Navigation of the pdf edition

It is assumed that readers will be making use of Adobe Reader and recommended that
they view the book as a two page spread by going to View - Page Display - Two Page
View. This will produce a view with odd-numbered pages on the right, as in a printed
book. To reach a particular page enter the number in the page number search box in
the toolbar if it is an odd number, but add 1 if it is an even number.
It is also possible to navigate through the document by opening the Comment
section on the right hand side of the screen and clicking on the desired page. It comes
supplied with links to the sections listed on the Contents page, but you can add as
many of your own as you wish.
THE BATTLE OF

HASTINGS
1066

M. K. LAWSON
First edition published 2002
Second edition published 2003
Third (pdf) edition published 2016
© M. K. Lawson 2002, 2003, 2016

The right of M. K. Lawson to be identified as the Author

of this work has been asserted in accordance with the
Copyright, Designs and Patents Act 1988.

All rights reserved.

 For
Rosamund, Blant, Irene
and Lilian

But to live outside the law, you must be honest;

I know you always say that you agree.

Bob Dylan Absolutely Sweet Marie

Choice Outstanding Academic Title

‘Highest recommendation. A marvelous text for classes in medieval

military history and historiography’, Choice

‘blows away many fundamental assumptions about the battle of

Hastings... an exciting and indispensable read’,
Professor David Bates, University of East Anglia, former Director
of the Institute of Historical Research

‘M.K Lawson knows more about the Battle of Hastings than most who
have lived since the twelfth century; this is the book that he owed to
history and he has paid his debt with aplomb’.
Times Literary Supplement

‘a deeply considered and provocative study, merits repeated reading’

Professor John Hudson, University of St Andrews

‘The greatest of the many strengths of M. K. Lawson's careful and

thorough study of the battle is the close attention he gives to source
criticism. For in the end, his aim is not to offer definitive answers but to
open up questions and possibilities, including about some surprising areas
where consensus has long reigned. The result is a very valuable book not
just about the battle itself, but about the historical process... Extensively
illustrated from contemporary sources and the author's own photos of the
site, thoroughly documented, and clearly written, this must remain the
definitive book on this famous battle until more evidence comes to light.’
Journal of Military History, April 2003.

‘M.K. Lawson does justice to all that is known of the battle of Hastings
and offers a fascinating, authoritative, well-constructed critique of all the
sources, as well as a thorough examination of the battlefield and of each
stage of the battle itself’, The Spectator, March 2004.
 CONTENTS

Abbreviations 9

Foreword 11

Foreword to the Third Edition 13

Introduction 17

Chapter One. The Prelude 21

Chapter Two. The Sources. Part One 43

Chapter Three. The Sources. Part Two 75

Chapter Four. The English Army 103

Chapter Five. The French Army 141

Chapter Six. The Battle 161

Chapter Seven. Epilogue 195

Appendix. The death of Harold and the authority of the

present form of the Bayeux Tapestry 203

Genealogies 1-4 215

Illustrations 219

The Lancelot Engravings of the Bayeux Tapestry 265

The Montfaucon Engravings of the Bayeux Tapestry 273

Select Bibliography 285

Index 309
 ABBREVIATIONS

Full details of these works will be found in the Select Bibliography.

AC Baudri of Bourgeuil, Adelae Comitissae, ed. Hilbert.

ANS Anglo-Norman Studies.
ASC Anglo-Saxon Chronicle, ed. Plummer.
Asser Asser, Life of Alfred, ed. Stevenson.
Barlow Barlow, Edward the Confessor.
Bernstein Bernstein, The Mystery of the Bayeux Tapestry.
BM The Battle of Maldon, ed. Scragg.
BR Brevis Relatio, ed. van Houts.
BT The Bayeux Tapestry, ed. Wilson.
B&W Brooks and Walker, ‘The Authority and Interpretation of the
Bayeux Tapestry’.
Carmen Carmen de Hastingae Proelio, ed. Barlow.
De Vries De Vries, The Norwegian Invasion of England in 1066.
DB Domesday Book.
EE Encomium Emmae, ed. Campbell.
EHD1 English Historical Documents Volume 1, ed. Whitelock.
EHD2 English Historical Documents Volume 2, ed. Douglas and
Greenaway.
EHR English Historical Review.
ESRO East Sussex Record Office.
FNC Freeman, The Norman Conquest.
GG William of Poitiers, Gesta Guillelmi, ed. Davis and Chibnall.
GND William of Jumièges, Gesta Normannorum Ducum, ed. Van
Houts.
GP William of Malmesbury, De Gestis Pontificum, ed.
Hamilton.
GR William of Malmesbury, De Gestis Regum Anglorum, ed.
Mynors, Thomson and Winterbottom.
HA Henry of Huntingdon, Historia Anglorum, ed. Greenway.

9
 ABBREVIATIONS

HÆ Orderic Vitalis, Historia Æcclesiastica, ed. Chibnall.

HH Henry of Huntingdon.
Illus. Illustrations following p. 219.
JW John of Worcester, Chronicle, ed. Darlington and McGurk.
LKE The Life of King Edward, ed. Barlow.
M&M Carmen de Hastingae Proelio, ed. Morton and Muntz.
ODNB Oxford Dictionary of National Biography.
OS Ordnance Survey.
OV Orderic Vitalis.
RR Wace, Roman de Rou, ed. Holden. All line numbers are
those of Book III, unless otherwise stated.
Survey English Heritage, An Earthwork Survey and Investigation of
the Parkland at Battle Abbey, East Sussex.
TNE The Normans in Europe, ed. van Houts.
TRHS Transactions of the Royal Historical Society.
WJ William of Jumièges.
WM William of Malmesbury.
WP William of Poitiers.
Writs Anglo-Saxon Writs, ed. Harmer.

10
 FOREWORD
I have incurred a number of debts in the preparation of this book. Two
friends of long standing, Sue Reedie and Dr John Hudson, read much of it
in draft, the former suggesting valuable amendments and taking me to
task for excluding the possibility that the designer of the Bayeux Tapestry
was a woman, the latter offering much-valued encouragement and a
number of acute corrections. Many of the illustrations have been
reproduced from volumes in the possession of the London Library in St
James’s Square, and the professionalism and care of their staff, and of
those of the History Faculty Library in Oxford and the East Sussex
Record Office in Lewes, have been greatly appreciated. Two successive
librarians of St Paul’s School, Gillian Drew and Alexandra Aslett, went
to considerable trouble acquiring books and articles through inter-library
loan, while my colleagues Richard Barker, Chris Fry, William Lines and
David May have shown cheerful good humour in easing my encounters
with computers and scanners, and Penny Holmes and Ian Tiley have
given valuable assistance on digital imaging. Specialist advice on the
geography of the battlefield is owed to Paul Littlewood, who has walked
over it with me, as have my friend Peter Sammut and my head of
department Mike Howat; I am also especially grateful to Mike for never
giving my academic work anything less than his unreserved support. John
Davie has eked out my Latin, Marie-José Gransard my Old French, and
Jennifer Attia and Jenifer Ball my modern French. This is a St Paul’s
volume in another and particularly important way too: between 1987 and
1995 boys in the Medieval History G.C.S.E. sets visited Battle every year
as part of their coursework on eleventh-century warfare. They (for the
most part) charged enthusiastically up the hill when ordered (in some
cases recalling this joyfully years later, long after their teacher’s
carefully-considered words of wisdom had been forgotten), frequently
asked questions to which he had no answers, and not infrequently
produced work which would have been worthy of publication; without
them, it is certain that the following pages would never have been written.

11
 THE BATTLE OF HASTINGS

My debt to Dr George Garnett is scarcely less. As the G.C.S.E.

battlefield visits were coming to an end, he pressed me to address the
students of the Norman Conquest Special Subject in Oxford, thereby
obliging an initially-reluctant speaker to get hitherto ill-assorted thoughts
into a degree of order. It was this that brought me to earlier work on the
subject, and especially to that of the great historians of the second half of
the nineteenth century. To open Professor Freeman’s remarkable The
History of the Norman Conquest of England, or follow the bitter
controversy after his death between his opponent J.H. Round and his
champions T.A. Archer and Kate Norgate, is not only to read some of the
most detailed work ever published on the battle, but also to be forcefully
reminded of the qualities of scholars whose works are today read mainly
by specialists. When this book was within two months of going to press
my understanding of the present state of the Bayeux Tapestry was
transformed by Dr David Hill’s address to the Manchester Conference on
Harold II and the Bayeux Tapestry. Since then I have profited greatly
from the assistance of Dr Hill and John McSween, and especially from
their generosity in allowing me to publish one of their photographs of the
drawings of the Tapestry made by Antoine Benoît in 1729. Finally, it
gives me pleasure to record my debt to Professor James Campbell, the
value of whose friendship can be appreciated only by those who have
enjoyed it, and whose generous hospitality never failed during the years
in which the material discussed in these pages was being researched and
analysed.

Chessington, Surrey.
June, 2002.

12
 FOREWORD TO THE THIRD EDITION

Late in 2014 my book on the battle of Hastings, originally published by

the now defunct Tempus, went out of print. Although I had signed a
contract with The History Press for it to be reissued they declined to go
ahead on the grounds that sales did not justify it. This meant that the
publishing rights returned to me and allowed the production of the
present edition. With a page size of A4 and no limit on the number of
illustrations in colour, it is at least in these respects significantly better
than its predecessors, the paperback first edition of 2002 and the hardback
second of the following year, subsequently also issued in paperback. As
an academic work which is not principally concerned with the history of
art, it is rather unusual in requiring a large number of pictures, something
which arises partly from the importance of the Bayeux Tapestry, but
much more from the need to illustrate various features of a possibly
extensive battlefield which play a key role in some of the arguments.
Moreover, not only is the reader entitled to be able to see exactly what is
being argued about, but it is the case that many, and especially those who
live outside the United Kingdom, will never be able to visit Battle
themselves, and may therefore find the maps and photographs in this
book by no means its least valuable feature. Added to this is the fact that
those who do go are likely to find it a confusing place, for what may be
only a part thereof is within the guardianship of English Heritage, and
their visitor trail guides people round only a fairly small portion of that,
while exploration on the ground is complicated by the existence of large
patches of dense scrub and woodland. To an extent, then, this volume lies
in the same tradition as Charles Lemmon’s The Field of Hastings (4th
edition, 1970), which guided my own early sorties to the site in the mid
1980s.
I have never regarded it as primarily a piece of military history, nor
myself as a military historian. It is really an exercise in the use of
complex sources to establish as much of the truth as is feasible about a
complex event, which as it happens was a battle of great historical
importance. Although the intended audience is therefore mainly an

13
 THE BATTLE OF HASTINGS

academic one, during the composition I went to some trouble to make the
text accessible to the intelligent general reader, knowing full well that it
could never be made so to a popular readership. Being aware of the likely
reaction to this, my Introduction issued a perfectly explicit warning about
it, but of course that has not prevented criticism from those who have
found the content simply too difficult. It is enough to point out that as the
events of 1066 have always attracted the attention of writers of popular
history and will doubtless continue to do so, those who prefer that kind of
material will find it readily available.
As already stated, the main difference between this edition and its
forerunners lies in the range and quality of the illustrations. When it
comes to the battlefield, I have provided a colour image of all the places
of possible significance, with the exception of that of the Malfosse
incident at the end of the conflict, which occurred at a location not
certainly identifiable. While needing to change very little of the essence
of what I wrote fourteen years ago, I have also updated the references to
secondary literature. A large amount of new material has appeared on the
Bayeux Tapestry, and especially the results of the scientific investigations
carried out in the early 1980s and published in 2004, but while I have
noted some of the latest opinions the fact is that much of what is written
about this source centres on establishing the identity of its patron, its
supposed political bias and where it was made, all matters which tend to
be peripheral to its use as a source for the battle. Virtually all agree that it
dates from the second half of the eleventh century; it would be quite
another matter were this to be seriously challenged.
A significant development in recent years has been the increasing
availability of valuable material on the worldwide web. This includes not
only academic articles but also the important engravings of the Tapestry
published by Bernard de Montfaucon in 1729 and 1730, which are now
available in high-resolution from the webpage of the Library of the
University of Heidelburg, and which have been added to this book. The
very important drawings of 1729 by Antoine Benoît upon which they
were based still await publication more than a decade after Dr David Hill
drew attention to their importance and to their existence in the
Bibliothèque Nationale.
Another issue upon which there has been no progress in recent years is
that of whether the battlefield contains features visible today which are
connected with the conflict. To be specific, might the mounds which line
the stream to the south of the Battle Abbey ridge and the depression in the
ground on the slope immediately below the abbey buildings be connected
with the burial of the French dead? I am less hopeful about the former
today than I was in 2003, when I was less hopeful than I had been in
2002, but although it is most likely that the mounds are the result of much

14
 FOREWORD

more recent dredging of the stream, I was advised by a geographer that

once such features become vegetated they change very slowly, and that
therefore it is unwise to assume too much simply from their appearance. I
did not need to be told that it is also unwise for an early medievalist to be
too sure about anything. As for the depression to the south of the abbey, it
is exactly where one might expect some of the French dead to lie buried,
but could be the site of a later quarry. Only excavation can establish the
truth, and my attempts to interest English Heritage in this have failed.
Meanwhile, battle re-enactors disport themselves and heavy vehicles
drive around upon ground which may be of great historical significance.
As it is being issued as a free download, and as some who use the
internet mistakenly believe that anything available on it is in the public
domain, I should stress that this book remains within the protection of the
copyright laws until (in the United Kingdom) 70 years after my death,
and that making more use of the text than they permit and reproducing
the illustrations without my permission are just as illegal as they would be
in the case of a printed book. The only exceptions to this are materials
published in the eighteenth and nineteenth centuries which are now in the
public domain. I have issued it in this form and manner because it is the
role of historians to both practise their craft and to make the results
available to as wide an audience as possible; and because the worldwide
web provides opportunities for dissemination which no producer of a
physical book can equal.

Sherborne, Dorset.
April, 2016

15
 INTRODUCTION

On Saturday, 14 October 1066 one of the most famous of all military

conflicts took place upon and around the ridge in Sussex now occupied in
part by the town named after it: Battle. Fighting is said to have lasted
almost all day and into the evening; by its end, the last Anglo-Saxon
king, Harold II, lay dead, while the remnants of his defeated army fled
into the forest to their rear. The victor, Duke William of Normandy, was
to face sporadic and sometimes serious resistance for the next five years,
but never again did the English face him upon a major field of battle, and
there thus flowed directly from his triumph the long-term historical
consequences with which students of the Norman Conquest are so
familiar: the replacement not only of an English king by a French one,
but of the native aristocracy by his followers; the rejuvenation of the
English church by foreign ecclesiastics led by Archbishop Lanfranc of
Canterbury, an Italian by birth; the slow decline under William and his
successors of the powerful systems of government developed by the late
Anglo-Saxon state; the end of the close political links with Scandinavia
which had characterised the pre-Conquest period, and that reorientation
of England towards France which meant that for almost the next half
millenium, until Queen Mary lost Calais in 1558, English monarchs had
possessions across the Channel and policies often designed to retain or
expand them.
Yet its considerable impact on subsequent centuries is far from being
the only reason why the battle of Hastings has been a much-visited
subject, and why 1066 is still one of the most well-known dates in
English history. It also has undeniable popular appeal – King Harold
being struck in the eye by an arrow ranks alongside King Alfred’s

17
 THE BATTLE OF HASTINGS

(supposed) burning of the cakes as one of the better-known episodes of

the Anglo-Saxon period – and the steady stream of visitors to the part of
the battlefield under the guardianship of English Heritage, and the
popularity of reconstructions of the fight involving hundreds of warriors
equipped with replica weapons and armour, suggest that public interest in
it has seldom been greater. That interest is also stimulated by one of the
most remarkable artefacts to have survived from the entire Middle Ages,
the Bayeux Tapestry, which shows both the battle and (as has long been
thought) Harold and the arrow. The Tapestry has much valuable evidence
to offer, and nor is it the only broadly contemporary source of which this
is true. Indeed, Hastings is generally so well supplied with sources that
more is known about it than any other engagement fought in this country
not only before 1066 but for centuries thereafter. Not least among them is
the battlefield itself, much of it still open ground, and at least part of it
arguably identifiable with a location shown in the Tapestry.
All this would seem to render it easy to write the history of what
happened there, and writers and historians have not been slow to do so,
for both popular and academic audiences. Unfortunately, the very natural
desire of the audience to know and of the informer to inform has often
led to descriptions of the conflict wearing an appearance of certainty
which the nature of the primary sources actually does little to warrant.
Battles are confusing affairs, not least for those involved in them, and
even were there testimony from a number of eye-witnesses it would
probably be difficult to write a coherent and convincing account of
fighting which lasted over eight hours and involved thousands and
perhaps tens of thousands of men on each side. In fact, there is not a
single eye-witness to be had, and the primary sources that do exist,
relatively plentiful as they are, all suffer from significant limitations
which need to be clearly understood before use can be made of them.
What emerges from this process, as this book will attempt to
demonstrate, is not that there is nothing that can be known about the
battle of Hastings, but that there are many important things that cannot
and never will be known. Thus, seekers after easy hard facts – how many
men there were on each side, for example, and where they were
positioned at different points during the day – will not find them in the
pages that follow. What they will find is the relevant evidence laid before
them, along with often-complex discussions of sources and issues, many
of which are to a degree incapable of resolution. The battle may have
ended the Anglo-Saxon period, but for the historian it shares one of the
latter’s crucial features – that the difficulties of dealing with the evidence
mean that one often ends up not with clear answers so much as a range of
possibilities, not one picture made up from the jigsaw, but a series of
possible pictures, all with resemblances to each other, and all with

18
 INTRODUCTION

significant numbers of pieces missing. It is one of the central theses of

this book that since the late nineteenth century, and the formative work of
the scholars J.H. Round, Wilhelm Spatz, Sir James Ramsay and F.H.
Baring, too many writers have followed their lead in unjustifiably
narrowing the range of possible interpretations of this crucial battle, and
that it may in some respects have been a more fascinating conflict than it
has often been given credit for, both in terms of the numbers involved
and the ways in which the fighting was conducted.
If it is true that Hastings has at times attracted a great deal of attention, it
is also true that there has been little real debate about it since the days of
the men named above. Indeed, the flurry of interest in the Conquest
occasioned by the 900th anniversary in 1966 has been followed by the
appearance of important new editions of a number of the major primary
sources, and almost all are now available in reliable translations.1 There
has also been a great deal of valuable secondary work upon the events of
1066 generally, and there is certainly no sign of any diminution of interest
in the Bayeux Tapestry, although there is an urgent need for a proper
critical edition of it, based upon rigorous examination of its fabric.
However, in the last twenty years surprisingly little has been written about
the battle itself,2 with re-enactors tending to show more interest in the
details of the fighting than historians. It may be that this is because the
latter, on the whole, have come to the conclusion that there is little more to
be said. If so, it is one of the purposes of this book to suggest that they
have been mistaken.

1
The Brevis Relatio, the Carmen de Hastingae Proelio, and the works of Henry of Huntingdon, John of
Worcester, Wace, William of Jumièges, William of Malmesbury and William of Poitiers.
2
Since 1970 there has been R. Allen Brown’s article ‘The Battle of Hastings’ of 1978, and Jim
Bradbury’s book of the same title from 1998. The Battle of Hastings, ed. Morillo (1996) is a collection of
reprints of earlier articles and translations of primary sources, including the editor’s own short but
valuable essay from 1990.

19
 I

THE PRELUDE
In 1948, less than a decade after England had faced the prospect of an
invasion even more terrible and destructive than that of 1066, Hope
Muntz published her fine historical novel The Golden Warrior: the story
of Harold and William. Dedicated to Winston Churchill, ‘In
Remembrance of 1940’, its prologue imagined Duke William of
Normandy passing through Dover during his visit to England in 1051:
When the Duke landed, the townsmen saw a host march through their
streets. They beheld Counts and Barons, Bishops like Princes, and
mailed Knights, riding as though to battle. The sound of marching filled
the town. Trumpets and clarions rang. The captains shouted commands
and the ranks wheeled and turned as one man at their bidding. The men
of Dover said: “If this is peace, how does he fare in war?”
In a story that develops with all the dreadful inevitability of Macbeth,
both Muntz and her readers knew that one day the men of Dover, and of
all England, would discover the answer to this question, and that it would
be delivered by one of the most determined, ruthless and successful
warriors of the Middle Ages, on one of its most celebrated battlefields.
The Conqueror, aged about twenty-three in 1051, was related to his
host, King Edward the Confessor, for his great-aunt, Emma, had married
two earlier rulers of the English, Æthelred II (often known today as the
Unready) and then in 1017 the Danish Cnut, who had conquered the
country the previous year. By Æthelred she had borne two sons, Edward
and Alfred, who lived in exile in Normandy after their mother’s second
marriage; by Cnut there was a son, Harthacnut, and a daughter, Gunnhild,
eventual wife of the future emperor Henry III of Germany. There is a
story that Emma only consented to a union with Cnut after securing his
agreement that any male offspring of hers should inherit the crown in
preference to his sons (Swegen and Harold Harefoot) by another

21
 THE BATTLE OF HASTINGS

consort,1 and in the years after Harthacnut’s birth Emma must have
become totally committed to his cause. In Normandy, however, Edward
and Alfred were not forgotten by her relatives; Cnut’s original desire to
marry a woman some years his senior may well have had a great deal to
do with fear of Norman support for them, and he later tried to strengthen
his position even further by uniting his sister Estrith with William’s
father, Duke Robert. Unfortunately, Robert eventually repudiated Estrith,
and by 1033 was recognising Edward as king of England and preparing a
naval expedition to establish him on the throne.2 However, this sailed and
came to nothing, and Robert died two years later while returning from a
pilgrimage to Jerusalem.
Cnut also died in 1035, and the succession dispute which followed
between his sons Harold Harefoot (ruled 1037-40) and Harthacnut (ruled
1040-2) was complicated by the arrival in England of both Emma's sons
by Æthelred. Edward raided the south coast with forty ships, fought a
battle at Southampton and then returned to Normandy laden with booty,
but appreciating that he needed more soldiers.3 Alfred’s fate was less
happy. He landed in Kent, and in attempting to reach his mother fell into
the hands of an Englishman who had waxed great under Cnut, Earl
Godwin of Wessex. Godwin brought him to King Harold and he was
eventually blinded, dying as a result of the knife used for the purpose
entering his brain.4 There is no evidence of popular support at this time
for either Alfred or his brother, and the possibility of Edward ever
ascending the English throne must have seemed remote; yet by 1041
King Harold, whose elder brother Swegen had predeceased him, was
dead and his half-brother King Harthacnut in uncertain health.5 The
latter’s mother Queen Emma may have feared the extinguishing of her
influence should he die, and her long-neglected son by her first husband
was accordingly summoned from Normandy to be sworn in as king. The
following year Harthacnut had a seizure while drinking at a marriage
feast, lingered speechless for some time, and expired on 8 June.6
In 1042, all three of Cnut’s sons and their father having died young,
King Edward thus found himself in a position which it must long have
1
EE, pp. 32-3; this work, the Encomium Emmae, was written for Emma probably in St Omer, Flanders,
c.1041. Swegen and Harold were the sons of Ælfgifu of Northampton, whose connection with Cnut
continued after his marriage to Emma; see Lawson, Cnut, pp. 131-2.
2
Ibid., pp. 109-12, and especially Keynes, ‘The Æthelings in Normandy’.
3
GND, ii. 104-7; GG (drawing on GND), pp. 2-3. On the Gesta Normannorum Ducum of William of
Jumieges, see below, pp. 80-1; on the Gesta Guillelmi of William of Poitiers, see below, pp. 81-8
4
Alfred died on 5 February 1036 or possibly 1037; see ASC, C and D texts, i. 158-60; EE, pp. 40-7;
GND, ii. 106-7; GG, (elaborating on GND), pp. 4-5; LKE, pp. 32-3. The Life of King Edward was
written in England by a Flemish cleric for Edward the Confessor’s queen, Edith, between 1065 and
1067 (ed. Barlow, pp. xxix-xxxiii, xliv-lix); on the Anglo-Saxon Chronicle, see below, pp. 51-3.
5
GG, pp. 6-7.
6
ASC, C and D texts, i. 162-3. The D text is often in this period a conflation of manuscripts closely
related to C and E, and therefore not at such points an independent witness.

22
 1 PRELUDE

seemed impossible he would ever occupy, and for which he was ill-
prepared. In exile for most of his adult life, he was scarcely acquainted
with the great figures of the time, the earls Godwin of Wessex, Leofric of
Mercia and Siward of Northumbria, or with the political system in which
they operated. Godwin, in particular, seems to have enjoyed considerable
influence over him. By 1066 the income from lands held by the different
members of the Godwin family, if added together, rivalled that from the
king’s estates.1 More to the point, in 1045 Edward married Godwin’s
daughter Edith, probably unwillingly, as subsequent events suggest. This
move was presumably designed to perpetuate the influence of the family
into future generations through the birth of a royal heir who would be
partly of Godwin blood; if so, it failed. The years slipped past, and by
1050 it may have been obvious that Edward, then in his mid-forties,2
would have no son. By now the king was finding the Godwin influence
burdensome. On 29 October, Archbishop Eadsige of Canterbury, once a
priest of Cnut’s household, died. A relative of Godwin was a member of
the monastic community which served Canterbury cathedral and was
clearly the candidate favoured by the earl to fill the vacancy in the richest
and most powerful bishopric in the country.3 The king had other ideas.
Following his accession, he had been joined from across the English
Channel by a number of men to whom he had given offices and lands;4
some of them were, if the Anglo-Saxon Chronicle is anything to go by,
highly unpopular with the English. Nevertheless, Edward in 1051 chose
as his new archbishop Robert of Jumièges, bishop of London since 1046,
and a Norman.
Nor was this all. It was necessary for Robert, like all late Anglo-Saxon
archbishops, to go to Rome for his pallium – an ecclesiastical vestment
received from the pope, without which he was not a recognised holder of
the see of Canterbury. During the journey he passed through Normandy,
and it may have been at this time that, according to the contemporary
Norman historian William of Jumièges, he took with him an offer that
Duke William should be Edward’s heir.5 Even though they had every

1
This is an issue which has attracted much attention in recent years, assisted by the statistics provided
by Domesday Book; see now Grassi, ‘The Lands and Revenues of Edward the Confessor’, who
concludes that in terms of landed revenue the predominance of the Godwins compared with the rest of
the nobility was unmistakable and ‘dangerously close’ to that of the king himself; however, if crown
income from other sources is added ‘Edward’s paramountcy is clear’.
2
He may have been born in 1005, Barlow, p. 29.
3
LKE, pp. 30-1.
4
See Lewis, ‘The French in England before the Norman Conquest’, who stresses that Norman
influence dated back to Emma’s marriage to Æthelred II, and that the French presence in England
before 1066 was neither confined to Edward’s court favourites nor ‘exclusively or distinctively’ (125)
Norman.
5
GND, ii. 158-9, which gives no date. WP, GG, pp. 20-1, says that Edward sent a son and grandson of
Godwin (mentioned again, pp. 68-9) to Normandy as hostages via Robert in confirmation of the

23
 THE BATTLE OF HASTINGS

reason to be biased on the issue, and say some things which are difficult
to believe,1 much of what followed becomes intelligible if this and other
statements by William of Jumièges and his fellow Norman writer William
of Poitiers, and by the probably contemporary French Carmen de
Hastingae Proelio (Song of the Battle of Hastings)2 are accepted, and
modern scholars have been convinced for the most part that Edward did
give some kind of undertaking to the duke at about this date.3 The same
year (1051) he was visited by his brother-in-law, Count Eustace of
Boulogne. The reason for Eustace’s journey is unknown, but is not
unlikely to have been his awareness of the promise, and a desire to press
his own claim to the throne as the second husband of Edward’s (dead)
sister, Godgifu. However, as things turned out, whatever he said to the
king was to prove less significant than events on his return journey: as he
and his men passed through Dover they became involved in a brawl with
the inhabitants in which deaths were sustained by both sides; as a result,
the king ordered Earl Godwin to harry the town, which lay in his
earldom. This must have put Godwin in a serious dilemma, as Edward
may well have intended – should he do as he had been told and thus lose
credibility with his own constituents, or should he defy the king’s express
command? Perhaps angered by what had happened both over the
archbishopric and the inheritance to the throne, he chose the latter course,
and as late Anglo-Saxon kings did not take kindly to disobedience, he and
his sons also began to raise an army. When this became known, Edward
called upon the earls Leofric of Mercia and Siward of Northumbria to do
the same, and the stage looked set for civil war. At the height of the crisis,
however, the nerve of Godwin’s men seems to have failed, and his
support fell away. The result was exile for him and all his sons, while the

promise, and with the consent of the English magnates; but this may have been in 1052, see below, p.
25. The consent of the English is also mentioned by the Carmen, ll. 292, 738.
1
Below, pp. 82-3..
2
On the provenance of the Carmen, probably by Bishop Guy of Amiens, see below, pp. 77-9.
3
But not M&M, pp. 65-8, who accepted no pledge made by Edward to William later than 1041, or
more recently Garnett, ‘Conquered England, 1066-1215’, pp. 62-8, who thinks that any promise by
Edward seems ‘highly unlikely’. Freeman (FNC, ii. 296-303, iii. 217-9, 684) rejected all the details in
the French sources, while accepting that Edward gave an undertaking during William’s visit in 1051;
for a similarly grudging provisional acceptance (‘even if Edward did nominate William as his heir in
1051’), see Barlow, pp. 106-9. Others have accepted the promise with less reservation: Douglas,
‘Edward the Confessor, Duke William of Normandy’, 534-9; Oleson, ‘Edward the Confessor’s
Promise’, 223; Stenton, Anglo-Saxon England, p. 561; Brown, The Normans and the Norman
Conquest, pp. 121-4; John, ‘Edward the Confessor and the Norman Succession’, 250-1; Bates, William
the Conqueror, p. 59; van Houts, introduction to GND, i. xlvii. The latter also points out (i. xlvi) that
WJ would have been in contact with Archbishop Robert, as after his exile from England in 1052 Robert
returned to Jumièges and died there in 1055. Garnett argues that Norman sources written after 1066
‘retrospectively imposed upon England in 1051’ a designation ceremony of the type frequently
practised by the dukes of Normandy in establishing their heirs. Yet as Edward had spent much of his
adult life there it would hardly be surprising if he decided to adopt the same practice, and of course he
had himself been designated king by Harthacnut in 1041. The article by Douglas is seminal here.

24
 1 PRELUDE

king consigned Queen Edith to a nunnery.1 It was now, according to the

D text of the Anglo-Saxon Chronicle, that William of Normandy visited
England, presumably to thank King Edward for the offer of the throne
and to accept it.2
The following year there was a dramatic reversal of fortune.3 Perhaps,
as Hope Muntz speculated, the sight of William and his men, together
with the knowledge of Edward’s promise, had not filled the English with
joy. At any rate, when Godwin and his sons returned from exile, by force,
their support seems to have been stronger than the previous year; once
again civil war was imminent, but avoided by negotiation, the upshot of
which was that the Godwins were fully reinstated in their lands and
offices, while the queen emerged from imprisonment in the nunnery to
take her place once more beside the king; some of his Norman followers,
including Archbishop Robert, fled. This must have been a humiliating
result for Edward, but it is possible that part of a face-saving deal struck
for the return of his enemies was that they would accept William as heir
to the throne, and that Archbishop Stigand of Canterbury (Robert’s
replacement) and the earls Godwin, Leofric and Siward swore an oath to
this effect, as William of Poitiers claims; it may have also been at this
point that two members of the Godwin family – Hakon, the earl’s
grandson by his eldest and deceased son Swegen, and his youngest son
Wulfnoth – were sent to Normandy as hostages.4 It is not known, of
course, that either oaths or hostages were seen by the senior members of
the Godwin family as placing significant limitations on their future
behaviour.
Shortly after this, in 1053, Godwin himself died, and was replaced as
earl of Wessex by his eldest surviving son, Harold. The following year

1
The main sources for the events of 1051 are the ASC D and E texts, i. 171-7; and LKE, pp. 30-7. The
latter attributes the crisis to friction between Godwin and Archbishop Robert which culminated in
Robert raising the matter of Godwin’s involvement in the death of Alfred (above, p. 22) and asserting
that he was now preparing the destruction of the king himself.
2
Douglas, ‘Edward the Confessor, Duke William of Normandy,’ 527-34, suggested that this is a late
and unreliable addition to D, and thought the duke too preoccupied with continental affairs to have
made such a visit in 1051 (similarly, Douglas, William the Conqueror, pp. 58-60, 169); Bates, William
the Conqueror, p. 34, thinks it ‘possible’ but ‘intrinsically unlikely’. I agree with Douglas that the
omission of the visit from the Norman sources is distinctly odd, but also with Hooper, ‘Edgar the
Ætheling’, 201, n. 18, that ‘it seems inconceivable that D’s account of the visit is fictional’; similarly,
Oleson, ‘Edward the Confessor’s Promise’, 221-2. On D, see further, below, pp. 52-3.
3
On 1052, see the ASC C, D and E texts, i. 177-83.
4
GG, pp. 18-21, 68-9, 120-1. This was the position of Oleson, ‘Edward the Confessor’s Promise’, 223-
4. WP implies that the oath was sworn and the hostages sent in 1051, but a degree of confusion on this
might be understandable; Barlow, p. 108, thinks that he fabricated the list of oath-takers to provide
circumstantial detail, and it is certainly possible that he knew the Carmen and was elaborating on its
statement (ll. 292-4) that Edward had made the promise to William with the assent of the people and by
the counsel of his nobles; for a discussion of the hostages and their identity, see Barlow, pp. 301-6. The
authority for them being sent to Normandy after the Godwins’ return is the post-Conquest writer
Eadmer, Historia Novorum, ed. Rule, pp. 5-6, who says that Edward was suspicious of Godwin and
would not make peace unless he first received hostages as security.

25
 THE BATTLE OF HASTINGS

the first moves were made to secure the return from exile in Hungary of
the son of Edward the Confessor’s half-brother Edmund Ironside, Edward
the Exile, who had left England at the beginning of Cnut’s reign, and was
the only surviving male member of the royal house (as far as we know)
apart from the Confessor himself. It is almost certain that his return was
connected with the question of the succession, and very unlikely that it
took place without the king’s permission, but we cannot say who took the
initiative in arranging it. Perhaps Harold, the rest of the Godwin family
and others hoped that the king might be persuaded to change his mind
about the duke1 and he was prepared to consider doing so, or Edward,
determined to prevent the accession of one of Godwin’s sons, was
providing himself with a second option should something go wrong with
his first choice.2 However, whatever hopes the return of the Exile may
have raised were to come to nothing, as he died shortly after arriving in
England in 1057, and before he was able to see the king. He left a son,
Edgar, aged about five,3 who was to survive all the great men of this
history; it is not likely that Edward ever considered making him his heir,4
although it need not follow that he was completely without support.
The next move on the succession issue was made when Earl Harold
went to Normandy, probably in 1064 or 1065.5 The French writers are
quite clear that the purpose of this visit was to confirm the promise of the
throne to the duke. The Carmen de Hastingae Proelio alludes to the
making of a pact of friendship between Harold and William, to Harold’s
oaths and subsequent perjuries, and to the fact that Edward sent via his
envoy a ring and a sword; this is a plausible detail, as these were two of
the objects with which English kings were invested when they were
crowned.6 William of Jumièges says that Edward sent Harold, the greatest
greatest of his earls, to swear fealty to William about the crown and to
confirm it with oaths. On the way Harold landed in Ponthieu (a small
county which lay on the coast between Normandy and Flanders to the
north), where he was captured and imprisoned by Count Guy, until
William’s pressure secured his release. He then stayed in Normandy for
some time, swore fealty about the kingdom with many oaths, and was
1
Oleson, ‘Edward the Confessor’s Promise’, 225-6, saw the initiative for Edward’s return as entirely
Harold’s.
2
This is Brown’s suggestion, The Normans and the Norman Conquest, pp. 126-7; WM, GR, i. 416-7,
reports that Edward sent for the Exile because he saw Godwin’s sons growing more powerful. His
return is described by ASC D and E, i. 187-8. See also, Körner, The Battle of Hastings, pp. 196-209;
Barlow, pp. 215-8.
3
Ibid., p. 218.
4
But see van Houts, 'The Norman Conquest through European Eyes', 845-6. JW, ii. 582-3, says that he
had decided that the Exile should be his heir, but on JW’s prejudices, see below, pp. 53-5.
5
The date is not certain, see FNC, iii. 227, 243-4, 694-6, in favour of 1064; Keats-Rohan, ‘William I
and the Breton Contingent’, 164-5, thinks that there is ‘no doubt’ that this is correct, and I use it
henceforth; John, ‘Edward the Confessor and the Norman Succession’, 259-60, argued for late 1065.
6
Carmen, ll. 233, 239, 295-300.

26
 1 PRELUDE

finally sent back to the king with many gifts.1 William of Poitiers
elaborates on this: according to him, Edward wished to prepare for death
by confirming his pledge with an oath and sent Harold, whose power
would be able to check the customary perfidy of the English should it
prove necessary; William secured his release after he fell into the hands
of Count Guy, and he swore fealty to the duke at a council held at
Bonneville, promising to be the vicar of William in Edward’s court, and
to do his best to ensure that England passed into William’s hand after
Edward’s death; in the meantime, he would fortify the castrum of Dover
for William’s soldiers, and fortify, garrison and provision other castra
chosen by the duke. Before the oath was sworn, but after Harold was
received as his host’s vassal, he was confirmed in his lands and powers;
he seems also to have agreed to marry one of William’s daughters.2 Then,
knowing his guest to be eager for renown, the duke furnished the
Englishmen with weapons and the best of horses, and took them on
campaign in Britanny. After their return, he kept them with him a little
longer before sending Harold home loaded with gifts and accompanied by
his nephew, previously one of the two hostages in his possession.3
A very similar impression is given by the Bayeux Tapestry, in so far as
can be judged from scenes intended for an audience familiar with the
story, and hence provided with only a sparse accompanying text. Edward
fairly clearly directs Harold to go to Normandy, he encounters Guy, is
delivered by William, and goes on campaign in Brittany; then, in a scene
which contemporaries would have recognised as indicating that he had
accepted the duke as his lord, he is shown being given arms by William,
before swearing the oath and returning to England. There are differences
of detail here – the oath is sworn at Bayeux rather than Bonneville, and
after the Breton campaign rather than before it – but the Tapestry, as one
would expect of a source produced for a Norman viewpoint,4 seems to
confirm the case presented by the two Norman chroniclers and the
Carmen.5 Is it true?
The most detailed treatment of Harold’s oath, as of many other events
of this period, remains that of Edward Augustus Freeman, whose
remarkable six volume The History of the Norman Conquest of England,
a fraction of the published output of a scholar possessed of a range which
allowed him to apply for the Camden Professorship of Ancient History in
Oxford in 1861 and its Chichele Professorship of Modern History in

1
GND, ii. 160-1.
2
GG, pp. 68-71, 156-7 (on the marriage); there is a shorter version in William’s alleged speech before
the battle of Hastings, pp. 120-1.
3
Ibid., 70-1, 76-7.
4
But see below, p. 29, n. 3.
5
See also OV, HÆ, ii. 134-7, who places the oath at Rouen, possibly as a result of misunderstanding
WP.

27
 THE BATTLE OF HASTINGS

1862, is seldom consulted without profit a century and a half after it was
first published. Even so, Freeman’s initial work was done during the
1860s, at a time of strong anti-French feeling, in an England which knew
much of Napoleon, French tyranny and the battle of Waterloo, but in
which the entente cordiale still lay in the future; there had been a recent
invasion scare, and he always referred to the contemporary French ruler
Napoleon III as the ‘tyrant’.1 It must be reflective of this, and of the
powerful nationalism of the Victorian period generally, that he was less
than sympathetic to Edward’s promise of the throne to William and
associated events, and concerned to give the French sources no more
credence than was absolutely necessary. Accordingly, in his discussion of
the oath2 he admitted that ‘there was some groundwork for the Norman
story’, and that Harold ‘made some engagement or other’ (finally
concluding that the ‘oath was primarily an oath to marry William’s
daughter… accompanied by an act of homage’), but refused to pledge
himself ‘to the accuracy of a single detail’, and rejected the French
explanation of the reason for Harold’s visit to the duke as ‘absolutely
fabulous’.3 One of his final comments, that he had been discussing ‘one
of the most perplexing questions in all history’, was to find something of
an echo many years later in Sir Frank Stenton’s assertion, near the end of
probably the most widely-read of all modern general histories of the
Anglo-Saxons, that ‘about the reason for Harold’s journey’ no
‘convincing answer has ever been given’.4 However, Stenton arrived at
this conclusion as the result of a discussion in which he considered the
evidence of the Bayeux Tapestry, but ignored the explicit statements of
the Carmen, William of Jumièges and William of Poitiers about the
genesis of that journey.
The reluctance of English scholars to accept that Edward required
Harold to guarantee William’s succession in Normandy did not end with
Stenton,5 and it is unlikely that there will ever be a consensus of opinion
about the matter. Still, the weight of the primary evidence does not favour

1
On Freeman, see Frank Barlow in ODNB; Bryce, ‘Edward Augustus Freeman’; and Stephens, The
Life and Letters. He was eventually appointed Regius Professor of Modern History in Oxford, in
succession to Stubbs, in 1884, and died in 1892.
2
FNC, iii. 215-53, 667-96.
3
Ibid., 243, 218, 693, 243, 219.
4
Ibid., 696; Stenton, Anglo-Saxon England, p. 578.
5
See Barlow, pp. 220-9, who, however, accepts the possibility that WP’s story ‘is basically true’;
Gibbs-Smith, The Bayeux Tapestry, pp. 10-11; Bates, William the Conqueror, pp. 60-1; Bradbury, The
Battle of Hastings, pp. 63-72; Higham, The Death of Anglo-Saxon England, pp. 152-62. Bradbury and
Higham (and B&W, 11) argue that as the Bayeux Tapestry shows a subservient-looking Harold
greeting the king on his return to England this indicates that something had gone wrong with his
mission; but it could equally mean that he had simply carried out Edward’s orders (Bernstein, plate
xxviii, is surely correct in saying that whether ‘Harold’s posture is meant to be reverential or apologetic
is unclear’); M&M, pp. 68-72, judged the oath simply one of ‘friendship and loyalty’, without
discussing how Harold came to be with William. DeVries, pp. 149-54, reviews recent opinion.

28
 1 PRELUDE

the doubters, even if it is impossible to be sure that they are mistaken.

Against the quite clear statements of the French sources about Edward’s
intentions, both in 1051 and prior to Harold’s journey, the broadly
contemporary English writers, including all three surviving texts of the
Anglo-Saxon Chronicle, can only oppose an ominous silence, apart from
an obscure reference in The Life of King Edward to the earl being ‘rather
too generous with oaths, alas!’, which may or may not be an allusion to
the one sworn in Normandy.1 Not until the post-Conquest Canterbury
monk Eadmer (writing over forty years later) do we actually get an
English account of Harold’s visit, in which he went (against King
Edward’s advice) to secure the return of the Godwin hostages, was driven
by a storm to Ponthieu, released from captivity on the duke’s orders,
informed by William of Edward’s promise, and encouraged by him to
swear the oath in return for the release of his nephew; Harold, perceiving
that there was danger whatever way he turned, agreed.2 That tales of this
type circulated in England after, and perhaps before, 1066 is not to be
wondered at, for the oath is likely to have been better known than the
truth of the circumstances in which it was taken; even so, they smack of
special pleading.3 It does not seem inherently likely, for example, that
Harold would have willingly put himself into the power of a great rival in
order to secure the return of two hostages who had already been in
Normandy for twelve years, and one of whom he was quite prepared to
abandon to his fate when he broke the oath not long afterwards. Even so,
this by no means exhausts the later reports. The early-twelfth-century
Anglo-Norman writer Orderic Vitalis says that when Harold returned to
England he lied to Edward by alleging that the duke had given him his
daughter in marriage and resigned his claim to the kingdom to his new
son-in-law; the sick king was amazed, but fell in with his wishes.4
William of Malmesbury, who knew both the Norman version of events
and that of Eadmer, tells an even wilder story (which he says he thought
plausible enough) to the effect that Harold went on a fishing expedition in

1
citius ad sacramenta nimis, proh dolor, prodigus; LKE, pp. 80-1.
2
Eadmer, Historia Novorum, ed. Rule, pp. 6-8. Eadmer has the duke date the promise to when Edward
and he were both young in Normandy; in fact, the king was more than twenty years William’s senior.
He also adds some familiar details, for example Harold’s agreement to marry William’s daughter.
Barlow, The Godwins, p. 76, is convinced by Eadmer’s claim that Harold took the oath against his will
and ‘could surely have argued in any tribunal that promises extorted under duress were invalid’.
3
The Bayeux Tapestry can be read in a way not incompatible with Eadmer, as Freeman noted, FNC,
iii. 676-7. Bernstein, pp. 115-7, argues that it was deliberately made ambiguous enough to suit both the
Norman and the English accounts (similarly, B&W, 10-11). The Norman version of events has been
accepted by Douglas, ‘Edward the Confessor, Duke William of Normandy’, 540-5, and William the
Conqueror, pp. 175-7; Oleson, ‘Edward the Confessor’s Promise’, 227; Brown, The Normans and the
Norman Conquest, pp. 127-32; John, ‘Edward the Confessor and the Norman Succession’, 260-2; van
Houts, introduction to GND, i. xlvii (but she has since changed her mind, compare TNE, p. 104).
4
HÆ, ii. 136-7. Chibnall (n. 1) regarded this as an embroidery ‘drawn from popular tradition or
possibly saga’; on OV, see below, pp. 92-4.

29
 THE BATTLE OF HASTINGS

the Channel, was blown to Ponthieu and imprisoned by Count Guy, and
persuaded William to release him by pretending that he had been sent to
confirm the promise of the throne.1 It is certainly not impossible that he
went to the continent of his own free will for diplomatic purposes which
cannot now be recovered,2 but there is not a scrap of evidence to support
this view. On the evidence we do have the balance is in favour of the
story told by the French being basically true: that he swore the oath, quite
simply, because King Edward had sent him to do so.
Inaccurate as some of the details in the Norman accounts may be, they
did not err in describing Harold as the most powerful magnate in England
at the time the visit was made; indeed, the power of the Godwin family
generally was by that time considerable. Not only had Harold succeeded
his father as earl of Wessex in 1053, but two years later his brother Tostig
received the earldom of Northumbria after the death of Siward. It was
perhaps in 1057 that their sibling Gyrth was given East Anglia, to which
was later added Oxfordshire, while the fifth of Godwin’s sons, Leofwine,
at some point received control of an earldom including at least Middlesex
and Hertfordshire.3 Such success is unlikely to have been achieved
without friction between the Godwins and other leading families. It is no
accident that the earls Leofric of Mercia and Siward of Northumbria
supported the king in the crisis of 1051, and that Leofric at least was
rewarded for doing so, his son Ælfgar receiving Harold’s earldom of East
Anglia,4 but being obliged to relinquish it upon the latter’s return the
following year. He recovered it when Harold was promoted to Wessex in
1053, but was outlawed in 1055, perhaps because he had overstepped the
mark in laying claim to the Northumbrian earldom given to Tostig.5
However, he allied with the Welsh king Gruffydd, defeated local English
levies near Hereford and was reinstated; in 1058, the year after he had
succeeded to the earldom of Mercia upon his father’s death, the same
process was repeated, Ælfgar’s expulsion being followed by restoration
with the help of Gruffydd.6 A period of calm seems (these years are not
well documented) to have followed; it did not last.
In 1065 Tostig’s earldom of Northumbria rose against him. All the
thegns of Yorkshire and Northumberland declared him an outlaw, killed
every one of his retainers they could find, seized all the weapons and
1
GR, i. 416-9; on WM’s strong pro-English bias, see below, pp. 58-60; HH, HA, pp. 380-1, says that
Harold was crossing to Flanders when he was driven ashore in Ponthieu.
2
This was Körner’s position, The Battle of Hastings, pp. 109-21, anticipated in The Bayeux Tapestry,
ed. Stenton, p. 15; see also the ingenious suggestions of Higham, The Death of Anglo-Saxon England,
pp. 160-2.
3
On Gyrth and Leofwine, Writs, pp. 562, 567.
4
ASC E, i. 177; see further Barlow, pp. 114-5.
5
As suggested, ibid., p. 193; ASC C, D and E, i. 184-6.
6
ASC D, i. 188; on Ælfgar, see Writs, pp. 546-7; JW, ii. 584-5, says that he also had the assistance of
the fleet which came from Norway in this year and joined him unexpectedly (below, pp. 35-6).

30
 1 PRELUDE

treasure in York, sent for Morcar, son of Earl Ælfgar, and chose him as
their earl; they then marched south, along with the men of
Nottinghamshire, Derbyshire and Lincolnshire, until they reached
Northampton, where they were joined by Morcar’s brother Edwin, who
had received the earldom of Mercia upon their father’s death.1
Negotiations followed, and the king accepted what had happened; Tostig,
his wife Judith and ‘all those who wanted what he wanted’ went into
exile in Flanders with Judith’s half-brother Count Baldwin V. Of course,
these events would seem to have weakened the Godwins’ position
considerably. Both Mercia and Northumbria were now in the hands of
brothers almost certainly unsympathetic to them, and it was probably
obvious that Tostig would eventually attempt a return by force, as the
whole family had done in 1052. However, it may be that Harold was not
entirely sorry to see the departure of a powerful and ambitious brother
who might have contested his own claim to the throne when the time
came, and The Life of King Edward claims that Tostig accused him of
inciting the rebellion.2 If this was Harold’s game he acted none too soon,
for on 5 January, 10663 the king at last died.
On his deathbed, he finally seems to have changed his mind about the
succession, although how far he was fully aware of, or responsible for,
what he was doing might, of course, be doubted.4 The C and D texts of
the Anglo-Saxon Chronicle (which are here one source rather than two)5
say that he ‘entrusted’ the kingdom to Harold, and The Life of King
Edward that he commended it to Harold’s protection.6 If these statements
statements stood alone we might be tempted to think that the earl was
intended to act as regent until William’s arrival, and the same impression
could be derived from Baudri of Bourgeuil’s statement of c.1100 that
during his final illness Edward dispatched legates to Normandy
confirming his promise to William in writing and that the Normans sent
to deliver the duke’s acceptance found him dead on arrival.7 However, the
the Chronicle E text says baldly that Edward granted Harold the throne,8
and this was fully admitted by the Normans. The Bayeux Tapestry shows
one of the men who offers Harold the crown pointing to the scene of King
Edward speaking to his fideles on his deathbed, thus apparently indicating

1
The date of this is not known, but was in 1062 or later.
2
ASC C, D and E, i. 190-5; LKE, pp. 74-83.
3
The date given by ASC C, D and E, but LKE, pp. 124-5, says 4 January; see Barlow, pp. 250, 253,
who suggests that Edward died on the night of 4/5 January.
4
See ibid., pp. 249-53, who thinks it ‘extremely doubtful’ whether Edward was ‘by modern
standards… of sound testamentary capacity’.
5
Above, p. 22, n. 6; below, p. 52.
6
ASC C and D, i. 194-5; LKE, pp. 122-3.
7
AC, ll. 271-8. As the legates are said to have sworn that the kingdom was William’s this may be a
confused version of Harold’s visit of 1064; on Baudri, see below, pp. 88-92.
8
ASC E, i. 197.

31
 THE BATTLE OF HASTINGS

the warrant for what was happening, and William of Poitiers mentions the
bequest to Harold twice. 1
By 1066, if not by the time of his visit to Normandy, Harold had
probably decided that he would take the throne whatever Edward’s
wishes, and he was crowned on 6 January, the day of the latter’s funeral.
He had, of course, broken his oath to William and must have known that
he would have to fight to maintain his position, but as a foreign invasion
would take time to organise the haste of the coronation is likely to have
stemmed from a fear of opposition in England which was to prove well-
founded. The C and D texts of the Anglo-Saxon Chronicle say that at
Easter (16 April) 1066 Harold travelled from York to Westminster. On
what he had been doing in Northumbria they (and the E text) are no more
forthcoming than on his oath in Normandy, and it is left to William of
Malmesbury’s twelfth-century Life of Bishop Wulfstan II of Worcester to
fill the deficit. According to this, Harold had travelled north because the
Northumbrians refused to accept him as king, and together with Wulfstan
he succeeded in overcoming their reluctance to be associated with
southern softness.2 It is also probable that they feared the restoration of
the new king’s brother Tostig as their earl, and that Harold gave
guarantees about this. How far their current earl Morcar and his brother
Edwin of Mercia were involved in these events is not known,3 but it is
known that Harold at some point married their sister Edith, and the period
following his coronation may have been that point.4
His return south was quickly followed by one of the most famous
events of this famous year – the appearance of Halley’s Comet, every
night of the week following Monday, 24 April.5 The Bayeux Tapestry
shows men marvelling at the star as the king sits uncomfortably on his
throne and spectral ships appear in the border below, while William of
Poitiers says that it was a prophecy of Harold’s destruction. 6 These

1
A point made by B&W, 21, and Bernstein, pp. 117-21; note, however, Cowdrey’s suggestion
(‘Towards an Intepretation of the Bayeux Tapestry’, pp. 101-2, 107-8) that the scene means that
Edward intended to keep to his earlier intention of making William king and that Harold then falsely
took the crown for himself; GG, pp. 118-9, 140-1.
2
WM, Vita Wulfstani, ed. Darlington, pp. 22-3; this follows closely a life written in Old English by the
Worcester monk Coleman at some date between 1095 and 1113 and now lost.
3
Freeman guessed that they were, FNC, iii. 59-60.
4
JW, ii. 604-5; OV, HÆ, ii. 138-9, 216-7; see FNC, iii. 625-7, iv. 755-6. As Edith, who had previously
been the wife of King Gruffydd of Wales (d. 1063), may have had two sons by Harold the marriage
could have been before 1066, unless, as Freeman thought, they were twins born posthumously.
However, a marriage alliance between Harold and her brothers could also be fitted plausibly into the
negotiations which culminated in Tostig’s expulsion in 1065, or the alliance could have been formed
earlier still as part of an understanding that Harold would be the next king. I am grateful to Stephen
Baxter for this last point, and for drawing Harold’s sons by Edith to my attention.
5
ASC, C and D, i. 194-5. It was also widely reported on the continent, van Houts, ‘The Norman
Conquest through European Eyes’, passim. The omission of any mention of it in this generally pro-
Godwinist (below, p. 51) part of the E text of the Chronicle may be no coincidence.
6
GG, pp. 142-3.

32
 1 PRELUDE

sources all had the benefit of hindsight, but such signs in the heavens
were in the Middle Ages often regarded as portents of great events, and
the comet is likely to have been seen by many in England as a sign of
God’s wrath about Harold’s broken oath and a very ill-omen for the
future. But this did not stop the king doing what he could to safeguard his
position against his enemies, including his own brother.
That brother, as events were to show, had been completely alienated
from Harold by the events of 1065, and while in Flanders he may have
been in touch with Duke William, whose wife Matilda, daughter of Count
Baldwin V, was like his own spouse a member of the Flemish comital
family. Orderic Vitalis claims that Tostig actually visited Normandy and
offered William assistance in his invasion, before sailing for England.1
Whatever the truth of this, not long after the comet’s appearance Tostig
landed on the Isle of Wight ‘with as big a force as he could gather’ and
was given money and supplies there. He then harried along the south
coast until he came to Sandwich in Kent. However, Harold was at the
same time gathering ‘greater naval and land forces than any king in this
land had ever gathered before’ to meet an expedition by William, and
when Tostig learnt that they would soon arrive in Sandwich he took
sailors from there and went north with 60 ships to Lindsey, whence he
was expelled by Edwin and Morcar. He then proceeded with twelve
vessels (the reduction in the size of his force presumably being indicative
of loss and desertion) to Scotland for a rendezvous with King Harald
Hardrada of Norway.2
Only the E text of the Chronicle states that Harold ‘went out against
William with a naval force’, and there is some reason to think that this
refers to rather more than the simple deployment of a fleet mentioned by
the C and D texts. Domesday Book for Essex lists an Ailricus who ‘went
away to a naval battle (abiit in navale proelium) against King William’
and fell ill on his return.3 Far away, in the monastery of Nieder-Alteich on
the Danube, an annalist writing in 1075 reported that, in the summer after
the appearance of Halley's Comet, Aquitanians fought the English in a
naval battle, and having defeated them subjected them to their rule; men
who had taken part in the campaign had said that 12,000 died on the side
of the victors, along with an incomprehensible number of the
vanquished.4 That there was a fight at sea is made more plausible by the

1
OV, HÆ, ii. 140-3, along with the statement (142-5) that Tostig also visited King Harald in Norway.
In his earlier additions to WJ’s GND (ii. 162-3, see below, pp. 92-3), Orderic claimed simply that
William sent Tostig against England.
2
ASC C, D and E, i. 194-7.
3
DB, ii. 14b; see FNC, iii. 338, 716-7; Writs, p. 302.
4
Van Houts, 'The Norman Conquest through European Eyes', 841-2. On Aquitanians in 1066, see
below, p. 141.

33
 THE BATTLE OF HASTINGS

writer having spoken to participants in the war.1 Perhaps Harold raided

the Norman coast at some time during the summer of 1066, while he lay
on the Isle of Wight awaiting the duke’s arrival; or perhaps there was a
bigger and much more significant conflict which has disappeared from
the sources almost completely.
All the summer and part of the autumn Harold waited, with land forces
stationed everywhere along the coast, but William did not attempt a
landing. By 8 September the English provisions had run out and the men
were allowed to go home, while the fleet returned to London. As it did so,
King Harald of Norway sailed into the Tyne with 300 ships, along with
Tostig, who had become his man. From there they proceeded into the
Humber and advanced upon York. The presence of Tostig virtually
guaranteed that this force would be opposed by his enemies Edwin and
Morcar. On Wednesday, 20 September they met the Norwegians in battle
on the Ouse to the south of York,2 and were defeated. The C text of the
Chronicle, the only contemporary source to give any detail about the first
of the three battles fought in England within twenty-five days, says that
the English made great slaughter, but themselves had a great number
slain, drowned or put to flight, and that the Norwegians remained in
possession of the field.3 There are more detailed accounts in Icelandic
sagas from the early thirteenth century of which the most famous is
Snorri Sturluson’s King Harald’s Saga, one of the components which
make up his Heimskringla (The Orb of the World).4 This may preserve
authentic traditions on a fight of which tales must have been told and re-
told in the north for many decades, but few historians would place much
trust in its statements today, and it is enough to simply record how the
ferocity of the Norwegian attack, led by Harald’s great banner ‘Land-
Waster’, is said to have carried all before it after initial English success,
and how the corpses of those of the English who fled into a marsh were
piled so high that the victors could cross it dry-shod; it is enough to
record it, and to pass on.
Following his victory King Harald accepted hostages from the citizens
of York and began negotiations with the Northumbrians intended to gain
their cooperation in his attempt to secure the English crown. 5 That
attempt was part of a long-standing order as far as the English were
1
Dr van Houts speculates that they may have been English.
2
Not until the Historia Regum attributed to the twelfth-century chronicler Symeon of Durham (ed.
Arnold, ii. 180) are we told that the site was Fulford; FNC, iii. 350, n. 1. HH (HA, pp. 386-7) says that
it was still pointed out on the south side of York, in a way which suggests that he may have seen it.
3
ASC, C, i. 196. The chronicler Marianus Scotus, writing in Mainz not long afterwards, says that
Harald killed more than 1,000 laymen and 100 priests in battle, van Houts, ‘The Norman Conquest
through European Eyes’, 839.
4
Trans. Magnusson and Pálsson, pp. 142-4. For a recent treatment, see DeVries, pp. 11-13, 255-9, who
discusses the Scandinavian sources, and places considerable credence in them.
5
ASC, C, i. 197. JW, ii. 602-3, says that each side gave 150 hostages to the other.

34
 1 PRELUDE

concerned. For more than two centuries Scandinavians had both raided
their shores and endeavoured to conquer parts of their territory, or the
whole. In 1013 King Swegen Forkbeard of Denmark had succeeded, as
had his son Cnut in 1016. Even after the Confessor’s accession in 1042
events in Scandinavia can seldom have been far from English minds. His
half-brother Harthacnut had been put in charge of Denmark by Cnut and
was not able to become king of England until five years after the death of
his father because of the threat to his position from King Magnus of
Norway.1 In the early 1040s Magnus was still attempting to conquer
Denmark, but was also seen as a threat to the English. A post-Conquest
source claims that Edward’s mother Emma was accused of inciting him
to invade England,2 and in 1045 Edward took a fleet to Sandwich because
his arrival was thought imminent.3 In 1047 he at last succeeded in taking
Denmark from Cnut’s nephew Swegen Estrithson (whose brother Beorn
at the time held an English earldom), but died in October of the same
year. This allowed Swegen to re-establish himself, while Norway was
secured by Magnus’ uncle, Harald Hardrada. Both sent embassies to
Edward in 1048, and Edward refused Swegen’s request for naval
assistance of at least 50 ships.4 It was fortunate for the English that the
following years saw much fighting between them, for both were
potentially dangerous. Swegen was later to claim that Edward had on two
occasions (both in the 1040s) promised him the English throne,5 although
although as he may have sent forces to assist Harold in 1066 this story
need not have been current before 1069-70, when he made an
unsuccessful attempt to deprive the Conqueror of his conquest. According
to a Danish source from the middle of the twelfth century, Harthacnut and
Magnus had agreed that the kingdom of the first to die would pass to the
other.6 As Harthacnut claimed England at the time, this (if true) would
have given Magnus and his successor some sort of claim also. Certainly
Norwegian forces crossed the North Sea well before 1066. In 1058 they
joined with the men of Orkney, the Hebrides and Dublin in a raid on
England led by Harald Hardrada’s son Magnus; although the author of
the D text of the Chronicle unfortunately found it too tedious to write an
1
Magnus’ father St Olaf had been expelled from Norway by Cnut in 1028, and killed in 1030 at the
battle of Stiklestad when attempting a return. Magnus established himself in Norway at about the time
of Cnut’s death, after expelling Cnut’s son Swegen.
2
See Barlow, p. 58, who thinks Emma’s guilt, but not the charge, unlikely.
3
ASC D, i. 165.
4
ASC D, i. 167.
5
Adam of Bremen, Gesta, pp. 135-6, 152; see Barlow, pp. 58, 93. Adam wrote in the 1070s, and had
met Swegen Estrithson; if the Danish king was telling the truth it would facilitate the interpretation,
contrary to that adopted here, that the Confessor had no fixed plan for William to succeed him, but
tended to play candidates off against each other as circumstances made convenient, perhaps with a
cruel and whimsical humour; for this view, see Barlow, ‘Edward the Confessor and the Norman
Conquest’, pp. 107, 110-1.
6
Chronicon Roskildense, ed. Gertz, i. 22.

35
 THE BATTLE OF HASTINGS

adequate account of what happened,1 it was clearly a harbinger of things

to come. Eight years later Swegen and Harald had made peace, and the
latter was preparing for action upon Edward’s death.
Harald, like William, did not sail until the autumn, and it is likely that
both were hoping to profit by doing so. Surely there would already have
been serious fighting involving the other, which would leave them to face
a weakened victor? Harald was clearly also hopeful that the
Scandinavianised part of eastern England known as the Danelaw would
be sympathetic to his cause, for it was only just over a century since a
Norse king, the celebrated Eric Bloodaxe, had ruled in York. It seems that
after the defeat of Edwin and Morcar the Northumbrians were prepared to
accompany him on a campaign in the south, and Harald withdrew to
Stamford Bridge, some seven miles to the east of York, and probably a
good communications centre, as it lay at the junction of several Roman
roads; once there, he awaited the hostages he had been promised from the
whole of Yorkshire, and grew careless.
Meanwhile, King Harold of England had gathered an army and was
marching north by day and night. On Sunday, 24 September he reached
Tadcaster, eight miles south-west of York, where he put his force in battle
array; the following day they passed through York and came upon the
Norwegians unawares beyond the bridge (i.e. on the east bank of the river
Derwent). In hard fighting, which went on until late in the day, both
Harald Hardrada and Tostig were killed, along with a countless number
of their Norwegian and English followers. Harold eventually gave quarter
to the survivors, but they were few. Twenty-four ships are said to have
sufficed for the return to Norway of a force which had arrived in 300; if
so, the slaughter had been very great.2 The contemporary sources are as
sparing of detail on the battle of Stamford Bridge as on that five days
earlier. The C text of the Anglo-Saxon Chronicle comes to an end
midway through its account, and it was left to a twelfth-century hand to
add the story of how a single Norwegian held the bridge against the
English until someone went underneath it and stabbed him from below.
Other post-Conquest sources tell a little more: Orderic Vitalis reported
that in his day a great heap of bones was still to be seen on the field; both
Henry of Huntingdon and William of Malmesbury knew of the lone
Norwegian hero, and the latter says that Tostig’s body was buried in York
after it was recognised by a wart between the shoulder-blades; Snorri
Sturluson claimed that in negotiations before the battle Hardrada was
offered seven feet of English ground, or as much more as necessary,

1
ASC D, i. 188-9.
2
ASC C and D, i. 197-9. JW, ii. 602-5, says the Norwegians came in more than 500 ‘great ships’ and
returned in twenty.

36
 1 PRELUDE

should he be taller than other men.1 Like the vast majority of English
battlefields that ground has yielded no evidence in modern times of the
bloody events which took place there. The recent digging by
archaeologists of three trenches prior to modern housing development
produced nothing. However, attention has been drawn to an undated mass
burial at Riccall Landing on the east bank of the Ouse, the area in which
the Norwegians left their ships, from which the remains of 39 individuals
were excavated in the 1950s and 6 more in the 1980s; the total number of
burials has been estimated at between 500 and 600, and some of the
bones recovered bear marks evidently caused by edged weapons and
arrows, although a few of the remains were those of women and children.
Clearly, they may be burials from the period of the battles of Fulford and
Stamford Bridge.2 Yet the latter, while doubtless celebrated in popular
legend for many of the hard years that followed, never had and never can
have the historian it deserved. Acting with a determination and speed of
which the most famous commanders would have been proud, Harold had
gained one of the greatest victories of Old English arms; and the last.
Three days later, on 28 September, William landed at Pevensey in
Sussex, and reached Hastings the following day.3 The very fact that he
was able to mount an attempt on the English throne at all in 1066 is
remarkable when one looks at his career up to this point. Born the
illegitimate son of Duke Robert of Normandy (and often called by
contemporaries William the Bastard), he had been recognised as his
father’s heir when Robert went on pilgrimage to the Holy Land in 1035,
and became duke following his death on the return journey. The ensuing
years must have played a major part in shaping the future Conqueror’s
personality.4 A period of great instability, characterised by extensive
feuding among the magnates, followed the death of Archbishop Robert of
Rouen, a brother of Duke Richard II, in 1037, and ten years later
William’s position was seriously threatened when there was a revolt
which sought to replace him with his cousin Count Guy of Brionne, the

1
HÆ, ii. 168-9; HA, pp. 386-9; GR, i. 420-1, 468-9; King Harald’s Saga, trans. Magnusson and
Pálsson, p. 150. HH gives details (perhaps drawn from a poem in Anglo-Saxon) that are not totally
consistent with the accounts in the Anglo-Saxon Chronicle, stating that fighting began at dawn, that the
great English numbers had pushed the Norwegians beyond the river by midday and that the hero held
the bridge until three in the afternoon. Snorri offers a detailed account of the battle which in some
respects echoes events at Hastings, except that here it is the English who are said to have used cavalry.
Most historians give it little credence, but see Glover, ‘English Warfare in 1066’, 5-7; and on the battle
generally, FNC, iii. 363-74, 720-8; Brooks, The Battle of Stamford Bridge; DeVries, pp. 262-96.
2
On evidence from English and Welsh battlefields generally, see the list which forms Appendix F of
Blood Red Roses, ed. Fiorato, pp. 211-39; on the 1950s finds at Riccall Landing, Wenham, ‘Seven
Archæological Discoveries in Yorkshire’, 301-7.
3
On the chronology of William’s crossing of the Channel, see further, below, pp. 152-3.
4
What follows is drawn from Douglas, William the Conqueror, pp. 31-80, 173-9; Bates, William the
Conqueror, pp. 25-43; ibid., pp. 90-8, provides a valuable summary of what is known about William’s
character.

37
 THE BATTLE OF HASTINGS

legitimate son of one of Richard II’s daughters. Fortunately for the young
duke, he was not only recognised by the French king Henry I, but Henry
came to his assistance in 1047, intervening with an army which played a
major role in defeating the rebels at Val-ès-Dunes, south-east of Caen.
Even then, William’s position was far from secure. It took another three
years for Count Guy to be banished from Normandy, and William also
began to have difficulties with Geoffrey Martel (The Hammer), count of
Anjou, whose ambitions now extended to the conquest of the county of
Maine, which lay between Anjou and Normandy, and which he occupied
in 1051. The following year he was reconciled with his former enemy
King Henry and both then turned against William, who was again facing
a serious rebellion within Normandy. In February 1054 a French army
invading eastern Normandy was defeated by Robert, count of Eu, at the
battle of Mortemer, and this forced the withdrawal of other troops led by
the king. He and William were briefly reconciled, but 1057 saw a further
royal invasion of the duchy along with Geoffrey Martel. William caught
them as they were crossing the river Dives near Varaville and apparently
succeeded in inflicting heavy losses on those who had been unable to do
so before the tide came in.1 Hostilities with Henry continued, but
deliverance was at hand. On 4 August, 1060 the king died, leaving France
to his young son Philip, who was under the guardianship of Willam’s
father-in-law, Count Baldwin V of Flanders. Only just over three months
later, Geoffrey Martel died too, and a resultant succession dispute
guaranteed that Anjou would not trouble William again for some time.
His marriage to Baldwin’s daughter Matilda had taken place c.1050,2
and its legitimacy was eventually recognised by the church, which had
originally opposed it for reasons which are not known. Useful as his
connection with Baldwin was, however, it was the twin fatalities of 1060
which brought about a dramatic improvement in his position. By the end
of 1063 he had made the most of his opportunities by conquering the
county of Maine, and he next turned his attention to Brittany,
encouraging some of the Bretons to rebel against Count Conan II, and
leading a campaign there, accompanied by his guest Earl Harold.3
Bretons were to accompany him to England in 1066, and William was
fortunate that by then he could leave Normandy without undue concern
about its safety while he was away. Had Edward the Confessor died
before 1060 it seems almost impossible that he would have been able to
back his claim to the English throne by the use of force. In 1066 he was
ready and able, and in more ways than one. The struggles of his youth
had produced a man with the charisma necessary for all medieval rulers,
1
GND, ii. 152-3. See further, below, p.156.
2
GG, p. 32, n. 6.
3
See Keats-Rohan, ‘William I and the Breton Contingent’, 162-6.

38
 1 PRELUDE

and also the required streaks of determination, ruthlessness and brutality.

William could be a cruel man, as those who opposed him often found out
to their cost, but in triumphing over adversity he had shown to
contemporaries that any undertaking in which he became involved stood
a good chance of success. It was this which drew to his banner men from
beyond the confines of Normandy,1 men confident of their leader’s ability
to deliver the rewards of victory, and fortified by the knowledge that God
was on their side.
The French vanguard at Hastings carried into battle a banner sent to
William by Pope Alexander II. It is mentioned by William of Poitiers,
who says that the duke sought the pope’s approval, and received the
standard with his blessing, so that he might attack his enemy with greater
confidence.2 It has been suggested that this author and other Norman
writers may have based much of their material on William’s claim to the
English throne on an account produced for the pope’s benefit;3 otherwise,
little is known about how the banner was acquired, although Orderic
Vitalis names Gilbert, archdeacon of Lisieux, as the envoy entrusted with
the mission to Rome, and describes it as ‘the standard of St Peter the
Apostle’,4 which may well be how the French soldiers were encouraged
to regard it: those who fought with the help of the greatest of the saints
would not expect to fight in vain, and in breaking an oath sworn on holy
relics Harold must have known that he would be vulnerable to action of
this kind. A letter written to William in April 1080 by Pope Gregory VII,
who as the archdeacon Hildebrand had been actively involved in the
negotiations of 1066, suggests that some had hesitated before sanctioning
the spilling of blood,5 but there is no evidence that any English embassy
contested the duke’s claims. Eventually, he would return like for like:
among the pope’s share of the spoils from the battle of Hastings was
Harold’s banner of the Fighting Man, upon which the image of a homo
armatus was worked in the purest gold.6 The idea that God might support
those fighting in a righteous cause was not new in the West: in the late
eighth century Charlemagne had sought the intercession of St Peter in his
campaigns against the pagan Saxons, and William may have been well
aware that in 1043 Pope Benedict IX had bestowed upon the emperor
Henry III a vexillum ex beati Petri parte prior to a campaign against the

1
Below, pp. 141-3.
2
GG, pp. 104-5. Morton, ‘Pope Alexander II and the Norman Conquest’, argued against any papal
approval of William’s expedition, and suggested (378-80) that WP, ‘an indefatigable but inept liar’ and
‘unsuccessful sycophant’, has antedated a (supposed) banner bestowed on him in 1070.
3
Van Houts, introduction to GND, i. xlvii.
4
HÆ, ii. 142-3.
5
Translated EHD2, No. 99; The Norman Conquest, ed. Brown, p. 165. Freeman gives the Latin text,
FNC, iii. 319, n. 2.
6
GG, pp. 152-3.

39
 THE BATTLE OF HASTINGS

Hungarians;1 thirty years after Hastings the papacy would launch in

God’s name the greatest of all the military expeditions of the eleventh
century – that now known as the First Crusade. In the meantime, the duke
had been given what contemporaries, many of whom needed no
convincing that heaven might intervene in human affairs in the most
decisive way, would have regarded as one of his most powerful weapons.
The months before his forces sailed for England may also have seen
diplomatic negotiations other than those with the pope. William of
Poitiers says that the duke made a pact with Henry IV, the king of
Germany, according to which the Germans would come to his assistance
if asked, and that envoys of Swegen Estrithson of Denmark promised
loyalty, although he was, as it turned out, to prove the friend of the duke’s
enemies.2
Freeman guessed that it was on Sunday, 1 October that Harold heard of
William’s landing on the south coast, but little is known of his actions
and whereabouts between the battles of Stamford Bridge and Hastings.3
The E text of the Anglo-Saxon Chronicle says that he fought the latter
before all his army had assembled,4 while William of Malmesbury reports
reports that many deserted him on the march south because he had
refused to share the booty from his victory, which he told Edwin and
Morcar to take to London; hence, he had with him his stipendiarios et
mercennarios milites (professional soldiers) but paucos… ex
prouintialibus (few shire levies).5 Orderic Vitalis suggests that it was in
London that Harold was told of the Normans’ arrival, and says that he
spent six days summoning the English to war and gathering an
innumerable multitude. He also tells the tale, which need not be
completely without foundation, of how Harold’s mother Gytha tried to
persuade him not to fight, and was supported by his brother Gyrth, who
brought up the problem of the oath to William, and offered to lead the
English army himself. This Harold is said to have angrily rejected.6
Certainly, he can have wasted no time in moving into Sussex; both
William of Jumièges and William of Poitiers say that he hastened his

1
Cowdrey, ‘Anglo-Norman Laudes’, 60; for other gifts of papal banners before 1066, see Morton,
‘Pope Alexander II’, 365.
2
GG, pp. 104-7. WP calls Henry Romanorum imperator, but he was not crowned emperor until 1084;
on the presemce of Danes in the English forces at Hastings, see below, p. 84.
3
FNC, iii. 733; see also the detailed discussion in Douglas, William the Conqueror, pp. 396-400.
4
ASC E, i. 198.
5
GR, i. 422-3, 468-9. Gaimar, L’Estoire des Engleis, ed. Bell, ll. 5245-8, says the booty was handed
over to Archbishop Ealdred of York, while one of the scholia added to Adam of Bremen’s Gesta, p.
196, notes that Hardrada’s mass of gold eventually came into the possession of William; on WM’s
evidence, see further, below, pp.58-60.
6
HÆ, ii. 170-3. Dr Chibnall comments (171, n. 4) that this ‘reads like a popular romance’. Orderic had
earlier added the same information to WJ’s GND, ii. 166-9. WM, GR, i. 450-3, places a very similar
story, here involving Gyrth but not Gytha, before Hastings. Other sources, HH for example (HA, pp.
388-9), claim that Harold was in York when he heard of William’s landing.

40
 1 PRELUDE

advance to take William by surprise,1 and given the success of such

tactics against Harald Hardrada this seems by no means unlikely. Harold
may well have been confident, and perhaps over-confident, but it is
probable that he considered the forces at his disposal adequate for the task
ahead.2
His adversary had not moved from the bridgehead he had established
on the south coast, with fortifications at Pevensey and Hastings. The
latter was a town with a mint, and may have had both effective ramparts
and a garrison;3 it was presumably occupied by the Normans, but of this
no source tells. William had everything to gain by allowing the English to
come to him, and in the fact that he did so we can probably discern his
experience as a commander. Staying where he was enabled him to protect
his fleet - William of Poitiers may err in claiming that there were 700
enemy vessels now in the Channel,4 but the safety of his ships must have
been one of William’s prime concerns – and to ensure that the English
would be as tired by their travels as his own men were refreshed by their
relative inactivity. How long he would have been able to maintain himself
in the area he had occupied had he been left to his own devices is a
question that cannot be answered, but Harold may have felt the adoption
of such a strategy too dangerous, and William was perhaps confident that
he would do so. His ravaging of English territory may also have provoked
Harold into action. Thus, by Friday, 13 October the English must have
been somewhere near the ridge which would one day be crowned by
Battle Abbey, for William of Jumièges says that they rode through the
night and appeared on the battlefield early the following morning.5 The
duke is likely to have already been in the vicinity,6 after advancing from
the coast when he heard of the English approach. He is said to have
ordered his army to stand by during the night in case of an attack; 7 there
was little danger that he would be taken by surprise, and Harold probably
erred in hoping otherwise. At around 9 o’clock on the morning of
Saturday, 14 October, the French moved forward; their vanguard bore the
papal banner, and it was later told that they began a song about Roland,
so that the great hero’s martial example might encourage those about to
fight.8

1
GND, ii. 166-7; GG, pp. 124-5; similarly, Carmen, ll. 281-2. As WP knew GND and perhaps the
Carmen he need not be an independent witness.
2
On the size of the English army, see further, below, pp. 126-31.
3
As my pupil Jonathan Hall has pointed out. Hastings was one of the forts of the Burghal Hidage
(below, pp. 107-8)) and one of the Cinque Ports.
4
GG, pp. 124-5; the Carmen, l. 319, gives 500; OV, HÆ, ii. 172-3, only 70.
5
GND, ii. 166-7.
6
On this, see below, pp. 158-9.
7
GND, ii. 168-9; the Carmen, l. 285, notes his watchfulness.
8
GR, i. 454-5.

41
 2
THE SOURCES: PART ONE
Few things are more difficult to describe than the events of the battle-
field. To say nothing of that lack of coolness which is essential to accurate
observation, an individual spectator commonly sees only a small portion
of the engagement, and is apt to overrate the incidents occurring in the
fore-ground of his view, while those which take place in the distance are
but slightly noticed. Acts comparatively insignificant thus become
magnified, while those of far greater importance are occasionally either
much distorted, or altogether overlooked. Allowances must also be made
for party prejudices, and for the flowers of rhetoric almost inseparable
from such descriptions. Now if even contemporary accounts of modern
battles are found to differ inter sese in some essential particulars, it must
be a matter of great difficulty to frame an intelligible history of the
sanguinary conflicts of ancient times from the materials furnished us by
partial and often incompetent chroniclers and written from oral traditions
at periods considerably subsequent to the transactions themselves.1

It would be difficult to improve upon these words, addressed by the

Sussex historian Mark Antony Lower to an assembly at Battle Abbey on
23 July, 1852, as a brief analysis of the difficulties involved in elucidating
events at the same location on 14 October, 1066. More is known about
Hastings than any battle fought in the West since the end of the Roman
empire, and any event in English history prior to 1066 and for another
century thereafter, until the murder of Thomas Becket by four of King
Henry II’s followers in Canterbury in 1170 produced a number of
contemporary accounts, several based on eye-witness testimony, and one
the work of a bystander who almost had an arm severed in the struggle.
There is no eye-witness on Hastings (and as Lower said, it would not be
an end of our problems if there were), but secondary writers such as
William of Jumièges, William of Poitiers, the author of the Carmen de
Hastingae Proelio, and Orderic Vitalis may well have spoken to
combatants, and some of them were working before King William’s

1
Lower, ‘On the Battle of Hastings’, 15.

43
 THE BATTLE OF HASTINGS

death in 1087; the same period probably also saw the production of the
Bayeux Tapestry, whose designer had access to details not included in
other contemporary sources. However, all this might tend to paint a
picture which is considerably rosier than the reality, for even
contemporaneity is no guarantee that medieval evidence will be easy to
use. In theory, source criticism seems a straightforward business: one
identifies the author’s background, his motive for writing and the likely
origin of his information, then his statements are compared with others to
establish his general reliability, and a rough rule-of-thumb worked out for
assessing his plausibility on matters with which he alone deals.
Unfortunately, such methods often break down in practice, partly because
of our imperfect knowledge and understanding of the world in which the
sources were produced. Thus, it can be difficult establishing exactly why
a particular piece was written and discovering very much about the author
concerned, even if his identity is known; it may be equally difficult cross-
checking his statements when other information on his subject is either
non-existent or inadequate, and even when statements can be checked it
tends to be characteristic of medieval authors, and probably of the
transmission of historical knowledge generally, that a source which is
wildly inaccurate on one point is not necessarily so on others. In general,
while there is enough evidence on the murder of Becket to establish the
main outline of what happened, this is not true with Hastings, in any case
a much more complex event, which is served by insufficient material of
proven reliability. Even so, nobody who seeks a proper understanding of
the battle, and of what can and cannot be known about it, is able to avoid
critical analysis of this material, and the intractable problems which
sometimes result. That is why this chapter and the one which follows are
the crux of this book.
By no means the least of our witnesses is the battlefield itself.
Important because it can to an extent be cross-referenced with the
documentary evidence, and because any discussion of the deployment of
the two armies needs to bear it in mind, it can also suggest much about
the strength of Harold’s position and why the battle seems to have been
so hard fought. Shortly after the conflict, and probably in 1067,1 Norman
bishops drew up a list of the penances to be served by surviving French
soldiers; this was subsequently confirmed by a papal legate, Bishop
Ermenfrid of Sion, and included the provision that:
Whoever does not know the number of those he struck or killed shall, at
the discretion of his bishop, do penance for one day a week for the rest of

1
On the date, I follow Cowdrey, ‘Anglo-Norman Laudes’, 59, n. 68; also Councils & Synods I, ed.
Whitelock et al., ii. 582. 1070 is an alternative.

44
 2 THE SOURCES: PART ONE

his life, or, if he is able, make amends either by building a church or by

1
giving perpetual alms to one.
King William may well have considered himself within the category of
those who knew not the number of those they had slain, and as a man of
conventional piety doubtless believed that God had, through the papal
banner, given him the victory that the righteousness of his cause
demanded. Thus it was probably early in his reign that he decided, as an
act of expiation and thanksgiving,2 to build Battle Abbey upon the site of
his triumph, although the church was not consecrated until early in 1094.3
The twelfth-century historian William of Malmesbury says that the altar
was positioned where Harold’s body had been found, while the author of
the Battle Abbey chronicle reports that it occupied the spot where his
standard had been seen to fall.4 There is little difficulty in believing that
these places were one and the same, or that William did issue instructions
to this effect. The abbey chronicle also tells how the monks, who came
from the French abbey of Marmoutier on the Loire, told the king that the
site was unsuitable for buildings, being on a waterless hilltop. They were
quite right, but he nevertheless insisted that this was where the abbey
should be,5 and as a result those erecting structures over the following
centuries often had to make substantial alterations to the original ground
level. Today, the slope to the south of the high altar ‘bears little relation
to the much steeper one up which the Norman cavalry would have had to
charge’, while further west the outer courtyard was extended and levelled
after the Dissolution by building up the hillside to the south.6 The
foundation of the abbey, at a time when relics of the battle, and perhaps
not least the grave-pits, must still have marked the field, is more than
adequate proof that the fight did take place in what has traditionally been
recognised as its location.7
1
Penitential articles issued after the Battle of Hastings, c. 1. I have quoted the translation in The
Norman Conquest, ed. Brown, p. 156.
2
The Brevis Relatio, p. 33, written in Battle 1114x20, says that William had the abbey built in memory
of his victory and for the absolution of the sins of all those killed there. Searle, Lordship and
Community, p. 21, suggests a date for the foundation not before 1071 and perhaps in the mid 1070s.
3
ASC E, i. 229. In the entry for 1087 this source also refers (i. 219) to the abbey’s foundation, ‘on the
very spot where God granted him the conquest of England’, in its obituary of the Conqueror.
4
GP, p. 207; Chronicle of Battle Abbey, ed. Searle, pp. 44-5. The latter was probably written in the last
third of the twelfth century (p. 8). I do not fully share Bradbury’s scepticism (The Battle of Hastings,
pp. 172-3) of its account. His statement that ‘the altar story does not emerge until a century after the
event’ takes no account of WM and is incorrect. In his GR, i. 492-3, WM reported that the principal
church at Battle was situated where Harold’s body was said to have been found among the heaps of
corpses. BR, p. 32, says that a Norman soldier told the duke before the fighting began that he thought
he could see Harold’s standard on the crest of the ridge.
5
Chronicle of Battle Abbey, ed. Searle, pp. 42-5.
6
Hare, ‘The buildings of Battle Abbey’, 80-2.
7
It would be unnecessary to stress this but for recent suggestions (Bradbury, The Battle of Hastings,
pp. 168-78; Grehan and Mace, The Battle of Hastings, pp. 91-5 and passim) that the battle was fought
on Caldbec Hill, a short distance to the north. Like Bradbury, the latter ignore WM on the location of
the altar.

45
 THE BATTLE OF HASTINGS

The battlefield on the Millenium Map. Reproduced by permission of Getmapping.plc. There is

a larger version of this image as Illus. 8, and 9-73 cover the battlefield in detail.

The battlefield according to the OS 6-inch map of 1931 (surveyed 1872/3, revised 1928). Sussex
(East) Sheet LVII. Not to the original scale. For the first edition of 1878, see Illus. 5.

46
 2 THE SOURCES: PART ONE

Harold drew up his men on and perhaps to the south of a slightly

curved ridge which, allowing for the curve, is roughly 1¼ kilometres (just
under a mile) long, and from which the ground falls away on all sides,
except at one point to the north, where there is a connecting neck of land
which is now the site of Battle High Street. The slope to the south was
steepest near the point at which he placed his standard, and this spot
occupied the crest of the ridge at a height above mean sea level of about
85 metres. From here the ground runs away both to the east and the south
west, in the latter case dropping about 15 metres over the approximate
250 metres to the western end of the abbey terrace, but then only one
metre more over the roughly 500 metres to the point at the westernmost
extremity of the ridge where the modern 2½ inch Ordnance Survey map
places a height marker of 69 metres.1 To the north of this is woodland
sometimes known as Long Plantation,2 and west of it, as stated, the ridge
comes to an end, dropping quite steeply at points into a drainage channel
which contains a small stream bordered on the east by Saxon Wood
(called Sacristy Copse on the Battle tithe map of 1859). Due west of the
marker the ground falls 24 metres in about 200 metres (an average of just
over 1 in 8),3 while to the south of it there is a fall of 29 metres in about
250 metres (an average of between 1 in 8 and 1 in 9). The visitor to the
field who now walks due east from the area of the marker through an area
of vegetation will come after slightly less than 500 metres to the north-
eastern edge of Horselodge Plantation at a height of between 50 and 60
metres.4 This covers some of a small but noticeable ridge which is part of
of the general fall of the ground to the south-west, and which at its
highest (north-eastern) point forms a distinct hillock. In the south-eastern
corner of the plantation are two small ponds, and to the rear of the hillock
a rather larger one;5 another pond, almost as big as the latter, forms to the
the west of it after a period of extended rainfall.6 Possibly the result of

1
OS Explorer Sheet 124, Hastings & Bexhill. The measurements quoted are all taken from this map.
2
It is so named on the OS 6 inch map, Sheet TQ 71NW, revised in 1974 and published in 1976, and on
earlier maps in the 6 inch series. Some OS maps give no name.
3
See further on the western end of the ridge, below, pp. 120-1 and Illus. 21-5.
4
I follow OS Explorer Sheet 124 in referring to the vegetation on the south-western face of the ridge
by this name, but earlier maps are both more specific and more revealing on its genesis. The tithe map
(Illus. 1) shows three plantations in this area, one centred on the hillock and named Mountain
Plantation, another slightly further west which is not named, and a roughly circular one north-west of
the latter and denoted Devil’s Plantation; the OS 6-inch map of 1931 (opposite) is similar, but gives
only the name Horselodge, which was evidently originally that of the central plantation of the three.
Their absence, with Long Plantation, from the estate survey of 1811 (ESRO, BAT 4435, Illus. 4) may
well mean that all were created between that date and 1859. However, the gorse on the hillock (see
below and Illus. 36-7) is likely to be natural.
5
All three are shown on the 2½ inch map.
6
This second pond appears on the modern 25 Inch OS map, but not on most other OS sheets. After a
period of dryness, it reappeared in the very wet autumn of 2000, but had disappeared by August, 2001.

47
 THE BATTLE OF HASTINGS

open-cast iron working1 at some unknown date, the two larger ponds, if
they existed in 1066, may have played some sort of role in the battle.2
Due north of the large ponds the slope rises 10 metres in about 125
metres before reaching the 70 metre contour line just south of the western
end of the abbey terrace. Due south of the hillock there is a very steep
gradient (down which English Heritage has provided a flight of steps)
leading to extensive low ground at a height of around 45 to 40 metres.
The amount of moisture which enters this area from the clay (see below)
of the western part of the battlefield is considerable, and today much of it
debouches into New Pond, constructed in 1815, and then eventually flows
down to Powdermill Lake and south to the coast. Before New Pond was
built the land to the west of it would have been subject to more flooding
than is now the case, and a map of 1724 shows that a distinct area of
marsh lay due south of the western edge of the main ridge, in ground now
occupied by the easternmost part of Warren Wood, gradually becoming
much larger as it extended downstream to the vicinity of Powdermill.
These areas are likely to be distinctly boggy even today, and nor were
there in 1724 apparently as many trees in this vicinity as there are now.3
To the east of New Pond the ground rises past the three Stew Ponds
which did exist in the early eighteenth century and which would have
reduced the amount of water flowing west still further, and when one
comes to the saddle which carries the road to Hastings the slope to the
north is much less severe than at the point where Harold was killed, rising
about 10 in 175 metres. Due east of the saddle, in the area which now
contains Battle railway station, the ground falls away again, and the slope
to the north gradually disappears as the eastern part of the ridge falls
away too, coming to an end more or less where Marley Lane, the road
which runs out of Battle to the east, crosses the railway line. The road
itself seems to be cut into the ridge (Illus. 63-5), and this is particularly
noticeable near the former National School of 1842, where the existence
and depth of the cutting suggest that there may originally have been a
gradient to the east so steep that wheeled vehicles could not readily
negotiate it. South of this part of the ridge rises a stream which flows
eastwards, but this drainage area, being situated on porous sandstone (see

1
The ironstone in the clay has been known to turn rainwater red. This was mentioned by the twelfth-
century chronicler William of Newburgh, who thought that it was blood, and in the nineteenth century
by the Duchess of Cleveland; see Stevenson, ‘Senlac and the Malfossé’, 301. On the Horselodge
Plantation ponds see further, below, p. 49, n. 2. There is also a pit which looks like the product of
human activity, and which sometimes contains water, in the most southerly part of Long Plantation, see
Illus. 18.
2
Below, p. 166, n. 4.
3
ESRO, BAT 4421 (7), one of a series of maps of the Battle Abbey estate drawn by Richard Budgen,
see Illus. 7. Apart from the easternmost part of Warren Wood this stretch of boggy ground lies outside
the land in the ownership of English Heritage.

48
 2 THE SOURCES: PART ONE

below) rather than impervious clay, does not carry the same volume of
water as its bigger brother to the west.
There is considerable vegetation on parts of the ridge today, but no
way of knowing the situation in 1066. Obviously, at least some of it was
sufficiently open for fighting to take place, and most of the battle scenes
in the Bayeux Tapestry show no trees. This is not negligible evidence, as
the Tapestry is not shy about such features (almost fifty appear in it in
all), and does depict trees both in the section where the French are
attacking English troops on an eminence thought by Freeman to be
identifiable with the Horselodge Plantation hillock,1 and in that where the
English are being pursued at the end. Nor, of course, despite the relatively
untouched nature of much of the field, can one be certain that its features
are exactly the same today as they were 900 years ago. The crest of the
ridge, as noted earlier, has undergone significant alterations, while a
recent survey by English Heritage has revealed signs of ridge-and-furrow
cultivation on parts of the battlefield, and that the ponds around
Horselodge Plantation were at some point dammed and provided with
sluices, evidently so that the stream into which they drained to the south-
west of the ridge could be put to industrial use further down its course;
the construction of New Pond, easily assumed to be primarily ornamental
in its purpose, may also have been connected with the same desire to be
able to hold water over the winter so that it could be released during a dry
summer to feed the mill-ponds at Powdermill and elsewhere, thus
facilitating continuous production of the gunpowder for which Battle was
famous between the seventeenth and nineteenth centuries. Still, apart
from the abbey the area to the east is the only one to have been built on,
much of it today containing private houses erected relatively recently.2
Orderic Vitalis, writing in the early twelfth century, repeatedly refers
to the site of the battle by the name of Senlac.3 This was adopted by
Freeman (although subsequent historians have preferred Domesday
Book’s ‘battle of Hastings’) on the not unreasonable grounds that as
Hastings is some miles away ‘a name for the spot is wanted’, and because

1
FNC, iii. 444, 490; his map of the field (opp. p. 443, and below, p. 113) shows it as ‘The English
Outpost’. See further, below, pp. 113-4.
2
The first edition of the OS 6 inch map, Sussex Sheet LVII (surveyed 1873-4, published 1878), shows
only the railway station, a gas works, some buildings on the Hastings road and the National School on
Marley Lane. The 1931 map, Sussex (East) Sheet LVII N.E., hardly differs. The preliminary results of
the English Heritage Survey were announced to their staff at a meeting at Battle on 26 March, 2002. I
am grateful to Mark Bowden, Senior Investigator of the National Monuments Centre, for inviting me to
attend it. On Battle gunpowder, see Blackman, ‘The Story of the old gunpowder works’; there were
five gunpowder mills on the stream, that at Powdermill being the largest, although one, at Farthing
Mills, was not served by the tributary which rises on the battlefield; production ceased in 1874. Rather
than being the result of iron extraction (see above, pp. 47-8) the surveyors raise the possibility that the
Horselodge Plantation ponds, and other apparent quarries on the battlefield, were created by the
extraction of sand and clay subsequent to the battle (Survey, pp. 11, 14, 16).
3
HÆ, ii. 172-3, 180-1, 185-6, 190-1, 266-7, 356-7.

49
 THE BATTLE OF HASTINGS

he knew that the term had survived in ‘the local names for the south-
eastern part of the town’.1 His suggestion that Orderic ‘cannot have
invented the word’, with the implication that it was of English origin, was
scornfully rejected shortly after his death by J.H. Round, who announced
that to ‘anyone acquainted with ‘Old English’ it must instantly occur that
‘Senlac’ is an impossible English name’.2 But with this Round
overreached himself, and in 1913 his position was challenged by W.H.
Stevenson, who showed that the name Sandlake ‘may be found scores of
times in deeds ranging from the twelfth century downwards’ in the
archives of Battle Abbey, and suggested that it did indeed partly derive
not from the Old French lac but from the Old English lacu – a stream or
watercourse; the Old English Sand-lace he thus interpreted as meaning ‘a
sandy brook or a brook that brings down sand’.3 His final conclusion,
based on documents as late as the sixteenth century, was that the area of
Sandlake, one of the divisions of the medieval town:
marched with the abbey precincts from the south-east corner to a spot
opposite the parish church on the north, and that some part… was south of
the abbey. Even if the abbey-site was never within… Sandlake, these facts
prove that the spot where Harold fell and where the high altar of the abbey
4
was erected was within a few yards at most of the limits of… Sandlake.
Yet whether Senlac must mean something as precise as ‘sandy brook’ is
far from certain, for the names Upper Lake and Lower Lake are still in
use in Battle for the road which runs to the north of the abbey and
continues down the southern slope of the ridge towards Hastings, and as
both are in positions too near its summit for them ever to have referred to
any substantial body of running water (except in the event of torrential
rain) the meaning of ‘lake’ here is probably that (deriving from lacu and
surviving in dialect use into modern times)5 of a small channel or drain.
Moreover, on the estate map of 1724 Lower Lake is actually named
‘Sanglake’.6 Certainly the geology of the site is no stranger to sand.7 The
1
FNC, iii. 746. He was not the first to use Orderic’s term, see Lappenberg, A History of England under
the Anglo-Saxon Kings, ii. 301; Lingard, The History of England, i. 188.
2
Round, anonymous article in the Quarterly Review of 1892, 11; repeated, Feudal England, p. 339.
3
Stevenson, ‘Senlac and the Malfossé’, espec. 293-6; similarly, Mawer and Stenton, The Place-Names
of Sussex, ii. 499. Ekwall, Concise Oxford Dictionary of English Place-Names, p. 412, gives sand-lacu
as ‘sandy brook’; Stevenson’s Sand-lace is the dative form. Bradbury’s apparent ignorance of
Stevenson’s article led to the erroneous statement that Senlac is ‘an otherwise unknown place name’
(The Battle of Hastings, p. 175) and has misled Grehan and Mace, The Battle of Hastings, p. 115. If it
is identifiable with the later area of Sandlake this is a problem for those who would place the battle on
Caldbec Hill, see p. 45, n.7.
4
Stevenson, ‘Senlac and the Malfossé’, 300; for an earlier commentary, see Lower, ‘On the Battle of
Hastings’, 37-8.
5
See OED under ‘lake’, ‘leach’ (both thought to be connected with the Old English verb leccan, to
moisten) and ‘letch’.
6
ESRO, BAT 4421 (6), another map in the series drawn by Richard Budgen, Illus. 3.
7
The following description is drawn from the British Geological Survey 1:50,000 Series, Sheet
320/321, Hastings & Dungeness; see Illus. 6 and Lake and Shephard-Thorn, Geology of the country
around Hastings and Dungeness. I am also grateful to Paul Littlewood for comment on the ground.

50
 2 THE SOURCES: PART ONE

western part of the ridge, and part of the northern face of its eastern part,
consist of Wadhurst Clay, which further south and west is eventually
replaced by Tunbridge Wells Sand. There is a small outcrop of this too at
the point where the monastic buildings were erected, and a fault line
divides this from the Ashdown Beds (also a sandstone) which form most
of the eastern portion of the ridge. The road now known as Lower Lake
cuts into this Ashdown sand (the surrounding ground surface being at
points several feet higher) and it may be that there was once a sandy
channel in this location from which the area known as Senlac, and
eventually the battle, took their names.1 Alternatively, if the name does
denote a sandy brook this could refer to either of the streams which rise
on each side of the saddle which carries the road to Hastings, the first
(and more insubstantial) flowing east and the second west to New Pond
and eventually Powdermill Lake, for the impervious Wadhurst Clay
which contributes so much to the volume of the latter, and through which
it runs, is at points very sandy in its nature, and the thickets of gorse
which today cover some of the area of Horselodge Plantation must be
growing on a superficial capping of sand. It is unlikely, but not
completely impossible, that both these derivations contain elements of
truth, the district of Sandlake perhaps being named after a sandy channel,
and the battle of Senlac after a sandy stream.
The documentary sources produced by the English are nothing like as
extensive as those which issued from their conquerors. The most
important are the two versions of the Anglo-Saxon Chronicle which deal
with the battle, that is those denoted D and E by historians. The latter is
the tersest of all the written evidence, and perhaps the most
contemporary. E is a manuscript written in the abbey of Peterborough in
the third decade of the twelfth century,2 but one of the sources of which it
made use was a chronicle produced in the monastery of St Augustine,
Canterbury during the mid-eleventh century.3 The St Augustine’s monks
knew a great deal about events in the south of England during the
Confessor’s reign and sometimes had strong views about them, for this
part of E tends to be distinctly pro-Godwinist. The last entry clearly
linked to St Augustine’s is that for 1061, but the information on Harold’s
naval expedition against William in 1066, which could have been
launched from Sandwich, hints that this predecessor of E was still in
Canterbury in that year, as does its use of the word ‘courageous’ to

1
The road need not originally have run in this channel. A number of commentators have noted the way
in which the present road skirts the abbey grounds and suggested that originally it took a more direct
line over or very near the spot where Harold was killed; see, for example, Lemmon’s map in The Field
of Hastings. However, in this case the gradient to the south would have been very much steeper, and
this suggests that the present line need not be far from that of the original.
2
E is Bodleian Library, MS Laud Misc. 636; Ker, Catalogue, pp. 424-6.
3
As Charles Plummer established, ASC, ii. xlviii-l.

51
 THE BATTLE OF HASTINGS

describe his victory at Stamford Bridge. Its very simple record of the
battle of Hastings, which confines itself to saying that he fought before all
his army had come, and was killed along with his brothers Gyrth and
Leofwine, also seems sympathetic.1 Under 1065, E comments that the
Northumbrians’ ravaging of Northamptonshire left it and neighbouring
shires the poorer for many years. This statement (which is also in D) is
obviously not contemporary, but may have been added at a subsequent
stage in the text’s transmission. Nothing is certain, but E’s account of the
battle, such as it is, was perhaps written at a time very close to that of the
event itself.
The Chronicle D text is more informative, even if its provenance is
more difficult to establish. Providing the significant information that the
English force was a large army (mycelne here), that it met at the hoary
apple tree, and that William came upon it by surprise before it was
properly ordered (Wyllelm him com ongean on unwær ær his folc
gefylced wære), it, too, is sympathetic to Harold, who is said to have
fought most resolutely with those who wished to follow him (a hint that
some did not?), and to have been killed with his brothers Gyrth and
Leofwine and many good men, after great slaughter on both sides. Rather
later in the entry there appears an idea which was also taken up by others
– that the misfortunes of the English were God’s punishment of sin. 2 The
date of the scribe responsible for it is uncertain,3 as is his location,
although Worcester is a strong possibility. He cannot have been writing
soon after the battle, as the entry for 1065 shares E’s comment on the
long-term effects of the ravaging of Northamptonshire by the
Northumbrian rebels, but like E he may (or may not) have been copying
older material on the battle itself. D uses texts of the C and E type
frequently for this period, and continues to use E into the 1070s (D itself
comes to an end in 1079). If our C did not break off before getting to
Hastings, it might well have shown that D’s account of the latter owed
something to a manuscript of the C type, as it certainly does in the earlier
part of its entry for 1066. If so, and as C itself is in this period a
contemporary manuscript written in the monastery of Abingdon,4 the
1
Ibid., i. 198.
2
Ibid., i. 199-200.
3
D is British Library, MS. Cotton Tiberius B. iv; Ker, Catalogue, p. 254, thought that at 1051 ‘the
manuscript is perhaps a contemporary record’, but Plummer (ASC, ii. lxxviii-ix) judged none of D
‘probably from much before 1100’, while Dumville, ‘Some aspects of annalistic writing’, 34, suggests
limits of 1080x1130 for its compilation. Its most recent editor, G.P. Cubbin, claims Ealdred (bishop of
Worcester c.1046-62, archbishop of York 1061-9) as ‘the person responsible for the creation of D’, and
that ‘D was compiled in the 1050s’ (Anglo-Saxon Chronicle…MS D, p. lxxix). Plummer argued that the
material on St Margaret of Scotland in 1067 is a late insertion, and that the entry as it now stands was
not written until after her daughter’s marriage to Henry I in 1100.
4
British Library, MS. Cotton Tiberius B. i. Ker, Catalogue, p. 252, commented that the ‘entries for
1065, 1066… appear to be contemporary, or nearly contemporary, with the events described’. On
Abingdon as the home of C, ASC, ii. xcii.

52
 2 THE SOURCES: PART ONE

authority of D’s information might be somewhat strengthened. Of course,

date of composition is only one factor to be taken into account when
assessing a source’s reliability; another is bias. Even so, D’s comments
on the battle are not obviously prejudiced, although the favour which this
source seems to show Harold’s family elsewhere – notably in its omission
from its C text exemplar of the information that Godwin was involved in
the opposition to Edward the Confessor’s brother Alfred in 1036 – can
perhaps be discerned within it.1
Not long after its composition, D, or a manuscript very like it, lay
before John of Worcester, who compiled a Latin chronicle on the basis of
various sources, including probably more than one manuscript of the
Anglo-Saxon Chronicle,2 in the first half of the twelfth century.3 John was
one of the famous quartet of Anglo-Norman historians which also
included William of Malmesbury, Henry of Huntingdon and Orderic
Vitalis, and which brought to the study of the past a professionalism
hardly equalled in England since the days of the great Northumbrian
scholar Bede (d. 735), and not always equalled even by Bede himself. He
assembled a wide range of sources, and sometimes draws the reader’s
attention to discrepancies between them (something Bede’s historical
works never do);4 he also provided his readers with carefully researched
lists of the bishops of all the English sees, and genealogies of the royal
families of the different Anglo-Saxon kingdoms, while his additions to
the eleventh-century entries in the Anglo-Saxon Chronicle are extensive
and have usually to be taken seriously. Doubtless many things
contributed to the golden age of English historical writing reflected by his
works and those of his fellows, but one of them was an awareness of the
changes brought about by the Norman Conquest and a desire to record the
proud history of the Anglo-Saxons before it passed forever into oblivion.
Among his sources was Eadmer’s Historia Novorum in Anglia (History of
Recent Events in England), in large part a biography of Archbishop
Anselm of Canterbury originally completed just after his death in 1109,
but with stories to tell about Harold’s visit to Normandy5 and the events
of 1066. Eadmer does not describe the battle itself in detail, but confines
himself to commenting that Harold fell in the line after heavy fighting
had been joined, and that the French who were there still said (clearly
implying that he had spoken to participants) that although fortunes had
fluctuated there was such a slaughter and flight of the Normans that their

1
A point made by Plummer, ibid., ii. lxxxii.
2
See Plummer’s remarks, ibid., ii. lxxxiii-vi.
3
His most recent editors suggest terminal dates of 1095x1106-1140x1143, and that ‘within these limits,
the 1120s and the very early 1130s seem crucial’, JW, ii. lxxxi.
4
For this and what follows, see Campbell, ‘Some Twelfth-Century Views of the Anglo-Saxon Past’,
pp. 213-4.
5
Above, p. 29.

53
 THE BATTLE OF HASTINGS

victory could only be attributed to the power of God in punishing

Harold’s broken oath.1 Eadmer, while a great friend of an Italian who had
become archbishop of Canterbury as a result of the Conquest, was
nevertheless strongly pro-English in his sympathies,2 and it is not difficult
to detect a native bias in these statements. Nor was John of Worcester
without his prejudices in the same kind of area, and they were prejudices
which unfortunately, and despite his virtues in other respects, to an extent
invalidate his comments on the same subject matter.3 Thus, he takes
considerably further the tendency of the D text of the Chronicle to offer
reasons, if not excuses, for the English defeat. The report of William’s
landing at Pevensey adds that he was accompanied by an innumerable
multitude of horsemen, slingers, archers and infantry, as he had brought
with him strong forces from the whole of France. King Harold is then
said to have moved his army to London in a great hurry and to have
advanced into Sussex as quickly as he could, although half his forces had
not yet assembled and he was well aware that the stronger men in all
England had fallen in the two earlier battles. He fought before a third of
his army was drawn up, on Saturday, 22 October (an error for Saturday,
14 October), but because his forces were arrayed in a narrow place many
withdrew from the line, and very few of constant heart remained with
him. Nevertheless, from the third hour (i.e. about 9 o’clock) until twilight
he most stoutly resisted his enemies, and fought so strenuously that it was
scarcely possible for the enemy to kill him, until, alas, at dusk, after many
had fallen on both sides, he himself fell; the earls Gyrth and Leofwine,
his brothers, were also among the dead, and the more noble of almost all
England.4
Internal criticism alone might cast doubt on the plausibility of this
account. For example, if Harold’s forces were really so inadequate why
did the battle last as long as it did? Yet when it is considered within the
context of this author’s remarks on the year as a whole it is not difficult to
see what has happened. It must be indicative of the passions which the
Conquest excited even fifty years later that John, who is said by Orderic
Vitalis to have been English,5 is (unlike his fellow historians) strongly in
favour of King Harold. He is also the only near-contemporary source to
assert that the latter was crowned by Archbishop Ealdred of York rather
than Stigand of Canterbury, whose position as archbishop was not
recognised by the papacy, and whom Norman sources identify as the

1
Historia Novorum, ed. Rule, pp. 8-9.
2
See Williams, The English and the Norman Conquest, pp. 165-8.
3
See further, ibid., pp. 168-9.
4
JW, ii. 604-5.
5
HÆ, ii. 186-7.

54
 2 THE SOURCES: PART ONE

bishop responsible in an attempt to further impugn Harold’s legitimacy. 1

Both sides had a motive to lie, and in this case it is impossible to be sure
which is telling the truth.2 Ealdred had been bishop of Worcester before
his promotion to York, and John may be using an authentic tradition. On
the other hand, the enthusiasm which he subsequently shows for King
Harold must give pause. Alleging him to have been the subregulus
(‘underking’) before Edward’s death, John continues by extolling the
excellence of his rule: he destroyed unjust laws, was a patron of churches
and monasteries, and was pious, humble and affable to all good men,
while detesting criminals; these his subordinates were ordered to seize, as
also to prepare for the defence of the country by land and sea.3 There is
nothing of this sort in the surviving texts of the Anglo-Saxon Chronicle,
and one might doubt whether John had much authority for it. His account
of Hastings seems to be biased in the same direction. Here, the Chronicle
E text’s statement that Harold fought before all his army had come looks
as though it has been elaborated into ‘before half his army had come’,
while D’s assertion that the English were attacked before they were
properly arrayed becomes ‘before a third of his army had been drawn up’.
Similarly, D’s statement that Harold fought alongside those who wished
to follow him may hint that some did not, and John apparently works this
up into many of the English withdrawing from the line because the space
was too narrow. Finally, and significantly, he omits altogether D’s
assertion that the English army was ‘large’. It is not possible to dismiss
all the details he supplies, because the restricted nature of the English
deployment seems to be confirmed by other writers.4 Nevertheless, one
could be forgiven for doubting whether much of what he says has any
great authority behind it.
The sympathy for the English shown by Eadmer and John is also to be
found in the work of a greater writer, their contemporary William of
Malmesbury, who visited Worcester and perhaps met John; certainly he
used the latter’s chronicle as one of his own sources. William apparently
wrote his Gesta Regum Anglorum (History of the Kings of the English)
as a result of a visit paid to Malmesbury by King Henry I’s queen,
Matilda, shortly before her death in 1118. As the granddaughter of

1
WP, GG, pp. 100-1. The Bayeux Tapestry shows Stigand standing beside the newly-crowned king,
thus apparently making the same point. OV, HÆ, ii. 136-9, says that Stigand performed the coronation
without the consent of other bishops; he also presents a much less rosy picture of Harold’s reign than
JW.
2
For a full discussion, see FNC, iii. 612-18, who predictably came out in favour of the Worcester
writer, and Chibnall’s note, HÆ, ii. 138, n. 1.
3
JW, ii. 600-1.
4
Below, pp.165-6, on the great density of the English line.

55
 THE BATTLE OF HASTINGS

Edward the Exile and great-granddaughter of King Edmund Ironside,1

Matilda (whose English name had been Edith) was a member of the Old
English royal family, and she asked William to write an account of her
royal predecessors; he agreed, being unwilling that the deeds of such men
should be forgotten.2 The Gesta Regum he claims to be the first history of
the English written in Latin since the days of Bede,3 and both here and in
his Gesta Pontificum, on the history of the English church, he used many
of the techniques of a modern historian, cross-checking his sources and
often drawing the reader’s attention to results which were not
satisfactory. He gives details, for example, of the murder of Edward the
Confessor’s brother Alfred which he claims to be well-known, but about
which he expresses reservations because they are not in the Anglo-Saxon
Chronicle. Similarly, he warns the reader that in his own day the English
and the Normans still disputed the rights and wrongs of the quarrels
between the Godwin family and King Edward’s Norman supporters, and
that the truth is therefore uncertain.4 His history of the abbey of
Glastonbury was researched there,5 and he visited many cathedrals and
monasteries in the course of his studies, examining their archives and
listening to their oral traditions. The list of classical and medieval sources
(including administrative documents) of which the Gesta Regum made
use either at first or second hand is a remarkable one.6 It is not therefore
surprising that William of Malmesbury’s comments on the battle of
Hastings are part of a more general treatment of the events of 1066 and
their significance, of a type considerably more ambitious than anything
attempted by the sources discussed thus far.
He was of mixed descent, probably having a French father and an
English mother, and he states explicitly what John of Worcester leaves
implicit – that the battle was a fatal day for England (‘our sweet country’,
as he calls it) because of the damage that was to come about as a result of
the change of lords. He then embarks upon some general reflections on
Old English history: the Anglo-Saxons had once been famous for their
practice of Christianity, but in the years before the Conquest their
attention to literature and religion had decayed, monks had ignored their

1
She was the daughter of Edward the Exile’s daughter Margaret and King Malcolm III of Scotland.
Henry had married her in 1100, probably to bolster his support among the English. ASC E (i. 236)
points out that she was of ‘the rightful royal house of England’.
2
GR, i. 6-9. This information is from a letter written to her daughter the empress Matilda; it survives in
only one manuscript of the Gesta Regum.
3
GR, i. 14-15.
4
GR, i. 336-7, 354-5. See further, Thomson, William of Malmesbury, esp. pp. 1-38, 72-5; Gransden,
Historical Writing in England, pp. 166-85; Williams, The English and the Norman Conquest, pp. 171-
4.
5
Campbell, ‘Some Twelfth-Century Views’, p. 214, describes it as ‘perhaps the most remarkable piece
of historical research of its day’,
6
See GR, ii. 458-68.

56
 2 THE SOURCES: PART ONE

Rule by wearing fine vestments, and the nobles had given themselves up
to good living, eating until they were surfeited and drinking until they
were sick. Indeed, entire days and nights had been spent in drinking, and
the vices which result from drunkenness had followed. Thus it was that
they engaged Duke William with more rashness than military science,
dooming their country to slavery through a single, easy victory. The
Normans, on the other hand, did not eat to excess; a race which could
hardly live without war, their arrival in England had revived its religious
life, so that churches appeared in every village and monasteries in the
towns, both built in a style previously unknown.1
These comments come after William of Malmesbury’s account of the
battle, and their apparent even-handedness is at first sight reflected in his
treatment of it. He had heard that the English had spent the previous night
drinking and singing while the Normans were confessing their sins and
receiving communion, and Harold he regards as a usurper; yet his
prowess as a warrior is admired, as is the band of Englishmen, few in
number but brave in the extreme, who fought and died for their country.2
It is thought that he used texts of the Anglo-Saxon Chronicle D and E
type,3 and he says much that evidently came from William of Poitiers’
Gesta Guillelmi, but he also had information from other sources, some of
them probably oral, which have not survived: only he, for example, tells
how Harold refused to share the booty from the battle of Stamford
Bridge and was accordingly accompanied to Hastings mainly by
professional soldiers, and how the Conqueror fell as he came ashore at
Pevensey, to be reassured by one of his men that he was holding England,
and would be king.4 Hence, his comments about the battle are both
visibly linked to those of other writers, and also independent of them in
ways that matter. He may have had the statements of William of Poitiers
(and perhaps also the Chronicle D text) in mind when he says that he
disagrees with those who assert that the English army was large, thus
diminishing their courage and bringing dishonour on the Normans, whom
they intend to praise; for it is hardly a commendation that they conquered
a people impeded by their numbers.5 He also states that the English used
a dense formation covered by shields (the shield-wall) and that this was
broken by a feigned retreat on the part of the Normans, who turned upon
the disordered ranks of their enemy. Knowledge of the latter probably
came from William of Poitiers, but this source says nothing of the
protective shields, and mentions two feigned flights following upon one

1
GR, i. 456-61.
2
Ibid., i. 454-7.
3
ASC, ii. lxxxvi-vii. In earlier passages of GR he refers to chronicles in Old English more than once.
4
GR, i. 422-3, 450-1.
5
Ibid., i. 422-3.

57
 THE BATTLE OF HASTINGS

real one, which the duke managed to halt. Even after this, according to
him, the English fought on doggedly, while William of Malmesbury
seems to have thought that their flight, once begun, largely continued; it
was during this process that, having occupied a hillock, they threw back
the Normans into the valley below, destroying them by hurling spears and
rolling down stones; also, they were able to avoid a deep ditch, but killed
so many of their enemies there that the bodies made the hollow level with
the surrounding ground. Of the hillock, William of Poitiers says nothing
whatever, while differences of detail make it improbable that the ditch
story owes very much to a similar tale placed at the end of the battle in
the Gesta Guillelmi.1 Similarly, William of Malmesbury’s account of the
duke’s military prowess, and of the three horses which were killed
beneath him, is shared with the earlier writer, while he added the
information that Harold fought so effectively that he could destroy horse
and rider with a single blow (possibly a detail drawn from a chanson de
geste), and that as none could come near him he was eventually killed
from a distance by an arrow.2
In short, not only is this a more detailed and ambitious account than
that of John of Worcester’s chronicle, it also has to be taken much more
seriously. Yet despite William’s qualities as a historian it cannot be
accepted at face value. Thus, while he may well have had sources which
have not come down to us, they need not have been particularly reliable.
The tale about Duke William’s fall as he came ashore, for example, is
very like one which the Roman historian Suetonius tells of Julius Caesar,3
and may have been taken from a source which (like William of Poitiers)4
sought to flatter the duke by making the obvious comparison. It has also
been pointed out that William of Malmesbury is sometimes more critical
of his sources when dealing with the distant past than when speaking of
more recent history; in particular, he gives more credence to miracle
stories than one can be at all comfortable with.5 This is a reminder that he
he was not free of the constraints imposed by his religious beliefs, and
raises the possibility that his account of the Conquest may have been both
influenced and to a degree distorted by his desire to see it as the
judgement of God on a sinful people, the kind of judgement natural to
any churchman, and one which he would in any case have come across
repeatedly in his perusal of Bede’s Ecclesiastical History; it is one which
he also expressed in his life of Bishop Wulfstan II of Worcester, who is
said to have prophesied to King Harold himself that disaster would befall
1
On ditches at Hastings, see further, below, pp. 166, 175-7, 186-7; there may have been more than one
incident involving them.
2
The suggestion on the chanson I owe to Dr John Hudson.
3
GR, ii. 230; Suetonius, Divus Iulius, chapter lix.
4
Below, pp. 83-4.
5
Campbell, ‘Some Twelfth-Century Views’, pp. 222-3.

58
 2 THE SOURCES: PART ONE

the country if it did not mend its evil ways. 1 William of Malmesbury’s
views were probably also shaped by the fact that, while he pays lip-
service to the Normans, his heart lay with their opponents. This is very
clear once one turns to other material in the Gesta Regum. For example, it
reports a portent which allegedly appeared on the border between
Brittany and Normandy – two women, one of a happy and the other of a
sad disposition, who were joined together at the waist so that what they
ate went into a single digestive tract. Eventually one died, and the other
had to drag her corpse around until nearly three years later she died too.
Some speculated that they represented England and Normandy, the
demands of the latter being supported (in William of Malmesbury’s day)
by the wealth of the former, until perhaps she would eventually succumb
to the violence of her rulers; the conquest was, he says, a calamity, which
had resulted in the extinction of the ancient nobility and the ravages of
taxation.2
In the light of this, one is entitled to ask if his treatment of 1066 is
quite as impartial as it appears. Had he really received information about
Harold’s prowess in the battle, for example, or did he insert it himself to
balance the material on Duke William’s martial exploits?3 Is it significant
that he tells how the duke punished a soldier who slashed at the English
king’s thigh as he lay on the ground, but omits William of Poitiers’
statement that the face of the corpse was so badly damaged as to be
unrecognisable, or that William of Poitiers says the Conqueror refused to
hand the corpse over to Harold’s mother in return for its weight in gold
while in the Gesta Regum it is handed over without ransom after a large
sum had been offered?4 Similarly, might there be a degree of nationalist
bias in his insistence that the English army was small but fought with
great bravery? If so, he could have found justification in sources which
might seem to lend the idea credence. His account of 1066 clearly owes
something to a text of the Anglo-Saxon Chronicle similar to E, which of
course says that the English fought before all their forces had come, and
may also have given rise to William’s rather surprising statement that
Harold took no military precautions before he knew of the threat from
Norway.5 One could certainly deduce this from E (which says nothing
about the preparations to meet the Normans on the south coast) if one

1
WM, Vita Wulfstani, p. 23. The wearing of long hair by men is one of the evils mentioned, it being
evident that those who chose to imitate women in their appearance would prove ineffective in defence
of their country; however, the bishop carried with him a little knife with which he trimmed the
offending locks whenever possible. WP (GG, pp. 178-81) says that the long-haired Englishmen taken
by William to Normandy in 1067 yielded nothing to the beauty of girls.
2
GR, i. 384-7.
3
As Thomson and Winterbottom suggest, GR, ii. 235.
4
GR, i. 456-7, 460-1; GG, pp. 140-1.
5
GR, i. 446-7.

59
 THE BATTLE OF HASTINGS

ignores its statement about the naval expedition against Duke William; 1 it
may even be that William of Malmesbury’s copies of the Chronicle did
not contain this sentence. Yet this will not quite do, for both the D text
and John of Worcester do describe Harold’s preparations, and William’s
familiarity with their material at this point is suggested by the fact that,
like them, he calls Harald of Norway Fairhair rather than Hardrada.2 He
would not have concurred with the favour John shows to Harold, but may
have been influenced by his efforts to diminish the size of the army which
fought at Hastings.3 It looks as though William of Malmesbury, like other
historians, sometimes took from his sources no more than the material
which suited his own interpretation – that the war was a minor one
because it was God’s purpose that the English would never again fight in
defence of their freedom.4
Not long after he began work on the Gesta Regum, his contemporary
Henry of Huntingdon turned to a similar task. Like William, Henry may
have been the product of a mixed marriage, being the son of Nicholas,
archdeacon of Huntingdon, and an unknown Englishwoman.5 He says
that he did the work at the command of Alexander, who became bishop of
Lincoln in 1123, and the first version of his Historia Anglorum (History
of the English) ended with 1129; there were also a number of subsequent
versions, the sixth and final one taking events down to 1154.6 Although a
secular cleric rather than a monk, he shared with William of Malmesbury
and others the idea that the Conquest was God’s judgment upon a sinful
people. Indeed, the whole of his history is one with a ‘serious and an
almost despairing moral purpose’,7 in which all five invasions of Britain –
– by Romans, Picts and Scots, Anglo-Saxons, Danes and Normans – are
viewed in this way.8 But it had other purposes too. One was to act as a
readily intelligible handbook to English history, an end which its author
sought to achieve by carrying out a thorough reordering of the accounts
of the early Anglo-Saxon period which he found in Bede and the Anglo-
Saxon Chronicle;9 another, subsidiary perhaps, but still important, was to
entertain. Written in fairly simple Latin, the ten books of roughly equal
length could each have been read at a single sitting, and contained plenty
1
Above, pp. 33-4.
2
GR, i. 422-3; ASC D, i. 199; JW, ii. 602-3. So do other twelfth-century sources, see Plummer’s
comments, ASC, ii. 256. WM’s reference to those who thought the English disorganised by their
numbers may well be an allusion to ASC D, see above, p. 52.
3
Above, p. 55.
4
GR, i. 422-3.
5
HA, pp. xxiv-vi. The Englishness of Henry’s mother his most recent editor deduces from his
familiarity with the English language and the theme of the HA; there is no direct evidence.
6
HA, pp. lxvi-lxxvii.
7
Campbell, ‘Some Twelfth-Century Views’, p. 213.
8
HA, p. lix. See also on HH, Partner, Serious Entertainments, pp. 11-50; Williams, The English and the
Norman Conquest, pp. 176-80.
9
Campbell, ‘Some Twelfth-Century Views’, pp. 212-3.

60
 2 THE SOURCES: PART ONE

of incident. His descriptions of battles, in particular, were couched in a

poetic style (he was a ‘not inconsiderable poet’) and sometimes enlivened
by rhetorical speeches of his own composition; one of these he eventually
put into the mouth of Duke William before Hastings.1
As Henry of Huntingdon saw the Conquest as the wages of sin, it is
hardly surprising that his treatment of the Godwin family is very
unsympathetic. He comments on the savagery of the behaviour of Harold
and Tostig, and illustrates this with a story of how they quarrelled at
Windsor in 1063, and of how Tostig then went to Hereford, where Harold
had prepared a banquet for King Edward, killed and dismembered his
brother’s servants, and placed their heads, arms and legs into the vessels
which contained the drink.2 His account of the events of 1066 begins with
with the death of King Edward and Harold’s usurpation of the throne,
ignoring the Chronicle E text’s notice3 of his designation by Edward and
subsequent election. Then he turns to the reasons why Duke William
decided to invade and describes William FitzOsbern tricking the other
Norman magnates into agreeing to participate, and the assembly of a very
large fleet at St Valéry. From here Henry follows the Anglo-Saxon
Chronicle with minor additions down to the defeat of Harald Hardrada
(he gives the story of the lone Norwegian hero who defended the river
crossing at Stamford Bridge) and then moves swiftly to Hastings, where
he says Harold drew up his lines in open country (in planis Hastinges)
and William deployed five companies of horsemen against him. In the
third version of the Historia, written c.1140,4 he prefaced his account of
the battle with a speech by Duke William which was supposedly so
successful in incensing his men that they charged the enemy before he
had finished. As the opposing forces came together, one Taillefer juggled
with swords before the English and killed one of their standard-bearers as
they stood stupefied, repeated the trick, and was then killed himself.
Next, clouds of arrows filled the air, but the English could not be broken
by the enemy because Harold had formed his men into a single very
dense formation, like a castle. William then ordered a feigned retreat, but
as his men retired they came to a cunningly-hidden and large ditch where
a great many of them fell. As the pursuit continued, however, the
Normans broke through the English centre, and the latter too then
suffered heavy casualties as they recrossed the ditch, where the greater
part of them died. Then the duke ordered his archers to shoot into the air
so that the enemy might be blinded, and this too caused heavy casualties.
As twenty of his strongest horsemen took a mutual oath that they would

1
HA, pp. lviii-lx.
2
HA, pp. 382-3.
3
For his use of texts of the C and E type, see ASC, ii. lv-lviii; HA, xci-iii.
4
On the date, see HA, pp. lxx-i, and n. 61.

61
 THE BATTLE OF HASTINGS

capture the royal banner known as the ‘Standard’, and did so despite
losses, Harold was hit in the eye by an arrow, and then killed along with
his brothers Gyrth and Leofwine when a mass of enemy cavalry broke
through. Thus the English were defeated.1
Several significant features of this story, for example the feigned
retreat and the deaths in a ditch, appear elsewhere, and it is important that
Henry of Huntingdon’s account cannot in fact be shown to have
borrowed anything directly from any other written source that has
survived. The documentary sources upon which he drew were much less
extensive than those of William of Malmesbury,2 and included neither the
histories of William of Jumièges and William of Poitiers, nor the Latin
Carmen, nor (probably) the Bayeux Tapestry.3 The Carmen certainly
mentions the juggler Taillefer4 (by the Latin name Incisor-ferri) and gives
an account of his exploits neither identical nor very dissimilar to Henry’s,
but it is likely that both were drawing on the same story or stories, and
Henry’s ability to give the name in the vernacular, and his use of the
French word ‘Standard’, may mean that his Latin account of the battle
drew on a chanson in French verse.5
Shortly before 1140 another secular clerk, Geoffrey Gaimar, applied
himself to a rather different process by turning the Anglo-Saxon
Chronicle into French verse for Constance, the wife of Ralph FitzGilbert,
an Anglo-Norman landowner in Lincolnshire. Gaimar’s L’Estoire des
Engleis is the earliest surviving history written in French,6 and a powerful
powerful reminder that churchmen were not the only people interested in
the past; for others, too, it provided both instruction and entertainment.
The desire to entertain was never far from the minds of writers like
William of Malmesbury and Henry of Huntingdon, and the extent of their
success is demonstrated by the considerable number of manuscripts of
their works which has survived, and by the works themselves, for their
authors were men of wide interests and knew how to tell a good story.
William, for example, does not stop at recounting how a Malmesbury
monk named Æthelmær prophesied that the appearance of Halley’s
Comet in 1066 portended the downfall of his country, but adds some
startling information about the exploits of the prophet’s youth: hoping to
fly like Daedalus, whose story he took to be true, he had fixed wings to
his hands and feet, jumped off a tower and flown for some distance,
before crashing to the ground and being crippled for life; he always

1
HA, pp. 384-95.
2
Upon them, see HA, pp. lxxxv-cvii.
3
On this, see below, pp. 64-74.
4
Carmen, ll. 389-404.
5
For this argument and other uses of ‘standard’, see Greenway’s comments, HA, pp. cvi-vii. The word
also appears in the rather earlier Battle Abbey Brevis Relatio, below, p. 94.
6
Campbell, ‘Some Twelfth-Century Views’, p. 211.

62
 2 THE SOURCES: PART ONE

regretted not having fitted a tail to his backside.1 The twelfth century was
not only a great age of historical writing, it was also a great age of story.
At the request of the same Bishop Alexander of Lincoln who interested
himself in Henry of Huntingdon, the Oxford cleric Geoffrey of
Monmouth wrote a work on the prophecies of Merlin which became
popular throughout Europe, and followed it up with his immensely
influential and almost entirely fictional Historia Regum Britanniae
(History of the Kings of Britain), of which the chief hero was King
Arthur. A treatment of the Historia in French verse was quickly produced
by Wace, and this was in turn paraphrased into English verse by
Layamon; and by the latter's time Chrétien de Troyes had already written
of Guinevere, Lancelot and Percival, and of the Round Table.
Gaimar, too, began by composing for Constance a version of Geoffrey
of Monmouth in French verse (it has not survived), and to this he later
added L’Estoire des Engleis. The dividing line between history and
fiction in such works might at times seem almost non-existent, and
Gaimar’s desire to entertain is very obvious, for the later sections of his
work in particular are full of romance, but this should not obscure the fact
that such writers could derive from popular songs and stories material
which contained authentic elements,2 and that Gaimar had heard tales
about the events of 1066 there is little need to doubt. By this point 3 his
debt to the Anglo-Saxon Chronicle, at least in the copies which survive
today, is not very extensive, and his accuracy not always dissimilar: early
in the year, for example, he wrongly has Queen Edith dying at the same
time as King Edward, and he has nothing at all to say about Harold’s
accession to the throne. However, he does give details on Tostig’s
harrying of England and his subsequent operations with Harald Hardrada
which are not in any other source. They are said to have had 460 ships 4
and to have found themselves opposed by the men of seven shires under
the earls Edwin and Morcar, whom they defeated after great slaughter on
both sides. Stamford Bridge he names Punt de la bataille (‘the bridge of
battle’) and says that nobody could count a half of those left on the field.
Then the French landed at Hastings with 11,000 ships,5 and when Harold
heard of it he gave to Bishop Ealdred (i.e. Archbishop Ealdred of York)
the booty taken from the Norwegians and then spent five days
summoning an army in the south, although he could obtain few troops
because of his earlier losses; still, accompanied by his brothers Gyrth and

1
GR, i. 412-5.
2
On the probable authenticity of some elements of Gaimar’s treatment of the early eleventh century,
see Lawson, Cnut, pp. 77-9. The same point is true of Wace’s Roman de Rou, upon which see below,
pp. 95-101.
3
What follows is L’Estoire des Engleis, ed. Bell, ll. 5129-338.
4
L’Estoire, l. 5200; two of the four surviving manuscripts give 470.
5
L’Estoire, ll. 5242-3; one manuscript reads 9,000.

63
 THE BATTLE OF HASTINGS

Leofwine and their men he advanced into Sussex. Gaimar has

surprisingly little to say about Hastings itself.1 The story of Taillefer
occupies the bulk of the lines which he originally devoted to it, although
his version differs from that of Henry of Huntingdon. Here, Taillefer
juggled with a lance, throwing it into the air three times and catching it by
the head before hurling it into the English lines on the fourth occasion
and wounding a soldier; then he juggled thrice with his sword before
approaching the enemy lines again, cutting off the hand of one of their
men and striking at another; but now he found that it was in an evil hour
that he had asked for the first blow, for the English killed both him and
his horse. Battle was then joined and lasted until the evening, but
eventually the English had the worst of it and Harold and his brothers fell.
Gaimar later inserted a section on the exploits of Alan (Rufus) of Brittany
into his comments on the battle and mentioned that the Conqueror
rewarded him with the fair castle of Richmond (Yorkshire),2 but it looks
as though he thought events in Sussex, like the character of King Harold,
of less interest to his audience than the fighting in the north which had
preceded them.
Having surveyed the most important written sources produced in
England by authors of native or semi-native origin within a century of the
battle, it is now time to turn to the Bayeux Tapestry. Not in fact a tapestry
at all, but a series of nine conjoined linen strips of various lengths (the
first two being much more extensive than the rest) embroidered with
wools of ten principal colours obtained from three dyes, it is today
roughly 68 metres long and of a height which varies between about 45
and 53 cm.3 It was once longer, as it is possible that a number of scenes
have been lost at the start4 and it breaks off at what is easily assumed to
be the end of its depiction of the battle of Hastings. Moreover, even what
remains was heavily restored in the nineteenth century and reworked at
earlier dates too.5 Certainly, it is easy to be unaware how much of the
Tapestry’s existing stitchwork is not original, and an edition which
addresses these problems is greatly to be desired. There was a time, for
example, when it seems to have been the custom of those who displayed
the work to visitors to allow them to remove small pieces as souvenirs, a
number disappearing in this way and eventually having to be made good.
According to Dr David Hill, whose work in this area is critical, there is a
1
Gaimar’s account of the battle is discussed by Eley and Bennett, ‘The Battle of Hastings’, 47-54, who
think it ‘anti-Norman’ and ‘an anglocentric epic of Harold’.
2
L’Estoire, ll. 5309-26; see Bell’s note on this section, p. 267.
3
More precise figures are available, but upon the difficulties with them, see Renn, ‘How Big is It?’, pp.
52-7. On the number of colours, Bédat and Girault-Kurzemann, ‘Étude technique de la Broderie de
Bayeux’, p. 91.
4
Renn, ‘How Big is It?’, pp. 56-7.
5
Lemagnen, ‘The Hidden Face of the Bayeux Tapestry’, pp. 37, 40, speaks of ‘many restorations over
the centuries’ and (re the beginning) ‘successive campaigns of restoration’.

64
 2 THE SOURCES: PART ONE

minimum of 379 places where the present version is or may have been
corrupted.1 Thus, there are points where the accuracy of the restorations
and the status of some of the accompanying inscriptions become matters
of the greatest importance and it is necessary to turn to early drawings
and to reproductions published in the eighteenth and early nineteenth
centuries, although even they do not always represent the original state of
affairs. In this respect the Tapestry is a much less straightforward source
than has often been assumed.2

Count Guy of Ponthieu apprehends Harold in a coloured engraving based on a drawing

by Charles Stothard, see n. 2 below. Modern photographs show that the missing piece
of upper border has since been restored.

How much further it took the story of the conquest there is no way of
knowing, but the common belief that it originally ended with William’s
coronation in December 1066 seems plausible enough. Although not
recorded before 1476, most commentators accept that it was made for (or
at least in the circle of) Bishop Odo of Bayeux, the Conqueror’s half-
brother, within about fifteen years of the battle (Odo was imprisoned by
William, for reasons which are not known, in 1082) and that it was made

1
Hill, ‘The Bayeux Tapestry’, p. 384. I am grateful to David Hill and John McSween for much
illuminating discussion on this matter. David Hill’s paper to the Manchester Conference on Harold II
and the Bayeux Tapestry in April 2002 was a revelation.
2
On this, the earliest drawings and the engravings published in 1729-30 and 1733, see the Appendix.
The Society of Antiquaries published coloured engravings of the Tapestry based on Charles Stothard’s
drawings between 1819 and 1823. Coloured engravings based on Victor Sansonetti’s work, and limited
to 100 copies, appeared in 1838. B&W, 25, say that the pen and ink and watercolour drawing now in
the possession of Mount Holyoke College in Massachusetts was made prior to nineteenth-century
restoration, but Brown, The Bayeux Tapestry, p. 155, simply dates it to that period and thinks it
probably based on Stothard or Sansonetti; Hill argued that both it and Sansonetti derived from
Stothard, ‘The Bayeux Tapestry’, pp. 394-6.

65
 THE BATTLE OF HASTINGS

in England and almost certainly in Kent, of which he had become earl. 1

Individuals who were probably three of his tenants in the shire – Wadard,
Vital and Turold – are shown and named within it,2 in addition to the
prominent role played by Odo himself, and this is strong evidence that its
audience included the kind of men who had fought in the battle, almost
certainly (because of the expense involved in acquiring the necessary
equipment) in metal armour3 and on horseback. It is thus hardly
surprising that war is one of its prime concerns, or that Hastings occupies
more than a quarter of its surviving length.4 Another central theme is
Harold’s oath and its breaking (the work begins with his visit to
Normandy), and this too would have been of great interest to men whose
lives were to a large extent defined by the obligation of loyalty to their
lord.5 Some of the Tapestry’s audience was also expected to be familar
with fables of the type associated with Aesop.6
Like most pictorial sources, the Tapestry is much better at showing
what things looked like than at recording events in any detail, and if it
were the only material to have survived on the Conquest historians would
be hard put to make much of it. Nevertheless, it does tell a story,
accompanied by a sparse commentary in Latin, and that story begins with
Harold’s visit to the continent, his seizure by Count Guy of Ponthieu,
release by William and arrival in the ducal palace. From there he joins the
Normans on a campaign against the Bretons, on the way distinguishing
himself by rescuing fellow soldiers from quicksand as they cross the river
Couesnon; the expedition then proceeds to Dol, from which its quarry
Count Conan flees, and then to Dinan, where he is obliged to surrender.
1
On the Tapestry’s provenance, see Wormald, ‘Style and Design’, pp. 29-34; B&W, 8-18; Bernstein,
pp. 37-50 (with good reproductions of manuscript illuminations whose similarity to Tapestry scenes
suggests a Canterbury origin). However, all acknowledge that manufacture in Normandy is a
possibility, and the Norman case is argued by Grape, The Bayeux Tapestry, pp. 44-65; for persuasive
criticisms of his position, see Gameson, ‘The Origin, Art, and Message of the Bayeux Tapestry’, pp.
162-74. Although I refer to the designer as male, this should not be taken to exclude the possibility that
the work was done by a woman.
2
B&W, 8 and n. 22 and Tsurushima, ‘Hic Est Miles’, pp. 81-91. Wilson (BT, p. 176) thinks Turold
may be the dwarf holding the horses rather than one of the messengers sent by Duke William to Guy of
Ponthieu in 1064, and is attracted by the idea that he ‘is the artist of the Tapestry’; see also EHD2, p.
257. The ‘dwarf’, however, may be an attempt at some kind of perspective.
3
It is unclear how much of the metal body armour of 1066 consisted of chain-mail made up of
interlaced iron rings or of tunics of leather or other fabric upon which rings or plates were sewn or
riveted, called in English byrnies. For a valuable treatment of this, and the view that chain-mail was in
the minority, see Renaudeau, ‘Problèmes d’interprétation du costume’, pp. 245-52.
4
The extent to which war dominates the subject matter of the Tapestry is stressed by, among others,
Gameson, ‘The Origin’, p. 207.
5
Dodwell, ‘The Bayeux Tapestry and the French Secular Epic’, pp. 52-9, noted the prominence of
themes of loyalty and disloyalty in the later French secular chansons de geste.
6
These scenes, which appear in the borders and whose concern with deceit is probably linked to
Harold’s oath, have been much discussed; see, for example, Cowdrey, ‘Towards an Interpretation of
the Bayeux Tapestry’, pp. 99-100, and ‘King Harold II and the Bayeux Tapestry’, pp. 4-5; Bernstein,
pp. 128-35; Gameson, ‘The Origin’, pp. 159-60 and n. 11; Lewis, The Rhetoric of Power in the Bayeux
Tapestry, pp. 59-73; Albu, The Normans and their Histories, pp. 91-105.

66
 2 THE SOURCES: PART ONE

After this we are shown William bestowing arms upon Harold as an

indication that he had become the duke’s man, the taking of the oath and
Harold’s return to England and appearance before King Edward. There
follows the king’s burial and death (with a deathbed scene which suggests
that Harold was there granted the throne),1 his coronation and the
appearance of Halley’s Comet. At this point the scene shifts to
Normandy, where William orders the building of ships; arms and
provisions are carried down to them and they cross the Channel to
Pevensey, the landing being followed by the plundering of the
surrounding district for food, which Bishop Odo duly blesses as they sit
at table. A castle is built at Hastings, the approach of Harold is
announced, and the French cavalry are shown leaving for the battle; on
both sides scouts report the arrival of the enemy, and after the duke has
exhorted his men to prepare themselves (the only example of reported
speech in the entire work)2 a long line of horsemen, broken only by four
archers, attacks the English shield-wall from left and right. At this point
the battle (or to be more precise, the bodies of the dead) overflows into
the lower border. The deaths of Harold’s brothers Gyrth and Leofwine
follow, French horses are in great difficulties near a watercourse, and
fighting takes place around a hillock, the legend at this point informing us
that Hic ceciderunt simul Angli et Franci in prelio (‘Here the English and
French fell at the same time in the battle’). It seems reasonable to assume
that French losses were by this stage serious, as Bishop Odo comforts the
boys and Duke William removes his helmet, apparently to show that he is
still alive. From here, and for no obvious reason in terms of the outcome
of the fighting so far depicted, the battle moves to its close; the lower
border is now occupied by French archers, while further attacks by the
horsemen result in the death of Harold and the flight of the remnants of
his army; as he is killed, men in the lower border are already stripping the
dead of their armour.
A summary such as this does scant justice to the power, artistry and
everlasting fascination of many of the Tapestry’s scenes, but as a record
of a battle which lasted many hours and involved thousands and perhaps
tens of thousands of men on each side it is inevitably somewhat lacking.
The limitations of the medium would have made it very difficult for it to
give much indication of strategy, tactics and the reasons for the French
victory, even if this had been the designer’s purpose, which it almost
certainly was not. Thus, it is hardly surprising that there is no sign of the
French retreat and English pursuit which feature so prominently in some
of the written sources.3 What its patrons required was a decorative
1
Above, pp. 31-2.
2
Bernstein, p. 258.
3
But see below, p. 179, n. 1.

67
 THE BATTLE OF HASTINGS

hanging which would justify the Conquest and show them in their hour of
glory. It is fair to assume that they would have expected a degree of
accuracy in terms of what soldiers looked like and the ways in which
individuals fought, and there is little reason to suppose that they were
disappointed in these respects.1 It has been suggested that one of the
Tapestry’s ‘most remarkable mistakes’ is the depiction of the wearing of
trousered mail by both sides when the French horsemen in fact used
skirted mail, and that it is therefore ‘a more dependable source for the
armour and weapons of the English than of the Normans’.2 This may be
so, but it is not the area in which the work is at its most misleading. A
more serious limitation, and one which arose inevitably from the nature
and interests of the aristocratic audience for which it was made, is the
impression that Hastings was won by French horsemen with the
assistance of archers, most of whom are relegated to the lower border,
where they do not impinge too much on the stirring depictions of military
prowess in the register above. Only near the beginning of the conflict is
an exceptionally long line of cavalry broken up by the figures of the only
archers to appear in the main register, and the individual wearing mail is
also one of the only two armoured French infantrymen shown in the
entire battle, the second being engaged in decapitating an unarmed
Englishman shortly before the death of Harold.3 That this cannot possibly
be an accurate reflection of the composition of William’s forces is
confirmed by William of Poitiers, who speaks of the duke’s men
advancing in three lines, the first and third being missile troops and

1
See Mann, ‘Arms and Armour’, and his comment (p. 57) that ‘One has confidence in the accuracy of
the military scenes by comparing the obvious naturalness of other scenes that are familiar to us today,
such as those of hawking, hunting...’. Even so, the Tapestry cannot be trusted implicitly. It is likely to
be inaccurate, for example, in suggesting that French horses crossed the Channel in vessels of such
shallow draught that they could simply step over the gunwale on landing; see Bachrach, ‘On the
Origins of William the Conqueror’s Horse Transports’, 506-9.
2
B&W, 19-20. Peirce, ‘Arms, Armour and Warfare in the Eleventh Century’, 238, suggests that the
Tapestry’s apparent French mail leggings simply depict the way in which skirted mail clung to the legs
and thighs. No surviving coat of mail of the eleventh or twelfth centuries is currently known, but Peirce
estimates the likely weight as 11.4 kg (25 lb); see ibid., 237-40. See also Renaudeau, ‘Problèmes
d’interprétation du costume’, pp. 253-4.
3
This is the figure used by Bernstein (pp. 166-78) to argue that the Tapestry designer incorporated the
idea that the Norman conquest of England was comparable with the Babylonian conquest of the Judah
of the Old Testament king Zedekiah, who like Harold violated an oath of fealty to his overlord and was
eventually blinded (in Harold’s case by an arrow); prior to his blinding, Zedekiah was obliged to
witness the killing of his two sons, and Bernstein suggests that the figure being decapitated in the main
register, and the one in the border below who has already had his head struck off, are an allusion to
this. His thesis is weakened by the probability that the Tapestry did not originally show Harold as
struck by an arrow (see the Appendix, below), but it is not impossible that its audience were even so
expected to be familiar with this event, and it is worth noting the similar parallel drawn between the
Conqueror and the biblical Jephthah (like the duke a bastard) in Fulcoius of Beauvais’ poem Jephthah
of c.1075; on Jephthah, see TNE, pp. 132-3, and van Houts, ‘Latin poetry’, 42. In counting the number
of archers in the main register I have omitted the apparent horse archer in the much restored final
scene.

68
 2 THE SOURCES: PART ONE

squadrons of horsemen, and the second being heavy infantry. 1 The role of
such infantry in the battle may well have been a significant one, but it
was ignored by the Tapestry designer in exactly the same way that he had
earlier given the impression that cavalry had been important in the assault
on Dinan during the campaign in Brittany; but few castles ever
surrendered as a result of horsemen charging their walls (which is what
the onlooker is encouraged to believe), as is tacitly acknowledged by the
depiction of the two dismounted figures who are setting fire to the
defences. Similarly, it may be unwise to assign too great a significance to
scenes in the Tapestry’s account of Hastings in which individual French
horsemen are engaging armoured and isolated English infantrymen, for
these may have been elements precisely tailored to the requirements of
the piece’s patrons; the accuracy of the depictions of cavalry charging the
English shield-wall and what look like light infantry on the hillock (one
of whom is being transfixed by a spear-thrust delivered by a trooper with
his spear couched beneath his arm) may well also be suspect, especially
in the case of the latter, as an eminence covered in some kind of
vegetation is not the most obvious ground for cavalry.
Yet even had their desire to see their own role in the battle celebrated
to the exclusion of almost everybody else not imposed limitations on the
Tapestry designer, the constraints of the medium in which he was
working would probably have pushed him in a similar direction. If French
armoured and unarmoured infantry had been depicted in any numbers, for
example, it would have been difficult to distinguish them from English
troops of the same type without making the battle scenes much more
complex than they actually are. The same argument may also work in
reverse. Only one English archer is shown compared with his many
French counterparts, but it would be dangerous to contend on these
grounds alone that there were fewer English missile troops present, when
the designer may have been simplifying life (as well as pandering to the
social prejudices of his patrons) by omitting them. Still, arguments such
as these emphasise only that his work must be used with great care, not
that it has nothing to offer. Far from it. None of the written sources can
equal it in giving a visual impression of what certain aspects of the
fighting were like, for this was something with which they assumed their
audience to be broadly familiar; none describe the wielding of the great
English battle axe, much less give the sort of impression of its fearsome
destructive power that can easily be derived from the embroidered image,
or show the shield-wall which the armies of the Anglo-Saxon state
employed time out of mind, and this last time. In these senses the
1
GG, pp. 126-7; see below, pp. 170-1. For similar lines of argument, see Morillo, ‘Hastings: An
Unusual Battle’, pp. 222-3; France, ‘The Importance of the Bayeux Tapestry for the History of War’,
pp. 292-5.

69
 THE BATTLE OF HASTINGS

Tapestry brings us closer to what happened that day than any other source
ever can. More prosaically, it is also important, as will become obvious
when discussing the deployment of Harold’s forces,1 in showing fighting
around the watercourse and the hillock, for no surviving source earlier
than William of Malmesbury says anything of the latter, while it will
eventually be necessary to consider whether the watercourse has any
connection with the ditches and deaths therein of which so much has
already been said, and has still to be said.
The Tapestry is a unique survival from the eleventh century, but was
probably far from unique in its own day.2 When it surfaces in the
historical record during the fifteenth century it was being hung around the
nave of Bayeux cathedral to celebrate the annual Feast of Relics, and as it
locates Harold’s oath (on relics) at Bayeux it can be argued that Odo
commissioned it for his cathedral church. This may be so, but nor is it
impossible, as has often been said, that it was originally intended for a
secular dwelling in England, and found its way across the Channel at a
later stage. Indeed, the prominence it gives to three of Odo’s tenants in
England rather points this way, for it is easier to see why a hanging
exhibited in, say Canterbury, might have included them than to
understand why a Bayeux audience should have had its attention drawn to
these particular men rather than anybody else.
Moreover, as a secular hanging it would fit into a known historical
context. Depictions of famous events, in various media, already had a
long history by the late eleventh century: sculptured reliefs from ancient
Mesopotamia can be set beside Trajan’s Column (an account of the
Roman emperor Trajan’s conquest of Dacia in the early second century
AD) and the paintings of biblical and classical scenes, together with
Charlemagne’s conquest of the (continental) Saxons, which graced the
walls of the Carolingian palace at Ingelheim; similarly, decorations in his
palace at Magdeburg celebrated the German Henry the Fowler’s victory
over the Magyars in 934. In England a stone fragment 69.5 cm in height
unearthed in excavations of Old Minster, Winchester in 1965 has been
interpreted as being from a sculptural frieze of great length which
recounted the story of the Scandinavian hero Sigmund and was perhaps
made during the reign of Cnut to celebrate the shared ancestry of the
royal houses of Wessex and Denmark. It may have had upper and lower
running borders framing the main scenes and if so these are not the only
elements which it and the Tapestry had in common.3 Yet it can hardly
1
Below, pp. 121-5.
2
A point made by B&W, 26.
3
See Illus. 75. The Golden Age of Anglo-Saxon Art, ed. Backhouse, No. 140. ‘The strip-like form of
the frieze, the break in scene caused by one figure turning his back upon another, and details of the
armoured warrior, all recall the Bayeux Tapestry’. See also English Romanesque Art 1066-1200, No.
97; Dodwell, Anglo-Saxon Art, pp. 137-8.

70
 2 THE SOURCES: PART ONE

have looked more like parts of the latter than does the tympanum above
the south door of the church of St George, Fordington, Dorset. Here the
mounted saint, who bears a great resemblance to horsemen depicted in
the Tapestry’s battle scenes, rides down not the usual dragon but a
footsoldier who has been speared in the mouth and lies tumbled among
other casualties in much the same way as the corpses which at points
litter the famous embroidery’s lower border; its broken weaponry is
echoed here by a broken spear, the round shields with a large central boss
which it shows on occasion being used by the English are present too. On
the left of the tympanum two men on their knees, like their adversaries
wearing conical helmets with nasal protection, seem to be invoking the
saint’s assistance, while kite-shaped shields and spears are stacked behind
them. It is difficult to believe that these two works of art are very remote
from each other in date and if, as is likely, the St George scene was
originally painted their resemblance must once have been close indeed.1

The tympanum above the south door of the church of St George, Fordington, Dorset.
Furthermore, there is from the Anglo-Saxon period a great deal of
evidence on the importance of textiles, some of them worked in gold and
ornamented with jewels, and of great worth.2 Ironically, the sight of
Harold’s banner of the Fighting Man, woven in the purest gold, would
have served to remind the French troops who faced it of the riches of the

1
The scene has sometimes been associated with St George’s reported appearance at battles during the
First Crusade. I am grateful to my friends Sue and Ian Reedie for bringing the Fordington tympanum to
my attention.
2
See the immensely informative material in Dodwell, Anglo-Saxon Art, pp. 129-69, 174-87. The only
surviving examples of Anglo-Saxon embroidery in gold and silk are those of c.800 now in Maaseik
(Belgium) and the tenth-century ecclesiastical vestments found in the tomb of St Cuthbert in Durham
cathedral; small but potent reminders of what has been lost.

71
 THE BATTLE OF HASTINGS

country they were attempting to conquer, and thus of the fruits of victory.
The English made great use of decorative hangings both in churches and
in their homes, as is revealed by wills which survive from the tenth and
eleventh centuries. At some point between 984 and 1016, for example, a
lady named Wulfwaru bequeathed to the monastery of Bath ‘a set of bed-
clothing with tapestry and curtain’, to her son Wulfmær ‘a hall-tapestry’,
and to her second son Ælfwine ‘a tapestry for a hall and tapestry for a
chamber’.1 The sources say much more about the existence of such works
than of the decorations which they bore, but there is an important
reference in the twelfth-century chronicle of the abbey of Ely (here
probably based on documents of the late tenth century) to a piece which
they received from the widow of Ealdorman Byrhtnoth, who was killed in
battle against the Vikings at Maldon in 991. Just after his death, she gave
them, in addition to a gold torque and various estates, a hanging which
showed the deeds of her husband. If these deeds did not include feats of
arms it would be surprising, and it looks therefore as though the piece
was similar to the Bayeux Tapestry and the decorations of the continental
palaces at Ingelheim and Magdeburg in being intended to celebrate the
achievements of its patron at the same time as it decorated his living
quarters.2
Another reference to what may have been ‘a continuing tradition of
heroic hangings in England’, even if in a slightly different context, is to
be found in the contemporary Life of King Edward when it describes a
ship given to the Confessor by Earl Godwin which had a sail worked in
gold and depicting the sea battles of noble kings.3 It is almost certain that
one reason for the survival of the Bayeux Tapestry is precisely the fact
that it was not worked in gold, and therefore was much less likely to be
destroyed in later times by those covetous of precious metal, and that its
use of wool would have meant that it was regarded by many
contemporaries not as a masterpiece or ‘a triumphal monument
unequalled in the West since the fall of the Roman Empire’, but on the
1
EHD1, No. 116; cited with other examples, Dodwell, Anglo-Saxon Art, p. 132..
2
et cortinam gestis viri sui intextam atque depictam in memoriam probitatis eius, Liber Eliensis, ed.
Blake, p. 136; see Dodwell, Anglo-Saxon Art, pp. 134-6, and the detailed and valuable treatment by
Budny, ‘The Byrhtnoth Tapestry or Embroidery’, who discusses the various possibilities raised by the
words intextam atque depictam, and questions whether the item had necessarily hung in Byrhtnoth’s
home. More complex explanations of the Bayeux Tapestry’s purpose have sometimes been advanced:
for example, Brown, ‘The Bayeux Tapestry: Why Eustace, Odo and William?’, argues that, at the time
of his disgrace in 1082, Odo had it made for the Conqueror to remind him of past services; Lewis, The
Rhetoric of Power, thinks he created it as a portable propaganda piece to justify his importance in
England after 1066 and the introduction of Norman feudalism. Few possibilities can be discounted
altogether, but such explanations seem to me to strain credibility. Lewis (p. 11) is also one of several
recent authors (notably, Brilliant, ‘The Bayeux Tapestry: A Stripped Narrative for their Eyes and Ears’)
who believe that the scenes and Latin inscriptions were used as the basis of an oral performance by an
interlocutor. They do not, however, cite other evidence of the use of European decorative wall-
hangings in this way.
3
Dodwell, Anglo-Saxon Art, pp. 136-7, citing LKE, pp. 20-1.

72
 2 THE SOURCES: PART ONE

contrary as fitting Professor Dodwell’s description of it as ‘quite

ordinary’.1 Certainly it is worth noticing in this context that ‘there are a
number of major defects in the weave that suggest it was not very
carefully made’2 and that the very obvious join between the first and
second strips, most noticeable in the upper border,3 implies a workshop
initially inexperienced in pieces of this type. Had the Old Minster,
Winchester sculptural frieze survived intact rather than as a fragment the
Tapestry’s true standing within its own time might have been more
obvious, and it is surely significant that when the poet Baudri of
Bourgeuil imagined a hanging depicting the Conquest in the chamber of
the Conqueror’s daughter Adela he said that it was of golden, silver and
silken thread embellished with gems and pearls, and that William of
Poitiers stresses the skill of Englishwomen in the weaving of gold
thread.4 It may well be that Bishop Odo would have required something
equally lavish in any commission of this type, and that the surviving
Tapestry was in fact made for one of his followers who settled in Kent,
whether one of the three featured in it or perhaps an associate.5 Moreover,
if it was at the time a rather less remarkable work than it seems today, and
produced in a world where textiles and other media which showed heroic
events were by no means uncommon, this raises the possibility that it was
far from being the only such depiction of the Conquest to come into
existence in the years following 1066. Given the resources of England’s
new French masters, and their probable pride in their exploits, it is not
unreasonable to suggest that there may have been many such items, and
certainly it is worth noting that in 1420 a tapestry ‘without gold’ which
depicted Duke William’s conquest of England existed in Dijon.6 In any
case, it would obviously be dangerous to make too much of apparent

1
Bernstein, p. 36; Dodwell, Anglo-Saxon Art, p. 139.
2
Lemagnen, ‘The Hidden Face of the Bayeux Tapestry’, p. 40.
3
BT, plate 15; Bédat and Girault-Kurtzemann, ‘Étude technique’, p. 97; Hicks, The Bayeux Tapestry,
pp. 47-8, who stresses the better quality of the later junctions.
4
Baudri, AC, ll. 211-12, 229-30; WP, GG, pp. 176-7; Dodwell, Anglo-Saxon Art, p. 139. Tilliette, ‘La
chambre de la comtesse Adèle’, 147-9, doubts whether Adela’s rooms could possibly have been as
lavishly furnished as Baudri imagined. The poet Serlo of Bayeux commented on the beautiful
decoration of Odo’s court, van Houts, ‘Latin poetry and the Anglo-Norman court’, 45; on Baudri, see
further below, pp. 88-92.
5
B&W, 18, claim that nothing ‘suggests that Turold, Wadard and Vital were such important lords that
they could commission so magnificent and vast a work’, but this begs several questions, not least the
fact that as the agency which actually produced the Tapestry is unknown (a workshop staffed by female
slaves, perhaps?) there is no way of balancing its cost against the resources of men who may have
profited greatly from the Conquest. Pastan and White, ‘Problematizing Patronage’, think Turold too
common a name to be identifiable and use ‘the austerity of the chosen medium’ and the known links of
Bishop Odo, Wadard and Vital with St Augustine’s Abbey, Canterbury to argue for the Tapestry’s
manufacture there. Clarke, ‘The Identity of the Designer of the Bayeux Tapestry’, thinks the designer
was Abbot Scolland.
6
Hicks, The Bayeux Tapestry, p. 66, who suggested that this was the Bayeux Tapestry itself. It seems
more likely that it was one made for an inhabitant of that area who took part in the conquest and
eventually returned home.

73
 THE BATTLE OF HASTINGS

similarities between the version of events given by the one which

happens to have survived and those of the written sources which happen
to have survived. Duke William’s use of missile troops and removal of
his helmet at a critical juncture appear in two of the earliest French
sources (the Carmen de Hastingae Proelio and William of Poitiers’ Gesta
Guillelmi), but they say nothing of fighting around the Tapestry’s hillock
and watercourse, although it does share what may well be the same
information with other, later authorities. The Tapestry’s designer may in
fact have drawn on a wide range of sources, many of them probably oral,
before going about his work. In dealing with the evidence which has
actually come down to us it is sometimes too easy to forget the very much
larger body of information which existed in the decades following the
battle, was available to those who created the small corpus which has
survived, and has since disappeared forever. Moreover, this is a point
which is just as important in dealing with the French material as when
considering that from England.

74
 3

THE SOURCES: PART TWO

The sources which probably or certainly originated in France within
twenty years of the battle are much more voluminous than those written
in England, but also present considerably greater problems. The earliest
may well be that known as the Carmen de Hastingae Proelio (Song of the
Battle of Hastings), attributed to Guy, bishop of Amiens, who was a
member of the ruling house of Ponthieu, being the brother of Count Hugh
II and uncle of Guy I, and related to the French royal family and the
counts of Boulogne.1 Around 1125, Orderic Vitalis stated that Guy was
the author of a poem in the style of Virgil and Statius which described the
battle of Senlac, condemning Harold and praising William, and that by
the spring of 1068, when he was the most eminent of the churchmen to
visit England in the entourage of William’s queen Matilda, it had already
been composed.2 Whether Orderic had himself seen the poem is not
certain, and although William of Poitiers, too, may have been familar
with it,3 the Carmen seems to have been little known in general, and long
remained so. Only in 1826 were two twelfth-century manuscripts (the
second a copy of the first) discovered in the Royal Library in Brussels
and identified as Guy’s work, partly because of the content matter and
partly in the belief that the L. W. salutat of the second line can be
expanded and translated as ‘Guy (Wido) salutes Lanfranc’ (Lanfranc,
abbot of Duke William’s monastery of St Stephen at Caen, and from
1070 archbishop of Canterbury).
If it is really Guy’s poem, the Carmen is one of the most detailed
contemporary sources available on the battle of Hastings. After a short
prologue, it opens by comparing the Conqueror to Julius Caesar and
recording how contrary winds hindered his crossing of the Channel. For a
fortnight he waited at St Valéry, at the mouth of the river Somme, before
setting sail and landing in England about 9 o’clock in the morning. Once

1
On what is known of Guy, see Barlow’s comments, Carmen, pp. xlii-liii.
2
HÆ, ii. 184-7, 214-5.
3
Below, pp. 86-7.

75
 THE BATTLE OF HASTINGS

there, he built a fortification to protect his ships and his men ravaged the
surrounding countryside. When Harold heard of this he was returning,
laden with booty, from the defeat of his brother; he made a speech to his
magnates and they agreed to fight the invaders, but first a monk was sent
to the duke’s camp to ask him to leave. During the resultant conversation
William referred to the earlier agreements he had made with Harold,
including the oath, and offered him the lands his father had held if he
agreed to be his vassal; then he sent an envoy to Harold who summarised
the grounds upon which he claimed the English throne and urged him not
to go back on his oaths. This was scornfully rejected, and as his
ambassador returned with the news that 500 English ships had been
dispatched to obstruct his passage home, the duke deployed his troops for
battle, having frustrated by his watchfulness Harold’s plan to attack him
during the night.
In his front rank were archers and crossbowmen, and although he had
intended placing cavalry behind the infantry the onset of the conflict
forbad this. The English poured forth from the forest to their rear and in a
dense formation and on foot, as was their custom, occupied a nearby hill
(mons) and valley (vallis) and land not cultivated because of its
roughness. Harold planted his standard on the crest of the hill and
strengthened his flanks with noble men, while William, who commanded
a better regulated line (moderantius agmen), climbed the steep slope. His
missile troops opened the battle with arrows and crossbow quarrels and
the infantry (galeati – literally, ‘helmeted men’) crashed shields against
shields. Meanwhile,1 a juggler on horseback named Taillefer (Incisor-
ferri) threw his sword into the air in front of the French lines, killed an
Englishman who dashed forward to meet him, decapitated the corpse and
displayed the head to his advancing comrades, who rejoiced that the first
blow was theirs and called upon the Lord.2 Then the archers and
crossbowmen opened fire3 and the French attacked the English left, the
Normans the centre and the Bretons the right.4 However, the English were
in a very dense formation in which the bodies of the dead could not fall,
and which could only be overcome by the military skill of William’s
men. The French5 pretended to flee and were pursued by the English,
whereupon the duke’s left and right wings turned upon the pursuers, as
did also those who had simulated flight. Many of the English – 10,000 is
1
Guy’s account of the battle is not always very coherent, and although he describes the incident which
follows after the beginning of the fighting he apparently thought that it preceded it.
2
This may be a reference to the Norman battle-cry Dex aïe, ‘May God help’, as suggested by M&M, p.
27, n. 2.
3
A repetition of the information given before the Taillefer story.
4
This is the most probable meaning of ll. 413-14; but it is not impossible that Guy intended to indicate
that the French were on the left and the Bretons on the right; see Carmen, p. lxxix, n. 266, and below,
p. 168, n. 1..
5
Barlow, Carmen, p. 27, n. 4, suggests that Guy means by this the Normans in the centre.

76
 3 THE SOURCES: PART TWO

the number given – were killed at this stage of the battle, but the rest
fought on, and used their superiority in numbers to force the Normans
into a real flight. At this stage Duke William took off his helmet and
made a speech to his men stressing that the way to survive was not by
running away but by conquering; they rallied, and their leader then killed
Harold’s brother Gyrth, who had unhorsed him with a spear cast, and
many others, including the ‘son of Helloc’, who had brought down a
second horse acquired from one of the duke’s fellow soldiers by force.
Count Eustace of Boulogne then gave him a third mount, and together
they cleared the field of the English. The French were already seeking the
spoils of war (i.e. stripping the dead) when Harold was seen on the crest
of the ridge repelling all who assailed him, and William and Eustace,
accompanied by Hugh, the noble heir of Ponthieu, and one Gilfard
(known by his father’s surname) attacked and slew him: the first speared
his chest through his shield, the second decapitated him with his sword,
the third spilt his entrails with a spear thrust and the fourth hewed off his
thigh.1 Upon the news of Harold’s death all English resistance collapsed,
and some fled while others asked for quarter, as evening was falling, in a
place where the duke had killed 2,000 and innumerable other thousands.
He spent the night on the battlefield while Hugh of Ponthieu pursued the
fugitives, and the following day buried his dead while leaving the English
strewn on the field as carrion; Harold’s corpse he ordered to be interred
on the seashore, having refused the request of the king’s mother that it be
handed over to her, if necessary in exchange for its weight in pure gold.2
The author of the Carmen does not conclude his account with the end of
the battle, but follows William to Dover, to London, and finally (in the
surviving texts) to the coronation on Christmas Day.3
Yet was he really Bishop Guy of Amiens, and if so how reliable is his
work? Ever since the discovery of the manuscript there have been doubts
whether the poem is truly that which Orderic Vitalis attributes to Guy,
and these found their fullest expression in an article published by R.H.C.
Davis in 1979. Despite the fact that its author addresses King William in
the present tense, thereby implying that he was still alive at the time of its
1
There has been much discussion of this difficult passage, in which ‘it is not easy to sort out the four
individuals among a welter of epithets in an unpunctuated text’; see M&M, pp. 116-20; Davis, ‘The
Carmen de Hastingae Proelio’, pp. 86-8; and Barlow’s comments, Carmen, pp. xxxv, lxxxii-v. I
follow Barlow, who thinks that the ‘better-supported of the interpretations’ makes the four: Duke
William, Count Eustace, Hugh of Ponthieu (younger brother of Count Guy I and nephew of Guy of
Amiens) and perhaps the French baron Robert son of Giffard, who witnesses charters in England after
1066; on the third and fourth, see the important work of van Houts, ‘Latin poetry and the Anglo-
Norman court’, 54-5. An alternative quartet (favoured by Davis) omits the duke and comprises Count
Eustace, the (in this interpretation, unnamed) noble heir of Ponthieu and the Norman barons Hugh II of
Montfort-sur-Risle and Walter I Giffard, who both fought at Hastings according to WP, GG, pp. 134-5.
2
On this element of the Carmen, which seems to reflect knowledge of the Iliad, see below, p. 88.
3
Like the Bayeux Tapestry, they break off incomplete, but the coronation seems a likely ending,
Carmen, p. xc.

77
 THE BATTLE OF HASTINGS

composition, Davis argued that it contains elements which are not

acceptable in a text supposedly written shortly after 1066, that it is
actually ‘a literary exercise’ composed ‘in one of the schools of northern
France or southern Flanders’ probably between c.1125 and 1140, and that
‘as a source for the history of the Norman Conquest it is simply
ridiculous’.1 Such views have not unnaturally tended to inhibit historians’
use of the Carmen, but cannot be said to have carried the day. Ten years
after they appeared Dr van Houts surveyed the considerable body of
surviving Latin poetry addressed to William I and his children and
commented that:
If Professor Davis is right in his conclusion that the Carmen is a school
product written fifty to seventy years after 1066 it would be the only
surviving poem of the period to pretend to an authenticity which it did
not possess. On the basis of my survey I find this extremely difficult, if
not impossible, to accept.2
Beyond this, Davis cannot be said to have provided any really convincing
reason for assigning the poem to the period he suggested, and some of his
argumentation seems flawed. He referred, for example, to the description
of the death of Harold as ‘the most improbable scene in the whole poem’
and judged it, like the story of Taillefer, to be ‘legendary’. 3 The difficulty
here is that not all material which might look legendary can be assigned
to a late date on that basis alone. It is possible, as can be widely
illustrated (one need only think of some of the wild rumours which swept
Britain during the First World War) and as some commentators on the
Carmen have stressed,4 for inaccurate reports to circulate very soon after
the event to which they relate, and the fact that the Carmen contains
statements which can only be paralleled in (surviving) sources of the
twelfth century does little to indicate that it must itself be of that date.
The case for Guy’s authorship of the poem has been restated in the recent
edition by Professor Barlow,5 and can be accepted in the absence of
strong arguments to the contrary, even though it is not watertight, and the
reasons why the piece was written are far from clear.6 However, if it is
true that Guy accompanied William’s queen to England in 1068 his

1
Davis, ‘The Carmen de Hastingae Proelio’, espec. pp. 98-9.
2
Van Houts, ‘Latin poetry’, 53. By ‘authenticity’ van Houts means the poem’s appearance of
contemporaneity in addressing William as still alive; if the W of L. W. salutat in the second line does
not refer to Guy (Wido) the text gives no reason to suppose that it must be the work which Orderic
attributed to Guy of Amiens.
3
Davis, ‘The Carmen de Hastingae Proelio’, pp. 86-8; Brown, ‘The Battle of Hastings’, 18, thought
that ‘the whole improbable incident recorded by the Carmen goes far to condemn that source itself’.
4
Engels, ‘Once more: the Carmen de Hastingae Proelio’, 9-10; and Chibnall’s contribution to the
discussion which followed this paper at Battle in 1979, 20; Orlandi, ‘Some Afterthoughts on the
Carmen de Hastingae Proelio’, 117, n. 2; Barlow, Carmen, p. xxxv.
5
Carmen, pp. xxiv-xl; supported by David Bates’ review, EHR, cxvi (2001), 173-4.
6
Carmen, pp. xli-ii; van Houts, ‘Latin poetry’, 56, suggests that it was addressed to Lanfranc so that he
could use his influence with the papacy on Guy’s behalf.

78
 3 THE SOURCES: PART TWO

relations with the Conqueror’s family were clearly close, and there is
little difficulty in believing that reasons there were, even if they cannot
now be identified.
As to his reliability, this cannot of course be taken for granted. Dr van
Houts, who thinks that the Carmen was composed in 1067, has suggested
that he obtained some of his information from his secular lord Ralph,
count of Amiens, whom he met at the French royal court in the middle of
1067, and who had himself been present at the Conqueror’s Easter court
at Fécamp in April, where English hostages and the spoils of war had
been on display, doubtless accompanied by a great deal of talk. She has
also noted that Guy knew three of the four men who are said to have
killed Harold – Eustace of Boulogne, Robert son of Gilfard and Hugh of
Ponthieu (his own nephew) and that he met Robert in August 1067 when
they both witnessed a French royal charter.1 It is not surprising, of course,
that he should assign a major role in the battle to his nephew Hugh, and
given his likely sources of information that the only Norman he mentions
by name is Duke William himself.2 As a poet producing a piece of
entertainment he may well have felt free to write what we would regard
as fiction,3 and certainly his description of the crown which William had
made for himself and of the coronation ceremony which followed have
been regarded as ‘obviously imaginary’ and ‘simply made... up’.4 On the
other hand, the insertion of fictional elements cannot be taken as
indicating that he knew nothing about the battle at all,5 and there is in any
any case adequate evidence to the contrary. Guy’s description of the hill
and valley occupied by the English army and of the roughness of the
ground, as of the forest which lay to the rear, are near enough to the truth
to suggest that they are not of his own invention, and so very probably is
his description of the breaking of their line by a French retreat, unless one
were to adopt the scarcely tenable position that all the later sources which
mention this feature of the battle have taken it at one or more removes
from his poem. Even all the elements of his account of the death of
Harold cannot be rejected with complete confidence.6 The Carmen may
not have been written for modern historians,7 but it is a detailed and
almost contemporary source which probably owes something to eye-
witness accounts, and despite all its attendant problems it cannot be
ignored.
1
Van Houts, ‘Latin poetry’, 54-5. Presumably Guy also knew Duke William.
2
This assumes that the third and fourth alleged slayers of Harold have been correctly identified in p.
77, n. 1, above..
3
Engels, ‘Once more’, 9, comments that ‘the margin for fiction is... much larger in poetry than in
prose’.
4
Barlow, in Carmen, pp. xxxvi-vii.
5
Another point made by Engels, ‘Once more’, 10.
6
Below, pp. 180, 184-5.
7
Van Houts, ‘Latin poetry’, 56.

79
 THE BATTLE OF HASTINGS

Whether or not the date of 1067 for the composition of the Carmen is
accepted, it cannot be assigned to one later than Guy’s death in 1074 or
1075. By then another surviving French account of the battle, that of the
Norman monk William of Jumièges, was also in existence. William is
thought to have written most of his Gesta Normannorum Ducum (Deeds
of the Norman Dukes), which in its early books draws heavily on an
earlier work by Dudo of St Quentin, in the 1050s, and to have
subsequently added a section on the Conquest; as this ends with the
Conqueror’s operations late in 1069 it may have been completed early the
following year.1 Despite the fact that King William, to whom the Gesta is
dedicated, attended the consecration of the new church of St Mary at
Jumièges in July 1067,2 its concluding section is rather brief, and it is
difficult to avoid the conclusion that William of Jumièges must have
heard far more about the Hastings campaign (not improbably from eye-
witnesses with events still fresh in their minds) than he chose to write
down.
His work comments on the undertakings given to Duke William on the
succession to the English throne by both King Edward and Earl Harold,3
and lays stress on the latter’s perfidy in breaking his oath. It then
mentions Halley’s Comet, the duke’s assembly of a force of 3,000 vessels
at St Valéry and the crossing to Pevensey, and the building of
fortifications both there and at Hastings. Harold is said to have hastened
to take him by surprise, gathered very large forces and ridden through the
night, appearing on the battlefield at about dawn. However, William had
adopted measures to deal with a night attack and his men had been
standing to arms. Early in the morning he formed them into three
divisions and advanced against the terrible enemy. Fighting began around
the third hour and went on until nightfall. Harold fell pierced with
wounds during the first onset and when the English heard of this they
turned, as night was falling, and began to flee. The duke returned to the
battlefield from the pursuit at about midnight, despoiled the enemy
corpses (they say that many thousands of the English were killed) and
buried his own dead the following morning, subsequently setting out for
London, where he was eventually crowned.4
This seems a sober enough account of the battle, and the only real
problem with it is the surprising statement that Harold was killed early in
the day, for this not only contradicts a number of other sources but also
appears internally inconsistent, as the English are said to have fled when

1
I follow here van Houts’ dating, GND, i. xxxii-v; for WJ’s sources, see xxxix –l, and for a recent
discussion of GND, Albu, The Normans in their Histories, pp. 51-81.
2
GND, ii. 172-3.
3
Above, pp. 23-4, 26-30.
4
GND, ii. 160-71. WJ’s ‘they say’ (referuntur) looks like an allusion to his sources.

80
 3 THE SOURCES: PART TWO

they heard of his death, yet as they did so night was falling. Dr van Houts,
the most recent editor of the Gesta, believes that William of Jumièges
made a statement that is ‘incontrovertibly false’ because he ‘was adding
these chapters... before accurate reports were available’,1 but given the
Conqueror’s visit to Jumièges in 1067 this is not entirely convincing, and
there is quite a lot to be said for Professor Gillingham’s suggestion that a
scribal error early in the transmission of the text produced a ‘first attack’
(in primo militum congressu) from an original ‘final attack’ (in postremo
militum congressu), especially as the oldest surviving manuscript dates
only from c.1100.2 This would both bring this account into line with other
sources on the time of Harold’s death, and also make better sense in
itself, as his fall late in the day would then explain why his men fled as
night was approaching. Even so, it is no more than a possibility, for it
should be noted that Orderic Vitalis3, who followed the existing text of
William of Jumièges on this matter, may well have had a manuscript
much closer to the author’s autograph than any now extant.
The number of surviving manuscripts suggests that the Gesta
Normannorum Ducum was a very popular work, and not long after it was
written it was used by a scholar who provides the most detailed
contemporary French account of the events of 1066, William of Poitiers.
His Gesta Guillelmi (The Deeds of William) was known to both Orderic
Vitalis and William of Malmesbury, but like the Carmen seems not to
have become widely disseminated, and the text has come down only
through a seventeenth-century printed edition taken from a manuscript
incomplete at both beginning and end, and which has since disappeared.
Orderic, who had a fuller copy and may have met its author, 4 says that it
was dedicated to King William to gain his favour, and that William of
Poitiers had been a soldier before entering the church; a Norman by birth,
he was called ‘of Poitiers’ because he had studied there, becoming
famous for his learning and an archdeacon in the diocese of Lisieux when
he returned home; he was also for a long time one of the king’s chaplains,
and attempted to describe the events which he had both seen and
participated in, although he was prevented by adverse circumstances from
continuing his work up to the king’s death (in 1087).5 Perhaps born
c.1020, he may have studied in Poitiers around 1050 and obtained his
archdeaconry in the 1070s; he seems also to have been one of the canons
of the church at St Martin at Dover, having received his share of the

1
GND, i. liii-iv.
2
Gillingham, ‘William the Bastard at War’, 146, n. 36; GND, i. c-ci, on the manuscript; van Houts
rejects this idea, GND, ii. 169, n. 3.
3
For Orderic’s revision of GND, see below, pp. 92-3.
4
As Chibnall suggests, GG, p. xvi.
5
HÆ, ii. 78-9, 258-61, 184-5.

81
 THE BATTLE OF HASTINGS

prebends (formerly held in common) from Bishop Odo of Bayeux. 1 It is

possible that he was too closely connected to Odo (of whom he speaks
well), and that the latter’s imprisonment by the Conqueror in 1082 had
something to do with the ‘adverse circumstances’ which prevented the
completion of his work.2
The Gesta Guillelmi as it exists today begins with the demise of King
Cnut in 1035 and ends in 1067, but Orderic says that his version
concluded with the death of Earl Edwin of Mercia (in 1071), and
elements of the lost ending can be deduced from his use of it. 3 William of
Poitiers was at work upon it in the 1070s, for the passage which refers to
the Conqueror then being aged around forty-five4 must date from about
1073, and he uses the present tense of Bishop Hugh of Lisieux, who died
in July 1077.5 However, he also refers to the dedication of the abbey of St
St Stephen at Caen which occurred in September 1077, and this raises the
possibility of later revision,6 as does Orderic’s belief that he intended
taking the work down to King William’s death. In saying that the Gesta
Guillelmi was, like the Gesta Normannorum Ducum, dedicated to
William to gain his favour Orderic is presumably referring to statements
in its (lost) opening sections, and there is no difficulty in believing him to
be correct; but it does not follow that William (who is not known to have
been a patron of learning) in any sense commissioned it. Whether he did
or not, of course, its author had every motive to flatter his subject, and
other eleventh-century works of a similar type sound timely warnings of
just how misleading such flattery might be. The Encomium written c.1041
for William’s great-aunt Emma of Normandy, for example, makes much
of her marriage to Cnut but fails to mention that she had previously been
the wife of his enemy King Æthelred II; similarly, Helgaud of Fleury’s
biography of the French king Robert the Pious (d. 1031) is less than open
about its subject’s marital affairs, and suppresses all mention of his first
wife.7 In these instances there are other sources of knowledge which
reveal the extent of the distortion, but the student of William of Poitiers is
not in such a fortunate position, for over many of his statements there is
either inadequate control or no control at all. Even so, he says things that
are difficult to believe. It was suggested in Chapter One, for example, that
the basic outlines of the story told by the Gesta Guillelmi and William of
Jumièges about the undertakings given to Duke William on the English
succession are probably correct, but it is not at all easy to credit the claim

1
See Chibnall’s comments, GG, pp. xv-vii.
2
Davis, ‘William of Poitiers’, pp. 120-3.
3
Ibid., pp. 129-30, guessed the contents of the lost parts of GG.
4
GG, pp. 16-17.
5
GG, pp. 92-5.
6
GG, pp. 84-5, and see Chibnall’s n. 3.
7
Campbell, ‘England, France, Flanders and Germany in the Reign of Ethelred II’, p. 192.

82
 3 THE SOURCES: PART TWO

that fear of the duke’s power played a significant role in establishing

Edward the Confessor on the English throne,1 given that at the time he
was having difficulty in asserting his authority over Normandy. Similarly,
the statement in connection with the duke’s campaign against Maine in
1063 that the county had formerly been under Norman control has been
described by his most recent editor as ‘untrue’.2
There are also problems arising from the extent of William of Poitiers’
classical education and the use to which he put it. He shared with two
other scholars who wrote on Hastings, Guy of Amiens and Baudri of
Bourgeuil (see below), the tendency of learned men of this period to
place within a classical framework their panegyrics of great figures, who
could then be lauded by the comparison of their deeds with those of the
heroes of antiquity. As far as the Conqueror was concerned the minds of
such writers turned naturally to Achilles and Aeneas (victors over Hector
and Turnus respectively), and even more naturally to Julius Caesar, who
had not only invaded Britain but also subsequently become involved in a
conflict with Pompey which could readily be likened to that between
William and Harold.3 It is thus unsurprising that the Gesta Guillelmi
should contain not only allusions to Virgil’s Aeneid, Statius’ Thebaid (on
the struggle for the throne of ancient Thebes), the works of the historian
Sallust and perhaps the Ilias latina, but also references to Suetonius’
Twelve Caesars, Lucan’s Pharsalia (on Caesar’s defeat of Pompey at the
battle of Pharsalus in 48 BC) and above all to Caesar’s own The Gallic
War. Nor was William of Poitiers’ use of these works at all lacking in
finesse. Professor Davis stressed that he:
was a classical scholar and stylist of distinction, intent on producing a
work of great literature... he was capable of imitating the style of quite a
number of ancient authors, choosing for each of his themes the model
which seemed most suitable, and often interrupting his narrative for a
passage of pure rhetoric... He thought himself into the relevant style and
used it boldly; in his comparison between the invasions of Julius Caesar
and William the Conqueror, which naturally favours the latter, he moves
about Caesar’s Gallic Wars with the ease of a master, using its facts
4
solely as they are relevant to his purpose.

1
GG, pp. 18-19. Davis, ‘William of Poitiers’, p. 110, thought that WP was using an earlier source here.
2
GG, pp. 60-1, and n. 3.
3
In addition to the major works under discussion here, other more minor ones compare William with
Caesar. These include an anonymous distich quoted by HH (pp. 410-11) and the anonymous Plus tibi
fama, attributed by van Houts to Godfrey of Rheims and dated by her to 1070; a distich by Bishop
Hugh-Renard of Langres of c.1075 contains what she describes as ‘an echo of the Caesar theme’; see
van Houts, ‘Latin poetry’, 41-2.
4
Davis, ‘William of Poitiers’, pp. 101-2. GG, pp. 189-90, lists WP’s citations from classical and
medieval sources. The Ilias latina is a Latin rendering of the Iliad in 1,070 lines believed to be from the
time of the emperor Nero; it is not thought that the full text of the Iliad was available in western Europe
in WP’s time; see further below, p. 88.

83
 THE BATTLE OF HASTINGS

These features of his work bring with them some intractable difficulties.
When he says, for example, that the Conqueror at one point concealed the
loss of men who had been shipwrecked by burying them in secret, is he
recounting something which actually happened, or being overcome by a
desire to compare Duke William with Xerxes? Similarly, the banquet
which he allegedly held at sea while waiting for the rest of his fleet to
catch up with him is sufficiently reminiscent of feasts described in
Virgil’s Aeneid for one to doubt whether it ever took place,1 while in the
account of events just before the battle of Hastings we are told that
French scouts reported the approach of the English and that the duke
hastily ordered those who could be found in his camp to arm themselves,
as many of his men were out foraging. The impression is given that the
latter played no part in the battle, but they may have been introduced into
the narrative at this point because William of Poitiers knew that foraging
troops feature prominently in Caesar’s account of his expeditions to
Britain in The Gallic War: in 55 BC the Seventh Legion was attacked
when foraging and had to be rescued by Caesar with troops assembled
from his camp, and the following year three of his legions and all his
cavalry were attacked while out foraging.2
The account of the battle itself3 begins by telling how the duke took
communion, hung round his neck the relics upon which Harold had sworn
his oath, and did not regard putting his coat of mail on back to front as a
bad omen. He then made a speech which William of Poitiers based on
one contained in the works of Sallust, and led his men into battle behind
the banner sent by the pope; in front were infantry equipped with bows
and crossbows, behind them armoured infantry, and at the rear squadrons
of horsemen, including the duke himself. Harold’s army, of immense size
and strengthened by help sent from Denmark, assembled on a hill (mons)
near the wood through which they had just passed; dismounting from
their horses, they adopted a dense formation on foot. The duke, who was
not dismayed by the roughness of the ground, climbed the steep slope and
the battle opened with an exchange of missiles, the English throwing
spears and projectiles of various kinds, murderous axes and stones tied to
sticks. After this the French horsemen weighed in with their swords, but
the fighting was evenly balanced and they gained no advantage, for their
enemies were assisted by the advantage of the higher ground, their great
and closely-packed numbers, and their weapons, which easily penetrated
shields and other defences (presumably a reference to the axes shown on
1
GG, pp. 108-9, 112-3 (and see 112, n.2, citing Aeneid, i. 695-747; vii. 107-34; viii. 175-83). Both
examples are given by Davis, ‘William of Poitiers’, p. 103; see also Professor Foreville’s discussion in
her edition of GG, pp. xxxviii-xliii.
2
GG, pp. 124-5; Caesar, The Gallic War, iv. 32, v. 17, pp. 220-3, 256-7. See further on the
unconvincing nature of WP’s account of the opening of the battle, below, pp. 158-9.
3
For WP’s account of the prelude to the conflict, see below, pp. 157-8.

84
 3 THE SOURCES: PART TWO

the Bayeux Tapestry). At this point William’s infantry, Breton horse and
such auxiliaries as were on his left wing began to give way, along with
almost all the rest of his line; there was a belief that their leader had
himself fallen, and he countered this by removing his helmet, making a
speech stressing that nothing was to be gained by flight and everything by
victory, and charging once more into the fray at the head of his men. At
this they rallied, surrounded and destroyed some thousands of the English
who had followed them, and continued to attack the rest, who were little
diminished in strength despite their losses.
Even now, they were so tightly packed that the dead could scarcely
fall, but inroads were made by the weapons of the strongest soldiers
(fortissimorum militum ferro),1 and the attackers included men of Maine,
Frenchmen,2 Bretons, Aquitanians and above all Normans. William of
Poitiers mentions by name the exploits of Robert, son of Roger of
Beaumont, fighting with his men on the right wing in his first battle, but
says that he does not intend to describe all the deeds of individuals, as
even an eye-witness (which he, by implication, was not) could hardly
have followed everything. He then moves to what appears the turning
point of the battle: the Normans twice feigned flight, again surrounded
and annihilated thousands of their English pursuers, and then renewed
their assault on the rest, shooting arrows and striking the enemy, who
were still so densely massed that the wounded could not withdraw. A list
of ten notable leaders on the French side is followed by a passage
extolling the military virtues of Duke William: he led from the front, and
swiftly avenged the loss of the three horses killed under him; astonished
at seeing him fight on foot, even soldiers who had been wounded took
heart from the sight.
As the day declined the English realised that they could no longer stand
against the Normans, for their losses had been heavy and included the
king and his brothers. Accordingly, they turned to flight and were
relentlessly pursued, until given renewed confidence by a broken rampart
and maze of ditches (praerupti ualli et frequentium fossarum). Here they
made a stand, and when William arrived on the scene he thought them
English reinforcements; despite being advised to withdraw by Eustace of
Boulogne, who was about to retire with fifty men, he charged the enemy
and destroyed them, although it was here that some of the nobler
Normans fell, their bravery brought to naught by the difficulty of the
ground. Returning to the main battlefield, he surveyed the English dead.
The king’s two brothers were found very near him, but Harold himself,
lacking all ornament (which presumably means that his corpse had been
despoiled) could not be recognised by his face, but only by certain signs.
1
On the translation of this phrase, see below, p. 179, n. 2.
2
By Francigenae WP meant men from the area around Paris controlled directly by the French king.

85
 THE BATTLE OF HASTINGS

He was carried into the French camp and eventually interred by William
Malet on the seashore, the duke having refused his mother’s offer of the
body’s weight in gold in exchange for it. William then, after burying his
own dead and allowing those of the English who wished to do so to
collect remains for interment, set out for Romney, where he punished
those responsible for killing Frenchmen who had landed there by mistake,
and then Dover.1 The submission of Canterbury and eventually London
followed, and his coronation. After its account of the new king’s
triumphant return to Normandy in 1067 the Gesta Guillelmi devotes
many lines to a detailed comparison of his successes against the English
and Caesar’s against the Britons, taking care to lay much stress on the
superiority of the former.2
An immediate problem with this account is whether it is directly
related to that of the Carmen, with which it has significant features in
common: both describe an exchange of monkish envoys before the battle
and the English, so numerous that their advance had turned rivers dry,3 as
densely-massed on a mons near a wood when it began; in both the French
missile troops led the way and opened the fighting, and there was a flight
which the duke halted by taking off his helmet, making a speech in which
he stressed that there was nothing to be gained by running away, and then
turned to advantage by leading a counter-attack which resulted in heavy
English losses; both say that William had horses killed beneath him and
emphasise his personal prowess, and that as evening came on and they
learnt of Harold’s death the English became dispirited; in both accounts
Harold’s body was buried on the seashore after William had refused an
offer of its weight in gold from his mother.4 The direction of the
borrowing obviously depends on whether the Carmen is really the poem
composed by Guy of Amiens: if it is, William of Poitiers may have made
use of it; if it is not, then the anonymous creator of Professor Davis’
twelfth-century literary creation perhaps used William of Poitiers. But
that direct and extensive borrowing there has been, few have doubted.
Engels has spoken of ‘a sort of dialogue between the Carmen and the
Gesta Guillelmi, perceptible in what seem to be reactions of William of
Poitiers to passages in the Carmen’, and joined earlier commentators in
stressing that the two have certain vocabulary in common: both say, for
example, that the duke held a hasta (‘spear’) when he rallied his troops
(the first use of the word in the Gesta Guillelmi, which has previously
employed lancea), and that Harold was buried where he could be the

1
GG, pp. 126-45. Harold’s mother, not named by WP, was Gytha, widow of Earl Godwin.
2
GG, pp. 169-75.
3
GG, p. 126, n. 5, notes the similarity with passages in Juvenal’s Satires and Justin’s Epitome.
4
Most of the similarities between WP and the Carmen are tabulated by M&M, pp. 98-108.

86
 3 THE SOURCES: PART TWO

‘guardian of the shore and sea’.1 Similarly, Morton and Muntz believed
that ‘William of Poitiers had studied the battle in the Carmen minutely’,
and that he revised it with the intention of exalting the Conqueror’s role
even further, and of diminishing that of his French allies, and especially
Count Eustace of Boulogne, compared with that of the Normans.2
Professor Barlow, the latest editor of the Carmen, also thinks that it was
known to William of Poitiers, just as Davis was prepared to believe the
reverse to be true, as the Gesta Guillelmi ‘has some passages which are
undoubtedly reminiscent of the Carmen’.3 As a royal chaplain, William
of Poitiers would presumably have have had little difficulty in acquiring
verse written by a bishop who accompanied Queen Matilda to England in
1068, and as a Latin scholar he would have been interested in doing so.
His familiarity with the Carmen is therefore easy to credit. Yet there is a
counter-case.4 At least some of the similarities between the two sources
might be explicable if it were thought that both men had spoken to
witnesses (or heard songs or seen written sources) telling similar sorts of
stories, and the differences between them might then be because the
stories varied rather than because one had consciously modified the
words of the other; that William took off his helmet when rallying his
troops, for example, also appears both in the Bayeux Tapestry and
elsewhere, and may have been widely reported. Equally, it has been noted
that verbal similarities between sources ‘are often due to common
knowledge of classical authors... do not imply direct imitation, and do not
help to solve the problems of whether either author knew the work of the
other’. Moreover, even if the strong possibility that William of Poitiers
did know the Carmen is accepted, it does not necessarily follow that his
own account of the battle depended heavily upon it, or had to become a
refutation or modification of it, for as a contemporary who may well have
known (or at least met) many of the principal men involved, and an ex-
soldier presumably with a lively interest in it, he could have had a great
many sources, and the poem is not known to have been of such
importance that he would have felt obliged to question or correct its
statements.5
Simply taken as it stands, the Gesta Guillelmi provides the most
detailed description of the battle to have survived, but of course this is not
to say that all its features are credible. It has already been noted that
1
Engels, ‘Once more’, 6-9; similarly, Foreville’s edition of GG, pp. xxxv-viii.
2
M&M, pp. 95-8.
3
Carmen, pp. xxix-xxx, xl, lvii; Davis, ‘The Carmen de Hastingae Proelio’, p. 83.
4
For what follows, including the quotation and the view that ‘WP’s sources were almost entirely oral’,
see Chibnall’s comments, GG, pp. xxviii-xxx.
5
Barlow, Carmen, pp. lvi-viii, seems to me to overstate the probable significance of the Carmen
among WP’s sources. Given the length of his account of the battle one can hardly agree that ‘the details
of the campaigns were not of much importance’, and given the extent of the oral material probably
available to him that the Carmen was used ‘faute de mieux’.

87
 THE BATTLE OF HASTINGS

legends can develop very quickly around famous events,1 and the story of
how the duke put on his mail-coat reversed may be something of the sort,
while it has been suggested that ‘the reality’ of the gold offered for
Harold’s remains was William of Poitiers’ ‘memory of Achilles and the
body of Hector’;2 this may well be so, for in the Iliad Achilles
specifically tells the dying Hector that he would refuse a ransom of ten or
twenty times its value or still more for the corpse, and that even if
Hector’s father Priam offered its weight in gold his mother would never
lay him on a bier where he might be lamented.3 There is, however, a
possibility that eleventh-century figures sometimes acted in ways
intended to imitate elements of stories known to them. 4 Had Harold’s
mother, too, heard of Achilles and Hector, and might she have felt that
acquiring her son’s remains in such a way was in itself a fitting
memorial? Whether this is so or not, the Gesta Guillelmi contains details
which are credible enough. The description of the battlefield (as with the
Carmen) is perfectly compatible with the present terrain, that of English
fighting methods and weaponry, as far as it goes, is borne out by other
evidence (including that of the Bayeux Tapestry), and the significance of
the French retreat is a recurrent feature of the sources, even if William of
Poitiers goes further than anyone else in describing one real withdrawal
followed by two that were feigned.
In offering their works to the Conqueror William of Jumièges and
William of Poitiers were adopting a practice not unusual among those
who composed the Latin poetry and prose of the period. Some years later
their contemporary Baudri, abbot of Bourgeuil near Tours, followed suit
in addressing the poem known as the Adelae Comitissae (To Countess
Adela) to King William’s daughter Adela, wife of Count Stephen of
Blois. Baudri was born near Orleans in 1046, studied probably in Angers
and then entered the monastery of St Peter in Bourgeuil, eventually
becoming abbot there in about 1089. After a failed attempt to obtain the
bishopric of Orleans he was eventually appointed to the archbishopric of
Dol in Brittany in 1107, but discovered that Breton manners were little to
his taste and between that date and his death in 1130 made a number of
journeys to Britain and Normandy. Over the course of his long life he
also produced a considerable body of written work, including saints’
1
Above, p. 78.
2
Davis, ‘William of Poitiers’, pp. 103, 113; GG, pp. 124-5, 140-1. WP, (GG, pp. 134-5), as my pupil
Jonathan Browning has reminded me, had already likened Harold to Hector and William to Achilles;
compare also the possibility that the Carmen (see Barlow’s comments, p. xxxii) intended to equate
William with Achilles.
3
Iliad, xxii. ll. 343-53. I am grateful to my colleague John Davie for this reference, which suggests that
Guy of Amiens and WP (if he did not simply take it from the Carmen) had a knowledge of the Iliad
which somehow went well beyond mere familiarity with the Ilias latina (above, p. 83,). Eventually, of
course (Iliad, xxiv), Achilles did allow Priam to ransom Hector’s body, but not for its weight in gold.
4
A point made by Davis, ‘William of Poitiers’, p. 114.

88
 3 THE SOURCES: PART TWO

lives, a history of the First Crusade and his Itinerarium, a description of

his foreign travels. Of his more than 250 poems, which with one
exception survive in their entirety in a single twelfth-century manuscript,
and which were mostly written during his time in Bourgeuil, two were
addressed to Adela. The first, the Adelae Comitissae, which refers to her
husband as absent from home but still alive, can be dated between Count
Stephen’s departure on the First Crusade in 1096 and death in 1102, and
perhaps specifically to 1101 or 1102.1 It is a lengthy treatment of the
purported furnishings and ornamentation of her great chamber, including
a ceiling which depicted the heavens complete with the signs of the
zodiac, stars and planets, and a floor which bore a map of the world
divided into Asia, Europe and Africa. Around the bed were statues of
Philosophy and her seven disciples, the Quadrivium of Music,
Arithmetic, Astronomy and Geometry being at its head, and the Trivium
of Rhetoric, Dialectic and Grammar at its foot; a third group of figures
represented Medicine, accompanied by Hippocrates and Galen. More
significant in the present context are the wall hangings, worked in silks
and gold and silver thread, embellished with pearls and jewels, and
accompanied by titles (in a manner presumably similar to that of the
Bayeux Tapestry). Their subject matter began with the Creation, the
Garden of Eden and the Flood, followed by further scenes from the Old
Testament and stories from Greek mythology (including the siege of
Troy) and Roman history. Finally, a marvellous piece, which hung
around Adela’s bed, depicted the deeds of her father, with particular
emphasis on the conquest of England. At the end of the poem Baudri
asked for a reward from ‘the goddess’ which he evidently did not receive,
as his request was eventually repeated in the second and (at 32 lines)
much shorter composition he addressed to her.
The early scenes of the supposed Conquest sequence dealt with
William’s birth and the struggles of his minority, but moved quickly to
Halley’s Comet and a speech made by the duke to his magnates in which
he set out his claim to the English throne (his consanguinity with Edward
the Confessor, the latter’s promise, his written confirmation of it as he lay
dying and the broken oath of Harold - called simply ‘a certain perjurer’)
and told them to prepare ships to be ready in the fifth month. Eventually,
3,000 vessels assembled and crossed the Channel, and the battle was
apparently fought immediately; not for Baudri the lengthy preliminaries
of some other sources, although he dwells on the building of the fleet and
its embarkation. He says that the very large enemy army abandoned its
horses and massed in a single dense formation whose spears looked like a
forest, and which would have been impregnable had it held together. The

1
On Baudri’s life and the date of the poem, see Ratkowitsch, Descriptio Picturae, pp. 19-20.

89
 THE BATTLE OF HASTINGS

duke sent his archers and crossbowmen against it and they brought about
the first casualties, as death seemed to the English to be falling from the
sky. As French missiles inflicted further losses, in ranks so tight-packed
that the dead were unable to fall, (some of) the enraged English broke
formation to launch a counter-attack, while the Normans simulated flight
and then intercepted and annihilated their pursuers. Eventually, the
French rushed upon the enemy in disordered ranks and were put to flight,
the rout being fuelled by the rumour that the duke had been killed. At this
point William removed his helmet and made a speech which stressed the
futility of withdrawal and the fact that victory was at hand. Then he
spurred his horse and charged into battle, bettering the feats of both
Hector and Achilles. Even so, much hard fighting followed, and the
ground ran with the blood of the slain, until Harold was hit by an arrow1
and his forces lost heart and began to retire, being harried mercilessly by
the Norman cavalry until nightfall interrupted the slaughter. The
following morning the duke made a speech urging his men to take no
heed of booty until the war was won, and it was subsequently brought to
a successful conclusion.
Baudri of Bourgeuil’s Adelae Comitissae, which may have been
intended for a wider audience than simply its subject,2 belongs within a
well-known classical, late antique and medieval tradition of depictions of
both real and imaginary buildings and works of art, including Homer’s
account of the shield of Achilles and Virgil’s of those of Aeneas and
Turnus, the lines in Ovid’s Metamorphoses which describe the palace of
the Sun and the wonderful engravings of the earth, sea and sky which
adorned it, and Ermold the Black’s detailed account, in a piece addressed
to the Carolingian emperor Louis the Pious, of the paintings to be seen in
the early ninth century on the walls of the church and secular hall of the
royal palace at Ingelheim;3 the latter comprised scenes from both the Old
and New Testaments and from antiquity: Cyrus, Phalaris, Romulus and
Remus, Hannibal and Alexander all had their place, together with the

1
Perforat Hairaldum casu letalis arundo, AC, l. 463; letalis arundo is a reference to Virgil’s Aeneid,
iv. 73. Tilliette, ‘La chambre de la comtesse Adèle’, 167-9, notes other references to Virgil, to Lucan’s
Pharsalia and to Statius’ Thebaid, and points out that for Baudri (as for WP, above, pp. 83-4) ‘on ne
donnera une idée de l’importance de la bataille d’Hastings qu’en l’assimilant à la bataille antique de
Pharsale ou aux combats mythiques devant les murs de Thèbes’. Ratkowitsch, Descriptio Picturae, pp.
60, 73, notes also Baudri’s use of the Psychomachia by the late Roman Christian poet Prudentius, and
argues that in thus stressing that the duke was a Christian warrior he sought to portray him as the
superior of Aeneas and Caesar.
2
It is the only one of Baudri’s poems surviving in part (ll. 749-946, on the world map) and in a variant
text in a second manuscript, written in York in the fourteenth century. Ratkowitsch, Descriptio
Picturae, pp. 18, 29, follows the suggestion of Gautier Dalché and Tilliette, ‘Un nouveau document’,
247-8, that this variant goes back to the copy sent to Adela, while the other version is that kept by
Baudri for insertion into his collected poems.
3
Iliad, xviii. 480ff; Aeneid, viii. 626-728, vii. 789-92; Metamorphoses, ii, 1-18. These examples and
many others are given by Ratkowitsch, Descriptio Picturae, pp. 9-11, 21-4.

90
 3 THE SOURCES: PART TWO

Christian Roman emperors Constantine the Great and Theodosius, while

the more recent triumphs of Frankish princes were represented by Charles
Martel’s victories over the Frisians, King Pepin I’s over the Aquitanians
and those of Charlemagne (Louis’ father) over the continental Saxons.1
Yet if the items decribed in Ermold’s poem have a certain amount in
common with those of the Adelae Comitissae, there are also significant
differences between the two pieces. Ermold was a member of Louis’
court and presumably familiar with the subject matter he was describing,
while by contrast it is far from clear how well Baudri knew Adela (and
others whom he addressed in familiar terms),2 and his poem itself
suggests that his account of the Norman Conquest hanging and therefore
probably of the entire chamber is in fact entirely fictional.3 Yet his
description of the battle cannot be regarded in the same light. Whether he
owed very much to other surviving sources is something upon which
scholars have failed to agree. The figure of 3,000 for William’s fleet
could well have come from William of Jumièges, it is conceivable (but
unlikely, see below) that he was familiar with the Bayeux Tapestry, and a
great deal of his material can also be found in the Carmen and William of
Poitiers, including that on the great density of the English formation, the
duke’s use of missile troops, the feigned followed by a real French flight
(this being the version of the Carmen rather than the Gesta Guillelmi),
William’s removal of his helmet to halt it and personal prowess in the
fighting,4 and the belief that this continued until nightfall.5 Nevertheless,
Baudri was no slave to such sources even if he did know them, for unlike

1
Ermold the Black, In honorem Hludovici Pii, ll. 189-284. On this poem and its context, see Godman,
Poetry of the Carolingian Renaissance, pp. 45-7.
2
See Phyllis Abrahams’ comments in her edition of AC, pp. xxiv-v.
3
In ll. 567-70, Baudri says (in Herren’s translation) ‘If you could believe that this weaving really
existed you would read true things on it, O writing paper. But you might also say: “What he wrote
ought to have been; a subject like this was becoming to the goddess”.’ See B&W, 26; Barlow, in
Carmen, p. lx; Brown and Herren, ‘The Adelae Comitissae’, espec. 71-3; Tilliette, ‘La chambre’, 153,
170. On Baudri, and poems addressed by others to Adela, see also van Houts, ‘Latin poetry’, 49-50.
Tilliette (161) shows that some passages of the poem would be unintelligible to a reader not possessed
of such considerable learning as was clearly attributed to Adela.
4
Ratkowitsch, Descriptio Picturae, p. 72, notes that the removal of the helmet is ‘ein altes episches
Motiv, das bei mehreren antiken römischen Epikern anzutreffen ist’, and suggests that Baudri modelled
it on Aeneid, v. 670ff, and x. 368ff; but these are not precise parallels, and in the second, where Pallas
atttempts to halt his routing Arcadians by making a speech similar to William’s and then charging into
battle like the duke, he does not first remove his helmet in order to be recognised. Note also the speech
allegedly (EE, pp. 20-1) made by Earl Thorkell to rally Danish troops fighting the English at Sherston
in 1016.
5
Tilliette, ‘La chambre’, 154, acknowledges the links with WP but thinks that ‘nous sommes plutôt ici
en présence d’une composition personelle de Baudri à partir de la “version officielle” diffusée par les
propagandistes de la nouvelle dynastie’. However, in concluding his discussion of the derivation of the
other subject matter of the poem, he comments (160) that ‘dans chaque matière, l’auteur a consulté
l’autorité reconnue et mis en vers son enseignement’. Ratkowitsch, Descriptio Picturae, pp. 59-61 is
sure that Baudri used WJ and WP and thinks he may have known the Carmen; she also suggests that he
was familiar with Godfrey of Rheims’ poem Lux mores hodierna, which includes a celebration of the
virtues of the Conqueror and Adela; on this, see van Houts, ‘Latin poetry’, 47-8.

91
 THE BATTLE OF HASTINGS

him they do not attribute to the duke speeches made as he landed in

England and at the battle’s close, and it is difficult to say when he was
drawing upon them and when he was simply utilising oral traditions
about the fight similar to their own. Moreover, if he had ever had any
contact with Adela the authority of those traditions may have been good.
It is of no small interest from this point of view that one of them had
Harold killed by an arrow. Baudri did not take this from either the
Carmen or the Gesta Guillelmi, which do not contain it, and there is no
obvious reason to think that he drew it from the history of the Normans
written in Italy c.1080 by Amatus, monk of Montecassino; this specifies
that the English king was hit in the eye.1 It looks as though the belief that
Harold had been killed by an arrow, which is also to be found in twelfth-
century sources, was one attached to the battle at an early date.
In 1075 Orderic Vitalis was born near Shrewsbury. The son of an
English mother and a French priest, he was offered at the age of ten to the
monastery of St Évroul in Normandy, and there he wrote the works upon
which his reputation as one of the great Anglo-Norman historians of the
first half of the twelfth century is based. From about 1095 he compiled a
set of annals for his abbey, and it may have been at about the same time
that he began revising the Gesta Normannorum Ducum of William of
Jumièges; he had completed most of it by 1109.2 Orderic’s additions were
very extensive, and sometimes he was concerned to correct the bias of his
sources: unlike William of Jumièges and William of Poitiers, for
example, he was prepared to write about the Conqueror’s illegitimate
birth, and when he dealt with England he took trouble to give a balanced
account. On the events of 1066 he inserted details of the almost complete
annihilation of the Norwegian forces at Stamford Bridge (while giving,
wrongly, the date of 7 October) and also tells the story of how Harold’s
brother Gyrth attempted to dissuade him from being involved in fighting
against the duke, a story which may have come from ‘a lost account,
1
Contre cestui (Harold) ala premerement Guillerme, et combati contre lui; et lui creva un oill d’une
sajete, et molt gent de li Englez occist; Amatus, Storia de’ normanni, ed. Bartholomaeis, pp. 11-12. The
Latin text does not survive, but was translated into Old French in the early fourteenth-century; the
accuracy of this version (and therefore whether the statement about the arrow stood in Amatus’
original) is not certain, although Bartholomaeis, its most recent editor, did not think the material on
William’s conquest of England interpolated. On Amatus, see Wolf, Making History, pp. 88-9, who
thinks the French version ‘not hopelessly corrupt’; Albu, The Normans in their Histories, pp. 109-10,
judges it ‘more an adaptation than a precise translation’ and thinks use of it ‘as if it were an eleventh-
century Norman text’ problematic. I am grateful to Dr Graham Loud for assistance on Amatus. There
are similarities between Baudri and the Tapestry, but Abrahams (n. 61, pp. 244-7) and Tilliette (‘La
chambre’, 150-2) argue, convincingly in my opinion, that they are insufficiently compelling to indicate
his knowledge of it; for a different view, see Brown and Herren, ‘The Adelae Comitissae’, and
Ratkowitsch, Descriptio Picturae, pp. 62-5, 74-6, but their case is weakened by the reference to the
arrow in Amatus and above all by the probability that the Tapestry did not originally show Harold
struck by one (see the Appendix, pp. 203-13).
2
For this and what follows, including comment on Orderic’s additions and alterations to WJ, see GND,
i. lxvi-lxxv.

92
 3 THE SOURCES: PART TWO

probably poetical’.1 On the battle of Hastings itself, he left William of

Jumièges’ text largely unaltered, but added to it a description of the
difficulties experienced by the French during their pursuit of the fleeing
English, when they rushed against an ancient rampart concealed by high
grass, fell against each other and were crushed to death. William of
Poitiers (whose work Orderic probably knew at this date, as he certainly
did later) has a similar incident at the same stage in the battle, but he
describes it in rather different terms, and it looks as if Orderic was
drawing on an independent tradition.2
Shortly after completing this work he turned to that known as the
Historia Æcclesiastica (History of the Church), and spent about ten years
on the book which includes a treatment of Hastings, the first he wrote,
which was finished c.1124.3 He reports that King Harold was a fine
physical specimen and famed for his eloquence and bravery, but criticises
him for usurping the English throne, and says that it was done without the
consent of the people and that the ensuing reign saw much oppression. He
then wrongly assigns the exile of Tostig to 1066, and claims that the earl
visited Normandy and encouraged William to cross the Channel with an
army. Next he moves to William’s calling of a council and dispatch of an
embassy to Rome (both drawn, with additions, from William of Poitiers),
and to Tostig’s adventures in England and supposed visit to Harald of
Norway. An extended treatment of the affairs of St Évroul (where the
abbot died at this time) intervenes, before Orderic returns to Tostig and
Harald and the battle of Stamford Bridge, and then William’s landing on
the south coast and Gyrth’s attempt to persuade his brother not to fight.
From this point down to the finding of Harold’s body and its burial his
account follows that of William of Poitiers fairly closely, while reducing
from 700 to 70 the number of English ships said to have been sent into
the Channel to prevent the French escaping, and adding a number of
details of its own: Thurstan, son of Rollo, is said to have borne the
standard of the Normans, and the name of the place is correctly given as
Senlac. William of Poitiers does not say when Harold was killed, and
Orderic therefore followed William of Jumièges in stating that it was
during the first assault, and he also repeated his own earlier account of the
disaster which befell the French during the pursuit, this time fusing it
with the version in the Gesta Guillelmi, and naming one of the dead (of
perhaps the battle as a whole) as Engenulf of Laigle. If he had heard

1
Orderic, in GND, ii. 164-71; Chibnall, HÆ, ii. xxiv.
2
In his later work he amalgamated this story with that of WP, see further, below, pp. 186-7.
3
HÆ, ii. xv. It had its origins as a history of his own monastery.

93
 THE BATTLE OF HASTINGS

stories about how Harold died, and it is difficult to believe that he had
not, he did not choose to repeat them.1
A work contemporary with this book of Orderic’s Historia
Æcclesiastica is that known as the Brevis Relatio de Guillelmo
nobilissimo comite Normannorum (Brief History of the most noble
William, count of the Normans), in fact a history of Normandy and
England from about 1035 to the early twelfth century written by an
anonymous but probably Norman monk of Battle Abbey between 1114
and 1120, and surviving in what is believed to be the author’s autograph.
His work was carried out in the time of Abbot Ralph (d. 1124), who had
known both the Conqueror (as a royal chaplain, like William of Poitiers)
and Archbishop Lanfranc of Canterbury, and who may well therefore
have been a good witness, even on events which he had not seen at first
hand; these no doubt included the battle of Hastings.2 His narrative of the
Conquest is broadly similar to that of William of Jumièges and William
of Poitiers, while containing a number of details which they do not.
Harold, for example, is said to have sworn his oath on a reliquary called
the Ox’s Eye, and to have been crowned at St Paul’s;3 when, after his
victory over his brother (Harald Hardrada is not mentioned), he heard of
William’s invasion, he came to London as quickly as possible and spoke
to the English and Danes who were with him, holding his oath to the duke
as nothing. Then he left London with all his army and came to the place
now called Battle, reportedly saying that he had never done anything so
willingly as come to this fight, and little knowing the punishment that
almighty God was about to inflict upon his pride. William and the
Normans came to a hill opposite that occupied by the English, and a
speech said to have been made by the most Christian duke as he was
putting on his armour is worthy to be transmitted to posterity. He was
accidentally offered his coat of mail back to front, and calmly told the
soldiers standing near him that if he believed in omens he would not go
into battle that day, but in fact he had never believed in such things, only
in the Creator. Once armed, he asked a soldier standing near him where
he thought Harold to be, and was told that he was believed to be in the
very dense formation of men on the crest of the ridge (in illo spisso
agmine quod erat ante eos in montis summitate), for the soldier thought
Harold’s standard could be seen there. William then called upon God to

1
HÆ, ii. 134-45, 168-79; on Senlac, above, pp. 49-50; on Orderic and the family of Engenulf, below,
p. 187, n. 3. Like other sources, Orderic misnames the Norwegian king Harald Fairhair, information
perhaps obtained when he visited Worcester, see HÆ, ii. 188-9, and Chibnall’s note, 186, n. 1.
2
See van Houts’ comments, BR, pp. 7-15.
3
BR, pp. 28-9. Van Houts thinks that the reliquary was among those bequeathed to Battle Abbey by the
Conqueror on his deathbed, ‘The Memory of 1066’, 167-9. The Chronicon Monasterii de Hida account
(see below, p. 102, n. 2) p. 290, which may have used BR, claims that the reliquary was one of St
Pancras, a detail rejected by Lewis, ‘The Earldom of Surrey’, 334.

94
 3 THE SOURCES: PART TWO

do him justice that day against Harold the perjurer, spurred his horse into
the English ranks and killed one of their number. Not much later a
Norman unit (cuneus) of almost a thousand horsemen attacked elsewhere
and rushed upon the English with a great impetus as if wishing to slay
them, but when they reached the enemy line pretended to flee as if they
were afraid. The English believed they were really fleeing and set off in
pursuit in the hope of killing them, but when they saw this the Normans,
who were more cautious in war than the English, soon turned back and,
placing themselves between the pursuers and their main line, quickly
annihilated them. Thus the Normans and the English began fighting each
other, but the battle lasted almost all day and into the evening, until the
defeated English fled, and those who could not flee remained to die. In
that battle Harold and his two brothers were killed, and with them the
greater part of the English nobility, and in that place Duke William
ordered an abbey to be built in memory of his victory and for the
absolution of the sins of all those who had died there.1
There is little sign that this description of the battle owes much to any
other that has survived, for although the story of the reversed armour
could have come from William of Poitiers it may be that both writers had
heard the same tradition, and frequent use in the Brevis Relatio of terms
like ‘they say’ (fertur) suggests that its author’s sources were often oral.2
This is hardly surprising, as Abbot Ralph is unlikely to have been the
only Battle monk with stories to tell. It is also significant that while the
feigned retreat appears here as elsewhere, the Brevis Relatio places it at
the beginning of the fight, and does not share William of Poitiers’ opinion
that it followed upon a real rout which the duke managed to halt. It does
share with other sources the belief that the fighting lasted virtually all
day, but if the community at Battle knew anything of deaths in ditches, or
of reverses of any kind suffered by the French forces, this author
preferred to remain silent about them.
The final source which needs to be considered by the historian of the
battle of Hastings is the longest and one of the most colourful, the Roman
de Rou (History of Rollo) by Master Wace. A native of Jersey (then part
of the duchy of Normandy) born c.1110, Wace was educated in Caen and
Paris, and eventually returned to Caen, where he produced a number of
verse pieces in the vernacular, including saints’ lives and his two major
works, the Roman de Brut and the Roman de Rou. The Brut, finished in
1155, was a translation into French of Geoffrey of Monmouth’s history of
the kings of Britain (written in the 1130s), while the Roman de Rou deals
with the deeds of the Norman dukes from the earliest, the tenth-century
Rollo, down to the battle of Tinchebrai (between the Conqueror’s sons
1
BR, pp. 31-3.
2
A point made by van Houts, BR, p. 22.

95
 THE BATTLE OF HASTINGS

Duke Robert of Normandy and King Henry I of England) in 1106. More

than 16,000 lines in length, it was begun in 1160, but Wace did not lay
down his pen until the middle of the 1170s, when he says that King
Henry II required him to hand the work over to a Master Benoît (probably
Benoît of St Maur); this was after the king had made him a prebendary of
Bayeux cathedral.1 He was a cleric, but wrote in French probably to
reach a lay audience which could not understand Latin; he also wrote to
entertain, but this is not to say that he was not a serious historian too, for
he knew and used a considerable number of the authors who have already
received our attention – including William of Jumièges, William of
Poitiers, Eadmer, Orderic Vitalis and William of Malmesbury; he may
also have been familiar with the Carmen and the Brevis Relatio, and the
Bayeux Tapestry, if it was in Bayeux in his day.2 These sources, as he
was aware, cannot always be reconciled, and this was something which
he was prepared to bring to his audience’s attention: when dealing with
the reason for Harold’s visit to Normandy during Edward the Confessor’s
reign, for example, he mentions the statements of the French sources, but
adds the English version of events given by the Canterbury writer Eadmer
and says that he does not know the truth of the matter.3 Thus, while it
may be necessary to acknowledge that his statements are based on what
Round called ‘a congeries of authorities, on tradition, and occasionally of
course, on the poetic invention of the trouveur’,4 the fact remains that
Wace went to some trouble to seek out information, and that his efforts
resulted in the preservation of material which would not otherwise have
survived. It has been suggested that he deliberately researched oral
memories of the events of 1066, and that during his visit to the abbey of
Fécamp in 1162 he was informed that the monk who acted as one of
Duke William’s messengers to Harold before the battle was named Hugh
Margot.5 On the size of the French fleet, he comments that as a boy he
had heard his father say that 696 vessels sailed from St Valéry, but that he
had also found it written as 3,000 (a figure given by both William of
Jumièges and Baudri of Bourgeuil); he also seems to have had reliable
information on the numbers of ships provided by some of the duke’s
magnates.6 Accordingly, when he gives details which are his alone they

1
I have followed Holden’s account of Wace, iii. 15-18; van Houts, ‘Wace as Historian’, pp. 103-6; and
the introduction to the new translation by Burgess. See also Eley and Bennett, ‘The Battle of Hastings’,
54-68, who describe the battle as ‘the centre of gravity of the entire work’.
2
B&W, 28, think that Wace ‘drew heavily on the Tapestry for his description of the Norman
conquest’; others have been less convinced, RR, trans. Burgess, p. xxxiv.
3
RR, ll. 5543-604; see above, pp. 29-30. On Wace’s frequently-expressed concern for truth, see RR,
trans. Burgess, pp. xxxvi-vii.
4
Round, ‘Wace and his Authorities’, 677.
5
Van Houts, 'Wace as Historian', 114-5.
6
RR, ll. 6423-32; below, pp. 146-7.

96
 3 THE SOURCES: PART TWO

cannot be dismissed simply on the basis of the relatively late date at

which he worked; equally, there is no way of checking them.
The events of 1066 occupy a major place in the Roman de Rou, taking
up more than 3,000 lines in all, while the account of the battle of Hastings
alone is about a tenth of the whole.1 This gives a clear indication of the
importance of this subject matter to Wace’s audience, which was also
thought to be very interested in the individuals involved: other sources
mention few participants in the battle on the French side, but Wace refers
to a great many Norman lords. His account of the events which were to
be of such interest to their descendants begins with William hearing of
Harold’s usurpation of the English throne and assembling a council of his
men, some of whom proved willing to undertake an expedition, others
not. Eventually they agreed to serve, and the duke then called on areas
bordering Normandy, promising lands and payment to those who would
follow him. He also sent to the pope for support, and received from Rome
a gonfanon (standard) and a most precious ring, which had beneath the
stone a tooth of St Peter; at much the same time Halley’s Comet
appeared. A description of the building of William’s fleet follows, and its
assembly at St Valéry, voyage across the Channel and arrival near
Hastings. Once there, the duke’s men constructed a fort out of material
they had brought with them, while the news of their landing was carried
to Harold beyond the Humber. He quickly reached London, ordering an
army to assemble from throughout England, and there encountered an
envoy of William, the Fécamp monk Hugh Margot, who rehearsed his
master’s claim to the throne and urged Harold to relinquish it; in reply,
his own envoy to the duke offered a large sum of gold and silver if the
French would depart, or if not a battle on the coming Saturday. Wace
then tells the story of how Harold’s brother Gyrth attempted to dissuade
him from leading the English forces, of Harold’s refusal, and of his
setting up of his standard on the site of Battle Abbey. He also had the
place surrounded by a good ditch with entrances on three sides, and the
following morning the two brothers rode out to reconnoitre the French
positions.2 The sight caused Harold to propose a delay until he had a
larger army, but Gyrth rejected this because of the effect it would have on
English morale, and they returned to camp to prepare for battle. Further
negotiations followed in which William offered Harold all Northumbria if
he would submit and Gyrth all the lands of their father Earl Godwin; he
also threatened any of the English who supported them with
excommunication, much to their concern. However, their spirits were

1
Bennett, ‘Poetry as History?’, 25.
2
Eley and Bennett, ‘The Battle of Hastings’, 60-1, note that Harold and Gyrth are here ‘a regular pair
of heroic companions from the chanson de geste, inviting comparison with Vivien and Girart, Raoul
and Bernier, Roland and Oliver’.

97
 THE BATTLE OF HASTINGS

restored by a rousing speech from Gyrth (in Wace’s version of events at

least as significant a figure as Harold himself) and, the talking being
done, both sides prepared to fight.
At this point Wace included William of Malmesbury’s account of how
the English spent the night before the battle carousing, while the French
made confession of their sins. A lengthy speech by the duke in which he
stressed the treachery of the English followed an arrangement that his
forces should form three divisions, and we are then given the tale, also to
be found in William of Poitiers and the Brevis Relatio, about William
donning his armour reversed, while Orderic Vitalis’ simple statement that
the Norman standard was borne by Thurstan, son of Rollo, is elaborated
into his carrying of the papal banner only after others had refused the
honour.1 Wace next, however, inserted material not from any known
source: the duke instructed Roger of Montgomery2 and William
FitzOsbern to command one of the wings, containing the men of
Boulogne and of the county of Poux in Picardy and soldiers of his own,
while the other consisted of troops from Brittany, Poitou and Maine under
the Breton Alan Fergant3 and Aimeri, viscount of Thouars; in the centre
was William himself. As for Harold, he had summoned men from castles,
cities, ports, villages and boroughs, and they included peasants armed
with clubs, iron forks and stakes; none had come from north of the
Humber, but the names of many of the shires and towns south of it are
listed as having sent contingents. Wace at this point rejects the
widespread belief that the English army was small, and gives it as his
own opinion that their numbers equalled those of the French. They had,
he says, erected a barricade to their front,4 and Harold ordered the men of
Kent to go where the first attack would be made, for it was the custom
that the first blow belonged to them,5 as it was also that the Londoners
should be stationed around the standard and guard the king himself. As
the French approached, Harold and Gyrth further disputed the wisdom of
fighting, it being the latter’s opinion that large as their army was (400,000
is the figure given) the peasantry was of questionable worth compared
with troops of the quality which now opposed them. However, they took
up position near the standard, and their men gave their battle-cries of
‘Holy Cross’ and ‘God Almighty’, while Taillefer rode before the duke
singing of Charlemagne, Roland, Oliver and those who died at the battle
of Roncesvalles, and asking for permission to strike the first blow. This

1
See further, Bennet, ‘Poetry as History?’, 33.
2
On Roger, see below, p. 99.
3
A mistake for the Breton Alan Rufus, van Houts, ‘Wace as Historian’, p. 118.
4
These were the lines (7793-8) disputed by Round and his opponents in the 1890s, see below, pp. 114-
5..
5
On this, see further, below, p. 137.

98
 3 THE SOURCES: PART TWO

he did before the English surrounded him, and as the lines closed there
began the battle which is still, says Wace, of great renown.
The opening lines of his description of it consist of stock phrases
enlivened only by the inclusion of the Norman battle-cry of Deus aïe! and
the English one of Ut! (‘Out!’), and the first concrete detail is of the
Normans being pushed back upon a ditch which they had earlier crossed,
and of horses and men being thrown into it; many of the English died
there, and more of the Normans at that time than any other, as those who
saw the dead said. Upon witnessing this the boys set to guard the baggage
began to panic until Odo of Bayeux rode up on a white horse to rally
them, before returning to a battle which raged from nine o’clock in the
morning until three in the afternoon without either side gaining an
advantage. It was then that the French archers began to fire into the air so
that their arrows fell upon the heads and faces of their enemies, and
Harold was struck on the right eye, which was put out; only after he was
wounded, according to Wace, did the Normans stage a feigned flight to
break up the enemy line. The English foolishly followed them, thus
abandoning a position in which they could hardly have been defeated, and
as their enemy turned to face them once again the Roman de Rou
becomes for many lines a recital of fighting involving the alleged actions
of 117 Norman lords, 74 of them simply designated by the territory which
they ruled and without a personal name being given.1 Even when names
are supplied they sometimes refer to people such as Roger of
Montgomery who are known not to have been present.2 Yet, while one
might believe that much of this catalogue of heroic deeds was invented by
Wace to please the descendants of those supposedly involved, he may
have eked out some of his comments with genuine traditions. He says, for
example, that Robert son of Erneis broke through the English in an
attempt to cut down their standard, but was surrounded, killed, and
eventually found lying at its foot; these details cannot be confirmed, but a
charter granted by his son to the monks of Fontenay does say that he was
killed in England in the Conqueror’s time.3 Recently, Dr van Houts has
gone further than this, and has suggested that Wace’s details of the
exploits of individual Normans, far from being largely fictional, were
actually based on research among his contemporaries upon the nature of
family traditions about the battle.4
Of the English themselves the men of Kent and Essex are supposed to
have fought well and pushed the Normans back, until a charge led by
1
The figures are from Douglas, ‘Companions of the Conqueror’, 131; see also Bennett, ‘Poetry as
History?’, 28-35, for the argument that Wace sometimes alludes to the politics of his own day.
2
Douglas, ‘Companions’, 132, 143-4; OV, HÆ, ii, 210-11, says that the duke left Roger in Normandy
to act as regent.
3
Douglas, ‘Companions’, 142.
4
Van Houts, ‘Wace as Historian’, espec. pp. 109-16.

99
 THE BATTLE OF HASTINGS

Duke William finally broke their countrymen’s resistance and his men
penetrated at last to the English standard. There the wounded Harold still
stood, until beaten down twice, first by a blow to his helmet and then by
one which cut his thigh to the bone. His brother Gyrth fell under a blow
from the duke himself, and then Harold too was killed, although Wace
says that the press around the king was so great that he knew not the
identity of the slayer. Even then, he reports that the English fought on
until the close of day, until news of their king’s death spread, and men
began to flee, not stopping before they reached London. As for William,
two horses had been killed beneath him, and as he removed his armour
and his attendants saw the damage it had sustained they exclaimed that
there had not been such a knight since the days of Roland and Oliver. He
raised his standard and tent, took his evening meal and slept at the place
where the English standard had been, and the following day the French
buried such of their friends as they could find, while native noblewomen
took the corpses of their relatives to be interred in village churches;
Harold himself was borne to Waltham (in Essex) for burial.
Such are the surviving sources on the battle of Hastings. Those from
the English (or, more strictly speaking, Anglo-Norman) side do not exist
in any quantity until the first half of the twelfth century, when authors
such as John of Worcester, William of Malmesbury and Henry of
Huntingdon were influenced by native resentment at the effects of the
Conquest, a tendency to make excuses for the defeat and the very
common medieval belief that disastrous events could only be the
judgement of God upon a sinful people. Nevertheless, these accounts are
relatively straightforward compared with the problems posed by the
significant body of almost contemporary material from French writers.
The Carmen de Hastingae Proelio may well be the work of Guy of
Amiens, but it and the writings of William of Jumièges, William of
Poitiers and Baudri of Bourgeuil are panegyrics of Duke William, and
shape their comment accordingly; neither substantial military difficulties
experienced by the Normans nor anything else to the duke's discredit are
likely to be reflected in their pages, and the temptation to draw parallels
between the events of 1066 and those of the classical period may have led
to distortion. Moreover, some of the similarities between the sources are
the result of direct borrowing, thus reducing their value as evidence:
Orderic Vitalis and Wace certainly used earlier material, for example, and
the accounts of the Carmen, William of Poitiers and Baudri of Bourgeuil
may well be related; similarly, both Baudri and William of Malmesbury
may have seen the Bayeux Tapestry.1 By the twelfth century, and the

1
But see p. 92, n. 1 above on Baudri, and for comment on WM and the Tapestry, below, pp. 176-7.

100
 3 THE SOURCES: PART TWO

works of Gaimar and Wace, we are to an extent in the realms of history as

fiction.
Nevertheless the picture is far from being an entirely negative one.
Wace, despite the fact that he was writing a century later, probably had
access to a certain amount of valuable information independent of our
other surviving sources, and the same is true of most of his predecessors.
Men who fought in the battle need not always have been particularly
reliable witnesses, both because of genuine confusion as to what course
events had taken and a desire to magnify their own achievements, but the
likelihood that the three contemporary French authors (Guy of Amiens,
William of Jumièges and William of Poitiers) had spoken to eye-
witnesses is high; the designer of the Bayeux Tapestry and Baudri of
Bourgeuil may have done so too. Moreover, one can hardly stress too
much that there was once a very much greater body of evidence which
has not come down to us. Some of this was committed to writing, for if
Sigebert of Gembloux was correct in saying in the early twelfth century
that Archbishop Lanfranc wrote a work about the deeds of Duke William1
it is likely that it referred to Hastings, and the fact that both Guy’s
Carmen and William of Poitiers’ Gesta Guillelmi have survived by a
hairsbreadth, and the full text of Baudri’s Adelae Comitissae only in a
single manuscript, is by itself a salutary warning of how much may have
disappeared. The songs and stories of the period have certainly done so,
but are known to have existed: Henry of Huntingdon says that as a boy he
had heard very old men speak of King Æthelred II’s massacre of Danes in
England on St Brice’s Day in 1002; William of Malmesbury that the
magnificence of the celebrations attending the marriage of Cnut’s
daughter Gunnhild to the future emperor Henry III in 1036 was still
celebrated in songs in his own time; Wace that he had heard his father
comment on the number of ships which sailed with the Conqueror from
St Valéry.2 It would be astonishing if one of the most celebrated battles of
the age was not also the subject of song, and of an immense body of
popular report, rumour and reminiscence which eventually passed away
without ever having taken any sort of literary form. This being so, one
should not be too ready to assume that similarities between different
sources are necessarily the result of direct borrowing, for what such
parallels may often reveal is that stories about the battle, wherever and by
whomever they were told, tended to group around certain themes: the
long and hard-fought nature of the conflict, the difficulty which the
French had in breaking the densely-packed English line, the significance
of a real and/or a feigned retreat by William’s men, the role of his
1
I owe this reference to Barlow, Carmen, p. xiv, n. 3, who comments that Sigebert’s words are ‘a good
description of GG, usually ascribed to William of Poitiers’.
2
HA, pp. 340-1; GR, i. 338-9; Wace, RR, ll. 6423-8.

101
 THE BATTLE OF HASTINGS

archers, deaths in ditches during the battle and/or after it, and the
importance of Harold’s own death (in some versions through being hit by
an arrow) late in the day. Moreover if these themes recur so frequently
because they reflect what was commonly said, it is no great distance from
this to the argument that what was commonly said was itself some sort of
reflection of what had actually happened, or at least what was believed to
have happened.1 Historians of more recent conflicts are frequently able to
establish a fairly precise timetable of events and say a great deal about the
individuals involved in the fighting. It will never be possible to
reconstruct Hastings on this kind of scale; its story can never be told as
can those of Waterloo and of Gettysburg, of the first day of the Somme
and of D-Day. Nevertheless, it is not completely beyond recovery as a
historical event.2

1
Note, by way of contrast, Barlow’s conclusion (Carmen, pp. xc-i) that of the ‘near-contemporary’
sources ‘all are variations on a common literary theme’, and that if ‘Guy of Amiens had not written his
poem, the story would probably have been quite different’. Given the great fame of the battle, and the
probable access of those responsible for the sources produced before c.1100 to eye-witness testimony
even if only at some removes, this elevation of the importance of the Carmen in the creation of later
accounts seems scarcely plausible. In earlier work, Barlow was willing to allow a significant role to
verbal witness in shaping the near-contemporary sources, although he also expressed considerable
scepticism about the details they provide, and suggested that Hastings is ‘almost as obscure’ as Fulford
and Stamford Bridge, ‘The Carmen de Hastingae Proelio’, pp. 189-90, 217; in The Godwins, p. 104, he
states that ‘we have very few facts’ about it.
2
Some sources which refer to the battle have not been discussed in detail here. The Chronicon
Monasterii de Hida account, pp. 293-4, dates from c.1128-34 and was perhaps written by a Norman
monk of the Cluniac priory of St Pancras at Lewes or by a Norman author working in the duchy itself;
see Lewis, ‘The Earldom of Surrey’, 330-4, and Gillingham, ‘Henry of Huntingdon’, Appendix. The
Draco Normannicus of Stephen of Rouen is a Latin verse treatment of Norman history derived from a
number of earlier writers including WJ and composed in the late 1160s (ll. 1427-95 on the battle). On
Benoît of St Maur’s account, written after that of Wace, whom he replaced in Henry II’s favour, see
Eley and Bennett, ‘The Battle of Hastings’, 68-77, and more generally, Blacker, The Faces of Time, pp.
45-52, 119-32, 185-90.

102
 4

THE ENGLISH ARMY

No matter with which we have to deal is darker than the constitution of
the English army on the eve of its defeat.
F.W. Maitland, Domesday Book and Beyond

In the popular consciousness, Harold being hit in the eye by an arrow at

Hastings is one of the most famous of historical events, and seemingly of
a piece with other gems from the period: Alfred burning the cakes at a
low point in his fortunes, Æthelred the Unready foolishly paying the
Danes to go away, all combine to give the poorest of impressions of the
military capacities of Anglo-Saxon government. Nor is it a picture which
a closer look might seem to contradict. Between about 800 and 1066
England was the scene of more fighting than at any other time in its
history, apart probably from the opening decades of the Roman invasion.
In the ninth century, after an initial period of coastal raiding, the Vikings
destroyed the kingdoms of East Anglia and Northumbria, and also
conquered half of the midlands realm of Mercia; Alfred’s alleged culinary
misfortunes1 occurred as they almost overcame that of Wessex too.
During the 870s, English rulers survived in England by only the
narrowest of margins. In the late tenth and early eleventh centuries, by
which time they were dealing with the united country created by Alfred’s
successors, the Scandinavians reduced the government of Æthelred II (the
Unready)2 to apparent impotence, and in 1016, after about twenty-five
years of fighting, Æthelred’s dynasty was deposed and replaced by that of
the Danish Cnut. Exactly fifty years later, following the interval which
saw the extinction of Cnut’s line and the succession of Æthelred’s son
Edward the Confessor, Harold II was defeated at Hastings, and the
country had the first of what was this time to prove a long line of foreign
kings.

1
On this legend, see Alfred the Great, ed. Keynes and Lapidge, pp. 197-202.
2
The term is a later distortion of the soubriquet unræd, ‘no counsel’, which is probably contemporary,
and was a malicious pun on the name Æthelred, ‘wise counsel’.

103
 THE BATTLE OF HASTINGS

Moreover, if all this gives an impression of military ineffectiveness it is

one which some of the sources would appear to confirm. Of all the battles
fought in England in this period we have reasonable accounts of only
three, and only one of those, King Æthelstan’s engagement against a
great coalition of his enemies at Brunanburh in 937,1 was a victory. The
other two were defeats – Hastings, and the fight between the local levies
of Essex and a Viking army at Maldon in 991, during the reign of
Æthelred. At Maldon the Vikings landed on an island, probably because
it offered some protection to their ships, and found the English facing
them at the other end of the causeway which connected it to the mainland.
The native commander, Ealdorman Byrhtnoth, rightly or wrongly allowed
them to cross the causeway without opposition so that the battle could
take place on equal terms, and was subsequently killed, whereupon part
of his army fled. According to the poem The Battle of Maldon the
remainder then affirmed, in some of the most moving words in Old
English literature, their determination to show their loyalty to their lord
by dying where they stood.2 Whether they actually did so is not known,
but certainly they were overcome, and the same year saw the payment of
the first of the very large tributes for which Æthelred’s reign is famous.
At Hastings defeat came about, if William of Poitiers can be believed,
through apparent indiscipline, when English forces three times left the
position which had served them so well to chase a French army which
then encircled and destroyed its pursuers. If there is added to this the
depiction on the Bayeux Tapestry of figures who are clearly poorly-
armed peasants, anyone might be forgiven for thinking that Anglo-Saxon
armies were little more than armed mobs which trusted to bravery and
luck more than to elaborately-organised logistics, military technology and
tactical skill.
Yet this cannot be so, as can be readily demonstrated by a rather more
detailed survey of late Anglo-Saxon history than that undertaken so far.3
Alfred may in the late ninth century have saved Wessex by the narrowest
of margins, but he did so by carrying out well-known reforms to its
military system which ensured that Viking raiders received a rather
warmer welcome in the later stages of his reign than they had done
hitherto. His son Edward the Elder, in campaigns which are recorded in
some of the most stirring and remarkable of all the entries in the Anglo-
Saxon Chronicle, subdued East Anglia and Danish Mercia, and by 927
his grandson Æthelstan, virtually forgotten today but one of the great
1
ASC, i. 106, 108-10. The location of the battle is not certain.
2
The poem is incomplete at both beginning and end, and is known only from an eighteenth-century
transcript, the manuscript having been destroyed by the fire in the Cotton Library in 1731. Perhaps
composed shortly after the battle, its historical reliability has been much disputed; see briefly, The
Battle of Maldon AD 991, ed. Scragg, pp. 34-5.
3
See, similarly, Hollister, Anglo-Saxon Military Institutions, pp. 128-9.

104
 4 THE ENGLISH ARMY

European figures of his time, was the first king of all England. In 934 he
led both an army and a navy into Scotland in a campaign which
foreshadowed those of Edward I over 300 years later, and received the
submission of northern kings and princes; on some of his coins appeared
a legend used by no other pre-Conquest king: Æthelstan rex tot(ius)
Brit(anniæ) (‘Æthelstan, king of all Britain’).1 Forty years later his
relative Edgar, another of the great forgotten figures of English history,
also achieved dominance over much of Britain, and perhaps of part of
Ireland too, being rowed on the Dee at Chester by six of his subordinates2
after a coronation earlier in the year which may have been intended to
emulate that of the western emperor, Otto I of Germany, crowned in
Rome in 962. Of course, Edgar’s son, Æthelred II, is not forgotten, and
for the wrong reasons, but that there were great inherent strengths in the
system of government which he inherited from his forebears, and was
unable to employ effectively, is suggested by the career of the Danish
Cnut, who after 1016 utilised English wealth and military resources to
create an empire which embraced not only parts of Britain but Denmark,
Norway and perhaps some of Sweden too. Like Edgar, Cnut copied the
German emperors, for there is a contemporary drawing of him in which
he is being given a crown which, with its arched bar, was a deliberate
imitation of the German imperial crown, and his daughter married the
future Emperor Henry III of Germany, who was eventually to request the
help of the English navy against one of his enemies, Count Baldwin V of
Flanders. In the same decade, the 1040s, King Swegen Estrithson of
Denmark also asked Edward the Confessor for the assistance of English
ships. If the late Anglo-Saxon military system was a farce, Swegen and
Henry III had not heard of it.3 Could it be then, that the defeats of the
period were not the inevitable result of an organisation which was
inherently chaotic, but accidental outcomes of personality and
circumstance of the sort which can bring down even the most well-
organised of military forces (for war is a risky business), and that it is to
such organisation, as well as good leadership, that one should attribute
the successes of Alfred, Edward the Elder, Æthelstan, Edgar and Cnut?
That this must to an extent be the truth can be powerfully suggested by
momentarily turning aside from military matters to what is known about
the late Anglo-Saxon period generally, and late Anglo-Saxon government
in particular. In the last thirty years this subject has been revolutionised
by the work of Professor James Campbell.4 He sees late Anglo-Saxon
1
See Walker, ‘A Context for ‘Brunanburh’?’.
2
JW, ii. 422-5. John’s source for this event may have been a tenth-century poem, Campbell, ‘Asser’s
Life of Alfred’, p. 138.
3
Hollister, Anglo-Saxon Military Institutions, pp. 125-6.
4
See P. Wormald, ‘James Campbell as Historian’, including the judgement (p. xxi) that ‘Campbell’s
importance lies... in how he has taught historians of early times to think’.

105
 THE BATTLE OF HASTINGS

England as economically developed and therefore effulgent with the

wealth which proved such a powerful attraction to invaders; the margin
between the king’s authority and that of the great lords may have been
significantly greater than it was on the continent; so may the size of the
political community (the number of people whose opinions in some sense
mattered politically); government was not a remote force which impinged
seldom and gently on the English people, but omnipresent in their lives;
the division of the country into shires and their sub-units the hundreds
ensured that the king’s will could be made known and implemented at a
very local level, the hundreds being used for the exercise of many
functions, including the fiscal, judicial and military; these organisations
probably ran alongside an extensive network of officials, the reeves, who
profited from doing the king’s business, and who seem in some respects
to have been able to exercise a degree of control over people’s lives
which is somewhat startling; if so, this is because we are always prone to
underestimate the capacities and achievements of so-called primitive
systems of government.1 Campbell’s work has been complemented by
that of numismatists,2 who have shown that late in his reign Edgar
introduced a coinage system of great complexity. This involved, among
other things, issuing a new coin type every six years or so (after 1035
every two or three), probably so that the authorities could make a
substantial profit by insisting on a favourable rate of exchange of old
coins for new. One might think that the population would simply have
ignored all this and continued using old coins, but the evidence is that
they did not, presumably because there was a date beyond which the
previous type was no longer legal tender, and all significant purchases
were witnessed by the king’s officials. When tenth-century laws state that
all trading must take place in towns before witnesses it is virtually certain
that this was not legislation which was amusingly ambitious and
ineffective, but deadly serious and strictly enforced. It is known what
strict enforcement might involve. Confiscation of all one’s possessions is
a common penalty for wrongdoing in this period, and there are several
examples of kings who ravaged parts of their own territory because the
inhabitants had incurred their displeasure. When the people of
Worcestershire killed two tax collectors Cnut’s son Harthacnut
despatched an army with instructions to burn Worcester itself, devastate
the entire shire and kill all the men. These orders were partly carried out.
Worcester was burnt and the shire (or at least some of it) ravaged, but few
1
See especially, Campbell, ‘Observations on English Government from the Tenth to the Twelfth
Century’; ‘Some Agents and Agencies of the Late Anglo-Saxon State’; ‘The Significance of the Anglo-
Norman State in the Administrative History of Western Europe’; ‘Was it Infancy in England? Some
Questions of Comparison’; ‘The Late Anglo-Saxon State: A Maximum View’.
2
The literature is very extensive, but for an introduction see Stewart, ‘Coinage and recoinage after
Edgar’s reform’.

106
 4 THE ENGLISH ARMY

were killed: they had fled in all directions; if late Anglo-Saxon

government was a farce, they had not heard of it. On the fifth day peace
was restored, the royal anger appeased and the ravagers returned home,
loaded with booty and doubtless glowing with virtue at their
implementation of the king’s wishes.1 Given, then, the determination of
kings to govern, and the evidence that they often did so very effectively
indeed, and given also the need for defence against external enemies,
which as noted earlier was very frequent, and must have seemed of
overriding importance, what one might expect to find is a military system
which was well-organised, well-equipped and a reflection of the
intelligence and power to be found in other aspects of the late Anglo-
Saxon state. Of course, these arguments are circumstantial: is there any
evidence which will put flesh on their bones?
It is likely that some of the administrative features of the late Old
English period had their origins in a remote past.2 Certainly, remarkable
feats of government were implemented before the time with which this
study is mainly concerned. In the late eighth century, for example, the
Mercian king Offa constructed a rampart of considerable length on the
Welsh border which may have availed itself of the services of ‘very many
thousands of diggers’3 by the imposition of labour service upon his
subject peoples. Nevertheless, a treatment of the military capacities of the
army which fought at Hastings can reasonably begin with Alfred, and two
elements of the reforms which succeeded in stopping the seemingly
overwhelming Viking advance in the late ninth century. Firstly, there was
the building, or perhaps renovation in some cases, of an extensive
network of burhs (fortified towns and fortresses) and their manning by
garrisons capable of looking after themselves; this meant that it became
much more difficult for the enemy to overrun areas than it had been
previously, as is evident from the entries in the Anglo-Saxon Chronicle
for the 890s, when the attackers never succeeded in penetrating Wessex
as they had done in the dark days twenty years earlier. Remains of some
of these defences survive above ground, as for example at Wallingford
and Wareham, and they do so together with a document, the Burghal
Hidage, which reveals something of the way the authorities had them
built and maintained. It lists many of the burhs, and says that if one hide
(the Anglo-Saxon unit of fiscal assessment) supplies one man then every
pole (that is about 5½ yards) of wall would be manned or maintained –

1
The harrying of 1041 is noted by ASC C and D (i. 162-3); for most of the details, see JW, ii. 532-3.
2
Campbell, ‘The Late Anglo-Saxon State’, pp. 2-8.
3
Ibid., p. 6. See Hill, ‘Offa’s Dyke: Pattern and Purpose’, who questions whether construction ran all
the way along the border from sea to sea, as claimed a century later by Bishop Asser, the biographer of
King Alfred.

107
 THE BATTLE OF HASTINGS

there are difficulties about the precise meaning of the word used1 – by
four men. It is possible using this formula and the other figures given to
work out the size of the individual burhs mentioned in the document, and
in a number of cases the result corresponds closely to the extent of the
surviving defences of the places concerned. It is also possible to calculate
the total number of men involved in their maintenance or defence -
27,671.2 And this, of course, as Alfred had other men who were not
involved in defending or maintaining burhs, immediately suggests a scale
of activity involving tens of thousands, perhaps, dare one say it, many
tens of thousands of men, which might seem surprising, but which other
evidence does not entirely fail to confirm.
Alfred is also well known for his interest in the West Saxon navy. It is
by no means clear that he founded it, or that the measures he took were
particularly successful, but there is the explicit and contemporary
statement in the Anglo-Saxon Chronicle that he was responsible for a new
type of ship, ‘neither of the Frisian design nor of the Danish, but as it
seemed to himself that they might be most useful’. Some had sixty oars,
some more.3 Now craft of thirty oars a side may have been very large
indeed, and are entirely absent from the archaeological record of the
period, although this is far from being extensive enough to provide any
substantial reason for thinking that they could not have existed. N.A.M.
Rodger has accepted the possibility that these vessels had crews of two to
three hundred men, being ‘a massive application of money and
manpower’ and among the largest ‘built in northern waters during the
Middle Ages’.4 Possibly so, but it does not necessarily follow that they
were particularly rare. In the reign of Æthelred II, Bishop Ælfwold of
Crediton bequeathed to the king a ship of sixty-four oars, ‘all ready
except for the rowlocks’.5 If it were common for sources from the period
to give precise information on the sizes of warships, and if such
information usually referred to smaller craft, it would be reasonable to
regard sixty-oared vessels as exceptional. But as such references are in
1
This word, the Old English waru, can refer either to physical defence or the discharge of fiscal
obligations. Dr Rumble (The Defence of Wessex, ed. Hill and Rumble, pp. 178-81) keeps the options
open by concluding that the formula refers to the defence of wall ‘in a military and/or a fiscal sense’;
Campbell, ‘Stenton’s Anglo-Saxon England’, p. 272, thinks that the similarity between the formula and
the Domesday Book entry on Chester (DB, i. 262d), which says that the city wall and bridge were
repaired by one man from every hide in the shire, makes it ‘nearly certain that what is in question is
maintenance’.
2
Davis, ‘Alfred the Great: Propaganda and Truth’, pp. 42-3. This seminal article is the foundation of
the best modern scholarship on Alfred.
3
ASC, i. 90.
4
Rodger, The Safeguard of the Sea, pp. 15-17; see also his very useful, ‘Cnut’s Geld and the Size of
Danish Ships’, passim. However, Rodger does not consider the possibility that ‘so large a number of
oarsmen (sixty or more) might have been accommodated in some manner other than simple alignment’,
for example with the oars double-banked; Swanton, ‘King Alfred’s ships’, 13.
5
EHD1, No. 122; Rodger, ‘Cnut’s Geld’, 400, suggests, not very plausibly in my view given the
Alfredian evidence, that some of the oars may have been spares.

108
 4 THE ENGLISH ARMY

fact very uncommon one is left wondering. Could it be that there were
times when great battleships crewed by hundreds of men, the
Dreadnoughts of their day, were by no means rare, and that this was not
unconnected with the desire of King Swegen of Denmark and Emperor
Henry III to seek English naval assistance in the 1040s?1
Alfred’s measures included the use of troops who travelled on
horseback so that they might more easily pursue Viking forces which did
likewise, and further evidence of activity on a large scale is to be found in
a law code of King Æthelstan which demands, in a provision which may
relate specifically to the obligations of the inhabitants of burhs, that he be
supplied with two well-horsed men from every plough. A ‘plough’ in this
context may be an assessment unit rather than an actual object – in other
words, however many ploughs a landowner actually had, it was not
necessarily the same as the notional figure at which the government
assessed him for tax, and which was presumably based on the perceived
wealth of the land concerned. Domesday Book, compiled on the
Conqueror’s orders in 1086, mentions both ploughlands and ploughteams,
although the results of associating Æthelstan’s order with either of them
are fairly staggering, for by the 1086 figures he would have been
demanding at least 120,000 mounted men from the former and 160,000
from the latter.2 Yet if association with these units is rejected, as it
seemingly must be, one would still think that there must have been many
‘ploughs’ in his kingdom, and even if not all were related to the military
obligations of burhs the likelihood that he was asking for many thousands
of well-horsed troops seems high. Moreover, that reliable sources such as
the Burghal Hidage and Æthelstan’s law code suggest large figures makes
one the more inclined to believe such totals in sources of a more dubious
nature. In the second decade of the eleventh century the German bishop
Thietmar of Merseburg recorded a report that when London was under
siege by the Danes in 1016 it contained 24,000 coats of mail. 3 Rather
significantly, he thought this incredible, but in the context of other very
sizeable totals it may not be.

1
See further on the Anglo-Saxon navy, Hooper, ‘Some Observations on the Navy in Late Anglo-Saxon
England’; Strickland, ‘Military Technology and Conquest: The Anomaly of Anglo-Saxon-England’,
373-8, with the comment that it was ‘the arm to which the Anglo-Saxons themselves attached great, if
not supreme significance’; Rodger, The Safeguard of the Sea, pp. 14-30. There were, of course, smaller
vessels too. DB, i. 1a, refers to the supply to the king of 20 ships for 15 days a year, each of them
containing 21 homines. Even if the latter were soldiers additional to the rowers they seem unlikely to
have been in very large craft. I owe this reference to James Campbell.
2
II Æthelstan 16, Die Gesetze der Angelsachsen, ed. Liebermann, i. 158; Darby, Domesday England,
pp. 95-136, 336. Abels, Lordship and Military Obligation, p. 111. See also the note in The Laws of the
Earliest English Kings, ed. Attenborough, p. 208: ‘the requirement stated here is exceptionally heavy;
but the explanation probably is that cap. 13-18 seem to have been intended for burgware, who may be
regarded as primarily a military caste’.
3
Thietmar, Chronicon, ed. Holtzmann, pp. 446-7; trans. EHD1, p. 348.

109
 THE BATTLE OF HASTINGS

If one turns from the administrative evidence mentioned so far to

accounts of actual campaigns (which are considerably more plentiful than
those of battles), it is to be expected that the apparently great capacities
hinted at in the one will also be found reflected in the other, and this is
indeed so. Take, for example, the events of 917,1 when Edward the Elder
was in the process of reconquering the eastern midlands, partly by the
building of burhs in enemy territory, from the Scandinavians who had
settled there in earlier decades. In that year he ordered such works to be
constructed at Towcester and Wigingamere, both of which then had to
withstand sieges until in the case of the former reinforcements arrived. In
the summer, a great force assembled from the men of nearby burhs and
took by assault the Viking fortress of Tempsford, while in the autumn
Colchester was captured by a great army consisting of garrisons and men
from Kent, Surrey and Essex; at much the same time the enemy made an
unsuccessful attack on Maldon in Essex, suffering ‘many hundreds’ of
dead in fighting its reinforced garrison. The king himself then led a West
Saxon army (thus far evidently inactive) to Passenham while Towcester
was fortified in stone, and once these men had gone home another force
assembled for the capture of Huntingdon. Perhaps it was the same army
which then restored the defences of Colchester. In the same year
Edward’s sister Æthelflæd, Lady of the Mercians, captured the Viking
stronghold of Derby. Even more impressive in its way is the effort which
Edmund Ironside, the son of Æthelred II, made against Cnut in 1016.
Although it came at the end of a long period of military failure, in which
the West Saxons themselves had been prepared to abandon their native
dynasty and accept a Danish king, Edmund was able to call up all the
people of England five times,2 and at least three major battles were fought
during the year, at Penselwood (probably in May or early June), Sherston
(25 June) and Assandun (18 October), with possibly a fourth near the
Forest of Dean before Edmund made peace and eventually died (30
November).3 This capacity for engaging in repeated conflict is
reminiscent of the fighting of 871, when the Anglo-Saxon Chronicle
claims that there were nine engagements between the West Saxons and
the Vikings; of these, the victory of Æthelred I and his brother Alfred at
Ashdown followed a mere four days after they had suffered heavy losses
at Reading, and was itself followed only two weeks later by a defeat at
Basing.4

1
ASC, i. 101-3. I use the revised date for this entry of EHD1, p. 214.
2
ASC, i. 151.
3
On the Forest of Dean battle and the dates of Sherston and Assandun, see Lawson, Cnut, pp. 78-9,
143.
4
ASC, i. 71-2.

110
 4 THE ENGLISH ARMY

What emerges from such material, which of course must bring to mind
the three battles of 20 September to 14 October, 1066, is that pre-
Conquest English military systems, far from being easily exhausted, were
capable of producing battleworthy forces over an extended period, and
despite loss and defeat. Of course, it might reasonably be objected that
almost 200 years separated 871 and 1066, and that even the events of
1016 do not necessarily prove anything about the military resources
available to Harold II fifty years later.1 This is true, but those fifty years
were far from uneventful, and such evidence as there is does not suggest a
marked decline in the effectiveness of English forces. Cnut used them in
Scandinavia at the battle of Holy River c.1026 (perhaps a defeat in the
midst of a successful campaign to repulse a Norwegian and Swedish
attack on his position in Denmark), in conquering Norway in 1028, and
doubtless also in securing the submission of other princes in the British
Isles.2 The reign of Edward the Confessor saw considerable fighting
against the Welsh. In 1052 Gruffydd of North Wales raided Herefordshire
and defeated local levies near Leominster; three years later he and Irish
forces assisted the return of the exiled Earl Ælfgar of Mercia, overcoming
local troops who had been made to fight on horseback under the king’s
(French) nephew Earl Ralph, and sacking Hereford. Earl Harold then
gathered men from almost all England who provided a screen while he
built an earthwork around the town, there was a reconciliation with
Ælfgar and peace was made. Nevertheless, Gruffydd remained a problem,
not least because he henceforth controlled all the Welsh. The following
year he defeated and killed Bishop Leofgar of Hereford and in 1058
assisted Ælfgar in a second return from exile. But in 1063, after an
interval about which little is known, he finally met his match. Harold
advanced to Rhuddlan and burnt the palace there, subsequently sailing
with a fleet from Bristol in a campaign coordinated with an army led by
his brother Earl Tostig; this was too much for the Welsh, and on 5 August
Gruffydd was killed by his own men and his head brought to Harold.
Whether or not this was the most successful invasion of Wales since the
Roman period and ‘the rebirth of England as a military power’,3 it
certainly suggests that there was little the matter with the country’s
military system, and this campaign was not the only successful venture of
Edward’s reign. In 1054 Earl Siward of Northumbria had invaded
Scotland with a navy and an army which included some of the king’s

1
Abels, ‘From Alfred to Harold II’, argues that ‘English military institutions eroded between the death
of Alfred in 899 and the Battle of Hastings’ and that Harold ‘possessed a military system that was
seriously flawed’ (pp. 16, 29).
2
Lawson, Cnut, pp. 96-101, 104-7.
3
Barlow, pp. 211-2, and more generally on Edward and Wales, pp. 204-13; also Barlow, The Godwins,
pp. 68-9. The main primary sources are the C, D and E versions of ASC, i. 176-91; JW, ii. 592-3; LKE,
pp. 86-7.

111
 THE BATTLE OF HASTINGS

housecarls, and defeated King Macbeth in a battle in which both sides

suffered heavy losses; having installed Malcolm as king of Cumbria and
Lothian, he returned home laden with booty. 1 Nor should the general
threat from Scandinavia be forgotten, and in particular the hostile fleet
which came from Norway in 1058.2 England was no safer in the
Confessor’s reign than in the previous tumultuous two and a half
centuries, and it is not probable that those responsible for its defences
were unaware of this, especially once it became clear that King Edward’s
death was likely to be followed by foreign invasion. Thus, the Anglo-
Saxon Chronicle C text’s statement that in 1066 Harold gathered greater
land and naval forces than had ever been seen before is likely to be an
accurate enough reflection of the scale of the preparations made to
receive William.
Of course, the size and nature of the army which fought at Hastings is
another matter. The battles of Fulford and Stamford Bridge had taken
place very recently and had involved losses, although the known ability
of English kings to assemble forces at short notice should give pause to
those who would assume that it must on that account have been
inadequate, and it should not be forgotten that most of the sources agree
that fighting was hard and lasted all day. Yet any attempt to be more
precise than this is fraught with difficulty, as will now become apparent.
None of the eleventh and twelfth century writers give any sort of
plausible figure for Harold’s army.3 Most of those from the English side
tend to minimise its size, from the Chronicle E text’s mild statement that
he fought before it had all gathered to what look like the deliberate
attempts of John of Worcester and William of Malmesbury to reduce the
impact of defeat by suggesting that their countrymen stood little chance
from the first.4 The exception to this is the Chronicle D version’s
statement that it was a large force, a view with which the Carmen,
William of Jumièges and William of Poitiers all concur. Naturally, the
French writers can be accused of wishing to exaggerate the number of
Duke William’s enemies in order to magnify his victory, and even if the
idea that their force was ‘large’ were to be accepted there is no way of
knowing what the word meant in eleventh-century terms. Yet none of this
has prevented historians of the battle from advancing opinions on the size
of the English army which have often worn a rather greater appearance of
certainty than the evidence actually warrants.

1
ASC, C and D, i. 184-5; Barlow, pp. 202-3. Malcolm showed little gratitude, ravaging Northumbria in
1061 and assisting Harald Hardrada in 1066; on housecarls, see below, pp.129-30..
2
Above, pp. 35-6.
3
The Carmen, ll. 223-4, gives twelve hundred thousand, which FNC, iii. 741, thought ‘the whole
military population of England’; Wace, RR, ll. 7861-3, has Harold claim to have 400,000 men.
4
Above, pp. 51-3.

112
 4 THE ENGLISH ARMY

Freeman's map of the dispositions of the two armies, FNC, iii. opp. 433. Not to the
original scale.
The most detailed scholarly treatment of Hastings remains that of
Freeman's Norman Conquest. Freeman enjoyed visiting historical sites,
and travelled to Battle from his home in Somerset on a number of
occasions. On 2 June, 1869 he walked over the ground in the company of
Captain Edward R. James of the Royal Engineers, whose father Henry
had been director-general of the Ordnance Survey since 1854, where he
pioneered the reproduction of maps by photozincography. Thus it was
that the publication of Freeman's third volume was delayed so that it
could include a zincographed fold-out map of the battlefield produced by
the Ordnance Survey; it showed the English army drawn up along most of
the Battle Abbey ridge in forty-nine individual units, with housecarls in
the centre and ‘light-armed’ on the wings; to their front was a palisade,
and there was an ‘outpost’ (which may have owed something to
awareness of Wellington’s use of Hougoumont and La Haye Sainte at
Waterloo) in what is now the area of Horselodge Plantation, on the south-
western face of the ridge.1 Freeman relied on Captain James to put the
extent and arrangement of the palisade into a ‘scientific military shape’,
and warned in his preface that:
1
On the topography of the battlefield, see above, pp. 47-9 and Illus. 8-73. On Freeman's map ‘b b b’
represents the palisade and ‘c c’ the outpost; he commented on the latter, FNC, iii. 444, 477.

113
 THE BATTLE OF HASTINGS

The relative position of the different divisions in the two armies seems
beyond doubt, but the extent of ground occupied by each division must be
matter of pure conjecture. The one absolutely certain point is the position
of the English Standard and the fact that it was against that point that the
1
main attack under William himself was made.
No one could quarrel with the final sentence, but elsewhere its author
spoke of an ‘immoveable wedge of men which, as if fixed to the ground
by nature, covered… every inch’ of the ridge, and of ‘ranks of men...
ranged... closely together in the thick array of the shield-wall’.2 Prudently,
he gave no estimate of the numbers involved,3 but there was an
inconsistency between some of these comments and the map, whose neat
little units bore no obvious relationship to an extended, densely-packed
line. Freeman died of smallpox in Spain on 16 March, 1892; only a few
months later his reputation, and in particular his account of the battle of
Hastings, had come under attack (anonymously) from John Horace
Round; within twenty years a very different view of the extent of the
English deployment on the ridge, and therefore of the scale of the battle,
had occupied a position from which it has not since been dislodged.
Indeed, so unassailable has this position apparently been deemed that
there has been no attempt to dislodge it.
The fact that Round never essayed a full and coherent account of the
fight about which he was eventually to write a great deal did not inhibit
some trenchant comment on what he took to be errors in that of Freeman,
whose reputation for historical accuracy, along with that of the acclaimed
Norman Conquest, he sought to destroy. Thus, he ridiculed both the
adoption of the name Senlac for the battle as the ‘truth is simply that the
site of the battle had no name at all’, and the idea that the English front
was protected by a palisade, based on a supposed mistranslation of a poor
authority, Wace’s Roman de Rou;4 further, he questioned the belief that
the English had formed up with the heaviest troops in the centre and the
lighter armed on the wings, as it would have been foolish of Harold to
place such forces on his right flank, which Freeman had himself
acknowledged to be the weakest point in his position. In Round’s
judgment, the most reliable native force, the shield-wall of ‘heavy
soldiery, clad in helmets and mail’, would have occupied their entire
front, with the ‘half-armed peasants’ behind it, able to throw missiles
over the heads of their colleagues and back them up in repelling an
attack.5 There followed a bitter controversy between Round and those

1
Ibid., iii. vi-vii. Many years later James disclaimed all responsibility for the way in which the details
of the battle had been shown, ‘The Battle of Hastings’, 19.
2
FNC, iii. 471.
3
Ibid., iii. 740-2.
4
RR, ll. 7793-804.
5
Round, anonymous Quarterly Review article of July, 1892.

114
 4 THE ENGLISH ARMY

who took up the cudgels on Freeman’s behalf, T.A. Archer and Kate
Norgate, with the latter eventually accusing her opponent of acting
dishonourably in attacking from ‘the shelter of his anonymity’ a ‘dead
man who had given him no provocation’.1
The period of this debate also saw the publication of the German
scholar Wilhelm Spatz’s small book Die Schlacht von Hastings. This
rejected both the notion that the English had employed such a difficult
formation as a shield-wall and the claim that the French had used a
feigned retreat, resulting, in Round’s words, in the belief ‘that both
armies were little better than armed mobs... incapable of the simplest
formation or manoeuvre’.2 But Spatz also attempted to arrive at estimates
of the sizes of both forces, and argued that as Harold had in the battle
only the military strength of southern England, and as he would probably
have attacked the French and not been defeated so heavily if his army had
been larger than theirs, he therefore probably commanded no more than
6,000-7,000 men, the number which Spatz believed William brought
against him.3 Only two years later Sir James Ramsay’s The Foundations
of England appeared. This concluded that 5,000 Frenchmen ‘would
satisfy our ideas of what was at all likely: 10,000 men we should consider
beyond credibility’, and although he gave no estimate for the English he
accepted that Harold fought before all his troops had arrived. He also
accepted (from William of Poitiers) that they were so densely massed that
the dead could not fall, and judged that this ‘points to a formation many
ranks deep’; referring too to John of Worcester on the narrowness of the
site, and alleging that the Battle Abbey Brevis Relatio states that the
Normans found the English in illo spisso agmine quod erat ante eos in
montis summitate (‘in that very dense formation which was before them

1
Norgate, ‘The Battle of Hastings. Part II’, 76. See further Round, anonymous Quarterly Review article
of July, 1893; ‘Wace and his Authorities’; and ‘Mr Freeman and the Battle of Hastings’; Archer, ‘Mr
Freeman and the “Quarterly Review”’; ‘The Battle of Hastings. Part I’. These are only the main
references in a controversy which lasted six years; a list of all the significant material produced
between 1892 and 1898 is given in Round, ‘The Battle of Hastings’, 63. He had little difficulty in
convicting Freeman of factual error (for he did not confine his assault to the battle), but emerged from
the fray by no means unscathed, or innocent of error, himself; he did not respond, for example, to Miss
Norgate’s rejection of his views on ‘Senlac’, and was eventually refuted by W.H. Stevenson (above,
pp. 49-50); his stance on Freeman’s supposed mistranslation of Wace was damaged when S.R.
Gardiner, editor of the English Historical Review, produced letters from two ‘eminent French
authorities’ supporting Freeman’s belief that Wace’s escuz de fenestres e d’altres fuz indicates ‘bien un
rempart, une sorte de palissade’; EHR, ix (1894), 260; Burgess has recently rendered the passage (RR,
ll. 7793-7): ‘they had made shields for themselves out of shutters and other pieces of wood. They had
them raised before them like hurdles, joined closely together; from them they had made a barrier in
front of them’.
2
Round, ‘The Battle of Hastings’, 61; to a French audience he presented Spatz’s rejection of the shield-
wall as ‘un exemple extrême de cette tendance à bâtir des théories en dehors des faits, à laquelle l’esprit
allemand est souvent trop enclin’, ‘La bataille de Hastings’, 70.
3
Spatz, Die Schlacht von Hastings, pp. 33-4; for Spatz and the French army, see below, pp. 143-4.

115
 THE BATTLE OF HASTINGS

on the crest of the ridge’),1 Ramsay concluded that their line must have
formed three sides of a square around that crest, with the rear ‘being left
open as sufficiently protected by the ravines, and in fact unassailable.2

Detail from Ramsay’s map of the dispositions of the two armies, The Foundations of
England, ii. opp. 24. Not to the original scale and with the key superimposed.

In making the perfectly reasonable suggestion that the English left

flank was turned back against the steep northern face of the ridge Ramsay
may have recalled Round’s comment that the line ‘might, for all we
know, have formed a crescent or semicircle, its wings resting strongly on
the rear-slopes of the hill’.3 It is not clear from this just where Round
thought the English right could have been positioned, but the map which
accompanied Ramsay’s book placed it roughly at the southern end of
what is now Battle High Street. There is, however, nothing in this area
that could reasonably be described as a ‘ravine’. Round had judged
Freeman’s map of the battlefield in need of substantial revision, and
Ramsay’s version was quickly joined by another, that of F.H. Baring.
1
This was rather misleading, as BR in fact simply records a report that Harold’s standard was to be
seen there; above, p. 94.
2
Ramsay, The Foundations of England, ii. 16-17, 22-6.
3
Round, anonymous Quarterly Review article of 1893, 101; Feudal England, p. 366, n. 100. Round
thought that Ramsay had made a ‘fairly strong case’ for his view that the English stood wholly ‘on the
little “plateau” of the summit of the hill’ and were not more than 5,000 strong, without wishing to
‘assert that the Normans were so few as this’, ‘The Battle of Hastings’, 62.

116
 4 THE ENGLISH ARMY

Detail from Baring's map of the disposition of the two armies, Domesday Tables,
endpaper. Not to the original scale and with an abbreviated key superimposed.

Baring concurred with Ramsay in refusing to join Freeman in allowing

the English a frontage along the entire length of the ridge, because
Charles Oman had already suggested that if one combined with such a
frontage Round’s belief that their formation consisted of a continuous
shield-wall then such a formation, if ten or twelve ranks deep, could have
contained 25,000 men.1 Baring thought that this was far more than
seemed likely, and therefore proposed a shorter front (if not quite as short
as Ramsay’s), arguing that as we are told that they were densely
compressed this cannot have been true of a line of the length that
Freeman had suggested, and that the English chronicles suggest an army
that was not large. He also agreed with Ramsay and Spatz that the French
army was not large (the figures given by William of Poitiers being ‘not to
be taken as arithmetic at all’), and thus arrived at an estimate of 10,000 to
13,000 for the English, ‘many of them... rustics’. This allowed each man
a frontage of two feet on a line 650-800 yards long, with ranks ten to
twelve deep. The French he guessed at perhaps 8,000 to 10,000, while

1
Oman, A History of the Art of War: The Middle Ages, pp. 154-5; he also estimated that there were
2,000 to 3,000 housecarls around the standard.

117
 THE BATTLE OF HASTINGS

considering Ramsay’s estimate of 5,000 ‘rather small, but not

impossible.1

Baring's map superimposed on the Millenium Map.

Baring reproduced a map made at a scale of 6 inches to the mile in

1907 for Major-General Edward R. James, who as a captain had walked
the field with Freeman nearly forty years before; this was naturally better
than the earlier map (which had used a scale of 4 inches to the mile and
much less informative contouring) on the relief, and of course showed the
smaller English frontage.2 Like Ramsay, Baring thought that its left did
not extend fully eastwards along the ridge, where the slope is less severe,
but was turned back so that it was protected from envelopment by the
steep gradient to the north; the English right he placed a little further west
than Ramsay, but agreed in showing it bent back towards (but some
distance from) what is now Battle High Street. However, he knew that
Ramsay was mistaken in positing the existence of a ravine in this area,
and commented instead that:

1
‘The Battle Field of Hastings’, 65-70; ‘On the Battle of Hastings’, pp. 217-20. But in n. 2 to p. 219 he
added on the English: ‘I incline to a position 600 yards long, rather than 800, and to 10,000 men or
less, rather than 12,000 or 13,000’.
2
The map, which showed the English deployment upon which both evidently agreed, was first
published in James’ article ‘The Battle of Hastings’, in which he acknowledged ‘much kind assistance’
from Baring. Comparison with modern OS maps suggests that it is not completely accurate, pushing
the lower reaches of the stream which runs west out of New Pond, for example, rather too far to the
south. .

118
 4 THE ENGLISH ARMY

a brook runs close behind the ridge and at 400 yards from the church the
bank at the back, though not so formidable as it is further west, is already
20 feet high; it has too a wide trench at its foot, cut by the head of the little
stream, so that there is quite enough to prevent any effective attack from
1
the rear at this point by the Norman horse.
As will become evident shortly, this was not particularly convincing.
Even so, Baring’s basic guide to the extent of the English deployment
was to carry all before it, being apparently adopted in Sir Frank Stenton’s
influential and widely-read Anglo-Saxon England in 1943, and featuring
in many other treatments of the battle down to the present day, including
R. Allen Brown’s article from 1980, which reproduced his map with little
comment.2 It is, of course, a concomitant of this that the relatively small
numbers first advanced by Spatz have also become part of the staple diet
of those who read about the battle.3 Yet that this should be so is not
because of any overwhelming evidence in their favour, for the truth, at
least as far as the English army is concerned, is that its size is simply not

1
‘On the Battle of Hastings’, p. 218.
2
Stenton, Anglo-Saxon England, pp. 594-5. With typical caution Stenton described Baring’s frontage
and referred to his map without specifically accepting them, having already warned, in words
reminiscent of Freeman’s, that the only certainty about the English dispositions was the location of
Harold and his standard; Brown, ‘The Battle of Hastings’, 4-5, 10-11. Lemmon, ‘The Campaign of
1066’, p. 96, shows an English front about 900 yards long, but with a western flank not bent back
towards the ‘brook’: ‘to refuse the flanks at all was really unnecessary because medieval attacks on a
flank were usually frontal’ (p. 100); similarly, Beeler, Warfare in England 1066-1189, p. 18. Fuller,
The Decisive Battles of the Western World, i. 377, opted for a slight curve, turning the English right
back to the ‘brook’, while there are varying degrees of curve in Morillo, Warfare under the Anglo-
Norman Kings, p. 163, and The Battle of Hastings: Sources and Interpretations, ed. Morillo, pp. xxiii-
xxx; Oman, who had once accepted Freeman’s long line, eventually went further than Ramsay and
Baring in showing the English position as almost a semicircle, with the western flank extending over
the ‘brook’, A History of the Art of War in the Middle Ages, i. opp. 164. Bennett’s maps (Campaigns of
the Norman Conquest, pp. 38, 42) do not trouble overmuch with the streams, as the flanks of the short
English front are shown resting upon two strategically-placed patches of woodland for which there is
no evidence.
3
Oman, who originally guessed 25,000 English and 10,000-12,000 French horse and 15,000-20,000
French foot, eventually spoke cautiously of ‘many thousands’ of English and perhaps 12,000 French, A
History of the Art of War, pp. 154-6, and (2nd edn.) i. 156-8; most other estimates have been closer to
Spatz’s roughly 7,000 a side, Der Schlacht von Hastings, pp. 26-34. Stenton, Anglo-Saxon England, p.
593, wrote of 7,000 English and rather fewer French; similarly his earlier William the Conqueror, p.
196, and Douglas, William the Conqueror, pp. 198-9. Burne, The Battlefields of England, pp. 37-8,
suggested nearly 9,000 Normans and ‘the English about the same at the start, with considerable
reinforcements flowing in during... the battle’. Lemmon, The Field of Hastings, p. 24, averaged eleven
earlier estimates to arrive at 8,800 English and 8,000 French; similarly, his ‘The Campaign of 1066’, p.
102, followed by Beeler, Warfare in England, pp. 16-17. Morillo, Anglo-Norman Warfare, p. 58,
thinks the ‘best estimates of army size are between 5000 and 7000’; Delbrück, History of the Art of
War, iii. 152, guessed 7,000 French and 4,000-7,000 English; Lot, L’Art Militaire, i. 284-5, announced
that the English could not have been more than 8,000 to 9,000 strong, but probably did not reach half
that, while the French numbered 7,000. Barlow, The Godwins, p. 101, thinks that for the French
‘20,000 men and 500 ships would seem to be the top limit; ten thousand effectives is probably closer to
the mark’. Brown, The Normans and the Norman Conquest, p. 150, n. 47, warned that figures of about
7,000 a side ‘are more or less rational guesswork’. The belief that medieval armies were almost always
small, widely-held since the days of Delbrück (to whom Spatz’s book was dedicated) and Lot, remains
so, see Verbruggen, The Art of Warfare in Western Europe, pp. 5-9. On the size of the French army,
see further below, pp. 143-51.

119
 THE BATTLE OF HASTINGS

known; consequently there is a wide range of possibilities, and the tens

of thousands of men who lie in the outer reaches of that range may in fact
be no less plausible a figure than the few thousand who inhabit the inner
reach. To the possibility of tens of thousands, spread over a considerably
greater area than the mere crest of the ridge, indeed to something not very
unlike Freeman’s position, it is now necessary to turn.
Firstly, it must be noted that the location of the English right wing is
not quite so straightforward as a casual reading of Baring’s comments
might suggest. The phrase ‘the bank at the back... already 20 feet high’,
which he used to describe the area where that wing rested, implies an
almost vertical drop when in fact there is no such feature in this vicinity,
as is evident from the ten-foot contour lines on his map.1 Moreover, the
‘head of the little stream’ actually contains very little water. The 6 inch
Ordnance Survey sheet to which he referred,2 and current maps, although
they do have a stream rising on the northern edge of this part of the ridge,
place its source rather further from the abbey gatehouse than did Baring,3
and examination on the ground shows that it is in fact not here a stream at
all, but what would be better described as a drainage channel; as it is so
near the top of the ridge it is virtually impossible for it to contain any
substantial body of flowing water, and even in the very wet autumn of
2000 it did not do so, producing little more than a trickle; 4 if there was a
significant obstacle to French cavalry seeking to outflank the English in
this area, the ‘head of the little stream’ can hardly have been it.
It is very likely from what is known of the battle that Harold did
occupy a position which William found difficult to turn, and that this is
why the conflict seems to have consisted of repeated frontal attacks; it is
thus not unreasonable to speculate on where his flanks may have lain, and
it might be said that as far as can be told from the existing ground there
seems a rather better position than the one selected by Baring and widely-
accepted ever since, which one could think rather too easy to simply ride
around, despite the undeniable gradient; it is to be found further south, at
the western end of the ridge, where there is a slope due west of just over 1
in 8, and the ground falls into the area of Saxon Wood. It also happens to
be Freeman's position. It is by no means impossible that in 1066, as
today, there was dense vegetation in this area, where the water to nourish
it collects and drains from the Wadhurst Clay, and that this prevented the
1
In his 1905 article, ‘The Battle Field of Hastings’, 66, Baring had stated that ‘the bank at the back,
though not so formidable as it is further on, is already over thirty feet high’ (my italics).
2
No. 57 N.E.; Baring, ‘On the Battle of Hastings’, p. 217, n. 2.
3
On the OS Explorer Sheet 124, the stream is shown rising just below the 65 metre contour line, to the
top left of the ‘B’ of ‘Battle Abbey’, and just to the west of Long Plantation.
4
See Illus. 14-15. It is noteworthy that James’ article on the battle argues for the same deployment,
which had been ‘the subject of the most careful consideration on the actual ground by Mr. Baring and
myself’, but says nothing of any stream, speaking instead of a slope of ‘1 in 8 on the right rear’, ‘The
Battle of Hastings’, 26.

120
 4 THE ENGLISH ARMY

English position being outflanked. However, there is, of course, no

certainty about this, and our ignorance of the precise extent of whatever
woodland or undergrowth there may have been elsewhere obviously
makes it very unsafe to place much weight on such an argument when
attempting to fix the extent of Harold’s deployment.
In any case, there is a better one to hand. The Bayeux Tapestry has a
scene in which the French attack a hillock, prefaced by the inscription
Hic ceciderunt simul Angli et Franci in prelio (‘Here the English and the
French fell at the same time in the battle’). To the left, under the word
simul, is a body of water, and to the left of that a French horseman faces
left, and thus has his back to the sequence which follows; this may well
be the Tapestry designer's way of indicating that the said sequence is
quite distinct from the scene showing the deaths of Harold’s brothers
Gyrth and Leofwine which precedes it. To the right of the water two
horses have been upended in one of this source's most powerful images;
to the right of them a mailed English foot-soldier holding a spear is
pulling down a horseman by seizing his mount's girthstrap, and another
horseman apparently prepares to hurl a spear at the defenders of the
hillock. These consist of two men with neither armour nor any kind of
weapon tumbling down the slope evidently in death, three men with
moustaches who have spears and shields but no helmets or coats of mail,
and a soldier with a beard but no moustache who holds a spear and is
receiving a spear-thrust from a French horseman who has his weapon
couched underarm.1 On the ground at the foot of the hillock there is a
figure equivalent to this last except that he has no beard and is regarding
the enemy with obvious trepidation, and to the left of him appears a man
of a rather different type, without armour, but holding an axe, wearing a
sword and with a very prominent moustache. To the right of the hillock
Bishop Odo is seen comforting the boys and Duke William removes his
helmet, almost certainly to prove that he is still alive, as recorded by
several of the written sources. The fact that Odo is turned away from the
hillock scene may or may not indicate that these events were thought
distinct from it.
This part of the Tapestry is important because it can be located on the
battlefield with a reasonable degree of plausibility, because this has
important consequences for the debate on the extent of the English
deployment, and because elements of it bear upon their fighting methods
in ways that are also of no small significance. Freeman identified the
hillock as the area of Horselodge Plantation, part of the subsidiary ridge
to the south-west of the main one, covered today with bushes and small
trees, and to the north of what is now New Pond; he also suggested that

1
As noted earlier (above, p. 69) the reliability of this element of the scene is suspect.

121
 THE BATTLE OF HASTINGS

Harold stationed troops there from the outset to protect the approach to
the main English position at a point where the gradient is relatively easy.1

The Tapestry watercourse and hillock scene in engravings published by Antoine

Lancelot in 1733 from drawings by Antoine Benoît of 1729. See the Appendix. All the
essential elements of the scene are present in Benoît’s drawing, where the moustaches
are quite distinct.
However, he thought that the water lay beyond the western end of the
ridge in the vicinity of Saxon Wood, and that the Tapestry scene itself
showed men cut off on the hillock during the annihilation of those of the

1
FNC, iii. 444, 477; on this part of the battlefield, see above, pp. 47-8, Illus. 32-3, 36-7.

122
 4 THE ENGLISH ARMY

English who left their positions to pursue the fleeing French, a view
apparently supported by a passage in William of Malmesbury’s Gesta
Regum referring to the occupation of a tumulus at that point in the battle.1
In fact, it seems much more likely that both water and hillock are really
part of the same scene, as Round noted,2 and that the former is not only
that which flows west in the area of New Pond and to the south of
Horselodge Plantation, but may also be identifiable with the ‘sandy
brook’ (if that is what the word Senlac means)3 which according to
Orderic Vitalis gave the battle its name; it is, at least, as is especially
noticeable today west of New Pond, distinctly sandy. Furthermore, it is
by no means certain that the men shown on the hillock did end up there
during the wreck of the pursuit, for while some can undoubtedly be
described as ‘half-armed peasants’4 the distinctly well-provided men with
moustaches look far more like a specialist group of light infantry, and the
moustaches themselves ‘as if some special significance attached to
them’.5 Perhaps this is a ‘picked company’ of the sort mentioned in the
Anglo-Saxon Chronicle poem on King Æthelstan’s great victory at
Brunanburh in 937.6 If so, then their light equipment, so suited to the
broken ground upon which they are found,7 suggests that they were
indeed deployed there from the beginning of the battle, an idea which
would be powerfully reinforced if Sir David Wilson is correct in thinking
that the serrated shapes above the water to the left represent ‘a defensive
work of sharpened stakes’;8 this would, of course, also go a long way to
1
FNC, iii. 490-1; GR, i. 454-5; among those who have followed Freeman on the location of the hillock
are Lemmon, ‘The Campaign of 1066’, pp. 96, 106, and The Field of Hastings, p. 48; Morillo,
introduction to The Battle of Hastings, ed. Morillo, p. xxvii; and Bennett, Campaigns of the Norman
Conquest, p. 38.
2
Round, anonymous Quarterly Review article of 1893, 78-80.
3
But this may not be the most likely interpretation, see above, pp. 50-1; other commentators have also
located the water in this area, for example, Lemmon, The Field of Hastings, p. 48. The maps in Burne,
The Battlefields of England, p. 25, and Barclay, Battle 1066, p. 58, marked the stream as ‘Sandlake
Brook’ and ‘Sandlake Stream’ respectively.
4
Brown, ‘The Battle of Hastings’, 20.
5
The Bayeux Tapestry, ed. Stenton, p. 175. Moustaches appear on Englishmen, including King Harold,
elsewhere in the Tapestry, and are almost certainly an indication of social status. Note also Norgate’s
comment (‘The Battle of Hastings’, 73) on the ‘light-armed warriors, whose accoutrement was
respectable enough to give them some title to rank as regulars’.
6
ASC, i. 108. The Old English eóred-cist is translated as ‘picked company’ here on the assumption that
the first element (sometimes used as the equivalent of the Latin word legio) refers to troops of any kind
and that the second derives from the verb céosan, to choose; see Plummer’s glossary, ibid., 328, and
The Anglo-Saxon Chronicle, trans. Swanton, p. 108, n. 1. The word cist can, however, also denote a
company, and eóred often refers specifically to a group of horsemen (hence its use by J.R.R. Tolkien
for the squadrons of the Riders of Rohan in The Lord of the Rings); thus, ‘mounted companies’ is an
alternative (adopted by EHD1, p. 219). Swanton states that eóred derives from eoh, ‘war-horse’.
7
Note also the comments in John of Salisbury’s Policraticus on Harold’s use of light infantry against
the Welsh in 1063, cited by DeVries, ‘Harold Godwinson in Wales’, pp. 83-4.
8
BT, pp. 192-3. It is, of course, impossible to be sure that they are not vegetation of some sort: Brown,
‘The Battle of Hastings’, n. 120, suggested ‘tufts of marsh grass’. Abels, Lordship and Military
Obligation, figs 6-7, notes that ‘the defenders of the hill are well-armed’, accepts the identification of
the shapes in the water as stakes, and makes the point that they indicate deployment on the hillock

123
 THE BATTLE OF HASTINGS

explain why the French horses are in such spectacular difficulties, and
together with the mailed English infantryman pulling down an enemy
horseman would indicate that not only the Horselodge Plantation area but
also that now occupied by New Pond, and perhaps ground further west
and south-west too, was held in strength against the French. The
construction in 1815 of New Pond, which has an overflow which allows
water into the channel to the west only when the Pond itself is full, must
have greatly reduced the wetness of this area, but in the early eighteenth
century the marsh which eventually extended down to Powdermill is
known from Budgen’s estate map of 1724 to have begun due south of the
westernmost part of the main ridge, that is at the point now occupied by
the easternmost part of Warren Wood, and this ground can be very boggy
even today. It is perfectly possible that Harold rested his right flank upon
that marsh,1 and that the Tapestry is showing French attempts to dislodge
his men from that area. Maps made before the construction of New Pond
do not suggest that the rather narrow stream need have been any wider in
1066 than it is today, and if the Tapestry is reliable in showing fighting
around a considerable body of water this may have been just to the north-
east of the modern Powdermill Lake, that is west of the track which today
forms the western boundary of the English Heritage site, and which
separates the rest of Warren Wood from its eastern tip.2 However, this is
far from certain, for the earliest detailed map of the area (that of 1724) is
still subsequent to the creation of the three Stew Ponds east of New Pond
and directly south of the abbey buildings which inhibit the flow of water
in the upper reaches of the little valley, and there may at the time of the
battle have been marshy pools in that vicinity; certainly, Budgen marked
what later became the site of New Pond as ‘Weanyer’s Pond’, although
no body of water is actually visible on his map, and there is even today a
fairly extensive area of marshy ground east of the point where two
separate little streams feed into the Pond.
If all this is to push the Tapestry’s evidence fairly hard, there would
seem to be good grounds for doing so. It is far from unlikely that Harold,
who knew that he would be facing heavy cavalry and must have
considered how he might best do so, decided to employ field defences
against them, and the hillock scene was presumably thought readily
recognisable by the Tapestry’s audience, some of whom may have
witnessed the fighting depicted. It is also arguably very significant that
what is probably our earliest detailed source on the battle, the Carmen,

before the battle began. Wilson’s suggestion was anticipated in 1816 by Hudson Gurney’s comment
(‘Observations on the Bayeux Tapestry’, 370) on ‘foot seeming to defend a kind of entrenchment
against horsemen’. Similarly, France, ‘The Importance of the Bayeux Tapestry’, p. 297.
1
As the map published in Behrens, Battle Abbey under Thirty-Nine Kings, p. 18, suggested.
2
The estate map of 1811 (ESRO, BAT 4435) shows a regularly shaped and therefore probably man-
made pond distinct from Powdermill Lake in this area, see Illus. 4.

124
 4 THE ENGLISH ARMY

says that the English occupied not only a mons (‘hill’) and rough ground,
but also a vallis (‘valley’),1 and although its author Guy of Amiens did
not elaborate on what happened in that area, and William of Poitiers, who
preferred to stress the reluctance of the English to meet William on level
ground,2 did not refer to it at all, their silence can plausibly be filled by
those writers who speak of deaths in ditches during the battle. William of
Malmesbury, in the passage noted by Freeman, says that the English
occupied a tumulus from which they hurled missiles and rolled down
stones on the enemy, killing so many of them near a deep ditch that the
bodies made it level with the plain;3 Henry of Huntingdon has the duke’s
men falling into a large and well-concealed ditch during their simulated
flight and the English also suffering heavy losses there as they retired;4
Wace mentions ditches several times, and says that during the battle the
Normans were pushed back to one which they had previously left behind
them, more dying at that time than any other, as those who saw the dead
said.5 Of course, these details are not completely consistent, but there is a
good chance that at least some of them reflect memories of fighting in the
drainage areas south of the main ridge. If so, there is an even stronger
likelihood that the initial English deployment extended to those areas, and
was not restricted to the crest of the ridge as Ramsay and Baring argued,
and their many followers have long believed. Moreover, this deployment
to the south, if it was not simply restricted to the area immediately below
the later abbey, may support Freeman’s contention that English troops
also occupied much of the ridge itself. More radically, and remembering
that the only certain point is the location of Harold’s standard, one might
wonder whether it was not simply that he rested his right flank on the
easternmost reaches of the marsh, but whether very significant fighting
may not also have taken place to the south-west of Horselodge Plantation,
in the relatively flat countryside which extends for just over half a
kilometre down to Powdermill Lake.
If there is a good chance that the scholars who wrote immediately after
Freeman were mistaken about the extent of the English position, could
they have erred on the question of numbers too? This is a much more
difficult matter, and one ultimately incapable of resolution. Even so,
discussion of it is not entirely fruitless. It is known from Domesday Book
that before the Conquest Berkshire sent a man to the army from every

1
Carmen, ll. 365-8.
2
GG, pp. 126-7, 170-1.
3
On WM’s possible use of the Tapestry, see below, pp. 176-7..
4
HA, pp. 392-5.
5
RR, ll. 6969-72, 7847-8, 8079-96. But, of course, Wace probably used both WM and HH and possibly
knew the Tapestry. To the west of New Pond there is visible today a series of mounds which extends
for something like 300 metres. While there are other and more likely explanations of their origin, there
is a possibility that they are grave mounds. See further, below, p. 200 and Illus. 29 and 31.

125
 THE BATTLE OF HASTINGS

five hides and that he was given four shillings from each one as his
maintenance and wages for two months.1 It has been thought probable
that the same system operated throughout much of England, and that in
the shires in the Danelaw assessed in carucates six of these units fulfilled
the same function.2 Thus, if the assessment of England as a whole was
something like 80,000 hides3 they ought to have supplied about 15,000
five-hide men, and such men were probably well-equipped, for when
Æthelred II raised a fleet in 1008 he required that every eight hides
should provide a helmet and a coat of mail.4 This levy system survived
for some time after the Conquest, and in 1094 William II (Rufus) is said
to have summoned 20,000 men for service in Normandy and then turned
them back at the coast after taking the half pound (ten shillings) each man
had with him. This sum suggests a uniform system of support
recognisable from the Berkshire Domesday, while the figure of 20,000 is
far from implausible on the five-hide calculations just mentioned.5
Moreover, such totals might also seem to fit an event such as the battle of
Æthelingadene in 1001, when the Vikings killed 81 men of Hampshire,6
for on the five-hide reckoning the defeated force would have numbered
about 550 men, raised from a shire assessed at 2,785 hides.
Other evidence they do not fit. It is very difficult, for example, to
believe that the peasants armed only with spears depicted on the Bayeux
Tapestry had been supplied by units of five hides,7 or that all the forces
raised by individual shires were made up in this way. There are many
examples in the Anglo-Saxon Chronicle of invaders being met by the men

1
DB, i. 56c.
2
Hollister, Anglo-Saxon Military Institutions, pp. 38-52, building on earlier work such as Round,
Feudal England, pp. 67-9, and Maitland, Domesday Book and Beyond, pp.156-7. Hollister (pp. 52-3)
thought the system in East Anglia impossible to reconstruct. Abels, Lordship and Military Obligation,
pp. 108-15, is less convinced of the universality of the five-hide and six-carucate rules (similarly,
Delbrück, History of the Art of War, iii. 160-1) but notes (p. 113) that the Domesday evidence suggests
that Bedfordshire ‘must have had some standard rule, similar to that practiced in Berkshire’.
3
For the Domesday figures, including the totals for individual shires referred to shortly, see Darby,
Domesday England, p. 336. They come in all to just over 69,000 hides, but Domesday does not deal
with lands north of Yorkshire, London and Winchester are omitted, and for many of the boroughs no
figures are given: compare the lists in ibid., pp. 364-8 and Ballard, The Domesday Boroughs, p. 65.
Thus, 80,000 may not be far from the true total.
4
ASC, i. 138. The Chronicle, of course, uses the Old English word for mailcoat, byrne, the equivalent
of the Carolingian (and Germanic) brunia. No Anglo-Saxon mailcoat of this period has survived, see
above, p. 66, n. 3 and 68, n. 2. Contamine, War in the Middle Ages, p. 51, estimates the number of men
available from the five-hide system as 20,000.
5
ASC, i. 229; similarly, Bachrach, ‘William Rufus’ Plan’, p. 55; Hollister, Anglo-Saxon Military
Institutions, pp. 43-4, regarded 20,000 as ‘a gross exaggeration’.
6
ASC, i. 132.
7
Abels, who does not believe that all freemen may have been liable for military duty in 1066 (see
below) or had ever been so, implies (Lordship and Military Obligation, pp. 178-9) that the Tapestry’s
figures can be equated with minor landholders of the type known to have fallen in the battle; this seems
to me most implausible. For general criticism of Abels’ thesis, see the review by Professor Loyn, EHR,
cvii (1992), 164-5, and for something close to an earlier statement of it on the battle of Hastings,
Delbrück, History of the Art of War, iii. 154-6.

126
 4 THE ENGLISH ARMY

of one shire only, and in some cases the totals produced by the five-hide
rule carry little conviction. In 1010, for example, the levies of East Anglia
and Cambridgeshire were defeated by a Viking force described as
‘immense’ when it arrived the previous year. Yet if the English thought
that they stood a chance of being victorious by engaging with only just
over 1,000 men the eleventh-century meaning of ‘immense’ was curious
indeed.1 In fact, it is possible that some local forces could summon every
able-bodied man if necessary, and on the Welsh border that is what
elements of the Domesday evidence suggest.2 Thus it would not be
surprising if in critical circumstances national armies were assembled in a
similar way, for when the Chronicle says that Edmund Ironside called up
all the English nation five times against the Danes in 1016 it is probably
referring to a higher rate of summons than one man from every five
hides.3 The flexibility of early medieval administrative methods would
have allowed infinite variation in what was required, as one is reminded
by material from Carolingian Francia, which had a system of government
sufficiently like that of the late Anglo-Saxon state for its evidence to be
worth noticing. In 806 Charlemagne ordered that in military action in
Spain or against the Avars five of the (continental) Saxons were to equip
a sixth, in the case of action in Bohemia two were to equip a third, but if
there was fighting against the neighbouring Sorbs one and all were to
come; similarly, when the Frisians were summoned counts, royal vassals
and horsemen were all to be present, but of the poorer men six were to
equip a seventh.4 In the circumstances of 1066 Harold seems likely to
have asked for more men rather than less, and this is the implication of
such scanty evidence as survives. Domesday Book records that two of the
three holders of an estate of four hides and one virgate at Tytherley in
Hampshire were killed at Hastings.5 It is possible, but seems unlikely,
that these two men were representing lands assessed at a total of ten
hides. If they were not, then the Hampshire levy was made up of more
soldiers than the five-hide reckoning, if it was ever used in the shire,
would have produced. Two other Domesday entries give the same sort of
impression. A freeman who held only twelve acres at Winfarthing in

1
ASC, i. 139-40. Cambridgeshire was assessed at 1,297 hides in 1086, Norfolk at 2,423 carucates and
Suffolk at 2,411 carucates. On the five-hide and six-carucate rules this would have produced around
1,065 men.
2
Hollister, Anglo-Saxon Military Institutions, p. 29, and on what Hollister called the ‘great fyrd’
generally, pp. 25-37.
3
ASC, i. 151. While one has to agree with Abels (Lordship and Military Obligation, pp. 176-7) that the
entry cannot refer to the summons of every able-bodied man in England, it does not seem likely that
Edmund’s armies ‘differed only in the number of shires involved, not in the personnel called forth’.
4
Capitularia, ed. Boretius, i. No. 49, cc. 2-3; Charlemagne, trans. King, p. 257.
5
DB, i. 50a.

127
 THE BATTLE OF HASTINGS

Norfolk also died in the battle, along with Breme, another freeman who
possessed 1½ carucates at Dagworth in Suffolk.1
Moreover, it is doubtful whether all those who served in Anglo-Saxon
armies did so because of the requirements of the hidage system or of a
more extended summons. In some cases they fought as one of the duties
of a relationship of mutual obligation between them and a lord. Very
early in the ninth century Charlemagne ordered that men summoned from
the counties were each to come with spear, shield and a bow with two
strings and twelve arrows, but that bishops, counts and abbots were to
bring homines in possession of metal body armour and helmets.2 The
homines of these lords were clearly distinct from men levied from the
land, not least in being rather better equipped, and such a distinction may
well have been evident in English armies too. When the men of Essex
assembled to face the Vikings in battle at Maldon in 991 their leader,
Ealdorman Byrhtnoth, positioned himself, according to the poem The
Battle of Maldon, where he knew his heorðwerod (‘household
companions’) to be most loyal. Following his death his thegn Offa was
killed, yet he had accomplished what he had promised his lord, and lay by
his side as a thegn should.3 When a thegn died he was required to give (or
more probably give back) military gear to his lord in what was known as
the heriot. Late in Edward the Confessor’s reign Ketel, a thegn of
Archbishop Stigand of Canterbury, provided for the return of a horse,
shield, spear, sword, helmet and coat of mail, while the laws of Cnut had
required that the heriot of a king’s thegn consist of four horses (two
saddled and two unsaddled), a helmet, a coat of mail, two swords, four
spears and four shields.4 Doubtless thegns who served in the army also
often represented five hides of land,5 but this need not always have been
so, for there may have been household companions of great lords who
held no land at all, and if so they would not have been dissimilar to many
of the most famous soldiers believed to have fought on the English side at
Hastings, the mercenaries known as housecarls.
Whether their fame is justified is another matter. Anglo-Saxon kings
had probably paid specialist troops at least since the days of Alfred, who

1
DB, ii. 275b, 409b. There is no way of knowing that these men did not have other and more extensive
lands elsewhere. The other casualties listed by Domesday are Edric the Deacon, a man of Harold, who
held an unspecified amount of land at Cavendish in Suffolk, and Ælfric, who held 16 hides in
Huntingdonshire from the Abbot of Ramsey (DB, i. 207a, 208a; ii. 449a). See also, below, p 189..
2
Capitularia, ed. Boretius, i. No. 77, c. 9; Charlemagne, trans. King, p. 244. The precise nature of
Carolingian metal body armour, usually denoted by the words brunia or lorica, is unclear, see
Coupland, ‘Carolingian Arms and Armor in the Ninth Century’, 38-42.
3
BM, ll. 23-4, 288-94. See Hollister, Anglo-Saxon Military Institutions, pp. 127-8; Abels, ‘English
Tactics, Strategy and Military Organization in the Late Tenth Century’, 146.
4
See Brooks, ‘Arms, Status and Warfare in Late-Saxon England’, who argues (p. 89) that Cnut’s rates
had been in force under Æthelred II; Ketel’s will is translated EHD2, No. 189.
5
Hollister thought that they always did so, Anglo-Saxon Military Institutions, p. 64.

128
 4 THE ENGLISH ARMY

according to his biographer Bishop Asser set aside part of his revenues
each year to be given to his bellatores (‘fighters’),1 but on his accession
in 1016 Cnut was once thought to have introduced a military fraternity of
regularly-paid élite troops governed by strict rules. Two Danish writers of
c.1200, Sven Aggeson and Saxo Grammaticus, say that they consisted of
3,000 men (Sven) or 6,000 in sixty ships (Saxo) and give the regulations
using versions of the Lex Castrensis (Law of the Retainers), which
according to Sven he had translated into Latin from a vernacular record
produced by Archbishop Absalon of Lund and believed to reflect the
practices of Cnut’s time.2 Unfortunately, the Lex Castrensis is now
thought worthless as a source for the eleventh century, 3 and this means
that it is necessary to depend on the sparse and unsatisfactory
contemporary and near-contemporary evidence. Housecarls became (or at
least the word was applied to) men who were more than simply hired
fighters. They are found in Domesday Book holding land and looking
very like thegns, and in Harthacnut’s reign collecting taxes.4 Yet that they
could also be important militarily there is no doubt. The Domesday entry
for Wallingford, which refers to fifteen acres where housecarls stayed,
hints at the existence of housecarl garrisons.5 John of Worcester says that
almost all Harthacnut’s housecarls participated in the ravaging of
Worcestershire in 1041 and the D text of the Anglo-Saxon Chronicle that
some of the Confessor’s were killed during Earl Siward’s invasion of
Scotland in 1054.6 Nor was the king the only lord to be served by such
men. Some of Siward’s own housecarls died on the Scottish expedition,
as did those belonging to Earl Tostig in the Northumbrian revolt of 1065,
while Domesday Book mentions individual housecarls of the earls
Leofwine, Ælfgar and Waltheof.7 Doubtless both the royal housecarls and
and those of Earl Gyrth, Earl Leofwine and other lords took part in the
battle of Hastings, but that is just about as far as one can go; they must
have been the ‘stipendiary and mercenary soldiers’ who according to
William of Malmesbury bore the brunt of the fighting,8 but if their deeds
were ever celebrated they made little impact on the authors of most of the
sources which have survived; if Wace had heard anything of them, he did
not choose to tell. Still, their numbers and those of the household thegns

1
Asser, p. 86. Payments were also made to his ministris nobilibus, i..e. probably thegns.
2
Sven Aggeson, Lex Castrensis, ed. Gertz, i. 64-92, at 68-9; Saxo, Historia, ed. Christiansen, i. 36-44.
3
For a revision of Larson, The King’s Household in England before the Norman Conquest, pp. 152-71,
see Hooper, ‘The Housecarls in England in the Eleventh Century’, 166-9; Abels, Lordship and Military
Obligation, pp. 161-70; and note also Eric Christiansen’s comments, Saxo, i. 154-5, and The Works of
Sven Aggesen, pp. 7-11.
4
Lawson, Cnut, pp. 179-80.
5
Campbell, ‘Some Agents and Agencies of the Late Anglo-Saxon State’, p. 204, n. 18.
6
JW, ii. 532-3; ASC, i. 185. The more contemporary C text does not mention them.
7
ASC C, i. 190; Larson, The King’s Household, p. 164.
8
GR, i. 422-3.

129
 THE BATTLE OF HASTINGS

would have added to those levied from the land by means of the hidage
and the more extended levy systems.1
To discuss the size of the English army and its deployment in the battle
is to discuss a range of possibilities. The inadequate and tendentious
nature of the evidence means that it is impossible to prove Ramsay,
Baring and others wrong in believing that Harold had no more, and
perhaps many less, than 10,000 men, who defended a restricted position
around the crest of the ridge. Indeed, the witness of the Chronicle E text,
John of Worcester and William of Malmesbury would all tend to support
this view, and because the clay of the western part of the field retains
water very well it could be argued that fighting around the Bayeux
Tapestry’s watercourse took place in the vicinity of a large pond on the
main ridge, and that the Tapestry's hillock is not Horselodge Plantation
but a feature on that ridge;2 this interpretation would also be likely to
think of the battle’s name of Senlac as referring to a sandy channel in the
area of the road known today as Lower Lake, and east of the location of
Harold’s standard;3 it could also be stressed that there are so many
streams and eminences in the vicinity that it is hazardous to identify any
with those in the Tapestry, or even suggested that the designer in this
instance created a scene which bore little relation to actual events. Yet
whether such arguments offer the most plausible interpretation of all the
surviving source material is another matter. The insistence of post-
Conquest English authors like John of Worcester and William of
Malmesbury on the inadequacy of the English forces may be the result of
a desire to minimise their countrymen’s defeat at a time when the effects
of the Conquest were still a very sensitive issue.4 Moreover, it is not only
contemporary French writers like Guy of Amiens and William of Poitiers
who say that the English army was large, but the D text of the Anglo-
Saxon Chronicle as well. The latter also has them under attack before
they were properly ordered, and if this was so when there had been time
for the construction of field defences this would seem more likely to have
been a problem experienced by a large army than a small one.
Furthermore, the area of low-lying land south of the main ridge,
containing a stream which is distinctly sandy, and which before the
construction of New Pond would probably have passed through an area
1
Bachrach, ‘Some Observations on the Military Administration of the Norman Conquest’, 4, 24-5,
judges that Harold had 50,000 men or more at his disposal late in 1066, at least 14,000 being from the
five-hide system along with ‘several thousand first-class warriors’ in the followings of the king and
great lords.
2
Bradbury, The Battle of Hastings, p. 189, suggests that it ‘is the artist simply showing the main hill of
the battle’ in a different view; similarly, Barlow, Carmen, p. lxxxi. Burne, The Battlefields of England,
p. 29, n. 2, thought the scene ‘refers to the closing stage of the battle when the English position was
contracted and attacked from each side’.
3
Above, pp. 50-1.
4
Above, pp. 52-60.

130
 4 THE ENGLISH ARMY

much more marshy than is the case today, is despite all doubts a distinctly
plausible location for the Tapestry’s water, and the statement of the
Carmen that the English occupied not only a hill but also a valley is also
of the greatest significance from this point of view.
All this may well imply that Freeman was correct in thinking that
Harold’s men occupied a great deal of the ridge too. Criticised after his
death because of the extent of the English deployment shown on his map,
the Tapestry’s mailed English infantryman pulling down a French horse
near the water is only part of the evidence that in fact he posited a
deployment that was not, in at least one respect, extensive enough. How
large a ‘large’ army may have been there is no way of knowing. The
losses at Fulford and Stamford Bridge so soon before Hastings may have
limited the numbers Harold could assemble, but the facility of the Anglo-
Saxon military system for dealing in surprisingly large numbers, at least
in the days of Alfred and Æthelstan, must to an extent call this into
question, as must also its known resilience.1 It is worth noting in this
respect that Mercian military resources may not have been exhausted by
the battle of Fulford, as a landowner from Worcestershire was evidently
killed at Stamford Bridge. Had the shire sent one contingent to join Earl
Edwin when he heard of the Norwegians’ arrival, and then another to
fight alongside Harold?2 If he, like Edmund Ironside, called up ‘all the
English nation’ to deal with the French, at a rate of considerably more
than one man for every five hides, then one could be in the area of tens of
thousands, especially as such a call-up would have been supplemented by
both the sworn followings of great lords and numbers of the not-
dissimilar housecarls. At the very least, tens of thousands are a
possibility, and possibilities are all that there is, for the only certainty is
that there is no warrant for accepting without demur the cosy consensus
established in the late nineteenth century and hardly questioned since.
The truth is that we do not know how large or small the English army
was, and that ignorance should be friend to many possibilities.

Whatever its size, what was its nature? Were Anglo-Saxon troops of
the period largely a poorly-armed rabble of ill-disciplined peasants
strengthened by the better weaponry of the five-hide men, thegns and
housecarls, or is there evidence for a much greater degree of
professionalism than this? On circumstantial grounds, one would expect
so. If any group in the early medieval west showed great skill in the

1
Above, pp. 107-12.
2
DB, i. 177d, speaks of an uncle of the abbot of Evesham who died in bello Heraldi contra Norrenses
(‘in Harold’s war against the Norse’). Of course, he need not have been a member of a second
Worcestershire levy, perhaps having joined Harold after fighting at Fulford, or holding land elsewhere
in England and marching with the men of another shire; see FNC, iii. 361.

131
 THE BATTLE OF HASTINGS

waging of war it was the Vikings, for this was how they earned their daily
bread.1 They profited in many ways from their operations, by plundering,
by being given tribute so that they would avoid plundering and the terror
and devastation that were doubtless its usual accompaniment, by
ransoming captives, by selling captives into slavery, and by serving
foreign kings as mercenaries; and all this depended on their ability to
defeat any who took the field against them. This was why opposing
forces found them very difficult to deal with, and why it is probable that
those which eventually gained a measure of success did so by achieving a
level of professionalism which matched or at least approached that of
their enemy. The sheer extent of the military measures taken by Alfred, in
terms of manpower, ships and fortresses, has already been mentioned, as
has the obvious scale of the forces deployed by his son Edward the Elder
in the reconquest of the Danelaw. Yet it is almost certain that the military
system utilised by them and their successors also contained an array of
refined and important details of which nothing is known not because they
never existed, but because of the inadequacy of the sources. Like
Charlemagne’s campaigns against the continental Saxons, the wars of
Alfred and his son often involved attacks on fortified sites which
probably included the use of siege machines. The more plentiful
Carolingian sources mention them, the English ones do not.2 Moreover,
Charlemagne’s order that his army was to be provided with fundibulas
(‘slinging machines’) and ‘such men as know how to make them throw
well’ and ‘stones for these on twenty pack-saddles’3 is only part of one of
a number of pieces of legislation which established in very precise terms
what he expected of his subjects in this sort of area. He also, as has
already been noted,4 required the men from the counties and the homines
brought by the lords to be equipped in a certain fashion, including in the
latter case the possession of helmets and metal body armour. Nor were
logistics neglected: tools were to be taken in abundance, while the
provisions of the king, and of his bishops, counts, abbots and magnates,
were to be transported in carts, and to include flour, wine and bacon.5 The
war-carts supplied by royal estates were to be well-constructed and to
have coverings of skins sewn together in such a way that when crossing
rivers there would be no ingress for water. They were to contain 12 modii
of flour, and 12 modii of wine were to be carried in the carts designated
1
On Viking methods, see below, pp. 138-9.
2
In the nineteenth century Benjamin Thorpe’s note to his translation of Lappenberg (A History of
England under the Anglo-Saxon Kings, ii. 297, n. 1) suggested that the lignis imposita saxa (literally,
‘stones placed on pieces of wood’) which WP (GG, pp. 128-9) describes the English as hurling at the
French referred to the use of field artillery. Unfortunately, ‘stones tied to sticks’ seems more plausible.
3
Capitularia, ed. Boretius, i. No. 77, c. 10; I have used the translation of Charlemagne, trans. King, p.
245.
4
Above, p. 128.
5
Capitularia, ed. Boretius, i. No. 77, c. 10; Charlemagne, trans. King, pp. 244-5.

132
 4 THE ENGLISH ARMY

for that purpose; each cart was also to have a shield, lance, quiver and
bow.1 The stewards of the royal estates were additionally to take great
care of the horses, almost certainly because these had a military role,2 and
counts were to reserve two thirds of the grass in their counties for the
needs of the army.3
English sources cannot match provisions as detailed as these, but
plausible extrapolation from what there is can take us quite a long way.
When Cheshire supplied one man from every hide to repair the walls of
Chester in the eleventh century,4 for example, it would be strange if they
were not expected to appear with specific tools. In 1066 the provisions of
the forces which Harold stationed on the south coast ran out early in
September, just as those of the West Saxons besieging a Viking army had
done in 893,5 but provisions there had been, and one assumes that they
had been transported to the required localities as a result of concrete
administrative orders and by methods which must have included carts and
draught animals. Similarly, it is not unlikely that those Anglo-Saxon
soldiers who travelled on horseback found, like their Carolingian
counterparts, that provision had been made for the feeding of their
mounts, and that when Æthelred II ordered that every 8 hides were to
provide a helmet and coat of mail for his navy6 this reflected a world in
which the five-hide man from Berkshire marched to the army not only
with his four shillings from every hide but also with clearly-specified
military equipment. If this resulted in a degree of uniformity in the
weaponry of English soldiers it would match the evidence of the Bayeux
Tapestry. The men in its shield-wall, for example, almost all have coats
of mail and helmets, kite-shaped shields and spears, and this is hardly
surprising in view of the evidence of heriots that helmets and mailcoats
were the normal gear for men of status (and doubtless housecarls).7
Similarly, the moustachioed light infantry on the hillock all have spears
and shields, and their peasant companions simply spears. Even clubs
could have been standard weapons, for Charlemagne's instruction that his
men were to have a bow rather than a cudgel implies that the latter had

1
Capitularia, ed. Boretius, i. No. 32, c. 64; The Reign of Charlemagne, trans. Loyn and Percival, p. 72.
This document, the Capitulare de Villis, may be attributable to Charlemagne’s son Louis the Pious.
2
Ibid., cc. 13-15; trans. Loyn and Percival, pp. 66-7. On the breeding of war-horses in this period, see
below, p. 155.
3
Capitularia, ed. Boretius, i. No. 77, c. 10
4
DB, i. 262d.
5
ASC C, i. 196; ASC A, i. 85-6.
6
ASC, C, D and E, i. 138.
7
One of the shield-wall men holds a small axe along with spears and a shield, another only a large axe,
and one of the standard-bearers simply his standard. Whether the Tapestry’s shield-wall men are
intended to represent housecarls is unclear. Another piece of evidence on the uniformity of equipment
is JW’s story (ii. 530-1) of how the ship given to King Harthacnut by Earl Godwin to atone for the
murder of the ætheling Alfred contained 80 picked soldiers, each with a loricam trilicem (‘triple coat of
mail’), a sword, a Danish axe, and a shield and a spear; on heriots, see above, p. 128.

133
 THE BATTLE OF HASTINGS

once been acceptable and perhaps indeed required equipment; it may still
have been so in England in 1066.1
Of course, this is perhaps to push the Tapestry’s evidence a great deal
too far; even so it has other features which are arguably significant. It is
noteworthy, for instance, that the heavily-armed English fighting only
with axes and those with round shields are operating in looser order than
the shield-wall men. This may be nothing more than the quirk of a
designer who wished to glorify the French cavalry by showing them in
individual comabt with armoured Englishmen,2 or merely a depiction of a
phase in the battle when the shield-wall had been broken. Yet it may also
be that the designer knew that the nature of a soldier’s equipment
sometimes reflected the role that the man was expected to perform, and
that the troops who comprised English armies and the formations which
they adopted extended beyond simply the shield-wall. The round shield
with its large boss may have been particularly suited to heavy infantry
fighting in open order, giving more all-round cover than the kite-shaped
variety,3 while the space required to swing a large axe with two hands
strongly suggests that many of its bearers must have been intended to
fight in this way too. Armoured men with spear and sword might also
have operated thus, as the individual defending the watercourse against
French cavalry indicates. Equally, the single archer who fronts the shield-
wall could be taken as representative of lines of missile troops who
skirmished in front of the heavy infantry before retiring to the rear as the
enemy approached.4
Yet if English armies contained specialist light infantry, missile troops
and heavy infantry who fought in formations other than densely-packed
units, the fact remains that the shield-wall is of considerable importance,
and a matter (for once) upon which a certain amount of information
survives. Its use by the English went back at least to the time of Alfred,
when according to Asser’s account of the battle of Ashdown in 871,
based on that of the Anglo-Saxon Chronicle, the Vikings split into two
divisions, forming shield-walls of equal size, and the West Saxons did the
same, while at Edington in 878 Alfred advanced cum densa testudine
(‘with a ?particularly dense formation’) to fight against the pagans and
won the victory;5 similarly, at Farnham in 893 his son Edward is said to

1
Capitularia, ed. Boretius, i. No. 77, c. 17; Charlemagne, trans. King, p. 245.
2
For the designer’s bias towards cavalry, see above pp. 68-9, and below, pp. 170-1.
3
See the comments of Peirce, ‘Arms, Armour and Warfare in the Eleventh Century’, 244.
4
The fact the Tapestry shows only one archer has often been remarked, and may be evidence of
Harold’s lack of such troops, but there is a danger of being misled here by the source’s limitations.
Lines of skirmishing English archers would not have been easy to represent in such a medium,
especially considering the danger of confusion with the many French bowmen in the lower border. The
designer also seems to have had little interest in depicting troops of modest social origin.
5
ASC, A, i. 70; Asser, pp. 28-30, 45-6; trans. Alfred the Great, ed. Keynes and Lapidge, pp. 79, 84.

134
 4 THE ENGLISH ARMY

have used an agmine denso against the enemy.1 The shield-wall appears
again at Maldon in 991,2 and at Hastings, where the insistence of the
more detailed sources on the impenetrability of the English lines must
refer to a body or (on the analogy of Ashdown) bodies of this type, which
are represented after a fashion in the Bayeux Tapestry. The Latin word
used by Asser to translate the Old English ge-truma is testudo, originally
meaning ‘tortoise’, but also denoting a formation used by Roman
legionaries in which they used their shields to protect themselves on all
sides, and from above.3 It was doubtless the implications of this parallel
that caused Wilhelm Spatz to deny the use of the shield-wall at Hastings
in significant terms:
Nur ein in langwierigen militärischen Exerzitien geschultes, also aus
taktischen Körpern zusammengesetztes Heer würde imstande sein, eine so
ausserordentlich schwierige Formation, wie sie ein lang fortlaufender,
4
gerader Schildwall ist, zu bilden.
Exactly so. The Tapestry’s depiction of the Hastings shield-wall is
inadequate to the extent that it almost certainly fails to convey the
formation’s depth, which was presumably of many ranks, and if so such
assemblies must have taken time to organize. When Asser says that
Alfred’s assault at Ashdown only took place testudine ordinabiliter
condensata (when the shield-wall had formed up properly) he had a
particular and probably fairly elaborate process in mind, and it may be
that at Hastings William was shrewd enough to begin proceedings before
Harold’s men had completed it, for this could well be the meaning of the
Anglo-Saxon Chronicle D text’s claim that the English were attacked ær
his folc gefylced wære (‘before his army was properly ordered’). Nor,
whether it was employed defensively as at Hastings or offensively as at
Ashdown, would it be easy to maintain in the stress of battle, as the ranks
adapted to the loss of the dead and wounded. Neither is it likely that
fighting in such units was solely the province of the well-equipped
household troops of great lords, or of the mercenary housecarls, as the
1
The Chronicle of Æthelweard, ed. Campbell, p. 49. This entry in Æthelweard’s translation of the
Anglo-Saxon Chronicle into Latin (of c.1000) is believed to be based on a revision made in the reign of
Edward (899-924) of the original Chronicle entry, see pp. xviii, xxviii-ix.
2
BM, ll. 101-2, 242, 277. The words used here are the Old English wihaga (‘battle-hedge’), scyldburh
(‘shield-fortress’) and bordweall (‘shield-wall’). The last is also employed of the Viking formation in
the poem on the battle of Brunanburh in 937, ASC, i. 106.
3
The ASC entry for 871 says that the Vikings split into two ge-fylce, translated as turmas by Asser, and
that Alfred and his brother King Æthelred then attacked their two ge-truma. Ge-fylce appears again in
the same entry of the Viking units involved in the fighting at Meretun. WM (GR, i. 452) also uses the
word testudo, in the phrase conserta ante se scutorum testudine, to describe the English shield-wall at
Hastings.
4
Spatz, Die Schlacht von Hastings, p. 45. ‘Only an army trained in protracted military exercises, such
as an army made up of tactical corps would be, would be capable of forming such an extraordinarily
difficult formation as a long, continuous shield-wall’. I am grateful to my colleague Paul Collinson for
verifying the translation of this passage. ‘Tactical corps’ was a term adopted by Spatz from Delbrück
and used to describe highly trained troops.

135
 THE BATTLE OF HASTINGS

Tapestry’s depiction of an armoured line might lead one to think, for it is

clear from its use by the men of Essex at Maldon and probably from the
Alfredian references that it was employed by levies raised from the shires
too. If so, it is virtually inconceivable that they did not receive a
considerable degree of appropriate training before being required to face
the enemy.1 English commanders may have been familiar with the
Epitome of Military Science by the Late Roman writer Vegetius, and have
reflected not only on his maxim that if you want peace you must prepare
for war, but also on that which stated that if you want victory you must
train soldiers diligently.2 There may be an oblique reference to such
training in the statement of The Battle of Maldon that Ealdorman
Byrhtnoth told his men how to stand and asked that they should hold their
shields properly, later asking them to form the ‘war-hedge’.3
Indications of the sort of methods used to form up dense bodies of
infantry have survived from the ancient world. Spartan hoplites fought in
close order with shield and thrusting spear in a manner probably not
dissimilar to that of the shield-wall, and in Xenophon’s time a hoplite
mora was made up of four lochoi, each of the latter having two
pentekostyes of two enomotiai apiece.4 The enomotia may have consisted
of 36 men and if so the lochos had 144.5 Yet, whatever the precise
numbers, it is the mere existence of sub-units which is important, for it
was these which enabled the phalanx to form up and manouevre, as was
also the case with the constituent parts (cohort, maniple and century) of
the Roman legion. Thus, it would be surprising (indeed, almost
incomprehensible) if the shield-wall did not contain equivalents, and
worth noting that Old English terms like eóred-heáp, eóred-þreát, eóred-
weorod, heáp, scild-truma, truma and weorod may occasionally have had
much more precise meanings than the simple ‘troop’ and ‘company’
which is as far as one can get in translating them today.6

1
Burne, The Battlefields of England, pp. 24-5, judged the ‘interlocked shield-wall... a poetic fantasy’;
Lemmon, The Field of Hastings, p. 43 (similarly, ‘The Campaign of 1066’, p. 101), that the ‘formation
was obviously designed to give maximum cover from missiles. For “in-fighting” it would have to be
broken up’. This is, however, neither the implication of the literary references nor of the scene in the
Bayeux Tapestry in which English infantry are clearly receiving French cavalry attacks with
overlapping shields and (mainly) spears used overarm.
2
Vegetius, Epitome, p. 63. The Epitome rivals Pliny the Elder’s Natural History in terms of the number
of manuscripts surviving from before 1300 (p. xiii).
3
BM, ll. 17-20, 101-2.
4
Xenophon, Spartan Society, trans. Talbert, p. 178. I am grateful to Mr J.R.M. Smith for these
references.
5
P. Connolly, Greece and Rome at War, pp. 39-41. However, this is speculation, and it should be noted
with regard to Xenophon’s remarks that ‘the size and precise relationship to one another of the units
mentioned both here and later are obscure’, Talbert, p. 187.
6
See the entries in An Anglo-Saxon Dictionary, ed. Bosworth and Toller, and Supplement. King
Alfred’s translation of Orosius equates truma with the Latin ‘cohort’; scild-truma eventually became
‘sheltron’, and was sometimes equated with the Latin testudo; see OED under ‘trume’ (dismissing a

136
 4 THE ENGLISH ARMY

Points such as these may (or may not) be half-answers on the many
complexities of the Old English military system, but there is a great range
of questions which can hardly be answered at all. When a shire levy faced
the enemy alone, for example, how much provision was there for
specialist troops such as archers1 and light infantry, and were all the rest
of its men part of the shield-wall? If so, is it to be assumed that those who
did not have shields, like the peasants depicted in the Bayeux Tapestry
bearing only spears and clubs, fought in the rear ranks? In national armies
where were the followings of great lords and the housecarls deployed,
and the specialist units referred to by the term eóred-cist, and from which
men were these units drawn? Did the Anglo-Saxons ever make use of
cavalry?2 When the levies of different shires coalesced in national armies
how did they form up on the field of battle? Upon this, at least, there is
just a little more to go on than the mere belief that some sort of system
there must have been, for John of Salisbury’s Policraticus, written
c.1160, says that the English army’s order of battle is that the men of
Kent form the first line and those of Wiltshire, with Devon and Cornwall,
come next.3 Not long afterwards Wace made a similar statement about the
Kentishmen, and added that the men of London formed the king’s
bodyguard,4 both of course with reference to the deployment at Hastings.
It is curious, and at the same time probably significant, that these late
comments about the prestige of the men of Kent find a distinct echo in the
report of Edward the Elder’s harrying of Danish East Anglia given by the
Anglo-Saxon Chronicle for 903. When he eventually wished to withdraw
he told the whole army that they were to set out together (the levies of
different shires had apparently separated) but the Kentish force lingered,
despite having received seven messengers, and was caught by the Danes.
In the bloody conflict which followed they lost several of their leaders,
including two ealdormen, and were driven from the field, but inflicted
heavy casualties on the enemy, King Eohric and Edward’s renegade
cousin Æthelwold being among the dead.5 If this reveals nothing else it
speaks of the pride (and perhaps greed for booty) of some shire levies,

connection with the Latin turma) and ‘sheltron’. Delbrück, History of the Art of War, iii. 153, rejected
any similarity between the capacities of the Greek phalanx and those of the English shield-wall.
1
BM, l. 110, refers to the use of bows at Maldon apparently by both sides. Whether ll. 70-1 and 269-70
also refer to archery depends upon whether the word flán is translated ‘arrow’ or ‘missile’.
2
Glover, ‘English Warfare in 1066’, was ahead of its time, if inclined to place too much weight on the
appearance of English horse in Snorri Sturluson’s account of Stamford Bridge; see more recently,
Hooper, ‘The Aberlemno Stone and Cavalry in Anglo-Saxon England’, 192-6; Bennett, ‘The Myth of
the Military Supremacy of Knightly Cavalry’, p. 310; Strickland, ‘Military Technology and Conquest’,
359-60; and for the possible use of cavalry at Brunanburh (noted by Hooper) and the term eóred-cist,
above, p. 123, n. 6.
3
Campbell, ‘The Late Anglo-Saxon State: A Maximum View’, p. 24.
4
RR, ll. 7819-30.
5
ASC A, i. 92, 94. I have used the date for this annal of EHD1, p. 208, which suggests that these events
actually took place late in 902.

137
 THE BATTLE OF HASTINGS

who doubtless carried their own standard into battle, and whose men may
have formed up beneath the banners of the shire’s individual hundreds.1
And if those levies sometimes contained peasants armed with no more
than spears or clubs one should not be too ready to assume that such men
had no military value.2 Were this so it is doubtful whether a polity as
concerned with warfare as the late Anglo-Saxon state would have had
them there at all,3 and if the Bayeux Tapestry shows one of them looking
apprehensively at the approaching enemy4 this may be no more than a
reflection of the social prejudices of its patrons. Nor should one forget the
known achievements of levies under professional direction at other times.
It was the modern conscript equivalents of the armed peasantry of the
Anglo-Saxon period who in 1918, with the assistance of artillery, tanks
and ground-attack aircraft, in one of the greatest of all the victories of
English arms, broke the elaborate German defence system known as the
Hindenburg Line.
Raised, equipped and probably trained in elaborate ways, how
elaborate were the tactical skills employed by those who led Old English
armies into battle? Presumably they varied in quality at different times
and under different commanders, but in considering these matters it is
useful to note the important article on early medieval warfare which
Professor Leyser presented to a conference on the battle of Maldon in
1991, and which was published only after his death. This is what he had
to say about the fighting methods of the Vikings in the late ninth century,
as they emerge from sources written on the continent:
The Northmen... possessed some very special skills of their own. They
could and frequently did use ground which Frankish armies always
avoided if possible. They massed, hid and then emerged from woods...
More important still was the Northmen’s habit of retreating into a stone
building, a church, a house or, on one occasion, the palace of Nymwegen,
and turning it into a fortress from which they could burst forth at their
chosen time when it looked as if they were at bay. Now and again they
attacked at night. Above all they had the capacity, the tools and the skills
to erect quick and effective field fortifications, dykes fortified by stakes,
palisades and advanced ditches. Time and again their enemies were
hampered by these works. The great count Henry, one of their most
ubiqitous assailants and admired everywhere as a great warrior, came to
grief when he rode into what might today be described as a tank trap. As

1
Campbell, ‘The Late Anglo-Saxon State: A Maximum View’, pp. 24-5.
2
An extreme example of this tendency is furnished by Delbrück, History of the Art of War, iii. 154,
who suggested that the hard-fought nature of Hastings was proof enough that the English army was not
composed of peasants, who would either have overwhelmed the enemy with their mass or immediately
taken to flight.
3
Hollister, Anglo-Saxon Military Institutions, p. 130, makes a similar point.
4
Above, p. 121.

138
 4 THE ENGLISH ARMY

he tried to get out, momentarily helpless and at a severe disadvantage, he

was slain.1
There is much that is important here. It is hardly surprising that some of
these features appear in the English sources, although given their relative
lack of detail on military campaigns the harvest in this respect is
inevitably scanty. Even so, there is the fondness for taking the enemy by
surprise (it was an unexpected attack in the middle of winter which
brought Alfred to his knees and almost finished him off in 878) and the
use of woodland, for in 893 the Vikings employed the forestry of the
Weald to screen themselves from English forces, and in 1009, during
Æthelred II’s reign, they passed through the Chilterns on their way from
London to Oxford. Some of their methods the English adopted (or were
already familiar with) themselves: Asser says that in 878 a force of West
Saxons surprised and defeated a besieging Viking force by bursting out
from their defences at dawn; in 893 Alfred’s men launched attacks on
them by both day and night;2 in 1066, Harold took Harald Hardrada by
surprise at Stamford Bridge, and is believed to have attempted to do the
same to the French three weeks later.3 If he also used field defences on
the battlefield,4 the ‘dykes fortified by stakes’ mentioned by Professor
Leyser, this is simply another indication of the intelligence with which
the Anglo-Saxons waged war. That Duke William remained in the
vicinity of Hastings awaiting the arrival of the enemy, and moved
cautiously even after his victory, is a sign not only of his own military
skill but also of the respect which he had for the foe.5
It would be both foolish and impossible to argue that the late Anglo-
Saxon military system always worked like clockwork, for there is no
doubt that it did not. In 1009 Æthelred II assembled a very large fleet at
Sandwich to prevent further Viking attack, but it returned to London
having achieved nothing after a major disagreement between the king’s
magnates; in 1066 Harold’s forces on the south coast eventually had to
disperse when their provisions ran out. Nevertheless, the more that
becomes known about the ways in which English armies assembled and
fought in this period the more it may become apparent that it saw the
mobilisation of the country and its resources for war to an extent that was
not to be repeated until the total wars of the twentieth century. The ability

1
Leyser, ‘Early Medieval Warfare’, pp. 48-9. Margrave Henry of Neustria was killed in 887. Note also
the future king Olaf of Norway’s alleged use of deep, narrow trenches against Breton cavalry in the
early eleventh century, WJ, GND, ii. 24-7. Vegetius warned his readers against sudden irruptions by
those under siege, Epitome of Military Science, p. 138.
2
ASC, i. 74, 84, 139; Asser, pp. 43-4.
3
ASC C and D, i. 197-8; Carmen, ll. 279-83; GND, ii. 166-7; WP, GG, pp. 124-5.
4
Above, pp. 121-4.
5
Glover, ‘English Warfare in 1066’, 2-4; Hollister, Anglo-Saxon Military Institutions, p. 129. Burne,
The Battlefields of England, p. 19, contrasted William’s behaviour with that of Julius Caesar in 54 BC,
who ‘pushed inland in search of his enemy on the very day of his landing’.

139
 THE BATTLE OF HASTINGS

to continue fielding forces despite loss and defeat, the employment of

conscript soldiers in dense and complex infantry formations, the existence
of élite companies, the use of field defences, none of these were issues
which an enemy could afford to regard lightly, and all may do much to
explain why Hastings was so hard fought. Overcome the Anglo-Saxons
may sometimes have been, for their enemies too were formidable, war is
always a lottery, and few systems are immune to problems of personality
and circumstance. Yet contemptible their military capacities were not.

140
 5
THE FRENCH ARMY
The army which was defeated at Hastings was the product of a complex,
wealthy and powerful system of government which had a long history of
success behind it; the one which overcame it was not. It was an ad hoc
force, assembled for the purpose not only from Normandy but also from
other parts of France and perhaps even further afield. The earliest sources
insist on this. Guy of Amiens’ Carmen tells how before the battle the
duke addressed men of Francia, of Brittany, and of Maine, as well as
Normans, apparently including men from their settlements in southern
Italy, and later of the Normans fighting in the centre and the French and
Bretons on the wings.1 William of Poitiers lists men of Maine,
Frenchmen, Bretons, Aquitanians and Normans, and has Breton horse and
infantry and other auxiliaries on the left, apparently Duke William and
the Normans in the centre, and presumably the French on the right; 2 the
annalist of Nieder-Alteich on the Danube also affirms the presence of
Aquitanians, as does the twelfth-century chronicle of the abbey of St
Maixent, in Poitou.3 William of Jumièges says nothing of the make-up of
the duke’s forces, but when Orderic Vitalis revised his work he described
it as a great army of Normans, Flemings, French and Bretons; when he
proceeded to write his Historia Æcclesiastica they became men of Gaul
1
Carmen, ll. 250-60, 413-4. However, Barlow (p. xxxix), while prepared to accept the possibility that
men from southern Italy were involved, prefers to emend the text so that they are referred to as subjects
of the Normans rather than their comrades; the Norman conquest of Sicily, begun in 1061 but planned
as early as 1059, was not finally completed until 1091, Loud, The Age of Robert Guiscard, pp. 146-73.
A horseman from Maine appears in Carmen, ll. 485-6; the messenger who announces William’s arrival
to Harold (ll. 159-60) speaks of Gauls and Bretons.
2
GG, pp. 126-31. William rode in medio of the cavalry which formed his third line, doubtless with
other Normans; Bretons and Aquitanians are also mentioned as being among his forces after his
coronation (pp. 160-1). It is not certain from the phrase pedites pariter atque equites Britanni, et
quotquot auxiliares erant in sinistro cornu, whether the foot and auxiliaries on his left wing as well as
the horse were Breton. WP does not specify the composition of the right wing, but has the Norman
Robert of Beaumont fighting there; Barlow (Carmen, p. lxxvi, n. 252) notes that he was heir to land in
the French Vexin and could have been brigaded with other French troops.
3
Above, pp. 33-4; La Chronique de Saint-Maixent, ed. Verdon, pp. 136-7; see Beech, ‘The
Participation of Aquitanians in the Conquest of England 1066-1100’, 16-20. WP refers to Aimeri of
Thouars (in northern Poitou) as an Aquitanian (GG, pp. 148-9), and it is possible that Poitevins were
the only such men present.

141
 THE BATTLE OF HASTINGS

and Bretons, Poitevins and Burgundians, and other peoples from north of
the Alps.1 Wace, whose Roman de Rou contains much (if of questionable
reliability) on the identities of the duke’s companions, says that he cannot
list the names of all the Norman and Breton barons in the company, and
that there were also many from Maine and Anjou, from Thouars and
Poitou, and from Ponthieu and Boulogne, together with soldiers from
many lands, some there for land and some for money.2 John of Worcester,
more succinctly, claims that William brought with him auxilaries from
the whole of Gaul, while when the Bayeux Tapestry twice describes his
men during the battle of Hastings they are on both occasions Franci:
French.3
There is nothing inherently unlikely in these statements about the
cosmopolitan nature of his forces. He already had a reputation as a
successful soldier and England was known to be very wealthy. Even,
therefore, if it is allowed that some of his non-Normans joined him
because they were political allies, like the Bretons, or subordinates, like
the men of Maine, this still leaves room for numbers of opportunists
ready to make their fortunes. Europe contained many such men in the
middle of the eleventh century. Tostig Godwinsson, for example, was
able to enlist Flemings in the forces with which he attacked England from
Flanders in 1066, and there is no difficulty in believing William of
Poitiers’ statement that as news of the duke’s expedition spread, warlike
men flocked to him, attracted by his very well-known generosity and
confident in the justice of his cause.4 They would have been further
encouraged when they heard of the granting of the papal banner, and in
fact an obvious parallel to their reaction is the assembly of the
international armies of the First Crusade thirty years later, forces which
were motivated not only by piety but also by a strong interest in the
taking of booty. That very few of Duke William’s non-Norman troops are
known by name – one cannot get much further than Count Eustace of
Boulogne, Guy of Amiens’ nephew Hugh of Ponthieu, the Poitevin
Aimeri of Thouars, the Breton Alan Rufus and perhaps Robert, son of
Giffard5 – is not of any great significance, for most of the surviving
names of those who fought in the battle come from William of Poitiers
and Wace, both writing for Norman audiences, and it is thus hardly

1
GND, ii. 164-7; HÆ, ii. 144-5.
2
Above, p. 99; RR, ll. 8659-68.
3
JW, ii. 604-5; the word Franci could be used in two senses at this date: for men who inhabited the
land around Paris controlled directly by the French king, and for those in the area which roughly
comprises modern France. The Tapestry seems to use this second sense, the authors mentioned above
(WP has Francigenae) the first.
4
GG, pp. 102-3; in one of his passages in praise of the duke, WP later claims (pp. 142-3) that he
conquered England without much outside help.
5
Above, p. 77, n. 1.

142
 5 THE FRENCH ARMY

surprising that they are predominantly Norman.1 It is in fact rare to have

more than a handful of names for participants in any battle of the period
before the crusades, and the number of identifiable Normans is not only
greater than that of non-Normans, but also greater than that of their
known English opponents. It is not therefore reasonable to argue that
William had few external auxiliaries simply because their precise
identities cannot be established.2 If the repeated assertions of
contemporary sources about the presence of such men on the expedition
are to be rejected, it must be on other grounds.
Even so, there is no way of knowing what sort of proportion of the
ducal army they formed, and when one comes to the question of its total
size there are dilemmas of a familiar kind to be faced.3 The sources do
have figures to offer: the Carmen speaks of 150,000 men, William of
Poitiers of 50,000 in France and 60,000 just before the battle, and the
Poitevin chronicle mentioned earlier of how ‘they say’ that William had
14,000 men; William of Jumièges gives no size for the army, but says
that it sailed in 3,000 ships.4 Nobody would believe the Carmen here and
hardly anybody William of Poitiers and William of Jumièges, for large
round numbers look rhetorical and thoroughly unreliable. Even 14,000
has on the whole been thought beyond the bounds of possibility, 5 for in
the last decade of the nineteenth century the scholars who favoured
limiting the size of the English army to less than ten thousand men
performed the same service for the French, and on what have long
seemed plausible grounds. Wilhelm Spatz’s Die Schlacht von Hastings,
published in 1896, went into the matter in what appeared a scientific
manner and came up with the conclusion that William led nothing like
14,000 men into battle. His reasons, in addition to the general observation
that medieval armies were smaller than modern ones, were as follows: as
William of Poitiers stresses that the forces delayed at Dives for a month
by contrary winds were so well supplied that they did not plunder the
surrounding country they cannot have been large; if William’s fleet is
1
On Wace’s list, see above, p. 99.
2
There is a detailed treatment of this subject in Körner, The Battle of Hastings, pp. 218-55, who rejects
the idea that ‘William’s army contained thousands of volunteers from all parts of Europe’ on much
these grounds. I agree with Bachrach (‘The Military Administration of the Norman Conquest’, 3, n. 5)
that Körner’s approach was ‘aggressively hypercritical’. It is a curious kind of argument, for example
(p. 234), which acknowledges the presence of Eustace of Boulogne but allows him not a single man
from his own county; on Flemish involvement, see Nip, ‘The Political Relations between England and
Flanders (1066-1128)’, 151-3.
3
Note not only the discussion of the size of the English army in the previous chapter, but also a recent
attempt to assess the numbers of those who went on the First Crusade, upon which far more sources
survive; France, Victory in the East, pp. 122-42.
4
Carmen, ll. 96-7; GG, pp. 102-3, 116-7; GND, ii. 164-5. On La Chronique de Saint-Maixent, see
Bachrach, ‘The Norman Conquest, Countess Adela, and Abbot Baudri’, 65-6.
5
It is accepted by Bachrach, ‘Some Observations’, 3-4, providing that it included 4,000 non-
combatants and represents the number present at Dives in the summer of 1066; he allows William
8,000 for the battle.

143
 THE BATTLE OF HASTINGS

estimated at 1,500 vessels, this and the Tapestry’s evidence that they
carried an average of seven men produces a total of 10,000-12,000; they
landed at Pevensey in a single day when in 1854 the disembarkation in
the Crimea of 60,000 men, just over 1,000 cavalry and 128 cannon took
five days in good weather; an army of 40,000 to 50,000 men could not
have marched the eleven kilometres from Hastings to Battle in the early
morning of 14 October, 1066; Domesday Book suggests that the
Conqueror provided about 4,000 Norman warriors with land in England,
and even allowing that some probably returned home and that many may
have been killed in the fighting, this suggests a force of no more than
7,000-8,000; but as he must have left men to protect his camp, William
probably had no more than 6,000-7,000 against Harold.1 Not much later
Sir James Ramsay judged that 5,000 ‘would satisfy our ideas of what
was at all likely’, and for similar reasons: the entire force was unshipped
in a short October day when it took Henry V three August days to land
8,000-10,000 men at Harfleur in 1415; Edward III never took 10,000 men
across the Channel, while Henry V and Edward IV each did so only once:
the ‘reader must be left to decide’ whether the resources of an eleventh-
century duke of Normandy could have equalled or exceeded those of
English kings of the thirteenth or fourteenth centuries.2 F.H. Baring
agreed that the large figures given by the sources are ‘not to be taken as
arithmetic at all’, being simply ‘rhetorical equivalents’ of ‘many’, and
endorsed Ramsay’s evidence from the Hundred Years’ War and Spatz’s
points about the speed of the French disembarkation and the significance
of a march of ‘at least six miles from their camp’ before the battle began.
He also added a point of his own: having traced William’s route to
London after the battle by the trail of apparently damaged estates
recorded by Domesday Book, he did not think it considerable enough to
imply the activities of any very large army. Thus, his estimate of the size
of the French forces ranged between 5,000 and 10,000 men. 3
Most of these arguments, so attractive on first acquaintance,
disintegrate on close examination, based as they are upon premises and
evidence far too flimsy to support them. Baring’s analysis of the
valuations of Domesday estates, for example, has recently been revealed
as seriously flawed when it comes to reflecting the line of William’s
march (and therefore the size of his army).4 Similarly, Spatz’s use of the
numbers of Domesday landholders of Norman origin proves little, given

1
Spatz, Die Schlacht von Hastings, pp. 27-30.
2
Ramsay, The Foundations of England, ii. 16-17. A slip, of course, for ‘fourteenth or fifteenth
centuries’.
3
Baring, Domesday Tables, pp. 207-20, 227.
4
Palmer, ‘The Conqueror’s Footprints in Domesday Book’. Note, for example (p. 29), that the
Oxfordshire figures ‘provide no evidence whatsoever that the Norman army passed this way, and
strongly imply that it did not. Yet we know that it did’.

144
 5 THE FRENCH ARMY

that those who received land did so over a period of twenty years, and
that there is little difficulty in believing that very large numbers of the
rank and file never did so, and may never have expected to.1 William of
Poitiers stresses the wealth of England, and says that after his coronation
King William gave part of it to those who had helped him win the battle; 2
nor should one forget the booty stripped from the dead, or the plunder
taken from the land before the fight and probably after it. Other points
can be disposed of even more easily. It may well be true that only a small
force could have marched from Hastings to Battle before 9 o’clock in the
morning, but there is actually no good reason to think that this is what
William’s army did, rather than awaiting Harold in the immediate vicinity
during the preceding night.3 Nor can one take seriously statistics based on
average ship capacity as deduced from the Bayeux Tapestry when it is
obvious that this source’s representations of numbers were never intended
to reflect real totals; thus, it is no more plausible to believe that William’s
vessels carried an average of seven men than it is to suppose that they
consisted of only eighteen ships, or that there was only a single English
archer in the battle, or that only two men fled on horses at the end of it.4
Moreover, when it comes to the question of the speed of the French
disembarkation it should be noted firstly that the fact that William is said
to have arrived on a particular day does not necessarily imply that he got
all his troops ashore in a matter of hours, and secondly (and more
significantly) that if we are to play the heady and dangerous game of
historical analogy there is an event not referred to by Spatz and Ramsay
which rather puts those of the Hundred Years’ War and the Crimea in the
shade. In 54 BC Julius Caesar, with five legions and 2,000 cavalry (that
is, probably something over 25,000 men) in over 800 ships, reached the
British coast about midday, landed without opposition (as did the
Conqueror, as far as is known), chose a suitable place for a camp and
marched against the Britons before midnight. It is thus necessary to be
very careful before using evidence from other periods to argue about
what William could or could not have done in 1066, because that
evidence may cut two ways, and be insufficiently analogous to be worth
very much in any case.5

1
See Douglas’ comments on the use of Domesday evidence, ‘Companions of the Conqueror’, 132-3.
2
GG, pp. 152-3.
3
Below, pp. 158-62.
4
I have included in these totals the empty ships which have been beached. The third fleeing horseman
on the restored Tapestry the Benoît drawing (below, p. 182 and the Appendix) shows to have originally
been an infantryman. James, ‘The Battle of Hastings’, 21-2, was also inclined to take the Tapestry’s
depictions of ship capacity literally.
5
Caesar, The Gallic War, v. 8-9, pp. 242-5. Caesar does not say in so many words that he marched
against the Britons on the day of his landing, but it is strongly implied, and in line with his customary
tactic of coming upon the enemy unexpectedly whenever possible. Of course, he disembarked on an

145
 THE BATTLE OF HASTINGS

This leaves only the point that he could not have supported tens of
thousands of men in France without them plundering the countryside
around, if William of Poitiers’ statement that they did not is to be
credited. A recent analysis by Professor Bachrach has suggested the scale
of the logistical problems involved in supporting an army of 14,000 men
and perhaps 3,000 war-horses for a month; he estimates that about 2,340
tons of grain would have been required (including 1,500 for the horses),
together with 1,500 tons of hay and 155 tons of straw, while someone
would also have had to cope with 5 million pounds of equine excreta and
700,000 gallons of equine urine.1 However, these figures have been
seriously challenged: R.H.C. Davis, who thought the horses considerably
smaller than does Bachrach, concluded that William’s forces may have
needed no more than 300 tons of grain and the same of hay while they
were at Dives.2 This not only shows the complexity, and to an extent
invalidity, of the assumptions involved in such elaborate statistical
exercises, it is also a reminder that there are actually no figures on the
resources of northern France, particularly in terms of population levels
and the production of livestock and foodstuffs, available to indicate that
Duke William’s forces could not have been substantially bigger than
14,000 men. Nevertheless, Bachrach’s article is a powerful indication of
the logistical problems that William would have faced, and even if his
totals are too high it is worth noting his conclusion that scholars:
have not fully appreciated the great resources in both wealth and
organisation that were available in Normandy only a generation before the
3
massive European-wide operations that sustained the first crusade.
There is one aspect of Conquest logistics, however, upon which what
looks like an administrative document, complete with some rather
startling statistics, does survive – the size of the duke’s fleet. William of
Jumièges’ assertion that it consisted of 3,000 ships looks impossible, if
not quite as ludicrous as Geoffrey Gaimar’s 11,000, and what seems a
much more plausible and very precise figure is offered by Wace, who
says that his father told him when he was a boy that 696 ships, excluding
the ships, boats and skiffs carrying arms and harness, sailed with the
Conqueror from St Valéry.4 However, to this evidence can be added that
open beach, whereas William’s vessels probably entered harbours in the main; see further on the
French landing, below, pp 152-3.
1
Bachrach, ‘The Military Administration of the Norman Conquest’, 11-15.
2
Davis, ‘The Warhorses of the Normans’, 69, 80. The considerable discrepancy between the figures is
partly, as Davis noted, because Bachrach miscalculated: a daily requirement of 14 tons of grain and the
same of hay does not produce monthly totals of 1,500 tons of each, but 434 tons for 2,000 horses and
620 tons for 3,000.
3
Bachrach, ‘Military Administration’, 21.
4
Above, pp. 62-4 on Gaimar; RR, ll. 6423-32. Wace also notes WJ’s figure. I have followed Edgar
Taylor’s translation of Wace here, but Burgess takes the (apparently ambiguous) meaning of the RR to
be that the vessels carrying arms and harness were included in the total of 696. I am grateful to my
colleague Marie-José Gransard for advice on Wace’s Old French.

146
 5 THE FRENCH ARMY

of the text known as the Ship List of William the Conqueror. It survives
in a hand of c.1130-60 in the Battle Abbey manuscript which also
contains the oldest copy of the house’s Brevis Relatio,1 and is a list of the
number of vessels owed to Duke William in 1066 by fourteen of his
followers, together in the cases of four of them with the number of milites
also due. Until recently treated with care because of the date of the
manuscript and the rather surprising nature of its evidence, it has been
rehabilitated in the work of Professor Hollister and Dr van Houts, the
latter believing that the copy goes back to a text created in c.1072 or
perhaps as early as December 1067 at Fécamp on the basis of information
compiled just before the Conquest by monks of that abbey in the ducal
service.2
The duke’s half-brothers, Robert of Mortain and Bishop Odo of
Bayeux, owed 100 and 120 ships respectively, others fewer, but the
(unstated) total comes to 776 and 280 milites. The list then says, wrongly
(though its enumeration of magnates may have become truncated during
transmission) that the total was 1,000, and that the duke also had many
other ships from certain of his men, according to their means. Now Wace,
too, had heard of William’s magnates supplying specific numbers of
vessels, and like the list assigns sixty to William FitzOsbern and unlike it
forty to Bishop Odo, adding that the bishop of Le Mans provided thirty
vessels and their crews (thirty being a number which appears once in the
list, attributed to Walter Giffard). As Dr van Houts has said, Wace and
the list differ sufficiently to appear independent of each other, and despite
the inconsistency in the numbers assigned to Bishop Odo this tends to
bolster confidence in their general reliability.3 On this reckoning, then,
the duke’s fleet was sizeable, for if the list’s statement that he was owed
1,000 ships by a portion of his men and more by others is accepted, and
the likelihood considered that some of his continental allies (including the
Aquitanians involved in the alleged naval battle of 1066)4 supplied their
own craft, then William of Jumièges’ total of 3,000 does not look quite so
ridiculous as it once did.
Moreover, such a conclusion might be supported by what is known of
the anchorages employed. William of Poitiers says that the fleet
assembled at the mouth of the river Dives and neighbouring ports (my
italics), was delayed for a month by unsuitable winds and then blown by
westerlies to St Valéry at the mouth of the Somme. Now the Gulf of

1
Oxford, Bodleian Library, MS e Museo 93; see BR, pp. 8-9; van Houts, ‘The Ship List of William the
Conqueror’, 164-6; and compare Bachrach’s comments on the probable keeping of written
administrative records in 1066, ‘Military Administration’, 19-20.
2
Hollister, ‘The Greater Domesday Tenants-in-Chief’, pp. 221-6; van Houts, ‘Ship List’, 166-8. It did
not, of course, escape Freeman’s notice, FNC, iii. 378-80.
3
RR, ll. 6119-22, 6163-7; van Houts, ‘Ship List’, 161-4, 168.
4
Above, pp. 33-4.

147
 THE BATTLE OF HASTINGS

Dives may at this date have provided a very large harbour indeed, for
Wace claimed that the river flowed into the sea close to Bavent, which is
some five miles from Dives-sur-Mer, where the entrance to the English
Channel is situated today;1 together with the ‘neighbouring ports’ it may
have been able to shelter many hundreds of vessels. The same is true of
the Somme estuary at St Valéry, and probably of the harbours around
Pevensey where William landed. The presence of sand and alluvium in
localities that are today well inland shows that they were once prone to
flood or were open water, and Domesday Book records the existence of
salt-pans, and thus presumably of salt-water, at (from west to east, in a
clockwise direction) Eastbourne, Willingdon, Bowley, Hailsham,
Netherfield, Ashburnham, Wartling and Hooe.2 Hailsham is now 5¼
miles from the seashore. One reconstruction of the coast that William
found (and doubtless knew that he was going to find) shows the bay itself
with an entrance four miles wide and penetrating inland for about six
miles, while the anchorage at Bulverhythe, west of Hastings, may have
had an entrance about two-thirds of a mile wide and a penetration of
about two and a half miles.3 While all this is far from certain, as much of
Pevensey Bay may in fact by 1066 have been saltmarsh,4 what matters is
the likelihood that there was more than enough space and protection for
the French ships. Did William use such large anchorages because the size
of his expedition required it?5
Even if there were a reliable figure for the numbers of vessels in his
fleet we could not deduce the numbers it carried without negotiating the
quicksand of average ship capacity, a matter upon which our ignorance of
the nature and range of eleventh-century vessels prohibits any worthwhile
conclusion; even so, just how far wide of the mark Spatz’s estimate of
seven men per ship might be is demonstrated by the account of a miracle
said to have taken place in the time of Abbot Baldwin of Bury St
Edmunds (1065-1097/8) involving a vessel carrying almost sixty men,
thirty-six beasts and sixteen horses laden with merchants’ baggage.6 Still,
1
GG, pp. 102-3, 108-9; RR, II, l. 2882; Gillmor, ‘Naval Logistics of the Cross-Channel Operation,
1066’, 107; Bachrach, ‘Military Administration’, 6, 10-11 (his useful map has an inaccurate scale); The
form of the lower Dives at this date is not, of course, known with any precision; see R.N. Sauvage,
L’abbaye de Saint-Martin de Troarn, pp. 245-52, who thought it un vaste lac envahi par les marées,
and his map from c.1760, below, Illus.74.
2
DB, 20c, 21a, 22b, 22a, 18d, 18b, 18a, 18a; see the map in Darby and Campbell, The Domesday
Geography of South-East England, p. 457. At Netherfield, at least, the salt-pans were probably some
way from the village, although alluvium reaches to Penhurst, a mile and a half to the south.
3
These calculations have been made from the map prepared by J.A. Williamson and reprinted M&M,
p. 110. Seventeenth-century maps show a significant inlet at Bulverhythe.
4
Brandon, The Sussex Landscape, p. 111.
5
The use of St Valéry may not have been part of his original plan, below, pp. 151-2.
6
Herman the Archdeacon, Liber de Miraculis Sancti Eadmundi, ed. Arnold, i. 72. Herman wrote after
Abbot Baldwin’s death. I owe this important reference to James Campbell. Bachrach, ‘On the Origins
of William the Conqueror’s Horse Transports’, argues for vessels carrying twenty men and horses plus
crew.

148
 5 THE FRENCH ARMY

Still, this is not quite the end of the Ship List. Professor Hollister noted
that William of Poitiers’ Gesta Guillelmi names seven prominent Norman
lay magnates who gave the duke the benefit of their counsel during 1066,
and that if Richard, count of Evreux, is replaced by his son William, who
fought at Hastings and is named in the list, then the seven correspond
precisely with the eight (including Odo, a bishop) premier suppliers of
vessels according to the List; thus one ‘might almost suppose that
William of Poitiers was writing with some such ship list at hand’. 1
Certainly one might, and one might also compare his comment that
Agamemnon set out with 1,000 vessels but the duke had more with the
list’s total of 1,000 and ‘many other ships’.2 Such ideas could be taken
further. If monks of Fécamp were with William’s army keeping records
(and one is said to have acted as a messenger to Harold before the battle)3
could this be where William of Poitiers obtained his figures of 50,000 for
the number supported by the Conqueror at Dives and 60,000 for those he
commanded just before Hastings? The credibility of numbers like 60,000
is very difficult to ascertain. Ferdinand Lot noted that 60, 600, 6,000 and
60,000 appear frequently in medieval chronicles, and nor are they absent
from works by Julius Caesar which William of Poitiers was eager to
imitate.4 But this is not to say that he was in this matter influenced by
Caesar or simply made use of random round numbers, for what might
seem like random round numbers also appear in early medieval
assessment systems and lists which were not works of exaggeration and
fantasy, but intended for the levying of taxation of various kinds, always
a serious matter requiring precise records. Thus, the Anglo-Saxon
document known as the Tribal Hidage deals repeatedly in recurrent round
numbers of varying orders of magnitude including 600, while 60 appears
six times in the Ship List (including one reference to milites) and is
known also from the Cartae Baronum of 1166, believed to record the
quotas of knights which William required from his tenants in England.5 If
If 60,000 was not the number of men he actually had in 1066, it could
have been the total recorded by a scribe in his service as the number he
thought he ought to have had.

1
GG, pp. 100-3; Hollister, ‘Greater Domesday Tenants-in-Chief’, p. 223.
2
GG, pp. 110-11. WP could have been thinking of WJ’s 3,000, although there is little evidence that he
used WJ on the events of 1066, Davis, ‘William of Poitiers’, p. 78.
3
GG, pp. 118-21; van Houts, ‘Ship List’, 168. Wace, RR, ll. 6757-61, claimed that the Fécamp monk
Hugh Margot was sent by William to Harold in London.
4
Lot, L’Art Militaire, i. 285, n. 2; The Gallic War: 60: v. 5, 23; 600: ii. 15, 28; iii. 22; v. 2; 6,000: i.
27, 48; ii. 29; iv. 37; viii. 17; 60,000: ii. 4, 28; v. 49; vii. 83; pp. 238-9, 262-3, 108-9, 126-7, 168-9,
234-5, 40-1, 78-9, 126-7, 226-7, 540-1, 94-5, 126-7, 298-9, 502-3. WP (GG, pp. 170-1) gives a figure
of 100,000 for the men who accompanied Caesar on the expedition of 54 BC, a figure which is
exaggerated if WP knew the size of a Roman legion (Caesar says that he had five); note also Round,
Feudal England, pp. 290-2.
5
Dumville, ‘The Tribal Hidage’, p. 227; Round, Feudal England, pp. 249, 251, 253.

149
 THE BATTLE OF HASTINGS

It is fair to say, however, that most historians would regard this as an

impossible figure for the size of his army, many of them being wedded to
the idea that medieval military forces can be safely assumed to have been
small unless there is overwhelming evidence to the contrary, an idea
which owes much to the views of the German scholar Hans Delbrück, to
whom Spatz’s book on Hastings was dedicated. On the face of it, of
course, this is reasonable enough, but in a period where overwhelming
evidence of any kind is often in short supply there is a danger that the
occasional existence of armies of tens of thousands of men may be
masked by the limitations of the sources, especially when they are linked
to the assumption that any large figure cannot be trusted. Also, while
analogies can prove nothing about the size of the French forces in 1066, it
is worth noting two cases where the dismissal of large figures as simply
wild exaggeration is not completely convincing. In 1106 the Conqueror’s
sons, Robert and Henry, met in battle at Tinchebrai in Normandy. The
latter’s forces were much the bigger, and were described shortly
afterwards in a letter by a priest of Fécamp as having gone into battle in
two lines of infantry, the second containing King Henry and his barons;
700 horsemen were placed with each line, and on the flanks were forces
from Maine and Brittany of about a thousand horse; the whole of
Henry’s army, says the priest, was about 40,000 men, compared with his
brother’s 6,000 and 700 cavalry.1 Is an author who gives plausible figures
for the cavalry really wildly inaccurate on the size of the forces as a
whole? Similarly, it is possible to believe from contemporary sources that
of the numbers engaged in the Wars of the Roses battle at Towton in
Yorkshire in 1461 some 28,000 were left dead on the field.2
Of course, Duke William may have been well aware that the Late
Roman military writer Vegetius warns against using forces that are over-
large, commenting on the logistical difficulties of keeping them supplied,
and stating that the ancients wished to have armies that were not so much
numerous as well-trained.3 Ultimately, one can only conclude that the
duke would have had a better idea than ourselves of the size of the
English force he might have to meet in battle, and presumably thought
that he was taking with him enough men to offer a chance of victory.
They may have numbered the 7,000 or so that is all most historians have
been willing to allow, but considering the widespread territories from
which they were drawn and the evidence that the fleet which carried them

1
Priest of Fécamp, Letter on the Battle of Tinchebrai.
2
Blood Red Roses, ed. Fiorato et al, pp. 15, 25; note also, by way of exception, K.F. Werner’s estimate
of the numbers available (from a very large area, of course) to Carolingian rulers in the ninth century
as 35,000 cavalry and 100,000 infantry, as against earlier suggestions of about 5,000 in all (cited
Contamine, The Art of War, p. 25); also Bachrach’s criticisms (‘William Rufus’ Plan’, p. 62) of
medievalists who are ‘in the thrall of Hans Delbrück’s obsession for small numbers’.
3
Vegetius, Epitome of Military Science, pp. 64-5.

150
 5 THE FRENCH ARMY

was very sizeable, tens of thousands are within the realms of possibility,
as indeed is the case with the forces fielded by their opponents. It is
perhaps also worth bearing in mind that contemporaries insist that the
Norman invasion of southern Italy in the same period was motivated
partly by over-population at home, that there is other evidence that the
duchy’s population was expanding at the same time as an economic boom
was taking place, and that such phenomena may also have had their effect
on the events of 1066.1
According to William of Poitiers, William took counsel with his men
when he heard of Harold’s usurpation of the English throne, disregarded
their opinion that a military expedition was beyond the resources of
Normandy, and oversaw the building and equipping of the ships. He
supplies no details of the mechanics of this process, to which the Bayeux
Tapestry devotes some of its most interesting scenes, but Orderic Vitalis
adds that clerics and laymen expended their energy and wealth in
construction, and this seems to tie in with the evidence of the Ship List.2
From there the sources move swiftly to William’s embarkation. William
of Poitiers, after recording the capture of an English spy and the duke’s
further refusal to be swayed by the doubts of his men, is the only
contemporary source to give any very detailed account of the
circumstances in which the French crossed the Channel; they were, he
says, delayed for a month at the mouth of the Dives by contrary winds,
and subsequently blown by westerlies to the port of St Valéry at the
mouth of the Somme, a number of vessels being wrecked on the way.
William then had the body of the saint brought from the church in an
attempt to secure a favourable wind, and eventually it blew. 3 According
to the Carmen, which also mentions the unfavourable conditions which
held him in Normandy, bad weather kept him in the Somme estuary for a
fortnight; William of Jumièges, however, says nothing of Dives or of
contrary winds at St Valéry.4 It is not, therefore, surprising that some
historians have disbelieved this element of the accounts by William of
Poitiers and the Carmen, arguing that the six weeks’ delay was deliberate,
and useful in allowing the duke to organise, provision and train his
motley force while Harold waited to no purpose on the south coast of
England.5 It may also be that William knew of Harald Hardrada’s
1
Loud, The Age of Robert Guiscard, pp. 84-5; Bates, Normandy before 1066, pp. 96-8.
2
GG, pp. 100-3; HÆ, ii. 144-5. Note also WM, GR, i. 448-9, and BR, p. 29; these references are given
by van Houts, ‘Ship List’, 159-64, who also discusses the account of William’s councils in Wace’s RR.
Councils were held at Lillebonne, Bonneville and (in June) Caen; Douglas, William the Conqueror, pp.
184-5.
3
GG, pp. 102-3, 108-11.
4
Carmen, ll. 40-77; GND, ii. 164-5.
5
GG, pp. xxiv-vi; Gillingham, ‘William the Bastard at War’, p. 155, notes the suggestion of Gillmor,
‘Naval Logistics’, 124, that the move to St Valéry came when William heard of the dispersal of
Harold’s forces on 8 September. It may also be a matter for concern, considering Guy of Amiens’ and

151
 THE BATTLE OF HASTINGS

expedition and hoped, as proved to be the case, that by delaying he would

eventually face a weakened enemy, whether English or Norwegian. Yet
none of this is certain. Purposely delaying a Channel crossing until late
September would have been taking very considerable risks with the
weather if conditions had been suitable earlier, and it has been argued on
nautical grounds that the diversion to St Valéry was unintentional, as
William of Poitiers implies, and put the duke in a thoroughly unenviable
position.1 It is true that it offered a shorter crossing to England than did
Dives, but this may have increased the risk of interception by the English
fleet, and sailing from the latter would have given him the option of
striking east to Pevensey Bay or Chichester Harbour, west to Portsmouth
or Southampton, or further west still to Poole, one of the biggest natural
anchorages in the world. On the other hand, a longer crossing may also
have meant, if the time necessary for embarkation and disembarkation is
added, that men and animals would have had to spend two nights on
board rather than one, something which all may have wished to avoid.2
Even if William’s fleet was a very large one the south coast of England
offered many possibilities, and Harold covered them as well as he could
by stationing himself centrally on the Isle of Wight, and levies
‘everywhere along the coast’.3 If William sailed for that coast in the
middle of September and was instead blown to St Valéry, incurring losses
on the way, then his entire enterprise came somewhere near to disaster,
and it says a great deal for his powers of leadership, not to mention
ultimate good fortune with the weather, that he was eventually able to
make a successful crossing a fortnight later. There must have been times
when he remembered how his father’s naval expedition against England
had been brought to nothing by a storm.4
The accounts of the crossing given by the Carmen and William of
Poitiers are very similar. The former implies that the French set sail late

William of Poitiers’ love of classical parallels (see above, pp. 83-4), that both Caesar’s crossings to
Britain took place at night (as the Conqueror’s is said to have done, see below) and that in 54 BC he
was delayed by contrary winds for about 25 days, The Gallic War, iv. 23, v. 7-8, pp. 210-11, 244-5.
1
C. and G. Grainge, ‘The Pevensey Expedition’; the arguments which these authors adduce from WP
and the Carmen, together with their stress on the nautical dangers of the lee shore which the area
around St Valéry represented in the prevailing westerly winds, seem convincing.
2
I am grateful to my colleague Stephen Lowden for this point. However, it may be, as Herman the
Archdeacon’s casual reference to a vessel carrying over fifty beasts rather implies (above, p. 148), that
historians have tended to overestimate the problems the French faced in transporting their horses.
3
ASC, C, i. 196; WP makes a similar statement about the levies, GG, pp. 106-7; OV, HÆ, ii. 168-9,
specifically names Hastings and Pevensey as being among the ports that Harold guarded. Note also
Cnut’s use of Wight as a naval base in 1022, Lawson, Cnut, pp. 92-4. The Grainges, ‘The Pevensey
Expedition’, pp. 133-4, 140, find it puzzling in nautical terms, but may underestimate French
willingness to beat against westerly winds to reach Portsmouth or Poole, or at least Harold’s fear that
they might do so; it is difficult, of course, knowing how far the vessels were rowed as well as sailed. In
the prevailing westerlies, being on Wight would also have put him advantageously to windward of
ships attempting a crossing to the east.
4
WJ, GND, ii. 76-9; above, p. 22.

152
 5 THE FRENCH ARMY

in the day, and says that nightfall was already close when William’s
vessel forged ahead of the rest; the fleet was then ordered to heave to
until dawn, and once under way again made its landing in England at 9
o’clock as a comet prophesied the defeat of the English. William of
Poitiers agrees that they anchored at sea, adding that William’s ship
subsequently became detached from the others, whereupon he calmly had
a good meal as he waited for them to appear; when they had done so, they
were all carried by a favourable wind to Pevensey, and there and at
Hastings they built fortifications to protect themselves and their ships;
William of Jumièges reports that those at Pevensey were strongly
entrenched and that William went rapidly to Hastings to build the
structures there; the Carmen, which gives no place-names, says that they
repaired the remains of earlier defences.1 The Bayeux Tapestry, having
shown the building of ships and weapons and provisions being carried
down to them, says that William himself sailed in a great ship and depicts
a number of vessels, most of which contain horses, landing at Pevensey;
the soldiers then hurried to Hastings to seize food, and after they had
enjoyed a meal blessed by Bishop Odo began building a castle there. The
D version of the Anglo-Saxon Chronicle reports that William came from
Normandy to Pevensey on Michaelmas Eve, 28 September, and the Battle
Abbey Brevis Relatio concurs; the E text says that he landed at Hastings
on the following day, that of St Michael; the Carmen agrees that
Michaelmas was approaching when the weather changed for the better,
and William of Poitiers (followed by Orderic Vitalis) has him leaving on
Michaelmas Day.2 These dates are not, of course, entirely consistent,
especially if he left in the evening and arrived the following morning.
Only at a later stage in his account does William of Poitiers mention men
who came ashore at Romney by mistake and were scattered by the
English after heavy losses on both sides, a clear indication both that the
crossing did not go entirely smoothly and that parts of the coast were
defended despite Harold’s absence in the north.3
The great bulk of the French force probably consisted of experienced
soldiers, few of whom are likely to have been as poorly equipped as
English peasants bearing no more than spears or clubs.4 However, nor
should one think of William’s army as consisting principally of well-
armed men in chain-mail5. The Bayeux Tapestry suggests that most of the
archers did not wear armour, and the same may have been true of much
of the rest of his infantry. It is not at all clear that there existed at this date
1
Carmen, ll. 78-126, 140-4; GG, pp. 110-15; GND, ii. 166-7; the Grainges, ‘The Pevensey
Expedition’, p. 135, think that William would have left St Valéry at high tide.
2
ASC, D and E, i. 198-9; BR, pp. 29-30. Carmen, ll. 76-7; GG, pp. 168-9; HÆ, ii. 170-1.
3
GG, pp. 142-3, and n. 6.
4
Above, pp. 121, 138.
5
On the nature of metal armour, see above, p. 66,. n.3.

153
 THE BATTLE OF HASTINGS

in Normandy a military system by which troops were levied from

landholders by means of fixed quotas imposed from above by the ducal
administration. Nevertheless, the duchy ‘was full of vassals and mailed
knights’ and of mercenaries and the military households of the duke and
the great lords;1 when William needed to field an army he summoned
men to his service,2 and just as he expected his magnates to provide ships
for his expedition to England in 1066 (as his father Duke Robert had
when he decided to support the brothers Edward the Confessor and
Alfred against Cnut)3 he must also have expected them to provide men,
although it is likely that many of these troops anticipated some sort of
remuneration for their activities sooner or later. The penitential drawn up
for the army shortly after the battle of Hastings appears more lenient to
those who fought in a public war (i.e. presumably as a result of an order
from the duke) than to those who fought merely for gain (i.e. probably as
mercenaries), and clearly some and perhaps many of William’s men came
into this second category.4 However, this is unlikely to have meant that
their leader regarded them as free agents. During hard campaigning in
England over the winter of 1069-70 men from Anjou, Maine and Brittany
asked to be discharged, but William refused, and when he did eventually
dismiss his army, which received lavish rewards, he detained the
fainthearts for an extra period.5 Many of those who sailed with him had
doubtless served previously, at least in the expeditions against Maine in
1063 and Brittany in 1064, and would have comprised a range of troop
types – archers and crossbowmen, engineers and miners6 capable of
laying siege to fortifications, light and heavy infantry, and of course
heavy cavalry. Such men were probably just as skilled in military
methods as their English opponents, even if far too little evidence
survives on Norman warfare before 1066 to make possible even the sort
of treatment attempted for those opponents in the previous chapter; it is a
particular problem that there is little evidence on how infantry were
deployed in battle. Even so, the tactic of bursting from a fortress to take
besiegers by surprise, noted by Professor Leyser as characteristic of the
Vikings, is known to have been used by the Norman garrison of Tillières
against a besieging force in the time of Duke Richard II, by Bishop Odo

1
See Chibnall’s very useful, ‘Military Service in Normandy before 1066’, (p. 66 for the quotation); and
for reservations about the translation of the word miles as ‘knight’, below, p. 179, n.2.
2
Note, for example, WJ’s comments on his assembly of Normans from all sides to deal with the
rebellion of Thurstan Goz in c.1043, GND, ii. 102-3.
3
WJ, GND, ii. 76-9.
4
Penitential articles issued after the Battle of Hastings, c. 5. Douglas, William the Conqueror, pp. 191-
2, used this as evidence that most of the duke’s non-Norman troops were ‘simple mercenaries’. There is
doubt over the precise meaning of this chapter, in a corrupt text.
5
OV, HÆ, ii. 234-7, probably using the lost ending of WP; see Chibnall, ‘Military Service’, pp. 89-90.
6
OV, HÆ, ii. 212-3, speaks of attempts to undermine the walls of Exeter in 1068.

154
 5 THE FRENCH ARMY

of Bayeux when defending Dover against the English in 1067, and by the
men holding the castle at Exeter during the English revolt of 1069.1
Along with evidence of their cavalry’s ability to feign flight (see below),
this can reasonably be taken as typical of the professionalism, intelligence
and competence with which the Normans waged war.
What sort of proportion of William’s force the heavy cavalry formed
there is no way of knowing.2 Their dominant role in the Bayeux Tapestry
should not lead to overestimation of their numbers, and certainly the
suggestion that a third of the entire army was cavalry is difficult to credit;
analogies prove nothing, but it is worth noting that under 10% of the men
whom Caesar led against Britain in 54 BC were horsemen, and that if the
priest of Fécamp’s figures are accepted far less of King Henry’s army
fought as cavalry at Tinchebrai.3 Yet whatever their numbers the threat
posed by William’s horse was something the English must have known
they would have to take seriously. Mounted on expensive animals
probably bred for the purpose, it is likely enough that they were highly
trained and operated in units of perhaps five or ten known as conrois.4
Doubts have sometimes been expressed whether they would have been
capable of carrying out the feigned retreat which many of the sources on
Hastings describe, but they are not very reasonable doubts given
extensive evidence of the use of such manoeuvres in the early medieval
period, and in particular William of Jumièges’ statement that during
fighting against the troops of King Henry of France in 1053 Normans led
some of them into a trap by feigning flight.5

1
Above, pp. 138-9; WJ, GND, ii. 22-5, 176-9; OV, ii. 228-9. De Boüard, Guillaume le Conquérant, p.
205, suggested that the rapid mobility characteristic of Norman warfare in the eleventh century was a
legacy of their Viking forebears.
2
Bachrach, ‘On the Origins of William the Conqueror’s Horse Transports’, n. 4, stresses that ‘there is
no source that even comes close to providing a reasonable figure’.
3
Morillo, Warfare under the Anglo-Norman Kings, p. 122, suggests an army of 5,000 to 6,000 men,
including 2,000 cavalry possessed of ‘perhaps 3000-6000 horses’, i.e. three horses per man. Brown,
The Normans and the Norman Conquest, p. 150, wrote of probably 2,000 to 3,000 horsemen out of
perhaps 7,000 men; Bachrach, ‘Military Administration’, 5, reckons with 2,000 to 3,000 horses. It
might be objected, of course, that 54 BC is not a fair comparison, as the Normans valued cavalry more
highly than did Caesar, who relied on heavy infantry; Vegetius, Epitome, p. 65, suggests that cavalry
might ideally have made up a fifth of Roman armies; also, while Henry and his barons fought on foot at
Tinchebrai, this was clearly regarded as unusual.
4
Verbruggen, The Art of Warfare in Western Europe, p. 12, gives evidence on the training of heavy
cavalry under the Carolingians; see also the valuable discussion of Carolingian cavalry by Bullough,
‘Europae Pater’, 84-90. The history of the conroi before 1066 seems thoroughly obscure, for the word
is a vernacular one and does not survive so early; Chibnall, ‘Military Service’, pp. 87-8, states that ‘not
until the vernacular became the language of narrative does the conroi clearly emerge from the classical
verbiage’, and that knights were trained in ‘small groups of five or ten, combined in larger units under
their magistri militum’; Morillo, Warfare under the Anglo-Norman Kings, p. 70, n. 145, rejects the idea
that there was ‘decimal organization for small groups of cavalry’. On the horses themselves, see Davis,
‘The Warhorses of the Normans’, who suggested that the standard size may not have been more than
fourteen hands tall.
5
GND, ii. 104-5; cited by Brown, The Norman Conquest, pp. 51, 171-2 (n. 147), and Gillingham,
‘William the Bastard at War’, p. 154, who notes that the duke had long practised ‘a type of warfare...

155
 THE BATTLE OF HASTINGS

The duke himself, almost forty at the time of the battle, had been
gaining experience of the stratagems of war since his youth. However,
this did not include the frequent command of sizeable forces in set-piece
conflicts, for in fighting in northern France in the middle of the eleventh
century these were relatively rare events. Professor Gillingham has
stressed the reluctance of commanders to stake their all on such
hazardous throws of the dice, and argued that as William was junior to
King Henry in their defeat of Norman rebels at Val-ès-Dunes in 1047,
and as Varaville (against Henry and Geoffrey of Anjou) in 1057 was ‘in
fact not a battle’, William had ‘no previous experience of command in a
set battle’.1 This is going a little far. It is true that William of Poitiers
seems to exaggerate the duke’s role at Val-ès-Dunes as against that of
Henry, and that William of Jumièges gives the impression that the French
king was the senior partner;2 nevertheless, he says that they launched a
counter-attack together, and if William was not in sole command he was
at least present and presumably responsible for the direction of his own
men in a serious engagement of great consequence. At Varaville, William
of Jumièges says that King Henry had crossed the river Dives and that the
incoming tide prevented the second part of his army following him,
whereupon William, who arrived upon the scene with alacrity, cut them
to pieces before his enemy’s eyes; as a result the latter withdrew. 3 He
might have been surprised to know that he had not witnessed a battle, and
one in which his opponent had seized a passing opportunity with great
determination. Nor should the parallel with Hastings, where William is
said to have attacked the English before they were properly ordered,4 pass
unnoticed. It might also be pointed out that sieges of fortifications, while
not field engagements, did often see heavy fighting.5
When he arrived in England the duke took a step that was very unusual
up to that time, at least among leaders intent upon conquering the whole
country: He established a fortified camp (in the best Roman6 and Viking
tradition) by building defences at Pevensey and Hastings, and waited for
Harold to attack him. This contrasts with Harald Hardrada’s immediate
advance on York, but whereas the latter had hopes that the

dependent upon good group discipline’ and that ‘it is hard to envisage a type of warfare in which tricks
like feigned flights would be more natural and more frequently practised’; for such flights more
generally, see Bachrach, ‘The Feigned Retreat at Hastings’. I agree with him in rejecting Lemmon’s
belief (‘The Campaign of 1066’, pp. 109-10) that medieval accounts of such events are merely the
covering up of real flights.
1
Gillingham, ‘William the Bastard at War’, pp. 143-4.
2
WP, GG, pp. 10-11; WJ, GND, ii. 120-3.
3
WJ, GND, ii. 150-3; similarly, WP, GG, pp. 54-7.
4
Below, p. 169.
5
Note, for example, that at the siege of Falaise c.1043, WJ, GND, ii. 102-3.
6
See, for example, Vegetius, Epitome, pp. 23, 79-83.

156
 5 THE FRENCH ARMY

Northumbrians would throw in their lot with him, 1 William probably

judged that he was not likely to achieve much until a decisive defeat had
been inflicted upon the enemy. He may also have feared that if he left his
ships relatively unattended they would receive unwelcome and potentially
disastrous attention from the English navy; William of Poitiers speaks of
Harold preparing a fleet of up to 700 ships (other sources of fewer), and
nothing is more likely than that there was an attempt to reactivate English
vessels once William had landed.2 During this waiting period the French
ravaged the surrounding countryside (the Bayeux Tapestry shows the
burning down of a house), partly to secure provisions (the Carmen says
that they took all the cattle) and partly perhaps to provoke the English
into action; certainly William of Poitiers claims that Harold hastened his
march on that account, and the damage done to the surrounding area is
evident in the statistics recorded by the Domesday commissioners twenty
years later.3 Probably the French also continued to train, as they waited
for their scouts to report the approach of the enemy.
If the earliest French sources are to be believed, that approach was
accompanied by exchanges of ambassadors. The Carmen says that Harold
sent an eloquent monk to scout out William’s camp and deliver the
message that he must leave immediately, a request which was, of course,
rejected. The following day William dispatched one of his monks to
Harold: he urged the justice of his master’s claim to the English throne
and received the reply that on the day following God would make his
own judgement; on his return to William he reported that Harold, who
was at the head of a vast army, hoped to catch him unawares; as he
listened, William was already deploying his men for battle.4 William of
Poitiers has a slightly different version of events: according to him, when
William met the English envoy he received Harold’s message while
pretending to be his own steward, and then the following morning
received it formally as part of an assembly of his followers. As he
returned to his master with the duke’s (negative) answer, the envoy was
accompanied by a monk of Fécamp who had been instructed by William
to lay out his claims to the throne and offer Harold single combat so that
innocent men should not die in battle. Harold received this message as he
was advancing and supposedly lifted his face to heaven and asked that
God would that day decide what was just between himself and William;
meanwhile, William’s scouts reported the immediate approach of the

1
Above, p. 34. Compare also the actions of the army which King Æthelstan defeated at Brunanburh in
937, and those of Swegen Forkbeard in 1013 and Cnut in 1015.
2
GG, pp. 124-5; the Carmen, l. 319, says 500 ships; OV, ii. 172-3, reduces the number to 70.
3
Carmen, l. 166; GG, pp. 122-5; Palmer, ‘The Conqueror’s Footprints in Domesday Book’, p. 26: the
rape of Hastings lost about three-quarters of its value after 1066, and its three coastal hundreds over
80%.
4
Carmen, ll. 203-334.

157
 THE BATTLE OF HASTINGS

enemy.1 These extensive preliminaries allowed both authors to heighten

their audiences’ anticipation of their accounts of the battle itself, to
remind them of the justice of the duke’s cause and to put Harold in a bad
light, but William of Poitiers specifically states that he had gone to some
trouble to discover the content of William’s message to the English king
rather than confecting it himself, and it is not particularly easy to regard
the monk of Fécamp as a complete invention; nor is it inherently unlikely
that as Harold approached there should have been negotiations between
two leaders who only about two years previously, during his visit to
Normandy, had spent some time in each other’s company, and ostensibly
been on good terms.
Both the Carmen and the Gesta Guillelmi agree that Harold attempted
to take William by surprise, and the same belief is also to be found in
William of Jumièges’ Gesta Normannorum Ducum and may well be
implied by the Anglo-Saxon Chronicle E text’s statement that he fought
before all his army had come. Round was correct in commenting that the
events immediately preceding the battle are ‘doubtful and difficult to
determine’.2 William of Poitiers says that when his scouts reported the
approach of the English William hastily assembled the men in his camps
(which must surely mean Pevensey and Hastings rather than camps
further north) as most had gone out foraging, took communion, hung
around his neck the relics upon which Harold had sworn his oath, put on
his mailcoat (at first reversed), made a speech to his men and then
advanced into battle.3 The implication of this, of course, is not only that
the foragers took no part in the battle, but that those who did must have
marched from the coast to Senlac before fighting began, although the
source says nothing of any march.4 Even if it stood alone this would not
be a particularly convincing account of events, which in its emphasis on
the French foragers, whom any competent commander would have called
in once he knew the English were in the vicinity, may be thought to have
been distorted by its author’s desire to draw parallels between the
Conqueror and Julius Caesar’s battles with the Britons in 55 and 54 BC,
in which Roman foragers figure rather prominently; at a later stage in his
history, William of Poitiers used his familiariy with the The Gallic War to
stress at length how the Conqueror’s achievements in England surpassed
those of Caesar.5 In any case, it is explicitly contradicted by William of
Jumièges’ statement that William ordered his men to stand to arms from

1
GG, pp. 116-23.
2
Round, ‘Wace and his Authorities’, 683.
3
GG, pp. 124-7.
4
In the translation of GG in The Norman Conquest, ed. Brown, the editor (p. 32) footnoted the day
upon which men were out foraging as Friday, 13 October, the day before the battle. This would get out
of the difficulty, but there is no warrant for it in WP’s text.
5
Above, pp.. 84, 86..

158
 5 THE FRENCH ARMY

dusk to dawn because of fear of a night attack, and that when dawn had
broken he drew them up and advanced against the enemy. This gives the
impression that the French had reached the vicinity of Senlac the day
before, if not earlier, and camped there overnight; nor is this at all
unlikely for other reasons. In 1066 the area which William’s army had
occupied probably formed a peninsula protected to the east by the estuary
of the Brede and to the north by marshland through which the only viable
route was the high ground which runs south through Battle to Telham and
then on to Hastings.1 Harold could no doubt have reached Pevensey by
making a wide sweep to the west around Pevensey Bay via the South
Downs, but William must by this stage have known that he was not doing
so. To protect his fortifications and ships, and because he needed a
decisive victory, it would therefore have made sense for him to block the
route along the high ground by advancing some way along it and offering
battle in that area. If his army bivouacked in the vicinity of Blackhorse
and Telham hills at a height of between 100 and 140 metres, and possibly
in the area lying south of the present Powdermill Lane (the road from
Battle to Catsfield), some of his troops may from the northern slopes (i.e.
perhaps in the vicinity of the modern Starr’s Green) have had a good
view of the ground beyond, dominated by Caldbec Hill and the ridge to
the south of it upon which Harold placed his standards.2 It is also not
inconceivable that if William’s scouts had gained some idea of enemy
numbers the duke was aware that they would have insufficient room to
deploy in the ground before him without their lines being cramped for
space.

1
See the map reprinted by M&M, p. 111; Tetlow, The Enigma of Hastings, p. 135. Those travelling the
Roman road which connected Hastings with Rochester in Kent must presumably have crossed the
Brede estuary either by bridge or ferry. If there was a bridge (which seems very likely given the
military role of the fortress at Hastings in King Alfred’s time) William may have either held it in force
or destroyed it.
2
Burne, The Battlefields of England, p. 45, and Lemmon, The Field of Hastings, p. 47, noted that
Harold’s standard would not have been visible from the summit of Blackhorse Hill, but only from the
lower slopes to the north (see the map, M&M, p. 111). The names of the hills in this area are likely to
cause confusion, as the eminence known as Telham Hill is some way to the west of Blackhorse Hill, to
the north of which the village of Telham now stands. Some OS maps mark as Battle Hill the area
between Starr’s Green and the south side of the railway line.

159
THE BATTLE OF HASTINGS

160
 6

THE BATTLE
The impossibility of giving any straightforward account of the battle of
Hastings will have been evident from the discussion of the primary
sources in earlier chapters. This one will therefore proceed by considering
the various statements relevant to each stage of the proceedings, and then
suggesting what seems to be the most plausible course of events.
Nevertheless, it remains true that, rather like a jigsaw made up of a series
of cubes which will produce six different pictures as their different faces
are turned upwards, and endless jumbles if individual blocks get out of
sequence, the impression one gets of the battle depends very much upon
which elements of the evidence one chooses to stress. Thus, the views on
offer include a large-scale engagement involving tens of thousands of
men on each side deployed over a considerable area, or a small-scale one
fought by perhaps less than ten thousand soldiers in total around no more
than the crest of the Battle Abbey ridge; a fight which was largely
decided, on the French side, by their cavalry and archers, or one which
was mainly a conflict of dense bodies of infantry on both sides; one in
which most attention is paid to the statements of contemporary sources
like the Carmen, and the works of William of Jumièges and William of
Poitiers, or one in which their accounts are significantly modified by
evidence drawn from twelfth-century authors such as John of Worcester,
William of Malmesbury, Henry of Huntingdon and Wace.
The prelude to the engagement, as we saw at the end of the last
chapter, involved Harold marching rapidly into Sussex in an attempt to
take the French by surprise and William advancing north from Pevensey
and Hastings to offer battle in the vicinity of the place called Senlac. It is
likely enough that negotiations of some sort preceded the actual conflict,
but if so they were probably prior to the point at which the two armies
came into sight of each other, a point which may have been rapidly
followed by the first fighting. William of Jumièges, whose account of the
timing of the opening of the battle is contemporary, more detailed and
coherent than that of the Carmen, and more convincing than that of

161
 THE BATTLE OF HASTINGS

William of Poitiers, says that the duke ordered his men to stand by their
weapons from dusk to dawn in case of a night attack, and that when dawn
had broken he deployed them and advanced against the enemy, battle
being joined at around the third hour; as for Harold, his forces had ridden
through the night and appeared on the battlefield early in the morning. 1
Late on Friday, 13 October a moon which had been full eight days earlier
rose at Battle at about 22:14 GMT and in clear conditions would have
provided a degree of illumination for the rest of the night. The following
day, Saturday, 14 October, the feast day of St Calixtus, there would have
been some light in the sky, unless the weather was very overcast, by 5:30,
and the sun appeared above the horizon at about 6:32.2 If dawn is
therefore reckoned as being around 6 o’clock, the third hour, at about the
time the battle began, would have been about 9 o’clock.
The precise line of the English advance is not known. A Roman road
ran down to Hastings from Rochester in Kent and the discovery in 1876
of a large coin hoard at Sedlescombe, a short distance to the west of it,
has sometimes been taken as evidence that Harold’s force utilised this
route on its way to Battle.3 Alternatively, it may have travelled down the
Roman road connecting London and Lewes, subsequently taking local
routes eastwards, for it is likely that there was a substantial roadway
which connected Hastings and the Lewes-London road, and passed
through the high ground upon which the battle was fought.4 It is possible,
of course, that levies from the southern shires, who must have known that
their services would be required immediately they heard of William’s
landing, in some instances joined him on the way or at an agreed
rendezvous, and Caldbec Hill, immediately to the north of the Battle
Abbey ridge, has occasionally been identified as both this spot and the
location of the ‘hoar apple tree’ which according to the D text of the
Anglo-Saxon Chronicle was where the battle took place.5 However, for
what it is worth the Carmen suggests that Harold’s men were already
fully assembled when the French saw spears glittering in the forest
(presumably that of the Weald, whose southern edge lay in the vicinity),

1
WJ, GND, ii. 166-9. The use of the preposition sub in the phrase sub hora diei tercia may indicate a
degree of vagueness about the exact time, the classical sense of the word being either ‘immediately
before’, ‘at the approach of’, or ‘in the course of’, ‘during’. The Ab hora tamen diei tertia of JW, ii.
604-5, ‘from the beginning of the third hour’, is less ambiguous; on WP, see above, p. 158.
2
The times of moonrise and sunrise have been computed using Focus Multimedia’s program Red Shift
4. The moon was still high in the sky to the south-south-west at 6:32, and eventually set around 13:30.
3
Lemmon, The Field of Hastings, pp. 41-2. The hoard ends with the penultimate coin type of Edward
the Confessor, and is not certainly connected with the events of 1066.
4
Lemmon, The Field of Hastings, pp. 18-19, offered suggestions on the line this route took between
Battle and the London-Lewes road.
5
Lemmon, The Field of Hastings, pp. 20, 40; Bradbury, The Battle of Hastings, pp. 173-8, who
suggests that the fighting was actually on Caldbec. In his ‘The Campaign of 1066’, pp. 97-8, Lemmon
assumed that Caldbec was the rendezvous and then criticised Harold for ‘the strategic blunder of
attempting to concentrate his forces within striking distance of the enemy’.

162
 6 THE BATTLE

for this is then said to have poured forth units in dense formation which
seized a hill, a valley, and rough untilled ground. William of Poitiers also
speaks of the seizure of a hill, in dense formation, by forces which had
come through a wood.1 The assumption that Caldbec Hill was the
rendezvous does not, therefore, have very much to recommend it.
The extent of the two armies’ deployment is something of an open
question, and it may be useful here to reiterate points made in the
discussion of this problem in Chapter Four.2 There are some grounds for
thinking that it covered no very great area, and this has been the received
wisdom among most historians of the battle for the last hundred years.
The E text of the Anglo-Saxon Chronicle says that Harold fought before
all his army had come; John of Worcester that his forces were only half as
big as they might have been and that they took up position in a place so
narrow that many withdrew from the ranks; William of Malmesbury that
they contained few shire levies and were mainly composed of
professional troops.3 Such evidence, combined with the idea that the
losses sustained against the Norwegians at Fulford and Stamford Bridge
must have had their effect, and a firmly-entrenched belief that medieval
armies were nearly always small, led in the late nineteenth century to a
scholarly consensus in favour of an English deployment extending over
600 to 800 yards rather than the much longer line which had been posited
by Freeman in the 1860s. Furthermore, those who have relied on William
of Poitiers’ account of the prelude to the battle have taken him to mean
what he does not actually say: that the French forces marched from
Hastings to Battle in the early morning and deployed all before 9
o’clock,4 something which seems impossible if they were of any great
size, and may well have been so even if they were not. C.H. Lemmon
followed William of Poitiers unquestioningly, but suggested that William
had previously advanced to the area of Baldslow, north of Hastings
(something which the Gesta Guillelmi specifically contradicts if, as seems
certain, the camps from which he assembled his men were those at
Pevensey and Hastings), and began his march at 6 a.m., dawn having
been at 5:30. This desire to both reduce the length of the march and to
allow it begin as quickly as possible may indicate a degree of unease on
Lemmon’s part about the entire timescale,5 and this is all the more reason
1
Carmen, ll. 343-4, 363-8; GG, pp. 126-9.
2
Above, pp. 113-25.
3
ASC E, i. 198, JW, ii. 604-5; WM, GR, i. 422-3.
4
The same impression could also be derived from the Bayeux Tapestry, which shows the French
cavalry leaving Hastings and then charging into battle; of course, a source which presents all events as
a seamless sequence is poor evidence on such a point.
5
Lemmon, ‘The Campaign of Hastings’, pp. 95, 102-3. Burne, The Battlefields of England, pp. 43-4,
had already said of the battle’s chronology that the ‘chief difficulty is that so much seems to have
happened in the 2½ hours between sunrise and the opening of the battle’, eventually suggesting that the
French march began at 6 a.m. and the fighting at 9:30. Another ex-soldier, Brigadier C.N. Barclay

163
 THE BATTLE OF HASTINGS

reason to credit the more plausible evidence of William of Jumièges,

which may well mean that the French were in the vicinity of Battle the
previous evening. However, having disposed of the early-morning march
does not by itself disprove the idea that both English and French deployed
over a small area, and there is in fact no way of proving that they did not.
Nevertheless, there is a counter-case which falls short of proof, but
provides good reason to doubt whether the long-accepted view of the
geographical extent of the battle is correct.
It can be stressed, for example, that both John of Worcester and
William of Malmesbury show clear signs of wishing to diminish the
impact of the English defeat, writing as they were in the first half of the
twelfth century at a time when it was still a very sore point, and that the
implications which their statements have for the size of the English army
need not therefore be taken very seriously.1 Moreover, even if the
Chronicle E text’s assertion that Harold fought before all his army had
come is less easily explained away, our knowledge of the military
capacities of late Anglo-Saxon government still renders it possible that he
commanded tens of thousands of men, and there is the same possibility
about the international army with which William had crossed the Channel
in what may have been a very large number of ships.2 To these general
points can be added a more concrete argument. The Bayeux Tapestry has
a scene in which French cavalry are being brought down by what look
like stakes in a body of water, and nearby is an eminence defended by
what may be a specialist unit of English light infantry. The locality
concerned cannot be identified beyond doubt, but those who have
suggested that the eminence is the area of Horselodge Plantation, on the
south-western edge of the Battle Abbey ridge, and the water the sandy
stream to the south and south-west of it which may even have given the
fight its name of Senlac, need not have erred; this may also fit Guy of
Amiens’ statement in the Carmen that William’s enemies occupied not
only a hill (mons), but also a valley (vallis). Moreover, this evidence that
the English deployment extended to an area well away from the crest of
the ridge raises the possibility that it also extended further still, and into
the relatively flat country traversed by the watercourse as it runs down to
the present Powdermill Lake, for if the depiction in the Tapestry is a
reliable indication of the volume of water involved (and of course it may
not be) the area concerned could be that which today lies immediately to
the north-east of that lake, the stream much further east perhaps being too
narrow, even in the days before the construction of New Pond, to be

(Battle 1066, p. 55), thought that ‘the time it would take William’s men to march six or seven miles’
and then deploy would have meant that the battle did not begin ‘before 9.30 or even 10 a.m.’.
1
Above, pp. 53-60.
2
Above, pp. 143-51.

164
 6 THE BATTLE

represented in this way; however, given that the three Stew Ponds at the
head of the valley, and almost due south of Harold’s standard, had not
been created in 1066 it is far from impossible that there were marshy
pools not far from what would one day be their location.1 As the more
cautious commentators on the battle have stressed, the only certainty is
the location of Harold’s standard, and beyond that there is only
speculation. It may be, as virtually all have believed, that his men
principally faced south-east to meet an attack coming mainly from the
direction of Blackhorse Hill and Starr’s Green, and that his left flank
rested somewhere in the vicinity of the former Battle National School,
where there was probably a steep gradient to the east,2 and his right, as
just suggested, on the marshy pools north-east of Powdermill. However,
another possible line of deployment would be one in which the English
army faced south-west to meet an attack coming mainly from the area of
Powdermill. Wherever they stood, they must have been readily visible to
their king, positioned on the highest point of the ridge behind them,
alongside a dragon standard of a type probably used by English armies
time out of mind, and his personal standard of the Fighting Man,
sumptuously worked in gold.
An aspect of John of Worcester’s account of the battle which seems to
find confirmation elsewhere is his stress on the constricted nature of the
English position, for the Carmen, William of Poitiers and other sources
refer to the closely-packed nature of their line.3 This renders it virtually
certain that many of Harold’s troops were formed into one or possibly
more shield-walls, and it may be that such formations sometimes fought
not only line abreast but also one behind, and in support of, another.
However, there is no detail to be had upon such matters from the French
sources which have most to say about the battle, English formations not
being a matter which they found of any great interest, apart from the
statement in the Carmen that the flanks were strengthened with noble
men (nobilibusque uiris – perhaps a reference to the professional
housecarls) and that having fixed his standard upon the crest of the ridge
Harold ordered other banners to be joined to his (whatever that means);
rather later, Baudri of Bourgeuil spoke of a rank of spears so dense that it
looked like a forest and of the English pressing close together in a single
wedge (cuneum densatur in unum), while Henry of Huntingdon believed
that Harold had placed all his men in a single very close line, and William
of Malmesbury that all the infantry possessed axes and formed an
impenetrable mass (impenetrabilem cuneum) with a wall of shields to

1
See above, p. 124 and Illus. 39-41.
2
Above, p. 48 and Illus. 63-5.
3
Carmen, ll. 415, 417-21; GG, pp. 128-9, 130-3.

165
 THE BATTLE OF HASTINGS

their front.1 There must, however, have been rather more to his
deployment than simply this, even if such references give no support to
the idea of multiple densely-packed formations. The presence on the
Bayeux Tapestry of a single archer in front of the shield-wall hints at
lines of skirmishing missile troops,2 and should the light infantry depicted
in its hillock scene have been in occupation of it from the beginning of
the battle they need to be taken into account too, as does the Tapestry’s
evidence that heavy infantry defended the line of the watercourse.
Moreover, as this source is most unlikely to be a complete guide to the
English dispositions, it follows that light infantry and mailed soldiers in
loose order may also have been stationed elsewhere on the field.
Furthermore, if the serrated shapes which it shows above the water are
stakes, as seems likely, this suggests that some of the English occupied
the prelude to the battle, which may mean much of the three hours or so
after dawn, by creating field defences to obstruct the advance of the
French, and especially that of their cavalry. Even a shallow ditch can act
as a major obstacle to horses, and it is not impossible that Harold’s men
went further than this by constructing concealed pits of the sort which
Vikings had employed to bring about the demise of Margrave Henry of
Neustria in 887.3 Henry of Huntingdon, at least, reports that during the
battle many of the French were laid low by a great ditch cunningly hidden
(foueam magnam dolose protectam), while William of Malmesbury
speaks of a ditch being filled to the brim with French bodies.4 Wace went
much further than this: saying at different points that Harold had the place
examined and protected by a good fosse with an entrance on each of three
sides; that the English had made a fosse which went across the field,
guarding one side of their army; and that during the battle the Normans
were forced back into a fosse which they had previously passed around
the side of, that horses and men kicked helplessly, that many of the
English fell there too, and that at no time in the day did as many Normans
die as perished in the fosse, as those who saw the dead said.5 How far
such details may derive from authentic tradition, rather than being a
muddled reflection of statements about ditches found in the earlier
sources known to Wace, is difficult to say.
To sum up: it is possible, as has long been believed, that Harold’s men
occupied a position around no more than the crest of the ridge, with the
1
Carmen, ll. 373-6; AC, ll. 403-6; HA, pp. 392-3. GR, i. 452-3. The Carmen, ll. 421, 428, 530, also
refers to the English ranks as like a forest.
2
See above, p. 69, on the danger of arguing from this single figure that the English were short of
archers.
3
Above, p.139.
4
HA, pp. 392-3; GR, i. 454-5. If the ponds to the rear of Horselodge Plantation (above, pp. 47-8 and
Illus. 32-3) existed in 1066 it would have been very easy for unsuspecting French troops coming
against this area from the north to have fallen into them.
5
RR, ll. 6969-72, 7847-8, 8079-96.

166
 6 THE BATTLE

flanks protected by vegetation and perhaps on the left by steep slopes to

the north and east in the vicinity of the former National School;1 it is also
possible, indeed more likely in the view of the present writer for reasons
already stated, that their forces extended over much of a ridge 1¼
kilometres long, and also down into the valley to the south of it, with at
least one watercourse being held in force against the enemy. The
orientation of the English units was probably either to the south-east,
facing Starr’s Green with Blackhorse Hill beyond, or to the south-west, in
the direction of Powdermill. There is no way of arriving at any
satisfactory estimate of the numbers of men under Harold’s command: it
is likely to have been a figure between a few thousand and many (perhaps
tens of) thousands; the same is true of Duke William’s army.
Both the steep slopes to be found on parts of the ridge, and especially
immediately to the south of the English king’s own position, and what the
Carmen calls the roughness of the uncultivated ground2 rendered the area
generally unsuitable for cavalry, as did the small streams which abound in
the vicinity and the probably distinctly marshy drainage area which
extended from east of the location of what is now New Pond down to that
of what is today Powdermill Lake. Geography probably made the English
position virtually impossible to outflank,3 and what is known of the battle
itself tends to confirm this belief. Wherever his flanks lay, Harold would
have been aware of the danger of their being turned by the French horse,
and there is no difficulty in believing that he stationed renowned men
there, as the Carmen says, and rested them on significant obstacles.
Otherwise, his troops consisted of a range of types, from those protected
by metal armour, helmet and shield, and with spears, swords and axes as
offensive weapons, to those with no defensive protection at all, and
wielding only bows, clubs or spears. There is, of course, no way of
knowing the relative proportions of the different types, or how the shire
levies, some of whom were probably very well armed, 4 were deployed
relative to each other and the professional housecarls, for the Bayeux
Tapestry, which can reasonably be taken as valuable evidence on what
English soldiers might look like, is of little help in answering questions
such as these. However, shield-walls are likely to have been at least
several and probably many ranks deep,5 and if the D text of the Anglo-
Saxon Chronicle is correct in stating that William launched his attack
before the English ranks were properly formed it looks as though the

1
On this area, see above, p. 48 and Illus. 63-5.
2
Carmen, l. 366: Et non cultus ager asperitate sui.
3
Lemmon, ‘The Campaign of 1066’, p. 100, stressed the difficulties which the ground presented to
cavalry, and suggested (like Burne, The Battlefields of England, p. 23) that Harold had reconnoitred the
ridge the previous summer.
4
Above, p. 126.
5
Above, pp. 134-6.

167
 THE BATTLE OF HASTINGS

battle began before everything was in suitable order. As the French are
said to have seen the enemy emerge from woodland and occupy the ridge
in dense formation one cannot be sure that the process had necessarily
taken all of the three hours which separated dawn from the beginning of
the fighting; even so, the more time it took the greater any estimate of the
numbers in this dense formation would be likely to be, and the time
involved in their deployment seems to have been at least great enough for
the construction of field defences elsewhere.
The sources have more to offer on the French deployment. The account
of the opening stages of the battle in the Carmen is both rather incoherent
and complicated by the insertion of the story of the heroic exploits of the
juggler Taillefer, but prior to this Guy of Amiens speaks of archers and
crossbowmen whom the duke had intended supporting with cavalry had
the onset of the battle not prevented it, and then of infantry armed with
arrows and quarrels (i.e. archers and crossbowmen) preceding him up the
hill, together with helmeted men who rushed to crash shields against
shields. The information on the missile troops is repeated again after the
Taillefer story, as is the point that shields were of no avail against the
bolts from crossbows. Then, Guy seems to say, the French attacked the
left (of the English) and the Bretons the right, while the duke and the
Normans occupied the centre. William of Poitiers’ belief that Breton
horsemen were on William’s left and Robert of Beaumont on the right
goes some way to support this,1 just as his more coherent account of the
French deployment is generally very similar to that of the Carmen. Here,
too, infantry in the front rank were armed with bows and crossbows,
while in the second line were steadier, armoured foot-soldiers (pedites...
firmiores et loricatos) and finally there were the squadrons of horsemen
(turmas equitum) with the duke himself in their midst.2 An advance in
three lines, but without further detail, is also reported by William of
Jumièges.3 It had plenty of classical precedents, and Caesar sometimes

1
Carmen, ll. 335-42, 381-414; GG, pp. 128-31. The use in the Carmen of the verb peto creates
difficulty here, as it can be translated either ‘occupy’ or (more plausibly) ‘attack’. As Barlow points out
(Carmen, p. lxxix, n. 266) ‘if the French ‘occupied’ the left and the Bretons the right, their positions are
clear enough. But if they attacked the English left and right wings... their position in the ducal army is
reversed’. His translation (p. 25) chooses the second option. On the possibilty that the Norman Robert
of Beaumont was fighting on the right with the French, see above, p. 141, n. 2.
2
GG, pp. 126-7; turmas equitum may be a conscious echo of the equitum turmas and turmisque
equitum to be found in Caesar’s The Gallic War, iv. 33, vii. 88, pp. 222, 506. On the use of crossbows
at Hastings, see M&M, pp. 112-5, who suggested that of the four archers who appear in the main
register of the Bayeux Tapestry the one wearing mail is a crossbowman. Baring,‘On the Battle of
Hastings’, pp. 226-7, used Wace’s Roman de Rou, which ‘no doubt followed a tradition which in this
matter was probably trustworthy’, as evidence that ‘the geographical principle was carried out in detail,
the various contingents being arranged from east to west according to the position of the places they
came from’, and that ‘individual lords with their followers were as a rule arranged by the
neighbourhood from which they came’.
3
GND, ii. 168-9.

168
 6 THE BATTLE

arranged his infantry thus,1 but there is no need to reject French

statements as groundless fabrications on this score, as the dispositions
described are plausible enough. However, a more obvious place for
cavalry would have been on the wings, where Vegetius was adamant that
it should be placed, and where King Henry I did place it in 1106 at the
battle of Tinchebrai, when a charge on the enemy flank decided the day. 2
That William does not seem to have done this may mean that he was
deploying over a frontage which was rather restricted relative to the
number of his men, or that the English flanks were so securely anchored
that he could see that there was no possibility of their being turned.
Stories, some of them fabrications, probably gathered around the battle
almost immediately, perhaps assisted by a demand for chansons which
celebrated the French victory. In this context, it is not surprising that the
tale of Taillefer should have found its way into a piece as early as the
Carmen is now believed to be, as it did later into the works of Henry of
Huntingdon, Gaimar and Wace. It is not necessarily completely fictitious,
but even if one of William’s men did taunt the English by juggling with
his sword and inflicting their first casualty, it is not likely that many were
deflected more than momentarily from the serious business of making
war. In this context the Anglo-Saxon Chronicle D text’s claim that
William attacked before his opponents were properly drawn up is
particularly important, for this is of course not only what a skilled general
might well have sought to do, but it is also the sort of stratagem the duke
had employed successfully on at least one occasion in the past, when at
Varaville in 1057 he caught and defeated his opponents by advancing
swiftly as they were crossing the river Dives; he was to act similarly
against northern English rebels in 1069.3 Nor is it impossible that the
apparently very constricted nature of the English lines was to an extent
the result of this tactic. Yet even if this is so it can have given the French
no very decisive advantage, as the sources are agreed that once joined the
fighting went on for many hours.
Any attempt to chart the course and extent of the conflict in detail over
such a lengthy period is bound to fail given the inadequate nature of the
evidence, and accordingly there will be no such attempt made here. Even
so, the statements of the sources do at points coincide in ways which
almost certainly reflect what might be termed real issues and events. One
of these is the French use of missile troops. Archers who did not know
how many they had killed and wounded were assigned their own penance

1
Caesar, The Gallic War, i. 49, 51, pp. 80-3.
2
Vegetius, Epitome, pp. 47, 98. On Tinchebrai, see further, below, pp. 172-3.
3
Above, p. 156; below, p. 196. Note also, ‘It is notable how often he appears to have surprised his
opponents and caught them off balance either on the battlefield, as at Varaville, or at sieges such as
Alençon or Arques’, Bates, William the Conqueror, pp. 42-3.

169
 THE BATTLE OF HASTINGS

by Norman bishops shortly after the battle,1 and they and their confrères
figure very prominently in the Bayeux Tapestry, even if they spend most
of their time in the lower border, and rather less prominently (for they are
quickly left behind) in the early written accounts. The Carmen, William
of Poitiers and Baudri of Bourgeuil all comment on the French assault
being preceded by discharges from archers and crossbowmen, the
Carmen noting that shields were no defence against the latter, while
Baudri suggests that the English had not faced the crossbow before. He
also thought that the losses they sustained from this missile attack led
them to abandon their dense formation in pursuit of the enemy, and that
the Normans then simulated flight and used their cavalry to surround and
slaughter their pursuers.2 The Carmen, however, which also speaks of a
feigned flight after the missile attack, suggests that infantry fighting,
when spear was met with spear and sword with sword, intervened
between the two, and the same impression is also to be derived from
William of Poitiers, if it is assumed that the foot-soldiers (pedites) who
killed and wounded many during an exchange of missiles with the
English, and who eventually received cavalry support, included the
armoured infantry mentioned by the same author at an earlier stage; it can
be noted that he also eventually describes the flight of pedites and Breton
horse and such auxiliaries as were on the left wing.3
The sources do their best to obstruct those who would seek out the role
of French infantry in the battle. It is, to begin with, extremely unlikely
that they consisted solely of missile troops and men in metal armour, for
it is probable that only a portion of the duke’s men could afford such an
expensive form of protection. Yet of unarmoured French infantry who
were not missile men the sources say nothing whatever. The designer of
the Bayeux Tapestry depicted at least 179 horses in the work’s main
register, and as R.H.C. Davis commented, ‘made it his business to show
William’s army as an army of cavalry’,4 but if it is reasonable to suppose
that this reflected the interests of the militaristic and aristocratic audience
for which the piece was made, it is far less so to conclude that they and
their mounts played as decisive a role in the fighting as the Tapestry
might lead one to think. The earlier depiction of the attack on Dinan
during the campaign in Brittany is significant here. Horsemen bearing
spears and shields gallop towards the fortification, while two of the
defenders stand outside the wall ready to hurl spears in response. Below
the walls two men attempt to set fire to the structure, and then the keys
1
Penitential articles issued after the Battle of Hastings, c. 6.
2
Carmen, ll. 337-40, 381-2, 409-12; GG, pp. 126-9; AC, ll. 409-21. I follow M&M, p. 115, who made
sense of an otherwise curious passage in Baudri by suggesting that his ‘arrow which they had not
known’ was the crossbow bolt.
3
Carmen, ll. 383-4, 416; GG, pp. 126-9.
4
Davis, ‘The Warhorses of the Normans’, 68.

170
 6 THE BATTLE

are handed over in token of surrender. Of course, this is the sketchiest of

representations of an attack on a fortified site (although not necessarily
much more sketchy than the Tapestry’s account of the battle of Hastings),
but the impression that men on horseback played any significant role in
assaulting the walls must be misleading, and even when it comes to
operations on foot these are performed by what look like dismounted
horsemen rather than the somewhat less glamorous infantry who must
surely have been responsible for such activities in reality. In the light of
this, it is reasonable to question the reliability of the Tapestry’s depiction
of horsemen with spears at the ready charging the English shield-wall at
Hastings, and it is worth noting that according to William of Poitiers
when they came up in support of the infantry they used their swords.
Nevertheless, such is the power of the Tapestry that some historians have
concluded that the battle represented a victory of cavalry and archers over
infantry. In his 1884 Lothian Prize Essay, for example, the young Charles
Oman compared the trials of the English array with that of the British
squares at Waterloo: ‘incessant charges by a gallant cavalry were
alternated with a destructive fire of missiles’. Eventually, unable to ‘bear
the rain of arrows’ many of the English followed the retiring cavalry
down the hill and were surrounded and cut to pieces. The remainder
fought on for another three hours, but the day was lost: the ‘tactics of the
phalanx of axemen had been decisively beaten by William’s combination
of archers and cavalry’.1 Not much later, J.H. Round went so far as to
suggest that the French army consisted only of horsemen and armoured
and unarmoured missile troops.2
Yet as far as cavalry goes, one can acknowledge that possession of it
gave William tactical options not available to Harold, and that at points in
the battle it may have been of great importance, without concluding that it
was inevitable that its actions would prove decisive. Its effectiveness on
the rough ground over which the battle was fought would have been
influenced by the location of the English position. If they occupied no
more than the crest of the ridge, this would have yielded a flattish area to
the west3 upon which horsemen could perhaps have galloped in the way
that the Tapestry suggests, but if they were stationed on something like

1
Oman, The Art of War in the Middle Ages A.D.378-1515, pp. 25-6. In his expanded 1898 treatment (A
History of the Art of War, pp. 149-64) Oman provided a much more elaborate account of the battle but
re-stated his basic conclusion, and while acknowledging the presence of French armoured infantry
judged the mailed knights ‘the really important section of the army’.
2
Round, Feudal England, pp. 368-70, drawing on his 1893 Quarterly Review article, 75-6, and upon an
interpretation of WP’s words (GG, pp. 126-7) that was unreasonable, as Kate Norgate observed, ‘The
Battle of Hastings’, 42-3; it was also a shift from the position of his Quarterly Review article of 1892,
17, where he spoke of the English being ‘first riddled by Norman arrows and then attacked by Norman
infantry’.
3
That is, that running from the western edge of the abbey terrace to the location of the OS height
marker of 69 metres, some 500 metres further west, above, p. 47 and see Illus. 45.

171
 THE BATTLE OF HASTINGS

the whole of it there would have been many gradients for the horses to
face. In this case, and while it would be foolish to be dogmatic given our
ignorance of the precise nature of the terrain, the ground as it now exists
provides good reason to wonder whether effective charges could have
been made over many of the southern slopes of the ridge, even allowing
that horses can climb very steep gradients indeed. It might also be
stressed that it is doubtful how useful horsemen would have been against
densely-packed, steady infantry even in more favourable conditions,1 that
the massed charge in which riders in close order couched heavy lances
beneath their arms had probably not been developed in 1066,2 that there is
no way of knowing what proportion of the French forces the cavalry
formed, that their steeds had no armour to protect them from English
missiles, and that even well-bred stallions carrying men in metal armour
must have tired as the day advanced (assuming, of course, that they found
much employment).
All this raises the question of whether Hastings was much more of an
infantry conflict than has sometimes been realised.3 At Tinchebrai in
1106, fought between William’s sons Duke Robert of Normandy and
King Henry I of England, Orderic Vitalis reports that the latter’s iron-clad
lines (which included Englishmen) were drawn up and then advanced in a
disciplined fashion in close order, with the first of the three under the
command of Ralph of Bayeux, while Robert’s front line was led by
William, count of Mortain, and his rear by Robert of Bellême. When the
opposing ranks met, however, they were so crowded together that it was
impossible from them to use their weapons properly, and the conflict was
decided when Henry’s cavalry charged the enemy’s flank and Robert of
Bellême took to flight. According to the priest of Fécamp, the fighting
had then lasted an hour.4 There are two points of interest about this. The
use of dense lines of infantry by both sides may have owed something to
Norman experience of facing the English at Hastings, or it may be that
their infantry had fought in this fashion in the past, including at Hastings
itself; and in the latter case, the length of time that the closely massed
foot-soldiers struggled without result at Tinchebrai could have
implications for Hastings, for in a situation where a flank attack by

1
See the comments of Verbruggen, The Art of Warfare, pp. 46-7; Morillo, Warfare under the Anglo-
Norman Kings, pp. 155-6; Strickland, ‘Military Technology and Conquest’, 360-1; and especially
Bennett, ‘The Myth of the Military Supremacy of Knightly Cavalry’, pp. 309-12, and his conclusion (p.
316) that ‘knightly cavalry were generally as impotent before well-trained foot as the French cavalry
against British infantry squares at Waterloo’. Vegetius, Epitome, p. 84, states that the strength of an
army depends mainly on its infantry.
2
See France, Victory in the East, p. 71.
3
The likely importance of the French infantry was stressed by Brown, ‘The Battle of Hastings’, 11;
similarly, Morillo, Warfare under the Anglo-Norman Kings, p. 164, n. 98.
4
HÆ, vi. 88-91; Priest of Fécamp, Letter on the Battle of Tinchebrai; HH, HA, pp. 452-5. On
Tinchebrai, see also David, Robert Curthose, pp. 245-8.

172
 6 THE BATTLE

cavalry seems not to have been a possibility it is not hard to see how a
battle of this type could have gone on for many hours.
Dr Morillo has commented that it is ‘difficult to find more than a
handful of ancient and medieval battles that lasted more than an hour or
two’,1 but in fact there are a number of engagements involving the Anglo-
Saxons of which one can be tolerably certain that this is true. Even
leaving aside John of Worcester’s assertion that Sherston, fought by
Edmund Ironside against Cnut in 1016, took up two days, Alfred’s defeat
of the Vikings at Ashdown in 871 is said to have lasted until nightfall,
while his fights against the same enemy in the same year at Meretun and
Wilton apparently both went on for much of the day. Similarly, the poem
on his grandson Æthelstan’s victory at Brunanburh in 937 speaks of the
slaughter extending from morning until sunset, and of the West Saxons
pursuing the enemy all day long, while Stamford Bridge in 1066 saw
fighting until late in the day.2 All these conflicts are likely to have seen
shield-wall pitted against shield-wall, a formation used by the Vikings as
well as the English, and perhaps involved considerable numbers of men.
Could it be, despite the prominence given by the Bayeux Tapestry to the
French cavalry, that Hastings was so long and hard fought because much
of the day was taken up by struggles between dense bodies of infantry on
both sides of a type with which the Anglo-Saxons had long been familiar,
and because the Normans had retained their Scandinavian ancestors’
practice of deploying their foot-soldiers in this fashion too?
Yet even if this were the case, the sources suggest that the battle had
other significant features. One was a French retirement which almost
turned into a rout, another the use of the feigned retreat. William of
Poitiers says that after a period of fighting in which the English were
greatly helped by their numbers, the density of their formation, their use
of weapons which easily penetrated shields and other defences, and the
advantage of occupying higher ground, the foot-soldiers, Breton cavalry
and auxiliaries on William’s left wing turned away, and almost all his line
began to give ground, for the Normans believed that the duke himself had
been killed. However, William dealt with this crisis by removing his
helmet to prove that he was still alive, urging his men on to victory, and
leading a charge which surrounded thousands of the pursuers and
annihilated them to a man. Subsequently, further heavy fighting was
followed by two feigned flights which drew forward and destroyed more
of the English forces.3 Guy of Amiens’ Carmen has a slightly different
version of events, claiming that the French would not have been able to

1
Morillo, ‘Hastings: An Unusual Battle’, p. 220.
2
JW, ii. 486-9; ASC, A text, i. 70, 72, 106, 108; C, i. 198. As Harold travelled some miles to reach
Stamford Bridge the battle cannot have begun early in the day.
3
GG, pp. 128-33, followed in its essential details by OV, HÆ, ii. 174-5.

173
 THE BATTLE OF HASTINGS

penetrate the dense English formation had it not been for their skill in
war, as they deliberately simulated flight and then turned on their
pursuers, while their two wings attacked an enemy now more dispersed
than formerly. Ten thousand of the English were killed at this stage, but
the rest fought on effectively enough to force the Normans into a real
flight, which William halted by removing his helmet (there is no mention
here of any belief that he had been killed), making a speech and leading
his forces in a renewed attack in which he performed great feats of arms,
killing Harold’s brother Gyrth who had unhorsed him with a spear cast,
forcing a trooper from Maine to yield up his steed and then being
unhorsed a second time, but once again wreaking vengeance on the man
responsible, the ‘son of Helloc’. The duke is then said to have received a
third mount from Count Eustace of Boulogne, with whom he cleared the
field of English soldiers, and from here the Carmen moves to the role of
both men in the killing of Harold.1
The authors of other sources, too, knew of the simulated flight.
According to Baudri of Bourgeuil, the English broke ranks because the
torment caused by the opening French missile attack apparently coincided
with a feigned Norman withdrawal; they were intercepted by cavalry and
slaughtered, but further attacks upon the English front resulted in a real
withdrawal fuelled by the rumour that the duke had been killed. He
responded to this by removing his helmet, making a speech similar in
tenor to those reported by William of Poitiers and the Carmen, and
leading his men in a further assault in which he surpassed the deeds of
Hector and Achilles; these deeds are followed by the report of Harold
being killed by an arrow.2 The Battle Abbey Brevis Relatio says nothing
of difficulties experienced by the French in the early stages of the battle,
only that not long after William had opened it by killing one of the
English a unit of almost 1,000 Norman horsemen rushed upon the enemy
line, but fled in apparent fear once they reached it; as a result some of the
English followed, wishing to kill them if possible, but the Normans, who
were more wary in war than their enemies, soon returned, interposed
themselves between their pursuers and the main English line, and quickly
killed all of them.3 William of Malmesbury and Henry of Huntingdon
connect the simulated French withdrawal with fighting around a ditch
(see below), while Wace does not mention it until after describing Harold
being wounded by an arrow,4 and the Bayeux Tapestry does not show it
at all; however, it does show Bishop Odo of Bayeux comforting the boys

1
Carmen, ll. 421-530.
2
AC, ll. 409-63.
3
BR, p. 32.
4
RR, ll. 8175-252.

174
 6 THE BATTLE

and William removing his helmet, presumably to prove that he is still

alive.
Of course, none of these sources is in complete agreement with any of
the others, the most similar being the Carmen and Baudri of Bourgeuil,
who both place the real French flight after a simulated one, in contrast to
William of Poitiers’ assertions that the latter preceded the former, and
that there were two simulated withdrawals, not one. It is impossible to
make any definite choice between these different versions of events, for
while William of Poitiers as a Norman with close contacts in the ducal
court1 ought to have been the most well informed there is a perhaps
suspicious neatness about his claim that the French learnt from the
advantage they had reaped from the outcome of the real flight. There are
also the problems, discussed in Chapter Three, about how far the
elements common to these sources may be the result of direct borrowing
from one to the other, or how far they may instead or in addition reflect
traditions about the battle which were widely known at the time and
accordingly found their way in different forms into different works.
Unless one is to take what seems the unreasonable view that the Carmen
was the only source ultimately behind William of Poitiers’, Baudri of
Bourgeuil’s and later sources’ knowledge of these events,2 the second
conclusion appears much the most likely basic explanation of the
similarities and dissimilarities of their statements about the real and
feigned French withdrawals. Indeed, it would not be at all surprising if
those who survived the conflict came away with confused impressions
and memories of what had happened, and if their reports were further
distorted by constant repetition and the passage of time to yield the
inconsistent picture which is all that now remains. Yet while this may not
provide any very clear narrative of the course of events, nor is it
completely useless. It looks as though in the years following the battle it
was said that there was a point at which Duke William managed to stem
the rout of significant numbers of his men only with difficulty and by
disproving the rumour that he had been killed by removing his helmet,
and that at another stage or stages the French managed to reduce the
impenetrability of the English lines by engaging in feigned retreats in
which their cavalry could to an extent come into its own.
While this takes us some way in establishing what happened during the
battle it is possible to get further still by turning to other sources which
describe elements of the fighting to which the contemporary and near-
contemporary French writers make no reference. The Bayeux Tapestry,
as has been mentioned several times already,3 shows what appear to be
1
Above, pp. 81-2.
2
For other comment on this, see above, p. 102, n. 1.
3
Above, pp. 121-4, 133, 139, 164-5.

175
 THE BATTLE OF HASTINGS

English light infantry defending a hillock near a watercourse in which

stakes seem to have been fixed to impede the French horse, and with
considerable success, judging by the legend ‘Here the English and the
French fell at the same time in the battle’ and the depiction of two horses
which have been brought down and of a third lying dead in the lower
border to their right. It is clear that the Tapestry’s designer, and
presumably the patrons for whom the work was being done, saw this as a
significant stage of the conflict, and nothing could sound a more effective
warning of the limitations of sources such as the Carmen, William of
Poitiers’ Gesta Guillelmi and Baudri of Bourgeuil’s Adelae Comitissae
than the fact that they are silent on these events, either through ignorance
or because they did not wish to stress elements of the fighting which
reflected little credit on Duke William and his men. Others are not silent.
Whatever reservations one might have about William of Malmesbury’s
account of the battle,1 he does speak of a feigned retreat which broke up
the English formation, and of how some of the troops who were put to
flight by this stratagem seized a hillock from which they threw down the
Normans into a valley, and by throwing spears and rolling down stones
upon them routed them to a man. Also, he says, making their way around
a deep ditch by a short cut known to them, the English killed so many of
the enemy that their corpses filled it to the brim. 2 Henry of Huntingdon
was another Anglo-Norman author of the first half of the twelfth century
who had heard about events involving a ditch. When Duke William
ordered his men to simulate flight, he states, they came to a cunningly-
hidden large ditch as they fled where a great number fell and were
crushed, but as the English continued their pursuit beyond it they saw the
main Norman line break through their own centre, and the greater part of
them fell around the same ditch as they retired.3 Wace’s several
statements about ditches, like those of William of Malmesbury and Henry
of Huntingdon, were noted earlier,4 but the final one is worth repeating:
that the Normans were forced back on a ditch into which horses and men
were thrown, that many of the English were killed there too, and that at
no time in the day did as many Normans die as in that ditch, as those who
saw the dead said.
Of course, these sources may not be completely independent of each
other, for Wace probably used the works of both William of Malmesbury
and Henry of Huntingdon (though they were not known to each other)
and it is not inconceivable that William derived some of his knowledge

1
Above, pp. 55-60..
2
GR, i. 454-5.
3
HA, pp. 392-5.
4
Above, p. 166.

176
 6 THE BATTLE

directly or indirectly from the Tapestry.1 However, he did not draw his
comments on the throwing of missiles and rolling down of stones from
the hillock, or that on the ditch being filled level to the brim with bodies,
from that source, and nor did Wace owe his statement on the recollections
of those who saw the dead to any authority that has survived. Hence, it is
most likely that the way in which these stories about fighting around
ditches and hillocks both resemble and do not resemble each other is
attributable to the same sort of reason mentioned above with reference to
the French flights and feigned flights: that there was a firm tradition that
events of this kind had taken place, but that with the passage of time this
reached different authors in different forms. It is a corollary of this, of
course, that again as with the feigned flights it is not possible to come to
any definite conclusions about the truth of what happened. It is the belief
of the present writer, as expressed earlier,2 that the Tapestry’s details are
strong evidence that hillock and watercourse were occupied by the
English at the start of the battle. It is difficult to say whether the
statements of William of Malmesbury, Henry of Huntingdon and Wace
are to be taken as closely-related to events depicted in the Tapestry’s
hillock scene, or whether they may also reflect other activity at different
times and/or different places. However, it would not be difficult to
believe that the English both defended the stream which flows west from
New Pond (if this is the Tapestry’s watercourse) at the beginning of the
battle, were eventually pushed back from it, and then fought around it
again when pursuing the enemy at a later stage. If so, Wace’s assertion
that more of the Normans fell there than elsewhere would be perfectly
credible.
If the French sources, and especially William of Poitiers, are to be
believed, the English were greatly weakened by the fact that many of
them abandoned their strong position to pursue a retiring enemy who then
used his cavalry to take advantage of breaks in their formation by
enveloping their flanks and destroying them. Were this so, there is no
denying that it may reflect indiscipline on the part of some of Harold’s
troops. Still, there are other possibilities. The reports that at one point
William’s men almost routed must be taken seriously, and it would not be
surprising if the English commanders decided that this was something
which had to be exploited in order to gain a decisive victory; this might
also fit Adam of Bremen’s almost contemporary report that the English
were at first victors but then were defeated by the Normans.3 Harold may
1
Brown, ‘The Battle of Hastings’, 20, was in little doubt that WM had seen the Tapestry, but B&W,
27-8, were more cautious; there is no convincing reason to believe so.
2
Above, pp. 121-4.
3
Adam of Bremen, Gesta, ed. Schmeidler, p. 197. Morillo, ‘Hastings: An Unusual Battle’, p. 224,
suggests that WP’s statement (GG, pp. 130-1) that ‘a great part’ of the English line moved forward is
evidence of a deliberate counter-attack.

177
 THE BATTLE OF HASTINGS

or may not have been aware of Vegetius’ warning that heavy infantry
should not follow a retiring enemy because of the danger that they may
become disordered and thus vulnerable to counter-attack, and that pursuit
is the role of light troops and cavalry.1 If he was, and if the heavily-armed
housecarls, thegns and shire-levies left the task to light-armed men who
were more mobile and intended for use in such circumstances, then one
could not necessarily take their pursuit as a gross tactical blunder.
Nevertheless, it may have been so, for it allowed the French horse to
come into its own, and points up the limitations imposed upon the
English by the fact that they were not employing cavalry themselves. As
Harold had apparently decided to stand on the defensive, as Edward III
would later do at Crécy in 1346, his son the Black Prince at Poitiers ten
years later and Henry V at Agincourt in 1415,2 it would seem that this is a
strategy he should have persisted in, and one might discern here the same
haste that led him to fight the battle when all his army had not yet come.
Beyond the damage caused to the English by the French retirements,
the sources have little more to offer in determining why they lost the
battle. The Bayeux Tapestry, whose designer did not, of course, intend it
as an analysis or explanation of this problem, shows the deaths of
Harold’s brothers Leofwine and Gyrth before the scene depicting fighting
around the watercourse and hillock, and Bishop Odo rallying the boys
and Duke William proving he is still alive after it; then French horsemen
fight mailed infantry described as ‘those who were with Harold’ (perhaps
a reference to his bodyguard), a mailed Frenchman on foot beheads an
unarmed and unarmoured enemy, and many French archers occupy the
lower border and the stripping of the dead begins as Harold is killed in
the main register (see below). One might guess that there were stories
about the deaths of his brothers which were familiar enough for the
designer of the Tapestry to think or know that its patrons would have
wished them included, but he seems to have been unaware of Guy of
Amiens’ statement in the Carmen that Gyrth was killed by none other
than Duke William himself; Wace, who has a great deal to say about
Gyrth at an earlier stage, also reports that he was felled by a blow
delivered by the duke, but more or less concurrently with the death of his
brother Harold.3 One might suspect that both he and Leofwine
commanded significant bodies of men in parts of the field distant from
their brother (although William of Poitiers says that their corpses were
found near his), but even so it is doubtful whether the way in which the
Tapestry shows their deaths is of much significance in terms of the

1
Vegetius, Epitome, p. 50.
2
At Poitiers, as James Campbell has reminded me, the Black Prince ordered his men to mount their
horses to deliver their final attack on the French.
3
Carmen, ll. 471-80; RR, ll. 8819-28.

178
 6 THE BATTLE

battle’s outcome.1 Nor is it impossible that the neatness of the way in

which it shows them dying together is itself rather suspect, and reflective
more of its audience’s desire to see their fall rather than of the precise
historical circumstances in which these events occurred.
William of Poitiers suggests that there was plenty of hard fighting even
after the destruction of those of the English who pursued the French
retirement which almost became a rout; still William’s men faced an
enemy who were closely massed, although inroads were made into their
lines by the weapons of the strongest soldiers.2 Moreover, despite the
success of the subsequent two (according to him) feigned retreats, the
English ranks are still said to have been so dense that the lightly wounded
could not escape and were crushed to death by the mass of their
companions. From this he moves to a list of the more notable of the
duke’s followers who took part in the battle and then to a paean, complete
with classical parallels, of the deeds of the duke himself, who is reported
to have had three horses killed beneath him. Only after this are we told
that as the day declined the English army knew that they could stand
against the Normans no longer, and that one of the reasons for this was
that their king had been killed.
The omission of the details of Harold’s death is not a feature of the
Gesta Guillelmi shared by other sources. The only information given by
William of Poitiers on the matter which is at all relevant is that the
corpse, which lacked all adornment, could not be recognised from the
face but only by certain marks.3 This is in itself perfectly plausible, and
there is something of a parallel to it in William of Malmesbury’s report
1
GG, pp. 138-41. See above, p. 134, for the possibility that they were with heavy infantry fighting in
loose order. Morillo (‘Hastings: An Unusual Battle’, p. 224) suggests that Harold may have ordered
them to lead the counter-attack and that their fall was one reason why it failed, but his claim that the
‘placement of their deaths in the Bayeux Tapestry is consistent with such an interpretation’ only
convinces if every scene from this point to that of William rallying his troops belongs to the same
sequence of events; however, because the stakes (if such they are) in the watercourse and hillock scene
must have been in position when the battle began, and the scene thus probably shows another distinct
stage in the fighting (as, on this view, do probably all those under discussion here) this seems to me
unlikely.
2
fortissimorum militum ferro. Some (for example, Davis and Chibnall’s recent edition, p. 131, and The
Norman Conquest, ed. Brown, p. 33) have translated miles here as ‘knight’, thus increasing the
apparent importance of such men in the fighting. However, the guerriers of Professor Foreville’s
edition (p. 193), ‘soldiers’ of EHD2, p. 241, and ‘fighters’ of The Battle of Hastings, ed. Morillo, p. 13,
are preferable. WP’s vocabulary is classical and as he accordingly tends to employ the terms pedites,
equites and milites in their classical sense, the latter probably usually means soldiers of any type. It is
true that in some cases it seems to denote people of higher status (see, for example, the reference to
Baldwin V of Flanders as Romani imperii miles, GG, p. 30) or specifically horsemen (for example, on
p. 16, where Duke William captures seven milites after fighting fifteen men who gloried in their horses
and arms), but note also the implications of milites uero mediae nobilitatis atque gregarios (translated
by Davis and Chibnall as ‘knights of middling rank and the common soldiers’, GG, pp. 158-9), and
when WP reports (p. 102) that the duke had 50,000 milites at Dives the word is clearly being used in its
more general sense; the same seems likely to be true of the comment about those who cut inroads into
the English line. See further the important study by Morillo, ‘Milites, Knights and Samurai’, pp. 175-7.
3
Ipse carens omni decore, quibusdam signis, nequaquam facie, recognitus est, GG, p. 140.

179
 THE BATTLE OF HASTINGS

that after Stamford Bridge Tostig’s remains were recognised by a wart

between the shoulder blades, and in the discovery at Riccall Landing of
bodies perhaps connected with the events of 1066 which bore the marks
of so many edged weapons as to suggest that they had been struck
repeatedly; similarly, one of the men exhumed in 1996 from a mass grave
connected with the Wars of the Roses battle of Towton in 1461 received a
blade wound which bisected his face when he was probably already
dead.1 Presumably it was not unusual for battles such as these to result in
a sort of fighting fury in which the dead might be savagely disfigured as
well, of course, as later (but sometimes before the conflict was fully over,
if the Tapestry is anything to go by) being stripped of items of value.
Nevertheless, if William of Poitiers has nothing to offer on the way in
which Harold died this is likely to reflect a reluctance on his part to
broach the subject, for whatever reason, rather than any shortage of
information upon it, for in the years immediately after Hastings it is
probable that it was a matter upon which many tales were told.
One of them apparently reached Bishop Guy of Amiens and was
inserted into the Carmen. According to this, Harold was cut down by
Duke William himself, accompanied by Count Eustace of Boulogne,
Hugh the noble heir of Ponthieu and Gilfard, known by his father’s
surname. The first speared the king through shield and chest, the second
beheaded him, the third pierced his belly with a lance and the fourth cut
off his thigh and removed it some distance.2 The Carmen agrees with the
Tapestry that the dead were already being stripped at this stage, but
otherwise its details have found little credence with historians, some of
whom have judged that they are evidence that the poem is not a
contemporary work.3 Certainly they are not convincing, even if the
statement about the thigh may reflect material also found in other sources.
If true, it would be remarkable that Harold happened to be killed by two
of the leading men on the French side – Duke William and Count Eustace
– together with the nephew, Hugh of Ponthieu, of the author of the
Carmen itself. It would also be difficult to see why such a renowned
event has not found its way into the work of other authors, for as R.A.
Brown said, ‘the feat of arms would have been bruited abroad in every
court and chanson in Latin Christendom’.4 This being the case it can be
rejected with some confidence, being perhaps evidence not so much of a
real occurrence as of the kind of legend to which events such as Hastings
can quickly give rise, possibly in this case with the assistance of a degree

1
Above, p. 37. Blood Red Roses, ed. Fiorato, p. 100.
2
Carmen, ll. 533-50.
3
Above, pp. 77-9.
4
Brown, ‘The Battle of Hastings’, 18. The problem could, of course, be avoided if this difficult passage
in the Carmen were read in such a way as to exclude the duke from the quartet, see above, p. 77, n. 1.

180
 6 THE BATTLE

of mendacity on the part of Hugh of Ponthieu and/or his uncle Guy of

Amiens. It might also be remembered in this connection that the reasons
which led Guy to write the Carmen, whatever they were, may also have
caused him to wish to flatter William by attributing to him the death of
the usurper king; that William would have known it to be literally untrue
may have mattered less than one might think to a man whom William of
Poitiers judged prepared to be likened to the greatest heroes of antiquity.
Determining the actual way in which Harold died is not easy. In one of
its most famous scenes, which has undergone much restoration, and
accompanied by the words Hic Harold rex interfectus est (‘Here King
Harold has been killed’), the Bayeux Tapestry shows a French horseman
approaching several Englishmen – first of all a dead standard-bearer lying
on the ground with the pole of his dragon standard beneath him, then a
man with a shield and a spear held overarm who stands next to a figure
carrying a round shield and holding a dragon standard; next to the
standard-bearer, and directly beneath the word Harold, is a soldier
wearing a scabbard whose left hand holds a shield and spear and whose
right hand grasps (as the present form of the Tapestry has it) an arrow
which has evidently entered his skull somewhere in the vicinity of the
right eye, the nasal of his helmet masking the precise area concerned. To
the right of this individual, beneath the words rex interfectus est, a
horseman has evidently just struck with his sword a man with an axe and
scabbard but no shield who is shown as either falling or lying on the
ground (perhaps the former, given that the horse’s left foreleg may be
interposed between the legs of the stricken figure),1 the lower part of the
Frenchman’s weapon apparently coming to rest just above the victim’s
right knee; further right still, a group of Englishmen is resisting an assault
by horsemen coming from the right.
Exactly what this scene is supposed to represent has been the subject of
much debate. Whether the figure grasping the arrow and the differently-
equipped one cut down to the right both represent Harold or whether only
one of them does so, and if so which one, has long seemed a major
problem. At the end of an exhaustive investigation of this matter, which
suggests that both the dead standard-bearer and the dead men near to
Harold’s brothers Gyrth and Leofwine in the scene depicting their demise
are duplicates of their standing counterparts, Brooks and Walker
concluded that both the standing and the fallen figure beneath the words
Harold rex interfectus est are intended to be the king, and that the
Tapestry is accordingly good evidence of the belief that he was hit in the

1
Benoît’s drawing of 1729 (see the Appendix, p. 211) is not easily interpreted at this point and differs
slightly from the Tapestry as subsequently restored; this shows the foreleg entirely behind the figure.

181
 THE BATTLE OF HASTINGS

eye by an arrow.1 Since then, Bernstein has pointed out that a series of
seventeen stitch holes fully visible only on the back of the Tapestry raise
the possibility that the second figure was originally depicted with an
arrow lodged in his skull. However, the nineteenth-century restorers of
the Tapestry muddied the water sufficiently for this to be extremely
doubtful (for the earlier pictorialisations of the scene do not show these
holes), and in any case the fact that this second figure carries equipment
markedly different to that of his supposed counterpart might alone raise
serious doubts as to whether they are really intended to be the same
person.2
Moreover, all this assumes that the first figure has indeed been hit by
an arrow, when it seems almost certain from an examination of the oldest
representation of this part of the Tapestry that in fact he has not. Shortly
before the publication of the second volume of his Les Monumens de la
Monarchie Françoise in 1730, the great French scholar Bernard de
Montfaucon sent the draughtsman Antoine Benoît to Bayeux to draw the
Tapestry as accurately as possible; from these drawings, which still
survive in the Bibliothèque Nationale in Paris, the engraved plates in the
second volume of Les Monumens were made, and shortly afterwards they
were also used to create a different set of engravings to accompany a
paper delivered to the French Academie Royale des Inscriptions et
Belles-Lettres by the historian Antoine Lancelot. The latter assured his
audience that his illustrations were a very accurate copy of Benoît’s work,
but in fact this was not really true either of them or of the engravings
done for Montfaucon, and on the figure supposedly struck by an arrow
they differ significantly. Montfaucon’s engraver showed an individual
holding in his right hand a missile (represented as a continuous line)

1
B&W, 23-34, with comment on the probable accuracy of the essential elements of the nineteenth-
century restoration of the Tapestry. Their argument is judged ‘almost certainly right’ by Wilson, BT, p.
194. Gibbs-Smith had previously contended (The Bayeux Tapestry, p. 15) that the first figure has been
hit by an arrow but is not Harold; this was followed (among others) by Brown, The Normans and the
Norman Conquest, p. 173, n. 154. In 1966 Gibbs-Smith (‘What the Bayeux Tapestry does not show’)
suggested that the designer originally intended this man to be throwing a spear but substituted an arrow
when the raised arm had already been embroidered because ‘he did not want any more horizontal lines
crossing his composition’. I owe this reference to Barlow, The Godwins, p. 107.
2
Some of the holes are visible in the plate in BT. Bernstein, pp. 148-52, with illustrations of both the
front and rear of the Tapestry, considers the possibility that they may be the subsequently-removed
work of a restorer. Wilson (BT, pp. 194-5) notes their absence from early reproductions of the Tapestry
and comments that ‘as there is no place where they are more obvious, they are presumably a fabrication
of the last century’. The different weaponry of the second figure is not dealt with by B&W, but their
discussion of parallels in manuscript depictions of biblical scenes and other supposed duplicates within
the Tapestry itself leans heavily on the argument that the figures (for example, the standard-bearer) are
the same person because they look the same; this is clearly not the case with the two figures in
question. Bernstein (pp. 144-7, and 218, n. 9) provides examples where, as (allegedly) here, the hose of
different figures intended to represent the same person changes colour, but weaponry is arguably
another matter, and a context in which the military-minded audience of the Tapestry may have viewed
inconsistency with some surprise; see below for a probable solution of this problem, and the Appendix,
pp. 203-13, for a more detailed treatment of what follows.

182
 6 THE BATTLE

which extends to the lower edge of his helmet, while Lancelot’s depicted
it as a broken line which does not touch the helmet and extends well to
the left of the hand in which it is held. In the latter case, then, this soldier
appears to be a figure possessed not of an arrow which has just hit him,
but of a spear which he is about to use against the enemy. Moreover, this
conclusion is substantially confirmed by the drawing from which the
engravings were made. Here, the missile reaches to the lower edge of the
helmet (as in Montfaucon) but also extends as a broken line further to the
left of the hand than Montfaucon’s engraver indicated, if not quite as far
as Lancelot’s did. It is both considerably longer than the arrows sticking
in the shield of the same figure and unlike them it has no flights. Thus, it
is probably not an arrow, but all that Benoît could make out of a spear, in
a scene that had clearly been damaged. The soldier who holds it flanks
the standard bearer on the one side and is apparently a close companion
of the very similar figure, who also holds a spear in his raised right hand,
on the other. Together the three may well represent the English king’s
bodyguard. Thus, there is no very strong reason to suppose that the
Bayeux Tapestry is part of the corpus of evidence that Harold was hit by
an arrow, and the only one of its figures which can plausibly be taken as
representing the king is the one struck by the French horseman; this, of
course, is why this man’s equipment is different to that of the right-hand
member of the trio.1 Benoît’s drawing also suggests the possibility that
the inscription did not originally read Hic Harold rex interfectus est but
perhaps Hic Harold rex in terra iactus est (‘Here King Harold has been
struck to the ground’); if so, this was a precise description of the figure
who has just received the horseman’s blow.
Yet if the apparent evidence of the present form of the Bayeux
Tapestry that Harold was struck by an arrow is very likely to be the work
of those who drew and restored it in the nineteenth century, an arrow
really is mentioned in connection with his demise in two sources more or
less contemporary with the Tapestry. Baudri of Bourgeuil says that he
was killed by an arrow without specifying where it struck him, while
Amatus of Montecassino wrote c.1080 that he was hit by an arrow in the
eye.2 These statements may represent a belief about Harold’s death that
was fairly widespread, and which some years later also reached William
of Malmesbury and Henry of Huntingdon, for the first says that Harold
was killed by an arrow which entered his brain, the second that he was hit
in the eye by one of the missiles fired by a group of archers which had
1
Dr David Hill generously sent me his photograph of the Benoît drawing of this part of the Tapestry,
one of a number of Benoît’s work taken in the Bibliothèque Nationale for himself and John McSween.
Although it has fallen to me to be the first to publish the results of their work, they had already arrived
at the conclusion outlined above. It will be obvious that the publication of a full set of photographs of
the Benoît drawings is much to be desired.
2
AC, l. 463; on Amatus, see above, p. 92.

183
 THE BATTLE OF HASTINGS

been ordered by William to shoot into the air in order to blind the enemy;
he then sank to the ground and was dispatched along with his brothers
Gyrth and Leofwine by a group of horsemen, and at much the same time
twenty of their companions broke through the English line and carried off
the standard, although it cost some their lives.1
William of Malmesbury thus agrees with Baudri in suggesting that the
arrow was the sole cause of Harold’s demise, and this is plausible
enough, for if a missile had entered the midbrain through an eye socket,
which would not have been difficult, death would have been almost
instantaneous.2 On the other hand it is also possible that it would simply
have caused a grievous wound, and that he might then have had to be cut
down in the way that Henry of Huntingdon describes. William of
Malmesbury says that a soldier hacked at his thigh with a sword as he lay
on the ground, for which disgraceful act William expelled him from the
ducal forces. Our earlier examination of the Gesta Regum established that
William of Malmesbury, while generally condemnatory of English
morals, regarded the Conquest as a disaster for his country and may have
manufactured or adapted from chansons details which allowed him to
present Harold’s fighting prowess as matching that of William.3 This
being so, one might wonder whether he would have been inclined to
report all that he may have heard about what happened to Harold after he
was hit by the arrow, the detail about the soldier who struck at his thigh
and was disgraced as a result possibly being a relic of more extensive
material which, had it been transmitted, would have contradicted the
belief that the courageous English king died very quickly. Baudri of
Bourgeuil, on the other hand, whose account of the battle is not a
particularly detailed one except in recording the exploits of Duke
William, the father of the lady to whom his poem was addressed, may not
have thought that Harold’s death merited any more space than he gave it;
alternatively, it is possible that both he and William of Poitiers had heard
gory accounts of Harold’s last moments and felt that it was not to
William’s credit or advantage to repeat them.
It may be then that Henry of Huntingdon and possibly the Carmen, if
there is any truth at all behind its account, are together enough to suggest
that the English king was wounded by an arrow and then cut down later,
his corpse eventually being so badly damaged that it became almost

1
GR, i. 454-7; HA, pp. 394-5. Greenway (HA, p. xxxv) says that it is ‘not obvious’ that HH was
acquainted directly with Baudri’s poetry.
2
I am grateful to Dr George Brown for discussion of this matter. Foys, ‘Pulling the Arrow Out’, by
adopting an attitude towards the complexities of extracting information from later sources which in my
opinion borders on the simplistic, rejects all the evidence on Harold and the arrow, from Amatus of
Montecassino and Baudri of Bourgeuil onwards.
3
Above, pp. 58-60.

184
 6 THE BATTLE

unrecognisable.1 Whether even the French soldiers responsible knew the

identity of their victim and therefore the significance of their action is
perhaps not certain, although it is not unlikely that in after times some
were willing to put forward spurious claims about these matters. If so,
none seems to have become universally accepted by posterity. Neither his
extensive reading in the documentary sources nor oral tradition seem to
have given Wace any clue as to who was really involved. He first reports
that Harold was wounded by an arrow which struck him on the right eye
and put it out before he speaks of the French feigned flight and of fighting
involving a great many of their lords. Only, he says, when they eventually
penetrated to the English standard did they find the wounded king still
standing there, and one of the soldiers struck him on the helmet and beat
him to the ground; as he tried to recover he was struck again by a blow
which pierced his thigh to the bone; eventually there was such a press of
men around him seeking his death that, says Wace, ‘I know not who it
was that slew him’, and, ‘I do not tell, and I do not indeed know, for I
was not there to see, and have not heard say, who it was that smote down
king Harold.2
The way in which the Tapestry shows these events can be taken to
mean that the king, with his own men to either side of him, was standing
within the shield-wall when he was struck down,3 and if so it is by no
means impossible that he was wounded some time before being killed, or
simply killed outright by an arrow with no further French intervention.
Despite the fact that William of Jumièges is one of our earliest sources it
is very difficult to accept the apparent statement that he perished in the
first onset,4 and the Tapestry and Carmen together speak fairly strongly
for the view that he died in the final stages of the conflict.5 It would not
be surprising if news of this event precipitated the final English rout, as
the Carmen and William of Malmesbury claim, and was thus the final

1
Shortly before 1820 Sir Godfrey Webster commissioned the artist Frank Wilkin to produce a painting
of the battle to hang in the Great Hall of Battle Abbey. The work originally measured 31 by 17 feet and
depicts the Conqueror (Sir Godfrey apparently sat for the figure himself) at the head of his men being
shown Harold’s body and offered his crown; the corpse exhibits scant sign of injury, but behind it a
French soldier holds up an arrow. Subsequently given to Hastings Corporation, which in 1979
presented it to English Heritage; it has recently been restored to its original location following
restoration. Sir Godfrey, who also built New Pond, evidently paid 2,000 guineas for the picture, and
later sold the medieval archives of the abbey (they are today in the U.S.A.) for a far smaller sum; the
nineteenth-century estate archive in the East Sussex Record Office contains much evidence of the
anguish of local tradesmen at his reluctance to meet the bills incurred during his extensive restoration
of his seat. I am grateful to Victoria Williams, Curator of Hastings Museum, for information on
Wilkin’s Battle of Hastings.
2
RR, ll. 8161-4, 8803-18, 8829-34, 8851-6. I have quoted from Edgar Taylor’s translation, pp. 254-5. It
is probable that elements of Wace’s account were drawn from sources such as WP, WM, HH and
perhaps the Tapestry (see above, pp. 95-7).
3
BT, p. 194.
4
Above, pp. 80-1.
5
Above, p. 180.

185
 THE BATTLE OF HASTINGS

turning point in the battle.1 William of Poitiers, however, atttributes it to a

variety of reasons: as the day was ending, he says, the English army
understood that it could stand against the Normans no longer, for they
knew that they had lost many men and that these included the king and
his two brothers with no small number of magnates; furthermore, those
who remained were virtually exhausted and they could expect no
reinforcement, while they saw that their enemies were not much
weakened by their casualties and that the duke’s ardour was such that he
would accept nothing less than victory; hence, they turned to flight, some
on horses they had taken, others on foot.2 The sun set at Battle on the
afternoon of Saturday, 14 October, 1066 at about 16:54 GMT; it would
have been fairly dark by 17:54 and completely so by 18:24, while the
moon did not rise until 23:12. William of Jumièges says that the English
turned tail at nightfall and John of Worcester agrees that fighting lasted
until dusk, when Harold himself fell.3 A last stand by his housecarls tends
to feature in re-enactments of the battle and doubtless will continue to do
so; it is a pity that the sources provide no grounds for believing that any
such thing occurred.4
Probably, then, much of the pursuit took place in the dark, and the
Carmen says specifically that night was falling as God made the duke the
victor, and that only darkness and flight through the thickets of the dense
forest saved the defeated English, as William rested on the battlefield
among the dead and awaited the return of daylight; however, Guy of
Amiens also asserts that his own nephew Hugh of Ponthieu spent the
night hunting down the enemy,5 and it is probable that the French now
reaped the maximum benefit from their cavalry, who may have killed as
many of the English in this final phase of the battle as at any earlier time.6
time.6 William of Poitiers reports that many left their bodies in dense
woods and that the Normans, although ignorant of the area, pursued them
zealously. Harold’s own forces had harried the Norwegians all the way to
their ships after the battle of Stamford Bridge, and Duke William was
evidently no less aware that an effective pursuit would render his victory
all the more decisive. However, his enemies made a stand when they
came to a broken rampart and a series of ditches; just as Count Eustace of

1
Carmen, ll. 551-4; WM, GR, i. 454-5;
2
GG, pp. 136-7.
3
GND, ii. 168-9; JW, ii. 604-5.
4
Freeman (FNC, iii. 500-1) did his best to argue otherwise, but there is no evidence for his ‘quarter
was neither given nor asked; not a man of the comitatus fled; not a man was taken captive’.
5
Carmen, ll. 557-66. I accept here Barlow’s identification of Hugh of Ponthieu as the Hectorides of the
Carmen.
6
The presence of the Weald to their rear would not have facilitated the English withdrawal, and one
might compare events after the battle of Towton in 1461, when the Lancastrian rout was hindered by a
stream in spate and they suffered heavy casualties during a pursuit which extended for ten miles to the
gates of York, Blood Red Roses, ed. Fiorato, p. 23.

186
 6 THE BATTLE

Boulogne was preparing to retire with fifty men the duke came upon the
scene, and although he thought the English newly-arrived reinforcements
ignored Eustace’s advice that he should withdraw; then he charged and
dispersed his opponents. Even so, Eustace himself received a blow which
caused the blood to gush from his nose and mouth and some of the more
noble Normans were killed there. Orderic Vitalis’ additions to William of
Jumièges assert that the pursuit lasted throughout the night and that high
grass concealed a rampart which caused the Norman horsemen to fall one
upon the other so that they were crushed. It was said that almost 15,000
fell there (whether the word ‘there’ refers to events around the rampart or,
as seems more likely,1 the battle as a whole, is not completely clear). In
his later Historia Æcclesiastica Orderic fused his earlier story with that of
William of Poitiers by claiming that it was the French disaster at the
rampart that caused the English to rally around this earthwork and the
series of ditches, where they inflicted heavy losses on the Normans.
There, he says (and again this may refer to the battle as a whole),
Engenulf of Laigle and many others fell, and those who were present said
that 15,000 Normans were killed. The roles of Count Eustace and the
duke in this final stage of the conflict, as reported by William of Poitiers,
Orderic decided to omit.2 There is a third account of a disaster to the
French in the closing stages of the battle in the late-twelfth-century Battle
Abbey Chronicle, which says that near the battlefield was a great ditch
concealed by brambles which engulfed large numbers, especially of
Normans, and that it is called Malfosse to this day as a result. One might
well, however, wonder whether this account, which says nothing of any
English involvement, and those of William of Poitiers and Orderic
Vitalis, do necessarily all refer to events at the same location.3

1
Similarly, Brown, ‘The Battle of Hastings’, 19.
2
GG, pp. 138-9; GND, ii. 168-9; HÆ, ii. 176-7.
3
Chronicle of Battle Abbey, ed. Searle, pp. 38-9. Stevenson, ‘Senlac and the Malfossé’, 302-3, noted
that Laigle is only a few miles from OV’s monastery of St Évroul and suggested that he had discussed
matters with members of Engenulf’s family or their men; similarly, Brown, ‘The Battle of Hastings’,
19, who also commented on the awkward way in which he later combined the story with that of WP.
There have been various attempts to locate the site of this incident or incidents, whose historicity
Brown clearly doubted (along with the accounts of deaths in ditches during the battle). Freeman (FNC,
iii. 502-3, 750, and map) thought the ditch one of the drainage channels on the north-eastern slope of
the Battle Abbey ridge; Baring (‘On the Battle of Hastings’, pp. 229-30) opted for a channel to the west
of the ridge later known as Manser’s Shaw, a name which in his opinion derived from Malfosse;
similarly, Burne, The Battlefields of England, pp. 25, 43, although he followed Stevenson in rejecting
Baring’s derivation of the later name. Chevallier, ‘Where was Malfosse?’, argued for a location over a
mile to the north of the battlefield at Oakwood Gill. This was accepted by Lemmon, The Field of
Hastings, pp. 33-5 and ‘The Campaign of 1066’, pp. 111-13, who thought the site the fruit of deliberate
‘reconnaissance and occupation by order of some unknown Saxon leader’, and supplied a number of
details of his own devising, such as the statement that Eustace of Boulogne ‘was struck down by a
battle-axe, presumably wielded by some Housecarl who had contrived to creep up unobserved’.
Bradbury, The Battle of Hastings, p. 210, describes Chevallier’s location as reasonably certain; see also
Clephane-Cameron, et al., The 1066 Malfosse Walk.

187
 THE BATTLE OF HASTINGS

The following morning, says the Carmen, the duke took up and buried
his own dead, while leaving the bodies of the English as carrion for birds
and beasts. Harold’s lacerated corpse he wrapped in purple linen and took
with him when he returned to his camp on the coast, where he
subsequently rejected a request that it be handed over for interment to the
dead king’s mother, despite being offered its weight in gold in return.
Later, a man of part Norman and part English descent, who was also
connected to Harold, carried out the duke’s order that it be placed on the
top of a cliff, and it was laid beneath a tombstone which read:
By command of the duke, you repose here, King Harold, so that you may
1
remain guardian of the shore and the ocean.
That the corpses of the rest of the English were left as carrion is credible
enough, for it seems to have been the practice of the period not to bury
the enemy dead, as evidence on the battle of Stamford Bridge and other
material indicates.2 William of Poitiers states that the battlefield was
covered far and wide with the flower of the English nobility and youth,
that Harold’s two brothers were found near him and (as noted earlier) that
he could only be recognised by certain signs; refusing the mother’s offer
of their weight in gold for the remains after they were carried into his
camp, Duke William allowed them to be laid to rest by William Malet,
considering it unseemly to accept gold in such circumstances and to fall
in with the mother’s wishes when many men lay unburied as a result of
her son’s greed; it was said jokingly that he should be buried as custodian
of the shore and ocean which he had previously, in his madness, occupied
in arms, and accordingly he was put in a tumulus on the seashore.3
William of Malmesbury chose to reject this part of the Gesta Guillelmi,
stating that the Conqueror sent Harold’s mother the corpse without

1
Carmen, ll. 567-92; see above, p. 77. The meaning in this context of compater, the word used to
indicate the man’s relationship to Harold, is uncertain. M&M, p. 39, translated ‘comrade’, Barlow
‘relative’.
2
Above, p. 36-7. Note, for example, the way in which poetry tends to speak of the bodies of the
defeated serving as food for birds and beasts, for example in the piece on the battle of Brunanburh in
937 (ASC, i. 109) and in Ottar the Black’s celebration of the victories of King Cnut in his Knútsdrápa
(EHD1, pp. 335-6); also, according to EE, pp. 28-9, after the battle of Assandun in 1016 the victors
buried their own dead while stripping the enemy and leaving them as carrion. WP’s claim (GG, pp.
142-3) that the duke allowed those who wished to collect the English dead for burial to do so may
indicate unease at such practices, but does not necessarily mean (pace Gillingham, ‘Holding to the
Rules of War’, 2) that many were actually buried. However, if the mass burial found at Riccall Landing
(above, p. 37) contains the bodies of Scandinavians who fell in 1066 it is not impossible that they were
buried by the English. There is a valuable treatment of evidence on Anglo-Saxon battlefield burial
practices in J. Hooper, ‘The “Rows of the Battle Swan”’, while Gillingham, ‘The Rules of War’, gives
a range of interesting ideas and references, often to material from the continent.
3
GG, pp. 138-41; on the possible ultimate derivation from the Iliad of the statement in the Carmen and
GG that William was offered the body’s weight in gold, see above, p. 88. Davis and Chibnall (141, n.
5) note Pompey’s burial in a tumulus on the seashore in Lucan’s Pharsalia, although the corpse, which
had no head, had first been cremated. WP’s account of these events is very close to that of the Carmen,
but clearly (as he is able to give the name of William Malet) not solely dependent upon it, if dependent
at all.

188
 6 THE BATTLE

payment, although she had offered much, and that she had it buried at the
church of Holy Cross at Waltham in Essex, a church which he had built
himself and filled with canons.1 Later, Waltham claimed that two
members of their community accompanied him to the battle and sought
his body from William after it in exchange for ten marks of gold; 2 the
gold the duke is said to have rejected, while granting their request for the
remains, which were buried at Waltham with great honour. Their late
twelfth-century chronicle, from which these details are drawn, may here
be almost entirely fictional, but we owe to it the story that the corpse
could not be recognised until one of the pair returned home and brought
back with him Edith Swan-neck, Harold’s handfast wife, who because of
her intimacy with the king could recognise the secret marks upon it better
than others.3
The numbers who fell in the battle are no clearer than those of the
totals engaged. In commenting on the entire campaign the annalist of
Nieder-Altaich on the Danube reports French casualties as 12,000, while
we have just seen that Orderic Vitalis is perhaps to be taken as meaning
that 15,000 (upon whether simply Normans or French as a whole he is
inconsistent) were killed at the battle of Hastings itself.4 Orderic’s
Engenulf of Laigle is the only Norman known to have died there, but
there is a slightly longer list available from Domesday Book and other
records of the named English dead, in addition to Harold and his two
brothers. Breme and Edric the Deacon, who both held land in Suffolk, are
known from William’s great survey to have been among them, along with
Ælfric, a man of some note who held sixteen hides in Huntingdonshire
from the abbot of Ramsey. The twelfth-century chronicle of the
monastery of Abingdon adds Godric, sheriff of Berkshire, and the local
landowner Thurkill, while the name of Abbot Ælfwig of New Minster,
Winchester is contributed by the records of his own house, and
(anonymous) dead of the men of the abbey of Bury St Edmunds are
mentioned in a writ addressed to Abbot Baldwin early in the Conqueror’s
reign.5

1
GR, i. 460-1.
2
Note the ten talents of gold which formed part of the ransom given to Achilles for Hector’s body,
Iliad, xxiv, l. 232.
3
The Waltham Chronicle, ed. Watkiss and Chibnall, pp. 50-7. The author of this work wrote shortly
after 1177, and naturally rejects tales that Harold lived in a cave at Canterbury after the battle and was
eventually buried at Chester. On the late legends, see ODNB on Harold II and Fellows-Jensen, ‘The
Myth of Harold II’s Survival in the Scandinavian Sources’. The notion that he had survived the conflict
could, of course, be paralleled in the histories of other leaders whose deaths have had ill consequences
for their people. On Edith Swan-neck, whose connection with Harold may have been of long standing
and produced many children, see FNC, iii. 763-5; Barlow, The Godwins, pp. 55-6, 120-1.
4
Above, pp. 33, 187.
5
DB, ii. 409b, 449a; i. 207a-208a; Chronicon Monasterii de Abingdon, ed. Stevenson, i. 484-5, 490-1;
Heads of Religious Houses, ed. Knowles, p. 81; Regesta Regum Anglo-Normannorum, ed. Bates, No.
37, p. 198 (trans. EHD2, No. 238). See also above, pp. 127-8, for unnamed English dead mentioned in

189
 THE BATTLE OF HASTINGS

The English defeat in 1066 can be attributed to many factors. The

fact that they had to prepare for two invasions rather than one, and that
both arrived much later in the year than might reasonably have been
expected, obviously tested the defenders’ resources to the limit. When the
provisions of the forces which Harold had stationed on the south coast ran
out on 8 September this probably does not represent a failure in the
workings of the late Anglo-Saxon administrative system so much as the
fact that the system itself was not designed to cope with such unusual
circumstances. If William brought this about through deliberately
delaying his arrival until he knew his opponents had dispersed then he
had played a clever and risky game with an iron nerve; if, as the French
sources say, he had just been held in port by contrary winds he was
simply extremely lucky. The reasons for the outcome of the battle itself
are naturally not easy to determine considering that we know neither the
absolute nor the relative size of the two armies and are ill-informed on
their composition and deployment on the battlefield; nor, of course, can
much be had in the way of detail about the course of fighting which lasted
(if the pursuit is included) well over eight hours. Nevertheless, the sheer
length of the conflict must reveal something, for neither side can have
had an overwhelming superiority of numbers or of military technique. It
is a commonplace about Hastings that Harold fought before he should
have done and that had he waited his men would have been both more
numerous and better rested after their exertions on the Stamford Bridge
campaign. It is also easy to believe that the earls Edwin of Mercia and his
brother Morcar of Northumbria were either not present or withdrew
earlier than they should have done, either because of their losses at
Fulford or because the force of old rivalries between their family and
Harold’s reappeared in October, 1066; once their enemy Tostig was dead
they were perhaps as ready to do a deal with Duke William as with King
Harold.1 One should probably also allow a role (if a completely
unquantifiable one) in reducing support for Harold to his broken oath, the
appearance of Halley’s Comet as a perceived harbinger of disaster and
the duke’s possession of a papal banner, carried into battle at the head of
his army. Yet while the significance of all these points can be

Domesday, and FNC, iii. 425-7, 730-1, although Freeman’s belief that Abbot Ælfwig was Harold’s
uncle now seems mistaken. A twelfth-century addition to the E text of the Anglo-Saxon Chronicle, i.
198, says that Abbot Leofric of Peterborough returned home sick and died on 31 October.
1
The only source to offer information upon their behaviour is JW, ii. 604-5, whose words (Cuius morte
audita, comites Eduuinus et Morkarus, qui se cum suis certamini substraxere, Lundoniam venere...) are
not without a degree of ambiguity. They mean either that Edwin and Morcar were not present at the
battle with their men and went to London when they heard of Harold’s death, or that they withdrew
from it during its course and went to London once they knew that Harold had been killed. Freeman
(FNC, iii. 421-2) had no doubt that the former meaning is correct, and suggested that they hoped to
divide the kingdom with William, themselves taking Mercia and Northumbria and leaving to him
Wessex and East Anglia.

190
 6 THE BATTLE

acknowledged, it is not obvious that he was short of men in the battle

itself, or that those he had were incapable of fighting with great
determination and vigour. It is quite unclear what proportion of his force
had been at Stamford Bridge, and those who had may well have travelled
to and from Yorkshire on horseback, for both the Carmen and William of
Poitiers say that the English dismounted from their horses before fighting
William.1 Certainly, repeated statements about the density of their lines
suggest that Harold’s army was more than large enough to cover the
frontage on which it fought, and nor does its morale seem to have been a
problem until after news spread that the king had been killed. It may be,
however, that shire levies of the less well-equipped type formed a greater
proportion of it than would have been the case if Stamford Bridge had
never been fought, or if Harold had waited longer for his army to gather.
Yet even on the timing of the battle it is impossible to be sure that he
erred, for according to the D text of the Anglo-Saxon Chronicle William
received reinforcements from oversea shortly after the battle, and Harold
may have both known that this was a possibility and acted with it in
mind.2
Indeed, as far as one can see the really critical events (and flaws in the
English king’s leadership, if such there were) took place not before the
conflict, but during it. Thus, while it may not have been unusual in this
period for infantry to fight in very close order indeed, and this was
perhaps often very advantageous, it may be that he did deploy in an area
too narrow to satisfactorily accommodate the numbers at his disposal, and
that (following the Anglo-Saxon Chronicle D text) William reaped the
maximum advantage from this by attacking before they were properly
arrayed. Even so, the English lines proved very difficult to break down,
and their tenure of a watercourse, which may well have been that running
west in the valley to the south, also gave the French considerable trouble,
for here the progress of their cavalry may have been hindered by
sharpened stakes planted in the bed of the stream, and specialist English
light infantry and perhaps heavy infantry too also fought in this area.3
Such features of the English conduct of the battle are a reminder that their
army was the product of a formidably-organized state with an impressive
past record in gearing its people for war, led by a commander who was
well aware that the French and their cavalry needed to be taken
seriously.4 In this, he was not mistaken. Duke William had more
experience of leading his men in war than his counterpart, and the troops
1
Carmen, l. 377; GG, pp. 126-9. See above, p. 123, n. 6, 128, 137, on the possible earlier use of forces
on horseback by the English and note also the men fleeing on horses in the final much-restored scene of
the Tapestry, although these mounts may, of course, have been plundered from the French.
2
ASC D, i. 200.
3
Above, pp. 121-4.
4
Above, pp. 104-12.

191
 THE BATTLE OF HASTINGS

who had gathered under his command had everything to gain by victory
and everything to lose in defeat. All are likely to have been reasonably
well-equipped, and while they may not have had the possibly years of
training given to its levies by the late Anglo-Saxon state, they had been
under the duke’s control for months and probably done as much
preparation in terms of manoeuvring and fighting together as the time
available allowed. The fighting capacities of this ad hoc force need
occasion no surprise: just over thirty years later an army of a similarly
confederate nature assembled as the result of the preaching of the First
Crusade and then proceeded to win the astonishing series of victories
over the Muslims which eventually delivered the city of Jerusalem into its
possession.
As already noted, the length of the battle strongly suggests that the two
forces were fairly equally matched, and had the English been content to
stand on the defensive all day it is doubtful whether they could have been
defeated. In this context their pursuit of an enemy who was retiring,
whether in a real or a simulated flight, is critical. The pursuers, who were
on foot, left their flanks exposed and were immediately vulnerable to
cavalry attacks which may have inflicted heavy losses upon them.
Whether the pursuit was a deliberate counter-attack by commanders who
ought to have stood their ground or the result of indiscipline on the part of
their troops is not clear, but the English fighting methods were not at this
point the equal of those of the French, and it looks as though this played a
major role in giving victory to their opponents. Contrary to the views of
some earlier historians of the battle, it was not their lack of cavalry which
proved decisive, but their failure or inability to persist in the defensive
strategy which may well have led to the eventual failure of the French
assaults launched against them. Had the day ended in a stalemate Harold
would still have been king of England and William’s credibility may have
been seriously damaged. It is a possible criticism of Harold, or of his
subordinate commanders and his troops, that they were too impatient to
secure a decisive victory, when the way in which the French had made
repeated attacks on a strong defensive position might have told them that
William needed such a victory just as much and probably more than they
did.
William’s army may not have been better than that of the English, but
as a commander he was probably Harold’s superior.1 Establishing a
bridge-head on the English coast and remaining there until his enemy
came to him was the act of a cautious commander and perhaps also of a
very shrewd one. It is questionable whether he could have maintained
himself in the Hastings area for much longer than he did, especially if
1
Note Morillo’s comment (‘Hastings: An Unusual Battle’, p. 227) that ‘Given essentially equal armies,
William simply outgeneraled Harold and had a bit more luck’.

192
 6 THE BATTLE

English naval forces had been sent to cut off his links with Normandy as
some sources say that they were;1 it may also be indicative of the state of
his men that when they eventually reached Dover some ate fresh meat
and drank water, with the result that many died of dysentery. 2 However,
he had everything to gain by staying where he was providing there was a
good chance that Harold eventually came to him, and that William was
willing to play this game may have stemmed from the knowledge of his
adversary’s character gained during the months the latter had spent in
Normandy not long before, months which may have told William more
about Harold than Harold learnt about William.
Once the appearance of the English army presented him with his great
opportunity he took it, advancing to prevent the enemy attacking his
camp and ships, moving against them on the morning of the conflict
before their lines were properly formed, repeatedly carrying the battle to
them during a long day’s fighting, putting both his archers and cavalry to
good use, and ensuring that there was a relentless pursuit as night fell and
his foes fled the field. Although William of Poitiers claims3 that he
directed everything by voice and gesture it may well be that during the
fighting much was owed to the skills of subordinate commanders, but the
duke can still be credited with halting the rout which almost cost him the
day and with the personal military prowess which is unlikely to be a
complete invention of the French sources. Of course, that there is much
less evidence on Harold’s role in the battle makes any detailed
comparison of the two men impossible. Even so, if William had been
lucky in being able to land on the English coast when his enemy was in
Yorkshire, he may also have had considerable luck in the battle itself,
passing unscathed through the fighting while Harold, who may never
have fought hand-to-hand at all,4 was either seriously injured or killed
outright by an arrow shot from a distance. This particular stroke of good
fortune may have finally turned the battle in the duke’s favour, but if so it
can hardly be said to have been undeserved, considering his skills as a
strategist, tactician and fighter; and married to them was the iron will
which would be evident again during his reign as king of England.

1
Above, p. 157.
2
WP, GG, pp. 144-5.
3
GG, pp. 126-7.
4
WM’s evidence to the contrary is unsafe, see above, p. 59.

193
 7

EPILOGUE
Although it might readily be thought that William conquered England by
fighting only one battle, in the months following Hastings he would not
appear to have been of that opinion himself. It looks very much as if he
feared that another army might be brought against him, and after
returning to his camp,1 and according to the D text of the Anglo-Saxon
Chronicle receiving reinforcements from oversea, he marched eastwards
to deal with Romney, the scene of a conflict in which those of his
followers who had landed there by mistake had been routed. From there
he moved to Dover, where the defences were surrendered despite the
large number of Englishmen present, and shortly afterwards the men of
Canterbury too submitted. William then made a circuit to the south and
west of London, crossing the Thames at Wallingford and ravaging as he
went until he reached Berkhamsted, where he was met by Archbishop
Ealdred of York, Edgar the Ætheling, the earls Edwin and Morcar and the
leading citizens of London; they gave hostages and swore oaths of fealty,
whereupon he promised them that he would be a good lord. This
submission was no foregone conclusion. It is known from the Chronicle
D text and John of Worcester that there had been a proposal that Edgar
should be made king, and that Edwin and Morcar had undertaken to fight
for him, although according to John they withdrew their forces as many
others were preparing for battle.2 If so, they may by this time have been
attempting to balance the risks of opposing William against the possible
benefits of coming to terms with him, and when they opted for the latter
course the duke, who cannot have relished the prospect of further
fighting, was no doubt willing to appear conciliatory. Orderic Vitalis says
1
The Carmen, l. 597, says that he stayed at Hastings a fortnight.
2
WP, GG, pp. 142-9; ASC, D, i. 199-200; JW, ii. 604-7. The latter lists the shires ravaged by the
French as Sussex, Kent, Hampshire, Middlesex and Hertfordshire. The Carmen, ll. 597-752, differs in
having William send to Winchester to secure the submission of Queen Edith and in suggesting that a
siege of London was necessary before its citizens accepted him, abandoning Edgar, whom they had
made king; WJ, GND, ii. 170-1, also mentions fighting against the Londoners, and recently-discovered
archaeological finds have been claimed as evidence of it (Carmen, pp. lxxxvii-viii). WP, GG, pp. 146-
7, agrees that Edgar had been chosen king; similarly, WM, GR, i. 416-7.

195
 THE BATTLE OF HASTINGS

that when he made peace with Edwin he granted him authority over both
his brother Morcar and almost a third of England.1 If so, this was not a
promise which he necessarily intended keeping, or which subordinates
eager to taste the full fruits of victory would have been happy for him to
keep.
The submission at Berkhamsted was eventually followed by his
coronation by Archbishop Ealdred in Westminster Abbey on Christmas
Day. Early the following year he returned to Normandy, taking care that
the remaining English leaders accompanied him, and his new kingdom
did not see him again until December, 1067. There had, however, been
trouble while he was away, for when his half-brother Odo of Bayeux and
Hugh of Montfort were beyond Thames with the greater part of their
men, Count Eustace of Boulogne, who was presumably disappointed with
whatever share of the spoils he had received for his exploits at Hastings,
and perhaps mindful that he might be able to claim the throne as the
husband of Edward the Confessor’s (dead) sister, accepted the invitation
of the English to launch an attack upon the castle at Dover. This failed,
but as an event it marked the beginning of several years of struggle for
the Normans as they attempted to completely sudue the country over
which William now ruled. In 1068, and with an army said to have
included Englishmen, he marched against Exeter, which was refusing to
accept him as king, and after a siege of eighteen days it prudently
submitted. He built a castle within its walls and subsequently moved into
the midlands and north, erecting further castles in Nottingham, York and
Lincoln. Yet these measures did not prevent a major rebellion north of the
Humber in 1069, almost certainly co-ordinated with risings in the south-
west and on the Welsh border, and eventually assisted by the arrival of a
fleet from Denmark. A force which included Edgar the Ætheling attacked
the castle at York but William came upon them swiftly and dispersed
them after what may have been virtually a pitched battle.2 He then built a
second castle in the city3 before going to deal with events elsewhere, but
the subsequent appearance of forces sent by King Swegen Estrithson of
Denmark brought another crisis, and both the castles in York fell to them
and their English allies. When William marched for York for the third
time he was determined that he would never need to do so again; delayed
for three weeks because of the difficulty of crossing the Aire near
Pontefract against opposition, when he eventually reached the city he
found that the Danes had fled, and then began the process of extensive
devastation now known as the Harrying of the North. Nowhere else, says

1
OV, HÆ, ii. 214-5. This may have been taken from the part of WP’s GG which is now lost.
2
OV, HÆ, ii. 222-3; similarly, ASC D, i. 203, which says that William came upon them unexpectedly
and many hundreds were killed.
3
I follow OV on this; ASC D, i. 202, says that he had built both the York castles in 1068.

196
 7 EPILOGUE

Orderic Vitalis, had he shown such cruelty, commanding that the region
should be stripped of all sustenance, with the consequence that more than
100,000 people died of famine; the figure may or may not be
exaggerated, but the effects of his action on the countryside were still
visible when his commissioners carried out the Domesday survey in 1086
and were obliged to record many villages as ‘waste’; a recent
examination of this evidence has suggested that ‘Yorkshire could well
have been a desert in 1070’.1
The Harrying of the North was intended to put an end to English
opposition and largely did so. The following spring the Danes returned
home probably after being bought off. In 1071 William received the
surrender of English rebels who included Earl Morcar at Ely, and the next
year campaigned in Scotland, where King Malcolm submitted. The Danes
were a threat again in 1075 and in 1085, in the former year in conjunction
with a plot against the king of which the ringleaders were the last
surviving native earl, Waltheof (the son of Edward the Confessor’s earl
Siward of Northumbria) and Earl Ralph of East Anglia, son of a Breton of
the same name who had also served the Confessor, together with the
Norman earl of Hereford, Roger. This was easily suppressed in the king’s
absence, with Ralf escaping abroad, Roger suffering imprisonment and
Waltheof being beheaded. In 1085 King Cnut of Denmark, son of
Swegen Estrithson, intended leading an expedition against England which
in fact never sailed, although William made extensive preparations to
deal with it. Few kings of this period were free of trouble for long and the
Conqueror was no exception, but it must have been evident to most
people that if the consequences of the events of 1066-1070 were to be
overturned it was not after the latter date likely to be solely by
Englishmen.
With the benefit of hindsight Hastings probably tends to look more
decisive today than it did to contemporaries. William did not destroy the
territory north of the Humber because he felt that English opposition was
of no account, and the willingness of his opponents to keep fighting
shows that they and their resources had not been broken by the three
battles of 1066. However, they had no leader worthy of them, and their
call for Danish assistance in 1069 suggests an awareness that the odds
were not in their favour. That the Danes were not willing to face the king
on a major battlefield when they arrived probably owed something to his
reputation for indefatigability, won over a long period, and as evident in
his ruthless suppression of the rebels as it had been in his youthful

1
OV, HÆ, ii. 230-3; Darby, Domesday England, pp. 248-52. More than half of those in the North
Riding and over a third in both the West and East Ridings were either wholly or partly waste. The
quotation is from Palmer, ‘War and Domesday Waste’ p. 274, who rejects suggestions that the harrying
was a relatively minor affair.

197
 THE BATTLE OF HASTINGS

struggle to assert his control over Normandy and on the field of Hastings.
The outcome of that battle itself must have discouraged some, and while
it is true that it had not ended all opposition it is also true that it was an
opposition in which large parts of the country were never involved. It did,
therefore, play a major role in winning England for William, and to this
extent its reputation as one of the most decisive events in the nation’s
history is not ill-deserved.

As such it will retain its fascination in the future, as it always has in the
past, assisted by the complexities which inevitably arise when dealing
with sources about which there will never be complete agreement, and by
the never-ending attractions of the Bayeux Tapestry. The study of
Hastings must ultimately be a study of the characteristics and the
strengths and weaknesses of those sources, and of what interpretations of
the battle they seem to render most plausible. Moreover, this source
criticism must necessarily be an elaborate process, and is not one ever
likely to yield definite conclusions. Nevertheless, it has been one of the
primary purposes of this book to question the validity of the scholarly
consensus upon the scale of the battle which was established in the years
around 1900 and has acted as a straitjacket upon interpretations of it ever
since. If the idea that it was contested by about 7,000 men on each side
simply around the crest of the Battle Abbey ridge cannot be proved to be
wrong, nor can it be proved to be right, and indeed no very convincing
evidence was ever produced in its favour. Yet, if in this sense less is
known about the battle than previously seemed the case, what has been
gained in compensation is that the possibilities about its nature have been
extended, and that those possibilities include a large-scale conflict fought
over a large area by well-trained and resourceful troops on both sides, the
English army being the product of a powerfully-organized state with a
long history of administrative competence behind it, and that of the
French being an international force under a great warrior.
The tendency of writers like Spatz and Baring to minimise not only the
scale of the battle but also the military competence of its captains and
soldiers found its most startling expression in Spatz’s refusal to allow the
English use of a formation as complex as the shield-wall, and was
ultimately grounded in a reluctance to grant to medieval political and
military systems any great degree of competence in these areas, or others.
This desire to simplify matters by denying the complexity and power of
early societies persists to this day, despite compelling evidence that it can
be mistaken. If such worlds did not have access to modern technology,
they had need of powerful systems of human organisation precisely on
that account. Just how remarkably effective such systems could be can be
widely illustrated, from the pyramids to the buildings of the Inca empire,

198
 7 EPILOGUE

and from Stonehenge to Offa’s Dyke. Tangible physical remains such as

these are inevitably, and to an extent rightly, more persuasive than mere
references to administrative organisation in written sources, but they are
also an indication that such references should not simply be swept aside
when there is no physical evidence to support them. Moreover, in the late
Anglo-Saxon period the witness of the coinage is a powerful indicator of
the ambitions and effectiveness of the state, and the physical remains of
Alfred’s burhs together with the document known as the Burghal Hidage1
proof that the same governmental competence was also to be found in the
organization of its military systems. These systems may or may not have
demanded more of the English people in martial terms than at any
subsequent date prior to 1914, but at least they committed three armies to
battle in a period of twenty-five days in the autumn of 1066, in the
process annihilating a Norwegian army under a famous commander and
holding the French for almost the whole of a long day nearly three weeks
later. While it may be true that the ‘organisation and the economy of a
state are largely reflected in and determined by its organisation for war’,2
it is very questionable how far defeat on the battlefield can be taken as
evidence of administrative incompetence. War brings many chances, not
least those of personality and circumstance, and the defeat of 14 October,
1066 is scarcely a more plausible witness to the ineffectiveness of the late
Anglo-Saxon military system than the conquest of South America by
small numbers of Spanish is of administrative failings in the Inca empire,
which despite lacking writing and the wheel is known to have been one of
the most impressive of ancient civilisations in terms of its ability to
organise public works. Hastings was decided by a narrow margin, in a
world hardly less rich and complex than our own.
Yet if the battle has to be left there for the time being, and the
complexities of the written sources are unlikely ever to be resolved, that
is not to say that there is no possibility of making further progress.
Fortunately, much of the battlefield was purchased for the nation when
finally sold by the Webster family in 1976, and is there to be investigated.
William of Jumièges says that William buried his men after despoiling
the English corpses, the Carmen that he took up the bodies of his own
dead and buried them in the earth while leaving his enemies as carrion,
William of Poitiers simply that he buried his own men and allowed those
of the English who wished to inter their fellow countrymen. 3 It is not
unlikely that some bodies were taken to local churches to be placed in
consecrated ground, and that this was at least a not unknown practice
with men of high social standing is suggested by William of
1
Above, pp. 107-8.
2
Campbell, ‘Some Agents and Agencies’, p. 201.
3
WJ, GND, ii. 170-1; Carmen, ll. 567-72; WP, GG, pp. 142-3.

199
 THE BATTLE OF HASTINGS

Malmesbury’s statement that King Harold’s brother Tostig was buried in

York after the battle of Stamford Bridge; it is also possible that skeletons
bearing marks of weapons injuries excavated at the church of St Andrew,
Fishergate in York may be some of the fallen from the battle of Fulford.1
Nevertheless, while none of the sources specifically say so, it seems very
probable that most of the French rank and file were laid to rest no great
distance from where they fell. Thus, somewhere upon or near the
battlefield there may be grave pits containing their remains, and if those
pits are ever located they could have evidence to offer on many things,
and not least on the question of the extent of the fighting.
It has been suggested above that the English may have deployed and
fought around the stream which flows west from New Pond, and it is
worth noting that for about 250 metres west of the Pond, mostly on the
south bank but then for a brief stretch on the northern one, lies a series of
rather prominent mounds, the easternmost (that is, east of the little bridge
which crosses the stream) being pierced by mature oak trees. It may well
be the case that these features, which need not all be the result of activity
of the same period, are of relatively recent origin, some of it connected
with management of the stream for the benefit of the Battle gunpowder
mills,2 but there is no way of being certain of this until we know what, if
anything, lies beneath them. Moreover, even if they do not contain
burials, it is worth noting Wace’s comment on the numbers of William’s
men who fell around a ditch, and the possibility that this ditch, the
watercourse shown in the Bayeux Tapestry, the ‘sandy stream’ after
which the battle may have been named and the sandy stream visible today
are all one and the same; if so, investigative surveys in this area might be
well worth the trouble, and even a negative result would at least tell us
more than is known at present. It is perhaps worth noting that there are
also what look like artificial hummocks on the extreme south-western
side of the Battle Abbey ridge, along the eastern fringe of Saxon Wood.
Whatever phenomena they represent, it seems unlikely to be the dredging
of the Saxon Wood stream, which is some distance away. However,
these features are probably nothing more than the root mantles of fallen
trees.
Wace also comments on the digging of ditches by the English as
defensive works. This may, of course, be completely erroneous, but it is
not inherently improbable, and such activity, like the interment of the
dead, could well have left traces which are there to be found. The recent
English Heritage survey of the battlefield has drawn attention to a linear
feature which it maps as Feature J and describes as ‘a prominent east-
west scarp’. It is clearly visible on the slope below the abbey ruins, but
1
Above, p. 37; Daniell, ‘Battle and Trial’.
2
Above, p. 49. For photographs of the mounds, below, Illus. 29 and 31.

200
 7 EPILOGUE

much less so where it seems to reappear not far from the western shoulder
of the ridge, just to the north of Horselodge Plantation.1 More than a
hundred years ago debate raged over whether Freeman had been right to
take Wace to mean that the English had constructed a palisade to their
front, and the result of that debate was that it largely disappeared from
historians’ accounts of the battle. Like other aspects of Freeman’s
treatment, it may one day have to be reinstated, for it is not impossible
that Feature J is nothing less than the remnants of a fieldwork used during
the battle and extending over a considerable frontage. Moreover,
immediately to the north of its easternmost surviving section are two
depressions in the ground which have been interpreted by the surveyors
as the remains of quarries:
The eastern depression, which measures up to 1.5m in depth, is
substantially larger than the other, with access routes to the south and
north-east. The western depression is cut along its western side by one of
the close boundary ditches. The form of these depressions, and the access
points, suggests that they were probably former quarries, perhaps for clay,
dating before the 18th century, that have later been ‘landscaped’ within the
2
parkland setting.
Quarries they may very well have been, but their location immediately
south of the abbey buildings, almost certainly the scene of some of the
heaviest fighting, must also raise the possibility that they in fact represent
the burials of at least a portion of the French dead. Of course, it may seem
ridiculous to suggest that the remains of William’s men are marked at one
point by mounds above ground level and at another by depressions below
it. Nevertheless, future archaeological investigations of the battlefield,
which should include an attempt to locate French graves over a wide area,
could hardly ignore either. Upon the fight of Senlac, whose renown has
always been so great, the last word should not be yet.

1
Survey, p. 9 and fig. 5. Illus. 7 below for the English Heritage plan and Illus. 41-5 for photographs of
Feature J.
2
Ibid., p. 10, and fig. 5. Illus. 42, below.

201
 APPENDIX
THE DEATH OF HAROLD AND THE
AUTHORITY OF THE PRESENT FORM
OF THE BAYEUX TAPESTRY

The earliest surviving drawing of the Bayeux Tapestry, which is coloured

but covers only the first thirty feet, was found among the collections of
N-J Foucault, governor of Normandy between 1689 and 1704, following
his death in 1721. It is now in the Bibliothèque Nationale in Paris and
may be the work of his daughter Anne, who was born in 1687 and is
known to have been possessed of some skill in drawing.1 It was brought
to the attention of the French historian Antoine Lancelot, who presented a
paper upon it, and the visit of Earl Harold to Normandy which obviously
formed its subject, to the French Académie Royale des Inscriptions et
Belles-Lettres on 21 July, 1724; this was published in the sixth volume of
their Memoires in 1729, along with an engraving of the drawing. Lancelot
introduced it to his fellow academicians because he believed it to be
contemporary with the Conqueror himself, but acknowledged that he had
been unable to discover the nature and whereabouts of the original from
which it had been taken, the requests for information he had sent to Caen
not having been favoured with a reply; however, he suggested that it
perhaps represented part of the tomb of William the Conqueror, which
had been located in the church of St Stephen, Caen and was destroyed by
the Huguenots in 1562.2 At this point the famous French classical scholar
Bernard de Montfaucon intervened. Being then engaged in assembling for
publication a collection of artistic and archaeological sources for the
history of medieval France, he wrote to Dom Romain de La Londe, prior
of St Stephen’s, in the hope of discovering the original upon which the
Foucault drawing was based. La Londe knew of nothing like it within the
abbey, but was advised by the oldest of his monks to make enquiries in
1
For what follows, see Huard, ‘Quelques Lettres’; Janssen, ‘La redécouverte de la Tapisserie de
Bayeux’; Brown, The Bayeux Tapestry, pp. 4-22; Hill, ‘The Bayeux Tapestry’, pp. 387-391; Hicks, The
Bayeux Tapestry, pp. 70-77.
2
Lancelot, ‘Explication d’un monument’, 739-40.

203
 THE BATTLE OF HASTINGS

Bayeux, and accordingly passed Montfaucon’s letter to Dom Mathurin

L’Archer, prior of the monastery of St Vigor. On 22 September, 1728
L’Archer replied that the item in question was a tapestry which belonged
to the cathedral of Bayeux and which tradition reported to be the work of
the Conqueror’s wife Matilda; he had himself seen it displayed there, but
not being aware that it concerned the deeds of Duke William had never
examined it at close quarters. Eventually, on 11 February, 1729, after a
delay which he said was not his responsibility, La Londe sent
Montfaucon readings of the Tapestry’s inscriptions made by L’Archer.
Montfaucon published engravings based on the Foucault drawing in the
first volume of his Les Monumens de la Monarchie Françoise the same
year, but when he knew of the Tapestry’s location sent the draughtsman
Antoine Benoît to Bayeux to produce an accurate copy of the rest of the
hanging.1 Benoît’s drawings, which were engraved for publication in
Montfaucon’s second volume, survive today in a fragile state in the
Cabinet des Estampes of the Bibliothèque Nationale.2
At much the same time as Montfaucon’s second volume appeared
Lancelot returned to the scene. On 9 May, 1730 he presented to the
Académie a further paper, accompanied by engravings which were also
based on Benoît’s drawings, but which differ from those done for
Montfaucon in significant respects. They were published in England in
Ducarel’s Anglo-Norman Antiquities in 1767, but have attracted little
attention for many years. This is surprising, because they have some
notable evidence to offer on the scene dealing with the death of Harold,
and specifically on the figure which many have thought to represent the
king immediately after he has been struck by an arrow. Montfaucon’s
engraving shows this individual holding a missile represented by a
continuous line which extends to the lower rim of his helmet. However,
that missile hardly looks like an arrow, for it does not have the flights
which feature so prominently on what are clearly arrows elsewhere
(including the three sticking in the same soldier’s shield) and it is also
distinctly longer. Lancelot’s engraving shows something both similar and
at the same time dramatically different. Here the missile is again without
flights and is depicted as a series of dashes (suggesting that the wool had
to an extent disappeared), extends well to the left of the Montfaucon line
and ends to the left of the figure’s helmet without touching it. Now
neither of these engravings is a completely accurate copy of the drawing
they were supposed to represent, for examination of that drawing
indicates that Montfaucon’s engraver was more correct in showing the
missile extending to the rim of the helmet, while Lancelot’s was more
1
Benoît’s work was complete by mid-July, 1729, by which time he was in Bec and short of the funds
necessary to return to Paris; Huard, ‘Quelques Lettres’, 370-2.
2
For their present condition, see Hill, ‘The Bayeux Tapestry’, p. 390.

204
 APPENDIX

accurate in having it extend well to the left of the hand in which it is

grasped, although he actually pushed it rather further than Benoît’s
drawing warrants. The crucial point, however, is that the Benoît drawing
tends to confirm the impression given by the Lancelot version of it: that
this is not an Englishman holding an arrow which has struck him in the
head so much as an Englishman who is about to use a spear against the
enemy. Thus, this figure to the right of the standard-bearer is a very close
companion of the very similar one to the left, both holding spears in their
raised right arms, and probably intended by the Tapestry’s designer to
represent, together with the standard-bearer himself, Harold’s bodyguard.
Of course, it may be objected that on the contrary the identity of this
individual is powerfully suggested by the way in which he stands directly
beneath the word Harold with his head separating the letters o and l, and
that this argument is strengthened by other scenes where names have
apparently deliberately been placed in close proximity to the figures to
which they refer: Bishop Odo, for example, rides in battle beneath the
words Odo and ep(iscopu)s, which are separated by his raised staff, while
to his right the duke raises his helmet to prove that he is still alive in very
close proximity to the word Wilel(mus); in the scene which shows
William and his half-brothers seated together after their landing in
England the three are all positioned beneath their names, and the duke’s
head separates the first l of Willelm from the second, while shortly
afterwards he receives news of the imminent arrival of the English army
with his head touching the letter e of the same word. These examples
could be multiplied, but there are also others which clearly demonstrate
that the designer of the Tapestry followed no single rule in such matters.
In the section which represents Count Guy of Ponthieu handing Harold
over to William in 1064 the word Haroldum is above its subject, but
Wilgelmum is above Guy’s head, which separates the m and the u, while
Guy’s own name, Wido, appears well to the left, above the second rank of
the cavalry escort. In a lengthy sequence involving five ranks of horses
(more, if one adds William’s escort) and with a lengthy inscription, it was
neither possible nor necessary to place names in close proximity to the
figures they represented, whose identity was clear enough. There is a very
similar example as the duke and his men set out for Brittany beneath the
words Hic Willem dux et exercitus eius venerunt ad monte(m) Michaelis
(‘Here Duke William and his army came to Mont St Michel’). Here the
fact that the inscription begins well to the left of the figure who is leading
the column means that William is positioned beneath the word monte(m)
while the e and m of Willem are separated by a spear carried by the
rearmost trooper. There is, however, no possibility of mistaking that
trooper for William or the figure of the duke for a common soldier. In this
context there seems no reason to come to any very different conclusion

205
 THE BATTLE OF HASTINGS

on the significance of the wording of the Harold death scene. The Hic of
Hic Harold rex interfectus est is well to the left of the trio which includes
the standard-bearer and if Harold is above the right-hand man of that trio
this would have caused no confusion if originally it was obvious that this
individual was a soldier using a spear rather than the king holding an
arrow, for it would have been clear enough that the word Harold could
only refer to the man struck by the French horseman.
Their separate identities also, of course, explain the distinctly different
equipment of the two figures. The much-favoured theory that both
represent the king, first being hit by an arrow and then being cut down, is
for all these reasons unlikely to be correct. In so far as we can judge from
the Benoît drawing, the Tapestry as it existed in 1729 did not show the
first figure holding an arrow at all. It is therefore hardly surprising that
neither Montfaucon nor Lancelot, who must both have seen Benoît’s
work, took the soldier to the right of the standard-bearer to be anything
other than one of Harold’s soldiers, despite the fact that Lancelot knew
and mentioned Wace’s evidence that the English king had earlier been
wounded in the eye by an arrow and elsewhere cited William of
Malmesbury, who speaks of an arrow in the brain causing his death,
while Montfaucon also knew evidence on the arrow and no more thought
that he was looking at a figure holding such an item than did Lancelot.1
If the present form of the Tapestry shows something different this may
be largely the responsibility of Charles Stothard, who went to Bayeux to
draw it for the Society of Antiquaries shortly after the end of the
Napoleonic Wars.2 Stothard may or may not have seen the stitch holes
visible in Benoît’s time, but he mentions the Montfaucon engraving of
Benoit’s work in a way that indicates his familiarity with it,3 and perhaps
decided that the missile being held by the individual concerned would
benefit from having flights attached to its left end. Alternatively, it is not
impossible that the flights had been added during a restoration prior to his
time but subsequent to that of Benoît. At any rate, the engraving produced
from Stothard’s drawing is fairly close to that of Montfaucon, but the
figure’s head is inclined back a little more to the right, thus decreasing the
angle between the shaft of the missile and nasal of the helmet and
increasing the impression that the head has been tilted in order to
facilitate the extraction of the ‘arrow’. It would not be surprising if
Stothard or earlier restorers were misled on this by Montfaucon’s
engraving, which had itself decreased the angle between the missile and
1
Lancelot (ibid. 668) suggested that the trio which includes the standard-bearer were members of
Harold’s bodyguard; Montfaucon, Monumens, ii. 29, thought the figure being struck by the French
horseman represented a Harold who had fallen from his own horse, and that the Tapestry here confirms
WM’s story that he was struck on the thigh as he lay on the ground.
2
On Stothard, see ODNB and Hicks, The Bayeux Tapestry, pp. 120-34.
3
Stothard, ‘Observations on the Bayeux Tapestry’, 185.

206
 APPENDIX

the nasal shown in Benoît’s drawing, and by their familiarity with sources
which speak of the king being hit by an arrow.1 At any rate, by Stothard’s
time it was perfectly possible to interpret this figure as King Harold being
struck by an arrow, and this Stothard’s wife duly did in her Letters
written during a tour through Normandy, commenting that the Tapestry
agrees closely with the documentary evidence on the battle, and that
‘Harold is represented receiving the arrow in his eye, he falls to the
ground; a soldier pierces him in the thigh, with a sword’.2 If the work thus
far had nevertheless stopped short of representing the ‘arrow’ as
embedded in the figure’s face, this was to be remedied by the restorers
who followed; accordingly the visitor to Bayeux today sees the ‘arrow’
terminating in the vicinity of the right eye. Of course, it is open to anyone
to argue that Benoît misrepresented the length of the shaft held by this
figure and also missed stitch holes which indicated that it had borne
flights, although these were visible to Stothard almost ninety years later.
This would restore the confidence of modern visitors to the Tapestry in
the basic authenticity of the present stitchwork, but hardly seems to be the
most plausible interpretation of the evidence.
At the time it first became known beyond the confines of Normandy
parts of the Tapestry were already seriously damaged, showing that it had
been subjected to hard usage at a period in its history which is now
completely obscure. There was a great tear near the beginning, 3 what is
today the end was in a very poor state and some of the inscriptions
elsewhere were difficult to read, a clear indication that other wool too had
been degraded or had disappeared altogether. Montfaucon published the
engravings in his Monumens along with his own commentary and a list of
the Tapestry’s inscriptions, this being partly derived from the list
produced by Mathurin L’Archer and sent to him by La Londe.4 However,
these readings do not always accord with what Benoît drew. L’Archer,
for example, had not been able to make out much of the inscription which
accompanied the depiction of Halley’s Comet, managing only the word
isti and the comment ‘il y a environ une ligne effacée’, while Montfaucon
offered ISTI MIRANTVR STELLAM (‘These marvel at the star’) and the
engraving in the Monumens taken from Benoît’s drawing showed ISTI
MIRANT S----A, but with traces of the missing letters which might easily
be reconstructed (as they have been) as TELL.5 The final inscription

1
I deduce Stothard’s familiarity with these sources from that of his wife; see the following note.
2
Mrs Stothard, Letters, p. 131. As far as I know this is the earliest expression of the idea that both
figures represent the king.
3
The patch which eventually replaced the material lost beneath the first occurrence of the word Harold
is obvious in modern reproductions.
4
There is a facsimile of the letter in Hill and McSween, The Bayeux Tapestry; Huard, ‘Quelques
Lettres’, 369, published the address but not the list of inscriptions.
5
I have not seen the Benoît drawing of this part of the Tapestry.

207
 THE BATTLE OF HASTINGS

recorded by L’Archer was Hic Harold Rex interfectus est and

Montfaucon followed this, although the plate which he published from
Benoît’s drawing goes only as far as the R of INTER(FECTUS), which is
followed not by the F one might expect but by a P partly in dotted lines;
in the register below, and also in rather faint dotted lines, can be seen IVS
(the I, however, being possibly the remains of a T or some other letter).
Finally, there is a colon and further marks which are difficult to make
much of at all. The engraving was here a reasonable representation of
Benoît’s drawing except in dealing with these marks, to whose
significance we shall return. While preparing the paper which he
presented to the Académie in 1730, Lancelot wrote to the bishop of
Bayeux asking for a further close examination of the inscriptions, and the
prelate entrusted the task to a competent man who carried it out,
according to Lancelot, ‘avec une attention aussi exacte qu’on pouvait le
souhaiter’.1 This may be so, for while L’Archer had written that ‘Il y a
encore deux ou trois lignes auplus entierement effacées au bout de la
tapisserie’ (implying, of course, that there was something to be seen
there) the bishop’s man not only confirmed the reading Hic Harold rex
interfectus est but added et fuga verterunt Angli (‘And the English turned
in flight’).2
Although both the prior of St Vigor and the man employed by the
bishop of Bayeux agreed on the reading Hic Harold rex interfectus est,
we should perhaps not be too hasty in following them. The wording was
obviously damaged, or Benoît would have concurred in showing it, and
the part of it which he does show could be expanded rather differently.
The P which he read after the R of INTER… is not easily turned into
Latin, but this would not be the case if it were in fact the truncated
remains of a second R. Indeed, if this were so one might guess that the
reading was not interfectus but in terra, and that the whole inscription
read something like Hic Harold rex in terra iactus est (‘Here King Harold
has been struck to the ground’), or, if iactus was a participle and the
sentence continued further, perhaps Hic Harold rex in terra iactus et fuga
verterunt Angli (‘Here King Harold struck to the ground and the English
turned in flight’). In either case, of course, it would have precisely fitted
the figure falling beneath the horseman’s blow.3 As presently restored the
F and E of interfectus are very close together, and it is conceivable that if
much of the E except the upper part of the ascender had disappeared
Benoît could have produced P by reading them together; also, the

1
Lancelot, ‘Suite de l’explication’, 602. Huard, ‘Quelques Lettres’, 348, names him as Louis Rustaing
de Saint-Jory, an écuyer of the bishop; he was admitted as a member of the Académie de Caen in 1731.
2
Lancelot, ibid., 668, noted the poor Latinity of this phrase (the verb verto should be in the passive)
and the same usage in the earlier Conan fuga vertit.
3
I am grateful to John Davie for discussion of possible Latin wording at this point.

208
 APPENDIX

difference between interfectus and in terra iactus is not great, but it

would require space for an extra i and a, and the fabric in this area does
not today exhibit the slightest trace of the stitch holes which might
suggest that these letters ever stood there. However, this may well be
much less significant than one might care to think. There is in fact plenty
of space in which these letters could once have stood, if one were to
assume that the restored E and C are not in their original positions, and
that the fabric of the Tapestry may once have exhibited stitch holes which
have since disappeared.
There is good reason to believe that the latter is almost certainly the
case. The missile held by the soldier to the right of the standard-bearer is
clearly shown by Benoît as extending to the rim of the helmet and it
seems likely that originally it also passed behind it and went on into the
lettering beyond; Stothard (if he did not simply copy the Montfaucon
engraving of Benoît) concurred on the line between the figure’s hand and
the helmet rim, yet there are no stitch holes to be seen on this line today,
at least on the front of the Tapestry. There is also the question of the
marks which Benoît drew after the colon following the word usually read
as interfectus. It is very difficult to believe that they are the remains of
work of a single period, for the lettering (assuming that it is lettering)
looks as though it has been revisited more than once, and thus that what
Benoît saw was a feature which had already been restored and then
suffered further serious damage before 1729. Naturally, it can be argued
that his drawing need not always have been accurate, but this would not
be an easy position to sustain in this particular case, for the tendency
would surely be to simplify a feature like this to make it intelligible rather
than to record it faithfully in all its confusion, yet the latter is what Benoît
seems to have done. It would also be necessary to completely reject the
features he drew if one were seeking to avoid the conclusion that the
Tapestry once bore stitch holes which are no longer to be seen, for around
the reconstructed word EST there is today no trace of the marks which he
saw there nearly three hundred years ago. One might wonder how many
other tell-tale signs the fabric once bore which have since been wiped
away forever. The effects of washing seem a likely explanation for the
disappearance of stitch holes, and it has recently been acknowledged that
the Tapestry was washed with soap before it was restored in the
nineteenth century.1
We have just seen evidence that the Tapestry may have undergone
changes before 1729 as a result of the lack of care with which it had

1
Information supplied to John McSween by Mlle Sylvette Lemagnen, curator of the Bayeux Tapestry.
If reproductions of the Benoît drawings are ever published it may be that detailed comparison of them
and the present fabric will eventually throw much light on the history of the hanging and the features it
bears, and that this will modify some of the thoughts offered here.

209
 THE BATTLE OF HASTINGS

evidently been treated at a time in its history which is now unknown, and
that it is not impossible that before that date parts of it had been damaged
and restored more than once. However, this was not to be the end of its
vicissitudes. During the French Revolution it narrowly escaped
destruction twice, and was exhibited in Paris when Napoleon was
preparing to invade England and then returned to Bayeux. In 1814 it was
seen by the Englishman Hudson Gurney:
It was coiled round a machine, like that which lets down the buckets to a
1
Well; and I had the opportunity of drawing it out at leisure, over a table.
Shortly afterwards it received a visit from Dawson Turner, who
confirmed Gurney’s evidence on the way in which it was exhibited:
The tapestry is coiled round a cylinder, which is turned by a winch and
wheel; and it is rolled and unrolled with so little attention, that if it
continues under such management as the present, it will be wholly ruined
in the course of half a century. It is injured at the beginning: towards the
end it becomes very ragged, and several of the figures have completely
disappeared. The worsted is unravelling too in many of the intermediate
portions. As yet, however, it is still in good preservation, considering its
2
great age, though, as I have just observed, it will not long continue so.
It was also in these years that the Society of Antiquaries sent Charles
Stothard to Bayeux to produce accurate drawings from which hand-
coloured engravings were eventually produced. He reported that the work
in some parts of the Tapestry was destroyed, especially towards the end,
which was a mass of rags, and that comparison with Montfaucon’s
engraving suggested that there had here been ‘much injury even since his
time’. However, he also said that he had been able to recreate much of
what was missing by identifying minute particles of coloured wool in the
stitch holes; where there were no such particles he simply recorded the
holes:
The restorations that I have made commence on the lower border with the
first of the archers. Of these figures I found scarcely one whose colours
of any kind remained perfect. In the upper border and historical part, the
restorations begin a little after, with the Saxons, under the word
“ceciderunt.” From the circumstance of the border being worked down
the side at the commencement of the Tapestry, it is evident that no part of
the subject is wanting; but the work in many places is defaced, and these
3
parts have been restored in the same manner as at the end.
As this gives no very precise guide to the amount he had re-created, we
shall never know just how much was readily visible to those who saw the
hanging at this date, just as the same is true of its state in Benoît’s time.
However, Thomas Dibdin, who described it soon after Stothard,
commented that:
1
Gurney, ‘Observations on the Bayeux Tapestry’, 359.
2
Turner, Account of a Tour in Normandy, ii. 242.
3
Stothard, ‘Some Observations on the Bayeux Tapestry’, 184-5.

210
 APPENDIX

The first portion of the needle-work, representing the embassy of Harold,

from Edward the Confessor to William Duke of Normandy, is
comparatively much defaced – that is to say, the stitches are worn away,
and little more than the ground, of fine close linen cloth, remains…
About the last five feet of this extraordinary roll are in a yet more decayed
1
and imperfect state than the first portion.
At some point after Dibdin’s time2 it was restored into more or less its
present condition. The Bayeux Tapestry is a historical monument of
immense interest and value, and it seems inconceivable that in its
essentials it differs significantly today from the piece which passed out of
the workshops in the late eleventh century; even so, attempts to interpret
its scenes and inscriptions need always to be preceded by doubts about
how far they are now as they once were.

Drawing of the Harold death scene by Antoine Benoît, 1729. From a photograph kindly loaned by Dr
David Hill and John McSween; reproduced by permission of the Bibliothèque Nationale, Paris.

1
Dibdin, A Bibliographical Antiquarian and Picturesque Tour, i. 378-9.
2
The Tapestry was put on display in a new gallery in 1842 after some conservation work, but a grant of
5000 francs in 1853 paid for further restoration, Hicks, The Bayeux Tapestry, p. 138. The synthetic
dyes used, in seventeen colours, apparently suggest that the work was carried out after 1860, Oger, ‘La
Tapisserie de Bayeux’, p. 121.

211
 THE BATTLE OF HASTINGS

The engraving of Benoît’s drawing of the Harold death scene published by Bernard de Montfaucon in
1730.

The engraving of Benoît’s drawing of the Harold death scene published by Antoine Lancelot in 1733.

212
 APPENDIX

A coloured engraving based on Charles Stothard’s drawing of the Harold death scene published by The Society
of Antiquaries as Plate 16 in their Tapestry series on 1st July, 1823.

213
 GENEALOGIES

Genealogical Table 1.

Kings of England: Æthelred II to Harold II. Kings are in bold; dates are regnal years.
Between 1035 and 1037 the throne was disputed between Harthacnut and Harold
Harefoot

215
THE BATTLE OF HASTINGS

216
 GENEALOGIES

Genealogical Table 4

The House of Godwin. Kings of England are in bold. Tostig was killed at the battle of
Stamford Bridge on 25 September, 1066; his brothers Harold, Gyrth and Leofwine at
Hastings on 14 October.

217
 ILLUSTRATIONS

Illus. 1. The western part of the battlefield on the Battle Tithe Map of 1859, ESRO TD/E 158. Reproduced
by kind permission of the East Sussex Record Office.
Next page: Illus. 2. Richard Budgen’s map of the area south of the western part of the Battle Abbey ridge,
1724. The Stew Ponds are visible at top right while the later New Pond covers some of the same area as
Weanyer’s Pond. The track which leaves the Catsfield Road and runs along the south-western side of
Eight-acre Meadow still exists and can be seen on the 1931 OS Map, above, p.46. Illus. 3. Budgen’s map
of the eastern part of the town of Battle, with the road now called Lower Lake marked as ‘Sanglake’.
ESRO BAT 4421 (7) and (6). Reproduced by kind permission of the East Sussex Record Office

219
THE BATTLE OF HASTINGS

220
 ILLUSTRATIONS

Illus 4. ESRO BAT 4435. Plans of the Several Farms Constituting the Estates of Sir Godfrey
Webster Bart. in Sussex according to Mr Willock’s Survey, 1811. No. 1 The Abbey Park.
Reproduced by kind permission of the East Sussex Record Office.
Next Page. Illus. 5. The battlefield on the OS 6 inch map first edition of 1878 (surveyed 1873-
4) Sussex Sheet LVII. Not to the original scale.

221
THE BATTLE OF HASTINGS

222
 ILLUSTRATIONS

Illus. 6. The geology of the area according to the British Geological Survey 1:50,000 Series,
Sheet 320/321, Hastings and Dungeness. Dark green: Wadhurst Clay; Yellow: Tunbridge Wells
Sand; Lemon: Alluvium; Blue-grey: Ashdown Beds; Grey: Head. See above, pp. 50-1.

Next page, Illus. 7. The plan which accompanied the important English Heritage ground survey
published in 2002. Reproduced by kind permission of English Heritage.

223
THE BATTLE OF HASTINGS

224
 ILLUSTRATIONS

Photographs of the battlefield

The extensive series of photographs which follows is arranged geographically;

the first sequence (Illus. Nos. 9-27) covers the western part of the Battle Abbey
ridge, land formerly part of the Battle Abbey Estate but not belonging to English
Heritage today; the second sequence (Nos. 28-51) is of the ground which is under
their guardianship; the third (Nos. 52-73) illustrates the town centre, the eastern
part of the ridge and the roads Upper Lake, Lower Lake and Marley Lane. Next
page, Illus. 8. The battlefield on the Millenium Map, reproduced by permission
of Getmapping plc.

225
THE BATTLE OF HASTINGS

226
 ILLUSTRATIONS

Illus. 9. The area dealt with in this section is reached by proceeding towards and beyond
the camera position, either by walking down the small road to the right of the abbey
gatehouse or by parking in the English Heritage car park (entrance on the right in this
photograph) and turning left upon leaving it. Passing through a small gate will produce
the view shown in Illus. 10.

227
 THE BATTLE OF HASTINGS

Illus 10. Long Plantation on the left. A small watercourse issues from it (bottom left), passes beneath
the track through a pipe and then flows south-west beyond the tree in the foreground. This area is
relatively flat and the gradient falling to the right only steepens at the furthermost point of this
photograph, centre left, see Illus.16.

Illus. 11. The watercourse just beyond the tree (here on the extreme right) in Illus. 9. These
photographs were taken in March 2015 when this area was wetter than I have ever seen it. This
perhaps had something to do with recent dredging of the channel in Long Plantation, see. Illus. 14.

228
 ILLUSTRATIONS

Illus. 12. The reverse view to Illus. 10, looking north-east, and again showing the relative flatness of
the ground.

Illus. 13. The watercourse (bottom of picture) inside Long Plantation, with a slope rising beyond it.
This is apparently the one described by Baring as 'the bank at the back ...already 20 feet high', see
above, p. 119.

229
 THE BATTLE OF HASTINGS

Illus. 14. The watercourse just within Long Plantation as it looked in March, 2015. Comparison
with the next picture probably indicates that the width of the channel is the result of recent
clearance work by English Heritage.

230
 ILLUSTRATIONS

Illus. 15. The same channel on 24th November, 2002, when there was less vegetation and
less water. None of these photographs suggest that there can have been any gradient or
watercourse in this vicinity in 1066 which would have been an obstacle to attacking troops.
See above, pp. 117-21.

231
 THE BATTLE OF HASTINGS

Illus. 16. The point where the slope steepens significantly, just visible from the camera position in Illus.
10. The top of the ridge at this point is within Long Plantation to the left and the gradient is therefore
interrupted by the modern track.

Illus 17. The reverse view, looking north-east from the top of the track as seen in Illus. 16. Some of the
buildings of the town of Battle can be seen on the horizon.

232
 ILLUSTRATIONS

Illus. 18. Further south still, a depression in the ground within Long Plantation is clearly the result of
human activity and a reminder that much more has happened on the site than simply warfare.

Illus. 19. At about the same point, looking down from the edge of Long Plantation into Saxon Wood.

233
 THE BATTLE OF HASTINGS

Illus. 20. The reverse view from the edge of Saxon Wood looking up to Long Plantation.

Illus. 21. Taken from the same position as Illus. 20 but looking across the shoulder of the western part
of the Battle Abbey ridge and giving a reasonably accurate impression of the extent of the gradient.

234
 ILLUSTRATIONS

Illus. 22. A different view of the vegetation visible in Illus. 21. The marsh-like tufts indicate the
presence of a rather boggy channel which drains into Saxon Wood from the top of the ridge.

Illus. 23. From the top of the ridge looking in a northerly direction with Saxon Wood on the left and
Long Plantation on the right.

235
 THE BATTLE OF HASTINGS

Illus.24. The extreme western end of the Battle Abbey ridge. The most southerly part of Long Plantation
is at the top.

Illus. 25. The reverse view looking south to Powdermill. Both photographs taken 4th January, 2002.

236
 ILLUSTRATIONS

Illus. 26. The stream which rises in the vicinity of the Stew Ponds to the east about to exit the western
boundary of the land belonging to English Heritage.

Illus. 27. And having flowed under the modern track enters Warren Wood on its journey to Powdermill.

237
 THE BATTLE OF HASTINGS

Illus. 28. Near the western end of the ridge, but now looking south within English Heritage land just
east of Long Plantation (to the right). The ground can be seen dropping away from the camera position.

Illus. 29. The stream in the valley bottom at a point where it would have been visible in Illus. 28 but for
the trees. There are mounds of raised earth on the south bank.

238
 ILLUSTRATIONS

Illus. 30. The stream close to the camera position of Illus. 29. The very sandy nature of the ground is
obvious here.

Illus. 31. Long mound on the north bank of the stream near the western English Heritage boundary and
the north-eastern tip of Warren Wood.

239
 THE BATTLE OF HASTINGS

Illus. 32. One of the ponds behind Horselodge Plantation on a wet day in February, 2003. The English
Heritage visitor trail can be seen at the bottom of the photograph.

Illus. 33. One of the more easterly stretches of water in the same area, banked and formerly having a
sluice gate to allow water to be held at will, see p. 49. Those who venture into this area should beware of
rabbit holes.

240
 ILLUSTRATIONS

Illus. 34. A small stream runs from the eastern edge of Horselodge Plantation down to the bigger
drainage area east of New Pond.

Illus. 35. The stream south of the small bridge visible in Illus. 34.

241
 THE BATTLE OF HASTINGS

Illus. 36. The dense gorse of the western part of Horselodge Plantation from the south.

Illus. 37. Horselodge Plantation near the most easterly part of New Pond. The English Heritage visitor
trail is visible bottom centre.

242
 ILLUSTRATIONS

Illus. 38. New Pond, built in 1815, looking west.

Illus. 39. The view east from the same point. Notice the volume of water. Both photographs taken 16th
November, 2005.

243
 THE BATTLE OF HASTINGS

Illus. 40. Another photograph taken east of New Pond. Note the sandy appearance of the water.

Illus. 41. Looking from the Stew Ponds to the abbey buildings, 4th January, 2002. Feature J is clearly
visible running across the centre of the picture. Perhaps because of concern about the danger of deep
water, English Heritage has since allowed vegetation of such density to grow around the Stew Ponds
that today the casual visitor may not even be aware of their existence.

244
 ILLUSTRATIONS

Illus.42. Feature J is in the lower foreground and beyond it a very distinct dip in the ground.

Illus. 43. The abbey terrace.

245
 THE BATTLE OF HASTINGS

Illus. 44. Feature J runs west. Horselodge Plantation is just off the photograph to the left.

Illus. 45. The westernmost surviving portion of Feature J runs diagonally immediately behind the
bushes in the right foreground before disappearing into the vegetation top left. This is not an easy
feature to find on the ground and on this occasion was only located by a number of people searching for
a number of minutes, despite having the English Heritage ground survey to hand.

246
 ILLUSTRATIONS

Illus. 46. The abbey terrace, created by the alteration of ground surfaces after the battle, looking west.

Illus. 47. View from the abbey terrace towards Telham Hill.

247
 THE BATTLE OF HASTINGS

Illus. 48. View from the abbey terrace looking towards Horselodge Plantation.

Illus. 49. A wall incorporating part of the wall of the south aisle of the eleventh-century abbey church,
the only part of that building still standing above ground level.

248
 ILLUSTRATIONS

Illus. 50. The steps leading from the terrace to the level of the church, with the monastic dormitory
to the right. The site of the high altar (Illus. 51), where what was believed to be Harold's body was
found, is a short way beyond the steps. In its original form this was one of the steepest gradients on
the entire battlefield.

249
 THE BATTLE OF HASTINGS

Illus. 51. No caption necessary.

Illus. 52. Battle High Street viewed from the abbey gatehouse.

250
 ILLUSTRATIONS

Illus. 53. The fourteenth-century abbey gatehouse, described by Sir Nikolaus Pevsner as ‘one of the
finest in England’.

251
 THE BATTLE OF HASTINGS

Illus. 54. The Church of St Mary, Upper Lake, Battle.

Illus. 55. Upper Lake runs west with the north wall of the abbey to the left and the High Street beyond.

252
 ILLUSTRATIONS

Illus. 56. Upper Lake descends eastwards towards the small roundabout (Illus. 57) which gives access
to Marley Lane to the left and Lower Lake to the right.

Illus. 57. Lower Lake descends the southern face of the Battle Abbey ridge.

253
 THE BATTLE OF HASTINGS

Illus. 58. Lower Lake climbs the slope to the small roundabout (Illus. 57).

Illus. 59. Looking north-east down Station Approach, with the railway station just visible on the right.
In the valley below the ridge the ground falls away from the Hastings road to both west and east, the
easterly slope here being evident, with the eastern part of the Battle Abbey ridge off to the left.

254
 ILLUSTRATIONS

Illus. 60. The view north from the footbridge over the railway line. The gradient is reasonably clear and
on the skyline the National School of 1842 is visible, see also Illus. 61.

Illus. 61. The eastern part of the Battle Abbey ridge c.1990. Newly-planted saplings prevented the
taking of an equivalent photograph in 2015. The National School of 1842 is visible centre left and the
fall of the houses to the right of it shows the line of Marley Lane. Notice the density of the vegetation in
the valley bottom.

255
 THE BATTLE OF HASTINGS

Illus. 62. And within that vegetation a small and sandy stream flows east.

Illus. 63. The top of Marley Lane, near the small roundabout (Illus. 57) which connects it with Lower
and Upper Lake. Notice the way in which the road begins to cut into the surrounding ground level as it
runs away from the camera position.

256
 ILLUSTRATIONS

Illus.64. Further down Marley Lane, the former National School of 1842, from above.

Illus. 65. And from below. The depth of the road cutting at this point suggests that there was once a
steep slope in the vicinity; this may have been where the English left wing rested in the battle.

257
 THE BATTLE OF HASTINGS

Illus. 66. The northern face of the eastern part of the ridge, photographed from a camera position close
to that of Illus. 63. For the reverse view, see Illus. 71 and 73.

Illus. 67 View of Caldbec Hill looking north from the top of Marley Lane.

258
 ILLUSTRATIONS

Illus. 68. And the reverse view from Caldbec. The houses in the middle distance are on Marley Lane.

Illus. 69. The northern face of the Battle Abbey ridge immediately behind the church of St Mary.

259
 THE BATTLE OF HASTINGS

Illus. 70. Another view of the northern face of the ridge taken from further east.

Illus. 71. The north face of the ridge with the westernmost part of Marley Lane on the horizon. In the
valley bottom is the sort of small stream and dense undergrowth so characteristic of the area.

260
 ILLUSTRATIONS

Illus. 72. The undergrowth in the valley bottom just north of Marley Lane.

Illus. 73. The slope which attackers would have faced had they managed to reach this point. For the
reverse view, see Illus. 66.

261
 THE BATTLE OF HASTINGS

Illus. 74. The estuary of the river Dives in Normandy as it was in the eighteenth century, Sauvage,
L’abbaye de Saint-Martin de Troarn, opp. p. 246. Bavent, near to which Wace says the river entered the
sea, is to the west of the island of Robehomme. The likely extent of the Dives estuary in 1066 can be
Illus. 74.
used to support the idea that Duke William assembled a large number of vessels within it, see above, pp.
147-8.

262
 ILLUSTRATIONS

Illus. 75. The stone slab found in excavations of Old Minster, Winchester in 1965 which
almost certainly depicts part of the story of the Scandinavian hero Sigmund, date
c.980x1093-4. Photograph copyright 2015 Winchester Excavations Committee/John
Crook; reproduced here by kind permission of The Winchester Excavations Committee,
Hampshire Cultural Trust and Winchester Museums.

263
 THE BATTLE OF HASTINGS

Illus. 76. A horseman on a capital in the crypt of Canterbury cathedral. The way in which he sits
distinctly low in the saddle is reminiscent of similar figures on the Bayeux Tapestry.

264
 THE LANCELOT ENGRAVINGS
OF THE BAYEUX TAPESTRY

In a sense, there is no ‘authentic’ version of the Bayeux Tapestry, given

that even the fabric now hanging in Bayeux is known to differ in certain
respects from that displayed there early in the eighteenth century. Of the
plates which follow, the first is an engraving of the Foucault drawing of
the first thirty feet of the Tapestry, and was published by Antoine
Lancelot in 1729; the remainder, published by Lancelot in 1733, were
modelled upon the drawings of the rest of the Tapestry made by Antoine
Benoît for Bernard de Montfaucon, and differ from those published by
Montfaucon himself (for them, see the next section). Lancelot’s
engravings omit the wavy line used to indicate the ground in the battle
scenes, the figures are rather freely drawn and the inscriptions seem
sometimes to stray from their proper positions. Even so, they are in other
respects more accurate than those of Montfaucon, and have some value
while the drawings upon which they are based remain unpublished. David
Hill stressed that they contain less detail than Benoît’s work and are
ultimately no substitute for it.1
These images were scanned by myself from a volume in the possession
of The London Library and as they are long out of copyright are in the
public domain.

1
Hill, ‘The Bayeux Tapestry’, pp. 390-1, where the first part of the Foucault drawing is reproduced.

265
THE BATTLE OF HASTINGS

266
LANCELOT ENGRAVINGS

267
THE BATTLE OF HASTINGS

268
LANCELOT ENGRAVINGS

269
THE BATTLE OF HASTINGS

270
LANCELOT ENGRAVINGS

271
THE BATTLE OF HASTINGS

272
 THE MONTFAUCON ENGRAVINGS
OF THE BAYEUX TAPESTRY

The images which follow are from the first two volumes of Montfaucon’s
Les Monumens de la Monarchie Françoise, published in 1729 and 1730.
They are in the public domain and have been placed online by the Library
of the University of Heidelburg under a Creative Commons Attribution-
ShareAlike License. In the first plate, based on the Foucault drawing, the
artistic style was adapted to suit contemporary taste in a way which
distorted the original,1 but this practice was abandoned in those which
followed.

1
Hicks, The Bayeux Tapestry, pp. 76-7. Hill, ‘The Bayeux Tapestry’, p. 399, noted that the
‘chiaroscuro engraving of 1729 disguises as much as it reveals, and some commentators have been
confused by it’.

273
THE BATTLE OF HASTINGS

274
MONTFAUCON ENGRAVINGS

275
THE BATTLE OF HASTINGS

276
MONTFAUCON ENGRAVINGS

277
THE BATTLE OF HASTINGS

278
MONTFAUCON ENGRAVINGS

279
THE BATTLE OF HASTINGS

280
MONTFAUCON ENGRAVINGS

281
THE BATTLE OF HASTINGS

282
MONTFAUCON ENGRAVINGS

283
 SELECT BIBLIOGRAPHY

Manuscript Sources

East Sussex Record Office.

BAT 4421, Nos. 6/7. Surveys by Richard Budgen of the eastern
part of the town of Battle and of the part of the estate south-west of the
abbey, 1724. Illustrated, p. xxx.
BAT 4435. Plans of the Several Farms Constituting the Estates of
Sir Godfrey Webster Bart. in Sussex according to Mr Willock’s Survey.
1811. No. 1: The Abbey Park. Illustrated, p. xxx.
TD/E 158 (Tithe Map of Battle). Plan of the Parish of Battle,
Sussex, surveyed by J.W. Cole, 1859. Illustrated, p. xxx.

Primary Sources

Adam of Bremen, Gesta Hammaburgensis Ecclesiae Pontificum, ed. B.

Schmeidler (3rd edn., Hannover, 1917); trans. F.J. Tschan, History of the
Archbishops of Hamburg-Bremen (New York, 1959).
Æthelweard, Chronicle, ed. A. Campbell (London, 1962).
Alfred the Great: Asser’s Life of King Alfred and other contemporary
sources, ed. S.D. Keynes and M. Lapidge (Harmondsworth, 1983).
Amatus of Montecassino, Storia de’ normanni, ed. V. de Bartholomaeis
(Rome, 1935).
Anglo-Saxon Chronicle, in Two of the Saxon Chronicles Parallel, ed. C.
Plummer (2 vols, Oxford, 1892-9); the 'D' text is also in The Anglo-Saxon
Chronicle: a collaborative edition. Volume 6. MS D, ed. G.P. Cubbin

285
 THE BATTLE OF HASTINGS

(Cambridge, 1996); there is a facsimile of E, The Peterborough

Chronicle, ed. D. Whitelock (Early English Manuscripts in Facsimile IV,
Copenhagen, 1954); the best translation of the Chronicle is in EHD1, No.
1 and EHD2, No. 1 (the entries for 1066 are reprinted The Battle of
Hastings, ed. Morillo, pp. 22-27; TNE, pp. 133-42; they are reprinted and
abridged in The Norman Conquest, ed. Brown, pp. 69-72); also trans. in
full in The Anglo-Saxon Chronicle, ed. M.J. Swanton (London, 1996).
References here are to Plummer’s edition.
An Anglo-Saxon Dictionary, ed. J. Bosworth and T.N. Toller (Oxford, 1898)
and Supplement.
Anglo-Saxon Writs, ed. and trans. F.E. Harmer (2nd edn., Stamford, 1989).
Annals of Nieder-Altaich. The Latin text of the entry for 1066 is given and
trans. by van Houts, ‘The Norman Conquest through European Eyes’,
841, n. 3; the trans. is reprinted, TNE, p. 131.
Asser, Life of King Alfred, ed. W.H. Stevenson (Oxford, 1904); trans. Alfred
the Great, ed. Keynes and Lapidge, pp. 67-110.
The Battle of Maldon, ed. and trans. D. Scragg, in The Battle of Maldon
AD991, ed. Scragg, pp. 18-31; also trans. EHD1, No. 10 (reprinted, The
Norman Conquest, ed. Brown, pp. 94-8).
Baudri of Bourgueil, Adelae Comitissae, in Baldricus Burgulianus:
Carmina, ed. K. Hilbert (Heidelberg, 1979), No. 134; Les Oeuvres
Poétiques de Baudri de Bourgueil (1046-1130), ed. P. Abrahams (Paris,
1926), No. CXCVI; trans. of ll. 207-572 of the Hilbert text (which is
superior to that of Abrahams) by M.W. Herren in Brown, The Bayeux
Tapestry: History and Bibliography, Appendix III, pp. 167-77; extracts
reprinted, TNE, pp. 125-8.
The Bayeux Tapestry. There are a number of important early reproductions
and treatments of the Tapestry. They are:
Lancelot, ‘Explication d’un monument de Guillaume le Conquérant’,
Memoires... de L’Academie Royale des Inscriptions et Belles
Lettres, vi (1729), 739-55.
Lancelot, ‘Suite de l’explication d’un Monument de Guillaume le
Conquérant’, Memoires... de L’Academie Royale des Inscriptions
et Belles Lettres, viii (1733), 602-68.
Montfaucon, B. de, Les Monumens de la Monarchie Françoise, i-ii
(Paris, 1729-30), i. plate xxxv, ii. plates i-ix.
Ducarel, A.C., Anglo-Norman Antiquities considered in a tour
through part of Normandy (London, 1767). Reproduces the
Lancelot engravings of 1733.

286
 SELECT BIBLIOGRAPHY

Stothard, C.A., ‘The Bayeux Tapestry’, Plates 1-17 were issued

individually by The Society of Antiquaries between 1819 and 1823
and later republished in Vetusta Monumenta, vol. vi.
La Tapisserie de Bayeux... édition variorum ... texte par Achille
Jubinal... dessins et gravures par Victor Sansonetti (Paris, 1838).
The Bayeux Tapestry, ed. F.M. Stenton (London, 1957).
The Bayeux Tapestry, ed. D.M. Wilson (London, 1985).
Brevis Relatio de Guillelmo nobilissimo comite Normannorum, ed. E.M.C.
van Houts, in Chronology, Conquest and Conflict in Medieval England,
Camden Miscellany XXXIV, Camden Fifth Series vol. 10 (Cambridge,
1997); reprinted and trans. E.M.C. van Houts, History and Family
Traditions in England and the Continent, 1000-1200 (Aldershot, 1999),
pp. 1-48.
The Burghal Hidage, ed. and trans. in The Defence of Wessex, ed. Hill and
Rumble, pp. 24-35; trans. Alfred the Great, ed. Keynes and Lapidge, pp.
193-4.
Caesar, J., The Gallic War, trans. H.J. Edwards (London, 1917).
Capitularia regum Francorum, i, ed. A. Boretius (Hanover, 1883).
Carmen de Hastingae Proelio, ed. and trans. C. Morton and H. Muntz
(Oxford, 1972); abridged reprint of the translation, The Battle of
Hastings, ed. Morillo, pp. 46-52; TNE, p. 129.
Carmen de Hastingae Proelio, ed. and trans. F. Barlow (Oxford, 1999).
Charlemagne: translated sources, trans. P.D. King (Kendal, 1987).
The Chronicle of Battle Abbey, ed. and trans. E. Searle (Oxford, 1980);
trans. extract, The Norman Conquest, ed. Brown, p. 121.
Chronicon Monasterii de Abingdon, ed. J. Stevenson (Rolls Series, 2, 1858).
Chronicon Monasterii de Hida, in Liber Monasterii de Hyda, ed. E.
Edwards (Rolls Series, 45, 1866), pp. 283-321.
Chronicon Roskildense, in Scriptores Minores, ed. Gertz, i. 3-33.
La Chronique de Saint-Maixent 751-1140, ed. and trans. J. Verdon (Paris,
1979).
Domesday Book, ed. A. Farley and H. Ellis (London, 1783-1816); also
published by the Phillimore Press in county volumes, with translations.
Dudo of St Quentin, History of the Normans, trans. E. Christiansen
(Woodbridge, 1998).
Eadmer, Historia Novorum in Anglia, ed. M. Rule (Rolls Series, 81, 1884);
trans. G. Bosanquet, Eadmer’s History of Recent Events in England
(London, 1964); trans. extracts, TNE, pp. 146-50.

287
 THE BATTLE OF HASTINGS

Encomium Emmae Reginae, ed. and trans. A. Campbell, with a

supplementary introduction by Simon Keynes (Cambridge, 1998). First
published as Camden Third Series, lxxii, 1949.
English Historical Documents Volume 1: c.500-1042, ed. D. Whitelock (2nd
edn., London, 1979).
English Historical Documents Volume 2: 1042-1189, ed. D.C. Douglas and
G.W. Greenaway (2nd edn., London, 1981).
Ermold the Black, In honorem Hludovici Pii, ed. and trans. in part by P.
Godman, Poetry of the Carolingian Renaissance (London, 1985), pp.
250-5.
Etienne de Rouen, see Stephen of Rouen.
Florence of Worcester, see John of Worcester.
Fulcoius of Beauvais, Jephthah, trans. TNE, pp. 132-3.
Gaimar, G., L’Estoire des Engleis, ed. A. Bell (Oxford, 1960); ed. and trans.
Sir T.D. Hardy and C.T. Martin (Rolls Series, 91, 1888-9).
Die Gesetze der Angelsachsen, ed. F. Liebermann (Halle, 1898-1916).
Guy, bishop of Amiens, see Carmen de Hastingae Proelio.
Henry of Huntingdon, Historia Anglorum, ed. and trans. D. Greenway
(Oxford, 1996); partial reprint of the translation, TNE, pp. 150-6.
Herman the Archdeacon, Liber de Miraculis Sancti Eadmundi, in
Memorials of St Edmund’s Abbey, ed. T. Arnold, vol. i (Rolls Series, 96,
London, 1890), 26-92.
John of Worcester, The Chronicle of John of Worcester. Volume II. The
Annals from 450 to 1066, ed. and trans. R.R. Darlington and P. McGurk
(Oxford, 1995); partial reprint of the translation, TNE, pp. 142-6. The
entry for 1066 is translated under the name Florence of Worcester, in
EHD2, No. 2; trans. extracts also in The Norman Conquest, ed. Brown,
pp. 58, 68, 72-3, 78.
The Laws of the Earliest English Kings, ed. F.L. Attenborough (Cambridge,
1922).
Liber Eliensis, ed. E.O. Blake (Camden Third Series, xcii, 1962).
The Life of King Edward who rests at Westminster, attributed to a monk of
Saint-Bertin, ed. and trans. F. Barlow (2nd edn., Oxford, 1992); partial
reprint of the translation of the first edition, The Norman Conquest, ed.
Brown, pp. 82-93.
The Norman Conquest, ed. R.A. Brown (London, 1984). A collection of
translated sources.
The Normans in Europe, ed. and trans. E.M.C. van Houts (Manchester,
2000).

288
 SELECT BIBLIOGRAPHY

Orderic Vitalis, Historia Æcclesiastica, ed. and trans. M. Chibnall (6 vols.,

Oxford, 1969-80); partial reprint of the translation, The Norman
Conquest, ed. Brown, pp. 100-14.
Penitential articles issued after the Battle of Hastings, in Councils & Synods
with other Documents Relating to the English Church. I. A.D.871-1204.
Part II 1066-1204, ed. D. Whitelock et al. (Oxford, 1981), 581-4; trans.,
The Norman Conquest, ed. Brown, pp. 156-7; EHD2, No. 81.
Priest of Fécamp, Letter on the Battle of Tinchebrai; Latin text, EHR, xxv
(1910), 296; trans. EHD2, No. 9. This translation omits the following
words on the size of Robert of Normandy’s forces, Comes uero ad vi
milia habuit, equites septingentos.
Regesta Regum Anglo-Normannorum. The Acta of William I (1066-1087,
ed. D. Bates (Oxford, 1998).
The Reign of Charlemagne: Documents on Carolingian Government and
Administration, trans. H.R. Loyn and J. Percival (London, 1975).
Saxo Grammaticus, Danorum Regum Heroumque Historia Books X-XVI, ed.
and trans. E. Christiansen (British Archaeological Reports, International
Series, vols. 84, 118, 1980-1).
Scriptores Minores Historiæ Danicæ Medii Ævi, ed. M. Cl. Gertz
(Copenhagen, 1917-22).
Ship List of William the Conqueror, ed. E..M.C. van Houts, in ‘The Ship
List of William the Conqueror’, ANS, X (1987), 175-6; trans. TNE, pp.
130-1.
Snorri Sturluson, King Harald’s Saga, trans. M. Magnusson and H. Pálsson
(London, 1966).
Stephen of Rouen, Draco Normannicus, ed. R. Howlett, Chronicles of the
Reigns of Stephen, Henry II and Richard I (Rolls Series, 82, 1884-9), ii.
589-781.
Sven Aggesen, Lex Castrensis, in Scriptores Minores ed. Gertz, i. 64-92;
trans. E. Christiansen, The Works of Sven Aggesen (London, 1992), pp.
31-43.
Symeon of Durham, Historia Regum, in Symeonis Monachi Opera Omnium,
ed. T. Arnold (Rolls Series, 75, 1882-5).
Thietmar of Merseburg, Chronicon, ed. R. Holtzmann (2nd edn., Berlin,
1955); trans. extract EHD1, pp. 347-50.
Vegetius, Epitome of Military Science, trans. N.P. Milner (Liverpool, 2nd
edn., 1996).
Vita Haroldi. The Romance of the Life of Harold, King of England, ed. and
trans. W. de G. Birch (London, 1885).

289
 THE BATTLE OF HASTINGS

Wace, Roman de Rou, ed. A.J. Holden (3 vols., Société des Anciens Textes
Francais, Paris, 1970-3); the section on William the Conqueror is trans. E.
Taylor, Master Wace, His Chronicle of the Norman Conquest from the
Roman de Rou (London, 1837); Holden’s text is reprinted and trans. in
full by G.S. Burgess, Wace. The Roman de Rou (St Helier, 2002).
The Waltham Chronicle, ed. and trans. L. Watkiss and M. Chibnall (Oxford,
1994); partial reprint of the translation, TNE, pp. 156-9.
William of Jumiéges, Gesta Normannorum Ducum, ed. and trans. E.M.C.
van Houts (Oxford,1992-5); partial reprint of the translation, TNE, pp.
113-7; translations of the passage on Hastings in EHD2, No. 3; (reprinted
The Battle of Hastings, ed. Morillo, pp. 18-19); trans. extracts, The
Norman Conquest, ed. Brown, pp. 3-15
William of Malmesbury, Gesta Regum Anglorum: The History of the
English Kings, ed. and trans. R.A.B. Mynors, R.M. Thomson and M.
Winterbottom (Oxford, 1998-9); partial translations, TNE, pp. 161-70;
The Norman Conquest, ed. Brown, pp. 115-7.
William of Malmesbury, De Gestis Pontificum Anglorum Libri Quinque, ed.
N.E.S.A. Hamilton (Rolls Series, 52, 1870).
William of Malmesbury, Vita Wulfstani, ed. R.R. Darlington (Camden
Society, Third Series, xl, 1928); trans. M. Swanton, Three Lives of the
Last Englishmen (London, 1984); partial reprint of the translation, TNE,
pp. 159-61.
William of Poitiers, Gesta Guillelmi, ed. and trans. R.H.C. Davis and M.
Chibnall (Oxford, 1998); partial reprint of the translation, TNE, pp. 118-
25; full translation of the passage on Hastings in The Battle of Hastings,
ed. Morillo, pp. 12-15; abridged translations in EHD2, No. 4; The
Norman Conquest, ed. Brown, pp. 32-5; full Latin text with French
translation, Histoire de Guillaume le Conquérant, ed. R. Foreville (Paris,
1952).
Xenophon, Spartan Society, trans. R.J.A. Talbert, Plutarch on Sparta
(London, 1988).

Secondary Sources

Abels, R.P., Lordship and Military Obligation in Anglo-Saxon England

(Berkeley, 1988).
Abels, R.P., ‘English Tactics, Strategy and Military Organization in the Late
Tenth Century’, The Battle of Maldon, ed. Scragg, pp. 143-55.

290
 SELECT BIBLIOGRAPHY

Abels, R.P., ‘From Alfred to Harold II: The Military Failure of the Late
Anglo-Saxon State’, The Normans and Their Adversaries at War, ed.
Abels and Bachrach, pp. 15-30.
Albu, E., The Normans in their Histories: Propaganda, Myth and
Subversion (Woodbridge, 2001).
Anglo-Norman Warfare: Studies in Late Anglo-Saxon and Anglo-Norman
Military Organization and Warfare, ed. M. Strickland (Woodbridge,
1992).
Archer, T.A., ‘Mr. Freeman and the “Quarterly Review”’, Contemporary
Review, March 1893, 335-55.
Archer, T.A., ‘The Battle of Hastings. Part I’, EHR, ix (1894), 1-41, 602-
608.
Armies, Chivalry and Warfare in Medieval Britain and France:
Proceedings of the 1995 Harlaxton Symposium, ed. M. Strickland
(Stamford, 1998).
Bachrach, B.S., ‘Some Observations on the Military Administration of the
Norman Conquest’, ANS, VIII (1985), 1-25.
Bachrach, B.S., ‘On the Origins of William the Conqueror's Horse
Transports’, Technology and Culture, 26 (1985), 505-31.
Bachrach, B.S., ‘The Feigned Retreat at Hastings’, Medieval Studies, 33
(1971), 344-7; reprinted The Battle of Hastings, ed. Morillo, pp. 190-3.
References here are to the reprint.
Bachrach, B.S., ‘William Rufus’ Plan for the Invasion of Aquitaine’, The
Normans and Their Adversaries at War, ed. Abels and Bachrach, pp. 31-
63.
Bachrach, B.S., ‘The Norman Conquest, Countess Adela, and Abbot
Baudri’, ANS, XXXV (2012), 65-78.
Ballard, A., The Domesday Boroughs (Oxford, 1904).
Barclay, C.N., Battle 1066 (London, 1966).
Baring, F.H., ‘The Conqueror’s Footprints in Domesday’, EHR, xiii (1898),
17-25; reprinted with additions and alterations as Appendix A of his
Domesday Tables, pp. 207-16.
Baring, F.H., ‘The Battle Field of Hastings’, EHR, xx (1905), 65-70;
reprinted with additions as Appendix B, ‘On the Battle of Hastings’, of
his Domesday Tables, pp. 217-32.
Baring, F.H., ‘The Malfosse at the Battle of Hastings’, EHR, xxii (1907),
69-72.
Baring, F.H., Domesday Tables (London, 1909).
Barlow, F., ‘Edward the Confessor and the Norman Conquest’, Barlow, The
Norman Conquest and Beyond, pp. 99-111.

291
 THE BATTLE OF HASTINGS

Barlow, F., ‘The Carmen de Hastingae Proelio’, Studies in International

History Presented to W. Norton Medlicott, ed. K. Bourne and D.C. Watt
(London, 1967), pp. 36-67; reprinted Barlow, The Norman Conquest and
Beyond, pp. 189-222. References here are to the reprint.
Barlow, F., Edward the Confessor (London, 1970).
Barlow, F., The Norman Conquest and Beyond (London, 1983).
Barlow, F., The Godwins: The Rise and Fall of a Noble Dynasty (Harlow,
2002).
Bateley, J., The Anglo-Saxon Chronicle: Texts and Textual Relationships
(Reading, 1991).
Bates, D., ‘The Character and Career of Odo, Bishop of Bayeux (1049/50-
1097)’, Speculum, l (1975), 1-20.
Bates, D., Normandy before 1066 (London, 1982).
Bates, D., William the Conqueror (London, 1989).
The Battle of Hastings: Sources and Interpretations, ed. S. Morillo
(Woodbridge, 1996).
The Battle of Maldon AD 991, ed. D. Scragg (Oxford, 1991).
The Bayeux Tapestry: Emboidering the Facts of History. Proceedings of the
Cerisy Colloquium (1999) ed. P. Bouet et al. (Caen, 2004). For the
French text see La Tapisserie de Bayeux. Articles by English-speaking
scholars are best read in the English version, those by French-speakers in
the French.
The Bayeux Tapestry: New Approaches, ed. M.J. Lewis et al. (Oxford,
2011).
The Bayeux Tapestry: New Interpretations, ed. M.K. Foys et al.
(Woodbridge, 2009).
Bédat, I. and Girault-Kurzemann, B., ‘Étude technique de la Broderie de
Bayeux’, La Tapisserie de Bayeux, ed. Bouet, pp. 83-109.
Beech, G., ‘The Participation of Aquitanians in the Conquest of England
1066-1100’, ANS, IX (1986), 1-24.
Beech, G.T., ‘The Breton Campaign and the Possibility that the Bayeux
Tapestry was Produced in the Loire Valley (St Florent of Saumur)’, The
Bayeux Tapestry:New Approaches, ed. Lewis, pp. 10-16.
Beeler, J., ‘The Composition of Anglo-Norman Armies’, Speculum, 40
(1965), 398-414.
Beeler, J., Warfare in England 1066-1189 (New York, 1966).
Behrens, L.B., Battle Abbey under Thirty-Nine Kings (London, 1937).
Bennett, M., ‘Poetry as History? The “Roman de Rou” of Wace as a source
for the Norman conquest’, ANS, V (1982), 21-39.

292
 SELECT BIBLIOGRAPHY

Bennett, M., ‘Wace and Warfare’, ANS, XI (1988), 37-57.

Bennett, M., ‘The Myth of the Military Supremacy of Knightly Cavalry’,
Armies, Chivalry and Warfare, ed. Strickland, pp. 304-16.
Bennett, M., Campaigns of the Norman Conquest (Botley, 2001).
Bernstein, D.J., The Mystery of the Bayeux Tapestry (London, 1986).
Blacker, J., The Faces of Time: Portrayal of the Past in Old French and
Latin Historical Narrative of the Anglo-Norman Regnum (Austin, 1994).
Blackman, H., ‘The Story of the old gunpowder works at Battle’, Sussex
Archæological Collections, lxiv (1923), 109-22.
Blood Red Roses: The archaeology of a mass grave from the Battle of
Towton AD 1461, ed. V. Fiorato et al. (Oxford, 2000).
Bond, G., ‘“Iocus Amoris”: The Poetry of Baudri of Bourgeuil and the
Formation of the Ovidian Subculture’, Traditio, 42 (1986), 143-93.
Boüard, M. de., Guillaume le Conquérant (Paris, 1984).
Bouet, P., ‘La Felicitas de Guillaume le Conquérant dans les Gesta
Guillelmi de Guillaume de Poitiers’, ANS, IV (1981), 37-52, 174-81.
Bradbury, J., ‘Battles in England and Normandy, 1066-1154’, ANS, VI
(1983), 1-12; reprinted in Anglo-Norman Warfare, ed. Strickland, pp.
182-93.
Bradbury, J., The Battle of Hastings (Stroud, 1998).
Brakspear, H., ‘Battle Abbey’, Antiquaries Journal, 11 (1931), 166-8.
Brakspear, H., ‘The Abbot’s House at Battle’, Archaeologia, lxxxiii (1933),
139-66.
Brandon, P., The Sussex Landscape (London, 1974).
Brilliant, R., ‘The Bayeux Tapestry: a stripped narrative for their eyes and
ears’, Word and Image, 7 (1991), 93-125; reprinted The Study of the
Bayeux Tapestry, ed. Gameson, pp. 111-37. References here are to the
reprint.
British Geological Survey, 1:50,000 Series, Sheet 320/321, Hastings and
Dungeness.
Brooks, F.W., The Battle of Stamford Bridge (York, 1956).
Brooks, N.P., ‘Arms, Status and Warfare in Late-Saxon England’, Ethelred
the Unready: Papers from the Millenary Conference, ed. D. Hill (Oxford,
1978), pp. 81-103.
Brooks, N.P., ‘Weapons and Armour’, The Battle of Maldon, ed. Scragg, pp.
208-19.
Brooks, N.P. and Walker, H.E., ‘The Authority and Interpretation of the
Bayeux Tapestry’, ANS, I (1978), 1-34, 191-9; reprinted The Study of the

293
 THE BATTLE OF HASTINGS

Bayeux Tapestry, ed. Gameson, pp. 63-92. References here are to the
original.
Brown, R.A., The Normans and the Norman Conquest (London, 1969).
Brown, R.A., Origins of English Feudalism (London, 1973).
Brown, R.A., ‘The Battle of Hastings’, ANS, III (1980), 1-21, 197-201;
reprinted Anglo-Norman Warfare, ed. Strickland, pp. 161-81, 273-7; The
Battle of Hastings, ed. Morillo, pp. 196-218. References here are to the
original.
Brown, S.A., The Bayeux Tapestry: History and Bibliography (Woodbridge,
1988).
Brown, S.A., ‘The Bayeux Tapestry: Why Eustace, Odo and William?’,
ANS, XII (1989), 7-28.
Brown, S.A and Herren, M.W., ‘The Adelae Comitissae of Baudri of
Bourgeuil and the Bayeux Tapestry’, ANS, XVI (1993), 55-73; reprinted
The Study of the Bayeux Tapestry, ed. Gameson, pp. 139-55. References
here are to the original.
Bryce, J., ‘Edward Augustus Freeman’, EHR, vii (1892), 497-509.
Budny, M., ‘The Byrhtnoth Tapestry or Embroidery’, The Battle of Maldon,
ed. Scragg, pp. 263-78.
Bullough, D., ‘Europae Pater: Charlemagne and his achievement in the
light of recent scholarship’, EHR, lxxxv (1970), 59-105.
Burne, A.H., The Battlefields of England (London, 1950).
Campbell, J., ‘Observations on English Government from the Tenth to the
Twelfth Century’, TRHS, Fifth Series, xxv (1975), 39-54; reprinted
Campbell, Essays in Anglo-Saxon History, pp. 155-70. References here
are to the reprint.
Campbell, J., ‘England, France, Flanders and Germany in the Reign of
Ethelred II: Some Comparisons and Connections’, Ethelred the Unready:
Papers from the Millenary Conference, ed. D. Hill (Oxford, 1978), pp.
255-70; reprinted Campbell, Essays in Anglo-Saxon History, pp. 191-
207. References here are to the reprint.
Campbell, J., ‘The Significance of the Anglo-Norman State in the
Administrative History of Western Europe’, Histoire Comparée de
l’Administration (IVe – XVIIIe Siècles), Beihefte der Francia, IX (Munich,
1980), 117-34; reprinted Campbell, Essays in Anglo-Saxon History, pp.
171-89. References here are to the reprint.
Campbell, J., ‘Some Twelfth-Century Views of the Anglo-Saxon Past’,
Peritia, 3 (1984), 131-50; reprinted Campbell, Essays in Anglo-Saxon
History, pp. 209-28. References here are to the reprint.

294
 SELECT BIBLIOGRAPHY

Campbell, J., ‘Asser’s Life of Alfred’, The Inheritance of Historiography,

350-900, ed. C. Holdsworth and T.P. Wiseman (Exeter, 1986), pp. 115-
35; reprinted Campbell, The Anglo-Saxon State, pp. 129-55. References
here are to the reprint.
Campbell, J., Essays in Anglo-Saxon History (London, 1986).
Campbell, J., ‘Some Agents and Agencies of the Late Anglo-Saxon State’,
Domesday Studies, ed. Holt, pp. 201-18; reprinted Campbell, The Anglo-
Saxon State, pp. 201-25. References here are to the reprint.
Campbell, J., ‘Was it Infancy in England? Some Questions of Comparison’,
England and her Neighbours, 1066-1453: Essays in Honour of Pierre
Chaplais, ed. M. Jones and M. Vale (London, 1989), pp. 1-17; reprinted
Campbell, The Anglo-Saxon State, pp. 179-99. References here are to the
reprint.
Campbell, J., ‘The Late Anglo-Saxon State: A Maximum View’,
Proceedings of the British Academy, 87 (1994), 39-65; reprinted
Campbell, The Anglo-Saxon State, pp. 1-30. References here are to the
reprint.
Campbell, J., ‘Stenton’s Anglo-Saxon England’, Stenton’s ‘Anglo-Saxon
England’ Fifty Years On, ed. D. Matthew (Reading, 1994), pp. 49-59;
reprinted Campbell, The Anglo-Saxon State, pp. 269-80. References here
are to the reprint.
Campbell, J., The Anglo-Saxon State (London, 2000).
Campbell, J., ‘What is not known about the reign of Edward the Elder’,
Edward the Elder, ed. Higham and Hill, pp. 12-24.
Chevallier, C.T., ‘Where was Malfosse? The End of the Battle of Hastings’,
Sussex Archælogical Collections, ci (1963), 1-13.
Chibnall, M., ‘Military Service in Normandy before 1066’, ANS,V (1982),
65-77; reprinted, The Battle of Hastings, ed. Morillo, pp. 80-92.
References here are to the reprint.
Chibnall, M., ‘Orderic Vitalis and the Bayeux Tapestry’, The Bayeux
Tapestry, ed. Bouet, pp. 127-33.
Clarke, H.B., ‘The Identity of the Designer of the Bayeux Tapestry’, ANS,
XXXV (2012), 119-39.
Clephane-Cameron, N. et al., The 1066 Malfosse Walk (Battle, 2000).
Cleveland, Duchess of (published anonymously), History of Battle Abbey
(London, 1877).
C.L.W.C., A Guide to Battle Abbey (Battle and Hastings, n.d.).
Coates, R., ‘On the Alleged Frankish Origin of the Hastings Tribe’, Sussex
Archæological Collections, 117 (1979), 263-4.
Connolly, P., Greece and Rome at War (London, 1981).

295
 THE BATTLE OF HASTINGS

Contamine, P., War in the Middle Ages, trans. M. Jones (Oxford, 1984).
Coupland, S., ‘Carolingian Arms and Armor in the Ninth Century’, Viator,
xx (1989), 29-50.
Cowdrey, H.E.J., ‘Bishop Ermenfrid of Sion and the Penitential Ordinance
following the Battle of Hastings’, Journal of Ecclesiastical History, xx
(1969), 225-42.
Cowdrey, H.E.J., ‘The Anglo-Norman Laudes Regiae’, Viator, xii (1981),
37-78.
Cowdrey, H.E.J., ‘Towards an Interpretation of the Bayeux Tapestry’, ANS,
X (1987), 49-65; reprinted The Study of the Bayeux Tapestry, ed.
Gameson, pp. 93-110. References here are to the reprint.
Cowdrey, H.E.J., ‘King Harold II and the Bayeux Tapestry: a Critical
Introduction’, King Harold II and The Bayeux Tapestry, ed. Owen-
Crocker, pp. 1-15.
Daniell, C., ‘Battle and Trial: Weapon Injury Burials of St Andrew’s
Church, Fishergate, York’, Medieval Archaeology, xlv (2001), 220-6.
Darby, H.C., Domesday England (Cambridge, 1977).
David, C.W., Robert Curthose (Cambridge, 1920).
Davis, R.H.C., ‘Alfred the Great: Propaganda and Truth’, History, lvi
(1971), 169-82; reprinted Davis, From Alfred the Great to Stephen, pp.
33-46. References here are to the reprint.
Davis, R.H.C., ‘The Carmen de Hastingae Proelio’, EHR, xciii (1978), 241-
61; reprinted Davis, From Alfred the Great to Stephen, pp. 79-100.
References here are to the reprint.
Davis, R.H.C., ‘William of Poitiers and his History of William the
Conqueror’, The Writing of History in the Middle Ages, ed. R.H.C. Davis
and J.M. Wallace-Hadrill (Oxford, 1981), pp. 71-100; reprinted Davis,
From Alfred the Great to Stephen, pp. 101-30. References here are to the
reprint.
Davis, R.H.C., ‘The Warhorses of the Normans’, ANS, X (1987), 67-82;
reprinted Davis, From Alfred the Great to Stephen, pp. 63-78. References
here are to the original.
Davis, R.H.C., The Medieval Warhorse (London, 1989).
Davis, R.H.C., Engels, L.J., et al., ‘The Carmen de Hastingae Proelio: a
discussion’, ANS, II (1979), 1-20, 165-7.
Davis, R.H.C., From Alfred the Great to Stephen (London, 1991).
Dawson, C., The “Restorations” of the Bayeux Tapestry (London, 1907).
Dawson, C., ‘The Bayeux Tapestry in the hands of the Restorers’, The
Antiquary, xlv (1909), 470.

296
 SELECT BIBLIOGRAPHY

The Defence of Wessex: the Burghal Hidage and Anglo-Saxon

Fortifications, ed. D. Hill and A.R. Rumble (Manchester, 1996).
Delbrück, H., Geschichte der Kriegskunst im Rahmen der politischen
Geschichte, iii (Berlin, 1923); trans. by W.J. Renfroe Jnr., History of the
Art of War Within the Framework of Political History (4 vols., London,
1975-85). References here are to the translation.
DeVries, K., The Norwegian Invasion of England (Woodbridge, 1999).
DeVries, K., ‘Harold Godwinson in Wales: Military Legitimacy in Late
Anglo-Saxon England’, The Normans and Their Adversaries at War, ed.
Abels and Bachrach, pp. 65-85.
Dibdin, T.F., A Bibliographical Antiquarian and Picturesque Tour in
France and Germany (3 vols., London, 1821).
Didier, M-H., ‘La Broderie, une oeuvre textile. Les expertises et les
analyses effectuées en 1982-1983: la mise en place de l’opération, La
Tapisserie de Bayeux, ed. Bouet, pp. 77-82.
Dodwell, C.R., ‘The Bayeux Tapestry and the French Secular Epic’,
Burlington Magazine, cviii (1966), 549-60; reprinted The Study of the
Bayeux Tapestry, ed. Gameson, pp. 47-62. References here are to the
reprint.
Dodwell, C.R., Anglo-Saxon Art: A new perspective (Manchester, 1982).
Domesday Studies, ed. J.C. Holt (Woodbridge, 1987).
Douglas, D.C., ‘Companions of the Conqueror’, History, xxviii (1943), 129-
47.
Douglas, D.C., ‘Edward the Confessor, Duke William of Normandy, and the
English Succession’, EHR, lxviii (1953), 526-45.
Douglas, D.C. William the Conqueror: the Norman impact upon England
(London, 1964).
Ducarel, A.C., see Bayeux Tapestry under Primary Sources.
Dumville, D., ‘Some aspects of annalistic writing at Canterbury in the
eleventh and early twelfth centuries’, Peritia, 2 (1983), 23-57.
Dumville, D., ‘The Tribal Hidage: an introduction to its texts and their
history’, The Origins of Anglo-Saxon Kingdoms, ed. S. Bassett (Leicester,
1989), pp. 225-30.
Edward the Elder 899-924, ed. N.J. Higham and D.H. Hill (London, 2001).
Ekwall, E., The Concise Oxford Dictionary of English Place-Names (4th
edn., Oxford, 1960).
Eley, P. and Bennett, P.E., ‘The Battle of Hastings according to Gaimar,
Wace and Benoît: rhetoric and politics’, Nottingham Medieval Studies,
xliii (1999), 47-78.

297
 THE BATTLE OF HASTINGS

Engels, L.J., ‘Once more: the Carmen de Hastingae Proelio’, ANS, II

(1979), 3-18, 165-7.
English Heritage, An Earthwork Survey and Investigation of the Parkland at
Battle Abbey, East Sussex (Archaeological Investigation Report Series
AI/13/2002, Swindon, 2002).
English Romanesque Art 1066-1200: Hayward Gallery, London 5 April-8
July 1984 (London, 1984).
English, B., ‘The Coronation of Harold in the Bayeux Tapestry; in The
Bayeux Tapestry, ed. Bouet, pp. 347-81.
Fellows-Jensen, G., ‘The Myth of Harold II’s Survival in the Scandinavian
Sources’, King Harold II and The Bayeux Tapestry, ed. Owen-Crocker,
pp. 53-64.
Focus Multimedia Ltd., Red Shift 4 (Rugely, 2000).
Foys, M.K., ‘Pulling the Arrow Out: The Legend of Harold’s Death and the
Bayeux Tapestry’, The Bayeux Tapestry: New Interpretations, ed. Foys,
pp. 158-75.
France, J., Victory in the East: a military history of the First Crusade
(Cambridge, 1994).
France, J., ‘The Importance of the Bayeux Tapestry for the History of War’,
The Bayeux Tapestry, ed. Bouet, pp. 290-9.
Freeman, E.A., The History of the Norman Conquest of England (2nd edn.,
6 vols., Oxford, 1870-79). Reprint of part of vol. iii, The Battle of
Hastings, ed. Morillo, pp. 150-64.
Fuller, J.F.C., The Decisive Battles of the Western World and their influence
upon history (3 vols., London, 1954-6).
Gameson, R., ‘The Origin, Art, and Message of the Bayeux Tapestry’, The
Study of the Bayeux Tapestry, ed. Gameson, pp. 157-211.
Garnett, G., ‘Coronation and Propaganda: some Implications of the Norman
claim to the throne of England in 1066’, TRHS, Fifth Series 36, 1986, 91-
116.
Garnett, G., ‘Conquered England, 1066-1215’, The Oxford Illustrated
History of Medieval England, ed. N. Saul (Oxford, 1997), pp. 61-101.
Garnett, G., Conquered England: Kingship, Succession and Tenure, 1066-
1166 (Oxford, 2007).
Gautier Dalché, P. and Tilliette, J-Y., ‘Un nouveau document sur la tradition
du poème de Baudri de Bourgeuil à la comtesse Adèle’, Bibliothèque de
l’école des chartes, 144 (1986), 241-57.
George, R.H., ‘The Contribution of Flanders to the Conquest of England’,
Revue belge de philologie et d’histoire, v (1926), 81-97.

298
 SELECT BIBLIOGRAPHY

Gibbs-Smith, C.H., ‘What the Bayeux Tapestry does not show’, The Times,
October 8, 1966, p. 9.
Gibbs-Smith, C.H., The Bayeux Tapestry (London, 1973).
Gillingham, J., ‘William the Bastard at War’, Studies in Medieval History
presented to R. Allen Brown, ed. Harper-Bill, pp. 141-58; reprinted
Anglo-Norman Warfare, ed. Strickland, pp. 143-60; The Battle of
Hastings, ed. Morillo, pp. 96-112. References here are to the original.
Gillingham, J., ‘Henry of Huntingdon and the Twelfth-Century Revival of
the English Nation’, Concepts of National Identity in the Middle Ages, ed.
S. Forde et al. (Leeds, 1995), pp. 75-101.
Gillingham, J., ‘Holding to the Rules of War (Bellica Iura Tenentes): Right
Conduct before, during, and after Battle in North-Western Europe in the
Eleventh Century’, ANS, XXIX (2006), 1-15.
Gillmor, M., ‘Naval Logistics of the Cross-Channel Operation, 1066’, ANS,
VII (1984), 105-31; reprinted The Battle of Hastings, ed. Morillo, pp.
114-28. References here are to the original.
Glover, R., ‘English Warfare in 1066’, EHR, lxvii (1952), 1-18; reprinted
The Battle of Hastings, ed. Morillo, pp. 174-88. References here are to
the original.
Godman, P., Poetry of the Carolingian Renaissance, see Ermold the Black
under Primary Sources.
The Golden Age of Anglo-Saxon Art 966-1066, ed. J. Backhouse et al.
(London, 1984).
Grainge, C. and G., ‘The Pevensey Expedition: Brilliantly Executed Plan or
Near Disaster?’, The Mariner's Mirror, 79 (1993), 261-73; reprinted The
Battle of Hastings, ed. Morillo, pp. 130-42. References here are to the
reprint.
Gransden, A., Historical Writing in England c.550 to c.1307 (London,
1974).
Grape, W., The Bayeux Tapestry: Monument to a Norman Triumph
(Munich, 1994).
Grassi, J.L., ‘T he Lands and Revenues of Edward the Confessor’, EHR,
cxvii (2002), 251-83.
Gravett, C., Hastings 1066: the fall of Saxon England (London, 1992).
Greenway, D., ‘Authority, Convention and Observation in Henry of
Huntingdon’s Historia Anglorum’, ANS, XVIII (1995), 105-21.
Grehan, J. and Mace, M., The Battle of Hastings 1066: The Uncomfortable
Truth. Revealing the true location of England’s most famous battle
(Barnsley, 2012).

299
 THE BATTLE OF HASTINGS

Grierson, P., ‘A Visit of Earl Harold to Flanders in 1056’, EHR, li (1936),

90-7.
Grierson, P., ‘The relations between England and Flanders before the
Norman Conquest’, TRHS, Fourth Series, xxiii (1941), 71-112.
Gurney, H., ‘Observations on the Bayeux Tapestry’, Archaeologia, xviii
(1817), 359-70.
Hare, J.N., ‘The buildings of Battle Abbey: a preliminary survey’, ANS, III
(1980), 78-95, 212-3.
Hare, J.N., Battle Abbey: the eastern range and the excavations of 1978-80
(Historic Buildings and Monuments Commission for England, London,
1985).
Hart, C., ‘The Canterbury Contribution to the Bayeux Tapestry’, Art and
Symbolism in Medieval Europe: Papers of the ‘Medieval Europe Brugge
1997’ Conference, V (1997), 7-15.
Hart, C., ‘The Bayeux Tapestry and Schools of Illumination at Canterbury’,
ANS, XXII (1999), 117-67.
Hart, C., ‘The Cicero-Aratea and the Bayeux Tapestry’, King Harold II and
The Bayeux Tapestry, ed. Owen-Crocker, pp. 161-78.
The Heads of Religious Houses England and Wales 940-1216, ed. D.
Knowles, et al. (Cambridge, 1972).
Higham, N.J., The Death of Anglo-Saxon England (Stroud, 1997).
Hill, D., ‘Offa’s Dyke: Pattern and Purpose’, The Antiquaries Journal, 80
(2000), 195-206.
Hill, D., ‘The Bayeux Tapestry:The Establishment of a Text’, The Bayeux
Tapestry, ed. Bouet, pp. 383-401.
Hill, D. and McSween, J., ‘The Storage Chest and the Repairs in the Bayeux
Tapestry’, The Bayeux Tapestry: New Approaches, ed. Lewis, pp. 44-51.
Hollister, C.W., Anglo-Saxon Military Institutions on the Eve of the Norman
Conquest (Oxford, 1962).
Hollister, C.W., ‘The Greater Domesday Tenants-in-Chief’, Domesday
Studies, ed. Holt, pp. 219-48.
Hooper, J., ‘The “Rows of the Battle-Swan”: The Aftermath of Battle in
Anglo-Saxon Art’, Armies, Chivalry and Warfare, ed. Strickland, pp. 82-
99.
Hooper, N., ‘The Housecarls in England in the Eleventh Century’, ANS, VII
(1984), 161-76.
Hooper, N., ‘Some Observations on the Navy in Late Anglo-Saxon
England’, Studies in Medieval History presented to R. Allen Brown, ed.
Harper-Bill, pp. 203-13.

300
 SELECT BIBLIOGRAPHY

Hooper, N., ‘Edgar the Ætheling: Anglo-Saxon prince, rebel and crusader’,
Anglo-Saxon England, 14 (1985), 197-214.
Hooper, N., ‘The Aberlemno Stone and Cavalry in Anglo-Saxon England’,
Northern History, 29 (1993), 188-96.
Howarth, D., 1066: the year of the Conquest (New York, 1978).
Huard, G., ‘Quelques Lettres de Bénédictins Normands à Dom Bernard de
Montfaucon pour la documentation des Monumens de la Monarchie
Françoise’, Bulletin de la Société des Antiquaires de Normandie, xxxviii
(1913), 343-75.
Hudson, J., ‘The Abbey of Abingdon, its Chronicle and the Norman
Conquest’, ANS, XIX (1996), 181-202.
James, E.R., ‘The Battle of Hastings, 14th October, 1066’, The Royal
Engineers Journal, v (1907), 18-34, and map opposite 60.
Janssen, A.L., ‘La redécouverte de la Tapisserie de Bayeux’, Annales de
Normandie, 11 (1961), 179-95.
John, E., ‘Edward the Confessor and the Norman Succession’, EHR, xciv
(1979), 241-67.
Keats-Rohan, K.S.B., ‘William I and the Breton Contingent in the non-
Norman Conquest 1060-87’, ANS, XIII (1990), 157-72.
Ker, N.R., Catalogue of Manuscripts containing Anglo-Saxon (Oxford,
1957).
Keynes, S.D., ‘The Æthelings in Normandy’, ANS, XIII (1990), 173-205.
Kiff, J., ‘Images of War: Illustrations of Warfare in Early Eleventh-Century
England’, ANS, VII (1984), 177-94.
King Harold II and The Bayeux Tapestry, ed. G.R. Owen-Crocker
(Woodbridge, 2005)
Körner, S., The Battle of Hastings, England, and Europe 1035-1066 (Lund,
1964).
Lake, R.D. and Shephard-Thorn, E.R., Geology of the country around
Hastings and Dungeness: Memoir for 1:50:000 geological sheets 320
and 321 (England and Wales) (London, 1987).
Lancelot, A., see The Bayeux Tapestry under Primary Sources.
Lappenberg, J.M., A History of England under the Anglo-Saxon Kings,
trans. B. Thorpe (London, 1845).
Larson, L.M., The King’s Household in England before the Norman
Conquest (Madison, 1904).
Lawson, M.K., Cnut: the Danes in England in the Early Eleventh Century
(Harlow, 1993); reprinted as Cnut: Englamd’s Viking King (Stroud,
2011}. Page references here are to the original.

301
 THE BATTLE OF HASTINGS

Lawson, M.K., ‘Observations upon a Scene in the Bayeux Tapestry, the

Battle of Hastings and the Military System of the Late Anglo-Saxon
State’, in The Medieval State, ed. Maddicott and Palliser, pp. 73-91.
Lemagnen, S., ‘The Hidden Face of the Bayeux Tapestry’, The Bayeux
Tapestry: New Approaches, ed. Lewis, pp. 37-43.
Lemmon, C.H., ‘The Campaign of 1066’, in D. Whitelock et al., The
Norman Conquest: its setting and impact (London, 1966), pp. 77-122.
Lemmon, C.H., The Field of Hastings (4th edn., St Leonards-on-Sea, 1970).
Lewis, C.P., ‘The Earldom of Surrey and the date of Domesday Book’,
Historical Research, lxiii (1990), 329-36.
Lewis, C.P., ‘The French in England before the Norman Conquest’, ANS,
XVII (1994), 123-39.
Lewis, S., The Rhetoric of Power in the Bayeux Tapestry (Cambridge,
1999).
Leyser, K.J., ‘Early Medieval Warfare’, Communications and Power in
Medieval Europe: The Carolingian and Ottonian Centuries, ed. T. Reuter
(London, 1994), pp. 29-71.
Lingard, J., The History of England from the First Invasion of the Romans
to the Accession of William and Mary in 1688 (6th edn., London, 1854).
Lot, F., L'Art Militaire et les Armées au Moyen Age en Europe et dans le
Proche Orient (2 vols., Paris, 1946).
Loud, G.A., The Age of Robert Guiscard: Southern Italy and the Norman
Conquest (Harlow, 2000).
Lower, M.A., ‘On the Battle of Hastings’, Sussex Archæological
Collections, vi (1853), 15-40.
Maitland, F.W., Domesday Book and Beyond (Cambridge, 1897).
Mann, J., ‘Arms and Armour’, The Bayeux Tapestry, ed. Stenton, pp. 56-69.
Maund, K.L., ‘The Welsh Alliances of Earl Aelfgar of Mercia and his
Family in the mid-Eleventh Century’, ANS, XI (1988), 181-90.
Mawer, A. and Stenton, F.M., The Place-Names of Sussex (English Place-
Name Society, vols. vi-vii, Cambridge, 1930),
The Medieval State: Essays presented to James Campbell, ed. J.R.
Maddicott and D.M. Palliser (London, 2000).
Montfaucon, B. de, see The Bayeux Tapestry under Primary Sources.
Morillo, S., ‘Hastings: An Unusual Battle’, The Haskins Society Journal, 2
(1990), 95-104; reprinted The Battle of Hastings , ed. Morillo, pp. 220-
27. References here are to the reprint.
Morillo, S., Warfare under the Anglo-Norman Kings 1066-1135
(Woodbridge, 1994).

302
 SELECT BIBLIOGRAPHY

Morillo, S., ‘Milites, Knights and Samurai: Military Technology,

Comparative History, and the Problem of Translation’, The Normans and
Their Adversaries at War, ed. Abels and Bachrach, pp. 167-84.
Morton, C., ‘Pope Alexander II and the Norman Conquest’, Latomus, 34
(1975), 362-82.
Muntz, H., The Golden Warrior: the story of Harold and William (London,
1948). A novel.
Neumann, J., ‘Hydrographic and Ship-Hydrodynamic Aspects of the
Norman Invasion, AD 1066’, ANS, XI (1988), 221-43.
Nip, R., ‘The Political Relations between England and Flanders (1066-
1128)’, ANS, XXX (1998), 145-67.
Norgate, K., ‘The Battle of Hastings. Part II’, EHR, ix (1894), 41-76, 608-
11.
The Normans and Their Adversaries at War: Essays in Memory of C.
Warren Hollister, ed. R.P. Abels and B.S. Bachrach (Woodbridge, 2001).
Oger, B., ‘La tapisserie de Bayeux: le bilan de l’expertise scientifique
(1982-1983), La Tapisserie de Bayeux, ed. Bouet, pp. 117-23.
Oleson, T.J., ‘Edward the Confessor’s Promise of the Throne to Duke
William of Normandy’, EHR, lxxii (1957), 221-228.
Oman, C.W.C., The Art of War in the Middle Ages: A.D. 378-1515
(Oxford, 1885).
Oman, C.W.C., A History of the Art of War: The Middle Ages from the
Fourth to the Fourteenth Century (London, 1898).
Oman, C.W.C., A History of the Art of War in the Middle Ages (2nd edn., 2
vols., London, 1924).
Ordnance Survey, Pathfinder Series, Sheet 1290 (TQ61/71), at 2½ inches to
1 mile, Battle & Herstmonceux.
Ordnance Survey, Explorer Series, Sheet 124, at 2½ inches to 1 mile,
Hastings & Bexhill.
Ordnance Survey, see also British Geological Survey.
Orlandi, G., ‘Some Afterthoughts on the Carmen de Hastingae Proelio’, in
Media Latinitas: A Collection of Essays to Mark the Occasion of the
Retirement of L.J. Engels, ed. R.I.A. Nip et al., (Turnhout, 1996), pp.
117-27.
Owen-Crocker, G.R., ‘The Interpretation of Gesture in the Bayeux
Tapestry’, ANS, XXIX (2006), 145-73.
Owen-Crocker, G.R., ‘Behind the Bayeux Tapestry’, The Bayeux Tapestry:
New Interpretations, ed. Foys, pp. 119-29.

303
 THE BATTLE OF HASTINGS

Palmer, J.J.N., ‘The Conqueror's Footprints in Domesday Book’, The

Medieval Military Revolution, ed. A. Ayton and J.L. Price (London,
1995), pp. 23-44.
Palmer, J.J.N., ‘War and Domesday Waste’, Armies, Chivalry and Warfare,
ed. Strickland, pp. 256-75.
Partner, N.F., Serious Entertainments: the Writing of History in Twelfth-
Century England (Chicago, 1977).
Pastan, E.C. and White, S.D., ‘Problematizing Patronage: Odo of Bayeux
and the Bayeux Tapestry’, The Bayeux Tapestry: New Interpretations, ed.
Foys, pp. 1-24.
Peirce, I., ‘Arms, Armour and Warfare in the Eleventh Century’, ANS, X
(1987), 237-57.
Pryce, T.D., ‘A Note on the Bayeux Tapestry’, The Antiquary, xliii (1907),
346-7.
Ramsay, J.H., The Foundations of England (Oxford, 1898).
Ratkowitsch, C., Descriptio Picturae: Die literarische Funktion der
Beschreibung von Kunstwerken in der lateinischen Großdichtung des 12.
Jahrhunderts (Vienna, 1991).
Renaudeau, O., ‘Problèmes d’interprétation du costume d’après la Broderie
de Bayeux’, La Tapisserie de Bayeux, ed. Bouet, pp. 237-59.
Renn, D., ‘Burhgeat and gonfanon: two sidelights from the Bayeux
Tapestry’, ANS, XVI (1993), 177-98.
Renn, D., ‘How Big is It - and Was It?’, The Bayeux Tapestry: New
Approaches, ed. Lewis, pp. 52-59.
Rodger, N.A.M., ‘Cnut’s Geld and the Size of Danish Ships’, EHR, cx
(1995), 392-403.
Rodger, N.A.M., The Safeguard of the Sea: A Naval History of Britain.
Volume One 660-1649 (London, 1997).
Round, J.H., Anonymous article in the Quarterly Review (Vol. 175, No.
349), July 1892, 1-37, reviewing E.A. Freeman, The History of the
Norman Conquest of England and The History of William Rufus.
Round, J.H., Anonymous article in the Quarterly Review (Vol. 177, No.
353), July 1893, 73-104, reviewing Poème adressé à Adèle, fille de
Guillaume le Conquérant, par Baudri, Abbé de Bourgeuil, ed. L. Delisle.
Round, J.H., ‘Wace and his Authorities’, EHR, viii (1893), 677-83.
Round, J.H., ‘Mr. Freeman and the Battle of Hastings’, EHR, ix (1894),
209-260.
Round, J.H., Feudal England (London, 1895).
Round, J.H., ‘La bataille de Hastings’, Revue Historique , 65 (1897), 61-77.

304
 SELECT BIBLIOGRAPHY

Round, J.H., ‘The Battle of Hastings’, Sussex Archæological Collections,

xlii (1899), 54-63.
Sansonetti, V., see the Bayeux Tapestry under Primary Sources.
Sauvage, R.N., L'abbaye de Saint-Martin de Troarn (Caen, 1911).
Searle, E., Lordship and Community: Battle Abbey and its Banlieu 1066-
1538 (Toronto, 1974).
Smail, R.C., Crusading Warfare 1097-1193 (2nd edm., Cambridge, 1995).
Spatz, W., Die Schlacht von Hastings (Berlin, 1896).
Stenton, F.M., William the Conqueror (New York & London, 1908).
Stenton, F.M., Article on J.H. Round in Dictionary of National Biography
1922-1930, ed. J.R. Weaver (Oxford, 1937).
Stenton, F.M., ‘The Historical Background’, The Bayeux Tapestry, ed.
Stenton, pp. 9-24.
Stenton, F.M., Anglo-Saxon England (3rd edn., Oxford, 1971).
Stephens, W.R.W., The Life and Letters of Edward A. Freeman (2 vols.,
London, 1895).
Stevenson, W.H., ‘Senlac and the Malfossé’, EHR, xxviii (1913), 292-303.
Stewart, I., ‘Coinage and recoinage after Edgar’s reform’, Studies in Late
Anglo-Saxon Coinage, ed. K. Jonsson (Stockholm, 1990), pp. 455-85.
Stothard, C.A., see the Bayeux Tapestry under Primary sources.
Stothard, C.A., ‘Some Observations on the Bayeux Tapestry’,
Archaeologia, xix (1821), 184-91; reprinted The Study of the Bayeux
Tapestry, ed. Gameson, pp. 1-6. References here are to the original.
Stothard, Mrs. Charles, Letters written during a Tour through Normandy,
Brittany, and other parts of France, in 1818... (London, 1820).
Strickland, M., ‘Military Technology and Conquest: the Anomaly of Anglo-
Saxon- England’, ANS, XIX (1996), 353-82.
Strickland, M., War and Chivalry. The Conduct and Perception of War in
England and Normandy, 1066-1217 (Cambridge, 1996).
Stroud, G. and Kemp, R.L., Cemeteries of St Andrew, Fishergate (York,
1993).
Studies in Medieval History presented to R. Allen Brown, ed. C. Harper-Bill
et al. (Woodbridge, 1989).
The Study of the Bayeux Tapestry, ed. R. Gameson (Woodbridge, 1997).
Swanton, M.J., ‘King Alfred’s ships: text and context’, Anglo-Saxon
England, 28 (1999), 1-22.
La tapisserie de Bayeux: oeuvre d’art et document historique, ed. L. Musset
(1989).

305
 THE BATTLE OF HASTINGS

La Tapisserie de Bayeux: l’art de broder l’Histoire. Actes du colloque de

Cerisy-la-Salle (1999), ed. P. Bouet et al. (Caen, 2004). The French text
of The Bayeux Tapestry: Embroidering the facts of History, ed. Bouet.
Articles by English-speaking scholars are best read in the English
version, those by French-speakers in the French.
Tetlow, E., The Enigma of Hastings (London, 1974).
Thierry, A., Histoire de la Conquête de L’Angleterre par les Normands (5th
edn., Paris, 1838).
Thomson, R.M., William of Malmesbury (Woodbridge, 1987).
Tilliette, J-Y., ‘La chambre de la comtesse Adèle: savoir scientifique et
technique littéraire dans le c. CXCVI de Baudri de Bourgeuil’, Romania,
102 (1981), 145-71.
Tilliette, J.-Y., ‘Culture classique et humanisme monastique: les poèmes de
Baudri de Bourgeuil’, La Litterature Angevine Médiévale: actes du
colloque du samedi 22 mars 1980 (Centre de Recherche de Littérature et
de Linguistique de l’Anjou et des Bocages, Université d’Angers, 1981),
pp. 77-88.
Tsurushima, H., ‘Hic Est Miles: Some images of three knights: Turold,
Wadard and Vital’, The Bayeux Tapestry: New Approaches, ed. Lewis,
pp. 81-95.
Turner, D., Account of a Tour in Normandy (2 vols, London, 1820).
Van Houts, E.M.C., ‘The adaptation of the Gesta Normannorum Ducum by
Wace and Benoît’, Non Nova, Sed Nove: Mélanges de civilisation
médiévale dédiés à Willem Noomen, ed. M. Gosman and J. van Os
(Groningen, 1984), pp. 115-24.
Van Houts, E.M.C., ‘The Ship List of William the Conqueror’, ANS, X
(1987), 159-83.
Van Houts, E.M.C., ‘Latin poetry and the Anglo-Norman court 1066-1135:
The Carmen de Hastingae Proelio’, Journal of Medieval History, 15
(1989), 39-62.
Van Houts, E.M.C., ‘The Norman Conquest through European Eyes’, EHR,
cx (1995), 832-53.
Van Houts, E.M.C., ‘The Memory of 1066 in Written and Oral Traditions’,
ANS, XIX (1996), 167-79.
Van Houts, E.M.C., ‘Wace as Historian’, Family Trees and the Roots of
Politics: The prosopography of Britain and France from the tenth to the
twelfth century, ed. K.S.B. Keats-Rohan (Woodbridge, 1997), pp. 103-32.
Verbruggen, J.F., The Art of Warfare in Western Europe during the Middle
Ages from the Eighth Century to 1340, trans. Col. S. Willard and Mrs.
R.W. Southern (Woodbridge, 1997).

306
 SELECT BIBLIOGRAPHY

Vidler, J. (published anonymously), Gleanings Respecting Battel and its

Abbey (Battel, n.d.).
Walker, I.W., Harold: The Last Anglo-Saxon King (Stroud, 1997).
Walker, S., ‘A Context for ‘Brunanburh’?’, Warriors and Churchmen in the
High Middle Ages: Essays presented to Karl Leyser, ed. T. Reuter
(London, 1992), pp. 21-39.
Wenham, P., ‘Seven Archæological Discoveries in Yorkshire: 3. Human
skeletons, Riccall Landing’, The Yorkshire Archæological Journal, xl
(1959-62), 301-7.
White, G.H., ‘The Battle of Hastings and the Death of Harold’, The
Complete Peerage, vol. XII, part I (London, 1953), Appendix L.
Whitelock, D., et al., The Norman Conquest (London, 1966).
Williams, A., The English and the Norman Conquest (Woodbridge, 1995).
Wolf, K.B., Making History: the Normans and Their Historians in
Eleventh-Century Italy (Philadelphia, 1995).
Wormald, F., ‘Style and Design’, The Bayeux Tapestry, ed. Stenton, pp. 25-
36.
Wormald, P., ‘James Campbell as Historian’, The Medieval State, ed.
Maddicott and Palliser, pp. xiii-xxii.

307
 INDEX
Illus. refers to the illustrations on pp. 219-64; GT to the numbers of the
Genealogical Tables, pp. 215-17. Abbreviations: abb = abbot, abp =
archbishop, bp = bishop, br = brother, ct = count, d = daughter, dk = duke, k
= king, q = queen. The characters Æ and æ have been indexed as AE and ae.
Abingdon Abbey, 52, 189 Alfred, br of Edward the Confessor, 22,
Absalon, abp of Lund, 129 25n, 53, 56, 133n7, 135n4, 139, 154;
Achilles, 83, 88, 90, 174, 189n2 GT1
Adam of Bremen, 35n5, 177 Alfred, k of Wessex, 104-5, 128-9, 131-2,
Adela, d of Dk William, 73 134, 139, 173
and Adelae Comitissae, 88-9, 91 and burhs, 107, 199
Ælfgar, s of Leofric, earl of Mercia, 30, and cakes, 17-18, 103
111, 129 and navy, 108
Ælfgifu of Northampton, 22n; GT1 Amatus of Montecassino, 92 and n1, 183
Ælfric, killed at Hastings, 128n1, 189 Angers, France, 88
Ælfwig, abb of New Minster, Anglo-Saxon Chronicle, 53, 60, 62-3,
Winchester, killed at Hastings, 189 104
Ælfwold, bp of Crediton, 108 C text, 22n6, 52-3, 61n3
Aeneas, classical hero, 83, 90 D text, 22n6, 25n2, 52-4, 57, 112
Aesop, 66 E text, 22n6, 51-2, 57, 59, 61n3, 112
Æthelflæd, Lady of the Mercians, 110 Anjou, county of, 38
Æthelingadene, battle of, 126 men of in William’s forces, 142, 154
Æthelmær, monk of Malmesbury, 62 Anselm, abp of Canterbury, 53
Æthelred I, k of Wessex, 110, 135n4 Aquitanians, 91
Æthelred II (the Unready), k of England, in naval battle 1066, 33
21, 82, 101, 103-5, 126, 133, 139; in William’s forces, 85, 141, 147
GT1-2 Archer, T.A., 12 , 114
Æthelstan, k of England, 104-5, 109, 123, archers,
131, 173 English, 69, 134, 137, 145, 166
Æthelwold, nephew of K. Alfred, 137 French, 61, 68-9, 76, 84-5, 90, 99, 153,
Agamemnon, classical hero,149 168-70
Agincourt, battle of, 178 Arthur, k, 63
Ailricus, 33 Ashburnham, Sussex, 148
Aimeri of Thouars, 98, 141n3, 142 Ashdown, battle of, 110, 134-5, 173
Aire, river, 196 Assandun, battle of, 110, 188n2
Alan Fergant, 98
Alan I (Rufus), lord of Richmond, 64, Bachrach, B.S., 146
98n3, 142 Baldslow, Sussex, 163
Alexander II, pope, 39 Baldwin, abb of Bury St Edmunds, 148,
Alexander, bp of Lincoln, 60, 63 189
Alexander the Great, 90

309
 THE BATTLE OF HASTINGS

Baldwin V, ct of Flanders, 31, 33, 38, name of Senlac, 49-51, 93, 114, 115n1,
105, 179n2 123
Baring, F.H., 19, 116-19, 130, 144, 229 Starr’s Green, 159, 165, 167, Illus. 8
map of battle, 117 Stew Ponds, 48, 124, 165, Illus. 2, 8,
Barlow, F., 87, 102n1 41
Basing, battle of, 110 Stream flowing east, Illus. 62
Bath Abbey, 72 Stream flowing west, Illus. 8, 26-7,
Battle Abbey, 45, 94-5, 97, 147, 185n1 29-31, 40
buildings, Illus. 8-9, 41-3, 46, 49-50, Tithe Map, 47, Illus. 1.
53, 55 Upper Lake, 50, Illus. 8, 54-6
Battle, Sussex, Warren Wood, 48, 124, Illus. 8, 27, 31
aerial photograph, 46, Illus. 8 Weanyer’s Pond, Illus. 2
description of battlefield, 47-9 The Battle of Maldon, poem, 104
Caldbec Hill, 45n7, 159, 162-3, Illus. Baudri, abbot of Bourgeuil, 73, 83
67-8 on the battle, 88-92
Devil’s Plantation, 47n4 Bavent, Normandy, 148, Illus. 74
English Heritage survey, 49, 200-1, Bayeux, Normandy, 27, 70
Illus. 7 abbey of St Vigor, 204
Feature J, 200-1, Illus. 41-2, 44-5 cathedral, 70, 96
geology, 50-1, Illus. 6 Bayeux Tapestry, 64-73, 89, 91, 92n1,
gunpowder manufacture, 49, 200 104, 198
High Street, Illus. 52 Lancelot engravings, 266-72
hoar apple tree, 162 Montfaucon engravings, 274-83
Horselodge Plantation, 47, 49, 51, 121- on Halley’s Comet, 32, 67, 207
5, 130, 164, 166n4, 201, Illus. 8, on Harold’s death, 181-3, 204-13
32-34, 36-7, 44, 48 on Harold’s oath, 27-8, 67
iron extraction, 47-8 provenance, 65-6, 70-3
Long Plantation, 47, 48n3, Illus. 10, reliability, 67-70, 138, 170-1, 178-9
13-24 restorations, 64, 209-11
Lower Lake, 50-1, 130, Illus. 3, 8, 57- trees, 48
8 watercourse & hillock scene, 67, 70,
Malfosse incident, 14, 186-7 121-4, 130-1, 164, 166, 175-7
Manser’s Shaw, 187n3, Illus. 8 Becket, T., abp of Canterbury, 43
Marley Lane, 48, 49n2, Illus. 8, 61, Bede, historian, 53, 56, 58, 60
63-5, 68, 71, 73 Bedfordshire,
Mountain Plantation, 47n4, Illus. 1 levy of, 126n2
National School. 48, 49n2, 165, 167, Benedict IX, pope, 39
Illus. 60-1, 63-5 Benoît, Antoine, 122, 182-3, 204, 265
New Pond, 48-9, 51, 123-4, 130, 164, Benoît of St Maur, 96, 102n2
167, 177, 185n1, 200, Illus. 8, 38 Beorn Estrithson, earl, 35; GT3
Oakwood Gill, 187n3 Berkhamsted, Hertfordshire, 195-6
OS 6-inch map, 46, Illus. 5 Berkshire,
Powdermill, 48-9, 51, 124-5, 164-5, levy of, 125-6, 133
167, Illus. 8, 25 Black Prince, s of K Edward III, 178
railway station, 48, 49n2, Illus. 59 Blackhorse Hill, Sussex, 159, 165, 167
Sacristy Copse, 47, Illus. 1 Bonneville, Normandy, 27, 151n2
St Mary’s Church, Illus. 54, 69-70 Boulogne, county of,
Saxon Wood, 47, 120, 122, 200, Illus. men of in William’s forces, 98, 142
8, 19-23 Bourgeuil,
abbey of St Peter, 88

310
 INDEX

Bowley, Sussex, 148 Charlemagne, k of the Franks and

Brede estuary, Sussex, 159 emperor, 39, 70, 91, 98, 127-8, 132
Breme, killed at Hastings, 128, 189 Charles Martel, ruler of the Franks, 91
Bretons, Chester, 105, 189n2
in William’s forces, 38, 85, 98, 141-2, fortification arrangements, 108n1, 133
154, 168, 170 Chichester harbour, 152
Brevis Relatio, 96, 147 Chilterns, 139
on the battle, 94-5 Chrétien de Troyes, 63
Bristol, 111 Chronicon Monasterii de Hida, 102n2
Brittany, 59, 150 Churchill, Sir W., 21
William’s campaign of 1064, 27, 38, Cnut, k of England, Denmark,
154 Norway and part of Sweden, 21-2, 35,
Brunanburh, battle of, 104, 123, 173, 70, 82, 105, 110-11, 129, 173, 188n2;
188n2 GT1-3
Budgen, R., 48n3, 50n6, 124, Illus. 2-3. Cnut, k of Denmark, son of K Swegen
Bulverhythe, Sussex, 148 Estrithson, 197
Burghal Hidage, 107-9, 199 coinage system, English, 105-6, 199
Burgundians, Colchester, Essex, 110
in William’s forces, 142 Coleman, monk of Worcester, 32n2
Bury St Edmunds Abbey, Conan II, ct of Brittany, 38, 66
men of killed at Hastings, 189 conrois, Norman cavalry units, 155
Byrhtnoth, ealdorman, 104, 128, 136 Constance FitzGilbert, 62-3
tapestry depicting his deeds, 72 Constantine the Great, 91
Cornwall,
Caen, Normandy, 38, 95, 151n2, 203 levy of, 137
abbey of St Stephen, 75, 82, 203 Couesnon, river, 66
Caesar, Julius, 58, 75, 83-4, 86, 90n1, Crécy, battle of, 178
139n5, 145, 149, 155, 158, 168-9 Crimean War, 144-5
Calais, France, 17 crossbowmen, 76, 90, 154, 168, 170
Caldbec Hill, Sussex, 45n7 Cuthbert, St., 71n2
Calixtus, St., 162 Cyrus, k of Persia, 90
Cambridgeshire,
levy of, 127 Dacia, 70
Campbell, J., 105-6 Daedalus, classical aviator, 62
Canterbury, Kent, 43, 66n1, 70, 86, Dagworth, Suffolk, 128
189n2, 195 Danelaw, 36, 126, 132
abbey of St Augustine, 51, 73n5 Davis, R.H.C., 77-8, 83, 86, 146, 170
cathedral, Illus. 76 Dee, river, 105
Carmen de Hastingae Proelio, see Guy Delbrück, H., 150
bp of Amiens Denmark,
Cartae Baronum, 149 troops from present at Hastings, 84
Catsfield, Sussex, 159 Derby, 110
cavalry, Derbyshire, 31
English, 37n , 111, 123n6, 137, 178 Derwent, river, 36
French (and role at Hastings), 68-9, Devon,
84-5, 95, 99, 155, 170-2, 176-8 levy of, 137
Cavendish, Suffolk, 128n1 Dibdin, T, 210-11
chansons de geste, 58, 61, 66n5, 97n2, Dijon, 73
169, 184 Dinan, Brittany, 66, 69, 170
Dives, Gulf of, 147-8

311
 THE BATTLE OF HASTINGS

Dives, river, 38, 147-8, 156, 169, Illus. 74 promise of throne to Swegen
Dives-sur-Mer, Normandy, 143, 146, Estrithson, 35
148-9 Edward the Elder, k of Wessex and
Dol, Brittany, 66, 88 Mercia, 104-5, 110, 132, 134, 137
Dover, Kent, 19, 24, 27, 77, 86, 155, 193, Edward the Exile, 26, 56
195-6 Edwin, earl of Mercia, 31-4, 40, 63, 82,
church of St Martin, 81 131, 190, 195-6; GT 4
Draco Normannicus, see Stephen of Ely Abbey, 72, 197
Rouen Emma, q, 21-2, 35, 82; GT1-2
Dreux, ct of the Vexin, GT1 Encomium Emmae, 22n1, 82
Dublin, men of, 35 Engels, L.J., 86
Dudo of St Quentin, 80 Engenulf of Laigle, 93, 187, 189
Durham, 71n2 English Heritage, 13, 15 ; see also under
Battle
Eadmer, monk of Canterbury, 29, 96 Eohric, Danish k, 137
on the battle, 53-4 Eric Bloodaxe, k of York, 36
Eadsige, abp of Canterbury, 23 Ermenfrid, bp of Sion, 44
Ealdred, abp of York, 40n , 52n3, 54-5, Ermold the Black, poet, 90-1
63, 195-6 Essex, 33
East Anglia, 104, 137, 190n1 levy of, 99, 104, 110, 128, 136
earldom of, 30 L’Estoire des Engleis, see Gaimar
kingdom of, 103 Estrith, sister of K Cnut, 22; GT2-3
levy of, 127 Eustace II, ct of Boulogne, 24, 77, 79, 85,
Eastbourne, Sussex, 148 87, 142, 174, 180, 186-7, 196 ; GT1
Edgar the Ætheling, s of Edward the Exeter, Devon, 154n6, 155, 196
Exile, 26, 195-6; GT1
Edgar, k of England, 105 Falaise, Normandy, 156n5
Edington, battle of, 134 Farnham, battle of, 134-5
Edith, q of K Edward the Confessor, Fécamp Abbey, Normandy, 79, 96, 147,
22n4, 23, 25, 63, 195n2; GT1, 4 149, 157-8
Edith, q of K Harold II, 32; GT4 First Crusade, 40, 71n1, 89, 142, 146, 192
Edith Swan-neck, handfast wife of K five-hide system, 125-8, 133
Harold II, 189 ; GT4 Flanders, county of, 26, 30n, 31, 33
Edmund Ironside, k, 26, 56, 110, 127, Flemings,
131, 173; GT1 in William’s forces, 141
Edric the Deacon, with Tostig, 142
killed at Hastings, 128n1, 189 Fontenay, abbey of, 99
Edward I, k of England, 105 Fordington, Dorset,
Edward III, k of England, 144, 178 church of St George, 71
Edward IV, k of England, 144 Forest of Dean, 110
Edward the Confessor, k of England, 72, Foucault, Anne, 203
105, 154; GT1, 4 Foucault, N-J, 203, 265, 273
and Harold’s oath, 26-30 Freeman, E.A., 12, 27, 113-14
attempt on the throne, 22 map of battle, 113
becomes king, 22-3, 83 on Harold’s oath, 27-8
death and deathbed, 31, 67 French,
exile in Normandy, 21 in William’s forces, 85, 141-2
grant of throne to Harold, 31-2 Fulcoius of Beauvais, 68n5
promise of throne to William, 23-32 Fulford, battle of, 34, 37, 102n1, 112,131,
163, 190, 200

312
 INDEX

Gaimar, Geoffrey, 101 Hannibal, Carthaginian general, 90

on the battle, 62-4 Harald, br of K Cnut, GT3
Galen, classical physician, 89 Harald Hardrada, k of Norway, 33-6, 40n,
Gaul, 41, 60, 63, 93, 139, 151, 156-7
men of in William’s forces, 141-2 Harfleur, Normandy, 144
Geoffrey Martel, ct of Anjou, 38, 156 Harold II, k of England, GT1, 4
Geoffrey of Monmouth, 63, 95 becomes earl of Wessex, 25, 30
George, St., 71 becomes king, 31-2, 67
Gesta Guillelmi, see William of Poitiers believed site of death, 45, Illus. 51
Gesta Normannorum Ducum, see William burial, 77, 86-7, 100, 188-9
of Jumièges coronation, 53-4, 94
Gesta Regum, see William of death, 77, 79-81, 90, 92, 98-100, 179-
Malmesbury 85, 204-12
Gettysburg, battle of, 102 Fighting Man banner, 39, 71, 94, 165
Giffard, see Robert and Walter military preparations 1066, 33-4, 112
Gilbert, archdeacon of Lisieux, 39 naval expedition 1066, 33-4 , 51
Gilfard, see Robert, son of Giffard role at Hastings, 58-9, 177-8, 191
Gillingham, J., 81, 156 size of his army at Hastings, 112-31,
Glastonbury Abbey, 56 163-7
Godfrey of Rheims, 83n3, 91n5 strategy in 1066, 41, 152, 191-2
Godgifu, d of K Æthelred II, 24, 196; visit to Normandy and oath, 26-30, 38,
GT1 65-6, 96, 158, 190
Godric, sheriff of Berkshire, visit to Northumbria 1066, 32
killed at Hastings, 189 Welsh campaign of 1063, 111
Godwin, earl of Wessex, 22-5, 53, 72, Harold Harefoot, k of England, 21-2; GT1
97, 133n7; GT1, 4 harrying, 106-7
Gregory VII, pope, 39 Harrying of the North, 196-7
Gruffydd, Welsh k, 30, 32n, 111 Harthacnut, k of England and Denmark,
Gunnhild, d of Cnut, 21, 101, 105 ; GT1 21-2, 24n, 35, 106-7, 129, 133n7; GT1
Gurney, H, 210 Hastings, Sussex (see also Battle), 37, 41,
Guy, bp of Amiens, 75, 83, 96 48-51, 63, 67, 80, 97, 144-5, 152n3,
authenticity of Carmen, 77-9 153, 156-9, 162-3
on the battle, 75-9 Hebrides, men of, 35
Guy, ct of Brionne, 37-8 Hector, classical hero, 83, 88, 90, 174,
Guy I, ct of Ponthieu, 26-7, 30, 65, 75, 189n2
205 Heidelburg, university of, 14
Gyrth, s of Godwin, earl of East Anglia, Helgaud of Fleury, 82
30, 40, 52, 54, 64, 67, 77, 92-3, 97-8, Helloc, son of, 77, 174
121, 124, 129, 174, 178-9, 181, 184; Henry I, k of England, 52n3, 55, 96, 150,
GT4 155, 169, 172
Gytha, wife of Earl Godwin, 40, 59, 77, Henry I, k of France, 38, 155-6
86, 88, 188-9 Henry I (the Fowler), k of Germany, 70
Henry II, k of England, 43, 52n3, 96,
Hailsham, Sussex, 148 102n2
Hakon, s of Swegen Godwinson, 25, 27; Henry III, emperor of Germany, 21, 39,
GT4 101, 105, 109; GT1
Halley’s Comet, 32-3, 62, 67, 80, 89, 97, Henry IV, emperor of Germany, 40
190 Henry V, k of England, 144, 178
Hampshire, 195n2 Henry of Huntingdon, 53, 62, 100-1
levy of, 126-7 on the battle, 60-2

313
 THE BATTLE OF HASTINGS

Henry, margrave of Neustria, 138-9, 166 Kent, 22, 33, 66, 73, 162, 195n2
Hereford, 30, 61, 111, 197 levy of, 98-9, 110, 137
Herefordshire, 111 Ketel, thegn, 128
heriots, 128, 133
Herleve, mother of William the La Londe, Dom R. de, 203-4, 207
Conqueor, GT 2 L’Archer, Dom M., 204, 207-8
Herluin, second consort of Herleve, GT 2 Lancelot, Antoine, 182-3, 203-4, 206,
Hertfordshire, 30, 195n2 208, 212; engravings of Bayeux
Hildebrand, see Gregory VII Tapestry, 266-72
Hill, David, 14, 64-5, 183n1, 211 Lanfranc, abp of Canterbury, 17, 75,
Hindenburg Line, 138 78n6, 94, 101
Hippocrates, classical physician, 89 Layamon, poet, 63
Historia Anglorum, see Henry of Le Mans, bp of, 147
Huntingdon Lemmon, C., 13
Hollister, C.W., 147, 149 Leofgar, bp of Hereford, 111
Holy River, battle of, 111 Leofric, abb of Peterborough, 189n5
Homer, classical poet, see also Iliad, 90 Leofric, earl of Mercia, 23-5, 30
Hooe, Sussex, 148 Leofwine, s of Godwin, earl, 30, 52, 54,
housecarls, 112-3, 128-30, 165, 167, 186 64, 67, 121, 124, 129, 178-9, 181, 184;
Hugh, bp of Lisieux, 82 GT4
Hugh II, ct of Ponthieu, 75 Leominster, 111
Hugh Margot, monk of Fécamp, 96-7, Lewes, Sussex, 162
149n3 priory of St Pancras, 102n2
Hugh of Montfort-sur-Risle, 77n1, 196 Lex Castrensis, 129
Hugh of Ponthieu, 77, 79, 142, 180-1, 186 Leyser, K.J., 138-9, 154
Hugh-Renard, bp of Langres, 83n3 Life of King Edward, 22n4
Humber, river, 24, 97-8 Lillebonne, Normandy, 151n2
hundreds, 106, 138 Lincoln, 196
Hungary, 26, 40 Lincolnshire, 31, 62
Huntingdon, 110 Lindsey, 33
London, 34, 40, 54, 77, 86, 94, 97, 99,
Iliad, 77n2, 88 126n2, 139, 144, 162, 195
Ilias latina, 83, 88n3 levy of, 98, 137
Inca empire, 198-9 St Paul’s Church, 94
Ingelheim, palace of, 70, 72, 90 siege of in 1016, 109
Ireland, 105 Lot, F., 149
Italian Normans, 141, 151 Louis the Pious, 90-1, 133n1
Lower, M.A., 43
James, E.R., 113, 118 Lucan, classical poet, 83, 90n1
James, Sir H., 113 Lux mores hodierna, see Godfrey of
Jephthah, k, 68n5 Rheims
Jersey, Normandy, 95
Jerusalem, 192 Maaseik, Belgium, 71n2
John of Salisbury, 123n7, 137 Macbeth, k of Scotland, 112
John of Worcester, 100 Magdeburg, palace of, 70, 72
on the battle, 53-5, 112 Magnus, k of Norway, 35; GT3
Judith, wife of Earl Tostig, 31 Magnus, s of Harald Hardrada, 35-6
Jumièges Abbey, Normandy, 24n3, 80-1 Magyars, 70
Justin, 86n3 Maine, county of, 38, 83, 150, 154
Juvenal, classical poet, 86n3 conquered by Dk William, 38, 83

314
 INDEX

men of in William’s forces, 85, 98, earldom of, 30

141-2, 154, 174 kingdom of, 36, 103
Malcolm III, k of Scotland, 56n1, 112, refusal to accept Harold as king, 32
197;GT1 revolt of 1065, 30-1, 52, 129
Maldon, Essex, Norway, 33-6, 59, 93, 105, 111-12
burh, 110 Nottingham, 196
battle of, 72, 104, 128, 135-6, 138 Nottinghamshire, 31
Malfosse, see Battle
Malmesbury, Wiltshire, 55 Odo, bp of Bayeux, 65-7, 70, 73, 82, 99,
Margaret, St., 52n3, 56n1; GT1 121, 147, 149, 153-4, 174, 178, 196,
Marianus Scotus, 34n3 205; GT2
Marmoutier Abbey, 45 Offa, Essex thegn, 128
Mary, q, 17 Offa, k of Mercia, 107
Matilda, wife of Dk William and q, 33, Offa’s Dyke, 107, 199
38, 75, 87, 204 Olaf, St, k of Norway, 35n, 139n1
Matilda, q of K Henry I, 55-6 Oliver, Frankish hero, 98, 100
Matilda, d of K Henry I and empress, Oman, C.W.C., 117, 119n2, 171
56n2 Orderic Vitalis, 29, 33, 36, 43, 49-50, 53,
McSween, J., 65n2, 183n1, 209n1, 211 75, 81-2, 100, 141-2
Mercia, 190n1 on the battle, 92-4
earldom of, 30 Orkney, men of, 35
kingdom of, 103 Orléans, France, 88
Meretun, battle of, 135n3, 173 Ottar the Black, poet, 188n2
Merlin, 63 Otto I, k of Germany and emperor, 105
Middlesex, 30, 195n2 Ouse, river, 34, 37
Mont St Michel, 205 Ovid, classical poet, 90
Montfaucon, Bernard de, 14, 182-3, 203- Ox’s Eye, reliquary, 94
4, 206-7, 212; engravings of Bayeux Oxford, 139
Tapestry, 274-83 Oxfordshire, 30, 144n4
Morcar, earl of Northumbria, 31-4, 36,
40, 63, 190, 195-7; GT4 Pallas, character in the Aeneid, 91n4
Morillo, S., 19n2, 173 Pancras, St.,
Mortemer, battle of, 38 reliquary of, 94n3
Mount Holyoke College, Massachusetts, Paris, France, 95, 210
65n2 Passenham, Northamptonshire, 110
Muntz, Hope, 21 Penhurst, Sussex, 148n2
Penitential, 44-5, 154, 169-70
Napoleon I, emperor, 28, 210 Penselwood, battle of, 110
Napoleon III, emperor, 28 Pepin I, k of the Franks, 91
naval warfare in 1066, Peter, St., 97
navy, English, 33, 105, 108-9, 112, 126, Peterborough Abbey, 51
133 Pevensey, Sussex, 37, 41, 54, 57, 67, 80,
Netherfield, Sussex, 148 144, 148, 152n3, 153, 156, 158-9, 161,
Nicholas, archdeacon of Huntingdon, 60 163
Norfolk, levy of, 127-8 Pevensey Bay, 148, 152, 159
Norgate, Kate, 12, 114 Phalaris, classical tyrant, 90
Northampton, 31 Pharsalus, battle of, 83, 90n1
Northamptonshire, 52 Philip I, k of France, 38
Northumberland, 30 Plus tibi fama, 83n3
Northumbria, 97, 112n1, 157, 190n1 Poitiers, France, 81

315
 THE BATTLE OF HASTINGS

Poitiers, battle of, 178 Robert, son of Giffard or Gilfard, 77, 79,
Poitou, county of, 142, 180
men of in William’s forces, 98, 141n3, Rochester, Kent, 159n1, 162
142 Rodger, N.A.M., 108
Pompey, Roman general and politician, Roger of Beaumont, 85
83, 188n3 Roger of Montgomery, 98-9
Pontefract, Yorkshire, 196 Roger, earl of Hereford, 197
Ponthieu, county of, 26, 29-30 Roland, Frankish hero, 41, 98, 100
location, 26 Rollo, dk of Normandy, 95
men of in William’s forces, 142 Roman de Rou, see Wace
Poole harbour, 152 Rome, 23, 105
Portsmouth, 152 Romney, Kent, 86, 153, 195
Poux, county in Picardy, Romulus, classical hero, 90
men of in William’s forces, 98 Roncesvalles, battle of, 98
Priam, k of Troy, 88 Rouen, Normandy, 27n
Prudentius, classical Christian poet, 90n1 Round, J.H., 12, 19, 50, 113-14

Ralph, abb of Battle, 94-5 St Brice’s Day Massacre, 101

Ralph, ct of Amiens, 79 St Évroul Abbey, Normandy, 92-3, 187n3
Ralph II Briquessart, vicomte of Bayeux, St Maixent Abbey, Poitou, 141
at Tinchebrai, 172 St Omer, Flanders, 22n
Ralph, earl of East Anglia, 197 St Stephen’s, Caen, see Caen, abbey of St
Ralph, earl of Hereford, 111, GT1 Stephen
Ralph FitzGilbert, 62 St Valéry, Ponthieu, 61, 75, 80, 96-7, 101,
Ramsay, Sir J.H., 19, 115-16, 130, 144 146-8, 151-2, 153n1
map of battle, 116 Sallust, Roman historian, 83-4
Reading, battle of, 110 Sandwich, Kent, 33, 35, 51, 139
Remus, classical hero, 90 Sansonetti, V., 65n2
Rhuddlan, 111 Saxons, continental, 39, 91
Riccall Landing, Yorkshire, 37, 180, Scolland, abb of St Augustine’s,
188n3 Canterbury, 73n5
Richard, ct of Evreux, 149 Scotland, 33, 105, 111-12, 129, 197
Richard I, dk of Normandy, GT2 Sedlescombe hoard, 162
Richard II, dk of Normandy, 37-8, 154 ; Senlac, see Battle
GT2 Serlo of Bayeux, French poet, 73n4
Richard III, dk of Normandy, GT2 Sherston, battle of, 91n4, 110, 173
Richmond, Yorkshire, 64 shield-wall, 57, 67, 69, 114-5, 117, 133-
Robert, abp of Rouen, 37 7, 165-7, 171, 173, 185, 198
Robert, ct of Eu, 38 Ship-List of William the Conqueror, 146-
Robert I, dk of Normandy, 22, 37, 152, 7, 151
154; GT2-3 shires, 106
Robert II, dk of Normandy, 96, 150, 172 Shrewsbury, 92
Robert the Pious, k of France, 82 Sicily,
Robert of Beaumont, 85, 141n2, 168 men of in William’s forces, 141n1
Robert of Bellême, 172 Sigebert of Gembloux, chronicler and
Robert of Jumièges, abp of poet, 101
Canterbury, 23, 24n3, 25 Sigmund, Norse hero, 70, Illus. 75
Robert of Mortain, 147 ; GT2 Siward, earl of Northumbria, 23-5, 30,
Robert, son of Erneis, 99 111-12, 129, 197
Snorri Sturluson, Icelandic saga writer, 34

316
 INDEX

Somme, battle of, 102 Tostig, s of Godwin, earl of Northumbria,

Somme, river, 147 30-4, 36, 61, 63, 93, 111, 129, 142,
Southampton, 22, 152 180, 190, 200; GT4
Spartan hoplites, 136 Towcester, Northamptonshire, 110
Spatz, W., 19, 115, 117, 119, 135, 143-5, Towton, battle of, 150, 180, 186n6
148, 150, 198 Trajan’s Column, 70
Stamford Bridge, battle of, 36-7, 52, 57, Tribal Hidage, 149
63, 92-3, 102n1, 112, 131, 137n2, 139, Troy, 89
163, 173, 180, 186, 188, 190-1, 200 Turner, D., 210
Statius, Roman poet, 75, 83, 90n1 Turnus, classical hero, 83, 90
Stenton, Sir F.M., 28, 119 Turold, tenant of Odo of Bayeux, 66,
Stephen, ct of Blois, 88-9 73n5
Stephen of Rouen, 102n2 Tyne, river, 34
Stevenson, W.H., 50 Tytherley, Hampshire, 127
Stigand, abp of Canterbury, 25, 54, 128
Stiklestad, battle of, 35n Ulf, earl, GT3
Stonehenge, Wiltshire, 199
Stothard, Charles, 65n2, 206-7, 210, 213 Val-ès-Dunes, battle of, 38, 156
Stothard, Mrs Charles, 207 Van Houts, E.M.C., 78, 81, 99, 147
Suetonius, Roman historian, 58, 83 Varaville, battle of, 38, 156, 169
Suffolk, Vegetius, late Roman author, 136, 150
levy of, 128 Vikings, 72, 103-4, 110, 126, 128, 154,
Surrey, 166, 173
levy of, 110 military methods, 132, 138-9
Sussex, 40, 54, 195n2 Virgil, Roman poet, 75, 83-4, 90
Swegen, s of Cnut, 21-2, 35n GT1 Vital, tenant of Odo of Bayeux, 66, 73n5
Swegen, s of Godwin, 25; GT4
Swegen Estrithson, k of Denmark, 35-6, Wace, 63, 114, 125, 129, 137, 142, 146-
40, 105, 109, 196-7 ; GT3 8, 161, 166, 169, 174, 176-8, 185, 200-
promised the English throne, 35 1, 206
Swegen Forkbeard, k of England and account of the battle, 95-101
Denmark, 35 ; GT3 Wadard, tenant of Odo of Bayeux, 66,
73n5
Tadcaster, Yorkshire, 36 Wallingford, Oxfordshire, 107, 129, 195
Taillefer, 61-2, 64, 76, 78, 98, 168-9 Walter I Giffard, 77n1, 147
Telham, Sussex, 159 Waltham Holy Cross, Essex, 100, 189
Tempsford, Bedfordshire, 110 Waltheof, earl, 129, 197
Thebes, 83, 90n1 Wareham, Dorset, 107
Theodosius the Great, 91 Wartling, Sussex, 148
Thorkell, earl, 91n4 Waterloo, battle of, 28, 102, 113, 171,
Thouars, 172n1
men of in William’s forces, 142 Weald forest, 139, 162, 186n6
Thurkill, killed at Hastings, 189 Webster, Sir G., 185n1, Illus. 4
Thurstan, son of Rollo, 93, 98 Westminster, 32
Tillières, Normandy, 154 Westminster Abbey, 196
Tinchebrai, battle of, 95, 150, 155, 169, Wight, Isle of, 33-4, 152
172 Wigingamere, 110
Tolkien, J.R.R., 123n6 Wilkin, F., 185n1
William, son of Richard, ct of Evreux,
149

317
 THE BATTLE OF HASTINGS

William, ct of Mortain, 172 York, 31-2, 34, 36, 40n, 90n2, 156,
William, dk of Normandy, k of England, 186n6, 196
GT2 St Andrew, Fishergate, 200
and Earl Tostig, 33 Yorkshire, 30
career prior to 1066, 37-9
coronation, 65, 79-80, 86, 196 Zedekiah, k, 68n5
crossing of English Channel, 152-3
papal banner, 39-41, 45, 84, 97-8, 142,
190
promised the English throne, 23-30
role at Hastings, 58-9, 76-7, 85, 90, 94-
5, 100, 168-9, 191-3
size of his army, 115-18, 143-51
size of his fleet, 146-8
strategy in 1066, 36, 41, 139, 151-2,
156-7, 159, 190
tomb of, 203
visit to England 1051, 21, 25
William II (Rufus), k of England, 126
William FitzOsbern, 61, 98, 147
William of Jumièges, 91-2, 96
on the battle, 80-1
revision by OV, 92-3.
William Malet, 86, 188
William of Malmesbury, 53, 62-3, 81
on the battle, 55-60, 112
William of Poitiers, 57, 81-2
on the battle, 82-8
possible use of administrative
documents, 149
Willingdon, Sussex, 148
Wilton, battle of, 173
Wiltshire,
levy of, 137
Winchester, Hampshire, 126n2, 195n2
abbey of New Minster, 189
Old Minster, 70, 73, Illus. 75
Windsor, Berkshire, 61
Winfarthing, Norfolk, 127-8
Worcester, 52, 55, 94n1, 106
Worcestershire,
harrying of 1041, 106-7, 129
levy of, 131
Wulfnoth, s of Godwin, 25, GT4
Wulfstan II, bp of Worcester, 32, 58-9
Wulfwaru, 72

Xenophon, Greek historian, 136

Xerxes, k of Persia, 84

318