not worry about this defection, as,
perhaps anticipating its occurrence, he had gathered a large army of other soldiers to
fight his war.23
Walter of Brienne was to fight this war on a field chosen by the Catalans. According to
Ramón Muntaner, the Catalan Company knew when the Athenian army marched
against them, and that in response, together 'with their wives and their children,' they
traveled to a 'beautiful plain near Thebes'. It was an easily defensible position,
Muntaner reports: 'in that place there was a marsh, and of that marsh the Company
made a shield for themselves.'24 Nikephoros Gregoras adds that the Catalan Company
further prepared the field by 'irrigating' it with more water brought in from the nearby
river. But the Company themselves were not to be bogged down by this water as
Gregoras also reports that the Company were able to form their line on 'a flat land,
green with grass.'25 To reach the Catalans, standing on this dry ground, the Athenians
would have to charge across a marsh, certainly an insurmountable obstacle.
So why did the Athenians choose to attack the Catalans at Kephissos? The answer is
not clear in the sources, except that the Aragonese Chronique de Morée seems to
indicate that the large size of the Athenian contingent gave the duke much confidence
in facing the Catalan Company, no matter what battlefield obstacles lay in his way.
This is also indicated in the speech given by Walter to the 500 Catalans who wished to
join their former compatriots: 'And the [duke] told them to go, and bad luck go with
them, that it was well that they should die with the others.'26 Furthermore, the
Company themselves may have encouraged this confidence first by the decision of the
Turks within the Company to refuse to fight, as Ramón Muntaner records, 'thinking it
[the battle] was done by an agreement between the two sides, in order to destroy
them,' and second by trying to make peace with the Athenians at the onset of the
battle. By refusing to make this peace, however, the duke forced the Catalans to
decide, in the words of the anonymous author of the Aragonese Chronique de Morée,
that it was 'better to die in battle than to surrender to his MerCy.'27
Choosing the field was only part of the Catalan tactics at Kephissos. They also ordered
themselves in a defensive formation, as a solid line beyond the marsh. Whether this
line included both cavalry and infantry or infantry with dismounted cavalry as in other
early fourteenth-century battles cannot be known from the original
22Libro de los fechos, pp. 119-20 and Ramón Muntaner, p. 576.
23 Ramón Muntaner, pp. 576-77.
24 Ramón Muntaner, p. 576.
25 Nikephoros Gregoras, pp. 419-22,
26Libro de los fechos, p. 120 and Ramón Muntaner, pp. 576-77. The speech of the duke is
reported only in Muntaner.
27 Ramón Muntaner, p. 577 and Libro de los fechos, p. 120: '& ellos, viendo que non los queria
recebir finon á mercer.'
Page 63

Battle of Kephissos
sources. However, all three sources are interested in and do describe the Athenian
offensive formation which charged at this line. This is reported to be a cavalry line
made up of knights200 French knights 'all with gold spurs' according to Muntanerwith
a few infantry following led by Walter of Brienne himself.
This charge failed miserably, although what caused its failure is in dispute. Ramón
Muntaner, for example, describes what happened simply and without many details:
What shall I tell you? The horses of the count, at the noise the almugavars made, turned towards
the marsh, and there the count and the banner fell, and all those who were in the van.28

But Nikephoros Gregoras has a more spectacular determination for why the charge
ended in defeat. After discussing the water-filled state of the battlefield, the Byzantine
chronicler writes: 'so that it slowed and stopped the unstable charge of the horses
whose hooves, mired in the mud, were able to be moved only with difficulty.'29
Despite the author's distance geographically and chronologically from the battle, it is
Gregoras' description of the charge's failure which has been accepted by most modern
historians; although, to be accurate, they get this depiction of events not from
Nikephoros Gregoras, but from the seventeenth-century Catalan Chronicle of
Francisco de Moncado. Attributing his narrative to Gregoras, Moncado expands the
original author's report of the Athenian charge:
28 Ramón Muntaner, p. 577.
29 Nikephoros Gregoras, pp. 421-22.
Page 64
The Duke with the first troop of the vanguard rushed to close with a squadron of infantry that
was on the other side of the submerged fields, the cavalry furiously pushed forward without
noticing that it was in the midst of the marsh. At the same time the Almugavars, swift moving
and unencumbered, with darts and swords fell upon those who, weighted down with iron, were
wallowing with their horses in the mud and slime. The rest of the troops came up to help the
Duke and fell into the same danger.30
Surely this could have occurred, but as it is not so mentioned in Muntaner's chronicle
or in the Aragonese Chronique de Morée, although such a colorful episode should
have elicited comment, there may be some doubt as to its accuracy. It could also be
that the marsh did not bog down and stop the charge but merely hindered it from
reaching the Catalan line with full force. In fact, this interpretation more easily allows
for the statements made by both Muntaner and the Aragonese Chronique after their
mention of the charge that the Company fought very well against the attacking force.
The Aragonese chronicler claims that the Catalans fought 'como hombres
desesperados,' while Ramón Muntaner writes that 'the battle was very hard; but God,
Who always helps the right, helped the Company.'31 These are hardly descriptions of
combat against an entrapped enemy. As well, at this point in the battle, seeing that it
was not simply a plan to destroy them, the Turks rejoined the Catalan Company, and
added their attacks against the Athenians.32 This too might not have happened if the
Athenian charge had been stopped in the mud.
In the initial charge, Walter of Brienne lost his life. Nikephoros Gregoras blames him
directly for the defeat:
Truly the opinion and arrogance of the prince differed much from correct reason. For he hoped
not only to defeat the Catalans, but also to take all of their lands and towns all the way to
Byzantium. But a while later the opposite occurred.33
Many others also fell at Kephissos. Ramón Muntaner numbers the dead infantry at
20,000, while Gregoras reports a loss of 6,400 cavalry and 8,000 foot, equalling the
total number which he says fought in the battle. Muntaner also claims that of the 700
French knights who fought with the Athenians, only two survived. Undoubtedly these
are all exaggerations, although the large numbers may simply demonstrate the
overwhelming defeat had by the Athenians that day.34
The defeat was significant and perhaps even could be defined as decisive. Ramón
Muntaner's concluding words are simple and direct, but they reveal much: 'And so
30 Francisco de Moncado, Expedición de los Catalanes y Aragoneses contra Turcos y
Griegos (Madrid, 1941), p. 230. This has been translated in The Catalan Chronicle of
Francisco de Moncado, trans. E Hernández, ed. J.M. Sharp (El Paso, 1975), pp. 217-18.
Moncado's explanation of battlefield events has been accepted by Setton, Catalan Domination,
p. 11; Pascot, p. 153; Setton, 'Catalans,' p. 171; Cheetham, p. 141; and Longnon, p. 271.
31Libro de los fechos, p. 120 and Ramón Muntaner, p. 577.
32 Ramón Muntaner, p. 577.
33 Nikephoros Gregoras, pp. 421-22.
34 Ramón Muntaner, p. 577 and Nikephoros Gregoras, pp. 421-22. See also Libro de los fechos,
p. 120.
Page 65
the Company . . . had won the battle and all the Duchy of Athens.'35 They would hold
the duchy for more than seventy years.
Despite the size and cavalry domination of the Athenian force, and despite the initial
Turkish unwillingness to fight with the Catalan Company, which significantly
decreased their numbers, the smaller army seems to have easily won the battle. Why?
In assessing the battle of Kephissos, two important factors must be considered: First,
the Catalan Company chose a battlefield with obvious natural defensive advantages.
In particular, a marsh ran across the field. This allowed the Company to order their
line in a defensive formation on flat land, with the Athenians forced, if they wanted to
fight a battle against their mercenary foes, to cross the marsh to reach them. The
Catalan Company may also have prepared this field further by adding even more
water from the nearby river.
Second, while all signs indicated that the Athenians should not charge their cavalry
into the Catalans across such a field, they did so anyway. The lure was not archery fire
as in some other early fourteenth-century battles, but pride. Walter of Brienne simply
could not understand how his large, noble, cavalry-based force could be defeated by a
much smaller, largely infantry army. As Gregoras indicated, this may have been
against 'correct reason,' but it did convince the duke that a charge could be successful.
It was not. Walter's pride cost him his life and lands and the lives and lands of many
of his retainers. For whether the charge became mired in the mud and stopped, thus
allowing for a more mobile and lighter infantry counterattack to defeat them, or
simply became disordered by the marshy conditions of the battlefield, thus allowing
the defending line to stop the confused and disrupted charge and to defeat it within
their defensive formation, the result was the same, victory for the Catalan Company.
35 Ramón Muntaner, p. 577.
Page 66

VI
The Battle of Bannockburn, 1314
With his victory at Loudon Hill in May 1307 and the death of the English king,
Edward I, in July 1307, Robert Bruce gained virtual control over all of Scotland.
Loudon Hill had given Bruce strong credibility as a leader, and even those who
opposed his right to the throne or doubted his ruling capabilities before his crowning
now accepted their new king without much hesitation. Also adding to Robert Bruce's
credibility was the fact that a strong English opponent in Edward I was replaced by a
much less strong heir, Edward II; the new English king's military, political, and
personal weaknesses were recognized long before his crowning. This, coupled with
the prevailing story in Scotland that in a vision an English knight had seen Edward I
residing and being tormented in hell, seemed to indicate that God was now favoring
the Scots and had given His divine approval to the reign of their King Robert I.1
From 1307 to 1310 Robert Bruce ruled Scotland with few complications or hindrances
from England. This gave the Scottish king time to solidify his kingdom and to unify
his followers. Those who continued to oppose Robert were dealt with without mercy.
For example, from the end of 1307 to the end of 1309 the Scottish king campaigned
against the Comyn-Balliol adherents throughout his kingdom, causing them to
surrender their castles and lands; some joined the side of Bruce, while others were
forced to flee to the protection of English borders. Indeed, so fervent was Bruce's
military enthusiasm during this campaign that when he fell ill in December 1307, he
continued to lead his army from a litter. Each foe he faced quickly wrote an appeal for
aid to Edward II, but on every occasion no relief came and defeat quickly followed.
By Christmas 1309, two-thirds of Scotland was securely under the control of Robert
Bruce, and almost all opposition to his rule was either neutralized or crushed. Only the
little populated northern hinterlands and the castles of Stirling, Perth, and Dundee,
able to be supported by the English navy, remained hostile to his authority.2
After 1309, Robert Bruce turned his efforts to more peaceful means of ruling.
Parliaments began to be held, the treasury filled, and positions allotted. Profiting
1 Nicholson, Scotland, p. 76.
2 Barrow, pp. 173-83; Scott, pp. 104-15; Barron, pp. 273-375; John E. Morris, Bannockburn
(Cambridge, 1914), p. 19; W.M. Mackenzie, The Battle of Bannockburn: A Study in Mediaeval
Warfare (Glasgow, 1913), p. 3-5; Nicholson, Scotland, pp. 77-81; Prestwich, Three Edwards, p.
53; May McKisack, The Fourteenth Century (Oxford, 1959), pp. 32-33; and A.A.M. Duncan,
'The War of the Scots, 1306-23,' Transactions of the Royal Historical Society 6th series, 2
(1992), 139-48.
Page 67
from this especially were the large number of lesser nobles and small landowners who
had supported the king since his ascension and now began to benefit from that
support.3 Diplomacy with France also began again, with new Franco-Scottish
relations agreed to. A Norwegian-Scottish treaty was signed on October 29, 1312. And
Robert Bruce also initiated diplomatic relations with Flanders and Germany, bringing
increased economic prosperity and security to his kingdom.4
In late 1310, Edward II finally decided to invade Scotland. But this expedition was
destined for failure as, despite a successful excursion into Scotland led by the king's
favorite, Piers Gaveston, few of the English forces were able or willing to move from
Berwick, remaining in garrison there until the summer of 1311.5 Despite its
impotence, Bruce seized upon the 'invasion' as an excuse to raid the marches of
England, attacking frequently during the next three years, carrying off large amounts
of booty, and forcing the inhabitants to agree to costly truces to protect their lands and
agricultural livelihoods. During these years Robert Bruce also conquered Perth and
Dundee.6 Finally, in May 1313 Edward Bruce, while his brother invaded the Isle of
Man, besieged Stirling Castle. By summer a truce had been agreed to: if an English
army did not come within three miles of the castle by Midsummer's Day 1314, Stirling
would peacefully capitulate. The next move was to be Edward's. He could undertake
an invasion of the Scots, something that he had been unable to do in 1310, and
perhaps win; or he could do nothing and not only lose Stirling Castle but also
continue to lose credibility at home among those who were awaiting an opportunity to
remove him from the English throne. The answer was simple: gather an army and
march to the relief of Stirling.7
Edward II used this incident to counter his baronial opposition in England, impressing
almost everyone of the necessity of participating in this military adventure. By March
1314 preparations began for the invasion, and by May, a large English army gathered
at Berwick and began to march into Scotland. By June 23, one day before the truce
was to expire, Edward's force came within three miles of Stirling Castle. However,
between him and the castle stood Robert Bruce's army,
3 Barrow, pp. 183-86; Scott, pp. 116-19; Barron, pp. 363-71; and Nicholson, Scotland, pp. 81-
82.
4 Barrow, pp. 183-84, 190, 198-200; Scott, pp. 120, 130 and Nicholson, Scotland, p. 82.
However, a formal Franco-Scottish alliance would not be signed until April 1326.
5 Scott, pp, 119-26; Barron, pp. 379-84; Morris, Bannockburn, pp. 19-20; Mackenzie, pp. 5-9;
Nicholson, Scotland, pp. 82-83; McKisack, pp. 11-12; and M.R. Powicke, 'Edward II and
Military Obligation,' Speculum 39 (1956), 97-98.
6 Barrow, pp. 190-95; Scott, pp. 128-36; Barron, pp. 392-418; Morris, Bannockburn, pp. 20-21;
Mackenzie, pp. 9-14; Nicholson, Scotland, pp. 83-84; Prestwich, Three Edwards, p. 53;
McKisack, pp. 32-34; Duncan, pp. 146-49; Jean Scammell, 'Robert I and the North of England,'
English Historical Review 73 (1958), 385-86; J.A. Tuck, 'War and Society in the Medieval
North,' Northern History 21 (1985), 35-36; and Michael Prestwich, 'England and Scotland during
the Wars of Independence,' in England and Her Neighbours, 1066-1453: Essays in Honour of
Pierre Chaplais, ed. M. Jones and M. Vale (Woodbridge, 1989), p. 195.
7 Barrow, pp. 195-202; Scott, pp. 136-42; Barron, pp. 418-28; Morris, Bannockburn, pp. 20-21;
Mackenzie, pp. 14-18; Nicholson, Scotland, pp. 84-86; Prestwich, Three Edwards, p. 54; and
McKisack, p. 34.
Page 68
outnumbered, but prepared, if needed, to fight the English force.8 A two day battle
ensued, and at the end, the unexpected had occurred. Edward II was defeated, barely
escaping his own capture, and the English soldiers, those who had not been killed or
taken as prisoners, were rushing back to their own borders.9 It would be the most
significant victory in Scottish history, a result which all future rebels would try to
imitate, but at which none would succeed.
Like Courtrai and completely unlike Loudon Hill, there is a large number and great
variety of impressive accounts of the battle of Bannockburn written by contemporary
or near contemporary authors from both sides. Obviously such a striking victory by a
smaller and less experienced force over an army made up of the most noble English
knights and led by their king fascinated and awed many. Scottish commentators tried
diligently to explain why they had won, while their English counterparts tried equally
as diligently to rationalize why they had lost.
Unfortunately, no contemporary Scottish account of the battle exists, the most detailed
report coming from John Barbour's c. 1375 narrative poem on Robert Bruce which
was of such great importance in explaining the Scottish king's tactics at Loudon Hill.
Other, much smaller accounts of the battle can be found in three Latin works, John
Fordun's Scotichronicon, written c. 1384, the Liber Pluscardensis, and Walter Bower's
continuation of Fordun's Scotichroniconwhich also contains much new material about
Bannockburnboth written in the early to middle fifteenth century. Strangely, Andrew
of Wyntoun's lengthy Orygynale Cronykil of Scotland contains nothing about the
battle, although a short account can be found in his Brevis cronica, written in Middle
English at the beginning of the fifteenth century.10
The English accounts of Bannockburn, although none as extensive as Barbour's
narrative, are both more numerous and more contemporary. Lengthy discussions of
the battle can be found in John of Trokelowe's Annales, written c. 1330, Geoffrey le
Baker's Chronicon, written starting in 1341, the anonymous Chronicon de Lanercost,
c. 1346, the Vita Edwardi II, written c. 1348 by an anonymous clerk at St. Paul's in
London, Thomas of Burton's Chronica monasterii de Melsa, written at the end of the
fourteenth century, and Thomas Walsingham's Historia Anglicana, written in the early
fifteenth century. Other, shorter descriptions are in Adam Murimuth's Continuatio
chronicorum, written and updated between 1325 and 1347, Henry Knighton's
Chronicon, c. 1360, The Anonimalle Chronicle, written c. 1381 in French at St. Mary's
8 Barrow, pp. 203-10; Scott, pp. 143-46; Barron, pp. 429-49; Morris, Bannockburn, pp. 57-
60; Mackenzie, pp. 19-41; Nicholson, Scotland, 86-87; McKisack, pp. 34-35; and Duncan, pp.
149-50.
9 Morris, Bannockburn, pp. 60-93; Mackenzie, pp. 42-111; Barrow, pp. 211-32; Scott, pp. 146-
65; Barron, pp. 450-77; Oman, (1905), II:84-100; Delbrück, III:438-42; Nicholson, Scotland, pp.
87-90; Prestwich, Three Edwards, pp. 54-55; and McKisack, pp. 35-39. Morris, MacKenzie,
Barrow, Scott, Barron, Oman, and Delbrück have all written extensive accounts of the battle. I
have chosen, however, to write my account of the battle solely from the original sources and have
used these other accounts only sparingly, even when these authors agree with my interpretation of
what occurred on the battlefield.
10 John Barbour, II:257-330; John Fordun, Chronica gentis Scotorum, ed. W.F. Skene (Edinburgh,
1871), I:346-47; Liber Pluscardensis, ed. F.J.H. Skene (Edinburgh, 1877), I:237-39; Walter
Bower, Scotichronicon, ed. D.E.R. Watt (Aberdeen, 1991), VI:349-77; and Andrew of Wyntoun,
Brevis cronica (Edinburgh, 1879), III:336.
Page 69
Abbey in York, the Eulogium historiarum, from the monastery at Malmesbury, which
was begun in 1367 and continued until 1413, and The Brut, which was written and
rewritten mostly in the fifteenth century.11 There are also three contemporary poems
on the battle of Bannockburn: one, composed in Middle English before 1352, by
Laurence Minot is quite renowned, while two others, one anonymous and one by
Robert Baston, both written in Latin, are less well known.12 Perhaps the most
interesting account of the battle is a long one written c. 1355 by Thomas Gray in his
Scalachronica. What makes this account so unusual is that Gray, a soldier who may
have fought against the Scots at Neville's Cross in 1346, began his work, written in
Old French, while in an Edinburgh prison. As well, his father, also named Thomas,
was captured at Bannockburn. Consequently, he includes some information which
appears nowhere else and which he may have learned from his captors.13
O famous race unconquered through the ages, why do you, who used to conquer knights, flee
from mere footmen? At Berwick, Dunbar, and Falkirk, you carried off the victory, and now you
flee from the infantry of the Scots.14
With these words the anonymous author of the Vita Edwardi II asked the question that
was on the lips of perhaps every Englishmen who learned of the defeat of their forces
against the Scots. How could the English have lost? They were a large and quite
experienced force, outfitted with the best arms and armor and led by the greatest
military leaders of the land. Moreover, they were a cavalry-based force, whose horsed
troops should have easily trampled down their rebellious and 'rustic' infantry
opponents.
With the siege of Stirling progressing and the Scottish ultimatum delivered south to
Edward II, the English king resolved to go north with a relief army. As Walter Bower
reports:
But the new King Edward, ablaze with strident anger on hearing of the illustrious deeds of King
Robert, and weighing up the innumerable evils and infinite losses
11 John of Trokelowe, Annales, ed. H.T. Riley (London, 1866), pp. 83-87; Geoffrey le Baker,
Chronicon, ed. E.M. Thompson (Oxford, 1889, pp. 7-9; Chronicon de Lanercost, ed. J.
Stevenson (Edinburgh, 1839), pp. 224-27; Vita Edwardi II, ed. and trans. N. Denholm-Young
(London, 1957), pp. 49-57; Thomas of Burton, Chronica monasterii de Melsa a fundatione
usque ad annum 1396, ed. E.A. Bonds (London, 1868), II:329-32; Thomas Walsingham,
Historia anglicana, ed. H.T. Riley (London, 1863), I:139-41; Adam Murimuth, Continuatio
chronicorum, ed. E.M. Thompson (London, 1889), pp. 20-21; Henry Knighton, Chronicon, ed.
J.R. Lumby (London, 1889), I:409; The Anonimalle Chronicle, 130 7 to 1334, ed. W.R. Childs
and J. Taylor (Leeds, 1991), pp. 86-89; Eulogium historiarum sive temporis, ed. F.S. Haydon
(London, 1863), III:194-95; and The Brut, or the Chronicles of England, ed. F.W.D. Brie
(London, 1908), I:207-08.
12 Laurence Minot, Poems, ed. T. Hall (Oxford, 1897). pp. 4-6; 'The Battle of Bannockburn,' in
Political Songs England, ed. T. Wright (London, 1839), pp. 261-67; and W.D. Macray, ed.,
'Robert Baston's Poem on the Battle of Bannockburn,' English Historical Review 19 (1904), 507-
08.
13 Thomas Gray, Scalachronica, ed. J. Stevenson (Edinburgh, 1836), pp. 140-43.
14Vita Edwardi II, p. 54: 'O gens inclita multis retro temporibus inuicta, cur fugis pedites que
uincere solebas equites? Apud Berewyke, Dounbar et Foukyrk, triumphum reportasti, nunc Scotis
peditibus terga dedisti.'
Page 70
brought upon him and his man by the same king, prepared new wars to punish the said acts.15

It was an impressive force, filled with 'a large number of worthy men,' according to
John Barbour. And while the exact number can never be knowncontemporary
chronicles exaggerate the total, placing the it between 100,000 and 340,00016it may
have equalled the English army which invaded France in 1340 both in size and noble
leadership. John of Trokelowe notes that 'all who owed him [Edward II] military
service' were assembled at Berwick for the Scottish invasion by the king.17 Indeed, of
all the leading English barons and knights, only four earls were absent from the force:
Lancaster, Warren, Arendale, and Warwick.18
Perhaps most significant for this study was the large number of cavalry troops
included among the English army. Of the 100,000 troops which John Barbour counts,
40,000 are cavalry with 3,000 of those equipped with horses covered with armor; of
the 340,000 Walter Bower tallies, 300,000 were cavalry.19 These are definite
exaggerations, but they do indicate the opinion of contemporaries that the English
force was composed of many horsed warriors. Whatever conflict this army was to
fight would undoubtedly be cavalry-based. John Barbour writes: 'There were so many
knights, that they believed that if they were to do battle, there would be no strength
which could withstand their might.'20
Justified or not by their large numbers, the English pride is remarked on by all
sources, English and Scottish. An anonymous poet remarks simply: 'There were in the
army many nobles, knights who were too showy and pompous.'21 The author of the
Vita Edwardi II is more eloquent:
There were in that company quite sufficient to penetrate the whole of Scotland, and some
thought if the whole strength of Scotland had been gathered together,
15 Walter Bower, VI:350-51: 'At Eadwardus novus rex acriori ira accensus audiens illustres
actus regis Roberti, perpendensque innumera mala et infinita dampna sibi et suis per eundem
regem illata, in vindictam premissorum nova bella parabat.' See also John Barbour, II:257-58;
Thomas Gray, p. 140; 'The Battle of Bannockburn,' p. 262; John Fordun, I:346-47; and Liber
Pluscardensis, I:237.
16 The tallies include: 100,000 (John Barbour, II:261); 300,000 (Andrew of Wyntoun, Brevis
cronica, III:336; Liber Pluscardensis, I:183; and a poem appended to Walter Bower, VI:357); and
340,000 (Walter Bower, VI:361). The Vita Edwardi II (p. 50) is much more realistic in its tally,
but only includes a count of cavalry, 2,000+, and records no figure of infantry or archers. Modern
historians have considerably decreased the numbers of English soldiers. Morris, for example, sees
only 17,500 troops with Edward II (Bannockburn, p. 41). Barron (pp. 430-31) tallies 17,000.
And MacKenzie (pp. 22-30) counts 20,000. Barrow (pp. 205-06) agrees with Morris.
17 John of Trokelowe, p. 83. This statement was used also by Thomas Walsingham, I:139.
18 John Barbour, II:260-61; John of Trokelowe, p. 83; Chronicon de Lanercost, p. 224; Vita
Edwardi II, pp. 49-50; Thomas Walsingham, I:139-40; and Robert Baston, p. 508.
19 John Barbour, II:262-64 and Walter Bower, VI:361. Modern historians have also decreased
this number. Barrow, for example, sees no more than 7,000 to 10,000 soldiers in the Scottish army
(pp. 208-09). Barron (pp. 432-33) and Mackenzie (pp. 31-32) give a smaller number of 7,000.
20 John Barbour, II:264: 'Thai war so cheuelrus, that thai / Trowit, gif thai corn to the ficht, / Thair
suld no strynth with-stand thar mycht.'
21 'The Battle of Bannockburn,' p. 263: 'Erant in excercitu plures generosi, / Milites in exitu nimis
et pomposi.'
Page 71
they would not have stayed to face the king's army. Indeed all who were present agreed that
never in our time has such an army gone forth from England.22
Although they perhaps could not anticipate the extremely large outpouring of English
support for the relief of Stirling Castle, the Scots had had a long time to prepare for
this invasion. Like Edward II, Robert Bruce also made a recruitment call to his
countrymen, encouraging them to fight 'manfully' for the 'fredome of this cuntre.'23
'The trumpet was sounded,' writes Robert Baston, and troops came to the sound.24
Weapons and armor were also gathered and manufactured.25
But the Scottish numbers would not, and perhaps in agreement with the author of the
Vita Edwardi II, could not equal those of the English force. Only the Scottish
chroniclers seem concerned with the strength of the Scottish army, with John Barbour
and Andrew of Wyntoun alone assigning them a number, 30,000, far smaller than their
numbers for the English army100,000 and 300,000 respectively.26 John Fordun and
Walter Bower are content simply to describe them as 'few men,' while the Liber
Pluscardensis uses a comparative tone: 'King Robert [had] an army small in
comparison to the multitude of the said king of England.'27
But the defending troops were very spirited, ardently willing to fight an invading army
in defense of their own lands. John Barbour describes at great length the religious
devotion of the Scots, an assertion echoed by the English Thomas of Burton. Burton,
a monastic chronicler who was not a supporter of Edward II, writes that while the
English were proud, 'confident in their strength and numbers,' the Scots were 'humble,
confessing, armed with the Eucharist, and placing their trust in God alone.'28
They also had good leaders, far more experienced in warfare through their years of
conflict in Scotland and on the English marches than their counterparts.29 Particularly
imposing in his military leadership abilities was Robert Bruce. By 1314 he was a hero
to his troops and a legend to his enemies. His presence alone could
22Vita Edwardi II, p. 50: 'Fuerunt in societate illa satis sufficientes ad penetrandum totam
Scotiam, et iudicio aliquorun, si tota Scotia collecta fuisset in unum, non exspectaret regis
exercitum. Revera hoc fatebatur tota comitiua, quod tempore nostro talis exercitus non exiuit ab
Angliam.' See also Geoffrey le Baker, p. 7 and Walter Bower, VI:351-52.
23 John Barbour, II:259 and Thomas Gray, p. 140.
24 Robert Baston, p. 508.
25 John Barbour, II:260.
26 John Barbour, II:266-67 and Andrew of Wyntoun, Brevis cronica, III:336.
27 John Fordun, I:347; Walter Bower, VI:353; and Liber Pluscardensis, I:237. The only English
chronicler reporting a description of the Scottish army is the Anonimalle Chronicle which refers
to them as a 'grant poer' (pp. 88-89). However, it is difficult to understand how the chronicler uses
the adjective 'grant' in describing the Scots as he follows this statement with a note that they
'strongly' fought the English in the battle.
28 John Barbour, II:271-73, 276 and Thomas of Burton, II:331: 'Angli vero summe elati, in viribus
et multitudine confidentes, Scotti vero contriti, confessi, viaticoque muniti, solum Deum
protectorem acclamantes.'
29 On the Scottish leaders see John Barbour, II:265-66.
Page 72
make up for some of the numerical advantage of the English, especially as they were
led by the 'inept' Edward II, to use Thomas of Burton's term.30
As there was to be no surprise to Edward's invasion, the English army moved with
much ostentation and violence through Scotland to Stirling Castle.31 With it was a
large logistical train, containing many wagons filled with armor, tents, furniture, wine,
wax, and food.32 But this did not slow it down much, as the army made rapid
progress, so rapid, insists the author of the Vita Edwardi II, that it was quite fatigued
by the time it arrived at Bannockburn; thus the Scots were at an advantage in the
battle:
Brief were the halts for sleep, briefer still for food; hence horses, horsemen and infantry were
worn out with toil and hunger, and if they did not bear themselves well it was hardly their
fault.33
While the English had to travel a large distance to arrive at the battlefield, Robert
Bruce's army was much closer. This enabled Bruce to locate a good battlefield site
and, as he had done at Loudon Hill, to prepare it to his army's advantage. Bruce's site
was known as New Park, a flat field bordered by trees and marshes through which the
English had to march on their way to Stirling Castle.34 According to John Barbour,
Robert Bruce needed the flat field for his defensive infantry formationif the Scots
fought on horseback, Bruce knew that the English would defeat them. However, the
Scottish leader also knew that Edward II would try to break his infantry line with
cavalry charges, and that a flat field would facilitate those charges unless some
modifications could be made to it.35 Ditches to narrow a cavalry charge, like those
that the Scots had dug at Loudon Hill, were not needed at Bannockburn, as the
bordering trees and marshes served that function. Yet, smaller ditches and pits dug in
the battlefield proper would serve to disrupt the cavalry charges, disordering them and
resulting in a loss of impact when the horses reached the infantry line. So they were
dug. Geoffrey le Baker describes these battlefield hindrances:
. . . the Scots selecting the battlefield for the greatest possibility of victory, dug extended
ditches three feet deep and three feet wide from the right to the left flanks of the army, filling
them with a brittle plait of twigs, reeds, and sticks, that is a 'trellis,' and covering them with
grass and weeds. Infantry might be aware of a safe passage through these, but heavy cavalry
would not be able to pass over them.36
30 Thomas of Burton, II:329-30. See also John Barbour, II:270-71; Walter Bower, VI:357; and
Henry Knighton, I:409
31Chronicon de Lanercost, p. 225; John Barbour, II:261-64; and Walter Bower, VI:352-53.
32 John Barbour, II:261-62 and Vita Edwardi II, p. 50.
33Vita Edwardi II, p. 51: 'Breuis erat mora capiendi sompnum, sed brevior erat mora sumendi
cibum; unde equi, equites et pedites, labore et fame fatigati, si minus bene rem gererent non erant
culpandi.'
34 John Barbour, II:268. There is some dispute on the site of this battle. For descriptions see
Morris, Bannockburn, pp. 60-66; Barrow, pp. 211-25; and Oman, II:85-86.
35 John Barbour, II:268.
36 Geoffrey le Baker, pp. 7-8: ' . . . Scoti, campi locum nacti victoribus maxime oportunum,
subfodiebant ad mensuram trium pedum in profundum et ad eiusdem mensure latitudinem fossas
protensas in longum a dextro in sinistrum cornu exercitus, operientes illas cum plexis fragilibus ex
virgulis et
(footnote continued on next page)
Page 73

Battle of Bannockburn
(First Day)
Walter Bower gives it a class interpretation; lesser individuals were able to defeat their
betters by using a simple 'machina plena malis':
A contrivance full of evils is fashioned for the feet of horses. / trenches set with stakes, so that
they cannot pass without tumbles. / The ordinary folk have dug the pits so that the cavalry may
stumble over them.37
With the ditches and pits dug, and hearing the approach of the English army, Robert
Bruce, who had previously split his troops into four divisions, under the command of
Thomas Randolph (Earl of Moray), Edward Bruce, Walter Stewart/ James Douglas,
and Bruce himself respectively, ordered them in their planned infantry formation. It
was a solid formation which stood ready for battle, waiting for someone to attack
them, states John Barbour.38
However, there was a delay in the English advance. Although their approach had been
apparent to the Scots since they had entered Scotland, as they neared the
(footnote continued from previous page)
viminibus sive cratibus, id est ''herdeles", cespite et herbis superstratis, peditibus quidem
perviis saltim consciis cautele, set equitum pondera non valentibus sufferre.' See also John
Barbour, II:271-72. A discussion of the nineteenth-century excavation of some of the
Bannockburn pits and ditches can be found in Morris, Bannockburn, pp. 62-63.
37 Walter Bower, VI:370-73: 'Machina plena malis pedibus formatur equinis, / concava cum
palis, ne pergant absque ruinis. / Plebs foveas fodit ut per eas labantur equestres.'
38 On Bruce's troop division see John Barbour, II:269-70. (Douglas and Stewart combined
leadership of the third division.) On the Scots order before battle see Barbour, II:273-74.
Page 74
battlefield they stopped. Why this occurred has always been somewhat of a mystery.
No original source discusses the reasons for the English delay, although Barbour gives
the impression that Robert Bruce was surprised by the lack of an attack, sending two
of his leaders, James Douglas and Robert Keith, to reconnoitre the enemy force. They
reported only their amazement at the size and splendor of their foe. (Bruce had them
lie to his troops for fear that they might flee when told what they had to face.39)
It was late in the day, June 23, perhaps too late for the English troops to advance on
the Scottish position. But if this was the reason for the lack of a prompt English
attack, Edward II blundered by sending a small number of troops (800) under the
leadership of Lord Robert of Clifford and Henry of Beaumont, a 'company,' to use
Barbour's term, to relieve the siege of Stirling Castle.40 Why he did this is yet another
mystery. Perhaps Edward believed that this small force could slip past the Scots, 'for
they knew well where the King [Robert Bruce] was,' and give relief to the castle. If
this was what he thought, however, the English king was gravely mistaken. For as the
company proceeded towards the castle, Thomas Randolph's division rushed from the
woods (his position in the main Scottish line was bordering the woods) and attacked
them.41 Randolph had only 500 men, but he was prepared to fight the larger force.
Still, the size differential was significant, and the Scots immediately ordered
themselves in a defensive formation, setting up 'back to back with their spear points
outward,' as the English troops moved in to surround them. Although outnumbered,
the formation was solid and strong. The English quickly began to take casualties,
including Sir William Dancourt, one of the noble leaders, as the Scots speared both
men and horses.42 Thus the English were forced to hold back from their accustomed
hand-to-hand combat and to throw 'spears, darts, and knives' into the Scottish line;
when these were gone, they even threw their maces and swords, all in an effort to try
and disrupt or break the Scottish formation. This stalemate continued for a long time,
the English surrounding an unwavering Scottish line. Finally, John Barbour reports
that both sides began to suffer from the heat of the day: 'On both sides they became so
fatigued / from the fighting and the sun's heat, / that their flesh became wet with
sweat.'43 The dust and smell also rose and blinded all in combat. But neither side
withdrew.
Seeing Randolph's troops surrounded and fatigued, James Douglas approached
39 John Barbour, II:274-76.
40 John Barbour (II:277-82),Thomas Gray (pp. 141-42), and Chronicon de Lanercost (pp. 225-
26) are the only original sources which discuss this initial phase of battle. (Other sources pick up
the story when the two main armies enter the fray.) Gray differs from Barbour in that he records
the movement of the English vanguard against the main Scottish force before the encounter
between Randolph and the English. The Chronicon de Lanercost gives no chronology. Despite
Gray's father, also named Thomas, being captured in this battle, I have accepted Barbour's
chronology because of the greater intricacy of his account.
41 The Chronicon de Lanercost (pp. 225-26) contends that Edward II knew that the Scots were in
the woods and that Clifford's expedition was in part trying to draw them out.
42 Thomas Gray (p. 142) reports that his father was captured at Bannockburn when his horse was
speared and he was brought down.
43 John Barbour, II:280: 'On athir half thai war so stad, / For the [rycht] gret heit that thei had, / Of
fechting and of sonnys het, /That all [thair] flesche of swat wes wete.'
Page 75
Robert Bruce and asked permission to enter the battle, hoping that such an action
might relieve the beleaguered Scottish line. Initially, Bruce refused Douglas' request,
fearing that a weakened main force might provoke the attack of Edward II, but when
Douglas repeated his request, the Scottish king gave in.44 Douglas' entry into the battle
moved it into a second, more violent phase.
Douglas' movement was noticed by the English troops, and it excited them. Barbour
insists that Edward II did not favor a fight that late in the day, early evening. But the
vanguard, under the leadership of Gilbert of Clare, the earl of Gloucester, and
Humphrey of Bohun, earl of Hereford, refused, or at least may have been too far
separated from the main body of English troops, to heed the king's words, for it
rushed after Douglas' soldiers.45 Perhaps, the Vita Edwardi II insists, the English
vanguard misinterpreted Douglas' advance in relief of Randolph's force as a flight
from the battlefield, and wishing to take advantage of the situation charged in pursuit;
the Scots being on foot and the English on horse made this a viable option.46 Thomas
Gray, however, maintains that the English vanguard was merely advancing onto the
battlefield without attempting anything against Douglas' force.47
Whatever the reason for their action, what the English troops found, to what most
chroniclers claim was their surprise, was the main Scottish body ready to fight against
them. Led by Robert Bruce himself, 'with his own axe in hand,' the remaining two
divisions of Scottish troops immediately broke from their formation and attacked the
surprised and somewhat trapped English vanguard.48 Seeing Robert Bruce on the
battlefieldhe wore a distinguishable crown on his bascinetgave at least one English
knight, Sir Henry of Bohun, the opportunity for heroism, as he rushed to strike the
Scottish leader. The two met: Bohun slashed with his weapon and missed; Bruce
buried his axe in Bohun's helmet, killing his foe with a blow so powerful that it broke
the handle of the axe. This display, Barbour reports, which was seen by almost all of
the troops on both sides, caused the Scots to fight with more enthusiasm and the
English to retreat from the field. The Scots pursued their foes, but most of the English
were able to escape.49
While this was going on, Randolph's force was still engaged with Clifford's. Douglas'
men approached the conflict to relieve it, but their presence alone seemed to turn the
tide of the battle in favor of the Scots. Indeed, John Barbour asserts that Douglas'
troops never even entered the combat, for once the English saw their approach, they
wavered and made an opening in their encirclement of Randolph's division. This gave
Randolph the advantage, and with Douglas halting his troops so that he might not
distract from the honor of victory Randolph deserved 'it would be a sin for him to
lose his well-earned honor'the Scots increased their fighting
44 John Barbour, II:281-82.
45 John Barbour, II:283.
46Vita Edwardi II, p. 51.
47 Thomas Gray, p. 141.
48 John Barbour, II:283-84.
49 John Barbour, II:284-86; Thomas Gray, p. 141; Vita Edwardi II, p. 51; and John Trokelowe, p.
84.
Page 76
intensity and broke through the surrounding English army, sending it in rout. Many on
both sides had been slain, but the English losses numbered many more.50
The Scots rejoiced in their victory. Although the main English body as yet had not
been encountered, the Scots had been victorious in both of the small conflicts of the
day. Thomas Randolph's troops had seen the most brutal of the fighting, and they
were justly praised for their victorious deeds. As well, Robert Bruce had himself been
embroiled in the combat, and he had proven his fighting capabilities against a
renowned English knight. He was rebuked by his nobles, who felt that he had
needlessly put himself in danger; from his troops he received only increased affection
and loyalty.51 As the Scots encamped for the night, they no longer feared the
upcoming day and its inevitable fight with the main body of English troops. They had
the assurance that God was on their side, and when Robert Bruce spoke to them,
warning them to be cautious against pride and asking them to be prepared for what
they would encounter the next day, he repeated the advantages that the Scots had over
the English: they had right on their side; they would gain much from the capture of
English riches; and they were fighting for their wives, children, and freedom. But they
would need discipline in order to win. Bruce believed that the English would try
cavalry charges against the Scottish infantry line, and this meant that the Scots must
defend themselves strongly and boldly with their spears. They must also not break for
looting until the battle was over. Finally, he assured those fighting that if they were to
die in the battle, their heirs would have immediate possession of their property
without fines, for honor required this. He concluded his speech by asking his soldiers
to remain prepared for battle throughout the night; they slept armed and in order.52

There was no joy in the English camp. Most had seen Gloucester and Hereford's
defeat, and when Clifford's return brought news of his lack of success, the English
troops became extremely downcast and disheartened.53 Some even expected the Scots
to attack during the night.54 Their feelings seem to have been expressed in Walter
Bower's recorded words of an English poet captured at the battle:
Ill-omened Sunday opens the preliminaries of the disaster / befalling the sons of England; hence
issues discord from its very mouth. / The dry ground of Stirling sustains the first conflicts.
/Splendid is the attacking host, but soon it takes a downward turn. / Great is the grief, grief
enhancing grief; / fierce is the frenzy, frenzy inflaming frenzy; / louder grows the clamour,
vanguard assailing vanguard; / feeble the valour, valour foiling valour; / fierier the ardour,
ardour firing ardour; / more hesitant the fighters, waverer reproaching waverer; / bewilderment
is audible, bewilderment redoubling bewilderment; / resistance is worn down, order
50 John Barbour, II:286-88.
51 John Barbour, II:288-89.
52 Robert Bruce's speech to his troops is recorded in John Barbour, II:289-95.
53 John Barbour, II:295-96.
54Vita Edward II, p. 51.
Page 77
losing order; / uproar uprises, blood shedding blood; / now fear is recognizable, fear afraid of
fear.55
The English nobles tried to boost the morale of their troops by declaring that the battle
on the next day, the one in which the main English body would take part, would have
a different result. But their words brought little comfort, as the English army spent a
miserable, sleepless night, made even more uncomfortable by their camp, on what
Thomas Gray describes as an 'evil, deep wet marsh.'56
The Scots began the next day by gaining the defection of Alexander Seton, a
Scotsman who had been before this time in the service of Edward II. According to
Thomas Gray, the only chronicler who reports his defection, Seton left the English
when he had witnessed their discouragement in losing to the Scots on the previous
day. This discouragement he reported to Robert Bruce; he also reported that the
English were planning a sudden, open attack.'57
No doubt Seton's news further excited the Scottish troops, and, although it probably
did not set the tactics which Robert Bruce had planned in fighting the English that day,
it certainly confirmed his plan. He would order his troops in an infantry formation and
prepare for English cavalry charges. The Chronicon de Lanercost describes the
Scottish formation as one of two lines and a rearguard; all other sources simply report
three lines. All commentators maintain that only infantry formed these lines. John of
Trokelowe and Thomas Walsingham claim that Bruce used only infantry because he
feared that if he were to use cavalry as the Scots had at Falkirk, the same result,
defeat, would occur. But Trokelowe also adds that the Scottish king himself was
dismounted, with his other leaders, 'so that danger having been equalized between the
nobility and the commoners no one thought about flight.'58 For whatever reason, the
Scots looked impressive in their infantry formation. They were well armed, notes the
Vita Edwardi II: 'each was furnished with light armor, not easily penetrable by a
sword. They had axes at their sides and carried
55 Walter Bower, VI:372-73: 'Dira dies solis pandit premordia molis / Angligene prolis; hinc
exit ab ore suo lis. / Arida terra gerit Strivelini prelia prima; / Splendida turba ferit, sed
tandem tendit ad ima. / Est dolor immensus, augente dolore dolorem; / est furor accensus,
stimulante furore furorem; / est clamor crescens, feriente priore priorem; / est valor arescens,
frustrante valore valorem; /est calor ardescens, urente calore calorem; / est gens demescens,
reprobante minore minorem; / est stupor auditus, geminante stupore stuporem; / est populus
tritus, perdente tenore tenorem; / surgit rugitus, fundente cruore cruorem; / nunc timor est scitus,
metuente timore timorem.' According to Bower, this English poet, 'the most famous poet in the
whole kingdom of England,' a Carmelite friar, had been taken by Edward II to Bannockburn
because he was certain of victory there. He had been captured after the battle and forced to
compose a poem on the battle in order to gain his release (VI:367).
56 Thomas Gray, p. 142. See also John Barbour, II:296-98 and Vita Edwardi II, p. 51.
57 Thomas Gray, p. 142.
58 John of Trokelowe, p. 84: 'Robertus vero le Brois, qui se Regem Scotiae clamitabat, pedes,
cum suis consortibus, totum exercitum suum praecedebant, ut sic, periculo inter majores et
minores coaequato, nemo de fuga cogitaret,' and Thomas Walsingham, I:140. On the Scottish
formation see also Chronicon de Lanercost, p. 225; Vita Edwardi II, p. 52; Geoffrey le Baker, p.
8; and Thomas Gray, p. 142. Thomas Gray (p. 140) uses the word 'schiltrom' to describe the
Scottish formation. However, John Barbour (pp. 299-300, 312) uses the same term to describe the
English formation.
Page 78

Battle of Bannockburn
(Second Day)
lances in their hands.'59 John of Trokelowe insists that the formation would be
difficult to defeat: 'These elect men, certainly spirited, set up an impenetrable line
armed with sharpened axes and other arms of war and with shields placed thickly
among them.'60 Before this formation, Robert Bruce unfurled his banners and set
them among the troops. He also knighted two men who had displayed valor on the
day before, Walter Stewart and James Douglas. Finally, he led his men in prayer.61
At the same time, the English also prepared for battle. However, their array seems to
have been more sloppy than the Scottish one. Using the term 'schiltrom,' John Barbour
records that the English were crowded together in a group, although whether this was
because of the terrain on which they were ordered or out of fear, the Scottish writer
does not know. Barbour does report that the English vanguard was well ordered and
anxious for conflict. Still, even without the harmony of the Scots, the English army
was imposing, their armor shining 'like angels,' and they were numerous, covering a
large amount of ground.62
There were two distinct groups of English soldiers. One consisted of infantry and
archers, with the other of cavalry. Not yet mounted, all sources agree that the cavalry
59Vita Edwardi II, p. 52: 'sed erat unusquisque eonun leui armatura munitus, quam non faciliter
penetraret gladius. Securim habebant ad latus et lanceas ferebant in manibus.'
60 John of Trokelowe, p. 84: 'viri quidem electi, animosi valde, securibus praeacutis caeterisque
armis bellicis decenter communiti, conserta ante se scutorum spissitudine, cuneum impenetrabilem
statuerunt.'
61 John Barbour, II:298-301 and Chronicon de Lanercost, p. 225.
62 John Barbour, II:299-300 and Geoffrey le Baker, p. 8.
Page 79
was to be the primary attackers in the battle, for, in the words of Thomas Gray, 'they
were not accustomed to fight on foot.'63 The English too were well armed and
anxious for the battle to begin.64
However, before the English could attack, Edward II called an impromptu council of
war. The reason, Barbour decides, was because the king was surprised at the Scottish
infantry formation, expecting it seems a cavalry array. 'How will we fight the Scots?'
he is reported to have said. One English knight, Sir Ingraham Umphraville, proposed
that the army retreat behind their camp, as if in flight, causing the Scots to break ranks
to loot the English tents. Then the more numerous English troops could sweep in and
easily defeat their opponents. Edward II dismissed this plan, as he refused to retreat.
Moreover, he had just noticed the Scots kneeling in prayer and believed that they were
kneeling for mercy. Thus he proposed an immediate cavalry charge, a tactic objected
to by both Umphraville and Gloucester. Ignoring their protestations, Edward mounted
his own horse and joined his cavalry line.65
Both sides now awaited the first attack. John of Trokelowe describes the action:
Both armies prepared for battle and at about the third hour of the day the lines of both parties,
approaching each other, showed a formidable sight to those who saw them. For the din of the
trumpets and clarions, the neighing of the horses, the motion of the standards, the calls of the
leaders which sounded in their midst, could frighten the hearts of even the bravest men.66
There may also have been an exchange of archery fire, short in duration and English
dominated, but only the Chronicon de Lanercost reports it.67
Who initiated the attack between the two main armies on June 24 is in dispute among
the original sources. John Barbour, John of Trokelowe, Walter Bower, and the
Chronicon de Lanercost all contend that it was the English cavalry who first charged
into the Scottish formation,68 while the Vita Edwardi II indicates a Scottish charge:
They advanced like a thick-set hedge, and such a line could not easily be broken. When the
situation was such that the two sides must meet, James Douglas, who commanded the first line
of the Scots, vigorously attacked the Earl of Gloucester's line.69
63 Thomas Gray, p. 142: 'qi nestoint my acoustomez pur descendre a coumbatre a pee.' See
also John Barbour, II:299-300; Geoffrey le Baker, p. 8; Thomas Walsingham, I:140; and Walter
Bower, VI:369. Walsingham orders the infantry in front of the cavalry, while Baker has the
horse before the infantry.
64 Walter Bower, VI:369.
65 Only John Barbour reports this council (II:300-02), with the Vita Edwardi II (p. 52) the only
chronicle noting the disagreement of Gloucester.
66 John of Trokelowe, p. 84: 'Uterque exercitus ad pugnam se parabat, et circa horam diei tertiam
acies utriusque partis, sibi invicem appropinquantes, formidabile spectaculum intuentibus
ostendebant. Clangor enim tubarum et lituorum, hinnitus equorum, motio vexillorum, clamor
exhortantium, qui in eorum congressu resonabant, corda poterant terruisse magnanimorum.'
67Chronicon de Lanercost, p. 225.
68 John Barbour, II:302; John of Trokelowe, p. 84; Walter Bower, VI:373; and Chronicon de
Lanercost, p. 225.
69Vita Edward II, p. 52: 'Ibant etiam quasi sepes densa conserti, nec leuiter potuit talis turma
(footnote continued on next page)
Page 80
Why one of the most contemporary chronicles would disagree both with these other
accounts and with Robert Bruce's standard infantry tactic, as shown both the day
previously in Randolph's stand and at Loudon Hill, is uncertain. However, the
anonymous author of this chronicle could be simply in error, especially as he will later
compare the battle to that fought by the Flemings at Courtrai.70
The first English attack struck the Scottish line with great impact but with no
penetration. The battle continued for a very long time. John of Trokelowe recounts
the scene:
On one side stood the magnates of England advancing with their troops strongly against the
Scots. On the other side, the leaders of the Scots stood defending themselves strongly. And the
troops clashing against each other fought a most bloody battle. The crash of the lances, the
ringing of the swords, the noise of hasty blows, the groan of the dying, the lamentation of the
wounded, being heard in this conflict seemed to split the air. For a long time it was fought at the
front of the lines with the ringing of many swords and both armies fighting strongly against each
other.71
Many English charges were made, and at times it seemed that the Scottish line would
break under the fierce attacks of the cavalry. Still, there was no penetration of the
solid infantry formation.72 The Scots fought 'with all their might and all their means;
they attacked like madmen,' writes John Barbour.73 Even the Scottish camp followers
(footnote continued from previous page)
penetrari. Cum autem ad hoc uentum esset ut congredi simul oporteret, Jacobus Douglas, qui
prime turme Scotorum preerat, aciem comitis Gloucestrie acriter in uasit.' N. Denhohn-Young's
translation of 'turma' as phalanx makes little sense in a medieval military context. I have chosen
to use the more generic 'line' in my translation.
70Vita Edwardi II, p. 56. There may be more to this account when we look also at Thomas Gray's
narrative of the battle. Gray's description of the charge in its original Old French language reads:
Lez auaunt ditz Escotez uindrent de tot aleyn en schiltrome, assenblerent sur lez bataillis dez
Engles, qi entassez estoint, qi rien remuerent deuers eaux tanque lours cheueaux estoient
enbuaillez dez launcez lez gentz dereir dez Engles (p. 142).
This has been translated by Herbert Maxwell as,
The aforesaid Scots came in line of 'schiltroms,' and attacked the English columns, which were
jammed together and could not operate against them [the Scots], so direfully were their horses
impaled on the pikes, (Glasgow, 1907, pp. 55-56)
and thus seems to agree with the Vita Edwardi II. But I think that Maxwell has mistranslated this,
that it is not the English which are jammed together, but the Scots.
71 John of Trokelowe, pp. 84-85: 'Stant hinc magnates Angliae, cum suis turmis Scotos fortiter
insilientes. Stant inde duces Scotorum, sese viriliter defendentes. Et collidentibus ad invicem
catervis, bellum cruentissimum commiserunt. Fragor lancearum, tinnitus gladiorum, ictuum
strepitus repentinorum, gemitus morientium, vulneratorum lamentatio, in ipso conflictu audita, aera
perturbare videbantur. Multo tempore nimis gladiis circa capita tinnientibus altercatum est, et
decertantibus utrinque viriliter partibus ex adverso.' See also John Barbour, II:302, 312;
Chronicon de Lanercost, p. 225; and Thomas Walsingham, I:140.
72 See, for example, John Barbour, II:303-05.
73 John Barbour, II:311: 'With all that mycht and all that mayne / Thai layd on, as men out of wit.'
On the battle in more detail see Barbour, II:302-14.
Page 81
chose a captain and joined in the attack.74 Walter Bower's English poet puts his report
in verse:
Scots no longer remote but close at hand. / The ordinary folk bang and clang; but, when the
impact touches them, / they beat their breasts. Battered by blows, their morale breaks. / The
overweening enemies of the Scots are shrunken.75
English and Scottish archers also engaged in the battle. But they had very little effect,
at least according to John Barbour, who is the only source which mentions their
participation in detail: the Scottish archers only annoyed the English cavalry, while the
English archers might have been more effective had not Robert Bruce sent a
contingent of horsed warriors under the command of his marshal, Robert Keith, to
successfully disperse them.76
As the battle raged on, the ditches and pits dug by the Scots prior to the fighting
became a factor. The English cavalry initially seems to have passed over these
impediments without any difficulty. But, as they had been unable to penetrate the
Scottish line and began to be pushed back, they also began to fall into the pits and
ditches, 'tumbling one over the other,' notes Thomas Gray.77 One ditch in particular,
after which the battlefield 'Bannockburn' was named, created special problems
according to the Chronicon de Lanercost:
Another misfortune which happened to the English was that, although they at first had crossed a
great ditch called Bannockburn, into which the tide flows, and now in confusion wanted to
recross it, in the press many nobles and others fell into it with their horses. And while some
escaped from it with much difficulty, many were never able to remove themselves from the
ditch. Thus Bannockburn was in the speech of Englishmen for many years after.78
Yet, perhaps the biggest blow to the English attacks came with the death of Gilbert of
Clare, earl of Gloucester. He was young, but his noble rank had given him command
over the vanguard on the day previously, and, although his fighting then had been
unsuccessful, it seems he had retained some leadership responsibilities on the second
day of battle. By losing on the first day, Gloucester had to prove his valor and military
capability, and so he pressed with as much courage and strength as he could muster. It
may have cost him his life. John of Trokelowe writes:
74 John Barbour, II:308-09.
75 Walter Bower, VI:372-73: 'admotos Scotos ab eis non longe remotos. / Plebs plangit clangit;
sed quam congressio tangit, / nunc pangit. Frangit vires quas ictibus angit. / Magnifici modici
Scotorum sunt inimici.'
76 John Barbour, II:307-09. See also Geoffrey le Baker, pp. 8-9. The existence of Scottish cavalry
obviously goes against the contention of most sources that all of the Scots were dismounted.
77 Thomas Gray, p. 142.
78Chronicon de Lanercost, p. 226: 'Aliud etiam infortunium accidit Anglicis, quia, cum paulo ant
transissent unam foveam magnam, in quam intrat fluxus maris, nomine Bannokeburne, et jam
confusi vellent redire, multi nobiles et alii prae pressura cum equis in illam ceciderunt, et aliqui
cum difficultate magna evaserunt, et multi nunquam se explicare de fovea potuerunt; et ideo
Bannokeburne in ore Anglicorum erat per multos annos sequentes.'
Page 82
Many fell having been fatally wounded. When Gilbert, count of Gloucester, saw this he was
very angry on account of the enemy's fierceness; he urged his comrades to rush in on the
effeminate Scots unaccustomed to fighting the gathered English; and wishing to win a name for
himself, he presented an example to them as a bellicose boar. In the heat of anger he attacked
the troops of the enemy with their blood inebriating his sword. For whomever he struck with
his sharp blade, he cut off a head or some other limb. Finally, thirsting so for their deaths, the
equilibrium of the entire battle was turned around to such a degree that the points of the lances
being applied to each part of his body, stabbing several places, he was knocked to the ground,
and his head was struck on all sides by the clubs of the enemy, until he breathed out his soul
under the horse's feet.
Seeing the death of one of their important leaders, English morale was broken and
several soldiers took flight. John of Trokelowe concludes: 'Seeing him thus to be
killed, all of the rest of his army, gripped with fear, fled, alas, leaving their lord
slaughtered on the battlefield.'79
Soon all of the English troops were fleeing from the Scots, leaving only a few valiant
soldiers to fight on in vain.80 Ultimately, even the king, Edward II, fled, having shown
his own expertise on the field. The Scots pursued him and all others who had not
been captured or killed; although the king escaped with difficulty, many others could
not and added their numbers to the Scottish prisoners.81 A large number of English
lay dead and dying on the battlefield. Only John Barbour reports a total number killed,
30,000,82 with most other sources which tally the deaths concerned only with the
number of nobles killed, a number which ranges from 154 to 700.83
79 John of Trokelowe, p. 85: 'Quod cum videret G[ilbertus] Comes Gloverniae, indignatus est
valde super eorum feritate; suos commilitones exhortatur, ut in Scotos effoeminatos,
congressiones Anglicanas expectare non solentes, irruant truculenter; et, nomen militiae sibi
volens acquirere, exemplum suis in seipso praetendens, ut aper bellicosus, in incendio irae
suae turmas hostium invasit, gladium suum cruori eorum inebriando. Que [m] cunque enim
mucrone suo attingebat, caput ei, aut membrum. aliquod, amputabat. In ipsum tandem, eorum
necem ita sitientem, pondus totius proelii conversum est; adeo ut lancearum cuspidibus ex omni
parte sui corporis applicatis, impingentibus hinc inde diversis, ad terram prosternitur, et caput
ejus clavis hostium undique malleatur, donec animam, cum sanguine, sub pedibus equinis
exhalaret. Reliqui vero omnes de exercitu suo, videntes eum sic occubuisse, timore perterriti,
fugam inierunt, dominum suum, in campo trucidatum, proh O dolor! relinquentes.' On
Gloucester's death see also Geoffrey le Baker, p. 8; Vita Edwardi II, pp. 52-54; Adam
Murimuth, p. 21; Thomas Walsingham, I:140-41; John Fordun, I:339; Liber Pluscardensis,
I:183; and Walter Bower, VI:375. On his death causing the English flight see Vita Edwardi II,
p. 54. There is absolutely no corroboration for the story recorded in the anonymous poem, 'The
Battle of Bannockburn,' that Gloucester was betrayed by one of his retainers, Bartholomew.
80 John Barbour, II:316.
81 John Barbour, II:316-22, 326-29; Thomas Gray, pp. 142-43; John of Trokelowe, p. 85;
Geoffrey le Baker, p. 9; Vita Edwardi II, pp. 54-55; Anonimalle Chronicle, p. 89; Adam
Murimuth, p. 21; Thomas of Burton, II:331; Thomas Walsingham, I:141; The Brut, I:208; John
Fordun, I:347; and Andrew of Wyntoun, Brevis cronica, p. 336. The ransoms paid to the Scots for
their noble prisoners are mentioned by several authors: Vita Edwardi II, p. 55; Anonimalle
Chronicle, p. 89; Adam Murimuth, p. 21; Thomas Walsingham, I:141; Eulogium historiarum,
III:195; John Fordun, I:347; Liber Pluscardensis, I:238; and Walter Bower, VI:353.
82 John Barbour, II:321.
83 These numbers include: 154 (Thomas Walsingham, I:141); 300 (Eulogium historiarum,
III:195); and 700 (John Barbour, II:323). Other chronicles which mention the large numbers of
English dead
(footnote continued on next page)
Page 83
The Scots' death toll was far less, with John Barbourthe only original source even
mentioning Scottish deadreporting only two slain knights.84
For the Scots the battle of Bannockburn was a sign from God that He was with them
in their search for independence from the English, and that Robert Bruce was His
'divine right' ruler of their blessed land.85 The English chroniclers have more
difficulty in judging why their forces lost. For the Chronicon de Lanercost, it was
Edward II's sinfulness which led to defeat for he 'did and said things to prejudice and
injure the saints.'86 For the Vita Edwardi II, it was the 'proud arrogance of our men
[which] made the Scots rejoice in victory.'87 Only John of Trokelowe looks to what
occurred on the field:
I do not know to which misfortune I should assign the cause of this defeat; unless perhaps
because the English advanced impetuously and more inordinately than was befitting. For their
men and beasts were exhausted on account of their great haste, and they were weak, indeed
made weak from hunger and lacking the revitalization of sleep. For the Scots, awaiting the
attack on their own ground, most speedily hastened to a place of battle known to them and
unknown to the English, revived to a fullness with food as well as sleep, organized for battle in
thick lines and troops.88
Trokelowe is mostly correct in his determination of what caused the Scottish victory at
Bannockburn. It is true that the English advance was impetuous, and that it became
quickly disordered and confused when it was unable to penetrate the Scottish line. It
is also true that the English troops were fatigued and hungry, both on their arrival at
Bannockburn and on the morning of the second day of the battle after spending a
sleepless, uncomfortable night encamped in the marshlands on the edge of the
battlefield.
It is also true that the Scots chose the battlefield and reached it first, were less fatigued
and hungry, and had ordered in thick lines of troops. But here Trokelowe does not
wholly understand how these factors enabled a Scottish victory. His assertions are
simply incomplete. Not only was it important that the Scots selected Bannockburn as
the field of battle because they knew the site and the English did not, but because they
selected a battlefield site which enabled their formation, on flat land, and hindered
their opponent's attacks, with trees and marshes narrowing
(footnote continued from previous page)
but do not give a count are: John of Trokelowe, p. 87; Geoffrey le Baker, p. 8; Vita Edwardi II,
p. 55; Anonimalle Chronicle, p. 89; Henry Knighton, I:409; The Brut, I:208; Liber
Pluscardensis, I:238; and Walter Bower, VI:53, 375.
84 John Barbour, II:323-24.
85 See, for example, Liber Pluscardensis, I:238-39 and Walter Bower, VI:355, 357, 359-63.
86Chronicon de Lanercost, pp. 224-25.
87Vita Edwardi II, p. 56: 'Certe superba nostrorum presumptio Scotos fecit gaudere triumpho.'
88 John of Trokelowe, p. 87: Causam qutem hujus ruinae, cui infortunio imputare valeam, ignoro;
nisi forte, quod Anglici impetuose, nimisque inordinate, quam deceret advenerunt. Homines
enimeorum, et bestiae, prae nimia festinatione lassi erant imbecilles, fame quidem tabefacti, et
somni recreatione carentes. Scoti quidem, adventum eorum in terra propria expectantes, ad locum
belli sibi notum, Anglicisque ignotum, tam cibo quam somno ad plenum refocillati, densis
agminibus catervisque ad proeliandum despositis, maturius accesserunt.'
Page 84
cavalry charges. This was further improved from the Scottish standpoint by the
digging of pits and ditches which may not have hindered the initial attacks of the
English, but did create hazards for the knights and horsemen once they were unable to
penetrate the Scottish lines and began to be pushed back into them. Judging from
most of the sources on the battle of Bannockburn, the ditches and pits caused many of
the English deaths.
Yet, the fact that the Scots were more rested and less hungry than their English
counterparts is only part of the reason for their spirited combat. The original sources,
especially the Scottish authors, all remark on the impressive numbers and nobility of
the English army and that this greatly frightened the Scottish warriors. But the Scots
did not give in to their fears and flee from their foes. Instead, they fought 'like
madmen,' says John Barbour, showing no fear and great morale. For this we must
recognize the leadership ability of Robert Bruce. His participation in the first day of
battle, even facing and killing an English knight, although reproved by his nobles and
other leaders, brought him favor among his troops. So too did his order for the
knights in his army to dismount and fight with the more common people on foot.
And, while there is some dispute among the sources as to where he placed himself on
the second day of battleThomas Walsingham puts Bruce in the front of the Scottish
formation while the Chronicon de Lanercost places him in the rear89there is no
dispute about his active leadership in the fight on that day.90 Moreover, Bruce's
speeches to his troops, primarily the one given by him at the end of the first day of
battle with its praises, cautions, and promises, was a great encouragement in building
his soldiers' morale.
Finally, the thick lines of Scottish troops were perhaps the greatest factor in their
victory. But again Trokelowe leaves out an essential part of the equation in not
including the fact that these were infantry lines standing in defensive formation. This
solid infantry formation, used both by Randolph's force on the first day and by the
main army on the second, was unable to be penetrated and thus presented an
impediment to victory that the English cavalry, no matter how well-armed, how noble,
or how large, was able to overcome. Their attacks proved impotent, and eventually
they were forced to try to save their lives by fleeing from the battlefield. It is no
wonder that both Thomas Gray and the Vita Edwardi II can see the comparison
between this battle and that at Courtrai, for as in the Flemings' victory there, at
Bannockburn the Scottish infantry defeated a more numerous and more noble cavalry
force. As the author of the Vita Edwardi II writes: 'Indeed I think it is unheard of in
our time for such an army to be scattered so suddenly by infantry, unless when the
flower of France fell before the Flemings at Courtrai.'91
Without a doubt Bannockburn was a great blow to the English. The defeat remained in
the thoughts of many English writers for many years, but before too long
89 Thomas Walsingham, I:140 and Chronicon de Lanercost, p. 225.
90 See, for example, John Barbour, II:302-14.
91 Thomas Gray, p. 142 and Vita Edwardi II, pp. 55-56: 'Siquidem a seculo recordor inauditum
talem exercitum coram peditibus tam subito dispersum, nisi cum flos Francie coram Flandrensibus
apud Coutray cecidit.'
Page 85
even this loss was forgotten. The English army would learn valuable lessons from
their defeat at Bannockburn, and the next time an English army faced the Scots, at
Dupplin Moor in 1332, they would not repeat the mistakes of their battle of 1314. Nor
would they repeat them at Halidon Hill in 1333 or at Neville's Cross in 1346 or in any
other battle between the two lands during the remaining centuries of the Middle Ages.
Indeed, John of Trokelowe may have been more of a prophet than he realized when
he concluded his account on the battle of Bannockburn with these words: 'But let this
outcome break no one, for the fates of battles are unknown. For the sword consumes
now these and now those, and thus with fortune turning its wheel, the victory
remained in this one turn to the Scots.'92
92 John of Trokelowe, p. 87: 'Sed neminem frangat ista res; varii enim sunt oventus proeliorum,
et fata eorum ambigua. Nunc enim hos consumit gladius, et nunc illos; et sic, fortuna rotam suam
volvente, victoria. Scotis illa sola vice remansit.' See also John Barbour, II:329-31.
Page 86

VII
The Battle of Boroughbridge, 1322
Two generalizations might be made about medieval rulership: first, strong and weak
rulers are almost always so because of their relative strong and weak military
leadership, and second, weak rulership brought not only attacks from outside, but
also, and sometimes more importantly, attacks from insidecivil war. These may be too
simplistic, but if they are accepted, then there is no better archetype for the weak ruler
than King Edward II of England.
No English king was less able to contend with the problems of his realm than was
Edward II. Already witnessed within these pages was Edward's inability to solve
Robert Bruce's 'outside' warfare, with English armies losing both at Loudon Hill in
1307 and at Bannockburn in 1314. But Edward also had to contend with attacks from
inside his realm, and no better example of this can be found than in the case of the
rebellion of Edward's cousin, the powerful lord and chief counsellor, Thomas, duke
of Lancaster; but in this instance, it was a rebellion which ended in victory for the
king at the battle of Boroughbridge fought on March 16, 1322.
It is unnecessary here to detail the conflict between Edward II and Thomas of
Lancaster, which at times appears to be a roller coaster of favor and disfavor. But a
few introductory comments must be mentioned to put the battle of Boroughbridge in
context. First, it must be noted that this was not the only nor even the first conflict
Edward II had with his nobles. Indeed, the baronial difficulties Edward had to a large
part were inherited from his father, Edward I, and perhaps even from his grandfather,
Henry III, and great-grandfather, John. But none of the difficulties Edward II faced, at
least until he met his end at the hand of his son, Edward III, can compare with the
1322 rebellion led by Thomas of Lancaster.
The Edward/Lancaster conflict dates from the hatred held by Lancaster and most of
the other barons of England towards the king's favorite, the handsome Gascon noble,
Piers Gaveston. Edward's affection towards Gaveston is legendary and needs not be
dealt with here. Suffice it to say, that when Gaveston was finally brought down, in
1312, it was Thomas of Lancaster, together with Guy Beauchamp, earl of Warwick,
who abducted him from his captors, Aylmer of Valence, earl of Pembroke, and John,
earl of Warenne; and it was Lancaster and Beauchamp who killed the king's favorite,
despite an earlier promise of safe-passage to Gascony given him by Pembroke and
Warenne.1
1 J.R. Maddicott, Thomas of Lancaster, 1307-1322: A Study in the Reign of Edward II
(Oxford, 1970), pp. 121-30; J.S. Hamilton, Piers Gaveston, Earl of Cornwall, 1307-1312:
Politics and
(footnote continued on next page)
Page 87
Despite later pardoning them, Edward II never forgave his two barons, but there was
little he could do about the situation. Thomas of Lancaster was perhaps the most
wealthy and powerful of all the barons. Already holding vast lands of his own, in
1311 he had inherited even more when his father-in-law, Henry de Lacy, earl of
Lincoln, died. At the time of Gaveston's murder, Lancaster held no fewer than five
earldoms: Lancaster, Leicester, Lincoln, Derby, and Salisbury.2
Nor does it seem that many other nobles in the kingdom approved of Lancaster and
Warwick's actions in the Gaveston affair. Especially Aylmer of Valence and John of
Warenne, whose promise of safe-passage to Gaveston had been violated by his
abduction and murder, seemed to believe that Lancaster and Warwick had chosen a
more harsh solution to the problem than was necessary3
Relations further soured in 1314 when Lancaster, again with Warwick in tow, refused
the summons to join the king's army against the Scots; their absence in the campaign
to some, including many of the other barons as well as the king, carried part of the
blame for the English loss to Robert Bruce at Bannockburn. Yet, at the same time, the
defeat at Bannockburn meant that there was a need for reconciliation between
Lancaster and the king. To preserve the peace and security of his kingdom, Edward
reluctantly decided to accept his wealthy baron's advice and to name him as his chief
counsellor.4
Lancaster's rise to chief counsellor did not, however, bring an end to his conflict with
Edward II or the other English barons. From 1316 to 1320 Lancaster wielded almost
monopolistic control over the English government. But he was not a good leader, and
his decisions, too often self-serving, continued to alienate the king and almost all of
the other English prelates. Sometimes this governmental turmoil even erupted into
open warfare, and on one occasion, in 1318, a major baronial civil war was averted
only by the intercession of Pope John XXII's papal mediators.5
In 1320 a new baronial contingent arose to oppose Lancaster. It was led by the
Despensers, Hugh the Elder and Hugh the Younger, father and son. Taking a cue from
(footnote continued from previous page)
Patronage in the Reign of Edward II (Detroit, 1988), pp. 93-99; T.F. Tout, The Place of the
Reign of Edward II in English History, 2nd ed. (Manchester, 1936), pp. 87-90; Prestwich, The
Three Edwards, pp. 84-85; McKisack, pp. 24-28; Keen, p. 54; Fryde, p. 22; and Phillips, pp.
31-37, On surrender terms between Pembroke and Gaveston see Phillips, pp. 33-34. The only
modern biography on Piers Gaveston used to be Hamilton, but recently a new biography, Pierre
Chaplais' Piers Gaveston: Edward II's Adoptive Brother (Oxford, 1994), has appeared which
argues against many of Hamilton's theses, but none which affect this chapter.
2 On Lancaster's estates, finances and retinue see Maddicott, pp. 8-66. See also James Conway
Davies, The Baronial Opposition to Edward II: Its Character and Policy (London, 1918), pp.
107-10; Prestwich, The Three Edwards, pp. 85, 91; and Keen, pp. 54-55. On contemporary views
of Lancaster see Fryde, pp. 19-20.
3 Maddicott, pp. 130-31, 149-50, 154-59; Phillips, pp.38-72; Hamilton, pp. 103-07; Prestwich,
The Three Edwards, p. 85; McKisack, pp. 28-31; and Keen, p. 55.
4 Maddicott, pp. 158-61; Tout, Edward II English History, pp. 90-94; Fryde, p. 24; Davies, pp.
412-19; Phillips, pp. 72-82; Prestwich, The Three Edwards, p. 86; McKisack, pp. 39, 45-47; and
Keen, p. 55,
5 Maddicott, pp. 190-208, 213-39; Tout, Edward II in English History, pp. 110-22; Fryde, pp. 20,
25-26; Davies, pp. 502-03; Prestwich, The Three Edwards, pp. 87-89; McKisack, pp. 51-56;
Keen, pp. 62-64; and Phillips, pp. 119-20, 154-77.
Page 88
Piers Gaveston before them (and perhaps from Lancaster as well), the Despensers
began to court the favor of the almost powerless king, who was unable to turn away
from their attendance, especially that of the young and fair Hugh Jr. Edward II began
to favor the Despensers as he had done Gaveston before them, especially when it
came to granting them lands. This, of course, brought immediate enmity from the
other barons of the realm, who conveniently forgot about their dislike of Lancaster
and turned instead to anger against the Despensers. Civil war broke out in April
1321.6
Initially at least, Lancaster remained outside of the anti-Despenser coalition, not
because he favored the Despensers, but because all of his enemies hated them.
However, no major step against them could be made without Lancaster's assistance
and, finally in May, he entered the conflict, calling together the anti-Despenser
coalition who, in their words, 'bound themselves together to preserve peace and to
defend the realm.' Action was taken. On July 15, 1321 they decided to march to
London to remove the Despensers, who fled from the city; on August 19, Parliament
sentenced them in absentia to a forfeiture of lands and banishment as 'evil and false
counsellors, seducers and conspirators, and disinheritors of the crown, and enemies to
the king and kingdom.'7
With the Despensers gone, the wrath of the barons (and also of the king) once again
was directed towards Thomas of Lancaster. Edward was especially incensed, for he
realized that this was the second of his favorites removed from him by Lancaster, and
he began to speak out against his chief counsellor. Finally showing some strength,
baronial adherents began to flock to the king, and Lancaster quickly found that only a
few nobles remained by his side. Now the anti-Lancaster coalition, led by the king
himself, began to march against Lancaster's lands and castles. Leeds, Chirk and
Wigmore fell. The hated duke could only flee to the north, in an attempt to seek
protection from the up-to-now neutral northern nobles or from the Scots
(contemporary historians seem uncertain about this). But instead of gaining refuge, on
March 16, 1322, Lancaster and his band were caught and defeated at Boroughbridge
by Andrew Harclay, the sheriff of Cumberland.8 All noble adherents to the rebellious
duke who survived the battle, including Lancaster himself, were taken to York and
tried for treason by the king. All were found guilty and were sentenced to
6 Maddicott, pp. 259-67; Tout, Edward II in English History, pp. 122-29; Fryde, pp. 37-45;
Davies, pp. 469-72; Phillips, pp. 177-201; Prestwich, The Three Edwards, pp. 89-90;
McKisack, pp. 58-61; and Keen, pp. 66-67. On the rise of the Despensers before 1320 see
Fryde, pp. 27-36.
7 Maddicott, pp. 267-80; Tout, Edward II in English History, pp. 129-30; Fryde, pp. 45-50;
Davies, pp. 472-84; Phillips, pp. 201-11; McKisack, pp. 61-64; and Keen, pp. 67-69. See also
S.L. Waugh, 'For King, Country and Patron: The Despensers and Local Administration,' Journal of
British Studies 22 (1983), 23-45.
8 Maddicott, pp. 292-311; Tout, Edward II in English History, pp. 131-34, 205; Fryde, pp. 50-57;
Davies, pp. 503-04; Phillips, pp. 214-24; Prestwich, The Three Edwards, pp. 90-91; McKisack,
pp. 64-67; and Keen, pp. 69-70. There is some dispute on the spelling of Harclay's name, with
Phillips for one wanting it to be spelled 'Harcla;' all other secondary authors use the spelling
'Harclay,' and thus so have I.
Page 89
death and the forfeiture of lands. Lancaster was beheaded on March 22. Others met
similar fates.9
The victory would be almost as short-lived for Andrew Harclay as the loss was for
Lancaster and his followers. For although being granted the earldom of Carlisle by the
king as a reward for his service at Boroughbridge, by an irony of historical
circumstance, within a year Harclay was arrested and executed for treason against
Edward II. His crimes were surprisingly familiar: he was accused of harboring
hostility against the Despensers and for allying with Robert Bruce.10
While the battle of Boroughbridge has engendered little interest among modern
writers,11 a similar lack of interest in what happened on the battlefield is not shared by
contemporary authors. Among English chroniclers reporting the reign of Edward II,
including all those who discussed the battle of Bannockburn, the battle at
Boroughbridge plays an important role in their narrative of the Edward/Lancaster
conflict. This is especially the case in the long accounts found in the Chronicon de
Lanercost and the Vita Edwardi II, both of which favor Edward II, and The Brut,
which favors Lancaster. Perhaps this is because Thomas of Lancaster's position of
importance in the English court as well as his relationship to the king made his
rebellion significant, or perhaps it is because the fight at Boroughbridge was one of
the few 'victories' had by the king; for whatever reason, the details of the battle and
the short campaign leading up to it can be fairly easily sketched out.

While all contemporary writers discuss other aspects of the Lancastrian conflict with
Edward, the campaign which would lead to Boroughbridge begins only after the exile
of the Despensers, when the king and his loyal earls set out to attack Lancaster and his
adherents. Lancaster discovered the king's intentions while he and his closest ally,
Humphrey of Bohun, earl of Hereford, were meeting at Lancaster's castle at Pontefract
in south Yorkshire. The Anonimalle Chronicle claims that Hereford had fled to
Pontefract with the news of Edward's approach because 'he thought that [Edward]
would destroy them as he had done others.'12 Thomas of Lancaster's
9 Maddicott, pp. 311-23; Fryde, pp. 58-68; Phillips, pp. 224-25; Prestwich, The Three
Edwards, p. 91; McKisack, p. 67; and Keen, p. 70. See also S.L. Waugh, 'The profits of
Violence: The Minor Gentry in the Rebellion of 1321-22 in Gloucestershire and
Herefordshire,' Speculum 52 (1977), 843-69. On these trials see Fryde, pp. 58-59; J.G.
Bellamy, The Law of Treason in England in the Late Middle Ages (Cambridge, 1970), pp. 49-
51; George Sayles, ''The Formal Judgments on the Traitors of 1322,' Speculum 16 (1941), 57-
63; and George L. Haskins, 'Judicial Proceedings Against a Traitor after Boroughbridge, 1322,'
Speculum 12 (1937), 509-11. On the punishments of Lancaster's adherents see Fryde, pp. 59-
86. For many years after the decapitation of Roger of Clifford, Lancaster's co-conspirator and
son-in-law, his body was hung in a gibbet from the wall of the castle of York, subsequently
giving the castle its more common name, Clifford's Tower.
10 Fryde, pp. 156-58. Thomas of Burton (II:347) cites Harclay's hatred of the Despensers as the
cause of his treason, while the Chronicon de Lanercost (p. 235) sees it as a result of his desire to
marry Robert Bruce's daughter.
11 Only T.F.Tout, 'The Tactics othe Battles of Boroughbridge and Morlaix,'English Historical
Review 19 (1904), 711-13 and J.E. Morris, 'Mounted Infantry in Mediaeval Warfare,'
Transactions of the Royal Historical Society 3rd set. 8 (1914), 86-91 mention much about the
battle.
12The Anonimalle Chronicle, 1307 to 1334, ed. W.R. Childs and J. Taylor (Leeds, 1991), p. 104-
05: ' . . . et penserent bien qil les voleit destruire si come il avoit fait autres . . . '
Page 90
immediate response to the threat of Edward's attack is somewhat of a mystery. At least
one chronicler, the anonymous author of the Chronicon de Lanercost, contends that
Lancaster initially wanted to stay and fight the king's troops, 'yet on the advice of his
people he retired with his army into the northern regions.'13 But this chronicle is alone
in that assertion, with all others reporting only that Lancaster and his force fled
immediately to the north. The Vita Edwardi II even asks the question: 'But why does
the Earl of Lancaster, so often accustomed to resist the king, now take to flight,
particularly as he had with him the Earl of Hereford and the flower of English
chivalry?'14
Beyond the Chronicon de Lanercost's response, there are five possible answers to the
Vita Edwardi's question. First, three chronicles mention the failure of Lancaster's ally,
Robert Holland (of Holland in Lancashire), to bring him reinforcements. It was
Holland who, according to the Vita Edwardi II, had been put in charge of Lancaster's
treasury and whose responsibility it was to hire an army 'of his best men.'15 But it was
a responsibility which Holland had failed to carry out. This brings an accusation of
treason by the pro-Lancastrian The Brut,16 while the Vita Edwardi II declares Robert
Holland to have been 'an apostate in his lord's cause [who] deserted to the lord king,' a
treacherous act even for the author of this pro-Edward treatise.17 In any case, the lack
of reinforcements must have significantly decreased Lancaster's expected army
strength and encouraged his flight.
A second possible reason for Lancaster's flight may have been the large size of
Edward's attacking force. At least this is the conclusion drawn by Thomas of Burton:
'But when the two counts saw that the king's troops were both more numerous and
stronger than theirs, they retreated.'18 And while the numbers of Edward's army
cannot be known for certainboth the Chronicon de Lanercost's estimate of 60,000 and
the Vita Edwardi II's of 300,000 are undoubtedly much exaggerated19it was probably
far larger than that of Lancaster and Hereford, especially if Robert Holland had failed
to come with reinforcements as had been anticipated.
Third, if it was not the size of the king's army which caused Lancaster to flee, perhaps
it was the ease with which the king, before Lancaster's march north, defeated a
Lancastrian force in a small battle near the Trent River at the town of Burton. All six
of the chronicles which mention this engagement agree that it led directly to the flight
of Lancaster's force. The Vita Edwardi II, which contains perhaps the most
13Chronicon de Lanercost, p. 242: 'Licet autem comes regem vellet ibi exspectasse et
pugnasse cum eo, de consilio tamen suorum divertit cum exercitu versus partes boriales.'
14Vita Edwardi II, p. 122: 'Sed quare fugit comes Lancastrie qui totiens solebat regi resistere,
precipue cum haberet secum comitem Herfordiae et clariorem militiam totiens Anglie?' See also
Henry Knighton, I:424; Adam Murimuth, p. 36; Chronicon de Lanercost, p. 242; Thomas of
Burton, II:341; Geoffrey le Baker, p. 13; Anonimalle Chronicle, pp. 104-05; Bridlington, p. 75;
and Brut, I:217.
15Vita Edwardi II, p. 122. See also Henry Knighton, I:424 and Brut, I:217.
16Brut, I:217
17Vita Edwardi II, pp. 122-23: ' . . . sed prefixo die Robertus non venit, immo prevaricator in
causa domini sui reddidit se domino regi.'
18 Thomas of Burton, p. 341: 'At, cum ipsi 2 comites acies regales plures esse suis et fortiores
conspexissent, terga verterunt . . . '
19Chronicon de Lanercost, p. 242 and Vita Edwardi II, p. 122.
Page 91
lengthy discussion of this battle, reports that the king in his march to the north came to
the bridge crossing the Trent at Burton where he sent 'a strong phalanx of cavalry and
infantry, wishing to know if there would be any opposition to his crossing [of it].' But
the king's investigating force found that Lancaster was defending the opposite side of
the river at this point; and so a small and inconclusive skirmishthe Vita calls it a
'conflictum'was fought for three or more days until the king's army found a ford
higher up the river and crossed there. The anonymous author concludes: 'When the
barons heard, and now saw for themselves that the king had crossed the river, they
left the bridge, took to horse, and fled.'20 None of the other sources is as explicit in its
details about this battle as is the Vita Edwardi II, and in fact none describes anything
more than the king winning the battle by force of arms.21 But undoubtedly this very
easy victory, whether accomplished by stealth or by strength, influenced Lancaster's
decision to flee to the north.
Fourth, both the Vita Edwardi II and the Vita et mors Edwardi II report that Edward
offered amnesty to any of Lancaster's rebels who wanted to return to his allegiance.
Although neither chronicle mentions how many of his adherents deserted from
Lancaster, the Vita et mors reports at least two desertions of note, Maurice Berkeley
and Hugh of Audeley.22 Lancaster's dwindling numbers, in the face of Edward's larger
force, as well as the possibility of further desertions must have encouraged his flight.
Finally, there is the question of Lancaster's ties to Scotland. Was he in fact fleeing to
the north because he had been promised aid by Robert Bruce? While no documentary
evidence has been discovered linking Edward's chief councillor with Robert Bruce,
and such an indictment may be merely a propaganda ploy, at least four chronicles
claim that Lancaster was indeed trying to flee to his allies across the border. As the
Bridlington chronicler puts it: 'it was common opinion that they had arranged to go to
Scotland;'23 and the Vita Edwardi II reports: 'they hoped to find a refuge in Scotland,
because Robert Bruce, as was said, had promised help against the king.'24 It is only
one later chronicle, The Brut, which claims that Lancaster was not trying to reach
Scotland. The anonymous author of this work reports instead that
20Vita Edwardi II, p. 122: 'Exinde conduxit. exercitum usque ad magnum fluuium qui dicitur
Trente. Est autem ibidem pons magnus qui uiam prebet transeuntibus. Premisit quoque rex ad
pontem cuneum fortem armatorum et peditum, scire volens an aliquis impediret transitum suum.
Venerat autem comes Lancastrie cum omni sequela sua in uillam de Burhtone ex parte alia.
Cumque iam fuisset compertum quod rex disposuit transire fluuium, misit comes viros fortes
armatos et pedites qui pontem defenderet. Verum cum per tres pluresque dies inter se partes
dimicassent, ac ad eundem conflictum in crastinum redissent, reperit rex uadum superius, ubi
transiit ipse et reliqua pars exercitus. Audientes itaque barones et iam uidentes quod rex flumen
transisset, pontem reliquerunt, equos ascenderunt et fugam inierunt.'
21 Thomas of Burton, II:341; Geoffrey le Baker, p. 13; Anonimalle Chronicle, pp. 104-05; Vita et
mors Edwardi II, p. 303; and 'A Chronicle of the Civil Wars of Edward II,' ed. G. L. Haskins, in
Speculum 14 (1939), 78.
22Vita et mors Edwardi II, p. 303 and Vita Edwardi II, p. 121.
23 Bridlington, p. 76: 'Opinio tamen communis vulgi fuit quod disponebant usque Scotiam
properasse.'
24Vita Edwardi II, p. 123: 'In Scotiam sperabant habere confugium, quia Robertus de Brutz, ut
dictum erat, contra regem promiserat auxilium.' See also Chronicon de Lanercost, p. 242 and
Annales Paulini,
(footnote continued on next page)
Page 92
Lancaster stopped short of going across the border because he did not want to be
perceived as a traitor. In a speech to his troops the earl presents his reasons for this
decision:
Lords . . . if we go towards the north, men will say that we have gone to join the Scots; and so
we shall be seen as traitors, because of the differences between King Edward and Robert
Bruce, who has taken the kingdom of Scotland. And therefore I say, for myself, that I will go not
further into the north.25
Whether an alliance with the Scots was the reason for Lancaster's flight to the north or
not cannot be ascertained from the narratives of the Boroughbridge campaign.
However, one thing is certain: if he was trying to reach the Scots, Lancaster failed to
make it to the border before meeting his demise.
The chronicles relate little about Lancaster's journey north until he reached the
Boroughbridge region. Only the anonymous Bridlington chronicler says anything at
all, and then it is to condemn the rebels, for they 'despoiled all the lands and did many
other criminal acts [on their journey to the north].'26
If it was his intention to do so, Lancaster should have easily reached Scotland too, as
he was some distance ahead of the king, but he and his troops stopped at
Boroughbridge, reports the author of the Vita Edwardi II, so 'that they might rest for
the night.'27 Here they encountered the unexpected: a sheriff, Andrew Harclay, who
with his northern levies, was loyal to the king. More importantly, and more
disastrously for Lancaster, these troops were very experienced in warfare, having
fought in a number of border skirmishes against the Scots; some probably also
participated at the battle of Bannockburn, eight years earlier.
Harclay was a capable leader, 'warlike and strong,' describes the Bridlington
chronicler, who also adds that 'he had been commissioned by the king because of his
valor to resist and stop those rebelling against the king.'28 He had been on his way
south to join up with the king's army in pursuit of the rebels when, according to the
Chronicon de Lanercost, while stopping at the town of Ripon, 'he learned from a spy
that the earl and his army were going to arrive the next day at the town of
Boroughbridge,' only a few miles from Ripon. This gave him the chance to surprise
(footnote continued from previous page)
in Chronicles of the Reigns of Edward I and Edward II, ed. W. Stubbs, vol 1, RS (London,
1883), p. 302.
25Brut, I:217: 'Lordes . . . if we gone toward pe * north, men wil seyn pat* we gon toward pe*
Scottes; and so we shul be holde traitoures, for cause of distaunce pat* is bituene Kyng Edward
and Robert pe* Brus, pat* made him Kyng of Scotland. And perfore* y say, as tochyng myself,
pat* y wil go no ferper* into pe* North.'
26 Bridlington, II:75: ' . . . patrias spoliantes et alia facinora facientes . . . '
27Vita Edwardi II, p. 123: ' . . . ut ibidem saltem una nocte requiescerent.' See also 'Chronicle of
the Civil Wars,' p. 78.
28 Bridlington, II:75: ' . . . Andreas de Harcla, bellicosus et strenuus, virtute commissionis regiae
sibi factae ad resistendum et refraenandum regi rebellantes . . . ' See also Thomas of Burton,
II:342 and Brut, I:217-28. On the commission of Edward II to Andrew Harclay, which was
principally to stop any Scottish foray into England, see Vita Edwardi II, pp. 120-21.
Page 93

Battle of Boroughbridge
Lancaster's force, and thus 'pressing forward, at night, he arrived before the earl and
occupied the bridge before him.'29
On the following day, March 16, Lancaster arrived at Boroughbridge, anticipating no
opposition to his crossing of the Ouse River and continuing his journey north. But
there stood Harclay's force, blocking Lancaster from further progress. The chronicles
report very little about Harclay's army at Boroughbridge, and what is reported
sometimes differs between these chronicles. For example, the Bridlington chronicler
and the anonymous Chronicle of the Civil Wars of Edward II say that some of
Harclay's force were knightsabout fifty according to the Chronicle and about twenty
according to Bridlington.30 However, beyond Harclay, only Sir Symond Ward, the
sheriff of York is mentioned by name, and then only in The Brut.31 Furthermore, the
Chronicon de Lanercost reports that most of these troops were from the counties of
Cumberland and Westmoreland, 'all who were able to bear arms,' while the Chronicle
of the Civil Wars records that they were almost all from the town of York.32
But the characteristic of Harclay's army which most impresses the contemporary
historians, and which all agree upon, is that almost all of these troops were infantry
soldiers armed with spears or pikes.33 As well, Harclay in arriving at the place of
battle ordered his men so that even the knights were forced to fight on foot. The
Chronicon de Lanercost recounts:
Sending his horse and those of his men to the rear, he ordered all his knights on foot and some
pikemen at the northern end of the bridge, and opposite the ford or
29Chronicon de Lanercost, p. 243: 'de nocte igitur consurgens praevenit comitem, et
praeoccupavit pontem de Burghbrigge.' See also Vita Edwardi II, p. 123 and Brut, I: 218.
30 'Chronicle of the Civil Wars of Edward II,' p. 78 and Bridlington, p. 76. The Vita Edwardi II
(p. 123) is the only chronicle which records the numerical strength of Harclay's force: 4000
soldiers. The Anonimalle Chronicle (pp. 106-07) notes only an army of 'grant poer.'
31The Brut, I:218.
32Chronicon de Lanercost, pp. 242-43 and 'Chronicle of the Civil Wars,' p. 78.
33 See Chronicon de Lanercost, p. 243; Bridlington, p. 76; 'Chronicle of the Civil Wars,' p. 78;
Vita et mors Edwardi II, p. 303; and Geoffrey le Baker, p. 14.
Page 94
passage of water, he ordered other pikemen in a schiltrom, according to the manner of the
Scots, to oppose the knights and the horses on which the enemy sat.34
Harclay's formation thus allowed him to take advantage of the fighting skill of his
experienced infantry warriors. It should also be noted, as remarked on by the author
of the Chronicon de Lanercost, that the formation around the ford was in the form of
a schiltrom, 'according to the manner of the Scots,' thus revealing from where, at least
in this author's opinion, this infantry formation had been learned.
Archers too are referred to among Harclay's soldiers, although how many there were
is not mentioned, and only The Brut declares their placement on the sides of the
infantry formation.35 Yet, the accuracy of this declaration is doubtful. What would be
the purpose of such a formation? Accurate archery fire into an opposing force across
the Ouse River would be nearly impossible. And there was no need for archers to be
placed on the flanks of the infantry to narrow the opposing cavalry charges as would
become standard in later English formationsat Dupplin Moor, Halidon Hill, and
Crécyas the bridge and ford would serve to narrow sufficiently any cavalry charges
made on these infantry positions.
Lancaster's army approached Harclay from the south side of the river, with the only
access to the opposite shore being the bridge and a nearby ford, both guarded by his
opponents. Lancaster's plan, according to the Chronicon de Lanercost, was to split his
force in two: the earl of Hereford and Lancaster's son-in-law, Sir Roger of Clifford,
would take their part of the army across the bridge to 'seize [it] from the pikemen
standing there,' while Lancaster himself would take the second part of the army across
the ford 'to seize the water and the ford from the pikemen' there. Both of these attacks
were to be made on horseback.36
At this point in the narrative of the battle The Brut adds an interesting story not
included in other contemporary writers' accounts. This chronicle reports that before
the battle began, Lancaster, when he had seen the great 'power' which Harclay had
brought with him and being 'very afraid,' called his opponent to him, hoping to
persuade the sheriff to abandon the king and join the rebellion. Lancaster offered
Harclay both land and leadership:
and we will give to you the best part of five earldoms that we have and hold, and we will give
you an oath that we will do nothing without your counsel and so that you would be well at ease
with us.
But Harclay refused Lancaster's entreaty, causing the earl to threaten the sheriff:
34Chronicon de Lanercost, p. 243: 'et, dimissis retro equis suis et suorum, statuit in pedibus
omnes milites et quosdam lancearios ad borialem partem pontis, et contra vadum sive transitum
aquae posuit alios lancearios in scheltrum, secundum modum Scottorum, ad resistendum
equitibus et equis in quibus adversarii considebant.'
35Brut, I:219. See also Chronicon de Lanercost, p. 243 and Vita Edwardi II, p. 124.
36Chronicon de Lanercost, p. 243.
Page 95
At your word, Sir Andrew, I tell thee that before the year is done you shall be taken and held as
a traitor, even more than you consider us now, and you shall die a worse death than any knight
of England ever suffered.37
It is doubtful whether The Brut's conversation with its prophecy of Harclay's eventual
treasonous demise ever took place. But if it did, it was a last, desperate attempt by
Lancaster to avoid battle, for it was necessary for him to fight the smaller force of
Harclay at Boroughbridge quickly rather than linger or march elsewhere and risk
running into Edward's larger force somewhere behind him.38 So the battle was
fought.
It is important at this point to note that neither Lancaster nor any of his lieutenants felt
that they were in any danger of defeat. Certainly their cavalry, with so many knights
among the force, should easily have defeated Harclay's infantry, despite, the fact that,
as Geoffrey le Baker insists, they were unaccustomed to fight against infantry.39 But,
as the Chronicon de Lanercost puts it, 'another thing happened.'40
It is unclear from contemporary sources if both Hereford and Lancaster launched
simultaneous attacks, or whether Hereford's force proceeded with the first assault. It
does seem apparent, however, that whether Hereford's attack was launched first or
not, it was the first to engage enemy troops, the bridge being easier to cross than the
ford. Sir Humphrey of Bohun, earl of Hereford and Essex, was 'a bellicose man in all
engagements, assuredly strong in body, most astute in mind and prudent enough in
council,' wrote Geoffrey le Baker; he was 'a worthy knight, renown throughout all
Christendom,' added The Brut.41 He was also the brother-in-law to Edward II, whose
friendship had wavered only in opposition to Piers Gaveston and the Despensers.
Now he made the initial attack on a force commissioned by the king to put down his
rebellion.
Hereford's attack was brief, and entirely unsuccessful; for not only was his charge
stopped by the infantry formation on the other side of the bridge, but the earl himself
was killed in the melee. Roger of Clifford, the second in command, was wounded,
and he and the rest of the company were driven back to their side of the bridge,
having
37Brut, I:218: 'Sire Andrew,' quod he, 'e mow wel vnderstonde how pat * our Lord pe* Kyng
is ladde and misgouernede by miche false conseil, prou* Sir Hugh pe* Spenser pe* fader, & Sir
Hugh pe* sone, & Sir John Erl of Arundel, and prou* Maistre Robert Baldok, a false pilede
clerc, pat* is in pe* Kyngus court duelling; wherfore y praye ow pat* e wil come wip* vs,
wip* al pe* power pat* e have ordeynede, and helpe to destroie pe* venemye of Engeland, and
pe* traitoures pat* bene perin*, and we wil if vnto ow pe* best part of v Erldomes par* We
havep* & holdep*; and We wil mak vnto ow an oth pat* we wil neuer do ping* wipout* our
consel, and so e shul bene as wele at ese wip* vs . . . ' 'Sir Thomas! pat* wolde nou t do, ne
consent perto*, for no maner ping pat* yhe might me eue, wipouten* pe* wil and
commaundement of our lord pe* Kyng; for pan* shulde y be holde a traitoure for evermore.' ' . .
. At on worde, Sir Andrew, y telle pe*, pat*, or pis* er be gon, pat* e shal be take and holde for
a traitoure, and more pan e holde vs nowe; and in worse deth e shul die, pan euer dede Knyght
of Engeland.'
38 No chronicle indicates how close Edward's army was to Lancaster on the day of
Boroughbridge, but because the king was able to arrive so quickly at the trials of the rebels after
their defeat, his army must have been fairly close.
39 Geoffrey le Baker, p. 14.
40Chronicon de Lanercost, p. 243.
Page 96
escaped 'with difficulty.' The Chronicon de Lanercost gives the best account of what
happened:
For when the earl of Hereford . . . and Sir Roger of Clifford and some other knights, in the
manner of lions, had bravely entered upon the bridge before the rest of the troops, and had
charged fiercely upon the enemy, pikes were thrust at the earl from all sides. And he fell
immediately and was killed on the bridge with his standard-bearer and [other] knights . . . Sir
Roger of Clifford, though grievously wounded with pikes and arrows, and driven back, escaped
with difficulty along with the others.42
It seems clear in this chronicle that Hereford's charge was made on horseback.
However, the Vita Edwardi II avers that Hereford's force was dismounted 'for the
bridge was narrow, and offered no path for horsemen in battle array.'43 While it may
seem more logical to agree with the Vita Edwardi II, it is in fact the Chronicon de
Lanercost which is the more detailed and accurate source for this battle and is the
narrative accepted by most historians for what occurred.44 The Chronicon de
Lanercost mentions only a cavalry charge by Hereford.
Lancaster's charge across the ford also met with disaster, although in this attack it
appears that archery played a more major role than against Hereford, as again stated in
the Chronicon de Lanercost:
The earl [of Lancaster's] cavalry, when they tried to cross the water, could not enter it because
of the number and density of arrows which the archers discharged onto them and their horses.45
No one appears to have been killed by the archery assault, but progress was slowed
and confused, and after hearing of Hereford's death on the bridge, according to the
Vita Edwardi II, 'their zeal for battle cooled off, and they at once retreated.'46
Completely overwhelmed, Lancaster was forced to seek a truce until the following
41 Geoffrey le Baker, p. 13 and Brut, I:219. See also Vita et mors Edwardi II, p. 303.
42Chronicon de Lanercost, p. 243: 'Cum enim comes Herfordiae . . . et dominus Rogerus de
Clifford cum quodam milite, more leonum, audacter ante alios pontem intrassent, et in adversarios
fortiter irruissent, infixae sunt comite lanceae circumquaque, et cito prostratus est et interfectus
supra pontem cum vexillario suo . . . dominus autem Rogerus de Clifford, graviter vulneratus cum
lanceis et sagittis et repulsus, cum aliis vix evasit.' On the death of Hereford see also Vita
Edwardi II, p. 124; Anonimalle Chronicle, pp. 106-07; Bridlington, p. 76; 'Chronicle of the Civil
Wars of Edward II, p. 78; Henry Knighton, I:425; and French Chronicle of London, p. 45. Thomas
of Burton (II:342), the Vita et mors Edwardi II (p. 303), and Brut (I:219) all claim that Hereford
was killed by a spear thrust from below the bridge, although neither of the most detailed accounts
of the battle, the Vita Edwardi II or the Chronicon de Lanercost, make such an assertion.
43Vita Edwardi II, p. 124: 'Erat enim ports strictus, nec uiam equitibus ad bellum procedentibus
prebere potuit.'
44 See, for example, Morris, 'Mounted Infantry' and Tout, 'Tactics.'
45Chronicon de Lanercost, p. 243: 'Equites autem comitis, qui voluerunt aquam transivisse, non
potuerunt eam intrare prae multitudine et spissitudine telorum quae a sagittariis mittebantur in eos
et in equos eorum.' See also Vita Edwardi II, p. 124.
46Vita Edwardi II, p. 124: 'Alii uero, dum uadum transire nituntur. ab ymbre saggitarum misere
atteruntur; sed post mortem comitis Herfordie sua virtus tepuit militie, et statim reuertuntur.' See
also Chronicon de Lanercost, pp. 242-43.
Page 97
morning so that he could regroup his remaining troops and take care of the
wounded.47 He had every intention to continue the battle on the following day, it
seems, but during the night a large number of his soldiers deserted their lords and fled
from the battlefield.48 Thus Lancaster awoke not to fight but to flee from the field,
realizing what would happen to him if was captured. Yet, he had nowhere to run, and
before the day was over he and sixteen other knights were apprehended and taken to
York to await trial and punishment. As the Vita Edwardi II recounts:
On that same night the Sheriff of York came with a large force to attack the king's enemies;
relying on his help Andrew Harclay entered the town [of Ripon] very early, and taking the Earl
of Lancaster and almost all the other knights and esquires scatheless, led them off to York and
imprisoned them. Some left their horses and putting off their armor looked round for ancient
worn-out garments, and took to the road as beggars. But their caution was of no avail, for not a
single well-known man among them escaped. O calamity! To see men lately dressed in purple
and fine linen now attired in rags, bound and imprisoned in chains!49
Almost all chronicles record the details and results of the trials of treason against
Lancaster and his fellow rebels; he was beheaded, being spared from the traditional
hanging, drawing and quartering, because of his kinship with the king.50
Thomas of Lancaster's rebellion had been short-lived, defeated as it was by a rather
rude bevy of northern infantry musters using a 'Scottish' formation. The Vita Edwardi
II lauds the victory:
[It was] a marvelous thing and one indeed brought about by God's will and aid, that so scanty a
company should in a moment overcome so many knights. For the Earl's side were more than
seven times as numerous as their adversaries. There were captured with the Earl of Lancaster
and the other barons more than a hundred valiant knights. The number of esquires no less
valiant was, I believe, much greater. Why therefore should they not have stood and fought
manfully for their safety? Indeed the criminal is always fearful and so less effective in action.
They saw that the whole countryside was up-in-arms in front of them, and thus their advance
was blocked. They knew that the king's army threatened them from the rear, and therefore their
retreat was not secure. Thus as men having no plan nor even time to deliberate, they fell into the
hands of their enemies, etc.51
47Chronicon de Lanercost, p. 244; Vita Edwardi II, p. 124; Bridlington, p. 76; and 'Chronicle
of the Civil Wars,' p. 78.
48Chrvnicon de Lanercost, p. 243 and Thomas of Burton, II:3422.
49Vita Edwardi II: pp. 124-25: 'Ipsa uero nocte vicecomes Eboraci cum magna cohorte uenerat
inimicos regis invadere; cuius auxilio fretus Andreas de Herkelee uillam intrauit summo mane, et
cepit comitem Lancastrie et onmes pene reliquos milites et scutarios sine uulnere, et perducens
Eboracum reclusit in carcarem. Quidam equos reliquerunt, et exeuntes arma sua ueteres attritas
uestes quesierunt sibi, et more mendicantium uiam incesserunt. Sed cautela non profuit, nam nec
unus quidem famosus ex omnibus euasit. O monstum! uidere uiros purpura et bisso nuper indutos
nunc attritis uestibus incedere, et uinctos in compedibus recludi sub carcere!'
50 See Thomas of Burton, II:342; Anonimalle Chronicle, pp. 106-11; Bridlington, p. 76;
'Chronicle of Civil Wars,' p. 78; Geoffrey le Baker, p. 14; Henry Knighton, I:425; and Brut, I:219-
20.
51Vita Edwardi II, p. 125: 'Res miranda et certe nutu Dei et auxilio promota, quod tam rara manu
subito superatur tanta militia. Pars enim comitis numero armatorum partem persequentium excessit
(footnote continued on next page)
Page 98
The Brut's epitaph is more succinct and more sorrowful, but the sentiment is nearly
the same: 'Alas, the shame and dishonor that the gentle order of knighthood had there
at that battle.'52
Both of these chroniclers misjudge the cause of Harclay's victory. The knights cannot
be blamed for their defeat. This was not a battle where the cavalry lost because of a
poorly executed charge, for neither Lancaster's nor Hereford's cavalry charges are
noted by the contemporary sources to have been in themselves disordered or
confused. Nor could the cavalry charges have been delayed, for Edward II's main
force, much larger than that of the rebels, was feared to be close at hand.
This was a battle won by infantry. Harclay's infantry formationin a solid line across the
end of the bridge and in a schiltrom, or semi-circle, around the end of the fordwas the
determining factor in his victory over the more 'knightly' cavalry force of the earls of
Lancaster and Hereford. No tally for Harclay's army is given by the chroniclers, but he
would not have needed many soldiers, even if facing a larger, more powerful troop.
For the bridge and ford caused a natural narrowing of the cavalry charges made
against Harclay, and this ensured that his formation was not flanked. And as his
infantry remained in formation, combined with the archers present among his troops,
the opposing cavalry was almost powerless to breach the 'Scottish' formation. Because
Edward's army was probably nearby, Lancaster's dual force of cavalry was forced to
charge, and these charges ended in failure, capped perhaps at their climax with the
death of Hereford. It was a final blow to the morale of an already demoralized rebel
army, for they must have felt that this would be an easy victory. Although no death
calculations are made for Harclay's soldiers by contemporary chroniclers, he probably
did not lose many. On the other hand, Boroughbridge cost Lancaster the rebellion.
Never again would an English army fight as it had at Bannockburn. From this time
until the end of the Middle Ages the chief tactic of the English was to dismount most
of the cavalry and to order them in a solid infantry formationin a line or, if warranted
by geography, as at the ford at Boroughbridge, in a schiltrom. This once powerful and
proud cavalry-based army had become little more than, as John E. Morris termed it,
'mounted infantry.' Was this recognized first at Boroughbridge in
(footnote continued from previous page)
in septuplum. Capti sunt enim cum comite Lancastrie et ceteris baronibus milites ualentes
centum et amplius. Sed et scutariorum non minus ualentium multo maiorem credo fuisse
numerum. Quare igitur non restitissent et pro salute sua uiriliter dimicassent? Revera cor
delinquentium semper est pauidum et ideo minus valens ad negotium. Videbant totam patriam a
fronte excitatam, et per hoc uiam eorum impeditam. Sciebant a tergo imminere regis exercitum,
et propter hoc cursum retrogradum non esse securum. Vnde quasi homines non habentes
consilium nec etiam tempus ad deliberandum, inciderunt in manus inimicorum, etc.'
52Brut, I:220: 'Allas pe * shame & despite, pat* pe* gentil ordre of Knyghthode pere* hade at
pat* bataile!' Other chronicles give different, less militarily oriented causes for defeat at
Boroughbridge. Geoffrey le Baker (p. 13) believes that defeat came because both sides had not
concentrated on defeating the Saracens in the Middle East, while the monk of Bridlington (p. 76)
saw the defeat of Lancaster as the fulfillment of Biblical prophecythe fulfillment of Revelation's
prophecy of the red horse riding forth in the last days, whose rider had been given the duty of
taking peace from the land (Revelations 6:4)and as the fulfillment of the prophecy of Gregory the
Great to Maximus of Salonathat those who are sinful will be punished by military conquest and
unhappiness.
Page 99
Harclay's brilliant victory over Lancaster? We probably cannot know for sure. But
perhaps it is better to recognize the influence of this positive image of infantry warfare
rather than the negative one received by the English at their defeat at Bannockburn,
the latter being credited most by modern historians for the change in English tactics in
the early fourteenth century.
Page 100

VIII
The Battle of Cassel, 1328
The peace of Athis-sur-Orge, described by fifteenth-century historian Adrien de Budt
as a peace of 'little stability,' brought almost no calm to the county of Flanders.1
Constant turmoil filled the region as the peace treaty with its heavy indemnities caused
a hatred of the French overlords unseen since before the battle of Courtrai. Although
Philip the Fair extended the treaty of Athis-sur-Orge and its indemnities in 1309 with
an added threat of a papal interdict and excommunication if it was broken,2 violence
erupted often in Flanders in the two decades following the battle of Mons-en-Pévèle.3
Most hard hit by the heavy indemnities brought down with the treaty of Athis-sur-
Orge were the peasants of the Flemish coastal regions. They resented having to
ransom nobles captured during the 1302-05 war who cared little about them and their
region. Robert of Bethune, who succeeded as count of Flanders in 1305, allowed the
payment of indemnities to lapse in an endeavor to keep his county peaceful and at
least a little loyal to him.4 But when Robert died and was succeeded in September
1322 by his grandson, Louis of Nevers, the payment of war indemnities became a
central issue which the new count wanted to enforce. Louis, who had married one of
the daughters of the French king, Philip V, was extremely loyal to the French crown,
and he immediately set out to make his county subservient to French royal wishes.5
1 Adrien de Budt, Chronicon Flandriae, in Corpus chronicorum Flandriae, i, ed. J.J. de Smet
(Brussels, 1837), p. 311.
2 See the documents contained in the Codex diplomaticus Flandriae, ed. Thierry de Limburg-
Stirum (Bruges, 1879), I:31-51, 65-69.
3 In 1309-10, a small rebellion arose in the Waasland. In 1311, at Ghent, the weavers of the city
arose against their local nobles. In 1319, they repeated this revolt and with other Ghentenaars
succeeded in attacking Rijsel as well. Finally, in 1320, a Brugeois contingent, under the command
of Nikolaas Zannekin, attacked Sluys. See Henri Pirenne, Histoire de Belgique, II:76-82; William
H. TeBrake, A Plague of Insurrection: Popular Politics and Peasant Revolt in Flanders, 1323-
1328 (Philadelphia, 1993), pp. 34-37; Jacques Sabbe, Vlanderen in opstand, 1323-1328:
Nikolaas Zannekin, Zeger Janszone en Willem de Deken (Bruges, 1992), pp. 14-15; and David
Nicholas, Medieval Flanders (London, 1992), pp. 195-96.
4 See Henri Pirenne, ed., Le soulèvement de la Flandre maritime de 1323-1328 (Brussels,
1900), pp. iii, iv, xi; Pirenne, Histoire de Belgique, II:71; David Nicholas, Town and
Countryside: Social, Economic and Political Tensions in Fourteenth-Century Flanders (Bruges,
1971), pp. 162-63; F.W.N. Hugenholtz, Drie boerenopstanden uit de viertiende eeuw ('s
Gravenhage, 1978), pp. 22-23; and Jacques Mertens, 'De boerenopstand onder Zannekin,' in
Nikolaas Zannekin en de slag bij Kassel 1328-1978 (Dixmude, 1978), p. 98.
5 See Lucas, The Low Countries, p. 46; TeBrake, pp. 45-50; Sabbe, p. 20; Nicholas, Medieval
(footnote continued on next page)
Page 101
Besides seeking the payment of these indemnities, Louis enacted other changes which
further alienated his subjects. Within a year of his ascension, Louis had supplanted the
influence of Bruges with that of Ghent, he had returned the pro-French party to its
political prominence, and he had nearly eliminated Robert of Bethune's policy of local
independence. The most sweeping blow to the Flemings (especially to the Brugeois)
came on July 13, 1323 when Louis granted a monopoly of the waterways to his great-
uncle, John of Namur, an outspoken foe of Bruges. This edict was in direct violation
of the privileges of the town of Bruges which had long controlled the waterways at
Sluys.6 Count Louis, perhaps misunderstanding the feelings of his subjects, had
delivered this edict while in Bruges, and when the townspeople heard it they revolted,
cast the count into prison, marched on Sluys, burned it, and captured John of Namur.
Another rebellion had begun.
The rebellion was directed by the weavers of Bruges, but fueled by the peasants of
maritime Flanders.7 Governmental changes were instituted, comital bailiffs were
replaced by revolutionary 'captains,' and Robert of Cassel, the uncle of the imprisoned
Louis of Nevers, and 'secretly hostile' to him, was asked to lead the people against
their French overlords.8 Only the town of Ghent, still loyal to the count and the king,
stood in the way of complete rebel control of Flanders.
Finally, in November 1325, the new French king, Charles IVPhilip V had died in
1322decided that the Flemish rebellion warranted some military action. He raised an
army and prepared a punitive expedition against the rebels. At the same time, he wrote
a letter to Robert of Cassel promising him forgiveness if he quit the rebellion; another
letter promised Ghent that it would not fall to the rebels.9 The threat of French
military involvement and the subsequent desertion of the rebellion by Robert of
Cassel was enough to force peace on the rebels. A treaty was formally signed at
Arques on April 19, 1326. Louis of Nevers was freed and restored to his rule. The
demolition of all fortresses constructed since 1323 was ordered. The indemnities
agreed upon at Athis-sur-Orge in 1305 were to be paid with the addition of an extra
10,000 livres coming to the count for his suffering. Robert of Cassel was
(footnote continued from previous page)
Flanders, pp. 209-11; Jan van Rompaey, 'De opstand in het vlaamse kustland van 1323 tot
1328 en de figuur van Nikolaas Zannekin,' in Nikolaas Zannekin en de slag bij Kassel 1328-
1978 (Dixmude, 1978), p. 104; and John Bell Henneman, Royal Taxation in Fourteenth
Century France: The Development of War Financing, 1322-1356 (Princeton, 1971), p. 55.
6 See Nicholas, Town and Countryside, p. 161; TeBrake, pp. 49-50; Pirenne, Soulèement, p. xv;
Sabbe, p. 20; Rompaey, p. 105; Nicholas, Medieval Flanders, 213; and Jules Viard, 'La guerre de
Flandre (1328),' Bibliothèque de l'école de chartes 83 (1922), p. 362.
7 See Henri Pirenne, Early Democracies in the Low Countries: Urban Society and Political
Conflict in the Middle Ages and the Renaissance, trans. J.V Saunders (New York, 1971), p. 153.
8 See TeBrake, pp. 51-98; Sabbe, pp. 25-53; Pirenne, Soulèvement, pp. xvi-xvii; Hugenholtz, pp.
28-30; Nicholas, Medieval Flanders, p. 131; J. Bovesse, 'Le Comte de Namur Jean Ier et les
événements du comté de Flandre en 1325-1326,' Bulletin de la commission royale d'histoire 131
(1965), 385-454; Henri Pirenne, 'Documents relatifs à l'histoire de la Flandre pendant la première
moitié du XIVe siècle,' Bulletin de la commission royale d'histoire 7 (1897), 477-93; and Henri
Pirenne, 'Un mémoire de Robert de Cassel sur sa participation à la révolte de la Flandre maritime
en 1324-1325,' Revue du nord 1 (1910), 45-50.
9 See Henneman, p. 55.
Page 102
pardoned, and the revolutionary captains were removed with the comital bailiffs
restored to their posts.10
But the treaty of Arques did not end the rebellion. The revolutionary captains refused
to surrender their positions to the returning bailiffs, and with a new intensity the
rebellion began again. Louis fled to Paris early in 1327 to ask Charles for more
military assistance, but the weak king had died and was succeeded on May 29, 1327
by Philip VI of Valois, Charles' cousin. Philip recognized the county of Flanders as an
area of rebellion in an otherwise supportive kingdom.11 Thus, as Edouard Perroy
writes, he 'zealously embraced the feudal cause' responding to Louis' appeal with a
resolve to punish the rebels and to avenge his loyal count.12 Leading a large French
army himself, Philip entered Flanders and encountered the rebel force. The battle of
Cassel was the result. It was fought on August 28, 1328.
At the end of this day, the French were victorious. Few rebels escaped the slaughter.
Philippe Contamine has calculated the Flemish loss at fifty percent of their force,13
while contemporary sources number the Flemish casualties as high as 20,000. The
French lost considerably fewer men, with possibly as few as seventeen knights having
died on the battlefield.
With the defeat at Cassel, the rebellion met its end. News of the French victory
traveled quickly to the northern rebels who fled to safety inside their town walls.
Ypres and Bruges rapidly submitted to Philip and Louis, and peace once again
returned to the French countryside.14 Philip was merciful. Perhaps he had seen too
much bloodshed at Cassel, or perhaps he felt that if he punished the Brugeois harshly
they might rebel further against him. In any case he refused to allow the destruction of
the towns and lands held by the rebels.15 Instead he launched a program of war
indemnities and land confiscations against the Flemings. Most of the landholding
rebels lost their property, but few lost their lives.16 Only Willem de Deken, the mayor
of Bruges, was taken to Paris to be tortured and executed.17
As with the battles of Courtrai and Mons-en-Pévèle, there are a number of
contemporary chroniclers who comment on the battle of Cassel. The anonymous
10 The text of the treaty of Arques can be found in Codex diplomaticus, II:385-403. See also
TeBrake, pp. 98-99; Pirenne, Soulèvement, pp. xxv-xxvi; Sabbe, pp. 52-53; Nicholas, Town
and Countryside, p. 161; Hugenholtz, p. 30; Rompaey, pp. 120-21; and Henneman, p. 56.
11 See TeBrake, pp. 108-19; Sabbe, pp. 55-64; Hugenholtz, pp. 106-07; and Nicholas, Medieval
Flanders, p. 215.
12 Edouard Perroy, The Hundred Years War, trans. W. B. Wells (London, 1951), p. 81.
13 Contamine, p. 258.
14 See Viard, pp. 375-80.
15 See Pirenne, Histoire de Belgique, II:90; Pirenne, Soulèvement, pp. xxx-xxxi; TeBrake, pp.
123-25; Sabbe, pp. 72-75; and Hugenholtz, pp. 32-34.
16 A scholarly 'cottage-industry' of confiscation records research seems to have developed after
Pirenne first published his records in 1900 (La soulèement de la Flandre maritime). Two later
additions have come from Jacques Mertens: 'La confiscations dans la Chatellenie du Franc de
Bruges après la bataille de Cassel,' Bulletin de la commission royale d'histoire de Belgique 134
(1968), 239-84 and 'De economische en sociale toestand van de opstandelingen uit het Brugse
Vrije, wier goederen na de slag bij Cassel (1328) verbeurd verklaard werden,' Revue Belge de
philologie et d'histoire 47 (1969), 1131-53. See also TeBrake, Appendix A, pp. 139-44.
17 See TeBrake, pp. 123-24 and Pirenne Soulèvement, p. xxxi.
Page 103
French authors of the Grandes chroniques de France and the Chronographia regum
Francorum, Gilles le Muisit, the bourgeois of Valenciennes, the second continuator of
Guillaume de Nangis' Chronicon, Willem Procurator, Giovanni Villani, and the
authors of the Chronicon comitum Flandriae, the Chronique de Flandre, and the
Rijmkroniek van Vlaenderen all include narratives of the battle.18 To these narratives
must be added the accounts written by two of the premier historical writers of the
fourteenth century, Jean le Bel and Jean Froissart.19 Also recounting the battle is
Richard Lescot, writing c.1344 at the monastery of Saint Denis, and an anonymous
work composed at Paris and now known as the Chronique Parisienne anonymé de
1316 à 1328.20 Another source for the battle of Cassel is a series of letters written
between September 6 and October 4, 1328 by Pope John XXII to Philip VI. Although
they do not give a narrative of the battle itself, these letters do report some details of
the conflict which are not mentioned in the other sources. More importantly perhaps,
they report what details were important enough to be recounted to the Pope in
Avignon only a few days after the battle was fought.21
However, unlike the battle of Courtrai and Mons-en-Pévèle, few contemporary
descriptions of the battle of Cassel favor the Flemings or mourn their defeat. Even
those sources written in Flanders, like the Chronicon comitum Flandriae, the
Chronique de Flandre, and the Rijmkroniek van Vlaenderen, are opposed to this
rebellion despite being favorable to the one fought in 1302-05. Perhaps this is due to
the fact that none of these chronicles were written in Bruges or in the maritime region
of Flanders, the areas involved in the 1323-28 rebellion. But, more likely there is an
absence of sympathetic records of the battle because of the social character of the
rebellion; it included peasants as well as townspeople. Nevertheless, it is still possible
to reconstruct what occurred at the battle of Cassel, and to understand why the
Flemish infantry lost to the French cavalry there.

Almost all of the contemporary accounts of the battle begin with the visit of Louis of
Nevers to Philip VI at the latter's coronation to plead for assistance against his
rebellious subjects. Louis needed this assistance, the continuator of Guillaume de
Nangis' chronicle insists, because 'he had not been of such power to be able to oppose
their evils and to extirpate the cause of the rebellions.'22 However, Philip did not
18Grandes chroniques, IX:79-95; Chronographia regum Francorum, II:2-10; Gilles le
Muisit, pp. 211-12; Récits d'un bourgeois de Valenciennes, pp. 150-52; the Continuatio
chronici of Guillaume de Nangis, II:90-102; Willem Procurator, pp. 218-25; Giovanni Villani
(ed. Muratori), cols. 656-57; Chronicon comitum Flandriae, pp. 202-07; Chronique de
Flandre, I:342-47, 534-37; and Rijmkroniek van Vlaenderen, pp. 822-23.
19 Jean le Bel, Chroniques de Jean le Bel, ed. J. Viard and E. Duprez, Société de l'histoire de
France (Paris, 1904), I:93-94 and Jean Froissart, Chroniques, in Oeuvres de Froissart, ed.
Kervyn de Lettenhove (Brussels, 1867), II:216-26.
20 Richard Lescot, Chronique, ed. J. Lemoine, Société de l'histoire de France (Paris, 1896), pp.
3-7 and Chronique Parisienne anonymé de 1316 à 1339, ed. A. Hellot, in Mémoires de la
société de l'histoire de Paris xi (1895), pp. 116-22.
21 These letters, nos. 2222, 2223, 2238, 2249 and 2258, are found in Lettres de Jean XXII (1316-
1334), ed. Fayen, in Analecta Vaticano-Belgica, vol. 2 (Rome, 1908).
22Continuatio chronici of Guillaume de Nangis, II:90-91: 'Circa istud tempus, comes Flandriae
suo
(footnote continued on next page)
Page 104
respond swiftly to Louis' request, choosing instead to deliberate for fifteen days
before deciding whether or not to involve himself in the count of Flanders' civil
war.23 In the end, with the advice of his council, Philip agreed to Louis' plea so that
the rebellion, in the words of the Chronicon comitum Flandriae, 'would not attract to
it other communities in Picardy and France.'24 Only then, Jean Froissart records, did
Philip promise Louis that he would 'never return to Paris until you have taken
peaceful possession of the whole of Flanders.'25
Once the decision to march against the Flemish rebels had been made, Philip began to
gather men and money to proceed with his war. On June 18 he announced an arrière-
ban which asked that both noble and non-noble knights assemble at Arras on July 31
to fulfill their military duty; if they could not come to Arras they were permitted to
send a monetary substitute so that others would be permitted to go in their place.26
The response was enthusiastic. Not only was there a 'large army' gathered in Arras at
the end of July, but it is also reported that more than 231,078 livres were collected.27
In fact, the army was so large, the Chronographia regum Francorum reports, that
Philip needed first to march to the towns of Lille, St. Omer, and Tournai in order to
gather food to feed them all.28 It is also reported that before entering Flanders Philip
visited Notre Dame Cathedral, the monastery of Saint Denis and several charitable
houses in order to gain the favor of God in his military venture; at Saint Denis he also
obtained the oriflamme to take with him into battle.29
Hearing of the approach of this large army, the Flemings too prepared for battle.
However, despite much anticipation, the advent of war was not met with much
(footnote continued from previous page)
domino regi Franciae, sicut debebat, fecit homagium. Quo facto, comes dicto regi plurium
subditorum suorum, et maxime de Brugis, de Ypra et de Cassello, et de pluribus aliis locis
intolerabiles rebelliones exposuit, et quod ipse tantae potestatis non esset quod solus posset
eorum malitiis obviare, et rebellionis materiam exstirpare. Unde et sibi a dicto rege fieri
auxilium humiliter supplicavit.' See also Chronographia regum Francorum, II:2-3; Chronique
Parisienne anonymé, p. 117; Richard Lescot, pp. 3-4; Chronique de Flandre, I:357; Jean
Froissart, II:217; Chronicon comitum Flandriae, pp. 202-03; and Chronique Normande, p. 36.
23 See Chronicon comitum Flandriae, p. 203.
24Chronicon comitum Flandriae, p. 203: 'Consilium fuit omnium, quod rex illuc exercitum
mitteret in adjutorium comitis Flandriae ad domandum rebelles, ne, si terminos suos exirent,
attrahere sibi possent communitates alias Picardiae et Franciae, et sic magnam confusionem facere
nobilibus atque et regno.'
25 See Jean Froissart, II:217: 'Dont parla li roys Phelippes, et dist: ''Loeis, biaux cousins, nous
vous tenons pour conte de Flandres, et par le digne unction et sacrement que nous recevons hui,
jammais ne renterons en Paris se vous avons mis en possession paisieulle de le content de
Flandres." '
26 For example, the town of Montpellier sent the king a subsidy of 2,000 livres as a substitute for
sending soldiers (C. Devic and J. Vaissette, Histoire générale de Languedoc (Toulouse, 1872),
I:676-80. See also the Continuatio chronici of Guillaume de Nangis, II:92 and Henneman, pp. 70-
71.
27 On the amount of money collected see Viard, p. 365. A number of contemporary sources are
impressed by the size of the French army. See the Chronique Parisienne anonymé, p. 117; Gilles
le Muisit, p. 211; and the Annales Tielenses, in MGH SS, xvi, ed. G.H. Pertz (Hannover, 1879), p.
26. There was resistance among some French towns to Philip's order to collect an armed force at
Arras. As seen in a document contained in C. Devic and J. Vaissette, I:671-676, the inhabitants of
Lunel refused to send troops or money to the king for his military expedition.
28 See Chronographia regum Francorum, II:3.
29 See the Continuatio chronici of Guillaume de Nangis, II:92-93 and the Grandes chroniques,
IX:80-81.
Page 105
enthusiasm by the Flemings. Still, a large army, drawn from the towns of Bruges and
Ypres as well as from many smaller villages, seems to have gathered to face Philip.30
According to the author of the Grandes chroniques, the Flemings were frightened at
the prospect of fighting the French king and 'they did not know where he would
attack.' Therefore, they divided their force into three separate groups: the rebels from
Bruges and the area around Bruges were marched towards Tournai, those from Ypres
and Courtrai towards Lille, and those from the areas of Veurne, Dunkirk, Cassel, and
Poperinge marched towards St. Omer.31 It was this last group which first saw the
French, and they moved swiftly to a defensive position, ordering their force on the
hill outside of the village of Cassel.32
There is some disagreement among the sources as to who led this army of rebels.
While most chronicles report that the Flemings at Cassel had no leader,33 Jean
Froissart names Nikolaas Zannekin, a long-time leader of the rebels. Not only does
Froissart name him as leader, but he also describes Zannekin in fairly positive terms as
a man 'marvelously arrogant, hardy and scurrilous' who promised those who followed
him that they would kill the king of France. He reports further that it was Zannekin's
purpose to keep Louis of Nevers out of Flanders, 'to confound all his enemies' and to
kill all those loyal to the king of France.34
The French army arrived at Cassel shortly after the Flemings, setting up their own
camp at the base of the hill. Philip, knowing that this was not the entire Flemish force,
also divided his army into three parts, sending Robert of Cassel with a small
contingent to St. Omer and Louis of Nevers with an equally small force to Lille. The
bulk of the French army stayed with Philip at Cassel.35
But Philip chose not to attack the Flemish army at Cassel. The Flemings had wisely
selected terrain which was easy to defend. A hillside, like ditches or a marsh, meant
that a cavalry charge would be disorganized and fatiguing, allowing a solid Flemish
infantry line to withstand easily the mounted shock combat of the knights. Thus,
unlike some of his predecessors, Philip chose to wait for a better opportunity to fight
against his opponents; he sat at the base of the hill, in effect besieging the Flemings
camped on top of it.
30 See the Chronique de Flandre, I:534. Jean Froissart claims that this force numbered 16,000
soldiers, but he is the only chronicler who provides a number for the Flemings.
31 See the Grandes chroniques, IX:82: 'Quant les Flamens virent que le roy avoit fait si grant
semonse, si s'assamblerent et virent qu'il n'avoient point de seigneur de qui il peussent faire
chevetaine, car touz les gentilz hommes du pays leur estoient failliz, et ne savoient de quel part le
roy les devoit assaillir, ne de quel part il devoit à eulz venir.'
32 See the Chronique de Flandre, I:344, 534.
33 See the Chronique Parisienne anonymé, p. 118; Grandes chroniques, IX:82; Récits d'un
bourgeois de Valenciennes, p. 150; and Willem Procurator, p. 222.
34 Jean Froissart, II:218: 'Quant le chapitainne de ces Flamens, qui se nommoit Clais Dennequins,
entendi que li rois de France, en sa nouvelle régnation, avoit juré que jamais il n'entreroit en
Paris, ne entenderoit à aultre cose si averoit remis en Flandres le conte Loïs et confondus tous ses
ennemis et nuisans, si s'en enfellona grandement, et dist que chils rois poroit bien fallir à ses
pourpos, et toutes fois pour lui brisier, il s'en meteroit en painne.' See Jacques Mertens, 'Zannekin
of de evolutie van het beeld van een volksheld,' De Frans Nederlanden. Les Pays-Bas Français
(1978), 24-37.
35 See Grandes chroniques, IX:83-85; Chronographia regum Trancorum, II:4-5; and Chronique
de Flandre, I:343.
Page 106
For three days the situation continued with little activity between the two armies.
There were a few skirmishes, but these were small and ineffectual. According to the
Chronicon comitum Flandriae, Philip also had brought or made a few artillery pieces
which were used to try to harass the Flemish army, but these too had little effect.36
Philip's army also pillaged and burned the land around Cassel, including the
monastery of Wastina (although the Chronicon comitum Flandriae insists that Philip
did not know of its destruction), and these fires were seen by the Flemings on the hill.
This was done, claims the Chronique de Flandre, that 'by this they wished to draw the
Flemings from the mountain.'37
The Flemings also tried to provoke their enemy to do battle during these three days.
Several chroniclers report the Flemings trying to taunt the French. For example, the
continuator of Guillaume de Nangis' chronicle reports that the Flemings 'for the
purpose of deriding and mocking the army of the king and of all France,' raised a
large dead cock on a colored spear and shouted that 'when this cock crows, the king
will take Cassel.' He also reports that the Flemings mocked Philip VI's assumption of
the French throne calling him an 'invented king.' This too was without effect as the
French king held his ground below the Flemings and refused to fight against their well
protected defensive order.38
It was the Flemish army which first decided to break from their defensive formation.
Impatient in having received no reinforcements, irritated by the burning of their lands
and the harassment of the French army, the Flemish troops had grown discouraged by
their inaction. They decided to charge down the hill and attack the French. It was
hoped that by surprising the French and perhaps even killing or capturing the king,
the Flemings might be able to defeat them, or at least to stop the devastation around
Cassel. To this end Nikolaas Zannekin spoke to his troops, an oration which the
Chronicon comitum Flandriae recounts:
Are we not strong men who have subjugated of Flanders, and are we accustomed to fear no
one? We are accustomed to fight no one more than the king of France, so that we might be able
to humble his pride. And behold this king is before us with but a few men; let us attack him in
our strength . . .
He was answered: 'Let us attack the king immediately!'39
36 On the skirmishes between the French and Flemish armies before August 28 see Grandes
chroniques, IX: 85-86; Chronographia regum Francorum, II:5-6; and Richard Lescot, p. 6. On
Philip's use of catapults see the Chronicon comitum Flandriae, p. 205.
37Chronique de Flandre, I:344: ' . . . et par ce cuidèrent traire les Flamens jus de mont.' See also
Chronicon comitum Flandriae, pp. 204-05; Grandes chroniques, IX:85-86; Chronographia
regum Francorum, 11:5-6; and Richard Lescot, p. 6.
38Continuatio chronici of Guillaume de Nangis, II:94-95: 'In dicto vero castro, in regis et totius
Francorum exercitus derisum et subsannationem, in quodam eminenti loco posuerant Flammingi
quemdam gallum permaximum de tela tincta, dicentes: "Quando gallus uste cantabit, rex Casselum
capiet vi armorum." . . . Unde et subsannantes regi, dicebant eum et vocabant Regem inventum.'
See also Richard Lescot, p. 6.
39Chronicon comitum Flandriae, p. 205: ' "Numquid non sumus nos viri fortes, qui nobis
subegimus totam Flandriam, et neminem timere consuevimus? nullum magis solebamus appetere
quam regem Francorum, ut ipsius superbiam humiliare possemus. Et ecce hic rex ante nos cum
paucis: adeamus eum in fortitudine nostra . . . " "Tunc omnes acclamaverunt:" "Regem protinus
invadamus." '
Page 107
Thus, at the hottest part of the day, after the French had retired to their tents to rest, to
play games, and to hold a war council, the Flemings swept down onto them and onto
their king who, the continuator of Guillaume de Nangis maintains, was taking his
'accustomed nap.'40 With an apparent sarcasm Gilles li Muisit describes the attack:
'The infantry attacked without noise and almost in silence. All had the intention of
approaching the tents of the king and taking and killing him. They passed through
many tents, killing and striking no one, and they did go silently.'41
The charge was initially effective in its surprise. After the battles of Courtrai and
Mons-en-Pévèle, the French forces were accustomed to having the Flemish infantry
hold its defensive position without an offensive attack. As the intended target of this
attack, no one was more surprised than the king. In this the attack almost succeeded,
for the Flemings nearly reached Philip. As the Chronique de Flandre recounts, the
attackers reached the king's tent but were beaten back by his soldiers. Still, Philip was
so shaken by the attack that he had difficulty arming himself, and he had to be assisted
by his aides.42 (Even the Pope knew of the threat to Philip's life, telling Philip that
God had preserved the king's life 'so that you might know most certainly that the
future victory would proceed from Him and that He might obligate you to his
devotion.'43)
Ultimately, however, the Flemish surprise attack failed. Although some of the French
did flee, the attack did not defeat them, and soon they were able to regroup and form
a counter-attack. Initially this counter-attack was led by Robert of Casselwho had
joined the main French force a few days previouslytogether with the marshals of the
French army, Gaucher de Châtillon and Robert Bertrand, with later assistance by
William of Hainault.44 But, as soon as he was able to recover from his attack, the king
himself entered the fray, and ultimately it was his presence that
40Continuatio chronici of Guillaume de Nangis, II:96-97. See also Grandes chroniques,
IX:86-87; Chronographia regum Fruncorum, II:6-7; Jean Froissart, II:220; and Récits d'un
bourgeois de Valenciennes, p. 150.
41 Gilles le Muisit, p. 211: 'Quasi circa horam nonae, sine strepitu et quasi sub silentio, pedites
omnes cum intentione accedendi ad sarcinas regis et ipsum capiendi vel occidendi, transierunt
plura tentoria, neminem occidentes au concutientes, sed tacite ibant.'
42Chronique de Flandre, I:345. See also Continuatio chronici of Guillaume de Nangis, II:98;
Grandes chroniques, IX:97; and Chronographia regum Francorum, II:7-8.
43 John XXII, letter 2222: 'Sane, fili dilectissime, cum horrore quodam percepimus quod hostes
ipsi usque ad tua tentoria . . . sed forsan permisit Altissimus ut certius cognosceres ab ipso
futuram victoriam processisse teque ad suam devotionem ejusque beneplacita exequenda et artius
evitanda que prohibuit, obligaret.' See also letter 2223.
44 There is some dispute among contemporary writers as to who led the counter-attacks which
defeated the Flemish surprise attack. Dutch chroniclers credit their count, William of Hainault and
Holland, with the leadership of the counter-attacking troops. (See Willem Procurator, p. 223 and
Jan Beke, p. 291.) This assessment is agreed to by the Liègeois Jean de Hocsem, p. 197. On the
other hand, several French and Flemish sources see the counter-attacks led by Robert of Cassel
and the French marshals, with William of Hainault's counter-attack coming later. (See
Chronographia regum Francorum, II:7; Grandes chroniques, IX:336; and Chronique de
Flandre, I:345.) Finally, the Chronicon comitum Flandriae (p. 205) claims that both Robert of
Cassel and William of Hainault led counter-attacks. An interesting aside to this incident is the
Tournaissien Gilles le Muisit's claim that it was the troops from Tournai who first saw the Flemish
surprise attack and sounded the alarm (p. 211).
Page 108
ensured victory. As the continuator of Guillaume de Nangis's Chronicon writes: 'And
behold when the king was led down through the battle with his few, the whole army
of knights, many fleeing the [Flemish] infantry, seeing the regal banners, all gathered
to him.'45
After the unsuccessful surprise attack, the remnants of the Flemish army fell back into
a crown formation to make a final defensive stand. But, according to the Chronicon
comitum Flandriae, this maneuver was futile; the battle was effectively lost:
But although in this order they resisted strongly and held themselves vigorously, this resistance
profited them nothing since the leonine descendants of the Franks stabbed at them with lances
wet with blood . . . Nevertheless, the Flemings in this resistance made a great slaughter of
horses, and they strongly resisted the French because another exit for flight was not open to
them, for they were surrounded on every side by knights.46
At the end of the battle the chroniclers report that the field was heaped high with dead
rebels. Contemporary estimates of Flemish dead range from a low of 9,000 given by
the Chronicon comitum Flandriae to a high of 22,000 recorded by the author of the
Chronique Parisienne anonymé. Among the Flemish dead was Nikolaas Zannekin. No
French casualty figures are given.47 Hearing of the massacre at Cassel, the other two
rebel armies quickly disbanded and returned to their homes hoping thereby to escape
punishment.
Philip had defeated the rebellion and returned Louis of Nevers to his comital throne.
At the same time, according to Guillaume de Nangis' continuator, Philip chastised
Louis for being unable to keep the county peaceful. In a rather lengthy castigation,
Philip cautioned the Flemish count to be just and to keep his lands at
45Continuatio chronici of Guillaume de Nangis, II:98: 'Et ecce interim cum rex per divium
cum suis paucis deducitur, visis regalibus insigniis totus exercitus, multis peditum fugientibus,
ad eum quasi unus homo congregatur.'
46Chronicon comitum Flandriae, pp. 205-06: 'Sed licet sic ordinati fortiter resisterent et
vigorose se haberent, nihil tamen eis profuit haec resistentia, dum Francorum leonina progenies
ipsos adeo lanceis cruore madentibus confoderet, quod ipsos vivere in tali perplexitate taederet.
Faciebant nihilominus Flamingi in resistentia illa magnam stragem in equis, et viriliter resistebant
Gallicis, eo quod non patebat eis aliquis exitus ad fugam ineundam: erant enim undique ab
equitantibus circumclusi.'
47 These are the Flemish death totals as tallied by contemporary sources (from highest to lowest):
22,000+ (Chronique Parisienne anonymé, p. 119); 19,800 (Grandes chroniques, IX:90; Richard
Lescot, p. 6); 16,000 (Jean le Bel, I:94); 15,000 (Récits d'un bourgeois de Valenciennes, pp. 151-
52; Jean Froissart, II:222-23); 14,000 (Gesta abbatum St. Trudoniensum, p. 420); 12,000 (Gilles
le Muisit, p. 211; Chronographia regum Francorum, II:9); 11,547 (Continuatio chronici of
Guillaume de Nangis, II:99); 11,000 (Chronique de Flandre, I:357; Rijmkroniek van Vlaenderen,
p. 823; Chronique Normande, p. 36); 10,000 (Jean de Hocsem, p. 197); and 9,000 (Chronicon
comitum Flandriae, p. 206). The only chronicle which mentions the French dead is the Chronique
de Flandre, and this chronicle notes only the names of a few slain knights without reporting a total
calculation (I:346). As well, only Jean Froissart (II:217) and the Grandes chroniques (IX:88-89)
call attention to Zannekin's death.
Page 109
peace so that the French troops would not need to return. For if he was forced to do
so, Philip is reported to have said, 'I will return not for your but for my utility.'48
To many modern historians, the battle of Cassel is a vindication of cavalry warfare in
the early fourteenth century. Sir Charles Oman, for one, believed that the victory at
Cassel confirmed the strength of the cavalry-based army against its infantry enemy,
reaffirming his opinion that the Flemish infantry victory over the French cavalry at
Courtrai was but an 'accident.'49 Edouard Perroy agreed, noting that the French
victory at Cassel was 'a fitting sequel to that [battlefield] on which the 'golden spurs'
had been defeated.'50 At least one contemporary chronicler agrees with this modern
opinion. The anonymous author of the Chronique Parisienne anonymé, referring to
the earlier French victory over the Flemings at Bouvines (in 1214) writes:
But our Lord Jesus Christ, merciful to and full of pity for the king of France and his troops, as
he otherwise had been in the battle against Ferrand of Flanders and others, did not allow the
Flemings to hold on longer to their iniquities.51
But can this battle be dismissed as a simple case of knight over foot, one which
affirms the continued success of mounted shock combat? Certainly it appears that
before the battle the Flemings had carried on their fourteenth-century military
tradition. They chose a defensive position, and while they did nothing to prepare the
ground for a cavalry assault as they had at Courtrai and Mons-en-Pévèle, the choice of
a hilltop for their infantry line provided suitable terrain to defeat any charge by
opposing knights. Furthermore, for three days they waited for this attack, continually
taunting the French in the hopes of drawing the knights into one of their accustomed
wild and disordered charges. This is all attested to by the author of the Grandes
chroniques who insists that the Flemings had realized that an offensive attack against
the French would be ineffective and thus had consequently chosen their defensive
positions at Cassel.52
Moreover, the attitude of Philip VI in initially encountering the Flemish army was to
respect their defensive positioning and not to make what would have been a suicidal
charge. He had no intention of repeating the mistakes made by Robert of
48Continuatio chronici of Guillaume de Nangis, II:101-02: 'Tota vero Flandria jam quasi
quietata et sub obedientia regis posita, rex in praesentia baronum convocari fecit coram se
comitem Flandriae, cui fertur sic dixisse: "Comes, ad requestam vestram huc veni, et forte quia
negligens fuistis de justitia facienda. Ut tamen vos scitis, venire non potui sine mei et meorum
maximis expensis et laboribus. Ecce de liberalitate totam terram vestram quietam et pacificam
vobis restituo, expensas condono; sed de caetero caveatis ne propter defectum justitiae
oporteat me redire, scientes quod si ob defectum vestrum rediero, non ad vestram sed ad meam
utilitatem redibo." '
49 Oman, I:118. This same opinion was repeated as late as the 1989 Annual Meeting of the
American Historical Association in a paper delivered by Franklin J. Pegues entitled 'Cavalry and
the Feudal Aristocracy in Thirteenth- and Fourteenth-Century England and France.'
50 Perroy, p.81.
51Chronique Parisienne anonymé, pp. 120-21: 'Maiz Nostre Seigneur Jhesuchrist, misericord et
piteable au roy de France et ès siens, comme aultrefoiz à este au fait de Ferrant de Flandrez et
allieurs, ne souffry pas les iniquitez dez Flamens plus à tenir.'
52Grandes chroniques, IX:82.
Page 110
Artois at Courtrai, Jacques of Bayonne at Arques, Aymer de Valence at Loudon Hill,
Walter of Brienne at Kephissos, Edward II at Bannockburn, and Thomas of Lancaster
at Bouroughbridge. Instead of attacking the Flemish infantry with his knights, he
chose essentially to besiege their camp, hoping to provoke them into breaking their
defensive order. This he accomplished by continually harassing the Flemings with
skirmishes and artillery combined with the burning and pillaging of the lands
surrounding the hill. In this the Flemish defensive position worked to their
disadvantage. Being the only high ground for miles around, the hill at Cassel gave the
Flemings an unhindered view of this destruction, and many whose homes and
families were nearby must have felt concern for them. Ultimately, this resulted in the
surprise attack which led to the demise of the Flemish infantry.
When the Flemish attack finally came, it proved to be less of a surprise than intimated
by contemporary chroniclers. Although the attack seemed to have caught some of the
French by surprise, including Philip VI, the fact that a counter-attack followed so
quickly with the participation of such high-ranking lieutenants as Robert of Cassel,
William of Hainault, and the two French marshals, Gaucher de Châtillon and Robert
Bertrand, seems to indicate that at least part of the army was on alert for just such an
occurrence. Moreover, the Flemish attack was rash and disorganized. The Chronique
Parisienne anonymé recalls the stupidity of the surprise attack: ' . . . around 20,000
Flemings, all inflamed by battle, in their fool-hardiness . . . descended from the hill.'53
What began then as a Flemish instigated surprise attack turned quickly into a French
cavalry charge which easily defeated a disordered and poorly led infantry.
Once the infantry did recover, reforming their defensive 'crown' formation on the hill,
although apparently not on the hilltop as before the attack, they had some success
against the French knights. But their numbers had now been so diminished and they
were so demoralized that the French eventually defeated even that formation.
Perhaps then it is inaccurate to credit the cavalry with this victory, as Oman and
Perroy have done; it may be more precise instead to blame the infantry for their own
defeat. Had they not broken from their defensive position into a confused and
disordered attack, Philip VI probably would not have attacked them. As he showed
later in attempting to raise the sieges at Tournai in 1340 and at Calais in 1347, Philip
was prepared to set up camp in the sight of the Flemings without encountering them
in baffle. As in those struggles, he would have waited at Cassel until the conflict was
solved by diplomacy; he would not have risked what might have resulted in at least a
large number of casualties if not a defeat.54 In this case, the inexperienced Flemings
53Chronique Parisienne anonymé, p.118: ' . . . XXm Flamens ou environ, tous enflambez de
batailler, par leur fol hardement, non pourveuz de conseil . . . descendirent aval la montaigne.'
Some modern authors have agreed that this ill-advised charge led to the defeat of the Flemings.
See Rompaey, p. 123; Lot, I:276; Verbruggen, Krijgkunst, pp. 111, 295; and Verbruggen,
'Historiographie,' p. 248.
54 At the siege of Tournai in 1340 the presence of a large French army within sight of the allied
force (English, Flemish, Brabantese and Hainaulter), coupled with the peace efforts made by
Jeanne de Valois, eventually caused Jean III, duke of Brabant, to desert his allies and the siege
failed. At Calais in 1347 Philip VI was not so lucky. Despite the peacemaking attempts of two
papal legates, the English army refused to raise its siege. Unable to fight the large English force,
Philip returned to France without
(footnote continued on next page)
Page 111
changed the tactics which had afforded them victory at Courtrai. They allowed
themselves to be drawn into a battle which they could not win, where cavalry did
indeed triumph over infantry.
(footnote continued from previous page)
a victory, and Calais fell to the English. See my 'Contemporary Views of Edward III's Failure
at the Siege of Tournai, 1340,' Nottingham Medieval Studies (forthcoming) and 'Hunger,
Flemish Participation and the Flight of Philip VI: Contemporary Accounts of the Siege of
Calais, 1346-47,' Studies in Medieval and Renaissance History n.s 12 (1991), 129-81.
Page 112

IX
The Battles of Dupplin Moor, 1332, and Halidon Hill, 1333
In January 1327, after a reign fraught with military defeats, like that at Bannockburn,
and baronial revolution, like that ending at Boroughbridge, Edward II was deposed.
He was succeeded by his fourteen year old son, Edward III; but because of the new
king's youth, the power in England was held by Edward III's mother, Isabella, and her
lover, Roger Mortimer.1
For Scotland, the deposition of Edward II presented some problems, but also some
opportunities. At the least, it meant that the treaty signed between the deposed king
and Robert Bruce in 1323, which recognized Bruce's rule and Scottish sovereignty,
was threatened.2 At the most, it meant a possible invasion of Scotland by an English
army. Robert Bruce himself was old and ill. Unwilling to risk his throne or his
country, and desiring to take advantage of the confusion caused by the deposition of
Edward II and the youth of Edward III, Bruce determined to take the offensive against
the English. On February 1, 1327, the day of Edward Ill's coronation, Bruce's army
entered the English border counties of Durham and Northumberland and began
besieging Norham Castle and raiding the countryside.3
An English army, led by Mortimer and the young king, travelled north to intercept the
Scottish force, but the campaign, known as the Weardale Campaign, met with absolute
failure. The English troops did cause the Scots to break off their siege of Norham, but
they were unable to capture or defeat Bruce's army. In what is recorded as one of the
most disastrous military expeditions ever attempted, the heavier English knights were
unable either to catch up with their lighter Scottish foes or to provoke them to do
battle. The English supply train bogged down in the wet northern countryside, the
supplies of bread carried by each English knight became inedible after being soaked
with their horses' sweat, and the accompanying Hainaulter mercenary force became
agitated and left the campaign. By August the English were forced to return home,
fatigued and bankrupt, and without even a minor victory over the Scots to buoy their
spirits.4
1 On the deposition of Edward II see N. Fryde, pp. 195-206; McKisack, pp. 84-94; Michael
Packe, King Edward III (London, 1983), pp. 23-31; and M.V. Clarke and V.H. Galbraith, 'The
Deposition of Edward II,' Bulletin of the John Rylands Library 14 (1930), 125-81.
2 See Barrow, pp. 251-52 and Prestwich, Three Edwards, p. 57.
3 See Barrow, pp. 251-52 and Nicholson, Scotland, p. 118.
4 On the Weardale Campaign see Ranald Nicholson, Edward III and the Scots: The Formative
Years
(footnote continued on next page)
Page 113
To prevent further attacks of the Scots on England, Isabella and Mortimer were forced
to negotiate the Treaty of Northampton with Robert Bruce. It was signed in the name
of Edward III on March 1, 1328. The 'shameful peace,' as it is now known,
acknowledged Bruce's rights as king and arranged the marriage between his very
young heir, David, and Edward's sister, Joan. The English renounced all claims over
Scotland, and in turn the Scots agreed to pay £20,000 to cover the costs of lost lands
and possessions.5 To the Scots the Treaty of Northampton was the final victorious act
in the heroic life of Robert Bruce. To the English the treaty was cowardly and
treasonous; Edward III even refused to attend the wedding of his sister and David
Bruce.6
Robert Bruce died on June 7,1329. He was succeeded by his son, David. However,
David was only five years old, and, as had been the case with Edward III, his minority
caused confusion which ultimately led to an attempt to unseat him as king.7 This was
not at the hands of Edward III, however, as he was still 'formally' bound by the Treaty
of Northampton. Instead, the threat to David's throne came from a group of men who
had unsatisfied claims in Scotland. Known as 'the disinherited,' these men supported
the royal claim of Edward Balliol, the son of John Balliol, the Scottish king who had
been forced from the throne in 1297. As well as Edward Balliol, they included David
of Strathbogie (claimant to the earldom of Atholl), Gilbert Umfraville (claimant to the
earldom of Angus), Richard Talbot, Ralph Stafford, Henry Ferrers, Alexander and
John Mowbray, and, perhaps most importantly, the rich and powerful Henry of
Beaumont (claimant to the earldom of Buchan). Their plan was to take an army into
Scotland, to seize their lost lands and titles, and to secure the throne for Balliol.8
Although Edward III did not openly support the disinherited, it is clear that their
designs presented the English king with a 'no -lose situation.' By being cautious and by
withholding his open permission of their expedition, Edward would be honoring the
treaty which had been made with Robert Bruce. At the same time, if the disinherited
were successful in their military venture, Edward would be willing to accept and to
profit from the crowning of Edward Balliol.9
By 1332 the disinherited were ready to undertake their conquest. They had raised
money for the expedition by leasing their English lands and by obtaining an advance
from the Archbishop of York who favored the defeat of David Bruce. They had also
raised an army of nearly 1,500 soldiers. With this small force the disinherited set sail
for Scotland on July 31. They landed, and after fighting a small battle on the beaches
(footnote continued from previous page)
of a Military Campaign (Oxford, 1965), pp. 21-41; Barrow, pp. 252-55; Nicholson, Scotland,
pp. 118-19; Fryde, pp. 210-13; McKisack, pp. 98-100; Keen, p. 17; and Packe, pp. 34-38.
5 On the Treaty of Northampton see Nicholson, Edward III, pp. 42 -56; Barrow, pp. 255-60;
Nicholson, Scotland, pp. 119-20; McKisack, pp. 98-99; Keen, pp. 77-78, 106; Prestwich, p. 58;
and Packe, p. 40.
6 On the English disapproval of the Treaty of Northampton see Nicholson, Edward III, pp. 54-56;
Nicholson, Scotland, p. 121; and McKisack, p. 115.
7 On the death of Robert Bruce and the inheritance of David see Barrow, pp. 322-24; Nicholson,
Scotland, p. 122; McKisack, p. 115; and Keen, p. 107.
8 On the names of the prominent disinherited and their claims in Scotland see Nicholson, Edward
III, pp. 57-74; Nicholson, Scotland, pp. 124-25; and Packe, pp. 65-66.
9 See Nicholson, Edward III, pp. 75-79 and McKisack, p. 116.
Page 114
near Kinghorn they proceeded inland towards Perth.10 On August 11, the disinherited
met a much larger Scottish army at Dupplin Moor. The battle was long and bloody,
but the disinherited prevailed, chasing the Scots from the field. A few days later they
occupied Perth, and on September 24, Edward Balliol was crowned king of Scotland
at the traditional coronation site of Scone.11
Edward III, who had moved to York to keep an eye on the events occurring in
Scotland, was pleased with Balliol's victory.12 Surprised by the results at the battle of
Dupplin Moor, he was nevertheless fully supportive of the new king, receiving his
homage in November 1332. However, Balliol's rule at this time was to be short-lived.
While he had defeated a large Scottish army at Dupplin Moor, the main Scottish force,
under the leadership of Andrew, Earl of Moray, and Sir Archibald Douglas, which
supported the kingship of David Bruce, was still active and gathering numbers.13
Balliol himself was attacked and nearly killed on December 17 and was forced to flee
south to York and Edward III's protection.
Edward III was now compelled, by his acceptance of Balliol's homage, to embrace
openly the leader of the disinherited. He outfitted Balliol with an English force, and
the erstwhile Scottish king travelled north in March 1333 to besiege Berwick Castle, a
major Scottish stronghold. Edward III joined him there two months later.14
It is clear from the events at Berwick that Edward III did not care about the fall of the
castle itself. The English king did not wish to repeat the 1327 campaign in which the
elusive Scottish army had continually frustrated English attempts to bring them to the
battlefield; he had to force Moray and Douglas to fight his army. He was confident that
he could defeat a Scottish force, having learned from Balliol how such a task had
been accomplished at Dupplin Moor. Therefore, he negotiated for the surrender of
Berwick unless the Scottish army would come to save it by July 11. He even took
hostages to ensure that this would occur. However, the Scottish army did not want to
come to battle against the English king. They too knew of their fallibility which had
been evidenced at Dupplin Moor, and thus they tried another tactic to raise the siege of
Berwick: they attacked Northumberland and even laid siege to Bamburgh castle,
where Edward's queen, Philippa of Hainault, was in residence. But Edward III refused
to be kept from his battle, and he began hanging the hostages on a gallows built
outside Berwick, beginning with Thomas Seton, whose father, Sir Alexander Seton,
was in command of the castle. In response, the Scottish army marched north to meet
the English in battle.15 It was fought near Berwick, at Halidon
10 See Nicholson, Edward III, pp. 79-82.
11 On the occupation of Perth see Nicholson, Edward III, pp. 91- 92 and McKisack, p. 116. On
the crowning of Balliol at Scone see Nicholson, Edward III, pp. 92-94; Nicholson, Scotland, pp.
126-27; and McKisack, p. 116.
12 See McKisack, p. 117 and Keen, p. 108.
13 On David Bruce's support in Scotland after the battle of Dupplin Moor see Nicholson, Edward
III, pp. 94-104; Nicholson, Scotland, p. 127; McKisack, p. 117; Packe, pp. 68-69; and Jonathan
Sumption, The Hundred Years War: Trial by Battle (Philadelphia, 1991), pp. 126-28.
14 See Nicholson, Edward III, pp. 119-23; Nicholson, Scotland, p. 128; and Packe, p. 69.
15 On the siege of Berwick see Ranald Nicholson, 'The Siege of Berwick, 1333,' Scottish
Historical Review 40 (1961), 19-42 and Nicholson, Edward III, pp. 123-29. This was, it seems,
the same
(footnote continued on next page)
Page 115
Hill, on July 19. The result was the same as that at Dupplin Moor: the English soundly
defeated their more numerous Scottish foes.
A few days later Berwick surrendered, and Balliol soon had pacified the rest of the
kingdom. David Bruce was forced to flee to France with several of his supporters
where he was granted asylum by Philip VI, whose allegiance to the defeated king was
based solely on his own fear of English invasion. Edward III again accepted Balliol as
king, but at the price of a large tract of land in southern Scotland.16 However, Balliol
was an ineffective ruler, and by the end of 1334 Edward III was forced to return to
assist his ally. He would return again in 1335, 1336, and 1337.17
There are many original sources which discuss the battles of Dupplin Moor and
Halidon Hill, and from them a fairly good understanding of what actually occurred on
the battlefield can be determined. The English sources include many authors who
recorded the battles of Bannockburn and Boroughbridge. And while John Barbour's
Bruce ends with the death of Robert Bruce, other Scottish chronicles, John Fordun's
Chronica gentis Scotorum, Andrew of Wyntoun's The Orygynale Cronykil of
Scotland and the Liber Pluscardensis, all written in the late fourteenth and early
fifteenth centuries, provide accounts of the battles of Dupplin Moor and Halidon Hill
from the Scottish perspective. As well as these sources, there are also two poems
written on the battle of Halidon Hill, one by Laurence Minot and the other by an
anonymous author, and a letter written on July 22, 1333 by Edward III to William of
Melton, the archbishop of York, delineating why the young king felt that they had won
the battle of Halidon Hill. However, these last sources contain little about the battle
itself, concentrating instead on the role played by God and Scottish pride in the
English victory.18

Nearly all original sources begin their discussion of the battle of Dupplin Moor with a
description of the disinherited army gathered by Edward Balliol. Although the
numbers of the troops differ, from a low of five hundred found in the Liber
Pluscardensis to a high of 3,300 found in Henry Knighton's Chronicon,19 most
describe the troops as a united body of confident warriors determined to regain their
(footnote continued from previous page)
Alexander Seton who defected from the English to the Scots before the second day of battle at
Bannockburn.
16 See Nicholson, Edward III, pp. 139-62; Nicholson, Scotland, pp. 129-30; and Prestwich, p.
60.
17 See Nicholson, Edward III, pp. 163-236; Nicholson, Scotland, pp. 130-35; and Keen, pp. 109-
10.
18 Laurence Minot's poem is found in Minot, pp. 1-4. The anonymous poem is found in Minot, pp.
95-97 and The Brut, I:287-89. Edward III's letter is found in Gesta Edwardi de Carnarvan, pp.
116-18 and in Thomas Rymer, ed., Foedera, conventiones, litterae, et cujuscunque generis acta
publica inter reges Angliae et alios quosvis imperatores, reges, pontifices, principes, vel
communitates (1101-1654) (London, 1704-35), II:866.
19 The numbers for Balliol's army are (from lowest to highest): 500 (Liber Pluscardensis, I:264);
1,500 (Gesta Edwardi de Carnarvan, p. 106; Chronicon de Lanercost, p. 268; Thomas of Burton,
II:362; Andrew of Wyntoun, II:382-83; Adam Murimuth, p. 66); 2,500 (Thomas Walsingham,
I:193; The Brut, I:275; the continuator of Walter of Guisborough, II:303); and 3,300 (Henry
Knighton, 1:461-62). Modern historians number the force at 1,500. See Oman, II:102 and
Nicholson, Edward III, pp. 80-81.
Page 116
lost possessions.20 They are depicted as a 'confederatyown' by Andrew of Wyntoun21
and as a 'small army of English soldiers' by the Chronicon de Lanercost.22 On the
other hand, the Anonimalle Chronicle describes the troops as 'a great army . . . with a
great force of Englishmen.'23 According to Robert of Avesbury, it was an army of
'strong men,' composed of some knights and archers, but mostly simple infantry
soldiers.24 However many were present, the number of disinherited was still quite
small in comparison to the number of enemy soldiers which they could expect to
encounter in Scotland. As the Chronicon de Lanercost exclaims: 'Oh what a small
number of soldiers was there for the invasion of a realm then most confident in its
strength.'25
Without the permission of Edward III, the disinherited sailed for Scotland planning to
land at Kinghorn and then to march to Perth.26 However, the Scots became aware of
their intentions and planned to keep them from landing. Their army, led by Duncan,
the earl of Fife, and numbering between 4,000 and 24,000, according to the English
sources,27 attempted to stop the disinherited from landing in Scotland. However,
Edward Balliol's force prevailed in this small engagement, killing a number of Scots
and chasing the rest from the battlefield.28 This was the first of many unexpected
events, one which shamed the earl of Fife who, The Brut reports, 'although amazed
20 On the disinherited's reasons for going to Scotland see Henry Kinghton, I:401; Geoffrey le
Baker, p. 49; and Adam Murimuth, p. 66.
21 Andrew of Wyntoun, II:382.
22Chronicon de Lanercost, p. 267.
23The Anonimalle Chronicle, 1307 to 1334, pp. 146-47: ' . . . ove grant ost et ove grant poer des
Engleis . . . '
24 Robert of Avesbury, p. 296. See also Geoffrey le Baker, p. 49 and Gesta Edwardi de
Carnarvan, P. III.
25Chronicon de Lanercost, p. 267: 'O quam parvus numerus bellatorum fuit iste ad invadendum
unum regnum de suis viribus plurimum tunc confidens!'
26 On Balliol's trip to Edward III see Robert of Avesbury, p. 296; Thomas of Burton, II:362;
Geoffrey le Baker, p. 49; Adam Murimuth, p. 66; and The Brut, I:275. All except for The Brut
insist that Edward III did not grant permission to Balliol for this expedition. The Anonimalle
Chronicle, while not describing Balliol's trip to Edward, nevertheless insists that it was because
'he did not wish to infringe or disturb the peace arranged between him and those of Scotland' that
he refused to allow the disinherited to travel through England to Scotland, thus necessitating their
sea travel (pp. 148-49).
27 See Gesta Edwardi de Carnarvan, p. 104; Henry Kington, II:4462; Thomas Walsingham,
I:193; Thomas of Burton, II:363; Adam Murimuth, p. 66; and Andrew of Wyntoun, II:384.
Estimates of Scottish strength at Kinghorn are (from lowest to highest): 4,000 (Chronicon de
Lanercost, p. 267); 10,000 (Anonimalle Chronicle, pp. 148-49; Henry Knighton, II:462; Thomas
Walsingham, I:193; the continuator of Walter of Guisborough, II:304; The Brut, I:275-76); 14,000
(Thomas of Burton, II:363); and 24,000 (Gesta Edwardi de Carnarvan, p. 104). However, the
two Scottish sources who mention the fighting at Kinghorn, the Liber Pluscardensis (I:265) and
John Fordun (I:354), refer to the size of the Scottish army only as a 'few.' Modern historians have
avoided making an estimate of this army's size. See Nicholson, Edward III, p. 83.
28 Although John Fordun (I:354) insists that only three or four Scots were killed in this
engagement, several English chroniclers insist a much larger number were slain there: 90 (the
continuator of Walter of Guisborough, II:304); 900 (Anonimalle Chronicle, pp. 148-49; Thomas of
Burton, II:363); and 1,000 (Thomas of Walsingham, I:193).
Page 117

Battle of Duppin Moor


at what had occurred, was sorry and full of shame, because such a small company had
defeated him.'29
From Kinghorn Balliol marched west to Dumferline where he replenished his food
supplies and weapons. Unable to rest for fear of Scottish action against them, the
disinherited army quickly continued their march north to Perth.30 At the Earn river the
Scots again tried to stop them, but as at Kinghorn Balliol's force prevailed, and they
crossed the river.31
As they crossed the Earn, the disinherited army could see a large Scottish army
gathered not too far in the distance at Dupplin Moor, an army described by The Brut
as 'many warriors well outfitted for war.'32 As the Anonimalle Chronicle notes: 'some
29 The Brut, I:276: ' . . . was po * wonder sory, and ful evel shamede pat* so litil company
hade him descomfitede.'
30 See Anonimalle Chronicle, pp. 148-49 and The Brut, I:276-77.
31Chronicon de Lanercost, pp. 267-68; Thomas of Burton, II:363; and Liber Pluscardensis,
I:265-66. The Liber Pluscardensis contains a rather curious story of Balliol's army at the Earn
river. This anonymous chronicler claims that the disinherited were initially kept from crossing the
river by the Scots. However, during the night of August 10-11 Balliol was shown a ford across the
river by the lord of Gask, and he was able to cross it, surprise the Scots (who had placed no night
sentries to prohibit a night crossing), and fight the battle of Dupplin Moor. The other sources
contain no reference to the ford or to the lord of Gask's leading Balliol to it.
32The Brut, I:277.
Page 118
of our people were greatly dejected by the small number of their forces.'33 The Scots
had chosen Dupplin Moor as the battlefield, although there appears to have been no
particular reason for this choice, and they had gathered there a large army, between
20,000 and 40,000 in number according to contemporary sources, almost all of whom
were infantry soldiers.34 Most contemporary authors maintain that the Scots were
prepared for the eventual battle. If nothing else, like the disinherited, they too could
see their opponents as they crossed the Earn, and as they observed the smaller force
which was coming against themthe disinherited army's numbers had changed very
little since their landing at Kinghornthey grew in confidence because of their own
much larger numbers.35 During the night before the battle, however, the Anonimalle
Chronicle indicates that the disinherited moved stealthily onto the battlefield, taking
over the moor to the left of the Scottish troops. So surprised were the Scots the next
morning that they 'marvelled at the manner in which the English had thus taken over
the moor.'36 This may be the origin of the Liber Pluscardensis's belief that the Scots
were surprised at the arrival of the disinherited; the anonymous author of this
chronicle claims that 'some were asleep, some unarmed, enjoying themselves and
playing and drinking and making merry.'37
The Scots ordered their army in either two or three divisions.38 Opposing them was a
force which would also fight on foot, except for a contingent of 40-44 German
mercenaries who remained on horseback.39 The disinherited infantry formed three
33Anonimalle Chronicle, pp. 148-49: 'et aucuns de noz gentz furent grantment descounfortez
pur la petit quantite des gentz.'
34 On the Scottish arrival at the battlefield see Anonimalle Chronicle, pp. 148-49; Thomas
Walsingham, I:194; and the continuator of Walter of Guisborough, II:304. Contemporary
chroniclers record the numbers in the Scottish army as: 20,000 (Robert of Avesbury, p. 297);
30,000 (Chronicon de Lanercost, p. 267; Liber Pluscardensis, I:265; Andrew of Wyntoun,
II:385); and 40,000 (Anonimalle Chronicle, pp. 148-49; Gesta Edwardi de Carnarvan, p. 106;
Geoffrey le Baker, p. 49; Henry Knighton, I:462; Thomas of Burton, II:363; Thomas Walsingham,
I:193-94; and the continuator of Walter of Guisborough, II:303-04). Because of the disparity of
numbers given by the original sources, modern historians do not attempt to give an estimate of the
Scottish numbers here. See Oman, II:102 and Nicholson, Edward III, pp. 84-85.
35 On Scottish morale before the battle see Liber Pluscardensis, I:265-66: The Anonimalle
Chronicle insists that the English could easily see the Scottish troops, observing them for an entire
day before approaching for battle (pp. 148-49).
36Anonimalle Chronicle, pp. 148-49: 'Et les Escoz se merveillerent coment les Engleis avoient
issint purpris la more.'
37Liber Pluscardensis, I:266.
38 The three sources which discuss the Scottish formation at Dupplin Moor all differ on the
number of lines in the formation and where they were placed on the battlefield. The Chronicon de
Lanercost (p. 268) contends that there were two Scottish lines, while The Brut (I:277) claims that
there were three lines. Both agree that these lines were ordered in front of the disinherited.
Andrew of Wyntoun agrees with the Lanercost chronicler on the number of lines, but he believes
that one was placed in front of the disinherited army while the other were ordered to the rear, in
effect surrounding Balliol's army (II:386).
39 All chroniclers, English and Scottish, record that the disinherited fought on foot. On the forty
mounted German mercenaries see Anonimalle Chronicle, pp. 150-51; Henry Kinghton, I:462; and
the continuator of Walter of Guisborough, II:304. Thomas Walsingham (I:194) also notes the
presence of the German mercenaries, but reports their number to be forty-four instead of forty.
Page 119
solid lines flanked on both sides by a relatively large number of longbowmen.40 Their
morale was also high, despite their obviously smaller numbers. In fact, as Andrew of
Wyntoun notes, the disinherited army's morale may have been strong because they
were trapped between the river and the Scottish army, 'as fysch in [a] net;' thus unable
to flee, they became determined to fight and to win.41 Their morale also grew with
battlefield orations delivered by Edward Balliol and John Burdon who promised them
victory.42
The Scots made the first charge. Confident in their overwhelming numbers and,
claiming, according to the Liber Pluscardensis, that 'they would drag the tailed
English by the tail and hang them,'43 they assaulted the disinherited order, rushing past
the archers and striking their dismounted opponents. Initially, this charge was
effective. The Chronicon de Lanercost. recounts what occurred: 'Moreover, in the first
onset, when the English and Scots were fighting with their spears firmly fixed against
each other, the Scots drove back the English some twenty or thirty feet.'44 However,
the English did not flee, but held their position, and spurred on by the encouragement
of Ralph Stafford, the earl of Stafford, who shouted 'You English! Turn your
shoulders instead of your breasts to the pikes,' the disinherited began to regain their
position and to push back the Scots.45
The infantry was greatly aided by the archers on their flanks. It seems that most of the
Scottish soldiers either wore no helmets or helmets unequipped with visors, and that
the disinherited archers, in the words of the Chronicon de Lanercost, 'blinded and
wounded the faces of the first division of the Scots by an incessant discharge of
arrows.'46 This may have caused little death, but in fact it so disrupted the Scots that
their attacks fell on the infantry with disarray and confusion. Thomas of Burton
describes this part of the fighting: 'The smaller squadrons, so cut by the archers, were
forced to cling to the larger army, and in a short time, the Scots massing together were
pressed one into the other.'47
The battle lasted from sunrise into the afternoon as the two infantry armies pushed
40 On the formation of the disinherited troops see The Brut, I:278. On the ordering of archers
along the wings of the infantry see Thomas of Burton, II:364.
41 Andrew of Wyntoun, II:386.
42 See Thomas of Burton, II:364 and The Brut, I:277-78.
43Liber Pluscardensis, I:265: ' . . . videntes dictum Edwardum tam paucos in exercitu suo
habentem, se stolide ex utraque parte in eorum multitudine gloriati sunt, dicentes quod Anglicos
caudatos per eorum caudas ad suspendium traherent.'
44Chronicon de Lanercost, p. 268: 'In primo tamen congressu, quando Anglici et Scotti, fixis in
alterutrum lanceis, confligebant, Scotti repellebant Anglicos quasi per viginti pedes vel triginta . .
. ' See also Thomas of Burton, II:364 and Andrew of Wyntoun, II:388.
45Chronicon de Lanercost, p. 268: ' . . . et tunc clamabat baro de Stafforde, ''vos Angli vertatis
contra lanceas vestros humeros et non pectus." 'See also Andrew of Wyntoun, II:388 and Thomas
of Burton, II:364.
46Chronicon de Lanercost, p. 268: ' . . . victi sunt Scotti maxime per sagittarios Anglicorum, qui
primam aciem Scottorum ita excaecaverunt et vulneraverunt in facie continuis ictibus saggitarum
quod non poterant se juvare . . . '
47 See Thomas of Burton, II:364: 'Minores vero turmae, per sagittarios nimium laceratae,
adhaerere magno exercitui compelluntur, et in brevi Scotti conglobati alius ab alio premebatur.'
Page 120
each other back and forth.48 Eventually, the disinherited soldiers began to dominate
their opponents, both because of the unity of their defensive formation and because of
their attendant archery fire which continued to harass the flanks and rear of the Scots.
Soon the disheartened Scottish leaders, seeing their armies on the verge of defeat,
mounted their horses and fled the battlefield.49 Their troops, for the most part, were
unable to follow. Many were so embroiled in the conflict that they continued to fight,
but by now the disinherited had begun a steady uniform advance against the
disorganized and now leaderless Scots. Ultimately, this ended the battle, as the
remaining Scottish soldiers either fled or were killed.50
In the words of John Fordun, the Scots 'perished in a no less astounding than
unhappy massacre.'51 The Scottish dead numbered in the thousands, described by
most authors as a pile of corpses measuring more than a spear's length high.52 Many
had died not of wounds but of suffocation, having been crushed under the feet of
both armies' soldiers. As the anonymous author of the Liber Pluscardensis writes:
`more died by being smothered than by the sword, falling one upon another in such
numbers that so sad a catastrophe is not recorded to have happened for a very long
time.'53 They were buried in a very large and deep ditch.54
The English losses were much smaller, numbering only two knights and thirty-three
soldiers, but according to the Anonimalle Chronicle, `no archer nor any footman.'55 It
is recorded that they were buried at Edward III's own expense.
The battle of Dupplin Moor was followed less than a year later by the battle of
Halidon Hill. At this battle Edward Balliol and the disinherited were joined by Edward
III and the `official' English army. In considering the battle of Halidon Hill, most
sources begin by discussing Edward III's decision to attack Scotland, defending this
decision despite the treaty which bound the English king to respect the sovereignty of
David Bruce. The treaty with Bruce had been signed when Edward III was a minor
and under the control of his domineering mother. Furthermore, Edward Balliol was in
fact the just king of Scotland, his father having been forced to relinquish his throne at
the point of a sword wielded by two `Jacobin' friars. Finally,
48 Robert of Avesbury (p. 296), Henry of Knighton (I:463), John Fordun (I:354-55), the
continuator of Walter of Guisborough (II:304), and the Liber Pluscardensis (I:265) all agree
that the battle began at dawn and continued for most of the day. However, they all disagree
about when in the afternoon it was concluded.
49 See Thomas of Burton, II:364.
50 See Henry Knighton, I:463.
51 John Fordun, I:347: ' . . . non minus morte stupenda, quam infelici perierunt . . . '
52 See Chronicon de Lanercost, p. 268; Robert of Avesbury, p. 297; Thomas of Burton, II:364;
Thomas Walsingham, I:194; the continuator of Walter of Guisborough, II:304-05; and The Brut,
I:279. On the numbers of Scots killed at Dupplin Moor see Anonimalle Chronicle, pp. 150-51;
Chronicon de Lanercost, p. 268; Thomas Walsingham, I:194; John Fordun, I:355; and Andrew of
Wyntoun, II:388.
53Liber Pluscardensis, I:266: ` . . . plures suffocacione quam gladio interierunt, unus super alium
incidentes, in tali multitudine quod tam dolorosus casus a multis retroactis temporibus in cronicis
non legitur accidisse.' See also Thomas Walsingham, I:194; the continuator of Walter of
Guisborough, II:304-05; and Andrew of Wyntoun, II:388.
54 See Thomas Walsingham, I:194.
55 See Anonimalle Chronicle, pp. 150-51; Chronicon de Lanercost, p. 269; Henry Knighton,
I:463; Thomas of Burton, II:364-65; the continuator of Walter of Guisborough, II:304; and The
Brut, I:278.
Page 121
Edward III's northern province of Northumberland had suffered constant raids and
invasions by Scottish troops, actions which clearly violated the Anglo-Scottish
treaty.56
Thus, Edward III traveled north with a large force and began his siege of Berwick.
The siege itself is of great interest to most contemporary sources, many of whom
outline all of the events which occurred there: the laying of the siege, the stubbornness
of the besieged, the continual negotiations for raising the siege, Edward III's taking of
hostages, and the eventual execution of Thomas Seton.57 However, it is clear in these
sources that the siege by itself was not important. Edward undoubtedly wished to fight
a battle against the Scottish army, and he was using the siege of Berwick only to draw
the Scots into a military conflict.58 He even promised the Scots safe and unhindered
passage until they arrived at Berwick to fight with the English.59 The Scots, on the
other hand, did not want to fight against the English King. Instead, they tried to draw
Edward from Berwick by raiding Northumberland and besieging the castle at
Bamburgh.60 This tactic did not work, and the Scottish army was soon forced to
respond to the English challenge.
The Scots decided not to attack the English at Berwick, determining to fight the battle
at nearby Halidon Hill. Their army was large; contemporary sources estimate their
numbers to be between 14,629 and 100,000, almost all of whom were infantry.61 They
were led by the most experienced and high-ranking nobles in Scotland, men tactically
wise enough to order their troops in several lines on top of the hill.62 The
56 See Anonimalle Chronicle, pp. 156-59; Chronicon de Lanercost, p. 273; Thomas Gray, pp.
162-63; Geoffrey le Baker, p. 50; and Adam Murimuth, p. 67.
57 On the siege of Berwick see Chronicon de Lanercost, p. 273; Thomas Gray, pp. 162-63; Henry
Knighton, I:459; Geoffrey le Baker, pp. 50-51; Gesta Edwardi de Carnarvan, pp. 111-14;
Anonimalle Chronicle, pp. 156-63; Thomas of Burton, II:367-68; Adam Murimuth, p. 67; Thomas
Walsingham, I:195-96; the continuator of Walter of Guisborough, II:306-08; Jean le Bel, I:112-13,
116-17; The Brut, I:281-83; and Andrew of Wyntoun, II:399.
58 See Anonimalle Chronicle, pp. 158-59; The Brut, I:282; and Andrew of Wyntoun, II:398.
59 See Thomas Gray, p. 163.
60 On the trek of the Scots into Northumberland see Robert of Avesbury, p. 298; Gesta Edwardi
de Carnarvan, p. 113; Geoffrey le Baker, p. 51; Henry Knighton, I:459; Thomas Walsingham,
I:196; Thomas of Burton, II:367; and Adam Murimuth, pp. 67-68.
61 The numbers of Scots are recorded as (from lowest to highest): 14,629 (the continuator of
Walter of Guisborough, II:308-09); 30,000+ (Chronicon de Lanercost, p. 273); 40,000 (Gesta
Edwardi de Carnarvan, p. 115); 60,000 (Liber Pluscardensis, I:269; Andrew of Wyntoun,
II:399); 80,000 (Anonimalle Chronicle, pp. 162-63); 90,000 (Thomas of Burton, II:369-70);
100,000+ (The Brut, I:285). As well as these calculations, four chroniclers simply report the
Scottish army to contain a large number of soldiers without giving a count: Geoffrey le Baker, p.
51; Robert of Avesbury, p. 297; Adam Murimuth, p. 68; and the continuator of Walter of
Guisborough, II:308. Finally, the anonymous poet writing on the battle also gives no tally for the
number of Scottish soldiers but does claim that they outnumbered the English by a 5:1 ratio (Minot
p. 95). On the fact that these were largely infantry troops see Henry Knighton, I:459.
62 There is disagreement on the number of Scottish lines ordered at Halidon Hill. The Chronicon
de Lanercost (p. 273), Henry Knighton (p. 459), Geoffrey le Baker (p. 51), Jean le Bel (I:116),
and Thomas of Burton (II:370) report a formation of three lines, while the Anonimalle Chronicle
(pp. 162-63) and the continuator of Walter of Guisborough (II:308-09) records four lines, and The
Brut (I:384-85) notes five lines of Scottish soldiers. On the leadership of the Scots see
Anonimalle Chronicle, pp. 164-67 and The Brut, I:383.
Page 122
Scottish morale was high, for they outnumbered their enemy and had the higher
ground. This was all explained to them in a speech delivered by Archibald Douglas,
who also commanded that no English prisoners were to be taken. Noting the knights
present in the English army, Douglas remarked:
Take comfort my brothers and take up your arms with vigor. Behold, the lords of our enemy are
on horseback, so that some of them after being put to flight might more easily arrive at the head
of those fleeing. Fortune has led them to this position, as if they were enclosed by boundaries.
At their back lies a large town armed with bellicose men. On their right is the spacious and
deep ocean. And on their left run the waters of the Tweed River which is filled to overflowing
by the tides. Therefore, take spirit, and let your hands kill. Do not allow the English to be
ransomed, but let both the lords and the infantry be slain equally on this day.63
The English were forewarned of the Scots' advance, and they could see them ordered
on the nearby hill.64 While they marveled at the large number which had gathered for
the battle, they did not waver in their resolve to fight. This was, after all, what Edward
III had desired, and his troops were prepared for the conflict. Some of the English
troops had been left to watch the castle. Those who were to fight at Halidon Hill
numbered considerably fewer than their Scottish opponents, although only Thomas of
Burton provides a numerical estimate of their strength, at 10,000 soldiers.65 They were
also all expected to fight on foot, despite Geoffrey le Baker claiming that this was
'against the ancient tradition of their fathers.'66 A large number of archers was also
present with the English army.67
Edward ordered his army similar to the formation used by the disinherited at Dupplin
Moor. He formed three lines of infantry at the base of the hill in the shape of an arc,
with the wings composed of archers. It was strictly a defensive formation.68
Morale among the English troops seems also to have been quite high. For despite the
Anonimalle Chronicle's belief that the English troops were 'very cast down' at
63 Recorded in Gesta Edwardi de Carnarvan, p. 115: 'Confortamini, fratres, et arma sumite
cum vigore; ecce! duces hostium nostrorum equites sunt effecti, ut ceteris nobis in praedam
dimissis fugae praesidium melius consequantur. Nunc alia fortuna tanquam inclusos illos in
nostris finibus huc adduxit, hinc, siquidem a tergo, villam optimam viris bellicosis munitam; a
dextris vero pelagum spatiosum et altum, a sinistris autem alveum fluminis de Twede cujus
intumescens mare riparum marginem jam implevit. Igitur [animos] resumite et manus ad
caedendum extendite, nec cuiquam redemptio concedatur, sed principes et pedites pariter
pereant isto die.'
64 See Chronicon de Lanercost, pp. 273-74 and Jean le Bel, I:116.
65 Thomas of Burton, II:367. See also Thomas Walsingham, I:196.
66 Geoffrey le Baker, p. 51: 'Ibi didicit a Scotis Anglorum generositas dextrarios reservare
venacioni fugencium, et, contra antiquatum morem suorum patrum, pedes pugnare.'
67 See Chronicon de Lanercost, p. 274; Gesta Edwardi de Carnarvan, p. 114; Henry Knighton,
I:459; Thomas of Burton, II:370; and The Brut, I:285.
68 There is another dispute over the number of English lines ordered at Halidon Hill. The
Chronicon de Lanercost (p. 274), Gesta Edwardi de Carnarvan (p. 114) and Thomas of Burton
(II:370) all record three lines of formation, while the continuator of Walter of Guisborough
(II:309) and The Brut (I:285) report four lines. On the formation in an arc see the continuator of
Walter of Guisborough, II:309. And, on the archers ordered on the wings see Gesta Edwardi de
Carnarvan, p. 114, and The Brut, I:285.
Page 123

Battle of Halodon Hill


the sight of such a large Scottish relieforce,69 Adam Murimuth writes that they 'were
comforted by a strong mood and a brave spirit.'70 Although being outnumbered by
the Scots, Edward III knew that an even more disparate number of Scottish soldiers
had been defeated quite handily by the disinherited at Dupplin Moor, and he reckoned
that his army could repeat this performance. The English king relayed this message to
his soldiers while recalling the recent defeats suffered by the English at the hands of
the Scots:
Consider, my comrades, what kind of people we fight today. For a long time these people have
rebelled against our ancestors. It also pains me to remember the subversions and massacres
which they have borne against the people and ecclesiastics of our time. For, God willing, the
day of vengeance has come, unless you are frightened by their numbers. For, trusting in the
leadership of God, we will be at equal strength to them in this battle.71
So the two armies stood opposite each other, both ordered in their defensive
69Anonimalle Chronicle, pp. 162-63.
70 Adam Murimuth, p. 68: ' . . . et omnes bono vultu et audaci animo confortante . . . '
71 As recorded in Gesta Edwardi de Carnarvan, p. 115: 'Considerate, commilitones mei, cum
quibus gentibus dimicaturi sumus isto die. Diu est quod progenitoribus nostris rebelles extiterunt.
Piget itaque referre plebis et religiosorum subversiones et strages quas nostro generi pluries
intulerunt. Jam, Deo propitio, dies instat ultionis, nec illorum multitudinem timeatis, quia de
Domini praesidio confidentes vobiscum pares erimus in conflictu.' See also Anonimalle
Chronicle, pp. 162-63; Geoffrey le Baker, pp. 51-52; and Laurence Minot, p. 4.
Page 124
formations, with neither wishing to break from them. Finally, Geoffrey le Baker
reports, the Scots challenged the English to single combat. The situation was, as this
English author sees it, similar to that of David and Goliath, with the Scottish army
itself playing the Philistine giant. Their combatant is not named, but he is described as
using as well as his more conventional weapons a 'certain black dog;' the English
chose Robert Benhale, a knight from Norfolk. After receiving the blessing of Edward
III, Benhale went against his opponent with sword and shield. He first met the dog,
which he killed with his sword, and then, almost as quickly, he slew and decapitated
the dog's master.72
Still, no battle was being fought. The Scots had chosen the field in order to protect
themselves. They had in fact thought that the English would charge them, but
Edward's intent was exactly the opposite. He wished to provoke the Scots to charge
his lines instead of making his own charge. Ultimately, the stalemate was broken. The
Scots, confident in their numbers and perhaps provoked a bit by the loss of their
champion in single combat, attacked the English. Trying to take advantage of their
larger numbers, they combined their first line into a large mass and charged down the
hill into the English formation attempting, in the estimation of the Chronicon de
Lanercost, to reach Edward Balliol who stood among the troops at the rear of the
English army.73
It was, according to John Fordun, a 'lachrimabile bellum,' which ended in an extended
hand-to-hand fight between two masses of foot soldiers.74 The Scots attacked
'manfully,' but in the end they were unable to break the English line. The second and
third lines of the Scots now also charged into the English infantry, but they met the
same fate as their comrades.75
The archers again played a role in this defeat. As at Dupplin Moor they attacked the
Scots as they rushed onto the infantry lines, and they continued to fire into their flanks
and rear as the fight continued. In this, as at Dupplin Moor, they blinded many of the
Scots, creating disorder in their ranks and aiding in their slaughter. The Lanercost
chronicler writes:
Now the Scots approaching in the first division were so grievously wounded in the face and
blinded by the host of English archery, just as they had been formerly at Glendenmore (Dupplin
Moor), that they were helpless, and quickly began to turn away their faces from the arrow
flights and to fall.76
72 Geoffrey le Baker, p. 51.
73Chronicon de Lanercost, p. 274.
74 John Fordun, I:356. See also Adam Murimuth, p. 68.
75 See Chronicon de Lanercost, p. 274; Gesta Edwardi de Carnarvan, pp. 115-16; Thomas of
Burton, II:370; Geoffrey le Baker, pp. 51-52; and Liber Pluscardensis, I:269. Only the
Anonimalle Chronicle differs with this scenario. This chronicle claims that it was not the Scots
who attacked but the English who 'advanced boldly' against their enemy. The Scots 'did not want
to fight against the king of England and his forces' because they were waiting for the River Tweed
to be at full flood tide, 'that they might drive the king of England and his men into the River Tweed
or into the sea to drown them.' However, this rather fanciful strategy did not occur before the
English were victorious in this battle (pp. 166-67).
76Chronicon de Lanercost, p. 273-74: 'Sed Scotti in prima acie venientes ita fuerunt a multitudine
saggitariorum Angliae vulnerati in facie et excaecati in hoc bello, sicut in priori apud
Gledenmore,
(footnote continued on next page)
Page 125
The Anonimalle Chronicle uses similar words: 'The English archers destroyed and
injured them so that they were in a short time as if choked and blinded, and soon they
were thrown into confusion.'77
In the end, the English prevailed, 'indubitablely' to quote Thomas Walsingham.78 The
Scots were overcome, and while some who were still embroiled in the battle
continued to fight, those able to do so fled from the field. Eventually, all Scots either
fled from the field or were killed. The Brut describes the victory: 'And thus it
happened, as God willed it, that the Scots on that day had no more numbers nor
strength to face the English than twenty sheep would have against five wolves.'79
The English knights remounted and pursued the fleeing Scots for as many as eight
miles. They continued to kill any soldiers they came across, and the Scots who eluded
their pursuers did so, claims Adam Murimuth, only with 'much difficulty.'80
Dead Scottish soldiers covered the battlefield. The Brut describes the scene: 'And
there might be seen many Scotsmen thrown dead down to the ground, their banners
broken and hacked into pieces, and many a good haubergeon bathed in blood.'81 The
contemporary English chroniclers estimate a death rate between 35,308 and 60,718
with the Scottish historians tallying a lower, although still significant 10,000 total. This
number included several notable Scottish leaders including the general in charge,
Archibald Douglas.82 The English death rate was notably smaller, estimated to be
under twenty men including only one knight and one man-at-arms.83
(footnote continued from previous page)
quod se ipsos adjuvare non poterant, et ideo cito faciem saggitarum ictibus avertere et cadere
inceperunt.'
77Anonimalle Chronicle, pp. 166-67: 'les archers Dengleterre les desbaretta et greva, ainsi quils
estoient en petite hure auxi come estuffez et envoegles, et tost perdirent lour contenance.' See also
Thomas of Burton, II:370 and The Brut, I:285. None of these chronicles indicates deaths by
archery fire.
78 Thomas Walsingham, I:196.
79The Brut, I:285: 'And pus hit bifelle, as God wolde, pat pe Scottis hade pat day no more
foisoun ne myght apeynes pe Englisshe-men, pan xx shepe shulde have a eyns v wolfes.' See also
John Fordun, I:356-57 and the anonymous poet in Minot p. 95. That some Scots continued to fight
after most had fled see Bridlington, p. 116.
80 Adam Murimuth, p. 68. See also Anonimalle Chronicle, pp. 168-69; Gesta Edwardi de
Carnarvan, p. 116; Thomas of Burton, II: 370; Liber Pluscardensis, I:269-70; The Brut, I:285;
Andrew of Wyntoun, II:401; and the anonymous poet in Minot p. 95.
81The Brut, I:285: 'And pere might men see meny a Scottisshe-man caste doun vnto pe erthe dede,
and hir baneres displaiede, & hackede into pices, and meny a gode habrigoun of stele in hir blode
bapede.' See also Andrew of Wyntoun, pp. 401-02.
82 English chroniclers record the Scottish death total as (from lowest to highest): 30,000+, not
including 'valiantz gentz et pedaille' (Anonimalle Chronicle, pp. 168-69); 35,308 (Henry
Knighton, I:459); 35,712 (The Brut, I:286); 36,320+ (Chronicon de Lanercost, p. 274); 38,745
(Thomas of Burton, II:370); 40,000 (Robert of Avesbury, p. 298); 50,302 (anonymous poet in
Minot p. 96); 60,000 (Geoffrey le Baker, p. 52); and 60,718 (The French Chronicle of London, p.
67). As well, Adam Murimuth calculates both the number who died and the number who fled as
80,000 (p. 68). The Gesta Edwardi de Carnarvan (p. 116), Thomas Walsingham (I:196), and the
continuator of Walter of Guisborough (II:309) all record no death tally, indicating instead only that
'many' were killed at Halidon Hill. As for the Scottish chroniclers, both Liber Pluscardensis
(I:270) and Andrew of Wyntoun (II:401-02) report 10,000 casualties. On the names of the
important Scots killed see Thomas Gray, p. 163 and John Fordun, I:356.
83 See Gesta Edwardi de Carnarvan, p. 116; Henry Knighton, I:459; Thomas of Burton, II:370;
and Thomas Walsingham, I:196. The anonymous poet reports only seven English dead (in Minot p.
95).
Page 126
The war with Scotland seemed at an end. Edward III had travelled north, and he had
defeated the largest and best army the Scots could gather. As Adam of Murimuth
writes:
And thus it was publicly reported that war in Scotland was finally at an end, because no one
remained in that land who could, knew how to, or wished to gather men for battle, or even if
they did, no one knew how to lead those who might be gathered.84
This report was, claims Geoffrey le Baker, 'false opinion,' as war would be renewed
later with David Bruce himself.85 Still, the conclusive nature of this battle and that at
Dupplin Moor should not be underestimated. These two battles are similar in tactics,
strategy, purpose, and result. They represent the character of Edward III's early reign,
showing a young king willing to use force to protect the interests of those he favored
and to correct the 'injustices' of his father's rule. This was certainly comprehended by
the contemporary authors who wrote about the battles, as many original sources pair
the two with little discussion of the other historical events which occurred between
August 11, 1332 and July 19, 1333.
These battles represent a definite shift in English battlefield tactics. The idea of using
dismounted knights in a defensive formation and of using archers on the flanks to
provoke an attack, a tactic forced upon the troops at Boroughbridge perhaps because
of geographic necessity rather than tactical forethought, appears now to be a 'practice'
rather than an 'accident.' It also indicates an army learning from its mistakes; the defeat
at Bannockburn in fact had led to military changes.
With the exception of digging ditches on the battlefield, the Scots began both these
battles without altering the tactics which had brought them victory at Loudon Hill and
Bannockburn. In both battles, they chose the field and seemingly had the better
defensive position.86 They also had the numerical advantage and were fighting to
protect their homeland from an English invasion.
Yet none of these usually overwhelmign advantages seemed to aid the Scots in these
two battles. For, ultimately, it was the tactical changes made by the English armies
which determined the outcome of these battles. The English knights dismounting and
acting essentially as foot soldiers, ordered in a solid defensive line, was new to and
not anticipated by the Scots. They had so influenced their enemy with the victory at
Bannockburn that the English changed their tactical formation and now stood awaiting
a Scottish charge, just as Robert Bruce had done nearly twenty years before.
The Scots appear not to have learned of this change in English tactics employed
84 Adam Murimuth, p. 68: 'Et sic dicebatur publice quod guerra in Scocia fuit finaliter finita et
terminata, quia nullus remansit de natione illa, qui posset, sciret, aut vellet homines ad
proelium congregare aut regere congregatos.'
85 Geoffrey le Baker, p. 52.
86 Richard Lescot, for example, in referring to the Scots' battlefield choice at Halidon Hill
determined that it was a wise choice, the Scots having picked the higher ground 'desiring to
protect themselves' (p. 32).
Page 127
so successfully at Boroughbridge, nor do they seem to have realized that the changes
which the disinherited used for the victory at Dupplin Moor would be similarly used
at Halidon Hill. This certainly seems to be evident in the speech given by Archibald
Douglas to his troops before the latter battle, for not only does he maintain that the
enemy knights would be on horseback, but he tries to draw a social contrast between
those on horseback and those on foot, claiming that the knights were mounted only so
that they would be able to flee more easily than those without horses. Thus it appears
that the Scots were caught without significant knowledge of how to fight the new
tactics of the English army.
Moreover, the English not only changed their tactics by dismounting the knights and
fighting on foot, they also changed their tactics by refusing to make an offensive
charge against their enemy. This too had been taught them at Loudon Hill and
Bannockburn. In both of those battles the English had taken the offensive initiative
and had charged unsucessfully into the Scottish line. At Dupplin Moor and Halidon
Hill they reversed this practice, choosing to wait for a Scottish charge rather than
initiating combat with their own attack. At Dupplin Moor they did not have to wait too
long, for the Scots, who so outnumbered the English that they in no way could
anticipate defeat, attacked the disinherited with little delay, despite their tradition of
taking a defensive stance and waiting for their opponent's charge. However, at Halidon
Hill, the Scots tried to revert to their proven defensive tactics, attempting, at least
according to Geoffrey le Baker, to provoke the English into an attack on their solid
infantry line.87 But in this battle, with Berwick castle suffering the hardships of a
siege, the Scots could not afford to wait for the possible English charge of their
infantry line. Thus, they again put their faith in their numerical advantage and
attempted their own, ultimately unsuccessful, attack of the English defensive
formation.
It was at this point in the two battles when the archers made their presence felt.
Although it may be too much to say, as Jonathan Sumption does, that these battles
'were [both] won by archers,' the archers did play a major role in the battle, although
not the oneas a decisive killing machinewhich has been bestowed upon them
traditionally by scholars.88 Without having ditches or some other hindrance to control
an offensive charge, the English had to do something to narrow the size of the
Scottish attack, otherwise the Scots might have outflanked the infantry line and
destroyed the advantage of the defensive formation. Therefore, it was the function of
the archers, ordered along the infantry flanks, to provide this hindrance and to narrow
the Scottish attack. In essence, they performed the same purpose as had the ditches at
Courtrai and Loudon Hill, the woods and marshes at Bannockburn, and the bridge at
Boroughbridge. In this duty they performed marvelously, so well in fact that this tactic
would become standard in all battles fought by the English for the next two hundred
years. For even if the archers did not kill many, which seems to be what occurred in
these two battles as in most others fought during the fourteenth and
87 Geoffrey le Baker, p. 51.
88 Sumption, p. 67. Sumption is in error on pp. 125-26 in describing 'some thousands of Scots
[dying] of arrow wounds.' There is no record of this in the contemporary battlefield narratives.
Page 128
fifteenth centuries, they so disrupted and disordered their opponent's charges that
when they finally did reach the English line their impetus was slight and the infantry
and dismounted knights were more easily able to withstand the attacks.
Perhaps then, Geoffrey le Baker is not simply using rhetorical license in his
comparison of the battles of Dupplin Moor and Halidon Hill to that fought between
David and Goliath. In both cases it was the larger entity which came forward to attack
while the smaller stayed back in defensive anticipation. And, as with the larger Goliath
whose weakness caused his defeat at the hands of the smaller David, in these two
battles the larger Scottish army was easily defeated by the smaller English force.89
89 Geoffrey le Baker, p. 51.
Page 129

X
The Battle of Laupen, 1339
With the disintegration of central power in the thirteenth-century Holy Roman Empire
there came a desire by many living on the 'frontier' of the Empire to seek
independence and self-rule. Of these frontier entities, the most obvious to modern
historians are undoubtedly the city-states of northern Italy. By the end of the thirteenth
century, Venice, Milan, Florence, and Genoa, to name only the most prominent of
these independence-seeking states, had shed their German overlords and begun the
process of self-government. To Jacob Burckhardt and his scholarly descendants this
was the beginning of the Italian Renaissance, a period which saw constant military
confusion coupled with the development of humanistic and artistic excellence.1
Nearly forgotten in comparison with the Italian experience is that of Switzerland.
Associated with the rest of the Holy Roman Empire since the time of Charlemagne,
but different in language, culture, society, and economy, during times of Imperial
solidarity, the medieval Swiss had usually remained faithful to the political powers-
that-were. But in times of governmental instability, the slightly populated Switzerland
was more likely to be overlooked than concerned with. Swiss towns, far smaller than
their German or Italian counterparts, learned to be relatively self-sufficient.2 To
survive they connected with neighboring rural enclaves to form independent political
organizations, known as cantons. In 1291, three of these contons, Uri, Schwyz, and
Unterwalden, allied to form a union, the first Swiss Confederation.3
While not generating much interest in the Holy Roman Empire as a whole, the Swiss
Confederation did concern the Austrian Habsburg family whose holdings
1 Jacob Burckhardt, The Civilization of the Renaissance in Italy, trans. L. Geiger and W. Götz,
2 vols. (New York, 1958). See also Eugenio Garin, Der Italienische Humanismus (Bern,
1947); Hans Baron, The Crisis of the Early Italian Renaissance, 2 vols. (Princeton, 1955);
and Denys Hay, The Italian Renaissance in Its Historical Background (Cambridge, 1961).
For a survey of this concept in Renaissance historiography see Albert Rabil Jr., 'The
Significance of ''Civic Humanism" in the Interpretation of the Italian Renaissance,' in
Renaissance Humanism: Foundations, Forms, and Legacy, vol. 1: Humanism in Italy, ed. A.
Rabil, Jr. (Philadelphia, 1988), pp. 141-74.
2 Denys Hay, Europe in the Fourteenth and Fifteenth Centuries, 2nd ed. (London, 1989), pp.
129-30.
3 Hay, Europe in the Fourteenth and Fifteenth Centuries, p. 130; E. Bonjour, H.S. Offler, and
G.R. Potter, A Short History of Switzerland (Oxford, 1952), pp. 69-81; W.D. McCrackan, The
Rise of the Swiss Republic, 2nd ed. (New York, 1901), pp. 83-91; Richard Feller, Geschichte
Berns (Bern, 1949), I:63-64; and George Holmes, Europe: Hierarchy and Revolt, 1320-1450
(London, 1975), p. 52. The date of confederation is not known. The first documentary evidence of
alliance is August 1291. (A transcription and translation of this document can be found in
McCracken, pp. 87-89.)
Page 130
included those independence-seeking lands. This concern increased further in 1292
when the Confederation united with Zurich and Bern to form an anti-Habsburg
league.4 However, Albrecht I of Austria, deprived, he felt, from his rightful
inheritance as Holy Roman Emperor, was tied up for the moment with a political
struggle to gain the Imperial throne, and he could not immediately attend to the Swiss
independence crisis.5 Indeed, it would not be until 1315, with Albrecht's sons,
Frederick and Leopold, on the throne of the Holy Roman Empire and Austria
respectively, that a concerted effort was made by the Habsburgs to break the
Confederation. At this time, with the Swiss Confederation supporting a rival candidate
to the Imperial throne, Lewis IV of Bavaria, Duke Leopold of Austria mounted a
major campaign against his rebelling Swiss lands and their inhabitants. In their 24
years of independence the Swiss had been preparing for an invasion which had never
come, but they had not dropped their guard, and when Leopold invaded the
geographically isolated Switzerland he found well maintained fortifications barring
easily accessible passes and pushing him to journey along more dangerous routes.
More importantly, he found a population which was not willing to give up their
independence nor willing to fight by conventional means. At Morgarten on November
14, 1315, while the Austrian army traveled through one of these more dangerous
passes, the Swissprimarily men from the Schwyz and Uri cantonsambushed and
massacred them.6
The battle of Morgarten was a trumpeting of Swiss independence and military success
which has continued to sound until this century.7 With their unanticipated defeat of
the Austrians in that battle, the Swiss Confederation initiated a peace with the Holy
Roman Empire and its affiliates which held for much of the following century.
However, peace was a relative condition, for while few invading armies proceeded
through the alpine passes against Swiss inhabitants, Swiss armies fought frequently
among themselves. Much of this was due to the desire for territorial expansion by the
larger Swiss towns, few of which had initially supported the Confederation or its
aspirations for independence, but all of which now profitted from what had taken
place at Morgarten. Among the most active in this regard were Lucerne, Zurich,
Freiburg, and Bern.8
It is estimated that in the early fourteenth century Bern had a population of only five
to six thousand, quite small in comparison to non-Swiss towns, but similar in size to
Zurich and twice as large as Lucerne and Freiburg.9 Bern had joined the Swiss
4 Bonjour, Offler, and Potter, pp. 81-85; McCracken, p. 116; and Feller, I:64-65.
5 Feller, I:101-09 and G. Barraclough, The Origins of Modern Germany, 2nd ed. (Oxford, 1947),
pp. 302-07.
6 Bonjour, Offler, and Potter, pp. 85-86; McCrackan, pp. 117-28; Feller, I:109-10; and Hans
Witzig, Von Mortgarten bis Marignano (Zurich, 1957), pp. 12-25. For a discussion of the battle
of Mortgarten see Appendix.
7 On the endurance of the battle of Mortgarten in Swiss political thought see Holmes, p. 52; Maria
Schnitzer, Die Mortgartenschlacht im werdenden schweizerischen Nationalbewusstsein (Zurich,
1969); and Fritz Wernli, Die Entstehung der schweizerischen Eidgenossenschaft (Zurich, 1972),
pp. 277-94.
8 Bonjour, Offler, and Potter, pp. 87-96; McCrackan, pp. 129-46; and Feller, I:110-28.
9 Bonjour, Offler, and Potter, pp. 93-94. Feller's 3,000 inhabitant tally for Bern in 1300 (I:78)
seems
(footnote continued on next page)
Page 131
anti-Habsburg alliance in 1292 and had always remained friendly with the Swiss
Confederation, even signing a mutual assistance pact with it, the treaty of Lungern, in
1323.10 (Bern would not, however, join the Confederation proper until 1353.)
Bern's friendliness with these Swiss 'rebels' was not approved of by its neighboring
town of Freiburg. Eighteen miles from Bern, Freiburg was much more tightly
controlled by a noble family, the Kiburgs, which had numerous smaller dynastic
attendants and which was also closely allied with the Habsburgs. Since 1315 both
Freiburg and Bern had sought to expand their territory beyond earlier boundaries.
Weak and unclaimed lands were occupied, and pacts were made with nearby villages.
As they were so close to each other, on many occasions the two towns encroached on
the other's lands; indeed, trespassing raids were frequent. But it was not until 1339 that
open warfare occurred. In that year the Freiburgers, together with their allies and
confident in the size and nobility of their army, determined to bring the Bernese to
battle. On June 10 they besieged the village of Laupen. Once held by the Kiburg
family, Laupen was now firmly under the control of Bern, with more than 600 Bernese
soldiers in garrison there. With haste, the Bernese reacted to the siege by mustering a
largely infantry army and marching to relieve Laupen; on June 21, less than two miles
from the village, they met and defeated the more noble, cavalry-oriented Freiburger
troops.11 The trend had continued: once again an anti-Habsburg, pro-independent
Swiss military force had proved victorious on the battlefield.
The battle of Laupen is not recorded in many contemporary historical sources,
although what does exist is fairly extensive in its coverage of occurrences on the
battlefield. Most important in this regard is the Conflictus Laupensis, written in Bern
shortly after the battle.12 Despite being judged as 'unjustifiably famed as an important
achievement of military history literature' by Hans Delbrück, who nonetheless uses
this narrative as his primary source for the battle of Laupen, the Conflictus Laupensis
is one of the very few medieval military narratives written within a year or two of
what it reports.13 And while it certainly is written only from the Bernese point of
view, its description of the battle is supported by later sources, nearly all of which also
support the Bernese postion, but none of which used the Conflictus Laupensis for
their own narratives. These include: Jean de Winterthur's Chronica, written c. 1347;
Heinrich von Diessenhoven's Chronicon, whose short account of the battle was
written c. 1361; the Cronica de Berno, which was completed in 1405 but
(footnote continued from previous page)
too small, especially as he later accepts a figure of 1,000-4,000 for the number of Bernese
soldiers at the battle of Laupen (I:134-39).
10 Bonjour, Offler, and Potter, pp. 94-95; McCrackan, pp. 157-59; and Feller, I:116-17.
11 Bonjour, Offler, and Potter, pp. 93-95; McCrackan, pp. 159-60; and Feller, I:129-34. On the
battle of Laupen see Delbrück, III:561-70; Oman (1905), II:242-47; Feller, 1:134-41; Witzig, pp.
27-32; G. Köhler, Die Entwickelung des Kriegswesens und der Kriegführung in der Ritterzeit
von Mitte des 11. Jahrhunderts bis zu den Hussitenkriegen (Breslau, 1886), II:605-13; and R.
von Fischer, Schweizerkriegsgeschichte, ed. M. Feldman and H.H.G. Wirz (Bern, 1915), I:9-21.
12Conflictus Laupensis, in Die Berner-Chronik des Conrad Justinger, ed. G. Studer (Bern,
1871), pp. 302-13.
13 Delbrück, III:562. Delbrück also disagrees with the number of soldiers on both sides reported
in the Conflictus Laupensis.
Page 132
with the section on Laupen undoubtedly written earlier; Conrad Justinger's Berner-
Chronik, written c. 1421; and an anonymous Stadtchronik of Bern, which was derived
from Justinger's chronicle and written in the middle of the fifteenth century.14

The Conflictus Laupensis begins its narrative with a lengthy discourse on the origin of
the conflict between Freiburg and Bern. Simply put, the anonymous author reports,
it is known that before the time [of battle] discord between Bern and Freiburg and their allies
had been seeded for many years and from many causes and on many occasions by both sides.15
Ultimately, in 1339, with the Bernese `working for the common peace and for the
conservation of the land,' two nobles associated with Freiburg, Gerhard of Valengin
and Peter of Aarburg, increased their malevolent rhetoric and raids on the lands of
Bern. The Bernese, frightened by the threats of these nobles, answered with an
expedition on May 16 against the fortifications of Aarburg; the expedition failed, as
the attackers were unable to draw the counts from the fortifications' protective
walls.16
This expedition, however, did prompt the Freiburgers to lay siege to Laupen, at least
this is the reason given for the siege in the Conflictus Laupensis. The other sources do
not record the failed assault on Aarburg, indicating rather that the siege of Laupen
occurred because the count of Kiburg had a large noble force which he felt could
easily defeat the army of Bern.17
Whatever the cause, the siege of Laupen took place, with all of the Freiburgers and
their allies attacking the town `with catapults brought on carts by horses.'18 The date
was June 10. The army of Freiburg was large and impressive. Although the
Conflictus' count of 16,000 infantry and 1,000 cavalry, the Cronica de Berno's tally of
24,000, and Conrad Justinger's number of 30,000 soldiers are most certainly
exaggerations based on the small population of the town,19 the presence of the counts
of Kiburg, Neuchâtel, Gruyères, Nidau, Valengin, Aarburg, and the lord of Vaud, son
14 Jean de Winterthur, pp. 162-64; Heinrich von Diessenhoven, Chronicon de Heinricus
Truchsess von Diessenhoven, in Fontes rerum Germanicarum, iv, ed. J.F. Bohmer (Stuttgart,
1868), pp. 32-33; Cronica de Berno, in Die Berner-Chronik des Conrad Justinger, ed. G.
Studer (Bern, 1871), pp. 299-300; Conrad Justinger, Die Berner-Chronik des Conrad
Justinger, ed. G. Studer (Bern, 1871), pp. 72-94; and Die Anonyme Stadchronik oder der
Königshofen-Justinger, in Die Berner-Chronik des Conrad Justinger, ed. G. Studer (Bern,
1871), pp. 352-72.
15Conflictus Laupensis, p. 302: `Sciendum est ergo, quod ante tempus prefatum fuit orta
discordia inter Bernenses et Friburgenses et adiutores eorundem, que ab annis multis et ex
diversis causis et occasionibus hinc inde ab utraque parte fuit seminata.' A discussion of these
causes is found on pp. 302-05.
16Conflictus Laupensis, pp. 305-06.
17 Jean de Winterthur, p. 162; Conrad Justinger, pp. 74-82; and Die anonyme Stadtchronik, pp.
353-56.
18Conflictus Laupensis, p. 306.
19Conflictus Laupensis, p. 306; Cronica de Berno, p. 300; and Conrad Justinger, p. 82. Hans
Delbrück (III:563) believes that these tallies are far from accurate and that in fact the Freiburger
army may have been smaller than the Bernese force.
Page 133
of the count of Savoy, as well as many lesser nobles, with all their retainers, as
recorded in the original sources, gives an impression of chivalry that is not present in
many other battles of the early fourteenth century: 'They gloried in their multitude and
power and in the many decorations of their new and costly vestments.'20 The
Conflictus Laupensis and Jean de Winterthur insist also that the besieging army
contained a number of Austrians.21
Opposing the Freiburg besiegers was a garrison of 600, a number which could not
hold out for long against such a siege.22 Those inside the town could also not hope
for any mercy from their attackers; as Laupen had once been part of the Freiburg
holdings and now sided with the enemy Bernese, the Conflictus Laupensis reports that
the Freiburgers were planning to destroy the town and massacre all of its
inhabitants.23
Thus it was up to the Bernese to enter into the conflict and to assist their confederates.
The leaders of the town met in council, but their meeting was brief as they quickly
agreed to attempt to raise the siege of Laupen. There was urgency in the matter as the
Bernese feared that if they dallied Laupen would fall and its inhabitants would be
killed. At the same time, they believed that God was on their side and would aid them
in their cause. To them it was a crusade, as their banners, a dark cross on a white
background, and the large number of ecclesiastics traveling with the army proved.24
Six thousand Bernese troops answered the call to arms, including 1,000-1,200 from
the three cantons.25 They were led by an experienced warrior, the knight Rudolf von
Erlach.26
The Bernese force marched to Laupen, but when they saw 'that the reported large size
of their enemy was valid,' the leaders decided not to attack the Freiburgers while they
were camped in the steep and forested terrain around the town. Instead, they
determined to order their troops, all infantry, as 'a small wedge'the Conflictus
Laupensis uses the term 'parvus cuneus'with troops facing out on all sides on top of a
small hill near to the town.27 Not only would this position give the Bernese the
20 See Conflictus Laupensis, p. 306; Jean de Winterthur, p. 162; Cronica de Berno, pp. 299-
300; and Die anonyme Stadtchronik, pp. 354-56. The quotation is from the Conflictus
Laupensis, p. 306: 'gloriantes de sua multitudine et potentia ac in ornatu vario vestium novarum
et pretiosarum.'
21Conflictus Laupensis, p. 306 and Jean de Winterthur, p. 162.
22 Conrad Justinger, pp. 82-86.
23Conflictus Laupensis, p. 307.
24Conflictus Laupensis, pp. 307-09 and Conrad Justinger, pp. 82-83. The Conflictus identifies a
number of Bernese leaders including Lord John of Bubenberg, a knight who is identified as the
'senior' of Bern; Burchard of Bennenwyle, the town secretary; Burchard, the master of machines;
Rudolf of Muleren, Peter of Balme, Peter Wentzschaz, and John of Herblingen, standard-bearers;
John of Seedorf, Berthold of Seedorf, Berthold Glockner; and Peter of Krantzingen.
25 The Conflictus Laupensis (pp. 308-09) counts 1,000 canton soldiers, with the Cronica de
Berno (p. 300) counting 1,200. Both total the Bernese force at 6,000. This number equals the
population of the town and thus appears too high; however, it must also have included troops from
throughout the Bernese controlled region of Switzerland.
26 Only Conrad Justinger (pp. 82-84) names Rudolf von Erlach as leader of the Bernese force.
The Conflictus Laupensis does not mention a leader of the Bernese. Delbrück's arguments in
support of Erlach as leader are convincing (III:564-65).
27Conflictus Laupensis, p. 309.
Page 134

Battle of Laupen
(First Phase)
higher-ground advantage during the battle, but it also allowed them to be warned of
the Freiburg approach. As well, Conrad Justinger insists that Rudolfvon Erlach wrote
to the leaders of the Freiburg army challenging their force to battle; the Freiburgers
readily agreed.28
This was after all the goal of the Freiburgers, and they quickly prepared for the fight.
Their 'new knights' were particularly enthused by the prospect of engaging the
Bernese in warfare: they waved their swords in the air in excitement and raced to
attack their opponents.29 Seeing this, some of the Bernese fled from their positions in
frightthese numbered some 2,000, according to the Conflictus Laupensis. But this still
left some 4,000 soldiers, all of whom stood in their formation ready to fight against
the oncoming Freiburg army.30
The Freiburg cavalry reached the Bernese formation first. The Conflictus recounts that
they initially rode 'frightfully' around the Bernese formation, waited for an infantry
assault to soften the target, and then attacked 'fiercely' the Bernese line themselves.
The Bernese reacted by standing solidly in formation against the attacks:
The Bernese, bursting the chains of all their fear, like Samson, received the assault of the
Freiburgers. And they took all of their Standards and slew the standard-bearers and many
others, all infantry, and the rest they put to a pitiable flight. When those who were circling with
their horses charged to their aid, the Bernese without delay either killed all of them or put them
to flight.31
28 Conrad Justinger, p. 83. See also Die anonyme Stadtchronik, pp. 356-58.
29Conflictus Laupensis, p. 309. What constituted a 'new knight' is not explained in the narrative.
30Conflictus Laupensis, p. 309. Conrad Justinger, on the other hand, contends that as these fleeing
troops looked back to see their comrades remaining to fight, most regrouped and returned to the
formation (p. 89). Delbrück doubts this contention (III:566-67).
31Conflictus Laupensis, p. 309: 'Ipsi Bernenses, more Sampsonis, quasi ruptis vinculis omnis
timoris, in se agressos ipsos Friburgenses receperunt, et omnia vexilla eorum protinus abstulerunt,
vexilliferis eorum et multis allis occisis ceterisque peditibus omnibus in fugam miserabilem
conversis;
(footnote continued on next page)
Page 135

Battle of Laupen
(Second Phase)
In this action, as indicated, large numbers of Freiburgers were killed. Those especially
targeted seem to have been the knights among their force; the counts of Nidau and
Valengrin were killed, as was the lord of Vaud. All others fled. Among the booty taken
by the Bernese were 27 banners and 80 crowned helmets.32
The Bernese had not lost as manyan estimated 1,000-1,500 of their force had
died33and they did not pursue their fleeing opponents for long. Instead, they returned
to Laupen and Bern where a rejoicing population presented honors to the victors and
said prayers for the victory.34 Never again would these towns come to battle, although
raids into each others' lands continued for many years.35
In comparison with other early fourteenth-century battles the Swiss conflict at Laupen
between the armies of the towns of Freiburg and Bern was quite small. Perhaps only
the battles of Loudon Hill and Boroughbridge studied here can compare with the few
troops which fought on both sides. Yet the tactics used by the Bernese infantry and the
outcome they caused against the Freiburger cavalry were very similar to almost all of
these other battles.
The Bernese, drawn into the battle by a siege of their allied town, Laupen, were able
to muster far fewer troops than those they faced. They were also all infantry, while
their opponents had a large number of noble cavalry soldiers. The look of these alone
frightened a good number of the Bernese into fleeing from the fight. But those
(footnote continued from previous page)
et ad auxilium eorum, qui sub equitibus erant circumdati, se convertentes, ipsi Bernenses sine
mora universos aut occiderunt aut in fugam verterunt.' See also Jean de Winterthur, p. 163 and
Conrad Justinger, pp. 90-92.
32Conflictus Laupensis, pp. 309-10; Jean de Winterthur, pp. 163-64; Heinrich von Diessenhoven,
pp. 32-33; Cronica de Berno, p. 300; Conrad Justinger, pp. 91-93; and Die anonyme
Stadtchronik, pp. 367-69. The Cronica de Berno is the only chronicle which numbers the
deceased at 4,000.
33 Jean de Winterthur (p. 163) claims a death total of 1,000, the Conflictus Laupensis (p. 309)
1,500.
34Conflictus Laupensis, p. 310.
35Conflictus Laupensis, pp. 310-13.
Page 136
who remained were able to choose the battlefield, selecting a small hill on which they
could order a solid defensive formation. This choice gave them a great advantage
against their foes, for although the Bernese troops could not alter the battlefield to
narrow the Freiburger charges and were destined ultimately to be surrounded, the
Freiburgers had to charge uphill into a 'wedge' of spear-wielding soldiers. This
became a task which first the infantry was unable to do and then the cavalry.
Had either the Freiburger cavalry or their infantry broken through the Bernese line, the
battle would have been quickly decided and the result would have been quite
different. With so few troops, and no reinforcements to fill the opened gaps, as the
Bernese stood almost back to back in their wedge, the formation easily would have
folded. However, it did not. Indeed, the infantry stood so solidly at Laupen that one
must wonder if their formation could ever have been defeated by simply charging into
it. At Cassel, fought eleven years earlier in Flanders, the French had chosen not to
charge up the hill into the Flemings, but instead to wait until they could draw them out
of their defensive formation. Ultimately, it worked, and the French were victorious.
The French had faced the Flemings before and knew what to expect; as such, they had
to humble the pride of their cavalry in the face of what appeared to be a 'lesser' army.
But at Laupen, the Freiburg leadership exulted in their chivalric numbers and saw no
need for any other strategy except for an immediate attack on the Bernese infantry.
Failure, coupled with large numbers slain, followed.
Page 137

XI
The Battle of Morlaix, 1342
While the Hundred Years War began with conflicts, largely skirmishes, fought between
French and English troops in 1337-1339, it was not until the year 1340 that the war
effectively began. At Sluys, on June 24, 1340, a large naval battle was fought and won
by England.1 However, the English king, Edward III, despite having a large army of
his own and a gathering of alliesmostly from the Low Countriesunprecedented in the
history of the Middle Ages, was unable to take advantage of this victory; his siege of
Tournai, undertaken between July and September 1340, was thwarted by financial and
political collapse at home and by the disintegration of his alliance in the southern Low
Countries.2 The year ended with a peace treaty, the Treaty of Esplechin, signed on
September 25 between Edward and King Philip Vl of France, the latter having arrived
at Tournai during the siege but having been satisfied to wait out its conclusion without
trying to relieve it militarily.3
Philip returned to Paris, in somewhat of a celebration, for although his navy was
destroyed, he had been able to bring peace to his land without involving his army.
Moreover, the Treaty of Esplechin had promised no warfare between England and
France for five years, and this would give the French king time to rebuild his fleet, to
reform alliances with his Low Countries' Lords, and to prepare his defenses for
further assaults.
Edward's return to England was not so happy. Although much of the kingdom's
financial crisis could be laid at the king's feet, Edward blamed it, and even the failure
at Tournai, on his chancellor, Archbishop John Stratford, and his banker, William de
la Pole.4 Edward's words to Pope Benedict Xll on November 18 show his anger: 'I
1 See Kelly DeVries, 'God, Leadership, Flemings, and Archery: Contemporary Perceptions of
Victory and Defeat at the Battle of Sluys, 1340,' American Neptune 55 (1995), 1-28.
2 See Kelly DeVries, 'Contemporary Views of Edward III's Failure at the Siege of Tournai, 1340,'
Nottingham Medieval Studies 39 (1995), 70-105.
3 Lucas, pp. 420-2 and Sumption, pp. 357-8.
4 For Edward's money problems at the siege of Tournai see Lucas, p. 418; G.L. Harriss, King,
Parliament, and Public Finance in Medieval England to 1369 (Oxford, 1975), pp. 278-82; T.H.
Lloyd, The English Wool Trade in the Middle Ages (Cambridge, 1977), pp. 155-56; Nathalie B.
Fryde, 'Edward III's Removal of his Ministers and Judges, 1340-1', Bulletin of the Institute of
Historical Research 48 (1975), 150-3; and E.B. Fryde, William de la Pole: Merchant and King's
Banker (+1366) (London, 1988), pp. 171-72. E.B. Fryde does insist, however, that 'it would be an
exaggeration to put the blame for the failure to capture Tournai solely on a lack of money . . . But
Edward would have been in a much stronger position towards his allies if he had been able to pay
a large part of the huge subsidies promised to them' (p. 172).
On the financial crisis in England in the years prior to the siege see Harriss, pp. 270-93; E.B.
Fryde,
(footnote continued on next page)
Page 138
believe that the archbishop meant by lack of money to see me betrayed and killed.'5
And later to the town of Ghent he wrote:
Certain of our false councilors and ministers in England have behaved in such a manner
towards us that unless we do something about it we shall be unable to sustain the war this
season and so we think that, if we do not help ourselves, the aforementioned ministers will
quickly put our people to mischief or to disobedience towards US.6
On November 30, Edward returned to England. What followed is known to modern
historians as the 'Crisis of 1341.'7 Using the failure at Tournai, the English king
eliminated the facets of government hostile to his plans for war against France. This
purging of anti-war elements was successful, and by the end of 1341, Edward's
finances were under control, and he was again planning an attack on France.8
But, 'legally' he could not; at least according to the Treaty of Esplechin, signed a little
more than one year previously, Edward had promised not to attack France. His royal
counterpart, Philip VI, surely believed that the English would return, but he was
counting on the period of peace to prepare for this return. Then, in April 1341, what
would have otherwise been only a minor event, changed both kings' plans for
pursuing warfare against each other: John III, count of Brittany, died without a male
heir. This produced a crisis of inheritance. One claimant to the ducal throne was Joan
of Penthiève, daughter of John III's brother, Guy. A second claimant was John of
Montford, step-brother to John III. Naturally, France and England could not both
support the same claimant to the leadership of Brittany. Philip VI chose to back Joan,
as she was married to the nephew of the French king, Charles, count of Blois; England
supported John of Montfort.9 (Thus occurred the historical irony that had Edward III
Sustaining his candidate on Salic legal grounds that no noble throne could pass to a
woman, a legal principle which he had denied in his own attempt to gain the French
kingship. At the same time, Philip VI was disregarding that same Salic Law that had
given him the throne in supporting Joan.)
Initially, the war in Brittany remained limited. Edward III was too busy with his
problems at home, and Philip Vl chose to wait until John of Montford made the first
move before sending soldiers to the region. John responded by entering Nantes, the
(footnote continued from previous page)
William de la Pole, pp. 135-69; and E.B. Fryde, 'Financial Resources of Edward III in the
Netherlands, 1337-40', Revue Belge de philologie et d'histoire 45 (1967), 1142-1216.
5 As quoted in Nathalie B. Fryde, 'Edward III' pp. 153-4.
6 In Nathalie B. Fryde, 'Edward III,' p. 154.
7 Probably the best account of the Crisis of 1341 can be found in Roy Martin Haines, Archbishop
John Stratford: Political Revolutionary and Champion of the Liberties of the English Church,
ca. 1275/80-1348 (Toronto, 1986), pp. 278-328. See also Nathalie B. Fryde, 'Edward III,' pp.
149-61; W.R. Jones, 'Rex et Ministri: English Local Government and the Crisis of 1341', Journal
of British Studies 13 (1973), 1-20; and E. Deprez, Les préliminaires de la guerre de cent ans
(Paris, 1902), pp. 349-52.
8 See Michael Prestwich, 'English Armies in the Early Stages of the Hundred Years War: a
Scheme in 1341', Bulletin of the Institute of Historical Research 56 (1983), 102-13.
9 Alfred H. Burne, The Crecy War (London, 1955), pp. 66-67 and Henri Denifle, La guerre de
cent ans et la désolation des églises, monastères et hospitaux en France (Paris, 1899), p. 18.
Page 139
largest town in the duchy, and then by attacking Joan's allies in the surrounding
countryside. Charles of Blois, leading Joan's supporting force, including a French
army under the leadership of Philip VI's son, John, countered John's move by
attacking and capturing Nantes in November 1341. Montford was captured and
imprisoned in Paris where he was to die in 1345.10
Before Charles could relish his victory, John's wife, Joan of Flanders, appealed to
Edward III for assistance in her husband's claim. Edward responded to her request by
sending a small force to Brittany under the leadership of Sir Walter Mauny. But the
expedition was delayed by poor weather and did not arrive until May 1342, landing at
Hennebout where the countess was under siege. Mauny was immediately successful.
The siege at Hennebout was quickly relieved, and in the next few days, Mauny
relieved another siege at Auray and then defeated a small Bloisian force at
Quimperlé.11 By July, however, Charles was again victorious; the fortifications of
Auray, Vannes, and Guémenée-sur-Scorff fell quickly to him.12
Despite these latest setbacks, news of Mauny's victories had reached London, and
Edward decided to take advantage of what this small force had done by increasing his
numbers in Brittany to more than 3,000.13 The Hundred Years War continued.
Edward's new army was led by William of Bohun, earl of Northampton, and included
Robert of Artois and the earls of Derby and Oxford. All three were experienced
generals and had been active participants in the 1340 conflicts in the southern Low
Countries. This new force landed at Brest on August 18, 1342. Charles of Blois was
then besieging the town, but withdrew at the appearance of the English. Northampton
cleared the countryside, arriving outside the town of Morlaix on September 3. Morlaix
was in Charles' hands, and Northampton began a siege.14
In the meantime, Charles' force had been growing, perhaps to as many as 15,000.15 He
moved to relieve the siege, but outside of the village of Lanmeur, seven miles from
Morlaix, he encountered the English army, ordered in a line to receive his attack. The
battle of Morlaix was fought on September 30, 1342. Although greatly outnumbered,
perhaps as many as four to one, by the end of the day the English had sent the French
fleeing from the field. But Northampton, undoubtedly still concerned about his
comparatively small numbers, refused to pursue Blois' fleeing troops; he returned to
the siege of Morlaix, although the town never did fall to him.16
As inconclusive as it was, the battle of Morlaix is less important for the history of the
Hundred Years War than for the history of military tactics during the first half of
10 Burne, p. 67 and Denifle, pp. 18-19.
11 Burne, pp. 67-69 and Denifle, pp. 19-20.
12 Sumption, p. 398 and Denifle, pp. 20-21.
13 Burne (pp. 71-72) contends that the total English force numbered fewer than 4,000. Denifle (p.
22) tallies between 4,000-5,000. And Sumption (p. 398), on the other hand, claims a total of
5,500, delivered to Brittany on 440 ships.
14 Burne, pp. 69-71; Sumption, pp. 400-02; and Denifle, pp. 21-22.
15 This is Burne's figure, which is in fact much lower than earlier historians had tallied (p. 71).
Sumption calculates an even lower number3,000 cavalry, 1,500 Genoese mercenary soldiers and
sailors, and a 'motley force of Breton infantry' (pp. 401-02).
16 Modern accounts of the battle of Morlaix include: Burne, pp. 71-76; Sumption, p. 402; and
Tout, 'Tactics,' pp. 713-15.
Page 140
the fifteenth century. Despite Geoffrey le Baker's contention that the soldiers on both
sides fought better than their counterparts at Halidon Hill, Crécy, or Poitiers,17 very
few modern accounts of the battle exist, with only T.F. Tout and Colonel Alfred H.
Burne giving any more than a paragraph to the conflict in this century. This seems to
be little more than a reflection of the lack of original sources for the battle. There is
perhaps no battle fought between the forces of England and France during the
Hundred Years War that received less comment than did this one. While the English
chroniclers Geoffrey le Baker, Henry Knighton, and Adam Murimuth all give accounts
of varying lengths of this battle, there is only one short French account, that found in
Cuvelier's poetic Chronique de Bertrand Guesclin, written 1380-87.18 Part of this is
undoubtedly because there are no contemporary Breton narrative histories; but that
Jean le Bel, Jean Froissart, and the author of the Grandes Chroniques, all of whom
otherwise extensively chronicle the events of the early Hundred Years War, have
nothing to report on the battle indicates perhaps its lack of importance, even in the
relatively minor Brittany campaign.
But for the history of infantry warfare, especially between 1302 and 1347, the battle of
Morlaix is of great importance, even if just to explain why the vastly outnumbered
English infantry army was not only able to keep from being annihilated by their
French cavalry opponents, but also to cause them to flee in rout from the battlefield.
That the earl of Northampton did not pursue the fleeing French, nor later take the
town of Morlaix is of little consequence to such a study.

Both Henry Knighton and Adam Murimuth begin their narratives leading to the battle
with Walter Mauny's expedition: Knighton recounts that this was because Brittany 'was
oppressed by the French king,'19 while Murimuth writes that Mauny led this
expedition, which contained 40 knights and 200 archers, 'with the permission of the
king [Edward III].' The latter chronicler also mentions that Mauny was successful,
taking three castles and forcing a truce with Charles of Blois, a truce which did not
please Edward and was therefore invalid.20 In response, reports Adam Murimuth,
Edward sent more troops, 50 knights and 1000 archers, under the leadership of the
earl of Northampton and other leading nobles.21
17 Geoffrey le Baker, p. 76: 'Pugnatum est fortiter ex utraque parte, ita quod contigit in illo
certamine quod nec in bellis, nec in Halydoneheil nec de Cressi nec de Petter, audivimus
contigisse.'
18 Cuvelier, Chronique de Bertrand Guesclin, ed. E. Charrière (Paris, 1839), II:439-45.
Geoffrey le Baker and Henry Knighton definitely contain the most extensive account of this battle,
with Cuvelier and Adam Murimuth recording much smaller accounts. Robert of Avesbury says
nothing about the battle itself, but does comment on the number of English troops in Brittany.
19 Henry Knighton, p. 23.
20 Adam Murimuth, p. 125.
21 Adam Murimuth, p. 125. All English chroniclers award Northampton the supreme leadership of
this expedition. Adam Murimuth also mentions the count of Devonshire, the Baron Radulph of
Stafford, and William of Killesby in leadership roles (p. 125). Geoffrey le Baker lists Killesby,
Hugh Despenser, and Richard Talbot (p. 76). Henry Knighton adds Count Robert of Artois and
John D'Arcy to Devonshire and Killesby (whom he calls Gaddesby) (p. 25). And Robert of
Avesbury lists the earl of Warwick with Despenser (p. 342).
Page 141
Northampton, whose army necessitated transport by 14 large ships and many small
ones, according to Murimuth, landed at Brest, where the duchess of Brittany and her
children were under siege by land and sea by Blois and 'a large force from the French
realm.' At this the French army fled from their siegeworks, and Northampton captured
and burned most of the besieging fleet.22
Few details of Northampton's march to Morlaix are available in the contemporary
sources. Henry Knighton simply moves to the siege of Morlaix,23 with Adam
Murimuth reporting that the English 'ran through the country indiscreetly, for the
French could find no one who could resist them.'24 But Geoffrey le Baker, who is the
most descriptive chronicler of the events leading to the battle at Morlaix, disagrees
with this. He claims that the English 'having gone into Brittany, took land, and, inviting
the resistence of their enemies, they had many violent conflicts with them.' Later he
adds: 'And they conquered all of the country up to Morlaix, part by surrender and part
by destruction.'25
Morlaix was not a chance target, declares Adam Murimuth, but one that was chosen
after Northampton 'had the counsel of those who knew the country well,' and that he
paid these to lead him to the town. There, as he tried to assault the town, he did not
fare well; many were hurt, and one knight, Jacob Lovel, was killed. Thus
Northampton began a siege, expecting, Murimuth continues, 'aid from the allies of
England,' which after fifteen days arrived. The chronicler does not describe further
what this aid was.26
Also moving towards his position was the army of Charles of Blois. Blois' force was
quite large. Henry Knighton numbers his army at 20,000; Adam Murimuth counts
3,000 knights, 1,500 'Janissaries,' and a nonprofessional infantry (Murimuth calls them
populari) 'without number;' and Geoffrey le Baker refuses to speculate, writing only
that the French were 'a huge army.'27 They came to relieve the siege of Morlaix, and
rushed to the site when they realized how few besieging soldiers there were.28 But, as
Henry Knighton reports, the English 'in the complete darkness of night moved to meet
their enemy.'29 They also arrived early enough for Northampton to select his position
and to prepare the field where the battle would be fought. Knighton writes:
22 Adam Murimuth, p. 126.
23 Henry Knighton, p. 25.
24 Adam Murimuth, pp. 126-27: 'ac discurrentes per patriam indiscrete neminem qui resisteret
invenerunt.'
25 Geoffrey le Baker, p. 76: 'Britanniam itaque profecti, invitis inimicis resistentibus, terram
ceperunt, et multos asperos conflictus contra ipsos habuerunt . . . et totam patriam partim redditam
et partim destructam sibi submiserunt usque ad villam Morleys.'
26 Adam Murimuth, p. 127.
27 Henry Knighton, p. 25; Adam Murimuth, p. 127; and Geoffrey le Baker, p. 76. Another
manuscript of Adam Murimuth's Continuatio chronicarum (p. 227) mentions only 1,200
'Janissaries,' although the calculation of knights and infantry remain the same.
28 Henry Knighton, p.
29 Henry Knighton, p. 25: 'et tota obscura nocte datit se in occursum inimicorum suorum.'
Page 142

Battle of Morlaix
And in the morning they took their position about one league from the enemy in front of a wood,
and they dug pits and ditches around them and covered them over with hay and grass. And after
the sun rose, they prepared themselves for battle.30
Part of this preparation, according to Adam Murimuth, was to dismount the knights
and have all the English soldiers fight in one line as infantry.31
In response, Blois ordered his troops in three lines. The first, commanded by Geoffrey
Charnys (the only named commander besides Blois) and containing many knights,
charged into the English line but were, in the words of Henry Knighton, 'Immediately
conquered . . . and sent into flight.'32 The two other lines did not follow, but held their
ground. Blois held a quick tactical planning session. But as the French still greatly
outnumbered the English, it was decided to attempt another charge, although whether
this was to be made by only the second line, as Henry Knighton seems to indicate, or
by a combination of the second and third lines, as suggested by Geoffrey le Baker,
cannot be determined from the original sources.33
In fact, Henry Knighton insists, the French had no thought other than to attack again:
'Seeing the small number of English soldiers, they were taken by an aggressive
30 Henry Knighton, p. 25: 'Et mane ceperunt locum suum quasi per unam leucam ab inimicis
prope unum boscum, et foderunt foveas et fossas circa eos, et cooperuerunt eas de feno et
herbagio; et post solis ortum paraverunt se ad bellum.' See also Adam Murimuth, pp. 127, 227.
31 Adam Murimuth, pp. 127, 227.
32 Only Henry Knighton (p. 26) mentions the formation of Blois' army (although Geoffrey le
Baker's claim (p. 77) that Blois fought three times that day may be a reflection of the French
order). He is also the only chronicler to report Charnys' failure: 'Et inimici se paraverunt in tres
acies divisas. In quarum prima fuit dominus Galfridus Charnys cum multis galletis; qui statim victi
sunt ab Anglicis et missi in fugam' (p. 26).
33 Henry Knighton only mentions two charges of French troops (p. 25), while Geoffrey le Baker,
who is far less detailed in his account of the battle, claims that the French charged three times that
day (p. 77).
Page 143
spirit and undertook to overwhelm the English troops.' Furthermore, they were on
horses, as cavalry, while they faced only infantry: 'And wishing to trample the infantry
under the feet of their very powerful horses, they charged dangerously at them.'34
However, this attack did not succeed. The French were pushed into the narrow line of
English infantry, their charge was halted, and they were driven back into the pits and
ditches which surrounded them. Knighton again describes the scene: 'But having been
drawn into a narrow cave . . . they fell one on top of the other into the concealed pits
which the enemy had made.'35 Being unable to penetrate the English infantry line and
to defeat them, the French retreated from the field. Over fifty French knights had been
killed, with many more populari also slain, and Geoffrey Charnys and many others
were capturedCuvelier writes that these were 'old and young, great and small.'36 The
English too retreated, 'on account of a [continuing] fear of the large numbers of
enemy,' writes Henry Knighton.37
It is perhaps easy to dismiss the battle of Morlaix in a study of military history,
because of the lack of sources, and because those that do exist are English only. But
the similarity of the tactics at Morlaix and those at other battles fought in the beginning
of the fourteenth century demonstrates the validity of the narrative here. It also
indicates that the battlefield tactical practice of the English, generalship for lack of a
better word, was by 1342 known and exercised at every military engagement. As at
Boroughbridge, Dupplin Moor, and Halidon Hill, the main tactic of the English troops
was to form a solid line or mass of infantry, with even the knights dismounted, and
then to force their opponents to charge into it. In this instance, the French chose to
make cavalry charges, either two or three of them, and each charge met with failure as
it was unable to penetrate the English infantry line or to cause it to rout. Finally, after
suffering many losses, and undoubtedly disappointed at their failure to defeat a
smaller force, the French fled from the field. The English had won the battle.
However, there is one important difference between this victory and most of the ones
which the English won before and after it. While archers played an important role in
the battles of Dupplin Moor, Halidon Hill, and Crécy, they do not play the same role at
Morlaix. At these other conflicts the archers were positioned along the flanks of the
infantry line to urge their opponents into a disorderly charge, to further confuse that
charge, to narrow it onto the infantry line, and perhaps, although not entirely
necessary, to kill or wound those charging. Although archers seem to have been
present at Morlaix, as all chroniclers who mention the numbers in both Mauny's and
Northampton's armies indicate a force that included mostly archers,38 they were
34 Henry Knighton, p. 25: 'Tandem videntes Anglicorum paucitatem et suorum multitudinem,
ferocitate animi ducti, opprimere Angliae gentes moliti sunt; et equorum suorum validorum
pedibus conculcare volentes capitose irruerunt in eos.'
35 Henry Knighton, 11:25: 'sed antris decepti obturatis, ut praedictum est, ceciderunt quilibet
super alium in foveis abinvicem confusi.'
36 Cuvelier, p. 439. See also Henry Knighton, p. 26; Adam Murimuth, pp. 127, 227; and Geoffrey
le Baker, pp. 76-77.
37 Henry Knighton, p. 26: 'propter metum multitudinis inimicorum.'
38 See nos. 20 and 21 above.
Page 144
not positioned as at these other battles. Had Northampton done so, with his small
numbers of troops, he would have made his infantry line so thin that the French
charges would probably have been successful. So he included the archers among the
other troops in the line of infantry, using weapons other than their longbows.
But how could Northampton make up for the loss of flanking archers to narrow and
confuse the French charges? In this he resorted to another tactic which was being
practiced in battlefield conflicts of the early fourteenth century, although not
previously by the English. He arrived first at the battlefield and prepared it for conflict
on the next day. He placed his troops with their backs against a dense wood, thus
discouraging their flight, and he dug ditches and pits along the flanks of the infantry
line and across the battlefield. As the Scots, Flemings, and Swissall without the benefit
of skilled longbowmenhad found, this tactic was able to narrow and confuse the
charge of their opponents so that when they met the infantry lines, they were easily
defeated. And so it happened also at Morlaix. Henry Knighton described it nicely; the
French were drawn into an 'antrum obturatum,' a 'narrow cave,'39 a cave in which
they met defeat.
39 See no. 35 above.
Page 145

XII
The Battles of Staveren, 1345, and Vottem, 1346
The success of the Flemings in 1302-03 and 1323-27 elicited fear among French
nobles that the spirit of rebellion would spread elsewhere in the realm, particularly
Bordeaux, Picardy, and Gascony. But with the exception of a small rebellion which
broke out in Bordeaux shortly after the battle of Courtrai, France remained relatively
calm.1 The Low Countries did not, however. In the 1330s and 1340s the Low
Countries were awash in popular rebellion. In Flanders from 1337 to 1345 a rebellion
was led by the Ghentenaar Jacob van Artevelde which succeeded in supplanting Louis
of Nevers's comital leadership. In 1344-45 a rebellion was fought by Frisians against
their lord, William, count of Hainault, Holland, and Zeeland. And from 1345 to 1347
the town of Liège led neighboring towns, Huy, Dinant, Bouvignes, and Saint-Trond, in
a revolt against the Prince-Bishop of Liège, Englebert de la Marck. It is these latter two
rebellions which concern us here. For although the Flemings involved themselves in
conflicts against the French at Sluys and Tournai in 1340 and would again be found
facing the French at Calais in 1346, these engagements were not battles as studied in
this work.2 On the other hand, both the Frisians and the Liègeois faced armies on the
battlefield which attempted to put down their rebellions. And in both cases, at
Staveren in 1345 and at Vottem in 1346, the rebellious armies, infantry forces,
defeated their cavalry-based opponents.
Of all the late medieval Low Countries, Friesland was the most sparsely populated and
economically backward. The Frisians even by the early fourteenth century had never
really recovered from their harsh conquest by Charlemagne. They were certainly a
proud and independent people, but as they were generally quite far from and of little
importance to any 'foreign' governmental leaders, Frisian inhabitants were usually left
alone without an outlet for their spirit of independence. Indeed, it was not until the
beginning of the fourteenth century, with the recognition of and struggle over the rest
of the northern Low Countries by France and the Holy Roman Empire, that Friesland
also began to seek associations with more powerful European
1 See Chronographia regum Francorum, I:119; the first and second continuators of Guillaume
de Nangis's Chronicon, I:324, 334; and Chronicon comitum Flandriae, p. 203.
2 On the Jacob van Artevelde rebellion see David Nicholas, Medieval Flanders, pp. 217-26;
Henri Pirenne, Histoire de Belgique, II:93-123; Henry Stephen Lucas, pp. 240-528; Hans van
Werveke, Jacques van Artevelde (Brussels, 1943); and David Nicholas, The van Arteveldes of
Ghent: The Varieties of Vendetta and the Hero in History (Ithaca, 1988), pp. 1-71.
Page 146
entities.3 In 1337 an alliance was agreed to with Philip VI of France, but it brought
negligible results. The French king was simply too far distant from this very northern
province. Therefore, in 1338, the Frisians turned to a closer ally, placing themselves
under the leadership of William III, the governor of the nearby counties of Holland
and Zeeland as well as the more southern county of Hainault.4
Why the Frisians did this, and in fact who among them had the right to do this, is very
much a mystery of medieval history. What is known is that before too long the
Frisians began to doubt the wisdom of their choice of governmental leaders. There
were some economic benefits, as the connection with Holland and Zeeland increased
trade and other commercial contacts, and, at least initially, William left the Frisians
alone.5 But in 1342-44, the count began to rule Friesland with more stringency and
immediacy. He became concerned by Hanseatic economic interest in his northern
lands, by Frisian ecclesiastical contests, and by what he perceived as a lack of service
and loyalty of his subjects there. Against this the Frisians rebelled, and on several
occasions William was required to use force and intimidation to pacify them.6
By 1345 the situation had deteriorated to such an extent that William decided to mount
a major expedition against Friesland. After preparing for this campaign in Amsterdam
for more than six months and after besieging Utrecht for more than six weeks (Utrecht
was also rebelling against his rule), on September 26 he crossed over the Zuider Zee
from Enkhuisen to Staveren, the 'capital' of Friesland. There the Frisian rebels stood
ready for battle, a battle which they would win, and which would even cause the
death of the count.7
In most ways the prince-bishopric of Liège was very different from Friesland. Neither
as populous or as economically prosperous as Flanders, Brabant, or Hainault, it was
still a powerful entity in the southern Low Countries, holding an important border-
position between the French concerns of the region and those of the Holy Roman
Empire. Long an ecclesiastical capital, Liège had been governed by a prince-bishop
since as early as the tenth century when Bishop Notger was appointed to the position
by Emperor Otto II.8 Since that time the prince-bishops had remained in control of
the province by retaining a strong and loyal alliance with the German emperors. Thus,
despite a few occasions when the Liégeois were forced to defend their lands against
the Brabantese, all of which were successful, the prince-bishopric almost always was
at peace with external powers.9
But internal peace, especially during the thirteenth, fourteenth, and fifteenth
3 See J.F. Niermeyer, 'Het Sticht Utrecht, Gelre en de Friese landen in de veertiende eeuw,' in
Algemene geschiedenis der Nederlanden, vol. 3: De late middeleeuwen, 1305-1477 (Utrecht,
1951), pp. 155-60.
4 Lucas, p. 505 and Niermeyer, pp. 156-57.
5 Lucas, p. 506,
6 See examples given in Lucas, pp. 506-08. On Frisian desires for independence see Niermeyer,
pp. 157-59.
7 Lucas, p. 508 and Jean van Malderghem, La bataille de Staveren (26 septembre 1345)
(Brussels, 1869).
8 Pirenne, Histoire de Belgique, I, 177-78, II:19 and Godefroid Kurth, La cité de Liège au
moyen-age (Brussels, 1909), pp. 21-52.
9 Pirenne, Histoire de Belgique, I:207-11; Kurth, I:110-28; and F. Vercauteren, 'Het prinsbisdom
(footnote continued on next page)
Page 147
centuries, was almost non-existent. In fact, there may have been no other region in
Europe during this period which suffered more popular rebellions, including
Flanders. For example, between 1298 and 1347 the major towns of the province were
almost continually in rebellion against the prince-bishop.10
The specific rebellion which would lead to the battle of Vottem originated at the death
of the Prince-Bishop Adolph de la Marck. Adolph de la Marck had not been a favorite
of his subjects. His heavy-handed rule was opposed by almost everyone in the
province, and he frequently answered this opposition with military force. Examples
are numerous: in 1314 he violently put down a rebellion in Saint-Trond, in 1318 in
Fosse, also in 1318 in Huy, in 1321 in Ciney, and in 1323 in Tongeren.11 Ultimately, in
1328 the towns of Liège, Huy, Tongeren, and Saint-Trond decided to oppose the
prince-bishop; forming a confederation and mustering an army, the towns fought three
battles against Adolph: Erbonne (May 27), Waremme (June 2), and Hoesselt
(September 25).12 Although these battles were by and large inconclusive, the idea of
armed rebels opposing his reign frightened the prince-bishop into some constitutional
concessions. Reforms of the municipal regime were undertaken in 1330 and 1331.
Known either as the Réformation d'Adolphe or the loi de murmure, the Liégeois were
to submit to the authority of the prince-bishop, but in turn they were to see a lessening
of local power'gouverneurs' were replaced by 'wardeurs'and an increase in the
autonomy of the guilds.13
For twelve years there was an extremely tenuous peace between the Liègeois and their
prince-bishop. The townspeople took as many privileges as they legitimately could
from the new constitutional reforms, and sometimes they took more. Eventually, in
1343, Adolph de la Marck repealed the loi de murmure and attempted to return to pre-
1330 governance. This might have provoked a new rebellion against the prince-
bishop, but before the confederation of towns could again come together, on
November 3, 1344, Adolph died.14 A solution to the problem might have presented
itself with the selection of a good new leader, but when the towns found on February
23, 1345 that their new prince-bishop was to be Englebert de la Marck, the nephew of
the object of their previous hatred, they immediately gathered an army and prepared
for war.15 Englebert also knew of the feelings of the Liégeois, and he too
(footnote continued from previous page)
Luik tot 1316,' in Algemene geschiedenis der Nederlanden, vol. 2: De volle middeleeuwen,
925-1305 (Utrecht, 1950), pp. 340-41.
10 Kurth, I:129-51, 179-215, 243-60, II:1-88; Claude Gaier, Art et organisation militaires dans
la principauté de Liège et dans le comté de Looz au Moyen Age (Brussels, 1968), pp. 262-306;
and Fernand Vercauteren, Luttes sociales à Liège, xiiie et xive siècles (Brussels, 1946).
11 Pirenne, Histoire de Belgique, II:16-17, 35-37; Kurth, II:1-17; Vercauteren, Luttes sociales,
pp. 81-85; Vercauteren, 'Het prinsbisdom Luik,' pp. 350-52; and J. Lejeune, 'Het prinsbisdom Luik
tot 1390,' in Algemene geschiedenis der Nederlanden, vol. 3: De late middeleewen, 1305-1477
(Utrecht, 1951), pp. 177-78.
12 Lejeune, pp. 179-80; Kurth, II:31-45; and Vercauteren, Luttes sociales, pp. 88-89. The best
description of these battles is in Gaier, pp. 276-85. I have chosen not to discuss them because of
the difficulty in discerning battlefield tactics from the original sources.
13 Pirenne, Histoire de Belgique, II:42-43; Lejeune, p. 180; Kurth, II:38-60; and Vercauteren,
Luttes sociales, pp. 88-91.
14 Lejeune, pp. 181-82 and Kurth, II:63-69.
15 Lejeune, pp. 182-83; Kurth, II:73-74; and Vercauteren, Luttes sociales, p. 94.
Page 148
gathered a force, an army composed of some of the greatest knights in the Holy
Roman Empire. The mustering of these two forces took time, and it was not until July
18, 1346 that they faced each other on the battlefield in a northern suburb of Liège
known as Vottem. As at Staveren, the rebels prevailed.16

Because of the death of Count William, the battle of Staveren is mentioned in several
chronicles, among them Jan Beke's Chronographia, Jean Froissart's Chroniques, the
continuation of Richard Lescot's Chronique, the Grandes chroniques de France, and
the Récits d'un bourgeois de Valenciennes. But it is only in the Récits and Froissart's
third redaction that a discussion of tactics on the battlefield can be found.17
In 1345 William was in his northern lands for several reasons but primarily to reassert
his control over the ecclesiastical capital of the north, Utrecht, and to put down the
insurrection in Friesland. Both, of course, required military force. Neither appeared to
be an easy task, but the count refused to be apprehensive, and he pressed forward
without delay, planning first to attack Utrecht and then Friesland. Utrecht proved to be
a difficult target, but after frequently assaulting the town and bombarding its walls
with 'many engines' for six weeks, it fell.18
Following this victory the count marched his army to Dordrecht, placed his troops on
ships, and sailed north to Friesland. The size of this force cannot be known from the
original sources. The Récits d'un bourgeois de Valenciennes does claim a total of
10,000 men-at-arms and 2,000 'haubergons' at the siege of Utrecht, but it also insists
that many stayed there under the command of Jean of Hainault, the count's uncle, and
did not go to on Friesland.19 Instead, the Récits and the other original sources all
indicate that William's army was quite small in comparison to the Frisian army which
was encountered at Staveren. Jean Froissart, for one, claims that the Frisians
outnumbered their foes by a twenty to one ratio.20 It was, however, a force composed
of the 'most noble chivalry of Hainault, Holland, Flanders, Brabant, Guelders, Jülich,
Namur, and Hesbaye' plus a few others who had either joined him 'on loan' from the
king of France or the German emperor. Included was the duke of Guelders, the count
of Namur, the count of Salm, the count of Eppenheim, and 35 other knights noble
enough to carry their own banners.21
The count's ships landed outside of Staveren, and on the morning of September
16 Lejeune, pp. 182-83.
17 Jan Beke, pp.301-03; Jean Froissart, IV:324-28 (the third redaction is pp. 326-28); Richard
Lescot, p. 68-69; Grandes chroniques, IX:257; and Récits d'un bourgeois de Valenciennes, pp.
202-05. There is also an anonymous poem on the death of William, but it also contains no details
about the battle: 'La mort du conte de Henau,' in Panégyriques des comtes de Hainaut et de
Hollande Guillaume I et Guillaume II, ed. C. Potvin (Mons, 1863).
18 The most detailed account of the siege of Utrecht is in Récits d'un bourgeois de Valenciennes,
pp. 196-97, 201-02. See also Jan Beke, p. 301, and Jean Froissart, IV:324-26.
19Récits d'un bourgeois de Valenciennes, p. 197.
20 Jean Froissart, IV:327. See also Jan Beke, p. 301 and Récits d'un bourgeois de Valenciennes,
p. 203.
21 The quotation comes from Jean Froissart, IV:325. See also Jean Froissart, IV:326; Récits d'un
bourgeois de Valenciennes, p. 197; and Richard Lescot, pp. 68-69. Some of these may have
stayed at Utrecht, as the original sources do not mention any of these at Staveren.
Page 149

Battle of Staveren
26 the soldiers heard mass and disembarked.22 William seemed to have known of the
presence at Staveren of his Frisian opponents. He may also have known that they
outnumbered him. But he did not fear them. Pride, present with almost all fourteenth-
century noble forces when facing 'lesser' opponents, also seems to have been common
among these knights. Jean Froissart asserts that William 'held little admiration for and
did not prize the Frisian power' at Staveren.23 And the Récits d'un bourgeois de
Valenciennes writes that William himself led his troops into battle, shouting 'follow me
for the honor of God and Saint George.'24
On the other side, the Frisians had prepared well for the count's attack. Jean Froissart
writes that they had fortified their position, outside of the abbey of Floricamp, while
the Récits d'un bourgeois de Valenciennes maintains that these fortifications were
ditches and embankments.25 But in seeing the charge of the count's troops the Frisians
abandoned their fortified positions and ordered a defensive infantry line in front of
them. The reason for this was simple, claims the Récits, for the Frisians could see that
the oncoming attackers were having difficulty with
22Récits d'un bourgeois de Valenciennes, p. 202.
23 Jean Froissart, IV:327: 'Chils contes qui fu de grande volonté, hardis et entreprendans oultre
mesure, et pour lors en la flour de sa jonèce, et qui petit amiroit et prisoit la poissance des Frisons
contre la sienne.'
24Récits d'un bourgeois de Valenciennes, p. 203: 'Sy me sieve en l'oneur de Dieu et de
monseigneur saint George!' See also Jean Froissart, IV:325, 327.
25 Jean Froissart, IV:326-27 and Récits d'un bourgeois de Valenciennes, p. 203.
Page 150
the dike-filled terrain and were approaching them so quickly that they had lost their
ordered composition; they came 'without array, without any order.'26
William's soldiers thus charged into this solid infantry formation in disarray, and their
attack met with fierce resistance. The Frisians fought well with their 'very large and
heavy pikes, their long swords and axes and falchions,' killing all they faced 'without
pity and without mercy.'27 Although the count's troops fought well and also killed
many, their lack of numbers eventually proved to be their downfall.28 Only a very
small number were able to flee to the safety of the ships. Most were left dead on the
battlefield.29
A short time after the battle, and evidently having heard of his nephew's defeat, Jean
of Hainault landed another party at the battlefield site. But this group did nothing but
collect the bodies of their dead and sail south. The Frisians had the victory.30

The battle of Vottem generated far more contemporary commentary than did the battle
of Staveren. Lengthy accounts of what occurred can be found in the chronicles of
Jean le Bel, Jean d'Outremeuse, and Mathias von Neuenberg, and in the Récits d'un
bourgeois de Valenciennes and the Chronique de l'abbaye de Saint-Trond. Shorter
accounts are found in the chronicles of Jean de Winterthur, Jean de Hocsem, Heinrich
von Diessenhoven, and Heinrich von Rebdorf, the Le miroir des nobles de Hesbaye
of Jacques de Hemricourt, and the Vita Clementis VI of Werner de Bonn.31
An account of the battle of Vottem begins with a discussion on the location of the
battlefield. Vottem is a suburb of the town of Liège; in 1346 it was situated just outside
the walls of the town, and was selected by the rebels as the place for battle because it
lay on the main road into Liège from Tongeren, where Archbishop Englebert de la
Marck had set up his court during the rebellion. Thus it was the route which would
have to be taken by the archbishop and his troops if they most easily wished to enter
Liège.32 According to the original sources, this was done by the rebels for two
reasons: first, they knew that news of the town's insurrection had been
26Récits d'un bourgeois de Valenciennes, p. 203: 'Et quant les Frisons virent qu'il y en eult
assés d'issus et qu'ils furent oultre les dicques et qu'ils venoient sy hastivement à grans flottes
et sans aroy, ne nul conroy, les Frisons yssirent hors de Stavres et de l'abaye de Flouricamp et
partout de leurs embucquemens.'
27 Quotations are from Récits d'un bourgeois de Valenciennes, p. 203. See also Jean Froissart,
IV:327.
28 Jean Froissart, IV:327 and Récits d'un bourgeois de Valenciennes, p. 203.
29 Jean Froissart, IV:327 and Récits d'un bourgeois de Valenciennes, pp. 203-04. The Récits
includes a list of the noble dead.
30Récits d'un bourgeois de Valenciennes, pp. 204-05 and Jean Froissart, IV:326, 327. The Récits
claims that there was a small battle fought between Jean and the Frisians but that it was
inconclusive.
31 Jean le Bel, II:140-41; Jean d'Outremeuse, Chronique abrégée, de 1341 à 1400, in Chroniques
Liègeoises, ii, ed. S. Balau and E. Fairon (Brussels, 1931), pp. 165-70; Mathias von Neuenberg,
Chronik, ed. A. Hofmeister, MGH nova ser., iv (Berlin, 1940), pp. 202-03; Récits d'un bourgeois
de Valenciennes, pp. 210-13; Chronique de l'abbaye de Saint-Trond, II:282-83; Jean de
Winterthur, pp. 263-64; Jean de Hocsem, pp. 341-45; Heinrich von Diessenhoven, Chronicon ab
a. 1316-1361, in Fontes rerum Germanicarum, iv, ed. J.F. Boehmer (Stuttgart, 1868), pp. 51-52;
Heinrich von Rebdorf, Annales imperatorum et paparum (1294-1362), in Fontes rerum
Germanicarum, iv, ed. J.F. Boehmer (Stuttgart, 1868), pp. 528-29; Jacques de Hemricourt, Le
miroir des nobles de Hesbaye in OEuvres
(footnote continued on next page)
Page 151

Battle of Vottem
received by the archbishop, that he had gathered an army, and that he was on his way
to the town by the quickest route possible. In the words of Jean de Hocsem: 'the
archbishop intended to punish the townspeople of Liège.'33 Second, the proximity of
the battlefield to the town walls boosted the morale of the rebels, as remarked on by
Jean de Winterthur, for they knew that if the archbishop got past them, there would be
little to prevent him from entering and punishing the town.34
It is difficult to know how large the rebel force was at Vottem. Only Mathias von
Neuenberg mentions a number for the soldiers, 40,000.35 What can be determined is
that these troops came also from outside of Liège, from Huy, Dinant, and
Bouvignes,36 that they contained very few knightsonly Godefroid de Harduemont,
Louis d'Agimont, Jean de Borgneval, Walthier de Hautepenne, and Arnoul de
(footnote continued from previous page)
de Jacques de Hemricourt, i, ed. C. de Borman (Brussels, 1910), p. 161; and Werner de Bonn,
Vita Clementis VI, in Vitaepaparum Avenionensium, i, ed. S. Baluze-Mollat (Paris, 1914), p.
546.
32 On the establishment of the archbishop's court in Tongeren see Jean d'Outremeuse, p. 165.
33 Jean de Hocsem, p. 341: ' . . . episcopus intendens se de Leodio vindicare . . . ' See also Jean
d'Outremeuse, pp. 165-66; Jean le Bel, II:140; Jean de Winterthur, p. 263; and Mathias von
Neuenberg, p. 202.
34 Jean de Winterthur, p. 263. See also Jean d'Outremeuse, pp. 165-66 and Werner de Bonn, p.
546.
35 Mathias von Neuenberg, p. 203.
36 Jean d'Outremeuse (p. 168) mentions troops from Huy, while the Récits d'un bourgeois de
Valenciennes (p. 210) adds Dinant and Bouvignes to Huy as providers of troops to the Liégeois
force.
Page 152
Hautepenne, lord of Villereau and Boilhe and mayor of Liège, are named by the
original sources37and that they greatly outnumbered their approaching opponents.
Mathias von Neuenberg expresses the latter by numbering Englebert de la Marck's
force at 9,000, more than four times smaller than his count for the Liégeois and their
allies.38 Jean le Bel is even more exaggerated in his comparison of the two forces,
claiming that the archbishop's army was only one-twentieth the size of the rebels'.39 It
is also known that the approaching army was composed of the 'lords of Germany',
including Adolph II, count of La Marck and brother to the archbishop; Adolph VIII,
count of Berg; Wilhelm I, count of Namur; John the Blind, king of Bohemia, count of
Luxembourg, and father to the emperor; Dietrich of Heinsberg, count of Looz;
Wilhelm V, marquis of Jülich; Reinhold III, duke of Guelders; and Dietrich IV, lord of
Falkenburg and burgrave of Zeeland.40 Infantry troops were also included in the
archbishop's force.41
The rebel army seems to have learned of the archbishop's approach sometime around
July 16, as Jean de Hocsem insists that they set up their position at Vottem two days
before the battle took place.42 They also seem to have fortified their position by
digging large ditches, not ones like those at Courtrai, made to disrupt a cavalry charge,
but like those at Staveren, made to hide in. According to the Récits d'un bourgeois de
Valenciennes, the ditches were also dug to hinder entry into the town if the
archbishop's force was to defeat the rebel army.43
On June 18 Englebert de la Marck's army encountered the rebels. Most of these were
in the ditches, 'standing well armed and tightly bunched to resist the lords,' writes the
anonymous author of the Chronique de l'abbaye de Saint-Trond.44 The archbishop
ordered his cavalry in three lines and began to charge the Liègeois positions.45 'The
bishop and his army proceeded incautiously,' writes Werner de Bonn simply,46 while
the Chronique de l'abbaye de Saint-Trond adds a recognition of the pride which
carried the horsemen to their task: 'They did not doubt that they would obtain victory,
since behold . . . they charged with impetus into the Liégeois.'47 But in all their
enthusiasm, the charge of the cavalry became disordered and
37 Jean d'Outremeuse, pp. 168-69. Other knights are made on the battlefield. For their names
see Jean d'Outremeuse, pp. 169-70.
38 Mathias von Neuenberg, p. 203.
39 Jean le Bel, II:140.
40 Jean le Bel (II:140) uses the term 'lords of Germany', while Jean d'Outremeuse (p. 168),
Mathias von Neuenberg (pp. 202-03), and Récits d'un bourgeois de Valenciennes (p. 210) name
some or all of them.
41 See Jean le Bel, II:141.
42 Jean de Hocsem, p. 342.
43Récits d'un bourgeois de Valenciennes, p. 211; Jean le Bel, II:140; Jean d'Outremeuse, p. 168;
and Chronique de l'abbaye de Saint-Trond, II:282.
44Chronique de l'abbaye de Saint-Trond, II:282.
45 Jean d'Outremeuse, p. 168.
46 Werner de Bonn, p. 546: 'seal episcopo cum suis incaute procedentibus . . . '
47Chronique de l'abbaye de Saint-Trond, II:282: 'non dubitabant de optinendo triumphum, cum,
ecce . . . cum impetu in Leodienses irruit.'
Page 153
confused.48 And seeing this, the rebels came out of their ditches, ordered themselves
in a solid defensive line, and faced the charging cavalry.49 As the Récits d'un
bourgeois de Valenciennes reports:
It was a great battle, perilous, cruel, and very felonious. The knights were strong and armed
with good weapons of iron and steel. So they fought well and shrewdly. And the Liègeois
defended themselves with axes and hammers of iron, of lead, of bronze, and of steel, and they
killed and knocked down the horses and the knights without pity and without offering ransom.50
Defeated, the archbishop's army fled from the field, those on horseback rushing
through and past their infantry. The rebels, inspired by the victory, broke from their
line and followed the fleeing troops, killing all who hesitated in their flight or
straggled behind.51 Included among these was Dietrich of Falkenburg, whose initial
refusal to flee allowed the pursuing Liégeois to pull him from his horse and kill him.
His name was added to the fifty-seven `knights' and more than three hundred other
soldiers who also met their end at Vottem.52
While entirely different in origin, goals, and participants, the battles with which these
two rebellions culminated, and which brought victory to both rebel armies, have many
similarities. There is of course the aspect of infantry-based armies facing cavalry-
based ones, which is prevalent so often in the early fourteenth century. But these
battles differ in the fact that the infantries which won both at Staveren and Vottem
greatly outnumbered their cavalry opponents. Could this mean that these victories
were acquired then simply because the larger armies overwhelmed the smaller ones?
Not if we are to accept what seems to be a fear of these smaller armies by the larger
ones. In both battles the rebels chose to construct field fortifications by digging
ditches large enough to give them some protection against a cavalry charge. This
certainly seems to indicate that they at least feared their smaller numbered foes and
refused to trust in the idea that larger numbers alone would bring them victory.
At the same time, it should also be noted that the smaller numbered cavalry-based
armies did not fear their larger infantry counterparts. At both the battle of Staveren
and the battle of Vottem the noble cavalries charged without hesitation on the rebel
48 See Jean le Bel, II:140; Mathias von Neuenberg, p. 203; and Récits d'un bourgeois de
Valenciennes, p. 211.
49 On the Liégeois leaving their ditches see Jean d'Outremeuse, p. 168. On their line see the
Récits d'un bourgeois de Valenciennes, p. 211 and the Chronique de l'abbaye de Saint-Trond,
II:282.
50Récits d'un bourgeois de Valenciennes, p. 212: `Et là y eult grande bataille périlleuse, cruese
et moult félonneuse; car les chevaliers estoient hardis et armés de bonnes armures de fer et
d'achier. Sy se combatirent bien et asprement. Et Liégois se deffendoient de haches et de marteaux
de fer, de plonc, de laiton, et d'achier, et en tuoient et assommoient ces chevaulx et ces chevalliers
sans nulle pité et sans point de renchon.'
51 Jean le Bel, II:140; Mathias von Neuenberg, p. 203; and Chronique de l'abbaye de Saint-
Trond, II:282-83.
52 Jean d'Outremeuse, pp. 169-70; Récits d'un bourgeois de Valenciennes, pp. 211-23; Jean de
Hocsem, pp. 342-43; Chronique de l'abbaye de Saint-Trond, II:282-83; Heinrich von
Diessenhoven, p. 52; Heinrich von Rebdorf, pp. 528-29; and Jacques de Hemricourt, p. 161.
Page 154
infantry positions and without regard for the fortifications in which the rebels hid
themselves. In both battles, as with so many other early fourteenth-century examples,
the pride of the cavalry forces induced them to ignore the size and defenses of their
opponents and provoked them to make unwise and eventually confused charges.
Once again this leads to a similarity between these two Low Countries' battles. For
seeing the cavalry charges become confused and disordered, both the Frisian and the
Liégeois rebel armies left the safety of their field fortifications, formed solid defensive
lines, and faced the attacks of the horsemen bearing down on them. Victory was
achieved when these charges failed to break the infantry formations and the cavalry
soldiers were forced to face more numerous opponents fighting with skill and
motivation. The infantry weapons, mentioned prominently in both battles, quickly put
an end to the `more noble' cavalry forces which they faced, and giving no quarter to
their opponents, the rebels in both battles killed many.
Page 155

XIII
The Battle of Crécy, 1346
The battle of Morlaix solved little in the 1342 conflict over Brittany. By the time
Edward III arrived at Brest on October 27, with an army numbering several
thousand,1 he found that the province was far from pacified. In response, the English
king conducted a chevauchée through the countryside, pillaging and devastating all
the land up to the town of Vannes. There, on November 25, he joined his army to that
of his renowned lieutenant, Robert of Artois, who had been besieging the important
Breton town for nearly a month. (Artois would be fatally wounded outside Vannes
before the end of the year.) Within days, English armies also attacked Nantes, Rennes,
and Dinan.2
A large French army, eventually led by the king, Philip VI, was raised and
immediately sent to the war-ridden province. The English army came together in
anticipation of a battle. But Philip refused to be drawn into combat, holding his army
away from the English for two weeks until papal legates arrived from Pope Clement
VI to broker a truce, the Truce of Malestroit. It would last for three years. On
February 22, 1343, Edward and his army returned home. The English would not
return again to the north of France until 1345.3
At the end of the truce, Edward III was again ready to make a claim on his `rightful'
throne of France. English armies attacked French outposts in Brittany and Gascony,
fighting there with much success during the next two years.4But neither English army
had the success of Edward III's main force which sailed to France under the
leadership of the king in spring 1346.
It is now apparent, although it may not have been at the time, that the 1345-46 English
campaigns in Brittany and Gascony were not to be the main thrusts of Edward's
military strategy in France. The armies in those two provinces were there
1 Adam Murimuth claims that Edward's army included 6,000 men-at-arms and 12,000 archers
which were transported on more than 400 ships (p. 127).
2 Burne, pp. 80-84 and Sumption, pp. 403-07. Nantes was beset by the earl of Norfolk, Rennes by
the earl of Northampton, and Dinan by the earl of Warwick.
3 On the signing of the truce see Burne, pp. 84-86 and Sumption, pp. 407-08. On the period of
1343-45 when the truce was in effect see Sumption, pp. 411-54. The truce kept both sides at status
quo, with the exception that Vannes became `neutral' for the duration of the truce.
4 On the breaking of the truce see Sumption, pp. 447-54 and Jules Viard, `La campagne de juillet-
août 1346 et la bataille de Crécy,' Le moyen âge, 2nd ser. 27 (1926), 1. On the war in Brittany
before Crécy see Burne, pp. 86-89; Sumption, pp. 471-73, 493-97; and Perroy, pp. 115-16. On
Gascony before Crécy see Burne, pp. 100-21; Sumption, pp. 455-59, 463-71, 473-88; Viard,
'Crécy,' pp. 1-3; and Henri de Wailly, Crécy 1346: Anatomy of a Battle (Poole, 1987), p. 11.
Page 156
only as precursors for a larger force which was to land at Normandy, fighting to
reclaim that part of France which had once been held by the English crown. Alfred
Burne goes one step further. He claims that it was Edward III's plan to bring all three
of his armies in France together, `advancing on three radii, as it were, all directed on
Paris, the centre.'5 However, this may be too speculative on the part of Colonel Burne,
for it does not take into account the very large French army which would probably
not be inclined to wait around Paris until all the `radii' of the English plan came to it.
Why should the French king not take on the smaller Gascon and/or Breton armies
before they could join up with the main English force? This very important question
must remain unanswered, however, for original sources do not explain either why the
king of England pursued his three-pronged strategy when conventional military
wisdom spoke against dividing one's force, or why the king of France chose not to
attack the smaller English armies, but instead to do battle with the main force at Crécy-
en-Ponthieu.
On April 6, 1346, Edward III ordered a large force to gather at Portsmouth for
transportation to the continent. Their destination was not known at the time, with most
expecting transit to Gascony or Brittany. The number of troops was large, perhaps as
many as 15,000, most of whom were infantry or archers; it is estimated that between
700 and 1,000 ships were needed to transport the force to France.6 Prepared to set sail
immediately, the Channel's weather was foul enough to keep the army in Portsmouth
until July 11. On the next day they landed at St. Vaast la Hogue, on the coast of
western Normandy.7
After a short break for reorganization, the English army proceeded quickly along the
coast of begin what some historians describe as a raid or chevauchée and what others
call a military campaign.8 On July 18 it captured Valognes, on July 20 St. Côme-du-
Mont, on July 22 Saint-Lô, on July 23 Torigny and Sept-Vents, on July 24
5 Burne, p. 136.
6 The numbers of English soldiers during the Normandy campaign vary between 7,000-10,000
(Sumption, p. 497) and 30,000 (Viard, `Crécy,' p. 8). Other tallies include 9,000 (Lot, I:344-47);
under 10,000 (Richard Barber, Edward, Prince of Wales and Aquitaine: A Biography of the
Black Prince (London, 1978), p. 48); 11,000 (Wailly, p. 11); 14,000-15,000 (Andrew Ayton, `The
English Army and the Normandy Campaign of 1346,' in England and Normandy in the Middle
Ages, ed. D. Bates and A. Curry (London, 1994), pp. 253-68); 15,000 (Burne, pp. 137-38 and
Christopher Allmand, The Hundred Years War: England and France at War, c.1300-c.1450
(Cambridge, 1988), p. 16); and 19,428 (George Wrottesley, Crecy and Calais from the Original
Records in the Public Record Office (London, 1898), pp. 9-10). For a discussion of English
numbers see Burne, pp. 166-68 and Ayton, pp. 253-68. Burne totals the English ships at 700 (pp.
137-38), with Viard counting 1,000 (`Crécy,' p. 8).
7 Burne, pp. 138-40; Sumption, pp. 500-02; Viard, `Crécy,' pp. 3-11; Wailly, p. 18; Barber, pp.
47-48; and Perroy, pp. 118-19.
8 Allmand (p. 15), Wailly (pp. 11-49), and, most recently, Clifford J. Rogers ('Edward III and the
Dialectics of Strategy, 1327-1360,' Transactions of the Royal Historical Society 6th ser. 4
(1994), 88-102) describe Edward's movements across Normandy as a raid, while Burne (pp. 100-
35), Sumption (pp. 471-525), and Viard (`Crécy,' pp. 1-67) all see it as a campaign. There are
numerous contemporary letters written by English participants of the campaign. A campaign diary
also exists, although only in fragmentary form. These are all most accessibly found in Richard
Barber, ed. and trans., The Life and Campaigns of the Black Prince (London, 1979), pp. 14-40.
For a commentary on these letters see Kenneth A. Fowler, `News from the Front: Letters and
Despatches of the Fourteenth
(footnote continued on next page)
Page 157
Torteval, and on July 25 Fontenay-le-Pesnel. Nowhere did they meet much
opposition.9 But at Caen, which was reached on July 26, a French garrison of between
300 and 1500 stood ready to defend their town. The castle of Caen had been built
largely by William the Conqueror and his son, Henry I. It was strong and would
necessitate siege machines for its fall, machines which Edward III did not have with
him. He chose instead to try to lure the castle's garrison from its protective walls by
attacking the town it defended. The attack, by both the land and sea, and in sight of
the castle, was cruel, but the cries of the beset townspeople finally caused the
garrison's surrender; more than 2,500 were massacred.10
The speed of march was again impressive. Over the next two weeks, August 1-13,
1346, more than 140 miles were traversed. Once more, there was little opposition to
the march, although Edward did come close to the French armywithin a dozen
mileswhich was gathered at Rouen.11 Several times it appears that he wanted to cross
the Seine River, but all of the bridges were destroyed and the narrow crossings were
heavily guarded. Only at Poissy was a lightly defended crossing found. The guard was
driven off and bridges quickly constructed. Philip VI, expected by Edward to be on
the opposite side, was not present, having traveled to Paris instead. He had left a small
force behind, but it was insufficient to keep the English army from crossing the
river.12
On August 15, Edward III crossed the Seine and travelled north, presumably to
rendezvous with an allied Flemish army moving southward to meet him.13 Philip was
given little option but to follow and perhaps to stop the English before they and the
Flemings could combine forces. (He did have some time before the two armies came
together as the Flemings had stopped to besiege Béthune, 125 miles north of
Poissy.14) Once more the English progress was swift. Edward's army never even
(footnote continued from previous page)
Century,' in Guerre et société en France, en Angleterre et en Bourgogne, xive-xve siècle, ed.
p. Contamine et al (Lille, 1991), pp. 78-80.
9 On the first phase of Edward's Norman campaign see Burne, pp. 140-44; Sumption, pp. 502-17;
Viard, `Crécy,' 9-19; Wailly, pp. 19-22; and Barber, pp. 44-53.
10 On the attack of Caen see Burne, pp. 144-47; Sumption, pp. 507-11; Viard, `Crécy,' pp. 19-32;
Wailly, pp. 20-25; Barber, pp. 53-55; Allmand, p. 15; and Henri Prentout, La prise de Caen par
Edouard III, 1346 (Caen, 1904). Viard claims a garrison of 200 men-at-arms and 100 Genoese
crossbowmen (`Crécy,' pp. 20-22); Sumption sees a garrison of between 1,000 and 1,500 (pp.
507-08).
11 On Philip VI's preparation for battle see Sumption, pp. 513-15, 518-19; Viard, `Crécy,' pp. 4-6,
36-39; Wailly, pp. 38-41; Barber, pp. 55-56; and Allmand, p. 15.
12 On the journey from Caen to Poissy see Burne, pp. 148-53; Sumption, pp. 512-18; Viard,
`Crécy,' pp. 32-50; Wailly, pp. 25-27; and Barber, pp. 55-59. Philip may have believed that
Edward's destination was to be Paris, lying not far from Poissy, but if so he was gravely mistaken.
Attacking Edward during his crossing of the Seine would have been much easier than what
occurred: chasing after him until the English king stopped and offered battle.
13 This is the assertion of Burne (pp. 148-49) and Viard (`Crécy,' p. 33). It may explain why
Edward did not stop or attack Paris. At least, according to Jonathan Sumption, the French military
leaders believed that this is what Edward planned (p. 524). Against the notion of Edward trying to
reach the Flemings see Barber (p. 62) who believes that had the English wished to rendezvous
with the Flemings that they could have done so. Oman also claims that Edward did not want to
reach the Flemish force, but that he only wished to have the way open for a retreat into Flanders if
it was needed ((1905), II:134).
14 On the movement of the Flemish army see Sumption, p. 519, and Viard, `Crécy,' pp. 33-37.
Page 158
paused to pillage or to attack important places, such as Beauvais, although some
wished to do so. Moreover, few encounters were fought with the French, so that by
August 24, Edward III was ready to ford the Somme River at Blanchetaque.
Resistance was minimal; although French numbers were not insignificant and
contained several Genoese crossbowmen, the river was crossed with little loss of
English life.15 Shortly thereafter, just north of the Somme in the woods of Crécy, the
English army camped and waited for the main French force to reach them.16
Why did the English king do this? Alfred Burne supplies an answer:
. . . three new factors decided [Edward III] to offer battle to his old opponent. In the first place,
he now had a fair chance of escape should he be worsted in the battle; for friendly Flanders
now lay behind him and so long as he did not allow Philip to outmarch him, his line of retreat
was secure. Secondly, he was now in Ponthieu, his grandmother's patrimony, on soil that he
considered his own; he would not give up this possession without a struggle. Thirdly, the
success of his army in crossing the Somme in the very face of the foe appeared, in that age of
faith, to be a miracle; the God of battle was evidently on his side, his cause was a just one in
the eyes of the Almighty, who would not allow them to be defeated.17
Their faith in God's favor would not be invalidated. On August 26, 1346, the famous
battle of Crécy was fought, and the French were soundly defeated, with many killed.
Philip VI fled from the field eventually to Paris, there to lick his wounds and attempt
to rebound from his defeat. Edward waited for a few days on the battlefield, resting
and nursing his soldiers and burying the dead of both sides. From there he traveled to
Calais where he besieged the town for the next eleven months.
The battle of Crécy is one of the most famous battles of the Middle Ages. Called
decisive by J.F.C. Fuller and Joseph Dahmus,18 the battle has been highlighted in
nearly every medieval military historical survey.19 Thus it should come as no surprise
15 On the English army's movement from the Seine to the Somme see Burne, pp. 153-58;
Sumption, pp. 520-23; Viard, `Crécy,' pp. 51-61; Wailly, pp. 27-31; Barber, pp. 59-61. On
crossing the Somme see Burne, pp. 158-62; Sumption, pp. 523-24; Viard, `Crécy,' pp. 61-66;
Wailly, pp. 31-33, 41-46; and Barber, pp. 61-62. Burne's numbers for the French army at the
Somme are 500 men-at-arms and 3,000 infantry, including Genoese crossbowmen (p. 158).
16 Burne, p. 162; Sumption, p. 525; Viard, `Crécy,' pp. 64-67; Wailly, pp. 47-49; and Barber, pp.
62-64.
17 Burne, p. 162. Wailly believes that Edward could not keep up a pace quick enough to reach the
Flemings before the French caught up to him (pp. 48-49). Allmand questions whether Edward
wanted to come to battle against the French (p. 15).
18 J.F.C. Fuller, A Military History of the Western World, vol. 1: From the Earliest Times to the
Battle of Lepanto (New York, 1954), pp. 444-68 and Joseph Dahmus, Seven Decisive Battles of
the Middle Ages (Chicago, 1983), pp. 169-96.
19 Delbrück, III:454-62; Oman (1905), II:133-47; Lot, I:340-57; Verbruggen, Krijgkunst, pp. 48,
58-59, 204, 529; Contamine, War, pp. 140, 155, 198, 257; and Philippe Contamine, `La guerre de
cent ans: Le XIVe siècle. La France au rythme de la guerre,' in Histoire militaire de la France,
vol. 1: Des origines à 1715, ed. P. Contamine (Paris, 1992), pp. 126-29. See also Burne, pp. 169-
203; Sumption, pp. 524-34; Barber, pp. 63-70; Perroy, pp. 119-21; Allmand, pp. 15-16; Viard,
`Crécy,' pp. 67-84; Desmond Seward, The Hundred Years War: The English in France, 1337-
1453 (New York, 1978), pp. 63-68; Jean Favier, La guerre de cent ans (Paris, 1980), pp. 110-20;
Philippe Contamine, `Crécy (1346) et Azincourt (1415): Une comparison,' in Divers aspects du
moyen age en occident:
(footnote continued on next page)
Page 159
that the battle was also very popular during the fourteenth century, with many
contemporary narrative sources detailing what occurred on the battlefield. In this it
rivals or even surpasses the battle of Courtrai. Lengthy accounts of the battle can be
found in Jean Froissart's Chroniques, Jean le Bel's Chronique, Gilles li Muisit's
Chronicon, Jean de Venette's Chronique, the Grandes chroniques, the Chronique des
quatre premiers Valois, the Chronographia regum Francorum, the Chronique
Normande, the Récits d'un bourgeois de Valenciennes, Geoffrey le Baker's Chronicon,
Chronicon de Lanercost, Thomas of Burton's Chronica monasterii de Melsa, the
Herald of Chandos' Life of the Black Prince, Henry Knighton's Chronicon, the
Anonimalle Chronicle, a later version of Adam Murimuth's Chronicon, the Eulogium
historiarum, the Chronicon comitum Flandriae, and Giovanni Villani's Istorie
Fiorentine, while shorter accounts of the battle are found in the Chronique de
l'abbaye de Saint-Trond, Richard Lescot's Chronicon, Thomas of Walsingham's
Historia Anglicana, The Brut, Jan de Klerk's Brabantse yeesten, the Breve chronicon
de Flandriae, the Liber Pluscardensis, Willem van Berchen's Gelderse Kroniek, and
Jean de Winterthur's Chronicon.20 As well, there remain several letters written by
English participants of the battle and three contemporary political poems.21 What this
means is that trying to decide what occurred on the battlefield of Crécy is a process of
sifting through and weighing many narrative pieces of evidence in order to try to
discover what happened on that August day in 1346, a day which in many ways
determined the course of English/French relations for more than one hundred years.
The difficulty of such an important task was recognized even by the contemporary
authors themselves, with many commenting on the challenge of determining a history
of this battle. Gilles li Muisit puts it well, although not succinctly:
(footnote continued from previous page)
ler Congrès Historique des Jeunes Historiens du Calaisis (Calais, 1977),pp. 29-44; and
Wailly. (N.B. Wailly's work is so fraught with errors that it is almost worthless for the study of
this battle.)
20 Jean Froissart, V:46-80; Jean le Bel, II:100-09; Gilles li Muisit, pp. 243-46; Jean de Venette,
Chronique, ed. and trans. R.A. Newhall (New York, 1953), pp. 42-44; Grandes chroniques,
IX:281-85; Chronique des quatre premiers Valois, ed. S. Luce (Paris, 1862), pp. 16-17;
Chronographia regum Francorum, II:231-34; Chronique Normande, pp. 80-82; Récits d'un
bourgeois de Valenciennes, pp. 229-35; Geoffrey le Baker, pp. 81-86; Chronicon de Lanercost,
pp. 343-44; Thomas of Burton, II:58-59; Herald of Chandos, Life of the Black Prince by the
Herald of Sir John Chandos, ed. and trans. M.K. Pope and E.C. Hodge (Oxford, 1910), pp. 7-11;
Henry Knighton, II:36-38; Anonimalle Chronicle, pp. 22-23; Adam Murimuth, pp. 246-48;
Eulogium historiarum, III:210-11; Chronicon comitum Flandriae, pp. 218-19; Giovanni Villani,
Cronica, ed. M.L. Ridotta (Florence, 1823), VII:161-73; and Jean de Winterthur, pp. 266-68;
Chronique de l'abbaye de Saint-Trond, ed. C. de Borman (Liege, 1877), II:283; Richard Lescot,
pp. 73-74; Thomas of Walsingham, I:268-69; The Brut, II:542-43; Jan de Klerk, II:573-74; Breve
chronicon de Flandriae, in Corpus chronicorum Flandriae, iii, ed. J.J. de Smet (Brussels, 1856),
p. 8; the Liber Pluscardensis, I:292-93; and Willem van Berchen, Gelderse kroniek, ed. A.J. de
Mooy (Arnhem, 1950), p. 3.
21 These letters can be found in Adam Murimuth, pp. 215-17; Robert of Avesbury, pp. 367-72;
Froissart, XVIII:289-90; Walter of Hemingburgh, Chronicon (London, 1848), II:423-26; and
Richard Barber, The Life and Campaigns of the Black Prince, pp. 23-25. The poems can be
found in Laurence Minot, pp. 21-27; Gilles li Muisit, pp. 246-63; A. Coville, ed., `Poems
historiques du début de la Guerre de Cent Ans,' Histoire litterature de France 38 (1949), 282-89;
and L. Leger, ed. and trans., `Un poème Tchèque sur la bataille de Crécy,' Journal des savants
(1902), 323-31.
Page 160
Since the events of war are dubious, and as battle is harsh, everyone fighting tends to conquer
rather than be conquered; and those fighting cannot consider anything going on away from them,
nor are able to judge well even those things which are happening to them. Yet afterwards the
events must be judged. Because many people discuss many things and either refer to the conflict
from the French side, with some supporting their arguments with things which they could not
know with certainty, and others refer to it from the side of the English, with some also
supporting their arguments with things which they do not know to be true; therefore, on account
of these diverse opinions, I will not write that which I cannot prove. But I will write only those
things which I have heard from certain trustworthy persons in order to satisfy the minds of
future readers, not however affirming them to be completely what happened.22
Hardly any contemporary source separates the battle of Crécy from the military
campaign which preceded it. Nor is there a separation between the battle and the siege
of Calais which followed it. But for the purposes of this study, it is necessary to focus
only on the events of the battle.
Most chroniclers begin their accounts of the battle of Crécy with the arrival of Edward
III and his army at the battlefield site. To them there is little doubt that the English
king chose with great care where he was to face the French army, and because the
battlefield still lies virtually undisturbed, we can agree even today with their
determination. It is a relatively flat plain, a `spacious field' writes Adam Murimuth,
which forms a valley in the triangle of roads leading to the villages of Crécy in the
west, Wadicourt in the east, and Fontaine in the south. Further to the south is a large,
impenetrable forest, and on the west runs the narrow, easily fordable Maye River.
Between Crécy and Wadicourt is a small ridge at the highest point of which a windmill
stood. North of the ridge grows a small wood, known today as the Bois de Crécy.23 It
was in this wood that Edward set up camp and along the ridge that he ordered his
troops; from the windmill he surveyed the battlefield. He did so hoping that there
indeed was to be a battle. As the anonymous author of the Chronicon comitum
Flandriae writes: `The king of England was anxious for battle, preferring to die
gloriously in combat, rather than fleeing from it in shameful disgrace.'24
22 Gilles li Muisit, pp. 243-44: `Quoniam eventus belli est dubius, et dum conflictus est
acierum, unusquisque bellans intendit plus vincere quam vinci, et non potest quispiam
considerare undique confligentes, neque bene de his, quae ibidem eveniunt judicare; sed exitus
acta probant, et idcirco quia multi multa dicunt et referunt de conflictu et pro parte regis
Franciae, et suorum aliqui sustinent ea, de quibus non potest sciri certitudo; et aliqui pro parte
regis Angliae et suorum sustinent etiam illa, quae de vero nesciuntur, et sic propter opiniones
diversorum nolo posteris demandare, quod probare non valeram, sed ea quae audivi a
quibusdam fide dignis personis proposui hic intellectui futurorum satisfacere, sic esse tamen
totaliter non affirmans.'
23 Adam Murimuth, p. 246. See also Récits d'un bourgeois de Valenciennes, p. 229. Good
modern descriptions of the battlefield can be found in Burne, pp. 169-70 and Oman (1905),
II:134-36.
24Chronicon comitum Flandríae, p. 218: `Sed rex Eduardus Angliae animatis suis ad proelium,
malens gloriose mori in proelio, quam cum pudoris verecundia fugere ab eodem.' See also
Geoffrey le Baker, p. 82; Jean Froissart, V:21-24; Grandes chroniques, IX:282; and Giovanni
Villani (ed. Ridotta), VII:163-64.
Page 161
Murimuth adds that the king even joked with his men on hearing of the approach of
the French that his wish for a fight was soon to be fulfilled.25
Because the French were still a day's march from the Englishten leagues behind them,
reports the Grandes Chroniques26Edward first set up his camp and rested for the
night. English tents were pitched in the woods, claim most sources, with the wagons
and carts which accompanied the troops linked to form a make-shift fortification, just
outside and to the east of the woods, reminiscent of that constructed by the Flemings
at Mons-en-Pévèle. Within the wagon fortress was placed the English baggage train
and the army's horses.27 Yet this fortification was built, records the Chronique
Normande, not only to protect the baggage and war horses, but also to keep the
French army from attacking the English lines in the rear.28 The army spent the next
several hours repairing their armor, feasting, sleeping, hearing Mass, and preparing for
battle.29
Only a few authorsThomas of Burton, Giovanni Villani, Jean le Bel, and Jean
Froissartrecord the number of English soldiers fighting at Crécy, and none of these
agree with another; in fact, in each of Froissart's three redactions different numbers
are reported. Thomas of Burton, for example, mentions only a total of 3,000 men-at-
arms without indicating a number of other troops. Giovanni Villani notes only the
number of archers, 3,000. Jean le Bel tallies 4,000 cavalry, 10,000 archers, and 10,000
Welsh and foot soldiers. Finally, Froissart's first redaction counts 2,000 men-at-arms,
5,200 archers, and 1,000 Welsh infantry. This increases in his second redaction to
4,000 men-at-arms, 11,000 archers and 4,000 infantry. And, in his third redaction,
Froissart numbers 3,900 men-at-arms with 15,000 generic infantry, including
archers.30 All of these, as well as several other sources which do not enumerate the
English troops, claim that these soldiers were greatly outnumbered by their French
opponents.31
On the following morning and with English spies and French `traitors' reporting that
Philip's army was quickly approaching the battlefield, Edward ordered his troops in
three solid defensive lines.32 All troops were to fight on foot, the cavalry
25 Adam Murimuth, p. 246.
26Grandes chroniques, IX:281.
27 On the English camp see the Anonimalle Chronicle, p. 22; Jean le Bel, II:105; Jean Froissart,
V:25; Jean de Venette, p. 42; Grandes chroniques, IX:281; and Chronique Normande, p. 80. Only
the Chronographia regum Francorum (II:231) contends that the English camp was not in the
woods but instead was located in the wagon fortification.
28Chronique Normande, p. 80.
29 Jean Froissart, V:25.
30 Thomas of Burton, II:58; Giovanni Villani (ed. Ridotta), VII:165; Jean le Bel, II:100, 105-06;
and Jean Froissart, V:31, 33, 35-36.
31 See also Herald of Chandos, p. 9 and Giovanni Villani (ed. Ridotta), VII:164.
32 On Edward's spies see Michael of Northburgh's letter in Barber, Life and Campaigns of the
Black Prince, p. 24 and Richard Lescot, p. 73. On the three line formation see Geoffrey le Baker,
pp. 83-84; Jean le Bel, II:101-02; Jean Froissart, V:31-36; Thomas of Burton, II:58; Adam
Murimuth, p. 246; Henry Knighton, II: 37-38; Giovanni Villani (ed. Ridotta), VII:163-64;
Chronique des quatre premiers Valois, p. 16; Chronicon comitum Flandriae, p. 218; and The
Brut, II:542.
Page 162
dismounting to stand alongside the rest of the infantry.33 Even the sixteen-year-old
Edward, the Black Prince, who was in command of the first line, was dismounted.34
Where the English archers were located in this formation is somewhat of a mystery
when one looks at the narrative accounts. Most chroniclers, among them Thomas of
Burton, Jean de Venette, and Gilles li Muisit, indicate the ordering of the archers
within the English formation, but not where they were placed in that order.35 Others,
Giovanni Villani, Geoffrey le Baker, the Récits d'un bourgeois de Valenciennes, and
Jean Froissart, are more specific, but not in agreement. Villani, the contemporary
Florentine civic official who often visited Flanders and France before his death in
1348, and whose account of Crécy is generally among the most accurate, claims that
the archers were unusually placed `come detto è addietro' or behind the English and
Welsh infantry.36 Baker, writing before 1356, and the Récits, written c. 1366, maintain
the more conventional English archer placement, along the flanks of each line.37
But it is Froissart who has most confused modern historians of the battle. Using a
word, `herce' or `erce,' not used before in conjunction with the ordering of archers,
Froissart describes the archers' placement at Crécy.38 But what does Froissart mean by
this term? In an 1895 article in the English Historical Review, E.M. Lloyd discussed
the many interpretations of `herce' and concluded that the term must mean either a line
of archers formed in front of the English infantry or, more likely, two lines of archers
formed along their flanks. In the next issue of the same journal, the views of Lloyd
were criticized by Hereford B. George. `Herce' was defined by George as a `harrow;' in
other words, the archers formed units placed both along the flanks of and as wedges
in between the infantry lines. The debate has continued: in 1897, John E. Morris
agreed with Hereford George; in 1905, Sir Charles Oman advanced the idea of a
flanking formation; in 1923, Hans Delbrück returned to the harrow order; in 1955,
Alfred Burne agreed with George, Morris, and Delbrück; in 1976, Robert Hardy also
accepted the harrow definition, an interpretation which he repeated in 1986, 1990, and
1992; in 1985, Jim Bradbury returned to Oman's flanking formation; and, finally, in
1990, Jonathan Sumption agreed with Oman and
33 Giovanni Villani (ed. Ridotta), VII:163-64; Geoffrey le Baker, p. 82; Adam Murimuth, p.
246; Herald of Chandos, p. 10; Récits d'un bourgeois de Valenciennes, p. 231; Chronicon
comitum Flandriae, p. 218; and Breve chronicon Flandriae, p. 11.
34 On the Black Prince's command of the first line see Adam Murimuth, p. 246; Herald of
Chandos, pp. 8-10; Chronographia regum Francorum, II:232; and Chronicon comitum
Flandriae, p. 218
35 Thomas of Burton, II:58; Jean de Venette, p. 43; and Gilles li Muisit, p. 244. See also Jean le
Bel, II:102, Chronographia regum Francorum, II:231-32; and Chronique Normande, pp. 80-81.
36 Giovanni Villani (ed. Ridotta), VII:165.
37 Geoffrey le Baker, pp. 83-84: `Sagittariis eciam sua loca designarunt, ut, non coram armatis,
set a lateribus regis exercitus quasi ale astarent, et sic non impedirent armatos neque inimicis
occurrerent in front, set in latera sagittas fulminarent.' Récits d'un bourgeois de Valenciennes, p.
231:` . . . le roy Édouart d'Engleterre fist et ordonna ossy briefment les siennes, et ne fist que II
batailles d'archiers à II costés en la manière d'un escut.'
38 This is found in only the first and third redactions of Froissart's Chroniques. Kervyn de
Lettenhove (V:48, 50) and Simeon Luce ((Paris, 1872), III:175, 416) transcribe this as `herce,'
while George Diller, who has only published the second and third redactions of the Chroniques,
has transcribed it as 'erce' ((Geneva, 1972), pp. 726-27.)
Page 163
Bradbury, although, in a misreading of the sources, he contended that the archers had
surrounded themselves with wagon fortifications.39
No doubt the debate over Froissart's meaning of `herce' will continue. But it certainly
should not be an issue at Crécy, as both the more contemporary chronicles of
Geoffrey le Baker and the Récits d'un bourgeois de Valenciennes claim a formation of
archers along the flanks of the infantry. Baker seems most certain in the archers'
placement:
The archers were placed in their order so that they stood not in front of their men-at-arms but on
the sides of the king's army like wings. Thus they would not get in the way of their men-at- arms
nor be attacked head-on by the enemy, but they would shoot their arrows from the flanks.40
And as this was the established formation used by Edward III as seen before at
Halidon Hill and later at Neville's Cross and Poitiersthese battle formations asserted
also by Froissartit seems odd to conclude that another formation, identified as a
`herce,' might be used at Crécy. The archers can only have been ordered along the
flanks of the infantry lines.
It seems certain that there were also some early gunpowder weapons with the English
forces which were used in the battle, although where these were placed in the English
formation cannot be determined from the original sources.41
The English army stood all day in their lines, `from dawn to vespers,' claim Geoffrey
le Baker and Henry Knighton.42 They ate a meal,43 and they listened to a speech from
their king. Edward exhorted his troops not to fear the attacks of their enemy, but to
put their faith in `God and the Blessed Virgin,' who would help them to be victorious.
He also told them that he wanted to fight the French cavalry and not the infantry,
although his reasons for this desire are not recorded. Finally, the king ordered that the
English soldiers were not to succumb to greed by taking any prisoners or seeking for
booty before the battle was over. If this was done, the English line would be
weakened and the battle might be lost.44
39 E.M. Lloyd, `The ''Herse" of Archers at Crecy,' English Historical Review 10 (1895), 538-
41; Hereford B. George, `The Archers at Crecy,' English Historical Review 10 (1895), 733-
38; John E. Morris, `The Archers at Crecy,' English Historical Review 12 (1897), 427-36;
Oman (1905), II:136-37; Delbrück, III:457-58; Burne, p. 172; Robert Hardy, Longbow: A
Social and Military History, 3rd ed. (London, 1992), p. 67; Jim Bradbury, The Medieval
Archer (New York, 1985), pp. 95-105; and Sumption, pp. 526-27. Bradbury's discussion is the
most detailed and convincing.
40 Geoffrey le Baker, pp. 83-84; `Sagitariis eciam sua loca designarunt, ut, non coram armatis, set
a lateribus regis exercitus quasi ale astarent, et sic non impedirent armatos neque inimicis
occurrerent in front, set in latera sagittas fulminarent.'
41 Giovanni Villani (ed. Ridotta), VII:163, 165-67; Grandes chroniques, IX:282; Jean Froissart,
V:46; and Istorie Pistolensi in Scriptores rerum Italicarum, xi, ed. L. Muratori (Rome, 1728), p.
516. For a discussion of the gunpowder weapons at Crécy see Burne, pp. 193-203 and T.F. Tout,
`Firearms in England in the Fourteenth Century,' English Historical Review 26 (1911), 671-73.
Even with such seemingly undeniable proof for the existence of guns at Crécy, there are still some
who doubt that they were there. See Oman (1905), II:142n2 and Wailly, p. 91.
42 Geoffrey le Baker, p. 83 and Henry Knighton, II:37.
43 Jean le Bel, II:106; Richard Lescot, p. 73; and Récits d'un bourgeois de Valenciennes, p. 232.
44 On Edward's speech see Jean le Bel, II:106-07; Geoffrey le Baker, p. 82; Adam Murimuth, p.
246;
(footnote continued on next page)
Page 164
During their wait for the battle to begin, the English also dug ditches in the battlefield,
if Geoffrey le Baker is to be trusted. He writes:
the English . . . quickly dug many holes in the earth in front of the first line, one foot deep and
one foot wide, so that if it happened that the French cavalry were able to attack them, the horses
might stagger because of the holes.45
Perhaps most importantly, the English troops watched as the French soldiers
continued to arrive. Ultimately, they saw a large force composed of numerous knights
in the most expensive and impressive military attire, large numbers of Genoese
crossbowmen, and a seemingly unending multitude of other infantry.46 Philip VI had
been ardently following the English army, realizing, Jean de Venette insists, that `he
had been duped and tricked and was sorrowful therefore.' He may also have hoped
that such a fervent pursuit of the English might cause them to flee from France
without coming to battle, at least this is the assessment of the Chronicon comitum
Flandriae.47 But if this was the French king's hope, it must have been shattered when
he learned from his scouts that Edward III's army had stopped at Crécy and was
preparing for battle.48 Nevertheless, he prepared his own troops for battle.
The French army was, by most medieval standards, massive, greatly outnumbering the
English troops.49 Contemporary estimations of French cavalry run from 12,000 to
30,000, with more than 60,000 foot frequently reported.50 Other sources simply note
the imposing sight of such a large force of armed men; `whoever saw the strength and
power of the king of France could only describe it as a great marvel,' writes the Herald
of Chandos.51 Among the French troops was also a large number of Italian mercenary
crossbowmen, 2,000-12,000 according to contemporary sources.52
(footnote continued from previous page)
and Chronique Normande, p. 80. The Chronique Normande is alone in insisting that Edward
wanted to fight the French cavalry, and Jean le Bel is alone in claiming that Edward ordered the
taking of no prisoners or booty before the battle's end.
45 Geoffrey le Baker, p. 83: `Anglici . . . effodierunt in parvo tempore multa foramina in terra
coram acie prima, profunditatem unius pedis et eandem latitudinem habente quolibet illorum, ut,
si, quod abfuit, equites Gallicorum ipsos nimis fuissent insecuti, equi ad foramina titubassent.'
46 Geoffrey le Baker, p. 82.
47 Jean de Venette, p. 42 and Chronicon comitum Flandriae, p. 218. See also Gilles li Muisit, p.
244; Jean le Bel, II: 100; Richard Lescot, p. 73; Grandes chroniques, IX:281; Chronique des
quatre premiers Valois, p. 16; and Thomas Walsingham, I:268.
48 Jean le Bel (II:101) and Jean Froissart (V:28-30) both claim that Philip sent two parties of
scouts to spy on the English position. The first reported that the English had stopped at Crécy,
while the second reported the English formation.
49 The Breve chronicon de Flandriae (p. 11) claims that the French army outnumbered the
English either 40 or 20 to 1. The Brut (II:542) declares that the French had more in one of their
four lines than the English had in their whole army. Both probably exaggerate the French to
English ratio but give the contemporary impression of an English victory over incredible odds.
50 The following numbers are given: 12,000 cavalry and 60,000 infantry (Richard Wynkeley in
Barber, The Life and Campaigns of the Black Prince, pp. 19-20 and Adam Murimuth, p. 246);
20,000 cavalry (Chronicon comitum Flandriae, p. 218 and Anonimalle Chronicle, p. 22); 20,000
cavalry and 100,000+foot (Jean le Bel, II:100); and 30,000 cavalry (Thomas of Burton, II:58).
51 Herald of Chandos, p. 9: `Qe veist venir la puissance / Et la poair du Roy de ffrance / Graunt
meruaille serroit a dire.' See also Geoffrey le Baker, p. 82; Henry Knighton, II:37; Récits d'un
bourgeois de Valenciennes, p. 230; Eulogium historiarum, III:210; and Willem van Berchen, p. 3.
52 The following numbers of Genoese crossbowmen are found in contemporary sources: 2,000
(footnote continued on next page)
Page 165

Battle of Crécy
(First Phase)
Known as Genoese crossbowmen, it is doubtful that they all came from that northern
Italian town. These would play a very important part in the battle. All of these troops
were led by the French king, his constable and marshals, and many other great lords
and knights of France, with some others, most notably the kings of Bohemia and
Majorca, coming to fight with the French as well.
As Philip VI approached the field he called his councilors together. The French king
told his nobles what he had learned from his scouts, that the English had stopped at
Crécy, and that they had ordered their troops in a defensive formation, one which they
were unlikely to break in order to attack the French army. The French councilors
advised the king to attack the enemy lines forcefully with his cavalry, but that this
attack should not take place until the following morning. The final part of this
recommendation was not heeded.53
To counter the English formation, Philip VI also ordered his troops for battle,
although the contemporary sources seem unclear as to how many lines he formed
with these troops. Between four and nine lines are mentioned by the chroniclers.54
(footnote continued from previous page)
(Chronographia regum Francorum, II:231); 10,000 (Gilles li Muisit, p. 244); and 12,000
(Jean le Bel, II:100). See also Jean de Venette, pp. 42-43.
53 On Philip's war council see Jean le Bel, II:101-02; Récits d'un bourgeois de Valenciennes, p.
230; Jean Froissart, V:38-56; and Richard Lescot, p. 74.
54 The following numbers of lines are noted by contemporary sources: 4 (The Brut, II:542); 5
(Récits d'un bourgeois de Valenciennes, pp. 230-31); 8 (Eulogium historiarum, III:210); and 9
(Geoffrey le
(footnote continued on next page)
Page 166
Philip placed himself in the rear line, according to the Récits d'un bourgeois de
Valenciennes, in the vanguard, according to Richard Wynkeley.55 As the course of
battle will show, the Récits is undoubtedly correct. Next to the king, the oriflamme
was unfurled. This was done, Geoffrey le Baker writes, so that `it was not lawful for
anyone, on penalty of death, to take any prisoners.'56
The pride of the French cavalry is noted directly or indirectly by almost all
contemporary authors. Two stories in particular stand out as examples of this pride.
The first comes from Geoffrey le Baker who insists that the French leaders were so
confident in their victory that they chose who they would be allowed to take as
prisoners:
So secure were the leaders of the French in the multitude of their army, that they asked for
specific Englishmen to be given to them as prisoners. The king of Majorca asked that the king of
England be given to him; others sought the prince, others the earl of Northampton, and others
other leaders, according to their noble rank.57
The second is recorded by Jean Froissart. He reports that after receiving the
recommendation of his councilors to wait until morning for battle, Philip VI
commanded his army to discontinue their advance. But while those in the front of his
army halted their march, those behind them refused to do so, continuing their move
forward. It soon became a question of honor. When those who had halted saw that
their comrades had pressed on, they took up the march again, wishing to prove their
own courage:
And thus a great pride and arrogance governed the events, because each wished to surpass his
companion . . . Neither the king nor his marshals were able to stop their troops, for there was
such a great number of soldiers and such a large number of great lords, each of whom wished to
demonstrate his power. They rode on in this way, without formation and without order, until
they approached the enemy and saw that they were in their presence.58
(footnote continued from previous page)
Baker, p. 82Baker also mentions 8 lines of order later in his chronicle, p. 83). See also Gilles
li Muisit, p. 244; Giovanni Villani (ed. Ridotta), pp. 164-65; and Chronique Normande, p. 80.
55Récits d'un bourgeois de Valenciennes, p. 231 and Richard Wynkeley in Barber, The Life and
Campaigns of the Black Prince, pp. 19-20.
56 Geoffrey le Baker, p. 82: `Set tirannus . . . iussit explicari suum vexilium quod vocatur
Oliflammum, quo erecto, non licuit sub pena capitis aliquem capere ad vitam reservandum.' See
also Gilles li Muisit, p. 244.
57 Geoffrey le Baker, p. 82: `Tantum securi fuerunt in multitudine sui exercitus heroes Francorum,
quod singuli pecierunt singulas personas Anglicas suis carceribus mancipandas. Rex Malogrie
peciit regem Anglorum sibi dari, alii principem, alii comitem Norhamptonie, alii alios, secundum
quod videbantur nobiliores.'
58 Jean Froissart, V:42: `Si chevaucièrent si doi mareschal, li uns devant, et li aultres derrière, en
disant et commandant as banerès: "Arrestés, banières, de par le roy, ou nom de Dieu et de
monsigneur saint Denis." Cil qui estoient premier, à ceste ordenance s'arrestèrent, et li darrainier
point, mès chevauçoient toutdis avant, et disoient que il ne s'arresteroient point jusques adont que
il seroient ossi avant que li premier estoient; et quant li premier veoient que il les approçoient, il
chevauçoient avant. Ensi et par grant orgueil et boubant fu demenée ceste cose, car cascuns voloit
sourpasser son compagnon . . . Ne ossi li rois, ne si mareschal ne peurent adont estre mestre de
leurs gens, car il y
(footnote continued on next page)
Page 167
Forced in confusion and disarray by their pride into a battle situation, the French army
could not effectively regroup or recover from their disorder before beginning the
battle. Nor could they keep from beginning the battle immediately.59
One other matter must also be discussed before looking at the battle of Crécy proper.
Sometime during the last phases of the French march to the battlefield, and while the
English army stood in their formations, there fell a heavy rain shower, complete with
thunder and lightening. Although it seems from the original sources that this rain had
dissipated by the beginning of the battle, it also seems certain that it made the field of
action muddyit had not rained previously for more than six weeks, according to
Thomas of Burton.60 There may also have been other effects of the rain, in particular
on the bowstrings of the Genoese crossbowmen. Both Jean de Venette and the
Grandes chroniques note that 'the strings of the Genoese crossbowmen . . . were
soaked by the rain and shrank, so that when it was time for them to be drawn against
the English, they were, woe is me! useless.' The same effect was not felt by the
English longbowmen, however, as 'they had quickly protected their bows by putting
the bowstrings on their heads under the helmets.'61
The battle began with the sound of trumpets, drums, and other French musical
instruments. These were sounded, claims the Récits d'un bourgeois de Valenciennes,
'to cause fear' among the enemy.62 But as a scare tactic they were ineffective,
especially as the English were able to answer back with their own noise-makers, their
cannons, which were discharged also at the beginning of the battle.63 Through it all,
Edward III's soldiers stood solidly in their formation.
Immediately upon hearing the sound of the instruments, the Genoese crossbowmen
began their approach towards the English formation. As seen in the battle of Courtrai,
this was a standard French tactic: the crossbowmen were to get within range of the
enemy's line, let loose at least one volley (or perhaps more if they had the time and
ability to reload), and in this way cause confusion and possibly flight among their
opponent's force. Ultimately it was hoped, and had often been proved, that this tactic
would enable the main part of the French army, the vaunted cavalry, to charge
(footnote continued from previous page)
avoit si grant nombre de gens et si grant nombre de grans signeurs, que cascuns par envie voloit
là monstrer sa poissance. Si chavaucièrent en cel estat, sans arroy et sans ordenance, si
avantque il approcièrent les ennemis et que il les veirent en leur présence.' See also Jean
Froissart, V:45 and Jean le Bel, II:102.
59 Others describing the French pride at Crécy include: Richard Lescot, p. 74; Jean Froissart,
V:39-40; and Giovanni Villani (ed. Ridotta), VII:163.
60 Thomas of Burton, II:58; Henry Knighton, II:37; Jean de Venette, p. 43; Récits d'un bourgeois
de Valenciennes, p. 232; Grandes Chroniques, IX:282; and Jean Froissart, V:48-49.
61 Jean de Venette, p. 43 and Grandes chroniques, IX:282. Both of these sources record the
shrinking of the Genoese crossbow strings, but only Jean de Venette records the English protection
of their strings. While some modern historians of the battle have either disregarded or determined
that the story of the wet and dry bow strings was merely an excuse for later Genoese inaction
(Oman (1905), II:141-42; Delbrück, III:458; Wailly, pp. 66-67; Barber, p. 66; Bradbury, pp. 106-
07), the relative difficulty of stringing a crossbow in comparison to the stringing and unstringing of
a longbow would give the English a quicker solution to the problem of rainfall than it would the
Genoese.
62Récits d'un bourgeois de Valenciennes, p. 232 and Henry Knighton, II:37.
63 Giovanni Villani (ed. Ridotta), VII: 163, 165-67; Grandes chroniques, IX:282; Jean Froissart,
V:46; and Istorie Pistolensi, p. 516.
Page 168

Battle of Crécy
(Second Phase)
on and defeat the confused and disordered defensive line. But, as the anonymous
author of the Chronographia regum Francorum succinctly puts it, 'sed aliter
evenitsomething else happened.'64 Jean Froissart describes the archery exchange:
When the Genoese had all been brought together and put in order, and after they had begun to
approach their enemy, they started to shout as loud as they could to frighten the English. But the
English remained quiet and did not move. Thus the Genoese shouted again and moved a bit
closer, but the English still remained silent and did not move. The Genoese again shouted, very
loud and very clear, and shortly thereafter they raised their crossbows and began to fire. The
English archers, when they saw this formation, took one step forward and let their arrows fly in
such large numbers that they fell on the Genoese so evenly that it seemed like snow. The
Genoese, who had not earlier faced such archers as those of the English, when they saw that the
arrows had pierced their arms, heads, and faces, became so confused that they cut their
bowstrings and threw away their crossbows. Then they turned and fled.65
64Chronographia regum Francorum, II:231.
65 Jean Froissart, V:49:`Quant li Génevois furent tout recueilliet et mis ensamble, et li deurent
approcier leurs ennemis, il commencièrent à juper très-haut que ce fu merveilles, et le fisent pour
esbahir les Englès, mès li Englès se tinrent tout quois, ne onques n'en fisent nul samblant.
Secondement encores jupèrent ensi, et puis alèrent un petit avant, et les Englès retoient tout quoi,
sans yaus mouvoir de leur pas. Tiercement encores jupèrent moult hault et moult cler, et passèrent
avant et
(footnote continued on next page)
Page 169
The Genoese were fatigued after already having marched for more than six leagues
that day, notes Jean Froissart; and their advance was too hasty, causing them to be
disordered and confused, insists Jean de Venette.66 Most importantly, claims Gilles li
Muisit, the Genoese could not withstand the English archery onslaught as they had no
armor and carried no shields.67 This led to large numbers of Genoese casualties lying
on the battlefield.68 The rest quickly fled from the massacre.
Seeing what was occurring in the archery exchange, the French constable and
marshals approached Philip VI and complained that their Genoese mercenaries were
unable to compete against the English troops and were now in flight.69 Philip
responded by commanding his cavalry to ride forward and kill the fleeing Genoese.
At least this is the contention of Gilles li Muisit, Jean de Venette, Jean Froissart, the
Grandes chroniques, and the Chronographia regum Francorum. Again Froissart tells
the story:
Between them [the Genoese] and the French there was a large body of men-at-arms, mounted
and richly armed, who saw the failure of the Genoese, and made it so they could not retreat.
Because the king of France, greatly angered when he had seen their disorder and then their
defeat, commanded the men-at-arms and said: 'Listen! Kill all that rabble: They hold us back
and block the road without reason!' Then the aforesaid men-at-arms rode among them and
attacked them, and many were cast down and fell, unable to save themselves.70
Jean le Bel, on the other hand, does not blame the king for this lethal plan. Instead, he
writes that the knights were simply so excited to enter the battle, being 'envious' of the
archers who had been able to make the first strike against the English, that
(footnote continued from previous page)
tendirent leurs arbalestres et commencièrent à traire. Et cil arcier d'Engleterre, quant il veirent
ceste ordenance, passèrent un pas avant, et puis fisent voler ces saïettes, de grant façon, qui
entrèrent et descendirent si ouniement sus ces Génevois que ce sambloit nège. Les Génevois
qui n'avoient point apris à trouver tels arciers que sont cil d'Engleterre, quant il sentirent ces
saïettes qui leur perçoient bras, tiestes et banlèvres, furent tantos desconfi, et copèrent li
pluiseur d'yaus les cordes de leurs ars, et li aucun les jettoient jus: si se misent ensi au retour.'
See also Jean Froissart, V:46-49, 51-53; Jean le Bel, II: 102; Gilles li Muisit, pp. 244-45; Jean
de Venette, p. 43; Geoffrey le Baker, p. 83; Grandes chroniques, IX:282; Chronique
Normande, pp. 80-81; Chronographia regum Francorum, II:231-32;and Giovanni Villani (ed.
Ridotta), VII:165-66. There are a few variations to Froissart's story: Gilles li Muisit insists
that the French foot also made this attack; Jean de Venette claims that the crossbowmen were
unable to fire any bolts; and the Grandes chroniques believes that the English cannons were
what frightened the Genoese from the battlefield.
66 Jean Froissart, V:48 and Jean de Venette, p. 43.
67 Gilles li Muisit, p. 244.
68Chronographia regum Francorum, II:232.
69 Gilles li Muisit, p. 245.
70 Jean Froissart, V:49: 'Entre yaus et les François avoit une grande haie de gens d'armes, montés
et parés moult richement, qui regardoient le convenant des Génevois, sique quant il cuidièrent
retourner, il ne peurent; car li rois de France, par grant mautalent, quant il vei leur povre arroi et
que il se desconfisoient ensi, commanda et dist: "Or tos, or tos! tués toute ceste ribaudaille: il
nous ensonnient et tiennent le voie sans raison." Là veissiés gens d'armes entoueilliés entre yaus
férir et fraper sus yaus, et les pluiseurs trébuchier et chéir parmi yaus, qui onques puis ne se
relevèrent.' See also Jean Froissart, V:52; Gilles li Muisit, p. 245; Jean de Venette, p. 43; Grandes
chroniques, IX:283; and Chronographia regum Francorum, II:232.
Page 170
they 'accidentally' rode down the Genoese crossbowmen.71 Whoever was to blame,
this maneuver, which the Chronique Normande insists killed many of the
crossbowmen, raised the English morale and caused confusion and disorder among
the French cavalry.72
After the slaughter of Genoese crossbowmen or, if Jean le Bel is to be believed,
concurrent with it, the French cavalry charged forward to attack the front line of
English infantry. Although the archery exchange had proved a defeat for the French
army, its main body, the cavalry, armed with lance and sword, was still an impressive
and formidable force, which in the past with its ordered charges had often caused
infantry foes to flee in panic before even encountering them. This had not, however,
happened at Courtrai, Arques, Mons-en-Pévèle, or Morlaix; nor would it happen at
Crécy. In fact, it seems that while several charges were made during the battle, none
could be called 'ordered', and this may be the cause of their ultimate failure.
The initial charging cavalry were first disrupted by the fleeing, dead, and wounded
crossbowmen over and through whom they were forced to ride.73 This was followed
by an archery onslaught from the English longbowmen who, flushed with confidence
by their success against the Genoese, continued to fire their arrows at the oncoming
troops. What was the result of this? Most contemporary battle commentators report
that the arrows of the English longbows caused the death of many men and horses.
Indeed, the Chronicon comitum Flandriae claims that the whole line of French
cavalry, which he numbers at 2,000, was 'entirely destroyed' by this tactic.74 However,
others, namely Geoffrey le Baker, the Grandes chroniques, and the Chronographia
regum Francorum, report only the wounding and slaying of horses during this part of
the attack. So many horses were wounded by arrows, asserts the author of the
Grandes chroniques, that 'it is pitiful and sad to record it.'75 No matter which account
is accepted, the effect on the charge was the same: increased disruption and
confusion. As Jean le Bel puts it:
On their side the archers fired so skillfully that some of those on horses, feeling the barbed
arrows, did not wish to advance, while others charged forward as planned; some resisted them
tirelessly, while others turned their backs on the enemy.76
71 Jean le Bel, II:102-03. See also Richard Lescot, p. 74.
72Chronique Normande, p. 81. See also Jean de Venette, p. 43.
73 Geoffrey le Baker, p. 84.
74Chronicon comitum Flandriae, pp. 218-19: ' . . . atque cum suis in Gallicos irruens ex una
parte cum gladiis et lanceis, et suis sagittariis cum sagittis equos et homines transverbantibus,
illam omnino devicit aciem.' See also Adam Murimuth, p. 247; Eulogium historiarum, III:211;
Jean Froissart, V:47, 49-50, 52-53; Récits d'un bourgeois de Valenciennes, p. 233; Chronique
des quatre premiers Valois, p. 16; and Laurence Minot, p. 24.
75Grandes chroniques, IX:283; Geoffrey le Baker, p. 84; and Chronographia regum Francorum,
II:232.
76 Jean le Bel, II:103: 'Et d'aultre part les archiers tiroient si merveilleusement que ceulx à
cheval, sentans ces flesches barbelées [qui] faisoient merveilles, l'ung ne vouloit avant aler,
l'aultre sailloit contremont si comme arragié, l'aultre regimboit hydeusement, l'aultre retournoit le
cul par devers les anemis.' See also the Chronographia regum Francorum, II:232.
Page 171
Thus by the time that the French cavalry actually encountered the English line they
were in complete disorder and their impetus had been lost. They simply had 'no
assembly,' reports the Chronique Normande.77 The opposite was true on the English
side. Jean le Bel writes that the English soldiers held their infantry line with strength
and solidarity, and that this was 'so wise and so intelligent that fortune turned to
them.'78 Geoffrey le Baker describes the action:
When fighting with the English men-at-arms, the French were beaten down by axes, lances, and
swords. And in the middle of the army, many French soldiers were crushed to death by the
weight of numbers without being wounded.79

It was a brutal fight, described by the bourgeois of Valenciennes as 'very perilous,


murderous, without pity, cruel, and very horrible.'80 The Herald of Chandos agrees:
'That day was there battle so horrible that never was there a man so bold that would
not be abashed thereby.'81 The French cavalry made a number of attacks on the
English line, but how many can not be determined by the original sources: two attacks
were repulsed claims Adam Murimuth; while, according to Richard Wynkeley, three
attacks were turned back; finally, Geoffrey le Baker asserts that 'at this time the French
three times shouted hostilely at our troops, and fifteen times they charged them.'82
Geoffrey le Baker adds further that those French horsemen 'killed, wounded, or
fatigued' were quickly replaced by 'fresh troops,' which continually kept the English
line engaged in combat.83
With the archers on the flanks, the French charges became directed at the center of the
English front line, the section commanded by the Black Prince. Indeed, the Prince
himself became the target of many direct attacks, and despite on one occasion being
'compelled to fight on his knees,' the Prince and his men held their position strongly
against the continual onslaughts of French cavalry charges.84
Many feats of arms are also reported on the other side of the conflict. This was not a
battle where one side's army fought with less courage than the other.85 For the
77Chronique Normande, p. 81. See also Grandes chroniques, IX:283 and Adam Murimuth, p.
246.
78 Jean le Bel, II:106: 'Et attendirent tant que les Françoys vinrent, et firent si sagement et sy à
point que la fortune tourna pour eulx.'
79 Geoffrey le Baker, p. 84: 'Cum Anglicis armatis confligentes securibus, lanceis, et gladiis
proternuntur, et in medio exercitu Francorum multi compressi a multitudine honerosa sine winere
opprimuntur.' See also Thomas Walsingham, I:268-69 and Laurence Minot, p. 23.
80Récits d'un bourgeois de Valenciennes, p. 232: 'et dura la bataille moult périlleuse, murdrière,
sans pité, creuse et trèsorrible . . . '
81 Herald of Chandos, p. 9.: 'Celuy iour ot il bataille / Si orible qe tout sanz faille / Unqes ne fuist
corps si hardis / Qe nen poeit estre esbahis.' See also Michael Northburgh's letter in Barber, Life
and Campaigns of the Black Prince, p. 24 and Willem van Berchen, p. 3.
82 Adam Murimuth, p. 246; Richard Wynkeley in Barber, Life and Campaigns of the Black
Prince, p. 20; and Geoffrey le Baker, p. 84: 'in quanto tempore ter Gallici nostros exclamaverunt
hostiliter, quindecies nostris insultum dederunt.'
83 Geoffrey le Baker, p. 84.
84 The quote is from Geoffrey le Baker, p. 84. See also Récits d'un bourgeois de Valenciennes, p.
233 and Anonimalle Chronicle, p. 22.
85 On French military prowess see Jean Froissart, V:54-55 and Thomas Walsingham, I:268.
Page 172
French, both Philip VI and John, the king of Bohemia, are particularly singled out for
their fighting prowess. The former, the Chronique Normande records, had two horses
killed from under him during the battle,86 while the latter, blind, is reported by Jean
Froissart to have asked his knights to take him into the middle of the fight: 'Lords, you
are my men, my friends, and my companions. Today I would ask and require you to
take me forward enough that I am able to strike a blow with my sword.' They did so,
having first tied their horses together so that the blind king would not get lost in the
chaos of fighting.87 Other French soldiers cited for their boldness on the battlefield
include the counts of Alençon, Flanders, Saint-Pol, and Blois, and the duke of
Lorraine.88 But valor and fighting prowess alone could not bring victory.
'Thus from sunset to the third quarter of the night was shown the horrid face of Mars.'
So wrote Geoffrey le Baker about the length of the battle of Crécy.89 Perhaps Jean
Froissart is correct in concluding that 'it was very late when the battle began,' and that
this hurt the French military cause 'for many men-at-arms, knights, and squires lost
their lords and leaders during the night; so they wandered the field and did not know
where to go.'90 On the other hand, the approaching darkness had little to do with the
fact that the English line had held so long against many serious French cavalry attacks.
More importantly, as darkness drew on, Philip VI chose to leave the battlefield. Why
he did this cannot be determined from the original sources. Thomas of Burton and
Gilles li Muisit claim that he fled after assessing that his army could not defeat the
English.91 Several English chroniclers, following the lead of Richard Wynkeley's
letter, maintain that Philip was wounded by an arrow which had struck his jaw, but no
French source confirms this.92 Wynkeley also asserts that the English troops had come
close enough to the French king to kill his standard-bearer and shred his standard.93
Other sources report that Philip had been taken from the field by Jean
86Chronique Normande, p. 81. See also Chronographia regum Francorum, II:233; Chronique
des quatre premiers Valois, p. 16; Adam Murimuth, p. 247; and Thomas Walsingham, I:269.
The Grandes chroniques (IX:283) claims that Philip wanted to fight Edward III at Crécy and
searched everywhere for him, but he could not find the English king.
87 Jean Froissart, V:53-54, 55-56: 'Signeur, vous estes mi homme et mi ami et mi compagnon à le
journée d'ui; je vous pri et requier très-espécialment que vous me menés si avant que je puisse
férir un cop d'espée.' On John of Bohemia see Walther Rose, 'König Johann der Blinde von
Böhmen und die Schlacht bei Crécy (1346),' Zeitschrift für historisches Waffenkunde 7 (1915-
17), 37-60.
88 Jean Froissart, V:63.
89 Geoffrey le Baker, p. 84: 'Sic a solis occasu usque ad terciam noctis quadrantem fuerat
vicissim orrida Martis facies ostensa.' See also Jean le Bel, II:103, 106; Gilles li Muisit, p. 245;
Grandes chroniques, IX:283; and Thomas of Burton, II:58.
90 Jean Froissart, V:68: 'car, quant la bataille commença, il estoit jà moult tart, et ce greva plus
les François que aultre cose, car pluisseurs gens d'armes, chevaliers et esquiers, sus la nuit,
perdirent lors signeurs et lors mestres. Si vaucroient par les camps et ne savoient où il aloient.'
See also Jean Froissart, V:61.
91 Thomas of Burton, II:58-59 and Gilles li Muisit, p. 245.
92 Richard Wynkeley in Barber, Life and Campaigns of the Black Prince, p. 20; Thomas of
Burton, II:58; Adam Murimuth, p. 247; Henry Knighton, II:38; and Eulogium historiarum, III:210-
11.
93 Richard Wynkeley in Barber, Life and Campaigns of the Black Prince, p. 20.
Page 173
of Hainault, who was worried about the king's safety.94 What is known for certain is
that the king and his party fled first to La Broye and then to Amiens.95
As their king fled, so too did the rest of the French army. The English, obviously
fatigued by the heavy fighting that had occurred during the day, did not pursue the
fleeing troops. Instead, they simply camped where they had fought, trying to get some
sleep in anticipation of more fighting that night or on the following day.96 But there
was to be no continuation of the battle. On the next day a few French troops were
found, principally by the earls of Northampton and Norfolk, although who and how
many they were is difficult to determine from the original sources. Jean le Bel calls
them 'common soldiers' who had no leaders, while Geoffrey le Baker and The Brut
declare them to be a large forcefour divisions claims Baker. Giovanni Villani gives
them a number, 3,000 cavalry and 4,000 infantry. And Jean Froissart writes that these
were new troops, from the towns of Rouen and Beauvais, who 'did not know about
the defeat which had occurred on the previous day.' All agree, however, that whoever
and how many they were, they were quickly killed or routed.97
The English had won the battle of Crécy. Although they had taken few prisoners and
little booty, and although they had not pursued the fleeing French forces, there was no
question of their victory.98 After praising God and holding a celebratory mass on the
battlefield, Edward III commended his soldiers and awarded some with knighthood.99
The dead of both sides were then buried in pits near where they had fallen. The
French dead were particularly numerous and are commented on by every
contemporary source. Among these were several notables, nine princes, more than
94 Jean Froissart, V:57-58; Chronographia regum Francorum, II:233; and Chronique
Normande, pp. 81-82.
95 Richard Wynkeley in Barber, Life and Campaigns of the Black Prince, p. 20; Giovanni Villani
(ed. Ridotta), VII:168; Gilles li Muisit, p. 245; Jean le Bel, II:103-04; Jean Froissart, V:57-59, 64;
Chronographia regum Francorum, II:232-34; Chronique Normande, pp. 81-82; Chronique des
quatre premiers Valois, p. 17; Breve chronicon Flandriae, p. 11; Willem van Berchen, p. 3;
Thomas of Burton, II:58; Adam Murimuth, p. 247; Henry Knighton, II:38; Anonimalle Chronicle,
p. 23; Eulogium historiarum, III:210-11; Chronicon de Lanercost, p. 344; Thomas Walsingham,
I:269; and Laurence Minot, p. 24.
96 Michael Northburgh in Barber, Life and Campaigns of the Black Prince, p. 25; Jean le Bel,
II:107; Giovanni Villani (ed. Ridotta), VII:168; Récits d'un bourgeois de Valenciennes, pp. 233-
34; Grandes chroniques, IX:284; Thomas of Burton, II:58; Henry Knighton, 11:38; and
Anonimalle Chronicle, p. 23. Thomas of Burton (II:58) also writes that during the night the
English feasted on the food supplies of the French and burned their captured weapons.
97 Jean le Bel, II:107; Geoffrey le Baker, p. 85; The Brut, II:542-43; Giovanni Villani (ed.
Ridotta), VII:168; Jean Froissart, V:71: 'Ce dimence au matin s'estoient parti de Abbeville et de
Saint-Rikieren-Pontieu les communautés de Roem et de Biauvais, qui riens ne savoient de le
desconfiture qui avoit esté faite le samedi.' See also Michael Northburgh in Barber, Life and
Campaigns of the Black Prince, p. 24; Grandes chroniques, IX:284; and Adam Murimuth, pp.
247-48. Thomas of Burton (II:59) and Henry Knighton (II:38) both claim that the French troops
encountered by the English were prepared to fight but fled in the face of superior numbers.
98 Jean le Bel (II:106-07) claims that there were no prisoners or booty taken, as they had been so
commanded by Edward III, but both the Grandes chroniques (IX:284-85) and the Récits d'un
bourgeois de Valenciennes (p. 235) claim that the English did capture some booty.
99 On the English mass after the battle see Jean le Bel, II:107; Giovanni Villani (ed. Ridotta),
VII:169; Geoffrey le Baker, p. 85; and Adam Murimuth, p. 248. On Edward's making of knights
see Chronographia regum Francorum, II:234.
Page 174
1,200 knights, and between 15,000 and 16,000 others, according to Jean le Bel;
included among these numbers were the counts of Alençon, Flanders, Blois, Harcourt,
Auxerre, Aumale, Savoy, Moreuil, Nevers, and Grandpré, the duke of Lorraine, the
archbishop of Xanten, and the bishop of Noyon.100 Perhaps the most celebrated death
was that of the king of Bohemia, whose body was found among the slain. His
blindness was known by the English, and his participation in the battle, seen as
bravery rather than foolishness, as well as his royal status accorded his corpse special
treatment. Brought to Edward III, Geoffrey le Baker writes that the body was washed,
wrapped in clean linen, placed on a horse-drawn litter, and returned to Germany; with
the English king and his earls present, the bishop of Durham, celebrated the office of
the dead over the corpse.101 The number of English soldiers killed was far smaller.102
Because of the numerous excellent original sources on the battle, the tactics of Crécy
are simple to discern. So too are the causes of victory for the English and defeat for
the French.
Edward III chose the battlefield with an eye to its defensibility. He found a ridge
between two woods on which to order his troops. The woods would protect his
flanks; to protect his rear he had a wagon fortress constructed. He also may have
ordered the digging of ditches to add further to the security of his position. (The rain
which fell just prior to the battle helped as well.) Edward's soldiers, fewer in number
than their French opponents, were ordered as infantry in a solid defensive formation
on this high ground. Finally, lines of archers were positioned along the flanks of the
English formation.
Edward knew what he wanted to happen at the battle of Crécy. He told his troops that
he wished to fight the French cavalry. He wanted to provoke them to charge uphill,
through the hindrances of a constant archery deluge and over ditcheswhich would
cause disorder among those charginginto his solid infantry formation.
Philip VI also knew what the English king desired. He recognized his opponent's
formation and understood its purpose. He could see it ordered on the ridge at Crécy,
and in 1328 he had encountered a similar formation at Cassel. Finally, his armies
100 Jean Froissart, V:73-77; Jean le Bel, II:108; Gilles li Muisit, pp. 245-46; Jean de Venette,
pp. 43-44; Grandes chroniques, IX:283-84; Chronique des quatre premiers Valois p. 16;
Chronographia regum Francorum, II:233; Chronique Normande, p. 82; Récits d'un bourgeois
de Valenciennes, pp. 233; Geoffrey le Baker, p. 85; Chronicon de Lanercost, p. 344; Thomas
of Burton, II:59; Herald of Chandos, p. 11; Henry Knighton, II:38; Anonimalle Chronicle, p. 23;
Adam Murimuth, pp. 247; Eulogium historiarum, III:210; Chronicon comitum Flandriae, p.
219; Giovanni Villani (ed. Ridotta), VII: 167-68; and Jean de Winterthur, p. 267; Chronique de
l'abbaye de Saint-Trond, II:283; Richard Lescot, p. 74; Thomas of Walsingham, I:268; The
Brut, II:542-43; Jan de Klerk, II:574; Breve chronicon de Flandriae, p. 11; the Liber
Pluscardensis, I:293; and Willem van Berchen, p. 3. Numbers vary as to how many French
soldiers were killed.
101 Geoffrey le Baker, p. 85. See also Jean le Bel, II:108; Giovanni Villani (ed. Ridotta),
VII:169; Jean de Venette, p. 44; Chronographia regum Francorum, II:234; Récits d'un bourgeois
de Valenciennes, p. 235; Herald of Chandos, p. 10; Minot, p. 24; and Jean de Winterthur, p. 267.
102 Only three sources mention the number of English notables dead: 3 (Adam Murimuth, p. 247);
40 (Geoffrey le Baker, p. 85); and 300 (Jean le Bel, II:108-09). No original source mentions
numbers of non-notable English dead. See also Gilles li Muisit, p. 246; Jean de Venette, p. 44; and
Michael Northburgh in Barber, Life and Campaigns of the Black Prince, p. 24.
Page 175
had also faced the English using this same formation at Morlaix. At Cassel, the French
king had won by following a policy of patience, by waiting until the Flemish rebels
could be induced to leave their defensive positions on the hill-top. (A similar policy
outside of the besieged town of Tournai in 1340 had also brought victory.) Should he
have tried patience at Crécy? Gilles li Muisit, for one, believes that he should have.103
But at Crécy, Philip could not follow this policy. The destruction of surrounding
landsa significant factor in provoking the Flemish troops from their Cassel
positionswould incite no similar feeling among the English. Moreover, the king had
also to consider the possible arrival of Flemings from the north to reinforce the
English. Therefore, patience was not an option. Instead of patience, Philip had as a
tactical advantage a large contingent of Genoese crossbowmen, whose intensive and
deadly fire into the English lines he reckoned should cause them either to break into
rout or to become so disordered that a cavalry charge could be effective against them.
To this point the French king and his soldiers had not done anything that could be
considered incautious. (To some it may seem that he should not have attacked on the
evening of August 26, but even that could not be considered incautious in that if
Philip's tactics had worked, the battle would have been over long before dark.) But
the attack of the Genoese crossbowmen failed miserably, as these archers were
completely stopped in their approach by a counter-attack of English longbowmen;
some sources indicate that the Genoese were unable to discharge even one shot. Once
this attack failed, Philip had lost his tactical advantage. What followed were errors
compounded by pride. The Genoese retreated, but were not allowed back to their lines
as Philip VI, perceiving their flight to be treasonous, ordered their destruction by his
own cavalry. This order was accepted with enthusiasm by the French cavalry, but all it
ultimately did was to add further hindrances to the many cavalry charges which
followed against the English infantry lines. Although the fighting was intense, these
lines received the disordered French charges without breaking, and eventually, as
darkness fell, Philip VI left the battlefield followed by his army.
It is perhaps no mistake that both Jean le Bel and Giovanni Villani compare the battle
of Crécy to the battle of Courtrai.104 There are certainly many similarities between the
two battles: Edward III choosing the battlefield, utilizing the natural terrain to protect
his flanks, and placing his troops, fewer in number than their French opponents, in a
solid defensive formation. There are also several differences, the most notable of
which was the participation on both sides of large numbers of archers. But the
similarity which impressed Le Bel and Villani was the way the two armies finally met
in combat, one a charging cavalry and the other a solid defensive line of infantry. As
at Courtrai, the infantry soldiers stood in unity, without giving way to their attackers.
It was a great victory, one which would eventually supersede in importance not only
the battle of Courtrai but almost all other medieval battles.
103 Gilles li Muisit, p. 245.
104 Jean le Bel, II:109 and Giovanni Villani (ed. Ridotta), VII:167.
Page 176

XIV
The Battle of Neville's Cross, 1346
From the successful battlefield of Crécy, Edward III moved to besiege the town of
Calais. This seems to have been part of the 'grand Strategy' of the English king, to
follow a victory over an opponent with a siege of one of that same opponent's major
towns:1 Edward had acted similarly in 1333 by following Balliol's victory at Dupplin
Moor with the siege of the town of Berwick, in 1340 by following the victory at Sluys
with the siege of Tournai, and he would do so again in 1356 by following the battle of
Poitiers with the siege of Rennes. This did two things for the king of England: first, it
allowed him to capitalize on his victory by acquiring territory, something that the
battlefield victory alone did not necessarily achieve. At Berwick, Calais, and Rennes,
he was successful in this strategy, while at Tournai he was unsuccessful. Second, by
besieging an important location in his enemy's lands, he was generally able to draw to
him another of his opponent's armies in an attempt to relieve the siege; this produced
the possibility of a more decisive victory. At Halidon Hill, he defeated the Scottish
army which had come to relieve the siege of Berwick, but at Tournai, Calais, and
Rennes, he failed to bring the French force to battle. Some modern historians, such as
Alfred H. Burne, have criticized this as short-sighted, but Edward seems never to have
wavered from this strategy.2
The siege of Calais was an impressive affair, with few medieval equivalents. Planning
for the siege to be extremely long in duration, the English destroyed the town's
suburbs and quickly constructed an intricate series of siegeworks, which included an
elaborate city known as Ville-à-Neuve complete with a market place and towers built
around the harbor.3 This allowed Edward's army, numbering around 32,000 menthe
largest army of the century, according to J.F. Verbruggenrelatively comfortable
surroundings, often duplicating more a garrison in England than a siege in northern
France.4 Indeed, Thomas of Burton claims that so many prostitutes were in Ville-à-
Neuve during the siege that God cursed the English troops with dysentery, effectively
halving the English force there.5
1 See my 'Hunger, Flemish Participation and the Flight of Philip Vl: Contemporary Accounts of
the Siege of Calais, 1346-47,' Studies in Medieval and Renaissance History n.s 12 (1991),
129.
2 For example, Burne believes that Edward should have sought the capture of Paris after his
victory at Crécy (pp. 206-07).
3 See DeVries, 'Calais,' p. 133 and Jules Viard, 'Le siege de Calais,' Le moyen âge 40 (1929),
178.
4 On the numbers at Calais see J.F. Verbruggen, 'La tactique de la chevalerie française de 1340 à
1415,' Publications de l'université de l'état à Elisabethville 1 (1961), 42 and Henneman, pp.
218-19.
5 Thomas of Burton, II:65.
Page 177
With so many troops in France, and with all of Edward's logistical powers being used
to supply these troops, what was still left in England? In fact, would this not entice the
Scots, who had been defeated only thirteen years before, to reassert their
independence, especially if urged to do so by the French king, who would desire such
a revolt at least to weaken the English effort and even perhaps to break the siege at
Calais? The first question is rhetorical, the answer to the second is yes.
Since their defeat at Halidon Hill the Scots had been anything but quiet. It was true
that there had been a loss of many high-ranking nobles in the battle, some killed and
some taken into custody. And, at least initially, there was the solid establishment of
Edward Balliol and the other disinherited in the ranks of Scottish government: Balliol
served as king; Hugh Beaumont became Earl of Moray and Buchan; Richard Talbot
became Lord of Mar; and David of Strathbogie, earl of Atholl, became Steward of
Scotland. By February 1334, Balliol was even able to hold a parliament at Holyrood,
which, although dominated by the disinherited, brought together many other Scottish
secular and ecclesiastical lords, all of whom gave him homage as their ruler.6 Finally,
the ten-year-old David Bruce, whose reign in Scotland had been supported by the
force defeated at Halidon Hill, had even taken refuge in France, invited to exile there
by Philip VI.7
But there were still many Scottish military leaders who were unwilling to be governed
by the English, in the person of Edward Balliol or Edward III. By the summer of 1334,
John Randolph, who before Halidon Hill was the Earl of Moray, had united the
remnants of the Bruce party and had begun to invade Balliol's territory; before the
English could mobilize their forces, in November, Randolph, with the assistance of Sir
Andrew Moray, William Douglas, and Alexander Ramsey, had recaptured much of
Scotland. For the next three years, Scottish and English troops often invaded each
other's territory. Raids, with their inherent pillaging and destruction, were constant,
with few clashes between troops taking place. Truces, pacifications, and peace
settlements frequently were made and just as frequently broken.8 Even a large
expedition mounted by Edward III in 1335 failed to bring the Scottish army to battle
and thus failed to recapture the territory which he and Balliol had held in 1333.9
Finally, in 1337, as war with France became imminent, Edward III began to lose
interest in his northern enemy. He turned command of his troops thereprobably
numbering no more than 3,500over to Thomas Beauchamp, earl of Warwick, and
settled down to plan an invasion of France.10 For the next few years, Warwick
6 For the most complete discussion of this period see Nicholson, Edward III and the Scots, pp.
139-62. See also Nicholson, Scotland, p. 129 and Packe, pp. 69-70.
7 Nicholson, Edward III and the Scots, p. 157; Nicholson, Scotland, p. 130; and McKisack, p.
117.
8 The most complete discussion of this is Nicholson, Edward III and the Scots, pp. 163-202. See
also Nicholson, Scotland, pp. 130-36; Packe, p. 70; and Bruce Webster, 'Scotland Without a King,
1329-1341,' in Medieval Scotland: Crown, Lordship and Community: Essays Presented to
G.W.S. Barrow, ed. A. Grant and K.J. Stringer (Edinburgh, 1993), pp. 224-30.
9 Nicholson, Edward III and the Scots, pp. 203-36 and Packe, pp. 70-71.
10 Nicholson, Scotland, p. 136 and Packe, p. 71.
Page 178
continued to defend the Balliol claims in Scotland, but mostly without success; the
power of the English in Scotland had begun to dissolve.11
Edward had found success at Sluys in 1340, but this had been followed by defeat at
Tournai, and by the beginning of 1341 he was back home, anxious to find a new
reason to attack France. He again paid no attention to Scotland. Scottish leaders took
advantage of this situation, and a new offensive was launched. On April 16,
Edinburgh Castle had been recaptured, and by the end of May, the kingdom was
secure to its old borders, with Balliol forced to retreat to England. This gave impetus
to the return of David Bruce, who landed in Scotland on June 2. Still Edward III did
little.12
From 1341 to 1346 an unsure truce persisted between the two old enemies. The
English troops along the Scottish border did not travel north, despite the wishes of
their commander, Edward Balliol, who had been given the lead of this force in 1344,
but neither did the Scots cross their southern border. However, when Edward III set
Sail for France again on July 12, 1346, a new opportunity for Scottish invasion was
presented. It would take place during the King's lengthy siege of Calais.13
While Edward III had paid no attention to Scotland during the 1340s, Philip VI of
France had taken a great interest in Scottish affairs. His hosting of David Bruce from
1333 to 1341 had been active, and since Bruce's return to Scotland, the French king
had supported his reinstatement on the throne with funds, friendship, and
encouragement.14 In 1346, after his defeat at Crécy, Philip stood in need of repayment
of these favors. Naturally, he did not expect military intervention in France, and he
obviously recognized the extreme poverty of a state whose finances almost entirely
went to supporting a defense of its borders. On the other hand, an attack of Scottish
troops into the poorly guarded north of England could perhaps draw at least some of
the English troops home, and such a weakened force might allow the French king to
relieve the siege of Calais.15
The Scots responded with fervor. On October 6, 1346, the Scottish army mustered at
Perth. All noble leaders but two were present, as was King David Bruce. The invading
army's progress to the south was swiftCumberland and Westmorland put up little
oppositionand by October 16, the Scottish force was encamped within sight of the
Durham Cathedral.16 But the English were not going to stand by and allow the Scots
easy access into their northern lands. Knowing early of the Scottish intentions,
perhaps even before the army had left Scotland, Lords William La Zouche, the
archbishop of York, Henry Percy, and Ralph Neville had recruited as many troops as
they could. It was not a very large number with so many English
11 Nicholson, Scotland, pp. 136-38; Webster, pp. 230-35; and Packe, pp. 71-74.
12 Nicholson, Scotland, pp. 139-44 and Packe, p. 72.
13 Nicholson, Scotland, pp. 144-45.
14 Prestwich, Three Edwards, pp. 167-68; McKisack, pp. 118-19; W.M. Ormrod, The Reign of
Edward III: Crown and Political Society in England, 1327-1377 (New Haven, 1990), p. 9; and
Scott L. Waugh, England in the Reign of Edward III (Cambridge, 1991), p. 14.
15 Nicholson, Scotland, p. 145.
16 Nicholson, Scotland, p. 146.
Page 179
soldiers serving in France, but its appearance and morale surprised the Scots. On
October 17, 1346, at Neville's Cross, outside the walls of Durham, the Scots were
soundly defeated. Almost the entire army was killed or captured, including David
Bruce, whose next residence was the Tower of London.17 Calais fell to Edward less
than ten months later.
There are several contemporary or near-contemporary accounts of the battle of
Neville's Cross. Many are quite lengthy and detailed. These include: the English
chronicles of Geoffrey le Baker, Thomas of Burton, Henry Knighton, John of
Reading, Robert of Avesbury, and the Anonimalle Chronicle; the Scottish chronicle of
the Liber Pluscardensis; and the 'foreign' chronicles of Jean le Bel, Jean Froissart, and
Giovanni Villani.18 Three poems, two anonymous and one by Laurence Minot, also
deal with the battle.19 Finally, an important, but unusual source for what occurred on
the battlefield is a letter written by Thomas Samson, a clerk of the archdiocese of
York.20

The contemporary commentators of this battle set the conflict at Neville's Cross in the
wider context of the English involvement at the siege of Calais. Jean le Bel exemplifies
this. He begins his account of the battle:
At the time when the noble King Edward was away on his good campaign against the king of
France, a short time after Crécy while he was besieging Calais, King David of Scotland
assembled a large number of men-at-arms to pillage and waste England, because he knew well
that King Edward was not there.21

David believed that 'the whole strength of arms were with King Edward,' declares
Adam Murimuth, leaving only 'farmers, herders, and chaplains, imbeciles and the
decrepit,' writes Henry Knighton; 'only the ecclesiastics and rustics, women and
children had been left behind,' claims John of Reading.22
Most chroniclers also report Philip VI's complicity in the Scottish invasion. To them
the French king was certainly guilty of encouraging David Bruce in the affair,
17 Sumption, pp. 550-53; Oman (1905), II:149-51; Burne, pp. 218-19; Nicholson, Scotland,
pp. 146-47; and Packe, p. 166.
18 Geoffrey le Baker, pp. 86-88; Thomas of Burton, II:60-67; Henry Knighton, II:41-45; John of
Reading, pp. 102-04; Robert of Avesbury, pp. 376-77; Anonimalle Chronicle, pp. 23-28; Liber
Pluscardensis, I:292-95; Jean le Bel, II:125-30; Jean Froissart, V:118-33; and Giovanni Villani
(ed. Ridotta), VII:186-87.
19 All three are published in Joseph Hall's edition of Minot's poems, pp. 30-33, 110-20.
20 Found in Oeuvres de Froissart, ed. Kervyn de Lettenhove (Brussels, 1869), V:489-92.
21 Jean le Bel, II:125-26: 'En celluy temps que le noble roy Edowart avoit eu celle belle avanture
contre le roy de France assez prez de Cressy, et au'il avoit assiegié Calaiz, assembla le roy David
d'Escoce grand nombre de gens d'armes pour venir gaster et exillier Angleterre, car il sçavoit bien
que le roy Edowart n'y estoit pas.' See also Jean Froissart, V:119, 122, 129; Robert of Avesbury,
p. 376; Adam Murimuth, p. 218; Henry Knighton, II:42-43; John of Reading, p. 102; Giovanni
Villani (ed. Ridotta), VII:186; and Liber Pluscardensis, I:292.
22Chronicon de Lanercost, pp. 348-49; Adam Murimuth, p. 218; Henry Knighton, II:42; John of
Reading, p. 102; and The Brut, II:299, 543.
Page 180
hoping to take pressure off of Calais, if not actually driving the English king back to
his kingdom.23 Geoffrey le Baker and Giovanni Villani also insist that Philip sent
troops and money to the Scots, but their claims seem to be unsubstantiated; at least no
French are recorded to have been killed or captured after the battle of Neville's
Cross.24
The Liber Pluscardensis asserts that some Scottish nobles, 'trusty men,' suggested that
David not take the advice of the French king, but that the Scottish king was 'inflamed
by youthful counsels,' and thus entered England.25 However, none of the more
contemporary chronicles concur with this account, and, indeed, the large number of
Scottish troops and nobles which joined with David in his invasion, casts doubt on
what the Liber proclaims. The Chronique Normande reports the Scottish numbers at
6,000. The Chronicon de Lanercost numbers the Scots at more than 2,000 knights and
men-at-arms, 20,000 'drawn from the villages, called ''hobelars",' and more than
10,000 foot soldiers and archers. Jean Froissart records the number of invading
Scottish troops at 3,000 knights and 30,000 other soldiers, a tally which includes all of
the major Scottish nobles. While Jean le Bel agrees with the former number but
increases the latter to 43,000. Thomas Samson claims a number of 2,000 men-at-arms,
20,000 other professional soldiers, and 40,000 'comunes,' non-professional milita
'which use lances, axes, and bows. Finally, the Récits d'un bourgeois de Valenciennes
reports a count of 80,000 soldiers.26 These six are alone in noting Scottish numbers,
and all but the Chronique Normande's are surely exaggerations, but the existence of a
large and powerful force seems certain as it is mentioned by all commentators. It was
'a very strong and excited army,' writes Henry Knighton.27
But it was not an army without opposition, as David Bruce had hoped. The English
had learned of the Scottish invasion early, maybe even before it left Scotland. Perhaps
they were also expecting it, for Giovanni Villani insists that Edward had left some of
his barons behind in England 'to guard his realm.'28 If the invasion was not expected
by the English, their speed of assembly was extremely impressive, for the
23 For example, Jean Froissart, V:120 and Giovanni Villani (ed. Ridotta), VII:186. See also
Robert of Avesbury, p. 376; Adam Murimuth, p. 218; Geoffrey le Baker, p. 86; the Anonymous
of Canterbury, p. 192; Liber Pluscardensis, I:292; Andrew of Wyntoun, II:470-71; and the
Chronique Normande. Also, Laurence Minot's poem on Neville's Cross, as well as an
anonymous companion piece, both accuse Philip VI of a strong involvement in the campaign
that would lead to David Bruce's failure (Minot, pp. 30-33, 110-12).
24 Geoffrey le Baker, p. 86 and Giovanni Villani (ed. Ridotta), VII:186.
25Liber Pluscardensis, I:293. Andrew of Wyntoun agrees somewhat with this; however, he only
names William Douglas as the advisor believing that it was unwise to attack England (II:472-73).
26Chronique Normande, p. 87; Chronicon de Lanercost, pp. 344-45; Jean Froissart, V:120, 123;
Jean le Bel, II:126; Thomas Samson in Froissart, V:489-90, 492; and Récits d'un bourgeois de
Valenciennes, p. 241. The Brut (II:543) does not number the Scots, but does indicate that they
outnumbered the English 3 to 1. See also Chronographia regum Francorum, II:242 and
Chronique Normande, p. 87. The Récits d'un bourgeois de Valenciennes also claims that the
Scottish numbers include archers (p. 241); no other source mentions archers among the Scottish
ranks.
27 Henry Knighton, II: 42. See also Robert of Avesbury, p. 376; Adam Murimuth, p. 218; John of
Reading, pp. 102-03; the Anonymous of Canterbury, p. 192; and Andrew of Wyntoun, II:472.
28 Giovanni Villani (ed. Ridotta), VII:186.
Page 181
Scots had gained little ground before they encountered a sizeable enemy army. But
who was responsible for such a reaction to the invasion? Contemporary chroniclers
seem somewhat uncertain. Most credit Archbishop William La Zouche of York with
raising the alarm and also the defending troops. As well, these all name Henry Percy
and Ralph Neville as instrumental in the battle.29 On the other hand, Jean Froissart,
Jean le Bel, and the Bourgeois of Valenciennes give their patron, Philippa, Edward's
queen, credit for the quick initiative against the Scots, claiming that she went so far as
to travel to Newcastle to urge on the English troops; then she even traveled with the
army to the battlefield itself. All came to fight the Scots, Jean Froissart writes, 'out of
love for their good queen, their lady.'30
The English army was not as large as its opponent's, perhaps numbering no more than
8,000. But it did include cavalry, infantry, and archers; some had also fought at
Crécy.31 Morale was high, and those preparing to fight were devoted to the defense of
their kingdom, with Henry Knighton noting that 'all unanimously agreed to be
prepared to give life or death for the salvation of the realm.'32 They also hoped by
swift action to save the town of Durham from attack.
The campaign of the Scots before Neville's Cross was quick and largely unopposed.
Only at Liddel did an English force try to resist their invasion, and this conflict, really
a siege of the town, is recorded by Thomas Samson, the Chronicon de Lanercost,
Robert of Avesbury, Geoffrey le Baker, the Anonimalle Chronicle, Andrew of
Wyntoun, and an anonymous poet. They claim that it was shortlasting only five or six
days and brought to an end by siege machinesand relatively bloodlessalthough it did
cost the life of Walter of Selby, 'a warrior of great honor,'
29 See Thomas Samson in Froissart, V:490; Chronicon de Lanercost, p. 347; Robert of
Avesbury, p. 376; Henry Knighton, II:42; Geoffrey le Baker, p. 87; Adam Murimuth, p. 218;
Thomas of Burton, II:61; Laurence Minot, p. 31; the Anonymous of Canterbury, p. 192;
Anonimalle Chronicle, p. 24; Chronographia regum Francorum, II:242; and Andrew of
Wyntoun, II:474.
30 Jean Froissart, V:122, 126-27. Froissart's first redaction mentions Philippa's presence on the
battlefield until asked to leave for fear of her capture. Reluctantly she returned to Newcastle. See
also Froissart, V:119-31; Jean le Bel, II:126; and Récits d'un bourgeois de Valenciennes, p. 241.
31 Several sources record the English numbers at Neville's Cross. Giovanni Villani ((ed.
Ridotta), VII:186-87) counts only a total of 1,000 soldiers, while Jean Froissart in his second
redaction numbers 8,000 soldiers (V:124). In his first redaction, Froissart divides and increases
his second redaction's total: 1,200 cavalry, 3,000 archers, and 5,000 infantry (V:126). The
Chronique Normande (p. 87) and Andrew of Wyntoun's (II:472) total of 1,400 men-at-arms and
20,000 archers should be disregarded as unsubstantiated by more contemporary chronicles. Also
suspicious is Thomas of Burton's 900 men-at-arms and 9,000 archers (II:61), although these are
almost the same as the 800 men-at-arms and 10,000 archers found in the Anonimalle Chronicle (p.
25). (Such an inordinate ratio of archers to men-at-arms seems unwarranted from the description
of fighting found in all contemporary sources.) Finally, the Récits d'un bourgeois de Valenciennes
claim an outlandish total of 70,000 English soldiers (p. 241). Other authors who mention the
different types of troops include Robert of Avesbury, p. 376 and Jean le Bel, II:126-27. On those
experienced at Crécy see Anonimalle Chronicle, p. 26.
32 Henry Knighton, II:42: 'omnes unanimi assensu parati vivere et mori pro salvatione regni
convenerunt.'
Page 182
and other defenders.33 All other chroniclers mention only the large amount of
destruction caused by the Scots both on their journey to and around Durham.34
But if there had been little resistance before Durham, thus confirming David Bruce
and Philip VI's prediction of all the English soldiers being in France, at Durham the
Scots found a relatively large English army which had traveled south from Newcastle
to stop their invasion. Did this surprise the Scots? Not according to John of Reading,
who holds that the time and place of the battle had been set before the actual
fighting.35 Nor according to Jean Froissart, who contends that the Scots had heard of
the gathering and movement of the English but had refused to believe it.36 Finally, the
author of the Anonimalle Chronicle does not believe that the English presence
surprised the Scots, reporting that prior to the appearance of the defensive force two
monks from Durham Cathedral had tried to persuade the Scottish troops not to
destroy their lands. While not accusing these monks of reporting the English advance,
the author of this work does claim that David Bruce thought that this was a ploy to
draw his army into an ambush by the English force, thus counselling his army to
avoid Durham. But the Scots disregarded their king's advice, claiming that they were
ready to fight.37
Only the Chronicon de Lanercost and the Liber Pluscardensis claim that the English
surprised the Scots at Neville's Cross. The anonymous authors of these works report
that while camping near Durham, the Scots 'being unaware of the approach of the
English,' had sent out a raiding party under the command of William Douglas to
pillage the nearby countryside. Instead, Douglas 'suddenly found himself unawares
almost in the middle of the English army' all ordered and ready for combat. This party
'hastily' returned to the main force and reported the incident.38
The only facet of this account which may be true is that the English were the first to
reach Durham and therefore had the opportunity to select a defensible battlefield site
at Neville's Cross. Although no contemporary author refers to the prudence of
33 Thomas Samson in Froissart, V:490; Chronicon de Lanercost, pp. 345-46; Robert of
Avesbury, p. 376; Geoffrey le Baker, pp. 86-87; Anonimalle Chronicle, pp. 23-24; Thomas of
Burton, II:61; Andrew of Wyntoun, II:472-73; and Poem IV in Laurence Minot, p. 113. On
Walter of Selby see Edward Miller, War in the North: The Anglo-Scottish Wars of the Middle
Ages (Hull, 1960), pp. 13-14. On the devastation of the Scots in north England see Chronicon
de Lanercost, pp. 344-47. Lanercost was one of the priories sacked by the Scots in 1346, and
the monks writing this chronicle appear to have been eyewitnesses to this devastation.
34 Jean Froissart, V:121, 124, 125; Jean le Bel, II:127; Geoffrey le Baker, p. 87; Adam Murimuth,
p. 218; Anonimalle Chronicle, pp. 24-25; The Brut, II:299; Anonymous of Canterbury, p. 192;
Récits d'un bourgeois de Valenciennes, p. 241; Chronographia regum Francorum, II:242;
Chronique Normande, p. 87; and Liber Pluscardensis, I:293.
35 John of Reading, p. 102: 'Assignatisque loco et die belli . . . '
36 Jean Froissart, V: 130. However, this appears only in Froissart's final redaction which was
written c. 1400.
37Anonimalle Chronicle, p. 26. The Chronicon de Lanercost (pp. 348-49) gives a different time,
purpose, and result for the visit of the Durham monks.
38Chronicon de Lanercost, pp. 348-49 and Liber Pluscardensis, I:294: 'Willelmus vero de
Douglas de mane, de appropinquacione Anglorum inscius, cure suo agmine, prout de sero
ordinatum. est, ad depraedandum devastandum patriam se festinando, confestim quasi in medio
exercitus Anglorum ignoranter cecidit. Quod percipientes primitivi equestres inimicos eorum. in
tribus aciebus belli ordinatos, paratos ad pugnam, reversi sunt festinanter ad regem David,
exponentes ei modum.'
Page 183

Battle of Neville's Cross


this choice, maps showing the physical features of the battlefield reveal the existence
of a high ground on which the English lines formed, their backs to the hill on which
Durham is located.39 Thomas of Burton confirms this when he reports that the Scots
'ascended' in their march towards the English.40 According to Jean Froissart and
Andrew of Wyntoun, there was also a hedge or dike which separated the Scots from
the English. To attack these lines, the Scots would have had to pass through this
impediment and charge uphill.41 On this site the English leaders ordered their army as
infantry in three lines.42 (Jean le Bel and one of Jean Froissart's redactions count four
lines.43) Each line was commanded by one baron and one ecclesiastical prelate.
According to Jean Froissart and Thomas of Burton, the archers were placed along the
sides and somewhat in front of the infantry lines.44
39 See, for example, the eighteenth-century map reprinted in Frank Graham, Famous Northern
Battles (Rothbury, 1988), p. 13.
40 Thomas of Burton, II:61.
41 Jean Froissart (V:132) uses the Old French word 'haie' and Andrew of Wyntoun (II:475) uses
the Middle English 'dykis'.
42 For a three line formation see Thomas Samson in Froissart, V:491; Chronicon de Lanercost,
pp. 350-51; Anonimalle Chronicle, pp. 26-27; Thomas of Burton, II:61; Froissart, V:124; Liber
Pluscardensis, I:294; and an anonymous poem, Poem IV, found in Minot, p. 115. The Anonimalle
Chronicle also insists that the Scots ordered in three lines first, with the English responding with
their own three line formation, but no other source corroborates this view. For four lines see Jean
le Bel, II:127 and Jean Froissart, V:126. This is found in the first redaction of Froissart; the
second redaction reduces the number to three lines; and the final redaction does not mention the
number of lines in the English formation.
43 Jean Froissart, V:126 and Jean le Bel, II:127. The first line was commanded by Percy and the
bishop of Durham, the second by Neville and the archbishop of York, the third by Mowbray and
the bishop of Lincoln, and the fourth by Edward Balliol and the archbishop of Canterbury.
44 Jean Froissart, V:124, 130 and Thomas of Burton, II:61.
Page 184
The Scots arrived on the battlefield and also ordered themselves in three lines.45 They
were confident in their numbers which were far greater than those opposing them,
and Giovanni Villani maintains that they wanted to attack the English immediately.46
But, according to Froissart and an anonymous contemporary poet, David Bruce and
William Douglas decided to hold off their attack in order to meet in council and to
speak to their troops.47 The Scottish leaders may also have wanted to take advantage
of the growing fatigue of the English army, which by that time, noon, had stood in its
lines for several hours.48 The English leaders too met in council, but their mood was
decidedly different from the confidence shown by the Scots. According to Jean
Froissart, they considered that the battle would not be fought until the next morning,
and thus they urged attentiveness throughout the night, reckoning that the Scots, 'a
great people,' could cause much damage. The English might even lose if the Scots
'came suddenly upon them.'49 They then 'placed their faith in God' and prepared for
battle.50
Undoubtedly, the English leaders also laid out their tactical plan for the battle,
although by this time it was well known and, at least since 1322, well practiced. They
were to stand in a defensive formation, as solid lines of infantry, awaiting the Scottish
charge. As Robert of Avesbury reports: 'They were to stand in the way of the Scots,
then they were to fight with them strongly.'51 The archers were to fire at the oncoming
troops, provoking, disorienting and slowing their charge.
As it was laid out so it happened. The Scots did or perhaps could not wait for the
night to pass. (Thomas of Burton claims that 500 archers provoked the attack by firing
arrows into the Scots while they waited to attack the English, but he is alone in this
assertion.52) They armed themselves, and whoever had them mounted their horses;53
at three o'clock Scottish horns sounded, signalling to their own troops that
45Chronicon de Lanercost, pp. 349-50; Henry Knighton, II:42; Anonimalle Chronicle, p. 26;
Jean Froissart, V:128; Liber Pluscardensis, I:229; and Andrew of Wyntoun, II:475.
46 Giovanni Villani (ed. Ridotta), VII:187. On Scottish confidence see Froissart, V:126, 128-29,
130.
47 Jean Froissart, V:130 and Poem TV in Laurence Minot, pp. 115-16.
48 Jean Froissart, V:130.
49 Jean Froissart, V:131: 'Li signeur et li prélat se remissent ensamble en consel, et dissent chil
liquel estoient le plus usé d'armes: "Se nous atendons jusques à la nuit, ces Escoçois, qui sont
grans gens, nous poront venir courir sus et porter trop grant damage. Si seroit bon que nous
envoions viers euls jusques à cinq cens lances pour euls atraire hors de lors logeis et que li nostre
se facent cachier, tout au lonch de celle haie, là où nostre archier seront mis et aresté, et se les
Escos viennent soudainement après nos gens, ensi que il sont bien taillet de ce faire (car il sont
chaut, boullant et orguilleus, et tant que pour l'heure il prisent moult petit nostre affaire), nostres
archiers qui sont frès et nouveauls, trairont sus euls et entre euls, et nous aussi, gens d'armes, les
requellerons ensi comme il apertient à faire. Par ce parti porons-nous bien avoir bonne aventure,
et se il se voellent tenir là où il sont, il donront à entendre que il nous vodront venir courir sus de
nuit; mais nous nos départirons avant et nous retrairons dedens le Noef-Chastiel, car pas ne nous
seroit proufitable à chi atendre et logier le nuit." '
50 Henry Knighton, II:42 and Anonimalle Chronicle, p. 26,
51 Robert of Avesbury, p. 377: ' . . . obviam sibi dantes, cum eodem fortiter proeliarunt.'
52 Thomas of Burton, II:61.
53 Jean Froissart, V:131.
Page 185
a charge was to commence, but also warning the English of that same charge;54 and,
with a war cry, they moved, 'fortified in their number by their polished helmets and
shields,' towards the infantry defensive formation.55 A 'grave and large' battle
followed.56
It was 'very cruel and very marvelous,' writes the Bourgeois of Valenciennes, and Jean
le Bel compares it to that fought by Roland and Oliver at Roncevalles.57 As the
Scottish troops passed through the hedge and progressed up the incline towards the
English army, they first encountered the arrows of the longbowmen. Jean Froissart
describes the action:
And when the Scots came through the hedge, the English archers commenced to fire very
strongly and at a great distance. And they impaled men and horses and caused a great
disruption.
Later he adds that the archers 'gave a great comfort to their men-at-arms and great pain
to the Scots.'58
Still, the 'great pain' given by the archers neither stopped the charge nor, at least
according to Geoffrey le Baker, did it seem to have been as effective at disordering it
as Froissart claims. Baker uses the verb 'frustrare' to describe this ineffectiveness: 'The
Scots . . . frustrated the English archers at the beginning of the battle.'59 What this
means is difficult to determine from this account alone. However, a solution may be
found in Thomas Samson's letter, as he writes that the charge of the Scots caused the
archers and non-professional militia to retreat twice from the battlefield, 'but our men-
at-arms fought well and lasted until the archers and 'communes' reassembled.'60
Geoffrey le Baker may be saying the same thing when he adds that 'the first line of
[English] soldiers greeted the enemy with deadly spears.'61
The Scottish soldiers, as a dense formation itself, hit the English lines with power and
strength; both sides fought wellas 'rabid lions' records one contemporary
54 Henry Knighton, II:42; Jean Froissart, V:129; and Poem IV in Laurence Minot, p. 117. The
time of the charge is recorded by Jean le Bel, II:127.
55 The quote is from Geoffrey le Baker, p. 88: ' . . . cassidibus politis et umbonibus numero
firmatis . . . ' On the Scottish war cry, see Jean Froissart, V:131. See also Henry Knighton, II:42;
Anonimalle Chronicle, p. 27; Thomas of Burton, II:61; and Jean Froissart, V:131. The Liber
Pluscardensis (I:294) contends that the English were the first to attack at Neville's Cross, but none
of the more contemporary writers corroborate this statement.
56 Jean Froissart, V:132.
57Récits d'un bourgeois de Valenciennes, p. 241 and Jean le Bel, II:127-28. See also Adam
Murimuth, p. 218 and Jean Froissart, V:124, 127.
58 Jean Froissart, V:132: 'et quant les Escos furent venu à celle haie, les archiers englois
commenchièrent à traire moult fort et moult roit et à enpaller hommes et cervaus et à mettre à grant
meschief . . . et donnèrent moult grant confort as gens d'armes de lor costé et grant painne as
Escoçois.' This description exists only in Froissart's final redaction. The account of his first
redaction is much shorter and does not describe any damage done by the archers (V:126-27). See
also Poem IV in Laurence Minot, pp. 115, 117.
59 Geoffrey le Baker, p. 88: 'Nacio Scotia . . . sagittas Anglicorum in primordio belli frustravit.'
60 Thomas Samson in Froissart, V:490-91: 'Deux foits se retraièrent les archers et comunes de
nostre part; mais nos gents d'armes se combatièrent et se continuèrent durment bien tant qe les
archers et communes reassemblèrent.' My thanks to Cliff Rogers for pointing me to this
interpretation.
61 Geoffrey le Baker, p. 88: 'Set armatorum acies prima ictubus letalibus hostes salutavit.'
Page 186
commentatorwith many displays of bravery.62 The Anonimalle Chronicle is almost
poetic in its description: 'In this conflict lances were broken, swords destroyed, armor
pierced, helmets and bascinets smashed, and shields split.'63 Although Thomas of
Burton claims that the English first line was initially pushed back, ultimately, the
defending force held against the Scottish onslaught, even, according to the
Anonimalle Chronicle, through a second and third charge.64 As the number of dead
rose, the Scots began to lose heart.65 Some of their nobles fled, and soon so did many
of the others.66 At the end, Geoffrey le Baker reports, only forty nobles remained,
standing in a circle around their king and prepared to fight to the death.67 David
Bruce, who was wounded at least once in the face by an arrow during the battle, chose
not to be killed.68 The English victory was greeted from the Cathedral tower by
prayers and shouts from monks who had watched the battle; their exhilaration was
heard by the troops below.69 When news of the victory reached Calais, the English
forces there echoed the Durham monks' cheers.70
Jean Froissart sums up the battle of Neville's Cross:
That day, as in that conflict which the king of England and his troops fought at the battle of
Crécy at which only fifty thousand men had defeated one hundred
62 The quotation is from Poem IV in Laurence Minot, p. 117. See also Henry Knighton, II:42;
Geoffrey le Baker, p. 88; Thomas of Burton, II:61-62; Jean Froissart, V:124, 127, 132; and
Andrew of Wyntoun, II:476.
63Anonimalle Chronicle, p. 27: 'En quele conflykte launces furrent frussez, espeis debrisez,
haubergeouns desmaillez, helmes et bacynetz availlez et escues desquarterez.'
64 Thomas of Burton, II:61-62 and Anonimalle Chronicle, p. 27. Andrew of Wyntoun reports two
Scottish attacks (II:475-76).
65 Only five sources record the number of Scots killed with the latter four obvious exaggerations:
Chronique Normande, p. 87 - 2,000; Adam Murimuth, p. 218 - 12,000; Thomas Samson in
Froissart, V:491 - 540 knights and 12,000 others; Henry Knighton, II:43 - 100 knights and 20,000
others; and Récits d'un bourgeois de Valenciennes, p. 241 - 40,000. See also Robert of Avesbury,
p. 377; Geoffrey le Baker, pp. 88-89; Anonimalle Chronicle, p. 27; Thomas of Burton, II:62; John
of Reading, p. 103; The Brut, II:299; the Anonymous of Canterbury, p. 192; Jean Froissart, V:125,
127; Chronographia regum Francorum, II:242; Chronique Normande, p. 87; Poem IV in
Laurence Minot, p. 118; Liber Pluscardensis, I:294-95; and Andrew of Wyntoun, II:476-77.
66Chronicon de Lanercost, p. 351; Henry Knighton, II:43; Geoffrey le Baker, p. 88; Anonimalle
Chronicle, p. 27; The Brut, II:543; Poem IV in Laurence Minot, p. 119; and John Fordun, I:367.
67 Geoffrey le Baker, p. 88.
68 That David was wounded see Thomas Samson in Froissart, V:491; Henry Knighton, II:44;
Adam Murimuth, p. 219; Thomas of Burton, II:62; Anonymous of Canterbury, p. 192; The Brut,
II:543; Chronique Normande, p. 87; and Liber Pluscardensis, I:294. (The Chronique Normande
claims that David Bruce was wounded twice in the face with arrows, with one arrowhead
remaining in his face for four years and the other for nine years before being removed.) On the
Scots captured see Thomas Samson in Froissart, V:491-92; Robert of Avesbury, p. 377; Henry
Knighton, II:43-44; Geoffrey le Baker, p. 88; Adam Murimuth, pp. 218-19; Anonimalle Chronicle,
p. 28; John of Reading, p. 103; The Brut, II:299-300; Anonymous of Canterbury, p. 192; Poem III
in Laurence Minot, pp. 110-12; Poem IV in Minot, p. 119; Jean Froissart, V:125, 128, 133; Jean le
Bel, II:128; Récits d'un bourgeois de Valenciennes, p. 241; Chronographia regum Francorum,
II:242; John Fordun, I:358; and Andrew of Wyntoun, II:476-77. A list of the Scottish leaders
captured at Neville's Cross and interred in the Tower of London is found in Rymer, Foedera,
III.1:95.
69 Henry Knighton, II:42-43.
70Anonimalle Chronicle, p. 28 and Jean le Bel, II:129-30.
Page 187
thousand, equally in the battle which I have just presented to you, a small force of English were
met by the Scots and defeated their enemy.71
Thus Froissart made the comparison between the two 1346 English victories. The
Hainaulter chronicler, never one to let subtlety be his guide, was perhaps claiming too
much, especially when we consider the disparity in sizes of the English and opposing
armies, the leadership differences, and the purposes for fightingone defensive and the
other offensive.
But there are several similarities apparent between the two battles. Most of these deal
with the tactics employed by the English on the field. The English chose the
battlefield: at Neville's Cross the choice of a high, defensible ground was significant in
the victory; the hedge through which the Scots were forced to pass also must have
caused some confusion and disruption among those charging towards the English
lines. Their leaders chose to dismount the cavalry: indeed, in none of the chronicles is
cavalry even mentioned outside of the categorization of them in an initial tally of
soldiers fighting for the English. The infantry was ordered in a formation of three or
four solid lines: these were to take on the charges of the Scottish troops, and it was
their ability to hold against these attacks that ultimately brought victory. And the
archers were set up along the sides of the infantry lines to provide disrupting
firepower: this narrowed the Scottish charge so that it struck the strength of the
English defensive formation, the infantry lines.
The archers proved very capable in narrowing the charge of the Scottish force, as no
contemporary commentator reports any flanking maneuver attempted by those
attacking the English line. However, unless Froissart's account is accepted over those
of the more contemporary Geoffrey le Baker and Thomas Samson, the archers were
unsuccessful in disordering the charge of the Scots so that it fell onto the English
infantry lines in an almost completely intact order.72 Yet that the archers were
unsuccessful in confusing and disrupting the charge of the Scots as they had done
against the French at Crécy and against the Scots at Halidon Hill proved of little
significance to the outcome of the battle. The English infantry at Neville's Cross
simply had to fight as the Flemings, Bernese, Catalan Company, Frisians, Liégeois,
and even Scots had since the beginning of the fourteenth century, with little or no
disordering confusion caused by archery. They found, as those other armies did in
their victories, that the role of anyone, including archers, was secondary to that of the
infantry soldier standing solidly in line against the attack of his foes. If the infantry
lines held, the charge could be, and at Neville's Cross was, defeated.
71 Jean Froissart, V:132: 'Che jour, ensi que de la belle aventure que li rois d'Engleterre et ses
gens orent de la bataille de Créchi et que euls quinze mille hommes en tout en desconfirent cent
mille, parellement en la bataille dont je vous parole présentement, un petit de gens que les
Englois estoient ou regart des Escos, desconfirent lors ennemis.'
72 Even Sir Charles Oman ((1905) II:150) must reluctantly admit that 'the matter was not entirely
settled by archery, as the masses of spearmen were seriously engaged for some time at close
quarters.' But Jonathan Sumption accepts Froissart's account without considering what Geoffrey le
Baker has to say (p. 553). Finally, Nicholson believes that it was the English archery which
caused the Scottish army to attack, although they 'wished to remain on the defensive' (Scotland, p.
147), but I can find nowhere in the original sources when this is suggested.
Page 188

Appendix
Three Infantry Ambushes:
The Battles of Morgarten, 1315, Auberoche, 1345, and La Roche-
Derrien, 1347
By only discussing those battles which began as 'two armies entirely assembled and
ordered against each other', three early fourteenth-century infantry victories by
definition have been excluded. The battle of Morgarten, where the Swiss defeated the
Austrians in 1315, and the battles of Auberoche and La Roche-Derrien, where the
English defeated the French in 1345 and 1347 respectively, all were fought by one
army, in each case the victorious one, surprising their opponents, and by using this
surprise, defeating them.
The battle of Morgarten followed the invasion of the rebellious Switzerland by Duke
Leopold of Austria.1 The Austrian army, composed largely of knightly cavalry and
numbering between 2,000 and 3,000, moved through Switzerland with what seemed to
them to be little difficulty. But in reality, by blocking more accessible passes, the Swiss
had directed the march of the Austrians over precipitous terrain and through
dangerous passes. In one of these, the pass of Morgarten, along the east bank of the
Aegeri Lake, a force of between 3,000 and 4,000 peasants lay in ambush.2
The narrowness of the pass road required the Austrians to reduce the size of their
cavalry column, but even after doing so, the vanguard of the column became
compressed and congested. At this point, the Swiss moved from their hiding spots,
high on the steep slopes of the pass, and sent large stones and logs down on the
surprised Austrians. Smaller stones and arrows also were showered onto the cavalry
column. This completely confused the Austrians, who tried vainly to get out of the
way of these disruptive devices. However, before they could regain their order, the
Swiss infantry charged down the hill into the confused column. Using their famed
1 Contemporary and near contemporary sources on the battle of Morgarten can be found in the
Chronicon Aulae Regiae, Rudolf von Radegg's Capella heremitana, Johann von Bictring's
Chronik, Jean de Winterthur's Chronicon, the Chronik of Mathias von Neuenberg, the
Oesterreichische Chronik, Jakob Twinger von Königshofen's Chronik, Konrad Justinger's
Berner Chronik, the Bürcher Chronik von 1428, and numerous other smaller historical works.
All of these original sources are edited in Theodor von Liebenau, 'Berichte über die Schlacht
am Morgarten,' Mitteilungen des historischen Vereins des Kantons Schwyz 3 (1884), 1-85.
Modern accounts of Morgarten can be found in Delbrück, pp. 551-60; Oman (1905), II:238-41;
Wernli, pp. 277-94; McCrackan, pp. 123-28; Bruno Meyer, 'Die Schlacht am Morgarten:
Verlauf der Schlacht und Absichten der Parteien,' Revue Suisse d'histoire 16 (1966), 129-79;
and Robert Durrer, Schweizer Kriegsgeschichte, ed. M. Feldmann and H.H.G. Wirz (Bern,
1915), I:74-90.
2 These numbers and those of the Austrians are from Hans Delbrück, pp. 553-54; they are much
smaller than those given in the original sources.
Page 189
'halberds', the Swiss easily cut down most of the knights from their horses, killing
them as they lay helpless on the ground. Those who could, and they were few, fled
out of the pass. Soon they encountered the rear-guard of the Austrian army, causing
them also to flee. Deaths were numerous, with Duke Leopold himself barely able to
flee.
The battle of Auberoche was fought on October 21, 1345 during the earl of Derby's
invasion of the Périgord region of Gascony.3 Derby, an experienced leader who had
been with the English army in Flanders in 1340, was sent to Gascony in June 1345 by
Edward III to defend those Gascon nobles who had remained loyal to the English
throne and thus were threatened by French incursions into their territory. He may also
have been sent there to keep Philip VI preoccupied while the main English force
prepared to attack Normandy.
The earl of Derby moved first against the French-controlled town of Bergerac which,
after a concerted defense by Bertrand, the count de l'Isle, fell to the English leader.
From there he proceeded north to Périgueux, the capital of the region, but found it too
strongly defended for his small army to attack. Thus he moved east to the castle and
village of Auberoche. The French army, made aware of the English progression by,
among others, the count de l'Isle, who had been able to escape from Bergerac,
gathered at La Réole and marched quickly to Auberoche. However, their approach
was seen by the English, who numbered no more than 1,200, and they were able to
successfully hide in nearby woods, The French army, numbering between 7,000 and
10,000, but not realizing the proximity of the English, camped very close to them.4
Derby, knowing that his army's supplies would not last long, decided that they should
break out of their hiding place and attack the French camp during the evening meal.
The timing of the English attack was perfect; with trumpets blaring and banners
unfurled, the English troops, under the cover of a constant archery barrage, charged
from the woods and reached the French camp, some 200-300 yards away, without
opposition. The French troops were completely surprised and utterly confused by the
English onslaught. Few were able even to grab their weapons let alone don their
armor. Casualties were extremely high, as both arrows and swords easily cut down the
unprotected French soldiers. A few French leaders were able to regroup, and they
attempted to restore some defensive order. But eventually they could do nothing but
retreat from the battlefield, saving their own lives. For the most part, the French army
in Gascony lay dead among their tents.
Nearly two years later, on June 20, 1347, the battle of La Roche-Derrien was fought in
Brittany between two secondary armies of the French and English, the main
3 Original sources on the battle of Auberoche include: Adam Murimuth, pp. 189-90; Henry
Knighton, II:31-32; Robert of Avesbury, pp. 356-57; Anonimalle Chronicle, p. 18; Jean
Froissart, IV:252-73; Grandes chroniques, IX:258-59; Chronique Normande, pp. 65-66;
Récits d'un bourgeois de Valenciennes, pp. 194-95; Villani (ed. Muratori), col. 927; and
Chronicon Vazatense, in Archives historiques du département de la Gironde, xv (Bordeaux,
1874), pp. 43-44. Secondary sources which recount the battle include: Burne, Crecy War, pp.
107-14 and Sumption, pp. 468-69.
4 These numbers are from Burne, Crecy War, pp. 107-08.
Page 190
forces of both kingdoms gathered at the siege of Calais.5 The leader of the French
force was Charles of Blois, the count of Brittany, and the leader of the English army
was Sir Thomas Dagworth. Both leaders were very experienced in warfare. In fact,
they had faced each other almost constantly since May 1345 when Dagworth first
landed with the earl of Northampton's army in France; Thomas Dagworth had also led
the English force at the small battle of St. Pol de Léon, fought on June 9, 1346, where
Charles of Blois had been defeated.
By spring 1347 Charles was ready to fight again. Hoping to avenge the recent defeats
of the French, including that at Crécy, and perhaps also hoping to raise the siege of
Calais by defeating an English force in Brittany, Charles gathered an army and in May
besieged the English-controlled town of La Roche-Derrien. Charles was patient at La
Roche-Derrien. He had time to take the town by hunger and thus did not press for an
attack on its walls. At the same time, he undoubtedly wanted to bring Dagworth,
whose army was much smaller than the count of Blois', to battlethe English force had
no more than 1,000 soldiers, including 300 men-at-arms and 400 archers.6
Sir Thomas Dagworth delayed his relief attack until June. Perhaps this delay lulled
Blois into believing that no relief force would come to La Roche-Derrien. If not, he
had unwisely determined that Dagworth's approach would come only along the
western bank of the Jaudi River. Dagworth came instead along the more dangerous
east bank, which necessitated a night march through woods and harsh country. Just
before dawn on June 20, Dagworth's army charged into the sleeping French camp.
The attack was completely unexpectedsentries had not even been postedand in the
dark the French soldiers were easily cut down by the English who knew each other
only by a secret password. Some French did recover from the surprise and were able
to make a couple of counter-attacks, one of which wounded Dagworth, but ultimately
the English gained the field. Charles of Blois, unable to don his armor, was also
wounded during the fight; taken by Dagworth's troops, he was sent to England and
imprisoned with David Bruce in the Tower of London. Large numbers of French
soldiers had been killed and many more captured, but part of the French army, in
itself larger than the English force, had remained inactive during the battle, besieging
the town on the opposite side of the river from the main French camp. Demoralized
by the victory over their general, they left the siege without further combat.
5 Original sources on the battle of La Roche-Derrien include: Henry Knighton, II:48-49; Robert
of Avesbury, pp. 388-90; John of Reading, p. 104; Jean le Bel, II:145-49; Jean Froissart,
V:164-77; Grandes chroniques, IX:298-309; Chronique Normande, pp. 90-91; and Récits d'un
bourgeois de Valenciennes, pp. 254-56. There also exists a letter from Thomas Dagworth to
the chancellor of England which discusses the battle of La Roche-Derrien. It is recorded in
Robert of Avesbury, pp. 388-90 and in Froissart, XVIII:299-300. Secondary sources on the
battle include: Oman (1905), II:254-55; Burne, Crecy War, pp. 89-99; and Sumption, pp. 572-
75.
6 The numbers are Burne's (Crecy War, pp. 91-92). See also pp. 96-97.
Page 191

Conclusions
There is but one main conclusion to this study: infantry armies fought battles during
the early fourteenth century using distinctive, and in most cases, decisive tactics. This
occurred not just in Flanders or in France or in Scotland, as some historians have
insisted, but throughout Europe. The number of battles fought in such a short period
of the Middle Ages is a historical fact. The constancy of infantry victories may
surprise some, but no less surprising should be the uniformity in tactics used by all
these infantry armies throughout such a large geographical region.
To begin to explain the constancy and uniformity of infantry tactics during the early
fourteenth century, we must start by characterizing the troops which won and lost
these battles. While each had different social backgrounds, military experience, and
motives for fighting, the only consistent characteristic of those armies described here
as infantry armies is that they primarily, and in most cases entirely, fought on foot. The
Flemings at Courtrai, Arques, Mons-en-Pévèle, and Cassel, the Scots at Loudon Hill
and Bannockburn, the Catalan Company at Kephissos, Andrew Harclay's force at
Boroughbridge, the disinherited at Dupplin Moor, the Bernese at Laupen, the Frisians
at Staveren, and the Liégeois at Vottem all fought using, as far as the original sources
are concerned, only infantry soldiers. Robert Bruce even went so far, according to
John of Trokelowe and Thomas Walsingham, to insist that his army at Bannockburn
only fight as infantry, so that they would not suffer a similar defeat as that at the battle
of Falkirk.
When cavalry also are reported to be present, as at Courtrai and at Kephissos, they are
said to be dismounted and fighting as infantry. So too were the English troops at the
battles of Halidon Hill, Morlaix, Crécy, and Neville's Cross, induced to dismount the
cavalry and order as infantry after suffering defeat against the Scots at Bannock-burn,
or, if we are to agree with T.F. Tout and J.E. Morris, influenced by Harclay's victorious
'Scottish' infantry tactics at Boroughbridge.1 Whatever the initial cause of their
dismounting, it was, as Geoffrey le Baker claims in his description of the battle of
Halidon Hill, 'against the ancient tradition of their fathers.'2 Soon, however, that
ancient tradition was completely supplanted by the new, victorious tactics of the
English army.
The troops these armies faced in battle were often, although not in every case,
cavalry-based forces. That is not to say that each of these armies was exclusively
cavalry; on the contrary, in the battles discussed here the cavalry did not stand alone,
but was joined by non-cavalry soldiers: for example, infantry, as with the Freiburger
cavalry at Laupen; archers, as with the French at Crécy; or infantry and archers, as
1 Tout, 'Tactics,' 711-13 and Morris, 'Mounted infantry,' 86-91.
2 Geoffrey le Baker, p. 51: 'Ibi didicit a Scotis Anglorum generositas dextrarios reservare
venacioni fugencium, et, contra antiquatum morem suorum patrum, pedes pugnare.'
Page 192
with the French at Courtrai or the English at Bannockburn. But it was, in almost every
battle, the cavalry of these armies charging against infantry opponents which led
directly to their defeat. Only at Dupplin Moor, Halidon Hill, and Neville's Cross was it
an infantry charge against infantry opponents which was the climactic point of the
battle, and in each of these conflicts the defensive infantry formation was able to
withstand the attacking infantry troops. As well, both at Courtrai and Laupen infantry
attacks were resisted before receiving cavalry charges. Finally, at Mons-en-Pévèle and
Cassel, the only indisputable defeats suffered by infantry armies studied here, it was
the attacks made by the defending troops which ultimately led to their defeats. At
Mons-en-Pévèle, this came late in the battle, after the Flemings had successfully
withstood many French cavalry charges, even reaching the French camp and
threatening Philip IV before being turned back. At Cassel, no cavalry charges were
made against the position of the Flemish force on top of the hill; instead, Philip VI
harassed the Flemings into a charge down the hill by artillery fire and by burning the
countryside around the battlefield. This attack also reached the French camp and
threatened the king before a counter-attack stopped the Flemings, killing those who
did not take flight.
Another characteristic of the troops which is similar in most of the battles described
here is that these infantry armies were usually smaller in number than their opponents.
Only at Arques, Mons-en-Pévéle, Staveren, and Vottem are the defending troops
reported to be larger in number than those attacking them, and only at Staveren and
Vottem does this size advantage aid in achieving victory. At Mons-en-Pévèle, the
larger Flemish numbers could not help them from being defeated. And at Arques,
while their larger numbers may have kept this later Flemish army from defeat, they
lost far more soldiers than their opponents in the battle and thus were forced to retreat
from the field without being able to take the town of St. Omer, their motive for
fighting the battle.
The next point in understanding the constancy and uniformity of infantry tactics in the
battles of the early fourteenth century is to identify what was done by these armies to
prepare to fight these battles. In all but four battles, Arques, Mons-en-Pévèle, Dupplin
Moor, and Halidon Hill, the infantry armies chose the site of the battle. At Courtrai, the
battlefield was chosen because of its proximity to the Lys River, which not only filled
agricultural ditches cutting across the field but also served as a non-penetrable shield
for the Flemish lines through which no one could attack and no one could flee. At
Loudon Hill, Robert Bruce found a flat, dry plain flanked on one side by a deep and
wide morass on which to order his troops, and at Bannockburn he gathered his army
at New Park, also a flat plain, but this time bordered on three sides by impenetrable
trees and marshes. At Kephissos, the Catalan Company placed themselves on a dry
plain surrounded by marsh and river. At Boroughbridge, Andrew Harclay ordered his
troops to block the bridge and ford across the Ouse River. Had Thomas of Lancaster
tried to move to another river crossing he would have encountered the main English
army. At Cassel, the Flemings ordered their soldiers on the top of a hill. Similarly, at
Laupen, the Bernese were allowed to march past the Freiburgers besieging the town
and place themselves on high ground nearby. At Morlaix, the English ordered
themselves in front of a wood.
Page 193
At Staveren and at Vottem, the Frisians and Liégeois chose to fight outside of the
largest town of the respective region, using the town itself as a protection against
attack from the rear. At Crécy, Edward III ordered his troops along a small ridge north
of which stood the Bois de Crécy. And at Neville's Cross, the English positioned their
lines on a high ground outside of Durham.
But choosing a good battlefield site was not the only important aspect of pre-battle
preparation for these armies. Arriving early on the battlefield gave them on several
occasions the opportunity further to ready the site for conflict. The most frequent
means of doing this was by digging ditches and pits, either along the flanks or in front
of the soldiers. Although these seem to have been a feature of battlefields since at least
the early Middle Ages, never were ditches more effective than during this period. In
all cases, they either disrupted a cavalry charge or narrowed its length preventing any
flanking of a defending line. They were constructed at Courtrai, where most were
filled with water and some were covered over with dirt and branches; even though
their position was known to the French they would prove to be disruptive to their
charge. They were also dug at Loudon Hill, where Robert Bruce had three large and
deep trenches made across the field leaving only small gaps for the English horses to
charge through. At Bannockburn, ditches and pits were dug in front of the Scottish
lines, a tactic which obviously impressed the English, as Geoffrey le Baker insists that
Edward III did the exact same thing at Crécy. (The muddiness of the battlefield of
Crécy resulting from the pre-battle rainfall must have made these especially
treacherous, although no contemporary writer confirms this.) And at Morlaix, the earl
of Northampton dug pits and ditches both on the flanks and in front of the infantry
which, when combined with the woods behind them, created what Henry Knighton
described as a 'narrow cave'.3 At Staveren and Vottem ditches were dug by the
infantry armies, but not for the purpose of disrupting or narrowing cavalry charges.
Instead, they were constructed as trenches for the defending army to hide in. In
neither case, however, did these troops remain in their trenches.
Another battlefield preparation was the construction of wagon fortresses at Mons-en-
Pévèle and Crécy. As described by the Annales Gandenses, at the battle of Mons-en-
Pévèle this fortress was built by circling the wagons and carts, removing one wheel
from each, and attaching them all together. The baggage train and horses were placed
inside to protect them, but the fortifications also served to guard the rear of the troops.
As well, if we are to believe the word of Jean de Paris, the Flemings built their wagon
fortress at Mons-en-Pévèle so that if needed they might be able to retreat there 'to
make a stand and fight inside such a circle.'4 No similar purpose for the fortress at
Crécy is mentioned.
Finally, at Kephissos, the site of battle was prepared for fighting when the Catalan
Company cut ditches into the banks of the river and flooded the already marshy field.
In many of these early fourteenth-century battles the selection of the site and the
3 Henry Knighton, II:25.
4 Jean de Paris, p. 643: 'Flandrenses autem astute fecerant quasi munitionem de suis curribus et
quadrigis, magnum terrae spatium hujus modi vehiculis circumdantes, ut infra talem circuitum
possent in necessitate consistere et pugnare.'
Page 194
preparing of the field meant that opponents had only one course of attack: a frontal
assault. At Loudon Hill, Kephissos, Boroughbridge, Morlaix, and Crécy, the charging
troops had no means of attacking from the flanks or the rear. The same was true for
the second day's battle at Bannockburn. At Mons-en-Pévèle, a French flank attack was
stopped by the Flemish wagon fortress. And at Courtrai, only the French garrison in
the castle could attack the Flemish troops from the rear. But the Flemings had planned
for such an action, and when it did occur, late in the battle, it was quickly countered
by a contingent placed there to stop such an attack. At other battles during the early
fourteenth century, there seems to have been the ability of the charging troops to
attack from the rear and flanks. However, in these conflicts either the attacking forces
chose, for several different reasons, only to make frontal assaultsat Dupplin Moor,
Halidon Hill, Staveren, Vottem, Neville's Cross, and the first two phases of the battle
of Arquesor the defenders altered their tactics and thus compensated for any flanking
and rear attacksat Laupen, the first day's battle at Bannockburn, and the final phase of
the battle of Arques.
The English in these battles had in the longbow a technological advantage which no
one else had, and they used this weapon not as 'a killing machine', as some historians
have contended, but instead to narrow their opponents' charges and to protect against
flank attacks. Indeed, using their archers as the English did, ordered along and
protruding out from the flanks of the infantry lines, as shown at the battles of Halidon
Hill, Crécy, Neville's Cross, and to a lesser extent at Bannockburn, Boroughbridge,
and Dupplin Moor, there was little need for the longbowmen to kill many of their
opponents. Their purpose was simply to narrow and confuse the attacker's charge so
that when it fell onto the infantry troops, it did so in a disrupted and relatively
impotent manner. (The English cannons at Crécy seem to have been used in the same
way.) The perfect staging of this may have occurred at Halidon Hill, where Thomas of
Burton describes what happened to the Scottish charge after encountering the English
longbow fire: 'The smaller squadrons, so cut by the archers, were forced to cling to
the larger army, and in a short time, the Scots massing together were pressed one into
the other.'5 And at Crécy, Jean le Bel notes:
On their side the archers fired so skillfully that some of those on horses, feeling the barbed
arrows, did not wish to advance, while others charged forward as planned; some resisted them
tirelessly, while others turned their backs on the enemy.6
Any deaths of men and horses which occurred would obviously have added to the
disruption of the charge. Finally, if the archers found themselves in range of the
opposing troops before the attack, their fire could hasten a charge which was being
5 See Thomas of Burton, II:364: 'Minores vero turmae, per sagittarios nimium laceratae,
adhaerere magno exercitui compelluntur, et in brevi Scotti conglobati alius ab alio premebatur.'
6 Jean le Bel, II:103: 'Et d'aultre part les archiers tiroient si merveilleusement que ceulx à cheval,
sentans ces flesches barbelées [qui] faisoient merveilles, l'ung ne vouloit avant aler, l'aultre
sailloit contremont si comme arragié, l'aultre regimboit hydeusement, l'aultre retournoit le cul par
devers les anemis.'
Page 195
formed, again effectively disrupting it; this occurred, if we are to trust Thomas of
Burton, at Neville's Cross.7 The longbowmen also proved themselves very effective
against the attack of the Genoese crossbowmen at Crécy, completely defeating their
archery counterparts before the Genoese could fire even one shot in reply.
After all pre-battle preparations had been undertaken and as the battle was about to
commence, the infantry soldiers were ordered in their defensive formation. This was
undoubtedly the most important tactic of every early fourteenth-century battle studied
here. The infantry and dismounted cavalry were to form one or more solid lines.
Sometimes there was a rearguard, and in other battles only a single solid line was
ordered. Most often the line was straight and thick, while at the ford at Boroughbridge
and at Halidon Hill it was slightly curved like an arc. At Courtrai, Mons-en-Pévèle,
Loudon Hill, Kephissos, the second day of Bannockburn, Boroughbridge, Cassel,
Morlaix, and Crécy, the infantry line was formed to accept and endure at least one and
perhaps more than one frontal cavalry charge. (The Frisians at Staveren and the
Liégeois at Vottem would also form a solid infantry lines, but these were ordered only
after seeing that the cavalry charges being made against the trenches where these
troops were hidden had become confused and disrupted in their attack.) At Dupplin
Moor, Halidon Hill, and Neville's Cross, it was to be an infantry charge.
At the third phase of Arques, the first day of Bannockburn, and Laupen, the charges
were made by cavalry, or, as at Laupen, by both cavalry and infantry, but they differed
from other early fourteenth-century battles as these charges came from all sides. In
these conflicts a different type of infantry line was ordered. At Arques, the Annales
Gandenses describes the formation as 'a circle like a bowl or a crown.'8 At
Bannockburn, John Barbour writes that the Scots set up 'back to back with their spear
points outward.'9 And at Laupen, the Conflictus Laupensis depicts the formation as a
'small wedge'.10 All authors report basically the same formation, a solid mass of
soldiers who faced out in all directions, so that there was no front or rear and no
flanks. Sometimes called a 'hedgehog', this infantry formation had the same function
and, at least in these battles, the same result as the infantry line which faced only one
direction.
The solid defensive infantry line was not only the most important tactic used by
infantry armies in these early fourteenth-century battles, but it may also have been the
most difficult tactic to achieve. While failing in only two battles, at Mons-en-Pévèle
and Cassel where the infantry forces forgot their defensive purpose and attempted
offensive attacks, the ability to form and then to keep an infantry line well ordered
seems to have been a difficult task and certainly denotes some adept generalship. The
tactic itself may not have been very intricate or elaborate, but the discipline needed to
make it successful must have taxed both leaders and men, especially in the case when
the soldiers charging down on them were ''the flower of
7 Thomas of Burton, II:61.
8Annales Gandenses, p. 40.
9 John Barbour, II:280.
10Conflictus Laupensis, p. 309.
Page 196
French chivalry', as the Flemings faced in so many battles, or those English knights
who were so 'brightly' armed that 'they appeared to be angels of a heavenly kingdom',
as faced by the Scots at Loudon Hill and Bannockburn. Similar descriptions
accompany the knights at Boroughbridge, Laupen, Staveren, and Vottem. Even when
dismounted English warriors faced equals in their battles against the French or 'lesser
soldiers' as against the Scots, the discipline required to succeed in this tactic, especially
as it was so new to English military practice, is impressive.
Perhaps this is why so many battle commentators acknowledge the importance of the
leaders of these armies, even when the leadership is vague or is not named. At
Courtrai it was John of Renesse and William of Jülich who rose to great leadership; at
Arques it was William of Jülich again; at Loudon Hill and Bannockburn, Robert Bruce
led his army to victory; at Boroughbridge, it was Andrew Harclay; at Dupplin Moor,
Edward Balliol defeated the Scots; at Halidon Hill and Crécy, the king of England,
Edward III, showed great generalship; at Laupen, Rudolf von Erlach led the Bernese;
at Morlaix, it was the earl of Northampton whose leadership is singled out; and at
Neville's Cross, it was either William of La Zouche, the archbishop of York, Henry
Percy, or Ralph Neville, who provided the leadership which achieved victory.
Often these leaders inspired their troops with battle orations. At Courtrai, John of
Renesse urged his troops to fight strongly, attacking horses and riders with their
weapons and pushing them into the ditches. He promises the Flemings that if they
would 'fight strongly for their wives and their children, for the laws and liberty of
their homeland,' that God 'would be merciful to those who were humble, giving
consolation and victory to them.'11 It was a speech which would be given, in differing
words, by Robert Bruce at Loudon Hill and Bannockburn, by Nikolaas Zannekin at
Cassel, by Edward III at Halidon Hill and Crécy, and by the English generals at
Neville's Cross. While the accuracy of their words may be questioned, it must be
concluded, as John R.E. Bliese has done in his study of battle orations from the
central Middle Ages:
The speeches, of course, are not verbatim reports of what the commanders actually said; they
are the rhetorical products of the chroniclers themselves. However, the speeches are not mere
flights of fancy. The tradition of rhetorical historiography in which the chroniclers were writing
demanded that devices of amplification and ornamentation such as speeches had to be
plausible. Battle orations thus contain much useful information because they are a recurrent
rhetorical form that concentrates the authors' conceptions of motivation and morale in war.12
11Chronicon comitum Flandriae, p. 168.
12 This quote comes from John R.E. Bliese, 'When Knightly Courage May Fail: Battle Orations in
Medieval Europe,' Historian 53 (1991), 491. See also his 'Rhetoric and Morale: A Study of
Battle Orations from the Central Middle Ages,' Journal of Medieval History 15 (1989), 201-26;
'Aelred of Rievaulx's Rhetoric and Morale at the Battle of the Standard, 1138,' Albion 20 (1988),
543-56; 'The Battle Rhetoric of Aelred of Rielvaulx,' Haskins Society Journal 1 (1989), 99-107;
'The Courage of the Normans. A Comparative Study of Battle Rhetoric,' Nottingham Medieval
Studies 35 (1991), 1-26; and 'Leadership, Rhetoric, and Morale in the Norman Conquest of
England,' Military Affairs 52 (1988), 23-28.
Page 197
By speaking to their troops, outnumbered and ordered as infantry against cavalry, the
leaders of these armies prepared the soldiers for combat, filling them with an
enthusiasm and a belief that they could win against anyone they faced. They became,
as Jean de Brusthem describes the Flemings at Courtrai, 'rejoicing and excited, roaring
in the manner of lions.'13 Ceremonies, especially religious and patriotic rites, also
boosted the enthusiasm of those about to be attacked.
On came the charges, mean and ferocious, led by proud warriors who believed that
they could not lose. As planned, the charges in every battle quickly became disordered
and confused. The impetus was lost, and the soldiers, cavalry and infantry, hit their
target with little force. Horses would not penetrate the infantry lines, and infantry
could not penetrate them. They had become like a 'hare' caught in a 'trap', as the
Middle English poem describes the French cavalry at Courtrai.14 Knights were pulled
from their horses and infantry soldiers were knocked down. There they became
vulnerable to attacks from their opponents' weapons; lances, spears, swords, axes,
halberds and godedags proved effective against all they faced, no matter who they
were or how well they were armored. Future charges, if there were any, also failed.
The number of dead rose and was always impressive to all contemporary writers. As
Guillaume Guiart, an eyewitness to the battle of Mons-en-Pévèle, recalled later, 'death
and blood dwelt there.'15
13 Jean de Brusthem, p. 60.
14 There are several editions of this poem. The one I have used is edited by Rossell Hope
Robbins in his collection, Historical Poems of the XIVth and XVth Centuries (New York, 1959),
pp. 9-13. Other editions can be found in Political Songs of England, Camden Society, vi, ed. T.
Wright (London, 1839), pp. 187-195 and Chants historique de la Flandre, 400-1650, ed. L. de
Baecker (Lille, 1855), pp. 161-72. The above reference can be found in Robbins, pp. 192-93.
15 Guillaume Guiart, p. 293.
Page 199

Bibliography
Primary Sources
Ancienne chronique de Flandre. In RHF, xxii. Ed. Guignant and de Wailly. Paris, n.d.
Annales Gandenses. Ed. Hilda Johnstone. London, 1951.
Annales Paulini. In Chronicles of the Reigns of Edward I and Edward II. Vol. 1. Ed.
W. Stubbs. RS. London, 1883.
Annales St. Jacobi Leodensis. In MGH, SS, xvi. Ed. G.H. Pertz. Hannover, 1850.
Annales Tielenses. In MGH, SS, xvi. Ed. G.H. Pertz. Hannover, 1879.
Anonimalle Chronicle, 1307 to 1334, The. Ed. W.R. Childs and J. Taylor. Leeds,
1991.
Anonimalle Chronicle, 1333-81, The. Ed. V.H. Galbraith. Manchester, 1927.
Anonyme Stadchronik oder der Königshofen-Justinger, Die. In Die Berner-Chronik
des Conrad Justinger. Ed. G. Studer. Bern, 1871.
Avesbury, Robert of. De gestis mirabilibus Edwardi III. Ed. E.M. Thompson. RS.
London, 1889.
Baker, Geoffrey le. Chronicon. Ed. E.M. Thompson. Oxford, 1889.
Barber, Richard, ed. and trans. The Life and Campaigns of the Black Prince. London,
1979.
Barbour, John. The Bruce, or the Book of Robert de Broyss, King of Scots (1286-
1332). Ed. W.W Skeat. 4 vols. EETS. 1870-77; rpt. Oxford, 1968.
Beke, Jan. Chronographia Johannis de Beke. Ed. H. Bruch. 's Gravenhage, 1973.
Bel, Jean le. Chroniques de Jean le Bel. Ed. J. Viard and E. Duprez. 2 vols. SHF Paris,
1904-05.
Berchen, Willem van. Gelderse kroniek. Ed. A.J. de Mooy. WUVG, xxiv. Arnhem,
1950.
Bonn, Werner de. Vita Clementis VI. In Vitae paparum Avenionensium, i. Ed. S.
Baluze-Mollat. Paris, 1914.
Bower, Walter. Scotichronicon. Ed. D.E.R. Watt. 8 vols. Aberdeen, 000.
Breve chronicon de Flandriae. In CCF, iii. Ed. J.J. de Smet. Brussels, 1856.
Bridlington, Auctor of. Gesta Edwardi de Carnarvan. In Chronicles of the Reigns of
Edward I and Edward II. Vol. 2. Ed. W. Stubbs. RS. London, 1883.
Brut, The, or the Chronicles of England. Ed. F.W.D. Brie. 2 vols. EETS. London,
1906-08.
Budt, Adrien de. Chronicon Flandriae. In CCF, i. Ed. J.J. de Smet. Brussels, 1837.
Burton, Thomas of. Chronica monasterii de Melsa a fundatione usque ad annum
1396. Ed. E.A. Bonds. 3 vols. RS. London, 1866-68.
Brusthem, Jean de. Chronique. In CL, ii. Ed. C.S. Balau. Liège, 1931.
Calendar of Documents Relating to Scotland. Ed. J. Bain. 4 vols. Edinburgh, 1883.
Canterbury, Anonymous of. Chronicon. Ed. J. Tait. Manchester, 1914.
Chandos, Herald of. Life of the Black Prince by the Herald of Sir John Chandos. Ed.
and trans. M.K. Pope and E.C. Hodge. Oxford, 1910.
Chants historique de la Flandre, 400-1650. Ed. L. de Baecker. Lille, 1855.
'Chronicle of the Civil Wars of Edward II, A.' Ed. G.L. Haskins. In Speculum 14
(1939), 73-81.
Chronicon comitum Flandriae. In CCF, i. Ed. J.J. de Smet. Brussels, 1837.
Chronicon de Lanercost. Ed. J. Stevenson. Edinburgh, 1839.
Page 200
Chronicon Rotomagensi. In RHF, xxiii. Ed. de Wailly, Delisle and Jourdain. Paris,
n.d.
Chronicon Vazatense. In AHDG, xv. Bordeaux, 1874.
Chronique anonymé Française finissant en MCCCLVI. In RHF, xxi. Ed. Guignant
and de Wailly. Paris, 1855.
Chronique Artésienne et chronique Tournaisienne. Ed. F. Funck-Brentano. Paris,
1898.
Chronique de Flandre. In Istore et croniques de Flandres. Ed. Kervyn de Lettenhove.
5 vols. Brussels, 1879-80.
Chronique de l'abbaye de Saint-Trond. Ed. C. de Borman. 2 vols. Liège, 1877.
Chronique des Pays-Bas, de France, d'Angleterre et de Tournai. In CCF, iii. Ed. J.J.
de Smet. Brussels, 1856.
Chronique des quatre premiers Valois. Ed. S. Luce. SHF. Paris, 1862.
Chronique Liegeoise de 1402, La. Ed. E. Bacha. Brussels, 1900.
Chronique Normande de xiv siècle. Ed. A. and E. Molinier. SHF. Paris, 1882.
Chronique Parisienne anonymé de 1316 à 1339. Ed. A. Hellot. In Mémoires de la
société de l'histoire de Paris 11 (1895), 1-181.
'Chronique Valenciennoise inedité, une.' Ed. Etienne Delcambre. In Bulletin de la
commission royale d'histoire de Belgique 94 (1930).
Chronographia regum Francorum. Ed. H. Moranville. 2 vols. SHF Paris, 1891-97.
Codex diplomaticus Flandriae. Ed. Thierry de Limburg-Stirum. 2 vols. Bruges, 1879.
Conflictus Laupensis. In Die Berner-Chronik des Conrad Justinger. Ed. G. Studer.
Bern, 1871, pp. 302-13.
Coville, A., ed. 'Poems historiques du début de la Guerre de Cent Ans,' Histoire
litterature de France 38 (1949).
Cronica de Berno. In Die Berner-Chronik des Conrad Justinger. Ed. G. Studer. Bern,
1871.
Cuvelier. Chronique de Bertrand Guesclin. Ed. E. Charrière. 2 vols. Paris, 1839.
Desnouelles, Jean. Chronique. In RHF, xxi. Ed. Guignant and de Wailly. Paris, 1855.
Diessenhoven, Heinrich von. Chronicon ab a. 1316-1361. In FRG, iv. Ed. J.F.
Boehmer. Stuttgart, 1868.
Eulogium historiarum sive temporis. Ed. F.S. Haydon. 3 vols. RS. London, 1858-63.
ffoulkes, Charles. 'A Carved Flemish Chest at New College, Oxford,' Archaeologia
2nd ser. 15 (1914), 113-28.
Fordun, John. Chronica gentis Scotorum. Ed. W.F. Skene. 2 vols. HS. Edinburgh,
1871.
Franchet, Guillaume de. Chronicon et continuationes. In RHF, xx. Ed. Guignant and
de Wailly. Paris, 1840.
French Chronicle of London, The. Ed. G.J. Augnier. CS. London, 1844.
Froissart, Jean. Chroniques. In Oeuvres de Froissart. Ed. Kervyn de Lettenhove. 29
vols. Brussels, 1867-77.
. Chroniques. 15 vols. Ed. S. Luce et al. SHF Paris, 1869-1975.
. Chroniques: Début du premier livre. Edition du manuscrit de Rome Reg. lat. 869.
Ed. G.T. Diller. Geneva, 1972.
Gallia christiana. Paris, 1744.
Gesta abbatum St. Trudoniensum. In MGH, SS, x. Ed. G.H. Pertz. Hannover, 1851.
Grandes chroniques de France, Les. Ed. J. Viard. 10 vols. SHF Paris, 1920-53.
Gray, Thomas. Scalachronica. Ed. J. Stevenson. Edinburgh, 1836. Trans. H. Maxwell.
Glasgow, 1907.
Gregoras, Nikephoros. Byzantina historia. In PG, 148. Ed. J.-P. Migne. Paris, 1865.
Gui, Bernard. Flos chronicorum necnon e chronico regum Francorum. In RHF, xxi.
Ed. Guignant and de Wailly. Paris, 1855.
Page 201
Guiart, Guillaume. Branche des royaux lignages. In RHF, xxii. Ed. Guignant and de
Wailly. Paris, n.d.
Guisborough, Walter of. Chronicon. Ed. H. Rothwell. CS. London, 1957.
Hemingburgh, Walter of. Chronicon. 2 vols. EHS. London, 1848.
Hemricourt, Jacques de. Le miroir des nobles de Hesbaye. In OEuvres de Jacques de
Hemricourt. Ed. C. de Borman. 3 vols. Brussels, 1910.
Historical Poems of the XIVth and XVth Centuries. Ed. R. H. Robbins. New York,
1959.
Hocsem, Jean de. La chronique de Jean de Hocsem. Ed. G. Kurth. Brussels, 1927.
Istorie Pistolensi. In SRI, xi. Ed. L. Muratori. Rome, 1728.
Ives. Pars ultima chronicon. In RHF, xxi. Ed. Guignant and de Wailly. Paris, 1855.
Justinger, Conrad. Die Berner-Chronik des Conrad Justinger. Ed. G. Studer. Bern,
1871.
Klerk, Jan de. Brabantse yeesten of rijmkroniek van Braband. Ed. J.F. Willems and
J.H. Bormans. 2 vols. Brussels, 1839.
Knighton, Henry. Chronicon. Ed. J.R. Lumby. 2 vols. RS. London, 1889-95.
Kronyk van Vlaenderen van 580 tot 1467. Ed. P. Blommaert and C.P Serrière. 2 vols.
Ghent, 1839.
Leger, L., ed. and trans., 'Un poème Tchèque sur la bataille de Crécy,' Journal des
savants (1902), 323-31.
Lescot, Richard. Chronique. Ed. J. Lemoine. SHF. Paris, 1896.
Lettres de Jean XXII (1316-1334). In Analecta Vaticano-Belgica, 2. Ed. Fayan.
Rome, 1908.
Liber Pluscardensis. Ed. F.J.H. Skene. 2 vols. HS. Edinburgh, 1877.
Libro de los fechos et conquistas del principado de la Morea compilado por
comandamiento de Don Johan Ferrandez de Heredia. Ed. A. Morel-Fatio. Geneva,
1885.
Macray, W.D., ed. 'Robert Baston's Poem on the Battle of Bannockburn,' English
Historical Review 19 (1904), 507-08.
Minot, Laurence. Poems. Ed. J. Hall. Oxford, 1897.
'Mort du conte de Henau, La.' In Panègyriques des comtes de Hainaut et de Hollande
Guillaume I et Guillaume II. Ed. C. Potvin. Mons, 1863.
Muisit, Gilles li. Chronicon. In CCF, ii. Ed. J.J. de Smet. Brussels, 1841.
Muntaner, Ramón. Crònica. Trans. H.M. Goodenough. 2 vols. London, 1920-21.
Murimuth, Adam. Continuatio chronicorum. Ed. E.M. Thompson. RS. London, 1889.
Nangis, Guillaume de. Chronicon et continuationes. Ed. H. Geraud. 2 vols. SHF.
Paris, 1843.
Neuenberg, Mathias von. Chronik. Ed. A. Hofmeister. MGH, SS (nova series), iv.
Berlin, 1940.
Outremeuse, Jean d'. Chronique abrégée, de 1341 à 1400. In CL, ii. Ed. S. Balau and
E. Fairon. Brussels, 1931.
Paris, Geoffroi de. Chronique rimée. In RHF, xxii. Ed. Guigant and de Wailly. Paris,
n.d.
Paris, Jean de. Memoriale temporum. In RHF, xxi. Ed. Guignant and de Wailly. Paris,
1855.
Passio Francorum secundum Flemingos. In Die Parodie im Mittelalter. Ed. Paul
Lehmann. Stuttgart, 1963, pp. 30-31.
Political Songs of England. Ed. T. Wright. CS. London, 1839.
Procurator, Willem. Chronicon. Ed. Pijnacker Hardwijk. WHGU, 3rd ser., 20.
Amsterdam, 1904.
Reading, John of. Chronicon. Ed. J. Tait. Manchester, 1914.
Rebdorf, Heinrich von. Annales imperatorum et paparum (1294-1362). In FRG, iv.
Ed. J.F. Boehmer. Stuttgart, 1868.
Page 202
Récits d'un bourgeois de Valenciennes. Ed. Kervyn de Lettenhove. Brussels, 1877.
Rijmkroniek van Vlaenderen. In CCF, iv. Ed. J.J. de Smet. Brussels, 1865.
Rymer, Thomas, ed. Foedera, conventiones, litterae, et cujuscunque generis acta
publica inter reges Angliae et alios quosvis imperatores, reges, pontifices, principes,
vel communitates (1101-1654). 20 vols. London, 1704-35.
Soulèvement de la Flandre maritime de 1323-1328, Le. Ed. H. Pirenne. Brussels,
1900.
Stiermarken, Otto von. Oesterreichische reimkroniek. In MGH, Deutschen Chroniken,
v. Ed. J. Seemuller. Hannover, 1893.
Stoke, Melis. Rijmkroniek van Holland. Ed. W.G. Brill. 2 vols. WHGU, n.s. x1 and
x1ii. Utrecht, 1885.
Thilrode, Johannes. Chronicon. In MGH, SS, xxv. Ed. G.H. Pertz. Hannover, 1880.
Trokelowe, John of. Annales. Ed. H.T. Riley. RS. London, 1866.
Velthem, Lodewijk van. Voortzetting van de Spiegel historiael (1284-1316). Ed. H.
Vander Linden et al. Brussels, 1922.
Venette, Jean de. Chronique. Ed. and trans. R.A. Newhall. New York, 1953.
Villani, Giovanni. Istorie Fiorentine. In SRI, xiii. Ed. L. Muratori. Rome, 1728.
. Cronica. Ed. M.L. Ridotta. 7 vols. Florence, 1823.
Vita Edwardi secundi. Ed. and trans. N. Denholm-Young. London, 1957.
Vita et mors Edwardi secundi. In Chronicles of the Reigns of Edward I and Edward
II. Vol. 2. Ed. W. Stubbs. RS. London, 1883.
Walsingham, Thomas. Historia anglicana. Ed. H.T. Riley. 2 vols. London, 1863-64.
Winterthur, Jean de. Chronicon. In MGH, SS (nova series), iii. Ed. F. Baethegen.
Hannover, 1924.
Wrottesley, George, ed. Crecy and Calais from the Original Records in the Public
Record Office. London, 1898.
Wyntoun, Andrew of. Brevis Chronica. HS. Edinburgh, 1879.
. Orygynale Cronykil of Scotland. 3 Vols. HS. Edinburgh, 1872, 1873, 1879.
Secondary Sources
Abels, Richard P. Lordship and Military Obligation in Anglo-Saxon England.
Berkeley, 1988.
Allmand, Christopher. Henry V. Berkeley, 1992.
. The Hundred Years War: England and France at War, c. 1300-c.1450. Cambridge,
1988.
Ayton, Andrew. 'The English Army and the Normandy Campaign of 1346.' In
England and Normandy in the Middle Ages. Ed. D. Bates and A. Curry. London,
1994, pp. 253-68.
. Knights and Warhorses: Military Service and the English Aristocracy under Edward
III. Woodbridge, 1994.
Bachrach, Bernard S. Fulk Nerra, the Neo-Roman Consul, 987-1040. Berkeley, 1993.
Barber, Malcolm. The New Knighthood: A History of the Order of the Temple.
Cambridge, 1994.
Barber, Richard. Edward, Prince of Wales and Aquitaine: A Biography of the Black
Prince. London, 1978.
Barker, Juliet R.V. The Tournament in England, 1100-1400. Woodbridge, 1986.
Barlow, Frank. William I and the Norman Conquest. New York, 1965.
Page 203
Baron, Hans. The Crisis of the Early Italian Renaissance. 2 vols. Princeton, 1955.
Barraclough, G. The Origins of Modern Germany. 2nd ed. Oxford, 1947.
Barron, Evan M. The Scottish War of Independence: A Critical Study. 2nd ed.
Inverness, 1934.
Barrow. G.W.S. Robert Bruce and the Community of the Realm of Scotland. 3rd ed.
Edinburgh, 1988.
Bartlett, Robert. 'Military Technology and Political Power.' In The Making of Europe:
Conquest, Colonization and Cultural Change, 950-1350 (Princeton, 1993), pp. 60-
84.
Bartusis, Mark C. The Late Byzantine Army: Arms and Society, 1204-1453.
Philadelphia, 1992.
Beeler, John. 'Towards a Re-Evaluation of Medieval English Leadership,' Journal of
British Studies 3 (1963), 1-10.
. Warfare in England, 1066-1189. Ithaca, 1966.
. Warfare in Feudal Europe, 730-1200. Ithaca, 1971.
Bellamy, J.G. The Law of Treason in England in the Late Middle Ages. Cambridge,
1970.
Bliese, John R.E. 'Aelred of Rievaulx's Rhetoric and Morale at the Battle of the
Standard, 1138,' Albion 20 (1988), 543-56.
. 'The Battle Rhetoric of Aelred of Rielvaulx,' Haskins Society Journal 1 (1989), 99-
107.
. 'The Courage of the Normans. A Comparative Study of Battle Rhetoric,' Nottingham
Medieval Studies 35 (1991), 1-26.
. 'Leadership, Rhetoric, and Morale in the Norman Conquest of England,' Military
Affairs 52 (1988), 23-28.
. 'Rhetoric and Morale: A Study of Battle Orations from the Central Middle Ages,'
Journal of Medieval History 15 (1989), 201-26.
. 'When Knightly Courage May Fail: Battle Orations in Medieval Europe,' Historian 53
(1991), 489-504.
Bonaparte, Louis-Napoleon and I. Favé. Du passé et l'avenir de l'artillerie. 6 vols.
Paris, 1856.
Bonjour, E., H.S. Offler, and G.R. Potter. A Short History of Swizterland. Oxford,
1952.
Bovesse, J. 'Le Comte de Namur Jean Ier et les événements du comté de Flandre en
1325-1326,' Bulletin de la commission royale d'histoire 131 (1965), 385-454.
Bradbury, Jim. The Medieval Archer. New York, 1985.
. The Medieval Siege. Woodbridge, 1992.
Burckhardt, Jacob. The Civilization of the Renaissance in Italy. Trans. L. Geiger and
W. Götz. 2 vols. New York, 1958.
Burne, Alfred H. The Agincourt War: A Military History of the Latter Part of the
Hundred Years War from 1369 to 1453. London, 1956.
. The Crecy War: A Military History of the Hundred Years War from 1337 to the
Peace of Bretigny, 1360. London, 1955.
Burns, R. Ignatius. 'The Catalan Company and the European Powers, 1305-1311,'
Speculum 29 (1954), 751-71.
Chaplais, Pierre. Piers Gaveston: Edward Ha Adoptive Brother. Oxford, 1994.
Cheetham, Nicolas. Mediaeval Greece. New Haven, 1981.
Chickering, Howell, and Thomas H. Seiler, eds. The Study of Chivalry: Resources and
Approaches. Kalamazoo, 1988.
Clarke M.V. and V.H. Galbraith. 'The Deposition of Edward II,' Bulletin of the John
Rylands Library 14 (1930), 125-81.
Page 204
Contamine, Philippe. 'Crécy (1346) et Azincourt (1415): Une comparison.' In Divers
aspects du moyen âge en occident: 1er Congrès Historique des Jeunes Historiens du
Calaisis. Calais, 1977, pp. 29-44.
. 'La guerre de cent ans: Le XIVe siècle. La France au rythme de la guerre.' In Histoire
militaire de la France, vol. 1: Des origines à 1715. Ed. P. Contamine. Paris, 1992.
. L'oriflamme de Saint-Denis aux XIVe et XVe siècles. Nancy, 1975.
. War in the Middle Ages. Trans. M. Jones. London, 1984.
Corfis, Ivy A. and Michael Wolfe, eds. The Medieval City Under Siege. Woodbridge,
1995.
Creasy, Sir Edward. The Fifteen Decisive Battles of the World. 36th ed. London, 1894.
Crouch, David. William Marshal. London, 1994.
Curry, Anne. The Hundred Years War. Houndmills, 1993.
Dahmus, Joseph. Seven Decisive Battles of the Middle Ages. Chicago, 1983.
Daniels, E. Geschichte des Kriegswesens. Vol. II: Das mittelalterliche Kriegswesen.
2nd ed. Berlin, 1927.
Davies, James Conway. The Baronial Opposition to Edward II. Its Character and
Policy. London,1918.
Davis, R.H.C. The Medieval Warhorse. London, 1989.
Delbrück, Hans. History of the Art of War Within the Framework of Political History.
Vol. III: Medieval Warfare. Trans. W.J. Renfroe, Jr. Westport, 1984originally
Geschichte der Kriegskunst im Rahmen des Politischen Geschichte. Vol. III:
Mittelalter. 2nd ed. Berlin, 1923.
Denifle, Henri. La guerre de cent ans et la désolation des églises, monastères et
hospitaux en France. Paris, 1899.
Deprez, E. Les préliminaires de la guerre de cent ans. Paris, 1902.
Devic, C. and J. Vaissette. Histoire généale de Languedoc. Toulouse, 1872.
De Vries, Kelly. 'Contemporary Views of Edward III's Failure at the Siege of Tournai,
1340,' Nottingham Medieval Studies 39 (1995), 70-105.
. 'God, Leadership, Flemings, and Archery: Contemporary Perceptions of Victory and
Defeat at the Battle of Sluys, 1340,' American Neptune 55 (1995), 1-28.
. 'Hunger, Flemish Participation and the Flight of Philip VI: Contemporary Accounts
of the Siege of Calais, 1346-47,' Studies in Medieval and Renaissance History n.s 12
(1991), 129-81.
. Medieval Military Technology. Peterborough, 1992.
Digard, Georges. Philippe le Bel et le Saint Siège de 1285 à 1304. 2 vols. Paris, 1936.
Douglas, David C. William the Conqueror: The Norman Impact upon England.
Berkeley, 1964.
Du Boulay, Caesar Egassio. Historia universitatis Parisiensis. 1668; rpt. Frankfurt,
1966.
Duby, Georges. William Marshal: The Flower of Chivalry. Trans. R. Howard. New
York, 1985.
Duncan, A.A.M. 'The War of the Scots, 1306-23,' Transactions of the Royal
Historical Society 6th series, 2 (1992), 125-51.
Durrer, Robert. Schweizer Kriegsgeschichte. Ed. M. Feldmann and H.H.G. Wirz. 2
vols. Bern, 1915.
Erben, W. Kriegsgeschichte des Mittelalters. Berlin, 1929.
Favier, Jean. La guerre de cent ans. Paris, 1980.
Felbien, Michel. Histoire de l'abbaye royale de Saint-Denis en France. 1707; rpt.
Paris, 1973.
Page 205
Feller, Richard. Geschichte Berns. 3 vols. Bern, 1949.
Fischer, R. von. Schweizerkriegsgeschichte. Ed. M. Feldman and H.H.G. Wirz. Vol. 1.
Bern, 1915.
Fisher, Andrew. 'Wallace and Bruce: Scotland's Uneasy Heroes,' History Today 39
(February 1989), 18-23.
Fletcher, Richard. The Quest for El Cid. Oxford, 1989.
Forey, Alan. The Military Orders: From the Twelfth to the Fourteenth Century.
Toronto, 1992.
Fowler, Kenneth A. 'News from the Front: Letters and Despatches of the Fourteenth
Century.' In Guerre et société en France, en Angleterre et en Bourgogne, xive-xve
siècle. Ed. R. Contamine et al. Lille, 1991, pp. 63-92.
Fryde, E.B. 'Financial Resources of Edward III in the Netherlands, 1337-40,' Revue
Beige de philologie et d'histoire 45 (1967), 1142-1216.
. William de la Pole: Merchant and King's Banker (+1366). London, 1988.
Fryde, Natalie B. 'Edward III's Removal of his Ministers and Judges, 1340-1,' Bulletin
of the Institute of Historical Research 48 (1975), 149-63.
. The Tyranny and Fall of Edward II, 1321-1327. Cambridge, 1979.
Fuller, J.F.C. The Decisive Battles of the Western World and their Influence upon
History. 2nd ed. 2 vols. London, 1970.
. A Military History of the Western World. Vol. 1: From the Earliest Times to the
Battle of Lepanto. New York, 1954.
Funck-Brentano, Frantz. Mémoire sur la bataille de Courtrai (1302, 11 juillet) et les
chroniqueurs qui en ont traité, pour servir à l'historiographie du règne de Philippe
le Bel. Paris, 1981.
. Philippe le Bel en Flandre: Les origines de la guerre de cent ans. Paris, 1896.
. 'Le traité de Marquette (Septembre 1304).' In Melanges Julien Havet. 1895; rpt.
Geneva, 1972, pp. 749-58.
Gaier, Claude. Art et organisation militaires dans la principauté de Liège et dans le
comté de Looz au Moyen Age. Brussels, 1968.
. 'La cavalerie lourde en Europe occidentale du XIIe au XVIe siècle,' Revue
internationale d'histoire militaire 31 (1971), 385-96.
. 'L'invincibilité anglaise et le grande are après la guerre de cent ans: un mythe tenace,'
Tijdschrift voor gescheidenis 91 (1978), 378-85.
Garin, Eugenio. Der Italienische Humanismus. Bern, 1947.
George, Hereford B. 'The Archers at Crecy,' English Historical Review 10 (1895), 733-
38.
Gillingham, John. Richard the Lionheart. London, 1978.
Graham, Frank. Famous Northern Battles. Rothbury, 1988.
Haines, Roy Martin. Archbishop John Stratford. Political Revolutionary and
Champion of the Liberties of the English Church, ca. 1275/80-1348. Toronto, 1986.
Hamilton, J.S. Piers Gaveston, Earl of Cornwall, 1307-1312: Politics and Patronage
in the Reign of Edward II. Detroit, 1988.
Hardy, Robert. Longbow: A Social and Military History. 3rd ed. London, 1992.
. 'The Longbow.' In Arms, Armies and Fortifications in the Hundred Years War. Ed.
A. Curry and M. Hughes. Woodbridge, 1994, pp. 161-82
Harriss, G.L. King, Parliament, and Public Finance in Medieval England to 1369.
Oxford, 1975.
Page 206
Haskins, George L. 'Judicial Proceedings Against a Traitor after Boroughbridge, 1322,'
Speculum 12 (1937), 509-11.
Hay, Denys. Europe in the Fourteenth and Fifteenth Centuries. 2nd ed. London,
1989.
. The Italian Renaissance in Its Historical Background. Cambridge, 1961.
Head, Thomas and Richard Landes, eds. The Peace of God: Social Violence and
Religious Response in France around the Year 1000. Ithaca, 1992.
Henneman, John Bell. Royal Taxation in Fourteenth-Century France: The
Development of War Financing, 1322-1356. Princeton, 1971.
Holmes, George. Europe: Hierarchy and Revolt, 1320-1450. London, 1975.
Housley, Norman. The Later Crusades, 1274-1580: From Lyons to Alcazar. Oxford,
1992.
Howard, Michael. War in European History. Oxford, 1976.
Hugenholtz, F.W.N. Drie boerenopstanden uit de viertiende eeuw. 's Gravenhage,
1978.
Jacoby, David. 'La ''compagnie catalane" et l'état catalan de Grèce. Quelques aspects de
leur histoire,' Journal des savants (1966), 78-103.
Jones, Archer. The Art of War in the Western World. Oxford, 1989.
Jones, W.R. 'Rex et Ministri: English Local Government and the Crisis of 1341,'
Journal of British Studies 13 (1973), 1-20.
Keegan, John. The Face of Battle. Harmondsworth, 1978.
Keen, M.H. Chivalry. New Haven, 1984.
. England in the Later Middle Ages: A Political History. London, 1973.
Kenyon, John R. Medieval Fortifications. New York, 1990.
Köhler, G. Die Entwickelung des Kriegswesens und der Kriegführung in der Ritterzeit
von Mitte des 11. Jahrhunderts bis zu den Hussitenkriegen. 3 vols. Breslau, 1886.
Kurth, Godefroid. La cité de Liège au moyen-age. 3 vols. Brussels, 1909-11.
Leclerq, Jean. 'Un sermon prononcé pendant la guerre de Flandre sous Philippe le
Bel,' Revue du moyen âge latin 1 (1945), 165-72.
Lejeune, J. 'Het prinsbisdom Luik tot 1390.' In Algemene geschiedenis der
Nederlanden. Vol. 3: De late middeleeuwen, 1305-1477. Utrecht, 1951, pp. 175-89.
Liddell Hart, B. H. Strategy: The Indirect Approach. New York, 1954.
Liebenau, Theodor von. 'Berichte über die Schlacht am Morgarten,' Mitteilungen des
historischen Vereins des Kantons Schwyz 3 (1884), 1-85.
Lloyd, E. M. 'The " Herse" of Archers at Crecy,' English Historical Review 10 (1895),
538-41.
Lloyd, T.H. The English Wool Trade in the Middle Ages. Cambridge, 1977.
Longnon, J. 'The Frankish States in Greece, 1204-1311.' In A History of The Crusades.
Ed. K.M. Setton. Vol. II: The Later Crusades. Ed. R. L. Wolff and H.W. Hazard.
Madison, 1969), pp. 235-74.
Lot, Ferdinand. L'art militaire et les armées au moyen âge en Europe et dans le
proche orient. 2 vols. Paris, 1946.
Lucas, Henry Stephen. The Low Countries and the Hundred Years' War, 1326-1347.
Ann Arbor, 1929.
Lyon, Bryce. 'The Role of Cavalry in Medieval Warfare: Horses, Horses All Around
and Not a One to Use,' Mededelingen van de Koninklijke Academie voor
Wetenschappen, Letteren en Schone Kunsten van België 49 (1987), 77-90.
Lyons, Malcolm Cameron and D.E.R. Jackson. Saladin: The Politics of the Holy War.
Cambridge, 1982.
Page 207
Mackenzie, W.M. The Battle of Bannockburn: A Study in Mediaeval Warfare.
Glasgow, 1913.
Maddicott, J.R. Thomas of Lancaster, 1307-1322: A Study in the Reign of Edward II.
Oxford, 1970.
Malderghem, Jean van. La bataille de Staveren (26 septembre 1345). Brussels, 1869.
Marshall, Christopher. Warfare in the Latin East, 1192-1291. Cambridge, 1992.
McCrackan, W.D. The Rise of the Swiss Republic. 2nd ed. New York, 1901.
McGuffie, T.H. 'The Long-bow as a Decisive Weapon,' History Today 5 (1955), 737-
41.
McKisack, May. The Fourteenth Century. Oxford History of England. Oxford, 1959.
Mertens, Jacques. 'De boerenopstand onder Zannekin.' In Nikolaas Zannekin en de
slag bij Kassel 1328-1978. Dixmude, 1978, pp. 96-103.
. 'La confiscations dans la Chatellenie du Franc de Bruges après la bataille de Cassel,'
Bulletin de la commission royale d'histoire de Belgique 134 (1968), 239-84.
. 'De economische en sociale toestand van de opstandelingen uit het Brugse Vrije, wier
goederen na de slag bij Cassel (1328) verbeurd verklaard werden,' Revue Belge de
philologie et d'histoire 47 (1969), 1131-53.
. 'Zannekin of de evolutie van het beeld van een volksheld,' De Frans Nederlanden.
Les Pays-Bas Français (1978), 24-37.
Meyer, Bruno. 'Die Schlacht am Morgarten: Verlauf der Schlacht und Absichten der
Parteien,' Revue Suisse d'histoire 16 (1966), 129-79.
Miller, Edward. War in the North: The Anglo-Scottish Wars of the Middle Ages. Hull,
1960.
Moncado, Francisco de. Expedición de los Catalanes y Aragoneses contra Turcos y
Griegos. Madrid, 1941. Translation: The Catalan Chronicle of Francisco de
Moncado. Trans. E Hernández. Ed. J.M. Sharp. El Paso, 1975.
Morillo, Stephen. Warfare under the Anglo-Norman Kings, 1066-1135. Woodbridge,
1994.
Morris, John E. 'The Archers at Crecy,' English Historical Review 12 (1897), 427-36.
. Bannockburn. Cambridge, 1914.
. 'Mounted Infantry in Mediaeval Warfare,' Transactions of the Royal Historical
Society 3rd ser. 8 (1914), 77-102.
. The Welsh Wars of Edward I. Oxford, 1901.
Neilson, George. John Barbour, Poet and Translator. London, 1900.
Nicholas, David. Medieval Flanders. London, 1992
. Town and Countryside: Social, Economic and Political Tensions in Fourteenth-
Century Flanders. Bruges, 1971.
. The van Arteveldes of Ghent: The Varieties of Vendetta and the Hero in History.
Ithaca, 1988.
Nicholson, Helen. Templars, Hospitallers and Teutonic Knights: Images of the
Military Orders, 1128-1291. Leicester, 1995.
Nicholson, Ranald. Edward III and the Scots: The Formative Years of a Military
Campaign. Oxford, 1965.
. Scotland: The Later Middle Ages. Edinburgh, 1974.
. 'The Siege of Berwick, 1333,' Scottish Historical Review 40 (1961), 19-42.
Niermeyer, J.F. 'Het Sticht Utrecht, Gelre en de Friese landen in de veertiende eeuw.'
In Algemene geschiedenis der Nederlanden. Vol. 3: De late middeleeuwen, 1305-
1477. Utrecht, 1951, pp. 125-60.
Nowé, Henri. La bataille des éperons d'or. Brussels, 1945.
Page 208
Oman, Charles. The Art of War in the Middle Ages A.D. 378-1515. Ed. J.H. Beeler.
Ithaca, 1953.
. A History of the Art of War: The Middle Ages from the Fourth to the Fourteenth
Century. London, 1898.
. A History of the Art of War in the Middle Ages. 2 vols. London, 1905.
Ormrod, W.M. The Reign of Edward III: Crown and Political Society in England,
1327-1377. New Haven, 1990.
Packe, Michael. King Edward III. London, 1983.
Partner, Peter. The Knights Templar and their Myth. Oxford, 1981.
Pascot, Jep. Les almugavares: Mercenaires catalans du moyen age (1302-1388).
Brussels, 1971.
Pernoud, Reginé. Joan of Arc: By Herself and Her Witnesses. Trans. E. Hyams. New
York, 1964.
Perroy, Edouard. The Hundred Years War. Trans. W.B, Wells. London, 1951.
Phillips, J.R.S. Aymer de Valence, Earl of Pembroke, 1307-1324: Baronial Politics in
the Reign of Edward II. Oxford, 1972.
Pirenne, Henri. 'Documents relatifs à l'histoire de la Flandre pendant la première
moitié du XIVe siècle,' Bulletin de la commission royale d'histoire 7 (1897), 477-93
. Early Democracies in the Low Countries: Urban Society and Political Conflict in
the Middle Ages and the Renaissance. Trans. J.V. Saunders. New York, 1971.
. Histoire de Belgique. Vol. 1: Des origines du commencement du XIVe siècle.
Brussels, 1903.
. Histoire de Belgique. Vol. 2: Du commencement du XIVe siècle à la mort de Charles
le Téméraire. Brussels, 1903.
. 'Un mémoire de Robert de Cassel sur sa participation à la révolte de la Flandre
maritime en 1324-1325,' Revue du nord 1 (1910), 45-50.
. 'La version flamande et la version française de la bataille de Courtrai, note
historiographie de XIVe siècle,' Bulletin de la commission royale d'histoire 4th ser.,
17 (1890), 11-50; 'Note supplémentaire,' Bulletin de la commission royale d'histoire
5th ser., 2 (1892), 85-123.
Pollard, A.J. John Talbot and the War in France, 1427-1453. London, 1983.
Pounds, N.J.G. The Medieval Castle in England and Wales: A Social and Political
History. Cambridge, 1990.
Powers, James F. A Society Organized for War: The Iberian Municipal Militias in the
Central Middle Ages, 1000-1284. Berkeley, 1988.
Powicke, F.M. The Thirteenth Century, 1216-1307. Oxford History of England.
Oxford, 1953.
Powicke, M.R. 'Edward II and Military Obligation,' Speculum 39 (1956), 92-119.
Prentout, Henri. La prise de Caen par Edouard III, 1346. Caen, 1904.
Preston, Richard A., Alex Roland, and Sydney F. Wise. Men at Arms. 5th ed. Fort
Worth, 1991.
Prestwich, Michael. Edward I. London, 1988.
'England and Scotland during the Wars of Independence.' In England and Her
Neighbours, 1066-1453: Essays in Honour of Pierre Chaplais. Ed. M. Jones and M.
Vale. Woodbridge, 1999, pp. 181-97.
. 'English Armies in the Early Stages of the Hundred Years War: a Scheme in 1341,'
Bulletin of the Institute of Historical Research 56 (1983), 102-13.
. The Three Edwards: War and State in England, 1272-1377. London, 1980.
Page 209
Pryor, John H. Geography, Technology, and War. Studies in the Maritime History of
the Mediterranean, 649-1571. Cambridge, 1988.
Rabil, Albert, Jr. 'The Significance of "Civic Humanism" in the Interpretation of the
Italian Renaissance.' In Renaissance Humanism: Foundations, Forms, and Legacy.
Vol. 1: Humanism in Italy. Ed. A. Rabil, Jr. Philadelphia, 1988, pp. 141-74.
Rees, Gareth. 'The Longbow's Deadly Secrets,' New Scientist 138 (June 5, 1993), 24-
25.
Rogers, Clifford J. 'Edward III and the Dialectics of Strategy, 1327-1360,'
Transactions of the Royal Historical Society 6th ser. 4 (1994), 88-102.
. 'The Military Revolutions of the Hundred Years War,' Journal of Military History 57
(1993), 249-51.
Rogers, R. Latin Siege Warfare in the Twelfth Century. Oxford, 1992.
Rompaey, Jan van. 'De opstand in het vlaamse kustland van 1323 tot 1328 en de
figuur van Nikolaas Zannekin.' In Nikolaas Zannekin en de slag bij Kassel 1328-
1978. Dixmude, 1978, pp. 104-32.
Rose, Walther. 'König Johann der Blinde von Böhmen und die Schlacht bei Crécy
(1346),' Zeitschrift für historisches Waffenkunde 7 (1915-17), 37-60.
Sabbe, Jacques. Vlaanderen in opstand, 1323-1328: Nikolaas Zannekin, Zeger
Janszone en Willem de Deken. Bruges, 1992.
Sayles, George. 'The Formal Judgments on the Traitors of 1322,' Speculum 16 (1941),
57-63.
Scammell, Jean. 'Robert I and the North of England,' English Historical Review 73
(1958), 385-403.
Schnitzer, Maria. Die Morgartenschlacht im werdenden schweizerischen National-
bewusstsein. Zurich, 1969.
Setton, Kenneth M. Catalan Domination of Athens, 1311-1388. Cambridge, 1948.
. 'The Catalans in Greece, 1311-1380.' In A History of The Crusades. Ed. K.M. Setton.
Vol. III: The Fourteenth and Fifteenth Centuries. Ed. H.W. Hazard. Madison, 1975,
pp. 167-224.
Smallwood, T.M. 'An Unpublished Early Account of Bruce's Murder of Comyn,'
Scottish Historical Review 54 (1975), 1-10.
Scott, Ronald McNair. Robert the Bruce: King of Scots. New York, 1982.
Seward, Desmond. The Hundred Years War: The English in France, 1337-1453.
New York, 1978.
Showalter, Dennis E. 'Caste, Skill, and Training: The Evolution of Cohesion in
European Armies from the Middle Ages to the Sixteenth Century,' Journal of Military
History 57 (1993), 407-30.
Smail, R.C. Crusading Warfare, 1097-1193. Cambridge, 1956.
Smet, J.M. de. 'Passio francorum secundum flemyngos,' De leiegouw (1977), 289-319.
Smet, Joseph de. 'Les effectifs Brugeois à la bataille de Courtrai en 1302,' Revue Belge
de philologie et d'histoire 12 (1933), 631-36.
Strayer, Joseph. The Reign of Philip the Fair. Princeton, 1980.
Sumption, Jonathan. The Hundred Years War. Trial by Battle. Philadelphia, 1991.
Talbot, Hugh. The English Achilles: The Life and Campaigns of John Talbot, 1st Earl
of Shrewsbury. London, 1981.
TeBrake, William H. A Plague of Insurrection: Popular Politics and Peasant Revolt
in Flanders, 1323-1328. Philadelphia, 1993.
Thompson, M.W. The Decline of the Castle. Cambridge, 1987.
. The Rise of the Castle. Cambridge, 1991.
Page 210
Tout, T.F. 'Firearms in England in the Fourteenth Century,' English Historical Review
26 (1911), 666-702.
. The Place of the Reign of Edward II in English History. 2nd ed. Manchester, 1936.
. 'The Tactics of the Battles of Boroughbridge and Morlaix,' English Historical Review
19 (1904), 711-15.
Tuck, J.A. 'War and Society in the Medieval North,' Northern History 21 (1985), 33-
52.
Vale, M.G.A. Charles VII. Berkeley, 1974.
van Creveld, Martin. Command in War. Cambridge, 1985.
Verbruggen, J.F. The Art of Warfare in Western Europe during the Middle Ages: From
the Eighth Century to 1340. Trans. S. Willard and S.C.M. Southern. Amsterdam,
1977.
. 'De goededag,' Militaria Belgica 1977, 65-70.
. 'De historiografie van de guldensporenslag,' De leiegouw (1977), 245-72.
. De krijgkunst in west-europa in de middeleeuwen (IXe tot XIVe eeuw). Brussels,
1954.
. 'De rol van de ruiterij in de middeleeuwse oorlogvoering,' Revue Belge d'histoire
militaire 30 (1994), 389-418.
. De slag der guldensporen, bijdrage tot de gescheidenis van Vlaanderens
vrijheidsoorlog, 1297-1305. Antwerp, 1952.
. 'La tactique de la chevalerie française de 1340 à 1415,' Publications de l'université de
l'état à Elisabethville 1 (1961), 39-48.
. 'La tactique militaire des armées de chevaliers,' Revue du nord 29 (1947), 161-80.
Vercauteren, Fernand. Luttes sociales à Liége, xiiie et xive siècles. Brussels, 1946.
. 'Het prinsbisdom Luik tot 1316.' Algemene geschiedenis der Nederlanden. Vol. 2: De
volle middeleeuwen, 925-1305. Utrecht, 1950, pp. 338-52.
Viard, Jules. 'La campagne de juillet-août 1346 et la bataille de Crécy,' Le moyen âge
2nd ser. 27 (1926), 1-84.
. 'La guerre de Flandre (1328),' Bibliothèque de l'ecole de chartes 83 (1922).
. 'Le siege de Calais,' Le moyen âge 40 (1929), 129-89.
Wailly, Henri de. Crécy 1346: Anatomy of a Battle. Poole, 1987.
Warner, Marina. Joan of Arc: The Image of Female Heroism. Harmondsworth, 1981.
Waugh, Scott. L. England in the Reign of Edward III. Cambridge, 1991.
. 'For King, Country and Patron: The Despensers and Local Administration,' Journal
of British Studies 22 (1983), 23-45.
. 'The Profits of Violence: The Minor Gentry in the Rebellion of 1321-22 in
Gloucestershire and Herefordshire,' Speculum 52 (1977), 843-69.
Webster, Bruce. 'Scotland Without a King, 1329-1341.' In Medieval Scotland: Crown,
Lordship and Community: Essays Presented to G.W.S. Barrow. Ed. A. Grant and K.J.
Stringer. Edinburgh, 1993, pp. 224-38.
Wernli, Fritz. Die Entstehung der schweizerischen Eidgenossenschaft. Zurich, 1972.
Werveke, Hans van. 'Les charges financières de traité d'Athis (1305),' Revue du Nord
22 (1950), 81-93.
-. Jacques van Artevelde. Brussels, 1943.
Witzig, Hans. Von Morgarten bis Marignano. Zurich, 1957.
Page 211

Index
A
Abernethy, Alexander 51-52
Africanus, Scipio 11
Agincourt, Battle of (1415) 6
Alexander II, King of Scotland 50
Alexander III, King of Scotland 49
Almugavares 58-59, 64
Ambushes 188-190
Ancienne Chronique de Flandre 11, 17, 21, 44
Annales Gandenses 11, 12, 14, 15, 21, 24-31, 33-35, 37, 38, 40-46, 193, 195
Anonimalle Chronicle, The 68-69, 80, 116-118, 120, 122-123, 125, 159, 179, 181, 182,
186
Antioch, Battle of (1098) 2
Archery, Crossbows 12-13, 35, 37, 164-165, 167-170, 175, 195
Archery, Longbows 5-7, 94, 118-119, 120, 124-125, 127-128, 140, 143-144, 161-163,
167-172, 174, 175, 180, 181, 183, 184-187, 189-192, 194-195
Arques, Battle of (1303) 2, 6, 23-32, 46, 48, 110, 170, 191, 192, 194, 195
Arques, Treaty of (1326) 101
Arsoef, Battle of (1191) 2
Artevelde, Jacob van 145
Artillery, Gunpowder 167, 194
Artillery, mounted crossbows 34-35
Artillery, trebuchets 34-35, 181, 192
Artois, Robert of 10, 14, 15, 17-20, 22, 109-110, 139, 155
Athens 6, 59, 61
Athis-sur-Orge, Peace of (1305) 33, 100-102
Auberoche, Battle of (1345) 6, 8, 188-189
Auray, Siege of (1342) 139
Austria, Albrecht, Duke of 130
Austria, Leopold, Duke of 130, 188
Avesbury, Robert of (chronicler) 116, 179, 181, 184
Axpoel, Battle of (1128) 2
B
Baker, Geoffrey le (chronicler) 68, 72, 95, 122, 126-128, 140-142, 159, 162, 163, 166,
170-174, 179-181, 186, 187, 191, 193
Balliol, Edward 113-117, 119, 120, 124, 176-178, 196
Balliol, John 113
Bannockburn, Battle of (1314) 6, 52, 57, 66-87, 110, 112, 115, 126, 127, 191-196
Barbour, John (narrative poet) 52-57, 68-85, 115, 195
Baston, Robert (narrative poet) 69, 71
Bavaria, Lewis IV, Duke of, Holy Roman Emperor 130
Bayonne, Jacques de 26-30, 110
Beauchamp, Guy, Earl of Warwick 70, 86-87
Beauchamp, Thomas, Earl of Warwick 177, 178
Beaumont, Henry of 74, 113
Beaumont, Hugh of, Count of Moray and Buchan 177
Beeler, John 3
Beke, Jan (chronicler) 148
Bel, Jean le (chronicler) 103, 140, 150, 152, 159, 161, 169-171, 173, 175, 179-181, 183,
194
Benedict XII, Pope 137-138
Benhale, Robert 124
Berchem, Willem van (chronicler) 159
Berg, Adolph VIII, Count of 152
Bern 130-136
Bertrand, Robert 107, 110
Berwick 67, 114-115, 121, 176
Berwick, Siege of (1333) 176
Bethune, Robert of, Count of Flanders 100-2
Black Death 7
Bliese, John R.E. 196
Blois, Charles of, Count of Brittany 139-144, 172, 174, 190
Bohemia, John the Blind, King of, and Count of Luxembourg 152, 165, 172, 174
Bohun, Sir Henry 75
Bohun, Humphrey of, Earl of Hereford and Essex 75, 76, 89, 94-96, 98
Bohun, William, Earl of Northampton 139-144, 166, 173, 190, 193, 196
Bois of Crécy 160, 193
Boniface VIII, Pope 10, 23-24, 49
Bonn, Werner de (chronicler) 150, 152
Bordeaux 23, 145
Boroughbridge, Battle of (1322) 7, 86-99, 110, 112, 115, 126, 127, 191-196
Bouvignes 145, 151
Bouvines, Battle of (1214) 2, 109
Bower, Walter (chronicler) 68-71, 73, 76-77, 79, 81
Brabant, Godfrey, Duke of 10
Bradbury, Jim 162-163
Brest 139, 141, 155
Breve chronicon de Flandriae 159
Bridlington, Auctor of (chronicler) 91-93
Brienne, Walter of, Duke of Austria 7, 59-65, 110
Page 212
Brittany 138-140, 155, 156, 189
Brittany, John III, Count of 138
Bruce, David 113, 114, 120, 126, 177-180, 182, 183, 186
Bruce, Edward 67, 73
Bruce, Robert 49
Bruce, Robert (King Robert I of Scotland) 49-57, 66-87, 89, 91-92, 112-113, 115, 126,
191-193, 196
Bruges 9, 10, 13, 37, 43, 101, 105
Brusthem, Jean de (chronicler) 15, 197
Brut, The 69, 89-95, 98, 116, 117, 125, 159, 173
Burckhardt, Jacob 129
Burgundy, Eudes, Count of 25-26
Burne, Alfred H. 1, 140, 156, 158, 162, 176
Burton 90-91
Burton, Thomas of (chronicler) 68, 71-72, 119, 122, 159, 161, 162, 167, 176, 179-180,
183, 186, 194, 195
C
Caen 157
Calais, Siege of (1346-47) 6, 110, 145, 160, 176-180, 186, 190
Carlisle 50, 51, 89
Cassel 23, 27
Cassel, Battle of (1328) 6, 7, 100-111, 136, 174, 175, 191, 192, 195, 196
Cassel, Robert of 101, 105, 107, 110
Casualties 18, 19, 30, 33, 43, 55, 56, 82, 83, 102, 108, 120, 125, 135, 150, 153, 169,
170, 173-174, 186, 188-189
Catalan Company 6, 58-65, 191-193
Cavalry, Dismounted 15-16, 28, 31, 62-63, 77-78, 84, 96, 98, 119, 126-127, 142-143,
161-162, 187, 191, 196
Chandos, Herald of (narrative poet) 159, 171
Charge, Cavalry 17-20, 28, 29, 38, 46, 47, 55, 56, 62-65, 79-81
Charlemagne 129, 145
Charles IV, King of France 101, 102
Charnys, Geoffrey 142, 143
Châtillon, Gaucher de 25, 26, 107, 110
Châtillon, Jacques de 20
Chieti, Philip of 33, 40
Chronicle of the Civil Wars of Edward II 93
Chronicon comitum Flandriae 11, 13-14, 16, 21, 37, 103, 106, 108, 159, 160, 164, 170
Chronicon de Lanercost 68, 77, 79, 81, 83, 84, 89, 90, 92-96, 116, 119, 124, 159, 180-
182
Chronicon Rotomagensi 22
Chronique anonymé Française finissant en MCCCLVI 40, 43
Chronique Artésienne 11, 13, 27, 29, 40
Chronique de Flandre 11, 25-28, 30, 31, 103, 106, 107
Chronique de l'abbaye de Saint-Trond 150, 152, 159
Chronique Morée (Aragonese Version) 60-62, 64
Chronique de Saint-Denis 42, 45
Chronique des quatre premiers Valois 159
Chronique des Pays-Bas, de France, d'Angleterre et de Tournai 30
Chronique Normande de xive siècle 37, 44, 159, 161, 170-172, 180
Chronique Parisienne anonymé de 1316 à 1339 103, 108-110
Chronique Tournaisienne 44
Chronographia regum Francorum 11, 17, 39, 41, 42, 44, 47, 103, 104, 159, 167, 169,
170
Clare, Gilbert of, Earl of Gloucester 75, 76, 79, 81-82
Clement VI, Pope 155
Clifford, Robert of 74-75
Clifford, Roger of 94-96
Comyn, John 50
Conflictus Laupensis 131-136, 195
Coninck, Pieter de 13, 43, 44, 47
Constantinople 59
Contamine, Philippe 3-4, 102
Courtrai, Battle of (1302) 2, 6, 7, 9-23, 25, 27-31, 33, 37, 38, 44-48, 56, 68, 80, 84, 100,
102, 103, 107, 109-111, 127, 145, 152, 159, 170, 175, 191-197
Crécy, Battle of (1346) 6, 7, 10, 99, 140, 143, 155-176, 178, 181, 186-187, 190, 191-197
Cressy, Sir Edward 6
Crisis of 1341 138
Cronica de Berno 131-132
Crossbowmen, Genoese 158, 164-165, 167-170, 175, 195
Cumberland 93, 178
Cuvelier (chronicler) 140, 143
D
Dagworth, Sir Thomas 190
Dahmus, Joseph 6, 158
Dalry, Battle of (1307) 51
Dammartin, Renaud of, Count of Boulogne 44, 48
Dampierre, Guy de, Count of Flanders 9, 23, 32
Daniels, E. 1
David and Goliath 10-11, 124, 128
Deken, Willem de, Mayor of Bruges 102
Delbrück, Hans 1, 34, 131, 162
Desnouelles, Jean (chronicler) 37, 40, 44
Despenser, Sir Hugh the Elder 87-89, 95
Despenser, Sir Hugh the Younger 87-89, 95
Diessenhoven, Heinrich von (chronicler) 131, 150
Digard, Georges 19
Page 213
Dinant 145, 151
Disinherited 113-125, 191
Ditches and Pits 13, 15, 17, 19-22, 42, 45-46, 53-57, 72-72, 81-82, 84, 126, 127, 141,
144, 149, 150, 152, 154, 164, 174, 192, 193
Douglas, Sir Archibald 114, 122, 125, 127
Douglas, James, Lord of 52, 73-75, 78, 79
Douglas, William 177, 182, 184
Dupplin Moor, Battle of (1332) 6, 85, 94, 112-120, 126, 128, 143, 176, 191, 192, 194-
196
Durham 112, 178-179, 181-183, 186, 193
Dysentary 176
E
Earn River 117, 118
Edinburgh 69, 178
Edward I, King of England 5-6, 9, 49-51, 66, 86
Edward II, King of England 7, 50-52, 57, 66-91, 95, 110, 112
Edward III, King of England 6, 86, 112-116, 120-127, 137-140, 155-175, 177, 178, 189,
193, 196
Edward the Black Prince, Prince of Wales 6, 162, 171
Effectives 12, 13-14, 26-27, 34, 35, 53, 58-61, 70, 71, 74, 90, 110, 116, 118, 121, 122,
132-134, 139-141, 148, 151, 161, 164, 173, 176, 180, 181, 186-190, 192
Erben, W. 1
Erbonne, Battle of (1328) 147
Erlach, Rudolf von 134, 196
Esplechin, Treaty of (1340) 137, 138
Eulogium historiarum 69
F
Falkenburg, Dietrich IV, Lord of, and Burgrave of Zeeland 152, 153
Falkirk, Battle of (1298) 7, 49-50, 77, 191
Fatigue 39-40, 44, 46-47, 74, 83, 173, 184
Fife, Duncan, Earl of 116
Flanders 9-48, 100-111, 136, 145, 146, 148, 157, 189, 191
Flanders, Joan of 139
Flor, Roger de, Leader of the Catalan Company 58, 59
Floricamp, Abbey of 149
Flote, Pierre 9, 10
Fordun, John (chronicler) 68, 71, 115, 120, 124
Frederick II, Holy Roman Emperor 58
Freiburg 130-136, 191, 192
Friesland 145-146, 148-150, 154
Froissart, Jean (chronicler) 22, 103, 105, 140, 148, 149, 159, 161-163, 166, 168, 169,
172, 173, 179-187
Fuller, J.F.C. 6, 158
Funck-Brentano, Frantz 20
G
Gascony 23, 24, 86, 144, 155, 156, 189
Gaveston, Piers 67, 86-88, 95
George, Herbert B. 162
Gesta abbatum St. Trudoniensum 21
Ghent 9, 11, 23, 31, 101, 138
Glen Trool 51, 52
Godedags 10, 12-13, 28, 31, 35
Golden Spurs, Battle of see Courtrai, Battle of
Grandes chroniques de France 11, 13, 14, 103, 105, 109, 140, 148, 159, 160, 167, 169,
170
Gray, Thomas (chronicler) 52, 69, 75, 77-79, 81, 84
Gregoras, Nikephoros (chronicler) 60-65
Guelders, Reinhold III, Duke of 148, 152
Guiart, Guillaume (narrative poet) 11, 25, 29, 33, 34, 36-38, 46, 197
H
Habsburgs 129-131
Hainault 24, 32, 146, 148
Hainault, Jean of 10, 148, 172-173
Hainault, Holland, and Zeeland, William III, Count of 7, 25, 26, 32, 35, 107, 110, 145,
146, 148-150
Halberds 188-189, 197
Halidon Hill, Battle of (1333) 6, 85, 94, 114-115, 120-128, 140, 143, 176, 177, 187, 191,
192, 194-196
Hanseatic League 146
Harclay, Andrew, Sheriff of Cumberland 88-89, 92-99, 191, 192, 196
Hardy, Robert 162
Hautepenne, Arnoul de, Lord of Billereau and Boilhe, Mayor of Liège 152
Hedges 183, 185, 187
Heinsburg, Dietrich of, Count of Looz 152
Hemricourt, Jacques de (chronicler) 150
Hennebout, Siege of (1342) 139
Henry I, King of England 157
Henry III, King of England 86
Henry V, King of England 6
Hocsem, Jean de (chronicler) 18, 150-152
Hoesselt, Battle of (1328) 147
Holland It, 24, 25, 32, 146, 148
Holy Roman Empire 129, 130, 145-148
Huy 145,147,151
I
Isabella, Queen of England 112-113, 120
Isle, Bertrand, Count d' 189
Ives (chronicler) 40
J
John, King of England 86
John XXII, Pope 87, 103, 107
Jülich, Wilhelm V, Marquis of 152
Page 214
Jülich, William of 9, 13, 23-28, 30-33, 40, 43-45, 47, 48, 196
Justinger, Conrad (chronicler) 132, 134
K
Keegan, John 1
Keith, Robert 74, 81
Kephissos, Battle of (1311) 6, 58-65, 110, 191-195
Kinghorn 114, 116-118
Klerk, Jan de (chronicler) 159
Knighton, Henry (chronicler) 68, 115, 140-143, 159, 163, 179-181, 193
Kohler, G. 1
Kronyk van Vlaenderen 13
L
Lacy, Henry de, Earl of Lincoln 87
La Marck, Adolphe I de, Prince-Bishop of Liège 147
La Marck, Adolphe II de, Count of 152
La Marck, Englebert de, Prince-Bishop of Liège 7, 145, 147, 150, 152
Lancaster, Thomas, Duke of 7, 70, 86-99, 110, 192
La Pole, William de 137-38
La Roche-Derrien, Battle of (1347) 6, 8, 188-190
Laupen, Battle of (1339) 7, 129-136, 191. 192, 195, 196
La Zouche, William, Archbishop of York 178, 181, 196
Lescot, Richard (chronicler) 103, 148, 159
Lessines, Siege of (1303) 24-25
Liber Pluscardensis 68, 71, 115, 118, 119, 120, 159, 179, 180, 182
Lichtenstein, Otto von (narrative poet) 11-12
Liddel, Attack on (1346) 181-182
Liddell Hart, Basil H. 5
Liège 145, 146, 150-154
Lille 9, 23, 33, 40, 42, 104, 105
Line, Infantry 7, 17-18, 28, 35-36, 46, 47, 62-65, 74, 77-78, 93-98, 118-119, 122-124,
127, 133-136, 142-143, 149-150, 153, 161-162, 170, 171, 174, 183-187, 191-192, 195-
197
Lloyd, E.M. 162
London, Tower of 179, 190
Lot, Ferdinand 1, 20, 34
Loudon Hill, Battle of (1307) 6, 7, 49-57, 66, 68, 72, 80, 86, 126, 127, 135, 191, 193-
196
Lucas, Henry Stephen 34
Lys River 13-15, 192
M
Malestroit, Treaty of (1342) 155
Margaret of Norway, Queen of Scotland 49
Marshes and Morasses 53, 55, 62, 83-84, 192
Marquette, Treaty of (1304) 33
Maury, Sir Walter 139, 140, 143
Melton, William of, Archbishop of York 113, 115
Methven, Battle of (1307) 51
Minot, Laurence (narrative poet) 69, 115, 179
Moncado, Francisco de 63-64
Mons-en-Pévèle, Battle of (1304) 2-3, 6, 7, 25, 30, 32-48, 100, 102, 103, 107, 109, 161,
170, 191-195, 197
Moray, Andrew, Earl of 114, 177
Morlaix, Battle of (1342) 6, 137-144, 155, 170, 191-196
Morlaix, Siege of (1342) 139-140, 174
Morris, John E. 98, 162, 191
Mortgarten, Battle of (1315) 7, 8, 130, 188-189
Mortimer, Roger 112-113
Mountford, John of 138, 139
Muisit, Gilles le (chronicler) 11, 15, 17, 103, 107, 159-160, 162, 169, 172, 175
Muntaner, Ramón (chronicler) 60-65
Murimuth, Adam (chronicler) 68, 123, 125, 126, 140-142, 159-161, 171, 179
N
Namur, Guy of 9, 13, 23-25, 32, 33, 35, 40
Namur, Henry of 33, 40
Namur, John of 101
Namur, Robert of 40
Namur, Wilhelm V, Count of 148, 152
Nangis, Guillaume de (chronicler), Continuators of 11, 14, 103, 106-108
Nantes 138, 139, 155
Napoleon III 19
Neuenberg, Mathias von (chronicler) 150-152
Nevers, Louis of, Count of Flanders 100-105, 108, 145, 172, 174
Neville, Ralph 178, 181, 196
Neville's Cross, Battle of (1346) 6, 69, 85, 163, 176-187, 191-196
Normandy 155-156
Northampton, Treaty of (1328) 113, 120-121
Northumberland 112, 114, 121
Notger, Prince-Bishop of Liège 146
Nowé, Henri 19
O
Oman, Sir Charles 1, 4, 19, 45, 109, 110, 162
Orations, Battle 15-16, 76, 119, 127, 163, 196-197
Oriflamme 35, 40, 104
Orrible, Pierre l' 15
Otto II, Holy Roman Emperor 146
Ouse River 93-96, 192
Outremeuse, Jean d' (chronicler) 150
Oxford Chest The 11, 21
P
Palaeologus, Andronicus II, Emperor of Byzantium 57
Page 215
Paris 11, 102, 103, 137, 139, 156-158
Paris, Geoffroi de (narrative poet) 11, 21
Paris, Jean de (chronicler) 36-38, 42, 193
Penthiève, Joan of, Duchess of Brittany 138-139, 141
Percy, Henry 178, 181, 196
Perroy, Edouard 102, 109, 110
Perth 66, 67, 114, 116, 117, 178
Philip IV (the Fair), King of France 9, 10, 13, 23-24, 26, 32-35, 38, 40-47, 49, 100, 192
Philip V, King of France 100, 101
Philip VI, King of France 102-110, 137, 138, 146, 155-175, 177, 178, 182, 189, 192
Philippa of Hainault, Queen of England 114, 181
Picardy 23, 104, 145
Pirenne, Henri 12, 19-20, 34
Poitiers, Battle of (1356) 6, 7, 10, 140, 163, 176
Pride 17, 20-21, 71, 166-167, 175
Procurator, Willem (chronicler) 12, 103
Prostitutes 176
Q
Quimperlé, Battle of (1342) 139
R
Ramsey, Alexander 177
Randolph, John, Earl of Moray 177
Randolph, Thomas, Earl of Moray 73-76, 80, 84
Reading, John of (chronicler) 179, 182
Rebdorf, Heinrich von (chronicler) 150
Réits d'un bourgeois de Valenciennes 11, 35, 44, 148-150, 152, 153, 159, 162, 163,
166, 167, 171, 180, 181, 185
Renesse, John of 13, 15-16, 196
Rennes, Siege of (1356) 176
Rijmkroniek van Vlaenderen 11, 103
Ripon 92, 97
Roland 11, 28, 185
Rouen 157, 173
S
St. Denis 11, 35, 103, 104
St. Omer 11, 25-31, 104, 105, 192
St. Pol, James, Count of 9, 10
St. Pol de Léon, Battle of (1346) 190
St. Trond 145, 147
St. Winoksbergen 27-31
Samson, Thomas 179-181, 185, 187
Schiltrom 78, 94, 98
Schwyz Canton 129, 130
Selby, Walter of 181-182
Seton, Sir Alexander 77, 114
Seton, Thomas 114, 121
Sluys, Battle of (1340) 6, 137, 145, 176, 178
Stadtchronik of Bern 132
Stafford, Ralph, Earl of 113, 119
Staveren, Battle of (1345) 7, 145, 148-150, 152, 153, 191-196
Steppes, Battle of the (1213) 2
Stewart, Walter 73, 78
Stiermarken, Ottokar von (chronicler) 11-12
Stirling Bridge, Battle of (1297) 7, 49
Stirling Castle 66, 67, 69-72, 74
Stoke, Melis (chronicler) 11
Stratford, John, Archbishop of Canterbury 137
Strathbogie, David, Count of Atholl 113, 177
Strayer, Joseph 19
Sumption, Jonathan 127, 162-163
Swiss Confederation 129-131
Switzerland 129-136
T
Talbot, Richard, Lord of Mar 113, 177
Tongeren 147, 150
Tournai 11, 24, 104, 105
Tournai, Siege of (1340) 6, 32, 110, 137, 138, 145, 175, 176, 178
Tout, T.F. 140, 191
Trent River, Battle of (1322) 90-91
Trokelowe, John of (chronicler) 68, 70, 77-85, 191
U
Umphraville, Sir Ingraham 79
Unterwalden Canton 129, 130
Uri Canton 129
Utrecht 146, 148
V
Valence, Aymer de, Count of Pembroke 50-57, 86-87
Valengin, Gerhard of 132, 135
Valois, Charles of 44, 48
Vannes 139, 155
Velthem, Lodewijk van (narrative poet) 10-15, 21
Venette, Jean de 159, 160, 162, 175, 179, 180, 184
Verbruggen, J.-F. 1-4, 7, 20, 30, 176
Villani, Giovanni (chronicler) 10, 21, 103, 159, 160, 162, 175, 179, 180, 184
Vita Edwardi II 68-72, 75, 77-79, 83, 84, 89-92, 96, 97
Vita et mors Edwardi II 91
Vottem, Battle of (1346) 7, 145, 148, 150-154, 191-196
W
Wagon-Fortresses 35-36, 39-40, 46, 161-163, 174, 193, 194
Wallace, Sir William 49
Ward, Symond, Sheriff of York 93
Waremme, Battle of (1328) 147
Warenne, John, Earl of 86
Walsingham, Thomas of (chronicler) 68, 77, 84, 125, 159, 191

Common questions

Powered by AI

Robert Bruce capitalized on English tactical disadvantages during the Battle of Bannockburn by effectively using defensive positioning and exploiting the terrain. He employed infantry in tightly packed, spear-wielding formations, known as schiltrons, to combat the English cavalry, which struggled on the uneven ground filled with concealed pits and trenches prepared by the Scots to disrupt horse movements . The Scots held a defensive position that forced the English to attack uphill, further hampering their mobility and effectiveness . Additionally, the English forces were less coordinated, partly due to Edward II's uncertain leadership, which contrasted sharply with Bruce's strategic planning and unity among his troops . Bruce's infantry tactics at Bannockburn effectively mirrored those he had previously employed at the Battle of Loudon Hill, demonstrating his understanding of how to neutralize cavalry despite being outnumbered .

The morale of English and French troops played a significant role in the outcomes of their battles. At the Battle of Courtrai in 1302, the Flemish troops, encouraged by John of Renesse's motivational speech, fought vigorously against the French, leading to a significant victory despite facing a numerically superior enemy . In contrast, during the Battle of Crécy in 1346, the English troops, leveraging a strong defensive position and employing long-range archery, demoralized the charging French cavalry. The French attacks became disordered, allowing the English to capitalize on the confusion and secure a decisive victory . The morale was also significantly bolstered by effective use of tactics, such as those seen at Bannockburn, where Scottish forces maintained high spirits and tactical discipline to defeat the English . Overall, high morale, often inspired by leaders' speeches and tactical superiority, consistently contributed to the success of troops in battle .

Historical narratives of key battles between French and Flemish forces often highlighted differences in perception and emphasis based on tactical details and outcome interpretations. In the depiction of battles such as Courtrai in 1302, narratives varied on the success of infantry tactics. The Flemish, who were largely infantry-based, were perceived by some accounts as having adeptly used disciplined infantry formations to withstand French cavalry charges, leading to their victory . In contrast, French sources often downplayed the decisiveness of these battles, focusing instead on the limited long-term political impact of such engagements, characterizing them as inconclusive despite tactical wins . This has led to a divergence in understanding the role of technological elements like the longbow, as some narratives attributed more importance to disciplined infantry lines rather than technological superiority . Overall, the differences in narratives stemmed from the varying emphasis on military tactics and the broader political implications of these battles during the period .

Scottish forces adapted to English numerical superiority by emphasizing strong defensive formations and selecting advantageous terrain that the English were unfamiliar with, such as the battlefield at Bannockburn which featured ditches and marshes that hindered English cavalry charges . The Scots formed defensive formations, often "back to back with their spear points outward," to counter English tactics and relied on their knowledge of the terrain, fatigue management, and hunger alleviation to maintain morale and effectiveness in battle . Despite being outnumbered, these formations allowed the Scots to withstand initial English attacks and counterattack effectively when the opportunity arose . However, these tactics were not always successful in every engagement as the English learned to dismount their knights, forming solid defensive lines with archers on the flanks that curtailed Scottish numerical advantages, which was evident in engagements like the battles of Dupplin Moor and Halidon Hill .

The French armor and tactics at the Battle of Mons-en-Pévèle played a significant role in their performance and ultimate victory. The French army, led by Philip IV, utilized a large number of knights complemented by infantry; this combination allowed them to mount a quick recovery after initial Flemish success in disarraying the French camp . The French had learned some lessons from previous battles, such as Courtrai, about engaging solid infantry formations with knights, but they largely continued traditional tactics. This can be seen in their use of cavalry to counter the Flemish attacks, which were initially successful in reaching the French camp . The French also had the advantage of a well-prepared site, having arrived first at the battlefield and setting up their forces strategically . Additionally, their use of artillery, including trebuchets and mounted crossbows, contributed to their battlefield tactics, although the precise impact of these is difficult to measure due to the lack of detailed information . The support of Philip IV himself and his marshals also provided strong leadership, which was crucial in turning the tide against the Flemish forces . Despite the large Flemish numbers, the French tactical response and leadership were decisive in their victory at Mons-en-Pévèle .

The Flemish forces at the Battle of the Golden Spurs utilized strategic advantages such as selecting the battlefield and preparing it with hidden ditches to disrupt the French cavalry charges . The Flemings fought on foot, which proved beneficial on the terrain and allowed them to withstand the heavily armored French knights, who were disadvantaged in their offensive due to fatigue and heavy armor . Additionally, the Flemish forces were motivated by a strong sense of local autonomy and superiority after early victories . In contrast, the French faced several strategic challenges. Their army was composed largely of mounted knights, which was less effective on the prepared and restrictive battlefield . The French cavalry charges over the hidden ditches resulted in significant disarray and high casualties, with contemporary estimates noting that between 40-50% of the French cavalry were lost . The leadership of Robert of Artois was also criticized, as his tactical decisions, including charging without regard to the field's preparations, were seen as a primary cause for the defeat . The tactical missteps, including the failed cavalry charges over ditches, and the strategic misjudgments in confronting a well-prepared and strategically placed Flemish military contributed to the French defeat, despite their numerical superiority and martial reputation .

Flemish leadership played a crucial role in sustaining their resilience during engagements with French forces by choosing effective defensive strategies, such as taking advantageous positions and utilizing terrain to their benefit. Leaders like William of Jülich, Guy of Namur, and John of Renesse were pivotal in coordinating Flemish forces, although often outnumbered . During key engagements, such as the battles of Courtrai and Mons-en-Pévèle, the Flemish demonstrated strong defensive maneuvers and initially successful offensives, such as well-ordered charges that led to early successes . However, in instances where tactical decisions led to disorganization, such as breaking formation in ill-advised attacks, Flemish forces suffered because they could not maintain defensive cohesion, ultimately leading to defeat . Despite this, the Flemish leadership's ability to withstand prolonged assaults and fatigue was significant, often forcing French troops to adapt their battle plans and demonstrating an ability to exploit French weaknesses, such as their reliance on heavy cavalry .

The Flemish success in the battle was initially due to their strong defensive position and the resilience of their infantry line, which held firm against repeated French cavalry charges . They used strategic fortifications like the wagon-fortress to protect their flanks and rear, preventing easy disruption by the French forces . Additionally, the Flemish employed tactical counter-attacks that effectively countered French maneuvers . The Flemish also benefited from the fatigue and disarray among the heavily armored French knights, who became quickly exhausted and were unable to maintain a sustained offensive . However, the early Flemish success eventually contributed to their downfall, as a disorganized attack into the French camp disrupted their defensive formation, allowing the French to rally and counter-attack decisively . Despite causing significant French casualties, the Flemish were ultimately unable to capitalize on their initial advantages, leading to their defeat as the French forces regrouped and launched a successful counter-attack .

Environmental conditions played a crucial role in medieval battles, affecting strategies and outcomes significantly. For instance, the battle of Bannockburn in 1314 illustrates how superior tactics in defense leveraging the landscape could lead to victory. The Scots effectively used the ground to their advantage, choosing a defensive position that neutralized the numerical superiority of the English . Similarly, the Battle of Morgarten in 1315 saw Swiss rebels utilizing the mountains to ambush and defeat the Austrian forces, highlighting how topography could dictate battle success . In battles such as Dupplin Moor and Halidon Hill, the landscape and environmental conditions forced tactical innovations, such as the dismounted English knights forming effective defensive lines against larger armies . Overall, the ability to use environmental conditions effectively often determined the winners in medieval conflicts, as terrain often provided defensive advantages that could counteract superior numbers or advanced weaponry.

Intelligence gathering and deception played a significant role in the French military strategy as indicated by their use of strategic site preparation and concealment during battles. For example, at the Battle of Courtrai, the Flemings strategically prepared the field by digging concealed ditches, which disrupted the French cavalry's ability to charge effectively, leading to heavy French casualties . This tactic of using terrain manipulation as a form of deception demonstrates the importance of non-combat strategies in medieval warfare. Furthermore, the French strategized their position by surveying the battlefield for days before engaging, although their ultimate defeat highlighted the effectiveness of Flemish defensive preparations . The significance of intelligence and deceit is further highlighted by the fact that Flemish military leaders successfully concealed these ditches from the French until it was too late, which was noted by multiple contemporary sources . Overall, these strategies illustrate the critical role of intelligence and deception in shaping the outcomes of medieval battles.

You might also like