Anglo-Norman Studies XXVI: Proceedings of the Battle Conference 2003 [26] 1843830728, 9781843830726, 9781846152054

Table of contents :
CONTENTS
......Page 6
EDITOR’S PREFACE......Page 8
ABBREVIATIONS......Page 10
The Norman Conquest and the Media......Page 14
Dudo of St Quentin and Norman Military Strategy c.1000......Page 34
Clergy in the Diocese of Hereford in the Eleventh and Twelfth Centuries......Page 50
England and the Irish-Sea Zone in the Eleventh Century......Page 68
Les abbés bénédictins de la Normandie ducale......Page 88
TheVita Ædwardi Regis: The Hagiographer as Insider......Page 100
The Warenne View of the Past, 1066–1203......Page 116
1088 – William II and the Rebels......Page 152
The Anglo-Norman Civil War of 1101 Reconsidered......Page 172
Epic and Romance in the Chronicles of Anjou......Page 190

Citation preview

ANGLO-NORMAN STUDIES 26 Proceedings of the Battle Conference 2003 John Gillingham Anglo-Norman Studies

ANGLO-NORMAN STUDIES XXVI PROCEEDINGS OF THE BATTLE CONFERENCE 2003

With one exception, the papers in this volume of Anglo-Norman Studies range in time from c.1000 to c.1200 and in place from Anjou to the Irish Sea zone. The exception, the Allen Brown memorial Lecture, surveys images of the Normans in historical and imaginative literature from the twelfth to the twentieth century. Other studies of narrative sources focus on Dudo of St Quentin’s history of the Norman dukes, on the Vita Ædwardi Regis, on the Fenland monastic chronicles, and on the chronicles of Anjou. Tombstone, charters and a chronicle are used to elucidate the Warenne view of the past. Two papers analyse both charter and chronicle evidence in re-considerations of the succession disputes following the deaths of William I and William II. In addition there are papers on the clergy of the diocese of Hereford, on Norman Benedictine abbots, and on Anglo-Irish relations in the eleventh century.

ANGLO-NORMAN STUDIES XXVI PROCEEDINGS OF THE BATTLE CONFERENCE 2003

Edited by John Gillingham

THE BOYDELL PRESS

© by contributors 2003, 2004 All Rights Reserved. Except as permitted under current legislation no part of this work may be photocopied, stored in a retrieval system, published, performed in public, adapted, broadcast, transmitted, recorded or reproduced in any form or by any means, without the prior permission of the copyright owner First published 2004 The Boydell Press, Woodbridge ISBN 0 84383 072 8

ISSN 0954–9927 Anglo-Norman Studies (Formerly ISSN 0261–9857: Proceedings of the Battle Conference on Anglo-Norman Studies)

The Boydell Press is an imprint of Boydell & Brewer Ltd PO Box 9, Woodbridge, Suffolk IP12 3DF, UK and of Boydell & Brewer Inc. 668 Mt. Hope Avenue, Rochester, NY 14620, USA website: www.boydellandbrewer.com A catalogue record for this series is available from the British Library Library of Congress Catalog Card Number: 89–646512

This publication is printed on acid-free paper Printed in Great Britain by Anthony Rowe Ltd, Chippenham, Wiltshire

CONTENTS EDITOR’S PREFACE ABBREVIATIONS

R. Allen Brown Memorial Lecture The Norman Conquest and the Media Richard Barber Dudo of St Quentin and Norman Military Strategy c.1000 Bernard S. Bachrach Clergy in the Diocese of Hereford in the Eleventh and Twelfth Centuries Julia Barrow England and the Irish-Sea Zone in the Eleventh Century Clare Downham Les abbés bénédictins de la Normandie ducale Véronique Gazeau TheVita Ædwardi Regis: The Hagiographer as Insider J. L. Grassi The Warenne View of the Past, 1066–1203 Elisabeth van Houts Textual Communities in the English Fenlands: A Lay Audience for Monastic Chronicles? Jennifer Paxton 1088 – William II and the Rebels Richard Sharpe The Anglo-Norman Civil War of 1101 Reconsidered Neil Strevett Epic and Romance in the Chronicles of Anjou Neil Wright

vii ix 1 21 37 55 75 87 103 123

139 159 177

EDITOR’S PREFACE The twenty-sixth Battle Conference, held in Pyke House and Battle Abbey from 24 to 28 July 2003, allowed us to celebrate the publication, timely as always, of the conference’s twenty-fifth annual volume of proceedings. To mark this milestone the conference opened with a reception on the terrace of Battle Abbey overlooking the battlefield; we then adjourned to the Library of Battle Abbey to listen to the R. Allen Brown Memorial Lecture delivered by Richard Barber. To him, as managing director of Boydell & Brewer, the conference’s debt had been incalculable, going back to the very beginning when, according to Allen Brown’s preface to the first (1978) volume, he ‘went far beyond a publisher’s call of duty in seeing this volume through the press’. There followed a special dinner in the magnificent Abbot’s Hall. From 1979 to 1998 the opening reception had always been held in the Abbot’s Hall, so we are very grateful indeed to Roger Clark, headmaster of Battle Abbey School, for giving us permission to re-invent and improve tradition. Thanks are also due to Jackie Perry for her help in organising a memorable evening. In addition to speakers and delegates, among those present, by invitation of the trustees of the R. Allen Brown Memorial Trust, were Mrs Vivian Brown, representatives of the Battle and District Historical Society and of the Battle Local History Museum, as well as some of those who have played key roles in the Conference’s history; in particular Gillian Murten, without whose help the conference might not have survived its early years (see Editor’s prefaces to volumes 1 to 10), and Julie Walker, Principal of Hastings College of Arts and Technology. The publication, in volume 25, of the proceedings of a conference held in Glasgow, reminded me very forcefully just how much is owed four years out of every five to the facilities and administrative backup provided by Hastings College and Pyke House. We also broke with 25 years of tradition. As an experiment, the conference ran from Thursday evening to Monday lunchtime instead of from Friday to Wednesday. This meant a shorter programme. The operation of sod’s law then ensured that on the eve of the conference a speaker fell ill and was unable to attend. One of our delegates, John Grassi, stepped into the breach. Not only had he brought a paper with him, but he was willing to undertake the last minute alterations needed to make it fit the Battle format. Then it turned out that two of our speakers were regrettably, and for very different reasons, unable to deliver written versions of their papers. Happily Richard Sharpe, whose paper given to the 2000 conference could not be published then, was kind enough to help out now. I am very grateful to both of them. Without their help, this slimmer than usual volume would have been even slimmer. This year’s conference outing took us to Bosham (where we did not discover King Harold’s bones), to Boxgrove Priory and to St Mary de Haura, Shoreham. Not even pouring rain (another break with Battle tradition!) could dampen the enthusiasm with which Tim Tatton-Brown, aided by Carol Davidson Cragoe, guided us around these three churches. It is on days such as this that Ian Peirce’s sterling qualities, without which no Battle Conference would be complete, are most in evidence. Thanks to the kindness of Anne Ainsley, we were also given the opportunity of a private view of Battle Museum of Local History in its fine new quarters. Pyke House and its staff, with a little help from The Chequers next door, made us

feel that it was good to be back in the Conference’s real home. For all their work and skill I am particularly grateful to Alison Martin and Bob Banks. For their speedy and efficient work in getting volume 26 into print on schedule my thanks as always go to Caroline Palmer, Pam Cope, Vanda Andrews and Helen Barber; also to Alison Coles for looking after our website: www.battleconference.com. I cannot close this preface without recording here an item of news that delighted all historians, and especially those interested in the work of the Battle Conference: the award in the New Year’s Honours List of an OBE to Marjorie Chibnall. As editor of ANS volumes 12 (1989) to 16 (1993) she ensured that a conference that had owed its existence to Allen Brown would survive and flourish after his death. Historians of the Anglo-Norman world, more conscious perhaps than most of the ‘succession problem’, will appreciate just how much we owe to her – and that even before she took on a ten year stint as secretary of the newly formed Memorial Trust. The OBE may not have been an award for sheer stamina in attending every Battle Conference since its inception in 1978, but is surely deserved on those grounds alone. John Gillingham

ABBREVIATIONS AA SS Actes Henri II AD ANS Antiqs. Journ. Arch. Journ. ASC ASC, Swanton ASE BAA BAR Bates, Regesta Battle Chronicle BIHR BL BN Cal. Docs France Cal. Pat. Rolls Carmen CCCM CCM CCSL CNRS Complete Peerage De gestis pontificum Diceto DNB Domesday Book

Acta Sanctorum (of the Bollandists) L. Delisle and E. Berger, Recueil des actes de Henri II, roi d’Angleterre et duc de Normandie, 4 vols, Paris 1909–27 Archives Départementales Anglo-Norman Studies The Antiquaries Journal (Society of Antiquaries of London) Archaeological Journal (Royal Archaeological Institute) Anglo-Saxon Chronicle, ed. D. Whitelock et al., London 1969 The Anglo-Saxon Chronicle, trans. and ed. M. Swanton, London 1996 Anglo-Saxon England British Archaeological Association British Archaeological Reports Regesta Regum Anglo-Normannorum: the Acta of William I (1066–1087), ed. D. Bates, Oxford 1998 The Chronicle of Battle Abbey, ed. Eleanor Searle, OMT, 1980 Bulletin of the Institute of Historical Research British Library Bibliothèque Nationale Calendar of Documents preserved in France . . ., i, 918–1216, ed. J. H. Round, HMSO, 1899 Calendar of Patent Rolls, HMSO, 1891– The Carmen de Hastingae Proelio of Guy, Bishop of Amiens, ed. F. Barlow, OMT, 1999 Corpus Christianorum, Continuatio Medievalis Cahiers de civilisation médiévale Corpus Christianorum, Series Latina, Turnhout 1953– Centre National de la Recherche Scientifique Complete Peerage of England, Scotland, Ireland, Great Britain and the United Kingdom, 13 vols in 14, London 1910–59 William of Malmesbury, De gestis pontificum Anglorum, ed. N. E. S. A. Hamilton, RS, 1870 Radulphi de Diceto Opera Historica. The Historical Works of Master Ralph de Diceto, Dean of London, ed. W. Stubbs, 2 vols, RS, 1876 Dictionary of National Biography Domesday Book, seu liber censualis . . ., i, ii, ed. A. Farley, 2 vols, ‘Record Commission’, 1783; iii, iv, ed. H. Ellis, 1816

Domesday People

K. S. B. Keats-Rohan, Domesday People: a Prosopography of Persons Occurring in English Documents, 1066–1166, i, Domesday Book, Woodbridge 1999 Eadmer HN Historia Novorum in Anglia, ed. M. Rule, RS, 1884 EEA English Episcopal Acta EHD English Historical Documents, 2nd edn, i, 500–1042, ed. D. Whitelock; ii, 1042–1189, ed. D. C. Douglas and G. W. Greenaway, London 1979–81 EHR English Historical Review English Lawsuits English Lawsuits from William I to Richard I, ed. R. C. van Caenegem, 2 vols, Selden Society CVI–CVII, 1990–91 EYC Early Yorkshire Charters, i–iii, ed. W. Farrer (Edinburgh, 1914–16), and iv–xii, ed. C. T. Clay, Yorks. Archaeological Soc., Record Soc., extra series I–X, 1935–65 Fasti, 1066–1300 John Le Neve, Fasti Ecclesiae Anglicanae, 1066–1300, ed. D. E. Greenway, London 1968– Fauroux Recueil des actes des ducs de Normandie (911–1066), ed. M. Fauroux, Mémoires de la Société des Antiquaires de Normandie XXXVI, 1961 Gesta Guillelmi The Gesta Guillelmi of William of Poitiers, ed. R. H. C. Davis and M. Chibnall, OMT, 1998 Gesta Regum William of Malmesbury, Gesta Regum Anglorum, ed. R. A. B. Mynors, R. M. Thomson and M. Winterbottom, 2 vols, OMT, 1998–9 Gesta Stephani Gesta Stephani, ed. and trans. K. R. Potter and R. H. C. Davis, OMT, 1976 Giraldi Cambrensis ed. J. S. Brewer, J. F. Dimock and G. F. Warner, 8 vols, RS Opera XXI, 1861–91 HBS Henry Bradshaw Society Historia Novella William of Malmesbury, Historia Novella, ed. E. King and K. R. Potter, OMT, 1998 HMSO Her Majesty’s Stationery Office Howden, Chronica Chronica Rogeri de Houedene, ed. W. Stubbs, 4 vols, RS, 1868–71 Howden, Gesta Regis Gesta Regis Henrici Secundi et Ricardi Primi, ed. W. Stubbs, 2 vols, RS, 1867 Howlett, Chronicles Chronicles of the Reigns of Stephen, Henry II and Richard I, ed. R. Howlett, 4 vols, RS, 1884–9 HSJ Haskins Society Journal Huntingdon Henry, Archdeacon of Huntingdon: ‘Historia Anglorum’, ed. D. Greenway, OMT, 1996 ITS Irish Texts Society JEH Journal of Ecclesiastical History JMH Journal of Medieval History John of Worcester The Chronicle of John of Worcester, ii–iii, ed. R. R. Darlington and P. McGurk, OMT, 1995–8 Journ. BAA Journal of the British Archaeological Association JRSAI Journal of the Royal Society of Antiquaries of Ireland Jumièges Gesta Normannorum Ducum of William of Jumièges, Orderic Vitalis and Robert of Torigni, ed. E. M. C. van Houts, 2 vols, OMT, 1992–5

Lanfranc’s Letters Liebermann Med. Arch. MGH MGH SRG MGH SS Monasticon MRSN MSHAB MTB Newburgh NMT ns OMT Orderic os PBA PL PR PRIA PRO Regesta

RHF RIA Royal Writs RS S

SATF ser. SHR Torigni Trans. TRHS

The Letters of Lanfranc, Archbishop of Canterbury, ed. H. Clover and M. Gibson, OMT, 1979 Die Gesetze der Angelsachsen, ed. F. Liebermann, 3 vols, Halle 1903–16 Medieval Archaeology Monumenta Germaniae Historica MGH Scriptores Rerum Germanicarum in usum Scholarum MGH Scriptores William Dugdale, Monasticon Anglicanum, ed. J. Caley, H. Ellis and B. Bandinel, 6 vols in 8, London 1817–30 Magni Rotuli Scaccarii Normanniae, ed. T. Stapleton, 2 vols, London 1840–4 Mémoires de la Société d’Histoire et d’Archéologie de Bretagne Materials for the History of Thomas Becket, ed. J. C. Robertson, 7 vols, RS, 1875–85 Historia Rerum Anglicarum of William of Newburgh, in Howlett, Chronicles, vols 1 and 2 Nelson’s Medieval Texts new series Oxford Medieval Texts Orderic Vitalis, Historia Ecclesiastica, ed. M. Chibnall, OMT, 1969–80 old series Proceedings of the British Academy Patrologiae cursus completus, series Latina, ed. J. P. Migne, Paris 1841–64 Pipe Roll (as published by Pipe Roll Society) Proceedings of the Royal Irish Academy Public Record Office Regesta Regum Anglo-Normannorum, i, ed. H. W. C. Davis, Oxford 1913; ii, ed. C. Johnson and H. A. Cronne, Oxford 1956; iii, ed. H. A. Cronne and R. H. C. Davis, Oxford 1968 Recueil des historiens des Gaules et de la France, Paris 1738–1904 Royal Irish Academy Royal Writs in England from the Conquest to Glanvil, ed. R. C. van Caenegem, Selden Society LXXVII, 1959 Rolls Series, London Anglo-Saxon Charters: an Annotated List and Bibliography, ed. P. H. Sawyer, London 1968; revised edn, ed. S. E. Kelly et al., available online at www.trin.cam.ac.uk/chartwww. Société des Anciens Textes Français series Scottish Historical Review The Chronicle of Robert de Torigni, in Howlett, Chronicles, vol. 4 Transactions Transactions of the Royal Historical Society

VCH Vita Ædwardi Wace Waltham

Victoria County History The Life of King Edward who Rests at Westminster, ed. F. Barlow, 2nd edn, OMT, 1992 Wace, Le Roman de Rou, ed. A. J. Holden, 3 vols, SATF, Paris 1970–73 The Waltham Chronicle, ed. and trans. L. Watkiss and M. Chibnall, OMT, 1994

The Norman Conquest and the Media

R. Allen Brown Memorial Lecture

THE NORMAN CONQUEST AND THE MEDIA Richard Barber William the First was the first of our kings Not counting Ethelreds, Egberts and things, And he had himself crowned and anointed and blest In Ten-Sixty-I-Needn’t-Tell-You-The Rest. But being a Norman, King William the First By the Saxons he conquered was hated and cursed, And they planned and they plotted far into the night, Which William could tell by the candles alight. Then William decided these rebels to quell By ringing the Curfew, a sort of a bell, And if any Saxon was found out of bed After eight o’clock sharp, it was Off with His Head. So at BONG NUMBER ONE they all started to run Like a warren of rabbits upset by a gun; At BONG NUMBER TWO they were all in a stew, Flinging cap after tunic and hose after shoe; At BONG NUMBER THREE they were bare to the knee, Undoing the doings as quick as could be; At BONG NUMBER FOUR they were stripped to the core, And pulling on nightshirts the wrong side before; At BONG NUMBER FIVE they were looking alive, And bizzing and buzzing like bees in a hive; At BONG NUMBER SIX they gave themselves kicks, Tripping over the rushes to snuff out the wicks; At BONG NUMBER SEVEN, from Durham to Devon, They slipped up a prayer to Our Father in Heaven; And at BONG NUMBER EIGHT it was fatal to wait, So with hearts beating all at a terrible rate, In a deuce of state, I need hardly relate, They jumped BONG into bed like a bull at a gate.* Before I am thrown out of these august surroundings for excessive levity, there is a serious purpose in quoting this rhyme, from Eleanor and Herbert Farjeon’s Kings and Queens of England of 1932. It offers a stereotype: the Normans are the oppressors of the simple Anglo-Saxons, who are shut up and deprived of their liberty. It is a point * Quoted by permission of the estate of Eleanor and Herbert Farjeon and Victor Gollancz Ltd.

2

Anglo-Norman Studies XXVI

of view typical of children’s history books of that period: but children’s books often tell us what their adult peers are thinking. This is pure popular tradition, prejudices, warts and all. It is not history; but it is a powerful myth. What I would like to examine today is something of the process by which this stereotype was established; I am less concerned with the truth or otherwise behind it than with tracing the writers and publications which created the image, what we would today term the media. The Norman conquest was hardly an event that chroniclers and historians could simply ignore. Yet it is perhaps surprising how few of them care to offer an opinion on it, instead of merely recording the political facts and perhaps the characters of the chief actors. It is not that medieval chroniclers were shy of offering racial stereotypes; they are in today’s terms splendidly politically incorrect. The Normans, to the medieval mind, were legitimate subjects for description and distinction. They were recognised as a race, and the idea of ‘races’ with distinctive traits is embedded in the tradition of medieval learning, beginning with Isidore’s definition of the gens in his Etymologies as ‘a multitude sprung from a single source’.1 How the Normans used their consciousness of their existence as a gens to create their own myths is already well-trodden territory, and I shall do no more than touch on it in the present paper.2 The Normans themselves liked to think of themselves as a fierce, warlike and aggressive people, so it is hardly surprised that something of this survives among the Anglo-Norman historians of English affairs. However, these can be divided into two broad groups, those favourable, or at least not hostile to the Conquest, and those who in their heart of hearts simply do not accept Norman rule. The mildest picture of the Normans is to be found in those historians who recognise that their rule has brought some benefits to England, despite the initial harshness of the invaders. This group includes the chronicles written under Henry I, and later ‘official’ chronicles. There are no hard and fast divisions between the chroniclers on grounds of Saxon or Norman origin. For example, Eadmer, whose name betrays his Saxon birth, believes that Harold was a perjurer, and that this fact and the decay of English monastic discipline due to the Danish invasions were sufficient to justify the conquest. Nonetheless, he admires Wulfstan as the ‘sole survivor of the old Fathers of the English people’;3 his real hero, and the subject of his books, is of course Anselm, who favoured the English customs and resisted the interference of the Norman kings in the government of the English church. Orderic Vitalis, as a Saxon monk in a Norman monastery, is similarly persuaded that there was good reason for the invasion: the sins of the royal house of Ethelred and his successors, particularly the murder of Edward the Confessor’s brother Alfred, are the cause of the divine intervention which the conquest represents. It is he who gives us one of the first contrasting pictures of the two races: The Normans are a warlike race, and unless restrained by firm rule, are always ready to make trouble. In all societies wherever they are found, they struggle for mastery, and disregarding all sanctions of truth and good faith are always moved by fierce ambition.4 He also denounces the Normans as oppressors, and accuses them of robbing men of their freedom: even though this is in the context of an attack on the conduct of the marcher lord, Robert of Rhuddlan, towards the Welsh, the vocabulary is significant: 1 2 3 4

Quoted by Marjorie Chibnall, The World of Orderic Vitalis, Woodbridge 1996, 213. See G. A. Loud, ‘The “Gens Normannorum” – Myth or Reality?’, ANS 4, 1981 (1982), 104–16. Eadmer, Eadmer’s History of Recent Events in England, trans. Geoffrey Bosanquet, London 1964, 46. Orderic iv, 8.

The Norman Conquest and the Media

3

the Welsh ‘still enjoyed their original liberty’ before Robert’s activities, and Orderic declares that ‘it is not right that Christians should so oppress their brothers’.5 However, Orderic is also harsh towards the English, whom he describes as a rough and almost illiterate people, drunken, gluttonous, lustful and effeminate.6 He recognises the eventual benefits of Norman rule, the peace and good order that William and more particularly Henry I were able to bring to England. William of Malmesbury, half Norman and half English by birth, declares that the Normans ‘have my loyalty, both for my own origins and for what I owe them’,7 and his verdict on them is therefore largely favourable. He reflects at length on the cultural effect of the Conquest, and also offers a contrasting portrait of Normans and English. He is harshly critical of the English nobles, who were covetous, lecherous and drunken, spending their money on orgies. ‘In small, mean houses they wasted their entire substance, unlike the French and the Normans, who in proud great buildings live a life of moderate expense.’8 The Normans are ‘well-dressed to a fault, and particular about their food, but this side of any excess. The whole nation is familiar with war, and hardly knows how to live without fighting. They charge the enemy with spirit, and if force has not succeeded, are equally ready to corrupt him with craft and coin.’ He portrays them as ambitious yet hospitable, great patrons of religion, devout and generous. As to oppression, ‘it perhaps provides some proper justification for the king’s policy if he was somewhat too harsh towards the English, that he found almost none of them trustworthy – behaviour which so exasperated his ferocity that he deprived the more powerful among them first of their revenues, then of their lands, and some even of their lives’. William refused to appoint Englishmen to vacancies within the Church, ‘driven to this, unless I am mistaken, by their ingrained prejudice against the king . . .’ In other words, William sees the Norman influence as almost entirely benign, and the English as ungrateful and treacherous. The late twelfth century historians are silent on the issue of Norman versus English, perhaps indicating that it is no longer such a burning question. Indeed, in the mid-thirteenth century Matthew Paris offers a very different view: he sees Harold as rash, untrustworthy and self-confident, and comes down firmly in favour of William ‘great-hearted in the business of war, circumspect, faithful to his word, sociable and cheerful in peace . . .’ He portrays him as widely accepted by the English as a bringer of peace, welcomed by them with the words ‘Blessed is he that cometh in the name of the Lord.’9 That he later acted as a tyrant is a reflection on him personally, rather than on the Norman government as a whole. Matthew Paris’s later praise for the English royal line in his life of Edward the Confessor has been read as a change of heart;10 but surely it is simply appropriate to his new subject? What is notable in Paris is that he never criticises the Norman government or laws. We now turn to the hostile, pro-English versions of the conquest, beginning with Henry of Huntingdon. He writes from a secular English standpoint, while admitting the failings of the Anglo-Saxons;11 and he adds to William’s account the theme of divine justice: 5 6 7 8 9

Orderic iv, 138–9. Orderic iv, 6. Gesta Regum i, 423. Gesta Regum i, 459. Matthew Paris, Flores historiarum, ed. H. R. Luard, 3 vols, RS, 1890, i, 599; this does not appear in the Chronica majora, from which the Flores is abbreviated. 10 R. Reader, ‘Matthew Paris and the Norman Conquest’ in The Cloister and the World, ed. J. Blair and B. Golding, Oxford 1996, 140–4. 11 John Gillingham, The English in the Twelfth Century, Woodbridge 2000, 128–9.

4

Anglo-Norman Studies XXVI For God had chose the Normans to wipe out the English nation, because He had seen that the Normans surpassed all other people in their unparalleled savagery.12

Nor do they keep faith: he criticises the Normans for their ‘mad treachery’ in the early years of the reign of Stephen, and even if he uses ‘Norman’ for a small group of insurgents within the royal court, he is nonetheless implying that this tendency towards treachery was widespread among their fellow-countrymen.13 And finally, Henry of Huntingdon raises the theme which is above all others the popular complaint about the Norman legacy: the injustice of their justice. He begins by echoing the Anglo-Saxon Chronicle, but adds his own comments in the second and third sentences: The more they spoke of right, the greater injustice was done. Those who were called justices were the source of all injustice. Sheriffs and reeves, whose office was justice and judgement, were more frightful than thieves and robbers, and more savage than the most savage.14 Another hostile account of the Conquest comes from a historian whom we now regard as a writer of fiction, Geoffrey of Monmouth; it is in his most inventive pages, containing the supposed prophecies of Merlin, that he gives us his view of the Conquest in the shape of an ex post facto prediction: There shall come a people dressed in wood and in iron corselets who will take vengeance on it for its wickedness. This people shall give their dwellings back to the earlier inhabitants, and the destruction of foreigners will be clear for all to see. The seed of the White Dragon shall be rooted up from our little gardens and what is then left of its progeny shall be decimated. They shall bear the yoke of perpetual slavery and they will wound their own Mother with their spades and ploughshares.15 The harshness of the Norman rule over the Saxons (‘the seed of the White Dragon’) is plainly depicted by Geoffrey, and he uses a phrase – iugum perpetue seruitutis16 – which was to become the epitome of the popular view of the relationship between Norman and Saxon. Between the twelfth and the fourteenth century we have no account of the Conquest in English: but when the English find their voice again, what they have to say is of great significance for the future. Indeed, I was tempted to subtitle this paper ‘The Revenge of the Anglo-Saxons’; for the next three centuries, English sources display a deep-seated antipathy to all things Norman. Three English verse chronicles, written between 1290 and 1340, give a very hostile account of the Norman invasion, attributing the harshness of contemporary life to the events of two hundred and fifty years earlier. Robert Mannyng of Brunne, Robert of Gloucester, and Thomas

12 13 14 15

Huntingdon, 403. Gillingham, English in the Twelfth Century, 130. Huntingdon, 403. Geoffrey of Monmouth, The History of the Kings of Britain, trans. Lewis Thorpe, Harmondsworth 1966, 173. 16 Geoffrey of Monmouth, The Historia Regum Britannie, I. Bern, Burgerbibliothek, MS. 568, ed. Neil Wright, Cambridge 1985, 75.

The Norman Conquest and the Media

5

Castleford all offer variants on the same theme.17 Robert Mannyng declares that William the Conqueror ‘set us in servage, of freedom fell the flower; the English through tallage live yet in sorrow full sour’, adding that because of Hastings ‘now they are in that bondage that was brought from overseas, now are they in servage who were once so free, our freedom that day forever took leave’.18 Robert of Gloucester’s chief complaint is not the loss of freedom, but the disadvantage of language: he bemoans the fact that men must learn the alien Norman tongue if they wish to rise in the world; and indeed one of the earliest French-English glossaries dates from this period.19 Thomas Castleford spends fifty lines lamenting the loss of English lands and liberty to the Norman invaders; William gives English land ‘to his aliens. From English blood England he tore, and left not an inch of soil to them . . . All the Englishmen, truth to tell, who lived in England from thenceforth, had to work, out of necessity, on other men’s land to get their food, dwelling as bondsmen and thralls, doing all that thraldom implies, living and working in bondage: they and their descendants were always in service. Englishmen who did not wish to live in such thraldom quit the kingdom and went where they wished – they would not dwell in England any longer.’20 He also protests forcibly against the use of French in law courts: He set up sheriffs and justices over everyone who wanted to rebel against him, to give judgement in the Frankish tongue, to judge folk both old and young; for the bondsmen of English lineage were not meant to know from their words how they were judged, for good or evil, but could only stand like beasts obedient to their will.21 After such declarations, one almost expects a call to arms, a summons to eject the oppressors; but this is not the line which Robert Mannyng takes; if one Norman conquest has brought ‘all this thraldom’, ‘bondage and distress’,22 failure to pay taxes in support of Edward III’s wars may bring down a second and worse oppression by the French invaders. The Norman conquest is a warning, not something to be undone in order to gain ancient freedoms. Mannyng’s protest should be read in conjunction with the political crisis of 1340 over the level of war taxes, and also with the statute of 1362 by which proceedings in court were to be in English, though they were to be enrolled in Latin. In fact, French survived in the law courts for a long time to come. If we turn to the great survey of English history written later in the same century, Ralph Higden’s Polychronicon, we find that he borrows from Henry of Huntingdon to emphasise in similar fashion the enslavement of the English: ‘Whanne this

17 Thea Summerfield, ‘Synthesis and Tradition in the Early Fourteenth-Century Verse Chronicles in Eng-

lish’, Thirteenth Century England VII, 143–52. 18 Robert Mannyng of Brunne, The Chronicle, Part 2, ed. Idelle Sullens, Medieval and Renaissance Texts

Series 153, Binghamton, NY, 1996, lines 1621–3, 1760–2. None of this is in Mannyng’s source, Pierre Langtoft. See The Chronicle of Pierre de Langtoft, ed. Thomas Wright, RS, 1866, i, 408–35; Douglas Moffat, ‘Sin, Conquest, Servitude: English Self-Image in the Chronicles of the Early Fourteenth Century’, in The Work of Work: Servitude, Slavery and Labor in Medieval England, ed. Allen J. Frantzen and Douglas Moffat, Glasgow 1994, 147. 19 The Metrical Chronicle of Robert of Gloucester, ed. W. A. Wright, RS, 1887, ii, lines 7500–1, 7537–45. 20 Castleford’s Chronicle, or The Boke of Brut, ed. Caroline D. Eckhardt, Early English Text Society os 306, ii, lines 31923–42 (my translation). 21 Ibid., lines 31943–9. 22 Mannyng, Chronicle, ed. Sullens, 2, 6316–8, p. 645.

6

Anglo-Norman Studies XXVI

William was kyng unnethe was there a lord in Engelond [an] Englische man; bote Englische men were i-made bonde, so that it was schame and despite to be i-cleped an Englischeman.’23 In many ways, Higden became the standard version of English history, if we judge by the number of surviving copies, and Caxton, looking for a general history to print, chose the Polychronicon in Trevisa’s translation. This view of the conquest must have been familiar to many readers. After Higden, it is not until the Tudor period that the history of England from earliest times, or indeed from the Norman conquest onwards, is of interest to writers once more. The first of the Tudor historians is of course a foreigner, the Italian Polydore Vergil, an acute and critical scholar who both cuts through much of the myth surrounding his subject, while preserving such tales as the gallant version of Edward III’s foundation of the Order of the Garter. He is immune to any nostalgia for an Anglo-Saxon past, and is generally favourable to William and the Normans, emphasising their qualities as lawgivers, and representing the Saxons as rebellious and disruptive of good order. It is he who is the first authority for the story of the Norman curfew, which he represents as an attempt by William to contain the natural contumacy of the Saxons.24 Polydore Vergil is writing from the standpoint of the Renaissance. It was the Reformation which was to have a much greater impact on the reputation of the Normans. For the Protestants in England, the Roman Catholic rite was seen as a Norman imposition, as William Tyndale avers in 1532, in his Practice of Prelates: King Harold exiled or banished Robert of Canterbury: for what cause, the English Polychronicon specifieth not: but if the cause were not somewhat suspect, I think they would not have passed over it with silence. This Robert gat him immediately unto king William the conqueror, then duke of Normandy: and the pope Alexander sent duke William a banner, to go and conquer England, and clean remission unto whosoever would follow the banner, and go with king William. Here mark how straight the pope followed Christ’s steps and his apostles’! They preached forgiveness of sins to all that repented, through Christ’s bloodshedding; the pope preacheth forgiveness of sins to all that will slay their brethren, bought with Christ’s blood, to subdue them unto his tyranny. Whatsoever other cause king William had against king Harold, thou mayest be sure that the pope would not have meddled, if Harold had not troubled his kingdom: neither should duke William have been able to conquer the land at that time, except the spiritualty had wrought on his side. What blood did that conquest cost England, through which almost all the lords of the English blood were slain, and the Normans became rulers, and all the laws were changed into French! . . .25 Tyndale was writing as a member of the radical opposition to the established order; but within three decades, this radical view would become the new orthodoxy. The chroniclers of Elizabeth’s reign paint a very different picture, for they were writing in a context which had changed dramatically. The Elizabethan church under Matthew Parker had a strong antiquarian agenda, the chief witness to which is his magnificent collection of manuscripts, bequeathed by him to Corpus Christi College, Cambridge. 23 John Trevisa, in Polychronicon Ranulphi Higden Monachi Cestrensis, ed. J. R. Lumby, RS, 1879, vii,

319. 24 See appendix below. 25 William Tyndale, Expositions and Notes on sundry portions of the Holy Scriptures, together with the

Practice of Prelates, ed. Henry Walter, Parker Society 1849, 294.

The Norman Conquest and the Media

7

The purpose behind this unrivalled collection was a simple one: to prove that the Anglo-Saxon church was an apostolic foundation, and that the Roman Catholic rite of the Norman kings, who had after all invaded under the banner of St Peter and with the pope’s authority, had supplanted a legitimate and independent church. Research into this period of history was intended to show the Anglo-Saxon church as the forerunner of the Elizabethan settlement. A letter in Matthew Parker’s correspondence indicates strong official support for his antiquarian activities: it is a circular from the Privy Council making clear the Queen’s interest in preserving ‘such historicall matters and monumentes of antiquitie, both for the state ecclesiasticall and ciuile gouernement’.26 The question of the religious impact of the Norman conquest was to remain a theme for debate; but it became very much secondary to the point which Tyndale makes, almost as an aside, about the legal boundary marked by the Conquest. This is the other major strand of the image of the Normans: not only were they oppressors, but they destroyed a golden age of Anglo-Saxon freedom to which all Englishmen should lay claim. This idea too had its roots in medieval literature; we have seen it – though implicitly rather than explicitly – in Castleford’s chronicle, and the lawyers themselves had begun to hold the view that the liberties of the individual derived from Saxon precedent. Besides Castleford, the other source for this idea of a primitive and virtuous Anglo-Saxon commonwealth is the Miroir des Justices, which is to be found in Andrew Horn’s collection of London customs, legal tracts and other matters, including a copy of Henry of Huntingdon, which he bequeathed to the corporation of London on his death in 1328.27 Horn had a particular interest in the Anglo-Saxons, as annotations in the manuscript show, but it was from a legal point of view, the search for precedents, rather than straightforward antiquarian curiosity. He copied the collection known as the Leges Anglorum, expanding and adding to it. The Miroir des Justices, however, is the key text in all this for our present purposes. In the third chapter, Of the original constitutions,28 Horn depicts the laws that he believed to have been in force in the reign of king Alfred, and sets these up as his ideal, and in the last part of the book, on abuses of the law, he looks back to Alfred’s time as one in which the law was enforced even-handedly, and false justices were duly punished. Horn claims to describe the usages ‘which have obtained since the time of king Arthur’,29 but in fact his starting point is ‘the coming of the English’, and he claims that parliament was established at that time to ‘hear and determine . . . all the writs and plaints concerning wrongs done by the king’.30 It was for this that subsequent generations were to cherish the Mirror of Justices; its law is more fiction than fact, drawn from imagination rather than researches into existing law-books, for even the genuine surviving Anglo-Saxon laws were not used by the author. Nor is the Conquest itself even mentioned. But what is evident is a strong libertarian streak, which claims to derive its authority from an invented scheme of the laws of the time of the ‘coming of the English’.

26 R. I. Page in Matthew Parker’s Legacy: Books and Plate, Cambridge 1975, 6. 27 See Jeremy Catto, ‘Andrew Horn: Law and History in Fourteenth-Century England’, in The Writing of

History in the Middle Ages: Essays presented to Richard William Southern, ed. R. H. C. Davis and J. M. Wallace-Hadrill, Oxford 1981, 367–91. 28 The Mirror of Justices, ed. William Joseph Whittaker, Selden Society 7, London 1895 [for 1893], 8–15. 29 Ibid., 3. 30 Ibid., 7.

8

Anglo-Norman Studies XXVI

Confirmation of this idea is found in the Burton annals, which at the end of the thirteenth century claim that the papal legate Pandulf reproached king John during the interdict of 1208–9, telling him that ‘. . . you enjoy and enforce the evil laws of William the Bastard, even the worst; and you spurn as worthless the laws of St Edward, even the best’. Like almost all direct speech in medieval chronicles, this of course reflects the views of the chronicler rather than those of the supposed speaker; but what it tells us is that there was a definite tradition which, as in the Miroir de Justices, looked back to the good old days.31 The theme of the alien nature of law written in French about which Castleford complained is to be found in the work of the Tudor lawyer, Thomas Starkey. In A Dialogue between Pole and Lupset, contemporary with Tyndale’s tract, he declares ‘that our commyn law ys wryten in the french tonge, & therin dysputyd . . . [it] ys also ignomynyouse and dyshonowre to our natyon, for as much as therby ys testyfyd our subjectyon to the normannys’.32 The laws are not only written in ‘thys barbairase langage’ but are themselves ‘barbarouse & tyrannycal . . . wherfor yf we wyl ever bryng in true cyvylyte in to our cuntrey . . . we must abrogate of thos lawys veray many . . .’33 ‘Civility’ in this context has overtones of the civil law, which Starkey had studied in Italy; the contrast is not with the ancient liberties of the Anglo-Saxons, but the more orderly and just world of the Roman tradition and of the Renaissance and Reformation. The Normans are a barbarous bygone, not the source of present tyranny; for this latter idea would have been a dangerous sentiment under Henry VIII. It is in the major Tudor chronicles that we find evidence of this new view of the Norman conquest. In support of it, they cite an obscure – and indeed still unpublished – Canterbury chronicle, that of Thomas Sprott, a monk of the abbey in the early thirteenth century.34 The version used by several Tudor writers was that of William Thorn, written at Canterbury at the end of the fourteenth century, but his text is almost identical with that of Sprott.35 Although Sprott is writing earlier than the English verse chroniclers, what he has to say is part of the same tradition, that the Normans’ chosen means of oppression was through the new laws which they introduced. The story was first found in a manuscript of Sprott’s chronicle by William Lambarde and published in his Perambulation of Kent, written in 1570 and printed in 1576. Lambarde was a lawyer who had studied with Laurence Nowell, famed for his knowledge of Anglo-Saxon and antiquities in general, and he uses the episode as part of his background to the foundation of the Norman legal system. He has his doubts about the authority of Sprott – ‘if hee shall seeme too weake to give sufficient authoritie to the tale’ – but says that ‘the matter itselfe is neither incredible, nor unlikely’ since the history of Kentish law confirms it. So he proceeds to tell it anyway.36

31 Annales Monastici, ed. H. R. Luard, RS, 1864, i, 213. 32 Thomas Starkey, A Dialogue between Pole and Lupset, ed. T. F. Mayer, Camden Fourth Series 37,

London 1989, 82. 33 Ibid., 129. 34 On the date of William Sprott’s work, see William Thorne’s Chronicle of Saint Augustine’s Abbey Can-

terbury, trans. A. H. Davis, Oxford 1934, xxvi. It is possible that Sprott wrote as late as 1272; the evidence is not at all clear. 35 William Sprott’s chronicle is in BL, Cotton Tiberius A ix. 36 See J. C. Holt, ‘The Origins of the Constitutional Tradition in England’, in Magna Carta and Medieval

The Norman Conquest and the Media

9

After such time (saith he) as Duke William the Conquerour had overthrowne King Harold in the field, at Battel in Sussex, and had received the Londoners to mercy, he marched with his armie towarde the castle of Dover, thinking thereby to have brought in subjection this countie of Kent also. But Stigande, the Archbishop of Canterbury, and Egelsine the Abbat of St Augustines, perceiving the daunger, assembled the countrie men together, and laide before them the intollerable pride of the Normanes that invaded them, and their own miserable condition, if they should yeeld unto them. By which meanes, they so enraged the common people, that they ran forthwith to weapon, and meeting at Swanscombe, elected the Archbishop and the Abbat for their captaines: this done, eche man got him a green bough in his hand, and bare it over his head, in such sort, as when the Duke approached, he was much amased therewith, thinking at the first, that it had been some miraculous woode, that mooved towards him: but they, as soone as he came within hearing, cast away their boughes from them, and at the sounde of a trumpet bewraied their weapons, and withall dispatched towards him a messenger, which spake unto him in this manner: The commons of Kent (most noble Duke) are readie to offer thee, either peace, or warre, at thyne owne choyse, and election: peace, with their faithfull obedience, if thou wilt permit them to enjoy their ancient liberties: warre, and that most deadly, if thou denie it them. Now when the Duke heard this, and considered that the danger of deniall was great, and that the thing desired was but small, he forthwith, more wisely than willingly, yeelded to their request. . . and they onely of all England . . . obtained for ever their accustomed priviledges.37 This story of the grant of liberty to Kent was quickly seized on; it was known to both Matthew Parker and Burghley, for the archbishop sent an advance copy of Lambarde’s Perambulation to Elizabeth’s minister in May 1573.38 Richard Grafton, the Protestant printer, was in fact the first to publish the story; he summarises Lambarde’s account and makes it more emphatic: ‘where as before the commyng of the said William there were no slaves or bondmen, now that all, aswell noble as meane men were brought under the perpetuall seruyle yoke of the Normannes . . .’, Kentish men, he concludes, retain ‘the ancient liberties of Englishe men, and their countries, lawes and customes’.39 Holinshed, famous as Shakespeare’s source, transforms the ‘perpetual servile yoke of the Normans’ into ‘the intellectual bondage of the Normans’;40 a little later, quoting Henry of Huntingdon and Matthew Paris as his sources, he elaborates his theme: Many of them were constrained (as it were for a further testimonie of Seruitude and bondage) to shaue their beards . . . Others utterlie refusing to susteine such an intollerable yoke of thraldome as was dailie laid upon them by the Normans, chose rather to leaue all both goods and landes . . . Whereupon it came to passe

Government (London and Ronceverte, VA, 1985), 9–13. Lambert was married to Juliana Horn, daughter of William Horn of London; it is tempting to speculate on a family connection with Andrew Horn. 37 William Lambarde, A Perambulation of Kent, Bath 1970, 20–1. 38 V. J. K. Brook, A Life of Archbishop Parker, Oxford 1962, 317, 322. 39 Richard Grafton, A Chronicle at large, and meere History of the affayres of England, London 1568, 155; Lambarde specifically claims in the Perambulation to be the first to have found the story. 40 Raphael Holinshed, The Chronicles of England, Scotlande and Irelande, London 1577, III, 1.

10

Anglo-Norman Studies XXVI within a while that no man might trauell in safetie from his owne house or towne to his next neighbours . . .41

Domesday Book also becomes a symbol of Norman heavy-handedness: it was exacted, Holinshed says, ‘to the great griefe and impouerishment of the people, who sore lamented the miserable estate whereinto they were brought and Hated the Normans in their harts unto the uerie death’.42 John Stow, Holinshed’s great rival, quotes Lambarde’s translation of Sprott in full, using the phrase ‘the perpetuall bondage of the Normans’. The greatest antiquary of the period, William Camden, gives this portrait of the Normans in his Britannia: What an insolent and bloudie victorie this was, the Monks that write of it, haue declamed with full mouth: neither is it to be doubted, but in this Victorie (as it hapneth in other) wickednes tooke head and bare the full sway. William the Conqueror, in token as it were of a Trophee for this conquest, abrogated some part of the ancient positive lawes of England, brought in some Customes of Normandie, and by vertue of a decree, commanded, That all causes should be pleaded in the French tongue. The English he thrust out of their ancient Inheritances, assigned their lands and Lordships to his souldiers, yet with this reservation to himselfe, that he should still remaine chiefe Lord and bind them to doe due service and homage unto him and his successors, that is to say, That all of them should hold their lands in Fee or fealty.43 Sprott’s original story has been reinforced and extended to a general depiction of the loss of Anglo-Saxon laws and liberty throughout the land. This is reflected in Edmund Spenser’s A View of the Present State of Ireland, where he is concerned that the law imposed by William the Conqueror might not be suitable for Ireland. It belongs, he says, to specific circumstances in England: For England before the entrance of the Conqueror was a peaceable kingdom, and but lately enured to the mild and good government of King Edward the Confessor, besides now lately grown unto a loathing and detestation of the unjust and tyrannous rule of Harold, an usurper, which made them the more willing to accept of any reasonable conditions and order of the new victor, thinking surely it could be no worse than the latter, and hoping well it would be as good as the former; yet what the proof of the first bringing in and establishing of those laws was, was to many fully bitterly made known.44 In the ensuing discussion, it becomes clear that Spenser does not favour the common law, but merely sees it as appropriate to the circumstances of the Conquest and the maintenance of order; his point is that a similar imposition of law will not work in the very different circumstances prevailing in Ireland. But the passage shows that the idea of the common law as originating with the Conquest was well established by the end of Elizabeth’s reign. Camden may not show any specific knowledge of the legal tradition of Anglo-Saxon liberty, but he moved in circles where history and law converged. The 41 Ibid., III, 5. 42 Ibid., III, 14. 43 William Camden, Britain, or a chorographicall description of the most flourishing Kingdomes,

England, Scotland and Ireland, trans. Philemon Holland, London 1610, 152. 44 Edmund Spenser, A View of the Present State of Ireland, ed. W. L. Renwick, Oxford 1970, 4.

The Norman Conquest and the Media

11

publication of Britannia in 1586, according to Joan Evans, ‘had so focused the interest of a little coterie of friends that they had fallen into the habit of regular meetings for discussion. Out of these meetings arose the first Society of Antiquaries in England’. Its members, so far as we can tell, probably included Robert Cotton, John Stow and Henry Spelman. Their discourses aimed to establish a ‘cultural longevity’ for Britain, to show that its institutions were ancient and authentic, and even if their arguments now seem romantic and heavily influenced by medieval myth, they were seen as highly relevant in political terms, in the unsettled atmosphere of the Stuart succession. The meetings of the society became less frequent around 1607–8, and in 1614 a revival was set in motion, but with the proviso, as Spelman noted, that ‘for avoiding Offence we should neither meddle with Matters of State, nor of Religion’. This was not enough: ‘his Majesty took a little Mislike of our Society’, and the society never met again. James was right to ‘mislike’ the society, for such antiquarian researches were to become a central plank of the Puritan revolution. Christopher Hill’s exposition of the idea of the ‘Norman yoke’ and its political influence is masterly, and anyone exploring this territory can only add the occasional gloss.45 But I do not think it has been much discussed in the context of this conference, and for that reason – and because it is central to my subject – I beg leave to go over some of the ground again. My question is slightly different from that set by Christopher Hill, whose main concern is how the history of the Conquest was used in seventeenth century politics: I want to know how the seventeenth-century radicals had learnt about the Conquest. The phrase ‘the Norman yoke’ was coined from Richard Grafton’s ‘the perpetuall servile yoke of the Normans’, which in turn originated with Geoffrey of Monmouth, but the concept that it describes contains two genuinely separate strands: the myth of Anglo-Saxon freedom stemming from the lawyers, and the myth of Norman oppression drawn from the chronicles,46 which do not, by and large, idealise the Anglo-Saxon past. Of course the two are intertwined, but it is important to note that nostalgia for a golden age is an important element in the political arguments, as well as the rejection of Norman – or Stuart – tyranny. Our analysis of the Tudor historians and of Camden shows us something of how the myth of the Norman yoke entered the political consciousness of the early seventeenth century. We are not looking at – as Michael Wood would like us to think – ‘a continuous literary and folk theme from the eleventh century to the Tudors, continuing on to the radicals in the seventeenth century’.47 Rather, the myth belongs to the legal and antiquarian tradition I have outlined, and its real home is in political units which enjoyed a degree of independence from the crown, with different customs and liberties, namely Kent and London. Here the general transfer of wealth and power to the Norman incomers had been less marked than elsewhere, and it is in the pragmatic world of the citizen and the yeoman that the record of the Saxon past, as a means of affirming hard-won and hard-kept rights against the centralising power of the monarchy, was most valued. If there is folklore present, it is in Thomas Sprott’s little

45 Christopher Hill, ‘The Norman Yoke’, in Puritanism and Revolution: Studies in Interpretation of the

English Revolution of the 17th Century, London 1958, 46–111; Christopher Hill, ‘The Norman Yoke’, in The Intellectual Origins of the Puritan Revolution Revisited, Oxford 1997, 361–6. 46 Frantzen and Moffat, Work of Work, 153 supports this. 47 Michael Wood, In Search of England, Harmondsworth 2000, 17.

12

Anglo-Norman Studies XXVI

fantasy (so reminiscent of Birnam wood) about the means by which Kent retained its customs. It is also arguable that the Norman conquest was seen in some quarters as a moment in British history analogous with the accession of the Stuart kings. James I declared that English kings had been the absolute owners of all property in England since 1066, an idea first put forward by Adam Blackwood in 1581.48 The Stuarts were certainly not averse to being told that the Norman kings were the founders of their dynasty, as in the dedication of John Hayward’s The Lives of the III Normans, Kings of England. William the first, William the second, Henrie the first. dedicated to Charles, prince of Wales: the author says that ‘the persons of whom it treateth are those most worthy ancestors of yours who laid the foundations of this English Empire’. And the anxiety of the aristocracy to prove descent from the Conqueror’s companions, which still haunts genealogists today, dates from the investigations of Tudor historians like John Stow, who printed tables of the Conqueror’s companions in their works, and thus encouraged the Elizabethan heralds to make often implausible connections to these names. ‘Norman’ was once more a synonym for ‘aristocratic’, and therefore a red rag to the radical bulls. This may help to explain the unexpected vehemence of the invective against ‘the Norman yoke’, but it is still surprising that the Normans should be so prominent in the language of politics. However, the Puritan writers were able to draw on a wide range of scholarly works as well. The work of Tudor antiquaries, beginning with Bale and Leland, had laid the foundation for a systematic knowledge of the writers of medieval England, and by 1600 the major sources for knowledge of Anglo-Norman history were available in print, and were relatively easy to obtain.49 And even unprinted manuscripts had a wider circulation than we would expect today: the Miroir des Justices, first printed in 1642, had already circulated among the Elizabethan antiquaries, and was a major source for Sir Edward Coke in his Institutes. The argument that the Norman conquest was the root of all evil in England was developed most fully by the sect called the Levellers, and in particular by Gerald Winstanley, though other writers such as John Hare blazed the trail. In essence, their stance was that all private ownership was based on robbery and force, and the defining moment for this transition to a society where property was paramount was identified as the arrival of William in 1066: James I’s ideas were adopted and turned against his successor. They declared that England was stolen from the English, and given to the Normans, who set up an elaborate legal system to defend their ill-gotten gains; because the law was in a foreign tongue, the English could not argue their case, and in any event the whole edifice of law and property was supported by the church, with its mixture of spiritual threats of damnation and secular taxes in the shape of tithes. And, as we have seen, all this depended on the idea of a primitive liberty which had once existed in Anglo-Saxon England. Anglo-Saxon laws were idealised by some thinkers, such as John Warr, who argued that if the law was in a known

48 Hill, Intellectual Origins, 362. 49 Matthew Parker himself edited the Flores Historiarum in 1567: in the same year William of Newburgh

was printed for the first time. Four years later Parker published the Chronica Majora of Matthew Paris, and this was followed by Asser’s Life of Alfred, and two works by Thomas Walsingham. The first major collection of chronicles was produced by Henry Savile in 1596, bringing William of Malmesbury, Henry of Huntingdon and Roger of Howden into circulation. Two further important additions came some twenty years later, John Selden’s edition of Eadmer and Duchesne’s collection of the Norman writers, which offered Dudo of St Quentin, William of Poitiers, Orderic Vitalis, the Encomium Emmae Reginae, and a quantity of fashionable genealogical material into the bargain.

The Norman Conquest and the Media

13

language, then anyone could plead his own cause; and this had not been the case for the mass of the people since the Conquest, for legal French still flourished.50 For the Levellers, the work of the Puritan revolution would not be accomplished until the whole structure of property was thrown down and replaced. Winstanley told the rulers of the Commonwealth: . . . you still lift up that Norman yoke, and slavish Tyranny, and hold the People as much in bondage, as the Bastard Conquerour himself, and his Councel of War. Take notice, That England is not a Free People, till the Poor that have no Land, have a free allowance to dig and labour the Commons and so live as Comfortably as the Landlords that live in their Inclosures. For the People have not laid out their Monies, and shed their Bloud, that their Landlords, the Norman power, should still have its liberty and freedom to rule in Tyranny in his Lords, landlords, Judges, Justice, Bayliffs and State Servants . . .51 The portrait of the Normans as alien tyrants is to be found in its purest form in the works of John Hare, who in 1648 actually entitled one of his tracts St Edward’s Ghost, or, Anti-Normanism. His theme is that all things ‘Norman’ should be destroyed: . . . the end of this whole discourse . . . is . . . to endeavour these following particulars: i. William ‘the conqueror’ ‘to be stripped of that insolent title’; no longer to stand ‘for the alpha of our kings in the royal catalogue’. ii. Charles ‘to derive his right from St Edward’s legacy’, to ‘restore the ancient English arms into the royal standard’. iii. ancient Norman nobility ‘to repudiate their names and titles brought over from Normandy’ and to adopt English names. iv. ‘all laws and usages introduced from Normandy be . . . abolished; and a supply made from St Edward’s laws, or the civil . . . and restored into the English or Latin tongue’. v. ‘That our language be cleared of the Norman and French invasion upon it, and depravation of it, by purging it of all words and terms of that descent; supplying it from the old Saxon and the learned tongues . . .’52 His political programme is engagingly vague as to what is to be done once Anti-Normanism has prevailed, an airy wave of the hand in the direction of Anglo-Saxon laws and language. For him the removal of all traces of things Norman is the paramount requirement: it is the most extreme view of the Normans as the hated enemy of the ordinary people of the seventeenth century. What had changed between the Tudor historians and the Puritan pamphleteers? The two sides of the argument, Norman oppression and Anglo-Saxon liberty, had coalesced to create the idea that the Norman conquest was a revolution, a point in history where a totally new system of law and property was introduced, the effects of which were still deep-seated in seventeenth century society: another revolution was needed in order to right the wrongs done in 1066. But with the coming of the Restora-

50 Christopher Hill, The World Turned Upside Down, London 1972, 219. 51 The True Levellers Standard, in Gerrard Winstanley, Works, ed. George H. Sabine, New York 1965,

259–60. 52 John Hare, St Edward’s Ghost, or Anti-Normanism, in The Harleian Miscellany, London 1808, 8, 104.

14

Anglo-Norman Studies XXVI

tion, the Normans were likewise restored to favour, and the radical views brushed aside. The apparently anti-Norman Argumentum Anti-Normannicum, published anonymously in 1682, takes the opposite view to Hare: the conquest was no conquest, but was subject to the approval of the English people, whose laws survived William’s arrival, and who retained both estates and a place on the king’s council. The author’s object was to bolster the parliamentary party in its opposition to Charles II’s absolutist tendencies, and he hoped ‘to see GREAT CHARLES happy in’s Parliament’.53 In 1695, Sir William Temple could write of ‘the happy circumstances of this famous Conquest’;54 and Sir Richard Baker’s popular Chronicle of the Kings of England, which went through several editions between 1643 and 1733, has this to say of William the Conqueror himself: And indeed all the states might in his very person finde something to make them apt to tolerate his Government. For first, the people might think themselves in a sort advanced, being now made members of a greater body, when the dukedom of Normandy should come to be annexed to the kingdom of England; and by experience of his good government being a Duke, they might well hope, he would not govern worse being made a king. And the Nobility might be well content, as having a King of their former King’s choosing; and though a stranger, yet no Alien, as having in him many veins of the same blood, and therefore likely also to have some veins of the same goodness as their good King Edward. But specially the Clergy could not choose but be content, as having a King who came commended to them, by a commending as strong as a commanding, the Pope’s benediction.55 This was the book which Joseph Addison’s fictional squire Sir Roger de Coverly kept always to hand ‘in his hall window’ for his historical reading, and its influence was probably greater than any of the more learned seventeenth century tracts we have quoted so far. Attitudes were of course complicated by the appearance of a new William, William of Orange, who was also a foreigner and also ‘conquered’ England; parallels between the raging tyrant of the Puritan revolution and the new ruler were obviously unwelcome, and the subject became one which was best left to the pens of satirists like Daniel Defoe, who to illustrate his theme that ‘He only justly holds a Government, That rules a People by their own consent’, calls William the Conqueror: Whore in his Scutcheon, Tyrant in his Face . . . Upon his Sword engraved the Right Divine. For Defoe, hereditary right was worthless, and the ‘true succession’ of the English crown could not depend on it, since it had been so often disregarded over the centuries. The Norman conquest was simply another usurpation of the people’s rights; whereas William III came with the people’s consent. In this welter of declarations and denials as to the nature of the Normans and their title to England, scholars still worked on the original materials for the history of the Conquest. The most notable of them was Robert Brady, a physician who became Keeper of the Records at the Tower, and who led the way in archival research; his

53 Quoted in David Douglas, English Scholars 1660–1730, London 1951, 122. 54 Sir William Tempest, Introduction to the History of England, quoted in David Douglas, The Norman

Conquest and British Historians, Glasgow 1971, 9. 55 Sir Richard Baker, Chronicle of the Kings of England, London 1670, 24.

The Norman Conquest and the Media

15

Introduction to the Old English History of 1694 is based on sound documentation. (Some of his admonitions are remarkably close to the mark; it is still all too easy to earn Brady’s reproach that ‘you have been partial in your Citations, taking some fragments or parcels of Sentences or sometimes a short sentence you thought might serve your turn, and always leaving what you could not but know would have destroyed your Notion and Argument . . .’56) Brady’s conclusions were anathema to those who felt that the Anglo-Saxon origin of contemporary English institutions was the bedrock on which the governance of England was founded. Brady firmly declared that there was no evidence for the presence of the commons in parliament before 1265, and that only tenants in capite had a collective presence in the affairs of government before then. The Normans and their doings were thus irrelevant to contemporary political arguments. Such a drastic historical revision was not immediately accepted, but permeated subsequent writing gradually, until by the end of the eighteenth century it had become almost an orthodoxy. It is worth noting that the manuscripts of the English verse chronicles were attracting attention at this time; the Yorkshire antiquary Nathaniel Johnston owned the unique copy of Castleford’s chronicle in the 1670s, and although he published little, he was interested in constitutional history. More interesting is the ownership at the same period of a copy of Robert Mannyng of Brunne’s chronicle by another Yorkshire antiquary, William Petyt, keeper of the records at the Tower and a member of the Inner Temple, whose work on the antiquity of parliament provoked Brady’s researches.57 However, ‘the Norman Yoke’ was still attractive to radicals, and the theme reappears briefly in the 1790s, most notably in Thomas Paine, who attacks the principle of English hereditary monarchy in Common Sense, partly by discrediting its very foundations: England, since the conquest, hath known some few good monarchs, but groaned beneath a much larger number of bad ones; yet no man in his sense can say that their claim under William the Conqueror is a very honorable one. A French bastard landing with an armed banditti, and establishing himself king of England against the consent of the natives, is in plain terms a very paltry rascally original. – It certainly hath no divinity in it.58 He goes on to brand William a usurper, and in The Rights of Man, takes him as a type of the government founded on power, a government of conquerors. He claims that the oppression of the Normans lives on in popular memory: ‘The hatred which the Norman invasion and tyranny begat, must have been deeply rooted in the nation, to have outlived the contrivance to obliterate it. Though not a courtier will talk of the curfeu-bell, not a village in England has forgotten it.’59 But from the beginning of the nineteenth century onwards, the Normans belong largely to the realms of fiction. Historians might fight over them, but they no longer carried weight as a political symbol. So my final section concerns the Normans in the historical novel. (And here I must confess to a problem: the historical novel exists in a kind of limbo, spurned – often rightly – by literary critic and historian alike; if I have missed an obvious example, it is because guides through this particular forest 56 Douglas, English Scholars, 126. 57 N. R. Ker, Medieval Manuscripts in British Libraries, Oxford 1969, i, 88. 58 Thomas Paine, Common Sense and The Rights of Man, edited and with a foreword by Tony Benn,

London 2000, 16–17. 59 Ibid., 182.

16

Anglo-Norman Studies XXVI

are almost non-existent.60) The first and most famous of these novels, Sir Walter Scott’s Ivanhoe, echoes Tom Paine’s views on the popular hatred of the Normans in a way which may seem surprising, given that their political views were at the opposite ends of the spectrum: Scott was as ardent a tory as Paine was a radical. Despite Paine’s attempts to make ‘the Norman yoke’ as powerful a political statement as it had been in the Civil War, Scott shows us that this moment of history no longer belonged to the political arena.61 Ivanhoe is an extraordinary book in its way. Relegated to the dustier shelves of secondhand bookshops today, it is remarkably modern: a Hollywood script before its time, totally anachronistic at moments, with sudden flashes of insight and a highly sympathetic portrait of the persecuted medieval Jews, and driven forward by a plot full of cliffhangers. It owes not a little to the twists and turns of the gothic novels that preceded it. History it is not; Scott was widely read, and was aware that he was in breach of historical truth, but nonetheless preferred to draw his inspiration from medieval literature. His proud and independent Saxons, though not blameless, are drawn from the myth of Anglo-Saxon liberty we have already discussed; Cedric, Athelstane, Rowena and Ivanhoe himself stand for a freer, more open and honest society than the endlessly treacherous Norman knights. Only Richard Coeur de Lion is exempted from this criticism, and Scott does not hesitate to blame him for his neglect of the good government of his kingdom. Norman scorn for the Saxons, and the hatred of the Saxons for their oppressors, drives the story forward. Scott’s introduction of Robin Hood into the story is interesting. It is made easier by the then accepted dating of Robin Hood to the reign of King John, but there is another element in the Robin Hood stories which is worthy of comment. The ballads are relatively few in number until the Restoration, when collections of the poems, both in manuscript and in print, become popular, the so-called Robin Hood Garlands. I would suggest, that although the first printed Garlands date from just after the Restoration, their true origin is as part of that image of a primeval Anglo-Saxon liberty which had had such a political vogue under the Puritans: Dobson and Taylor note that the first Garland ‘probably originated in the 1650s’.62 Robin Hood essentially represents a fictionalised anti-Normanism, transmuted into popular literature from a partisan reading of the medieval chronicles. The novelists of Victorian England had no doubt as to whose side they were on: perhaps I have missed a major work of fiction with a Norman as its hero, but I do not think it is to be found – nor, to the best of my knowledge, is such a book to be found in the twentieth century either. Instead, it is Harold and Hereward the Wake who dominate the historical fiction of the time. Bulwer Lytton, hugely popular in his day, but now consigned to a largely unread oblivion, wrote Harold, the last of the Saxon kings in 1848 ‘at a heat’, as he himself said. It is sadly easy to mock a book which begins ‘Merry was the month of May in the year of our Lord 1052. Few were the boys, and few the lasses, who overslept themselves on the first of that buxom month.’ His preface to the third edition, however, is germane to our present subject. In it he describes the problems of bringing before the public a period of which they knew little, and gives his view of the Anglo-Saxons as ‘a decrepit monarchy and a fated race’, in contrast to which 60 The best is still Helen Cam’s now rather outdated Historical Association pamphlet of 1961, The His-

torical Novel. 61 Sir Walter Scott, Ivanhoe, Waverley Novels: The Border Edition vol. 9, London 1906. 62 R. B. Dobson and J. Taylor, ‘The Legend since the Middle Ages’, in Robin Hood: an Anthology of

Scholarship and Criticism, ed. Stephen Knight, Woodbridge 1999, 170.

The Norman Conquest and the Media

17

I have to bring forcibly before the reader the vigorous attributes of the coming conquerors, – the stern will and deep guile of the Norman chief – the comparative knowledge of the rising Norman Church – the nascent spirit of chivalry in the Norman vavasours; a spirit designed to emancipate the very people it contributed to enslave, associated, as it imperfectly was, with the sense of freedom: disdainful, it is true, of the villein, but proudly curbing, though into feudal limits, the domination of the liege.63 If this sounds more impressive than the subsequent text of the novel, it is because Lytton’s other avatar was as a reforming politician, and there is more than an echo of the old controversies over the political legacy of the Conquest here. And indeed the difference between Lytton and Scott is that – the opening apart – Lytton is writing with the chroniclers at his elbow rather than the romances, even if he reads them otherwise than we might today. His last scene, the burial of Harold, has a footnote discussing the controversy over Harold’s grave in which he cites William of Poitiers and Giraldus Cambrensis. Charles Kingsley’s Hereward the Wake has survived rather better than Lytton’s novel, though it bears a similar subtitle, Last of the English.64 It too has claims to a scholarly pedigree, for the dedication is to Thomas Wright, one of the pioneering Victorian editors of Middle English; Kingsley claims that Wright disinterred Hereward ‘long ago, when scarcely a hand or foot of him was left standing out from beneath the dust of ages’. And when Kingsley comes to describe the oppression of the English, he evidently draws on the series of lectures on the Norman conquest which he was giving as Regius Professor of History at Cambridge at the time when he wrote the novel: For now broke out in England that wrong-doing which endured as long as she was a mere appanage and foreign farm of Norman kings, whose hearts and homes were across the seas in France. Fitz-Osbern, and Odo the warrior-prelate, William’s half-brother, had been left as his regents in England. Little do they seem to have cared for William’s promise to the English that they were to be ruled still by the laws of Edward the Confessor. . . Oppression began, lawlessness and violence; men were ill-treated on the highways; and women – what was worse – in their own homes; and the regents abetted the ill-doers.65 The vision of the book, though, is that of an ancient English freedom and an ancient English pride; the Normans are little more than stock figures, greedy and treacherous, eventually luring Hereward into their ranks, only to kill him by stealth years later. If Kingsley has read his historians, he has also, and chiefly, read the Gesta Herewardi; it was this that Wright had disinterred and printed as an appendix to his edition of Gaimar in 1850. If Victorian writers had no great love for the Normans, none of them matched the odium which a rather different user of words, the professor of Anglo-Saxon at Oxford, bore towards things Norman. John Earle, in his highly successful Philology of the English tongue, which ran to five editions between 1866 and 1892, has this to say of the evolution of English in the middle ages:

63 Lord Lytton, Harold: The Last of the Saxon Kings, London [1902], 16. 64 Despite Kingsley’s claims, the story of Hereward was not entirely unknown: C. Macfarlane had pub-

lished The Camp of Refuge: A Tale of the Conquest of the Isle of Ely in 1844, presumably using Francisque Michel’s edition in his Chroniques Normandes, Rouen 1836–40. 65 Charles Kingsley, Hereward the Wake, “Last of the English”, London 1883, 219.

18

Anglo-Norman Studies XXVI One particular class of words shall be noticed in this place as the result of the French rule in England. This is a group of words which will serve to depict the times that stamped them on our speeech. They are the utterance of the violent and selfish passions. Most of the sinister and ill-favoured words which were in the English language at the time of Shakspeare, owed their admission to this unhappy era. The malignant passions were let loose, as if without control of religion or of reason; men hotly pursued after the objects of their ambition, covetousness, or other passions, till they grew insensible to every feeling of tenderness and humanity; they regarded one another in no other light but as obstructive or auxiliaries in their own path. Such a society supplied the nascent English with a mass of opprobrious epithets . . .

Good rhetoric, perhaps, but the four or five pages of examples which follow scarcely make his case; he adduces the French origin of gambling terms, and a few not very severe insults. Modern philologists would probably argue the reverse: our terms of abuse are generally Anglo-Saxon. My next and final author brings us just into the twentieth century; he too depends on a legendary history, the Vita Haroldi, and his tale concludes two of the shrewdest books on England’s past that have ever been written, whatever their faults, and whatever the faults of excessive patriotism too often – and unjustly – imputed to its writer. I mean, of course, Rudyard Kipling’s Rewards and Fairies, and the tale entitled ‘The Tree of Justice’. But before we look at that, let us turn first to ‘Old Men at Pevensey’ in Puck of Pook’s Hill,66 where de Aquila, one of the Conqueror’s companions, reflects on the troubles of Henry I’s reign from the point of view of the Norman barons: They say Henry is overly English for their stomachs, because he hath married an English wife and she hath coaxed him to give back their old laws to our Saxons. (Better ride a horse on the bit he knows, I say!) . . . William crammed us Norman barons full of good English acres after Santlache. I had my share too . . . but I warned him . . . that he should have bidden the Barons give up their lands and lordships in Normandy if they would be English lords. Kipling’s theme is the forging of Englishness, looking to a day ‘when there will be neither Norman nor Saxon, but all English’, and when the lords of the land, like de Aquila, will ‘think for England’. He sees Normandy as a blight on English affairs, and the Normans only as a building block for the future greatness of their new home. ‘The Tree of Justice’ similarly treats of reconciliation between Norman and Saxon, but at the highest level. Drawing on the Vita Haroldi, in which, of course, Harold survives Hastings to become a hermit at Chester, Kipling weaves a new ending to the legend: Rahere, Henry I’s jester, meets Harold, now ‘a witless man that journeyed without rest to all the shrines of England’, and takes him under his protection. But he and other villagers are brought in to drive the deer for the king’s sport, and one of them calls out to beware of the arrow that killed William Rufus. The king furiously demands to know who it is, and it proves to be the old and – supposedly – witless man. Because he is Rahere’s man, Rahere answers for him, and brings him before the king. Rahere, using the jester’s licence to speak unpalatable truths, shows Henry and his barons that it is indeed Harold who sits among them, and makes them

66 Rudyard Kipling, Puck of Pook’s Hill, London 1906, repr. 1983, 105–34.

The Norman Conquest and the Media

19

all admit to the fact: none of them mock him for his claim to have survived ‘Santlache fight’, and Hugh, the one Saxon in the company, honours him, with impunity, as his king. At this moment, Harold falls forward in his chair, and dies in Hugh’s arms, recognised and honoured by his Norman foes. Norman or Saxon no longer matters: Rahere, in the last moments of Harold’s life, calls both Henry and Harold ‘of England’, and in the jester’s words, as always, there is truth.67 Are the Normans at last part of England’s past in a way that is both remembered and yet no longer controversial, assimilated into our collective memory? Or do we still instinctively owe our prime loyalty to the Anglo-Saxons? You may read my final evidence either way: it is a line of Latin, on the monument at the British war cemetery at Bayeux, recording the Normandy landings of 1944: GULIELMO VICTI VICTORIS PATRIAM LIBERAVIMUS.68

APPENDIX After the lecture, John Gillingham pointed out that the history of the curfew was very obscure, particularly in relation to the idea that William I had imposed it as a means of oppressing the Saxons. This brief note may inspire someone else to research the subject in depth; much of what follows comes from John Gillingham’s knowledge of the byways of medieval history as well as from my own investigations. The word ‘curfew’ is only recorded a century and a half after the Conquest. The earliest reference in the Oxford English Dictionary is to the statutes imposed on London by Edward I in 1285, where no-one is to wander, without very good cause, the streets after curfew tolled at St Martin’s-le-Grand.69 However, the word occurs in the Customary of the Abbey of Bury St Edmunds, dated by Antonia Gransden to 1234, where the writer notes that the bell rung at compline ‘is called curfu’.70 And ecclesiastical customs take the word even further back in Latin: ‘ignitegium’ is found in the statutes of Lichfield cathedral laid down by Bishop Hugh of Nonant (1185–1198): the ‘pulsus ignitegii’ or curfew bell is to be rung for the time that it would take to journey for a mile.71 It seems to have been an institution found in towns – Bury St Edmunds was of course relatively unusual in being a monastery within the lordship of the neighbouring town – as the earliest known use of ‘ignitegium’ shows. This is in the ‘peace of Valenciennes’ of 1114, which is a grant of certain rights to the town; in these it is ordained that if a disturbance should arise in the town the ‘ignitegium’ and the bell summoning the town militia to arms should both be rung.72 The idea that an evening bell was a signal for movement in the streets to cease is found before the concept of curfew or ‘ignitegium’: at a church council at Caen in 1061 it was decreed that after the evening bell people should shut their doors and no longer wander abroad. ‘Ut quotidie sero signi pulsu ad preces Deo fundendas quisque invitaretur, 67 68 69 70

Rudyard Kipling, Rewards and Fairies, London 1910, repr. 1983, 311–36. I owe this reference to Professor Celia Applegate. Statutes of the Realm i, 102. The Customary of the Benedictine Abbey of Bury St Edmunds in Suffolk, ed. Antonia Gransden, Henry Bradshaw Society XCIX, Chichester 1973 [for 1966], 45. 71 Monasticon vi, 1256. 72 MGH SS XXI, 607 col. 2 line 44.

20

Anglo-Norman Studies XXVI

atque occlusis foribus domorum, ultra vagari amplius vetitum admoneretur.’73 Similar ordinances are found fairly widely in the later middle ages, for example in the laws and customs of the German towns. The specific name ‘cover-fire’ implies that it probably belongs to the rudimentary fire precautions of medieval cities, so prone to disastrous fires; it is analogous to the building regulations found in medieval new towns, where houses were separated by narrow alleys (andrones) which acted as firebreaks.74 Interestingly, the French word which is the source of the English, cuevrefeu, does not appear until the early thirteenth century. The association of the curfew with the Normans, and its depiction as an instrument of repression, is first mentioned in Polydore Vergil, in the 1546 edition of his Anglica Historia: ‘Item ut ferociam populi ad otium perducetet [sic], omnibus arma ademit, statuitque ut quisque paterfamilias uesperi circiter horam octavam, post meridiem, tecto cineribus igne, dormitum iret; et ad id signum uicatim dari uoluit, per campanas.’75 This passage does not occur in the first edition of 1534 or in the manuscript.76 Unfortunately, it is impossible to identify Polydore Vergil’s source, as his revision of 1546 seems to have consisted of adding various snippets of information which had come to his attention, rather than reworking his text from one particular new source. He concentrates on removing ‘statements which had become politically undesirable’, and improves ‘the notices of English institutions for English readers’.77 The entry for the curfew would come under the latter heading. All we can say is that the idea must have been current in print or in oral tradition in the 1540s, and that it was rapidly picked up by other Tudor historians. By the end of the eighteenth century, Paine could claim that ‘not a village in England has forgotten it’ and Thomas Gray’s ‘Elegy in a Country Churchyard’ had immortalised it as part of the English pastoral scene: The curfew tolls the knell of parting day. . . But whatever its subsequent fame, the origins of the curfew, and its association with Norman repression, remain stubbornly obscure.

73 Guillaume Bessin, Concilia Rotomagensis Provinciae, Rouen 1717, 48, printed from Mabillon, who

cites a charter of the monastery of ‘Pratellensis’; the synod was of the ‘bishops, abbots and peers of all Normandy’. Bessin claims that the bell known as ‘ignitegium’ was first instituted by William I ‘ut furtis nocturnis caveretur’. 74 The purpose of these is probably, but not certainly, as a firebreak: see Alain Lauret, Raymond Malebranche and Gilles Séraphin, Bastides: villes nouvelles du moyen-âge, Toulouse 1988, 99. 75 Polydore Vergil, Anglica Historia, 151. The quotation is from the Basle edition of 1555, repr. Menston 1972. 76 Denys Hay, Polydore Vergil: Renaissance Historian and Man of Letters, Oxford 1952, 189. 77 Ibid., 82–3.

Dudo of St Quentin and Norman Military Strategy

DUDO OF ST QUENTIN AND NORMAN MILITARY STRATEGY c.1000 Bernard S. Bachrach For almost a century and a half, scholars have questioned the historical accuracy of De Moribus et Actis primorum Normanniae ducum written by Dudo of St Quentin early in the eleventh century.1 This work, which is sometimes referred to by scholars as Gesta Normannorum, is now widely recognized to be littered throughout with unreliable details in regard to political matters and this is the case especially in regard to the early history of the Norman ducal family.2 Whether Dudo’s general failure to get the facts right, as many scholars believe, was intentional or whether his inaccuracies resulted from trying to compensate for his ignorace by filling in ‘gaps’ with fiction today still remains controversial.3 It is clear, nevertheless, that the contempo1

Francis Palgrave, History of Normandy and England, 4 vols, Cambridge 1852–64, ii, 908, questioned Dudo’s accuracy in a fundamental manner as early as 1857. A quarter-century later Henry Howarth, ‘A Criticism of the Life of Rollo, as told by Dudo de St Quentin’, Archaeologia 45, 1880, 223–50, made clear many of his delicts. Finally, Henri Prentout, Étude critique sur Dudon de S. Quentin et son Histoire, Paris 1916, mustered the massive detail which demolished De Moribus as an source for early Norman political history. Prentout also reviewed much of the earlier scholarly literature (pp. 3–13; 114–32). This critical tradition is dealt with in considerable detail by David. C. Douglas, ‘Rollo of Normandy’, EHR 57, 1942, 417–36, and less focused in David C. Douglas, ‘The Rise of Normandy’, PBA 33, 1947, 101–30. Both are reprinted in David C. Douglas, Time and the Hour, London 1977, 95–119, and 121–40, respectively. 2 The basic edition remains Dudo: De Moribus et Actis primorum Normaniae ducum, Caen 1865–72. Gerda C. Huisman, ‘Notes on the Manuscript Tradition of Dudo of St Quentin’s Gesta Normanorum’, ANS 6, 1983 (1984), 122–35, has promised a new edition. In addition, Eric Christiansen, Dudo of St Quentin, History of the Normans, Woodbridge 1998, has provided a complete translation with a useful introduction and helpful notes. Unfortunately, in dealing with technical military matters the translation is sometimes misleading and occasionally wrong. This is especially the case, for example, with regard to the term miles which he does not seem to recognize as a problem (p. xxxvi). See on this topic Bernard S. Bachrach, ‘The Milites and the Millennium’, HSJ 6, 1994, 85–95, 85–95, and reprinted in Bernard S. Bachrach, Warfare and Military Organization in Pre-Crusade Europe, London 2002, with the same pagination. For the continued problematic use of miles into the Anglo-Norman period and beyond see the decisive discussion by William Delehanty, ‘Milites in the Narrative Sources of England, 1135–1154’ (unpublished Ph.D. diss., Minneapolis, MN, 1975). 3 Emily Albu [Hanawalt], ‘Dudo of Saint-Quentin: The Heroic Past Imagined’, HSJ 6, 1994, 113, correctly observes that this negative view by modern scholars is encouraged by the fact that Dudo ‘apparently had access to archival materials and to the duke’s own memories and family records’ but chose rather to write a ‘Vergilian’ epic of no historical value in tracing the political narrative of the Norman rulers. Some scholars, however, have tried to vindicate Dudo as an historian who reported the facts in an essentially accurate manner. The strongest proponent of this view was Johannes Steenstrup, ‘Normandiets Historie under de syv første hertyger’, Det Konelgige Dansk Videnskaberess Selskabs Skrifter, Historisk og Filosophisk Afdeling (Mémoires de l’Academie royale des sciences et des lettres de Danemark, Section des lettres), 7th ser., V. 1, Copenhagen 1925, 1–310. This effort, it is widely agreed, has failed. As David Bates, Normandy before 1066, London 1982, 10–11, observes: ‘A series of studies by Professors Lucien Musset and Jean Yver, published over the last forty years and culminating in the latter’s magisterial survey of the province’s institutions, have consigned Dudo’s opinions to an oblivion from which they will surely never return.’ A recent popular work by François Neveux, La Normandie des ducs aux rois Xe–XIIe siècle, Rennes 1998, 17–59, and esp. pp. 22–7, 40, relies strongly on Dudo and disparages previous critiques. By contrast, a growing number of scholars have argued that it is essentially wrong to judge Dudo by the

22

Anglo-Norman Studies XXVI

rary ruling family in Normandy, Dudo’s patrons, commissioned him to write a story in praise of their ancestors that would be plausible to its intended audience at the ducal court.4 In the context of plausibility, it is likely to be of some importance that Dudo was writing within a rhetorical tradition that flourished in the northern parts of Francia Occidentalis during the latter part of the tenth century and especially at Liège where he was educated.5 Indeed Dudo himself opines, from time to time, on the importance of rhetoric.6 Thus, fundamental questions must be asked: how did Dudo maintain or perhaps more accurately try to maintain what scholars call ‘rhetorical plausibility’ with his audience? How did he go about trying to convince his listeners to pass over the factual inaccuracies with which the text was larded, whether intentionally or in a heroic effort to fill in the gaps, while at the same time maintaining rhetorical plausibility? Finally, at least in the eyes of some potential listeners, the Normans may have been treated in a too favorable or praiseworthy light in relation to what Dudo’s contemporary listeners were likely to know or believe about the subject.7 standards normally applied to historians, even eleventh-century historians, because he was trying to write in the Norse saga tradition. See, for example, August Nitschke, ‘Beobachtungen zur normannische Erziehung’, Archiv für Kulturgeschichte 43, 1961, 273–82; Frederick Amory, ‘The Viking Hasting in Franco-Scandinavian Legend’, in Saints, Scholars and Heroes: Studies in Medieval Culture in Honour of Charles W. Jones, 2 vols, ed. Margot H. King and Wesley M. Stevens, St Johns, MN, 1979, 265–86; Elisabeth van Houts, ‘Scandinavian Influence in Norman Literature of the Eleventh Century’, ANS 6, 1983 (1984), 107–21; Eleanor Searle, ‘Fact and Pattern in Heroic History: Dudo of St Quentin’, Viator 15, 1984, 75–85; Jan Ziolkowski, Jezebel: A Norman Latin Poem of the Early Eleventh Century, New York 1989; and Marcello Meli, ‘Dudoni di S. Quintino e la preistoria vichinga’, in Dudone di San Quintino, ed. Paolo Gatti and Antonella Degl’Innocenti, Trent 1995, 29–48. Christiansen, Dudo, pp. xvi–xix, effectively points out the weaknesses inherent in the saga-model and in efforts to demonstrate that early French vernacular written and/or oral accounts provided Dudo with his sources and models. This tradition of ‘romance’ origins goes back to the nineteenth century and was still being argued late in the twentieth century: e.g. R. H. C. Davis, The Normans and their Myth, London 1976, pp. 58–9. Karsten Friis-Jensen, ‘Dudo of St. Quentin and Saxo Grammaticus’, in Dudone di San Quintino, 11–28, casts strong doubt on the idea that Dudo uses a Scandinavian model. However, it should be noted that van Houts, ‘Scandinavian Influence’, makes some useful points in regard to connections between Normandy and Scandinavia even though the importance of a few of her examples is exaggerated. 4 Scholars have provided a broad spectrum of arguments regarding the aims of Dudo’s patrons. For example, Prentout, Étude critique, 19–23, saw the patrons of De Moribus as wanting a work of ‘propaganda’ that was intended to make the Normans respectable; Searle, ‘Fact and Pattern’, 122–3, sees the patrons as desiring a work of family praise in the ‘Germanic’ tradition of the ‘literature of house and Kindred’; while Felice Lifshitz, ‘Dudo’s Historical Narrative and the Norman Succession of 996’, JMH 20, 1994, 101–20, argues that Richard I wanted a work that was intended to help secure the succession of his son and heir Richard II. See Christiansen, Dudo, xxiii–xxvii, where previous scholarship is well summarized. 5 Richard W. Southern, ‘Aspects of the European Tradition of Historical Writing, 1: The Classical Tradition from Einhard to Geoffrey of Monmouth’, TRHS 5th ser. 20, 1970, 173–96. Geoffrey Koziol, Begging Pardon and Favor: Ritual and Political Order in Early Medieval France, Ithaca, NY, 1992, 139–55, groups Dudo with the other rhetorical historians of the later tenth and early eleventh centuries. See also Leah Shopkow, History and Community: Norman Historical Writing in the Eleventh and Twelfth Centuries, Washington DC 1997, 130; and Christiansen, Dudo, xxi–xxiii. 6 See, for example, De Moribus, bk IV, ch. 106 (p. 269); and bk IV, ch. 117 (p. 280); and regarding the importance of rhetoric in this intellectual milieu see Southern, ‘Aspects of the European Tradition’, 191–2. 7 Regarding rhetorical plausibility see, for example, the discussions by Nancy Partner, ‘The New Cornificius: Medieval History and the Artifice of Words’, 12, and Roger Ray, ‘Rhetorical Scepticism and Verisimilar Narrative in John of Salisbury’s Historia Pontificalis’, 66, 83–4, both in Classical Rhetoric and Medieval Historiography, ed. Ernst Breisach, Kalamazoo 1985. See also a series of essays by John Bliese: ‘Rhetoric and Morale: A Study of Battle Orations from the Central Middle Ages’, JMH 15, 1989, 201–26; ‘The Courage of the Normans – A Comparative Study of Battle Rhetoric’, Nottingham Medieval Studies 35, 1991, 1–16; and ‘Rhetoric Goes to War: The Doctrine of Ancient and Medieval Military Manuals’, Rhetoric Society Quarterly 24, 1994, 105–30. Cf. Christiansen, Dudo, xxi, who asserts that Dudo’s

Dudo of St Quentin and Norman Military Strategy

23

Dudo, modern scholars generally agree, knew that he was introducing a considerable element of false or at least unsubstantiated information into De Moribus and this is especially the case in his treatment of the pagan period. However, the modern scholar must ask whether the types of inaccuracies introduced by Dudo were of such a nature that Dudo’s contemporaries were likely to know they were implausible or inaccurate. For example, Dudo provides an extended discussion of the sacking of Luna, a city he locates somewhere in the Mediterranean, by the Viking leader Hasting.8 Was this event, which may never have taken place, likely to have resulted in loud murmers through the audience suggesting the possibility much less the probability of an authorial ‘fabrication’?9 In this context, it is important to note that for more than two hundred years the Vikings had been raiding far and wide throughout what had been the western half of the erstwhile Roman empire.10 Indeed, they even had operated in the Mediterranean11 and on the Atlantic coast of Spain.12 The general pattern of this raiding behavior was likely well known during the early eleventh century.13 Yet, with particular regard to the Luna episode, even modern scholars have been unable to identify the city with certainty, much less Hasting’s operations there.14 In a similar vein, Rollo, the putative founder of the Norman ruling house, is credited with laying siege to Paris. However, he is described as failing to take the city.15 In the course of Viking military operations in the regnum Francorum there were various sieges.16 There was, indeed, a very famous and well documented investment of Paris that failed in 888.17 In addition, there were numerous Viking military strikes all along the course of the river Seine, especially in the environs of Paris and against various of

‘first consideration was aesthetic’. However, on pp. xxix–xxxiv, where Christiansen provides a very satisfactory treatment of Dudo’s rhetorical techniques and observes (p. xxxii): ‘The plain passages are no less artificial than the coloured, since they observe the rhetorical plan of inspiring confidence by lucidity, verisimilitude, and brevity . . .’. 8 De Moribus, bk I, chs 5–7 (pp. 132–6). 9 Whether Lunx urbem with variations such as Lunxc and Lux found in the manuscripts of De Moribus, bk I, ch. 5 (p. 132) is, in fact, the Italian city of Luni is controversial despite Dudo’s gloss quae Luna dicitur. Regarding this story as a fabrication see Prentout, Études critiques, 53–7; and the extended the discussion by Silvia Busch, ‘Luni in the Middle Ages: The Agony and Disappearance of a City’, JMH 27, 1991, 283–96. The story of the raid on this town and Hasting’s ruse became a staple of Viking legend and was presented in a number of variations. See Rory McTurk, Studies in Ragnars Saga, Oxford 1991, 227; and for additional literature, Christiansen, Dudo, 184 n. 88. 10 The most thorough general treatment of Viking military operations remains Walther Vogel, Die Normannen und das fränkische Reich bis zur Gründung der Normandie, 799–911, Heidelberg 1906. See also Horst Zettel, Das Bild der Normannen und der Normanneneinfälle in westfränkischen, ostfränkischen und angelsächsischen Quellen des 8. bis 11. Jahrhunderts, Munich 1977; and the review of Zettel’s work by Bernard S. Bachrach in Speculum 55, 1980, 613–15. 11 Vogel, Die Normannen, 171–8. 12 With particular attention to Spain see Reinhard Dozy, Histoire des Musulmans d’Espagne jusqu’à la conquête de l’Andoalousie par les Almoravides, 3rd edn, 3 vols, ed. É. Levi-Provençal, Leiden 1932, ii, 250–371; and É. Levi- Provençal, Histoire de l’espagne musulmane, 3 vols, Paris–Leiden 1950–1967, i, 218–25, 310–12. 13 Zettel, Das Bild der Normannen, provides a source focused account. From a methodological perspective it would be helpful to try to ascertain how the wide base of information available through a broad spectrum of written sources was disseminated through oral discourse. 14 Cf. Busch, ‘Luni in the Middle Ages’, 287, who believes it is Luni in northwestern Italy; and Christiansen, Dudo, 184 n. 88, would seem to agree. 15 De Moribus, bk II, chs 16–17 (pp. 157–8). 16 See, for example, Vogel, Die Normannen, 160–6, 237–44. 17 E. Favre, Eudes, comte de Paris et roi de France (882–898), Paris 1893, 17–68, remains the best treatment of this operation; but cf. Vogel, Die Normannen, 320–38.

24

Anglo-Norman Studies XXVI

the city’s assets.18 Thus, it may be asked, who in Dudo’s audience was likely to know, almost a century after Rollo’s death, whether or not this particular ancestor of the Norman duke had undertaken a failed military effort against Paris? These and the many other possible factual ‘inaccuracies’ in De Moribus may possibly have evoked the suspicion of one or another member of Dudo’s audience from time to time and led them to the conclusion that some erroneous information was being purveyed. It is unlikely, however, that a large number of people would have questioned historical information of this type presented to them by Dudo. Indeed, when the Norman chronicler William of Jumièges, writing two generations after Dudo, abridged De Moribus for his Gesta Normannorum, he gave his audience and by extension his modern readers no reason believe that he questioned in any significant way the accuracy of Dudo’s largely inaccurate political narrative.19 Robert of Torigny, writing in the twelfth century, would appear to have trusted Dudo’s account without questioning it in a serious manner.20 It is also clear that vernacular writers such as Wace and Benoît of St Maure relied heavily on Dodo’s De Moribus without questioning its accuracy.21 Further, those who made new manuscript copies of Dudo’s De Moribus during the eleventh, twelfth, and thirteenth centuries appear to have made no noteworthy much less systematic efforts to excise ‘inaccuracies’ from the political narrative and thus correct the text.22 In short, few members of Dudo’s

18 E. A. Freeman, ‘The Early Sieges of Paris’, in his Historical Essays, First Series, 4th edn, London

1886, 212–56; Einar Joranson, The Danegeld in France, Rock Island, Ill., 1923, 141–52. 19 See Jumièges, 6–7, where William indicates that he omitted from Dudo’s account some of Rollo’s

activites that took place while he was still a pagan Viking chief because deeds done under such circumstances did not bring honor and were not useful. In short, William had serious reservations about flattering, i.e. engaging in adulatoria of a pagan. William’s aim here was not to attack the veracity of Dudo’s account but rather to assert the view that flattering a pagan was unseemly. Cf. Elisabeth M. C. van Houts, ‘The Gesta Normannorum Ducum: A History without an End’, ANS 3, 1980 (1981), 112, where she notes that Robert [of Torigny, who rejected William’s abridgement of Dudo’s account on these matters] seems to prefer the pagan hero as presented by Dudo to the Christian duke as portrayed by William of Jumièges. But in Jumièges, i, pp. xxxv–xxxviii, lxvii, she continues to treat William’s ‘editing’ of Dudo as though he were acting according to modern critical values. Shopkow, History and Community, 129–30, takes a different view and argues effectively that ‘William’s objections to Dudo’s narrative cannot arise from what we would understand to be historical criticism’. She goes on to suggest that it was Dudo’s ‘unabashedly exuberant rhetoric’ that caused a problem for William. 20 Van Houts, ‘The Gesta Normannorum Ducum’, 110, noted that Robert of Torigny ‘reinstates the stories that William of Jumièges had left out.’ I would suggest that this is clear evidence not only that Robert believed Dudo’s account regarding Rollo’s career but also that he believed that this information was utilis and brought an appearance of honor (species honesti). In consequence Robert omitted William’s statement as to why he had edited out material from Dudo regarding some aspects of Rollo’s pagan career. Concerning this omission by Robert, van Houts observed (p. 110): ‘Consequently Robert was forced to omit from the dedicatory letter the sentence in which William of Jumièges manifested doubt about the veracity of these stories.’ At this date, however, van Houts had translated speciem honesti vel utilis as ‘neither useful nor honest’, rather than, as in her subsequent edition, not offering ‘a model of what is honourable or edifying’. Cf. Koziol, Begging Pardon and Favor, 379 n. 36. 21 See Hugo Andresen, ‘Über die von Benoît in seiner normannischen Chronik benutzen Quellen, insbesondere über sein Verhältnis zu Dudo, Wilhelm von Jumièges und Wace’, Romanische Forschungen 1, 1883, 327–412; 2, 1886, 477–538; Francine Mora, ‘Dudo of Saint-Quentin et ses deux traducteurs Français, Wace et Benoît’, in Dudone di San Quintino, 77–102. With regard to Benoît see also Peter Damian-Grint, ‘En nul leu nel truis escrit: Research and Invention in Benoît de Sainte-Maure’s Chroniques des Ducs de Normandie’, ANS 21, 1998 (1999), 11–30. 22 Huisman, ‘Notes on the Manuscript Tradition’, 124, indicates that at least ten of the surviving fourteen manuscripts were copied either in the eleventh or twelfth century. It is worth mentioning in this context that although, as noted above, William of Jumièges eliminated some material regarding Rollo’s career when the latter was a pagan, the complete version of Dudo’s De Moribus was appended to the A version of William’s Gesta Normannorum, which was redacted sometime at the end of the eleventh century

Dudo of St Quentin and Norman Military Strategy

25

audience, much less his patrons and their entourage, were likely to be sufficiently well-schooled concerning the details of early Norman political history to identify a great many of his inaccuracies in regard to the minutiae that so animate modern scholars.23 However, Dudo appears, in addition, to have had an overarching schema for casting the Normans in a favorable light which he would seem to have pursued with a combination of at least three models that modern scholars arguably have identified. Emily Albu, for example, has noted: ‘He [Dudo] had heroes who came by sea . . . find a new homeland where they would settle and intermarry . . . to create a dynamic civilization with the best of both worlds . . .’24 Dudo’s debt to Virgil is uncontroversial and has been exhaustively documented.25 While Dudo’s use of a Virgilian model is well established, Lifshitz has contended that the Hebrew Bible, in Latin translation, also provided a model. Allusion is likely being made to the Normans as the ‘New Israel’ conquering Canaan after wandering in the symbolic desert.26 Finally, Shopkow has argued that the structure of De Moribus was shaped by hagiographical models.27 Whether many in Dudo’s aristocratic audience at the Norman court would have recognized one or more of these models or overarching schema, which modern scholars have worked so diligently to identify and concerning which they often disagree in very fundamental ways, may certainly be thought to be problematic.28 What is clear, however, is that Dudo provided ‘facts’, however simplified, that have enabled modern scholars to argue for each of these hypotheses. In my view, however, these models or schema should not be treated exclusively. Rather, they should be conflated to suggest that Dudo was presenting a general or overarching classical-Christian view of early Norman history. The Normans had come by sea, they had been in the desert of paganism, they were converted to Christianity, they conquered Normandy, intermarried with the Franks, their leader was martyred (or at least foully murdered) and they founded a polity that by the early eleventh century

(post-1092) or shortly thereafter. The A version omitted books I–IV from William’s text and thus the abridgements of Rollo’s career as a pagan discussed above (nn. 19–20). 23 Albu [Hanawalt], ‘Dudo of Saint-Quentin’, 112, notes that Dudo ‘was so convincing that virtually everyone believed him for over 800 years’. 24 Ibid., 113. 25 Most recently the Virgilian debt has been catalogued by Pierre Bouet, ‘Dudon de Saint-Quentin et Virgile: L’Eneide au service de la cause normande’, Cahiers des Annales de Normandie 23, 1990, 215–36, who also provides references to the relevant scholarly literature. 26 ‘Dudo’s Historical Narrative’, 109–11. 27 Leah Shopkow, ‘The Carolingian World of Dudo of St. Quentin’, JMH 15, 1989, 19–37; and the firm structure provided for this argument by Christiansen, Dudo, xxi–xxii. 28 Recently, for example, Lars Boje Mortensen, ‘Stylistic Choices in a Reborn Genre’, in Dudone di San Quintino, 79–102, questions the hagiographic model and looks to Widukind (Widukindi monachi Corbeiensis rerum gestarum Saxonicarum libri tres, ed. Paul Hirsch, MGH SRG, Hanover 1935), for comparisons in the tradition of ‘origin’ history or ethnogenesis. Even more recently, Shopkow, History and Community, p. 181, has emphasized that, among other things, Dudo is making a case for the development of ducal policy in a period of ‘rapid social change’; while Victoria B. Jordan, ‘The Role of Kingship in Tenth-Century Normandy: Hagiography of Dudo of Saint-Quention,’ HSJ 3, 1991, 54–62, strongly argues for the hagiography model but seems unaware of Shopkow’s earlier work. For other interpretations see: Davis, The Normans and their Myth, 52, who has Dudo writing a panegyric of sorts to Norman ducal power; Koziol, Begging Pardon and Favor, 147–60, who sees Dudo as a commentator on royal power; for van Houts, ‘The Gesta Normannorum Ducum’, 110, Dudo is writing Gesta history or serial biography; Albu [Hanawalt], ‘Dudo of Saint-Quentin’, 111–18, argues for Dudo as a parodist of Virgil; see also above, n. 3, and Antonella Degl’Innocenti, ‘Bibliografia su Dudone de San Quintino’, in Dudone di San Quintino, 209–18.

26

Anglo-Norman Studies XXVI

was both militarily strong and economically thriving. In short, they acted in an appropriate manner as seen by Dudo’s audience and, as a result, God had given them victory.29 Dudo’s factual inaccuracies of the type discussed above, unless identified in substantial numbers, likely were not the types of errors that would create a level of cognitive dissonance sufficient to undermine the rhetorical plausibility of his story as a whole. Indeed, it was not until the mid-nineteenth century or thereabouts, that scholars began to grasp the magnitude of Dudo’s likely falsification of fact. In a similar vein, modern scholars have found great difficulty in trying to identify Dudo’s grand scheme, if he had one, whether provided by Virgil, the Bible, hagiography or some eclectic combination thereof, which seems more probable. These difficulties experienced by highly educated medievalists may well sustain the inference that it is rather unlikely that many in Dudo’s audience grasped his grand scheme, whatever it may have been, either wholly or even in large part. Thus, Dudo’s listeners were unlikely to have been in a position to take issue with such a grand scheme, if it existed, either on factual or on philosophical grounds. Indeed, those medieval writers, e.g. William of Jumièges, Robert of Torigny, Wace, and Benoît of St Maure, who, as noted above, used De Moribus, either plundered it for ‘facts’ and/or copied from it wholesale. They have given us little or no reason to believe that they appreciated whatever grand scheme or schemes Dudo pursued much less that these putative models were the subject of serious objection. The type of inaccuracy that could undermine rhetorical plausibility in the medieval context, as specialists generally acknowledge, was unlikely to have been a grand intellectual scheme that was difficult to apprehend. Similarly, largely unverifiable inaccuracies, that were passed off as fact, probably would not attract the attention of Dudo’s audience. By contrast, rhetorical plausibility can be undermined, it was understood then as now, if a substantial corpus of information about which the audience is fundamentally secure in its knowledge is handled in a manner that attracts focused negative attention. With apologies to Gilbert and Sullivan, an air of realistic verisimilitude is given even to a potentially bald and unconvincing narrative, which Dudo’s De Moribus clearly is not, when familiar or well known matters are treated in a way that does not raise suspicion of massive falsification or consistent inaccuracy. To put it another way, when the furniture on the stage is familiar, the audience is less likely to question the truth or accuracy of actions on stage that otherwise are unknown or less well known.30 Dudo’s academic training in rhetoric at Liège no doubt prepared him to understand the need to write plausibly in relation to what he knew to be the level of information that was available to his audience. For example, Emily Tabuteau has shown that Dudo represents matters of law or custom in regard to land transfers and their attendant obligations with considerable accuracy.31 As a landholder in Normandy 29 Christiansen, Dudo, xxiii, would seem to go too far in suggesting that Dudo disguised and dignified

Norman history ‘beyond recognition’ by contemporaries. Indeed, as noted above, not only were Dudo’s inaccuracies often accepted but so was his general picture whatever contemporaries made of it. See Lifshitz, ‘Dudo’s Historical Narrative’, 101–20, who observes (p. 118): ‘All evidence for the reign of Richard I bespeaks a region in which any western Frank would have felt at home . . .’. This was true, she avers, in economy, administration, law courts, toll collecting, and ecclesiastical matters among others areas of life. 30 Cf. Suzanne Fleishmann, ‘On the Representation of History and Fiction in the Middle Ages’, History and Theory 22, 1983, 278–310. 31 Transfers of Property in Eleventh-Century Norman Law, Chapel Hill 1988, 12 and n. 39 (p. 383); Tabuteau cites the traditional wisdom: ‘Dudo . . . is extremely untrustworthy . . .’. However, she goes on to

Dudo of St Quentin and Norman Military Strategy

27

and perhaps as cancellarius and/or capellanus for Duke Richard II, Dudo had the opportunity to obtain accurate knowledge in regard to these legal matters.32 In a similar vein, Annie Renoux has shown that Dudo provides a detailed and accurate physical description of the Norman court at Fécamp. These details, according to Renoux, have been confirmed by archaeological excavations.33 The examples noted by Tabuteau and Renoux may perhaps be taken to demonstrate the practical results of Dudo’s adherence in De Moribus to a program of rhetorical plausibility.34 No less significant in this context is Dudo’s emphasis on the crucial role played by the English and their king in the success enjoyed by Rollo and his men.35 We, of course, do not know whether this was true and in all likelihood it was not.36 However, Dudo, who spent many years at the Norman court, certainly was aware of the positive relations between the rulers of Normandy and the kings of the English during the late tenth and early eleventh centuries. The Treaty of Rouen agreed to by the envoys of Aethelred II and Duke Richard I, initially was confirmed by the English king and then on 1 March 991 by the Norman ruler.37 In 1002, the effort to maintain positive relations between England and Normandy was strengthened by the marriage of Emma, the sister of Richard II, Richard I’s son and successor, to King Aethelred.38 These major diplomatic accommodations may well have provided a certain credibility to Dudo’s adumbration of strong and positive relations between the English and ancestors of the present Norman rulers a century earlier.39 The interest of Dudo’s audience at the Norman court in preparation for war, warfare, and the aftermath of war appears obvious from his extended focus on military matters in De Moribus. In addition, the ducal family’s patronage of Dudo’s work surely suggests that the aristocratic values of the Norman court, with a focus on politics and war, would be given not only extensive but accurate attention.40 To put the matter simply, Dudo had to get contemporary military matters right in order to keep demonsrate in considerable detail (pp. 51–5, 61–2, 103, 198, 302–3, 330–1) that in regard to the legal or institutional aspects of property transfers he is by and large quite trustworthy. 32 Concerning the possibility that Dudo was cancellarius, see Tabuteau, Transfers of Property, 8, 275 n. 12; for his land holding see Shopkow, History and Community, 36–7; and as chaplain, see Christiansen, Dudo, xi. Regarding Dudo’s close knowledge of contemporary documents see M. Fauroux, ‘Deux autographes de Dudon de Saint-Quentin (1011, 1015)’, Bibliothèque de l’Ecole des Chartes 111, 1953, 229–34. The basic treatment of Dudo’s life remains the unpublished doctoral dissertation of Barbara Vopelius- Holtzendorff, ‘Dudo von Saint-Quentin: der erste Geschichtsschreiber der Normandie, 987–1017’, Göttingen 1967, whose argument (pp. 98–9) that Dudo headed the cathedral school at Rouen is not strongly supported. She is also likely wrong in arguing (pp. 25–82), that Dudo was educated at Rheims under Gerbert. See Shopkow, ‘The Carolingian World of Dudo’, 24–7, for the more plausible argument that Dudo was educated at Liège. 33 ‘Fouilles sur le site du château ducal de Fécamp (Xe–XIIe siècle)’, ANS 4, 1981 (1982), 138–9, 144. 34 See also van Houts, ‘Scandinavian Influence’, 108–11, who identifies several examples of Dudo’s putative reliability accepted by both Prentout and Steenstrup, cited above (nn. 1 and 3). 35 De Moribus, bk I, chs 5–8, 17–20 (pp. 132–8, 156–60). 36 It is hard to reconcile the relations of Alfred the Great and his immediate successors with the Vikings with the tale told by Dudo concerning Rollo. In this context, see the observations by Richard Abels, Alfred the Great: War, Kingship and Culture in Anglo-Saxon England, London 1998, 290. 37 Frank M. Stenton, Anglo-Saxon England, 2nd edn, Oxford 1962, 370–1; and Bates, Normandy before 1066, 37. 38 Bates, Normandy before 1066, 37, 67; and Stenton, Anglo-Saxon England, 374. 39 For additional background information with regard to English-Viking relations, see Theodore Andersson, ‘The Viking Policy of Aethelred the Unready’, in Anglo-Scandinavian England: NorseEnglish Relations in the Period before the Conquest, ed. John D. Niles and Mark Amodio, New York 1987, 1–12. 40 Graham A. Loud, ‘The Gens Normannorum, Myth or reality?’, ANS 4, 1981 (1982), 108, emphasizes the ‘secular, noble and courtly audience’ for whom Dudo wrote.

28

Anglo-Norman Studies XXVI

his intended audience from questioning other information that he provided that might be inaccurate. I have tested this hypothesis in a previous paper and it is clear Dudo treated the matter of military organization c.1000 in an accurate manner.41 However, getting the facts right in regard to military organization was a rather simple task for Dudo. He not only was close to the Norman ruler, the military commander par excellence of the duchy, but he had access to and used numerous contemporary and near contemporary historical works, e.g. the Annales of Flodoard, that provided accurate information regarding these matters.42 In this study, I will examine Dudo’s treatment of long term strategy, a rather more complex area of military thinking, in order to ascertain if here too he told a story that was likely to be plausible to his patron and the latter’s entourage.

Long Term Strategy During the earlier Middle Ages, writers in the regnum Francorum understood the notion of a long-term military strategy that was pursued over several decades and even longer. Indeed, it may be concluded with considerable certainty that neither these writers nor their audiences were inclined, prima facie, to find the idea of long-term strategy inherently absurd.43 It is clear, for example, that the author of the Annales Mettenses priores, who wrote for Charlemagne’s court early in the ninth century, identified a long-term strategy, which he alleges was initiated during the later seventh century by Pippin II. This long-term strategic effort is depicted in the Annales as having focused upon the reconstruction of the regnum Francorum under Carolingian dynastic rule as part of a Divine plan which was successfully executed by Charlemagne’s ancestors with God’s help.44 In a similar vein, Dudo identified a long-term military strategy that undergirded the creation of what was to become the Norman duchy. He argued, in effect, that Rollo, the likely founding ancestor of the present Norman ruling house, whose contemporary leaders patronized the writing of De Moribus, pursued a rather complicated long-term strategy. In Dudo’s view, however, this strategy is depicted as having evolved over a rather short period of time and to have had a two-fold goal. First, it was Rollo’s aim to establish his regnum in Gaul and the second was to assure that his dynasty would rule the region in perpetuum. As Dudo begins the story, Rollo and his younger brother Gurim lead a revolt in Dacia, the putative Urheimat of his people. However, their effort fails, Gurim dies in combat, and Rollo is forced into exile with the remnants of the rebel army which flee in six ships to the Island of Scanza.45

41 See Bernard S. Bachrach, ‘Dudo of St Quentin as an Historian of Military Organization’, HSJ 12

(2003), 165–85. 42 Regarding Dudo’s contemporaries who wrote history see the interesting remarks by Christiansen,

Dudo, xx. Concerning Flodoard’s treatment of military matters see the Introduction to the forthcoming translation of his Annales by Bernard S. Bachrach and Steven C. Fanning. In regard to the role played by Flodoard’s Annales in Dudo’s thinking see Emily Albu, The Normans and their Histories: Propaganda, Myth and Subversion, Woodbridge 2001, 21. 43 Bachrach, Early Carolingian Warfare, Philadelphia 2001, 3–4, 44–9, 206–7, 242–5. 44 Annales metteneses priores, ed. B. von Simson, MGH SRG X, Hanover 1895. The basic modern work on this text was done by Hartmut Hofmann, Untersuchung zur karolingian Annalistik, Bonn 1958, 9–68; and Irene Haselbach, ‘Aufstieg und Herrschaft der Karolinger in der Darstellung der sogenannten Annalen Mettenses priores’, Historische Studien 412, 1970, 1–208. For a disucssion of this text in the context of a long-term strategy see Bachrach, Early Carolingian Warfare, 47–8. 45 De Moribus, bk II, chs 1–5 (pp. 140–5).

Dudo of St Quentin and Norman Military Strategy

29

While in exile on Scanza, Rollo tries for some time to figure out what he is going to do in the near future. This thinking was geared, according to Dudo, toward carrying out what then was Rollo’s intention, i.e. to return to Dacia and to establish himself there as ruler. However, Rollo cannot figure out how to do this and Dudo claims that the Viking leader was informed in a dream that the first step in his plan to return to Dacia must begin with a journey to England.46 Upon landing in England with his war fleet, Rollo meets and defeats local opposition and then ponders three options. Should he go back to Dacia immediately? Should he launch an attack on Francia? Or should he carry out military operations in England? At this point, it is clear that Rollo’s long term goal regarding Dacia has not been abandoned but as yet no long term strategy had been decided. Dudo characterizes Rollo as anguishing over three possible options.47 However, it is important to emphasize that for the first time in Dudo’s presentation of the situation, Francia is depicted as appearing on Rollo’s horizon as an area for potential military operations. It is perhaps of some importance, in addition, that Dudo provides no reason for this augmentation of Rollo’s options. After laying out Rollo’s dilemma in brief, Dudo, acting like a chorus that is informed about the future (he uses poetry rather than prose for this purpose), asks a series of rhetorical questions. The answers to these questions indicate that Rollo should have no worries or doubts about the future. Because Rollo has survived the situation in Dacia, Dudo indicates that the Danish leader is ‘eternally Christian’, i.e. God is looking out for his wellbeing and success despite the fact that he is a pagan. In addition, Dudo indicates that Rollo is a patricius of exceptional merit, who ultimately, i.e. in heaven, will earn a crown greater than that of rule in Francia. Thus, as Rollo is depicted as brooding over what his long term strategy will be, Dudo is foreshadowing his decision both in terms of indicating that Francia is the goal and that ultimately the Norse leader will become a Christian. However, Dudo is also foreshadowing the result that this strategy, well known to God, that Rollo will not be a king but only a great noble, i.e. patricius, with substantial landed holdings.48 Rollo’s regnum, as the audience undoubtedly was expected to recognize, was to be the Norman duchy. At this point in Dudo’s narrative, Rollo receives the submission of the local English population in the region that he has invaded. Then the Danish leader has yet another dream that sets a more certain course for his long term strategy. In the vision of the dream, as interpreted for Rollo by one of the English Christians whom he has taken captive, he learns that his future lies in Francia. Indeed, in this vision, Rollo sees himself living in a house that is built in the Frankish style. The dream, as interpreted, sets out what will be Rollo’s long term strategy. He will gain control of extensive lands in Francia with large numbers of soldiers from different regions serving under his leadership. Indeed, special attention is given to restoring the walls of the fortress cities that have been destroyed. However, it is made very clear that Rollo must obtain the support of the church and he is told that in order to carry out this strategy of establishing himself and his dynasty in Francia, he will accept baptism.49 Dudo now has made clear that Rollo’s has decided that his strategic goal will be regnum in Francia and that in order to bring this plan to fruition he will become a Christian at some time along the way. At this point, yet another stage in Rollo’s strategy becomes evident. He arranges to become a supporter of the English king. 46 47 48 49

Ibid., bk II, ch. 5 (p. 145). Ibid., bk I, ch. 5 (p. 145). Ibid., bk II, ch. 5 (p. 145). Ibid., bk II, ch. 6 (pp. 146–7). See the discussion by Albu, The Normans and their Histories, 17–18.

30

Anglo-Norman Studies XXVI

The latter, in turn, makes available to Rollo the all that is necessary to sustain his army while in England and agrees to provide logistic support for the invasion of Francia. Throughout the winter, preparations are made with the help of the English and, it is important to note, Rollo is given permission by the king to recruit additional fighting men from among his subjects in England. Clearly, as Dudo’s audience understood, the crews of the six ships that fled Dacia would be inadequate for the purpose of making significant territorial conquests in Francia much less to take control of the regnum Francorum as a whole.50 When the time is appropriate, Rollo’s much reinforced flotilla sails for the continental mainland and attacks the wealthy region of Walcheren. As the invasion waxes, two dozen additional ships, twelve troop carriers and twelve supply transports, arrive from England to support Rollo.51 As presented by Dudo, Rollo’s success in the Walcheren region and beyond into greater Flanders, would appear to have assured the Danish leader that he could undertake successful military operations against the native population of Francia. However, these victories in the northeast would not get him a new homeland in which to settle. Indeed, Dudo makes clear that this region does not conform to the vision that Rollo had in his dream of the place where he would impose his regnum.52 Thus, as Dudo tells the story, Rollo moved west, sailed into the mouth of the Seine and encamped on the bank oposite the monastery of Jumièges. When the people of the nearby city of Rouen learned of the landing, they sent Bishop Franco to deal with Rollo. The latter, in his turn, went to scout out Rouen and its region. It is in this context that Dudo suggests that Rollo decided on settling his people in the Rouennais because the region conformed to the vision that he had experienced earlier.53 What is important here is that this long term strategy appears to Rollo in dreams and that these dreams, or perhaps more accurately, visions within dreams, clearly have a Christian underpinning. This permits Dudo’s audience to conclude that God has inspired this plan. First, from a religious perspective, Rollo’s military adventures, when properly focused on establishing his regnum in a part of Francia, are justified or just in as much as they adhere to God’s plan. One might perhaps even go so far as to suggest that whatever wars Rollo pursued in this context were, in fact, just wars.54 Secondly, and in a more prosaic vein, the dreams provide a way for Rollo to have obtained the needed intelligence regarding places, peoples, and potential military efforts. As a result, Dudo does not have to provide human links, i.e. specific factual information that he did not have and which might arouse suspicion of inaccuracy among his listeners if he concocted such links. From Dudo’s approach to this topic, it is clear that he expected his audience to believe that useful and accurate information of a military nature was conveyed by God in dreams or visions and that such a method of transmission was not prima facie implausible to early eleventh century Christians.55 50 With regard to the problem of numbers see Bernard S. Bachrach, ‘Early Medieval Military Demogra-

phy: Some Observations on the Methods of Hans Delbrück’, in The Circle of War, ed. Donald Kagay and L. J. Andrew Villalon, Woodbridge 1999, 3–20. 51 De Moribus, bk II, ch. 9 (pp. 149–50). 52 Ibid., bk II, ch. 10 (pp. 34–5). 53 Ibid., bk II, ch. 10 (pp. 34–5). 54 Frederick H. Russell, The Just War in the Middle Ages, Cambridge 1975, remains the best introduction to this subject. 55 This obviously is not the place to examine early medieval ideas regarding dreams or the cognitive process by which people formulate ideas in dreams from information that they have acquired while conscious. In this context, however, see the illuminating observations of Paul Edward Dutton, The Politics of Dreaming in the Carolingian Empire, Lincoln, Nebraska, 1994.

Dudo of St Quentin and Norman Military Strategy

31

Once Rollo’s long term strategy is adumbrated, the nature of the information provided by Dudo and the problems involved in the execution of the military operations, that were needed to sustain this effort, are made a great deal more specific. However, Dudo strengthens the key point regarding the choice of Rouen as the strategic goal and has Rollo ask his close advisers what he should do. These men, Dudo makes clear, replied ‘as though they had foreknowledge of future events (quasi futuroram praescii)’ and he suggests that God provided the inspiration that gave Rollo’s advisers this foreknowledge (divinaeque inspirationis praesagio imbuti).56 As a result of their presumed divinely inspired foreknowledge, Dudo has these men reply to Rollo in a strong affirmation of his long term strategy. They emphasize the economic value of the land with regard to its fertility and the abundance of fish and game.57 Then the advisers echo what Rollo has already learned from his reconnaisance and they indicate that the region lacks first class military men (milites) and even well armed support troops (armigeri).58 The Danish advisers continue in their reply to Rollo, and Dudo, through direct discourse, outlines the strategy of conquest by indicating the need to capture the oppida, vici, castra, and urbes of the pagus.59 Dudo clinches his point regarding the strategic significance of the dream when he concludes the advisers’ reply to Rollo by having them observe that: ‘Perhaps the interpretatio of your visio has directed us to the land within these very borders (fines).’60 Dudo’s description of the military situation at Rouen provides crucial insights into what he wanted his audience to believe was important to Rollo when the latter made his decision to establish himself in the region. First and foremost, Rollo is depicted as reinforcing the vision that he received in his dream by having ascertained through a reconnaisance that there were no highly trained military forces established in the city of Rouen or in its environs. He is reported by Dudo to have learned that the region lacked milites and armigeri, as noted above.61 In fact, the only armed people in the region were the local levies of lightly armed militia men, i.e. inermes or men who lacked armor. The garrison within the urbs itself is considered to have been small and also composed of inermes.62 Now that Rollo’s goal of identifying a homeland for the future has been fixed, his strategy becomes more focused. First, Dudo makes it very clear that Rollo understood that he must subdue the region, he simply could not appear on the scene and claim the Rouennais without undertaking a serious military effort. This conquest involved a series of military operations. However, the reported lack of readiness of 56 De Moribus, bk II, ch. 12 (p. 153). 57 Ibid., bk II, chs 12 and 26 (pp. 153, 166). Loud, ‘The Gens normannorum’, 108, argues that this is a lit-

erary topos because eighty years later Amatus of Montecasino averred that the Norman lands in Italy were filled with milk and honey. However, Loud refers only to ch. 26 and ignores ch. 12 of Dudo’s account which is very balanced. 58 De Moribus, bk II, ch. 12 (p. 153). The translation of milites by the word ‘knights’ by Christiansen, Dudo, 36, is a serious anachronism and thus very misleading. Regarding this problem see Bernard S. Bachrach, ‘Magyar-Ottonian Warfare: à-propos a new minimalist interpretation’, Francia 13.1, 2000, 218. 59 De Moribus, bk II, ch. 12 (p. 153). 60 Ibid., bk II, ch. 12 (p. 153). 61 Ibid., bk II, ch. 12 (p. 153). 62 Ibid., bk II, chs 11 (pp. 152–3). Christiansen, Dudo, 35, translates inermes as ‘weaponless’. This is misleading, especially when used to denote men serving as members of an urban garrison. Concerning the use of the term inermes to mean lacking in armor see Bernard S. Bachrach, ‘The Northern Origins of the Peace Movement at Le Puy in 975’, Historical Reflections/Reflexions Historiques 14, 1987, 405–21 and reprinted with the same pagination in Bernard S. Bachrach, State Building in Medieval France: Studies in Early Angevin History, London 1995, 417, with the literature cited there.

32

Anglo-Norman Studies XXVI

the men of the Rouennais to defend the region would appear to have been somewhat exaggerated. Thus, Dudo depicts Rollo and his advisers as making clear to the troops that they must win control through battle of the fortified places, i.e. oppida, vici, castra, and urbes, throughout the region.63 Clearly, the putative lack of first class soldiers (milites) and perhaps even second class troops (armigeri) in the region during this period would not result in a walkover by the Danish invaders. Rather, the militarized general population or, as modern scholars refer to them, the local levies of each pagus, i.e. the analogue of the English ‘great fyrd’, had to be taken into consideration and this was especially the case when the capture of fortications was at issue.64 For obvious reasons, poorly armed and comparatively poorly trained local levies likely were to be more effective in a defensive deployment by shooting arrows or crossbow bolts and throwing spears and other missiles at invaders from the protection of the walls of oppida, vici, castra, and urbes than in battle in the open field. In this regard, Dudo makes the point regarding the military value of local levies on several occasions in the context of sieges where, despite their lack of heavy armament, the militia, i.e. the cives, living within the walls of a city, were able to operate effectively. Dudo wants his audience to understand that both he and Rollo fully appreciated the importance of siege warfare. He does this especially through a detailed account of the defense of erstwhile Roman fortress cities, for example, Paris, Bayeux, and Chartres, in order to make this point unambiguously. Indeed, in these cases the Danish besieging force either failed totally or, at least initially, to take its objective.65 Whether or not Rollo and his successors at Rouen had pursued a long-term strategy of military conquest as adumbrated in De Moribus is certainly problematic. However, by the time Dudo wrote, the Norman duchy was a fact of political life and exactly how it came into being, much less whether Rollo had a long term strategy, certainly was not well known.66 Thus, the potential focus of criticism by Dudo’s contemporaries regarding his representation of Norman strategy would likely not have been in regard to its long-term aspects. However, Dudo could have left his presentation on this matter open not only to criticism but even to rejection if he had treated the details of the strategy of conquest in a manner that those men at the Norman court who were knowledgeable in military matters were likely to see as seriously flawed. In short, if Dudo wanted his audience not to question his picture of a long-term Norman military strategy of conquest, he had to provide, at the least, a realistic or rhetorically plausible appreciation of how such efforts generally were executed as his contemporary audience understood these matters. Verisimilitude in regard to what was known was and still is presumed to have the effect of assuaging possible doubts concerning the presentation of what was or is not known.

63 De Moribus, bk II, ch. 12 (p. 153). 64 Ibid., bk II, ch. 12 (p. 153). Concerning the ‘local levy’ see Bachrach, Early Carolingian Warfare,

52–4. 65 Dudo De Moribus, bk II, ch. 17 (p. 156) for Paris; bk II, ch. 23 (p. 162), for Chartres, and for Bayeux,

bk II, ch. 16 (p. 157). 66 Indeed, the origins of the duchy remain obscure and this is the case in large part because of Dudo’s failure to provide the kind of information that historians require. See, for example, the discussion by Bates, Normandy before 1066, 2–13.

Dudo of St Quentin and Norman Military Strategy

33

Siege Warfare As modern specialists in medieval military history are well aware, wars, and more particularly wars of conquest, were dominated during the Middle Ages by sieges.67 It is important to emphasize that the decades on either side of the millennium were not an exception.68 This is very well demonstrated by the strategy employed in the western parts of Francia Occidentalis by Fulk Nerra, count of the Angevins 987–1040, who earned the sobriquet le grand bâtisseur, from posterity.69 In general, it is noteworthy that substantial territorial conquest was accomplished by the capture of the great fortress cities that had been constucted during the later Roman empire.70 Smaller fortifications, e.g. castra and castella, also were the subject of sieges.71 Throughout much of the earlier Middle Ages, such comparatively minor strongholds often were built as strategic satellites of the cities, which, in general, were the focuses of political, religious, and economic power, the places from which the civitas or pagus was usually governed.72 In the light of Dudo’s diligent efforts to provide a plausible depiction of Norman military organization, modern readers hardly should be surprised that he describes military strategy in a manner that contemporaries would both recognize and accept as accurate. Indeed, Dudo provides almost a mantra-like cadence in his discussion of the strategy of conquest. He has various of the dramatis personae espouse the idea: ‘Invade et obside, atque cape . . . civitatem’;73 ‘obsidere et capere . . . urbes’;74 ‘capiam urbes et castra, atque munitissima loca’.75 The conqueror is described as ‘obsidens, et capiens urbes’ and as a result, he is ‘reported’ to have subjugated the populus.76 Dudo has Rollo besiege many cities, urbes and civitates, both in England and in Francia.77 Dudo also recognizes the importance of lesser strongholds as he mentions not only castra and oppida but also has those undertaking efforts at 67 Jim Bradbury, The Medieval Siege, Woodbridge 1992; and Bernard S. Bachrach, ‘Medieval Siege Warfare: A Reconnaissance’, JMH 58, 1994, 119–33; and reprinted in Bernard S. Bachrach, Warfare and Military Organization, with the same pagination. 68 See, for example, Bernard S. Bachrach, ‘Fortifications and Military Tactics’, Technology and Culture 20, 1979, 531–49, with the scholarly literature cited there; and from a more technical perspective see Bernard S. Bachrach, ‘Early Medieval Fortifications in the “West” of France: A Technical Vocabulary’, Technology and Culture 16, 1975, 531–69. 69 Bernard S. Bachrach, ‘The Angevin Strategy of Castle-Building in the Reign of Fulk Nerra, 987–1040’, American Historical Review 88, 1983, 533–60; and in a broader context, Bernard S. Bachrach, Fulk Nerra the Neo-roman Consul, 987–1040: A Political Biography of the Angevin Count, Berkeley– London 1993. 70 Bernard S. Bachrach, ‘Early Medieval Europe’, in War and Society in the Ancient and Medieval Worlds: Asia, the Mediterranean, Europe, and Mesoamerica, ed. Kurt Raaflaub and Nathan Rosenstein, Cambridge, MA, 1999, 271–307; and with specific focus on the great fortress cities see Bernard S. Bachrach, ‘Imperial Walled Cities in the West and their Early Medieval Nachleben’, in City Walls: The Urban Enceinte in Global Perspective, ed. James D. Tracy, Cambridge 2000, 192–218. 71 See, for example, Bernard S. Bachrach, ‘Military Organization in Aquitaine under the Early Carolingians’, Speculum 49, 1974, 1–33 and reprinted with the same pagination in Armies and Politics in the Early Medieval West, London 1993. 72 Bernard S. Bachrach, ‘On Roman Ramparts, 300–1300’, in The Cambridge Illustrated History of Warfare: The Triumph of the West, ed. Geoffrey Parker, Cambridge 1995, 64–91; and Bachrach, ‘Early Medieval Europe’, 271–307. 73 De Moribus, bk IV, ch. 110 (p. 273). 74 Ibid., bk IV, ch. 107 (p. 270). 75 Ibid., bk I, ch. 2 (p. 142). 76 Ibid., bk I, ch. 4 (p. 143). 77 Dudo, Ibid., bk II, chs 18, 19 (pp. 158–60), concerning England; and bk II, ch. 22 (p. 162): Rollo ‘civitatem Carnotis . . . cum magno exercitu obsedit’. Regarding the siege of Chartres see Prentout, 191–6; and Jules Lair, Le siège de Chartres par les Normands, Caen 1902.

34

Anglo-Norman Studies XXVI

conquest aver: ‘occupemus castra et oppida illorum’,78 or the Norman leader in the process of conquest ‘subvertit plurima loca munitionum’.79 A Viking force is described as ‘invadens saepissime plurima castra’.80

Diplomacy More examples can be adduced to demonstrate that Dudo’s depiction of the strategy of territorial conquest was focused on the capture of enemy strongholds through the use of siegecraft. In fact, this was a fundamentally accurate appreciation of the means by which territories were conquered in early medieval Europe. However, the strategy of conquest depicted by Dudo, in his pursuit of providing a plausible military setting for his story, is more complex than the mantra ‘invade et obside, atque cape’. Rather, he makes clear that commanders often engaged in both diplomatic and military preliminaries for the purposes of gaining control of enemy cities and lesser strongholds which were necessary to assure territorial conquest. This was done, in large part, to avoid establishing lengthy and costly sieges that might fail.81 Indeed, the very first encounter between Rollo’s army and the forces of the French king begins with a diplomatic prelude. According to Dudo, information concerning Rollo’s landing in the lower Seine reaches the royal government which sends a legation to learn the purpose of this incursion. The Frankish delegates offer to provide Rollo’s men with extensive grants of land to sustain them and their families if they are willing to enter royal service and recognize themselves as subjects of the king.82 The Danes refuse and build a fortified encampment on the banks of the Seine, which the Franks attack unsuccessfully. This provides Dudo with the opportunity to claim that the Danes have been attacked unjustly and to justify their subsequent military operations in the Frankish kingdom.83 These military operations undertaken by Rollo’s forces lead to a number of victories and some setbacks. However, at this point in the narrative, Dudo leaves the impression that Rollo gradually focuses his efforts on territorial conquest as he plans to bring to the negotiating table a chastened Frankish king, whose forces not only have been defeated in the field but also have seen important fortifications taken by the Danes. Thus, Rollo is seen first to attack targets of opportunity, e.g. Meulan and Paris.84 However, as the enemy’s resistence is weakened, especially after a defeat in the field, Rollo turns his attention to Bayeux and Évreux which ultimately would become part of Normandy and were then part of the ecclesiastical province of Rouen.85 At the same time, Rollo imposes tributes upon various regions into which he brings his army.86 However, as Dudo tells the story, Rollo appears to be in need of more troops in order to bring his plans in Francia to fruition and thus he obtains permission from the English king to recruit fighting men in England. Dudo indicates

78 79 80 81

De Moribus, bk II, ch. 14 (pp. 155–6). Ibid., bk III, ch. 41 (p. 185). Ibid., bk III, ch. 54 (p. 198). With regard to negotiations see, for example, concerning truces De Moribus, bk II, ch. 20 (p. 160); bk II, ch. 26 (p. 167); bk IV, ch. 119 (p. 282). 82 Ibid., bk II, ch. 13 (pp. 154–5). 83 Ibid., bk II, ch. 14 (p. 156). 84 Ibid., bk II, ch. 16 (p. 157). 85 Ibid., bk II, ch. 16 (p. 157). 86 Ibid., bk II, ch. 16 (pp. 157–8).

Dudo of St Quentin and Norman Military Strategy

35

that a force ‘too numerous to tell’ thus joined Rollo’s army.87 Here, of course, Dudo, who is not writing epic but must maintain rhetorical plausibility in an historical genre, does not want to use huge and unbelievable figures which would cast doubt on his narrative. However, he does want everyone to believe, in consonance with the realities of contemporary affairs, that the English king gave overwhelming support to Rollo at crucial points along the latter’s road to success. With his army now reinforced by a substantial force of recruits from England, Dudo reports that Rollo divided his forces and sent units to ravage various parts of the regnum Francorum while he led a contingent to harass the Paris region.88 As a result of this many pronged offensive, Dudo indicates that the Franks decided to negotiate and as a result requested a truce. It seems clear that it was Rollo’s strategy to bring the Frankish king to negotiate by undermining his ability to protect and govern his realm. Indeed, the effectiveness of Rollo’s campaign is recognized and Dudo has the king observe in direct discourse, ‘The kingdom that I rule has been devastated. The cultivators are not plowing the land, the res publica is being held captive, and every day I lose more and more men, killed and captured.’ He then sends Bishop Franco to Rollo to negotiate a three month truce which the latter grants on the prelate’s promise that a final agreement would be very much to the benefit the Danes.89 Rollo’s strategy would seem to have been working. However, after the three month truce ended, the Franks, rather than grant the Danes a fair settlement, again went to war. This phase of the conflict ended, however, with yet another round of devastation visited on the Frankish realm by Rollo’s troops. As Dudo describes the situation, the king’s honor and his power were undermined. The land has been reduced to an unproductive desert, much of the population either had been put to the sword or died as a result of famine, while still others have been sold into slavery. Finally, Dudo makes clear that the Franks finally got Rollo’s message. Thus, the king is advised to ‘make peace through negotiation’.90 As Dudo’s story progresses, the king agrees with this new strategy of negotiation and his advisers map out a broad geographical area which is to be offered to Rollo. Fundamentally, Dudo’s description is of the area that ultimately will form the basis for the Norman duchy. As a result, it may be hypothesized that his audience will readily accept the accuracy of the story. In addition, Dudo indicates that the king is advised to arrange to have his daughter Gisla marry Rollo and the latter is to become a Christian. Finally, King Charles promises that an offspring of this union between Gisla and Rollo will exercise regnum over this region in perpetuity.91 When Rollo brings this offer to his advisers, they recognize, according to Dudo, that it fits the long term strategy that had been made manifest by the interpretation of the dream.92 Conclusion As Dudo, c.1000, tells the story of the Normans’ acquisition of their duchy, he attributes a long term strategy to Rollo. This strategy is depicted as having appeared to Rollo in dream-visions which begin after his flight from Scanza. The reader is led to 87 88 89 90 91 92

Ibid., bk II, ch. 19 (p. 160). Ibid., bk II, ch. 20 (p. 160). Ibid., bk II, ch. 20 (p. 160). Ibid., bk II, ch. 25 (p. 165). Ibid., bk II, ch. 25 (p. 166). Ibid., bk II, ch. 26 (p. 167).

36

Anglo-Norman Studies XXVI

believe that these dreams were inspired in the pagan Dane by God, i.e. the Christian God, whose foreknowledge of events made clear that Rollo would rule in a part of Francia and that his descendants would rule the duchy in perpetuity. But, in addition and very importantly, Rollo would become a Christian and his successors would be loyal and pious Christians. In this context, two points require emphasis from a military perspective. First, the notion of a long term strategy of territorial conquest was an idea that would seem to have been plausible to Dudo’s audience at the Norman court. Whether, William the Conqueror, two generations later, thought in such terms in regard to his relations both with Edward the Confessor and Harald Godwinson is a subject that may be worthy of serious consideration. Secondly, it seems that Dudo assumed that his audience believed that information of value, indeed, even information of military value, could be transmitted to a worthy person by God through the medium of a dream. Constantine’s dream prior to the Battle of the Milvian Bridge, which certainly was widely known, may well have been something of a locus classicus in regard to dreams or visions of this type. In addition to providing good reason to believe that there was a general acceptance among military men c.1000 that long-term strategic thinking was part of the Norman armoury, Dudo also provides an accurate understanding regarding the strategy that undergirds territorial conquest. Consistent with early medieval writers who make clear the centrality of siege warfare, Dudo provides a mantra that territorial conquest is dominated by siege warfare, e.g. ‘Invade et obside, atque cape . . . civitatem’, ‘obsidere et capere . . . urbes’, and ‘capiam urbes et castra, atque munitissima loca’. Finally, Dudo demonstrates his understanding that in the complicated milieu of conquest, where the result of military success requires legitimization, negotiations are necessary in addition to taking control of the enemy’s territory and the fortifications that are used to defend it. Thus, Dudo makes clear that through a process of harrying the land and undermining the king’s regnum, Dudo brought Charles to the negotiating table. This, according to Dudo, not only resulted in a treaty which defined the boundaries of Normandy but also assured to Rollo’s successors perpetual possession of the region. As modern scholars have long emphasized, Dudo may have known little of the early history of his patrons’ family in regard to their political and military efforts. It is clear, in addition, that he introduced inaccuracies into his narrative in this regard. However, it also seems likely that the story he told to his patrons and their supporters was accepted, in general terms, and that there was little opposition or revision introduced until modern times. As David Bates put it, ‘Dudo of St-Quentin was as ready as anyone to present the past in terms of the conditions which existed in his own day.’93 It has been argued here that the details of military operations in regard to contemporary ideas about strategy c.1000 helped to provide the realistic verisimilitude which made convincing an otherwise unsupported narrative. While this may perhaps help us to understand why many of the stories detailed in De Moribus went unchallenged for so long, it also gives the modern researcher confidence regarding not only Dudo’s understanding of an important aspect of warfare in pre-Crusade Europe but also permits the inference that his audience shared this understanding.

93 Normandy before 1066, xiv.

Clergy in the Diocese of Hereford

CLERGY IN THE DIOCESE OF HEREFORD IN THE ELEVENTH AND TWELFTH CENTURIES Julia Barrow In the 1140s Bishop Robert de Béthune of Hereford issued a charter adjudging two churches to the Norman abbey of Lire and its abbot Hildier who had come all the way to Hereford to press claims to them. After explaining the validity of Lire’s claim to the church of Fownhope, Robert went on to say ‘Similarly I grant to [Hildier] the church of Much Marcle, anciently subjected to Lire, in which in my time I found two priests, one having a concubine and the other not only having a concubine but also being a simoniac, which was recognised and proved in my presence and that of my chapter.’ Robert had the two priests ejected and allowed the abbey of Lire to take direct control of the church.1 The document spotlights two clerics being forced to come to terms with changing attitudes among the ecclesiastical authorities to clerical marriage and to the ownership and transmission of churches. How typical were they of the clergy of the diocese of Hereford in the twelfth century? and, in wider terms, what sort of people became clerics in the diocese in the eleventh and twelfth centuries? which families did they belong to? how were they educated? how did they obtain preferment? how prevalent was clerical marriage and how did it help them to fit into local networks? Posing these questions about the diocese of Hereford in the eleventh and twelfth centuries, the period when the Gregorian reform movement was beginning (rather slowly, it must be said) to have an impact, brings into play two separate areas of historical enquiry. One of these is a long-standing interest in the prosopography of the medieval clergy. Clerics, who had a much clearer and much more structured career pattern than laymen in the middle ages, are an obvious group to isolate and study. In particular work has been done to collect materials: hence Emden’s Biographical Registers for the universities of Oxford and Cambridge, and series like the new version of Fasti Ecclesiae Anglicanae (for English cathedrals) and Fasti Ecclesiae Gallicanae (for French ones).2 The collection of information is only the starting point of prosopography: information once collected can then be used to answer questions by sifting through individual entries to trace underlying trends: what can we learn of, for example, family

1 2

EEA, vii, Hereford 1079–1234, ed. J. Barrow, Oxford 1993, 32–3, no. 34. A. B. Emden, A Biographical Register of the University of Oxford to c.1500, 3 vols, Oxford 1957–9; idem, A Biographical Register of the University of Cambridge to c.1500, Cambridge 1963; John le Neve, Fasti Ecclesiae Anglicanae 1066–1300, new edition by D. E. Greenway and others, 9 vols, London 1968– , in progress; John le Neve, Fasti Ecclesiae Anglicanae 1300–1541, new edition by J. M. Horn and others, 12 vols, London 1962–7; Fasti Ecclesiae Gallicanae 1200–1500, general ed. H. Millet, 6 vols to date, Turnhout 1996– , in progress. For monastic cathedrals, see Joan Greatrex, Biographical Register of the English Cathedral Priories, c.1066 to 1540, Oxford 1997. The compiling of prosopographical information about the clergy is easier for the period after c.1200, for which much fuller royal and papal records are available.

38

Anglo-Norman Studies XXVI

background, training or patronage? which clerics were best placed to become bishops? How many clerics were priests, or likely to become such? How many clerics moved between dioceses, and were, for example, clerics coming from outside the diocese of Hereford more likely than those born within it to obtain the best preferment within the diocese? (I fear I am not going to answer these questions fully in this paper – I am merely setting them up as examples of what a prosopographical approach can help to elucidate.) The other strand of scholarship which helps us to set the context for our exploration of Hereford clergy is the study of the development of the parish. In parochial terms Hereford shows the normal pattern of progression from a fairly thin network of mid-Anglo-Saxon minster churches with large parishes to numerous small parishes, usually serving not much more than one village, by the mid-twelfth century. Although the diocese has not been fully surveyed in this respect, work has been done by John Blair on the parish of Bromfield in southern Shropshire, by Brian Kemp on the very large minster parish of Leominster, by Jane Croom on the parishes of south-eastern Shropshire and by Steve Bassett on the parishes on the border of the dioceses of Hereford and Lichfield.3 This line of research bears more indirectly than the prosopographical one on the lives of the medieval clergy, because it is topographical in approach – the discovery of the network of the minster churches of the Anglo-Saxon period and their large ancient parishes, essentially by working backwards through twelfth-century charters to look for evidence of lesser churches owing obligations to senior ones (for example, the retention by the latter of burial rights). The subordinate churches are overwhelmingly likely to be younger foundations, set up in the eleventh and twelfth centuries – mostly in the twelfth century, Herefordshire and Shropshire being, in comparison with more easterly areas of England, slow to see a burgeoning of small, village-sized parishes.4 (The pattern of settlement in the area, which was often dispersed rather than in the form of nucleated villages, is of course one of the reasons behind this.5) This study of parishes and settlements helps us to understand the lives and career patterns of the clergy because it provides the framework within which we can place them. It is now necessary to say something in more detail about the diocese of Hereford itself. The diocese of Hereford as it existed in the high middle ages (as we see it for example in Pope Nicholas IV’s Taxatio of c.12916) consisted of (1) southern Shrop3

On Bromfield: John Blair, ‘Secular Minster Churches in Domesday Book’, in Domesday Book: a Reassessment, ed. Peter Sawyer, London 1985, 104–42 at 128–31; on Leominster: B. R. Kemp, ‘Some Aspects of the Parochia of Leominster in the 12th Century’, in Minsters and Parish Churches ed. J. Blair, Oxford 1988, 83–95; on south-eastern Shropshire: Jane Croom, ‘The Fragmentation of the Minster Parochiae of South-East Shropshire’, in ibid., 67–81; on the border between the dioceses of Hereford and Lichfield: S. Bassett, ‘Medieval Ecclesiastical Organisation in the Vicinity of Wroxeter and its British Antecedents’, Journ. BAA 145, 1992, 1–28; idem, ‘Church and Diocese in the West Midlands: the Transition from British to Anglo-Saxon Control’, in Pastoral Care before the Parish, ed. John Blair and Richard Sharpe, Leicester 1992, 13–40. In general, see John Blair, The Church in Anglo-Saxon Society, Oxford forthcoming. 4 This is discussed, with reference to Herefordshire, by David Parsons, ‘Early Churches in Herefordshire: Documentary and Structural Evidence’, in Medieval Art, Architecture and Archaeology at Hereford, ed. David Whitehead, BAA Conference Transactions 15, Leeds 1995, 60–74. See also Blair, Church in Anglo-Saxon Society. 5 In general, see Margaret Gelling, The West Midlands in the Early Middle Ages, Leicester 1992, 182. For an example of dispersed settlement in Worcestershire but near the border of Herefordshire, see Christopher Dyer, ‘Dispersed Settlements in Medieval England. A Case Study of Pendock, Worcestershire’, Med. Arch. 34, 1990, 97–121. 6 Taxatio Ecclesiastica Angliae et Walliae Auctoritate P. Nicholai IV, circa 1291, ed. T. Astle, S. Ayscough and J. Caley, Record Commission, London 1802, 157–77. The Taxatio can also be con-

Clergy in the Diocese of Hereford

39

shire, essentially all the area south and west of the Severn, though excluding the very large ancient parish of Condover, which lay to the south of the river, and including some lands lying to the north of the river associated with Much Wenlock (2) the north-western corner of Worcestershire, containing part of the valley of the Teme (3) most of Herefordshire, but not Ewias in the far south-west of the county, which formed part of the diocese of St Davids, and (4) the part of Gloucestershire on the far side of the Severn estuary eastwards as far as the River Leadon – the Forest of Dean, in short. With the exception of Archenfield (the south-western corner of Herefordshire immediately to the south of Hereford), which was absorbed into the diocese only late in the eleventh century (the Book of Llandaff appears to date Llandaff’s loss of Archenfield to the reign of William Rufus7), the territory of the diocese is ancient. It was originally founded in the late seventh century when the great Mercian diocese was subdivided to provide sees for various subkingdoms within the area of Mercian overlordship, one of which was the kingdom of the Westerna (later known as the Magonsaete).8 Although records for the seventh-century territory of the Magonsaete are extremely sparse, we can guess that it included Leominster (the major house-monastery, founded about 660, of the kings of the Magonsaete, just as Winchcombe was for the Hwicce),9 Much Wenlock, where a nunnery was set up in the seventh century for St Mildburg, daughter of Merewalh, king of the Magonsaete,10 and the western side of the Malvern hills, which, according to a very garbled twelfth-century account in a bull of Innocent II, had been given by ‘King Meredith’ (in other words King Milfrith, a successor of Merewalh) to Hereford cathedral.11 Milfrith himself was apparently buried at Hereford, but the text which records this (verses from an inscription recorded by William of Malmesbury) does not name the place, and the earliest genuine documentary reference to Hereford is no

sulted on a website allowing access to Jeffrey Denton’s database version of the text: see http://www.taxatio.man.ac.uk/info.html . 7 The Liber Landavensis: Llyfr Teilo, ed. W. J. Rees, Welsh Manuscripts Society, Llandovery 1840, 263–6; for an English translation of part of the passage, see Episcopal Acts and Cognate Documents Relating to Welsh Dioceses 1066–1272, ed. James Conway Davies, 2 vols, Historical Society of the Church in Wales 1 and 3, 1946, 1948, i, 150, ii, 610–12. 8 On the early history of the diocese, see S. Keynes, ‘Diocese and Cathedral’, in Hereford Cathedral: a History, ed. G. Aylmer and J. Tiller, London 2000, 3–20 at 3–7; F. M. Stenton, ‘Pre-Conquest Herefordshire’, in idem, Preparatory to Anglo-Saxon England, Oxford 1970, 193–202; P. Sims-Williams, Religion and Literature in Western England, 600–800, Cambridge 1990, 39–53; K. Pretty, ‘Defining the Magonsaete’, in Origins of Anglo-Saxon Kingdoms, ed. S. R. Bassett, Leicester 1989, 171–83. 9 Sims-Williams, Religion and Literature, 55–6, 77, 94, 101, 114, 118, 149, 398 is cautious about the evidence for Leominster in the early middle ages; Joe Hillaby, ‘Leominster and Hereford: the Origins of the Diocese’, in Medieval Art, Architecture, ed. Whitehead, 1–14, and idem, ‘The Early Church in Herefordshire: Columban and Roman’, in The Early Church in Herefordshire: Proceedings of a Conference Held in Leominster in June 2000, ed. Ann Malpas, Leominster 2001, 41–76; John Blair, ‘The Anglo-Saxon Church in Herefordshire: Four Themes’, in ibid., 3–13; Keynes, ‘Diocese and Cathedral’, 4. 10 Sims-Williams, Religion and Literature, 29, 42–4, 98–101, 110–11, 117–18, 144–5; H. P. R. Finberg, The Early Charters of the West Midlands, Leicester 1961, 197–216; Paul Hayward, ‘The Miracula Inventionis Beate Mylburge Virginis attributed to “the Lord Atto, Cardinal Bishop of Ostia” ’, EHR 114, 1999, 543–73 at 543–5; Gelling, West Midlands, 71. 11 For the bull, dated 20 December 1134/5, see Papsturkunden in England, II, Die kirchlichen Archive und Bibliotheken, ed. Walther Holtzmann, Abhandlungen der Gesellschaft der Wissenschaften zu Göttingen, Philologisch-Historische Klasse, 3. Folge 14, Berlin 1935, 152–3, no. 15, and Charters and Records of Hereford Cathedral, ed. W. W. Capes, Cantilupe Society 3, Hereford 1908, 6–7; for discussion, see Fasti, viii, Hereford, ed. J. S. Barrow, London 2002, pp. xxi–xxii; on Milfrith see also Sims-Williams, Religion and Literature, 50–1.

40

Anglo-Norman Studies XXVI

earlier than the very early ninth century.12 Nonetheless it is likely that the see was based at Hereford from the beginning.13 From the scraps of information about Leominster and Much Wenlock, together with later references to the province of the Magonsaete in Anglo-Saxon sources,14 we can form some impression of the territory ruled by its early princes: clearly it included southern Shropshire (Much Wenlock) and most of Herefordshire, with the Malverns acting as a border between it and the neighbouring land of the Hwicce.15 It seems reasonable to suppose that the kingdom and the early diocese shared the same bounds, as has also been assumed in the case of the kingdom of the Hwicce and the diocese of Worcester.16 The eastern border of the see of Hereford was defined in detail by Bishop Æthelstan in the first half of the eleventh century, evidently in response to some possible hostile claim by Worcester.17 In surface area the diocese was of middling rank in English terms, somewhat smaller than those of Worcester and Winchester and somewhat larger than those of Wells and Chichester. All were much smaller than the giant sees of Lincoln, York and Coventry and Lichfield.18 In terms of population density Hereford was some way behind its English neighbours. Settlement was thinner, especially in southern Shropshire, which had significantly less arable land than neighbouring areas (much of Herefordshire, by contrast, was fertile);19 villages certainly were plentiful by the twelfth century, but were less likely to be nucleated than elsewhere, and towns were few.20 At the time of Domesday, there was only one town in the diocese of any size: this was Hereford itself, which had been a fortified place since the later eighth century, and which by the later twelfth century, when we can compare dues paid by different towns, came behind Bristol and Gloucester, but far ahead of Worcester, Shrewsbury and Bridgnorth, in the local pecking-order: when Gloucester paid a farm of £50, Hereford paid £40, Worcester £24, Shrewsbury £20 and Bridgnorth 10 marks.21 By the late twelfth century Leominster, 12 For the verses, see De gestis pontificum, 299; for the earliest documentary references to Hereford, see

Keynes, ‘Diocese and Cathedral’, 7–13. 13 For discussion see works cited in nn. 8–9 above; C. N. L. Brooke, ‘The Diocese of Hereford,

676–1200’, Transactions of the Woolhope Naturalists Field Club 48, 1994, 23–36 at 26–7, thinks that the see was not originally sited at Hereford. 14 For example S677, a charter of King Edgar of 958. 15 Finberg, Early Charters of the West Midlands, 208–12 on the early endowments of Much Wenlock; on the great estate of Leominster see B. Coplestone-Crow, Herefordshire Place-Names, BAR, British Series 214, 1989, 6–9, and Blair, ‘The Anglo-Saxon Church in Herefordshire’, 5, 10. 16 The early territory of the Hwicce can be mapped with rather more certainty than that of the Magonsaete because of the rich documentation for the early endowment of Worcester Cathedral: see Steven Bassett, ‘In Search of the Origins of Anglo-Saxon Kingdoms’, in The Origins of Anglo-Saxon Kingdoms, ed. Steven Bassett, Leicester 1989, 3–27, at 6–17. Shire boundaries in the west midlands of the early eleventh century and later did not correspond closely to diocesan boundaries: see maps in Steven Bassett, ‘The Administrative Landscape of the Diocese of Worcester in the Tenth Century’, in St Oswald of Worcester: Life and Influence, ed. Nicholas Brooks and Catherine Cubitt, London 1996, 147–73 at 152–4. 17 M. Förster, Der Flussname der Themse und seine Sippe, Munich 1941, 769; discussed by Finberg, Early Charters of the West Midlands, 225–7, and see also comment by J. F. A. Mason, ‘The Norman Earls of Shrewsbury: Three Notes’, Transactions of the Shropshire Archaeological Society 57, 1961–4, 152–61, at 158. 18 Mapped on the South Sheet of Ordnance Survey, Monastic Britain, Southampton 1978. 19 H. C. Darby, Domesday England, Cambridge 1977, 127. 20 Cf. Gelling, The West Midlands, 180–2. 21 For analysis of the Domesday material, see Darby, Domesday England, 296–7 (Domesday Book refers to boroughs within the territory of the diocese at Quatford, Wigmore, Clifford and Hereford); for Hereford see M. D. Lobel, ‘Hereford’, 1–11, in Historic Towns, i, ed. M. D. Lobel, London and Oxford 1969, each

Clergy in the Diocese of Hereford

41

Ledbury, Ross on Wye and Ludlow had emerged as a second tier of towns – all quite small, but active.22 However, although the diocese was not swift to become urbanised it enjoyed reasonable communications with the outside world. Hereford itself lay on the main route from central southern Wales, via Carmarthen and Brecon, to Worcester, which meant that the two dioceses with which its bishops had closest contacts were Worcester and St Davids.23 Above all the area had a vital military role. It had to protect the southern west midlands from possible attacks by the southern Welsh princes. These had been serious in the early and mid-eleventh century: in 1055 Hereford had been sacked by a Welsh invasion. The political situation was also precarious for the English in the years immediately following Henry I’s death.24 In offensive, rather than defensive, terms, Hereford functioned as a base for Anglo-Norman colonisation westwards,25 and, more peacefully, as a place for trade and negotiations between English and Welsh.26 There was Welsh immigration into Hereford itself,27 and in Archenfield the population remained Welsh-speaking until the eighteenth century,28 but it is important to note that Herefordshire and Shropshire were essentially, and rather consciously, English. Legally and administratively they operated exactly as did the neighbouring shires to the east, and had done for centuries.29 We now need to turn to the ecclesiastical geography of the diocese. For a study of the clergy in the diocese it is the parochial framework which is most important, but before looking at that some comment on monasticism in the diocese would be helpful to provide greater understanding of how ecclesiastical patronage operated. Monastic life in the diocese was, throughout our period, extremely limited. In this respect Hereford differed very greatly from the neighbouring see of Worcester, both before and after the Conquest. In the mid-eleventh century, when the diocese of Worcester could boast numerous major Benedictine houses, including, by then, its own cathedral chapter,30 Hereford had only one, Leominster, which, like several other late town separately paginated; Hereford City Excavations, ed. R. Shoesmith, 3 vols, Council for British Archaeology Research Reports 36, 46, 56, London 1980–5; Gelling, The West Midlands, 159–64; for Hereford’s position in the economic hierarchy of the valleys of the Severn and Wye, see J. Barrow, ‘The Canons and Citizens of Hereford c.1160–c.1240’, Midland History 24, 1999, 1–23 at 10. 22 Cf. Charters and Records, ed. Capes, 8, 13, 19; D. M. Palliser, T. R. Slater and E. Patricia Dennison, ‘The Topography of Towns 600–1300’, in The Cambridge Urban History of Britain, i, 600–1540, ed. D. M. Palliser, Cambridge 2000, 161, 167; Christopher Dyer and T. R. Slater, ‘The Midlands’, in ibid., 620. 23 Cf. EEA, vii, nos 29, 35, 134, 185; St Davids Episcopal Acta 1085–1280, ed. Julia Barrow, South Wales Record Society 13, Cardiff 1998, nos 42, 61, 63, 66, 77, 80, 81, 88 and Appendix, nos I–IV. 24 On the 1055 invasion, see The Chronicle of John of Worcester, ed. R. R. Darlington and P. McGurk, 2 vols to date, Oxford 1995–8, ii, 576; on the events of 1135–6 see R. R. Davies, The Age of Conquest: Wales 1063–1415, Oxford 1991, 45–6. 25 Cf. Joe Hillaby, ‘Hereford Gold, Irish, Welsh and English Land. The Jewish Community at Hereford and its Clients, 1179–1253. Part 1’, Transactions of the Woolhope Naturalists Field Club 44, 1984, 358–419. 26 Lobel, ‘Hereford’, 7. 27 Barrow, ‘Canons and Citizens’, 10, n. 73. 28 Welsh was spoken in Kentchurch until c.1750, and later in those parts of Herefordshire which lay in the diocese of St Davids: John Duncumb, Collections towards the History and Antiquities of the County of Hereford, 8 vols, London etc. 1804–1915, vi, part 3, Hundred of Wormelow, Upper Division, Part 1, by John Hobson Matthews, 5. 29 Cf. entries for Herefordshire and Shropshire in the Pipe Rolls. Note that the term normally used by Pipe Roll scribes for Herefordshire, ‘Herefordshire in Wales’ (e.g. ‘Herefordscr’ in Walia’, PR 5 Richard I, 86), was for administrative convenience, to ensure that it was not confused with Hertfordshire. 30 Cf. VCH Worcester, ed. J. W. Willis-Bund, H. A. Doubleday and William Page, 4 vols, London

42

Anglo-Norman Studies XXVI

Anglo-Saxon nunneries, formed part of the queen’s estates.31 Godwin’s son Swein, who visited Leominster in 1046, when he abducted its abbess,32 had presumably decided to stay there because it was his sister’s property. The scandal caused by the abduction did not lead to the closure of the house, since an abbess and nuns are recorded in 1086 in Domesday,33 but equally it did nothing to shake Edith’s, and thence William I’s, control of the house and its huge and wealthy estates. The history of the church is obscure after 1086, but clearly it remained in royal hands until Henry I gave it to his new foundation of Reading Abbey in 1121;34 not long afterwards it became a dependent priory.35 The diocese of Hereford saw only a limited growth in numbers of monastic houses after the Norman Conquest: by the 1230s there were some twenty Benedictine and Augustinian priories, two Cistercian abbeys, a Hospitaller preceptory, a Grandmontine priory and a house of Franciscans.36 Even the largest and richest of these houses, Abbey Dore, was only in the second rank of English monastic establishments.37 Most were dependencies of major monasteries elsewhere, especially of Gloucester Abbey, which owned the priories of St Guthlac’s in Hereford, Kilpeck and Bromfield.38 Indeed, in the twelfth century the most significant monastic influence inside the diocese was operated from outside it, chiefly by Gloucester Abbey, Llanthony Priory and the Norman abbeys of Lire and Cormeilles. Lire and Cormeilles had received several parish churches and tithes in the diocese from Earl William fitzOsbern soon after the Conquest;39 Gloucester acquired an extensive range of churches in Herefordshire, mostly in the time of Abbot Serlo (1072–1104),40 and Llanthony presumably began to acquire churches soon after its foundation as a priory in 1108.41 By 1291 Lire had the advowson of eight churches in

1901–26, ii, 94–136, and VCH Gloucester, ed. William Page, N. M. Herbert and C. R. Elrington, 9 vols, London 1907– , in progress, ii, 53–72; on the process by which Worcester cathedral became Benedictine see Julia Barrow, ‘The Community of Worcester, 961–c.1100’, in St Oswald of Worcester, ed. Brooks and Cubitt, 84–99. 31 Pauline Stafford, ‘Cherchez la femme. Queens, Queens’ Lands and Nunneries: Missing Links in the Foundation of Reading Abbey’, History 85, 2000, 4–27 at 9–10 and 11–13; R. V. Lennard, Rural England 1086–1135, Oxford 1959, 400–1; Sarah Foot, Veiled Women, 2 vols, Aldershot 2000, ii, 103–7; J. Hillaby, ‘Early Christian and Pre-Conquest Leominster: an Exploration of the Sources’, Transactions of the Woolhope Naturalists Field Club 45, 1987, 557–685. 32 ASC, 109: C, s.a. 1046. 33 Domesday Book, i, fol. 180a–c and discussion in Stafford, ‘Cherchez la femme’, 9. 34 Stafford, ‘Cherchez la femme’, 5–6; Reading Abbey Cartularies, ed. B. R. Kemp, 2 vols, Camden Society, 4th ser. 31 and 33, 1986–7, i, 13–19. 35 Reading Abbey Cartularies, ed. Kemp, i, 262–3, 287–8; on the difficulty of dating the foundation of the priory at Leominster see B. R. Kemp, ‘The Monastic Dean of Leominster’, EHR 83, 1968, 505–15 at 506; Medieval Religious Houses: England and Wales, ed. David Knowles and R. Neville Hadcock, 2nd edn, London 1971, 56, 74. 36 Ibid., maps at end of volume. 37 Ibid., 115 shows that Abbey Dore’s wealth at the end of the thirteenth century was considerably less than half that of the wealthiest English Cistercian houses such as Meaux and Fountains (ibid., 119, 122). 38 Ibid., 52, 54–5; for context see Emma Cownie, Religious Patronage in Anglo-Norman England 1066–1135, Woodbridge 1998, 61, though she dates the merger of St Guthlac’s and St Peter’s over forty-two years too early. 39 For Lire’s churches in the diocese recorded in Domesday, see Domesday Book, i, fols 181b, 182d; for Cormeilles’ churches and tithes in the diocese see ibid., fols 179cd, 180c, 182d, 184d; Fownhope was given to Lire by Aitrop the patron in the time of Bishop Gerard of Hereford (EEA, vii, no. 34); the Norman abbey of Conches had the church at Monkland, Domesday Book, i, fol. 183b. 40 Historia et Cartularium Monasterii Sancti Petri Gloucestriae, ed. W. H. Hart, 3 vols, RS, 1863–7, i, 58, 62, 66, 67, 71, 72, 84–7, 88, 91, 92, 109, 116, 118–19; Cownie, Religious Patronage, 54–65 on Gloucester abbey, especially 57–61 on its holdings in the diocese of Hereford. 41 F. G. Cowley, The Monastic Order in South Wales 1066–1349, Cardiff 1977, 30–1 on the foundation

Clergy in the Diocese of Hereford

43

the diocese, Gloucester of fourteen and Llanthony Prima of thirteen, and most of these were probably early acquisitions.42 Monastic patrons did not, however, necessarily present their own choice of incumbents for admission: a similar situation prevailed in the diocese of Hereford to that in the diocese of Norfolk, where, as Christopher Harper-Bill has shown, bishops often persuaded monasteries to present episcopal protégés.43 From Domesday we know that the diocese contained some sixty-four churches in 1086.44 Domesday probably understates, but not necessarily by very much. By the end of the thirteenth century we know from the great list of English churches drawn up for Pope Nicholas IV’s taxation in 1291 that the diocese contained just under three hundred parish churches and a further thirty-odd chapels lacking full parochial status.45 By way of comparison, Worcester diocese had over 150 churches recorded in 1086 and over 360 in 1291, Coventry and Lichfield had about five hundred churches in 1291, and the much more populous East Anglian see of Elmham (later Thetford and then Norwich) had seven hundred churches in 1086 and 1349 in the thirteenth century:46 in other words, the diocese of Hereford, which in 1086 had been very ‘poorly churched’, had effectively caught up.47 Over half of the churches the Taxatio records for the diocese of Hereford are mentioned in episcopal acta of the twelfth and early thirteenth century, but in fact it is likely that the overwhelming majority of the churches and chapels of 1291 were in existence by the second half of the twelfth century, for specific references to the foundation of new churches, though few overall, are most plentiful in the episcopates of Robert de Béthune (1131–48) and Gilbert Foliot (1148–63), and after about the middle of the twelfth century it became progressively harder to split up existing parishes.48 and early growth of Llanthony (Llanthony Prima), and the establishment of Llanthony Secunda (Lanthony by Gloucester) in 1137. 42 Taxatio P. Nicholai, 158–61, 165–6 and see also 170–2. 43 C. Harper-Bill, ‘The Struggle for Benefices in Twelfth-century East Anglia’, ANS 11, 1988 (1989), 113–32, at 130–2, and see below at n. 109. 44 Awre and Dymock in Gloucestershire; Tenbury Wells and Mathon (the latter was disputed between the dioceses of Hereford and Worcester) in Worcestershire; Linton, Kingsland, Eardisland, Much Marcle, Cleeve, Leominster, perhaps Edwyn Ralph, just possibly Aymestrey, Monmouth, Pontrilas, Dewsall, Hereford Cathedral, Llanwarne, Bishop’s Frome, Ross on Wye, Upton Bishop, Ledbury, Bosbury, Cradley, Bromyard, Little Hereford, Hereford St Peter’s, Hereford St Guthlac’s, Avenbury, perhaps Ashperton, Weston Beggard, Letton, Weobley, King’s Pyon, Wormsley, Collington, Burghill, Brinsop, Pencombe, Much Cowarne, Bodenham, Fownhope and Wellington in Herefordshire; Lydbury North, Onibury, Wenlock, Bromfield, Morville, Corfham, Chirbury, Stretton, Stottesdon, Chetton, Munslow Aston, Glazeley, Patton, Bitterley, Woolston, Holdgate, Wotherton, Burford, Cleobury Mortimer, Leintwardine, Stanton Lacy, Aldon in Shropshire (Domesday Book, i, fols 163a, 164a, 174b, 175c, 179c, 179d, 180a, 180d, 181a–d, 182a, 182cd, 183a, 183c, 184b, 184d, 185a, 186ab, 186d, 187b, 252b–d, 253bc, 254a, 255ab, 256c, 258bc, 259c, 260a–d). 45 Taxatio P. Nicholai, 157–67 and 175–6. 46 For Worcester, see discussion in J. Barrow ‘Wulfstan and Worcester: Bishop and Clergy in the Early Eleventh Century’, in Wulfstan, Archbishop of York, ed. Matthew Townend, Turnhout forthcoming, and table in E. L. Cutts, Parish Priests and their People in the Middle Ages in England, London 1914, 385; for East Anglia, see James Campbell, ‘The East Anglian Sees before the Conquest’, in Norwich Cathedral: Church, City and Diocese, 1096–1996, ed. Ian Atherton, Eric Fernie, Christopher Harper-Bill and Hassell Smith, London 1996, 3–21, at 20; Cutts, Parish Priests, 394, and W. Hudson, ‘The “Norwich Taxation” of 1254 so far as it relates to the Diocese of Norwich’, Norfolk Archaeology 17, 1908–10, 46–157 at 69. See also Taxatio P. Nicholai, 78–93, 115–23 (Norwich), 216–224, 239–40 (Worcester), and 241–9 (Coventry and Lichfield). 47 The expression is used by John Blair to describe Herefordshire in 1086: ‘The Anglo-Saxon Church in Herefordshire’, 11. 48 160 parishes and chapels with parochial title in the diocese occur in the index to EEA, vii, 321–53, covering the period 1079–1234; for dedications, see n. 80 below.

44

Anglo-Norman Studies XXVI

An administrative framework for the diocese had been provided by the post-Conquest bishops, chiefly Robert the Lotharingian (1079–95), who had established the two archdeaconries by the mid-1080s,49 and Robert de Béthune, under whom the earliest rural deaneries emerge.50 These structures helped the bishops to delegate duties and to keep an eye on the swiftly increasing numbers of parish clergy. Pre-Conquest bishops had presumably supervised churches on their own estates and assumed that the king and other lords would supervise their own clergy, and since the number of clergy was very small, doubtless this system worked adequately. The period from 1080 to the mid-twelfth century also saw the transformation of some of the richer pre-Conquest minsters into Benedictine or Augustinian priories or abbeys, and, in addition, as we have seen, the growing influence within the diocese of Hereford of several monasteries sited outside the diocese, which acquired large numbers of parish churches within it. This obviously had an impact on the clergy serving parish churches, affecting patronage patterns and removing much of the income which otherwise would have gone wholly to the incumbents. Domesday in 1086 provides us with a marker against which further developments can be set. Although the information it supplies about churches is less full than we would wish nonetheless it gives vital clues to the status of churches. Clues suggesting senior status for churches have been discussed by John Blair in Peter Sawyer, ed. Domesday Book: a Reassessment: particularly significant for our purposes are specified endowments for churches, multiple numbers of clergy, references to church dedications and entries for churches as tenants in chief.51 Applying these criteria to Domesday’s record of ecclesiastical establishments in the diocese of Hereford shows that just under two-thirds of the entries fall below this level. Many of these references to ‘minor’ establishments refer simply to priests with no mention of a church, with the priest counting as one of the lesser manorial tenants, classed with the villani and bordars and perhaps at most having a share in a plough-team. In some instances, mostly on manors which had been royal or comital TRE, a church is mentioned as well: perhaps in these cases the structure of the church was more elaborate or more permanent, making it worthier of record.52 It is probably right to see in many of these instances a new generation of eleventh century parish foundations. However, since one of the above churches, Much Marcle, was later recorded to have had two priests, as we have seen,53 it is possible that some of these ‘new’ foundations were not necessarily very different from the old ones. Side by side with these, however, are the churches with more elaborate endowments or personnel. The largest of these was Hereford Cathedral, or, as Domesday phrases it, the Church of Hereford, since Bishop Robert had, probably deliberately, drafted his returns for the Domesday Survey to suggest that the church estates were run as a whole, with no separation into episcopal and capitular possessions.54 Indeed 49 50 51 52

Fasti, viii, pp. xxiii and n., 23, 26. Cf. EEA, vii, nos 27–8, 50–2. Blair, ‘Secular Minster Churches’, 106. Church Stretton (TRE Earl Edwin), Bitterley (TRE Godwine, a free man), Holdgate (TRE held by Genust, Alweard, Dunning and Ælfgifu: the church may well have been built by Helgot after the conquest to accompany his castle) and Leintwardine (Edward the Confessor) in Shropshire, Pencombe (TRE and 1086 Alvred of Marlborough), Linton (Edward the Confessor), Kingsland (Edward the Confessor), Eardisland (TRE Earl Morcar), Much Marcle (TRE Earl Harold) and Cleeve (Edward the Confessor) in Herefordshire (Domesday Book, i, fols 254a, 256c, 258c, 260d, 186a, 179cd). 53 See n. 1 above. 54 J. Barrow, ‘A Lotharingian in Hereford: Bishop Robert’s Reorganisation of the Church of Hereford 1079–1095’, in Medieval Art, Architecture, ed. Whitehead, 29–47, at 34–9.

Clergy in the Diocese of Hereford

45

he viewed the members of the cathedral community as his clerks and chaplains (they are often referred to in Domesday as clerks or chaplains ‘of the bishop’) and in his 1085 land grant charter for Roger de Lacy several members of the cathedral community, who can be identified as such from entries in the cathedral obit book, are listed as ‘men of the bishop’.55 Nonetheless Robert had instigated some separation on the Hereford estates by carving small tenancies of about a hide apiece for each canon of the cathedral out of some of the larger manors. Probably forming part of the cathedral community were the four episcopal chaplains mentioned in Domesday, who presumably served Robert’s new Romanesque chapel.56 Bishop Robert also controlled several of the other old minster churches in the diocese – Lydbury North in Shropshire where his episcopal clerk William (the first archdeacon of Shropshire) presided over a church with priests endowed with land and a plough-team,57 and several Herefordshire churches, most significant among which were Ledbury (one priest with 2½ hides), Bosbury (one priest with one hide), and Bromyard (two priests with one hide). The last-mentioned was a minster going back at least to the ninth century when it was mentioned in a charter of Bishop Cuthwulf.58 The other major churches existing in the diocese of Hereford in 1086, apart from Hugh de Lacy’s new church of St Peter’s in Hereford59 and the church attached to Monmouth Castle, which belonged to St-Florent-ès-Saumur,60 had mostly been in royal hands in 1066, though a few – Wenlock, for example, which belonged to the Mercian earls,61 or Fownhope, which was owned by Thorkell the White62 – belonged to members of the Anglo-Saxon aristocracy. With the exception of Leominster, still clearly with some vestiges of its community of nuns as late as 1086,63 all these churches were, in 1066, communities of secular clergy. Several of them, notably Wenlock, St Guthlac’s in Hereford, Bromfield and Morville, were quite generously endowed, with a hide or more of land per canon.64 Edward the Confessor, however, had made use of St Guthlac’s and Bromfield to show especial favour to his clerk Spirites, who had obtained much bigger shares in both churches than the other canons could have held, and Spirites’ disgrace had led to the alienation of his holdings from Bromfield.65 St Guthlac’s suffered likewise: by 1086 the lands which Spirites had held were held separately from St Guthlac’s by one of William the 55 Ibid., 35; EEA, vii, no. 2, and see n. 69 below. 56 Barrow, ‘A Lotharingian’, 33–5; on the chapel, see most recently Richard Plant, ‘English Romanesque

and the Empire’, ANS 24, 2001 (2002), 177–202 at 186–7 and literature there cited. 57 Domesday Book, i, fol. 252b (entry for Lydbury North); Fasti, viii, 26. 58 Domesday Book, i, fol. 182a (entries for Ledbury and Bosbury) and 182c (entry for Bromyard); for the

charter of Bishop Cuthwulf concerning Bromyard, see Charters and Records, ed. Capes, 1–2, and comment by Keynes, ‘Diocese and Cathedral’, 12–13 with facsimile ibid., 13; large-scale facsimile of face and dorse of Cuthwulf charter in Facsimiles of Anglo-Saxon Charters, ed. S. Keynes, Anglo-Saxon Charters Supplementary Volume i, Oxford 1991, plate 3. 59 Domesday Book, i, fol. 182d (entry for St Peter’s, Hereford). 60 Ibid., i, fol. 180d (entry for Monmouth); the church at the castle was an old one, dedicated to St Cadog; by the start of the twelfth century there was also a priory at Monmouth, dependent on St-Florent: Cowley, Monastic Order, 14–15. 61 Domesday Book, i, fol. 252c–d (entry for Wenlock), and see n. 73 below. 62 Ibid., i, fol. 187b (entry for Fownhope). 63 Ibid., i, fol. 180a–c (entry for Leominster) and see nn. 31–5 above. 64 Ibid., i, fos. 182d (St Guthlac’s, 18 hides), 183a (Nigel the physician, 19 hides, mostly or wholly land which had belonged to St Guthlac’s), 252c–d (Much Wenlock, just over 57 hides), 252d (Bromfield, just under 21 hides in 1086 but just under 31 hides before the disgrace and dispossession of Spirites, who had held 10 of the 20 hides at Bromfield) and 253b (St Gregory’s church, Morville, which had had 8 hides with 8 canons TRE). 65 Ibid., i, fol. 252d (entry for Bromfield) and see n. 3 above.

46

Anglo-Norman Studies XXVI

Conqueror’s favoured clerks, Nigel the physician.66 In Shropshire the royal and comital churches came into the hands of Earl Roger, who used some of their endowments for monastic foundations, but nonetheless retained the remainder for his household clergy. In the final two decades of the eleventh century both Bishop Robert and Earl Roger maintained significant groups of household clergy. Bishop Robert’s following attended when he made his grant of Holme Lacy to Roger de Lacy in 1085.67 In November 1086 Earl Roger set up a new collegiate church for his household clergy at Quatford.68 Whereas the clergy serving the smaller old minster churches in the diocese are not known to us by name in this period and probably had essentially a local and pastoral role, Bishop Robert’s and Earl Roger’s clerks must have had administrative duties. Robert and Roger made use of collegiate foundations to support their staff. Robert’s household clergy can almost all be identified as members of the cathedral chapter. They were headed by his brother Gerard, probably Hereford cathedral’s earliest dean, and included the two earliest archdeacons, Heinfrid and William, and also several clerks with Old English names – Leofwine, Alfward, Saewulf, Alfwine.69 By contrast, Earl Roger’s clergy were, as far as we know their names, exclusively continental – Godebald, Herbert the Grammarian and Odelerius of Orléans, the last-mentioned better known to us as the father of Orderic Vitalis.70 Nonetheless two of this group, Godebald and Odelerius, tried to anchor themselves in the local community through marriage.71 A similar group of household clergy, again with continental names, appears in the entourage of Roger de Lacy in Bishop Robert’s 1085 land-grant – Ralph, Geoffrey, Odo and Gerald. Probably they served the new church of St Peter’s in Hereford, built by Roger’s father.72 This relatively simple state of affairs, with smallish numbers of clergy operating often in small communities at points scattered widely across the diocese, and sometimes acting in household groups tagging along behind the good and the great, changed radically in the twelfth century. On the one hand, the clergy came under attack from the Gregorian reformers for not leading the simple lives of the Apostles, and while this threat rarely operated against them at a personal level it did operate against many of their institutions, as landowners with rights over old minster 66 Ibid., i, fol. 183a (entry for Nigel the Physician in Herefordshire). 67 EEA, vii, no. 2, and see n. 69 below. 68 Quatford was founded for six canons: W. G. Clark-Maxwell and A. Hamilton Thompson, ‘The College

of St Mary Magdalene, Bridgnorth, with Some Account of its Deans and Prebendaries’, Arch. Journ. 84, 1927, 1–87 at 2–3, with facsimile of the early eighteenth-century copy of Roger’s charter on plate 1, facing p. 1; for the date of Roger’s charter, 14 November 1086, see J. F. A. Mason and P. A. Barker, ‘The Norman Castle at Quatford’, Transactions of the Shropshire Archaeological Society 57, 1961–4, 37–62 at 39–41; on the transfer of the collegiate foundation from Quatford to Bridgnorth, see A. T. Gaydon, ‘The College of St Mary Magdalen, Bridgnorth’, in VCH Shropshire, ed. William Page, A. T. Gaydon and G. C. Baugh, 7 vols, London 1908– , in progress, ii, 123–8 at 124. 69 EEA, vii, no. 2; Fasti, viii, 8 (Gerard), 23 (Archdeacon Heinfrid), 26, 93 (William), 63 (Alfward and Alfwine), 64 (Ansfrid the priest), 76 (Leofwine) and 88 (Saewulf). 70 J. F. A. Mason, ‘The Officers and Clerks of the Norman Earls of Shropshire’, Transactions of the Shropshire Archaeological Society 56, 1957–60, 244–57 at 252–3; Marjorie Chibnall, The World of Orderic Vitalis, Oxford 1984, 7–8. 71 Ibid., 8. 72 EEA, vii, no. 2, and comment by T. S. Purser, ‘The Origins of English Feudalism: an Episcopal Landgrant Revisited’, Historical Research 73, 2000, 80–92 at 91. St Peter’s was given to Gloucester Abbey in 1101–2 by Roger’s brother Hugh, but Hereford canons occur as priests there in 1125 (Oxford, Balliol College MS 271, fol. 77v, no. 330, whose date is given as 1225 by the copyist, though the names of the witnesses show that the document must be a century earlier) and Gloucester Abbey only got introitus in 1134 (EEA, vii, no. 17).

Clergy in the Diocese of Hereford

47

churches decided, where there were sufficient resources, to convert them into Benedictine or Augustinian priories. Wenlock became Cluniac c.1080,73 Morville became a cell of Shrewsbury Abbey c.1138,74 St Guthlac’s in Hereford merged with St Peter’s Hereford to form a dependent priory of Gloucester Abbey in 1143,75 Bromfield became a dependency of Gloucester in 1155,76 and Chirbury became an Augustinian priory at the very end of the twelfth century.77 At one level therefore there was a reduction in preferment for secular clerics. It is worth noting that there was no serious move to preserve and strengthen some of the richer secular collegiate churches in the diocese of Hereford as there was in the diocese of Coventry and Lichfield in the cases of Stafford, Warwick and Wolverhampton; the collegiate community at Quatford moved to Bridgnorth in the diocese of Coventry and Lichfield, and became a royal free chapel with exemption from episcopal intervention in its parishes.78 However, several of the less wealthy churches served by more than one clerk, for example Ledbury and Llanwarne, survived as churches with portionists.79 On the other hand, a much larger number of positions was opened up to the clergy through the foundation of parish churches and chapels. During the period 1086–1291, as we have seen, the number of churches grew by up to five-fold, and most of this growth probably occurred in the twelfth century, when references to dedications and foundations, though they only exist in a minority of cases, are to be found in significant numbers in the charters of Bishops Robert de Béthune (1131–48) and Gilbert Foliot (1148–63).80 Although on the face of things a parish cure might be viewed as less attractive than a prebend in a collegiate church, in practice much depended on the endowment of the parish church and the value of its tithes, in other words on the agricultural productivity of the parish, nor did the tenure of a parish prevent the holder from holding other positions (preferably, according to canon law, ones without cure of souls, though even this restriction could be overlooked at times). Everything depended, therefore, on the cleric’s ability to attract the notice of favourable patrons. For the twelfth-century clergy, the information at our disposal is much greater than for the eleventh century. We can name a sizeable minority of the parish clergy in the diocese and probably pretty well all the cathedral canons in the period. On the other hand, very often the information does solely consist of names, for example names of people listed as witnesses to charters. Names alone are not very helpful in

73 M. Chibnall, ‘The Abbey, later Priory, of Wenlock’, in VCH Shropshire, ii, 38–47 at 39–40. Some of

Much Wenlock’s property was used by Earl Roger to fund his clerks: Stoke Milborough temporarily in 1086 and Eardington more permanently (ibid., 39). 74 M. Chibnall, ‘The Priory of Morville’, ibid., 29–30. 75 EEA, vii, no. 21; C. N. L. Brooke, The Church and the Welsh Border in the Central Middle Ages, Woodbridge 1986, 55, n. 19. 76 Blair, ‘Secular Minster Churches’, 128–31; M. Chibnall, ‘The Priory of Bromfield’, in VCH Shropshire, ii, 27–9. 77 The community was originally established at Snead, c.1190, and moved to Chirbury at the end of the twelfth century: M. Chibnall, ‘The Priory of Chirbury’, in VCH Shropshire, ii, 59–62; EEA, vii, no. 184 and cf. nos 295, 319. 78 J. H. Denton, English Royal Free Chapels 1100–1300: a Constitutional Survey, Manchester 1970, 94, 119. 79 For Ledbury and Llanwarne see The Register of Thomas de Cantilupe, ed. R. G. Griffiths with W. W. Capes, Cantilupe Society 1, Hereford 1906, 27, 141–2, 185; The Register of Richard de Swinfield, ed. W. W. Capes, Cantilupe Society 4, Hereford 1909, 464, 538. 80 EEA, vii, nos 31, 44, 50–1, 56, and The Letters and Charters of Gilbert Foliot, ed. A. Morey and C. N. L. Brooke, Cambridge 1967, nos 290, 301, 334–7 for dedication charters of Robert and Gilbert (often concerning the dedication of graveyards).

48

Anglo-Norman Studies XXVI

trying to work out professional or family relationships, but nonetheless can be quite suggestive. Among the beneficed clergy, the group about which it is possible to know most, of course, is the cathedral chapter of Hereford itself, which is quite fully recorded, especially from c.1150 onwards, principally through its own impressive charter collection and its obit book.81 Although the canons would not all have been permanently resident, since several of them were employed by kings or sometimes other authorities as administrators, or had other ecclesiastical offices elsewhere, most of them would have spent at least some time each year in Hereford itself and would probably have set the tone for the rest of the clergy in the diocese.82 In an official context, the two groups would have met at synods, which began to be held in the diocese of Hereford under Robert de Béthune in the first year of his pontificate (1131–2),83 but there would have been many other more informal contacts as well. About 115 canons can be identified from the period from the mid-eleventh century to c.1200.84 For the period from the eleventh century to the middle of the twelfth century we probably lack many the names, but thereafter it is likely that most, perhaps practically all, the members of the chapter are known to us. By the 1130s canons at Hereford, at about the same time as canons at other English cathedrals, began to use by-names or surnames, which increasingly came to be necessary where a single given name was shared by several canons; thus, from the 1130s–40s, we find Master Hugh de Clifford, Master Hugh of Northampton and Hugh Partes.85 Bynames give more clues than do given names as to the social or geographical origins of their bearers. Nonetheless even in the period before the 1130s we can form some ideas about social background simply from the given names. Of the canons likely to have lived in the late eleventh and the very early twelfth century, rather over half had Old English names, one had a Scandinavian name popular in England, and slightly under half had Continental Germanic names.86 This latter group apparently included some Lotharingians, for example Bishop Robert’s brother Gerard, as well as Normans.87 The mixture suggests that Hereford cathedral, like some other major churches, for example St Paul’s in London, acted as a haven for English clergy while simultaneously bringing in incomers from the Continent.88

81 The original charters of Hereford cathedral are preserved in Hereford Dean and Chapter Muniments;

the cathedral’s three medieval cartularies are in the Bodleian Library in Oxford (Bodl. MS Jones 23; Bodl. MS Rawlinson B329, fols 121–76; Bodl. MS Rawlinson B329, fols 1–120); Hereford cathedral obit book (Bodl. MS Rawlinson B328, fols 1–54r) has been published twice, first as an appendix (paginated separately) to R. Rawlinson, The History and Antiquities of the City and Cathedral Church of Hereford, London 1717, and more recently in Fasti, viii, Appendix I, 99–158. 82 Barrow, ‘Canons and Citizens’, 7, n. 44. 83 EEA, vii, nos 33, 55. 84 Calculated from Fasti, viii. Of these, slightly more than 90 can be dated to between c.1130 and c.1200. 85 Ibid., 72–3; for canons at other English cathedrals, see volumes of Fasti, passim. 86 Old English names: Leofwin, Alfward, Alfwine, Saulfus (Saewulf) (all occ. 1085), Alulfus (Æthelwulf), Edric, Edwin, Eilwin, Ordmaer, Wulfric, Wulfwin, another Wulfwin (not dated); Scandinavian: Ketelbern (occ. 1101/2); Continental Germanic: Gerard, Heinfrid, Ansfrid, William (all occ. 1085), Bernard, Walter (occ. 1101/2), Erchemar, Durand, Eustace, Odo (occ. 1107x1115), Geoffrey (d.1120), Gurinus and perhaps Ralph and Richard (not dated) Fasti, viii, 8–9, 23, 26, 63–7, 70, 76, 78–9, 82–3, 88, 91, 97–8. 87 Ibid., 8 for Gerard; possibly Heinfrid, Ansfrid and William were also Lotharingian (ibid., 23, 26, 64). 88 Fasti, i, St Paul’s, London, ed. Greenway, 4, 8, 23, 27, 29, 32, 36, 40, 47, 49, 57, 65, 71, 73, 77, 79, 85, 87, 89; for discussion of the composition of St Paul’s chapter see C. N. L. Brooke, ‘The Composition of the Chapter of St Paul’s, 1086–1163’, Cambridge Historical Journal 10, 1950–2, 111–32 at 122, and Ann Williams, The English and the Norman Conquest, Oxford 1995, 127.

Clergy in the Diocese of Hereford

49

From the 1130s onwards we can form some idea of social and geographical origins for many of those canons with by-names. Of the ninety or so canons who can be placed in the period c.1130–c.1200, fifty-three, just over half, had by-names. Some of these by-names are helpful in suggesting where the canons had originated: five came from families of Norman origin settled in the diocese, a further six bore names of places within the diocese, and about twenty-two others evidently originated outside the diocese, some from neighbouring areas of the west midlands and some from further afield, including France and possibly even Italy.89 One canon was described as the son of a dean, and therefore could be viewed as local.90 Some surnames provide pointers to social status – quite a few canons belonged to established knightly or even baronial families (Cumin, Foliot, de Freine, de Lacy, Clifford, de Clare, de Barri, Vere).91 Some canons used, as by-names, names of English towns which might suggest an origin in the burgess class (Northampton, Colchester, Totnes, Ledbury, Salisbury) but equally this might mean some other connection with these places, perhaps through education.92 Education was fairly important in securing advancement: about a third of all canons over the period from the 1130s to c.1200 bore the magister title, suggesting that they had completed a course of study at a higher school,93 while one of the archdeacons, Ralph Foliot, was an experienced royal judge and had probably acquired a comparable level of education through his knowledge of common law.94 The cathedral prebends were in the gift of the bishop, or in episcopal vacancies of the king, and the clerics most likely to attract the notice of twelfth-century bishops or kings were those with a good education who had also shown skill in law or administration. Already under Gilbert Foliot (1148–63) two of his household clergy became canons at Hereford and by the end of the twelfth century under Bishop William de Vere (1186–98) the intake of episcopal clerks into the chapter was becoming significant – five or six out of a total intake under William of about fourteen.95 None of the twelfth-century Hereford bishops had family ties in the diocese and they would not have felt bound to give preferment to clerks from families living in the diocese. In fact most Herefordshire and Shropshire knightly and baronial families were not represented in the chapter at all in the twelfth and thirteenth centuries, and in the twelfth century only the de Lacy, de Clifford and de Freine families made much of an impact.96 Two knightly families, de Chandos and Parvus, tried to secure a foothold in the chapter by establishing new prebends but their control of these prebends was broken after the deaths of the first canons to hold them.97 Some of the incoming canons, however, put down roots in local families through marriage, especially in the city of Hereford itself, for example William

89 Calculated from Fasti, viii; J. Barrow, ‘Origins and Careers of Cathedral Canons in Twelfth-Century

England’, Medieval Prosopography 21, 2000, 23–40 at 29–30. 90 Fasti, viii, 44 (Master Ranulf son of Dean Erchemar). 91 Ibid., 14, 21, 24, 26, 68–9, 83, 85–6, 89, 92, 94. 92 Ibid., 72, 75, 82, 83, 97. The use of Northampton and Salisbury as by-names might possibly refer to

schooling rather than to origin: both Northampton and Salisbury had schools of some distinction. 93 Ibid., passim; J. Barrow, ‘Education and the Recruitment of Cathedral Canons in England and

Germany 1100–1225’, Viator 20, 1989, 117–38 at 138. 94 Fasti, viii, 24, 82. 95 Barrow, ‘Origins and Careers’, 38. 96 Fasti, viii, 68, 72, 83–4, 86, 89, 92. Examples of Herefordshire landowning families not represented in

the chapter in the twelfth and thirteenth centuries are Baskerville, Blez (Bliss), Ferrers, Gamages, Kinnersley, Mucegros, Mynors, Pichard, Saucey, Sollers, Tosny, Wafre. 97 Ibid., 47–8, 59–60, and Barrow, ‘A Lotharingian in Hereford’, 41.

50

Anglo-Norman Studies XXVI

Foliot, precentor at the end of the twelfth century, who married (or at least produced a family with) Margaret, daughter of a Hereford moneyer.98 Turning from the cathedral chapter to the parish clergy we see some similar developments but operating with a very long time-lag. The observations which follow are made on the basis of a list of eighty-four clergy in the diocese of Hereford in the twelfth century whose livings can be identified, but this list represents work in progress and the figure of eighty-four is capable of being enlarged.99 The parish clergy are much less well documented than the cathedral canons: the latter can occur frequently as witnesses to charters and in other guises, the former perhaps at most two or three times, often only once. In yet another significant respect the material is less tractable than in the case of cathedral canons: it took much longer for by-names to be used for parish clergy than for the latter. (This is a general phenomenon, but parish clergy in eastern England were quicker to adopt the use of by-names than those in the west midlands.) Even so this lack of evidence is itself a clue, suggesting that the social horizon was so narrow in the twelfth century that, say, William, priest of Eastnor, witnessing a charter, would feel that the mention of his church gave him enough of an identity.100 The lack of by-names may also suggest that the parish clergy in the diocese of Hereford were not geographically mobile. This may also be borne out by the fact that Old English names still occurred quite often in the middle of the twelfth century, though already from the 1130s they were heavily outnumbered by Continental Germanic names. Welsh names occur rarely in the twelfth century, probably because the Welsh-speaking parts of the diocese were poorly documented after Llandaff lost control of Archenfield, but the mysterious Novis, occurring as priest of Kilpeck probably before 1134, must have been a Welshman, bearing the name Nobis. For the second half of the eleventh century the Book of Llandaff records the names of many clerics, mostly members of clerical dynasties, serving in churches in Archenfield.101 The only parish clergy with by-names in the middle of the twelfth century are Robert Albus – a nickname which conveys little – and Walter de Freine. About the latter, who was already incumbent of Presteigne when his kinsman, Thomas de Freine, gave the church to St Guthlac’s Priory in Hereford in 1145, we can say rather more: the de Freines were established at Freen Court in Sutton, and several members of the family were cathedral canons in the twelfth and thirteenth centuries, Walter among them. Probably the church at Presteigne was the opening move in his career.102 By the late twelfth century by-names were slightly more frequent, but here too it was often cathedral canons or episcopal clerks who were referred to by them – Archdeacon Ralph Foliot was parson of Cradley in the 1180s,103 Canon Simon Foliot was parson of Holme Lacy in the 1190s,104 Master Nicholas of Wolverhampton was

98

J. Barrow, ‘Hereford Bishops and Married Clergy, c.1130–1240’, Historical Research 60, 1987, 1–8, at 7–8; Fasti, viii, 75; Barrow, ‘Canons and Citizens’, 12. 99 Work in progress, drawn from EEA, vii, Balliol College MS 271 and other sources. 100 For examples of clergy with by-names in the diocese of Norwich in the second half of the twelfth century, see EEA, vi, Norwich 1070–1214, ed. Christopher Harper-Bill, Oxford 1990, nos 86, 95, 107, 141, 165–6; for William of Eastnor, see EEA, vii, nos 28, 116. 101 For Old English names see EEA, vii, nos 20, 28, 50, 55, and The Letters and Charters of Gilbert Foliot, ed. Morey and Brooke, nos 304, 312, 334; for Novis see Balliol MS 271, fol. 53v; for eleventhcentury Archenfield see n. 7 above. 102 EEA, vii, nos 25, 27, 47; Fasti, viii, 92. 103 EEA, vii, no. 196. 104 Ibid., no. 209.

Clergy in the Diocese of Hereford

51

parson of Ditton Priors by the early thirteenth century105 and Master William of Calne, a clerk of Bishop William de Vere, was parson of Stottesdon.106 Nonetheless the use of surnames was percolating downwards among the clergy by c.1200, even those who did not enter the cathedral chapter, as evidenced by Walter of Stockton, the vicar serving the parochial altar of Holy Cross in Leominster Priory, and Peter of Hopton, incumbent of Aston Eyre.107 Scanty though our material is, it does provide some clues about preferment. Clearly Walter de Freine had owed his tenure of the church of Presteigne to his family. Peter of Hopton at Aston Eyre had been proposed by the lord of Aston Eyre, though in fact he was presented, in a compromise agreement, by Shrewsbury Abbey which was claiming the advowson of the church. Walter of Stockton was a local man, probably from Stockton in Kimbolton, Herefordshire, being presented by Leominster Priory. William of Kilpeck, evidently also local, who became a canon of Hereford, began his career with the help of the patronage of St Guthlac’s Priory as incumbent of St Peter’s church in Hereford.108 There was, unsurprisingly, a tendency among secular landowners to favour relatives (as in the presentation of Walter de Freine to Presteigne), but nevertheless both secular and monastic patrons would frequently favour episcopal clerics or Hereford cathedral canons (as in the case of Master Nicholas of Wolverhampton, presented to Ditton Priors church by Much Wenlock Priory), and as a result a growing number of parish clergy, by the end of the twelfth century, were outsiders to the diocese. On the whole, however, change was slow for most of the twelfth century. The dynastic principle, so noticeable in the Book of Llandaff, remained strong. Whole dynasties of clergy might receive patronage from religious establishments – Leominster Priory, as Brian Kemp showed, only broke the hold of the family of incumbents of Eye church (Adam, Osbert, Osbert the younger and Roger) in the 1250s, and even then Roger’s son Philip had been allowed to become parson of Yarpole.109 Similarly, Hereford Cathedral allowed Roger the son of William parson of Allensmore to succeed his father in 1200 or 1201, and it is just possible that the Gilbert son of Payn who was presented by Lire Abbey to Lydney in the late twelfth century was the son of a priest, since Payn the chaplain occurs in the witness list. Another dynasty of clerks existed in Ledbury, though it is not certain if they were beneficed.110 Only slowly did a more fluid pattern of patronage come into play. One pointer towards this is the increasing number of parish clergy with the magister title, which is a trend which began in the twelfth century. By the early thirteenth century magistri were quite a noticeable proportion of the whole, perhaps as many as a sixth.111 The significance of the title from the point of view of the social historian is that clerics had gone to the trouble to obtain an education so that they would then more easily obtain support from a patron. Another straw in the wind was the presentation of Stephen the king’s notary by St Guthlac’s Priory to Bartestree in the reign of 105 106 107 108

Ibid., nos 228, 359. Ibid., nos 231–2. Ibid., nos 211, 215–16, 279 (Walter of Stockton), and 234–5 (Peter of Hopton). Balliol College MS 271, fol. 18r–v, nos 13–14; EEA, vii, nos 208, 258, 260, 286, 289, 293, 308–9, 311; Fasti, viii, 14–15, 96, 126. 109 B. R. Kemp, ‘Hereditary Benefices in the Medieval English Church: a Herefordshire Example’, BIHR 43, 1970, 1–15. 110 Barrow, ‘Hereford Bishops and Married Clergy’, 1–8 at 3 for the clergy at Allensmore; EEA, vii, no. 221 for Gilbert son of Payn, and ibid., no. 192 for a possible dynasty at Ledbury. 111 E.g. The Letters and Charters of Gilbert Foliot, ed. Morey and Brooke, no. 326; Charters and Records, ed. Capes, 14–15, 26–7; EEA, vii, nos 136–8, 143, 152–3, 194, 228, 231.

52

Anglo-Norman Studies XXVI

Henry II: during the thirteenth century, from John’s reign onwards, a more sustained royal interest in parish livings emerges into clearer view thanks to the enrolment of royal writs.112 It was also at about this point that various religious houses, too, started to take a slightly more adventurous approach to choosing candidates for presentation, at any rate to judge from the increasing use of surnames and the fact that several of these are names of places outside the diocese. St Guthlac’s Priory, a dependency of Gloucester Abbey, gave preferment to two clerics ‘from Gloucester’; Lire presented Master John of Godstow to Tidenham and Haughmond Abbey presented Mr John of Worcester to Culmington, all in the early decades of the thirteenth century.113 As these names suggest, the degree of mobility was very small: the clerics concerned came from neighbouring or at least nearby dioceses: nonetheless even so the emergence of a slightly wider framework of patronage is noticeable. Concurrently with this, several knightly families continued to obtain patronage for some of their kin in local parish churches. Thus Alan de Bullers was a prebendary of Chirbury and William de Tregoz was rector of Kentchurch (the Tregoz family acquired much land in Archenfield through Robert de Tregoz’s marriage to Sibyl of Ewias).114 If the parish clergy are shadowy in the twelfth century, this is even more the case with unbeneficed clerics. However, opportunities for the latter were on the increase in the twelfth century, thanks to three developments, firstly the expansion of prayers for the dead, secondly the need to provide deputies for the liturgical duties of frequently absent canons, and thirdly the great increase in the use of writing to record even quite minor legal transactions. By the end of the twelfth century, Hereford Cathedral, though admittedly more slowly than other cathedrals, was beginning to build up a small group of vicars to assist in its services.115 Many of these are recorded only in the obit book, which means we cannot date them more precisely than within the twelfth and thirteenth centuries, but a variety of vicars choral, chaplains, sacrists and even an organist evidently existed during the thirteenth century at least.116 Meanwhile a largely illiterate lay population was coming to see the benefits of the written word for its property transactions, leading to plenty of employment for scribes. Within the city of Hereford quite a sizeable group of freelance clergy can be spotted from the very end of the twelfth century onwards in the cathedral’s charter collection. Some of these, for example John son of Terricus, briefly reeve of the city in the early thirteenth century, and Thomas, clerk of the castle, who owned valuable houses in the city, were men of some standing in the community.117 Similarly the clergy in the bishop’s household, many of them highly educated, were well-positioned. This group, however, had advantages lacking to the others: for them the fast track to higher clerical promotion opened up with obtaining employment in the bishop’s household, after which they could hope for preferment directly from the bishop himself or at least indirectly from his episcopal colleagues, to whom, with luck, he might put in a good word. 112 EEA, vii, no. 152; for an example of thirteenth-century royal patronage, see Rotuli Litterarum

Patentium, ed. T. D. Hardy, Record Commission, London 1835, 62b. 113 EEA, vii, nos 203 and 262 (Geoffrey of Gloucester, Arnold of Gloucester), 312 (Master John of

Godstow), 334 (Master John of Worcester). 114 EEA, vii, no. 295; Historia et Cartularium Monasterii Sancti Petri Gloucestriae, iii, 269. 115 Philip Barrett, ‘The College of Vicars Choral’, in Hereford Cathedral, ed. Aylmer and Tiller; J.

Barrow, ‘The Origins of Vicars Choral to c.1300’, in Cantate Domino, ed. David Stocker, Council for British Archaeology, forthcoming. 116 Fasti, viii, 107, 108, 110, 112, 114, 116, 117, 119, 121, 122, 123, 125, 128, 132, 137, 151; see also Barrow, ‘Origins of Vicars Choral’. 117 Barrow, ‘Canons and Citizens’, 10 (Thomas), 12, 14 (John son of Terricus).

Clergy in the Diocese of Hereford

53

In conclusion, we can return to Robert de Béthune’s charter for Lire to see how typical it was of the relationship between Hereford bishops and their clergy in the twelfth century. Clearly it was atypical: among all patrons of churches there was a strong preference for local incumbents, and family networks, even dynasties of clergy, were usually allowed to flourish. The bishops of Hereford, even Robert de Béthune, who was perhaps the staunchest supporter of the Gregorian reform of them all, were not disposed to go against the general wishes of the diocese save where (as in the case of Much Marcle in the 1140s) the patrons insisted, which usually only happened when they saw their interests being threatened. On the whole, bishops preferred to allow the traditional system to prevail as long as they could promote their household clerics; these mostly could be satisfied with the churches in the bishop’s gift, but, where necessary, this pool of patronage could be enlarged by persuading monastic patrons to present episcopal protégés to their own livings.

England and the Irish-Sea Zone

ENGLAND AND THE IRISH-SEA ZONE IN THE ELEVENTH CENTURY* Clare Downham Many historical studies have been written about Anglo-Irish relations in the years immediately after the English invasion of Ireland in 1169.1 That the invasion should have an important place in research is understandable, given its long-term impact and its implications in recent historical and political debate.2 In contrast, very few publications have focused on Anglo-Irish political interaction in the eleventh century.3 In this paper, I hope to draw more attention to this somewhat neglected field of enquiry. The emphasis of historical scholarship on the invasion and its aftermath has perhaps influenced the interpretation of earlier events. The issues in the eleventh century which have been studied most are those which can be seen to foreshadow the later invasion. These include Canterbury’s claims of ecclesiastical primacy, and the alleged ambitions of Knútr or William the Conqueror to dominate Irish rulers.4 Meanwhile, research on a wider range of issues has been lacking. The resulting narrative gives a rather selective view of events. This hindsight perspective has, I suggest, meant that England’s domination of Irish rulers in the eleventh century has tended to be exaggerated. Furthermore Ireland’s impact on England has generally been underestimated.5 In this paper I seek to highlight Ireland’s significance in English affairs from the reign of Æthelred the Unready to that of William Rufus. From the late ninth century Ireland’s main contacts with England were through the viking towns of Dublin, Waterford and Limerick.6 In the late tenth century, these ports were dominated by the Hiberno-Scandinavian dynasty of Ívarr. Members of this family also ruled the Isle of Man and the Hebrides. Thus they were a significant power in Irish Sea politics. However, divisions within the dynasty enabled Irish provincial overkings to win increasing influence over the viking towns from the late tenth century.7 The wealth and military resources of these ports came to be regarded * I should like to thank David Dumville and Chris Lewis for commenting on a draft of this paper, and Howard Clarke and John Gillingham for helpful remarks at the Battle conference. I am grateful to the Dublin Institute for Advanced Studies for funding and research facilities. 1 For use of the term ‘English’, see R. H. C. Davis, The Normans and their Myth, London 1976, 13 and 131; John Gillingham, The English in the Twelfth Century, Woodbridge 2000, 145–60. 2 E.g. Anthony Carty, Was Ireland Conquered? International Law and the Irish Question, London 1996. 3 More work has focused on Hiberno- and Anglo-Welsh relations. Recent surveys include: Seán Duffy, ‘Ostmen, Irish and Welsh in the Eleventh Century’, Peritia 9, 1995, 378–96; Colmán Etchingham, ‘North Wales, Ireland and the Isles: the Insular Viking Zone’, Peritia 15, 2001, 145–87; K. L. Maund, Ireland, Wales, and England in the Eleventh Century, Woodbridge 1991. 4 Mark Philpott, ‘Some Interactions between the English and Irish Churches’, ANS 20, 1997 (1998), 187–204, at 188. 5 This in part reflects a bias in post-Conquest English sources: Gillingham, The English, 3–18, 145–50. 6 Mary A. Valante, ‘Reassessing the Irish “Monastic Town” ’, Irish Historical Studies 31, 1998–9, 1–18, at 12. 7 Donnchadh Ó Corráin, ‘The Vikings in Ireland’, in The Vikings in Ireland, ed. Anne-Christine Larsen, Roskilde 2001, 17–27.

56

Anglo-Norman Studies XXVI

as significant to those Irish leaders who sought to rule as much of Ireland as possible. As Seán Duffy has noted, Irish kings tended to become increasingly involved in Irish Sea affairs in the eleventh century, as a consequence of their ambition to rule these ports.8 This Irish dimension in foreign affairs is sometimes underrated, as overkings tended to manage Irish Sea affairs through the agency of Hiberno-Scandinavian fleets which they ruled or hired.9 During the eleventh century, the nature of contacts between England and Ireland can be broadly divided under the headings of trade, religious and intellectual links, and political relations.10 Most of this article will be devoted to political contacts. However, a brief overview of other links across the Irish Sea provides a context in which political events can be interpreted. A large amount of contact between England and Ireland took place through the medium of trade. The vikings’ colonies and their network of external contacts stimulated the import and export of goods to and from Ireland. English ports appear to have been the main trading partners with the viking towns of Ireland in the eleventh century.11 Archaeological and written evidence indicate that major exports included animal skins, grain, timber, slaves, fish, metal jewellery and antler combs. Slaves are perhaps the best-documented trade, and Dublin in particular seems to have been a hub for importing and exporting human traffic.12 It is notable that many of these exported goods would have been brought to the coastal towns from the interior of Ireland.13 An insight into this inland network of economic contacts is provided by the location of silver hoards. These demonstrate that Irish kings and merchants engaged in, and derived benefit from, trade across the Irish Sea.14 Recent studies have fostered a greater awareness of wider Irish involvement in foreign commerce. This has led to revision of the statement made by Gerald of Wales that the Irish were too lazy to trade.15 Many English ports had commercial links with Ireland. However, the geographical pattern of these relations altered over time. The political association of Dublin and York in the early tenth century encouraged trade. Their economic contacts continued, albeit on a reduced scale, in the eleventh century.16 The port of Chester appears to have been the main trading partner with Ireland from the 920s to the 970s, 8

Seán Duffy, ‘Irishmen and Islesmen in the Kingdoms of Dublin and Man, 1052–1171’, Ériu 43, 1992, 93–133, at 100. 9 This can be shown during the career of Gruffudd ap Cynan. See below, notes 146–7. Duffy (‘Ostmen’, 382–4) has also challenged the assumption that most fleets from Ireland were viking fleets. 10 A less tangible transmission of cultural practices and ideas took place within these exchanges. An example is the transfer of oral tales suggested in literature: Ó Corráin, ‘Viking Ireland – Afterthoughts’, in Ireland and Scandinavia in the Early Viking Age, ed. H. B. Clarke et al., Dublin 1998, 421–52, at 447. 11 Patrick F. Wallace, ‘The Economy and Commerce of Viking Age Dublin’, in Untersuchungen zu Handel und Verkehr der vor- und frühgeschichtlichen Zeit in Mittel- und Nordeuropa, Teil IV: Der Handel der Karolinger- und Wikingerzeit, ed. Klaus Düwel et al., Göttingen 1987, 200–45, at 229. 12 Poul Holm, ‘The Slave Trade of Dublin, Ninth to Twelfth Centuries’, Peritia 5, 1986, 317–45. 13 Wallace, ‘The Economy’, 201–5; Mary Valante, ‘Dublin’s Economic Relations with Hinterland and Periphery in the Later Viking Age’, Medieval Dublin 1, 2000, 69–83. 14 Marilyn Gerriets, ‘Money among the Irish: Coin Hoards in Viking-Age Ireland’, JRSAI 115, 1985, 121–39. 15 ‘Giraldus Cambrensis in Topographia Hibernie: Text of the First Recension’, ed. John J. O’Meara, PRIA 52 C, 1948–50, 113–78, at 175; Gerald of Wales, The History and Topography of Ireland, trans. John J. O’Meara, Harmondsworth 1982, 122. For the influence of this statement in the debate concerning urban origins in Ireland, see John Bradley, ‘The Interpretation of Scandinavian Settlement in Ireland’, in Settlement and Society in Medieval Ireland, ed. John Bradley, Kilkenny 1988, 39; Wallace, ‘The Economy’, 202. 16 Ibid., 230.

England and the Irish-Sea Zone

57

as numismatic evidence suggests.17 This town had developed largely in response to viking’s trading networks. While Chester remained significant in the eleventh century, the Severn estuary seems to have become increasingly important for Irish traders. This shift may be observed in the efforts made by the dynasty of Ívarr to dominate the coasts of South Wales, and in the deposition of silver hoards in this region.18 One notable consequence of these developments was the rise of Bristol. This port emerges in the historical record in the mid-eleventh century. It became the major partner in Irish trade throughout the late eleventh and twelfth centuries.19 This geographical re-orientation of trade in the eleventh century is also represented by stronger contacts with other towns in southern England. Documented examples include London, Gloucester, Exeter, and Cambridge.20 This change has been linked with the political decline of northern England and the increased significance of trade across the English Channel.21 Goods exported from England to Ireland at this time include jewellery, pottery, high quality woollens, and basic foodstuffs. More exotic goods also came from the Continent via England.22 Direct trading links between Ireland and north-west France are also well attested from the mid-eleventh century through the evidence of pottery imports.23 One consequence of this trade was that foreign merchants settled in major ports. Patrick Wallace has discussed the evidence for a settlement of English merchants along Fishamble Street in Dublin.24 In London, the dedication of a church to St Bride (Brigit) in Fleet Street might suggest the presence of an Irish community in the eleventh century.25 Other merchant colonies have been posited elsewhere.26 In addition, these trading links must have fostered multilingualism among Irish Sea merchants who might wish to converse in English, Norse, or Irish to effect transactions.27 Traders can be seen as a culturally, economically, and therefore politically significant group, whose interests were vested in contact across the Irish Sea. People were also prompted to cross the Irish Sea for religious reasons. This group included lay pilgrims and clerics. The main route to Rome was through England and Flanders.28 In 1941, the eminent ecclesiastical scholar, Aubrey Gwynn, drew atten17 A. T. Thacker, ‘Early Medieval Chester, 400–1230’, in VCH Cheshire, v, part 1, ed. C. P. Lewis and A. T. Thacker, London 2003, 16–33, at 20–21. 18 Mark Blackburn and Hugh Pagan, ‘A Revised Checklist of Coin Hoards from the British Isles, c.500–1100’, in Anglo-Saxon Monetary History, ed. M. A. S. Blackburn, Leicester 1986, 291–313 (revised edn, available online at www.-cm.fitzmuseum.cam.ac.uk/Coins/hoards/index.html), nos 172, 187c, 203; ASC MSS CDE, s.a. 997; Annales Cambriae, A.D. 955–1097: Texts BC in Parallel, ed. and trans. David N. Dumville, Cambridge (forthcoming), s.aa. 999, 1001. 19 ASC MS D, s.a. 1052; Aubrey Gwynn, ‘Medieval Bristol and Dublin’, Irish Historical Studies 5, 1946–7, 275–86; Wallace, ‘The Economy’, 231. 20 Ibid., 227–32. 21 Ibid., 239. 22 Ibid., 215–20. 23 Audrey Gahan et al., ‘Medieval Pottery’, in Late Viking Age and Medieval Waterford, ed. Maurice Hurley et al., Waterford 1997, 285–336; Alan G. Vince, ‘Early Medieval English Pottery in Viking Dublin’, in Keimelia: Studies in Medieval Archaeology and History in Memory of Tom Delaney, ed. Gearóid Mac Niocaill and Patrick F. Wallace, Galway 1988, 254–70. 24 Patrick F. Wallace, ‘The English Presence in Viking Dublin’, in Anglo-Saxon Monetary History, ed. Blackburn, 201–21. 25 Christopher N. L. Brooke and Gillian Keir, London 800–1216: The Shaping of a City, London 1975, 139–40. 26 Wallace, ‘The Economy’, 225. 27 Vita Ædwardi, 22–3, suggests that some English nobles, as well as merchants, learnt Irish. An ‘average standard’ (mediocritatem) is referred to, which the audience of the text is assumed to have encountered. 28 Annals of Ulster (to A.D. 1131), ed. and trans. Séan Mac Airt and Gearóid Mac Niocaill, Dublin 1983,

58

Anglo-Norman Studies XXVI

tion to chronicle records of royal pilgrims travelling from Ireland to Rome in the eleventh century.29 Although these high ranking pilgrims desired to perform a conspicuous act of piety, their journeys can also be seen as a political exercise. They networked with the prominent ecclesiastical and lay households where they stopped en route. Ultimately some may have desired to win the ear of the pope on particular issues, as the papacy became increasingly influential in North European politics. Gwynn initially identified seven kings who set off for Rome from Ireland from 1028 to 1064. He attributed this vogue to Knútr’s visit to Rome in 1027. On this trip he obtained guarantees of safe passage and freedom from tolls for other pilgrims travelling to Rome.30 Gwynn identified the first royal pilgrim from Ireland as Sigtryggr, king of Dublin; and two more kings of Dublin also went to Rome.31 He concluded that royal pilgrimage to Rome began in Ireland as a Scandinavian phenomenon which was then copied by indigenous rulers. Later in the same year, Gwynn published another article which brought newly discovered material to light.32 He identified two more Irish kings who undertook this pilgrimage. The first travelled in 1026, thus preceding the journeys of Sigtryggr and Knútr. Furthermore this evidence demonstrated that most kings travelling to Rome were Irish rather than Hiberno-Scandinavian. Therefore this trend cannot be seen as a primarily Scandinavian phenomenon, and native Irish rulers were not lagging behind in their contacts with the rest of Christendom.33 Clerical links further demonstrate that direct links existed between Ireland, England and the Continent at this date. Irish churchmen had a strong presence in the Rhineland and Lorraine where reformed Benedictinism had taken hold.34 The comings and goings of Irish ecclesiastics, and mention of their influence at the Imperial court, can be seen in the chronicle of Marianus Scotus, an inclusus at Mainz.35 Many of the clerics travelling to the Continent would have travelled via England. Some would have returned or sent books and letters bringing news and ideas to Ireland.36 Nevertheless, a better-documented era of ecclesiastical contacts began during the pontificates of Lanfranc and Anselm at Canterbury. These archbishops consecrated several clerics, including Irishmen trained in England, as bishops of Dublin and Waterford.37 There was a strong political dimension to this aspect of reform, and it was associated with Canterbury’s claim of primacy over the churches of Ireland. I

s.a. 1034.2; Philip Grierson, ‘The Relations between England and Flanders before the Norman Conquest’, TRHS 4th ser. 23, 1941, 71–112, at 72–3; Aubrey Gwynn, The Irish Church in the Eleventh and Twelfth Centuries, Blackrock 1992, 37–8. 29 Aubrey Gwynn, ‘Ireland and Rome in the Eleventh Century’, Irish Ecclesiastical Record 5th ser. 57, 1941, 213–32. 30 John of Worcester, s.a. 1031. 31 Gwynn, ‘Ireland and Rome’, 224–6. 32 Gwynn, The Irish Church, 84–98. 33 Marie Therese Flanagan, Irish Society, Anglo-Norman Settlers, Angevin Kingship, Oxford 1989, 11. 34 Gwynn, ‘Ireland and Rome’; Jean-Michel Picard, ‘The Cult of Columba in Lotharingia (9th–11th Centuries): The Manuscript Evidence’, in Studies in Irish Hagiography: Saints and Scholars, ed. John Carey et al., Dublin 2001, 221–36. 35 Mariani Scotti Chronicon, ed. Georg Waitz, MGH SS 5, Hanover 1844, 481–564. 36 The royal dynasty of Uí Briain were prominent patrons of reform. This interest may have begun with Donnchad son of Brian: Annals of Inisfallen (MS Rawlinson B 503), ed. and trans. Séan Mac Airt, Dublin 1951, s.aa. 1040.6, 1050.2, 1064.5; [Annala Rioghachta Eireann], Annals of the [Kingdom of Ireland by the] Four Masters, ed. and trans. John O’Donovan, 2nd edn, 7 vols, Dublin 1856, s.a. 1050; David N. Dumville, Councils and Synods of the Gaelic Early and Central Middle Ages, Cambridge 1997, 35–6. 37 Flanagan, Irish Society, 7–55.

England and the Irish-Sea Zone

59

will treat this topic in greater depth below, while discussing political links between England and Ireland. It can be shown that clerics and pilgrims, as well as traders and merchants, were significant groups crossing the Irish Sea in the eleventh century. Contacts across the Irish Sea were not exclusive to the Hiberno-Scandinavian population of Ireland; and they increasingly included the native Irish. A consideration of political events highlights the importance of exiles, mercenaries, and diplomats, who also sailed across the Irish Sea. It is to this topic that I wish to devote the rest of the paper. From an Irish perspective, the crucial players in these relations were the viking dynasty of Ívarr, who ruled Dublin and the Hebrides, and the Irish provincial overkings who sought to bring Dublin under their control. However, I shall discuss these relations within the familiar framework of the reigns of successive English kings, beginning with Æthelred the Unready. Æthelred’s reign coincided with the onset of the so-called Second Viking Age in Britain. Viking attacks began in the 980s, and they led ultimately to the conquest of England by Sveinn, king of Denmark, in 1013. It can be argued that there was an Irish Sea dimension to these events. Simon Keynes has suggested that some of the raids against western England, at the beginning of Æthelred’s reign, were directed from viking colonies in the Gaelic world.38 It may be worth noting that viking attacks on Cheshire and Southampton in 980 coincided with a campaign against Wales by the Hiberno-Scandinavian king of Man and the Hebrides, Guðrøðr Haraldsson.39 Guðrøðr’s main enemy at this point, Hywel, king of Gwynedd, was allied with Ælfhere, earl of Mercia. This would have made Cheshire fair game for an attack. Guðrøðr also assailed Dyfed in 982, and his southerly campaigns may be linked with raids across the Bristol Channel, in Devon and in Cornwall, in 981 and 982.40 The early raids against western England could therefore represent the escalation of a political dispute which began between the dynasty of Ívarr and kings in Wales. Irish chronicles report the appearance of warships from Scandinavia in the Irish Sea from 986.41 This new fleets are identified as Danair, ‘Danes’.42 Their arrival was to have serious consequences for England. It was not long before an alliance developed between some of the Danish fleets and the Hiberno-Scandinavian dynasty of Ívarr. In 987, some joined forces with Guðrøðr Haraldsson on the Isle of Man.43 Their alliance may have continued into the following year, when raids were recorded against Devon and South Wales.44 Contingents from Ireland also joined in viking raids against south-west England. Numismatic evidence shows that coin dies taken from mints at Bath, Watchet and Lydford were used to produce some of the earliest coins minted at Dublin from about

38 Simon Keynes, ‘The Historical Context of the Battle of Maldon’, in The Battle of Maldon AD 991, ed.

Donald Scragg, Oxford 1991, 81–113, at 85. 39 ASC MS C, s.a. 980; Brut y Tywysogyon or The Chronicle of the Princes – Red Book of Hergest

Version, ed. and trans. Thomas Jones, 2nd edn, Cardiff 1973, s.a. 980. 40 ASC MS C, s.aa. 981, 982; Brut y Tywysogyon, s.a. 982. 41 Annals of Ulster, s.aa. 985[=986].2 and 3, 986[=987].1 and 3, 989[=990].1; Annals of the Four

Masters, s.aa. 985 [=986], 986 [=987], 989 [=990]. After 990 the term Danair does not reappear until 1014. 42 Dictionary of the Irish Language, Compact Edition, ed. E. G. Quin et al., Dublin 1983, 182, cols D: 82–4. This term seems to be introduced in chronicles initially to distinguish vikings from Scandinavia from vikings settled in Ireland. From the early eleventh century the label describes vikings in general. Its meaning later extends to foreigners and bandits. 43 Annals of Ulster, s.a. 986[=987].1. 44 ASC MSS CDE, s.a. 988; Brut y Tywysogyon, s.a. 988.

60

Anglo-Norman Studies XXVI

995. It is likely that these dies were stolen in raids.45 Furthermore, the supply of booty from England may have served as an economic stimulus as it was traded through the port. This may have influenced the decision to mint coin regularly in Dublin at this time. In 990 Scandinavian fleets also co-operated with Dubliners during their campaigns in Ireland.46 This highlights a sense of connection between events in Ireland and Britain. It is a striking that after 990 there are no more attacks by Danair (‘Danes’ rather than Hiberno-Scandinavians) recorded against Ireland.47 Nevertheless the number of raids on England dramatically increased during these years. England offered richer pickings and a more centralised system of government which would have enabled greater tributes to be raised; but this cannot be the whole story. The pattern of attacks is radically different from that of the First Viking Age when Ireland suffered in no less measure than England. It seems possible that the Danish fleets agreed, or decided, not to mount further attacks within the sphere of influence of the dynasty of Ívarr. The Scandinavian attacks on England from 991 were more focused on eastern areas of the kingdom, and they were on a bigger scale than those of the 980s.48 Nevertheless, these years appear to have been marked by increasing competition for control of the western sea-lanes. Brian Bóruma, overking of Munster, temporarily brought all the major viking ports in Ireland under his control, and he may have sought to extend his authority in Man and the Hebrides. He may have allied with, or subdued, Røgnvaldr, king of the Isles, as Røgnvaldr died in Munster in 1005.49 If so, Brian’s activities may have provoked Sveinn to intervene in Irish Sea affairs. He attacked the Isle of Man in the 990s, and some of his ships were wrecked off the Welsh coast in 1011.50 These actions may have then prompted English fleets to invade Strathclyde and Man in 1000 and Dyfed in 1011 to try to offset Sveinn’s influence there.51 According to Adam of Bremen, Sveinn also took shelter with a rex Scothorum, but the identity of this king, whether Scottish or Irish, is unknown.52 In 1000 a fleet which had been ravaging Wales and southern England, departed to Normandy.53 This event raises a further question of links between vikings based in the Irish Sea region and the rulers of Normandy. The Gesta Normannorum Ducum, written by William of Jumièges, may shed some light on this issue. This history was composed within living memory of the Danish conquest of England. However, the evidence sometimes poses problems of interpretation.54 William asserted that Knútr, son of Sveinn of Denmark, sought military help from Lagmann, king of the Swedes, and Óláfr, king of the Northmen, after the death of his

45 K. Bornholdt, ‘Myth or Mint? The Evidence for a Viking-Age Coinage from the Isle of Man’, in

Recent Archaeological Research on the Isle of Man, ed. P. J. Davey, BAR 1999, 199–213 at 204; Wallace, ‘The English Presence’, 212. 46 Annals of the Four Masters, s.a. 989 [=990]; Annals of Ulster, s.a. 989[=990].1. 47 Furthermore there is a general decrease in viking activity in Ireland in 1001×1013. 48 Keynes, ‘The Historical Context’, 88. 49 Etchingham, ‘North Wales’, 180; Annals of Inisfallen, s.a. 984.2. 50 Annales Cambriae (B), s.aa. 992, 1011; Brut y Tywysogyon, s.a. 995. 51 ASC MS E, s.a. 1000; Annales Cambriae (BC), s.a. 1011; Brut y Tywysogyon, s.a. 1012. 52 Adam of Bremen, Gesta Hammaburgensis Ecclesiae Pontificum, ii, 32 (ed. Bernard Schmeidler, MGH SRG 1917, 95; Adam of Bremen, History of the Archbishops of Hamburg-Bremen, trans. Francis J. Tschan, New York 1959, 78). 53 ASC MS E, s.a. 1000. 54 Jumièges i, xx.

England and the Irish-Sea Zone

61

father in 1014.55 Óláfr has been identified as Óláfr Haraldsson, future king and patron saint of Norway.56 However there was no king of Sweden called Lagmann. Adigard des Gautries recommended the emendation Sudrorum for Suauorum, thus identifying Lagmann as a king of the Hebrides.57 This interpretation has been espoused by Elisabeth van Houts who has noted that William of Jumièges was prone to make ‘ingenious identifications of peoples’.58 Thus Lagmann may have been a member of the Hiberno-Scandinavian dynasty of Ívarr which was active in the Irish Sea. Some support for this theory comes from the name. Lagmann is derived from Old Norse lögmadr (‘lawman’). This was used as a personal name in Ireland, the Northern Isles and the Hebrides from the tenth century, but it is not attested in Scandinavia.59 According to the Annals of Ulster and an early twelfth-century Irish saga Cocad Gaedel re Gallaib, Óláfr son of Lagmann Gudrødsson fought at Clontarf in 1014, alongside a contingent of warriors from the Hebrides.60 James Henthorn Todd concluded that Lagmann was a son of Guðrøðr, the king of the Isles and a member of the dynasty of Ívarr, who died in 989.61 Thus two independent lines of scholarship serve to identify Lagmann as a king of the Hebrides and Man in the early eleventh century. Nevertheless his career has been neglected by historians. William of Jumièges reported that, after Lagmann and Óláfr assisted Knútr, Richard, duke of Normandy, courted their support in his struggle against Odo, count of Chartres. The Capetian king Robert II intervened to arrange a peace between the two sides. As a result, Lagmann and Óláfr were presented with gifts and persuaded to return to their own countries.62 The precise chronology of these events is uncertain. However, it does seem that Lagmann was rubbing shoulders with leading figures in England, Normandy and Scandinavia. Lagmann was not the only viking from the Irish Sea zone whom Continental accounts identify as co-operating with Danes. The chronicler Ademar of Chabannes wrote that a joint fleet from Denmark and Ireland raided Aquitaine and then fought in Ireland, probably at the battle of Clontarf on Good Friday in 1014.63 In this conflict the vikings of Dublin and their allies opposed the troops of Brian Bóruma and his allies. Welsh and Scandinavian sources report that Sigtryggr, the king of Dublin, hired foreign fleets to fight for him in the battle.64 Doubtless there were mercenary 55 Jumièges ii, 18–21. 56 Jumièges ii, 24–9; Encomium Emmae Reginae, ed. and trans. Alistair Campbell, 2nd edn, Cambridge

1998, 76–82. Cf. Adam of Bremen, Gesta Hammaburgensis Ecclesiae ii, 51 (ed. Schmeidler, 112; trans. Tschan, 90). 57 Adigard des Gautries, Les Noms de personnes scandinaves en Normandie de 911 à 1066, Lund 1954, 69, note 12. 58 Jumièges i, xxxv, li and ii, 20, note 1. 59 Carl Marstrander, Bidrag til det Norske Sprogs Historie i Irland, Oslo 1915, 74; Donnchadh Ó Corráin, ‘The Vikings in Scotland and Ireland in the Ninth Century’, Peritia 12, 1998, 296–339, at 308. 60 Annals of Ulster, s.a. 1014.2; Cogadh Gaedhel re Gallaibh: The War of the Gaedhil with the Gaill, §78 (ed. and trans. James Henthorn Todd, RS 1867, 136–7). 61 Cogadh, ed. and trans. Todd, 271. 62 Jumièges ii, 24–7. 63 Ademari Cabannensis Chronicon iii, 53–5 (ed. P. Bourgain et al., CCCM 1999, 172–3); David N. Dumville, ‘Images of the Viking in Eleventh-Century Latin Literature’, in Latin Culture in the Eleventh Century, ed. Michael W. Herren et al., 2 vols, Turnhout 2002, i, 250–63, at 255. The date 1018 is editorial. It is likely, but not certain, that the battle described is Clontarf: Lesley Abrams, ‘England, Normandy and Scandinavia’, in A Companion to the Anglo-Norman World, ed. Christopher Harper-Bill and Elisabeth van Houts, Woodbridge 2003, 43–62, at 53; Benjamin Hudson, ‘Knútr and Viking Dublin’, Scandinavian Studies 66, 1994, 319–35, at 321. 64 Brut y Tywysogyon, s.a. 1014; Njáls Saga, §155 (ed. Einar Óline Sveinsson, Reykjavík 1954, 445;

62

Anglo-Norman Studies XXVI

fleets available in Insular waters following the Danish conquest of England. This battle in Ireland was reported in Britain, Scandinavia and the Continent, and this may reflect the wide geographical range of the participants hired to fight in this conflict. The events leading up to the Danish conquest of England therefore indicate significant contact between England, Ireland and the Hebrides. After Knútr became king in England in 1016, he maintained an interest in Irish Sea affairs. His claims in Denmark and in Norway involved him in foreign campaigns for part of his reign.65 These concerns may have encouraged Knútr to secure the western fringes of England from attack. One of Knútr’s skalds, Óttarr svarti, asserted that Knútr succeeded in becoming king of the Danes, the Irish, the English and the Isles.66 Rather much has been made of this statement by historians. To explore it further I shall discuss Knútr’s relations first with Ireland and then with the Hebrides. Benjamin Hudson has provided the fullest argument that Óttarr svarti called Knútr king of the Irish because he was overking of Dublin.67 The argument is based on a wide range of evidence, some of which is rather tentative. While there seem to have been close links between England and Dublin, it is possible that Knútr’s power over Dublin has been exaggerated.68 For example, it is far from clear that the Sitric dux who witnessed a charter (or charters) in 1026x1031, was Sigtryggr, king of Dublin, who ruled until 1036.69 Furthermore Hudson suggested that the church of St Bride in Fleet Street near was founded by Irish merchants brought to the city under the sponsorship of Knútr. However, there is no evidence to support this assertion: there is no clear evidence when the church was founded.70 The argument that Canterbury enjoyed primacy over Dublin during Knútr’s reign, which was advanced by Aubrey Gwynn, also needs to be questioned. This theory has been challenged in recent years, by Marie Therese Flanagan and others, due to the lack of contemporary evidence. Canterbury’s ecclesiastical authority over Dublin is first witnessed unambiguously in 1074 when Lanfranc consecrated a bishop of Dublin. A letter sent by Lanfranc, in this year, alludes to the custom of previous archbishops. However, this may or may not refer to the custom of consecrating bishops of Dublin.71 During Knútr’s reign, Æthelnoth of Canterbury took an active interest in foreign churches, which included his consecration of a Danish bishop.72 Thus it is possible, but far from certain, that a bishop of Dublin was consecrated in England during Knútr’s reign. Such a consecration need not have required formal recognition of Canterbury’s metropolitical

trans. Magnus Magnusson and Hermann Pálsson, Harmondsworth 1960, 344); cf. Benjamin Hudson, ‘Brjáns saga’, Medium Ævum 71, 2002, 241–68. 65 M. K. Lawson, Cnut, The Danes in England in the Early Eleventh Century, Harlow 1993, 81–116. 66 Den norsk-islandske skjaldedigtning, ed. F. Jónsson, 4 vols, Copenhagen 1908–15, A, i, 299 and B, i, 275. 67 Hudson, ‘Knútr’. 68 Aubrey Gwynn, ‘The First Bishops of Dublin’, Reportorium Novum, Dublin Diocesan Record 1, 1955–6, 1–26, at 3–6; Flanagan, Irish Society, 10; Lawson, Cnut, 106. 69 S 962, 963, 971. All these charters appear to have been drawn up at Crediton. Only one of the witness lists may be contemporary with the grant it subscribes. 70 The late tenth- or eleventh-century date assigned to this church is based on conjecture: Brooke, London, 138–40; Hudson, ‘Knútr’, 330–1. 71 Flanagan, Irish Society, 9–19; Philpott, ‘The English and Irish Churches’, 191–2; Martin Holland, ‘Dublin and Reform of the Irish Church in the Eleventh and Twelfth Centuries’, Peritia 14, 2000, 111–60, at 111–12. 72 Hudson, ‘Knútr’, 327; Lesley Abrams, ‘The Conversion of the Scandinavians of Dublin’, ANS 20, 1997 (1998), 1–29, at 27–8.

England and the Irish-Sea Zone

63

authority, as Lesley Abrams has pointed out.73 It should not be assumed that the assertions of primacy made by Canterbury after 1066 would have applied in Cnut’s reign. Circumstances within the English Church under Lanfranc were very different.74 While some aspects of Hudson’s argument are debatable, he has drawn attention to important evidence of contact between Knútr and Sigtryggr of Dublin. Hudson has highlighted the significance of a coin of Sigtryggr of Dublin minted from a die probably produced in Chester. This bears the legend Sihtric rex Irum (‘Sigtryggr king of the Irish’).75 It seems likely that this die was produced with Knútr’s consent. It is tempting to draw a parallel with the coin minted at Chester in 946×950 for the Welsh king Hywel Dda, during the reign of Eadred.76 Sigtryggr’s coin indicates strong trading links and co-operation between England and Dublin. This impression is reinforced by the large number of English coins found in Ireland, deposited from about 1000 to 1030. Hundreds of coins of Dublin from this period have also been found in Scandinavia.77 Hudson has drawn attention to collaboration between Sigtryggr and Knútr in their foreign policy. This is witnessed by the report in the Annals of Tigernach that the English and the men of Dublin led a joint attack on Wales in 1030.78 It seems that the two kings had complementary interests in Wales. Knútr seems to have encountered opposition from Wales throughout his reign. Russell Poole has argued, using the evidence of the skaldic verse Liðsmannaflokkr, that Welsh troops opposed Knútr during the siege of London in 1016.79 M. K. Lawson has also noted that Knútr granted lands to men in his service with greater density along the Welsh border, suggesting particular concern for the security of those regions.80 Knútr may have chosen to join forces with Sigtryggr in 1030 as his naval resources and influence in the Irish Sea might help to curb the power of troublesome Welsh kings. Co-operation between England and Dublin in Wales could have begun as early as 1022. Welsh chronicles report the arrival in this year of Rhain Scotus, a pretender to the kingship of South Wales.81 Rhain claimed to be a son of Maredudd (r.988–89). It seems that Rhain had spent some time in Ireland and gathered military support there. Nevertheless, it is far from clear who supported him.82 The Hiberno-Scandinavian dynasty of Ívarr had a long history of intervening in Welsh affairs. Nevertheless, Seán Duffy has convincingly argued that Welsh chronicles identify Rhain’s supporters as Gaels rather than vikings from Ireland.83 Was this Irish contingent therefore seeking to establish a candidate in Wales contrary to the interests of the dynasty of Ívarr? If so, potential supporters of Rhain include Niall, overking of Ulaid (who was locked in a feud with Dublin at this date), Donnchad of Brega and Donnchad of Munster.84 In 73 Lesley Abrams, ‘The Anglo-Saxons and the Christianization of Scandinavia’, ASE 24, 1995, 213–49,

at 227. 74 Flanagan, Irish Society, 13–16. 75 Hudson, ‘Knútr’, 323–4. 76 C. E. Blunt, ‘The Cabinet of the Marquess of Ailesbury and the Penny of Hywel Dda’, British Numis-

matic Journal 52, 1982, 117–22. 77 Wallace, ‘The English Presence’, 208–9. 78 Annals of Tigernach, s.a. 1030 (ed. and trans. Whitley Stokes, 2nd edn, 2 vols, Felinfach 1993). 79 Russell Poole, ‘Skaldic Verse and Anglo-Saxon History: Some Aspects of the Period 1009–1016’,

Speculum 62, 1987, 265–98. Lawson, Cnut, 165. Annales Cambriae (BC), s.a. 1022; Brut y Tywysogyon, s.a. 1022. David E. Thornton, ‘Who was Rhain the Irishman?’, Studia Celtica 34, 2000, 131–48. Duffy, ‘Ostmen’, 382–4. Annals of Ulster, s.aa. 1022.4, 1023.2, 1026.1.

80 81 82 83 84

64

Anglo-Norman Studies XXVI

the same year as Rhain’s coup, an English fleet attacked South Wales.85 It seems unlikely that this was a coincidence. Earl Eilaf, who led the expedition, was high in royal favour at the time and probably had Knútr’s support.86 The attack may have been timed to challenge Rhain and his Irish supporters. Knútr was perhaps suspicious of letting an Irish ruler extend his power into Britain at this date. He may have also wished to protect the trading interests of Dubliners along the Bristol Channel. As I have already mentioned, the English and the men of Dublin attacked Wales in 1030. The expedition probably targeted Rhydderch ap Iestyn, a southern king who had recently extended his rule over Gwynedd.87 Both Knútr and Sigtryggr of Dublin had cause to resent Rhydderch’s growing power. This mighty Welsh king could threaten the English border, while Sigtryggr’s interests in North Wales would be affected by Rhydderch’s possession of Gwynedd. According to the thirteenth century Welsh ‘Life of Gruffudd ap Cynan’, Sigtryggr’s son Óláfr founded a fort on Anglesey before 1034.88 As the Annals of Tigernach report that Óláfr paid 120 Welsh horses to the king of Brega in 1029, the fort could have been founded before that date.89 To cement Dublin’s influence in Wales, Óláfr’s daughter married Cynan ab Iago, one of Rhydderch’s rivals in Gwynedd.90 It seems that Knútr was willing to allow Sigtryggr to extend his influence in Wales as a counter-balance to Rhydderch.91 Knútr was clearly more powerful than Sigtryggr. Nevertheless, historians may have exaggerated the extent of England’s dominion over Dublin in these years. I have already discussed the evidence for collaboration between Danes and Dubliners prior to Knútr’s reign. The alliance between Sigtryggr and Knútr may have been a natural extension of this friendship. The two rulers co-operated closely in trade, in their policy towards Wales, and perhaps in religious matters. However, there is nothing in English or Irish sources which demonstrates that Knútr was the overking of Sigtryggr. Knútr’s relations with Echmarcach, king of Man and the Hebrides, appear to have been less agreeable than his friendship with Sigtryggr. The Anglo-Saxon Chronicle reports that Knútr received the submission of Echmarcach in 1031, along with Mael Colaim of Alba and Macbeth of Moray.92 Echmarcach was a member of the Hiberno-Scandinavian dynasty of Ívarr. This may have led Óttar svarti to assert that Knútr held ‘Irishmen’ under his sway. Knútr may have chosen to demand Echmarcach’s obedience, because he was an enemy of Sigtryggr. Knútr may also have feared that he might ally with a foreign king. Echmarcach and Sigtryggr were rival members of the dynasty of Ívarr. Sigtryggr’s activities in Wales curtailed Echmarcach’s influence in the Irish Sea. Previous kings of Man and the Hebrides had played an active role in Wales.93 This may have prompted Echmarcach to develop an alternative power-base on the Rhinns of

85 Annales Cambriae (BC), s.a. 1022. 86 Simon Keynes, ‘Cnut’s Earls’, in The Reign of Cnut, King of England, Denmark and Norway, ed.

Alexander R. Rumble, London 1994, 43–88, at 58–60. 87 Maund, Ireland, 162. 88 A Mediaeval Prince of Wales: The Life of Gruffudd ap Cynan, ed. and trans. D. Simon Evans, Felinfach 1990, 24, 55. Paul Russell is preparing an edition of the newly discovered twelfth-century Latin version. 89 Annals of Tigernach, s.a. 1029; Etchingham, ‘North Wales’, 162. 90 A Mediaeval Prince, 24, 54. 91 Brut y Tywysogyon, s.aa. 1033, 1035. The Irish killed Rhydderch and the English killed his son Caradog. 92 ASC MS E, s.a. 1031; Benjamin T. Hudson, ‘Cnut and the Scottish Kings’, EHR 107, 1992, 350–60. 93 Mariani Scotti Chronicon, s.a. 1065.

England and the Irish-Sea Zone

65

Galloway, which was mentioned at his death.94 Echmarcach ousted Sigtryggr from Dublin soon after Knútr’s death. It may be that Echamarcach was prevented from acting sooner because of his oath to Knútr. Echmarcach’s predecessor Lagmann had been allied both to Duke Richard II of Normandy and to Óláfr, king of Norway. Neither power was well disposed towards Knútr in 1030.95 Knútr may have taken action in 1031 to ensure that Echmarcach did not continue an alliance either with Richard or with a rival faction in Norway.96 Local rivalries in the Irish Sea zone therefore appear to take on a broader political dimension during the reign of Knútr due to a network of external relations. During the long reign of Edward the Confessor in England, Anglo-Irish relations assumed a different character. The Vita Ædwardi Regis stresses the harmony which Edward enjoyed with his neighbours at the beginning of his reign, ‘all Britain, together with the jagged islands of adjacent kingdoms and monarchies, settling down in the calm of peace’.97 Nevertheless, the image may be more literary ideal than reality. Unlike Knutr, Edward did not enjoy good relations with Dublin. Instead Ireland became a haven for subjects who flouted the authority of the English king. The most important aristocratic faction in England during Edward’s reign was the house of Godwine. There is evidence that the family had important and lasting connections with Ireland. Godwine’s daughter Edith was fluent in Irish. In Vita Ædwardi her skill is presented as highly prestigious and equal in significance to her knowledge of Danish and French (perhaps hinting that other English nobles at this date sought to learn these languages). Members of the family of Godwine fled to Ireland in political exile on more than one occasion.98 It appears that the families of Godwine and the overking of Leinster, Diarmait, also exchanged gifts.99 Nevertheless, some unnecessary doubt and surprise has been expressed about this evidence.100 Consequently, the importance of this connection has been somewhat underestimated. The relationship of the house of Godwine with Ireland may have originated in common economic interests. Godwine controlled ports in Sussex, which he inherited from his father, and ports in Wessex, where he was made earl by Knútr.101 These interests would have brought him into contact with Irish merchants from an early date. In addition, he probably benefited from growing trade between Ireland and the south and west of England during the eleventh century. The Anglo-Saxon Chronicle reports that Godwine exported slaves, and his wife is also reputed to have done so.102 Tolls from Sussex ports for the sale of slaves are also mentioned in Domesday

94 95 96

Francis John Byrne, ‘Onomastica 2: Na Renna’, Peritia 1, 1982, 267. Simon Keynes, ‘The Æthelings in Normandy’, ANS 13, 1990 (1991), 173–205; Hudson, ‘Cnut’, 359. For Duke Richard’s links with the Irish Sea, see: Rodulphus Glabri Historiarum libri quinque, ii, 3, ed. and trans. J. France, Oxford 1989, 54–7; De Moribus et Actibus Primorum Normanniae Ducum, §103, ed. Jules Lair, Caen 1865, 265; Dudo of St Quentin, History of the Normans, trans. Eric Christiansen, Woodbridge 1998, 140. 97 Vita Ædwardi, 118–19: ‘ut quiescente in pacis quiete uniuersali Britannia, cum adaicentium regnorum monarchiarumque angularibus insulis’. 98 Vita Ædwardi, 22–3, 40–1; Ben Hudson, ‘The Family of Harold Godwinson and the Irish Sea Province’, JRSAI 109, 1979, 92–100. 99 Annals of Inisfallen, s.a. 1068.5; Frank Barlow, The Godwins, the Rise and Fall of a Noble Dynasty, Harlow 2002, 120. 100 Ibid., 24; Vita Ædwardi, lxvi; Frank Barlow, Edward the Confessor, London 1970, 120. 101 Keynes, ‘Cnut’s Earls’, 70–4. 102 ASC MSS CD, s.a. 1036; Gesta Regum i, 363 (bk II, c. 200); David Pelteret, ‘Slave Raiding and Slave Trading in Early England’, ASE 9, 1981, 99–114.

66

Anglo-Norman Studies XXVI

Book.103 As Dublin was a centre for the slave trade in Northern Europe, Godwine’s dealings probably encouraged contacts across the Irish Sea. As Godwine’s family received further promotion during Edward’s reign, their economic links with Ireland would have grown stronger. When Edith married King Edward, she received Winchester and Exeter as morning gifts. Both towns enjoyed considerable trade with Ireland and Edith benefited from their wealth.104 Godwine’s sons Swegn and Harold also received earldoms in the 1040s. Significantly, Swegn’s earldom included control of Bristol. This port was of pivotal importance for Anglo-Irish trade in the eleventh century.105 The family of Godwine thus derived income from areas which had trade with Ireland, and which might be vulnerable to attack from Ireland.106 In the light of these circumstances, Edith’s knowledge of Irish and her family’s contacts with Ireland may be interpreted. The first evidence of contact between the family of Godwine and Dublin is, however, hostile. The two parties initially supported opposing factions in Wales. From 1039 a claimant to the throne of Gwynedd, Cynan ab Iago, took refuge in Dublin.107 He was exiled during the reign of Gruffudd ap Llywelyn, king of Gwynedd, who succeeded Cynan’s father.108 The Dublin vikings led campaigns against Gruffudd in 1042 and 1044 and provided support for his enemies in South Wales.109 Nevertheless, in 1046 Gruffudd ap Llywelyn allied with Swegn Godwinesson, and they campaigned together in Wales.110 Perhaps in retaliation, a viking fleet from Ireland joined ranks with a rival of Gruffudd ap Llywelyn in 1049. They ravaged English territory beyond the Usk while Swegn Godwinesson was in exile in Flanders.111 Perhaps as a result of this hostility, the house of Godwine allied with the overking of Leinster, Diarmait mac Maól na mBó, who controlled the viking ports of Wexford and Waterford. The same king also conquered Dublin in 1052. In 1051, a connection between Diarmait and the house of Godwine is first witnessed when the family was outlawed by Edward. Two sons of Godwine, Harold and Leofric, fled to Diarmait. They took a ship from Bristol which had been prepared by their brother Swegn. Meanwhile Godwine and another son, Tostig, went to Flanders.112 The family’s prior links with Flanders are well attested.113 This raises the question why some of the family chose to seek refuge elsewhere.114 They may have decided to split for security reasons, should a host prove false, or in case enemies sought to pursue them. The division also enabled the family to seek support in different locations. Flanders and Ireland were both places where a mercenary fleet could be recruited. Frequent commerce between the three countries would also 103 104 105 106

Domesday Book, Sussex (26a) 12, 1. Domesday Book, Devon (100a) C, 2. Gwynn, ‘Medieval Bristol’. Peter A. Clarke, The English Nobility under Edward the Confessor, Oxford 1994, 164–205; David Hill, An Atlas of Anglo-Saxon England, Oxford 1981, 100–5. King Edward would also have benefited significantly from Irish trade. 107 Duffy, ‘Ostmen’, 386. 108 Annales Cambriae (C), s.a. 1039; Brut y Tywysogyon, s.a. 1039. 109 The terminology used in Welsh chronicles indicates that vikings from Ireland opposed Gruffudd ap Llewelyn in 1042 and 1044. Other unidentified vikings (whom it is tempting to associate with the kingdom of the Isles) waged war against Gruffudd’s enemies in the south in 1039 and 1042: Annales Cambriae (BC), s.aa. 1039, 1042, 1044; Brut y Tywysogyon, s.aa. 1039, 1042, 1044. 110 ASC MS C, s.a. 1046; Maund, Ireland, 127–8. 111 ASC MS D, s.a. 1049. 112 ASC MSS CDE, s.a. 1051. 113 Grierson, ‘The Relations’, 95–100. 114 Barlow, Edward, 120.

England and the Irish-Sea Zone

67

enable messages to pass between members of the family during their exile. By using these opportunities, the family of Godwine was able to organise a two-pronged attack on England in 1052. While Harold sailed from Ireland with nine ships, Godwine set out with a fleet from the River Yser. Their campaign successfully and dramatically restored the house of Godwine to power.115 Diarmait no doubt hoped to gain from his timely support. His link with the house of Godwine would have brought some prestige. He may also have sought economic benefits. The Anglo-Saxon Chronicle reports that Harold seized cattle, captives, and property in Somerset.116 Some of this loot may have made its way back to Ireland. Diarmait may have hoped also for future benefits from this relationship. The next prominent outlaw who fled from England to Ireland was Ælfgar, son of the Mercian earl Leofric. Ælfgar had fallen foul of the house of Godwine, and he was accused of treachery in 1055. He promptly withdrew to Ireland and collected eighteen warships. He then sought help from Gruffudd ap Llywelyn, king of Gwynedd, and they invaded Herefordshire.117 Harold Godwinesson was sent to put down the invasion, but he failed to engage his opponents in battle. In the truce which resulted, Ælfgar was restored to his former status. Three years later Ælfgar was exiled again. On this occasion he allied with Gruffudd ap Llywelyn and a fleet under the command of the Norwegian prince, Magnús, son of Haraldr hardráði. This included vikings from the Orkneys, the Hebrides, and Dublin.118 The combined forces attacked England, and Ælfgar was again restored. There has been some speculation as to where Ælfgar recruited support in Ireland. Kari Maund regarded the approach of Irish kings as opportunistic and mercenary in their involvements in England. She credited Diarmait mac Maól na mBó with providing help first to the family of Godwine in 1052, and then to their enemies in 1055.119 Colmán Etchingham has nevertheless questioned this conclusion. He has suggested that Ælfgar may have had recourse to Diarmait’s great rival, Donnchad, overking of Munster in 1055.120 During his second exile Ælfgar received help from an Irish Sea fleet under Norwegian control. At this date the Norwegian crown seems to have backed a branch of the Hiberno-Scandinavian dynasty of Ívarr which was opposed to Diarmait. The alliance which Diarmait had fostered with the family of Godwine appears to have weathered the events of the 1050s. They made common cause against Gruffudd ap Llywelyn after Ælfgar’s death in 1062. In 1063, Harold Godwinesson and his brother Tostig led a major campaign against Gruffudd by sea and land. The Dubliners had sheltered Gruffudd’s rival Cynan ab Iago and Diarmait may have assisted Cynan’s return to Wales at this point. According to the Annals of Ulster the son of Iago killed Gruffudd.121 The king’s head and the prow of his ship were sent to Harold as gifts. These trophies were then forwarded to King Edward.122 The success of this campaign may be a consequence of co-operation between Diarmait and the family of Godwine. The assistance given by Magnús Haraldsson, Hebrideans and Dubliners to Ælfgar

115 116 117 118 119 120 121

Vita Ædwardi, 40–5. ASC MSS EF, s.a. 1052. ASC MSS CDE, s.a. 1055; Maund, Ireland, 136. Annals of Tigernach, s.a. 1058. Maund, Ireland, 135, 165. Etchingham, ‘North Wales’, 154. Annals of Ulster, s.a. 1064.8; Benjamin T. Hudson, ‘The Destruction of Gruffudd ap Llywelyn’, Welsh Historical Review 15, 1990–1, 331–50. 122 ASC MSS DE, s.a. 1063.

68

Anglo-Norman Studies XXVI

in 1058 also seems to have been motivated by something other than immediate gain. Magnús was attempting to extend Norwegian authority in the Irish Sea by backing a faction opposed to Echmarcach.123 It could be argued that Magnús was attempting to strengthen Norwegian power in the Isles as a future springboard for the conquest of England.124 The seriousness of Norwegian designs on England were demonstrated following the death of Edward the Confessor in 1066 when Haraldr hardráði invaded England.125 Adam of Bremen mentioned the participation of an Irish king and a Scottish king in this campaign. This indicates the presence of an Irish Sea contingent at the Battle of Stamford Bridge.126 It is striking that during the reign of Edward the Confessor political exiles could rally support in Ireland which helped to restore them to power. These events suggest the increasing willingness of Irish rulers to participate in events in England. In the reign of William the Conqueror, this threat from Ireland continued. In the immediate aftermath of the battle of Hastings, two of Harold’s sons fled to King Diarmait in Leinster.127 Harold’s wife Ealdgyth initially travelled to Chester, and his mother Gytha fled to Exeter.128 Both were ports of departure for Ireland.129 At this stage Gytha may have made contact with her nephew, the Danish king Sveinn Estrithson. Two sons and a daughter of Harold fled to his court. From these refuges, the family plotted their return to power. Diarmait supported two attempts by the sons of Harold to win control of England. The first was made in 1068. Godwine Haroldsson led a ‘pirate-fleet’ up the River Avon to Bristol. However, the townsmen made a successful defence. The fleet then landed in Somerset, where an indecisive battle was fought, and Godwine returned to Ireland.130 A second invasion was planned in 1069. This coincided with an invasion from Denmark and a series of local rebellions against William the Conqueror. Orderic Vitalis related that Diarmait helped to raise a fleet of sixty-six ships under the command of two sons of Harold. This fleet then sailed to Exeter, but it was heavily defeated. According to Orderic only two small boatloads of survivors made their way back to Ireland.131 After the failure of these uprisings, William vigorously and infamously set about subduing northern parts of the kingdom. The efforts of the sons of Harold had failed. Their supporters, Diarmait, overking of Leinster and Dublin, and Sveinn of Denmark, had taken a chance in supporting them. Nevertheless, the gamble could have reaped significant rewards, had it paid off. Hope that William the Conqueror could be defeated seems to have persisted, at least in Denmark. Sveinn Estrithson planned another invasion of England in 1075. He may have made overtures to Dublin for support. It is interesting that a warship made from Dublin oak c.1042 and repaired in the Irish Sea region c.1075 has been recovered from Roskilde

123 Cf. Chronica Regum Mannie et Insularum, s.a. 1047 [=1066] (ed. and trans. George Broderick,

Belfast 1979); Duffy, ‘Irishmen’, 106–8. 124 Maund, Ireland, 166, has suggested that the Irish Sea was an odd place to begin an invasion of

England. However, it was a good recruiting ground for military support. ASC MSS CDE, s.a. 1066. Gesta Hammaburgensis Ecclesiae, iii, 51 (ed. Schmeidler, 356; trans Tschan, 159). Orderic ii, 224–5. Ibid.; John of Worcester, s.a. 1066; Hudson, ‘The Family’, 93. J. O. Prestwich, ‘Military Intelligence under the Norman and Angevin Kings’, in Law and Government in Medieval England and Normandy: Essays in Honour of Sir James Holt, ed. George Garnett and John Hudson, Cambridge 1994, 1–30, at 5–8. 130 ASC MS D, s.a. 1068. 131 Orderic ii, 224–5. 125 126 127 128 129

England and the Irish-Sea Zone

69

fjord.132 It is possible that Diarmait’s successor in Dublin, Guðrøðr Óláfsson, was providing warships for Sveinn. Nevertheless, Guðrøðr died in the same year, and Sveinn’s expedition never landed in England.133 Benjamin Hudson has argued that Toirrdelbach Ua Briain of Munster, the new overking of Dublin, followed a more friendly policy towards William than either Diarmait or Guðrøðr.134 This theory is supported by numismatic evidence. From about 1074 until about 1088 the Dublin mint imitated English coin-issues. The weight ratio of Irish to English coins in this period also remained constant at the workable rate of 3:2.135 Hudson has interpreted this as a sign of better trade relations between the two kingdoms. William also seems to have encouraged closer ecclesiastical links between England and Ireland. Nevertheless these links do not necessarily indicate friendship between the kings of England and Munster. Lanfranc consecrated bishops of Dublin in 1074 and 1085, and letters have survived from him to Guðrøðr, king of Dublin, and to Toirrdelbach Ua Briain of Munster and his clergy.136 Toirrdelbach Ua Briain may not have been the prime mover in securing the consecration of a bishop of Dublin in Canterbury in 1074, as Guðrøðr, a member of the dynasty of Ívarr, ruled there until 1075. Guðrøðr may have considered that Bishop Patrick’s consecration in England was desirable as it avoided accepting the authority or intervention of one of the Irish ecclesiastical familiae. A more distant ecclesiastical overlord may have been preferred.137 In the text of Patrick’s profession of obedience to Canterbury, the use of terminology is politically charged. Lanfranc used the title ‘primate of the British Isles’. Dublin is called ‘the metropolitan church of Ireland’. This served both to flatter the powers of Dublin and to strengthen Lanfranc’s claims.138 It is unlikely that the clergy of Toirrdelbach Ua Briain would have accepted these assertions. In practice the titles were quite unfounded but they were perhaps meant to establish a precedent for future claims. After Toirrdelbach took the overkingship of Dublin in 1075, Lanfranc sought to extend his own influence in Munster. He wrote to the leading bishop of the province and his clergy.139 In 1085, a new bishop of Dublin, Donngus, was consecrated in Canterbury. Flanagan has pointed out that the wording of his profession of obedience was very different from his predecessor’s.140 Lanfranc was styled merely ‘archbishop of Canterbury’, and Dublin was not referred to as a ‘metropolitan church’.141 It may 132 Sean McGrail, ‘Ship Timbers from Wood Quay, Dublin and Other Medieval Sites in Ireland’, Bullán

1, 1994, 49–61, at 50; N. Bonde and O. Crumlin-Pedersen, ‘The Dating of Wreck 2 from Skuldelev, Denmark’, Newswarp 7, 1990, 3–6; Tinna Damgård Sørensen, ‘Les bateaux Skuldeler (Roskilde) et leurs répliques modernes’, in L’héritage maritime des Vikings en Europe de l’Ouest, ed. Elisabeth Ridel, Caen 2002, 199–227, at 204. 133 ASC MSS DE, s.a. 1075; Annals of Inisfallen, s.a. 1075.2. 134 Benjamin Hudson, ‘William the Conqueror and Ireland’, Irish Historical Studies 29, 1994–5, 145–58. 135 Ibid., 153–4; Wallace, ‘The English Presence’, 209. 136 Lanfranc’s Letters, nos 9, 10, 49. 137 In the late tenth century, kings of Dublin developed links with the church of Iona and the familia of Columba. The growing influence of their rivals, Clann Cholmáin, over the church of Kells, which headed the familia, may have encouraged Dubliners to foster links elsewhere: Edel Bhreathnach, ‘Columban Churches in Brega and Leinster: Relations with the Norse and the Anglo-Normans’, JRSAI 129, 1999, 5–18. 138 Canterbury Professions, ed. M. Richter, Torquay 1973, no. 36; Philpott, ‘The English and Irish Churches’, 193–7; Holland, ‘Dublin and Reform’, 112–16. 139 Lanfranc’s Letters, no. 49. 140 Flanagan, Irish Society, 19; but cf. Holland, ‘Dublin and Reform’, 118–19. 141 Canterbury Professions, no. 51.

70

Anglo-Norman Studies XXVI

be that Toirrdelbach Ua Briain and his bishops objected to Lanfranc’s claims and that they had to be toned down. Nevertheless, Toirrdelbach encouraged the consecration, perhaps to serve his own political agenda: the church of Armagh claimed primacy over the churches of Ireland, and its relations with Toirrdelbach were perhaps cooling at this date.142 Lanfranc’s intervention may have been used by Toirrdelbach to snub Armagh. Lanfranc’s influence over the Irish Church was encouraged by the papacy. Gregory VII wrote to Lanfranc in the 1070s, mentioning rumours of the immoral marriage practices of the Scotti.143 Lanfranc was vested with authority to punish these crimes. As the Normans appear to be the generators of anti-Irish propaganda from the eleventh century, it is tempting to attribute to them the rumours of Irish immorality which incensed Gregory.144 After all, adopting the moral high ground had served as a useful political weapon in 1066 when the papacy blessed the banners of William the Conqueror.145 Lanfranc may have hoped that by reporting on the poor state of the Irish Church he would be assigned responsibility to sort it out. It seems that there was scope for competition between Uí Briain overkings and William the Conqueror, as well as common interest. In particular, Uí Briain seem to have resented the rapid conquest of territory in North Wales achieved by the Normans.146 Perhaps in an effort to counter this intrusion, Uí Briain supported the claims of Gruffudd ap Cynan to the throne of Gwynedd. In 1075 Gruffudd defeated his Welsh enemies with the support of Muirchertach Ua Briain, who was then king of Dublin.147 However, a revolt soon forced Gruffudd back to Ireland. Another invasion was made in 1081, with the support of Muirchertach’s brother.148 As a result Gruffudd was restored to power in the north, and his ally Rhys ap Tewdwr was placed in a much stronger position in the south.149 Both rulers were thus indebted to Irish support. These actions may have provoked William to intervene directly in Welsh affairs. In 1081 he visited St Davids and seems to have struck a deal with Rhys.150 Robert Babcock has highlighted evidence in Domesday Book that Rhys agreed to pay William an annual tribute of forty pounds, in exchange for recognition of his power in South Wales.151 This agreement drew Rhys within William’s sphere of influence. Gruffudd was subjected by harsher means, as he was captured by Hugh, earl of Chester, and imprisoned for twelve years.152 The speed with which Irish intervention in Welsh politics brought about William’s reaction, suggests that he was not entirely trusting of their motives. William may have perceived Ireland as a potential threat throughout his reign. Orderic Vitalis identified the Irish as among his most dangerous foes.153 The island

142 Armagh was allied with Cenél nEogain, who became a threat to Uí Briain, following the accession of

Domnall Mac Lochlainn, overking of Cenél nEogain, in 1083. Friendship between Uí Briain and Armagh was restored in 1103: Ó Corráin, Ireland, 141, 147; Holland, ‘Dublin and Reform’, 133. 143 Lanfranc’s Letters, no. 8. 144 Dumville, ‘Images of the Viking’. 145 It is notable that fewer royal pilgrims travelled from Ireland to England after 1066. Gwynn, The Irish Church, 37–8, attributed this to worsening relations between England and Ireland. 146 Hudson, ‘William’, 148. 147 A Mediaeval Prince, 28–9, 59; Annales Cambriae (C), s.a. 1075; Brut y Tywysogyon, s.a. 1075. 148 A Mediaeval Prince, 36–8, 67–9; Annales Cambriae (BC), s.a. 1081; Brut y Tywysogyon, s.a. 1081. 149 Robert S. Babcock, ‘Rhys ap Tewdwr, King of Deheubarth’, ANS 16, 1993 (1994), 21–35, at 26. 150 Annales Cambriae (BC), s.a. 1081; Brut y Tywysogyon, s.a. 1081; ASC MS E, s.a. 1081. 151 Babcock, ‘Rhys’, 21. 152 A Mediaeval Prince, 38, 69. 153 Orderic iv, 42–3.

England and the Irish-Sea Zone

71

had long served as a refuge for political enemies and as a recruiting ground for mercenaries and fleets.154 The concern of William’s earls regarding attacks from the Irish Sea was very tangibly manifested in the castles they built along the coast.155 William may have sought to resolve these dangers though diplomatic channels. According to the Anglo-Saxon Chronicle, ‘If he had lived only two years more he would have conquered Ireland by his astuteness and without any display of force.’156 This has led to speculation. It evokes a similar boast of the Roman governor Agricola in AD 81. Tacitus wrote: ‘I have often heard Agricola say that Ireland could be reduced and held by a single legion and a few auxiliaries, and the conquest would also pay from the point of view of Britain, if Roman arms were in evidence on every side and liberty banished off the map.’157 Nevertheless, the Romans did not conquer Ireland. Similarly, the statement in the Anglo-Saxon Chronicle should not be used to suggest that Ireland’s subjection by the English in the 1170s was an inevitable progression of events from the reign of William the Conqueror.158 In the reign of William Rufus, contingents from across the Irish Sea continued to oppose the extension of Norman power in Wales.159 Nevertheless, ecclesiastical links remained strong as bishops from Dublin and Waterford were consecrated at Canterbury.160 Relations between Muirchertach Ua Briain and England became more strained during the reign of Henry I. In a fascinating sequence of events, Muirchertach married one of his daughters to the rebel Arnulf de Montgomery. He then married another to Magnús, king of Norway, while Magnús was seeking to establish a power base in the Irish Sea.161 This prompted Henry to impose a temporary embargo on Irish trade.162 After 1101 Muirchertach Ua Briain stopped the consecration of bishops of Irish ports in England.163 This practice was resumed only after his death in 1119.164 In conclusion, it seems that Ireland had an important role in English affairs in the eleventh century. This relationship was expressed through trade and religious links, and through the political involvements of Irish kings in English affairs. There are

154 155 156 157

Flanagan, Irish Society, 48–9, 59–60. A Mediaeval Prince, 39, 70. ASC MS E, s.a. 1086. Tacitus, Agricola, §24 (Tacitus, Dialogus, Agricola, Germania, ed. and trans. William Peterson, London 1914, 211). This was perhaps imitated by Gerald of Wales, Itinerarium Cambriae, ii.2 (Giraldi Cambrensis, vi, 108–9; The Journey through Wales and the Description of Wales, trans. Lewis Thorpe, Harmondsworth 1978, 169). 158 John Gillingham, ‘A Second Tidal Wave? The Historiography of English Colonization of Ireland, Scotland and Wales in the 12th and 13th Centuries’, in Historiographical Approaches to Medieval Colonization of East Central Europe, ed. Jan M. Piskorski, Boulder 2002, 303–27, at 315, has noted that the use of the label ‘Norman’ to identify the invaders of 1169 has the implied effect of ‘linking subsequent conquests in Wales and Ireland with the Norman Conquest of England in 1066’. 159 Annales Cambriae (BC), s.aa. 1088, 1091; Brut y Tywysogyon, s.aa. 1088, 1091. 160 Flanagan, Irish Society, 20–1. 161 Anthony Candon, ‘Muirchertach Ua Briain, Politics and Naval Activity in the Irish Sea, 1075 to 1119’, in Keimelia, ed. Mac Niocaill and Wallace, 397–415; E. Curtis, ‘Muirchertach O’Brien, High King of Ireland and his Norman Son-in-law, Arnulf de Mongomery’, JRSAI 51, 1921, 116–24. Curtis perceived Arnulf as a precursor of Strongbow, whose example planted the idea of conquest ‘in the minds of the restless and greedy Norman race’. 162 Gesta Regum i, 738–9. bk V, c. 409. This passage exaggerated Muirchertach’s obedience to Henry. Similarly, Chronica Regum Mannie, s.a. 1098, exaggerates his obedience to Magnús berfœttr (‘Barefoot’ or ‘Barelegs’). 163 Muirchertach improved relations with Armagh: Annals of Ulster, s.aa. 1103.5, 1105.3, 1106.6; Holland, ‘Dublin and Reform’, 128–33. 164 Flanagan, Irish Society, 25–30.

72

Anglo-Norman Studies XXVI

recurrent themes in these relations, such as the passage of exiles or enemies across the Irish Sea to seek help, and the attempts of rulers on both sides of the Irish Sea to influence Welsh politics. While initially these links focused on the HibernoScandinavian population in Ireland, increasingly in the eleventh century the Irish themselves played a crucial role in events across the Irish Sea. It seems appropriate to ask why Irish rulers in the eleventh century sought greater involvement in English affairs. To some extent this can be explained by political developments within Ireland. As power became more centralised, claimants to the overkingship of Ireland may have sought to validate their power through external recognition. Furthermore, as Irish kings controlled viking towns from the late tenth century, their interests were naturally drawn into the external political and economic interests of these ports.165 However, the question also has a broader dimension. As Michael Richter has demonstrated, Irish kings were not only interested in England. They were interested in external affairs in general.166 This awareness was manifest early on by the pilgrimages to Rome which were undertaken from the 1020s. Irish rulers were not exceptional in their attitude. Both at the fringes of Christendom and at the core, the eleventh century was characterised by growing aristocratic mobility and communication across regional boundaries.167 Ultimately these developments sprang from increasing opportunities. The growth of trading networks (which was to some degree indebted to vikings) facilitated external communication.168 In Ireland, as elsewhere, individuals responded to rising opportunities for external contact, encouraged perhaps by ambition and natural curiosity. I have argued that modern narratives of Anglo-Irish relations in the eleventh century are sometimes overshadowed by the consequences of the invasion of 1171. By projecting back the origins of English overkingship in Ireland, and downplaying Ireland’s earlier political significance, historians may imbue the invasion with an air of predictability (and, in extreme cases, justification). Comparison may be made with the debate over conditions within Ireland in the eleventh and early twelfth centuries.169 One perspective is that Ireland was economically and politically backward. Therefore invasion by a more developed and progressive neighbour was ‘inevitable’.170 Historians opposed to this view have stressed that Ireland was becoming part of a European mainstream in the eleventh century.171 In order to contribute to, or to break away from, these debates, more research is necessary in the relatively

165 See above, note 8. 166 Michael Richter, ‘The European Dimension of Irish History in the Eleventh and Twelfth Centuries’,

Peritia 4, 1985, 328–45. 167 Bartlett, The Making of Europe: Conquest, Colonisation and Cultural Change 950–1350, London

1993. One example, to illustrate this trend, is the Kingdom of Alba. Macbeth (1040–57) undertook a pilgrimage to Rome and allowed Normans to settle in his territory. His successor Maelcolaim III (1058–93) married foreign royal ladies, first, Ingibjorg, and then Margaret. His sons famously laboured to increase Scotland’s contacts with Continental Europe. As a rather conspicuous sign of foreign contacts, his son Edgar (1097–1107) sent a camel to Muircertach Ua Briain. Annals of Inisfallen, s.a. 1105.7. 168 David Griffiths, ‘Exchange, Trade and Urbanisation’, in From the Vikings to the Normans, ed. Wendy Davies, Oxford 2003, 73–104. 169 Marie Therese Flanagan, ‘Strategies of Lordship in Pre-Norman and Post-Norman Leinster’, ANS 20, 1997 (1998), 107–26, at 109–10. 170 G. H. Orpen, Ireland under the Normans, 4 vols, Oxford 1911–20, i, 20–37; Beckett, The Anglo-Irish Tradition, 17–18. Cf. Carty, Was Ireland Conquered?, 18–19. 171 Donnchadh Ó Corráin, ‘Nationality and Kingship in pre-Norman Ireland’, Historical Studies 11, 1975, 1–35; Dáibhí Ó Cróinín, Early Medieval Ireland 400–1200, London 1995, 291–2.

England and the Irish-Sea Zone

73

neglected fields of Irish history and Anglo-Irish relations in the eleventh and earlier twelfth centuries.172 The diversity of political links between England and Ireland requires greater recognition. The role of these interactions within a broader network of economic, cultural and religious contacts offers many opportunities for future discussion.

172 Several commentators have drawn attention to the lack of research on this period: Ó Corráin, Ireland,

201–2; Duffy, ‘Irishmen’, 93; Denis Bethell, ‘English Monks and Irish Reform in the Eleventh and Twelfth Centuries’, Historical Studies 8, 1969, 111–35, at 111; Flanagan, Irish Society, 2.

Les abbés bénédictins de la Normandie ducale

LES ABBES BENEDICTINS DE LA NORMANDIE DUCALE Véronique Gazeau Cet article constitue un résumé d’une partie du dossier présenté en vue de l’obtention de l’Habilitation à diriger des recherches en décembre 2002.1 L’enquête sur les abbés bénédictins normands se propose d’être une large contribution aux recherches historiques relatives à la principauté normande pendant la période ducale (911–1204). Les ducs de Normandie, à la suite des souverains carolingiens dans la continuité desquels s’inscrit leur politique religieuse, restaurent le monachisme en relevant plusieurs maisons bénédictines au Xe siècle et au début du XIe siècle, Saint-Ouen de Rouen, Jumièges, Saint-Wandrille, Saint-Taurin d’Evreux, Le MontSaint-Michel peut-être, Fécamp et Cerisy. Le relais est ensuite pris par l’aristocratie qui restaure et fonde entre 1030 et 1080, comme l’avait montré David Douglas, quinze établissements. On y a ajouté l’abbaye de Notre-Dame de Lonlay située à la limite septentrionale du Passais, conquis par Guillaume le Bâtard vers 1050. Au début du XIIe siècle, quelques collégiales de chanoines sont transformées en abbayes bénédictines. La principauté dispose de trente-trois abbayes bénédictines. La recherche consiste en deux volumes. Le premier est une synthèse en trois parties sur le recrutement des abbés, leurs origines et les relations entre les abbés bénédictins et les ducs de Normandie. Le second, une enquête prosopographique, comporte les notices des abbés bénédictins des trente-trois maisons. Dans le second volume, j’ai adopté un classement par abbaye et me suis, préalablement, interrogée sur la validité des listes abbatiales de chaque établissement. En effet, le Gallia christiana, le Neustria pia, deux ouvrages qui furent au départ de l’étude et sans lesquels l’entreprise eût été rendue très compliquée, et avant ces ouvrages, deux rouleaux des morts qui circulèrent en Normandie en 1113 et 1122, ceux de Mathilde et de Vital de Savigny,2 ainsi que le De immutatione ordinis monachorum de Robert de Torigni, proposent des listes qui, à l’examen, se sont souvent avérées fausses. On y trouve des erreurs, des lacunes, des inversions. Ainsi, dans la liste du Gallia christiana de Saint-Pierre-sur-Dives, trouve-t-on deux abbés Alveredus, un nom peu fréquent. Il résulte de l’enquête que le premier n’a jamais existé. Dom Laporte, le moine-historien de Saint-Wandrille, avait considéré qu’au Mont-Saint-Michel, les deux abbés qui portaient le nom d’Hildebert, ne formaient qu’un seul personnage. Avec l’aide de Katharine Keats-Rohan, je prouve définitivement qu’il y a effectivement deux abbés, l’oncle et le neveu. Dans la liste du Gallia christiana, toujours, à la suite d’une lecture de Syméon de Durham, a été placé un abbé Hugues de Coilly, dont je montre qu’il s’agit en réalité d’une confusion avec l’abbé Henri de Sully, qui exerça l’abbatiat à Fécamp. Cette confusion a été reprise par Raymonde

1

Le dossier d’habilitation fut présenté à l’Université de Paris I – Panthéon-Sorbonne le 14 décembre 2002 devant un jury composé de Mathieu Arnoux, David Bates, René Locatelli, Claude Lorren, François Neveux et Michel Parisse. 2 L. Delisle, Rouleaux des morts du XIe au XVe siècle, Paris 1866.

76

Anglo-Norman Studies XXVI

Foreville qui évoque un Hugues de Soldi à la tête des moines de Caen, dans L’Eglise et la royauté sous Henri II Plantagenêt. Le Gallia indique un abbé Hugues à La Trinité-du-Mont de Rouen, intercalé entre le deuxième abbé Rainier et Gautier. En réalité les auteurs du Gallia ont suivi Hugues de Flavigny qui a confondu Gautier avec un Hugues qui n’a jamais existé. La succession des abbés et la constitution des listes abbatiales furent un travail difficile en raison de l’absence de datation des actes de la pratique dans la Normandie ducale. Je dois avouer qu’un grand nombre d’abbés sont restés dans l’ombre et que certains sont demeurés un simple nom. Revenons aux notices abbatiales: au nombre de 327, elles sont organisées en cinq rubriques.3 La première rubrique a trait à la datation des abbatiats, aux circonstances de la fin des abbatiats, à l’inscription des abbés dans la documentation nécrologique, à l’organisation de la memoria abbatiale. La deuxième traite des origines familiales des prélats. La troisième porte sur la carrière laïque ou religieuse des abbés avant leur accession à la charge abbatiale. La quatrième est relative aux circonstances dans lesquelles le religieux devient abbé (élection, bénédiction, installation). La cinquième enregistre les Gesta des abbés. La principauté a compté 327 abbés, ainsi que plusieurs religieux qui, sans porter le titre abbatial, ont exercé, généralement au moment des fondations monastiques, des charges qui s’apparentent à celle d’abbé. Tous les abbatiats ne sont naturellement pas documentés de la même manière et un très grand nombre restent hors de notre connaissance. Je ne peux ici développer les sources qui sous-tendent l’enquête, car je me suis efforcée d’utiliser toute la documentation possible: sources de la pratique, documentation narrative, hagiographique, nécrologique, liturgique, correspondance, les coutumiers monastiques, les données de l’archéologie et de l’architecture. Une mention spéciale doit être donnée à Orderic Vital, sans l’Histoire ecclésiastique duquel l’entreprise aurait été impossible. Les abbés en général et les abbés normands en particulier n’ont pas donné lieu à des études d’ensemble. Pour la Normandie, seules quelques monographies de figures abbatiales prestigieuses permettent de porter un regard sur des individus. Ainsi connaît-on les biographies d’Anselme par Richard Southern, de Lanfranc par Margaret Gibson, de Guillaume de Volpiano par Neithard Bulst, de Jean de Fécamp par Jean Leclercq et de Guillaume Bonne-Ame par David Spear.4 La réalisation des notices a permis d’orienter la recherche vers une thématique d’histoire sociale et politique. En effet, les informations fournies par la prosopographie n’autorisent pas une étude d’histoire religieuse. Elles ne montrent pas vraiment l’abbé aux côtés des moines, se livrant à la réalisation de l’Opus Dei. La première partie du premier volume consiste en l’étude des élections abbatiales, c’est-à-dire des circonstances de l’accession à l’abbatiat. La règle de saint Benoît en son chapitre 64 stipule la libre élection de l’abbé par la communauté. Les ducs et les papes ont accordé des privilèges à quelques maisons normandes, mais quoi qu’il en soit, les moines devaient élire librement leur supérieur. Les sources montrent qu’au

3

Les références se trouvent dans Recherches sur l’histoire de la principauté normande (911–1204) I. Les abbés bénédictins de la principauté normande (911–1204) II. Prosopographie des abbés bénédictins (911–1204). Elles seront indiquées dans la publication de la recherche (Publications du CRAHM/ Brepols). Un exemple de la réalisation des notices des abbés de Saint-Etienne de Caen est déjà publié dans Mélanges P. Bouet, Cahier des Annales de Normandie no 32, 2002, 93–105. 4 R. W. Southern, Saint Anselm. A Portrait in a Landscape, Cambridge 1990. M. Gibson, Lanfranc of Bec, Oxford 1978. N. Bulst, Untersuchungen zu den Klosterreformen Wilhelms von Dijon (962–1031), Bonn 1973. J. Leclercq et J.-P. Bonnes, Un maître de la vie spirituelle au XIe siècle: Jean de Fécamp, Paris 1946. D. Spear, ‘William Bona Anima, Abbot of St Stephen’s of Caen, 1070–79’, HSJ 1, 1989, 51–60.

Les abbés bénédictins de la Normandie ducale

77

début de la période certaines abbayes ont élu leur abbé à l’unanimité.5 Puis, dans le courant du XIIe siècle, les abbés furent élus selon le scrutin dit par compromis; un collège électoral réduit à une douzaine de moines se réunissait pour choisir un abbé. Quel que soit le type de scrutin, des « tractationes » ne manquaient pas de se produire. On les observe nettement à Jumièges et à Saint-Evroult. Néanmoins, les pratiques simoniaques furent pratiquement inconnues en Normandie. Je n’ai relevé que cinq cas: deux abbés du Mont-Saint-Michel, Suppo et Raoul, venus d’Italie et de Fécamp dans la première moitié du XIe siècle, un abbé clunisien à Jumièges au XIIe siècle, l’abbé français de Saint-Denis, Robert, que Robert Courteheuse imposa à Saint-Pierre-sur-Dives, et l’abbé Annon de Jumièges que ses frères requirent à Saint-Mesmin de Micy près d’Orléans au milieu du Xe siècle, après lui avoir acheté la charge abbatiale. En revanche, quelques abbés ont dû pressentir un successeur. Anselme souhaita que Guillaume de Beaumont lui succédât au Bec; à Lyre, trois frères exercent successivement l’abbatiat dans le deuxième moitié du XIIe siècle. On observe des vacances des sièges abbatiaux tout au long de la période, l’abbaye étant alors généralement dirigée par le prieur. Les ducs ont tiré profit de ces périodes de crise pour exploiter la mense abbatiale, notamment Henri II. Une fois élus, vers l’âge de quarante ans en moyenne, les abbés se rendaient auprès du prince normand, porteurs de lettres épiscopales. Jusqu’au début du XIIe siècle, les abbés se firent remettre le bâton pastoral par le duc de Normandie, signe tangible de la concession matérielle de l’abbaye. Après 1107, la pratique se transforme. Dorénavant, le bâton est remis par l’évêque, mais les ducs purent réclamer un hommage aux prélats bénédictins. Toutefois, on ne conserve aucune trace documentée d’un quelconque hommage pour toute la période. On est tenté de considérer qu’à la différence de ce qui se passe en Angleterre à la même époque, et à l’instar de ce qui se passe au Bec où les abbés ont toujours refusé l’hommage au duc-roi, les abbés normands ne s’y sont pas résolus. Les abbés élus devaient également être bénis, mais pas consacrés comme le sont les évêques, par l’ordinaire du diocèse au cours d’une messe. La cérémonie se déroulait ainsi: l’évêque prononçait une prière pour faire venir l’Esprit saint sur l’élu. Une série de cinq questions étaient posées à l’élu, relatives à ses futures responsabilités. L’abbé s’engageait à observer et faire observer la règle de saint Benoît, à conserver le patrimoine de son abbaye, à exercer l’humilité et la patience, à respecter les liens avec l’Eglise et la sujétion à l’égard de l’évêque diocésain. L’évêque bénisseur lui remettait ensuite les insignes de sa charge, le bâton et la règle bénédictine. L’évêque ou un abbé installait ensuite l’abbé béni dans sa nouvelle abbaye. Le nouvel abbé se présentait pieds nus devant ses moines. Après une oraison et le chant du Te Deum laudamus, le prélat chargé de l’installation, suivi de la communauté, lui donnaient un baiser. Après quoi, l’abbé prenait un repas avec ses moines au réfectoire. Certains évêques réclamèrent à la fin du XIe siècle un serment d’obéissance aux nouveaux abbés élus, ce qui donna lieu à des refus à de nombreuses reprises. Entre 1088 et 1136 plusieurs prélats, soutenus par les ducs, refusèrent de se plier à la prestation de ce serment.6 Une véritable crise eut lieu au début des années 1130 entre l’abbé de Saint-Wandrille, Alain, soutenu par le duc-roi Henri Ier, qui s’appuyait sur les coutumes héritées de son père, et l’archevêque Hugues d’Amiens mollement 5

L’analyse de la désignation des abbés repose en grande partie sur des sources qui proviennent du Bec (coutumier, traités) et de Saint-Evroult (Orderic Vital). Mais des informations sont parfois disséminées dans une charte isolée, dans des documents tardifs, dans des chroniques . . . 6 C’est l’évêque Odon de Bayeux qui le premier réclama en 1088 à l’abbé Arnoul de Troarn un serment écrit d’obéissance.

78

Anglo-Norman Studies XXVI

encouragé par le pape Innocent II.7 Il n’est pas certain du tout que les abbés normands ont effectivement prêté un serment d’obéissance à leur évêque. En tout état de cause, comme pour l’hommage qu’auraient pu prêter les abbés au duc, aucune trace de serment n’est décelable dans la documentation. Malgré les prescriptions bénédictines et les privilèges pontificaux, les princes normands ont cherché à placer leurs candidats à la tête des abbayes. Dans l’ensemble, les ducs du Xe siècle et leurs successeurs, Robert le Magnifique, Guillaume le Conquérant et Henri Ier, ont particulièrement veillé aux désignations abbatiales. Dans le sillage de l’étude de Judith Green,8 qui propose de réhabiliter Robert Courteheuse, on peut dire que le successeur de Guillaume le Conquérant en Normandie, n’a vraisemblablement pas cherché à imposer systématiquement ses candidats à la tête des établissements bénédictins. Pendant les quatre années où Guillaume II le Roux tint en mains la Normandie, de 1096 à 1100, deux abbés durent être remplacés à Jumièges et à Saint-Pierre-sur-Dives. Dans les deux cas, il s’est immiscé dans les opérations électorales. Seules, en raison du manque de sources, les pratiques des trois derniers principats d’Henri II, de Richard Cœur de Lion et de Jean sans Terre demeurent mal connues. Le terme d’un abbatiat, en moyenne au bout de vingt ans, est très normalement la mort de l’abbé, mais nombre des supérieurs ont résigné leur charge ou ont été contraints à l’abandon, au prétexte d’une maladie ou d’une incapacité physique, ou bien encore chassés par les moines ou révoqués par la disgrâce ducale. Un seul abbé, Simon de Conches, dut abandonner sa charge après un procès en cour de Rome, procès qui dura plusieurs années, à l’extrême fin du XIIe siècle. Force est de constater que les abbés chassés par les moines, avaient été désignés dans des conditions anormales au regard de la règle. Certains d’entre eux sont même considérés par leurs moines comme des étrangers. Les raisons évoquées par les sources de ces abandons prématurés résident à tort ou à raison dans des pratiques comme la dilapidation des patrimoines monastiques, des accusations de simonie et le manque de respect de la règle, trois loci communi de la littérature monastique. Quelques abbés connaissent des promotions à l’épiscopat en Normandie et en Angleterre; le deuxième abbé de Caen, Guillaume Bonne Ame, devint en 1079 archevêque de Rouen. Serlon d’Orgères, abbé de Saint-Evroult, devint évêque de Sées en juillet 1091. Robert Champart de Jumièges monta sur le trône épiscopal de Londres en 1044. Trois abbés devinrent archevêques de Cantorbéry: Lanfranc de Caen en 1070, Anselme du Bec en 1093 et Thibaud du Bec en 1138. Raoul de Saint-Martin de Sées abandonna son abbatiat vers 1103/1104, à la suite de quoi il passa en Angleterre et devint évêque de Rochester en 1108, plus tard archevêque de Cantorbéry. Deux abbés furent pressentis pour devenir évêques: Gradulphe de Saint-Wandrille mourut avant d’avoir accédé au siège de Rouen, le duc Guillaume ayant suggéré qu’il devînt le vicarius de l’archevêque Mauger. Renaud de Saint-Evroult fut choisi par les chanoines de Sées pour devenir leur évêque, mais le choix de ces derniers fut entravé par Jean sans Terre. Deux autres abbés faillirent monter sur le trône de Cantorbéry: Roger du Bec, puis Martin de Cerisy, en avril 1173, auraient pu succéder à Thomas Becket, les circonstances ne le permirent pas. Un seul abbé devint abbé en Angleterre: Vital de Bernay fut en effet désigné par

7

Le dossier d’Alain est solidement documenté par une série de lettres de l’archevêque Hugues d’Amiens, du duc-roi Henri Ier et du pape. Alain finit par abandonner sa charge d’abbé de Saint-Wandrille, vraisemblablement dans le courant de l’année 1133. 8 J. Green, ‘Robert Curthose Reassessed’, ANS 22, 1999 (2000), 95–116.

Les abbés bénédictins de la Normandie ducale

79

Guillaume le Conquérant à l’abbatiat de Westminster en 1076. Quelques abbés changent d’abbayes. L’enquête se tourne ensuite sur la question de l’inhumation abbatiale. La messe de funérailles est généralement célébrée par un abbé ami du défunt après que le corps a subi un traitement particulier réglé selon les prescriptions des coutumiers. Au début du dernier quart du XIe siècle, les sarcophages furent inhumés dans la salle capitulaire qui devint le lieu de la memoria. Celle-ci est entretenue dans les communautés par des épitaphes, des inscriptions au nécrologe ou sur les rouleaux des morts, et notamment dans le cas des premiers abbés par un service spécial d’anniversaire. Ainsi l’ordinaire de l’abbaye de Saint-Pierre-sur-Dives, dont la rédaction du XIIIe siècle entérinait des rites en vigueur dès le XIe siècle, contenait un office spécial prévu au 13 janvier, jour anniversaire de la mort d’Ainard, le premier abbé.9 Au chapitre, les moines devaient prononcer à haute voix les mots « Ainardus abbas ». Après le cantique du Salve Regina, trois moines entonnaient solennellement l’antienne Dirige. Le sixième répons était chanté par deux autres moines, le neuvième avec les trois versets par trois moines. Après les vêpres du jour, une procession était effectuée auprès du tombeau du fondateur, en portant deux cierges et en chantant le répons Libera me. Après ce répons, on chantait Placebo, c’est-à-dire les vêpres des morts, sur le ton solennel, auprès du tombeau.10 Le célébrant, revêtu d’une aube parée et du manipule, chantait les oraisons Adjuva nos, Deus indulgentarium et Fidelium. Le lendemain, on chantait à la messe matutinale l’office qui commençait par ces mots Requiem. Il se faisait en aubes. Orderic Vital utilise la formule memoriam celebriter exolvunt, pour caractériser la célébration annuelle par les moines de Saint-Evroult de l’anniversaire de leur premier abbé, Thierry de Mathonville, mort à Chypre.11 Le coutumier du Bec comporte un office propre à l’anniversaire du premier abbé Herluin.12 La messe matutinale était chantée à l’autel majeur, en aubes parées. Une centaine de pauvres étaient nourris. En revanche, les autres abbés n’avaient le droit qu’à une messe dite sur un autel mineur.13 La deuxième partie est consacrée aux parcours d’hommes qui parviennent à l’abbatiat en Normandie. Trois abbés seulement pouvaient se prévaloir de liens avec la dynastie régnante du duché: Ansfroi, le premier abbé de Préaux en 1040, Nicolas de Saint-Ouen de Rouen en 1042, le fils de Richard III, et Henri de Fécamp, un neveu du roi Etienne de Blois en 1140. La haute aristocratie normande ou implantée dans la principauté ne fournit que sept abbés: Annon, le deuxième abbé de Jumièges appartient aux Le Riche, un lignage français implanté dans le Pincerais, la région de Bellême et les marges de l’Evrecin. Il est apparenté aux Bellême. Robert de Grandmesnil succède à Thierry de Mathonville à Saint-Evroult en 1059. Guillaume Bonne Ame de Caen est apparenté aux Fleitel, une famille de l’entourage de Robert le Magnifique. Vital et son frère Osberne qui furent abbés de Bernay sont fils de la famille de Creully. Leur père Richard avait été le fondateur du prieuré fécampois de Saint-Gabriel de Creully. Il a quelque raison de penser que le deuxième abbé de Saint-Martin de Sées, Raoul d’Escures, fut un parent des Bellême. La mère de 9 J.-B. Blin, Ordinal de Saint-Pierre-sur-Dives, Bar-le-Duc 1887, 60. 10 Le coutumier de Saint-Ouen de Rouen indique qu’avant les vêpres du jour de la mort de Nicolas, on

récita Placebo et Dirige (E. Martène, De antiquis ecclesiae ritibus, Rouen 1700–1702, III, De obitu abbatis S. Audoeni Rotomagensis, 652). 11 Orderic ii, 74. 12 M.-P. Dickson, ed., Consuetudines Beccenses, Siegeburg 1967, 212. 13 Dickson, Consuetudines Beccenses, 212–13.

80

Anglo-Norman Studies XXVI

Guillaume du Bec, le successeur d’Anselme, était une fille d’une sœur de Roger de Beaumont. Plusieurs abbés venus de l’extérieur de la principauté appartiennent à des familles de l’aristocratie: Mainard de Saint-Wandrille, Guillaume de Volpiano, Aumode du Mont-Saint-Michel, Herluin et Anselme du Bec, Lanfranc, Gerbert de Saint-Wandrille et Geoffroy de Notre-Dame de Grestain viennent d’illustres familles saxonne, italiennes, mancelle, scandinave, allemande et gasconne. Tous ceux que je viens d’énumérer, ont en commun, hormis Henri de Fécamp, d’avoir exercé l’abbatiat au XIe siècle. Par la suite, au XIIe siècle, la très grande majorité des abbés, dont l’origine sociale ou familiale est identifiée, provient de la moyenne voire petite aristocratie des vassaux des fondateurs ou de familles possessionnées dans les environs des abbayes. Des Xe–XIe siècles à la fin du XIIe siècle, le recrutement subit des modifications assez sensibles. On assiste progressivement à un abaissement du niveau social des abbés, qui va de pair avec une métamorphose de la formation préalable à l’abbatiat. Les abbés du XIe siècle ont acquis des connaissances dans les écoles cathédrales parfois, monastiques toujours, qui portent sur les disciplines du Trivium et du Quadrivium. Au XIIe siècle, la formation tend à devenir plus pratique, plus technique. Expérimentés grâce à l’exercice d’un priorat ou d’un office claustral, les abbés se préoccupent davantage de questions économiques, voire administratives et judiciaires, et beaucoup moins de théologie et de liturgie. Prenons l’exemple des abbés de Saint-Etienne de Caen: Lanfranc et ses successeurs, Guillaume Bonne Ame et Gilbert, furent au XIe siècle des théologiens, des maîtres de l’école monastique, des liturgistes, tandis que Robert II et Sanson, les deux derniers abbés du XIIe siècle semblent avoir passé beaucoup plus de temps à siéger à l’Échiquier que dans le cloître. Il en est de même pour les abbés de Saint-Pierre-sur-Dives. Le premier, Ainard, déjà cité, est un abbé liturgiste et musicien, qui composa des offices en l’honneur de saint Kilian, de la Vierge Catherine et du Créateur. Les derniers abbés sont préoccupés prioritairement par des litiges relatifs à des patronages d’églises et de dîmes. De très nombreux abbés ont été choisis à l’extérieur de leur maison, en Normandie ou à l’extérieur du duché. Au Xe siècle, cela s’explique par l’absence d’un vivier de moines normands dans une principauté qui a connu les incursions vikings et les luttes intestines des grands de l’aristocratie. Les structures religieuses ont été mises à mal. Mais ce qu’il convient de souligner, c’est que ce type de recrutement va se poursuivre bien après le renouveau du monachisme. Au XIe siècle, l’Italie et l’Allemagne expédient des abbés prestigieux vers les sanctuaires normands. Quatre abbés sont venus d’Empire: Isembert de La Trinité-du-Mont de Rouen, Ainard de Saint-Pierre-sur-Dives, Mainard et Gerbert de Saint-Wandrille. D’Italie, vinrent Guillaume de Volpiano, le restaurateur de Fécamp, Suppo du Mont-Saint-Michel, Jean de Fécamp, Lanfranc et son neveu, Lanfranc, abbé de Saint-Wandrille, et Anselme du Bec. L’Angleterre fournit au XIIe siècle cinq abbés: Guillaume Huboud de Grestain vint d’Exeter, Gui de Saint-Sever de Chester, Roger de Saint-Ouen du prieuré de Bermondsey, Richard de Saint-Evroult de Leicester, Gautier de Saint-Wandrille d’un endroit indéterminé. L’abbé Sanson de Saint-Ouen de Rouen, apparenté à l’archevêque anglais Gautier de Rouen, à la fin du XIIe siècle, est peut-être venu d’Angleterre. Quatre abbés vinrent d’Ile-de-France, Robert de Saint-Pierre-sur-Dives, Gérard de Saint-Wandrille, né dans la région de Mantes et préalablement abbé de Crépy-en-Valois, Gradulphe et Robert de Saint-Germain-desPrés à Saint-Wandrille. La vallée de la Loire fournit des abbés au XIIe siècle: les abbayes de Saint-Laumer de Blois, de Fleury, de Marmoutier et de Saint-Julien de Tours. Coulombs et Saint-Père au diocèse de Chartres furent aussi pourvoyeuses d’abbés.

Les abbés bénédictins de la Normandie ducale

81

Il n’est pas certain que l’arrivée en masse de huit abbés extérieurs au duché, Foucher du Tréport (1138–1153), Henri de Fécamp (1140–1189), Fraterne de Saint-Ouen de Rouen (1141–1157) Richard (1147–1149) puis Gilbert (1149–vers 1178) de Troarn, Guérin de Saint-Pierre-sur-Dives (vers 1149–?), Pierre de Jumièges (1154–1169), puis Roger de Saint-Ouen de Rouen (1157–1167), en provenance de maisons clunisiennes, a modifié en profondeur la vie religieuse des six maisons dont ils eurent la charge.14 Cluny était déjà implantée dans le duché sous la forme de prieurés.15 La simultanéité de ces arrivées est à mettre en relation avec la période d’anarchie qui régna en Normandie après la mort du duc-roi Henri Ier en 1135. L’archevêque de Rouen, Hugues d’Amiens, ancien moine clunisien,16 en accord avec le roi d’Angleterre, Etienne de Blois favorable à Cluny, encouragea la venue de ses confrères, à une époque où le monachisme cistercien prenait pied dans le duché. C’est précisément en 1137 que fut fondé le premier monastère cistercien normand, Mortemer, affilié à Clairvaux17 et en 1147 que les maisons de Savigny rentrèrent dans le giron de Cîteaux.18 S’il était besoin d’aller chercher des abbés dans d’authentiques maisons clunisiennes, c’était sans doute parce que les abbayes qui avaient adapté les coutumes dijonnaises ou même clunisiennes ne méritaient pas leur brevet de loyalisme clunisien.19 Parmi les six abbayes qui eurent recours à un abbé clunisien, Fécamp, Jumièges, Troarn, suivaient des coutumes proches de celles du monastère clunisien. Pour les autres, on ignore si l’arrivée d’un supérieur clunisien se traduisit par l’obligation de suivre des coutumes nouvelles. En 1132, Pierre le Vénérable avait modifié les statuts de Cluny. Les sources ne renseignent pas sur leur éventuelle introduction dans les maisons normandes. L’opiniâtreté conduisit l’action de l’archevêque Hugues puisque l’introduction d’abbés clunisiens se répéta dans deux maisons. A Saint-Ouen de Rouen, Fraterne choisit son successeur, Roger. A Troarn, Gilbert remplaça Richard. Trois actes d’Hugues d’Amiens furent en effet donnés à Saint-Wandrille, au Bec et à Mortemer entre le 8 décembre 1140 et 1141, per manum Fraterni cancellarii.20 Plusieurs arguments conduisent à accorder du crédit à l’hypothèse selon laquelle le chancelier de l’archevêque, Fraterne, devint l’abbé de Saint-Ouen. Ce nom reste assez peu porté en Normandie; la disparition dans l’entourage de l’archevêque du chancelier nommé Fraterne en 1141, au moment où un personnage portant ce nom parvient à la tête de Saint-Ouen ne doit pas relever du seul hasard. L’origine clunisienne de Fraterne, comme celle de l’archevêque, conforte l’hypothèse. Nos recherches nous ont conduite à voir aussi l’œuvre d’Etienne de Blois, présent en Normandie à l’automne 1137 au moment précis où s’y trouve également l’abbé Pierre le Vénérable. Or, c’est en 1138 qu’arrive au Tréport le premier abbé clunisien. Cependant, la fièvre clunisienne retomba après la mort

14 Richard de Méré, abbé clunisien du Mont-Saint-Michel, est un Normand, nommé par Henri Ier en

1123 ou 1124, c’est-à-dire bien avant la vague massive d’abbés clunisiens des années 1138 et suivantes. L’accession de Richard de Méré au Mont-Saint-Michel précède de peu la fondation clunisienne de Reading par Henri Ier. 15 Sainte-Foy de Longueville[-sur-Scie] au diocèse de Rouen fut fondée en 1093 par Gautier Giffard II. Saint-Côme-du-Mont était un prieuré situé dans le Cotentin depuis le début du XIIe siècle. 16 Après avoir été prieur de Saint-Martial de Limoges en 1115, Hugues d’Amiens part pour Lewes en Angleterre, puis devient le premier abbé de la fondation du roi Henri Ier en 1125 à Reading. 17 Gallia christiana XI, 663 D. 18 Gallia christiana XI, 545. 19 Fécamp, Troarn, Conches, Sées, Saint-Evroult, Jumièges. 20 T. Waldman, Hugh ‘of Amiens’, Archbishop of Rouen (1130–1164), Oxford 1970, no 149 (acte en faveur de Saint-Wandrille daté du 8 décembre 1140), no 9 b (acte en faveur du Bec daté du 12 juin 1141) et no 84 (acte en faveur de Mortemer daté de 1141).

82

Anglo-Norman Studies XXVI

d’Hugues d’Amiens. Ce dernier mourut en 1164 et après la mort en 1167 et 1169 de Roger de Saint-Ouen de Rouen et de Pierre de Jumièges, il n’y aura jamais plus d’abbé clunisien dans les monastères bénédictins en Normandie. En tout, durant toute la période, 48 des 327 abbés, soit 14,75% sont venus de l’extérieur du duché. Le choix des abbés s’est aussi largement porté sur des moines de maisons normandes, mais qui n’étaient pas issus de la communauté. Jumièges a fourni 7 abbés à 7 abbayes: 1 à Saint-Pierre-sur-Dives,21 1 à Montebourg (le deuxième), 1 au Mont-Saint-Michel,22 1 à Saint-Sauveur-le-Vicomte (le premier), 1 à Saint-Sever (le premier), 1 à Saint-Evroult (le premier), 1 à Saint-Wandrille.23 Fécamp a fourni 24 abbés à 9 abbayes. Parmi ces 24 il faut enlever les 7 de Bernay, les 5 de Saint-Taurin d’Evreux. En effet, Bernay et Saint-Taurin d’Evreux étaient des abbayes de plein exercice, mais qui depuis toujours étaient dans l’obligation d’élire un abbé fécampois. Il en reste 12 qui se répartissent ainsi: 2 (dont le premier) à Conches, 1 à Ivry, 3 à Jumièges,24 3 au Mont-Saint-Michel,25 1 à SaintOuen,26 1 à Troarn27 et 1 à Saint-Wandrille.28 Saint-Ouen de Rouen a fourni 16 abbés à 6 abbayes. Parmi ces 16 il faut soustraire les 4 de La Croix-Saint-Leufroy et les 8 de Saint-Victor-en-Caux. La Croix-SaintLeufroy et Saint-Victor-en-Caux avaient l’obligation de choisir un abbé dans la communauté de Saint-Ouen. Il en reste 4, soit 1 pour Cerisy (le premier), 1 pour Saint-Pierre-sur-Dives,29 1 pour Jumièges30 et 1 pour La Trinité-du-Mont (le premier). Saint-Etienne de Caen a donné 15 abbés à 8 abbayes: 1 à Saint-Pierre-sur-Dives,31 1 à Fécamp,32 3 à Fontenay,33 4 à Lessay,34 2 à Lonlay,35 1 au Mont-Saint-Michel,36 2 à Saint-Ouen de Rouen,37 1 à La Trinité-du-Mont de Rouen.38 Le Bec a fourni 28 abbés à 16 abbayes: 2 à Bernay,39 2 à Caen (dont le premier),40 1 à Conches,41 4 à Cormeilles,42 1 à Grestain (G. Huboud déjà compté comme moine étranger au duché), 4 à Ivry,43 1 à Jumièges,44 3 à Lessay (dont le premier),45 1 à Etard remplaça Benoît à Saint-Pierre-sur-Dives à la fin du XIe siècle. Roger II du Mont était le prieur de Jumièges, quand il fut désigné en 1106. Il s’agit du rector Enfulbert de la fin du Xe siècle. Guillaume de Volpiano, Thierry et Tancard. Thierry, Suppo et Raoul, entre 1023 et 1051–1053. Nicolas (1042–1092). Durand Ier, le premier qui porta le titre abbatial à Troarn, avait été oblat à Saint-Wandrille, puis moine de Fécamp. 28 Gerbert (1063–1089). 29 Benoît remplaça Foulque à Saint-Pierre-sur-Dives en 1092. 30 Robert II Champart avait été moine de Jumièges, puis prieur de Saint-Ouen. Il abandonna la charge pour partir vers l’Angleterre où il devint évêque de Londres, puis archevêque de Cantorbéry. 31 Renouf (2e moitié du XIIe siècle). 32 Guillaume de Rots (1079–1107). 33 Herbert, Robert et Geoffroy II. 34 Robert Ier, Raoul, Roger II et Thomas, tous au XIIe siècle. 35 Renouf et Lanfry. 36 Roger Ier. 37 Helgot et Samson. 38 Dreu. 39 Goscelin et Guillaume (entre 1169 et 1177). 40 Lanfranc et Guillaume Bonne Ame. 41 Gilbert II (v. 1130 – avant 1140). 42 Guillaume, Benoît, Hardouin et Durand. 43 Durand, Guillaume, Norman, Raoul. 44 Roger Ier (1169 – 15 ou16 août 1176). 45 Roger, Geoffroy et Garin. 21 22 23 24 25 26 27

Les abbés bénédictins de la Normandie ducale

83

Lyre,46 2 au Mont-Saint-Michel,47 1 à Saint-Ouen de Rouen,48 2 à Préaux,49 1 à Saint-Evroult,50 1 à La Trinité-du-Mont,51 1 au Tréport52 et 1 à Saint-Wandrille.53 Ce sont incontestablement Le Bec et Saint-Etienne de Caen qui sont les deux pépinières d’abbés. Il est vrai que Fécamp et Saint-Ouen étaient obligées de fournir les abbés de Bernay et Saint-Taurin d’Evreux, d’une part et de La Croix-SaintLeufroy et de Saint-Victor-en-Caux d’autre part, ce qui laisse sans doute peu de moines disponibles pour aller vers d’autres abbayes. Il est intéressant de remarquer que Le Bec a pourvu en abbés des abbayes qui n’auraient pas dû se tourner vers elle. C’est le cas de Bernay qui y trouve 2 abbés et d’Ivry censée choisir ses supérieurs à Coulombs. C’est pendant toute la période que Le Bec envoya des abbés dans les autres établissements normands. Sous le principat d’Henri II, il y eut comme un regain de départs depuis l’abbaye de la vallée de la Risle. Le roi affectionnait les moines du Bec, mais l’on sait l’intérêt qu’il porta au Mont-Saint-Michel et la confiance qu’il plaça en son abbé, Robert de Torigni. Cependant, les moines du Mont ne devinrent pas pour autant des abbés. Certes Robert de Torigni avait été moine à l’abbaye du Bec de 1128 à 1154. La troisième partie a trait aux relations qu’ont entretenues les abbés et les ducs normands. Elles peuvent être appréciées par le biais des souscriptions abbatiales, mais aussi grâce aux récits des narrateurs qui mettent en scène des rencontres entre les supérieurs d’abbayes et les princes, et à la production diplomatique; les princes ont su s’entourer des conseils de certains abbés, même si la mission de conseil appartient en principe davantage aux évêques. L’étude a consisté à évaluer le rôle de ces abbés, proches de la curia et appelés dans des circonstances précises, des cérémonies de fondation ou des règlements de conflits. Il convient de s’arrêter sur la méthode contestée et contestable dite des souscriptions. Ainsi l’abbé Lanfranc de Caen n’a-t-il jamais souscrit d’acte du duc Guillaume, quand ce dernier ne cesse d’avoir recours à ses conseils en matière religieuse. A en croire Orderic, il aurait été pressenti pour monter sur le trône métropolitain de Rouen. En revanche, la comparaison des souscriptions abbatiales et le rang des abbés dans les listes de souscriptions, sous les principats de Robert le Magnifique et de Guillaume duc puis roi, montre que Jean de Fécamp fut le conseiller le plus écouté jusque dans les années 1063–1066, date à laquelle Nicolas de Saint-Ouen de Rouen le détrône auprès du duc. A partir de 1078, Anselme est l’abbé le plus actif auprès du duc-roi. Au XIe siècle, les abbés ont participé à la restauration du monachisme par la création de nouveaux établissements aux côtés des ducs. Ils ont été sollicités pour leurs conseils en matière de recrutement, voire appelés à des fonctions en Angleterre. Dans la deuxième moitié du XIIe siècle, plusieurs prélats ont rempli des charges qui les apparentent à des officiers ducaux: fermier de vicomtés comme les abbés de Fécamp, réformateur de la fiscalité anglaise comme Robert II de Caen, pilier de l’Echiquier de Caen comme les derniers abbés de Caen, garant de traités de paix comme Raoul, le 46 47 48 49

Raoul (1130–1142) qui apporta les coutumes du Bec. Bernard (5 février 1131 – 8 mai 1149) et Robert de Torigni (1154–1186). Haimeric, cellérier du Bec (avant le 26 février 1167 – 11 février 1171). Michel (21 décembre 1152 – 15 ou 16 décembre 1167) et Henri (entre fin décembre 1167 et avant le 24 avril 1168 – 26 février 1182). 50 Robert II. 51 Guillaume (après le 10 septembre 1174 – après 1179, le 26 novembre ou le 27 décembre avant le 16 octobre 1196). 52 Guillaume II. 53 Lanfranc, le neveu de Lanfranc de Caen (1089–1091).

Les abbayes bénédictines de Normandie aux Xe–XIIe siècles

Les abbés bénédictins de la Normandie ducale

85

dernier abbé de Fécamp. A cet égard, durant toute la période, en raison du prestige attaché à la fondation ducale, les abbés de Fécamp ont joui d’une grande réputation qui les plaçait aux premiers rangs des honneurs. Ils furent les seuls à obtenir du pape les insignes pontificaux. Leur choix ne fut jamais dû au hasard. Au milieu du XIIe siècle, Henri de Sully était le neveu du roi Etienne d’Angleterre. En réalité, les ducs regardèrent de plus en plus souvent vers l’ouest du duché. Et d’autres abbés d’établissements moins prestigieux ont pu faire émerger des prélats dont l’action semble avoir été efficace auprès des ducs. Ainsi, Mainier de Saint-Evroult ou Foulque de Saint-Pierre-sur-Dives apportèrent au XIe siècle leur concours au duc Guillaume lors de plaids importants. Les abbés Anselme du Bec sous Guillaume le Conquérant, Boson du Bec sous Henri Ier, Robert de Torigni sous Henri II ont eu des liens d’amitié avec les ducs. J’évoquerai pour terminer quelques unes des conclusions majeures auxquelles cette étude est parvenue ainsi que les pistes de recherches qui en sont le prolongement. En premier lieu, j’ai réévalué le rôle de Guillaume de Volpiano dans la restauration du monachisme normand au commencement du duché et jusqu’aux années 1030. Ainsi Martin et Annon, qui furent les deux premiers abbés de Jumièges, ne sont certes pas restés longtemps à la tête de leur maison, mais ils avaient joué un rôle important dans la restauration du monachisme poitevin et avaient gravité dans l’entourage de l’évêque Frotier de Poitiers dans les années 930. On ne peut exclure que la réforme de Jumièges ait été menée par Guillaume Tête d’Etoupe et par l’évêque de Poitiers. Ensuite, Gérard de Saint-Wandrille qui prend ses fonctions en 1008, apporte coutumes et liturgie de Chartres, lesquelles demeurent en vigueur à Saint-Wandrille, qui restaure Préaux et participe à la fondation de Grestain. Il est le contemporain de Guillaume de Volpiano et seul le manque de sources empêche d’en savoir davantage sur Gérard. Orderic Vital est quasiment muet sur Guillaume de Volpiano dont l’inscription dans les nécrologes n’est pas fréquente. En deuxième lieu, le principat d’Henri Ier apparaît comme un moment de rupture. A cette époque, apparaissent les transformations dans le type de recrutement, dans la formation des abbés, dans les relations avec l’épiscopat. Les familles modestes de la petite aristocratie, des novi homines, fournissent la majorité des abbés. Ceux-ci acquièrent certes, toujours une formation dans les écoles monastiques, mais elle est complétée par un passage dans un prieuré ou dans un office claustral. Les tâches des abbés semblent parallèlement acquérir une dimension supplémentaire. Certes, toujours voués à l’administration du temporel et à l’édification de leurs moines, les supérieurs bénédictins sont de plus en plus appelés à l’extérieur des cloîtres pour trancher dans des affaires judiciaires, très souvent comme juges arbitres pour le compte de la papauté. En troisième lieu, au total, l’abbatiat normand se présente non pas comme un corps constitué, mais comme un ensemble d’hommes d’une haute qualité intellectuelle et technicienne, d’une compétence reconnue par les princes qui ont recherché leur proximité et qui les ont parfois associés à l’exercice du pouvoir. Mais l’étude a manqué de points de comparaison avec les abbés bénédictins d’autres principautés ou même anglais, simplement parce qu’ils n’existent pas. Elle s’inscrit néanmoins dans les recherches historiographiques actuelles autour de la question des élites et de leur place dans la société, des lieux de pouvoir et de l’application de la réforme de l’Eglise. En quatrième lieu, il convient de s’interroger sur les réseaux monastiques en Normandie. Je n’envisagerai qu’une hypothèse de travail. On constate qu’en règle générale, le recrutement des abbés de nombreuses abbayes ne semble obéir à aucune

86

Anglo-Norman Studies XXVI

logique: ainsi par exemple, à Lyre, où les premiers abbés sont venus de SaintEvroult, à partir de 1142, on choisit un abbé au Bec qui introduisit les coutumes de son monastère d’origine. A Saint-Pierre-sur-Dives, les treize abbés vinrent de huit abbayes différentes. Enfin, on peut reconsidérer la question de la crise du cénobitisme dans la Normandie du XIIe siècle. Les abbés bénédictins n’ont rien perdu de leur influence auprès du duc, de l’archevêque et des grandes familles, même si ces dernières n’ont pas cherché à placer leurs fils. Les cisterciens et les chanoines réguliers semblent constamment en arrière-plan.

Vita Ædwardi Regis

THE VITA ÆDWARDI REGIS: THE HAGIOGRAPHER AS INSIDER* J. L. Grassi In the preface to his second edition of the Vita, Frank Barlow has commented that ‘it has received little critical attention’ in recent years and that ‘its lack of a certain author and indisputable date makes it slightly disreputable, something to be acknowledged only with half-averted eyes’.1 It is not quite certain that Barlow himself has looked the work straight in the eye. While emphasising the significance of the Life for our knowledge of the reign of Edward the Confessor,2 he has entered reservations that seem somewhat to undermine his confidence in the work3 whose ‘evidence has to be used with caution’.4 Above all, he does not think that the author had access to inside information.5 The burden of this paper is that the author did have such access to inside information and that, as a source for the reign of Edward the Confessor, the Vita is to be taken at least as seriously as John of Worcester, William of Malmesbury or the versions of the Anglo-Saxon Chronicle. To go further: in those matters on which the author wrote, there should be an a priori presumption that the witness of the author of the Vita be preferred to that of any other source with which it has a clear difference. The Vita contains some forty items of information for which it is either the unique or the original source. A large part of this information has been readily and generally accepted – the physical description of Edward;6 the importance of Robert of Jumièges and his actions to the crisis of 1051;7 Tostig’s accusation of Harold’s complicity in the Northern rebellion of 10658 – but acceptance has been somewhat arbitrary and there has been a general rejection of information provided by the Vita * I would like to thank Dr Ann Williams and Professors Frank Barlow, David Dumville and John Gillingham, who have read drafts of this paper for their help and comments. 1 Vita Ædwardi, v. One writer who has made considerable use of the Vita is Pauline Stafford, Queen Emma & Queen Edith, Oxford 1997. She gives a full summary of its contents, pp. 40–8, but is sparing in her own assessment of the work. She recognises that Edith commissioned it (p. viii) and, although Edith’s voice can be heard in it (p. 41), Edith is not dominant in the work as a whole (p. 28 n. 1). See also her further comments, 264–9, 271–4 and elsewhere. 2 E.g., Vita Ædwardi, lx–lxi. 3 Compare his acceptance that the author had conversations with Queen Edith (Vita Ædwardi, xliv); that he was ‘well acquainted with those children of Godwin who were in England in the later years of Edward’s reign’ and that of these ‘he knew Edith and Tostig best’ (p. xlc); and that ‘we can believe that he knew Queen Edith personally and that he had at least seen the king’ (p. xli) with his remark that the author, ‘probably correctly [my italics], claimed to be acquainted with the queen’ (F. Barlow, Edward the Confessor, 2nd edn, New Haven 1997, 83), and his hesitation in ‘if this work is based on information provided by the queen . . .’ (Vita Ædwardi, xli). 4 F. Barlow, The Godwins, London 2002, 8. 5 Barlow, Edward, 83. 6 Vita Ædwardi, 6, 18–20. 7 Ibid., 30–6. 8 Ibid., 76–82.

88

Anglo-Norman Studies XXVI

where it contradicts information in other sources, particularly that in the Anglo-Saxon Chronicle; for instance, the question of which nunnery Edith was sent to in 1051, or in certain of its unique items of information, such as whether Tostig was accompanied by his mother when he went into exile in Flanders at the end of 1065. It should be carefully considered whether such a rejection is justified. To state the dilemma in its starkest terms: when the Vita is at odds with the Anglo-Saxon Chronicle, why should the Chronicle be preferred? As we have seen, Barlow thought that the Vita’s ‘lack of a certain author and indisputable date makes it slightly disreputable’. Do any of the versions of the Chronicle have a more certain author or a less disputable date? Who did compose the Worcester, Abingdon or Peterborough versions of the Chronicle and when were their passages on the reign of Edward composed? What reasons are there to regard their sources as more authorative than those of the Vita? The authorship of the Vita may not be certain and its accuracy may be doubted but its authenticity is not open to question, nor is the date of what is conventionally styled ‘Book I’, thanks largely to Barlow’s work. That part of the work was written in the last year or so of Edward’s reign and was brought to an abrupt and incomplete end by the events in the last months of the king’s life in 1065 although there is evidence of some later, minor, revisions.9 While the identity of the author is not known, he was not only contemporary with the events which he recorded but, like the equally unknown writer of the 1087 annal in the ‘E’ version of the Chronicle, could claim to be writing of some things which he had himself witnessed. He was acquainted with Edith and those of her brothers who were in England from about 1064 and knew, at least by sight, Edward himself. His sources were those who had taken part in the events he describes or had information from those who had taken part or, indeed, his own observation. The work was composed at the behest of Queen Edith and written with her active participation. Time and again this is emphasised: the work is to concern ‘my lady’s fame (ad laudes . . . domine)’; it is she who has ‘restored [him] as from death’ and made him ‘put back the pens [he] had thrown away’.10 The author goes on to emphasise that it was Edith, not her family, whom the work was intended to praise; ‘To whose praise . . . to her (cuius laudi . . . huic)’. His Muse tells him ‘It is she whom you will acclaim as long as you live’,11 and, again, in the poem which opens the second part of the work, his Muse tells him that he will write of the queen who first helped him and urges that ‘whatever he writes should be to her praise’.12 She is ‘the illustrious mistress whom we chiefly serve in this account’.13 It is not impossible that the passage beginning ‘You shall be the first to sing King Edward’s song’ originally read ‘You shall be the first to sing Queen Edith’s song’ with appropriate text leading to ‘You will write that Godwine, her father . . .’ and that the passage was altered when the work was skewed into an encomium on Edward, with a repetition of the reference to Edward’s song at the end of the Muse’s contribution to the opening poem of the second part of the Vita.14 At the beginning of the

9

The most obvious of these is the reference, at the beginning of the account of the events of 1051, to Edward as ‘of sacred memory’, ibid., 28. 10 Ibid., 4. 11 Ibid., 4–6. 12 Ibid., 88–90. 13 Ibid., 66. 14 Ibid., 90.

Vita Ædwardi Regis

89

second part of the Vita, his Muse reminds the writer that ‘it was to honour her that you chose to write the earlier part’ and that he should continue since Edith ‘might wish him to (cumque velit repetes)’.15 The author, through his Muse, states unequivocally that Edith often spoke to him about Edward.16 Taking all this together, it would not be pressing his words too hard to take them as meaning that not only did Edith commission the work, but that she took an active part in its composition; that she was a major source, perhaps the main source, of his information and pace Barlow, that the author did have access to inside information. On the other hand, it might be asked what were the sources of those writers whose information contradicts or conflicts with the information in the Vita. The need for choice between the evidence of the Vita and that of other sources is clearly seen in one detail of the events of 1051, and the discrepancy is worth considering at length. The author of the Vita states that, on the disgrace of the family of Godwine, Edith was sent to the convent at Wilton, while both ASC ‘D’ and ‘E’, followed by John of Worcester17 and William of Malmesbury,18 state that she was sent to Wherwell. There seems to be a common source for this section of the two versions of the Chronicle,19 while John and William either relied directly on that source or on one of the versions of the Chronicle. There is, therefore, only a single source which stated that Edith was sent to Wherwell against another single source, the Vita, which says that she went to Wilton. Why should the former be preferred to the latter? The weight of scholarly tradition has favoured the Anglo-Saxon Chronicle. Plummer dismissed Wilton as ‘probably a mere slip’ by the author of the Vita.20 Freeman21 and Stenton22 sent her to Wherwell, while Pauline Stafford, seeking to reconcile the conflict, sent her first to Wherwell and then to Wilton.23 Barlow in his biography of Edward has her go to Wilton on the grounds that ‘the author of the Vita should have known the truth’,24 but was embarrassed by the conflict of evidence; ‘a possible factual error concerning Wilton is suspicious . . . It is . . . an awkward disagreement’25 and in a recent work joins Stafford in running with the hare and hunting with the hounds; Edith was sent to ‘either Wherwell . . . or Wilton . . . or perhaps to one before the other’.26 Yet one can almost hear Edith giving the anonymous author of the Vita the version of events as he narrated them with its emphasis on the malignity of Robert of Jumièges. If Edith was, as she must have been, the source for this account, she was also the source for the information that she was sent to Wilton, which is named three

15 Ibid. Barlow has translated this as ‘and since she wills, proceed’, but the mood is subjunctive, not

active; therefore, ‘she may wish’. Barlow’s translation implies that the continuation was at her command, which it may have been, but that is not what the text says; the continuation is optimistic, not prescriptive. 16 Ibid. 17 John of Worcester ii, 562. 18 Gesta Regum ii, 360 (199, 7). 19 Two of the Saxon Chronicles Parallel, ed. C. Plummer, 2 vols, Oxford 1892–9, ii, pp. lxxvi and cxvi. G. P. Cubbin, The Anglo-Saxon Chronicle: A Collaborative Edition, vol. 6: MS D, Cambridge 1996, liii–iv, has followed Ker and Keller in thinking that ‘D’ from perhaps 1031 to 1052 was a conflation compiled in the 1050s and for that period was based on ‘C’. 20 Plummer, Two of the Saxon Chronicles ii, 238. 21 E. A. Freeman, The History of the Norman Conquest of England, 6 vols, 3rd edn, Oxford 1876, ii, 156. 22 F. M. Stenton, Anglo-Saxon England, 3rd edn, Oxford 1971, 565. 23 Stafford, Queen Emma, 264–5. 24 Barlow, Edward, 115–16. 25 Vita Ædwardi, l. 26 Barlow, The Godwins, 43.

90

Anglo-Norman Studies XXVI

times (pace Plummer’s ‘mere slip’ and Freeman’s ‘clerical error’).27 If Edith did not know where she had been sent, then no one did. It could be argued that in the triple insistence on Wilton the lady protested too much and it might also be asked why it should have mattered so much to Edith that it be believed she had been sent to Wilton and never to Wherwell, but what would such an untruth have profited her? Of much greater hurt to Edith’s pride than whether she had been sent to one convent or another would have been the total rejection which the divorce Robert proposed would have signalled. Yet it is only the Anonymous who clearly mentions such a possibility, although the ‘E’ version of the Chronicle’s use of the verb forlaeten, i.e. ‘to forsake, relinquish’ and John of Worcester’s repudavit might be construed as implying the possibility of divorce.28 If the author (and Edith) wished to suppress a truth, that more humiliating one would have been the truth to suppress. The conclusion must be that the single source behind the Anglo-Saxon Chronicle and the works based on it are in error; that Edith was sent to Wilton but never to Wherwell. If this argument is accepted, it has implications for the authority of the Vita beyond its accuracy on this detail. While on the question of Edith’s place of seclusion there is a clear and apparently irreconcilable difference between the Vita and the other sources, there is another, and greater, matter on which the Vita differs from all the other sources. This is the matter of the cause or causes of the crisis of 1051. On that question the differing sources may not be so much contradictory as complementary. The crisis of 1051–2 was the central, pivotal, event of Edward’s reign. AngloSaxon Chronicle ‘D’ and ‘E’, followed again by John of Worcester and William of Malmesbury, have very similar accounts. These accounts differ entirely from that given in the Vita.29 All four begin with the visit of Eustace of Boulogne and its consequences: his visit to Edward and the violent quarrel with the citizens of Dover, either when coming (‘D’ and John) or when going (‘E’ and William). They all continue with Godwine’s refusal to obey Edward’s order to harry his province, the meeting of the witan at Gloucester, the challenge to Edward by Godwine and his sons, the second meeting of the witan at London and the consequent exile of Godwine and his family. Of all this – except that there were two meetings of the witan – there is not a single word in the Vita. There is no mention of Eustace, the men of Dover or of the incursion of foreigners into the sphere of influence of the family of Godwine: there is only the malign influence of Robert of Jumièges and Edward’s desire for revenge for the death of his brother so many years before. It was these factors which were the business, the Vita says, of the two meetings of the witan. That there was any such long-standing background to the clash between Edward and the family of Godwine is something which is completely absent from the accounts in the Anglo-Saxon Chronicle and John of Worcester but is spelt out in detail in the Vita and, later, by William of Malmesbury who was able to use the Vita’s narrative but who also added a number of embellishments in this account. The emphases placed by historians on these two versions of the causes of the 1051 crisis are noteworthy. Stenton ignored the Vita completely:30 others have mentioned the disputed election

27 Vita Ædwardi, 36 and 44. 28 Plummer, Two of the Saxon Chronicles i, 176; John of Worcester ii, 562. 29 The account in ASC ‘C’ is a curious one. It dismisses the whole episode in four or five lines in contrast

to its very full account of the return of Godwine and his sons the next year. 30 Stenton, Anglo-Saxon England, 565, although he used the Vita Ædwardi for details of the manoeuvres

of the Godwinist forces in 1052.

Vita Ædwardi Regis

91

to Canterbury and the influence of Robert of Jumièges,31 but none except Stafford, give serious consideration to the Vita’s account of the causes of the crisis.32 All versions of the Chronicle and John of Worcester give accounts of the numerous ecclesiastical changes in 1050–1, although none directly link these with the causes of the crisis, which the Vita alone does.33 It reports that Robert of Jumièges, who had come to England with Edward on his return, ‘was, they say, always the most powerful confidential adviser of the king’.34 It is, incidentally, a testimonial to the fair-mindedness of the Anonymous that he acknowledged that by Robert’s counsel ‘a great many good things as well as bad were done in the realm’ despite the consequent casting of Robert as the villain of the piece. After he became bishop of London (1044), Robert ‘involved himself (immersit se) more than was necessary in directing the course of the royal council and acts’.35 This led Edward ‘to begin to neglect more useful advice’, and many nobles (including, presumably, Godwine) were offended. There were disagreements over the filling of ecclesiastical vacancies, with some wanting them filled by their (?native) candidates and others wanting them filled by foreigners (alienis).36 Now Canterbury fell vacant and ‘all the sons of all the church’ joined the monks of Canterbury in electing ‘by general consent and by petition according to the rule and canon law’ one of Canterbury’s own. This was Æthelric, a kinsman of Godwine, whose support for his candidacy was sought, not only because he was Æthelric’s kinsman but because Godwine was ‘by royal favour ruling in that part of the kingdom’. Godwine made his plea to the king but was rebuffed because ‘the good (pius) king leant his ear more to the rival party in those days’ and ‘while all the clergy protested with all their might against this wrong’, Edward promoted Robert to Canterbury. It is worth pausing to consider the implications of this version of the events leading up to Robert’s accession to Canterbury. The account in the Vita is precise and very emphatic; although Æthelric was, it appears, no more than a monk of Canterbury who held no office there and was a somewhat light-weight candidate, he was, the Anonymous says, positively elected both by the monks of Canterbury and by the church as a whole in a way which satisfied both the monastic rule and canon law, and there was general outrage at his rejection.37 No other source so much as mentions that there was dispute over the succession to Canterbury, and any such dispute does not seem to have troubled Leo IX who gave Robert his pallium; yet the account in the Vita is so specific that there can be no question of slips of the pen or mistaken information, and it is difficult to see how there could even be a question of exaggeration. It follows that this account must be 31 Freeman, Norman Conquest ii, 119–23, 130; F. Barlow, The Feudal Kingdom of England,1042–1216,

3rd edn, London 1972, 62–7; Barlow, Edward, 104–17; Ann Williams, Kingship and Government in pre-Conquest England, London 1999, 139–41. 32 Stafford, Queen Emma, 262–4, combines the two versions but with no comment on the exclusivity of information in each of them. In an earlier work, Unification and Conquest, London 1989, 90–1, she makes no reference to the Vita’s account. Barlow, The Godwins, 40, has opined that ‘the basic cause was Edward’s desire to get the Godwins off his back’ with the Canterbury election as the immediate cause. 33 Vita Ædwardi, 30. 34 As Barlow has pointed out, ibid., 29 n. 61, William of Malmesbury seems to have used this passage from the Vita. 35 Ibid., 28. Barlow has translated immersit se as ‘intruded himself’, which is too loaded. 36 Ibid., 30. 37 For Æthelric, see Ian W. Walker, Harold, the Last Anglo-Saxon King, Stroud 1977, 203–4, and Barlow, The Godwins, 40. Both suggest that this Æthelric was the man of this name who became bishop of Selsey in 1058. It is worth noting that even after Robert’s expulsion, it was not Æthelric who became archbishop.

92

Anglo-Norman Studies XXVI

accepted as truthful or rejected as an invention: if it be rejected then all other unsubstantiated information in the Vita must be rejected. If accepted, it gives more support for the rest of its account of the causes of the rift between Edward and Earl Godwine. The promotion of Robert to Canterbury would certainly have caused Godwine a considerable amount of irritation, not only with Robert but also with Edward, and the Vita provides further reasons for antagonism between these competitors for the king’s ear. Now that he was archbishop, Robert ‘began to provoke and oppose [Godwine] with all his might’, and the Anonymous offers examples. One specific example involved lands of Christ Church which, it was claimed, Godwine had invaded, in which matter the Anonymous recognised that right was on Robert’ side and for this there is other evidence.38 There are claims, unspecified, of other provocations by Robert against Godwine to which the ‘active’ and ‘peaceable’ earl refused to rise. These are followed by what the author of the Vita asserted to have been the major and, indeed, the only cause of the rift between king and earl and one which was provoked by Robert. The archbishop accused Godwine of advising Harold Harefoot to kill Edward’s brother, Alfred the atheling, and his men in 1036 and that Godwine ‘was now planning in the same way . . . Edward’s ruin’.39 The passage concludes with the curious statements that ‘with continual persuasion [Robert] got the king to give more credence to this than was right’ and that the king ‘was moved by these accusations more than was just’. Again, one must accept or reject the Vita’s evidence that Robert did make the accusation against Godwine and, again, if that is rejected it must follow that the Vita is to be regarded as an untrustworthy source. If, on the other hand, it is accepted that the charge was made, the Vita must be accepted as trustworthy. It is very improbable that such a charge would have been in any way a revelation to Edward – there is other evidence that Godwine had faced such a charge in the time of Harthacnut40 – and Robert had a listener who probably already believed the accusation: he would have no need to ‘turn the king’s mind against Godwine’ on that matter. It is probable that all Robert did was provide Edward with a long-desired opportunity or excuse to re-open the matter. The Vita is quite unambiguous in stating that the charge of responsibility for Alfred’s death and the alleged plotting against Edward himself was the reason for the summoning of the royal council at Gloucester and it gives no other reasons. ‘As the king was moved by these accusations more than was just, all the nobles and earls from the whole of Britain assembled in the royal palace at Gloucester.’41 It is manifest that this charge against his patroness’s parent caused great distress to the Anonymous; he twice describes Godwine as ‘guiltless’ (insontem ducem and insons);42 he refers to Godwine’s lack of guilt in the poem which ends this part of his work and refers to his innocence on other occasions.43 Despite these reiterations of innocence, it is striking that the author does not straightforwardly declare the charges to be false or groundless; instead he says that Edward gave the charges ‘more credence than was right’ and that he was ‘moved by these accusations more than was just’. In all honesty, the Anonymous could not completely dismiss the accusations 38 39 40 41 42 43

Vita Ædwardi, 32 with the corroboratory evidence on this dispute quoted in n. 68 there. Ibid., 32–4. John of Worcester ii, 530 (s.a. 1040). Vita Ædwardi, 34. Ibid. Ibid., 34, 36, 38.

Vita Ædwardi Regis

93

and his distress is as patent as it is later when he was faced with Tostig’s accusations against Harold. Most striking of all is his account of Edward’s answer to Godwine’s attempts to win the king’s peace: He could hope for the king’s peace when and only when he gave him back his brother alive together with all his men and all their possessions intact which had been taken from them quick and dead. On hearing this, Godwine pushed away the table at which he sat and rode off to Bosham and exile.44 There is an immediacy in the venom of Edward’s answer and in the detail of the pushing away of the table which argues strongly for this to be an account that the author had been given by someone who had witnessed the incident; it rings with truth. The Vita’s account shows what Edward thought: for fifteen years he had carried the thought that Godwine was responsible for his brother’s mutilation and death and, at last, given the opportunity to destroy Godwine, the bitterness and anger were brought out into the open and Edward could become his own man. One is reminded of similar long-awaited opportunities for revenge by later kings: the ten years Edward II waited to wreak his vengeance on his cousin, Thomas of Lancaster, at the battle of Boroughbridge and Thomas’ subsequent execution, and the twelve years which Richard II spent in picking off, one by one, the Lords Appellant, although in the end it cost him his throne. If this be a true reading of what happened to Alfred and Edward’s response to that tragedy, then it must determine our reading of the characters of Edward and Godwine, the nature of the relationship between them and our interpretation of Edward’s policies, of the events of his reign and of the motives behind those policies and events. What then of the version of events given by Anglo-Saxon Chronicle ‘D’ and ‘E’, John of Worcester and William of Malmesbury? That there was an incident involving Eustace of Boulogne and citizens of Kent cannot be doubted; nor that Eustace complained to Edward about that fracas, nor that Edward ordered Godwine to harry in Kent. It was all very public, involved all the leading persons in the realm and, despite the Vita’s silence on all this, was, at the least, the ostensible reason for the calling of the meeting at Gloucester. The difference between the two versions is due to emphasis and motivation. The events surrounding Eustace were the occasion for the eruption of the crisis but not its cause. If for the world at large that was the cause, those most intimately involved knew that there were deeper issues and emotions involved. Using Robert’s accusations, Edward took the opportunity to humiliate and destroy Godwine and Godwine knew that that was what he was doing. Unlike the matter of Edith’s exile to Wherwell or Wilton, both these versions of the outbreak of the crisis of 1051 are accurate; each of them is true, but neither of them had the whole truth. There are many other places in the Vita in which the intimate connections of the author with the family of Earl Godwine are evident. Perhaps the most significant and certainly the most obscure of them are the puzzling references to the ‘four streams’ which flowed from Godwine and the identity of the member of the family styled the voratrix. Four times – twice in the initial poem,45 again in the second poem,46 and yet

44 Ibid., 36. 45 Ibid., 6. 46 Ibid., 26.

94

Anglo-Norman Studies XXVI

again in the poem at the start of the second book47 – the writer speaks of four children of Godwine whom he also allegorises as four streams. In one place he seems to divide them into two groups – pars hec and pars illa48 – the first of which brings honour to the family and the other, the voratrix, born of a lifeless mother (de non animata matre), destruction for the family. The Latin is not easy to construe and the author’s meaning is, perhaps deliberately, extremely difficult to uncover but the point was obviously of some considerable significance to the purpose of the book. Barlow has translated the key passage as The one part mounts to the skies, to heaven twined, And tends its race’s hope in tree-top nest. The other, gulping monster, seeks the depths, Attacks its roots and mouths the parent trunk, And holds, until, as doomed, the breath of life Creates a creature from a lifeless dam; And losing grip, pursues again its prey.49 Barlow twice tried to unravel the meaning of the ‘four streams’,50 but his attempts were frustrated by the apparent division into two groups. It seems clear from the poem which begins the second part of the Vita that the author had in mind, at least when that part of the work was written, Godwine’s four senior surviving sons at the end of 1065, Harold, Tostig, Leofwine and Gyrth. These ‘four streams’ were separate from and additional to Edith, that gem at the kingdom’s centre and lover of all virtues;51 they form the praiseworthy pars hec. The pars illa is the voratrix. I offer a translation slightly different from Barlow’s: One part (pars hec) climbs to the sky, harnessed to the heavens and fosters the hope of its kind from [its] tree-top nest. The other (illa), she the enemy who devours, swimming (tranans) seeks the deep, causing damage to its very roots (sue stirpis) is poised, holding in [her] jaws the parent trunk until in due course the living breath of life creates a living being from a lifeless mother: from that time forth (inde) (?since when) [she] occupies [herself] (studet) with her plunder (rapinis) unrestrained (resolute). Although the author is, to say the least, opaque in this section of the poem, he must have expected his meaning to be understood – by Edith, even if by nobody else. He was being obscure but he cannot have meant to be impenetrable, and it is worth trying, therefore, to uncover that meaning. The first, and most important, thing to be noted is that the voratrix – the ‘gulping monster’ or ‘devourer’ – was female. The word voratix is not grammatically feminine in the way that poeta or agricola are; there is a masculine form, vorator,52 and it must be deduced, therefore, that it was a woman to whom the author was referring. It is, however, difficult in the extreme to make any sense of the second part of the sentence. As the text stands, it reads as if there are three separate people – the voratrix, the ‘living being’ and the ‘lifeless

47 48 49 50

Ibid., 84. Ibid., 26. Ibid. 27. Ibid., 6–8 n. 10 and 26–7 n. 57. He has made a third attempt, The Godwins, 86–7, tentatively identifying Tostig as the good pars hec and Harold as the evil, pars illa, the devourer. 51 Vita Ædwardi, 26. 52 R. E. Latham, Revised Medieval Word-List, Oxford 1980, 517 gives a ninth-century reference for vorator and ‘c.1070’, presumably from the Vita, for voratrix.

Vita Ædwardi Regis

95

mother’ – which seems excessive, although that may be exactly the meaning which the writer wished to convey. What is more probable is that the writer became entangled, logically and grammatically, in the convolutions of his poetry and his allegory and was speaking not of three, but of two, people. Godwine had three known daughters; Edith, of whom nothing more needs to be said; Gunnhild, who seems to have led a blameless, unmarried, life, eventually going into exile with her mother, Gytha, and Ælfgifu. Of the last, there is only one certain reference and that an oblique one. Domesday Book records that before 1066 a sokemen who held land at Waldridge (Bucks.) was the man of Alueve soror Heraldi comitis.53 That she is described as ‘Earl Harold’s sister’ rather than as ‘Earl Godwine’s daughter’ might imply that she was illegitimate54 and it might be further surmised that her mother died giving her birth and was thus the ‘lifeless dam’. Then there is the mysterious Ælfgifu of the Bayeux Tapestry who was obviously and inevitably in Barlow’s mind when he was puzzling over the meaning of this part of the poem; he concluded his note on it with the remark, ‘But this passage like pl[ate] xv of the Bayeux Tapestry, must remain a historical mystery.’55 Considering the context in which Ælfgifu appears, it is possible that she was Harold’s sister; she and the episode in which she appears were clearly germane to the relationship between Harold and William and to the events related in the Tapestry. That the episode involved some sexual scandal seems certain, as does the implication that it in some way affected the negotiations between Harold and William. If it involved a sister of Harold it would seem to have been to his disadvantage and to the disadvantage and discredit of his family – Edith’s family – and would justify branding her as voratrix.56 At this remove of time it is impossible to uncover the whole story but that does not reduce the importance or the significance of the voratrix to the understanding of Edward’s reign. There is a very large hole in the fabric. Other members of Edith’s family also appear in the Vita; her father, Godwine, features very prominently, as do two of her brothers, Harold and Tostig. Leofwine is referred to in the events of 1051 and Gyrth rather more prominently, as being given control of a shire (comitatum ei dedit) in East Anglia at some date,57 and there was a third reference to him in a portion of the work now lost. Of Swein there is no mention at all despite the prominence given to him in the accounts of 1051 in other sources; and, unless she also featured in the lost portion of the work, there is remarkably little reference to Gytha. She is described as Cnut’s sister – but not named – on the notice of Godwine’s marriage to her58 and is stated by the Anonymous to have accompanied Tostig on his exile to Flanders at the end of 1065.59 This last statement, though of no great significance in itself, unless indicative of Gytha’s possible support for Tostig against Harold, is worthwhile considering as bearing once again on the trustworthy nature of the unique information in the Vita and on the tendency to downgrade such information in comparison with that of other

53 54 55 56

Domesday Book i, fol. 144c. I owe this suggestion to Ann Williams. Vita Ædwardi, 26 n. 57. For discussion of the identity of the Tapestry’s Ælfgifu, see among others, F. M. Stenton, The Bayeux Tapestry, London 1957, 10; Freeman, Norman Conquest iii, 710; and J. Bard McNulty, ‘The Lady Aelgyva in the Bayeux Tapestry’, Speculum 55, 1980, 659–68. 57 Vita Ædwardi, 50. As Barlow has pointed out, p. 50 n. 122, this is not referred to as an earldom. 58 Ibid., 10. 59 Ibid., 82.

96

Anglo-Norman Studies XXVI

sources for Edward and his reign. In a footnote Barlow has rejected it simply on the basis that ‘no other writer mentions Gytha’s departure’ and that Gytha was in Exeter in 1068 and he has proposed an emendation of and addition to the text to fit.60 That Gytha was in Exeter in 1068 does not mean that she could not have gone to Flanders with Tostig in 1065, returning to England at some later time, perhaps when he went to Norway. That no other source mentions her going to Flanders is also no reason to doubt the accuracy of the statement; there are many pieces of information mentioned only by a single source which are, nevertheless, accepted as accurate. The Anonymous at that point offers another detail which has been doubted. He says that Tostig and Judith took with their infant children (lactentibus liberis); although neither Freeman61 nor Barlow62 rejected this statement they both cast doubt on it on the ground that Tostig and Judith had been married for fourteen years with the implication that they were unlikely to have unweaned children at that time. Perhaps Barlow was right to think that ‘there seems to be some exaggeration here’, but it is not unknown for marriages to be long barren but eventually productive.63 One of the oddest features in the Vita is that, although Godwine is the dominant figure in the earlier pages of the work, when the writer comes to his death he is most reticent; the earl ‘of happy memory’ died, the people sighed and ‘showed great grief’ with ‘many tears’ and that is that.64 The author then turns to Godwine’s eldest surviving sons, Harold and Tostig with an interesting comparison of the two brothers and their characters, the accuracy of which was borne out by subsequent events. Harold was easy-going, mild-tempered, affable and open, quick of understanding; Tostig appeared altogether less genial, perhaps a somewhat dour character who kept his own counsel: whereas Harold sought popularity (felicitatem), Tostig wanted only success;65 but there is no doubt which of them the Anonymous preferred. There is good reason to think that the writer was very well acquainted with Tostig: he gives a detailed and vivid account of Tostig’s visit to Rome which is so full that it seems highly probable that he was himself a member of the party and that this is his personal account.66 There then follows a lengthy poem full of classical, and sometimes obscure allusions, the burden of which is a lament that discord should have arisen between these ‘two Hercules, the kingdom’s sacred oaks’, the tone of which suggests very strongly that it was written after the conflict arose between them late in 1065 but before its final and fatal conclusion.67 The occasion and, indeed, the cause of the discord between Harold and Tostig was the Northumbrian rising against the latter.68 As in the account of the crisis of 1051–2, the Vita gives a version different from other sources of the causes and course of the rising. Anglo-Saxon Chronicle ‘C’ gives

60 61 62 63

Ibid., 82 and nn. 199 and ‘a’. Norman Conquest ii, 507n. Vita Ædwardi, 82 n. 200. Louis XIV’s parents were in their late thirties and had been married for more than twenty years before he was born. 64 Ibid., 76. 65 Ibid., 46–50. 66 Ibid., 52–6. 67 Ibid., 58–60. 68 The causes and course of the rebellion are dealt with in detail by Freeman, Norman Conquest, 487–507 and note WW, 711–16; Barlow, Edward, 234–44, and The Godwins, 83–8; William E. Kapelle, The Norman Conquest of the North, London 1979, 86–100; Richard Fletcher, Bloodfeud, London 2002, 153–62.

Vita Ædwardi Regis

97

as the causes that Tostig ‘had promoted injustice because he robbed God first and then despoiled of life and land all those over whom he could tyrannize’ which was, presumably, a reference to the death of the Northumbrian thegn, Gospatric, allegedly by order of Edith at Edward’s court on 28 December 1064 and of Gamel, son of Orm, and Ulf, son of Dolfin, allegedly by order of Tostig, seemingly in 1063, in his own chamber at York while under safe-conduct (sub pace federe) as detailed by John of Worcester.69 It can be questioned whether those deaths were actually a cause of the rebellion in October 1065 or were only an excuse. Gospatric died apparently in the previous December and John’s ‘preceding year’ for the deaths of Gamel and Ulf seems to mean 1063: the anger at these three deaths seem to have taken a remarkably long time to come to the boil, although they did undoubtedly play a part in the resentment against Tostig. More potent, perhaps, was the tax which he levied which exceeded the normal beneficial rate which Northumbria enjoyed both before and after that levy. It was presumably these factors which the author of Anglo-Saxon Chronicle ‘C’ had in mind when he wrote that Tostig ‘promoted injustice’ and ‘despoiled of life and land all those over whom he could tyrannize’, but for his other charge that Tostig ‘robbed God first’ there would appear to be no other evidence: on the contrary, Tostig would appear to have been a benefactor of churches, especially Durham.70 There is little difference between the sequence of events as related in the Vita and that given by other sources but a great difference in their accounts of causation. The impression given by the Anglo-Saxon Chronicle and John of Worcester is that there was a general uprising of his earldom against Tostig and the injustice of his rule; but the Vita has a very different perspective on it: one man’s injustice can be another man’s justice. For the Anonymous, this was the work of a faction of nobles which Tostig had pressed, or repressed, by the rightfully heavy yoke of his rule because of their misdeeds. He speaks of the group’s ‘savage rashness’, ‘mad conspiracy’ and ‘the wickedness of a few nobles’. They sought to give legitimacy to their cause by making Morcar of Mercia their ‘leader and lord’ and invited his brother, Edwin, to join them, counting on ‘the hatreds from a long-standing feud between them and Tostig’.71 There were no limits set to the killing and people were slain for private grudges.72 The result was that the region which ‘had so long rested in the quietness of peace through [Tostig’s] strength and justice . . . was driven to self-destruction’. The Anonymous set out to justify Tostig’s policy: Siward had been greatly feared because of the severity of his justice but had been unable to curb the lawlessness of a cruel and godless people yet Tostig, ‘the son and lover of divine peace’ had brought them to heel and cleared the country of robbers by mutilating or killing them and by sparing no one, however noble, who was guilty of crime. It is worth considering whether Gospatric, Gamel and Ulf were among those who were, in the writer’s mind, to be numbered among the criminal nobles.73 With Tostig driven out and disciplinary 69 John of Worcester ii, 598–9. The phrase sub pace federe is translated as ‘under cover of a peace-treaty’

which is perhaps not quite the same thing. On the significance of Gospatric’s death and its possible connection with the extended account of Tostig’s visit to Rome, see Stafford, Queen Emma, 44–5. 70 Kapelle, Conquest, 89–90; Barlow, The Godwins, 27, 63, 65 n. 45, and 94–5. 71 Vita Ædwardi, 74–80. 72 The text, ibid., 76, reads ‘Nullus ergo modo fit in occasione’, i.e. ‘There were no bounds/limits to the opportunity.’ Barlow, ibid., 76 note b, suggests emending occasione to occisione which would suit the sense of the passage much better, and I prefer it. 73 Ann Williams has pointed out to me that the great crucifix which Tostig gave to Durham was cast down and despoiled in the rising of 1069–70 by Earl Gospatric who was almost certainly a kinsman of the

98

Anglo-Norman Studies XXVI

strictness gone, there was a return to old wicked ways. The half-century which followed – or, perhaps, continued with – the murder of Earl Uhtred in 1016 more than justifies the Vita’s remarks on the nature of Northumbrian society: the events of 1065 were almost certainly part of that story.74 The Vita has a detail of some significance:75 Edward offered redress for any injury which the rebels could prove Tostig to have done to them. It does not say that any such proof was presented. If the Anonymous knew of such proof, he suppressed that information, but if he did suppress it, why did he raise the matter in the first place? The implication must be that no such proof was forthcoming. Edward called a meeting of the whole witan and the Vita has two passages which give weight to the essential honesty and trustworthiness of the Anonymous and his account. First, he corroborates, at least in part, the charges which, according to the ‘C’ version of the Chronicle, were brought against Tostig by stating that at that meeting not a few (nonnulli) charged Tostig ‘with being too cruel; and he was accused of punishing disturbers more for desire of their property which could be confiscated than for love of justic’. What was disputed, it seems, was not what Tostig had done but his motives for doing it. It must be wondered who the nonnulli were; presumably they included Morcar and his brother. Then the Anonymous added his most sensational detail: although he found it an almost incredible charge (si dignum esset credere), it was claimed that the rebellion against Tostig was undertaken ‘at the artful persuasion’ of Tostig’s own brother, Harold. The Anonymous was horrified (quod absit) and unable to restrain himself from a personal intervention in the narrative; ‘I dare not, nor might I be willing, to give credence to such a detestable wickedness by such a prince to his own brother’, but he had to accept it because it was Tostig himself who made the accusation publicly before the king and his courtiers. The phrase the Anonymous uses, frequentibus palatinis, i.e. ‘courtiers’, may be significant, implying the king’s inner circle rather than members of the witan. Harold cleared himself of this charge by an oath but ‘he was, alas (proh dolor), far too free and easy with his oaths’. That this information should be given in this source, of all sources, compels acceptance of the truth of it. It is significant that whereas all versions of the Chronicle and John of Worcester emphasise that it was Harold who led the negotiations on the king’s behalf, he is conspicuously absent from the account of those negotiations in the Vita. It is clear that Tostig felt that Harold had sold him down the river, having perhaps made a deal with Edwin and Morcar involving Harold’s marriage to their sister and the succession to the throne, and that his resentment against Harold was total. The Vita then goes on to relate, and is the only source to do so, how Edward wished to resist the rebels’ demands by force and, paralleling his strategy in 1051, issued an edict, presumably to summon the fyrd. He was frustrated by various factors, but above all by widespread refusal to use force. If Harold and the Mercian brothers were, as they must have been, among those who resisted his will, it is no wonder that he was unable to enforce it and had to accept the deposition and exile of his favourite. Thwarted, as in 1052, and with a ‘raging spirit’, Edward ‘fell ill and from that day to the day of his death he bore a sickness of the mind’. John of Worcester says that after all this he ‘began gradually to weaken’, but it is the Vita Gospatric killed, allegedly, on Tostig’s behalf by Edith. See Symeon of Durham, Libellus de exordio atque procursu istius, hoc est Dunhelmensis, ecclesie, ed. David Rollason, OMT 2000, 176, 186–92. 74 Fletcher, Bloodfeud, has given the most detailed account of the whole story which fully justifies the sub-title he gave his book, ‘Murder and Revenge in Anglo-Saxon England’. 75 Vita Ædwardi, 78.

Vita Ædwardi Regis

99

which makes the connection between Edward’s choleric nature, his frustration and his fatal illness. Edward bitterly resented his inability to impose his will and called upon God to right that wrong. He was not mentally incapacitated and was able to be profoundly distressed at his helplessness (merens nimium quod in hanc impotentiam deciderit) and to heap many gifts on Tostig before he departed. For her part, Edith was confounded by this quarrel between her brothers but, bereft of support because of Edward’s helplessness (impotentia), could do nothing.76 The information, explicit and implicit, given by the Vita on the causes and course of this final tragedy of Edward’s reign presents an interpretation radically different from the alternative version encapsulated in the versions of the Anglo-Saxon Chronicle and the chronicle of John of Worcester, as it did in its interpretation of the causes and course of events in 1051–2. As then, there are powerful reasons for preferring it. When was this account written? The reference to Edward’s illness continuing until his death points to at least a minor revision or insertion after that event, but the narrative concludes with an account so unbalanced in its fullness of detail of Tostig’s reception in Flanders and of his settling in St Omer ‘a few days before Christmas Day’ that it gives a very strong sense of immediacy – of being written within days of the reception of that news: the last few lines on Edward’s death soon after (cum mox) ‘languishing from the mental illness which he contracted’ give a strong sense of having been tacked on. The death of Edward is dealt with in full detail in the second part of the Vita, which is concerned only with him and the miracles and visions which testified to his sanctity. The description of the scene at his deathbed and the striking similarity of the pictorial depiction of it in the Bayeux Tapestry are well known and have been much commented on. The only point about it that needs to be made here is that the Anonymous unequivocally claims that he had the details from those who were present (sicut testantur hi qui aderant presentes).77 Despite this, Barlow has stated that there is ‘a lack of authority’ in this account;78 but the verbatim report by the Anonymous of the things said there will almost certainly have come from the lips of Edith herself, including the commendation of the care of Edith (?and the kingdom) to Harold.79 Particularly striking is the report of Stigand’s sceptical aside to Harold on Edward’s wandering mind which has immediacy and the smell of truth. This must be coupled with the extraordinarily long and uncharacteristic musing on the significance of the vision of the green tree with its obscure references to the shortcomings of the English aristocracy and ecclesiastical hierarchy.80 Were these the thoughts of the writer or those of his patroness, and were Harold and Stigand those mainly being thought of? That Edward, whether in full possession of his faculties or not, did have this vision and did say the things attributed to him must be accepted or the Anonymous’s informant or informants lied and all the details given to the writer on Edward’s last moments must be rejected, although it is difficult to see what purpose such misinformation could have served. Perhaps the most extraordinary feature in the Vita is its claims about what might be described as Edward’s title to the throne. At the beginning of the second section of

76 Vita Ædwardi, 80. Should the word impotentia be translated as ‘helplessness’ in both instances? Edith

might very well have been distressed at Edward’s physical ‘helplessness’ after a stroke, but would Edward not have been more distressed at his ‘powerlessness’ to impose his will? 77 Ibid., 116. 78 Barlow, Edward, 83. 79 Vita Ædwardi, 122. 80 Ibid., 116–18.

100

Anglo-Norman Studies XXVI

the work, after the introductory poem, the Anonymous claims that Edward ‘was chosen by God before the day of his birth, and consequently was consecrated to the kingdom less by man than, as we have said before, by Heaven’.81 This refers back to two assertions very early in the work;82 that, while Emma was pregnant with Edward, ‘all the men of the country took an oath’ to accept the child, if male, as their future ruler. To this was added a vision which came sometime in the reign of Cnut or of his sons to Brihtwold, bishop of Wiltshire (Ramsbury), in the watches of the night at Glastonbury, of which he was sometime abbot. In the vision he saw St Peter consecrate as king a man to whom he assigned a celibate life (celibem ei vitam designare)83 and a fixed number of years of rule: when the man asked about his successors he was told that God had already made his choice.84 All this raises a number of questions. They are extraordinary, not to say astounding, statements which, if true, could only have been included to justify Edward’s right to the throne; but by the time the book came to be written Edward had been on the throne for some twenty-four years. Who at that time was to challenge his right? Magnus of Norway died in 1047, Edward the atheling died in 1057, and his son, Edgar, even if a claim could be made for him, was a minor. The reason for the inclusion of these statements seems to have been not in the present but in the past. A second question, if they are true, concerns the Anonymous’s sources for the statements. Brihtwold was very long-lived and, not dying until 1045, lived to see Edward on the throne and, perhaps, even to see him married. He himself was probably the source for the tale of his vision and there would have been many still alive in the mid-1060s who would have heard him tell of it.85 Bishop by 1005, if not from 995, Brihtwold might have been one of those who took the oath to accept Emma’s unborn child as king, if such an oath was taken. There were doubtless others still living in the 1040’s who could have been among those who took such an oath and, of course, many more in the next generation who could have been told of such an oath by those who had taken it. It should, however, be pointed out that according to Emma’s apologist in the Encomium, Edward maintained that he was unable to help his mother in her exile in 1037 on the grounds that ‘the English nobles had sworn no oath to him’, which could possibly be a reference to this alleged oath.86 There was one source which was certainly available to the Anonymous for both statements. That was Edward himself, directly or indirectly. He had been exiled in his childhood and had seen the ancestral throne usurped by Cnut; he had had to suffer the indignity of his mother’s preference for his half-brother, the son of the usurper, and there was always the knowledge that there was a senior line of descent from his father. He needed to assure himself of the soundness of his title. Although the rise of

81 Vita Ædwardi., 90–2. 82 Ibid., 12–14. 83 In his first edition of the Vita, 9, Barlow translated this phrase as ‘mark out for him a life of chastity’,

but in the second edition, 15, as ‘assign him the life of a bachelor’. Chastity or celibacy and bachelorhood are not the same thing, but see below, note 85. 84 William of Malmesbury gave an account of the vision, adding that it occurred in the reign of Cnut, specifically referring it to Edward and fixing his reign at twenty-four years. Osbert of Clare also described the vision, but in much the same unspecific way as in the Vita. Barlow prints the three accounts in parallel in Vita Ædwardi, App. A, 128–9. 85 If Brihtwold died before Edward was committed to marriage with Edith, then bachelorhood, not celibacy, might very well have been what he had in mind. 86 Encomium Emmae Reginae, ed. Alastair Campbell, with a supplementary introduction by Simon Keynes, Cambridge 1998, 48–9.

Vita Ædwardi Regis

101

Godwine under Cnut is recorded with great satisfaction, the tone of the comments on Cnut and his line is far from approving. The insistence in the earlier part of the Vita on the signs of divine approval of Edward as foreordained king does not appear to be the result of a later revision.87 The Vita then goes on to give a version of events concerning the circumstances of Edward’s accession to the throne which, once more, seems to contradict that in other sources. The major versions of the Anglo-Saxon Chronicle and John of Worcester have accounts which all seem to be based on a common source, though perhaps not entirely on that common source. They all agree that Edward arrived in England in 1041 and that Harthacnut having died on 8 June 1042, Edward was crowned king at Winchester by Archbishop Eadsige almost a year later on Easter Day 1043. The Vita never mentions Harthacnut and, while agreeing with John on the role of Godwine in Edward’s accession, states that Edward had to be brought from abroad and that he was raised to the throne and crowned at Christ Church, Canterbury.88 Attempts have been made to reconcile these two versions, proposing a second coronation, which may have been what happened.89 A possible solution to the problem is that neither of the versions gives a complete account of the sequence of events; that Edward did return to England in 1041 but was temporarily on the continent when Harthacnut suddenly and unexpectedly died and so had to be recalled. There is considerable reason to think that the accession of Edward as king of England was far from straightforward. The ‘E’ version of the Chronicle, and presumably the source upon which it, and John of Worcester and other recensions of the Chronicle were based, states that before Harthacnut was buried ‘the whole nation chose Edward as king in London; may he reign as long as God grants him’,90 which seems to indicate that there was some need for great speed. Anglo-Saxon Chronicle ‘D’ and ‘C’ also state that ‘the whole nation received Edward as king’ with ‘C’ adding ‘as was his right by birth’. ‘C’ and ‘D’ also thought it necessary to comment on Edward’s return to England that he had been ‘driven from the country many years before but was nevertheless sworn in as king’. John of Worcester found it necessary to lay out Edward’s ancestry back to Alfred.91 With all this the Vita is in accord. Having told of the oath on Edward’s birth and the vision of Brihtwold, it immediately turns to Edward’s accession; at Godwine’s instigation Edward ascended the throne ‘that was his right by birth’; he was acclaimed ‘everywhere loyally and the whole kingdom was settled under native rule’.92 This insistence by all sources on Edward’s right to the throne leads one to suspect that there were some who doubted that right, and it would explain the haste with which he was chosen: a hurried sacring at Canterbury and then a later coronation ‘with great ceremony’ by the two archbishops ‘and almost all the other English bishops’ at Winchester.93 None of this explains why, a quarter of a century later, it should be felt necessary to tell of oaths and visions nearly a half of a century before; but clearly it mattered to someone in the last days of Edward’s reign, probably to Edward himself.

87 88 89 90 91 92 93

Vita Ædwardi, 10, 14. Ibid., 14. Ibid., 14–15, nn. 31 and 32. The last phrase clearly indicates that this entry, or its original, was written not long after the event. John of Worcester ii, 534. Vita Ædwardi, 14. ASC ‘D’ and ‘E’ mid mycclum wurðscipe; John of Worcester ii, 534.

102

Anglo-Norman Studies XXVI

There remains one over-arching question about the Vita, which is why it was written at all. Although it has hagiographical features, it does not fit into the hagiographical tradition, and that was not its original purpose. Barlow discusses its purpose in his introduction to the Vita94 and more fully in Appendix A of his biography of Edward.95 With some hesitation and reservations he thought that although its original and apparent intention was to praise the family of Godwine, Edith had an ulterior motive in commissioning the work. That was to prepare for the time when the childless Edward died; that it could be used either as a justification for herself to take the throne and then be followed by one of her brothers or for the family to be king-makers, possibly pushing the atheling Edgar as the anonymous successor of whom Brihtwold had dreamt; ‘but it is possible that in 1065 Edith and her brothers had not decided whom to make king’. He did not think that Edith could have been ‘so simple, so politically naïve’ as to have commissioned the work ‘simply [for the] glorification of herself and her relatives’. He concluded that, ‘owing to the change in the author’s plan, it is impossible to be sure what exactly was the original political purpose of the Vita’. Stafford suggested that the Vita was intended as an exculpation of Edith and was, in some way, meant to claim a role for Edith in the post-Conquest settlement.96 It is inevitable that historians seek to find a hidden agenda, a deeper motive, in such a work as the Vita, but it is just possible that Edith was, indeed, ‘so simple, so politically naïve’ as to have no other motive in commissioning the work than ‘the glorification of herself and her relatives’: that it was, in large part, an act of filial devotion to her father, Earl Godwine, is obvious. Edith may also have been motivated to imitate her predecessor and mother-in-law, Emma, and her Encomium. None of the suggestions of possible political reasons proffered convince and perhaps none existed. If the argument be accepted that the information in the Vita is generally trustworthy and is based on evidence provided directly by some of the major figures of Edward’s inner circle, then it must follow that that information should be given rather more weight in the interpretation of the events and the personalities of Edward’s reign than it has hitherto been given; that, on those matters of which it speaks, it should be given pride of place over the other of our sources for Edward’s reign with the consequent changes which that involves.

94 Vita Ædwardi, xxii–xxiv. 95 Barlow, Edward, 291–300. 96 Stafford, Queen Emma, 45.

The Warenne View of the Past

THE WARENNE VIEW OF THE PAST 1066–1203 Elisabeth van Houts This paper is concerned with an heiress, Isabelle of Warenne (c.1130–c.1203), a tombstone, a chronicle and various clusters of charters, which between them throw light on the Warenne view of the past.1 The heiress Isabelle of Warenne was born in the 1130s as the great-granddaughter of William I of Warenne (d.1088), the family’s Norman ancestor and his Flemish wife Gundrada.2 Isabella succeeded her father William III of Warenne after his death in the Holy Land in 1147 or 1148. Around this time she was married by King Stephen (1135–54) to his second son William of Blois, a lad of about 13 years old, who became the fourth earl of Warenne, but died in 1159. In 1164 Isabelle remarried. Her new husband was Hamelin of Anjou, the illegitimate half brother of King Henry II (1154–89). Hamelin, fifth earl of Warenne died in 1202 and, well into her seventies, Isabelle followed him probably a year later in 1203. During Isabelle’s long life three sources were produced that shed light on the Warenne family’s perceptions of the past. The oldest of them is the tombstone of Isabelle’s great-grandmother Gundrada (d.1085), which was refashioned in the mid-twelfth century.3 Of the same period dates the Warenne Chronicle, an incomplete historical narrative known as the so-called ‘Hyde’ Chronicle.4 Covering the period from 1035 until 1120, the text of the unique manuscript breaks off abruptly.5 A careful analysis as part of the preparation of a new edition has established a date, a potential author and a raison-d’etre for its composition. The third source consists of two clusters of charters by or for Hamelin, alone or in conjunction with his wife, which throw not only interesting light on their commemorative traditions, but also reveal a renewed interest in their combined possessions in Touraine and Flanders. But before I place each of these sources in their Warenne context, I will briefly remind you of the fortunes of the Warenne family.

1

I am most grateful to the Arts and Humanities Research Board for the award of a Research Leave Grant in 2001 and to the Master and Fellows of Emmanuel College for allowing me to accept it. During the preparation of this paper I have received valuable advice from David Bates, Charles Coulson, David Crouch, Judith Everard, Judith Green, Edmund King, Chris Lewis, Daniel Power, Kathleen Thompson, Steven Vanderputten and Tessa Webber, for which I am much indebted to them. I should like to thank John Gillingham in particular for his editorial advice. 2 EYC viii, 13–14 (Isabella), 14–18 (William IV), 18–24 (Hamelin). For a genealogical chart of the Warenne family, see Appendix 1. 3 F. Anderson, ‘ “Uxor mea”, the First Wife of the First William of Warenne’, Sussex Archaeological Collections 130, 1992, 107–29 and, for a different interpretation, E. van Houts, ‘The Epitaph of Gundrada of Warenne’, in Nova de Veteribus. Festschrift für Paul Gerhard Smidt, ed. E. Stein and A. Bihrer (forthcoming). 4 E. Edwards, ed., Chronica monasterii de Hida juxta Wintoniam, in Liber monasterii de Hyda, RS 1866, 283–321. My new edition and translation is forthcoming in the OMT series. All references to the text will be by paragraph to my new edition and by page numbers to the Edwards edition. 5 London, BL, MS Cotton Domit. A xiv, fols 3–21v.

104

Anglo-Norman Studies XXVI

The Warenne Family The first Warenne was William, a younger son of Rodulf of Warenne, a landholder of some local importance in northern Normandy, who could boast kinship with the dukes.6 William accompanied the Conqueror in 1066 to England and became one of the richest magnates in England with extensive lands in Sussex, Cambridgeshire, Norfolk, Suffolk and later in Yorkshire, totalling more than £1000 annually.7 Probably around the time of the conquest William had married Gundrada, a distant relative of Queen Matilda, and one of the children of the hereditary advocate of the monastery of Saint-Bertin at Saint-Omer.8 Gundrada’s brother Gerbold was briefly earl of Chester before leaving England in 1071, while her other brother Frederick was killed around the same time by the Fenland rebel Hereward.9 It was Frederick’s legacy that provided Gundrada and her husband with the lands in Cambridgeshire and Norfolk. The advocacy of Saint-Bertin passed on to their sons, but it is unclear, as we shall see below, to what extent any Flemish estates actually reached them, and if so, how long they remained in the family.10 For after 1096 the link seems to have collapsed and it was not revived (if this is the right expression) until the time of Earl Hamelin. Shortly before his death in 1088 William I of Warenne was made earl of Surrey by King William Rufus.11 His wife Gundrada having died in 1085, he remarried the sister of Richard Guet, but his second wife does not feature at all in the later recollections of the Warenne family.12 There is no doubt that William exploited his distant kinship and close association with the new king of England as much as his links with his first in-laws. His fortunes derived in part from those of his wife’s Flemish connections, which in turn profited from her distant kinship with Queen Matilda’s family, the counts of Flanders.13 Already at the time of the earliest history of this new family we recognise a wish to be part of the royal orbit, a trend that became stronger as time went on. William I was succeeded by his eldest son William II, whose active career spanned half a century from 1088 until his death on 11 May 1138.14 Leading the life of a typical iuvenis he produced several illegitimate children by a series of concu6 7

EYC viii,1–7; D. Bates, Normandy before 1066, Harlow 1982, 112. Domesday Book i, fols 26r–27v, 196r–v; ii, fols 158r–173v, 400r–v. For a commentary, see J. Green, The Aristocracy of Norman England, Cambridge 1997, 39, 58, 80, 85 and 94. 8 The date of the marriage is suggested in Green, The Aristocracy 352; for Gundrada’s Flemish origins and the identification of her as belonging to the Oosterzele-Scheldewindeke family of the hereditary advocates of Saint-Bertin at Saint-Omer, see EYC viii, 40–6; Anderson, ‘ “Uxor mea” ’, 107–8, and van Houts, ‘The epitaph’. 9 C. P. Lewis, ‘The Formation of the Honour of Chester 1066–1100’, in The Earldom of Chester and its Charters. A Tribute to Geoffrey Barraclough, ed. A. Thacker, Journal of the Chester Archaeological Society 71, Chester 1991, 37–68; E. van Houts, ‘Hereward and Flanders’, ASE 28, 1999, 201–23. 10 EYC viii, 45–6; D. Haigneré, ed., Les chartes de Saint-Bertin, 4 vols, Saint-Omer 1886–99, i, nos 87 and 94. The advocacy of Saint-Bertin is discussed by P. Feuchère, ‘Les avoués de Saint-Bertin’, Société Académique des Antiquaires de la Morinie 17, 1948, 193–207, and very sketchily by A. Giry, Histoire de la ville de Saint-Omer et des institutions jusqu’au XIVe siècle, Paris 1877, 130–1. 11 C. P. Lewis, ‘The Earldom of Surrrey and the Date of Domesday Book’, BIHR 63, 1990, 329–36. 12 EYC viii, 5 note 12, and E. O. Blake, ed., Liber Eliensis, Camden 3rd ser., London 1962, 202–3. 13 Although there is no proof of Gundrada’s kinship with the counts of Flanders other than the assertion in her epitaph, for which see below, pp. 108–9, the name of her brother Frederick featured in the comital family of Luxembourg to which Matilda’s paternal grandmother Ogiva belonged (van Houts, ‘Hereward and Flanders’, 216). Otherwise, it is important to bear in mind the substantial gift of the manor of Carlton (Cambs., see Domesday Book i, fol. 196r) given by Matilda to Gundrada (who gave it to St Pancras at Lewes, see EYC viii, 43), which is usually interpreted as evidence that Gundrada may have belonged to Matilda’s household. Such an interpretation is entirely consistent with kinship. 14 EYC viii, 7–12.

The Warenne View of the Past

105

bines while pursuing potential spouses. His ambition was to catch a royal wife. In the mid-1090s he unsuccessfully wooed the daughter of King Malcolm III of Scotland, Edith/Matilda, who was later to marry Henry I.15 As a result of this unhappy pursuit, some friction remained between William II of Warenne and Henry I, as was later reported by Wace.16 It may also have contributed to the brief fall-out between the king and the earl in the early reign of Henry I, some time between 1101 and 1103. Either just before or just after the disagreement, plans for William’s wedding with one of Henry I’s illegitimate daughters collapsed, but not until after Henry I took advice from Archbishop Anselm who counselled against it on grounds of consanguinity.17 Twenty years later William II finally caught a royal bride when he married the widow of Robert of Beaumont, Isabelle of Vermandois, amidst gossip of intrigue and even accusations of violence.18 Their son William III was born within a year of Robert of Beaumont’s death in 1118, an event that could have given rise to speculation as to the circumstances of the marriage. As the granddaughter of King Henry I (1030–60) by his son Hugh of Vermandois, Isabelle was of royal descent and she was proud of it. She passed on her royal heraldic insigna to both sets of her children, those by Robert of Beaumont and by William II of Warenne.19 In the aftermath of the Battle of the Standard in 1139, the Warennes revived their interest in a royal Scottish connection by marrying off William II’s daughter Ada to Henry, son of king David of Scotland (1124–53), expecting her to become in due course queen of Scotland.20 Since Henry predeceased his father by one year this expectation remained unfulfilled and Ada had to be content to become the mother of two successive Scottish kings, Malcolm IV(1153–65) and William II (1165–1214) instead. There was no royal bride for William III of Warenne, the son and successor of William II.21 The nearest he came to royalty was to exploit his mother Isabelle’s royal cousins by joining his second cousin King Louis VII (1137–80) on crusade in 1147. He marrried Ela, daughter of William Talvas, and named after her mother Ela of Burgundy.22 It was their daughter Isabelle, called after her paternal grandmother Isabelle of Vermandois, who is the heiress and centrepiece of today’s paper. During the career of Isabelle the heiress, the Warenne fortunes were determined by the family’s links with two rival branches of the royal family: the Blois-Boulogne branch of King Stephen (1135–54) versus the Angevins represented by Henry II (1154–89). When King Henry I fell ill in December 1135, William II of Warenne was one of the magnates at his bedside. After the king’s death ‘wise men’ (sapientes) charged him

15 Orderic iv, 272. 16 Wace. The Roman de Rou, trans., G. S. Burgess, ed. A. J. Holden, annot. G. S. Burgess and E. van

Houts, Société Jersiaise 2002, 320–1. 17 F. Schmitt, ed., S. Anselmi Cantuariensis archiepiscopi opera omnia, 6 vols, Edinburgh 1946–63, v,

no. 424, pp. 369–70; W. Frölich, trans., The Letters of Saint Anselm of Canterbury, 3 vols, Kalamazoo 1990–4, iii, 197–8. For a commentary, see K. Thompson, ‘Affairs of State: The Illegitimate Children of Henry I’, JMH 29, 2003, 129–51 at 139, 147–8 and 150. 18 Huntingdon, 598: ‘contigit quendam alium consulem sponsam ei tam factione quam dolosis uiribus arripuisse’ (my translation: it happened that another count stole his (Robert de Beaumont’s) wife as much by intrigue as by lamentable force); cf. Orderic vi, 20–1, 46–7. 19 D. Crouch, The Beaumont Twins: The Roots and Branches of Power in the Twelfth Century, Cambridge 1986, 211–12. 20 EYC viii, 11; D. Crouch, The Reign of King Stephen 1135–1154, Harlow 2000, 89. For Ada, see V. Chandler, ‘Ada de Warenne, Queen Mother of Scotland (c. 1123–1178)’, SHR 60, 1982, 119–39. 21 EYC viii, 12–13. For the crusade, see V. G. Berry, ed., Odo of Deuil. De profectione Ludovici VII in Orientem. The Journey of Louis VII to the East, New York 1948, 54–5 and 122–3. 22 Jumièges ii, 266–7.

106

Anglo-Norman Studies XXVI

with the defence of Rouen and the Pays de Caux.23 Almost immediately he switched to Stephen, whom he supported both in Normandy and in England against Henry I’s daughter, Empress Matilda. William III continued to support Stephen’s cause after his father’s death in 1138.24 Following his untimely death on crusade in 1147/8, King Stephen took the opportunity to marry the Warenne heiress Isabelle to his second son William of Blois, who became William IV of Warenne.25 At his stage there was no reason to suppose that he would ever be the next in line to succeed King Stephen. When, however, his older brother Eustace died in 1153, Stephen arranged a truce with his cousin Henry and accepted him as his heir and successor. William was made count of Mortain and Boulogne and was offered a compensation package which in the corporate world of Anglo-Norman politics was one of the most generous. The terms were laid out in the so-called Treaty of Winchester negotiated between King Stephen and Duke Henry in late 1153, which as a result turned William IV and Isabelle into the richest couple in England and Normandy.26 Four years later in May 1157, however, Henry II, now securely positioned on the English throne, rescinded most of these concessions. The couple’s fortunes were dramatically reduced. The Warenne lands were curtailed by taking away their Norman castles as well as the English grants of Pevensey, Norwich, half the income of Norfolk, and other benefices given in 1153. William IV was confirmed in the possessions his father Stephen had held in 1135, which were Mortain, Boulogne, Lancaster and Eyre.27 In good corporate fashion William and Isabelle decided to sue but before doing so they appealed for support to the pope. What exactly happened is unclear but it looks as if before the matter was heard in the courts (papal and royal), William IV died in 1159 on the return journey from the siege of Toulouse, leaving Isabelle a childless widow.28 Again, as a rich heiress she became a useful royal commodity; except this time it was not King Stephen but his successor Henry II who took advantage of his rights over widows and heiresses. There is evidence that Henry II allowed his younger brother William of Anjou to pursue a marriage alliance with Isabelle. This would solve two problems in one stroke: what to do with a desirable young childless widow and how to provide for a landless younger royal sibling. Unfortunately, the courtship took place against the background of increasing animosity between the king and his chancellor Thomas Becket, who saw an opportunity to frustrate the royal match on grounds on consanguinity. In despair William went to Normandy to consult his mother Empress Matilda, but as Stephen of Rouen movingly tells us in the Draco Normannicus, the young man fell ill and died.29 The archiepiscopal ban was not forgotten, for one of Thomas Becket’s murderers is said to have lifted his fateful sword with the words: 23 24 25 26

Orderic vi, 448–50 at 450: ‘prouido consultu sapientum’ (by the prudent advice of wise men). For the Warenne support for King Stephen, see Crouch, The Reign, 89 and 195. The marriage had taken place by the time EYC viii, no. 50 dated to 1148 had been issued, or soon after. Regesta iii, no. 272; EYC viii, 15–16; J. C. Holt, ‘1153: The Treaty of Winchester’, in The Anarchy of King Stephen’s Reign, ed. E. King, Oxford 1994, 291–316; G. J. White, Restoration and Reform 1153–1165: Recovery from Civil War in England, Cambridge 2000, 6 and 62. 27 Torigni, 192–3; E. Amt, The Accession of Henry II in England: Royal Government Restored 1149–1159, Woodbridge 1993, 27–8; White, Restoration, 115. 28 EYC viii, 17–18. After Wiliam’s death Eye escheated to the Crown, Lancaster folowed in 1164; in 1160 Boulogne went to William’s sister Mary who was released from her abbacy at Romsey in order to marry Matthew of Flanders; Boulogne’s English lands remained in the king’s hand as did Mortain. 29 EYC viii, 14 notes 1 and 2; R. Howlett, ed., ‘The “Draco Normannicus” of Etienne de Rouen’, in Howlett, Chronicles ii, 589–762 at 676; M. Chibnall, ‘The Empress Matilda and her Sons’, in Medieval Mothering, ed. J. Carmi Parsons and B. Wheeler, New York and London 1996, 297–4 at 287. William died 30 January 1164 and was buried at Le Bec’s priory of Notre Dame du Pré (Torigni, 221).

The Warenne View of the Past

107

‘Take this for the love of my lord, William, the king’s brother.’30 Determined to bring the Warenne inheritance into his family King Henry II produced as Isabelle’s new spouse another brother, Hamelin, the illegimate son of his father Count Geoffrey, a man of whom until that moment nothing is known.31 Isabelle and Hamelin were married in the late summer of 1164 with Henry paying for Isabelle’s expenses. The couple produced six children, thereby safeguarding the family’s succession.32 Thus, it had taken four generations for a Warenne to go from a marrriage with a kinswoman (Gundrada) distantly related to the ruling queen (Matilda), to a Warenne daughter (Isabelle) married first to a man who might have become king (William IV) and then to the half-brother (Hamelin) of the ruling monarch (Henry II). How did this series of extraordinary events affect their view of the family past? Let us now turn to the first of the sources produced during Isabelle’s lifetime, namely the tombstone of her most famous ancestor Gundrada.

Gundrada’s Tomb The tombstone is preserved at Southover church, having been recovered from the ancient burial grounds of St Pancras Priory at Lewes, founded by William I and Gundrada in the 1070s. The tombstone is made of black Tournai marble, is 1.935mm long and 620mm wide tapering to a width only 485mm at the bottom.33 It is decorated with two bands of palmette-like plant motifs. Each band contains eight of such units. The foliage is linked by finely carved lions’ heads, many of which are shown full face. Along the four sides of the stone as well as length-wise through the middle runs an inscription. Due to damage to the stone, the bottom (right hand) part has been restored including four of the palmette motifs, but the text of the inscription there is missing. Minor damage in the bottom left hand corner has also resulted in a few lost letters. The tombstone and the inscription show a striking similarity to those of Queen Matilda (d.1083), who is buried in Holy Trinity at Caen (Normandy).34 Her tombstone, however, does not contain any decoration and instead is entirely plain. The comparison with Matilda’s tomb raises the question of the date of Gundrada’s tomb. Stylistically, its decoration dates from the middle of the twelfth century, a date which has been sensibly linked with the refurbishment of the new church of St Pancras in the years from1142–47 onwards.35 It is assumed that sometime during this 30 31 32 33

MTB iii, 142. EYC viii, 18; Torigni, 221. For the expenses, see Pipe Roll 10 Henr II, p. 20 (Michaelmas 1164). EYC viii, 20–4. R. Gough, Sepulchral Monuments in Great Britain, 3 vols, [London] 1786, i, 8–10; R. B. Lockett, ‘A Catalogue of Romanesque Sculpture from the Cluniac Houses in England’, Journ. BAA 3rd ser. 34, 1971, 43–61, at 53–4; G. Zarnecki, ‘Gundrada’s Tombstone’, in English Romanesque Art 1066–1200. Hayward Gallery, London 5 April – 8 July 1984, exhibition catalogue published in association with Weidenfeld and Nicolson, ed. G. Zarnecki, J. Holt and T. Holland, London 1984, 181–2; Anderson, ‘ “Uxor mea” ’, 107–29; van Houts, ‘The Epitaph’. 34 M. Baylé, ‘L’Abbaye aux Dames à Caen’, Art de Basse-Normandie 93, 1986, 1–40, at 14, and Zarnecki, ‘Gundrada’s Tomb’, 182. 35 Zarnecki, ‘Gundrada’s Tomb’, 181; EYC viii, no. 32; W. H. Blaauw, ‘On the Early History of Lewes Priory, and its Seals, with Extracts from a Ms Chronicle’, Sussex Archaeological Collections 2, 1849, 7–37, at 11–12; B. M. Crook, ‘General History of Lewes Priory in the Twelfth and Thirteenth Centuries’, Sussex Archaeological Collections 81, 1940, 68–96; F. Anderson, ‘St Pancras Priory, Lewes: its Architectural Development to 1200’, ANS 11, 1988 (1989), 1–35, and M. Lyne, Lewes Priory: Excavations by Richard Lewis 1969–82, ed. M. Gardiner, Lewes 1999.

108

Anglo-Norman Studies XXVI

period in the late 1140s or early1150s (or as Anderson unconvincingly argues as late as the 1160s),36 but most likely when the church was dedicated in 1147, Gundrada’s remains as well as those of her husband William – originally buried in the chapter-house – were deposited in two lead caskets, each engraved with their name, and reburied in front of the main altar of the new church. At the same time the couple probably received new tombs, a hypothesis which is difficult to verify because of the loss of William’s tombstone. Gundrada’s has not only survived, but is in excellent condition.37 What was Gundrada’s original tombstone like? It probably was similar to that of her husband William, whose remains c.1088 were buried in a tomb of white stone, with an epitaph engraved on it, as we know from Orderic Vitalis.38 Even though we cannot prove it, it seems probable that in the mid twelfth-century William, like his wife, was given a new tomb. Does it follow, however, that both tombs were provided with new twelfth-century epitaphs, or were the originals, written at the time of their deaths, copied?39 Gundrada’s epitaph, like William’s, is written, as was common, in dactylic hexameters.40 The message of both epitaphs is the fact that St Pancras, patron saint of Lewes Priory, had become the heir of lands given by William and Gundrada.41 William’s verse states that the gifts of Earl William, the priory’s founder, were freely and gladly given in return for which the saint as his heir may give him a seat in heaven.42 The same sentiment can be found in Gundrada’s epithaph, where in line 7 she is said to make the saint her heir, and whose clemency is appealed to. The reference to Gundrada’s motherhood has a double meaning. It is an allusion to the fact that on 27 May [1085] she had died in childbirth at Castle Acre (Norfolk), where her 36 Anderson, ‘ “Uxor Mea” ’, 112. 37 This is due to its re-use as a table tombstone at Isfield church, where it formed part of the tomb of

Edward Shirley who had died on 16 March in 1550. Two centuries later, in 1774, Gundrada’s tombstone was discovered and put on display. Subsequently, in 1845, it was reunited with the lead caskets found by accident during the building of the Lewes railway, which cut across the ruins of St Pancras priory. Because St John the Baptist Church in Southover was the nearest church to the priory’s ruins, it was chosen as the most suitable location for the preservation of the tombstone and the caskets, and for the reinterment of the couple’s bones. 38 Orderic iv, 180–1. 39 Anderson, ‘ “Uxor Mea” ’, 112, 114 seems to imply that the inscription dates from the time of the mid-twelfth-century tomb. 40 Van Houts, ‘The Epitaph’ contains a critical edition and commentary. Here follows my reconstructed text: 1 Stirps Gundrada ducu[m] dec[us] eui, nobile germen:/ 2 intulit ecclesiis Angloru[m] balsama moru[m]:/ 3 marti[ris] [. . .]/ 4 [. . .]it miseris fuit. Ex pietate Maria./ 5 Pars obiit Marthe sup[er]est pars magna Marie./ 6 O pie Pancrati tes[tis pietat]is et equi/ 7 te facit herede[m] tu clemens suscipe matrem/ 8 sexta kalendaru[m] iunii lux obuia carnis./ ifregit alabastu [. . .]. Translation: ‘Gundrada, distinguished offspring of dukes and noble shoot in her own time, brought to the churches of the English the balms of the martyr’s (?) traditions [. . .] to those in misery she was in her piety a Mary. The part of Martha [in her] died, the greater part of Mary survives. O, pious Pancras, witness of piety and justice, she makes you her heir; may you in your clemency accept the mother. Her light faded on the 27th of May, when she broke the alabaster [vase . . .].’ 41 A. Z. Huisman, Die Verehrung des heiligen Pancratius in West- und Mitteleuropa, Amsterdam 1938, 42, and P. Franci de’ Cavalieri, Hagiographica, Studi e Testi 19, Rome 1908, 75–120. 42 Orderic iv, 180–1: ‘Hic Guillelme comes locus est laudis tibi fomes./ Huius fundator et largus sedis amator,/ Iste tuum funus decorat, placuit quia munus,/ Pauperibus Christi quod prompta mente dedisti./ Ille tuos cineres seruat Pancratius heres,/Sanctorum castris qui te sociabit in astris./ Optime Pancrati fer opem te glorificanti,/ Daque poli sedem, talem tibi qui dedit aedem.’ Translation: ‘Earl William, in this place your fame is kindled,/ You built this house and were its generous friend:/ This was a gift freely and gladly given/ To the poor of Christ; it honours now your urn./ The saint himself, Pancras, who guards your ashes,/ Will raise you to the mansions of the blessed./ Saint Pancras, give, we pray, a seat in heaven/ To him who for your glory gave this house.’ Orderic introduces the epitaph by saying that the verses were carved on white stone, either sandstone or, less likely, white marble.

The Warenne View of the Past

109

child may well have survived.43 On the other hand, it also reads as a reminder of Gundrada as the priory’s founder and, therefore, as mother of monks. The use of the present tense in line 7, concerning Gundrada’s bequest to St Pancras, strongly suggests an immediacy and actuality which suit the contemporary circumstances of her death and a date of composition in 1085, rather than for a possible revision in the mid-twelfth century. While the theme of the couple’s bequest of their land to St Pancras is the same in both epithaps, that of Gundrada contains the additional theme of penance. Her epitaph couples the bequest to St Pancras with the description of the saint as patron of justice and piety and the appeal to his clemency.44 The penitential theme is reinforced by the comparison of Gundrada with Martha and Mary, known from the Gospel according to John as the sisters of Lazarus.45 The first three words of the epitaph, ‘stirps Gundrada ducum’, describing Gundrada as either stem or progeny of dukes, has attracted much attention from historians. Whereas the conventional interpretation has favoured the sense of Gundrada as ducal offspring,46 the most recent suggestion by George Zarnecki prefers to see her as stem of dukes, or in other words as the mother of a line of earls.47 Inevitably, the two interpretations bring us back to the issue of the epitaph’s date. Zarnecki implies a date sometime in the 1140s by which time, indeed, three earls of Warenne had succeeded one another: Gundrada’s husband William I, their son William II (1088–1138) and their grandson William III (1138–48). However, the immediacy of the bequest and the theme of penance hint at the circumstances of 1085 when the notion of Gundrada as the origin of a Warenne dynasty of earls would have been highly anachronistic. At that time, a mother of at least two sons and one daughter, no one could have foreseen that she would become the progenitor of earls. In the first place her husband William had not even been made an earl; his elevation to that rank did not take place until the period between Christmas 1087 and Easter 1088.48 From these observations it follows that ‘stirps Gundrada ducum’ should be interpreted as Gundrada descending from a line of dukes (or counts). It also follows that the epitaph was written contemporaneously with Gundrada’s death and not in the mid-twelfth century. The reburial of the caskets with the remains of William I and Gundrada is, as I have said already, very likely to have taken place during the dedication ceremony of the new church of St Pancras in 1147 just before William III of Warenne went off on crusade. A charter issued on this occasion testifies to the family’s awareness of the need to leave tangible objects behind as pegs for memorial traditions. William III and his brother Reginald both had their hair cut by Bishop Henry of Winchester who deposited the locks on the altar as evidence for the earl’s confirmation of the priory’s possessions. Thus the tombstone and the locks of hair as well as the continuation of the use of Gundrada’s name guaranteed that neither she nor her gifts would be forgotten.49 However, not very long after the dedication ceremony at St Pancras in 43 EYC viii, 5–6. The surviving child may well have been her daughter Edith, who married first Gerard of

Gournay and later Drogo de Monchy. 44 For St Pancras as an avenger of perjury and treason, see C. P. Lewis, ‘The Earldom of Surrey and the

Date of Domesday Book’, BIHR 63, 1990, 329–36, at 332, as well as Huisman, Die Verehrung, 142–4. 45 John 12:1–11; cf. Matt. 26:6–13. 46 Gough, Sepulchral Monuments, 9; EYC viii, 40–6 contains a lucid overview of the various interpreta-

tions. 47 Zarnecki, ‘Gundrada’s Tomb’, 181. 48 Lewis, ‘The Earldom’, 330–1. 49 Before the middle of the twelfth century her name appears only once in the Warenne charters (EYC

viii, no. 15 dated 1106x1118); such explicit naming was perhaps not yet necessary as her name, and thus

110

Anglo-Norman Studies XXVI

1147 events took a momentous turn, which within a decade necessitated the recording of the history of the Warennes in much greater detail than an epitaph or charters allowed, or than the memory of a lock of hair could provide. The Warenne Chronicle, so I will argue, was written around this time of crisis and it is to the chronicle that we will now turn.

The Warenne Chronicle (also known as the ‘Hyde Chronicle’) Let us fast forward to the year 1157 when probably in May Henry II took away the many possessions he had given to King Stephen’s son William of Blois and his wife Isabelle of Warenne as part of the Treaty of Winchester in 1153. The couple protested and there is good evidence that William IV appealed to Rome to the pope who at that time was none other than Adrian IV (1154–9), the Englishman Nicholas Breakspeare. It comes in the form of John of Salisbury’s undated letter 46.50 It states that Archdeacon B, commonly identified as Baldwin of Boulogne, archdeacon of Sudbury, had spread the story that the pope had instructed Baldwin to collect 700 marks from William IV in return ‘for listening [by the pope] to their [William’s and Isabel’s] appeals against the king’.51 Although the nature of the appeals is not specified it is commonly assumed and persuasively argued that they relate to Henry II’s seizure of William’s lands in England and Normandy. The legal case put forward presumably was that of royal breach of the treaty of 1153, in other words the likely accusation would have been that Henry II had committed perjury. Intriguingly, Robert of Torigni in his Chronicle introduces the very passage on Henry II’s treatment of William IV by characterising it as similiter to the king’s dealings with King Malcolm IV, which he had reported in the previous paragraph.52 The similarity, according to Robert, lies in the fact that Malcolm too was forced to hand back to the king lands – no less than the northern shires of England – and importantly he hints here that this action meant breach of an oath.53 Since Henry II had justified his actions by saying that all he did was to restore the lands under his authority to their status on the death of his grandfather Henry I, it seems reasonable to assume that any Warenne appeal would have comprised information as to the state of Warenne affairs in Henry I’s reign. It is this assumption that warrants a reconsideration of the text known as the ‘Hyde Chronicle’ or, now, the Warenne Chronicle. Is it possible that it might have formed part of a historical dossier put together as evidence for the support

presumably knowledge as to who she was, survived amongst her granddaughters. Her daughter Edith and her first husband, Gerard of Gournay (d. after 1104), named their child Gundrada, who c.1118 became the wife of Nigel d’Aubigni (d.1129); she died a widow in c.1154. Secondly, William II of Warenne and Isabel of Vermandois named one of their daughters Gundrada; she married first Roger of Warwick and secondly, after 12 June 1153, William of Lancaster; she herself died after 1166. This Gundrada, too, passed on the family name to a Gundrada who died in 1200 or 1208. Interestingly, Isabelle the heiress of Warenne never named any daughter after her great-grandmother; instead, she called her daughters Ela and Isabelle. 50 W. J. Millor, H. E. Butler and C. N. L. Brooke, eds, The Letters of John of Salisbury, vol. i: The Early Letters (1153–61), OMT, Oxford 1986, no. 46. 51 The Letters of John of Salisbury i, no. 46, p. 82: ‘. . . dcc marcas reciperet [sc. Baldwin], quas pro audiendis petititonibus suis contra regem’. 52 Torigni, 192. I am most grateful to John Gillingham for drawing my attention to this passage. 53 Both William of Newburgh (Newburgh i, 70) and Roger of Howden (Howden, Chronica i, 211) report that Henry had promised these lands to Malcolm’s father David in 1148/9, perhaps on the occasion of Henry’s knighting by King David on 22 May 1149 (Torigni, 159–60). For a commentary, see Crouch, The Reign, 323–4.

The Warenne View of the Past

111

the Warennes had given to the Norman kings including Henry I, in an attempt to persuade King Henry II not to take away the ancestral lands of the Warennes? In order to answer this question we have to remind ourselves of the narrative and its contents. It is a relatively short history of approximately forty pages in print and covers the period from 1035 to 1120, when the narrative breaks off abruptly in the account of the White Ship disaster. The only surviving manuscript, BL Cotton Domit. A xiv ff 3–21v, in a late thirteenth-century hand, reveals that the copyist had an exemplar that was already incomplete. It is impossible to say what the original text looked like. Did it extend until the death of King Henry I in 1135? Did it continue beyond his reign? Or, was composition halted, perhaps suddenly, because events had overtaken the reason for its composition? Apart from some potential anticipatory remarks, the author has left virtually no clues as to the dating of the tract.54 Only a close analysis of the text, and in particular a literary comparison of the various historical anecdotes with other versions by Orderic Vitalis, William of Malmesbury and others, suggests that in all cases the text of the Warenne Chronicle is the latest version. Let me give you an example. The death of Earl Godwin, father of Harold, in 1053 is narrated here in the longest known version, full of direct speech, elaboration and otherwise unknown detail. All of this points to literary embroidery.55 The pattern is the same elsewhere, e.g. the murder of the aetheling Alfred, Countess Gytha walking barefoot to Mass everyday, the siege of York in 1066, Bishop Odo of Bayeux’ s papal pretensions and William the Conqueror’s cursing of Robert Curthose in 1079. For this reason we are looking for a date of, say, sometime in the1140s or probably the1150s, well after the composition of the main Anglo-Norman narratives. What were the circumstances of the one family whose deeds are highlighted in the text? It has long been recognised that the earls of Warenne, especially William I and William II occupy a prominent place in the narrative. William I is said, uniquely, to have been made earl after the battle of Pevensey in 1088, and his burial place next to his [unnamed] wife, is given at Saint Pancras at Lewes.56 He is also said to have divided his inheritance amongst his two sons with the English lands going to William II and the Flemish lands going to Reginald.57 William II is said to have temporarily deserted Henry I, but to have done so for not as long as William of Mortain and Robert of Bellême;58 otherwise the text contains two eulogies of him as a staunch supporter of Henry I.59 There is, moreover, uncorroborated evidence for William II’s dealing with the northern Norman magnates,60 and there is plenty of incidental detail not found elsewhere relating to members of the extended Warenne family. In partic54 E.g., in par 13 (ed. Edwards, 297) the author comments that ensuing time shows how effective William

the Conqueror’s curse of his son Robert had been, a remark that could have been made any time after 1106 but the curse’s fulfilment can only have made sense after Robert’s death in 1134. In par 28 (ed. Edwards, 307) after Tinchebrai, according to the author, there were terrible consequences for ‘him [Henry I] . . . and those who came after him/ his successors (contra eum . . . et contra posteros eius)’, i.e. Stephen and Henry II, which takes us to after 1154. On various occasions the author refers to the memory of people ‘until the present day’, e.g. Waltheof (par 11, ed. Edwards, 295), and ‘the laws of Edward’ (par 7, ed. Edwards 290), a moment in time difficult for us to establish. Cf. also the reference to the future after Edith/Matilda (par. 31, ed. Edwards, 311–12). 55 As was already pointed out by C. Wright, The Cultivation of Saga in Anglo-Saxon England, Edinburgh–London 1939, 233–6 and 296–8. For more examples, see my forthcoming edition. 56 Par 16 (ed. Edwards, 298–9) and 17 (ed. Edwards, 299). 57 Par 17 (ed. Edwards, 299). 58 Par 24 (ed. Edwards, 304–5). 59 Par 31 (ed. Edwards, 311–14 at 313) and 34 (ed. Edwards, 316–18 at 316–17). 60 Par 27 (ed. Edwards, 307) and 36 (ed. Edwards, 319–21 at 320).

112

Anglo-Norman Studies XXVI

ular, there are unique details about King Stephen’s family of Blois-Chartres and of their Boulogne links.61 There is also new material on Mortain.62 All this information points to someone writing at a crucial time for the Warenne family in conjunction with a relative of the Blois-Chartres/Boulogne families. I suggest that the juncture is the period of the marriage of William IV, son of King Stephen and Queen Matilda heiress of Boulogne, and Isabelle of Warenne. Within the period stretching from 1147/8, when they married, and 1159, when William IV died, the real crisis came in May 1157 with King Henry’s actions against the couple. If the crisis of 1157 can be identified as the cataclyst for the compostion of the text, who might have been its author? The most likely person to be responsible for a brief chronicle setting out the case for Warenne support for Henry I is someone with intimate knowledge of the Warenne family, of William IV son of King Stephen and of the wider kin, who also had access to knowledge about the funeral arrangements for Queen Edith/Matilda in 1118 whose commemorative details have, so famously and uniquely, survived in the Warenne Chronicle.63 The best candidate must surely be Master Eustace of Boulogne, also known as Eustace the Chancellor.64 He was a royal chaplain at the court of King Stephen and Queen Matilda, and was most probably appointed as chancellor to William IV after his marriage in 1147/8 to Isabelle of Warenne. His first occurrence, as ‘Eustacius clericus comitis Willelmi filii regis’, comes in the St Pancras charter of 1148 drawn up shortly after the marriage and while details of Earl William III’s whereabouts were still unknown.65 He witnessed seven more of William’s charters in the period 1148–1159 concerning Warenne properties in England, Normandy and France.66 His early connection with the royal court can be established by identifying him as Eustace, brother of Baldwin of Boulogne, well known to Thomas Becket.67 The two secular clerks, attached for while to St Martin-le-Grand in London, were probably introduced to royal service through Queen Matilda, herself heiress of Boulogne.68 It is particularly significant that Eustace’s brother Baldwin is the very same Baldwin whom we encountered earlier as the Archdeacon Baldwin who was to collect the papal fee for listening to William’s and Isabel’s appeals against Henry II.69 61 For Theobald IV of Blois–Chartres, see par 30 (ed. Edwards, 309–11) and for Mary of Boulogne, see

par 25 (ed. Edwards, 306). 62 Par 25 (ed. Edwards, 306) and 28 (ed. Edwards, 308). 63 Par 31 (ed. Edwards, 311–14 at 312–13); for a commentary on the queen’s commemorative passage,

see now L. L. Huneycutt, Matilda of Scotland: A Study in Medieval Queenship, Woodbridge 2003, 145–6. 64 EYC viii, 50. I should like to acknowledge my debt to David Crouch’s contribution to the unravelling

of Eustace’s authorship. 65 EYC viii, no. 50; I am grateful to Edmund King for allowing me to read an unpublished chapter of his

forthcoming study of King Stephen. 66 EYC viii, 48–9, nos iv, vii, viii, xi and xii. The editors’ emendation of ‘cancellarius’ (in no. xi) is almost

certainly right, in the light of a new original charter that can be added to the series. Amiens, Bibliothèque municipale, MS 992 contains an original charter of William IV as count of Boulogne, confirming an agreement between the abbot of Beaulieu and Beaulieu’s founder, Aitropus. The charter, datable to 1157, contains a series of witnesses, one of whom is none other than ‘magistro eustac[hi]o cancell[ario]’. I am immensely grateful to Steven Vanderputten of the University of Gent for providing me with a photograph of the charter. A brief reference to this charter was given by D. Haigneré, in Mémoires de la Société Académique de l’Arrondisement de Boulogne-sur-Mer 13, 1882–6, 388–9. 67 Regesta iii, ix and nos 117 and 541, and MTB iii, 15: ‘per duos fratres Bolonienses Baldewinum archidiaconum et magistrum Eustachium, hospites plerumque patris ejus [sc. Thomas Becket]’. 68 R. H. C. Davis, ‘The College of St Martin-le-Grand and the Anarchy, 1135–54’, London Topographical Record 23, 1974, 9–26. 69 Fasti ii, 69, and EEA vi, xl, lxxxvi and 361. The identification fits well with the author’s reference to

The Warenne View of the Past

113

If my suggestion for the authorship and circumstances of the composition of the Warenne Chronicle some time after May 1157 but before 1159 can be accepted, the chronicle becomes a very interesting testimony to the family memories, naturally carefully selected to project William I and William II of Warenne as staunch supporters of Kings William I, William Rufus and Henry I. The narrative, as I will show in the annotation prepared for the edition, is full of details that must have survived orally. To shape this mostly oral tradition into a more authoritative framework, the author used various Anglo-Norman chronicles such as the Gesta Normannorum Ducum in Robert of Torigni’s version and those by William of Malmesbury and Henry of Huntingdon. However, surprisingly, his greatest debt is to Orderic Vitalis. For a substantial number of passages, there is no doubt that the author of the Warenne Chronicle used Orderic. The links between the two texts are so close, including some literal quotes,70 that it is simply not credible that two such narratives could have sprung up independently; nor have I found evidence for a lost source that both could have used. In this context, it is surely not a coincidence that we can actually find Eustace the Chancellor in Normandy in the mid-1150s in southern Normandy (in 1155–8 at Coutances and in 1158 at Tinchebrai), from where it would have been easy to arrange a visit to the library of St Evroult for a consultation of Orderic’s manuscripts.71 The Warenne Chronicle is not a dynastic history, nor is it a legal document. It is a historical narrative that highlights the Warenne family and reveals special pleading on their behalf because of the loss of ancestral lands. If, as I have suggested above, their argument against the king was that he had committed perjury, a chronicle highlighting the terrible results of the Norman conquest for the English and in particular the wrath of St Pancras against King Harold, the perjurer of 1066, was an eloquent warning to King Henry II of the potential dangers he was facing.72 Because it is incomplete, we do not know how it might have ended. At this point I should like to reiterate my earlier remark that it is not impossible that the narrative was never finished because in 1159 William IV died and the Warenne situation was under review. Nevertheless, there is no doubt in my mind that it was composed in the context of fear of losing the very castles in northern Normandy that had been in the Warenne family since the mid-eleventh century.73 himself as non-English, see par 25 (ed. Edwards, 288): ‘regio que lingua eorum dicitur Sudsexia’ (‘the region which in their language is called Sussex’). 70 The link with Orderic is particularly striking in the sections on Waltheof and Gyrth (par 11, ed. Edwards, 293–5 at 294–5; cf. Orderic ii, 320–1), on Gerbod of Chester (par 12, ed. Edwards, 295–6 at 296; cf. Orderic ii, 264–5), on the siege of Rochester and Pevensey (par 15, ed. Edwards, 298; cf. Orderic iv, 128–9 and 134–5), on Helias of St Saens and William Clito (par 28, ed. Edwards, 308; cf. Orderic vi, 92–3 and 162–3, 164–6) and on Brémule (par 34, ed. Edwards, 316–18 at 316; cf. Orderic vi, 234–7). For a striking verbal echo, compare par 15 (ed. Edwards, 198, line 15) ‘cuius rei fama tota Anglia concutitur’ with Orderic iv, 134 ‘cum supra dicta tempestate uehementer Anglia undique concuteretur’. Moreover, in pars 12, 14, 23–5, 28, 30–1, 34 and 36 there is too much agreement between Orderic and the Warenne chronicler for it to be explained by coincidence. However, in several cases Orderic and the Warenne chronicler disagree, with the latter giving the more plausible or correct information, as when Orderic mistakenly identifies Gundrada as William I’s widow (Orderic iv, 180–1 and par 17, ed. Edwards, 299), or misplaces both Reginald’s release from prison in 1106 (Orderic vi, 88–9 and par 27, ed. Edwards, 307) and that of William of Nevers in 1117 (Orderic vi, 258–9 and par 30, ed. Edwards, 309–11). 71 For Eustace’s presence in south-western Normandy, see EYC viii, 48–9 nos xi and xii. 72 For Harold’s oath sworn in c.1064 on relics, uniquely identified as those of St Pancras in the Warenne Chronicle, and subsequently famously broken in 1066, see par 6 (ed. Edwards, 290). 73 No return was made for the Norman castles in 1172 (H. Hall, ed., The Red Book of the Exchequer, 3 vols, RS 1896, ii, 644) but at a later date in the twelfth century the castles were back in Warennne hands (L. C. Loyd, ‘The Origin of the Family of Warenne’, The Yorkshire Archaeological Journal 31, 1932–4, 97–113 at 110).

114

Anglo-Norman Studies XXVI

Having considered the tombstone and the chronicle it is now time to turn to the third and last group of sources to be discussed, namely the two clusters of charters. Both emanate from Earl Hamelin and help to provide the framework for the Warenne view of the past. The first cluster concerns the commemoration of Warenne ancestors in general and his own Angevin origins in particular, while the second cluster centers on the Warennes’ Flemish origins.

The Charters Hamelin’s marriage with Isabelle in 1164 began a new trend in the commemorative traditions of the Warenne family. During the time of the first three Williams charters contain standard phrases referring to previous earls, but they were very short and laconic. Thus, for example, we find brief references to William I and William II and only one charter named Gundrada.74 This is no surprise because oral tradition and living memory meant that names and stories were still fresh in people’s minds and easily passed on. With the arrival of William IV not very much changed in this respect, except for the snippets of information we find about the Blois-Chartres and Boulogne connection in the Warenne Chronicle. But once Hamelin joined the family, the commemorative passages in the charters become more extensive. They gradually cover more generations and thus present a longer genealogical line. They focus on the Warenne ancestors and for the first time consistently name not only the earls of Warenne, but also their countesses, listing them as Gundrada, Isabelle (of Vermandois) and Ela, though not the elusive sister of Richard Guet, William I’s second wife.75 Isabelle’s own interest can be found in the Warenne’s charters, which reserve a special place for her mother Ela and, after her mother’s remarriage to Earl Patrick of Salisbury, also for her half-siblings.76 Thus the documents bear testimony to the continued warm relations Isabelle maintained with her mother until her mother’s death in 1174. Another testimony to this is the inclusion of Salisbury material in the Warenne Chronicle. Knowledge of the crusading career of Ernulf of Hesdin, great-grandfather of Earl Patrick, is uniquely preserved there.77 Hamelin’s own interest resulted in the commemorative passages in the Warenne charters embracing the Angevin counts. In one early original charter of Earl Hamelin for St Pancras at Lewes, possibly datable to 1166 – because it refers to children in the plural – he specifically asked for prayers for the souls of his father and mother, though neither are named, making this the only document that explicitly commemorates his unknown mother.78 Later charters regularly name his father as Count Geoffrey of Anjou.79 As I have pointed out before, we know nothing about Hamelin before his marriage to Isabelle. Yet, one charter has survived that lifts perhaps a tip of the veil over his past. As a staunch supporter of his half-brother King Henry II, Hamelin occurs regularly (twelve times) as a witness to royal charters including one to St Pancras at Lewes in 1170.80 His attestations peak with the accession of Richard I, when he witnessed fourteen times in the period 16 September 1189 to 74 75 76 77 78 79 80

EYC viii, no. 15 (dated to 1106–1118). EYC viii, no. 64 (1189–1202) and no. 80, p. 123. EYC viii, no. 111 (dated 1159–64). Par 21 (ed. Edwards, 301–2). EYC viii, no. 77. EYC viii, nos 58, 60, 64–6, 74, 76 and no. 80, p. 123. What follows is based on information kindly provided by Judith Everard of the Henry II Acta project at

The Warenne View of the Past

115

27 June 1190.81 Very shortly afterwards, probably in late June or early July 1190, potentially – judging by the charter of 27 June 1190 (Charter 1117R) – while at Montrichard, Hamelin arranged an interesting exchange of lands with King Richard I. Hamelin received from Richard I the town of Thetford in Norfolk, in return for his own lands at Ballan, Colombiers and Chambray-lès-Tours.82 These three estates are all situated on the southern fringe of Tours strategically on the route to Chinon, which suggests that they were comital possessions of the counts of Anjou. As Judith Everard has pointed out to me, they are interesting because they are the very places to which in the summer of 1189 Henry II retired shortly before he died on 6 July at Chinon. It suggests he could rely on Hamelin’s loyalty.83 The question arises when Hamelin received these lands. One possibility, suggested by Judith Everard, would be that he had received them from King Henry II, his halfbrother, either early on in c.1164 around the date of his Warenne marriage, or perhaps much later in c.1189 from the king on his deathbed. Alternatively, I would suggest, Hamelin’s possession may go back further in time and may constitute not his half-brother’s but his father Geoffrey’s gift, in which case it must have predated 1151, or it could be property that belonged to his maternal kin.84 Hamelin is a frequent name in Anjou, Vendôme and Maine, which may suggest that his father’s concubine originated there. Whatever the origin of Hamelin’s Touraine posessions, it it significant that Richard was keen to get these strategically important places back as soon as he had become king. Clearly, for Hamelin, having been absorped into the Warenne orbit, the acquisition of the prosperous town of Thetford was an important asset to the Warenne lands in East Anglia. Valued at £35 per annum, any surplus had to be handed over to the royal exchequer, but any shortfall had to be made up by the king. Thetford thus was of more immediate advantage to Hamelin as earl of Warenne than the far away estates in the Loire valley. Hamelin’s administration of the honour of Warenne is perhaps best known for the vast sums he spent on what is generally agreed to be his most prestigious undertaking, the constuction of Conisbrough castle (Yorks.) dated to the 1180s.85 Significantly, perhaps due to his search for extra income to finance this building project, his responsibility for the Warenne interests had already forced him to look to northern France to the prosperous town of Saint-Omer,86 where in the early 1180s he seems to Cambridge. The charters are nos 1192H, 4290H, 2702H, 388H, 1748H, 1776H, 1177H, 885H, 74H (for St Pancras), 4585H, 1260H, 2037H and 2053H. 81 I owe this information, again, to Judith Everard. The charters are 133R, 1379R, 137R, 305R, 1320R, 918R, 1186R, 1002R and 1246R, 453R, 3544R, 472R, 2632R, 3990R, 1117R. 82 EYC viii, no. 82 where Ballan and Colombiers are identified. The identification of Chambrayles-Tours is Judith Everard’s. 83 This is based on Judith Everard’s unpublished ‘Itinerary of Henry II’, s.a. 1189. According to Gerald of Wales (Giraldi Cambrensis viii, 286–7) on 2 July Henry II was in Ballan and a number of authorities place him on 4 July at Colombiers. 84 Bernard Bachrach suggests a link with the Hamelini at Langeais, the castle built by Fulk Nerra 24km downriver from Tours (B. S. Bachrach, Fulk Nerra, the Neo-Roman Consul 987–1040, Berkeley 1993, 49–51, 58–9). A Hamelin de Langeais became first lord of Montatoire by right of his wife, see O. Guillot, Le comte d’Anjou et son entourage au XIe siècle, 2 vols, Paris 1972, ii, no. C216. Colombiers (com. Villandry, arr. Tours) was in existence by 1093 and belonged at that time to Paien de Mirebeau, son of Jean de Chinon; he held both castles on behalf of the count of Anjou (Guillot, Le comte d’Anjou i, 300 note 118 and 323 note 199). 85 M. W. Thompson, Conisbrough Castle, HMSO 1959, 1–2. I am most grateful to Charles Coulson for drawing my attention to Hamelin’s building operation, and for providing me with a copy of Thompson’s booklet. 86 For the enormous expansion of Saint-Omer as a prosperous harbour and trading centre, see A. Derville, Saint-Omer des origines au début du XIVe siècle, Lille 1995, 47–82. For the contacts with

116

Anglo-Norman Studies XXVI

have single handedly restored the old Warenne link with the abbey of Saint-Bertin. This brings us to the second cluster of Hamelin’s charters. We may recall that Gundrada of Warenne was almost certainly a daughter of Gerbod of Oosterzele-Scheldewindeke, the hereditary advocate of Saint-Bertin during the mid-eleventh century. Through her the advocacy passed to her two sons, William II of Warenne and Reginald, whom we can trace as advocates of the abbey as late as the 1090s: surviving Saint-Bertin charters show Reginald there in 1091 and William II of Warenne in the summer of 1096.87 Subsequently, presumably due to the increasing tensions between Normandy, England and Flanders, the Warenne connection collapsed and the advocacy seems from then on to have alternated between the counts of Béthune and the family of Fauquembergues, hereditary castellans of Saint-Omer. It comes then as something of surprise when we find Earl Hamelin in c.1182, seemingly out of nowhere, re-appearing as advocate of Saint-Bertin and granting the abbey three mesura of land at Rocquetoire, known to have belonged to Gundrada’s family until 1096. The document itself is known only from a copy made in 1775 by Charles-Joseph Dewitte, whose ‘grand cartulaire’, now Saint-Omer BM MS 803, is the nearest we can come to the original charters of Saint-Omer (all lost during the French Revolution).88 Moreover, another charter dates from around the same time, issued by Count Philip of Flanders (1168–91) and confirming an arrangement between Earl Hamelin and Renaud of Aire, the abbey’s vassal.89 Over the next few years various other transactions with Saint-Bertin, or with its tenants, involving Hamelin, Isabelle and their son William took place. They can only have happened with the agreement of the kings of England, Henry II, Richard I and John, as also with that of the counts of Flanders, Philip (1168–91), Baldwin VIII of Flanders (= Baldwin V of Hainault 1191–94) and Baldwin IX (1195–1206).90 How do we explain this renewed interest in Flanders? And what was known by Hamelin and Isabelle about the Warenne past in Flanders? We have established that the tombstone refers to Gundrada’s noble background but without specifying her country of origin. The epitaph in 1085 did not need to specify what was a well known fact. In the early twelfth century Orderic Vitalis, using information from Warenne-related monks at Saint-Evroult, referred to

England, see G. Dept, ‘Les marchands flamands et le roi d’Angleterre (1154–1216)’, Revue du Nord 12, 1926, 303–24, and Amt, Accession, 83–99. 87 D. Haigneré, ed., Les chartes de Saint-Bertin, 4 vols, Saint-Omer 1886–99, i, nos 87 and 94. 88 The charter can be found in ed. Haigneré, i, no. 325, where on p. 144 Haigneré accepted the dating of c.1182, proposed by Dewitte, who described the charter as an original with seal. As for the early history of Rocquetoire, in 1096 Arnulf and Gerbod, older sons of Earl Gerbod of Chester, sold their allod at Rocquetoire (20km south-east of Saint-Omer) for 14 mark of silver to Abbot Lambert of Saint-Omer (ed. Haigneré, i, no. 96); cf. EYC viii, 45–6. Note that on 30 July of that year William II of Warenne had been at the abbey in his capacity of advocate (ed. Haigneré, i, no. 94). Since this is the last known occurrence of Warenne dealings with the abbey one wonders whether William had a say in the sale of Rocquetoire. 89 Ed. Haigneré, i, no. 327, with on p. 145 the date of c.1183 and described as ‘original with seal’. 90 Sometime before 1191, Earl Hamelin, his wife and son had agreed to the sale of one third of the tithe of Rincq, as well as land at Vaux held by Baldwin of Haverskerque, to the abbey of Blendeques for which Count Philip had paid restitution to Earl Hamelin (W. Prevenier, ed., De oorkonden der graven van Vlaanderen (1191 – aanvang 1206), 3 vols, Brussels 1964–71, ii, no. 100, pp. 216–18). In 1197 William of Warenne, son of Hamelin and Isabelle, acted as guarantor for Richard I’s treaty with Baldwin IX of Flanders, repeating earlier Anglo-Flemish treaties (P. Chaplais, ed., Diplomatic Documents Preserved in the Public Record Office, Volume I: 1101–1272, London 1964, no. 1, pp. 1–4 (1101), no. 2, pp. 4–8 (1110), no. 3, pp. 8–12 (1163), no. 4, pp. 12–13 (money fiefs 1163) and no. 7, pp. 18–20 at 19). For discussion, see G. Dept, Les influences anglaise et française dans le comté de Flandre au début du XIIIe siècle, Gent–Paris 1928, 21–3.

The Warenne View of the Past

117

Gundrada as sister of Gerbod, whom he explicitly linked with Flanders.91 Master Eustace of Boulogne, the putative author of the Warenne Chronicle, repeats Orderic’s information and, as we have seen, adds the detail that William I divided his inheritance in England and Flanders between his sons William II and Reginald, information that is, as we have seen, confirmed by contemporary charter material from Saint-Bertin.92 Presumably he relied on the Warenne family itself, or, conceivably, on local Boulogne connections since Boulogne had many dealings with nearby Saint-Omer. Otherwise, there is no surviving record of Warenne interest in Flanders.93 But what about Saint-Bertin itself? In the 1130s the Gesta abbatum sancti Bertini, written originally by Folcuin of Saint-Bertin in the late tenth century, was updated by the historian Simon of Gent (d.1145).94 He complained about the absence of sources for the eleventh century and clearly only had oral tradition and a handful of charters to go by. Simon tells us that in 1056 there arose a disagreement between the then advocate of Saint-Bertin, Gerbod of Oosterzele-Scheldewindeke, and Abbot Bovo. His story is clearly based on a series of transactions preserved in a charter from 1056 which records an agreement between the abbot and the advocate.95 Thereafter Simon does not once refer to Gundrada’s family, even though he includes a fair amount of Anglo-Norman information. He does not even reveal that Gerbod and the Warennes are linked. Perhaps he did not know that there was such a link. In turn, Simon’s narrative, which ends in 1145, was continued in the late1180s by another (anonymous) Saint-Bertin monk.96 According to his account, the abbey experienced troubled times with estates being lost, advocates being negligent – both those of the Béthune family and the Fauquembergue castellans of Saint- Omer – and the abbey suffering from the see-sawing politics between Count Philip of Flanders and the kings of France, Louis VII (1137–80) and Philip Augustus (1180–1223). The anonymous author, like Simon of Gent before him, had done his historical research in that he continuously cross references lost estates with surviving charters. The Saint-Bertin narrative refers to papal correspondence revealing that from 1178 onwards the monks protested against their abbot, Simon II (1176–87).97 The worst time came in 1182 when Abbot Simon, already ‘weak in religion’ (tepide in religione), fell ill with a fever and argued that only the finest wines and food could keep him alive. Worse still, the prior followed his example by using the same excuse for culinary delights.98 As a result all monastic accounts went into debt, especially those of the cellerar and the monk in charge of the wine supply (the vinetarius). After four years the abbot and prior resigned as a result of the monks’ protest to Bishop Desiderius of Therouanne (1169–91). It is therefore significant that in this time of 91 Orderic iii, 118–19 and 226–31; Orderic’s main informant was the monk Roger of Coulances, son of

William I of Warenne’s sister by Erneis of Coulances. Other members of Roger’s family were benefactors. 92 Above, note 87. 93 Both Giry, Histoire de la ville de Saint-Omer, 130–1, and Feuchère, ‘Les avoués’, 193–207, tend to

argue backwards from late twelfth-century information which shows that advocates (including Hamelin) were involved in transactions within and outside the town that may represent ancient advocacy rights of ownership. This may well be the case but the late evidence cannot prove continuity of an active Warenne connection with Saint-Omer between 1096 and c.1182. Although Derville, Saint-Omer, is silent on the advocacy, he refers interestingly to a ‘motte de Warenne’ (p. 37) as an example of toponymic traces which may be old, i.e. dating from before 1200. I suspect that this reference represents Hamelin’s or his son William’s work rather than eleventh-century activity. 94 O. Holder-Egger, ed., MGH SS 13, 635–63. 95 Ibid., 638–9; ed. Haigneré, i, no. 76. 96 MGH SS 13, 663–73. 97 Copies are in fact included in the Gesta abbatum, 673–4. 98 Ibid., 673.

118

Anglo-Norman Studies XXVI

internal crisis and, more in particular, at the time of Abbot Simon II’s illness at Saint-Bertin, we can place both Earl Hamelin’s charter to the glutinous abbot and Count Philip’s charter confirming the arrangement between Earl Hamelin and the abbey’s tenant Renaud d’Aire. The abbey needed money and protection and Earl Hamelin was prepared to revive old interests. Yet, however neat this solution sounds, it still does not explain how so suddenly out of the blue Hamelin was in a position to strike up friendship with Abbot Simon II and, more importantly, receive the backing of Count Philip of Flanders. International politics combined with Hamelin’s knowledge of the Warenne links with the monastery may explain his presence in the area. In the spring of 1182 Henry II renegotiated relations with Count Philip of Flanders and in particular the exchange of English money fiefs in return for Flemish military support.99 This was nothing new for a series of Anglo-Flemish treaties had been concluded in 1101, 1110 and 1163 and in that year too individual money fiefs had been sanctioned by Henry II and Count Thierry of Flanders.100 In April 1182, however, they coincided with a series of ‘international summits’ at Gournay and la Grange Saint-Arnoult, meetings attended not only by the count of Flanders and Henry II but also by Henry the young king as well as King Philip Augustus (1180–1223). These royal meetings had been hastily organised following the death on 26 March 1182 of the countess of Flanders, Elisabeth, the estranged and childless wife of Philip, and joint heiress of Vermandois.101 A peaceful settlement of the inheritance of Vermandois (with Amiens and Artois) was the issue highest on all participants’ agenda.102 Elisabeth’s sister Eleanor, married to her fourth husband, but still likewise childless claimed the inheritance, as did Count Philip as widower. King Philip Augustus also had an interest.103 It was decided that Eleanor would be accepted as the heir, but that Count Philip should be allowed to take a life interest. After his death Vermandois would revert to Eleanor, who, in case she died childless – which was very likely – would recognise the king of France as her heir. The settlement was agreed and confirmed three years later at the Treaty of Boves and again after Count Philip’s death in 1191. Although none of the records concerning these events refer to Earl Hamelin it is conceivable that he was involved in the negotiations due to the fact that his wife Isabelle was the granddaughter of Isabelle of 99

There is an interesting comment by Gerald of Wales on the endemic fighting and the locals’ animosity against the English when he travelled through the area around Saint-Omer and Cambrai in 1179 (Geraldi Cambrensis iii, 239–40). 100 See above, note 90. 101 For a schematic representation, see Appendix 2. The fundamental narrative for what follows remains A. Cartellieri, Philip II August König von Frankreich, 4 vols, Leipzig–Paris 1899–1922, i, 120–7. See also J. W. Baldwin, The Government of Philip Augustus: Foundations of French Royal Power in the Middle Ages, Berkeley 1986, 80–1, and T. de Hemptinne, ‘Aspects des relations de Philippe Auguste avec la Flandre au temps de Philippe d’Alsace’, in La France de Philippe Auguste: Le temps des mutations. Actes du colloque international organisé par le CNRS (Paris, 29 septembre – 4 octobre 1980), ed. R.-H. Bautier, Paris 1982, 255–62. I have also benefited from L. Napran, ‘Marriage Contracts in the Southern Low Countries and the North of France in the Twelfth Century’ (unpublished PhD thesis, University of Cambridge, 2001), 90–4. The terms of the treaty can be reconstructed on the basis of L. Vanderkindere, ed., La Chronique de Gislebert de Mons, Brussels 1904, 147–9. Dr Napran’s English translation with historical annotation of Gislebert of Mons’ Chronicle is forthcoming from Boydell and Brewer. 102 Le Colonel Borrelli de Serres, La réunion des provinces septentrionales à la couronne par Philippe Auguste: Amiénois-Artois-Vermandois-Valois, Paris 1899, iv–v. 103 The king as overlord had royal rights, but he also had another claim arising from his wife Elisabeth’s dowry. Philip Augustus had married Elisabeth of Hainault, niece of Count Philip. Their marriage contract formally assigned Artois (including Saint-Omer and Aire) to the new queen. However, there is no sign that between 1180 and 1191 the French king administered any rights in Artois; all surviving evidence points to Count Philip as being in control (Borrelli de Serres, La réunion, xxiii, xxv).

The Warenne View of the Past

119

Vermandois, aunt of Elisabeth and Eleanor. His half-brother King Henry II, at that stage the grand old man in western Europe, had been repeatedly asked to be present by envoys sent from France and Flanders.104 As elderly stateman his prestige, advice and meditation carried considerable weight.105 In fact it carried so much weight that the final peace settlement in 1182 was concluded, according to Roger of Howden ‘by the advice and will of the king of England’.106 In the absence of any male Vermandois heir,107 Henry II may have put up Earl Hamelin as a claimant by right of his wife.108 From our perspective, interestingly, Artois was included in the Vermandois settlement because it had been part of Elisabeth of Vermandois’s dowry. The county of Artois had been newly formed around the time of the marriage of Elisabeth and Philip of Flanders in 1159. One of its principal towns was Saint-Omer, and Aire belonged to it too.109 Having the usufruct of his deceased wife’s inheritance, including Artois, Count Philip thus had access not only to the income from the town but also a say over the advocacy of Saint-Bertin. My guess is that on this occasion in April 1182, he was persuaded by King Henry II, with agreement of King Philip Augustus, to hand over the advocacy, and whatever lands or rights the Warennes felt they still could claim, to Earl Hamelin.110 If the contemporary issues were as I have tried to identify them, it remains virtually impossible for us now to determine who revived the Warenne interest in Flanders and what, apart from the advocacy, was exactly at stake. As to the ‘who’, was it Earl Hamelin and Isabelle and their advisers? Was it the cash strapped abbot of Saint-Bertin whose monks had recently put their archive in order, updated their house history and so (re) discovered the Warennes? Or, was it any of the advisers of the main players Count Philip of Flanders, King Philip Augustus, or King Henry II? We will never know for sure who took the initiative. As for the stake, this requires further research into the Anglo-Flemish-French relations at the end of the twelfth century in order to determine precisely the role of the Warennes. In the end we can safely conclude that it was old and new interests, of monks and lay people, of men and women, whose collective memories brought forth an ingenious solution for the troubled times at Saint-Bertin at Saint-Omer in the spring of 1182. After what appears to be an absence of almost a century the Warennes once again held a foothold in Flanders, but whether the wool from the Warenne estates in East Anglia would from now on be woven into Flemish cloth at Saint-Omer is part of the further research that has to be reserved for another occasion. 104 Howden, Gesta i, 244–7, 250, 277 and 283–6 illustrate the cumulative effect of the many appeals

made for Henry II’s mediation from 1180 onwards. I am most grateful to John Gillingham for drawing my attention to Howden’s testimony, and for suggesting to me the international importance of King Henry II’s position in the first place. 105 Howden, Chronica ii, 267, and Diceto ii, 10–11. 106 Howden, Gesta i, 285. Once again, I owe this observation to John Gillingham. 107 The older brother of Elisabeth and Eleanor, Ralph II of Vermandois was a leper who had ruled Vermandois from 1152 to 64 when he handed over control to his sister Elisabeth and her husband Philip of Flanders; he died in 1167 (Borrelli de Serres, La réunion, xi–xii). 108 There is no evidence that any of the Beaumont/Leicester grandchildren of Isabelle of Vermandois (d. after 1138) by her first husband Robert of Beaumont/Leicester (d. 1118) were on the scene. Any presence would be highly unlikely, as John Gillingham pointed out to me, as a consequence of the earl of Leicester’s rebellion against King Henry II in 1173. His fortunes were not fully restored until Richard I’s reign. 109 Giry, La ville de Saint-Omer, 69–72, and Borrelli des Serres, La réunion, iv–v. 110 After Count Philip’s death in 1191, Saint-Omer and Aire were unsuccessfully claimed by his second wife, Matilda of Portugal, as belonging to her dower. The new Count Baldwin VIII (1191–4) kept them (Baldwin, The Government, 81 and 478 note 11). Artois and the other counties devolved to King Philip Augustus, but were given back by him to Baldwin IX of Flanders in January 1200 as a reward for having broken off contact with King John (ibid., 82, 92, 95, and Giry, La ville de Saint-Omer, 70–1).

120

Anglo-Norman Studies XXVI APPENDIX ONE Descent of the Warenne Inheritance

William I of Warenne (d.1088) = 1 Gundrada of Oosterzele-Scheldewindeke (d.1085) = 2 sister of Richard Guet (still alive 1098)

William II of Warenne (d.1138) = [c.1118] Isabelle of Vermandois (d.after 1138) (first wife of Robert de Beaumont d.1118)

William III of Warenne (d.1147/8) = Ala of Bellême/Ponthieu (d. ?1174) (married 1151 Patrick of Salisbury)

Isabella of Warenne (d. c.1203) = [1147/8] 1 William IV (d.1159), son of King Stephen = [1164] 2 Hamelin (d.1202), half-brother of King Henry II (c.1163 marriage plans with William of Anjou, brother of King Henry II)

William V of Warenne (d.1240) = Maud, daughter of William Marshal

The Warenne View of the Past

121

APPENDIX TWO Vermandois Settlement in April 1182 (confirmed in 1185 and 1191) Philip I of France 1060–1108

Hugh = Adela of Vermandois

Ralph I of Vermandois 1117–52 steward of king = 1 Eleanor of Champagne = 2 Petronilla of Aquitaine

Ralph II of Vermandois Elisabeth 1152–64 (d. of leprosy 1167) d. 26/3/1182 = = Margaret of Flanders Philip of Flanders 1168–91 Handed over Vermandois to Philip of Flanders in 1164

Isabelle/Elisabeth of Vermandois d. after 1138 = 1 Robert de Beaumont d.1118 = 2 William II of Warenne d.1138

Eleanor d.1213 = 4 husbands

William III of Warenne d.1147/8

Isabelle = [1164] Hamelin

4 April 1182 at a place between Gerberoy (Oise, arr. Beauvais) and Gournay-en-Bray (Seine-Mar. arr. Neufchatel) (Cartellieri, i, 125 note 1). Present: Count Philip of Flanders, Count Baldwin of Hainault, King Henry II, Henry the young king, King Philip Augustus 11 April 1182 at La Grange St Arnoult, between Senlis and Crépy (Cartellieri, i, 125 note 3). Present: The same people as listed above, plus Archbishop William of Reims, Count Theobald of Blois, Count Stephen of Sancerre, Countess Mary of Troyes, Duke Hugh III of Burgundy, Countess Eleanor of Beaumont-sur-Oise [sister of Elisabeth of Vermandois], Count Ralph of Clermont, Count Ralph of Coucy, Henry of Albano (papal legate) and various unnamed abbots and bishops. Whether or not Earl Hamelin was present at any of these meetings is unknown.

Textual Communities in the English Fenlands

TEXTUAL COMMUNITIES IN THE ENGLISH FENLANDS: A LAY AUDIENCE FOR MONASTIC CHRONICLES?* Jennifer Paxton Introduction In the second half of the twelfth century a number of Benedictine houses in England compiled histories of their own communities. This trend was particularly pronounced in the Fenlands, where three important abbeys – Ely, Peterborough, and Ramsey – produced house-histories at almost exactly the same time. These house chronicles do not seem to have enjoyed wide circulation. The manuscripts of these texts seem to have stayed within the monastic precincts, and there is little or no sign of their being consulted by outsiders. Not unreasonably, most historians have assumed that these texts were intended primarily for the monks themselves, and indeed, house histories were certainly designed to address the internal concerns of the monastic community.1 Nevertheless, a close look at these texts reveals that they include another dimension. They contain evidence of efforts the communities were making in the twelfth century to reach out to the wider lay community, and, if I may push my argument a little bit, I think the chronicles themselves were intended to serve as a repository of knowledge that would equip the brothers (or at least the most prominent among them) to meet the outside world on terms advantageous to themselves. This appeal to the past was intimately linked to a whole campaign of lay outreach that promoted the abbey through the writing of hagiographic texts, the physical reconfiguration of shrines and the active solicitation of donations associated with various forms of monastic services rendered, from supernatural cures to burial within the abbey precincts, and even, in one case, local public works. In this paper I will explore what these texts reveal about how monastic communities were using their own past to shape relationships of patronage and power, and I will suggest that the chronicles may have been intended to serve as the basis for a sort of ‘textual community’ that would draw the surrounding laypeople into the orbit of the monastic house. While they were undoubtedly read by only a few, even of the monks themselves, the audience for the chronicles was in a real sense much wider than that. The many references throughout the chronicles to the ‘reader or hearer’ of the text strongly suggest that they were intended to be read aloud, perhaps in chapter. A particularly good example is the exhortation in the preface to the Liber Benefactorum ecclesiæ Ramesiensis directed to ‘whomever should read or hear the present work’.2 The fact that this phrase occurs in the preface to the chronicle implies that the compiler of the * The author wishes to thank Sarah McNamer, Emily Francomano, the members of the Washington Area Medieval History Colloquium, and the audience at Battle for helpful comments and suggestions. 1 Jennifer Paxton, ‘Charter and Chronicle in Twelfth-Century England: The House-histories of the Fenland Abbeys’ (PhD dissertation, Harvard University, 1999). 2 ‘Universitatem vestram quicunque præsentia legitis aut auditis’. W. Dunn Macray, ed., Chronicon abbatiae Rameseiensis, RS 1886 (hereafter LB), Preface, p. 5.

124

Anglo-Norman Studies XXVI

text expected it to reach beyond the small circle of monks who would normally have had access to such a work, perhaps even to the laity.3 This wider audience for the chronicles accords well with Brian Stock’s model of a ‘textual community’, by which he means a group of people gathered around a text, or an interpreter of a text, who share certain assumptions flowing from that text and are animated by them. Oral performance of the text or of versions of the text could then form an integral part of the group’s activity.4 While Stock applies this concept mainly to small groups of heretics who concentrated on interpreting certain parts of the Bible in a particular way, it is not unreasonable to extend the idea to the monasteries of the Fens. The authors wrote, I would argue, partly to equip their internal audience with the ‘text’ they would need to represent the abbeys’ interests on the outside. One distinctive feature of the Fenland chronicles (though they share it with some others, such as the Abingdon Chronicle), is their seemingly miscellaneous nature. Alongside narrative they also include miracle stories, charters, and other sorts of documents, making them somewhat ‘lumpy’ and even indigestible to the modern reader. In fact, this aspect of their construction is integral to the purpose for which they were compiled, which is to provide warrant for all sorts of ‘evidence’ of the community’s supernatural power: miracles, relics, and even the tombs of holy people. The texts are also designed, of course, to record the property deeds for the abbey’s estates and to provide records for future claims. Thus, they are multivalent texts that satisfied a variety of needs that these communities felt pressing upon them in the mid to late twelfth century. This essay will therefore merely explore the texts from a different angle to see what they tell us about the message the monks were attempting to convey to the outside world. There have been many studies of twelfth-century saints’ cults, of course, but most have focused somewhat narrowly on hagiographic texts alone, particularly on miracle collections.5 Emma Cownie has done a superb study of the Bury cult in the post-Conquest period.6 Other scholars have focused more on what miracle texts can reveal about the sociology of saints’ cults and how monastic communities fit into a local landscape,7 proceeding from the assumption, I think not at all controversial now, that we can read miracle stories as reflections in at least some sense of what was really going on at a shrine.8 What I will do here, however, is to add to an analysis of miracles a study of other kinds of evidence that have not usually been read together

3

Emma Cownie makes a similar argument in her study of the cult of St Edmund. Emma Cownie, ‘The Cult of St Edmund in the Eleventh and Twelfth Centuries: The Language and Communication of a Medieval Saint’s Cult’, Neuphilologische Mitteilungen 99, 1998, 177–97, at 179, 189. 4 Brian Stock, The Implications of Literacy: Written Language and Models of Interpretation in the Eleventh and Twelfth Centuries, Princeton 1983, 88–92. 5 For miracles generally, see Benedicta Ward, Miracles and the Medieval Mind: Theory, Record and Event, 1000–1215, Philadelphia 1987; and Pierre-André Sigal, L’Homme et le miracle dans la France médiévale (XIe et XIIe siècles), Paris 1985. For the relationship between hagiography and historiography with specific regard to miracles, see Pierre-André Sigal, ‘Histoire et hagiographie: les Miracula aux xie et xiie siècles’, Annales de Bretagne et des pays de l’ouest 87, 1980, 237–57. 6 Cownie, ‘Cult of St Edmund’. 7 D. W. Rollason, ‘The Miracles of St Benedict: A Window on Early Medieval France’, in Studies in Medieval History presented to R. H. C. Davis, ed. Henry Mayr-Harting and R. I. Moore, London and Ronceverte 1985, 73–90; Henry Mayr-Harting, ‘Functions of a Twelfth-Century Shrine: The Miracles of St Frideswide’, in Studies in Medieval History presented to R. H. C. Davis, 193–206. Mayr-Harting places the growth of shrines in England in about 1180, when a number of important miracle collections were produced, but Cownie’s work and Hugh Thomas’s article on ‘Miracle Stories and the Violence of King Stephen’s Reign,’ HSJ, forthcoming, shows that this phenomenon should really be dated much earlier. 8 See, for example, Sigal, ‘Histoire et hagiographie’, 237.

Textual Communities in the English Fenlands

125

with miracles, and the reason is, as I just pointed out, that these other kinds of evidence were juxtaposed with the miracles by the twelfth-century monastic authors themselves. I will be dealing with three kinds of evidence: miracles, charters, and tombs. These are not evenly distributed among the three houses, as their chronicles do not follow exactly the same literary form. By far the largest number of miracles comes from Ely,9 and the most extensive notes on burials come from Ramsey, with Peterborough lying in between. First, the miracles.

Miracles The most obvious place to see the interaction between the texts and the outside world is in the miracles contained in the Liber Eliensis and, to a lesser extent, in the chronicle of Hugh Candidus of Peterborough.10 Laymen and laywomen are often the protagonists of these stories, and there is a good representative sampling of men and women of every age and social status. Scholars such as David Rollason and Henry Mayr-Harting have argued that the presence of a large number of laypeople in a miracle collection might indicate that the laity are the intended audience of the text. For example, Rollason suggests that the increase in the number of laypeople who appear in the miracles of St Benedict at Fleury in the twelfth century reflects Fleury’s efforts to penetrate deeper into the lay community: ‘it is difficult to escape the conclusion that [the miracles] were envisaged also as material suitable for preaching to the laity who frequented the shrines’.11 On what sort of occasions would the material included in the texts be transmitted to a lay audience? Again, the miracles provide a clue. Sacred objects, to which stories had become attached, might be used to solicit lay patronage.12 The chronicles often record those stories, and then demonstrate the objects being used to effect cures. The text thus provides a double historical warrant for believing in the efficacy of the objects by describing first how they came to have holy associations and then how they work miracles in more contemporary times. For example, the compiler includes the story of how St Æthelthryth and St Benedict effect the miraculous release from captivity of a man named Bricstan, who upon declaring his intention to enter the monastery of Ely had been falsely accused of stealing from the king. The compiler then describes how the shackles from which he was released were put on display at Ely: ‘They hung the shackles with which he had been held captive before the altar where the people could see them as a memorial to this great miracle in praise of our Lord Jesus Christ, to whom be honor and glory forever and ever.’13 Immediately

9

The cult of the Ely saints in the twelfth century has recently attracted much scholarly attention. See especially Susan Ridyard, The Royal Saints of Anglo-Saxon England: A Study of West Saxon and East Anglian Cults, Cambridge 1988, 50–60, 176–210; Monika Otter, ‘The Temptation of Æthelthryth’, Exemplaria 9, 1997, 139–63; Diana Webb, Pilgrimage in Medieval England, London 2000, 39–41; and Virginia Blanton-Whetsell, ‘Tota integra, tota incorrupta: the Shrine of St Æthelthryth as Symbol of Monastic Identity’, Journal of Medieval and Early Modern Studies 32, 2002, 227–67. 10 The LB is notably lacking in miracles, though the Miracula S. Ivonis (Macray, ed., Chronicon abbatiae Rameseiensis, lix–lxxxiv) may have circulated widely. This text awaits a thorough investigation, a large project in itself, and is thus not discussed here. The text has been translated by S. B. Edgington, The Life and Miracles of St Ivo, St Ives 1985. 11 Rollason, ‘Miracles of St Benedict’, 89. 12 Blanton-Whetsell, ‘Shrine of St Æthelthryth’, 230. 13 ‘Boias autem quibus compeditus fuerat in eadem ecclesia ante altare suspensas quasi memoriale tanti

126

Anglo-Norman Studies XXVI

following this episode, there is the story of the brother of an Ely provost who contracts a terrible toothache. He visits a tooth-drawer in vain, and after storming out in frustration, he remembers the miracle (recounted immediately before in the text) of Bricstan’s liberation from captivity, and this convinces him to go straight to the church of Ely.14 There he applies the shackles with which Bricstan had been held to his aching tooth, and is cured. Now, how did the provost’s brother know the story? The provost may have told him, of course, but he as well as others could presumably be told about Bricstan by one of the monks at the shrine itself. The story of someone close to the church of Ely recalling the miracle of Bricstan and resorting to his shackles for a cure is a powerful incentive for others to do the same. It is also evidence for the kinds of family ties that connected the Ely monks to the local community. While there are no contemporary records to prove the efficacy of Ely propaganda about this relic, the thirteenth-century sacrist rolls do record donations de Boiis, or ‘from the shackles’.15 The dissemination of this miracle in the twelfth century was no doubt the origin of this income recorded in the thirteenth. The second example is even nicer for my purposes, because it gives direct evidence of the kind of preaching to the laity that I am suggesting the monks engaged in.16 This one occurs during the reign of Bishop Nigel (1133–69), where the LE records a miracle in which the tunic of St Æthelthryth is used to heal a sick child. Earlier in the text, the LE has given the history of this relic. It was presumably one of the linteamina of the saint that were found to be uncorrupted when her tomb was opened by her sister Sexburga 16 years after her death, and that worked many cures.17 This is not all, however, for the chronicle also records supernatural proof of the relic’s authenticity. In Book One of the LE, the chronicler gives the story of an archpriest who is punished by the saint for presuming to verify the uncorrupt state of the saint’s vestments. This story serves to warn visitors to Ely to accept the word of the brothers on faith, and also testifies to the truth of the stories they told about the vestments.18 It may tell us something about the task the monks face – they have to authenticate their holdings for their audience, who are presumably laypeople (and other clerics) less invested in believing in the relics than the community is, so they take more convincing. Susan Reynolds has pointed out that ‘miracle stories are full of scoffers . . . who get their comeuppance, but they would not be in the stories if such people had not existed in real life and needed to be converted’.19 The really interesting point about the tunic here is the means by which it was brought in contact with the sick child. In this case, in contrast to the shackles of Bricstan, which stayed in the church, the relic was itself on tour. A certain Brother miraculi ad populi spectaculum posuerunt ad laudem domini nostri Iesu Cristi, cui sit honor et gloria per seculorum secula. Amen.’ E. O. Blake, ed., Liber Eliensis, London 1962 (hereafter LE), bk iii, 266–9. The story is also found in Orderic iii, 346–58. 14 ‘Reducens igitur ad memoriam qualiter beata Æðeldreða paulo ante captivum quendam a compedibus liberavit, qualiter easdem compedes per eius merita gloriosa virtus divina confregerit, quam pia et quam potens in hoc negotio extiterit, ad eius ecclesiam celeri gradu properavit.’ LE, bk iii, 270. 15 LE, bk iii, 269, n. 2, citing F. R. Chapman, ed., Sacrist Rolls of Ely, Cambridge 1907, i, 118–19; ii, 35. The sacrist rolls are discussed in Ben Nilson, Cathedral Shrines of Medieval England, Woodbridge 1998, 154–6. 16 Giles Constable discusses preaching to lay, clerical and mixed audiences in ‘The Language of Preaching in the Twelfth Century’, Viator 25, 1994, 131–52. 17 LE, bk i, 47, in a passage derived from Bede’s Historia ecclesiastica, bk iv, c. 19 (17). 18 LE, bk i, 60–1. See discussions of this episode in Otter, ‘Temptation of St Æthelthryth’, 161, 163; and Blanton-Whetsell, ‘Shrine of St Æthelthryth’, 232–3. 19 Susan Reynolds, ‘Social Mentalities and the Case of Medieval Scepticism’, TRHS 6th ser. 1, 1991, 21–42, at 29.

Textual Communities in the English Fenlands

127

Brithmar had brought it with him to visit the house of William of Flanders in order to raise funds for the upkeep of a bridge by preaching to the faithful; while there, he was able to save a child of the house who had been deathly ill by wrapping him in the tunic of St Æthelthryth.20 One can try to imagine who this William of Flanders must have been. Perhaps he was a locally prominent figure who would gather people together in his home for such an occasion. We do know a tiny bit about who Brother Brithmar was. He is listed in the LE as one of three monks who witnessed an inventory of the treasures of Ely conducted by Bishop Nigel c.114021 and was thus presumably an important member of the community; his death was commemorated in the Ely necrology on 8 December.22 Now, the miracle does not specify the matter of Brother Brithmar’s sermon, but besides stories about the tunic itself, one possible subject can be found a bit earlier in the text, at Book Three, chapter 32, in which St Edmund appeared to a resident of the town of Exning in Suffolk and asked him to hasten to the bishop of Ely (Hervey) and tell him to make a road so that the saint could visit his lady, St Æthelthryth.23 When told of the vision, the bishop rejoiced, and asked many people if this could be done. When no one came forward, a simple monk named John pledged that he could perform the task, with God’s help, and constructed a causeway.24 It is virtually certain that the bridge referred to in the miracle of the child wrapped in the saint’s tunic is the same causeway as the one in the vision.25 It is at least possible, if not positively likely, that the story of St Edmund’s apparition formed part of Brother Brithmar’s sermon in the house of William of Flanders. Perhaps the monk also described the martyrdom of St Edmund at the hands of the Danes; the LE helpfully inserts a passage from Abbo’s Passio S. Edmundi in its treatment of the devastation of East Anglia by the pagans.26 The stories in Book One of the LE about the tunic would have equipped Brother Brithmar to explain the efficacy of the relics he had brought with him, and the stories about the causeway and St Edmund would have provided him with persuasive arguments for contributing to the maintenance fund. Now, Brother Brithmar did not have the chronicle itself available to him, of course, but it would certainly be convenient to bring all the relevant information into one place so that monks charged with a similar fund-raising errand in the future could study their brief. A similar desire to prove the authenticity of sacred objects clearly motivated Hugh Candidus of Peterborough. The most important relic Peterborough possessed was the arm of St Oswald (the Northumbrian king, not the bishop of Worcester), which the abbey had acquired in suspicious circumstances that Hugh passes over largely in silence.27 He does, however, record an incident in which the relics were miraculously 20 ‘Habuit eandem tunicam secum cum aliis reliquiis, eo quod pontis curam ageret et fidelium auxilia

predicando postularet.’ LE, bk iii, 306. 21 LE, bk iii, 289. 22 Jan Gerchow, Die Gedenküberlieferung der Angelsachsen, Berlin 1988, 350. 23 Bury St Edmunds and Ely were closely linked by history. The precise nature of the relationship

between the houses in the twelfth century is, however, far from clear. The LE, bk ii, 155, claims that Bury was founded by monks from Ely, and there is a general tone of condescension or, at best, rivalry towards Bury. See Antonia Gransden, ‘The Legends and Traditions Concerning the Origins of the Abbey of Bury St Edmunds’, EHR 100, 1985, 81–104. 24 LE, bk iii, 265–6. Blake (n. 2) notes that there are records of the cost of maintaining the causeway in Chapman, Sacrist Rolls of Ely i, 112 and ii, 132. 25 H. C. Darby, The Medieval Fenland, Cambridge 1940, 108. 26 LE, bk i, 53. 27 W. T. Mellows, ed., The Chronicle of Hugh Candidus, Oxford 1949 (hereafter HC), 70. Hugh seems to have constructed the list of saints’ resting-places that he includes in his chronicle based on earlier materi-

128

Anglo-Norman Studies XXVI

saved for the community from marauding Danes at the time of Hereward’s revolt, when the prior had to steal them back from the thieves (a classic case of a furtum sacrum – Hugh’s very words – that provides supernatural warrant for relics that the community perhaps had only a dubious right to). Later, Hugh provides proof of the authenticity of the relic by recounting the story of how the arm of St Oswald was displayed to the assembled crowd by Bishop Alexander of Lincoln at the dedication of the newly rebuilt church of Peterborough on the feast of St Peter.28 Abbot Martin wanted to see it, Hugh says, either out of curiosity or because he doubted whether it was intact.29 It was reverently washed, and the water effected many cures.30 Hugh himself claimed to have kissed it and washed it himself on this occasion.31 This story could be used to give a historical context to the holy water that was available at Peterborough and, by raising the very subject of doubt in the relic’s authenticity, to quell it by the authority of the abbot himself. There seems to have been a need to address the issue, since no less a figure than William of Malmesbury had his doubts.32 The point of establishing that a community possessed important relics was, of course, to attract pilgrims. Many of the Ely miracles stress the need for those who wish to be cured to come to Ely in person, presumably because only there could the monks expect them to make substantial donations to the upkeep of the community. These offerings were extremely important to the monks. They constituted the steady if unspectacular lubricant of the monastic economy. A measure of their importance is the indignation that the compiler of the LE expresses when Bishop Nigel uses three years’ worth of pilgrims’ offerings to pay for his hawking and hunting.33 On the other hand, Hugh Candidus notes that after the disastrous reign of Abbot Thurold, the monks of Peterborough used the revenue from their altar to pay the king 300 marks of silver for the right to elect their own abbot.34 Clearly, the shrines of saints were an important part of the monastic income. Pilgrims might be drawn to the shrines by extreme measures. Sometimes they arrive seeking not just healing but relief from the effects of the Ely saints’ chastisements. The LE includes the story of a man who was punished for disregarding his parish priest’s order to observe the feast of St Æthelthryth.35 He suffered terribly until her feast day had come around again, when his people persuaded him to go to Ely.36 Note the importance of persuasion by his friends and family. This was exactly the sort of local peer pressure the Ely monks would have been hoping to foster. The saint proved hardhearted, and he kept a whole day and night’s vigil at her

als, personal researches at neighboring monasteries and ‘written enquiry further afield’. He puts the remaining relics of St Oswald at Bardney, whereas they had been broken up and the trunk moved to Gloucester in 909, but this may have been due to his desire to belittle Gloucester’s contemporary claims. L. A. S. Butler, ‘Two Twelfth-Century Lists of Saints’ Resting Places’, Analecta Bollandiana 105, 1987, 87–103, at 98, 95. David Rollason surveys the evidence for the cult of St Oswald after 1066 in ‘St Oswald in Post-Conquest England’, in Oswald: Northumbrian King to European Saint, ed. Clare Stancliffe and Eric Cambridge, Stamford 1995, 164–77. 28 The year of the dedication is unknown, since Mellows points out that Hugh Candidus’s reckoning puts it at 1130, before Abbot Martin took office. HC, xvii, n. 1. 29 ‘. . . aut per curiositatem aut quia dubitauit illam esse integram.’ HC, 106. 30 HC, 106–7. 31 HC, 52. 32 Webb, Pilgrimage in Medieval England, 38. 33 ‘Insuper oblationes et redditus, totum quod venerat ad altare Sancte Æðeldreðe, per tres annis in aucupibus et venatoribus suis expendidit.’ LE, bk iii, 339. 34 ‘Et pene omnia de altari acceperunt.’ HC, 86. 35 Note the good example set to other clergy, another potential audience for these miracles. 36 LE, bk iii, 305–6.

Textual Communities in the English Fenlands

129

tomb in vain, until on the return journey he made an explicit prayer for forgiveness and promised to keep her feast in the future. One long and extremely odd story concerns a young girl named Reinburgis who is blinded in one eye by order of the saint because she hesitates to follow through on an initial promise to go to Ely to give thanks for relief from a more minor complaint. Her sight is not fully restored until she makes the long journey from her home in Wallingford to Ely, where she has been instructed to keep vigil at a specific spot near the shrine of the holy women.37 If it was advisable to bring pilgrims to the church, it was also evidently useful to be able to demonstrate the long reach of a particular saint. Hugh Candidus claims a universal portability for the water taken from the washing of St Oswald’s arm, noting that it had been carried to St Paul’s in London, where it effected many cures.38 Of course, this water was a renewable resource, well worth publicising. If the relics could travel, then so could the saints themselves, of course. The LE shows St Æthelthryth appearing on the other side of the country, in Shropshire, and interestingly, here again the miracle concerns the need to appear in person at a church dedicated to the saint.39 The texts are thus concerned both with boasting of their saints’ universal powers and requiring that they be sought at the abbey church itself (or another church under the saint’s protection). The miracles discussed so far have dealt, by and large, with the effort to attract pilgrimage and to ensure the observance of the saints’ feasts. These texts were just as important, however, in providing warnings against behavior by laymen (and secular clerics) that could threaten the welfare of the monks more directly than a failure to keep a saint’s day. The Ely compiler cites a number of chastisements of laymen who attempted to get the better of the Ely monks in land transactions. A perjurer who is punished for his crime, and men and women who suffer the saints’ wrath for reneging on land transactions with the abbey served as object lessons for those with whom the abbey had business dealings.40 The Ely saints insisted on the proper respect both of their spiritual powers and their property rights.

Charters While the miracles included in these texts were clearly intended to draw pilgrims, these houses also sought other means of financial support. All three chronicles provide examples of generosity from the past, and I believe that these notices are included not just as a means to secure title to the estates but also to encourage benefactions in the present. In the late Anglo-Saxon period, monasteries such as Ely and Ramsey had benefited from the protection of powerful laymen, such as Æthelwine and Byrhtnoth, although they also had many lay enemies, whom the compilers do not hesitate to criticise. After the Norman Conquest, however, such protectors virtually vanish. The powerful men who arrived in East Anglia immediately after 1066 had no

37 LE, bk iii, 307–12. See my translation of this miracle in ‘The Book of Ely’, in Medieval Hagiography:

An Anthology, ed. Thomas Head, New York and London 2000, 459–94, at 467–73. 38 HC, 108. 39 LE, bk iii, 281–3. 40 LE, bk iii, 363–4, 368–70. Blanton-Whetsell discusses these miracles in ‘The Shrine of St

Æthelthryth’, 256–7. For chastisement miracles in French sources, see Pierre-André Sigal, ‘Un aspect du culte des saints: le châtiment divin aux XIe et XIIe siècles, d’après la littérature hagiographique du Midi de la France’, in La religion populaire en Languedoc du XIIIe à la moitié du XIVe siècle, ed. M.-H. Vicaire, Toulouse 1976, 49–59.

130

Anglo-Norman Studies XXVI

ties to the Fenland abbeys. They gave little land to these monasteries, and thus they probably felt no special kinship with them, at least in the first years of the Norman regime.41 Whereas in the Anglo-Saxon period, the ealdormen felt a special connection to and obligation toward the monasteries to which they were personally tied, their Norman successors, the sheriffs, were notorious for encroaching upon monastic property, although, ironically, they also founded monasteries of their own.42 Gilbert Picot is the most famous example of this dual phenomenon.43 Monasteries often had to rely on their own abbots or on the king himself for protection from rapacious lords. There were signs, however, that monasteries were making deliberate efforts to reach out to the laity, and much was made of the few successful connections forged. Hugh Candidus notes that during the reign of Abbot Ernulf, a certain Robert de Castre came to Peterborough in his infirmity and was made a monk along with his son, in exchange for £50.44 The LB records more lay grants, such as that of Over (Cambs.) by Gilbert son of Ingulf,45 but there are still precious few to set against the benefactions of the pre-Conquest period. It may have been partly an effort to remedy this deficiency that led to one of the most interesting lies in any of the chronicles. The compiler of the LE inserts into the narrative of Bishop Hervey’s reign, which began in 1109, the story of a certain Harscoit Musard, described as a bonus decurio who was greatly favored by the king.46 Harscoit is said to have heard round about the praises of St Æthelthryth as an intercessor for the forgiveness of sins,47 and so to have journeyed to Ely, where he was received with reverence and was greatly impressed by the devotion of the brothers.48 Unfortunately, he fell ill, and sensing death approaching, he asked the monks to receive him into their number, and offered the manor of Aston in exchange, which gift he solemnised with his seal before the clergy and knights of the church.49 Alas, the compiler notes, after the church had held this estate freely for a long time (diu libere), Bishop Hervey took it away and gave it to a member of his family. This story is a complete fabrication, though we cannot be entirely sure when it was invented. It is contradicted by information contained in a document that is included in a thirteenth-century manuscript of the LE (O) but omitted from the earlier of the two contemporary manuscripts of the LE (E) and only added in the margin of the slightly later of the two (F) by a thirteenth-century hand, probably that of O.50 It is also omitted from the twelfth-century cartularies, but was almost certainly preserved 41 Ramsey fared only slightly better than Ely and Peterborough. See Emma Cownie, Religious Patronage

in Anglo-Norman England: 1066–1135, Woodbridge 1998,109–17. 42 Cownie, Religious Patronage, 189. 43 See especially LE, bk ii, 210–11. Ely’s dealings with Picot are discussed as part of the larger phenome-

non of post-Conquest depredations in Edward Miller, The Abbey and Bishopric of Ely: The Social History of an Ecclesiastical Estate from the Tenth Century to the Early Fourteenth Century, Cambridge 1951, repr. 1969, 66–70. 44 HC, 91. 45 LB, 242; see also Cownie, Religious Patronage, 114–17. 46 LE, bk iii, 277. 47 ‘Audierat sane opinionem beate Æðeldreðe undique citra regionum fines cum meritorum preconio maxime propagari.’ LE, bk iii, 277. 48 ‘Ubi et venerando susceptus est obsequio et fratrum valde delectatus devotione.’ LE, bk iii, 277. 49 ‘Donum suum proprio sigillo signavit ecclesie in conspectu clericorum ac militum.’ LE, bk iii, 277. 50 LE, bk ii, 207–8. Blake believes the hand is that of the compiler of a late thirteenth-century version of the LE that Blake designates as O; it is preserved in Oxford, Bodleian Library, MS Laud. Misc. 647. To the scribe of O in the thirteenth century, all urgency had gone out of the matter, and the desire to be comprehensive was paramount. It is unfortunate that Blake has inserted this charter into the main text of his edition rather than placing it in a footnote, since it was clearly important to the Ely community (or at least to the compilers/planners of the text) that it not be there at all.

Textual Communities in the English Fenlands

131

at Ely then, since it was available to the scribe of O a century later. This brief writ records that William Rufus is granting to Bishop Hervey the land at Aston that Harscoit Musard had given to the king. It is dated 1096×1098, probably 1097, and the latest view is that it is completely authentic.51 Thus, the land went to Bishop Hervey a decade before he came to Ely as bishop, and there is no question of its having gone directly from Harscoit to the monks of Ely. The compiler of the LE has suppressed the genuine writ for several reasons. First, he wishes to create a retrospective grievance against Bishop Hervey and stake a claim for the return of the estate to the control of the convent. The utility of this story for reaching out to the laity, though, is that the compiler depicts a pious layman who has heard about the merits of St Æthelthryth and come to Ely from, again, relatively far away (the estate is in Gloucestershire, so it is obviously more plausible for it to have been given to Bishop Hervey while he was bishop of Bangor) to make a significant donation. Each instance that could be cited would presumably increase the likelihood of other donors following suit. Also, the compiler is careful to mention that the gift was confirmed by the donor’s seal, but he cannot, of course, produce any genuine document to prove this. It was important to the Ely monks that they be able to tell the story of this gift and assert that it was warranted by a written instrument, even though they could not, of course, actually show it to anyone. This is evidence, of course, for the growing importance of written documents in English society.52 Besides soliciting donations by example, another purpose of recording documents in the charter-chronicles may have been simply to reassure the laity that any gift they made to the churches would in fact be remembered in those very texts. This is made explicit in the introduction to Part Two of the LB, where the compiler notes that lands that are given in alms are sometimes lost to the church due to the simplicity or negligence of prelates or their subordinates.53 Sons sometimes repudiate their fathers’ gifts. Those who are supposed to defend instead attack. Power conquers, and poverty is oppressed. Yet, the compiler reassures his reader, whatever treachery the sons perform, the merit and the memory due to the father will not be lost: For this reason, all the benefactions that were made to this church from its first foundation have been collected and marked off in this work, both those that were lost and those that were saved or exchanged by public sale, along with the names of the benefactors, which we have inserted, so that even though by some chance they should be lost to us, the memory of the just will nevertheless continue to be praised.54 This is a declaration that the monks of Ramsey will do their duty by any benefactors who choose to favor them; they will use this book to make sure they can keep their promise to remember their patrons. Even when they can do so imperfectly, they will at least make the attempt. For example, after a long series of rather tall tales about the acquisition of estates by Bishop Æthelric, the compiler must rather lamely admit that 51 Richard Sharpe, personal communication, 7 July 2003. 52 Michael Clanchy, From Memory to Written Record: England 1066–1307, 2nd edn, Oxford 1993,

especially 146–9. 53 ‘Sæpe quidem contingit ut possessiones et prædia quæ viris ecclesiasticis et religiosis a sæcularibus in

eleemosynam conferuntur, per simplicitatem aut incuriam tam prælatorum quam subditorum, violentia calumniantium prævalente, ab ecclesiis alienentur.’ LB, 46. 54 ‘Eapropter præsenti hujus operis distinctioni universa quæ memoratæ ecclesiæ a prima ejus fundatione collata sunt beneficia, tam servata quam amissa vel publica cambitione commutata, cum benefactorum nominibus duximus inserenda, quatinus, licet ipsa nobis aliqua perierint occasione, memoria tamen justorum cum laudibus perseveret.’ LB, 47.

132

Anglo-Norman Studies XXVI

he has no information about how the bishop obtained certain others that he names, but he wants to be sure they are set down to the credit of their donor.55 This small admission strongly suggests the value the compiler placed on knowing the history of individual acquisitions. While he states that Æthelric will get ‘credit’ for them, he is clearly distressed that he has no colorful story to tell. There could scarcely be a better argument for preserving the abbey’s records, lest a similar oblivion eclipse the memory of other gifts. Also, by recording the pious circumstances of these gifts, the compiler casts even more obloquy on those who have tampered with the abbey’s lands, or might do so in the future. It is purely speculative, of course, to imagine the Ramsey monks telling potential donors that their gifts will be recalled, just as those of the past have already been, but the monks might have shown them the necrology that they produced c.1150–60, at about the same time as the LB, where many of the gifts described in the LB are listed with the dates on which their donors were commemorated.56 At any rate, the LB and the necrology together testify powerfully to the seriousness with which the house took its obligations to remember the past. The same connection can be seen at Ely as well. Hans Gerchow has pointed out the very close connection between the Ely calendar (which again was compiled at the same time) and the LE. It appears that the necrology was composed with the aid of the chronicle, specifically Book II, which runs from the refoundation of the abbey in 970 down to the elevation of the see to a bishopric in 1109. The entries for lay donors in the necrology record details of the grants they made to the abbey that come right from the chronicle. Gerchow argues, and I agree with him, that the two texts, the historical and the liturgical, were really two sides of the same coin, reinforcing each other’s authority.57 This obligation to commemorate donors may partly explain the inclusion of charters in the body of the chronicles. By the middle of the twelfth century, charters had assumed such an importance that their presence or absence was commented upon, as we have seen. It may be this need to vouch for the presence of the charter that explains its integral inclusion in the text.

Tombs Donors might be attracted to a shrine by hope of a miraculous cure, and then persuaded to donate land by means of a written instrument, but they could also be helpful to the monastic community in one final way: by dying. This is because then as now, charitable bequests were often designed to take effect upon the donor’s death, but in this case, they were often linked also to burial within the abbey precincts. The desire to attract patrons who wished to be buried at the abbey, and who would pay handsomely for the privilege, might explain the strong emphasis on burials of all kinds in all three Fenland chronicles. The monastic authors took pains, I would argue, to provide extensive information on the famous figures who were buried at their respective houses not just to encourage visits to these tombs but also to encourage contemporary men and women to choose burial among such exalted neighbors.

55 ‘Quas ille terras quomodo aut ubi adquisierit nos licet ignoremus, in his tamen commendabile ejus

beneficium utilitate non deficienti experimur.’ LB, 144. 56 Gerchow, Gedenküberlieferung, 342–3. 57 Ibid., 280–9, at 288–9.

Textual Communities in the English Fenlands

133

There was a great emphasis in the Fenland chronicles on the burial of famous local figures and the gifts they brought with them. Indeed, these worthies seem to have been such prizes that there was competition between the various houses for the possession of promising corpses. Hugh Candidus, for example, is eager to note the many famous men who gave both their bodies and their worldly goods to Peterborough, not only earls and wealthy men but especially those bishops and archbishops who paid the church the compliment of choosing it over their own sees, and he claims this fact as proof that Peterborough was much loved and admired.58 Archbishop Ælfric of York, for example, chose burial at Peterborough and donated to the abbey many rich vestments as well as two candelabras, a gold altar with relics and his very own staff.59 Another archbishop of York, Cynesige, an ex-monk of Peterborough who had become estranged from his cathedral community, was also buried at Peterborough;60 he gave the church both land and treasure, but the latter was taken away by Queen Edith.61 Peterborough’s run of luck with the archbishops of York was probably a welcome development for the abbey, because the archbishop who had served just before Ælfric and Cynesige, Wulfstan, was supposed to have been buried at Peterborough but had changed his mind. Hugh claims that Wulfstan had promised himself to Peterborough, but he notes, rather laconically, that he died while on a visit to Ely, so he was buried there.62 Now, in the LE, there is a much fuller discussion of Wulfstan’s career and death.63 After noting his episcopal activities, particularly his services to successive English kings, the compiler says that when he came to Ely to pray, he was met with due reverence by the brothers, and while processing into the church, his episcopal staff suddenly plunged halfway into the earth. Wulfstan took this as a sign that he should be buried there, and thereafter loved Ely, gave it many ornaments, and signed many of Ely’s documents first in the list.64 This is yet another case in which burial, treasures, and land are linked together. When his time came, he ordered his body to be brought to Ely from York and buried where his staff had entered the earth. The compiler then describes how the archbishop’s tomb was moved when the new church at Ely was built, and he was finally reinterred in the new burial site.65 Needless to say, there is no mention of any promise Wulfstan may have made to be buried at Peterborough. The Ely necrology notes in its entry for May 25: ‘Ob(iit) do(m)n(us)

58 ‘In temporibus illis ita carum et amatum erat monasterium sancti Petri de Burch, propter magnam

religionem et caritatem que ibi erat, ut non solum comites et diuites, set etiam archiepiscopi et episcopi relinquerent sedes proprias et se suaque omnia deo et sancto Petro traderent.’ HC, 72. 59 ‘. . . sicut Eleuricus archiepiscopus [de Eboraco] qui dedit albam de purpura cum optimis aurifrisiis paratam et duas capas optimas et stolas et dalmaticam albam, et altare cum reliquiis optime cum auro paratum, et duo pallia et duo magna candelabra de argento que postea furata sunt, et baculum suum, omnia optima.’ HC, 72. 60 Susanne Schäfer, Die Tradition der mittelalterlichen Bischofssepulturen in Canterbury und York, Frankfurt am Main 1996, 271. This study does not mention the Peterborough account of the archbishops’ burial. 61 HC, 73. 62 ‘Set et Wlstanus archiepiscopus se et omnia sua dederat in eodem loco. Set cum isset ad uisitandum sanctorum [loca] et uenisset ad Eli, ibi infirmatus est et mortuus et sepultus.’ HC, p. 73. 63 LE, bk ii,156–7. 64 ‘De quo signo spiritualiter commonitus, futuram ibi sui corporis requiem pluribus audientibus Davitico vaticinio sic prenuntiavit: “Hec requies mea, in seculum seculi hic habitabo [Ps. 131:14].” Unde et locum istum, quoad vixit, vehementer dilexit, ornamentis ditavit et plurima munimenta nostra primus inter primos subscriptione sua roboravit.’ LE, bk ii,156. Blake notes that only one charter, appearing in the LE as bk ii, c. 82, lists Wulfstan first. 65 LE, bk ii, 157.

134

Anglo-Norman Studies XXVI

Wlfstan(us) arch(i)ep(is)c(opus) q(u)i ap(ud) nos iac(et) sepult(us).’66 This is the only time that burial at the abbey is specifically mentioned in the necrology, probably because there had been a dispute about the matter. Ely could boast other impressive burials, and the community seems to have been eager to exploit them in the mid-twelfth century. For example, Book Two of the LE records that Bishop Æthelstan of Elmham (995×997–1001), after giving many gifts of treasure, accepted burial at Ely because he was attracted by the devotion of the brothers and the wonders performed by the saints resting there.67 Ely could boast other famous bishops among its dead. Bishop Ælfwine of Elmham (1021–3×1038), who had been an oblate at Ely, retired to the abbey again at the end of his life and was buried there.68 We have evidence that these burials were still being commemorated in the abbey at the time the chronicle was written. The Ely necrology that is bound with the earliest manuscript of the LE records Bishop Ælfwine’s death on 12 April and Bishop Æthelstan’s death on 7 October; the latter entry also includes a notice of the bishop’s gifts of land and treasure to the abbey.69 A sign of the importance of these tombs is the fact that the monks of Ely translated Bishop Æthelstan from his original resting place to a new tomb among the relics of the patron saints of the abbey during the reign of Bishop Nigel (1133–69).70 This translation may have occurred at the same time as a mass translation of burials, described in the following chapter of the LE, which took place under Prior Alexander as a means of providing access to tombs that were previously hard to find. The LE states explicitly that these translations were performed to honor those who had been vital to the material and spiritual success of the community: When the glorious and most pious king of the English, Stephen, had reigned for a long time, under the direction of Prior Alexander the remains of the venerable men whose gifts caused our church to grow and who helped our community in holy religion were translated into the northern part of the church from their old tombs, which could be found only with great difficulty because they were in an out-of-the-way place, and set up in their own spots with inscriptions clearly marking them off.71

66 Gerchow, Gedenküberlieferung, 346. 67 ‘Crebris sanctarum hic quiescentium virtutibus fratrum religione ad diligendum hunc locum attractus.’

LE, bk ii, 137. 68 LE, bk ii, 155. These burials are extremely significant to Ely, because they stake the church’s claim to

independence from the bishops of Lincoln, who were their diocesan bishops until the erection of the new see at Ely in 1109. The first of these accounts states explicitly that abbots of Ely always made their professions before the bishops of Elmham, not Lincoln, as the Ely archives showed: ‘Quare ob indicium antique dignitatis servate sunt huc usque in scriniis nostris priscorum fratrum scripte professiones que facte sunt in presentia domini Ædelstani Helmanhensis episcopi, nulla vero in presentia alicuius episcopi Lincolniensis.’ LE, bk ii, 137. 69 Gerchow, Gedenküberlieferung, 345, 349. 70 ‘Hunc etiam de antiquo tumulo tempore Nigelli episcopi inter reliquos honorifice transtulimus.’ LE, bk ii, 155. 71 ‘Glorioso itaque rege atque piissimo Anglorum Stephano post multa tempora regnante, in septentrionalem partem ecclesie nostre venerabilium reliquie virorum, quorum beneficiis locus noster adcrevit et quorum cetus noster in sancta religione profecit, cura Alexandri prioris de veteribus sepulcris translate sunt, que, antiquitus in profundo posite et magna difficultate, tandem cum certis signis, invente, singulorum loculos cum scriptionibus acceperunt.’ LE, bk ii, 155–6. According to the antiquarian James Bentham, these graves and their identifying inscriptions were still to be seen in 1769, when building work was done on the choir of the cathedral. James Bentham, History and Antiquities of the Conventual and Cathedral Church of Ely, 2nd edn, Norwich 1812, Addendum, 23–4.

Textual Communities in the English Fenlands

135

Who needed easy access to the tombs? Presumably the brothers themselves, but one suspects they were not the only audience. The information contained in the LE, which was being compiled at the time this translation took place, was no doubt useful to the brothers in their attempt to explain in more detail who was buried in the tombs and why they merited reverence. The emphasis on burials is if anything strongest at Ramsey, where the Ramsey monks also set great store by obtaining the burial of famous bishops. The LB includes the story of Ælfweard, an ex-monk of Ramsey and abbot of Evesham who, by an unusual series of events, ended up being buried at Ramsey. In fact, his burial there seems to be the main reason for including his rather long story. First, the text recounts how he is named bishop of London (while retaining the abbacy of Evesham), and then the compiler tells how Ælfweard and his companions were saved from shipwreck by appealing to St Ecgwine. Ælfweard promises to remedy any defect in the saint’s cult at Evesham and to build a new reliquary to house him in the future.72 Shortly thereafter, the bishop was forced to give up his episcopal duties because of illness, an illness which some believed to have been due to his violation of the tomb of St Osgith at Evesham, in an effort to abstract relics. The compiler is not so sure: ‘But since the causes by which man’s health can be compromised are many, I can by no means confirm changeable opinion about such matters.’73 Ælfweard’s condition caused the monks of Evesham to refuse him refuge, and he returned to his mother house, Ramsey, which as a result received all his ecclesiastical ornaments and liturgical books, as well as the jawbone of Evesham’s patron St Ecgwine (an acquisition that surely both served Evesham right for rejecting the bishop and provided confirmation of Ælfweard’s tendency to violate the tombs of saints) and the cowl of the martyred archbishop Alphege of Canterbury.74 We can catch a glimpse of the Ramsey monks as hosts to pilgrims when the compiler notes that there are still spots of blood on the cowl, which can be seen and kissed.75 Thus, says the compiler, Ramsey did very well out of Evesham’s lack of hospitality.76 It provides a sort of advertisement for potential donors of the reverence with which they might be treated at Ramsey. Of course, turnabout is fair play. In another case, Ramsey lost one of her own great patrons when Bishop Ædnoth I, another Ramsey ex-monk, was buried at Ely, and perhaps the biggest loss of all, another occasion when Ely beat Ramsey to the punch, is the famous episode in which Ramsey’s refusal of hospitality to Ealdorman Byrhtnoth led to his burial at Ely instead.77 72 ‘Siquid hactenus per meam vel vicariæ sollicitudinis incuriam circa reverentiam tui cultus negligentius

actum est, accuratius amodo omnimodæ emendationi operam dabo . . . thecam argenteam in qua venerabiles corporis tui excipiantur reliquiæ operose faciam fabricari.’ LB,149–50. 73 ‘Sed quoniam multiformes sunt accidentium causæ quibus vana hominum corrumpitur salus, nutantem super his opinionem nequaquam confirmamus.’ LB, 157. It is significant that the compiler says that the chronicles are silent about the cause of his illness. The D manuscript of the Anglo-Saxon Chronicle s.a. 1045 [1044] merely records the date of his death and the fact that he had been abbot of Evesham, where he had done much good, before retiring to Ramsey, where he died. G. P. Cubbin, ed., The Anglo-Saxon Chronicle: A Collaborative Edition, Vol. 6: MS D, Cambridge 1996, 67. The Evesham Chronicle and John of Worcester record essentially the same story, but there are no signs of verbal parallels between these two accounts and the LB, though John of Worcester adds that some said the bishop had brought to Ramsey not just items that he had given to Evesham himself but also some given by others. W. D. Macray, ed., Chronicon abbatiae de Evesham ad annum 1418, RS 1863, 85; and John of Worcester ii, 540–1. 74 LB, 157–8. 75 ‘Quas usque hodie et oculis conspicabiles et osculis attrectabiles ostendit.’ LB, 158. 76 ‘Sicque fratres Eveshamiœ, dum infirmo patri mansionem simul et alimenta impudenter denegarunt, nobis plurimum utilitatis propriæ profuerunt dispendio.’ LB, 158. 77 LE, bk ii,133–6; LB, 116–17.

136

Anglo-Norman Studies XXVI

What might explain the emphasis on burials at the abbeys? Burial at a religious house was something of a fashion among the elite, in the late Anglo-Saxon period as at other times. The LB records at least one instance of donors explicitly requesting that they be buried at Ramsey. A pre-Conquest charter of Ærnketel and his wife Wlfrun, from the reign of Æthelred, states that the couple will give to Ramsey land at Hickling and Kilvington (Notts.) after their deaths. They wish the brothers of Ramsey to bring their bodies to the abbey for burial, no matter where they have died, and they offer a sum of money and chattels in exchange.78 There may have been an implied relationship between donation and burial even when no document exists to state it explicitly. For example, there is the story of another donor of both land and treasure, Turgunda, who had promised Ramsey the estate of Sawtry (Hunts.), with the permission of her husband, Thurkill of Harringworth. Upon her death, Thurkill brought her body to Ramsey for burial, and, the text strongly implies, on that very occasion carried through the bequest by offering the estate on the great altar before a large crowd of clerks and laypeople.79 The presence of a large crowd of witnesses, of course, is intended to warrant the bequest and publicise it. The promised burial of a donor at the church may have helped to ensure that the expected legacy was safely transferred to the control of the monastery. There is evidence that by the early twelfth century the abbeys were becoming systematic in their efforts to attract burials. In a paragraph that was incorporated into the thirteenth-century version of Hugh Candidus’s text by Robert of Swaffham as a footnote to his treatment of the reign of Abbot Ernulf (1107–14), there is a notice, probably dating from Ernulf’s reign, which states that Ernulf made an agreement with the knights of Peterborough that any knight who gave two parts of his tithe to the sacristy of Peterborough would, in exchange for a third of his property, receive honorable burial at Peterborough, with the whole convent going out to meet the body, and the deceased and his family would be party to all the benefits of the church.80 Even those who did not pay the tithe could still be buried at Peterborough with the convent performing the offices (perhaps without the solemn procession and confraternity?).81 This agreement suggests that by the early twelfth century, one church at least was actively soliciting burials in exchange for donations. The emphasis in the chronicles on the presence in the churches of many famous figures who had also given lands and treasure to the abbeys at their burials would serve to attract others who would like to be associated with them in their place of rest. There are hints of how this worked in practice. The LB records Henry I’s confirmation of an agreement made between Abbot Aldwin of Ramsey and Reinald de Argentoin about the mill and estate of Icklesford, which Reinald was to hold for his life, with reversion to the abbey. In exchange for 10 shillings, Reinald was also to be buried at Ramsey, if he should die in England.82 Another document preserved in the same text records that Abbot Reginald granted the church of Upwell (Norfolk) to

78 ‘Ubicunque alter nostrum vitam finierit fratres Ramesiæ venientes, cum custamento tam suo quam

amicorum nostrorum, corpus defuncti Ramesiam deferant tumulandum.’ LB, 66–7. 79 LB,175–6. 80 ‘Conuentus autem contra corpus honorifice procedat, et ab omnibus pro eo plenarie celebretur

officium, et ut particeps sit ipse, et uxor eius et filii eorum, omnibus beneficiis que imperpetuum (sic) deo annuente in ecclesia sancti Petri fient.’ HC, 90. 81 HC, 90–1. 82 ‘Et si Reinaldus morietur in Anglia sepelietur in prœdicta abbathia.’ LB, 216.

Textual Communities in the English Fenlands

137

Folcard, in exchange for becoming the abbot’s liege man. His parents will be buried at Ramsey, and will give the church whatever they have in money at that time.83 Burial was thus linked with patronage. It behooved the monasteries to advertise that they were the site of famous tombs, and that burial in their precincts had in the past and would in the future go hand in hand with donations of land and treasure. While many of these burials, such as Reinald’s, would have been far humbler than those of the renowned eleventh-century bishops, the monasteries were constrained to work with what was available to them. They used their past glory to help provide their present daily bread.

Conclusion In conclusion, then, the charter-chronicles were partly designed to pull together all the information the brothers of the house would need in order to solicit the patronage of pilgrims and other lay donors as well as to discourage attacks on their resources by both miraculous and legal means. They needed to put the various assets of the monastery into a historical context that would both warrant their association with the monastery and suggest to potential donors that they might expect a similar commemoration. Only by truly understanding the history of their own house could the monks adequately act as guardians of the legacy left to them by their benefactors and represent that past to the benefactors of the future. Now, the missing piece of the puzzle here is the lay reaction. We can catch only tantalising glimpses of how successful the monks were in using their past to create links to the surrounding community. But it is significant in itself, I believe, to suggest that the Fenland chronicles are not just directed at a narrow audience of monks with a partisan interest in glorifying their past. They do not represent merely records of title or efforts to shore up internal morale. They were also part of an effort to open a dialogue with the surrounding community that involved both legal and supernatural forms of intimidation as well as appeals to vanity, low humor, and simple curiosity. For example, the LE includes the story of a pagan Dane who bored a hole in the tomb of St Æthelthryth, thinking to find treasure inside.84 This hole, the compiler notes, is still visible.85 One can easily imagine a brother of the church of Ely displaying the hole to visitors and telling the story. For all that these houses advertised the inaccessibility of their locations, they were very much part of the landscape of the Fens.

83 ‘Pater autem ejus et mater quum de hoc seculo migraverint, corpora eorum debent ferri ad

Ramesiensem ecclesiam, cum ea parte pecuniæ quæ tunc temporis pertinebit, et ibi sepeliri.’ LB, 256. 84 LE, bk i, 55–6. This episode is discussed at length in Otter, ‘Temptation of St Æthelthryth’. 85 ‘Usque manens cernitur.’ LE, bk i, 56.

1088 – William II and the Rebels

1088 – WILLIAM II AND THE REBELS* Richard Sharpe William Rufus celebrated Christmas for the first time as king at London.1 No diploma from that occasion has preserved the names of the great men who attended, though Henry of Huntingdon mentions the names of the bishops present.2 His most probable source may have been a diploma for Bishop Remigius of Lincoln, now lost but accessible to Henry at Lincoln. Last of those named (apart from Remigius himself, who would not have witnessed such a diploma but must be counted as Henry’s addition) was Odo, bishop of Bayeux. William Rufus had restored him to his earldom of Kent, and he is addressed as earl in one surviving writ, appointing Wido as abbot of St Augustine’s abbey, Canterbury, in the closing months of 1087.3 A source from Canterbury tells us that Odo was present in Canterbury when Abbot Wido was installed by Archbishop Lanfranc.4 If the much later writer William Thorne has correctly given the date as 21 December, they may have travelled from there to London together for the king’s Christmas court.5 After his coronation William of Malmesbury tells us that ‘for the rest of the winter’ King William ‘enjoyed peace and popularity’,6 though none of the national chroniclers mentions that in January or February the king, ‘surrounded by a great multitude of the great men of the palace’, travelled north to York, something that must be gathered from Stephen of Whitby’s account of the founding of St Mary’s abbey. Stephen names Archbishop Thomas of York, Bishop Odo, Bishop Geoffrey of Coutances, Bishop William of Durham, Count Alan of Richmond and Count Odo of Champagne (both

* I am grateful to colleagues working with me on the project to collect and edit the charters of Henry I for their reading drafts of this paper: Nicholas Karn, Hugh Doherty, and Mark Hagger. 1 ASC, 166. 2 Huntingdon, 408. Those present were Archbishops Lanfranc and Thomas, Bishops Maurice of London, Walkelin of Winchester, ‘Galfridus’ (an error for Osbern, presumably guessed expansion from a misread initial) of Exeter, ‘Wlnod’ (for Wulfstan) of Worcester, William of Thetford, Robert of Chester, William of Durham, Odo of Bayeux, and Remigius of Lincoln. 3 Regesta i, no. 304. 4 Acta Lanfranci, ed. J. Earle and C. Plummer, Two of the Saxon Chronicles Parallel, Oxford 1892–9, i, 290; translated in EHD, 679. 5 William Thorne, Chronica, ed. R. Twysden, Historiae Anglicanae scriptores X, London 1652, col. 1793, dates the blessing of Abbot Wido to the feast of St Thomas the Apostle, 21 December (which was a Tuesday). The Acta Lanfranci, without this precise date but linking the blessing to the consecration at Canterbury of Bishop Godfrey of Chichester and Bishop John of Wells (elsewhere dated to July 1088), says that on the next day Odo was present when Lanfranc installed Abbot Wido at St Augustine’s. F. Barlow, William Rufus, London 1983, 67 and n. 66, inferred that the installation took place on 22 December, but the Acta goes on to describe the monks’ rebellion and Lanfranc’s actions, after which ‘Lanfranc returned home’. One must wonder how much of this could have happened on the eve of Christmas with the king. Wido was certainly installed before July, because he was sent as an envoy from the king to Bishop William in Durham (De iniusta uexatione Willelmi episcopi, ed. H. S. Offler, Camden Society 5th ser. 10, 1997, 76; English Lawsuits, 92). 6 Gesta Regum i, 544.

140

Anglo-Norman Studies XXVI

with large territorial interests in Yorkshire), William de Warenne, and Henry de Beaumont.7 Trouble began in the early spring of 1088. By then Bishop Odo was plotting rebellion against the king, and Bishop Geoffrey and Bishop William were in on the plot. From Easter or earlier Odo of Bayeux was at the centre of a widely distributed rising by many of the leading nobility, even perhaps receiving oaths of fealty from some of them on behalf of the king’s brother, Duke Robert,8 who was to have invaded England and added the kingdom to the duchy he had inherited. There is reason to regret that Stephen did not give a more complete list of the laymen present. The principal primary source for our knowledge of the events of the spring and summer of 1088 is the Anglo-Saxon Chronicle kept at that time at St Augustine’s abbey in Canterbury. William of Malmesbury, John of Worcester, and Henry of Huntingdon provide accounts, which elaborate in different ways on what the Chronicle says; John’s contains the most detail about some parts of the campaign, but it leaves King William besieging his uncles Odo and Robert at Pevensey and never completes the story. While broadly following the Chronicle’s account, Orderic Vitalis stands a little apart from these sources, adding further information as well as his usual vivid drama; he begins with a clear discrepancy, indicating that the plot was hatched in Normandy between Bishop Odo, Eustace of Boulogne, Robert of Bellesme, and many others, and suggesting that these magnates crossed to England after Christmas.9 Since Henry of Huntingdon’s more circumstantial statement has Odo in London with the king at Christmas, we can probably infer that Orderic was resorting to conjecture for the obscure early stages of the conspiracy. When so many of the leading magnates were involved, we might assume that the planning required a good deal of communication – unless it was done while many of them were at court for Christmas or in the course of the king’s northern journey. The events have been much discussed, most recently by Frank Barlow, and what remains at issue is questions of detail.10 It will suit our purposes here to quote the Canterbury chronicler’s spare account in modern English:11

7

The form of words includes comital titles for the last two: ‘Interfuerunt etiam fundationi nostre ecclesie multi ex primoribus palatii quorum ista sunt nomina Thomas archiepiscopus, Odo Baiocensis episcopus ipsius regis patruus, Gosfridus Constantiensis episcopus qui eo quoque tempore Northahimbrorum consulatum regebat, Willelmus Dunelmensis episcopus, comes Alanus, Odo comes de Campania, Willelmus comes de Warenna, Henricus comes de Bello Monte, et multi alii nobiles quos hic inserere longum est’; it is likely that this is a minor anachronism on the part of the writer. Stephen’s text was printed in Monasticon iii, 544–6, from an inferior copy, Bodleian Library, MS Bodley 39; an improved text based on the twelfth-century copy, BL, MS Add. 38816, has been prepared by Dr N. E. Karn, to whom I am grateful for letting me use it. The king’s visit to York was first noted by E. C. Norton, ‘The Buildings of St Mary’s Abbey, York, and their Destruction’, Antiquaries Journal 74, 1994, 256–88 (at pp. 280–1), and again by J. E. Burton, The Monastic Order in Yorkshire, Cambridge 1999, 40–1. 8 This is Barlow’s inference, William Rufus, 71 and n. 82, from De iniusta uexatione; he cites pp. 174, 189, in Arnold’s edition (below, n. 28). In the first passage, Bishop William protests that ‘nulli fiduciam uel sacramentum feci uel ab aliquo recepi’, and in the second, ‘nulli fiduciam feci uel ab aliquo recepi’ (ed. Offler, 76, 91); one must take a guess as to what promise or oath he felt accused of giving or taking, and a conspiratorial commitment is the likely interpretation. Douglas’s translation obscures the evidence (‘I have neither made nor received any pledge’, ‘nor have I ever associated myself with any scheme’, EHD, 655, 664); Van Caenegem is clearer (‘I have made or received no promise or oath to the detriment of your body, lands, or honour’, ‘I have given or received no surety thereon to anyone’, English Lawsuits, 93, 101). 9 Orderic iv, 120–34; for their crossing to England ‘post natale Domini’, p. 124. Orderic would himself treat these three as the main losers from the plot (Orderic v, 208), and hindsight may be at work. 10 E. A. Freeman, The Reign of William Rufus, Oxford 1882, i, 22–89, and ii, 465–83; J. H. Ramsay, The Foundations of England, London 1898, ii, 157–62; C. W. David, Robert Curthose, Duke of Normandy, Cambridge, MA, 1920, 44–52; Barlow, William Rufus, 69–93. 11 ASC, 166–8.

1088 – William II and the Rebels

141

In this year this country was very much disturbed and filled with great treachery, so that the most powerful Frenchmen who were in this country intended to betray their lord the king and to have as king his brother Robert, who was count of Normandy. At the head of this plot was Bishop Odo, with Bishop Geoffrey, and Bishop William of Durham. The king treated the bishop so well that all England went by his counsel and did exactly as he wished; and he thought to treat him just as Judas Iscariot did our Lord; . . . The identity of the bishop in this sentence is unclear. Usage in the rest of the annal makes one suppose that the bishop of Bayeux was intended, and that is how Henry of Huntingdon takes it, though there is no other evidence for Odo’s prominence in the king’s counsels.12 Word-order points to the bishop of Durham, who played a greater role in guiding the king but even so deserted him. Certainly William of Malmesbury interpreted it as referring to Bishop William, adding that ‘Odo saw that he was no longer as in former times to have everything in the country all his own way, the conduct of public business having been entrusted to William, bishop of Durham; and in his jealousy he himself deserted the king and infected many others by the same insinuations’.13 John of Worcester too identifies the bishop as William, ‘At this time the king (who knew him well) relied on Bishop William’s wisdom as a true counsellor, and the affairs of all England were managed by his advice.’14 The Canterbury writer goes on: . . . and Earl Roger was also in this conspiracy, and a very great number of people with them, all Frenchmen, and this conspiracy was plotted during Lent. As soon as Easter was reached, they marched and ravaged and burned and laid waste the king’s demesnes, and they ruined the lands of all those men who were in allegiance to the king. And each of them went to his castle and manned it and provisioned it as best he could. Bishop Geoffrey and Robert de Mowbray went to Bristol and ravaged it and carried the plunder to the castle, and then went out of the castle and ravaged Bath and all the surrounding area, and laid waste all the district of Berkeley. Those who were the chief men of Hereford and all the shire with them and the men of Shropshire with a large force from Wales came and ravaged and burned Worcestershire until they came to Worcester itself, and intended to burn the town and plunder the monastery and get the king’s castle by force into their hands. Seeing these things the reverend Bishop Wulfstan was much distressed in mind because the castle had been committed to him to hold; nevertheless, the members of his household marched out with a few men from the castle and, through God’s mercy and the bishop’s merits, killed and captured five hundred men and routed all the rest. The bishop of Durham did whatever damage he could everywhere in the north. One of them was called Roger, who threw himself into the castle of Norwich and did always the worst of all throughout the country. There was also one Hugh who did not mend matters at all, neither in Leicestershire nor Northampton. Bishop Odo, with whom all these affairs originated, went to Kent to his earldom and injured it severely, and

12 The modern translators of the Chronicle thought it worth adding a note to the effect that Odo is meant.

This was the sense put on the sentence by Huntingdon, 412–13, ‘Odo, bishop of Bayeux, the chief man and governor of England’; discussed by Barlow, William Rufus, 61 n. 35. 13 Gesta Regum i, 544. 14 John of Worcester iii, 48.

142

Anglo-Norman Studies XXVI

they utterly laid waste the king’s land and the archbishop’s, and he carried all the goods into his castle at Rochester. When the king understood all these matters and what treachery they were committing against him, he grew much disturbed in mind. Then he sent for Englishmen and explained his need to them and asked for their help, and promised them the best law that there had ever been in this country, and forbade every unjust tax, and granted the people their woods and hunting rights – but it did not last any time. But nevertheless the Englishmen came to the help of their liege lord the king. They marched towards Rochester and intended to capture Bishop Odo – they thought that if they had the one who had been the head of the plot they could the better get hold of all the rest. Then they came to the castle at Tonbridge. Odo’s soldiers were then in the castle and many others who meant to support him against the king. But the Englishmen proceeded to storm the castle, and the men in it made truce with the king. The king with his army marched towards Rochester, and they thought the bishop was there, but it became known to the king that the bishop had gone to the castle at Pevensey. And the king went in pursuit with his army and besieged the castle with a very large army for a full six weeks. Meanwhile the count of Normandy, Robert, the king’s brother, collected a very large force and thought to conquer England with the help of the men who were opposed to the king in this country. And he sent some of his men to this country and meant to follow himself, but the English who guarded the sea captured some of his men and killed and drowned more than anyone could count. Afterwards their food failed in the castle; then they asked for a truce and rendered it up to the king, and the bishop swore he would go out of England and never more come into this country unless the king sent for him and that he would render up the castle at Rochester. Just as the bishop went and was to render up the castle, and as the king sent his men with them, the men who were in the castle rose up and seized the bishop and the king’s men and put them in prison. In the castle were very good knights – the young Eustace and Earl Roger’s three sons and all the highest born men in this country or Normandy. When the king understood these matters, he followed with the army that he had there, and sent over all England and ordered that everyone who was not a scoundrel should come to him, French and English, from town and country. Then a great company came to him and went to Rochester and besieged the castle until those inside made a truce and rendered up the castle. Bishop Odo with the men who were in the castle went overseas, and the bishop thus relinquished the dignity that he had in this country. Then the king sent an army to Durham and had seige laid to the castle; and the bishop made a truce and rendered up the castle and relinquished his bishopric and went to Normandy. Also many Frenchmen relinquished their lands and went overseas, and the king gave their lands to the men who were loyal to him. Taking a little gloss on the names from the Latin historians, we have here a clear indication of the personnel of the rebellion.15 Odo, bishop of Bayeux and earl of Kent, was the ring-leader in England, Robert Curthose the intended beneficiary in Normandy. John of Worcester adds the name of Odo’s brother, Robert, count of 15 The versions will be found in John of Worcester iii, 48–57; Gesta Regum i, 544–9; Huntingdon,

412–15. The account in the Durham Historia regum, in Symeonis Monachi Opera Omnia, 2 vols, ed. T. Arnold, RS 1882–5, ii, 214–17, is similar; it is in part copied from John of Worcester, but it includes information from the Anglo-Saxon Chronicle omitted by John.

1088 – William II and the Rebels

143

Mortain. Roger de Montgomery, earl of Shrewsbury, was privy to the conspiracy, but it is only from John of Worcester that we learn that he was in his castle at Arundel, waiting for Duke Robert to cross the channel. Bishop Geoffrey of Coutances and his nephew Robert de Mowbray were active in Gloucestershire and Somerset. John of Worcester says that Robert failed to capture Gloucester itself, and he adds the name of William of Eu as attacking Berkeley. And only John mentions that Bernard de Neufmarché, Roger de Lassy, Ralf de Mortemer, together with men of Earl Roger of Shrewsbury, led the attack on the towns of Hereford and Worcester. One manuscript of the Worcester version adds that Robert fitz Baldwin rebelled in Exeter.16 William of Saint-Calais, bishop of Durham, was not with King William when rebellion broke out but he was present when the king began to take steps to deal with the situation. At that point he deserted the king who had trusted him. Roger Bigod in East Anglia had been a constant courtier of William I; his rebellion now is further attested by a source from Bury St Edmunds.17 The active part taken by Hugh of Grandmesnil in Leicestershire is less clear. In Kent itself, apart from Odo, there was Gilbert of Tonbridge at Tonbridge castle, whose name enters John of Worcester’s and Henry of Huntingdon’s accounts; John adds that Gilbert was wounded and yielded himself and his castle after a siege of only two days. And further south at Pevensey John of Worcester adds the name of Robert, count of Mortain, Odo’s brother; his name is corrupted to ‘Rogerum consulem’ by Henry of Huntingdon. In the final conflict at Rochester, the town was defended by Eustace III, count of Boulogne, and three sons of Roger of Montgomery, usually identified as Robert de Bellesme (he at least is named by John of Worcester), Hugh de Montgomery, and (it has been both presumed and denied) Roger of Poitou. The spin is clear. The great men from Normandy were all disloyal, the English stood by the new king. Of those leading figures who did not desert William, only Bishop Wulfstan is mentioned. Orderic is the one source to name several other loyal magnates:18 Hugh of Avranches, earl of Chester, William de Warenne, and, more surprisingly, Robert de Mowbray, whom the Anglo-Saxon Chronicle and John of Worcester named among the rebels. Orderic also mentions Robert fitz Haimo, son of the sheriff of Kent, as a loyal courtier, while John of Worcester is the only chronicler to mention that Archbishop Lanfranc and the most of the optimates of Kent stood by the king. These are hardly the Englishmen of the Chronicler’s slanted perspective, and historians can take a more detached view of how the Anglo-Norman aristocracy took sides in this crisis over their allegiances to two princes. A comparison with Domesday Book shows that six of the ten wealthiest magnates in England favoured rebellion and the cause of Duke Robert: Bishop Odo, Count Robert of Mortain, Roger de Montgomery, Gilbert fitz Richard of Tonbridge, Bishop Geoffrey, and Count Eustace of Boulogne. Three stood by the king: William de Warenne, Count Alan of Richmond, and Earl Hugh of Chester.19 The missing name from the top ten is Geoffrey de Mandeville, whose position is not clear. Some other prominent figures remained invisible in Normandy, among them Roger de Beaumont and his elder son

16 John of Worcester iii, 313. 17 ‘Hermannus archidiaconus’ (identified by A. Gransden as Bertram the Archdeacon), Miracula S.

Eadmundi, ed. F. Liebermann, Anglo-Normannische Geschichtsquellen, Strassburg 1879, 202–81 (at pp. 268–9); ed. T. Arnold, Memorials of St Edmund’s Abbey, RS 1890–96, i, 26–92 (§ 39, pp. 79–80). 18 Orderic iv, 128. 19 C. W. Hollister, ‘Magnates and curiales in Early Norman England’, Viator 4, 1973, reprinted in his Monarchy, Magnates, and Institutions in the Anglo-Norman World, London 1986, 97–115 (at pp. 99–100). Hollister overlooked the evidence for Count Alan’s loyalty in De iniusta uexatione.

144

Anglo-Norman Studies XXVI

Robert of Meulan.20 For an overview of the court of William I that had been divided in 1087, one can hardly improve on the picture presented by the list of witnesses to the second dedication charter of St Stephen’s at Caen from the 1080s.21 There are three figures whose roles appear ambiguous in this bald summary. First, it is not apparent why Orderic alone thought that Robert de Mowbray, earl of Northumbria, adhered to the king when other sources depict him as in rebellion with his uncle Bishop Geoffrey. This is probably Orderic’s mistake. Orderic also mentions that Robert of Rhuddlan was fighting alongside his uncle Hugh de Grandmesnil in the midlands, but Robert’s lord was Earl Hugh of Chester, and Orderic may have made a bad guess; elsewhere he suggests that Robert had actually taken part in the siege at Rochester.22 Second, and more interestingly, Roger de Montgomery, earl of Shrewsbury, was at Arundel in Sussex when the rebellion broke out after Easter, according to John of Worcester; his men took part in the fighting in Worcestershire, and, as the Anglo-Saxon Chronicle puts it, he was ‘in this conspiracy’. William of Malmesbury has him and his men fighting at Worcester, but a few lines later he contradicts himself, saying that Roger had not made public his complicity in the conspiracy but was still attending on the king, in circumstances where William was able to talk him over to the royal cause.23 It looks as if Roger’s personal involvement at Worcester was a careless assumption on the part of William of Malmesbury. If Earl Roger was keeping a low profile at Arundel, he could hardly have avoided attending on the king as he rode south from Tonbridge to Pevensey; William of Malmesbury implies that he was with the king even earlier. It is possible that he had joined the king already when the siege of Pevensey began, or that he came over during the six weeks of the siege. In Orderic’s narrative, he was with the king’s forces at Rochester, where his sons were besieged, and even secretly tried to help the defenders.24 Of the besieged sons who had arrived from Normandy with Count Eustace of Boulogne, only Robert de Bellesme is named, though the Chronicle says ‘three sons’. It has been assumed since Freeman’s work that the others were Hugh de Montgomery and Roger of Poitou.25 Our third ambiguous figure is this Roger of Poitou, who was 20 Robert’s whereabouts in 1088 are unknown, but the prominence of his younger brother Henry de

Beaumont on the king’s side would surely have attracted the attention of chroniclers to Robert if he too had been in England. Orderic iv, 204, refers to his arrival in Normandy from England, apparently in 1090, and S. N. Vaughn, Anselm of Bec and Robert of Meulan, Berkeley, CA, 1987, 98–9, infers from this that he was in England in 1088, but the lapse of time is too great for this to be decisive. 21 Bates, Regesta, 258–63 (no. 54). An anomaly compromises the exact validity of this list, since it was a renewal; Bishop Hugh of Lisieux was already dead, and others may have been, though none of the following, whose names appear in the discussion here: Duke Robert and King William as the king’s sons, Archbishop Lanfranc, Bishop Odo, Bishop Geoffrey, Roger de Montgomery, Robert of Mortain, Roger de Beaumont and his sons Robert of Meulan and Henry de Beaumont, Walter Giffard, Earl Hugh of Chester, Robert de Bellesme, Hugh de Montfort, Hugh de Grandmesnil, Bishop William of Durham, Count Alan, William de Warenne, and Henry de Ferrers. 22 Orderic iv, 124; Barlow, William Rufus, 81 and n. 140, draws attention to the fact that a few pages later, iv, 136, Orderic refers to Robert of Rhuddlan, ‘de obsidione Rofensi rediens’, as if he had taken part in the siege. 23 Gesta Regum, § 306. 4 (Roger attacks Worcester), § 306. 5–6 (riding with the king, he is talked round). 24 Orderic iv, 126–8. 25 Freeman, William Rufus i, 57 n. 3, 93. J. F. A. Mason, ‘Roger de Montgomery and his sons (1067– 1102)’, TRHS 5th ser. 13, 1963, 1–28 (at p. 16) warns against this, but, as Freeman himself observed, of the younger sons from his first marriage (Orderic iii, 138) Philip had no English interests and Arnulf is not known to have taken any active role until 1093. Ramsay, Foundations ii, 158, says evasively that Roger ‘kept pretty loyal’, and Barlow, William Rufus, 77, hedges his bets behind opaque punctuation, ‘and three sons of Roger de Montgomery – Robert of Bellême, and two of the younger sons, Hugh of Montgomery, Roger of Poitou – and Arnulf’, though ibid., 91, he more persuasively explains Roger’s wobbly allegiance and his recovery of his English estates.

1088 – William II and the Rebels

145

certainly acting on behalf of the king in September 1088, when he was sent north with Count Alan of Brittany and Count Odo of Champagne to bring the bishop of Durham to the king.26 There is no precise evidence for his having taken any part in the rebellion with his brothers, but he was certainly in England earlier in the summer, since we shall see that he witnessed more than one royal diploma. It is perhaps more plausible that he joined the king and his father after the siege of Rochester was over than that he was with the king all summer.27 Apart from the Chronicle and its translators, there is one other near-contemporary account, which provides an important side-light on the main narrative, the Libellus de iniusta uexatione Willelmi episcopi.28 This allows an insight into the bishop of Durham’s role, which was rather more complicated than appears from the Chronicle, and adds the names of others loyal to the king, Count Alan who held the honour of Richmond and Count Odo, kinsman of the count of Champagne. In spite of the detail available from these sources, it is not possible to establish a definite timetable of events. ‘An exact timetable’, as Barlow observed, ‘cannot be constructed from De iniusta uexatione.’29 There is one key problem, which makes it impossible to harmonize this text with the chroniclers: the crucial charge against the bishop, put by an eye-witness, ‘H. de Bellomonte’, presumably Henry de Beaumont, was that, when rebellion broke out, and the king began to organize himself to fight Bishop Odo and Earl Roger de Montgomery, the bishop of Durham left his court and deserted him.30 If the rebellion did not get underway until Easter, 16 April 1088, it is impossible to square this with the explicit date in the libellus for the king’s ordering the disseisin of Bishop William, 12 March. The bishop’s claims to have defended Dover, Hastings, and London for the king before he left the court appear to push the outbreak of hostilities to a date so early as to conflict with all other evidence.31 The manuscript tradition has a tendency to make mistakes at crucial points; if we boldly emend quarto Idus Martii (Sunday, 12 March) to quarto Idus Maii (Friday, 12 May), 26 De iniusta uexatione, ed. Offler, 79–80; English Lawsuits, 95. Roger acted uice regis ‘on the king’s

behalf’ in receiving the oath of seven of the bishop’s men. 27 C. P. Lewis, ‘The King and Eye: a Study in Anglo-Norman Politics’, EHR 104, 1989, 569–89, at

pp. 572–3, 575–6, reads back from Roger’s part in escorting Bishop William from Durham to guessing that he may have been in the party escorting the captured Bishop Odo from Pevensey to Rochester when Odo was rescued and his escort taken prisoner at Rochester. He goes against the evidence that the king turned rebels into followers by assuming that ‘one of the Rochester rebels would not have been given such a commission so soon afterwards’ (p. 575). 28 De iniusta uexatione, ed. Offler, 49–104; translated (from Symeonis Opera, ed. Arnold, i, 170–95) in EHD, 652–69 (no. 84), and in English Lawsuits, 90–106 (no. 134). 29 Some of the difficulty is encapsulated in his dating the fall of Rochester to the end of May in one discussion and July in another: quotation from William Rufus, 83n, where he favoured July; compare Barlow, The English Church 1066–1154, London 1979, 282, ‘at the end of May Pevensey and Rochester surrendered’. 30 De iniusta uexatione, ed. Offler, 83–4; English Lawsuits, 97. The name of the accuser is five times given as Hugo de Bellomonte or just Hugo; Barlow, William Rufus, 77 n. 111, proposed that this was an error for Henry, a suggestion I find persuasive. Offler, 84 n. 45, denied the possibility of five-fold error, adding that Henry would have been accorded his recent comital title, and proposed instead Hugo de Bellocampo. In favour of Barlow’s conjecture is the clear evidence that the writer (as in so many royal acts) used initials for baptismal names, some of which remain unexpanded (‘G. Constantiensis episcopus’, ‘T. Eboracensis archiepiscopus’, ‘W. de Merlaio’, pp. 84–5, 93), while others were incorrectly expanded (‘Rogerum Paganellum’ for Ralf Paynell, p. 82 and n. 37; ‘Radulfus Piperellus’ for Ranulf Peverell, p. 89 and n. 63; ‘Reginaldus Paganellus’ for Ralf Paynell, p. 93 and n. 75; ‘Rogero de Molbraio’ for Robert, pp. 95–6 and n. 91). The copyist of the archetype of our extant witnesses did not make five mistakes but consistently took H. for Hugo. 31 Barlow, William Rufus, 74–6, implausibly links this with a miracle story from Fécamp as evidence for ‘an early outbreak of hostilities’.

146

Anglo-Norman Studies XXVI

the difficulties vanish, and we are spared the smaller problem that the disseisin would have been ordered on a Sunday. We can now propose a timetable of events that keeps the sources in harmony. The chroniclers indicate that the rebellion broke out at Easter, when the rebels, instead of attending the king’s court, defended their castles and commenced attacks on the royal demesne. We do not have evidence for where the king spent Easter, but a depleted court may have been his first inkling of the outbreak.32 At that date Bishop William appears to have been at London and communicated with the king, who was very likely at Winchester where it was usual to observe Easter. He later claimed to have helped preserve for the king the castles of Dover and Hastings in the troubled counties of Kent and Sussex, and to have imposed loyalty on Londoners by taking twelve leading citizens as hostages.33 We should probably allow two weeks or more for the attacks to take place in various parts of the country. The assembly summoned in London thereafter,34 the king’s arrival, the promise of good laws, and the raising of forces to defend the realm may have taken place around the end of April or the beginning of May, and it was at this moment of crisis that the bishop of Durham declined to fight and left the king’s court without leave. His disseisin was ordered on 12 May, by which date he was well on his way northwards; he was able to reach Durham, though behind him in Lincolnshire and Yorkshire the king’s order took effect. Meanwhile the king went into Kent and captured Tonbridge. The six-week siege of Bishop Odo and Count Robert in Pevensey castle must have lasted through the rest of May and much of June. Whitsuntide fell on 4 June, but the king can hardly have held court. During the siege William de Warenne, recently made earl of Surrey for his loyalty to the king, was injured; he was taken to his foundation at Lewes priory, where he died on 24 June 1088.35 It must have been the middle of June at the earliest when Odo was snatched from his captors outside Rochester and the king began his second siege. Since early May, more than seven weeks had passed in the exchange of letters between Bishop William and the king until, in early July, the king, still in Kent, sent Abbot Wido from Canterbury to Durham to bring the bishop to face charges.36 It must have been late July when the bishop at last went south to the king – we do not know where – but failed to engage in discussion, because Archbishop Thomas and the other bishops present were unwilling to deal with him against the king’s instructions. The libellus here points towards a summer court, for which we shall see other evidence in royal diplomas. The bishop did not stay long at court but went back to Durham, and in August the king sent three counts, Alan of Richmond, Odo of Champagne, and Roger of Poitou to fetch him. They promised him safe conduct on 8 September, but the bishop and his escort remained in the north until Michaelmas, and his trial at Salisbury did not commence until November. There

32 John of Worcester iii, 48. 33 De iniusta uexatione, ed. Offler, 91; English Lawsuits, 101. 34 Henry writes, ‘rex autem congregato Anglorum populo reddidit uenatus et nemora legesque promisit

exoptabiles’ (Huntingdon, 414); John writes, ‘mittit legatos, uocat quos sibi credit fidos, uadit Londoniam’, where ‘statuens leges, promittens fautoribus omnia bona’, he prepared to march on Rochester. The assembly at London to whom good laws were promised may have been essentially military in character. 35 Chronicle of Hyde Abbey, ed. E. Edwards, RS 1866, 298–9; the exact date derives from the Lewes cartulary but is accepted as reflecting his obit by Clay, EYC viii, 4; Complete Peerage xii.1, p. 494. Orderic iv, 128, names William de Warenne among the loyal magnates but places his being created earl and his death rather later, iv, 180. 36 De iniusta uexatione, ed. Offler, 75; English Lawsuits, 92.

1088 – William II and the Rebels

147

Henry de Beaumont appears to have been the principal accuser, but two former rebels took part in the debate, Bishop Geoffrey of Coutances and Roger Bigod.37 The role of Henry de Beaumont is important, since like William de Warenne, he seems to have been created earl in the course of the year 1088, taking Warwickshire where he already held his father’s estates.38 It is assumed that this was a reward for his service to the king during the rebellion. It would appear that William was made earl sooner than Henry, for in a charter for Lewes datable after William de Warenne’s death he witnessed as Henry de Beaumont.39 In documents to be discussed below, datable to the summer of 1088, he witnessed as Henry, earl of Warwick, or as Henry of Warwick.40 One is tempted to guess that William de Warenne was made earl in the context of the siege of Pevensey, and that a distinguished role at the later siege of Rochester, without fatal consequences, earned a similar reward for Henry de Beaumont. If so, it is a curious insight into the reasons that induced King William to create earldoms: it begins to look more like a rank of precedence and less like a responsibility in the shires of Surrey and Warwickshire; the tenurial consequences, if any, are a matter of guesswork. Duke Robert’s intentions are altogether less well attested. Much importance has been attached to an act of Duke Robert, dated 7 July 1088. As Barlow puts it, ‘On 8 July he granted a charter to the abbey of Fécamp on the Norman coast and added to the dating clause, “on the day when I should have crossed to England”.’41 He cites C. W. David’s study of Duke Robert, who tells us that ‘a charter by Duke Robert in favor of La Trinité of Fécamp is dated 7 July 1088, quando in Angliam transire debui’.42 If it were correct that a contemporary dating-clause invoked an unfulfilled intention, then already by 7 July Robert had abandoned the plan to invade. David in his turn cites Haskins, where we find that this is a fusion of two texts: first, a charter dated Friday, 7 July 1088, by which Duke Robert granted certain lands to Fécamp, with an annexe of further grants, and then the narratio of a third charter refers back to the Duke’s grant, ‘et quod ego eam quando in Angliam transire debui cum aliis terris ecclesie reddiderim’.43 At this stage, not before 1089, Duke Robert came to Fécamp and delivered seisin of the lands in question. His original grant on 7 July 1088 was not on the day when Robert was due to cross but did not, but rather was around the time when he intended to cross. It seems strange to recall that failed plan a year or more later: perhaps it was meant only to signal the circumstances in which he had then been at Fécamp, if he was indeed there – the act of 7 July does not actually indicate where the duke was. The only layman to witness alongside Duke Robert on that date was his brother Count Henry, who, we learn from Orderic (see below, 150), did cross to England soon afterwards. The last sentence of the Anglo-Saxon Chronicle’s narrative is translated by Henry

37 38 39 40

De iniusta uexatione, ed. Offler, 83–4; English Lawsuits, 97. D. B. Crouch, The Beaumont Twins, Cambridge 1986, 10. Regesta i, no. 325. The argument is made in Complete Peerage, xii.2, Appendix A, that Henry was made earl of Warwick after 24 June 1088 and before he witnessed Regesta i, no. 302 for Rochester (see below), ‘unlikely to be very long after the surrender of Rochester, the date of which cannot be closely determined but which was probably some time in July or a little later’. 41 Barlow, William Rufus, 81 and n. 139. 42 David, Robert Curthose, 51–2 n. 52. 43 C. H. Haskins, Norman Institutions, Cambridge, MA, 1918, 287–9; Haskins makes the fusion in his heading, ‘7 July 1088, Robert, when about to cross to England, restores to Fécamp and frees from all secular dues the land of William of Bec’ [&c.]. This rendering is picked up by Chibnall, Orderic iv, 134 n. 1, and is explicitly amended by Barlow as quoted.

148

Anglo-Norman Studies XXVI

of Huntingdon, but William of Malmesbury reports a different conclusion to the rebellion: ‘Odo, taken prisoner for a second time, forswore England in perpetuity; the bishop of Durham voluntarily went overseas, being allowed to leave the realm unharmed out of respect for their former friendship.’ Ceteri omnes in fidem recepti ‘all the rest were admitted to take the oath of allegiance’.44 Orderic says a little more:45 He [the king] bided his time and waited for the right moment to take vengeance. He punished some of the rebels with severe penalties and deliberately turned a blind eye to the guilt of others. He shrewdly spared the older barons, although the conspiracy had temporarily weakened their allegiance to him, out of love for his father whom they had served long and faithfully, and through respect for their grey hairs. In any case he knew that disease and speedy death would soon put a term to their activities. Consequently some of those who had gone furthest in their treachery obeyed him with all the more devotion in the years that followed and endeavoured to appease him with gifts and services and flattery. To test these contrasting views of how King William treated the rebels, we may consider a small group of charters. Their witness-lists tend to support William of Malmesbury’s view that the king was lenient toward the rebels. The three documents of most interest are two for Rochester cathedral priory and one for the abbey of Le Bec, for which I shall argue a date in July or perhaps August 1088. The act for Lewes priory, already cited, would seem to have been issued earlier, at the end of June or beginning of July. The usefulness of these documents for the present argument is that they all contain substantial witness-lists. In form they are all diplomas rather than writ-charters, and while this provides us with longer lists of witnesses, it deprives us of place-dates; it also raises anxieties, since in this category of act forgery is more readily suspected and less easily proven. There is also the imponderable question why we should have a cluster of diplomas rather than writ-charters from around July 1088. Without an answer to that, it is not possible to know how far these witness-lists reflect an official perception of the attestation as the few select witnesses to a writ-charter would. I take first a short diploma confirming King William’s gift of St Mary’s church, Lambeth, to the church of Rochester in compensation for loss caused during the attack on Rochester Castle.46 There are six named witnesses: presentibus testibus et assentientibus de baronibus meis Thoma archiepiscopo Eboracensi, Rogero comite Scropesbyriensi, Henrico comite VVaruuicensi, Henrico de Ferrariis, Rogero Bigoto, Iuone Taillebosc, Willelmo Peurell, et aliis pluribus. It is only formally datable after Henry de Beaumont was made earl of Warwick, apparently around July 1088, and before the death of Roger de Montgomery, earl of Shrewbury, on 27 July 1093.47 The reference to making amends for damage caused 44 Gesta Regum i, 548. 45 Orderic iv, 134. 46 Regesta i, no. 302. The principal source is Textus Roffensis (Maidstone, Kent Record Office, MS

DRc/R1), fol. 211r–v (no. 202), published in facsimile, ed. P. H. Sawyer, Copenhagen 1957–62; the text was printed by T. Hearne, Textus Roffensis, Oxford 1720, 213–14; again by J. Thorpe, Registrum Roffense, London 1769, 459 [from BL, MS Cotton Domitian X (s. xiii), fols 107v–108r, now fols 109v–110r (no. xviii)]; and in Monasticon i, 173 (no. xliii). 47 This last date has been muddled in the twentieth century. His death is reported by Orderic as happening on 27 July (Orderic iv, 302); the year is unspecified but Orderic says that he died six years after King

1088 – William II and the Rebels

149

during the siege of Rochester surely points to a date soon after the siege ended, for which we might plausibly guess July 1088. Roger de Montgomery had been complicit in the rebellion earlier in the year, but here, in what may be one of the king’s first acts after the rebellion was finally quelled, he witnesses alongside the king. Roger Bigod, another witness here, had held Norwich castle for the rebels and raided around East Anglia, and there is no clear evidence as to how or when he was brought back into the king’s friendship. Alongside them, loyal Henry de Beaumont is given his new title but not allowed precedence over the more senior earl of Shrewsbury. Ivo Taillebois had been with Henry de Beaumont in Normandy after William I had died;48 both must have come to England soon after, and William Peverell was presumably also a loyal courtier.49 Also from the Textus Roffensis we have another act in diploma form confirming to the church of Rochester the manor of Haddenham (Bucks) which Archbishop Lanfranc had bought from William I. This act comes with an imposing list of twenty-two witnesses. Its text, which is arguably authentic, was not differentiated in Regesta from a much later inflation of it, though the definitely spurious version abbreviated the witness-list and so reduced the most persuasive means of authenticating the act.50 The Textus Roffensis also includes a narrative of how Archbishop Lanfranc and Bishop Gundulf secured this royal confirmation by a process of negotiation with Henry de Beaumont, earl of Warwick, and Robert fitz Haimo, a loyal friend of the king’s and son of the sheriff of Kent.51 The king had demanded a fine of £100 in return for approving this transfer of a valuable manor, which the archbishop held as part of his lay fee, into the ownership of the church of Rochester. The figure is not an unusually high one, and it happens to be the same amount of money as the church of Rochester received in the distribution of William I’s charitable bequests in 1087. Lanfranc and Gundulf are represented as unwilling to pay so much, and at the suggestion of the king’s negotiators, the fine was commuted: instead Bishop Gundulf, an experienced builder, would raise up the castle at Rochester in stone to William I (Orderic iii, 148). The DNB therefore favoured 1093. G. H. White in the Complete Peerage xi, 687n, thought this statement was not inconsistent with his dying in 1094; he dated Earl Roger’s death to July 1094, which has generally found favour. White was, however, misled by what is no more than a demonstrably mistaken guess by H. H. E. Craster: in 1930 when Craster first published Regesta, no. 338a (Regesta ii, p. 401), he saw a long witness list and the place-date Gloucester, when William II for the first time kept his Christmas court there (ASC, 171). He guessed, therefore, at a Christmas date, which was in conflict with the date of Earl Roger’s death; he accordingly redated the latter to 1094. The earlier date for Earl Roger’s death draws strong support from John of Worcester (iii, 64–6 and n), who quite explicitly places the earl’s death in the same year and context as those of Wido, abbot of St Augustine’s, who died on 9 August 1093, and Paul of Caen, abbot of St Albans, who died on 11 November 1093. Craster overlooked the fact that in 1093 the king was at Gloucester throughout Lent; he also overlooked Regesta i, no. 338 for St Mary’s abbey, York, dated at Christmas, datable to 1093, and witnessed by Hugh de Montgomery, earl of Shrewsbury. 48 They witnessed together alongside Duke Robert; Haskins, Norman Institutions, 285. 49 Presumably William Peverell the elder of Dover, whose earliest certain attestation is in a royal confirmation for Bermondsey priory, dated at Windsor, Whitsuntide 1095 (Regesta i, no. 362). The first explicit attestation by William Peverell the elder of Nottingham is in a writ-charter for Peterborough, c.1094x1098 (Regesta i, no. 409). 50 Regesta i, no. 301 failed to distinguish this document, Textus Roffensis, fols 212r–213r (nos 205, 206), from the inflated version, BL, Campbell Charter vii.1; M. Brett, ‘Forgery at Rochester’, in Fälschungen im Mittelalter, MGH Schriften 33, 1988, iv, 397–412, describes this pseudo-original as ‘a conflation and adaptation of two apparently genuine charters of William II’ [those discussed here concerning Lambeth and Haddenham] ‘clearly written later than c. 1150’ (p. 403). 51 Textus Roffensis, fols 173r–174v (no. 88). The less controversial writer of the Vita Gundulfi episcopi, § 27, ed. R. M. Thomson, Toronto 1977, 50, mentions the king’s gift of Lambeth and his grant of the transfer of Haddenham without alluding to such negotiation.

150

Anglo-Norman Studies XXVI

replace William I’s wooden structure. We may be tempted to conjecture that the circumstances in which this was something the king wanted would most likely be the aftermath of the siege. The witness-list provides evidence in support of this conjecture: + Signum Willelmi regis Anglorum + Signum Lanfranci Cantuariensis archiepiscopi + Signum Thom” Eboracensis archiepiscopi + Signum Remigii Lincoliensis episcopi + Signum Walcelini Wentoniensis episcopi + Signum Mauricii Lundoniensis episcopi + Signum Osmundi Serberiensis episcopi + Signum Rodberti Herefordensis episcopi + Signum Baldeuuini abbatis sancti Eadmundi + Signum Henrici fratris regis + Signum Philippi filii Rodberti comitis Flandri” + Signum Alani comitis + Signum Hugonis comitis + Signum Heinrici comitis + Signum Willelmi comitis + Signum Eudonis dapiferi + Signum Rogerii Bigotis + Signum Gosfridi de Magnauilla + Signum Rodberti filii Haimonis + Signum Hugonis de Monte forti + Signum Gisleberti de Tonebrigge + Signum Hugonis de Bello campo + The list of bishops is considerable, suggesting that the king had called what amounted to a council, though where the act was completed is unknown. The key evidence for the date is the presence of Henry, the king’s brother, who was scarcely ever in England during William II’s reign. We do know, however, that he came to England in the summer of 1088. On 7 July he was still in Normandy, when he witnessed alongside Duke Robert, as already mentioned. From Orderic we learn that,52 when positive news of the surrender of Rochester was heard in Normandy, Prince Henry, count of the Cotentin, crossed to England and asked his brother for his mother’s lands. King William received him kindly as a brother should and fraternally granted his petition. In the autumn, after completing the business that had brought him, he said good-bye to the king and began preparations for returning to Normandy with Robert de Bellesme, who had already been reconciled to the king through the intercession of powerful friends. Orderic goes on to explain how Duke Robert arrested Count Henry on his return to Normandy, along with Robert de Bellesme, and that when news of this reached England, Robert’s father Roger de Montgomery left England for Normandy to protect his castles against the duke. If Henry was in England only from July to September, then this act must have been issued during that period. As with the act granting St Mary’s Lambeth to Rochester, so here we have a mixture of King 52 Orderic iv, 148.

1088 – William II and the Rebels

151

William’s allies and rebels among the witnesses. The most senior laymen of the realm are his allies Count Alan of Richmond, Earl Hugh of Chester, Earl Henry of Warwick, Earl William de Warenne of Surrey – and this, it seems, must be William II de Warenne, already styled earl. Eudo Dapifer and Robert fitz Haimo come lower down the order, and between them are the former rebel, Roger Bigod, and Geoffrey de Mandeville, who has been mentioned as a rebel, though that is no more than a historian’s conjecture.53 Later in the list is Gilbert of Tonbridge, whose estates in Suffolk were greater than Roger Bigod’s, who had more land in Surrey than William de Warenne, and who took part in the rebellion in Kent; he had been wounded during the short siege of Tonbridge at the start of the king’s campaign in Kent, and he had remained under guard there when the king’s forces moved on to Pevensey.54 These diplomas appear to bear out the story that King William treated the rebels with clemency and received them back into his service. The same implication runs through another document, which must be dated to very much the same period. This is a royal confirmation in Norman style of a gift to the abbey of Le Bec by Walter Giffard; it survives as an original diploma with the signa of the witnesses and the royal seal, and also as a sealed exemplification.55 The act is unusual in form for William II, though it would not have been at all unusual for his father; its seals are authentic and the witness-list unimpeachable. William II’s seal was reproduced from this act.56 The business of the transaction is relatively minor. Walter Giffard had given the manor of Blakenham (Suffolk) to the abbey of Le Bec, a transaction attested by four of his men; this was then confirmed by the king and attested with seventeen signa in rows below the text: + Henricus de Guaruuic + Signum regis + Rogerus de Monte gomerici + archiepiscopus de Ebroic + Rogerius Pictauensis + Eudo dapifer + Gislebertus filius Ricardi + episcopus de sancto Laudo + episcopus de Lincholensi + episcopus de Herefort + comes Alanus + comes Hugo de Cestre + Rogerius Bigot + Guillelmus de Perceio + Hugo de Montfort 53 No source attributes any role to him at this date. Barlow, however, has brought together the claim made

by Bishop William of Durham that there had been outbreaks of rebellion at Dover, Hastings, and London, and the assumption that Geoffrey de Mandeville had custody of the Tower of London (William Rufus, 76). 54 John of Worcester iii, 52. 55 Regesta i, no. 320. H. J. Ellis dated this act to the vacancy 1089x1093 (Complete Peerage ii, 387n), on the strange assumption that the absence of Lanfranc’s name implies that he was dead; this assumption seems to have lingered. The editors of Regesta give the date c.1091, following the terms set out in a note on the facsimile published by the New Palaeographical Society (pl. 45a), that William Rufus was not reconciled with Walter Giffard until 1090 nor with his brother Henry until February 1091, and that Henry was in England briefly that year. 56 T. A. M. Bishop and P. Chaplais, Facsimiles of English Royal Writs, Oxford 1959, pl. xxx; Barlow, William Rufus, pl. 1b. These give the date-range August 1091 x May 1092; the first term is unexplained, the second is the death of Remigius, bishop of Lincoln.

152

Anglo-Norman Studies XXVI + Henricus filius regis + episcopus de Wincestre

There is again no indication of where this transaction took place, but it is a weighty witness-list. Next to the king, we have Henry de Beaumont, already styled ‘of Warwick’; Roger de Montgomery, sometime rebel who went over to the king early in the rebellion; Archbishop Thomas of York (Ebroic in Norman French); Roger of Poitou, a possible rebel; Eudo dapifer, one of William’s father’s loyal curiales; the bishop of Saint-Lô is none other than Geoffrey, bishop of Coutances, who had rebelled at Bristol in the spring; the bishops of Lincoln and of Hereford; Count Alan of Richmond and Earl Hugh of Chester, both loyal throughout; Roger Bigod who rebelled in East Anglia, the only disloyal sheriff; William de Percy, patron of Whitby abbey;57 Hugh de Montfort, a royal official from William I’s court who witnessed only these two acts for Rochester in William II’s time;58 Henry the king’s son; and finally Walkelin, bishop of Winchester. The irregular sequence reflects the haphazard placing of the signa and not the order of signing. Apart from the king himself, twelve names are shared by this act and the confirmation of Lanfranc’s gift to Rochester of Haddenham, a high proportion. These are (following the more orderly precedence of the Rochester diploma): Archbishop Thomas of York, Bishop Remigius of Lincoln, Bishop Walkelin of Winchester, Bishop Robert of Hereford (a very rare witness), Henry the king’s brother (surely to be equated with the otherwise uninterpretable Henry the king’s son in the act for Le Bec), Count Alan of Richmond, Earl Hugh of Chester, Earl Henry of Warwick, Eudo dapifer, Roger Bigod, Hugh de Montfort, and Gilbert fitz Richard of Tonbridge. Among these names, again we have two former rebels, Roger Bigod and Gilbert of Tonbridge. There is a prima facie case for thinking that these two acts may have been drawn up on the same occasion, and the number of bishops present suggests a significant occasion. The presence of Henry the king’s brother provides the main route to a dating during July–September 1088.59 Taken in conjunction with the narrative of De iniusta uexatione we can further reduce the possible data-range. By 8 September Count Alan of Brittany, who witnessed both acts, and one possible converted rebel, Roger of Poitou, who witnessed for Le Bec, had reached Durham and made arrangements to bring Bishop William south to meet the king. They appear to have remained in the north for at least three weeks, since it was Michaelmas before the bishop began his journey south.60 It would appear likely that they had set out for the north before the end of August, so that both acts must be dated to the months of July or August 1088. Within this time-frame we know that Bishop William had briefly visited the

57 There is a grant of liberties to Whitby (Regesta i, no. 228, though the sheriff Ralf Paynell argues for

William II’s time), whose substance is suspicious, but whose address to Archbishop Thomas, Count Alan, and Ralf Paynell, and attestation by Archbishop Lanfranc and Bishop Osmund and William de Percy at York, are not incompatible with a date around late January or early February 1088. They may have been taken from an authentic act. 58 William de Percy and Hugh de Montfort are both among the many witnesses to an impressive forgery from Lincoln, dated 1090 (Regesta i. no. 328). Stenton thought it derived from authentic documents, since the rarer names included could not be ‘mere inventions’ (in C. W. Foster, The Registrum Antiquissimum of the Cathedral Church of Lincoln, vol. i, Lincolnshire Record Society 27, 1930, 10–11). The inclusion of Robert, count of Mortain, there, and also Bishop Robert of Hereford makes it plausible that another act from the summer of 1088 was among the sources used. There are, however, sufficient anomalies among the bishops to rule out the possibility that the witness-list was taken over as a unity from a single source. 59 This was recognized by C. W. Hollister, Henry I, New Haven, CT, 2001, 61 n. 149. 60 De iniusta uexatione, ed. Offler, 79–80.

1088 – William II and the Rebels

153

king, whose company at the time included Archbishop Thomas and several other bishops. A fourth act, not as it stands authentic, has a witness-list that contributes to the same picture. The beneficiary in this case was St Peter’s abbey in Gent, in Flanders, and King William confirmed to it the lands it held in Kent. The witness list is not so long, but there are again some notable similarities: Signum Willelmi regis. S(ignum) Gosfridi episcopi de sancto Laudo. S(ignum) comitis Roberti de Moritonio. S(ignum) Odonis comitis de Campania. S(ignum) Philippi comitis Roberti Flandrensis filii. S(ignum) Eudonis dapiferi regis. S(ignum) Ostonis Flandrensis. S(ignum) Raingoti de Aldenarda. Here we have the rebel Bishop Geoffrey of Coutances, who witnessed the act for Le Bec, and Philip, son of Count Robert of Flanders, who witnessed the act for Rochester. Neither was habitually in England. Another peculiar similarity is the form of Bishop Geoffrey’s witness; in William I’s time he frequently attested with the title bishop of Coutances, but only in these two acts does he use the title bishop of Saint-Lô, a larger town in his diocese about halfway between Coutances and Bayeux. Eudo dapifer, one of the king’s steadfast courtiers, appears in all three lists. Two other names, however, further demonstrate the swiftness of the king’s leniency: Count Odo of Champagne had been loyal, but he went to Durham with Count Alan of Richmond and Roger of Poitou, so this act again seems likely to fall into the same narrow range of dates (July–August 1088); another witness, Robert, count of Mortain, the king’s uncle, brother of Bishop Odo, took part in the rebellion and was besieged at Pevensey. He attests very rarely in William II’s reign and is only known to have been in England during 1088.61 The two names at the end of the list were presumably members of the Flemish contingent in England, accompanying the son of Count Robert of Flanders, among whose party may also have been the representatives of St Peter’s abbey who sought the confirmation. In spite of the doubts attaching to the tenor of this act, the witness-list would appear to have followed one that could well have been authentic from England in the summer of 1088. It is hardly possible to suppose that forgers at Rochester at the time when the Textus Roffensis was drawn up and forgers at Gent later in the twelfth century had been able to collude and had information available to them drawn from an unusual act such as the diploma for Le Bec. Two versions of a forged confirmation from St Mary’s abbey at York present different versions of a single witness list which, if any confidence could be placed in it, would also have to date from the summer of 1088.62 The presence of Bishop 61 The four acts with his name among the witnesses are this one for St Peter’s, Gent; the conflated

witness-list of a Lincoln forgery (Regesta i no. 328; see above, n. 58); a very dubious confirmation for Rochester (Regesta i, no. 451); and a writ for Abbot Herbert of Ramsey, datable 1087 x May 1089. 62 Neither of these texts has been printed, but a further version is Regesta i, no. 313 (with only two witnesses). From a twelfth-century York manuscript, BL MS Add. 38816, fols 21r–22r, the list contains eleven names: Archbishops Lanfranc and Thomas, Bishops William of Durham and Hildebert (recte Hoel) of Le Mans, Cardinal Albert, Abbot Anselm of Le Bec, Abbots Paul of St Albans and Serlo of Gloucester, Count Alan of Richmond, Earl Henry of Warwick, Earl William II de Warenne, and many others. Roger Dodsworth has preserved a longer version, Bodl. MS Dodsworth 10, fols 110r–111v, with twenty-three names: Archbishops Lanfranc and Thomas, Count Alan, Bishops Geoffrey of Coutances, Remigius of Lincoln, Godfrey of Chichester, Gundulf of Rochester, and William of Durham, Earl Henry of Warwick,

154

Anglo-Norman Studies XXVI

Godfrey of Chichester, who died on 25 September 1088, a few months after his consecration, is particularly distinctive. The attestation of Anselm of Le Bec is also remarkable: he is not known to have been in England in 1088, though the confirmation of Walter Giffard’s gift to the abbey suggests that there was a deputation in England. The witness of an otherwise unknown Cardinal Albert, however, diminishes confidence here. If more reliance could be placed on these two copies, we should have to date them to the summer council at which Bishop William of Durham briefly attended on the king and his fellow bishops. It would be his first datable attestation since he was with the king in York in January or February of 1088. At this point I should like to go back and consider the Lewes diploma in more detail.63 If the absence of Henry de Beaumont’s comital title can be trusted, this must predate the two acts for Rochester cathedral priory that deal with issues arising from the siege there. It can hardly be later than July 1088, and yet already in among the witnesses here, we find signs that rebels are being drawn back into the king’s circle. It is attested by the king and eighteen other witnesses, and, unusually for any witness-list of that length, there are only two clerks included: Signum + Willelmi regis Signum + Gunnulfi episcopi Signum + Willelmi Teforensis episcopi Signum + Rogerii comitis Signum + Roberti comitis Signum + Walteri Gyphardi Signum + Henrici de Ferrariis Signum + Henrici de Bellomonte Signum + comitis Alani Signum + comitis Hugonis de Cestria Signum + Reotgerii de Busliaco Signum + Bernardi de Nouo Mercato Signum + Walterii de Aincurte Signum + Ricoardi Signum + Radulfi de Caisneto Signum + Ricardi Signum + Hugonis filii Golde Signum + Godefridi de Petraponte Signum + Hugonis de Wanciaco The terminus post is secured by the reference in the text of the document to the day when William de Warenne was uiuus et mortuus, 24 June 1088. The absence of Henry de Beaumont’s title as earl of Warwick argues for a date before the siege of Rochester was concluded. We may therefore have a glimpse of those with the king as he travelled from Pevensey to Rochester or even as Rochester was besieged. The presence of Bishop Gundulf in these circumstances is not surprising; he was never disloyal, but during the siege he was able to move to and fro between the two sides.64

Earl William de Warenne, Count Roger, the chancellor Gerard, Abbot Anselm of Le Bec, Abbots William of Cerne, Ingulf of Crowland, Ranulf of Saint-Vincent at Le Mans, Serlo of Gloucester, and Paul of St Albans, Cardinal Albert, Milo Crispin from Le Bec, Henry de Ferrers, Bishop Hildebert (recte Hoel) of Le Mans, and Ernulf de Hesdin. 63 Regesta i, no. 325. 64 Vita Gundulfi episcopi, § 28, ‘uirum Dei tamen et licenter exire et licenter cum uolebat urbem permittebat intrare’ (ed. Thomson, 50–1).

1088 – William II and the Rebels

155

Bishop William of Thetford’s presence is unexpected; we have not met him in the king’s company since Christmas in London at the beginning of the year, and this is his only attestation. Earl Roger de Montgomery of Shrewsbury and Earl Robert de Mowbray of Northumberland are former rebels. (If we accept Count Roger of Poitou’s involvement, he was presumably still defending Rochester and raiding in Kent at this date.) Walter Giffard had witnessed frequently for the king’s father, and he would witness Henry I’s coronation charter, but this is his only attestation in William II’s time outside a definite forgery from Tewkesbury. The document for Le Bec, however, which we have already considered, confirmed his gift, so we may infer that he was in England around this time. Henry de Ferrers was another courtier of William I who in William II’s time witnesses only here and in the diploma soon afterwards that granted the church of Lambeth to Rochester. Henry de Beaumont we have already discussed; Count Alan and Earl Hugh of Chester were fighting with the king. Roger de Builli otherwise witnesses only William II’s diploma granting Bath abbey to John of Tours, bishop of Wells, in January 1091.65 Bernard de Neufmarché had rebelled in the Welsh borders in the spring, and yet here he appears at the king’s side around the end of June or beginning of July; this is his only attestation during William II’s reign. Walter d’Aincourt also attests only here, but he must have been loyal; later in the summer of 1088 he acted in the king’s name, giving instructions to the bishop of Durham, whose men (‘homines episcopi qui in castello suo sunt’) had stolen from the king’s custody two hundred cattle belonging to Bishop Geoffrey of Coutances.66 It would appear that before the siege of Rochester was over, several of those who had broken their faith to King William had gone to join him, to make their peace, and presumably to renew their oath of fealty. Something of the circumstances of Roger de Montgomery’s return we may accept from the story told by William of Malmesbury. His son Roger of Poitou seems to have returned to the fold as the siege of Rochester drew to a close, while Gilbert of Tonbridge was under guard in Kent and could have been brought to the king in July 1088. Robert, count of Mortain, would surely have been a bigger prize, captured with his brother Odo at Pevensey; the only sign of his witnessing is among the names attached to the Gent forgery. None of the sources explains how the rebellion in East Anglia was suppressed, but Roger Bigod also appears here among the king’s courtiers. How Bishop Geoffrey was brought to heel, we do not know, but he too is here in attendance on the king. Bernard de Neufmarché was thwarted at Worcester by Bishop Wulfstan, but there is no word as to whether he and the other leaders in the border shires were captured and brought to the king. Did such men decide that their rebellion had failed – perhaps as soon as it was clear that resolute action in Kent and Sussex was working – and come of their own accord to the king in the hope of favourable terms? That cannot be known, but evidence shows that soon after the rebellion in Kent was defeated, the king was allowing as witnesses to his diplomas men who had committed acts of violence in his realm and in support of a rival claim to the throne.67

65 Regesta i, nos 314, 474; see also nn. 46, 55. 66 De iniusta uexatione, ed. Offler, 93–4; English Lawsuits, 103. This episode was mentioned at the end

of Bishop William’s trial by one W. de Merlaio, possibly William, steward of Bishop Geoffrey (as Offler notes). 67 Where Barlow, William Rufus, 94–5, presents a round-up of names from witness-lists of 1088–9, no distinction is drawn between those who had been loyal or disloyal; ‘steadfastness’ is attributed even to Roger de Montgomery, Robert of Mortain, Roger of Poitou, Roger Bigod, and Bernard de Neufmarché.

156

Anglo-Norman Studies XXVI

The evidence for his leniency is clear. It includes the seniors who, as Orderic expressed it, would soon be dead: these must include Robert, count of Mortain, and Roger de Montgomery. His son Roger of Poitou was of the younger generation, and he was able to secure his position in William II’s time. Old Bishop Geoffrey and his nephew Robert de Mowbray were allowed to retain their estates. Gilbert of Tonbridge granted the church of Rotherfield to the church of Rochester, a gift confirmed by King William in an act witnessed at Winchester by Roger Bigod. And Roger himself granted the church of Felixstowe, in Suffolk, to the church of Rochester, a gift again confirmed by King William at Winchester in an act witnessed by Eudo dapifer.68 It is not surprising that Gilbert should be a benefactor of the cathedral priory so near to Tonbridge, but Roger’s interests were not in Kent. Were they perhaps both making some condign gesture of regret – a gift to a charity of the king’s choosing – within a few weeks of the siege of Rochester? We should probably also reflect on how much money was paid to the king to secure this lenient treatment.69 Roger Bigod, although also a trusted official in the king’s father’s time, seems certainly to have belonged to Orderic’s category of those who, forgiven for their disloyalty, became more faithful than ever. Punishment is not much in evidence here. The Chronicle’s notion that the lands of the disloyal were made over to those who helped the king is certainly not generally borne out. The nearest to it, one might think, is that Queen Matilda’s English lands, which the king’s brother Henry had hoped to inherit, were given to the landless Robert fitz Haimo.70 Odo was exiled, and so were the leading defenders of Rochester, Eustace of Boulogne and Robert de Bellesme, but their lands remained in the king’s possession. Bishop William, we know, was punished in a very public way, tried at Salisbury before being sent into exile, but even he, who betrayed the king’s trust and deserted him in his hour of need at the very start of the campaign against the rebels, had climbed back into the king’s favour by 1091 and been restored to his position. And if Bernard de Neufmarché was already being reconciled to the king before the siege of Rochester was concluded, then it can hardly be said, in Orderic’s phrase, that the king ‘bided his time and waited for the right moment to take vengeance’.71 I wondered whether Orderic was here thinking ahead to 1095, when William of Eu, a rebel for a second time, was mutilated and died as a result, but the response is in Orderic’s own text. Years later, he returns to report on the fortunes of some of the rebels of 1088, when Urban II preached the crusade in 1098, and Duke Robert pledged his duchy to King William. Bishop Odo had lost his earldom and withdrawn to Normandy to live out his last years as bishop of Bayeux. ‘So great was the enmity between him and the king as a result of a past strife’ (pro transactis simultatibus), writes Orderic, ‘that it was impossible for any mediators to bring about a reconciliation.’ Odo had persuaded his brother Count Robert of Mortain to join the rebellion, but ‘when besieged by the king his nephew’, Robert had made peace and ‘been restored to his friendship after surrendering the castle’. Gilbert of Tonbridge and his brother Roger had fortified Tonbridge castle against the king, but it fell at the first assault; there is no word of their reconciliation here, but Gilbert certainly was forgiven. Those who held out against the king’s two siege-towers at Rochester, 68 Regesta i, nos 450, 452, both from Textus Roffensis; both have defective address clauses, which might

be attributed to the twelfth-century compiler, but I remain a little suspicious about them. 69 In 1095 Hugh de Montgomery, earl of Shrewsbury, paid £3,000 to buy his way back into King

William’s favour (Orderic iv, 284). 70 Orderic iv, 220. 71 Orderic iv, 134; quoted above, 148.

1088 – William II and the Rebels

157

Eustace of Boulogne and Robert of Bellesme were disinherited by William, whose final triumph, in Orderic’s tale, was his taking control of Normandy from Duke Robert. Bishop Odo now found it preferable to join the duke on the crusade and died at Palermo in February 1099.72 It was William of Malmesbury who took the gentlest view of William’s treatment of the rebels. Apart from Odo and Bishop William, he said, in 1088, ‘all the rest were admitted to take the oath of allegiance’.73 We have not been able to establish exactly when or where, but the evidence of an unusual series of diplomas from different archives has illustrated how several of the rebels were quickly restored to the king’s favour. The process began before hostilities had even really ended. Two new earls had been made in the context of sieges, and some rebels – Bernard de Neufmarché is the first of those who had actually attacked the king’s towns – were already restored to good standing. He and Gilbert of Tonbridge were never active as courtiers, nor were Robert of Mortain or Bishop Geoffrey, so that their attestation in 1088 is all the more conspicuous. More rebels were perhaps rehabilitated at Rochester, but sometime in late July or August, before Count Alan and Count Odo and Count Roger went north to deal with the bishop of Durham, it appears that there was a large public occasion, attended by both archbishops, several other English bishops, and many of the nobility, when business might be done to deal with the consequences of the war in Kent and to bring back some sense of unity again to the court. If one could rely on the writ-charters that confirmed two gifts from former rebels to the church of Rochester, one might hazard that this happened at Winchester. If there is a trade-off between writ-charters with place-dates and diplomas with long witness-lists, we have not done too badly from the odd documents surviving from the summer of 1088. They cannot tell us where the court was held to restore some kind of normality, but they confirm William of Malmesbury’s understanding of the king’s conciliatory approach in 1088. William Rufus had neutralized the major problem of separating England from Normandy; the rebels who had caused him most difficulty were excluded from England, and he secured their English lands for himself withoug having to distribute them to buy others’ loyalty. He had shown enough strength in this, and in his military successes at Pevensey and Rochester, that other rebels were thankful enough to have been forgiven their temporary disloyalty. Normal government in England was quickly restored.

72 Orderic v, 208–10. 73 Gesta Regum i, 548–9.

The Anglo-Norman Civil War of 1101

THE ANGLO-NORMAN CIVIL WAR OF 1101 RECONSIDERED Neil Strevett In July of 1101, Robert Curthose, duke of Normandy and the eldest son of William the Conqueror, landed in England with the intention of challenging his younger brother, Henry I, for the English throne.1 Though contemporaries recognised a good story when they saw one, modern historians have shown a reticence to consider the episode with only three detailed studies devoted to the campaign of 1101. The first came from E. A. Freeman in the nineteenth century, who in characteristically nationalist terms saw a ‘listless’ Curthose momentarily dazzled by the prospect of the English throne, with the English rallying to support the king they had freely chosen.2 The second came from C. W. David, who considered the campaign as part of his biography of Curthose, published in 1920, in a chapter entitled ‘The Failure to Gain the English Crown’. As one might expect, David relied exclusively upon narrative sources, which meant he saw the outcome of 1101 as much a result of Curthose’s personal failings as Henry’s success.3 The most important and far reaching analysis has been that of C. Warren Hollister, who turned his attention to what he termed the Anglo-Norman Civil War in the early 1970s.4 Like his twelfth-century predecessors, Hollister immediately recognised the importance of the episode. ‘Duke Robert Curthose’s invasion of England’, wrote Hollister, ‘might have changed the course of twelfth century history. But in fact it ended anticlimactically, in a truce rather than a battle, and the newly won crown was saved without a blow.’5 The posthumous publication of Hollister’s biography of Henry I reiterated many of these views, which continue to have an impact on the wider historiography.6 The central space occupied by Hollister’s discussion should come as no surprise. The 1960s and 1970s represented an exciting and innovative period in the study of 1

I am grateful to Professor John Gillingham for the invitation to present this paper to the Battle Conference. I am also grateful to Drs Stephen Marritt, Matthew Strickland and Stuart Airlie who, together with Ms Eileen O’Sullivan, have provided many valuable conversations that have helped to shape my thinking. Versions of this paper have been presented at the Leeds International Medieval Congress, the Glasgow Centre for Medieval and Renaissance Studies and the University of Edinburgh; my thanks are due to the participants at these occasions for comments and observations which have helped to refine my thinking. Finally, thanks are due to my supervisor Professor David Bates for comments on an earlier draft and his unstinting support of my research. The Department of History and the Graduate School Board of the University of Glasgow were generous enough to fund my attendance and participation at the Conference. 2 E. A. Freeman, The Reign of William Rufus, 2 vols, London 1882, ii, 392–415. 3 C. W. David, Robert Curthose, Duke of Normandy, Cambridge, Mass., 1920, 137. 4 Hollister’s articles are most readily accessible in an edition of collected essays, Monarchy, Magnates and Institutions in the Anglo-Norman World, London 1986. For the events of 1101 see, ‘Anglo-Norman Civil War: 1101’, 76–96; ‘Magnates and “Curiales” in Early Norman England’, 97–115. 5 Hollister, ‘Anglo-Norman Civil War’, 77. 6 C. Warren Hollister, Henry I, London and New Haven 2001, 103–48; D. Crouch, The Normans: The History of a Dynasty, London 2002, 165–9. My thanks are due to Professor Crouch for a personal communication in which he discussed the extent to which Hollister’s ideas influenced his account. See also, C. Holdsworth, ‘Peace Making in the Twelfth Century’, ANS 19, 1996 (1997), 3 n. 8.

160

Anglo-Norman Studies XXVI

Anglo-Norman history, particularly with regard to the cross-Channel aristocracy, with the application of new prosopographical methods and theoretical concepts. Central to this was John Le Patourel’s vision of a homogeneous cross-Channel aristocracy, with its concern for a single cross-Channel ruler. Initially developed in the mid 1960s, the concept was given its most persuasive form with the publication of The Norman Empire in 1976.7 Hollister fully subscribed to Le Patourel’s views and worked independently towards similar conclusions.8 Indeed, Hollister took the argument a stage further than Le Patourel in arguing for an Anglo-Norman regnum.9 Within this framework, William the Conqueror’s death in September 1087 is regarded as a moment of crisis for the aristocracy. None of the Conqueror’s sons or the most senior members of the aristocracy accepted the Conqueror’s deathbed bequest and subsequent division of Normandy and England. Within a year a period of instability ensued, which would last until Henry’s victory over his eldest brother at the battle of Tinchbray in 1106.10 Curthose, in alliance with a powerful coalition of cross-Channel magnates, attempted to take England from Rufus in 1088. Among his supporter’s, Curthose could count on his uncles, the Conqueror’s half brothers Robert of Mortain and Odo of Bayeux. Also involved were Bishop Geoffrey of Coutances and his nephew Robert de Mowbray, earl of Northumberland, Roger de Montgomery, earl of Shrewsbury and his son Robert de Bellême, Gilbert de Clare, William, son of Robert, count of Eu and Esutace, count of Boulogne.11 In contrast, Rufus owed his survival to those members of the aristocracy whose interests and lands were predominately based in England. Especially prominent were those Normans who had prospered in royal service, and in particular, those who had assumed the office of sheriff. Their organisation and leadership of the local militia was crucial in containing and defeating the rebellion before it had time to coalesce and develop momentum.12 Support also came from those members of the cross-Channel aristocracy whose Norman lands lay on the fringes of the duchy, where ducal power had always been difficult to enforce, in particular, Hugh d’Avranches, earl of Chester.13 However, William de Warenne’s involvement is also notable, as his lands lay close to the heartlands of ducal power.14 Nor were the sides static. Negotiations detached Robert de Mowbray and Roger de Montgomery from Curthose’s cause by the time of the siege of Rochester, towards the end of the rebellion in England.15 7

J. Le Patourel, ‘Norman Barons’, in Feudal Empires Norman and Plantagenet, London 1984, VI, 27; Normandy and England, 1066–1144, Reading 1971, 9; The Norman Empire, Oxford 1976, esp. 179–221. 8 Hollister, ‘Anglo-Norman Civil War’, 77–8. 9 ‘Normandy, France and the Anglo-Norman Regnum’, in Hollister, Monarchy, Magnates, 17–58. 10 The most recent discussions can be found in J. A. Green, The Aristocracy of Norman England, Cambridge 1998, 274–83; R. Bartlett, England under the Norman and Angevin Kings 1075–1225, Oxford 2000, 4–21; Crouch, The Normans, 117–28, 130–5, 165–9. 11 ASC, E, 1088; Orderic iv, 128; John of Worcester iii, 48; Hollister, ‘Anglo-Norman Civil War’, 79–80. 12 I intend to provide a full appraisal of the role of the sheriffs in the rebellion of 1088 in my Ph.D. thesis, ‘The Anglo-Norman Aristocracy under Divided Lordship: 1087–1106. A Social and Political Study’ (currently in progress); Frank Barlow, William Rufus, London 1983, 73. 13 L. Musset, ‘Les fiefs de deux familles vicomtales de l’Hiémois au XIe siècle, les Goz et les Montgomery’, in Revue historique de droit français et étranger 48, 1970, 431–3; A. Nakamura, ‘The Earls of Chester and their Family in Normandy and England from the Early Eleventh Century until 1120’, unpublished University of Glasgow M.Phil. thesis, 1997. 14 ‘The Taming of a Turbulent Earl: Henry I and William of Warenne’, in Hollister, Monarchy, Magnates, 137–44; Green, Aristocracy, 31; L. C. Loyd, The Origins of Some Anglo-Norman Families, ed. C. T. Clay and D. C. Douglas, Baltimore 1985, 111; L. C. Loyd, ‘The Origin of the Family of Warenne’, Yorkshire Archaeological Journal 31, 1934, 97–113. 15 Orderic iv, 128.

The Anglo-Norman Civil War of 1101

161

After Rufus’s death in August 1100, Henry took the opportunity to seize the English throne. In his challenge to his younger brother, Curthose was supported by essentially the same coalition of magnates and families who had supported him in 1088, although the intervening years had seen several deaths among the major nobility, with sons inheriting their fathers’ titles and cross-Channel estates. Robert de Bellême had succeeded to his father’s earldom of Shrewsbury, and was joined by his brothers Roger and Arnulf, while William of Mortain had succeeded to his father’s lands and title. In addition, William II de Warenne followed a different course from his father and joined the ducal party, together with Walter II Giffard, earl of Buckingham, Ivo de Grandmesnil, Robert de Lacy, lord of Pontefract and Eutace III, count of Boulogne.16 Henry, like Rufus before him, found support among his officials and sheriffs. Other supporters can be seen to have had close personal ties to Henry stretching back to the late 1080s, when he attempted to establish himself in western Normandy, including Richard de Redvers; or came from families whose lands were concentrated mainly in England or lay outside of Normandy, as was the case of the Beaumont brothers, Robert count of Meulan and Henry, earl of Warwick. Also listed as partisans of Henry were Roger Bigod and Robert fitz Hamon.17 As in 1088 the king could rely upon the support of the Church, with Archbishop Anselm especially prominent.18 Finally, and rather sweepingly, Orderic noted that the ‘all English’ supported Henry, adding that they did so because they did not recognise the ‘rights of the other prince’.19 Overall, a remarkable degree of consistency is apparent over a thirteen-year period in the decisions made by many members of the aristocracy to support either, Rufus and Henry, or Curthose. Significantly, apart from Odo of Bayeux, many of those who were to initiate so much instability in 1087 had unblemished records of loyalty to Conqueror’s regime. The formulation of the concept of a civil war to help explain this was a significant departure in the existing historiography. Hollister’s civil war thesis was in reality part of the wider analysis of relations between the Anglo-Norman aristocracy and the sons of the Conqueror that he undertook in the 1970s. A core component of Hollister’s civil war argument was his belief in a change in the nature of politics under Rufus. Using a methodology based upon a comparison of witness lists to surviving royal acta from the Conqueror’s reign to that of his son, Hollister concluded that a dangerous schism had been created between the cross-Channel magnates and a newly risen administrative elite, whom Hollister termed curiales. The prominence of these curiales in the surviving texts could only be explained by the gradual eclipse of the great magnates at the heart of the royal entourage and therefore the centre of political power. At the time of Rufus’s death, the split between magnates and curiales was as pronounced as ever, manifesting itself in the decisions made by the aristocracy to support either Henry or Curthose. As Hollister succinctly summarised ‘the war of 1101 pitted the curiales of the previous reign against the non-curial magnates’.20 Except it may be doubted that this is indeed what happened.21 At the outset, it 16 17 18 19

Orderic v, 308; ASC, E, 1101; Hollister, ‘Anglo-Norman Civil War’, 79–80; idem, Henry I, 132. Orderic v, 298; Gesta Regum i, 716. Gesta Regum i, 716; Eadmer HN, 127. Orderic v, 314, ‘. . . omnes quoque Angli alterius princips iura nescientes in sui regis fidelitate perstiterunt, pro qua certamen satis optaverunt’. 20 Hollister, ‘Magnates and “Curiales” ’, 114. 21 For critiques of the statistical approach undertaken by Hollister see, Barlow, Rufus, 210–13; D. Bates,

162

Anglo-Norman Studies XXVI

must be recognised that the so-called Anglo-Norman Civil War was a war without any fighting. The prevailing mood among the aristocracy in 1101 was quite clearly one where the avoidance of conflict was regarded as a priority. Many of the sources that record the events of 1101 play a variation on a theme of negotiation and reconciliation. Eadmer stated that Archbishop Anselm was appointed as a mediator between the nobility and the king, before Curthose had landed in England, and during the period just after Whitsuntide, when sections of the nobility were beginning to openly desert Henry.22 After Curthose’s landing negotiations between him and Henry continued. The Anglo-Saxon Chronicler noted that ‘the chief men’ went between Curthose and Henry and reconciled them.23 William of Malmesbury recorded that ‘wiser heads’ among the aristocracy were keen not to break the ‘law of natural affection’ between brothers.24 Orderic inverted the order of things by suggesting that the aristocracy actively sought war and it was the intervention of Henry, who negotiated with his brother on a face to face basis, that avoided this calamity.25 John of Worcester noted that ‘sensible discussions’ resulted in a peace, while Wace named three of the baronial negotiators: Robert de Bellême and William of Mortain, presumably for Robert Curthose, and Robert fitz Hamon for Henry.26 Overall, the observations of these writers constitute a formidable body of analysis. They suggest that the impetus behind the search for a negotiated settlement in 1101 was well understood several decades later when many of these accounts were written and lay much deeper than an understandable fear of war. The observations of William of Malmesbury, Orderic and the other writers who concerned themselves with the nature of Curthose’s challenge to Henry’s kingship need to be seen as part of a much wider set of discussions and disagreements over succession and political legitimacy that had been a general feature of political life for generations.27 More specifically, the events of 1101 have to be seen in the context of the violence and instability created after the division of England and Normandy in 1087. Orderic constructed a rhetorical scene where he presented the arguments put forward by a rebellious aristocracy to justify its actions in 1088. It was claimed that Curthose was the first born, weaker and more pliable in character, and the aristocracy had already sworn fealty to him for their Norman lands. They doubted whether it was possible to serve two lords who were in the eyes of the conspirators, so different and lived so far apart.28 The evidence from this passage would suggest that substantial sections of the aristocracy clearly doubted whether the decision taken to divide Normandy and England in 1087 was either legally correct or politically viable. Quite clearly the Conqueror’s deathbed bequest was contrary to the political preferences of many members of the cross-Channel elite, and whose response was to think in terms of violence to overturn the Conqueror’s bequest.

‘The Prosopographical Study of Anglo-Norman Royal Charters’, in Family Trees and the Roots of Politics: The Prosopography of Britain and France from the Tenth to Twelfth Century, ed. K. S. B. Keats-Rohan, Woodbridge 1997, 89–102. Especially important is the appendix to Henry I, where Hollister maintains the value of his approach, Hollister, Henry I, 499–506. 22 Eadmer HN, 127. 23 ASC, E, 1101. 24 Gesta Regum i, 716–18, ‘Sed satagentibus sanioris consilii hominibus, qui dicerent pietatis ius uiolandum si fraterna necessitudo prelio concurreret, paci animos accommodauere, reputantes quod, si alter occumberet, alter infirmior remaneret, cum nullus fratum preter ipsos superesset.’ 25 Orderic v, 318. 26 John of Worcester iii, 98; Wace ii, 270–1. 27 See the comments in D. Bates, ‘The Conqueror’s Adolescence’, ANS 25, 2002 (2003), 5. 28 Orderic iv, 122–6.

The Anglo-Norman Civil War of 1101

163

Recognition of this immediately launches any revision of the so-called Anglo-Norman Civil War onto a different historiographical trajectory than that previously taken; one that is firmly rooted in the wider concepts of succession politics, but also in the historiography of the ‘Feudal Revolution’.29 This offers an interpretative framework that acknowledges political change and the recourse to violence, yet also the need for conflict resolution and the peaceful co-existence of legitimate centres of power, which might exercise overlapping claims on the loyalties of a trans-regional aristocracy. In this context, the peace settlement of 1101 is far from the anti-climax that Hollister thought. It must be viewed as part of a genuine and ongoing attempt by the aristocracy to solve the problems of instability, by seeking to accommodate the rights and responsibilities of both Henry and Curthose within a framework that stressed the separate political and legal co-existence of England and Normandy. Arguably, this is what the Conqueror had attempted in 1087, and was attempted again by the aristocracy within the terms of the treaty of Rouen in 1091.30 This treaty might be thought to be an even more remarkable settlement than the one negotiated in 1101, in that it had to contend with the ambitions of three brothers and the legacy of Rufus’s unprecedented intervention into Normandy.31 Overall, given the conceptual framework in which Hollister viewed the cross-Channel aristocracy and the methodology he employed to interpret surviving royal acta, Hollister’s analysis undoubtedly underplayed the anxieties that contemporaries felt on the issue of succession, an anxiety that continued to be expressed in the histories written in the first quarter of the twelfth century. The general framework within which the aristocracy is now discussed has moved on considerably from the 1970s, and leaves little room to doubt that it was a truly heterogeneous entity, whose overall concern for a single cross-Channel ruler must have been variable.32 An emphasis on the diversity of aristocratic interests would help to 29 The literature on both topics is vast. However, for the politics of succession especial note should be

taken of G. Garnett, ‘ “Ducal” Succession in Early Normandy’, in Law and Government in Medieval England, ed. G. Garnett and J. Hudson, Cambridge 1994, 80–110; J. Martindale, ‘Succession and Politics in the Romance-speaking World c. 1000–1140’, in her Status, Authority and Regional Power: Aquitaine and France, 9th to 12th Centuries, Aldershot 1997, 19–41. On the concepts of the ‘Feudal Revolution’ see T. N. Bisson, ‘The “Feudal Revolution” ’, Past and Present 142, 1994, 6–42. The interplay of aristocratic violence and social change has proved controversial: see D. Barthélemy, S. D. White, T. Reuter, C. Wickham and T. N. Bisson, ‘Debate: The “Feudal Revolution” ’, Past and Present 152, 1996, 196–223; 155, 1997, 177–225. The wider literature of La mutation de l’an mil is vast. Of particular relevance in this context is, D. Barthélemy, La société dans le comté de Vendôme de l’an mil au XIVe siècle, Paris 1993; idem, La mutation de l’an mil: a-t-elle eu lieu? Servage et chevalerie dans la France des X et XIe siècles, Paris 1997; P. A. Stafford, ‘ “La mutation familiale”: a Suitable Case for Caution’, in J. Hill and M. Swan, eds, The Community, the Family and the Saint, Turnhout 1998, 103–25. An overview of the existing literature and debate can be found in D. Bates, ‘England and the “Feudal Revolution” ’, Il Feudalesimo nell’Alto Medioevo, Spoleto 2000, 611–49. 30 The major commentaries on the treaty of Rouen can be found in David, Curthose, 59–63; Barlow, Rufus, 281–6. For the treaty of Winchester see Le Patourel, Norman Empire, 199–200; Hollister, Henry I, 141–5; David, Curthose, 133–7. More generally, see J. A. Green, ‘Robert Curthose Reassessed’, ANS 22, 1999 (2000), 110–12. See also Freeman, William Rufus ii, 522–8, 688–91. The terms of the treaty of Rouen can be found in ASC, E, 1091; Orderic iv, 236, v, 252; Gesta Regum i, 548; John of Worcester iii, 58. On the veracity of John’s statement regarding Mont St Michel see Barlow, Rufus, 282 n. 84; Hollister, Henry I, 78 n. 216. On the nomenclature of the respective treaties see Barlow, Rufus, 281 n. 77; Hollister, Henry I, 141. 31 In general see Barlow, Rufus, 263–88; see further my forthcoming thesis. 32 The literature on this topic stands as a testament to the stimulus given to Anglo-Norman history by the work of both John Le Patourel and Warren Hollister. Criticisms of the centripetal nature of the relationship between England and Normandy and of the notion of a homogeneous aristocracy can be found in D. Bates, ‘Normandy and England after 1066’, EHR 104, 1989, 851–80; J. A. Green, ‘Unity and Disunity in the

164

Anglo-Norman Studies XXVI

explain the continued support given to Rufus and Henry by those individuals and families whose interests were predominately based on England, and in particular, those who had prospered in royal service. The historiography of many of these individuals, men such as Hugh de Port, Haimo dapifer, Urse d’Abetôt, Durand des Pîtres and many others, is well established and certain general traits can be discerned. Many of the sheriffs who are identifiable as office holders in both 1088 and 1101 started their careers as tenants of those closest to the centre of power in pre-1066 Normandy, to whom they found themselves in opposition in 1088 and again in 1101.33 The development of extensive cross-Channel estates often facilitated their introduction into England as tenants of their Norman lords. However, the opportunity for royal service in England acted as a counter weight to these ties, and gave these men an independent power base from which they were able to construct careers of local importance.34 For men such as these the core issue in 1087 and again in 1101 was the harsh reality of the continuity of career across a change of regime. As their careers demonstrate, acting as the king’s representative in the localities was a potent source of power and influence, both in articulating royal authority and in manipulating it to their own advantage. The prospect of a Curthose kingship and a return of former lords and patrons would have been distinctly unappealing. There are, however, several crucially important aspects to this situation. In a society where social mobility was marked and alternative forms of patronage and power can be seen to complement one another, none of these issues were new. The revolt of 1075 and the arrest of Odo of Bayeux in 1082 had raised these issues for many individuals, though admittedly on a much reduced scale. By 1087, let alone 1101, the careers and backgrounds of many sheriffs had given them experience in negotiating precisely the sort of problems evident after 1087. Moreover, consideration of these issues suggests that the division of the aristocracy into curial and non-curial magnates is somewhat artificial. Most of the visible sheriffs and royal officials in the thirty or so years after the Conquest were clearly well known to the ruling elite and often utilised existing relationships with the elite to advance their careers, or Anglo-Norman State’, Historical Research 63, 1989, 115; eadem, ‘Lords of the Norman Vexin’, in War and Government in the Middle Ages: Essays in Honour of J. O. Prestwich, ed. J. C. Holt and J. Gillingham, Woodbridge 1984, 47–61; eadem, ‘King Henry I and the Aristocracy of Normandy’, in La France anglaise au moyen âge, Actes du 111e congrès national des sociétés savants, Poitiers 1986, 161–73. The wider corpus of literature produced in the 1990s saw further erosion of the notion of homogeneity through detailed individual studies of the aristocracy and the diversity of aristocratic experience. In particular see D. Crouch, ‘Normans and Anglo-Normans: A Divided Aristocracy?’, in England and Normandy in the Middle Ages, ed. D. Bates and A. Curry, London 1994, 51–67; E. Cownie, ‘Religious Patronage and Lordship: the Debate on the Nature of the Honor’, in Family Trees and the Roots of Politics, ed. Keats-Rohan, 133–46; B. Golding, ‘Anglo-Norman Knightly Burials’, in The Ideals and Practice of Medieval Knighthood, I, Woodbridge 1986, 35–48; J. C. Holt, ‘What’s in a Name? Family Nomenclature and the Norman Conquest’, in his Colonial England 1066–1215, London 1997, 179–97; P. Stafford, ‘Women and the Norman Conquest’, TRHS 6th ser. 4, 1994, 221–49; Green, Aristocracy, 126–40. 33 On the role of the post-Conquest sheriffs in general see J. A. Green, ‘The Sheriffs of William the Conqueror’, ANS 5, 1982 (1983),129–45; eadem, English Sheriffs to 1154, PRO, London 1990; R. Abels, ‘Sheriffs, Lord-Seeking, and the Norman Settlement of the South-East Midlands’, ANS 19, 1996 (1997), 19–50. For individual sheriffs see K. S. B. Keats-Rohan, Domesday People: A Prosopography of Persons Occurring in English Documents, 1066–1166: I. Domesday Book, Woodbridge 1998. For Urse d’Abetôt’s background and early career see E. Mason, ‘Magnates, Curiales and the Wheel of Fortune:1066–1154’, ANS 2, 1979 (1980), 136. For Durand des Pîtres see D. Walker, ‘The “Honour” of the Earls of Hereford in the Twelfth Century’, Transactions of the Bristol and Gloucestershire Archaeological Society 79, 1960, 178; Green, ‘Sheriffs’, 136. For Haimo dapifer see the introduction to Domesday Monachorum of Christ Church, Canterbury, ed. D. C. Douglas, Royal Historical Society 1944. 34 See the comments on Hugh de Port in D. Bates, ‘Kingship, Government and Political Life to c. 1160’, in The Twelfth and Thirteenth Centuries, 1066–c.1280, ed. B. Harvey, Oxford 2000, 79.

The Anglo-Norman Civil War of 1101

165

used family connections to access the royal or ducal households.35 Quite clearly these men were able to move and make careers for themselves in a world of serious political power. Service to the English king was an engine of social change and advancement, as indeed had been service to the Norman duke, though the scale of the process was radically different by the twelfth century. Nor should it be forgotten that much of this discussion relates to a world that was essentially local in its outlook. There is no evidence in the sources that might lead one to view the events of 1101 as a civil war essentially fought out between royal servants and great magnates. In the list given by Orderic of Henry’s most prominent supporters, only Roger Bigod had a career based upon royal service, while Robert fitz Hamon came from a family with a history of ducal service.36 However, the participation of these men stands as testament to their rapid scaling of the social ladder. More generally, the noun used by Orderic to describe the men who surrounded Henry and gave him support in 1101 was optimates, and Henry’s chief counsellors in 1101 were the Beaumont brothers.37 Orderic’s comments reveal that common threads of lifestyle, aspirations and outlook bound a king and the aristocracy together, and provided the mechanisms to facilitate complex relationships. One such relationship, that between Robert de Bellême and Rufus, is a case in point.38 Within the aristocracy as a whole there existed a degree of respect for the legitimacy of a king, even when in opposition to him. Examples of a complete breakdown of relationships are relatively few and far between, the most obvious being that between Rufus and Robert de Mowbray in 1095. Yet even here, the evidence of the sources suggests that the reasons behind this breakdown were well known and understood.39 Significantly, many of those involved in de Mowbray’s revolt found enough common ground with Rufus to be received back into royal favour once regicide found its way on to the agenda.40 No source suggests that regicide was a factor in 1088 or 1101, except for a dubious comment by Eadmer who suggested Henry feared for his life and had to be calmed by Anselm.41 Indeed most sources stress the opposite and focus upon the respective rights of each brother’s claim to the throne. Orderic stated that the English supported Henry because they did not recognise the rights of Curthose. He did not state that Curthose was acting in a tyrannical manner in attempting to usurp the throne.42 35 For example, the means of Urse d’Abetôt’s introduction to royal service is not clear. However, his

elder brother, Robert the Dispenser, made a notable career in the royal household, Orderic iv, 172. Urse and his brother appear to have co-operated closely throughout their careers, with Urse eventually inheriting his brother’s lands, Mason, ‘Magnates’, 136. Durand des Pîtres’ brother, Roger, had been settled in England as a follower of William fitz Osbern, and was sheriff of Gloucester by 1071. Though Roger was dead by 1086, when Durand was listed as sheriff, it is possible that Durand was sheriff before this or alternated the office with his brother. See Bates, Regesta, nos 4, 135; Domesday Book i, 169a; Green, ‘Sheriffs’, 136; eadem, Aristocracy, 61. Durand’s son, Roger, continued the family tradition of royal service until killed at the siege of Falaise, Orderic vi, 80. 36 For a general survey of the Bigod family see A. Wareham, ‘The Motives and Politics of the Bigod Family, c. 1066–1177’, ANS 17, 1993 (1994), 223–42. See also Bates, ‘Kingship’, 79; Barlow, Rufus, 61–2. 37 Orderic v, 316, ‘Universi optimates Henrico regi assistentes verba consulis collaudaverunt, et regem ut monitis eius obsecundaret cohortati sunt.’ 38 Orderic noted that Robert surpassed all others in ingenuity and devotion to Rufus while he had custody of Normandy: Orderic v, 214, ‘Rodbertus Belesmensis princips militiae huius erat cuius favor erga regem et calliditas prae ceteris uigebat.’ See also K. Thompson, ‘Robert de Bellême Reconsidered’, ANS 13, 1990 (1991), 263–86. 39 Orderic iv, 278–86. 40 Orderic iv, 280–2. 41 Eadmer HN, 127. 42 Orderic v, 314.

166

Anglo-Norman Studies XXVI

In truth, the whole weight of Hollister’s concept of a civil war rested upon the statistical analysis of the witness lists to surviving royal acta for Rufus, calendared in the first volume of the Regesta Regum Anglo-Normannorum, and the errata and addenda of volume two.43 In the context of the late 1080s and 1090s, issues of documentary loss, the overwhelmingly decentralised nature of the production of writs and the preservation policies of religious institutions, makes statistical analysis essentially meaningless. The temptation is to read the texts in a way that confuses form with function, and in the context of witness lists, to confuse those entrusted with supplying documentary authority to a text, with those who may be thought to have exercised an influence on the direction of royal policy.44 Diplomatic scholarship now stresses the social, political and legal context in which texts were produced. There is recognition of the value of reading these texts as narratives, and for the barriers between them and other sources to be broken down.45 Within this framework, it appears as though the beneficiaries of writs in the 1090s valued them as supplying a form of warranty in the preservation of lands, rights and privileges. Many of the concessions granted or confirmed to institutions were extremely minor, and would hardly have come to the attention of the royal administration had it not been for the efforts of the beneficiaries themselves.46 If read as a narrative on the social conditions prevalent in England in the 1090s they reveal a society still struggling to come to terms with the trauma of defeat and conquest. The historiography on the gradual expansion of royal involvement in local affairs and the link between the Domesday inquest, writs and royal administration is well established.47 However, the language of many of the writs suggests a cultural shift in the perceptions of their beneficiaries’ vis à vis royal administration, stimulated by the Domesday inquest itself. The abbey of Bury St Edmunds provides the clearest example. Bury had a tradition of obtaining confirmatory writs on the accession of each new abbot or king, and two writs early on in Rufus’s reign confirmed Abbot Baldwin in his lands with sake and soke.48 The difference from previous writs, however, lay in the language of the injunction that the abbot was to have his lands as they were on the day when the king’s father was alive and dead, a reflection of the linguistic formula used in the Domesday Book.49 Nor are the Bury St Edmunds writs exceptional. Other Bury St Edmunds charters display the influence of Domesday in their drafting, particularly in the use of the clause tempore patris mei.50 This undoubtedly reflected a conceptual link to the use of tempore regis Edwardi and tempore regis Willelmi within 43 Regesta Regum Anglo-Normannorum i, ed. H. W. C Davis, Oxford 1913; ii, ed. H. A. Cronne and

C. Johnstone, Oxford 1956. 44 Barlow, Rufus, 211. 45 D. Bates, Re-ordering the Past and Negotiating the Present in Stenton’s First Century, Reading 2000,

4. 46 R. Fleming, Domesday Book and the Law: Society and Legal Custom in Early Medieval England,

Cambridge 1998, 33. 47 For example see Fleming, Domesday Book, 68–83; D. Bates, ‘Two Ramsey Abbey Writs and the

Domesday Survey’, Historical Research 63, 1990, 337–9; D. Roffe, Domesday: The Inquest and the Book, Oxford 2000. 48 Feudal Documents from the Abbey of Bury St Edmunds, ed. D. C. Douglas, The British Academy, Records of the Social and Economic History of England and Wales 8, London 1932, nos 12, 13; calendared, Regesta i, nos 291, 293. Cf. Bates, Regesta, no. 34; F. E. Harmer, Anglo-Saxon Writs, Manchester 1952, repr. Stamford 1989, nos 8, 11, 12. 49 Feudal Docs., nos 12, 13, ‘. . . die qua pater meus vivus et mortus’; Regesta i, nos 291, 293. 50 Feudal Docs., no. 15, ‘. . . et omnes illos hominess quos habuit in tempore regis Eadwardi aut in tempore regis patris mei . . .’; Regesti, no. 294.

The Anglo-Norman Civil War of 1101

167

Domesday, and is often used in conjunction with these and similar clauses in the Bury texts.51 Moreover, this language can be found in writs drawn up at other institutions. Three writs preserved at Abingdon abbey used the same expressions as Bury St Edmunds to refer to the time of William the Conqueror and Edward the Confessor.52 Writs drafted at Lincoln, Ramsey, Westminster and Thorney Abbey use similar terms.53 The adoption of this language also appears to have been used to express episcopal authority. A writ to Bishop Robert Bloet excused him from pleading for any churches or lands that Bishop Remegius had been in possession of on the day when he was alive and dead.54 Undoubtedly this is much more than simply the adoption of a new administrative lingua franca. The language used in the writs emphasised a continuum of legitimacy across the Conquest and two changes of regime. As such, institutions that appear to have had a policy of record keeping under the Conqueror continued to keep records under Rufus and may have increased their rates of preservation.55 In some instances it is possible to link the preservation of texts with evidence of sophisticated archival practices.56 The practice of witnessing writs by prominent members of the king’s entourage had developed slowly over the Conqueror’s reign to convey the impression that the writ in question reflected the king’s will.57 Moreover, these writs were drafted in order to be read out in local assemblies and courts, and thus presented a means to articulate increasing royal involvement in a local world, whose structures of law and government were still dominated by great magnates, but also open to abuse by the sheriffs and other royal officials.58 In this context it is possible to explain why the majority of royal acta that survive

51 Feudal Docs., no. 15, ‘. . . sicuti Ædwardus rex et post eum Willelmus rex pater meus sibi concessit’;

Regesta i, nos 294, 392. 52 J. Hudson, Historia Ecclesie Abbendonensis, 2 vols, Oxford 2002, ii, nos 19, 31, 51, ‘. . . tempore regis

Eadwardi et patris mei . . .’; calendared Regesta i, nos 289, 359, 390. 53 The list is far from exhaustive. For Lincoln see Registrum Antiquissimum of the Cathedral Church of

Lincoln, ed. C. W. Foster, 12 vols, Lincoln Record Society, Lincoln 1932, i, nos 12, 8, 9; calendared Regesta i, nos 406, 305, 467; Ramsey, Cartularium Monasterii de Rameseia, ed. W. H. Hart and P. A. Lyon, 3 vols, RS 1884, nos 146, 148; calendared Regesta i, nos 295, 296; Westminster, J. Armitage Robinson, Gilbert Crispin, Abbot of Westminster, Cambridge 1911, 137 no. 10, 140 no. 17; calendared Westminster Abbey Charters 1066–c.1214, ed. E. Mason, London Record Society, London 1988, nos 49, 50; Regesta i, nos 306, 436. 54 Registrum Antiquissimum, no. 9; Regesta i, no. 467, ‘. . . de quibus Remigius episcopis saisitus fuit de qua vivvus et mortuus fuit’. 55 For example, Ramsey abbey has two surviving writs for the Conqueror, while this figure jumps dramatically to nineteen under Rufus. Bates, Regesta, nos 221, 222; Regesta i, nos 295, 296, 321, 322, 329, 330, 331, 332, 354, 373, 383, 413, 419, 447, 448, 449, 461, 462, 469. 56 Bury St Edmunds was an abbey with an incentive to keep good records in the years after 1066 given the scale of encroachments on its estates. See E. Cownie, Religious Patronage in Anglo-Norman England, Woodbridge 1998, 66–79. Bury appears to have had an active preservation policy across the regimes of both the Conqueror and Rufus. Five Latin and four Old English writs were preserved from the Conqueror’s reign, and seven writs for Rufus, Bates, Regesta, nos 35, 36, 37, 38, 40, 41, 42, 43, 54; Feudal Docs., nos 12, 13, 14, 15, 16, 17, 19; calendared Regesta i, nos 291, 293, 392, 294, 393, 395, 394. The abbey preserved a writ of the Conqueror on the grounds that it confirmed a bilingual diploma that upheld the monastery’s rejection of the claims by Bishop Arfast for the abbey to be his see. According to the narrative of the diploma, Arfast lost his claim because he could not produce documents or witnesses in support. This sophisticated bureaucracy continued into Rufus’s reign, with one writ granting sake and soke and all the customs of the abbey. It was drafted with explicit reference to the same grants that were contained in previous writs of Edward the Confessor, William the Conqueror and Rufus, Bates, Regesta, nos 39, 40; Feudal Docs., no. 18; calendared Regesta i, 292. 57 Bates, ‘Prosopographical Study’, 100. 58 Abels, ‘Sheriffs, Lord-Seeking’, 19–50.

168

Anglo-Norman Studies XXVI

for Rufus are in the form of writs, which in itself means that the witnesses to these texts have to be seen not only in the context of the diplomatic form of the texts, but also in the context of a local world where religious institutions were vulnerable to infringements from great magnates and royal officials alike, and whose response to the Domesday inquest was to adopt its concepts as the means to give added emphasis and impact to an existing policy of seeking writs as a form of warranty. Great magnates attested fewer documents under Rufus, not because they were being systematically excluded from power, but because a far higher number of documents were preserved that would not have ordinarily required their attestation. These preliminary observations suggest that the aristocracy may have been divided in 1101, but it was not along the lines of curial and non-curial magnates. The conflicting claims of Henry and Curthose to the English kingship, and the choices made by the aristocracy in deciding whom to support belonged to a different world. Henry’s actions in 1100 presented many of the writers who dealt with this issue, and who were also admirers of Henry, with some tricky problems. Henry’s actions caused many of the issues that had been so prominent in 1087 to resurface. His dash to Winchester on the death of Rufus and seizure of the treasury split the aristocracy. William de Breteuil, who maintained that an oath of loyalty had been taken to Robert Curthose, and by right it ought to be maintained, immediately opposed Henry.59 The nature and date of this oath is uncertain. In contrast, Henry’s claim was that of the heir who was on the spot. The importance of being in the right place at the right time is shown by Henry’s so called ‘election’ by supporters who were with him at Winchester. This, together with his hurried coronation only three days later, had all the characteristics of a palace coup, and was regarded as such by Robert Curthose.60 A legal argument to bolster Henry’s actions was found in the doctrine of porphyrogeniture. This had not been a factor in 1087 and must be regarded as a retrospective justification. Curthose’s claim to the English kingship was discussed within the context of his position as the eldest son of the Conqueror. Much of the recent historiography has seen the whole issue of succession and division within the wider framework of developing aristocratic inheritance practices, where land could be apportioned on the basis of the distinction between acquisitions and patrimonies.61 The implication of this wider discourse is to see the events of 1101 within this framework: the claims of an elder son against those of a younger son to their father’s acquisition. On this point it appears that Hollister acknowledged the historiography that had developed since the 1970s and juxtaposed primogeniture and porphyrogeniture to assess the relative strengths of each brother’s claim to the throne.62 Yet in a very real sense much of the existing historiography’s preoccupation with aristocratic inheritance practice is something of a red herring in the context of 1100. As the concerns of Orderic, William of Malmesbury and many other writers clearly show, what concerned contemporaries most were the politics of succession and the criteria used in selecting a king. Nor was this anything new. Majorie Chibnall has demonstrated how William of Poitiers constructed a detailed legal argument in the

59 Orderic v, 290. 60 Jumièges ii, 220; Orderic v, 300, 306–8. 61 See especially E. Z. Tabuteau, ‘The Role of Law in the Succession to Normandy and England, 1087’,

HSJ 3, 1991, 141–69; Garnett, ‘ “Ducal” Succession’, 110. At an earlier stage George Garnett was apparently more sceptical of the application of acquêts and propres to the circumstances of 1087, G. Garnett, ‘Coronation and Propaganda: Some Implications of the Norman Claim to the Throne of England in 1066’, TRHS 5th ser. 36, 1986, 115; cf. Barlow, Rufus, 47–9. 62 Hollister, Henry I, 105.

The Anglo-Norman Civil War of 1101

169

Gesta Guillelmi to support the Conqueror’s claim to the English kingship in 1066, which had to be proved right and just ‘by every law known to be learned’.63 It seems inescapable that in 1087 and especially in 1100, there was a similar need to legitimise the succession of one of the Conqueror’s sons to the English throne. The observations in many of the sources on Curthose’s status as eldest son of the Conqueror undoubtedly reflect the legal and political ties that existed between Curthose and many members of the cross-Channel aristocracy. They had done homage to him for their Norman lands. He had been invested with the duchy of Normandy before 1087, and as Orderic has the Conqueror say, once the ‘honor’ had been granted it could not be taken away.64 In 1087 substantial sections of the cross-Channel aristocracy quite clearly favoured the maintenance of union between England and Normandy under Curthose, and it would appear as though a growing recognition of the benefits of maintaining a union had emerged in the years leading up to 1087. During his rebellion against his father in the late 1070s and 1080s, Curthose was able to attract a coterie of youthful supporters from some very influential families, many with substantial cross-Channel links, who would retain a loyalty to him after 1087.65 Crucially, the advice of these supporters as recorded by Orderic, was for Curthose to claim a ‘share’ of England from his father, who was denying him the riches of his inheritance.66 William of Malmesbury commented that as a result of the dispute, Curthose forfeited both his father’s blessing and his inheritance, failing to secure England and only just retaining Normandy.67 These passages do not necessarily mean that Curthose had been designated as the heir to England, but must be read as passages indicating that he was one possible heir, and on occasion may have been considered as the obvious candidate for the throne, at least by the cross-Channel aristocracy. This almost certainly explains Robert of Torigni’s use of the verb restituere in the context of 1087, when Curthose was urged by his supporters to re-conquer the kingdom of England.68 It is likely that the implications of Curthose’s relationship to the aristocracy was recognised fairly quickly after 1066. The conflict between contemporary political imperatives and longstanding obligations had been thrown into sharp relief by the Conqueror’s own experience of Edward the Confessor’s deathbed bequest to Harold in 1066.69 There is sufficient evidence to suggest that fairly rapidly after 1066 the Conqueror acted in a manner which suggests that he may have had doubts over the feasibility of one individual being both duke of Normandy and king of England. The division of the fitz Osbern inheritance in 1071 is often taken as evidence of the development of aristocratic inheritance practices based upon acquisitions and patrimo-

63 M. Chibnall, ‘ “Clio’s Legal Cosmetics”: Law and Custom in the Work of Medieval Historians’, ANS

20, 1998 (1999), 36. 64 Orderic iv, 92, ‘Ducatum Normanniae antequam in epitimio Senlac contra Heraldum certassem

Roberto filio meo concessi, cui quia primogenitus est et hominium pene omnium huius patriae baronium iam recepit concessus honor nequit abstrahi.’ 65 Orderic iii, 96–100. See also W. M. Aird, ‘Frustrated Masculinity: the Relationship between William the Conqueror and his Eldest Son’, in Masculinity in Medieval Europe, ed. D. M. Hadley, London 1999, 39–54. 66 Orderic iii, 96, ‘Eia viriliter exurge, a genitore tuo partem regni Albionis exige, aut saltem ducatum reposce Normanniae, quem tibi iam dudum concessit coram optimatum qui ad huc praesto sunt agmire.’ 67 Gesta Regum i, 503. 68 Jumièges ii, 204, ‘Cumque sui fideles eum exhortarentur ut regnum Anglie sibi a fratre prereptum velocius armis sibimet retitueret . . .’. 69 See H. E.J. Cowdrey, ‘Death-bed Testaments’, in Fälschungen im Mittelalter, MGH Schriften, vol. 33, iv, Hanover 1998, 716–24.

170

Anglo-Norman Studies XXVI

nies.70 But William’s action in dividing the family into Norman and English branches could equally be thought to be a reflection of his own attitudes towards the possible shape that the future would take. The very process of elevating the Conqueror to the English kingship had created tensions throughout the body politic, which made the continuation of a crossChannel political complex open to question. In one of Orderic’s rhetorical set pieces Earl Roger of Hereford outlined his reasons for the participating in the revolt of 1075. The attitude of some Normans towards the kingship appears to have been to treat it as an office that could be assumed or discarded.71 Crucially, Rufus’s attitude towards his kingship initially appeared to have echoed this perception. William of Malmesbury recorded a scene during the revolt of 1088 where Rufus declared that he would gladly resign the crown if it were thought to be the right course of action by the advisors appointed by his father.72 The revolt of 1075, the arrest of Odo of Bayeux in 1082, and the aristocratic dismay at Rufus’s apparent willingness to contemplate the hanging of the rebels besieged in Rochester in 1088 are but three examples which suggest that the potential and limitations of royal power were being consciously worked out after 1066, and resulted in a few aristocratic casualties along the way.73 Unsurprisingly perhaps, the question of who would succeed the Conqueror is likely to have assumed increasing importance as time went by. If the pro-Curthose views of the senior members of the cross-Channel aristocracy were known before 1087, then there seems little doubt that the Conqueror looked beyond the aristocracy to find the means to facilitate the succession of one of his other sons to the English throne. The evidence of the Conqueror’s deathbed speech would suggest that he had doubts over the means by which he could transmit the English kingship to his preferred choice. He recognised that he had won his crown through military violence, that being a king was a completely new departure for the Norman ducal family, and crucially that the transmission of the kingship through hereditary right was not possible.74 This is in direct opposition to the arguments presented by William of Poitiers who emphasised that the Conqueror’s coronation ensured that his children would succeed him by ‘lawful succession’.75 The reality of the situation in which the Conqueror found himself in 1087 is betrayed by his actions. He may not have been sure of the legalities of the situation, but there was no doubt that in Rufus he had identified the son he wanted to succeed him and was well aware of the strength of opposition Rufus was likely to meet. Fearing that a rebellion would break out once news of his death was known, the Conqueror ensured that Rufus left Normandy before his death, carrying a letter to Archbishop Lanfranc.76 William of Malmesbury described Lanfranc as the ‘moving

70 Orderic ii, 282–4; cf. Tabuteau, ‘Law in the Succession’, 157–8. 71 Orderic ii, 310–22, esp. 314, ‘Unus ex nobis sit rex et duo duces; et sic nobis tribus omnes Anglia

subicientur honores.’ 72 Gesta Regum i, 546, ‘Seorsum enim ducto magnam ingessit invidiam, dicens libenter se imperio

cessurum si illi et aliis videatur quos pater tutores reliquerat.’ 73 Orderic iv, 132. See also J. L. Nelson, ‘The Rites of the Conqueror’, in her Politics and Rituals in Early

Medieval Europe, London 1986, 400–1. 74 Orderic iv, 90–4. 75 Gesta Guillelmi, 150. 76 Orderic iv, 96, ‘. . . epistolam de constituendo rege fecit Lanfranco archiepiscopo . . .’. The translation

by Chibnall as ‘a letter to secure the recognition of the new king addressed to Archbishop Lanfranc’ should be compared to the translation by Douglas and Greenaway as ‘a letter . . . on the appointment of a successor to the throne’, EHD ii, 312.

The Anglo-Norman Civil War of 1101

171

spirit’ in Rufus’s coronation.77 The Acta Lanfranci is more explicit. It noted that Lanfranc chose Rufus to succeed as king as his father had desired.78 As a churchman, Lanfranc naturally looked to canon law for the means to enact the Conqueror’s wishes. As George Garnett has pointed out, Lanfranc’s own copy of the PseudoIsidorian Decretals, brought from Bec to Canterbury and now in Trinity College Cambridge, contains a mark in the margin beside canon 75 from the Fourth Council of Toledo in 633, which lays out the mechanisms by which ‘the bishops and the head men of the people’ would decide who should succeed to a kingdom.79 The inviolability of the king was reinforced by the stress placed upon the effects of anointing by the primate. The canon goes on to threaten with excommunication anyone who disrupts the process through a tyrannical presumption.80 Another marginal mark appears in the manuscript beside a section from the first canon from the Seventh Council of Toledo, which states that anyone speaking or conspiring against a king would be liable to excommunication.81 These marks could have been made at any time after 1075, when the manuscript was certainly at Canterbury, and possibly as early as 1070.82 However, the appropriateness of these two canons to the circumstances of 1087 seems more than coincidental. The two-week delay between Rufus arriving in England in September 1087 and his coronation by Lanfranc undoubtedly involved the English bishops and the ‘head men of the people’ in negotiations.83 Just who might be thought to comprise ‘the head men of the people’ in the circumstances of 1087 is open to question. Crucially, the Conqueror’s half brothers and uncles to Rufus were in Normandy at this point. In 1101, a similar situation presented itself. Henry had been ‘elected’ and crowned before the news of Rufus’s death reached Hugh, earl of Chester, Robert de Bellême and ‘many other magnates’, all of whom, according to Orderic, were in Normandy at the time, most probably awaiting the imminent return of Robert Curthose.84 A crucial point connecting the successions of 1087 and 1100 is that on both occasions many members of the senior aristocracy who were most affected by the decision were unable to express their views and preferences. The exclusion from this process of some of the most important members of the aristocracy created doubts that hung over the legality of the kingship of both Rufus and Henry. The fact that in both 1088 and 1101 opposition to Rufus and Henry did not incur the penalty of excommunication suggests that contemporaries recognised that those who opposed Rufus and Henry were not attempting to disrupt the process through a ‘tyrannical presumption’ 77 Gesta Regum i, 542–4. 78 EHD ii, 679; cf. Eadmer HN, 24. Eadmer described Lanfranc as a ‘vir divinae simul et humanae legis

peritissimus’, whose advice the Conqueror always relied upon. 79 Trinity College MS B.16.44, 328: ‘primatus totius gentis cum sacerdotibus successorum regni concilio communi constituant . . .’; Decretals Pseudo-Isidoriannae et Capitula Angilramni, ed. P. Hinschius, Leipzig 1863, 373–4; Garnett, ‘Coronation and Propaganda’, 91–116. See also M. Philpott, ‘Lanfranc’s Canonical Collection and “the Law of the Church” ’, in Lanfranco di Pavia e l’Europa del secolo XI, ed. G. D’Onofrio, Rome 1993, 131–47. 80 Garnett, ‘Some Implications’, 108–9; Trinity MS B.16.44, 328: ‘. . . aut praesumptione tyrannica regni fastigium usurpaverit, anathema sit in conspectu dei patris et angelorum, atque ab ecclesia catholica quam perjurio profanaverit efficiatur extraneus et ab omni coetu christianorum alienus . . .’. 81 Trinity MS B.16.44, 336. 82 M. Gullick, ‘The English Owned Manuscripts of the Collectio Lanfranci (s.xi/xii)’, in The Legacy of M. R. James: Papers from the 1995 Cambridge Symposium, ed. L. Dennison, Donnington 2001, 101. 83 For the chronology of events see Barlow, Rufus, 55–7. 84 Orderic v, 298, ‘Hugo Cestrensis comes et Rodbertus Belemensis ac alii optimates qui erant in Normannia . . .’.

172

Anglo-Norman Studies XXVI

as stated in the Pseudo-Isidorian decretals. This stands in direct contrast to the action of Lanfranc in 1075 when he excommunicated earl Roger of Hereford and his supporters.85 The nearest one comes to evidence of ecclesiastical sanctions is with Eadmer, who records that Anselm impressed on the Henry’s supporters that any desertion of the king would incur God’s curse.86 Moreover, and although this can only be a speculation, the support given to Curthose by Bishop Geoffrey of Coutances in 1088 may not be totally unconnected with the role he played in the Conqueror’s coronation, where he presided over a joint ceremony with Archbishop Ealdred of York, ensuring that when the Conqueror was presented to the assembly within the abbey, both French and English speakers could acclaim the new king.87 In these circumstances the concerns expressed by many of the chroniclers were perfectly understandable. Robert of Torigni said that Curthose’s first words on hearing of his brother’s coronation in 1087 reflected his ‘usual simplicity’, speaking ‘almost as a fool’. ‘ “By the angels of God, if I were in Alexandria, the English would have waited for me and they would never have dared to make him king before my arrival. Even my brother William, whom you say has dared to aspire to the kingship, would never risk his head without my permission.” ’88 Torigni’s version of Curthose’s response is clearly a rhetorical device, but makes the point that expected avenues of consultation were not being followed. In 1101, according to Torigni Curthose’s response was merely one of anger at Henry’s seizure of the kingdom.89 Nor was he the only writer to concern himself with these issues. Orderic drew a comparison between the situation in 1087 and the division that befell the Israelites under Rehoboam, Solomon’s son and designated successor as king of a united kingdom of Israel.90 In the Old Testament account, Rehoboam had travelled to Shechem to be confirmed as king. Before this could happen, delegates from the Ten Tribes of Israel demanded an end to the levy of forced labour as a condition of accepting Rehoboam. Rehoboam sought advice from the ‘old men’, as the counsellors who had served his father are described, and from the youths he had grown up with. He rejected the advice given to him by his father’s advisers to end the levy, in favour of the advice of the youths who advocated more oppression. On hearing of Rehoboam’s decision, nine of the Ten Tribes of Israel rose in rebellion and elected Jeroboam, one of Solomon’s generals, as their king, with the result that the nation of Israel divided in two, with only the tribe of Judah maintaining its loyalty to the Davidic dynasty.91 The suspicion has to be that Orderic was attempting more than a simple biblical comparison. Rehoboam’s fate and that of the united kingdom of Israel had been sealed by Solomon’s own transgression of God’s law, for which God had judged that he would ‘tear the kingdom . . . out of the hand of your son’.92 Rehoboam’s decision

85 86 87 88

Lanfranc’s Letters, 33A. Eadmer HN, 127–8. Gesta Guillemi, 150. Jumièges ii, 204, ‘Cumque sui fideles eum exhortarentur ut regnum Anglie sibi a fratre prereptum velocius armis sibimet retitueret, simplicitate solita et, ut ita dicam, imprudentie proxima, repondisse fertur: “Per angelos Dei, si ego essem in Alexandria, expectarent me Angli, nec ante adventum meum regem sibi facere auderent. Ipse etiam Willelmus, frater meus, quod eum presumpisse dictis, pro capite suo sine mea permissione minime attenaret.” ’ 89 Jumièges ii, 218. 90 Orderic iv, 122. See also E. Mégier, ‘Divina Pagina and the Narration of History in Orderic Vitalis’ Historia Ecclesiastica’, Revue Bénédictine 110, 2000, 106–23. 91 I Kings 12, 1–15. 92 I Kings 11, 11–13.

The Anglo-Norman Civil War of 1101

173

not to follow the advice of his father’s counsellors resulted in the prophesied split, yet as chapter 12 verse 15 states, ‘for it was a turn of affairs brought about by the Lord that he might fulfil his word’.93 It would appear as though Rehoboam is as much a victim of his father’s choices, as he is of his own. Rehoboam’s mistake in listening to ‘youthful counsels’ had compounded his father’s mistakes in transgressing God’s law by rejecting the advice of those mature counsellors who could foresee the dangers ahead. The implication here is that Orderic felt the Conqueror had made a mistake in dividing England from Normandy by not listening to those who wanted to maintain a union. As part of his rhetorical set piece explaining aristocratic motives in 1088, Orderic has members of the cross-Channel aristocracy form an inviolable league to oppose Rufus and avoid being destroyed by ‘youthful counsels’.94 In 1101, Henry avoids ‘youthful counsels’ by adhering to the advice of his mature counsellors, especially Robert, count of Meulan.95 The emphasis Orderic placed on avoiding making a bad decision worse, suggests that he looked to the upper echelons of the aristocracy to deal with the consequences of the Conqueror’s decision and work for a cross-Channel union. With Henry as king, support from members of the crossChannel aristocracy for Curthose was something of an embarrassment that it had not been under Rufus. However, those members of the aristocracy who supported Henry were still regarded by Orderic as working to mitigate the effects of the Conqueror’s decision. Though less rhetorical than Orderic, William of Malmesbury was equally concerned with the problems caused by division. Though the Conqueror’s decision was at odds with the preferences of most of the senior members of the aristocracy, it appears as though some sort of framework was established to oversee the transfer of power. William describes Roger de Montgomery as a tutor to Rufus, appointed by the Conqueror to advise him. References to a tutor in the narrative sources usually occur in the context of a ducal minority, clearly not applicable to Rufus in 1087. A quality associated with being a tutor is a wider role in the care and development of the duke, indicating a degree of ‘wisdom’ on the part of the tutor.96 This framework might also account for William’s statement that Odo confirmed Curthose in possession of Normandy once he had been released from prison following the Conqueror’s death.97 The evidence would point towards a situation where the initial role of some of the senior members of the aristocracy following the Conqueror’s death was in helping to implement a decision that they were not a party to and did not agree with. 93 94 95 96

I Kings 12, 15. Orderic iv, 122, ‘. . . prudenter precauere ne per consilium iuvenile pereamus’. Orderic v, 298. Gesta Regum i, 546. William of Jumièges uses the term when commenting on the Conqueror’s minority: ‘Is itaque dux in puerilibus annis patre orbatus, sagaci tutorum providential liberalium morum instituebatur ad incrementa.’ William also noted that Duke Robert entrusted his son to his tutors and guardians: ‘. . . dux sub tutoribus et actoribus sapienta uigentibus illum adusque legitimam etatem subegit . . .’, Jumièges ii, 92, 80. William’s guardians are listed by Orderic as Count Alan III of Brittany, Gilbert of Brionne and Osbern the steward, Orderic iii, 86; iv, 82. The tutors mentioned are Turold, Ralph the monk and Master William, Fauroux, nos 220, 259, 262. Orderic also mentions a Thurkill as nutricium to William, Orderic iv, 82. The use of two separate terms by William of Jumièges suggests a division in responsibilities, with actors fulfilling a more public role in the exercise of power and tutors exercising a more pastoral role. However, set against this is Orderic’s statement that William selected Ralph de Gacé as his tutor and commander of his forces on Count Gilbert’s death: Jumièges ii, 98, ‘Rodulfum de Wacceio ex consultu maiorum sibi tutorem eligit, et principiem militie Normannorum constituit.’ 97 Gesta Regum i, 545, ‘Namque cum ille, ut dixi, solutus a uinculis Rotbertum nepotem in comitatu Normanniae confirmasset . . .’.

174

Anglo-Norman Studies XXVI

In this context, unsurprisingly perhaps, the senior members of the aristocracy felt that recourse to violence offered the chance to overturn the Conqueror’s settlement, a sentiment expressed once more in 1101. However, the crucial difference in 1101 was that the lessons of the failure of 1088 had been learnt, and the ability of the aristocracy to cope with a divided lordship during the intervening years had created a platform for a negotiated settlement. Recourse to violence and the threat of violence were clearly limited in what they could achieve, and the dangers were obvious. Within this framework the peace settlement drawn up between Henry and Curthose can be reassessed. The text of the treaty of Winchester has not survived and only its main terms can be discerned in the narrative sources.98 Henry was to give up all of his possessions in Normandy except for the town and castle of Domfront and was to pay three thousand marks of silver annually to Curthose, who in turn gave up his claim to England. All those who had suffered forfeiture on account of Curthose were to have their lands restored. Each brother pledged to assist the other to recover all the lands of their father. Finally, provision was made for each brother to succeed the other in the event of one of them dying, unless the deceased had an heir from a lawful marriage. The agreement was guaranteed by oaths from twelve magnates on each side. Orderic alone added a further provision: anyone working to stir up discord was to be punished.99 After the treaty had been concluded Curthose remained in England for some time, during which he issued a separate confirmation of a grant by Henry of the city of Bath to Bishop John.100 Both Warren Hollister and Judith Green have noted some of the difficulties and contradictory provisions within the treaty.101 Yet the importance of the treaty lies in its recognition that both Henry and Curthose had claims to the English throne, which needed to be separated and settled. The crucial clause here is not so much Curthose’s renunciation of his claim to the English throne, but the provision that related to future succession. Christopher Holdsworth has suggested that this ‘represented no very significant concession’ for either brother.102 Yet Henry’s wife was approaching her fourth month of pregnancy at the time the treaty was negotiated, and Curthose had been married for a year and could probably expect to produce a child in the near future.103 It seems incredible that the negotiators would insert this particular provision, and in these circumstances, unless their intention had been to achieve what the Conqueror had attempted to do in 1087, and had been attempted once more in 1091; namely establish the future means to transmit the English crown as smoothly as possible. As with the treaty of Rouen in 1091, the double confirmation from the period immediately after the treaty, suggests that both treaties were regarded as settling the issues at stake, with each brother being assigned their respective rights and responsibilities.104 That this settlement ultimately failed was due to the fact that

98 Orderic v, 318–20; John of Worcester iii, 90; Huntingdon, 450; ASC E, 1101. 99 Orderic v, 320. 100 D. Bates, ‘A Neglected English Charter of Robert Curthose, Duke of Normandy’, BIHR 59, 1986,

121–4. Hollister, ‘Anglo-Norman Civil War’, 92–3; Green, ‘Robert Curthose Reassessed’, 112. Holdsworth, ‘Peace Making’, 3. Hollister, Henry I, 142. After concluding the treaty of Rouen, Rufus and Curthose issued double confirmations of an exchange between the abbot of Saint Etienne and William de Tournebu: Caen, Bibliothèque de l’Université, fonds normand, Cartulaire de Saint-Etienne de Caen, fol. 50; Bates, ‘A Neglected Charter’, 123.

101 102 103 104

The Anglo-Norman Civil War of 1101

175

many members of the aristocracy had seriously underestimated Henry’s desire and ability to reconstitute his father’s cross-Channel dominions. The paradox that emerges from this discussion is that Hollister’s labelling the events of 1101 as a civil war is very nearly correct, if it is remembered that the aristocracy was split as a result of Henry’s actions, though not on the basis of curial and non-curial factions. There is no evidence to suggest that Rufus enjoyed relations with his magnates that were any worse than those enjoyed by the Conqueror. Underneath the high politics of succession disputes the concerns of a local world continued. The texts that survive from this world need to be seen in this context, and as evidence of wider cultural change. Henry’s actions, and those of his supporters who initially chose him as king, represented another turn in a longstanding discourse on succession and legitimacy that for generations had been, and would long continue to be, a feature of political life. Nor was any of this unique to the Anglo-Norman world. The issues that so concerned Orderic and William of Malmesbury also concerned many other writers in Europe, in particular Abbot Suger of St Denis.105 The real and consistent nature of the support given to Rufus, Henry and Curthose, related to the way in which the ambitions of the Conqueror and his sons were at odds or in tune with the political preferences of a heterogeneous aristocracy. Any reassessment of the aristocracy’s response to Curthose’s challenge to Henry’s kingship must acknowledge that contemporaries recognised that both brothers had a right to the throne and that in consequence, realising the limitations of violence as a means of effecting long-term change, they preferred to search for a negotiated settlement that would open the way to co-existence, and a permanent solution to the problem of divided lordship.

105 See Martindale, ‘Succession and Politics’, 19–22.

Epic and Romance in the Chronicles of Anjou

EPIC AND ROMANCE IN THE CHRONICLES OF ANJOU Neil Wright In this paper I shall consider two anonymous twelfth-century Angevin chronicles, the Chronica de gestis consulum Andegauorum and the Gesta Ambaziensium dominorum.1 My prime concern will be to identify new sources in these texts, through close reading of some relevant passages which will, I hope, shed light on various aspects of both the texts and their sources. One question raised will be the influence exerted by sources in the vernacular – the romance of my title – as well, as we shall see, as that exercised by Latin epic. This latter question will lead in a surprising direction, to an (as far as I am aware) unexpected source. One of my aims is in part to set out for future students, and indeed editors, of the two texts the use made of this source by the Angevin chroniclers. Along the way I shall also consider evidence for the particular manuscript tradition of the source text employed by the chronicles. However, my chief purpose is historiographical: to explore what recognition of embedded borrowings can do for our understanding of the chronicles, of their authors’ intentions and, indeed, of audience response. Above all, I hope to share the enjoyment of reading these chronicles as texts.

Chronica Let us begin, appropriately enough for the papers of the Battle Conference, with a spirited battle passage. At one point in its account of Geoffrey Greymantle, count of Anjou 960–87, the Chronica de gestis consulum Andegauorum (hereafter Chronica) describes a (quite unhistorical) battle which Geoffrey and Hugh Capet supposedly fought near Soissons against a joint force of Northmen (note that the chronicler rather anachronistically calls the Normans Dani) and Flemings. The action itself is depicted in the following emotive passage (Chronica, pp. 42–3):2

1 Ed. Loius Halphen and René Poupardin, Chroniques des comtes d’Anjou et des seigneurs d’Amboise, Paris 1913, 25–73 and 74–132. I should like to express my gratitude to the staff of the Cambridge University Library, who allowed me to borrow the text while it was being repaired (and of St John’s College, Cambridge, who also offered to step into the breach). I would also like to thank Nick Paul, a Cambridge graduate student, who brought a passage from the chronicles to one of my translation classes. When we discovered our common interest, he kindly allowed me to consult an unpublished seminar paper (see n. 14 below); we were both pleased and relieved to find that our researches did not duplicate, but complemented, one another. Finally I dedicate this paper to my partner, Susie Halksworth, for her unfailing support and encouragement. 2 ‘As trumpets sounded and bugles replied, a cry was heard from both armies; now shield was clashing against shield and boss against boss; spears were shattered and swords broken as the legions of Northmen and Flemish came to close quarters; their supports advanced and began to drive the front line back with heavy losses. Unable to resist the attack of so many nations, the French were forced to stumble and yield. So great were the cries, and the noise and hail of weapons that the sky itself grew dark. The king, all eyes as he everywhere watched over his men, groaned and said: “Bring aid to your Frenchmen, Christ!”, and to

178

Anglo-Norman Studies XXVI

Litui clangebant, bucine reboabant, utriusque multitudinis clamor audiebatur, et iam clipeo clipeus, iam umbone umbo repellebatur; hastis confractis enses mutilabantur et cominus utreque Danorum et Flandrensium instabant legiones, et ipsorum superuenere suscenturie et grauiter primos ceperunt repellere. Nequibant enim impetum tot nationum sustinere, sed titubantes cogebantur cedere. Tantus enim erat clamor et strepitus telorum et imber ut ipsum etiam aerem obnubilarent. Ingemuit rex, qui undique prospiciebat eis tanquam occulatus, et ait: ‘Christe, tuos sustenta Francos!’, et Gosfrido, qui suum detulerat uexillum, per nuncium adiecit: ‘Gosfride, rapidum calcaribus urge cornipedem et Francis titillantibus esto iuuamen. Memor esto, obsecro, parentum nostrorum, ne liuidaueris in aliquo titulum Francorum.’ Gosfridus sancte crucis signo munitus et auxiliaribus constipatus manipulis, properus affuit Danisque miles audacissimus obstitit. Interdum enim perfidos agressus est illos, ut uexilli regis lingulas in ora Danorum uolitare faceret altoque clamore suo eos aliquantulum deterreret. Ad illius primipilaris impetum Franci, animo resumpto, in Danos inruunt unanimes et pugionibus uibratis instabant efferatius et instantes. Fragor armorum multus erat et ab ereis cassidibus ignis elucubratus multus scintillabat. Vulnera uulneribus illidebantur et campi nimio sanguine purpurabantur. Intestina uideres dependentia et cesa capita et trunca corpora passim oppetentia. Exterriti sunt autem Dani pre timore nimio et repente, cuneis eorum labantibus, fuge se commiserunt. Persecuti sunt eos Franci sternendo, proterendo mactando, et cesi sunt ibi multi milites et pedites, adeo ut duces ipsorum inuenti sint postmodum mortui in medio quinque millium mortuorum. Magno autem tropheo Franci potiti, leti reuersi sunt ad suos, secum adducentes equos multos spoliaque multa, que sibi ipsi manu sua in prelio pepererent. Factum est igitur gaudium magnum in Francia Deoque dignas omnes edidere gratias. The epic character and inspiration of this passage is immediately apparent in its vivid language and action, gruesome emphasis on wounds, and the heroic stature of Geoffrey Greymantle’s dramatic intervention in the fighting. Such epic qualities as these led Ferdinand Lot to propose in an article published over a century ago that, for their entire account of Geoffrey Greymantle, the Chronica relied on a romance-language source.3 The portion of the Chronica devoted to Geoffrey is quite short, running to only some eight pages in the printed edition. It

Geoffrey Greymantle, who was carrying his standard, he added by means of a messanger: “Geoffrey, spur on your swift steed and bring support to our faultering Frenchmen. I beg you to remember our forefathers and let no blot stain the reputation of the French.” Geoffrey, protected by the sign of the cross and surrounded by reinforcing troops, arrived swifty and opposed the Northmen like a daring knight. At times he attacked those faithless men, making the streamers of the king’s banner flutter in the faces of the Northmen, and cowed them to a degree with his lusty shouting. Upon the onset of this commander, the French recovered their spirits and all together rushed upon the Northmen, waving their daggers and attacking more fiercely and persistently. With a huge crash of weapons, great sparks were flashing, struck from bronze helmets. Wounds were inflicted on wounds, the field was red with pouring blood. One could see guts hanging out, severed heads and mutilated bodies dying all around. The Northmen, overcome by great terror, suddenly turned in flight as their formations collapsed. The French pursued, overthrowing, crushing and butchering them. Many knights and foot soldiers were slaughtered there, so that their leaders were afterwards found dead amidst five thousand corpses. The French, having won a great victory, returned joyfully to their compatriots, taking with them many horses and much booty, which they had taken with their own hands in the battle. And so there was great rejoicing in France, where all rendered up due thanks to God.’ (All translations are my own.) 3 ‘Geoffroi Grisegonelle dans l’épopée’, Romania 19, 1890, 377–93.

Epic and Romance in the Chronicles of Anjou

179

falls into three main sections: first, a single combat with a giant Northman, quaintly named Ethelwulf, whereby Geoffrey also earns his distinctive nickname ‘Grisa Tunica’, ‘Greymantle’ or ‘Grisegonelle’; second, the engagement near Soissons, culminating in the passage we have just considered; and third, Geoffrey’s equally epic duel against an enemy champion, Bertold, in a later war against the Germans. Lot maintained that all three of these episodes are ultimately derived from Old French epic. No romance-language epic poem or poems devoted to Geoffrey Greymantle in fact survive. To support his hypothesis, Lot advanced instead a number of parallels with other surviving chansons de geste. With the combat against the giant Ethelwulf, for example, he compared a similar episode in the Moniage Guillaume, suggesting that at some stage Geoffrey Greymantle had in the putative romance source usurped the role of the stock epic hero Guillaume d’Orange. In the third episode by contrast, Lot placed more emphasis on the presence of epic commonplaces, commenting:4 ‘La soudaineté de la novelle guerre, le conseil des anciens, le duel judiciaire qui assure la couronne au suzerain du champion vainceur . . . tout cela est encore dans le goût de l’épopée féodale.’ Likewise in the case of the second episode, the battle at Soissons, and of our battle passage in particular, Lot adopted a similar approach; he identified as typically epic the hail of weapons which overshadow the sky, the descriptions of wounds and Geoffrey’s heroic courage. With rather greater precision, he also suggested that Geoffrey’s defiant waving of the royal gonfalon in the faces of the Northmen chimes well with the appearance of Greymantle in Chanson de Roland, line 106, as ‘Gefriez d’Anjou, il rei gunfanuniers’.5 All the same, once the shift has been made, as Lot did, from similarities of plot and narrative (as in the first episode) to commonplaces and topoi of epic poetry (as in the Soissons passage), the issue then becomes a much broader one. For twelfthcentury writers and readers the epic tradition was by no means restricted to the romance languages; Latin epic too represented a rich vein for them to tap, and authors like Lucan and Statius – who, as we shall see, have their own part to play in the chronicles of Anjou – were also replete with the commonplaces, conventions and imagery of epic. Indeed, if we return for a moment to our passage, two elements (highlighted in italics) serve to bear this out. First, near the beginning, when battle is joined, we read, ‘iam clipeo clipeus, iam umbone umbo repellebatur’, ‘now shield was clashing against shield and boss against boss’; and later the king bids Geoffrey, ‘rapidum calacaribus urge cornipedem’, ‘spur on your swift steed’. Both phrases exhibit hexameter structure and rhythm (though slightly recast in the case of the first) and also vocabulary (in the second for example, the steed, a cornipes, is more literally, a ‘hornfooted one’, a favourite periphrasis for a horse in Latin epic); both phrases thus immediately betray their origins in Latin poetry. The first is borrowed, with minor recasting, from Statius’ Thebaid, VIII.398, ‘iam clipeus clipeis, umbone repellitur umbo’, whilst the second comes verbatim from Prudentius’ Psychomachia, 253–4, ‘talia uociferans, rapidum calcaribus urget/ cornipedem’. In other words, in the case of at least two of the epic elements to which Lot drew attention, the inspiration came not from Old French epic but directly from two Latin poets, one classical, the other Christian.

4 5

Ibid., 391. Ibid., 390–1.

180

Anglo-Norman Studies XXVI

These verbal reminiscences of Statius and Prudentius serve to remind us that the position is more complex than Lot suggested. He was quite right to identify the influence of epic in the Chronica, and it may well be that the their account of Geoffrey Greymantle did indeed draw at some basic level on lost vernacular sources – without their survival it is impossible to be sure. However, we can be certain that the Latin tradition also, as we have seen, had a part to play in that account. Indeed, the editors of the Chronica, while they acknowledged Lot’s hypothesis, commented on our battle passage that it was ‘évidemment inspiré d’un texte classique’, though they were not able to identify a particular source.6 Their instinct was, however, quite correct; the passage is in its entirety derived from a single source, but it is neither classical nor poetic, though its author is probably best remembered today as a poet working within the classical tradition. In addition to his poetry, however, Baudri of Bourgueil, archbishop of Dol, also wrote, sometime between 1107 and 1130, the Historia Jerosolimitana (hereafter Historia), an account in four books of the First Crusade.7 It is from a passage in the Historia’s narrative of the siege of Antioch that the Chronica drew its vivid battle description almost verbatim (including, that is, the reminiscences from epic that we have just been considering, which are ultimately Baudri’s handiwork).8 The Chronica’s modifications are chiefly limited to replacing Bohemond and his standard-bearer with Hugh Capet and Geoffrey Greymantle, and substituting armies of French and Northmen for crusaders and Turks. This is by no means the only borrowing from Baudri that can be detected in the Soissons episode. Before battle is joined, Hugh Capet harangues his barons and, when they demur, finds support from Geoffrey Greymantle (Chronica, pp. 40–41):9 His auditis, loquutus rex ad optimates sic demum ora resoluit: ‘Videtis, optimates, quod sine profundis singultibus enucleare non possum quantis calamitatibus et incommoditatibus populus Francorum percellitur. Quid plebeios homines commemorem, cum plures ex uobis, ex illustri sanguinis stemmate orti, inedia palleatis et grauis lues Danorum uestros labores 6 7 8 9

Halphen and Poupardin, Chroniques, 41 n. 1. Ed. in Recueil des historiens des croisades: historiens occidentaux IV, Paris 1879, 1–111. Historia, II.15 (pp. 46H–47E). ‘At this news the king, addressing his nobles, embarked at last on the following speech: “Nobles, you see what disasters and difficulties assail the French people. I cannot express them without deep sighs. Why need I mention the common folk when many among you, who spring from the stock of noble blood, are pale with hunger, while this plague of Northmen blights your efforts. Already your fields have been reduced to deserts, tilled by few, if any, ploughs. Do not, I beg you, allow the reputation of the French to be sullied by our inaction. Do not be afraid! We are an unbeaten and unconquerable race. The crisis has arrived: we face a fight, the enemy is near and in strength. Bravest of knights, awake! Behold the moment for battle; ready your hands for war and display the might of your ancestors, as the occasion demands. Why need I talk? Now is the time for each man to speak for himself.” The nobles were unsure what to advise their king. Some of them replied: “We do not advocate fighting now, but desire and suggest that the battle be postponed and this affair settled until we get more troops.” But Geoffrey Greymantle voiced his opinion and gave this advice: “My lord counts and nobles, you are the light, the flower of France, a glorious example of militant knighthood. Fight for your own sakes; lay down your lives for your brothers. For how long will we look on while those who have entrusted themselves to us and our king die unavenged. Thanks be to God! I see that you are all united, nor ought there in this affair to be any dissention in our ranks. There can be no distinction between lord and servant, noble and commoner, rich and poor, knight and footsoldier, unless our plans aid and our help supports those over whom we rule. If the Northmen are going to lord it over me without a fight, then I wish to live no longer. Were we to die without glory, it would be the same as being compared to foolish animals, like brute beasts. All of you should be eager for battle, since you all believe that it will benefit our common safety. I desire it, I demand it ardently. Let us not, I beg you, die like slothful weaklings, lest we be dishonored and disgraced in the eyes of all nations.” ’

Epic and Romance in the Chronicles of Anjou

181

contaminet? Iam dudum agri uestri, in solitudinem redacti, uel nullo uel raro uomere excolluntur. Ne, queso, deturpetur propter nostram neglegentiam laus Francorum. O genus infractum, o gens inuictissima, ne terreamini! Res in arcto est, bellum ex aduerso est, hostis multus in proximo est. Expergiscimini, fortissimi milites! Ecce dimicandi tempus est; bellicosas manus exerite uiresque auitas dum tempus est ostendite. Quid opus est uerbis? Iam nunc sibi quisque loquatur.’ Nobilitas igitur quid regi consuleret anxiabatur. Quorum quidam responderunt: ‘Nullam ad presens pugnandi dare possumus sententiam, sed uolumus et conlaudamus quatinus, ista re indutiata, pugnam procrastinemus, donec maiores uires habeamus.’ Gosfridus tamen Grisa Tunica, suam expromens sententiam, consilium adiecit: ‘Vos, domini consulares et illustres uiri, lux et flos uictoriose Francie, decus et speculum pugnatricis militie, pro uobis ipsis decertate et pro fratribus uestris animas ponite. Nam quoad populum, qui se regi et nobis commisit, inultum mori conspicabimur? Video uos, Deo gratias, omnes unanimes, nec aliquis in hac re debet ab alio dissidere. Quid differt dominus a seruo, nobilis a plebeio, diues a paupere, miles a pedite, nisi nostrum, qui presidemus eis, prosit consilium et patrocinetur auxilium? Si Dani mihi dominabuntur impune, nolo amplius uiuere. Tantundem est si moriamur inglorii ac si comparemur iumentis insipientibus, brutis assimiles animalibus. Omnes quidem anhelare debetis ad pugnam, quia omnes id ad communem creditis profuturum salutem. Ego uero idipsum collaudo, uehementer efflagito, rogo ne sicut segnes moriamur uel imbecilles, ne simus improperium uel omnium infamia gentium.’ Both these speeches are also lifted from Baudri: the king’s opening address combines elements from two speeches of Bohemond, the first delivered during the siege of Antioch, the second before a slightly earlier engagment; and Geoffrey Greymantle’s reply is drawn from another speech by Bohemond, again delivered during the siege of Antioch.10 Indeed, the short passage which links the two speeches with our initial battle passage is also borrowed from Baudri.11 Comparison of the Chronica passage, just cited, with its sources reveals the ways in which the compiler adapted Baudri’s text. The opening of the king’s s speech, for example, follows Baudri quite closely, simply adapting the dramatis personae to the new setting and making minor verbal changes. Conversely, the second part of the speech and the whole of Greymantle’s reply are more freely handled, not so much by making verbal changes as altering the order of sentences, reshuffling the cards in the rhetorical pack. In between, Baudri’s bald statement of the reservations of the other nobles (‘Nobilitas autem de bello disputare ausa est’) is considerably expanded into a plea for the fighting to be postponed.12 These modifications reveal both the care and the freedom with which Baudri’s text has been adapted to its new setting. 10 Respectively Historia, II.19 (pp. 53H–54A), ‘His auditis . . . excoluntur uomere’; II.1 (p. 34A–C),

‘Fortissimi . . . quisque loquatur’; and II.14 (pp. 45E–46A), ‘Nobilitas . . . animas ponite’ (the verbatim borrowings from Baudri are in italics in the passage cited in the main text). 11 Chronica, pp. 41–2. The passage, ‘His dictis, . . . ciebantur’, comes from Historia, II.14 (46D) and ‘Locuti sunt . . . excepturus’ from slightly later in the same episode (46F–H, where it introduces the passage on which our epic battle was based). However, the intervening passage, ‘Venerunt . . . ordinauerunt’, is drawn from a later, quite different passage, Historia, IV.19 (108A). The ease with which the compiler is able to combine disparate elements of Baudri’s narrative is a clear indication of his ready familiarity with his source. 12 While revising this paper, it has come to my attention that the chronicler also draws here on Historia, III.23 (p. 83F), ‘Nullam . . . re pacifice induciata’, another instance of his combining two passages from Baudri.

182

Anglo-Norman Studies XXVI

Nevertheless, the Chronica’s text is in many places so close to Baudri’s as to permit us to speculate about the manuscript traditon on which it drew. In both the passages cited from the Greymantle episode, there are a number of places where the Chronica shares a readings with MS G, sometimes in the company of other manuscripts, but also with G alone.13 It seems likely then that the author of the Chronica was using a manuscript akin to G, if perhaps not G itself (with whose readings he occasionally disagrees). This is sigificant because MS G, viz. Bibliothèque Nationale MS lat 5513, contains a number of additions, some of which, as Nick Paul has recently put it, ‘suit the political and commemorative imperatives of the lords of Amboise’, who were important vassals of the counts of Anjou.14 It would therefore make good sense if G or a close relative were the manuscript used by the Chronica, since the chronicler’s borrowings from Baudri can, as we shall see, be viewed as certainly of a piece with, if not perhaps part of the same process as, the additions in G. To recapitulate, then, almost the whole of the second episode devoted to Geoffrey Greymantle is borrowed verbatim from Baudri, and so transfers narrative from during and before the siege of Antioch to the quite different setting of a tenth-century battle fought rather anachronistically against Northmen. Why did the author of the Chronica borrow from Baudri in this manner? On one level answer is simple, particularly if we assume that the silent appropriation of Baudri’s text was meant to go unnoticed. Baudri’s prime aim in writing his Historia was to improve on the comparatively bald and unpolished narrative of his source, the Gesta Francorum, whose style he judged quite unworthy of its subject-matter. To that end, he applied all his considerable literary and rhetorical skill to impart the high style to his narrative, including of course epic reminiscences such as those of Statius and Prudentius which, as we have seen, were absorbed wholesale into the Chronica. Indeed at the beginning of our second cited passage (Chronica, pp. 40–41, above), the tag ‘ora resoluit’ (in bold) recalls Vergil, Georgic, IV.452; and a different but related kind of reminiscence can be found near the end of the passage, where the phrase ‘comparemur iumentis insipientibus’ (also in bold) is drawn from Ps. 48, verse 13. These echoes of Vergil and the Bible, which like those of Statius and Prudentius, combine classical and Christian reminiscence, thus passed wholesale into the Chronica as part of its author’s extensive verbal borrowing. The elevated style of Baudri’s prose, an effect to which such echoes contribute, is moreover reinforced by the emotive oratory deployed in his speeches, which are carefully constructed and crafted, making effective use at sentence endings of the accentual rhythms of the medieval cursus system. Baudri’s literary embellishments may not endear him to the ‘scientific’ historian, but they were much to the taste of twelfth-century readers. What then was more natural than that, when seeking to inject exciting rhetoric and

13 In particular, G and the Chronica share (in our first passage) the readings, ostendite for ostentate, and

animalibus for pecudibus; and (in the second) the omissions of ait, imperterritus and uideres – the shared reading eos (for illos) may be no more than coincidental. Conversely, the Chronica at one point agree with the other witnesses in reading illius (against G’s cuius), which, if it too is not simply coincidental, may corroborate that G was not itself the Chronica’s source. It is also worth noting that corruption has occurred at two places in first passage from the Chronica: in the sentence beginning ‘Interdum’, Baudri’s ‘In tantum’ makes better sense, whilst the awkward sentence ending, ‘efferatius et instantes’, is seemingly a misreading of Baudri’s ‘efferatius in instantes’ [G et al.]. G itself and its additions are briefly discussed in the preface to the Recueil edition, pp. xiii–xiv. 14 ‘Crusade and family memory in later twelfth-century Amboise: Bibliothèque nationale de France MS lat. 5513 and the Gesta Ambaziensium dominorum’ (unpublished seminar paper).

Epic and Romance in the Chronicles of Anjou

183

battle description in the Chronica, its author should turn for inspiration to Baudri? His imitation represents the sincerest of flattery. What, however, if the chronicler’s borrowing is detected? Do we, because the borrowing is so extensive – as opposed to short, pointed verbal reminiscences such as those from Statius and Prudentius – dismiss him as a hack, tricking out his text in borrowed feathers, or, more anachronistically, condemn him as a mere plagiarist, despite the fact that he was elsewhere quite able and willing to rely on Latin of his own composition? The chronicler’s use of a classical historian perhaps provides an instructive parallel. As the editors of the Chronica have amply demonstrated, the text is studded with similar (if usually somewhat shorter) borrowings from the histories of Sallust; in the earlier sections of the work for instance, Sallustian passages, often striking character analyses, are reapplied to the first counts of Anjou. Again, this lends stylistic depth and gravitas, even if the source is not recognised. Sallust was, however, one of the classical historians best known in the middle ages, and we can be reasonably sure that at least some twelfth-century readers recognised, and were meant to recognise, these allusions, which served to set the Chronica, their author and his subject-matter firmly within the traditions of classical historiography. Let us assume for a moment that twelfth-century readers of Baudri (such as the chronicler himself evidently was) might recognise the Chronica’s borrowings. What were they to make of them? Before considering this question, allow me to try to forestall any doubts about the existence of our posited readers. For me, with my interest in intertextuality, the presence of copious quotations, borrowings, allusions and reminiscences in the work of major historians such as William of Malmesbury presupposes that at least some readers were intended to recognise them. Consider the case of Jean of Marmoutier. Jean was the author of the Historia Gaufredi ducis Normannorum et comitis Andegauorum, a biography of Henry II’s father, which he probably composed between 1173 and 1187; he was also responsible for one of the later recensions of the Chronica, for which he composed an elaborate prologue. We can assume, then, that, as a later redactor, he knew the text of the Chronica well. Early in the Historia Gaufredi, Jean too presents a bloody battle description: ‘franguntur haste fraxinee, emutilantur enses. Iam pede pes teritur, umbone repellitur umbo’, ‘ash-wood spears were shattered and swords broken; foot was pressed on foot, shield-boss was clashing against boss’.15 The shield-bosses, the spears and swords are instantly recognisable from our initial battle passage, but the pressing of foot on foot is new. However, you will recall that the Chronica were (via Baudri) influenced by Statius, Thebaid VIII.398 here; and the following line (399) runs, ‘Ense minax ensis, pede pes et cuspide cuspis’, thus incuding ‘pede pes’, ‘foot pressed on foot’. In other words, when reworking the Chronica, Jean signals that he has recognised the Statian echo by introducing a further, independent echo from the very next hexameter. Jean of Marmoutier for one was evidently sensitive to intertextual borrowing. What we cannot unfortunately know is if he was trumping the Chronica, or Baudri, or both. To return now to our initial question: if the Chronica’s borrowing from Baudri is recognised, the intertextual resonanances are considerable. Capet and Greymantle effectively become precursors of Bohemond and the crusading lords. Their battle against the Normans (who are, you will recall, called Dani and are suspiciously Viking and pagan-seeming) thus takes on some of the qualities of a crusade (note that Geoffrey is described as ‘sancte crucis signo munitus’). Indeed, their battle becomes 15 Halphen and Poupardin, Chroniques, 182.

184

Anglo-Norman Studies XXVI

one link in the history of salvation, though with a distinctively French slant. Capet and Greymantle’s heroism not only mirrors that of Bohemond and the crusaders, in a sense it made the later crusade possible: without the French defeat of Norse attacks like that at Soissons (and of earlier Vikings), the chronicler implies, there might well have been no crusade. And if Lot was right to suggest that an ultimate inspiration for Greymantle’s exploits was chansons de geste, then the historical sweep becomes yet broader. If we invoke Old French epic, then the victories of Charlemagne (or even Charles Martel) can be seen as part of the same historical process; the defeat of the Arabs in the eight century, and later of the Northmen prepare for the ultimate triumph of the First Crusade, and Graymantle is one link in the chain. The kudos is moreover reciprocal. Just as Baudri’s elevated Latin style made new epic heroes of Bohemond and the crusaders, so the Chronica ensure that Greymantle’s exploits foreshadow, and are simultaneously enhanced by, the crusaders’ epic stature; at the same time, Greymantle’s own arguable connections with vernacular epic serve to link the crusaders back to a glorious French past. To help convince you that borrowings such as that in the Chronica may well work in this allusive way, I should like now to pass on to our second main text. Before we leave the Chronica, however, I should remark that borrowings from Baudri are not limited to the section devoted to Geoffrey Greymantle; they are also prominent in a later chapter on Geoffrey Martel (d.1060). Since I intend to discuss these extensive borrowings elsewhere, I shall restrict myself here to a single short example which has decisive bearing on the question of the source manuscript of Baudri employed by the Chronica. The following passage (Chronica, p. 61) is part of another battle passage which relies heavily on Baudri:16 Globus etiam Turonorum militum subsequentium dominum suum multos strauit et uexillum ipsius ducis prosternit, quod pedites sequentes rapuerunt et retinuerunt quod non mediocrem eis incussit timorem. The chronicler here draws on Historia, IV.20 (109B): [Globus etiam militum subsequentium dominum suum multos strauit et stantarum [sic] prosternit, quod pedites, uiriliter equites sequentes, rapuerunt et retinuerunt. G only] Quod non mediocrem gentibus incussit timorem. However, as you can see, this particular sentence is for the most part taken from one of the unique additions which, as I have already observed, are found in MS G, viz. BN lat. 5513. In this instance, therefore, we can be sure that the chronicler was using either G itself or a close relative which included its additions; and, furthermore, those additions had already been made to the G-recension of Baudri’s Historia before it served as a source for the Chronica (possibly sometime in the mid 1150s). Embedded sources have their uses for text history. Gesta Let us move on now from the Chronica to the closely related Gesta Ambaziensium dominorum (hereafter Gesta). This history of the lords of Amboise is transmitted in manuscript with the Chronica, and directly influenced two of the latter’s surviving 16 ‘The mass of knights from Tours, following their lord, killed many men and overthrew the standard of

the leader himself, which was snatched up and captured by the supporting footsoldiers, filling the enemy with great terror.’

Epic and Romance in the Chronicles of Anjou

185

recensions. Like the Chronica, the Gesta borrow frequently from poetic and historical sources (often, indeed, the same ones as the Chronica). It is not therefore surprising that the influence of Baudri’s Historia is again apparent in the Gesta (which incidentally also probably employed the G-recension of Baudri’s text, though the borrowings I have so far detected do not permit absolute certainty). Here I should like consider the contribution of two borrowings in the Gesta’s account of Sulpice II of Amboise, whose recent capture in 1153 by Thibaud of Blois is lamented in the preface of the work. Since the lordship and fall of Sulpice later provide the finale of the narrative, they effectively frame the text. The first, and shorter, borrowing helps to set the moral tone for Sulpice’s fall (Gesta, p. 129):17 Sepe pro Supplicio populus suus lamentabatur, qui plaudente fortuna multoties triumphauerat. Tales tamen sunt bellorum euentus, tales sunt uicissitudines et hominum et temporum: nulli unquam semper successit feliciter, nemo umquam de continua prosperitate uel letabitur uel letatus est. Hac de re et timenda et cauenda est in prosperis aduersitas, et speranda est in aduersis prosperitas. Such observations on the vicissitudes of fate immediately recall the philosopher Boethius, who is a favourite source of both the Gesta (and the Chronica); but, although a lengthy citation of Boethius’ De consolatione philosophiae does indeed immediately follow in the Gesta, these initial observations come directly from Baudri.18 With Baudri’s help, then, Sulpice is represented as an exemplum of the mutabilty of fate, a great man broken on fortune’s great wheel. A second, and much longer borrowing from Baudri reminds us that it is never a good idea to disregard a mother’s advice (Gesta, pp. 126–7):19 Dum predicta aguntur, Elizabeth, mater Supplicii, senex et plena dierum, utpote presaga futurorum, de his que audierat pro filio gemebunda, Ambazio in domo sua iuxta ecclesiam sancti Thome ipsum acciuit. Flebilis igitur ad filium aiebat: ‘Ut quid me inconsulta, fili, negotium adgessus es bellicum? An quoiam sum decrepita me disipuisse putasti? Crede mihi, uiget sensus effetis in uisceribus, 17 ‘Sulpice’s subjects often lamented for him, since he had so frequently triumphed when fortune smiled

on him. But such is the outcome of war, such the mutability to which men and time are subject: no one has ever enjoyed unbroken success, no one has ever rejoiced, or will ever rejoice, in unchanging good fortune. We should therefore fear and beware of failure in times of success, and hope for success in times of failure.’ 18 Historia, II.16 (p. 49G–H). 19 ‘While this was going on, Elizabeth, Sulpice’s mother, summoned him to her home beside the church of St Thomas in Amboise. She was old and full of days and groaned for her son over what she had heard, since she foresaw the future. In tears she said to him: “Why, son, have you undertaken an affair of arms without consulting me? Did you think that I have grown foolish because I am so old? Believe me, my intellect is still sharp within this worn body, and my loose skin, wrinkled with age, still nourishes a lively mind. After all, you could not have revealed your plan to anyone who loved you more dearly or could advise you more prudently. What can compare with a mother’s love? Robert, the French king’s brother, and many other nobles are accompanying Thibaud from Blois with an invincible force. You should have considered what a serious affair you had undertaken; you should have recognised the invincibility of the nation you would be fighting; you should have realised that the forces you had summoned from all sides cannot be relied upon. Flesh of my flesh and bone of my bone, your fame has spread far in our own lands, but your arm has not yet been tested against these foes. If you took your mother’s advice, you would desist from this rashness and disengage yourself and your men from this presumptious enterprise. It is not clear on which side Mars has smiled, but I am more fearful for ours.” When the old woman, overcome by tears, had finished speaking, her son made a short and boastful reply: “I am amazed, mother, how boldly you predict our enemies’ victory, when their spirit cannot be compared with that of our nation. My men too are powerful and warlike. Our enemies are human beings just as we are. So, dearest mother, await the outcome and hope for the better.” ’

186

Anglo-Norman Studies XXVI

et laxa cutis et ruga senilis uiuacem adhuc fouet animum. Denique nulli poteras reserare consilium que te uel arctius diligeret uel que tibi discretius consuleret. Quid enim affectui materno comparari poterit? Robertus, frater regis Francorum, et alii multi proceres cum insuperabili militia, cum Teobaldo Blesis ueniunt. Debueras preponderare quam grande aggressus es negotium: decuerat te metiri cum qua insuperabili gente dimicaturus es; oportuerat te perpendere quod copie quas undiquaque corrogasti nullatenus tibi fideles sunt. Nostris in regionibus, os meum et caro mea, nomen tuum satis dilatatum est, sed brachium tuum in istos nondum approbatum est. Porro, si matri consulenti adquiesceres, ab hac temeritate absisteres teque gentemque tuam huic presumptuoso labori subtraheres. Dubium enim cui parti potius Mars arriserit, nostris tamen magis timeo.’ Veterana illa, in lacrimis deficiens, loquendi finem fecerat, cum filius iactabundus sic paucis respondit: ‘Miror, mater, qua fronte de uictoria prenuntiaris, cum ipsi animis genti nostre non debeant comparari. Et ego uiros potentes et bellicosos habeo; ipsi uero homines sunt sicut et nos. Idcirco, mater dulcissima, euentum rei meliora desiderans exspecta.’ The anonymous author, much as we have already seen in the Chronica, incorporates this affecting interview almost verbatim from Baudri.20 Evidently, the pathos of Baudri’s imagined exchange between mother and son was one motive for its incorporation. What is interesting is the change in context. In the Historia, it is the commander of the crusaders’ enemies, Corbarannus as Baudri calls him, who rejects his canny old mother’s advice. The effect of alligning Sulpice with Corbarannus in this way is twofold. First, it clearly works to the former’s disadvantage. Though Baudri’s references to Corbarannus’ ‘paganism’ are excised, the implicit parallel between infidel and Christian counts evidently against Sulpice, just as his boastful dismissal of his mother does. The second effect is more subtle. Baudri was at pains to paint a convincing psychological portrait of Corbarannus; he is a doomed anti-hero, a ‘desert fox’ to repel yet fascinate his readers. If the borrowing is recognised, Sulpice too grows in complexity. His fall, lamented in the preface, is presented as a disaster for Amboise, yet, as well as fate, Sulpice’s own character has also unwitting played a part in his downfall; he is presented as just as doomed as was Corbarannus. In addition to pathos, this episode thus injects a note of the tragic into the Gesta. I maintain, then, that like the Chronica, the Gesta borrow extended passages from Baudri in a similarly allusive way. In case, however, such borrowings from Baudri, no matter how high or emotive their style, threaten to take us too far from the epic of my title, I shall conclude with two examples of allusive borrowing from Lucan. A glance at the notes in the standard edition of the Gesta shows how fond the author was of citing classical poetry, and Lucan’s epic in particular. In fact he draws on Lucan even more extensively than has been recognised, as the following unnoticed examples will demonstrate. In the first, Lucan provides him with much of the detail of another battle narrative, in this case between Sulpice and the rival forces of Renaud of Châteaurenaud and Jean, count of Vendôme (Gesta, pp. 120–21):21 20 Historia, III.4 (pp. 62F–63G). 21 ‘Therefore Sulpice, trusting in the strength of his men, often troubled the spirits of the foresaid men by

laying waste their territory; and, because ease always causes uncertainty of mind and great fear can be concealed by daring, he sent his troops into the lands of Renaud. When this news reached Jean, count of Vendôme, who was at Châteaurenaud with his forces, he was encouraged, thinking that he was predestined for martial glory and victory. So he stealthily hurried his troops on, concealing them as much as possible by an enforced silence. He himself advanced with a vanguard to join battle first and draw out the enemy, having held back his main force in a hollow valley; Renaud followed slowly with the remaining forces. But

Epic and Romance in the Chronicles of Anjou

187

Idcirco Supplicius, robore suorum confisus, sepe predictorum animos terram ipsorum uastando sollicitauit; et quoniam otia semper uariam mentem dant et audendo magnus timor tegitur, agmina sua in terram Raginaldi emisit. Que postquam sunt a Johanne comite audita, qui Castro Raginaudo cum suis copiis erat, estimans quod gloria belli sibi reseruata esset et uictoria, letus efficitur. Itaque agmina sua furtim rapit atque per iussa silentia quoad potest obscurat. Ipse uero, ut prima prelia lacessat et eliciat, retentis maioribus uiribus in caua ualle, in primo agmine processit; Raginaudus uero de Castro cum ceteris pedetemptim subsequitur. At Supplicius agmina suorum militum ex diuersis partibus aggregata apertis campis instruxit, quorum agmine emisso, campi sono statim tremuere terraque soluta turbine pulueris tenebras traxit. Vindocinenses econtra, dum miscere manus et precedere licuit, parum stetere; sed cito agmen eorum frangitur, cum in primo impetu Johannes consul eorum capitur. Ut uero fortuna belli in pedites incubuit, qui fugere non potuerunt, ligati Supplicio et suis leta spectacula prebuerunt; sicque, paucis peremptis multisque captis, uictores ad propria redierunt et Iohannem captum in arce Calui Montis incluserunt. If one compares the following passages drawn from Lucan, De bello ciuili, IV 702–85, it is immediately apparent that the chronicler has made several elements of his battle closely resemble Lucan’s: Audendo magnus tegitur timor; arma capessam Ipse prior. Campum miles descendat in aequum, Dum meus est; uariam semper dant otia mentem. Tristia sed postquam superati proelia Vari Sunt audita Iubae, laetus, quod gloria belli Sit rebus seruata suis, rapit agmina furtim, Obscuratque suam per iussa silentia famam.

(702–4)

(715–18)

Mittitur exigua qui proelia prima lacessat Eliciatque manu . . . Ipse caua regni uires in ualle retentat. (720–21 and 724) At uagus Afer equos ut primum emisit in agmen, Tum campi tremuere sono, terraque soluta, Quantus Bistonio torquetur turbine, puluis Aera nube sua texit traxitque tenebras. Ut uero in pedites fatum miserabile belli Incubuit . . . (765–70) . . . neque enim licuit procurrere contra Et miscere manus. (772–3) Non tam laeta tulit uictor spectacula Maurus Quam fortuna dabat. (784–5) Sulpice, having assembled his troops of knights from various directions, drew them up on the open field; when they charged in a body, the fields at once shook with the noise and the ground, stirred up in a cloud of dust, was overshadowed. Their opponents from Vendôme resisted for a short time, while they could trade blows and advance, but soon their army was broken, Count Jean having been captured in the first attack. And when the foot soldiers had to face the fortunes of war, they could not escape and, having been bound, they provided a fine spectacle for Sulpice and his men. And so with few casualties and many captives, they returned home victorious and kept Jean prisoner in the castle of Chaumont.’

188

Anglo-Norman Studies XXVI

These passages certainly reinforce the lesson of how unwise it would be to assume that medieval chronicles necessarily provide the historian with exact and entirely trustworthy details of the events and tactics of battles which they record. What I would like to consider, however, is once again the interplay beween the Gesta’s borrowings and their new context. In Lucan, the context is of course that of civil war. Curio, Caesar’s general in Africa, is opposed to the wily African prince Juba, whose cavalrymen trick and slaughter Curio’s legionaries. The Gesta’s borrowings reflect the fractured and confused situation in this civil war. Initially Sulpice is cast in the role of the Roman Curio, concealing fear with daring and avoiding the dangers of ease (cf. De bello ciuili, IV.702–4), whilst Jean of Vendôme’s stealthy advance and spoiling tactics are made to recall those of Juba (715–18 and 720–23). Once battle is joined however, the roles are reversed: the charge of Sulpice’s men mirrors that of Juba’s (765–8), and the defeat of the enemy foot-soldiers recalls that of Curio’s legionaries (769–70 and 784–5). One might see this as an indication that the borrowing is not intended to work on an intertextual level, that the epic diction and exciting details are all that count. However, it is, I think, suggestive that in this passage, Sulpice, for part of the time at least, plays the role of an African commander whom Lucan depicts as a traditional enemy of Rome. This alignment with ‘the other side’ is by no means surprising if we recall that with Baudri’s Corbarannus, which it in a sense anticipates. Moreover, and more subtly, whichever role Sulpice plays, either way he is, or will be, a loser. He starts as the doomed Curio; he finishes as the seemingly victorious Juba. But Juba’s triumph is only temporary; he too will eventually be defeated by an avenging Caesar. Likewise Sulpice’s success will not last. We know that in the end he will be captured at Maindray and die in captivity, and, as we saw in his interview with his mother, his arrogance and overconfidence, fed by seeming victories such as that over Jean and Renaud, will play a large part in his downfall. In our second passage, past and present, the literary and the ‘real’ interpenetrate and blur. The narrative of the Gesta ends with the author musing on the sixth sense people can have about impending disaster (Gesta, p. 131):22 Mirum et uerum est quod mens presaga malorum homini data est. Nam sub quacumque parte mundi Ambasiensis affuit, eo die quo Supplicius capitur meret et causas ignorat animumque dolentem corripit, nesciens quid in proditione Mindraii perdat. Quidam namque iuuenis ex nostris, qui hospes in Apulia manebat et alter Andegauis mulierque Calui montis in Bitirico pago existens hec sibi contigisse mihi retulerunt. At first sight, it seems that we are dealing here with something drawn directly from the author’s own experience: three contemporaries reporting the unexplained feeling of doom they had when Sulpice, unbeknown to them, was captured at Maindray. These witnesses may well have been real, but the way in which the chronicler expresses their feelings of unease, and his speculation about its cause is borrowed directly from Lucan (De bello ciuili, VII.185–91):23 22 ‘It is both surprising and true that mankind is endowed with a mind that can foresee misfortune. For on

the day that Sulpice was captured, in whatever part of the world natives of Amboise were living, they felt grief without knowing the reason, and rebuked their sorrowing hearts, unaware what they were losing through treachery at Maindray. A young man of Amboise who was staying in Apulia, and another in Anjou, and a woman from Chaumont resident in Berry have all told me that they had this experience.’ 23 ‘Why is it surprising that peoples who face their final day have felt mad fear if mankind is endowed with a mind that can foresee misfortune? Romans, whether they were staying in Phoenician Cadiz or

Epic and Romance in the Chronicles of Anjou

189

Quid mirum, populos quod lux extrema manebat, Lymphato trepidasse metu, praesaga malorum Si data mens homini est? Tyriis qui Gadibus hospes Adiacet Armeniumque bibit Romanus Araxen, Sub quocumque die, quocumque est sidere mundi, Maeret et ignorat causas animumque dolentem Corripit, Emathiis quid perdat nescius aruis. According to Lucan, Romans everywhere felt irrational grief during the fighting between Caesar and Pompey at Pharsalia, even though they were quite unaware that the battle was taking place; their experience thus neatly parallels and foreshadows that of our chronicler’s twelfth-century psychics. For our chronicler, the experience thus becomes a universal, the example from the first-century B.C. validating and enriching that from his own period. Moreover, as a patriotic man of Ambroise, he is enabled by his reminiscence of Lucan to present the downfall of Sulpice as just as catastrophic for his own province as, according to the poet, the battle of Pharsalia was for the Roman republic. For Lucan, Pharsalia sealed the fate of his flawed, yet sympathetic anti-hero, Pompey. By subtly linking Sulpice and Pompey, yet another loser, the chronicler is able to reinforce the tragically flawed nature of the lord of Amboise. Our passage of Lucan will eventually conclude by proclaiming his own role in ensuring Pompey’s posthumous fame, as he promises to win over his readers, ending (line 213) ‘Magne, fauebunt’ – ‘they will be on your side, Magnus’. Our chronicler, by choosing to invoke Lucan at the end of his text, adopts a similar strategy: he presents Sulpice’s fall as a tragic disaster, a fall, as we have seen, by no means unmerited, but which wrung the hearts of men and women from Amboise wherever they were, and which will wring the hearts of his readers in Amboise and maybe elsewhere. The note of pathos on which the chronicler chooses to end his narrative owes more to Lucan than just its form; like Lucan’s, it also is also designed to win his audience’s sympathy. The chronicler’s echo of Lucan, then, is a vital part his rhetoric of persuasion; if recognised, it triggers associations which increase its impact. At the same time, the chronicler also uses Lucan to validate his text and indeed vice versa: recall that he began ‘it is both surprising and true’ (my italics). In the Chronica, as the Gesta, history and literary history become inextricably interwined, and that is what I have tried to demonstrate and explore in this paper.

drinking the Araxes in Armenia, beneath whatever clime or star of the world, felt grief without knowing the reason, and rebuked their sorrowing hearts, unaware what they were losing on the fields of Thessaly.’