Colonial England, 1066-1215 1852851406, 9781852851408

The process of colonization that followed the Norman Conquest defined much of the history of England over the next 150 y

437 14 22MB

English Pages 362 Year 1997

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Cover
Contents
Acknowledgements
Abbreviations
Preface
Introduction
1 Colonial England, 1066–1215
2 Domesday Book, 1086–1986
3 1086
4 Domesday Book and Beyond
5 Feudalism Revisited
6 The Introduction of Knight-Service in England
7 The Carta of Richard de La Haye, 1166
8 Politics and Property in Early Medieval England
9 Feudal Society and the Family in Early Medieval England, I: The Revolution of 1066
10 What's in a Name? Family Nomenclature and the Norman Conquest
11 Feudal Society and the Family in Early Medieval England, II: Notions of Patrimony
12 Feudal Society and the Family in Early Medieval England, III: Politics and Patronage
13 Feudal Society and the Family, IV: The Heiress and the Alien
14 The Treaty of Winchester 1153
15 Magna Carta, 1215–1217: The Legal and Social Context
16 The Casus Regisr: The Law and Politics of Succession in the Plantagenet Dominions, 1185–1247
Index
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
Y
Пустая страница
Recommend Papers

Colonial England, 1066-1215
 1852851406, 9781852851408

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

COLONIAL ENGLAND 1066-1215

This page intentionally left blank

COLONIAL ENGLAND 1066-1215

J.C. HOLT

THE HAMBLEDON PRESS LONDON

AND

RIO

GRANDE

Published by The Hambledon Press, 1997 102 Gloucester Avenue, London NW1 8HX (UK) PO Box 162, Rio Grande, Ohio 45674 (USA) ISBN 1 85285 140 6 © T. C. Holt 1997 A description of this book is available from the British Library and from the Library of Congress

Typeset by The Midlands Book Typesetting Company, Loughborough Printed on acid-free paper and bound in Great Britain by Cambridge University Press

Contents

Acknowledgements Abbreviations Preface Introduction 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Colonial England, 1066-1215 Domesday Book, 1086-1986 1086 Domesday Book and Beyond Feudalism Revisited The Introduction of Knight-Service in England The Carta of Richard de La Haye, 1166 Politics and Property in Early Medieval England Feudal Society and the Family in Early Medieval England, I: The Revolution of 1066 What's in a Name? Family Nomenclature and the Norman Conquest Feudal Society and the Family in Early Medieval England, II: Notions of Patrimony Feudal Society and the Family in Early Medieval England, III: Politics and Patronage Feudal Society and the Family, IV: The Heiress and the Alien The Treaty of Winchester 1153 Magna Carta, 1215-1217: The Legal and Social Context The Casus Regisr. The Law and Politics of Succession in the Plantagenet Dominions, 1185-1247

Index

vii ix xi xiii 1 25 31 59 71 81 103 113 161 179 197 223 245 271 291 307 327

This page intentionally left blank

A cknowledgements

The author and publisher are grateful to the original publishers for permission to reprint essays that appeared originally in the following places: 1 2 3 4 5 6 7 8

9 10 11 12 13 14 15 16

This appears here for the first time. Cambridge Magazine, 19 (1986), pp. 23-28. Domesday Studies, ed. J.C. Holt (Woodbridge, 1987), pp. 41-64. Foreword to F.W. Maitland, Domesday Book and Beyond (Cambridge, 1987), pp. v-xvii. Economic History Review, second series, 14 (1961), pp. 333-40. Anglo-Norman Studies, 6 (1983), pp. 89-106. English Historical Review, 84 (1969), pp. 289-97. Past and Present, 57 (1972), pp. 3-52; 65 (1974), pp. 130-32; reprinted in Landlords, Peasants and Politics in Medieval England, ed. T.H. Aston (Cambridge, 1987), pp. 65-114. World Copyright: The Past and Present Society, Corpus Christi College, Oxford, England. Transactions of the Royal Historical Society, fifth series, 32 (1982), pp. 193-212. Stenton Lecture, University of Reading (1981). Transactions oj'the Royal Historical Society, fifth series, 33 (1983),pp. 193-220. Transactions of the Royal Historical Society, fifth series, 34 (1984), pp. 1-25. Transactions of the Royal Historical Society, fifth series, 35 (1985), pp. 1-28. The Anarchy of King Stephens Reign ed. Edmund B. King (Oxford, 1994), pp. 291-316. Law in Medieval Life and Thought ed. Edward B. King and Susan J. Ridyard, Sewanee Mediaeval Studies, 5 (Sewanee, Tennessee, 1990), pp. 1-19. Law in Mediaeval Life and Thought ed. Edward B. King and Susan J. Ridyard, Sewanee Mediaeval Studies, 5 (Sewanee, Tennessee, 1990), pp. 21-42.

This page intentionally left blank

Abbreviations

Anglo-Saxon Wills Annales ESC BN Cal. Charter Rolls Cal. Docs. France

Cal. I.RM. Chronica majora Chron. Abingdon Complete Peerage Curia Regis Rolls DB De legibus

DNB EHD

EHR Exon EYC Gesta regum

D. Whitelock, ed. Anglo-Saxon Wills (Cambridge, 1930) Annales: economies, societes, civilisations Bibliotheque Nationale Calendar of the Charter Rolls Preserved in the Public Record Office, 6 vols (London, 1903-27) Calendar of Documents Preserved in France, ed. J.H. Round (London, 1899) Calendar of Inquisitions Post Mortem Preserved in the Public Records, 14 vols (Rolls Series, London 1904-54) Matthei Parisiensis Chronica Majora, ed. H.R. Luard, 7 vols (Rolls Series, London 1872-83) Chronicon Monasterii de Abingdon, ed. J. Stevenson, 2 vols (Rolls Series, London 1858) Compkte Peerage of England, Scotland, Ireland, newedn., ed. V. Gibbs and others, 12 vols (London, 1910-59) Curia Regis Rolls Preserved in the Public Record Office, 17 vols (London, 1922-91) Domesday Book Tractatus de legibus et consuetudines regni Anglie qui Glanvilla vocatur, ed. G.D.G. Hall (London, 1965) Dictionary of National Biography English Historical Documents, 12 vols, general ed. D.C. Douglas (London, 1953-), voli, ed. Dorothy Whitelock (1955,1979); vol. ii, ed. D.C. Douglas and G.W. Greenway (1953, 1981) English Historical Review Liber Exoniensis Early Yorkshire Charters, eds W. Farrer and C.T. Clay, 13 vols (Yorkshire Archaeological Society, record series, extra series, 1914—16) William of Malmesbury, Gesta regum, ed. W. Stubbs, 2 vols (Rolls Series, London 1887-89)

X

Colonial England, 1066-1215

HMC Historical Manuscripts Commission Orderic Vitalis, ed. Chibnall The Ecclesiastical History of Orderic Vitalis, ed. Marjorie Chibnall, 6 vols (Oxford, 1969-85) Pipe Roll Pipe Roll 31 Henry I (London, 1929); Pipe Roll 2-3-4 Henry II (London, 1930); Pipe Roll 5 Henry //to 34 Henry II, Pipe Roll Society, vols 1-38 (1884-1925); Pipe Roll 1 Richard I, Record Commission (London, 1844); Pipe Roll 2 Richard I to 5 Henry III, Pipe Roll Society, new series, vols 1-48 (1925-1990) PRO Public Record Office RADN Recueil des actes des dues de Normandie, 911-1066, ed. Marie Fauroux (Caen, 1961) Red Book The Red Book of the Exchequer, ed. Hubert Hall, 3 vols, Rolls Series, London, 1986) Rot. Chartarum Rotuli Chartarum, ed. T.D. Hardy, Record Commission (London, 1837) Rot. Curiae Regis Rotuli Curiae Regis, ed. Sir Francis Palgrave, 2 vols (London, 1835) Rotuli de liberate ac de misis etpraestitis, ed. T.D. Hardy, Rot. de Liberate Record Commission (London, 1844) Rotuli de oblatis et finibus, ed. T.D. Hardy, Record Rot. de Oblatis et Finibus Commission (London, 1835) Rotuli litteratum clausarum, ed. T.D. Hardy, 2 vols, Rot. Litt. Glaus. Record Commission (London, 1833-34) Rotuli litterarum patentium, ed. T.D. Hardy, Record Rot. Litt. Pat. Commission (London, 1835) Regesta regum Anglo-Normannorum, 4 vols, vol. i, ed. RRAN H.W.C. Davis (Oxford, 1913); vol ii, ed. C.Johnson and H.A. Cronne (Oxford, 1956); vols. iii and iv, ed. H.A. Cronne and R.H.C. Davis (Oxford, 1968, 1969) Victoria County History VCH

Preface

The essays in this volume were written over a period of more than thirty years. They have been arranged in an order fitting to the title of the book. Work on them was more continuous than the dates of publication indicate. For example, the paper on the 'Introduction of Knight-Service' was presented, already near its final form, to an Anglo-American Conference in London a dozen years or so before the publication date of 1983. It was foreshadowed in an address on Early Feudal Lincolnshire delivered to the Lincoln Record Society in 1960 or thereabouts; this is not included. Likewise, 'Politics and Property in Early Medieval England' was read to, and improved by, many tolerant audiences both in England and the United States from 1966 onwards. I have left these and the other early writings couched in the terms then current. Quite apart from the impossibility of changing their ambience, there is some interest in leaving the language of history as it then was. In any case, thirty years on I might alter the terminology here and there to suit current tastes, but not the arguments or the use of evidence. I have made few alterations, amounting to little more than corrections of factual or typographical slips. To mention and discuss all that has been written on these topics since the essays were composed is impossible within the scope of a volume such as this. I have therefore confined myself to appending notes to some of the essays; these are concerned solely with later work specifically relevant to my arguments. I am grateful to all those who took part in discussions of many of these papers and helped me to refine what I was trying to say. Any remaining error or muddle is my own. Mr Paul White of the Midlands Book Typesetting Company was particularly helpful in dealing difficulties arising in the proofs. Finally I am deeply grateful to my wife, Dr Elizabeth Holt, who helped with the proofs and the index. Fitzwilliam College Cambridge September 1996

J.C. Holt

This page intentionally left blank

Introduction

Collections of essays frequently take their tide from the first item, especially when, as here, it has not been published before. In this case also something further is intended, for all the essays in this volume are concerned in one way or another with conquest, settlement and the longer processes of innovation and compromise which began in 1066.1 I have carried the collection to 1215 and even later. Magna Carta put a seal on many of the developments of the previous 150 years. The Charter was also the starting point of much that was new. So it is a convenient terminus to the colonial period of English history.2 Why 'Colonial'? What results from the use of the word? One effect is that the formal, somewhat artificial terms in which discussion of the Conquest has usually been set - continuity or change, English versus Norman influences, and so on - cease to matter very much. Only one of these essays, the earliest, deals with these traditional problems, and that is a review-article (Chapter 5) with an extraneous agenda. Instead the Conquest becomes a question of necessity and convenience. To take but one example, the followers of Duke William, within a few years of 1066, supplanted the Anglo-Saxon aristocracy south of Tees. With them they brought a dynastic family structure radically different from that known in England hitherto. It implanted new practices of inheritance, and ultimately a new law governing the tenure, devolution and descent of landed property. This was a break with the English past- a revolution (Chapter 9). But the establishment of the dominion which underlay it could only be achieved within the existing administrative and legal structure. Hence English jurors in English courts of hundred and shire bore witness which determined the detail of the territorial setdement: past ownership, bounds, disputed lands, disjuncture between land and legal dependence. The Conqueror, too, responded to necessity and made grants to his men of the whole or part of the estates of English nobles, and of the whole or part of existing administrative units - shires, hundreds, wapentakes. So the allocation of land reflected more ancient tenures and structures. And it was more than an administrative process. William usurped 1

1 am not forgetting French intrusions and influences prior to 1066. But 1066 is more than a convenient starting date: without Hastings the story would have been very different. 2 I am concerned mainly with insular developments. Hence 1215 is a much better date than 1204, which marked the break with the Norman homeland. In any case, this was neither so sudden nor so complete as is often argued.

xiv

Colonial England, 1066-1215

the throne of England and he and his men the realm. Legitimacy became beautiful in their sight. William traced it to the Confessor, his men to antecessores throughout the land. So by 1086 English jurors might well bear witness to the legitimacy of the Norman intruder - his claims to the lands of an antecessor, or whether he had occupied the land by force, or whether they had seen a writ to warrant his tenure. Now whether English or Norman influences predominated in this process is hard to determine; and there is no need to try, so long as the complicated process of settlement is properly understood; so long, also, as a salient fact is kept in mind - whatever the most convincing and interesting link with the English past, there was a conquest, an overwhelming victory for the conquerors and a calamity for the vanquished. The convenience of the conquerors determined what survived. Was England much changed as a result at the end of the initial settlement say by 1100? In its overall management by the king and the new nobility, yes, though the new Norman king still called on the powers of the old Anglo-Saxon monarchy; in its physical traits - castles, cathedrals, monasteries, strikingly so; in its trade and towns considerably; in its rural economy not fundamentally, though probably more than is apparent; in its local government and local courts, scarcely at all, although where there was proprietary jurisdiction, there were new owners with new ambitions and assumptions. So the Anglo-Scandinavian world survived intact, or was blended with the Norman, or was overlain or replaced thereby. The whole process was subject to geographic variation. In the north especially conquest and settlement came slower and later and with much more fusion of old with new. All these matters concern but one aspect of the settlement, namely the relationship between the victors and the vanquished. Language apart, this soon lost importance. There was another matter, vital from the start, which soon predominated, namely the interrelations of the conquerors, interrelations within the family, between families, between lord and vassal and between the king and all the rest; relationships expressed in terms of control of property, the security of tenure and inheritance, the definition of services and incidents, and the dependence of the whole upon the royal will. Almost all the essays in this book deal with these aspects of colonial England. In memory, perhaps also to serve later political ends, all of this was soon telescoped and simplified: We decree likewise and firmly enjoin, that all earls, barons, knights, tenants by sergeanty and all free men throughout the whole of our realm aforesaid shall always keep themselves well supplied with arms and horses, as is fitting and right, and they shall always be ready and well prepared to perform and fulfil the whole of their service to us, whenever the need shall arise, in accordance with their legal duty to us, by virtue of their fiefs and holdings, and in accordance with what we have ordained for them, with the general approval of our whole realm aforesaid, and have granted

Introduction

xv

and conceded to them as fiefs by hereditary right. This decree shall not be violated in any way, under pain of incurring the full fine for insubordination to us.3

That was how the Conqueror's law looked to a legal writer, with axes of his own to grind, working in London c. 1200. It is not entirely delusory. Many of the baronies which our legal writer would have known had been established by the Conqueror; so also had many of the quotas of knight-service due to the king (Chapter 6). But it was not until 1100 that the varied problems which were bound to arise in the establishment of a new body politic began to surface. The so-called coronation charter of Henry I revealed and sought to solve some of these second-generation problems of the Conquest.4 The settlers no longer needed to worry overmuch about control of the English: that had been secured by military dominance and the measures taken by the Conqueror. The topics of the day were now reliefs, rights of marriage, dower, testamentary dispositions, amercements. Alongside these it soon became clear that feudal notions about justice, which turned on men being justiciable in their lords' courts, were not going to work; alienation, sale and the accidents of descent were producing too many cross-tenancies. Henry Fs ordinance of 1109-11 brought such cases into the county courts. These were the seeds of the common law, which was a unique response to particular colonial circumstances. The outcome was that by 1215, when Magna Carta pulled the threads together, the realm of England had become very insular in character, idiosyncratic beyond recall within the western world. It has remained so. Some early schemes were dropped. Monetagium, for example, the Norman fouageor recoinage tax, was abandoned in the 'charter' of Henry I. Some changed out of all recognition. The quotas of knight-service imposed by the Conqueror, essentially a means of underpinning the Conquest, became in time a fiscal imposition and a convenient method of assessment. Some changes were quite beyond human control. No one could have predicted that the descendants of the French invaders would end up speaking English or that the English they would come to speak would be so different from the languages of Anglo-Saxon England. There was no way of regulating this slow linguistic shift, although late in the day, in the fourteenth century, futile attempts were made to do so. We still need some overall explanation of why the language of the conquered ultimately predominated in England, but not in Scotland, Wales or Ireland. Royal power bestrode the country, especially in the immediate aftermath of the Conquest. William was well enough aware that not all his followers were Norman. Throughout his written ordinances they are contrasted with the English not as Normans but as Frenchmen. Nevertheless, when it came to practical measures it was Norman precedent and no other which he applied: 3 Willelmi articuli retractati, cap. 8. These appear in the much debated London legal collection of 1200-1204. See F. Liebermann, Uber die Leges Anglorum saeculi XIII ineunte Londoniis Colkctae, (Halle, 1894). I have preserved the flavour of the translation in AJ. Robertson, The Laws of the Kings of England from Edmund to Henry I (Cambridge, 1925), p. 247. 4 For some brief comment on 'so-called' see below, pp. 17-18 and nn. 82, 83.

xvi

Colonial England, 1066-1215

If the Englishman declines trial by combat, the Frenchman, who is accused by the Englishman, shall clear himself against him with an oath supported by those who are qualified by Norman law to be his compurgators.5

This principle applied in every field of law and government. Not even the considerable Breton contingent implanted any of their local customs; their origin was reflected, and then only for a few generations, simply in their personal names. So varied French customs, admittedly still in their formative stages, were reduced to a single Anglo-Norman code of practice. Furthermore, the whole army of settlers and the conquered population were brought under a single discipline. All land was held, directly or indirecdy, of the king, and all fealties ascended to him. These principles were firmly and finally sealed in Domesday and the Salisbury Oath of 1086 (Chapters 2 and 3). Hence William combined the authority of the old English monarchy with a feudal lordship all the more clear cut because it had not grown gradually as on the Continent but had been imposed and defined amidst the urgencies of conquest. This combination of powers was more formidable than either old English kingship or the contemporary dukedom of Normandy. William was not beset by earls as the Confessor had been; Henry I could drive rebels from their English estates and yet scarcely touch their ancestral estates in Normandy. As the twelfth century advanced the monarchy so developed this combination of ancient authority and feudal superiority that control became an imperative requirement: the church sought it first, then the nobility and rest of free society. So from 'Norman versus English' we can easily slip into 'king versus subject' -from one artificial mode of argument into another. Yet the king also was enfolded in the consequences of conquest; he was himself part of the society busy devising its own rules concerning tenure, inheritance and descent; and in the case of the Conqueror he was entrapped by his own insistence on legitimacy. William was a bastard who succeeded to his duchy at a point where rules of succession in the Frankish world were hardening and necessarily excluding illegitimate succession; two generations or so later and he would scarcely have stood a chance. Then his succession to England was based, force apart, on a concocted, trumped-up claim, advanced so successfully by Norman propaganda that it has confused historians ever since. Yet he divided his lands between his eldest and younger son, patrimony from acquisition, just like any other well-ordered Norman lord (Chapter 8). Custom and practice prevailed. And so it continued; if the rules failed, something new was invented, still cast in legal terms. Up to 1272, the accession of Edward I, a period of 200 years, every succession to the English crown was disputed or vulnerable to a rival claim, with the sole exception of Richard I's in 1189. In 1135 conflicting claims were advanced in the female line derived from the daughters of the Conqueror and Henry I. The law provided no solution; war did, and it ended with the fiction that the descendant of the one, Henry fitz Empress, was heir to the descendant of the other, King Stephen. Thereafter, under Henry II, 5

Robertson, The Laws of the Kings of England, p. 233.

Introduction

xvii

patrimony and acquisitions were tossed around among his sons; it was only by genealogical accident that a single heir emerged in 1189 in Richard Lionheart. But there was a residue; John's succession in 1199, involving his own claim as cadet son of Henry II against a representative grandson, Arthur, dogged him throughout his reign (Chapter 16). So the ruling line moved, within little more than a century and a half, from a bastard succession to one of the most difficult issues to arise in the practice of primogeniture. It was a remarkable passage from crudity to sophistication, a sophistication part acquired, part invented, and applied ad hoc. Such a progression from crudity to sophistication was the general experience of Norman and Angevin England. In 1066 England, it has been remarked, was 'an old country'6 It now ceased to be so. After 1066 a new political order was imposed; the church was reorganised; monasteries and monastic orders proliferated; and landowning society was changed in character. Several recent studies of royal patronage have dwelt on the tabula rasa available to the Conqueror for the distribution of land to his followers. His followers enjoyed a similar opportunity; it was available indeed to the newcomers all the way down the social scale; and it was prolonged well into the twelfth century by forfeitures and fresh grants. Still in John's reign tenants took precautions against incoming lords as estates changed hands. So turbulence did not subside rapidly. The adventurous could still find opportunity, though within an increasingly corseted political and legal world, and at increasing financial and political risk. Still they settled down. Nothing illustrates this better than the resort by families of the second and third rank to English toponymic family names. This was apparent by the middle of the twelfth century and became the norm by the end (Chapter 10). With them came minor castles, fortified or unfortified manor houses, rebuilt parish churches, involvement in local government and all the rest of the social paraphernalia of a settled landed class. I have sketched a broad approach. Detailed arguments and qualifications will be found in the essays which follow. Many important aspects of colonial England are scarcely discussed at all. I have not tried to deal with the effect of the Conquest on the peasantry (although there is one brief hint in Chapter 1). I have not discussed the survival of English landowners. I have not tried to draw the obvious parallels with the settlement of the Normans in southern Italy and Sicily. These topics, and indeed others, cry out for further research. So my tide should not be taken to indicate an attempt at complete coverage. Except for the first item, these essays were not written with the title in mind. Nevertheless they all fit into it, and they are best read as random comments made over the years on different problems all of which are part of a single vision which the tide expresses. The essays came first, the tide and the vision later.

* Reginald Lennard, Rural England, 1086-1135 (Oxford, 1959), p. 1 and chapter 1.

This page intentionally left blank

1

Colonial England, W66-12I51

No book has ever been written before with the title of this book and of this essay.2 This is very remarkable, all the more so because so many American historians have written of England in this period. For the United States began as colonies - thirteen of them - and in dealing with their origin and development American historians have laid great emphasis on freedom, especially in the frontier studies exemplified by the work of Frederick Jackson Turner.3 So far as I am aware, very little of this has influenced study of the French colonisation of England which followed Duke William of Normandy's victory at Hastings in 1066. Indeed our interpretation of the history of the American colonies lies clean contrary to our interpretation of the French colony of England, for it is frequently argued that the ruling house was much more domineering in England than it had been or could ever be in Normandy. It is only occasionally that we have fully grasped that, for the French, England was a land of adventure and opportunity, a land of new freedom and initiative, and that not just for the new Norman king.4 It is only in frontier studies, initiated by T.F. Tout in 1902,5 and continued in the last ten years or so by Professors Barrow, Bartlett and Rees Davies, sometimes with direct acknowledgement of a debt to Jackson and the American experience, that a growing interest has developed into something

1 A paper read at the Anglo-Japanese Conference of Historians held at the Institute of Historical Research, London, 29 September to 1 October 1994, and therefore very much a pieced'occasion. I am indebted to Dr DJ. Keane and Professor David Palliser for comments made during the discussion. 2 A close approximation of title is John Le Patourel, The Norman Colonization of Britain', in INormanni e la loro espansione inEuropa nell'alto medioevo, Settimane di studio del centre italiano de studi sull'alto medioevo, 16 (Spoleto, 1969), pp. 409-38, but this is largely concerned with the process of settlement. So also is Brian Golding, Conquest and Colonisation: The Normans in Britain, 1066-1100 (London, 1994). 3 FJ. Turner, The Frontier in American History (New York, 1902). 4 The outstanding example is G.W.S. Barrow, The Anglo-Norman Era in Scottish History (Oxford, 1980). See especially chapter 1, 'A Land for Younger Sons'. See also Willaim Kapelle, The Norman Conquest of the North (London, 1979). 5 T.F. Tout, 'Wales and the March during the Barons' War, 1258-1267', in Historical Essays by Members of the Owens College, Manchester, ed. T.F. Tout and James Tait (London, 1902), pp. 76-136; reprinted in T.F. Tout, Collected Papers, ii (Manchester, 1933), pp. 47-100. Tout took up the theme again in T.F. Tout, The Place of the Reign of Edward II in English History (Manchester, 1914).

2

Colonial England, 1066-1215

more coherent.6 But this work has largely been concerned with the expansion of Anglo-Norman England into Celtic lands beyond the old English borders. We still await a study of colonial England. This is a preliminary sketch of some aspects of such a study. Let me suggest what I am about with two quotations from Turner's great book: The West, at bottom, is a form of society, rather than an area. It is the term applied to the region whose social conditions result from the application of older institutions and ideas to the transforming influences of free land. By this application, a new environment is suddenly entered, freedom of opportunity is opened, the cake of custom is broken, and new activities, new lines of growth, new institutions and new ideals, are brought into existence.

And again: The history of our political institutions . . . is not a history of imitation, of simple borrowing; it is a history of the evolution and adaptation of organs in response to changed environment, a history of the origin of new political species.7

Now I do not want to follow Turner too closely;81 looked him up at a late stage in developing my argument, and in any case Norman England was not the American West and the lands which the Normans acquired were not exactly free. Nevertheless, Norman England was, in Turner's sense, a new political species. With this in mind, I intend to deal with three topics. I have selected these almost at random. Some of the problems of the Conquest, such as the origins of feudal tenures, are viewed as settled by some (including me) and as 'old hat' by others. Other problems, such as the peculiar mixture of cultural and historical fictions and traditions whereby the invaders established for themselves an English identity, are quite beyond the scope of a single essay. But my topics do have a cultural bearing. They will be building, language and law. But first a caveat. The west African in the fifteenth and the American Indian in the seventeenth century were surely aware that something surprising and astonishing was being done to them. In 1066 this was not so. Between English and Norman there was no great cultural contrast, certainly nothing comparable to that created by the European in Africa or America. There were differences, 6 Among many other works see especially Barrow, The Anglo-Norman Era; Robert Bartlett and Angus Mackay, eds, Medieval Frontier Societies (Oxford, 1989); R.R. Davies, Domination and Conquest: The Experience of Ireland, Scotland and Walts, 1100-1300 (Cambridge, 1990); and Robert Bartlett, The Making of Europe (London, 1993). 7 Turner, The Frontier in American History, pp. 205-6. 8 Turner's work is discussed by Robert I. Burns in Bartlett and Mackay, Medieval Frontier Societies, pp. 307-30; and also by Margaret Walsh, 'FrederickJackson Turner and the American Frontier', The Historian, 36 (1992), pp. 3-8.

Colonial England, 1066-1215

3

and there has been argument in plenty about their nature and extent, but they were not as great as those between the Anglo-Norman and the Welsh or Irish, or between Christian and Moslem in the Mediterranean world. So no Englishman in 1066 would have thought that the Norman triumph at Hastings would bring about immense changes. Indeed, he would be likely to view William's coronation simply as a change of dynasty. If he had reached the ripe old age of seventy, he would have remembered the wars of Cnut and Edmund Ironside, the succession of Cnut in 1035, the end of his Scandinavian dynasty in England in 1042, and the return of the old English line with Edward the Confessor. In 1066, with Edgar Aetheling still alive and active, why should Norman William's reign be any more enduring than Danish Cnut's had been? And if he thought a little deeper he would recollect that although Cnut's invasion had brought war, treachery, the elimination of leading English nobles, the rending of the network of kin which sustained the English royal house, yet the resulting intrusion of Scandinavians had been absorbed with little disturbance of the underlying social structure.9 So why should he imagine that a Norman intrusion would be any different? True, if he moved in court circles he would have seen Frenchmen brought into influential positions by Edward the Confessor, but the more he knew the more likely he would be to see this intrusion as just another ripple on the political waters, largely settled after the events of 1052. Even if he lived in the marches he would have to possess a remarkably prophetic eye to see that Pentecost's castle and its like presaged doom.10 So, there had been a battle, a king had been killed and replaced by another, and his followers would no doubt expect and get reward; but those who had not fought against him could make their peace. That was all. It was not a cataclysm. He could not have been more wrong. We know that now. But foreknowledge is only the reverse side of hindsight; to expect the one is to exercise the other. And lack of foreknowledge in the English had important results. In southern England there was no national resistance after Hastings, merely piecemeal aristocratic rebellion. There was more prolonged resistance in the west, along the Welsh borders and especially in the north, but even there the best hope of reversing the verdict of Hastings lay in renewed Danish invasion. With that defeated, all that was left was material for heroic legend — Hereward the Wake and Eadric the Wild. Now the legend of Hereward is all very well, but consider also the Vita Haroldi. Nothing demonstrates the collapse of English morale, the lack of urgency in a dire national emergency, than this sorry tale that Harold 9

For varying assessments of the effects of Cnut's invasion see Katharin Mack, 'Changing Thegns: Cnut's Conquest and the English Aristocracy', Albion, 16 (1984), pp. 375-87; Ann Williams, '"Cockles amongst the Wheat": Danes and English in the Western Midlands in the First Half of the Eleventh Century', Midland History, 11 (1986), pp. 1-22; Robin Fleming, Kings and Lords in Conquest England (Cambridge, 1991),pp. 21-104; and P.A. Clarke, TheEnglish Nobility underEdward the Confessor (Oxford, 1994), pp. 42-44. 10 Our contemporary observer would not enjoy the hindsight which we can now bring to the 'foreigners' castle and the harm thus brought recorded in the E version of the Anglo-Saxon Chronicle under 1051-52, The Anglo-Saxon Chronicle, ed. Dorothy Whitelock (London, 1961), pp. 119, 125.

4

Colonial England, 1066-1215

was rescued from the field of Hastings, then become a hermit (in Cheshire) and then a missionary - the quietist literature of a defeated nation. In that moral collapse, lack of foreknowledge, of understanding of the consequences of defeat, was debilitating. It spelt inaction. There were yet more serious consequences. The English landowner, from the highest noble, to the lowest freeman, could scarcely see that Hastings would bring devastation to the social structures he knew and understood. For him the relationship between lord and man was a personal tie sealed by commendation. That loose arrangement would go. Before 1066 a man might hold his land from one lord and be commended to another. That would go. Again, he might or might not be able to 'go with his land where he will', and so sell without reference to his grantor. That, largely, would go. The land might or might not owe services to the grantor. That would go. These practices had been left largely untouched by Cnut's intrusion. They were still in a healthy state in 1066. They would now be enveloped, as if by blanket weed, smothered, destroyed, so that in southern England, at any rate, not much more than the detritus remained to be recorded in Domesday Book as surviving in 1086. In their place a new, tightly knit structure was imposed: all land now in lordship; dependency which, except for some households offices, was at once personal and tenurial; and no tenure without service. There was a further consequence of immense importance. The English practice was for a man's land to be distributed, on his death, to members of his kin, both descendants and collaterals, either by custom or specific testamentary provision. That, too, stopped. Tenancies held of the new Norman lords, like their own tenancies held of the King, now descended unfragmented to a single male heir. Freedom of manoeuvre now depended on the lord's power to dispose of heiresses and the tenant's power to treat his acquisitions with greater freedom than his patrimony; when he did so, his dispositions were for the benefit of younger sons, not of collaterals or distant kin. Under this new dispensation the old English tenant's power to make testamentary disposition of his lands to his kin was ignored, by-passed and overridden. In this conflict of tenurial cultures the material bonding of the old kin-group was dissolved. It marked the passing of an age. This took place quite apart from the pattern and procedure whereby the Norman succeeded to his new lands. It mattered little whether he came in in a relatively neat and orderly manner on the authority of the king's writ as successor to an AngloSaxon antecessor, or sought to provide the appearance of a more ancient tide to his succession by marriage to an English heiress, or benefited from a grant of consolidated territory based on local administrative arrangements or military needs, or whether he accumulated his lands by force and guile and by luring English occupiers into the net of his new lordship. These were administrative and procedural variants which might soften or exaggerate the impact of Norman rule. It took effect just the same.11 11

The recent revival of old arguments has not led me to change the views expressed in 1982 in Chapter 9 below. Since then the matter has been effectively settled by Fleming, Kings and Lords in

Colonial England, 1066-1215

5

The process began in 1066 when men submitted to the Conqueror, performed fealty and became his men, and received confirmation of their lands and liberties. They gained the protection of a lord in the old way. He acquired vassals in the new way.12 It was the church which preserved the old ways and the old claims strongest and longest; yet the church was subjected to the new burden of knight-service within six years of the Conquest.13 By 1086 the process was almost complete. The final record of Domesday Book only records AngloSaxon lordship at all consistently in one of its six circuits; for many counties there is no record at all. It was no longer important. The one circuit which does provide details indicates that where the pattern of English personal lordship differed from that of tenurial dependency it was usually the latter which determined the allocation of land and rights to the new Norman lords.14 The change was insidious. There was nothing to hit at. There were numerous rebellions and outbreaks against the new Norman rule, but not one carried a banner for the restoration of the old ways. Instead they were directed against the material instrument of the new lordship, the castle, and only two of these, at York, ever fell to the attack of the irate English and their allies.15 It was all the more insidious because the Normans were quite prepared to use and adapt English institutions and practices. They depended on the shire court, the hundred juries, the sworn testimony of ownership and rights in the time of King Edward, and they absorbed the territorial jurisdiction of the Anglo-Saxon sake and soke. So the tenurial revolution was not a total bouleversement rather, it took hold like a parasite within a host. The host, nevertheless, was excoriated. If this was at first hidden from the English eye other matters were plain for all to see. First of all, building. Much Norman building, of course, survives for us still to see, and it is difficult to look upon it without absorbing its antiquity. This is misleading. Forget it. Forget the moss-grown castle walls and Tennyson's dying echoes and think instead of a new housing estate, or Canary Wharf ten years ago, or the Barbican thirty years ago, or Tokyo in the era of post-war continued

Conquest England, and Clarke, The English Nobility under Edward the Confessor. I have drawn on both of these in the above paragraph. See also Eleanor Searle, 'Women and the Legitimization of Succession at the Norman Conquest', Proceedings of the Battle Conference on Anglo-Norman Studies, 3 (1981), pp. 159-70; and Paul Dalton, Conquest Anarchy and Lordship: Yorkshire, 1066-1154 (Cambridge, 1994). A different view is taken by P.H. Sawyer, '1066-1086: A Tenurial Revolution?', in P.H. Sawyer, ed., Domesday Book: A Reassessment (London, 1985), pp. 71-85; and by David Roffe, 'From Thegnage to Baronage: Sake and Soke, Tide and Tenants-in-Chief, Anglo-Norman Studies, 12 (1990), pp. 157176. Apart from the matter of sake and soke, on which there is no unbridgeable distance between the views of Dr Roffe and Professor Fleming, the closely related argument of Sawyer and Roffe does not withstand the case which she and Dr Clarke, quite independently, present. 12 The most extended discussion of the significance of the submissions still seems to be E.A. Freeman, The History of the Nor man Conquest of England, iv (Oxford, 1871), pp. 22-27. 13 This assumes, of course, that the service demanded from the abbey of Evesham in 1072 was a general requirement. 14 Fleming, Kings and Lords in Conquest England, pp. 127-32. 15 At York the garrison was caught outside the castle walls.

6

Colonial England, 1066-1215

reconstruction. The Normans rivalled this. Ninety castles are recorded by 1100, twenty-five of them royal, mostly motte and baileys, structures of earth and timber, but some of them already built in stone. By the same date all English cathedrals were being rebuilt or newly constructed. So were many monasteries. In the next fifty years the effort spread out to parish churches. Today we only see a small proportion of this vast effort. If we could see it whole it would quickly be apparent that it represented a vast capital investment, probably a greater capital investment^capita than this country has ever seen, at least prior to the industrial revolution of the nineteenth century. It was almost the Normans' trademark. It occurred wherever their rule took root; indeed it was the outward manifestation of that rule. Would this great efflorescence of building have occurred any way without the Normans? No doubt some will point to the Confessor's new abbey at Westminster, others to the greater minsters of Saxon England, others again to the halls of Anglo-Saxon kings and lords or to the first signs of castle-building (by Frenchmen) in Edward's reign. But there is really no comparison. In secular building the castle was a novelty. In ecclesiastical building the new cathedrals were of immense dimensions by comparison with what had gone before.16 Moreover, the number of buildings under construction increased enormously and the effort continued beyond the period of immediate settlement into the twelfth and later centuries. An attempt to review this great building effort would be quite out of place.17 I confine myself to those features which are relevant to my general line of argument. First, the castle was an instrument of domination - a blunt instrument. The great church was an equally blatant assertion of power. All but Worcester were built under bishops of continental origin. Five were on new sites, at the centre of reorganised dioceses: Chichester, Coventry, Lincoln, Norwich and Salisbury.18 At Rochester, Bishop Gundulf began work on his new church with a massive defensive tower, almost a keep. At Lincoln also some have argued cogently that Bishop Remigius incorporated defensive work in his new cathedral on the hill at Lincoln, and it seems likely that some of the masons he employed on the west front also worked on the west gate of Lincoln castle.19 At all events, 16

The first post-Conquest cathedral at Old Sarum is an exception; on this see below, p. 8. For recent studies see David Parsons on stone and Julian Munby on wood in John Blair and Nigel Ramsay, eds, English Medieval Industries (London, 1991), pp. 1-27, 379-405; E.G. Fernie, 'Architecture and the Effects of the Norman Conquest', in David Bates and Anne Curry, eds, England and Normandy in the Middle Ages (London, 1994), pp. 105-16; also L.F Salzman, Building in Endland down to 1540 (rev. edn, Oxford, 1967), and D. Knoop and G.P.Jones, The Mediaeval Mason (Manchester, 1933). 18 Exeter might be added to this number, but the move thither from Crediton was planned in 1050. 19 R. Gem, 'Lincoln Minster: Ecclesia Pulchra, Ecclesia Fortis', in Medieval Art and Architecture at Lincoln Cathedral: The British Archaeological Association Conference Transactions for the Year 1982, 8 (1986), pp. 9-28. See also David Bates, Bishop Remigius of Lincoln, 1067-1092 (Lincoln, 1992); and Peter Kidson, in Dorothy Owen, ed., A History of Lincoln Minster (Cambridge, 1994), pp. 14-24. 17

Colonial England, 1066-1215

7

in the words of Henry of Huntingdon, Lincoln Minster became 'a beautiful church in a beautiful place, and a strong church in a strong place, both welcome to the servants of god and inaccessible to any attacker' .20 At some of these sites the new cathedrals involved the destruction of older churches, possibly at Lincoln and Exeter, most dramatically of all at Winchester. Here the old royal palace, in which Anglo-Saxon kings had held court from Egbert to the Confessor, was replaced by a new Norman palace which was under way by 1070. Nine years later Bishop Walkelin began work on a cathedral of immense dimensions, at 162 metres the longest in Europe. As the new church was built the Old Minster was destroyed, and the new cathedral was so sited that the second Anglo-Saxon foundation, the New Minster was shouldered aside and had to find a new home outside the city walls at Hyde.21 This was a real conquest, emphatic, domineering, ruthless, and supremely confident. Winchester and Battle Abbey together - Battle, which commemorated ^battle (the present place-name reflects the lack of a definite article in Latin) - display all the characteristics of the new regime. They registered triumph, firmly rooted in the soil, symbols of conquest like the great railway terminals which the Victorians built in India. Secondly, the castle-building was intimately related to the change in aristocratic family structure which the Normans brought with them. The castle began as a fortress, a mechanism of conquest; it also became an administrative centre and a lavish residence - an exercise in conspicuous consumption.22 In -all these functions the castle and the dynastic family went together. True, there are examples in the later middle ages, particularly in Germany, of castles shared among collateral heirs. The most dramatic is Burg Eltz, which by 1268 was shared between three branches of the Eltz family, each with its own apartments and kitchens, sharing a common court and making common decisions in a Rztterhalle. But such arrangements, the product of a sophisticated law of partible inheritance, were unknown in Norman England. Here castles descended to a single heir (or heiress). And in this material content of the inheritance there was a sharp difference from what had gone before. The Anglo-Saxon landowner not only distributed his land among the kin, he also disposed of precious goods: armlets, bracelets, headbands, broaches, jewelled belts, buckles, purses and inlaid swords, bowls, cups ornamented with gold, jewelry in diverse form, and quantities of gold. Every distant cousin, favoured relative of wife, heir of the half-blood, might have a piece, each according to his or her deserts and standing, and the decedent's favour. With the Norman all this, or nearly all, went into the meltingpot of investment and was transformed into stones and mortar. The Norman 20

Henry of Huntingdon, HistoriaAnglorum, ed,TA Arnold, Rolls Series (London, 1879), p. 212. Martin Biddle and D J. Keane, in Martin Biddle, ed., Winchester in the Early Middk Ages, Winchester Studies, 1 (Oxford, 1976), pp. 306-21. 22 For examples of the castle as a residence see T.A. Heslop, 'Oxford Castle, Nostalgia and Sophisticated Living', Architectural History, 34 (1991), pp. 36-58; also Philip Dixon and Pamela Marshall, The Great Tower at Hedingham Castle: A Reassessment', Fortress, 18 (1993), pp. 16-23. To these may be added Castle Rising and others. 21

8

Colonial England, 1066-1215

lord was as likely to hand down debt as treasure, especially debt to the crown.23 The projects were so vast and so numerous and the effort so continuous that architects, engineers, masons, sculptors, and carpenters throve and multiplied. Some came at first from Normandy, especially from Caen, for work at Canterbury and possibly at Lincoln. But home-bred architects and masons soon came to play a part with increasing confidence in their own ways. As in any rapidly expanding industry risks were taken, much was done in haste, there was sloppy design and workmanship, and some rank bad planning. Amongst the last Old Sarum ranks high. William of Malmesbury tells us that the new cathedral there was built in a castle rather than a city, an unknown place where water was so scarce that it was bought and sold.24 It is apparent, now as then, that the site is compressed, windswept, arid amidst chalk upland, and, to cap all, too near to the castle, 'an ark of God shut up in a temple of Baal', as Peter of Blois put it.25 The need for a defensible site may be advanced as an excuse for the first patron of the work, Bishop Herman.26 Herman was not a Norman but a Lotharingian; he had been bishop of Ramsbury (from 1045) and then Sherborne (from 1058) before the see was transferred to Sarum. His new cathedral was a compromise in which traditional English design did much to determine its size.27 There is less excuse for his successor Osmund, who brought the first church to consecration in 1092, and none at all for Roger, Henry I's great minister, who rebuilt and extended Osmund's work and, as William of Malmesbury tells us, 'beautified it in such a manner that it yields to none in England'.28 Except as a fortress (and that may still have been in Bishop Roger's mind), it was still in the wrong place. It was not until 1219 that papal permission was obtained to correct the error and to move downhill to the present site in the valley of the Avon.29 By the side of the major mistake at Salisbury there were many other disasters. Let us leave on one side the many fires and note just the structural failures.30 23 The argument could be pursued further. It is worth asking how far the change which I described above led to a change in demand and hence in the direction of artistic endeavour. With one or two striking exceptions, such as the Gloucester candle-stick, early Anglo-Norman works in precious metal are few and far between. There was a remarkable efflorescence of ivory carving in the first half of the twelfth century, chiefly using walrus ivory. Had gold become scarce and too expensive? And what encouraged the fashion for walrus ivory? 24 Willelmi Malmesburiensis monachi de gestis pontijlcum Anglorum, ed. N.E.SA. Hamilton, Rolls Series (London, 1870), p. 183. 25 Registrum S. Osmundi episcopi, ed. W.H. Rich Jones, Rolls Series (London, 1883-84), ii, p. xx. 26 The suggestion is made by R. Gem in Medieval Art and Architecture at Lincoln Cathedral, p. 25. 27 Idem, The First Romanesque Cathedral at Old Salisbury', in Eric Fernie and Paul Crossley, eds, Medieval Architecture and its Intellectual Context: Studies in Honour of Peter Kidson (London, 1990), pp. 9-18. 28 Gesta regum, ii, p. 484. 29 Calendar of Papal Registers (London, 1893), i, p. 46; Registrum S. Osmundi, ii, pp. 5-6. For a succinct and useful account of the history of the first site at Old Sarum, see F.WJ. Hill in VCH, Wiltshire, vi, pp. 51-67. 30 Fires occurred as follows: Bath (1137), Canterbury (pre-1086,1111,1174), Chichester (1114, 1187), Crowland (1091), Exeter (1111), Glastonbury (1184), Gloucester (1122), Hyde (1140),

Colonial England, 1066-1215

9

At Old Sarum, just five days after Bishop Osmund's consecration of his new church in 1092, a storm brought down the roof of the tower.31 At Winchester in 1107 even worse befell: the whole tower collapsed.32 A storm brought down a tower at Ely in 1111, causing great damage to the church.33 At Worcester in 1175 yet another tower fell to the ground - this a new one.34 In 1185 there was serious damage at Lincoln; the vaulting collapsed in the choir probably as a result of overloading.35 Finally in 1210 a great storm brought down towers at Evesham, Bury St Edmunds and two at Chichester.36 The plain fact is that much Norman building simply fell down.37 No doubt natural disasters - earthquakes, lightning and the like, Acts of God in the parlance of insurance companies - account for some of these collapses. But the sources turn too readily to such causes. Our monastic annals record fifteen earthquakes in England between 1076 and 1201 ;38 in my life-time, since 1922, I have so far experienced just one very slight tremor. Some structural collapse probably came from technical failures in over-ambitious projects - a common enough feature of building booms.39 But there was also some bad engineering and poor workmanship - in fact, jerry-building. Some of the evidence remains before our eyes - at no less a place than Westminster. At the time, Westminster Hall was probably the largest in Europe (73 metres long, 20 metres wide). When William Rufus saw it for the first time and held his court there in 1094 he commented that it was only half the size it should have been.40 But there were other causes for dissatisfaction: the external buttresses and arcades were misaligned by 4 feet, with consequent displacement of the interior posts

continued

Lincoln (1124), Peterborough (1115), Tewkesbury (1178), Winchester (1102), Worcester (1113, 1147, 1189), York (1137). In this somewhat casually collected list it is not always clear whether the evidence refers to a fire in the church, or town, or both. Fires specifically restricted to towns have been excluded. According to Gervase of Canterbury, the fire at Canterbury in 1174 began in the neighbouring houses, but it is possible that it was a deliberate act of arson by the monks; see Peter Kidson, 'Gervase, Becket and William of Sens', Speculum, 68 (1993), pp. 969-91. Acts of war have also been excluded. It should be noted that there was a frequent occurrence of fire in the first half of the twelfth century when much new building was coming to completion. How many were caused by workmen's braziers? 31 Gesta regum, ii, p. 375. 32 Annaks monastiri, ii, p. 43. 33 Ibid., ii, p. 44. 34 Ibid., iv, p. 383. 35 Peter Kidson in Dorothy Owen, ed. A History of Lincoln Minster, pp. 25-26. 36 Annaks monastiri, iii, p. 32. 37 Similar and later examples are to be found in Salzman, Building in England down to 1540, pp. 25-29. 38 Annaks monastici, v, index, s.n.. 39 This was probably the case in the failure of the vaulting at Lincoln. For commentary on the whole problem in the modern world, see H. Petroski, Design Paradigm: Case Histories of Error and Judgement in Engineering (Cambridge, 1994). 40 Henry of Huntingdon, Historia Anglorum, p. 231; William of Malmesbury, Gesta regum, ii, p. 374.

10

Colonial England, 1066^1215

and roof beams, and much of the masonry was of poor standard.41 Despite that, Westminster Hall still stands.42 Worse befell at Abingdon. There in 1091 work on the enlargement of the choir undermined the foundations of the tower at the point of junction and the whole tower collapsed during the night office; the monks escaped only through the divinely inspired instinct of the prior who had cleared the church. Date and time were long remembered: 28 March, just after the response to the third reading.43 St Albans did not even get that far in the rebuilding of the west front of the abbey under Abbot John (1195-1214). John benefited from the bequest of 100 m. from his predecessor. Then: A number of builders were chosen; Master Hugh of Goldcliff was in charge. Though an outstanding architect, he proved to be a false and deceitful man. In digging and laying the foundations he soon spent the 100 marks along with considerable other sums, not counting daily allowances, and this before the work was above ground level. On Hugh's malign advice, much decorative carving was added, which was unnecessary, trifling and immeasurably expensive. And so it happened that the abbot became vexed and alarmed and finally grew weary of the scheme. So the work came to a halt before it had risen to roof level (?tabulatum domitialem). The walls were left open to the winter weather, the stonework fractured as if it were soft, all to no purpose, and a wall completely collapsed under its own weight, columns, capitals, pediments and all, like a delapidated ruin, so that the rubble of figure and floral carvings became the laughing stock of all who saw it. The masons quit in despair, without any wages to reward their labours.44

Plus fa change. We might describe that builder as a cowboy did we not ourselves know what great expense an unrestrained architect can incur. So there has been disaster in many forms. Our Norman monuments are the residue of a much larger number which has been reduced by fire, tempest, structural inadequacy and the changing requirements and novel styles of later ages. They have survived intense natural selection. In considering the building programme as a whole, therefore, say over the first century of Norman rule, we have to consider what has been lost or overlain as well as what survives. In total it represents a truly vast effort, now immeasurable, both physical and financial. The most striking effect must have been visual. But the most important consequences were economic and social. The industry was labour-intensive. At Canterbury the monk Gervase remarked on the ingenuity with which William of Sens used the windlass to load and unload ships and shift stone and mortar, as if it were at the sharp end of technological innovation.45 The main effort was 41 H.M. Colvin, in H.M. Colvin, ed., History of the King's Works, i (1963), pp. 46-47. Dr Keene has reminded me that some of the early stonework at Winchester was also of poor workmanship. 42 The present hammer-beam roof is the work of Richard II's reign. 43 Chronicon manasterii de Abingdon, ed.J. Stevenson, Rolls Series (London, 1858), ii, pp. 23-24. 44 Gesta abbatum S. Albani, i, ed. H.T. Riley, Rolls Series (London, 1867), p. 219.1 have condensed the passage somewhat in translation. 45 Gervase of Canterbury, i, p. 7.

Colonial England, 1066-1215

11

human.46 The early castles were built by forced labour, royal castles by extending the old English duty of burh bot into castle-work, which fell as a burden on the shire.47 Some lords claimed or enforced similar services. But churches were not run up in a few days as the early motte might be. Moreover, labourers might be forced, but not architects, masons, sculptors, stone-layers, carpenters, wood and metal workers, and all the other skilled and semi-skilled men involved. They received payment. So did the shipmen and their crews who brought stone from Caen, or Quarr, the wagoners who hauled it from the local quarry or carried timber from the forest far or near. At Abingdon, Abbot Faricius brought roof-beams from Wales 'at geat expense and effort'. He employed six wagons, each with twelve oxen, and the round trip to Shrewsbury and back took six to seven weeks.48 Further back in the process there were the quarrymen who extracted and cut the stone, the lumbermen who cut the trees and the sawyers who prepared it for use. It is difficult, in short, to imagine the industry working at all without a contracted or salaried staff directing a largely waged labour force (I use modern jargon quite deliberately) - a labour force which must have been to a considerable extent seasonal and migratory.49 And as the money they received cascaded down it generated demand for ships, wagons, tools and equipment of all kinds - trowels, axes, adzes, chisels, hammers, something as simple as a rope - and in so doing encouraged further investment in the cordwainers shed, the carpenter's shop and the mason's yard. It engendered credit transactions; Aaron of Lincoln himself boasted that he had built the shrine of St Alban.50 It led to the exploitation of new natural resources, sea-coal supplementing timber to fuel the lime-kilns manufacturing mortar.51 It encouraged new methods - stone cut to standard size so as to suggest some system of mass production at the quarries.52 The building effort, in short, acted as a Keynesian Multiplier, as an economist might still describe it. The fluid element was money; and a lot of that money, as it spread out, ended in the hands of the peasantry, the suppliers of necessities, selling their produce in the markets which grew up alongside any major building project, the suppliers also of the labourers who might return home with coin in their purses at the end of the year's building season. Some sixty years ago M.M. Postan, in a famous article, demonstrated that the first half of the twelfth century saw a shift from labour services on the manor to money rents and wage labour.53 In all the discussion which has taken place 46

See Salzman, Building in England down to 1540, plates 4, 5. Colvin, History of the King's Works, i, pp. 24-26. 48 Chronicon monasterii de Abingdon, ii, p. 150. 49 All these matters are examined in great detail by Knoop and Jones and in the other works cited above, n. 17. Much of the documentation naturally comes from the later middle ages. 50 Gesta abbatum S. Albani, i, pp. 193-94. 51 Salzman, Building in England down to 1540, p. 150. 52 H.M. Taylor, Anglo-Saxon Architecture, iii (Cambridge, 1978), p. 754. 53 M.M. Postan, The Chronology of Labour Services', Transactions of the Royal Historical Society, fourth series, 20 (1937), pp. 169-93; and in revised form in M.M. Postan, Essays on Medieval Agriculture and General Problems of the Medieval Economy (Cambridge, 1973), pp. 89-106. 47

12

Colonial England, 1066-1215

since few have wondered what the landlords did with their cash renders or how the peasants acquired the money to pay them.54 The building industry provides an answer. Although the castle was the instrument of oppression, as the obituary poem of William the Conqueror preserved in the Peterborough Chronicle has it,55 and although the great church expressed moral domination, as Henry of Huntingdon saw it,56 the building programme as a whole was nevertheless a powerful liberating force, engendering demand, encouraging investment, generating credit, dissolving rigid patterns of employment. It also created opportunity. Already in 1086 carpenters and masons, working in the king's service, had been rewarded and retained with grants of land.57 Leading men in industry, then as now, became landed gentry. To sum up then, Norman building was an expression of dominance. At the same time it refurbished the economy of England. The process was very far from simple, certainly not as simple as my brief sketch allows. At Winchester, for example, the very focal point of change, William the Conqueror commonly wore his crown in the Old Minster, thereby emphasising the legitimacy of his succession to the Confessor (quite apart from that there was as yet nowhere else in Winchester where he could go). Likewise, as the Old Minster was demolished, its stones used in the new cathedral, the relics of St Swithin and other English saints, along with the bodies of the English kings, were translated with honour into the new church. And as these changes were taking place, Winchester itself grew and prospered; it depended in no small degree on these new capital ventures and declined when they ceased to constitute a capital.58 So at Winchester, as indeed elsewhere, there was English stone and English sculpture, and there were English craftsmen, tradesmen and labourers. By the late twelfth century such intermingling was far advanced. When William of Sens was brought in as the architect for Canterbury in 1174 he rebuilt the choir and presbytery in what we call a 'French' style. We do so despite the fact that he was succeeded by William the Englishman who continued the work in the same mode and to William of Sens' design, and we do so because by then the work of both Williams looked distinctly intrusive, alien, by comparison with contemporary designs at Wells, Glastonbury and Ripon.59 These embodied an insular style, a home-grown product, not simply of racial intermingling, but of a shared response to the genius loci, one also in which continental elements were 54

See, for example, Rosamund Faith, 'Demesne Resources and Labour Rent on the Manors of St Paul's Cathedral, 1066-1222', Economic History Review, second series, 47 (1994), pp. 657-78. 55 The Peterborough Chronicle, 1070-1154, ed. Cecily Clark (Oxford, 1970), p. 13; EHD, ii, p. 164. 56 See above p. 7. 57 Colvin, History of the King's Works, i, pp. 26-27. 58 Martin Biddle, 'Early Norman Winchester', in J.C. Holt, ed., Domesday Studies (Woodbridge, 1987), pp. 311-31. See also Biddle and Keene, in Biddle, ed., Winchester in the Middle Ages, p. 312. 59 It should be noted, however, that Oakham Hall and the church of St Peter, Tickencote (Rutland) share a French style similar to Canterbury. For the argument that the choice of both architect and style at Canterbury was deliberate and closely linked to the manoeuvring of archbishop and chapter, see Kidson, 'Gervase, Becket and William of Sens', pp. 969-91.

Colonial England, 1066-1215

13

blended as fresh continental designers and artists were attracted to the adventure.60 If we encountered its analogue in the American colonies we should call it a colonial style. Instead, in England, we have to be satisfied with 'Early English', which, outside the esoteric definitions of Art History, is wonderfully misleading in its insularity.61 I have now moved into my second topic: language, and it is to this that I now want to turn, but only briefly, as an interlude, because it has been well clarified for the historian by Dr Clanchy and further studied by Professor Rothwell and Professor Short.62 To sum up a very complex development very briefly, the Norman Conquest murdered the written Anglo-Saxon language as surely as it demolished the Old Minster at Winchester.63 It was the predominant language of government under King Edward.64 Very rapidly, by the 1070s, it was swamped by Latin, and Latin became the main language of record for the rest of the middle ages. But Latin was a clerical language and from the start it is clear that when royal government wished to reach out to as wide a political public as possible it supplemented Latin with a vernacular: Anglo-Saxon under William the Conqueror, Anglo-Saxon under Henry I, French for the Magna Carta of 1215, both French and English in the period of baronial reform of 1258-9; both again in 1298 in the movement against Edward I.65 Of the two, French was used in a wide range of business transactions, and it was not until the middle of the fourteenth century that English began to forge ahead.66 60 The point is developed, quite independently, by Lindy Grant in 'Architectural Relationships between England and Normandy, 1100-1204', in Bates and Curry, eds, England and Normandy in the Middle Ages, pp. 117-42. For continental influences in a broader context see Elizabeth Salter, English and Continental: Studies in the Literature^ Art and Patronage of Medieval England (Cambridge, 1988), pp. 4-28. Among many local illustrations of the increasing sophistication of building and its decoration see Rita Wood, 'The Romanesque Doorways of Yorkshire, with Special Reference to that at St Mary's Church, Riccall', Yorkshire ArchaeologicalJournal, 66 (1994), pp. 59-90. 61 Yet another term - 'transitional' - is equally dubious, though on other grounds. 62 M.T. Clanchy, From Memory to Written Record (London, 1979; 1993). The works of Professor Rothwell and Professor Short are mentioned below. 63 Only one OE diploma and perhaps two writs survive from William II, seven from Henry I, one from Stephen and five from Henry II. Almost all these are bilingual (with Latin) and a high proportion are the work of beneficiaries, especially the cathedrals of Canterbury and Rochester, David A.E. Pelteret, Catalogue of English Post-Conquest Vernacular Documents (Woodbridge, 1990), nos 40-55. Two circumstances delayed the extinction of OE: (a) it was used locally (the evidence coming from Bath and Exeter) for manumissions and the like, which were often attested by priest, or reeve, and sometimes by the hundred court (ibid., nos 70-148, passim); (b) it was occasionally used for administrative purposes in the north. See the writ of Ranulf Flambard, ibid., no. 64; and H.S. Offler, ed., Durham Episcopal Charters, Surtees Society, 179 (1968), pp. 89-91. 64 F.M. Stenton, The Latin Charters of the Anglo-Saxon Period (Oxford, 1955), pp. 43-49, 84-91; F.E. Harmer, Anglo-Saxon Writs (Manchester, 1952). 65 AJ. Robertson, The Laws of the Kings of England from Edmund to Henry I (Cambridge, 1925), pp. 22S-24, texts at 232-37, J.C. Holt, Magna Carta and Medieval Government (London, 1985), pp. 15n., 242-43; idem, Magna Carta (Cambridge, 1992), pp. 474-77; William Stubbs, Select Charts, ed. W.H.C. Davis (Oxford, 1921), pp. 490-93. 66 For the best summary, see William Rothwell, 'Language and Government in Medieval England', ZeitschriftfurFranzosische Sprdche und Literatur, 93, (1983), pp. 258-70; see also M. Dominica Legge, Anglo-Norman Letters and Petitions, Anglo-Norman Text Society, 3 (1941).

14

Colonial England, 1066-1215

The local variant of French in England (which we can now call AngloNorman) came to diverge markedly from the French of the He de France. By 1200 it had become something of a joke; correct French was now a language which boys were taught at school; sons of nobles were sent to tutors and households in France so that they might learn to speak correctly.67 English also changed; the old formal language was replaced by something simpler (which we can call Middle English), derived in part from spoken English. Throughout the catalyst was speech, not writing. It was through these fluid media (alongside Latin) that England was governed. One result was that particular languages were used each for its own purpose. In the law courts in the thirteenth century the record was kept in Latin, pleas were made in French, jurors responded in English, arguments to the pleas were also couched in English.68 Earlier there was even more variety. The invaders of 1066 brought with them not simply Norman French, but Breton, Poitevin and Flemish speech, not to mention the contribution of a few Germans. Then on the English side there were local dialects which in the north amounted almost to a different language, and on the fringes of the new realm the Normans encountered Welsh and Gaelic. So, to start with, kings and lords used interpreters, latimarii, who enjoyed a respected and privileged position in the Anglo-Norman world.69 Seven of them appear as Domesday tenants. But in south-east England, at least, this phase soon passed and in the twelfth century the linguistic ability of great churchmen attracted much contemporary comment; even kings apparently acquired a smattering of English.70 The evidence on the varying usage of French and English is kaleidoscopic and specific to time and place; no single piece establishes that either the one or the other was generally predominant by a particular date.71 But it does settle one matter with reasonable certainty. The difference

67

Mildred K. Pope, From Latin to Modern French with Especial Consideration of Anglo-Norman (Manchester, 1934), pt 5, especially pp. 420-26; on which see William Rothwell, 'From Latin to Modern French: Fifty Years On', Bulletin of theJohn Ry lands Library, 68 (1985), pp. 179-209; and idem, 'Stratford atte Bowe and Paris', Modern Language Review, 80 (1985), pp. 43-44. 68 Clanchy, From Memory to Written Record, pp. 208-9. 69 Constance Bullock-Davies, Professional Interpreters and the Matter of Britain (Cardiff, 1966). 70 The instances are brought together by R.M. Wilson, 'English and French in England, 11001300', History, new series, 28 (1943), pp. 37-60, especially 49-52. 71 The balance between French and English has been much debated. See ibid., and M. Dominica Legge, 'Anglo-Norman and the Historian', History, 26 (1941), pp. 163-75. The current weight of opinion is certainly on the side of English as the 'mother tongue' even of the aristocracy in the late twelfth century. See especially Michael Richter, Towards a Methodology for Historical Sociolinguistics', Folia linguistica historica, 6 (1985), pp. 41-61; and Cecily Clark, Towards a Reassessment of "AngloNorman Influence" on English Place-Names', Language Contact in the British Isles, Linguistische Arbeiten, 238 (1991), pp. 275-93. The argument has turned on a small number of pieces of evidence and it may well be that these have been seen as more decisive than they really are. For example, Gerald of Wales's well-known charge that his nephew was lazy and slipshod in learning his French, though establishing that French had to be taught and learned, does not establish that the starting-

Colonial England, 1066-1215

15

between French and English became one of rank, rather than race. French was for the landlords and officers, English for the peasants and privates. One of the more dramatic miracles in twelfth-century hagiography is the cure of the dumb. This was made all the more dramatic when the sufferer found his voice in French. It happened to a deaf and dumb boy at Beverley 1101-9. St Godric of Finchale (d. 1155) found he could do it, and the power was attributed also to St Wulfric of Haselbury (d. 1185). The names, note, are all English, and context is English speaking; the gift of French enhances the miracle. But in the last case another Englishman, Brihtric, priest of the parish of Haselbury, enters the story, for he complained to Wulfric, asking why he had not given the gift of French to him, too; as it was, he had to stand dumb in the presence of his bishop and archdeacon.72 So parish priests spoke English, bishops and archdeacons French. The same assumption is made in the Vita Haroldi: Harold's anchorite assistant, Sebricht, is rusticanus, a peasant ignorant of any language but English.73 Nevertheless, in the interchange between the two languages, English had most of the advantages. The Normans needed English to command. The English had no need for French unless they belonged to the landed or professional classes. Moreover, there were compulsive pressures favouring English. The effect of intermarriage of Norman lord with English lady has probably been exaggerated. Some, like Alfred's mother, may have taught their native language to their children but I suspect that most of them were too busy acquiring status by learning French. The influence of the household is a different matter. It was from an English nurse that the young Norman lord would most likely learn to speak, and from the servants in hall and stable that he would acquire some of his vocabulary.74 The resulting interchange of words was complex. For example,

point was English rather than the crude, vulgar and insular French which he contrasts with the elegant French spoken by John Blund. See William Rothwell, 'The Teaching of French in Medieval England', Modern Languages Review, 63 (1968), pp. 37-46; and Ian Short, 'On Bilingualism in AngloNorman England', Romance Philology, 33 (1980), pp. 470-72. Similar problems, involving the interpenetration of spoken English and Anglo-Norman, arise with the well-known Tretez of Walter de Bibbesworth, which was intended as a primer of French for estate management. See Clanchy, From Memory to Written Record, pp. 197-200; Rothwell, 'Language and Government in Medieval England', p. 269; idem, 'The Role of French in Thirteenth-Century England', Bulletin of the John Rylands Library, 58 (1976), pp. 459-60. The famous warning in English to Hugh de Morville, which A.L. Poole accepted as evidence that 'English was . . . familiarly spoken in at least some AngloNorman knightly households in the first part of the twelfth century', From Domesday Book to Magna Carta (Oxford, 1951), pp. 252-53, has now been totally undermined. See Roger Dahood, 'Hugh de Morville, William of Canterbury, and Anecdotal Evidence for English Language History', Speculum, 69 (1994), pp. 40-56; and for earlier criticism of this tale see M. Dominica Legge, 'AngloNorman asaSpoken Language', Proceedings ojthe Battle Conference on Anglo-Norman Studies, 2 (1979), pp. 108-17, at p. 111. 72 Ian Short, 'Bilingualism', pp. 474-78. 73 Vita Haroldi, ed. Walter de Gray Birch (London, 1885), p. 32. 74 Cecily Clark, 'Women's Names in Post-Conquest England: Observations and Speculations', Speculum, 53 (1978), pp. 223-51; idem, 'Towards a Reassessment of "Anglo-Norman Influence'", pp. 279-80.

16

Colonial England, 1066-1215

much is sometimes made to the English word 'knight' invading the very heart of post-Conquest Norman society as if the Norman knight and the English thegn were alike.75 In some respects they were. Nevertheless, the explanation lies elsewhere. The Normans knew that diey were the men on horses, the chevaliers:, they had no need to tell each other so. It was the English, striving to set them in context by relating them to the mounted officers of the Anglo-Saxon household, who called them knights. So it would very likely be below stairs in the household that the Norman youth first heard the word. But if the English gave us 'knight' French gave us 'chivalry', which the Normans did need to talk, sing and read about, and did so in the Anglo-Norman literature of chivalry for the next centuries. There are other examples to bear in mind. Parliament is at present in recess. Shortly newspapers will tell us of a new session and'of ministers and secretaries of state reassembling with members or representatives in the Commons. We shall not read that the stallers are foregathering with the wise men in a moot. In 1991 Dr Tony Hunt published a massive work of scholarship with the forbidding tide, Teaching and Learning Latin in Thirteenth-Century England.™ This is a study of the vernacular glosses to educational texts and for the first time it gives us objective, systematic evidence of linguistic interchange. The texts were in Latin. They were glossed by the teachers with class-notes, and these notes are in French or English as the subject and its particular vocabulary required. In short the teachers were language-switching.77 It follows that the ordinary pupil in the class, son of knight, gendeman or merchant, could do the same. This sheds a flood of light on much else. What language did Queen Emma speak? Daughter of Duke Richard I of Normandy she was raised in the Norman court in the last two decades of the tenth century when the duke and his entourage were absorbing all things French. At a guess French was probably her native language, although she may also have picked up Danish. Then she married King Aethelred II of England, then Cnut of Denmark. After Cnut's death she spent seven years in exile in Flanders, then the last twelve years in England where she died in 1052 to rest beside Cnut in the Old Minster at Winchester. So — French, English, Danish, Flemish, English - we surely have to assume that she could communicate in all these languages and there is evidence for at least one small part of this. To her end she used her English name Aelfgifu in documents, but also preserved her given name Emma frequently enough for it to acquire the English form of Imma, and it was as Emma that she was remembered.78 Now unless Emma was a brilliant linguist, conversant in four distinct languages, she was surely language-switching. And she is merely one example of the general mix of nations, or tribes (call diem what you will) with their different languages and cultures which pervaded north-west Europe at this time. Far lower in society than Emma the same imperatives applied. The Domesday Survey was a 75

Most recently, ibid., p. 282. A. Hunt, Teaching and learning Latin in Thirteenth-Century England, Svols (Woodbridge, 1991). 77 Ibid., i, pp. 12-16, 434-35. 78 Encomium Emmae, ed. Alistair Campbell, Camden third series, 72 (1949), pp. xl-1, 55-61.

76

Colonial England, 1066-1215

17

multilingual operation. It ended as a record in Latin, but originated in spoken statements which must have been rendered in a range of English dialects and on occasion in French. What language then was spoken by the Cambridgeshire jurors, to judge from their names, half of them English and half French?79 One thing is certain; however crudely, they must have spoken intelligibly to each other. The same applies in the vast building operations which I have discussed. At Rochester Bishop Gundulf had to instruct his builders what they had to do, and they in turn had to issue orders which worked down in the end to the labourers, forced or paid: somewhere down the line French was turned into English; word-of-mouth communications entered a frontier zone in which words, phrases, constructions even, of the different languages intermingled. It is a zone of interchange not a line sharply defined by philological considerations. Men were fudging together a method of communication, not setting out to learn a foreign language; soldiers in any era in any foreign land have done the same. They were making a mass of particular choices, like flip-flops falling this way for English and that way for French. There was no attempt at formal regulation until 1362 when parliament passed a statute aimed at restricting the use of French in favour of English in the law courts. The statute itself was in French. It had no apparent effect. It was in the arcana of the law that French survived longest, though much debased, even to the eighteenth century.80 This brings me to my final topic - law; and to some simple comments on the steps taken for the settlement of the land after 1066. First, a series of documents: William Fs writ for the Londoners confirming the laws of King Edward; William I's ordinances concerning exculpation; William I's writ ordering the separation of the lay and ecclesiastical courts.81 All these were issued in English (the London writ) or in both English and Latin. Then followed, issued in Latin only apparently, the judicial instructions which pass under the name of the ten articles and include the arrangements for the murdrumfine, a protection of the French ensured by imposing a penalty on the local community when a victim could not be proved to be English. Then Domesday Book. Together these are the documents of primary settlement. To them we should add the 'coronation charter' of Henry I. The inclusion of this last item may seem surprising but it is in fact part of the sequence. It survives only in copies and it is worth noting that it did not describe itself as a charter nor was it set precisely in charter form. In one of the earliest versions, in the Textus Roffensis, it is entitled Institutiones regis Henriti, no different in short from the institutions of King William.82 The best of all contemporary writers, William of Malmesbury, tells us that it was an

79 Inquisitio comitatus Cantabrigiensis, ed. N.E.S.A. Hamilton (London, 1876), index rerum, s.n. juratores. 80 J.H. Baker, Manual of Law French (Letchworth, 1979), pp. 9-14. 81 Robertson, Laws of the Kings of England, pp. 230-43.1 leave on one side the Willelmi mrticuli retractati and the Leis Wilklmens not strictly contemporary (ibid., pp. 225-27; texts at 244-75). 82 Robertson, Laws of the Kings of England, p. 276.

18

Colonial England, 1066-1215

edict.83 Others refer more vaguely to verbal promises. It is best read, therefore, as a straightforward continuation of the Conqueror's work. As such it fits. Itwas only in the course of the twelfth century that it acquired extra status as a charter of liberties. All these documents are concerned with how the French were to live besides the English in a conquered land and how the conquered land was to be occupied and managed. The London writ and the regulations concerning exculpation were immediate measures urgently necessary in the circumstances of 1066. The rest are more leisured and in these consultation and consent were expressly stated and effectively practised (as in the case of Domesday). Taken together they were highly effective measures of occupation. But they were designed primarily for the first generation of settlers. The last in the series, the Charter of Henry I, shows that in the second generation further problems had to be tackled: reliefs and the succession to estates, wardship and the marriage of minors, the marriage of widows; and buried in these problems were the rights of one lord against another, especially the rights of the chief lord, the king, and his rights and powers as king, against the rest. In all these matters an English, insular, solution had to be found which was applied across all the different customs and instincts which invaders of varied origin brought with them. It is these matters, or course, which link the Charter of Henry I with Magna Carta. It soon became apparent, too, that the territorial settlement sealed in Domesday was very intricate, involving all kinds of cross tenures. This, along with the consequences of further enfeoffments, marriages between noble houses, sequestration and redistribution consequent on rebellion, produced an exceptionally volatile land market. Henry I stepped in to bring all disputes about land where the disputants held of different lords into the county courts. It was a momentous step, the first towards the common law of England. I must leave what followed on one side - with great regret because the whole structure of the common law was a unique response to a colonial situation, the work of men in new circumstances inventing their own methods of resolving the disputes about tenure and tide which accrued as the land settled down under new rulers. The impetus was continental. The English had no land law for the dynastic family: in Normandy the Normans were beginning to move towards one. But the solution in the end was insular, quite unlike anything else in western Europe.84 For the circumstances of conquest made a double contribution. They allowed the roots of primogeniture to strike deep because the immense acquisitions of the Conquest allowed for easy satisfaction of the claims and possible objections of younger sons. And they involved a mass of feoffments and ensuing disputes which led on to legal and procedural innovation. The settlers had to choose: the common law (as it became) or chaos. It was not the whole of law. Rather it was a highly vigorous graft on the older stock of English law. At first sight law looks quite different from language. 83 84

Gesta regum, ii, p. 470. For aspects of this see below, pp. 118-27.

Colonial England, 1066-1215

19

Language grows; law may be imposed. But it is almost as difficult for invading settlers to change a country's existing law as to change its language - and there is often no need to. This was so in England. The old structure of courts and the old law remained, and they remained for a very long time, regulating local communities long after the assizes and all the associated procedures had been established, and still resistant, though with rapidly diminishing strength, to the jurisdiction of the justices of the peace in the sixteenth century and later. Most cases, remember, never came to the assizes or the courts of common law. They were settled locally in manor, hundred and borough. This sustained old procedures. Some litigants still chose wager of law, preferring oath-helpers to a jury, in the borough court of Nottingham in the mid fifteenth century.85 This old procedure of compurgation was not finally abolished until 1833.86 Local courts gave a structure to local communities, and local procedures gave them an active identity, each to its own ways. I will confine myself to one example: the town of Halifax in the manor of Wakefield. Now Wakefield was a vast manor which extended from Wakefield up the Calder valley to the Pennine watershed. Discontinuously, it spanned thirty miles from east to west and twenty from north to south.87 From the twelfth to the fourteenth century it was held of the crown by the Warenne earls of Surrey, thereafter by the earls and dukes of Lancaster. It was ajurisdictional liberty which as a court leet combined manorial jurisdiction with the criminal jurisdiction of the local court of hundred or wapentake. The manor was an enclave within the ordinary government of the county and it remained so throughout the middle ages. It was a mixed community. In the fourteenth century Wakefield itself was a borough with a borough court; there was much coal mining in the area; the moors carried sheep and the valleys of the Calder and the Colne were becoming by the fifteenth century one of the chief centres of the woollen industry. Culturally it made its mark. Wakefield itself was the centre of one of the great cycles of mystery plays. Richard Rolle, one of the first theologians to write in middle English, was a near neighbour at Hampole. But administratively it was relatively isolated. It lay to the west of the main communications north and south through Castleford and Ferrybridge. It was sealed off in some degree by itsjurisdictional privilege from the processes of royal government. The king's writ ran, but it was executed by the earl's officers; royal justices did not enter the liberty. Kings rarely came to Wakefield or to the earl's castle of Sandal. In short the manor was an enclosed local community. The rolls of the court of the manor of Wakefield still survive in an almost

85 Trevor Foulds, Jill Hughes and Michael Jones, The Nottingham Borough Court Rolls: The Reign of Henry VI (1422-57)', Transactions of the Thoroton Society of Nottinghamshire, 97 (1993), pp. 74-87, at p. 78. 86 F. Pollock and F.W. Maitland, History of English Law (Cambridge, 1898), i, p. 150. 87 For maps, see The Court Rolls of the Manor of Wakefield, i, ed. C.M. Fraser and Kenneth Elmsley, Yorkshire Archaeological Society (Leeds, 1977) pp. xxiv-xxv; improved and amended by D J.H. Michelmore, ibid., iv (1983), pp. viii-ix; and ibid., vi (1985), frontispiece.

20

Colonial England, 1066-1215

unbroken series extending from 1274 to 1925: 670 rolls in all.88 They are a major source for the study of a local society and its law. That law, in the main, was the common law of England. There were local variants certainly. The Wakefield court, for example, gave short shrift to confessed usurers. It pilloried common litigants. It was very sensitive to the removal of cases to the ecclesiastical courts at York. On occasion the rolls draw a distinction between common law and the ancient custom of the manor. But in the main it operated very much as any other local court, and it can be seen simply as one of the many minor variants in a national system of criminal and civil justice founded in royal assizes, confirmed by royal statutes, ultimately responsible through the lord of the liberty to royal authority and kept up to scratch by the example of royal justice. That works except at one point. Suppose that you committed a theft, a felony, and in the subsequent legal game of snakes and ladders you missed all the ladders and landed on the snakes. You had no money to buy your way out, or you were a stranger without local support, or, though a native, your neighbours on the jury finally decided that they had had enough of you. You would be condemned and sentenced. If the court were sitting at Wakefield, Kirkburton or Brighouse, you would be hanged. If on the other hand you were condemned in a session at Halifax, you would be beheaded; from some point, most probably early in the sixteenth century, you would be guillotined. The guillotine is still there. It stands where it stood in the sixteenth and seventeenth centuries at the top of Gibbet Street. It is, of course, a replica, but the platform is probably original and the original blade may still be seen in the Piece Hall Museum. We do not know when the guillotine first came into use in Halifax. One story is that a headsman got squeamish some time in the reign of Henry VIII - an unlikely time to have qualms about the axe. We do know on firm evidence that it was last used in 1650. Prior to that the parish registers, which begin in 1538, record fifty-six executions and in the whole run of court rolls back to 1274 nearly a hundred decapitations have so far been found, the first in 1286 within a dozen years of the first surviving roll.89 The instrument seems to have been unique in England. The nearest rival machine was at Edinburgh, where the Scottish Maiden, as it was known, was still in use in the seventeenth century. It crops up here and there in western Europe in the late medieval and early modern period - in Germany, in northern Italy and in the south of France. In some of these areas it was reserved for noblemen as a more humane and dignified end than the hangman's noose 88

The rolls for 1274^1331, with the exception of 1311,1326 and 1329, are published in Yorkshire Archaeological Society, record series, 29,36,57,78,109, ed. or trans. W.P. Baildonjohn ListerJ.W. Walker (1901-45); and rolls of various dates in Wakefield Court Rolls, Yorkshire Archaeological Society (Leeds), general editors Richard Vaughan and C.M. Foster, 1-9 (1977-93), and continuing. 89 Edward Armitage 'Halifax Gibbet Law', Transactions of the Halifax Antiquarian Society (1948), pp. 1-17, at pp. 14-15 for the executions after 1538. See also T.W. Hanson, The Halifax Gibbet Custom', ibid., pp. 47-58; for the earliest examples pp. 48ff. The first certain evidence locating decapitation within Sowerbyshire (on which see below, p. 24) comes from 1314 (Hanson The Halifax Gibbet Custom', p. 49).

Colonial England, 1066-1215

21

or the blundering headsman's axe. It is frequently illustrated, most dramatically perhaps by Lucas Cranach in 1539.90 Later Dr Guillotin cribbed it and the French promptly gave his name to it. In 1708, fifty-eight years after the last use of the machine, a small octavo volume, Halifax and its Gibbet Law, was published in Halifax by William Bentley.91 He was probably also the author; as clerk to the local justice of the peace and to the parish church, he was one well qualified to comment. He sought to justify the old procedure. Local custom, he argued, was better adapted to local circumstances than statute law. He then described the local circumstances and gives us a fine old social ritual: There being, as hath been said, three Meeting, or Market Days in Hallifax, every Week, for Traffick in all sorts of Commodities, saving Cloth, which is only bought and sold on the Saturdays: On every such Meeting Day, the Felon is set in the publick stocks, and either upon his Back, if the Thing stoln be portible, or if not, then before his Face the Goods are so placed, that they may be noted by all Passengers. And this is done, both in Terror to others, that they may take Warning by his wicked Deeds, never to commit the like. And also that he being known for a Felon, it may engage any Person that hath ought against him to bring in their Accusation. And on the day of the execution: The Prisoner being brought to the Scaffold by the Bailiff, the Ax being drawn up by a Pulley, and fasten 'd with a Pin to the Side of the Scaffold; the Bailiff, the Jurors, and the Minister chosen by the Prisoner, being always upon the Scaffold with the Prisoner, in most solemn manner. After the Minister hath finished his Ministerial-Office, and Christian Duty: If it was a Horse, an Ox, or Cow, etc. that was taken with the Prisoner, it was thither brought along with him to the place of Execution, and fasten'd by a Cord to the Pin that stay'd the Block, so that when the time of the Execution came (which was known by the Jurors holding up one of their Hands) the Bailiff, or his Servant whipping the Beast, the Pin was pluck'd out and Execution done. But if there be no Beast in the Felon's Case, then the Bailiff, or his Servant cut the Rope.92 Plainly it was conducted with a flourish. It was a social ritual; proceedings were apparently enlivened by the town bagpiper, accompanying the 4th psalm - 'I will both lay me down in peace and sleep' - a somewhat grim text in the circumstances.93 90

Armitage, 'Halifax Gibbet Law', p. 17. The volume was republished on various occasions, the last apparently byj. Horsfall Turner. The tide underwent some changes, in particular The History of the Famous Town of Hallifax in Yorkshire: With a True Account oftheAntient Odd Customary Gibbet Law (E. Tracy, London Bridge, 1712). It is from this edition that all quotations below have been taken. 92 Bentley, Halifax and its Gibbet Law (Halifax, 1708), pp. 30-31, 67-68. 93 The piper and the psalm are not mentioned in Bentley. They first appear in Thomas Wright, The Antiquities of the Town of Halifax in Yorkshire (Leeds, 1738), p. 202. Wright was curate of Halifax: 91

22

Colonial England, 1066-1215

To explain all this our writer first gives us some social and economic history. The area had become wealthy and populous. It attracted strangers and newcomers some of whom were 'wicked and ungovernable'. They fell to stealing: tempted therunto by Satan and their own Wants, together with the Opportunity laid before them of taking their Neighbours Cloth from off the Tenters, which they conclude may be done without Discovery, in regard those Tenters are at some Distance from their Houses, and their Houses at a much greater Distance from Neighbours, to discover their Wickedness, which thing hath but given too frequent occasion unto Men who want means to supply their extravagant Expences, to rob honest Men, and most commonly they do it unto those who are least able to bear the loss.94

That was the explanation adopted by some local antiquaries in the years before the Second World War: the local weavers exercising peoples'justice. By itself it will not do, for two reasons: first what the writer of 1708 was describing was the ancient process of infangtheof, the right to execute a thief caught in the act, still used on occasion in the seventeenth century in cases of manifest grand larceny. At Halifax the jury was simply asked two questions: was the accused in possession of the stolen goods; and were they worth at least thirteen and a halfpence? If the answer is 'yes' to the guillotine he (or she) goes. But that was merely a local refinement of the ancient procedures whereby a thief pursued and caught with the goods on him was executed after summary procedure in a local court. In Northumberland also he was beheaded. At Dover he was thrown over the cliffs. At Sandwich he was buried alive.95 In other townships besides Halifax the stolen goods were bound to his back. Secondly, the r^al interest of the Halifax procedure lies in its concern for the stolen goods. Our writer gives the game away with an explanation much more characteristic of his native heath. Victims, he says: cannot prosecute such Felons at the common Law, according to their Merit, if they should do it, what by reason of their Distance from the Place of the Assizes, which is held but twice in the Year, and are also so vastly Expensive, that they have neither Money nor Leisure wherewith to prosecute the Felon: Besides, well knowing that if they should do it, they do but spend both their time and their Money, to recover those Goods they must never enjoy, as being forfeited to the Crown. continued

his book was little more than a gloss on Bentley although he claimed to correct many of his predecessor's errors. In earlier days the ritual apparently took a more primitive and popular form. William Harrison, writing 1577-87, gives us: 'when the offender hath made his confession and hath laid his neck over the nethermost block, every man there present doth either take hold of the rope or putteth forth his arm so near to the same as he can get, in token that he is willing to see true justice executed, and pulling out the pin in this manner, the head block wherein the ax is fastened doth fall down with such violence that if the neck of the transgressor were so big as that of a bull it should be cut in sunder at a stroke and roll from the body by an huge distance'. See William Harrison, The Description of England, ed. George Edelen (Ithaca, NY, 1968), p. 191. 94 Bentley, Halifax and its Gibbet Law, pp. 34-35. 95 Pollock and Maitland, History of English Law, ii, p. 496n.

Colonial England, 1066-1215

23

Hence the gibbet law permits: the Party injured to have his Goods restored to him again, with as little loss and damage, as can be contrived; to the great Encouragement of the Honest and Industrious, and as great Terror to the Wicked and Evil-doers.96

There we have it. The objective was immediate recovery of the stolen goods. Usually the felon's chattels were forfeit to the king. Increasingly the courts tended to accept that the felon's goods included those he had stolen. An appellant might sue for them - there was an appropriate writ - and if he pressed his case urgendy (which involved costs) he had some hope of recovery. If, however, the felon had been indicted by jury, the victim had no such hope of recovery until the law was changed in his favour in 1529.97 Halifax by-passed all these complexities. Restitution was made immediately on execution of the malefactor. So the procedure was not all ritual. It was intimately related in a very hard-headed way to the procedures of the common law and the long arm of royal justice. The ancient procedure of infangtheof, sustained in a local court, had become a forfeiture evasion system. How did it begin and how did it end? The second is easier to answer. The last executions occurred in 1650; in 1708 Bentley's interest was antiquarian. The procedure had come to the attention of central government over the previous century, regarded as somewhat odd, but allowed to continue. 'Halifax Law', signifying summary justice, was common parlance between Robert, earl of Leicester, and the great Lord Burghley.98 Oliver Cromwell and the Major-Generals stopped it. To them it was part of ancient ritual to be jettisoned along with all the old feasts and celebrations of the medieval world and the church of Rome. Moreover, it ran counter to the puritan objection, in which Oliver himself shared, to imposing the death penalty for petty theft. So it ended and henceforth felons went for trial to the assizes at York; thus the Luddite leaders 160 years later, some hanged, some transported. Nevertheless the guillotine was long remembered. Mrs Gaskell instanced decapitation as an example of the ancient barbarity of the folk in the hinterland of Haworth." The decay of memory was slow, a combined result perhaps of John Wesley and the Lancashire and Yorkshire Railway. The origins are more obscure and more interesting. The guillotine was already well known when William Harrison compiled his Description ofEnglandfor Holinshed's Chronicles between 1577 and 87.100 That seems to be the first reference to the machine. The first instance of decapitation, as I have indicated, comes from 1286. The practice was obviously ancient, more ancient indeed than that 96

Bentley, Halifax and its Gibbet Law, pp. 35, 36. Pollock and Maitland, History of English Law, ii, pp. 164-66; T.F.T. Plucknett, A Concise History of the Common Law (London, 1948), p. 426. 98 Armitage, 'Halifax Gibbet Law', pp. 10-11; Hanson, pp. 57-58. 'I have had Halifax Law, to be condemned first and enquired upon after', Leicester to Burleigh, March 1586, Calendar of State Papers Foreign, 1585-86, p. 452. 99 Mrs Gaskell, Life of Charlotte Bronte (London, 1891), p. 13. 100 See above n. 93. 97

24

Colonial England, 1066-1215

date suggests. For there were geographic bounds to it. If you escaped over Hebble Beck (which runs through Halifax just below the church on its way down to join the Calder) then you had the satisfaction of knowing that you might be hanged in Wakefield rather than beheaded in Halifax. At that point therefore there was a precise boundary. For the rest the jurisdiction of the axe did not coincide with the parish of Halifax, which was established no earlier than the twelfth century. Indeed it extended over a much wider area, embracing the whole of upper Calderdale as far as the crest of the Pennines and the border with Lancashire.101 In the rolls this area is frequently described as the Forest of Hardwick or the Forest of Sowerby, the hunting preserve of the Warenne earls. But behind that there lies another much older term: Sourebysdr.102 Sowerby, which unlike Halifax lies vantageously in the Calder valley, must have been the ancient centre, and Sourebyscirtakes us back to the ancient administrative divisions of the earliest Anglo-Saxon settlements (some would push scirs even further back) still represented east of the Pennines by Hallamshire and Richmondshire (also Howdenshire, Allertonshire, Norhamshire and others). The use of the axe was circumscribed by the bounds of this ancient unit. Beheading, therefore, was probably an ancient practice. It is a thin thread of argument but a strong one, tested in many other cases both in England and on the Continent.103 The most likely conclusion is that decapitation was the practice adopted by the English settlers of the Upper Calder valley. Who knows why? It endured for at least three-quarters of a millennium. The horseman or packman of the eighteenth century entering Calderdale from Lancashire and travelling the old road from Blackstone Edge to Ripponden, might still pick up a grim reminder of the past. As he descended towards Soyland there was a hill high to his left which bore the name Man's Head End.104 Today the name remains, still there to remind the passing motorist that he is entering Souerbyscir. 101

For the bounds and the townships which lay within the forest of Hardwick see Bentley, Halifax and its Gibbet Law, pp. 24-25, 27, and for further discussion, Hanson, The Halifax Gibbet Custom', pp. 55-56. 102 Armitage, 'Halifax Gibbet Law', p. 8; Hanson, The Halifax Gibbet Custom', pp. 53-54. See also Early Yorkshire Charters, 8, ed. C.T. Clay, Yorkshire Archeological Society, record series, extra series (1949), pp. 21, 207, 233-36. For a recent account of the area, see Bernard Jennings, ed., Pennine Valley: A History of Upper Calderdale (Otley, 1994). 103 It has become a favourite line of attack in the study of the origins of French coutumes. Hanson, The Halifax Gibbet Custom', p. 54, pursued it quite independently. Even more strikingly, J. Horsfall Turner commented in The History ofBrighouse, Rastrick and Hipperholme (Idel, 1893), p. 107: 'It is as yet unaccountable that Sourby graveship should have exercised the "privilege" of gibbetting thieves caught within its boundaries, whilst Hipperholme and Rastrick graveships seem to have had no such privilege. Even the Sourby culprit was free if he could escape across the Halifax beck, so long as he did not return. The custom must have come down from Saxon times before the Manor of Wakefield included the Halifax portion, or all would have had the same rigorous treatment.' 104 Thomas Jeffreys, The County of York (1775), ed. J.B. Harley and J.C. Harvey (Lympne Castle, 1973), plate xii.

2 Domesday Book, 1086-1986

Why all the fuss about it? After all it amounts simply to two large manuscript volumes. Why celebrate? Certainly 900 years is a long time, but there are plenty of manuscripts of greater beauty which are older than that. The occasion, to be sure, was exploited commercially, but what is not? And the shade of William the Conqueror would scarcely object. The plain fact is that Domesday Book, like the battle of Hastings, Magna Carta and the Great Armada, is part of what every schoolchild knows, or at least used to know when schoolchildren were still so called. The name has helped. Those in the twelfth century who began to call it 'Domesdei' chose well, far better than liberjudidarius which was the Latin of officialdom; and Domesday is what it has remained 'because its decisions, like those of the Last Judgement are unalterable'. Thus Richard fitz Neal, treasurer to Henry II, writing in 1179. When that is said it remains a remarkable document. It is of course impressive in its bulk. It was already bound in book form before 1100 and until its rebinding in 1986 it survived in two volumes: Great Domesday, dealing with the whole of England south of the Tees, except for Essex and East Anglia which constitute Little Domesday. Great Domesday comprises nearly four hundred folios measuring 15 inches by 11 inches written in double format; Little Domesday comprises 450 folios measuring 11 inches by 8 inches written in single format. They are businesslike volumes. Both are rubricated, Great Domesday lavishly so, and to a purpose, namely to facilitate rapid reference. But both in size and appearance they are not out of the way; bulky manuscripts are not uncommon, some of them much more impressive in their calligraphy and decoration. It is the content which matters. As a comprehensive survey Domesday is unrivalled, not only in medieval Europe but anywhere at any time. It was subjected to a statistical economic analysis by John McDonald and G.D. Snooks in Domesday Economy (Oxford, 1986). They concluded that 'the Domesday data contain considerably less measurement error than modern economic survey or census data' and that 'while the methods of processing data have improved radically since 1086, in many respects the methods of data collection have deteriorated'. These judgments were based on an analysis of two counties: Cambridgeshire and Wiltshire. Whether they will stand for the whole country remains to be seen. But it may well be doubted whether the modern Domesday put together in celebration of the novocentenary will inspire such confident comment 900 years hence. William the Conqueror did better. His Domesday was systematic.

26

Colonial England, 1066-1215

Domesday was made like a pyramid or like the mottes of the Normans' castles, in layers. The base was made up of local information: some of it written, most commonly in the form of earlier tax assessments, some of it oral provided by the priest, reeve and six villeins of each village; all of it corroborated by the sworn testimony of juries in the county courts. The counties were combined into seven circuits, each with its own commissioners, responsible both for executing the enquiries and assembling the information. Hence the next layer or stage was to combine the county information within each circuit into one. One example of this stage survives - the Exon Domesday for the south western counties, still preserved at Exeter cathedral. These circuit volumes, or more probably fair copies of them, were then used to construct the final volume, Great Domesday. This was almost certainly completed at an agreed centre, possibly Winchester, where the Book soon came to reside. One such circuit return survives, Little Domesday for the eastern counties, which was never embodied in the final volume. So there were at least three stages: collection and ordering of information in the county, construction of the seven circuit returns; embodiment of six of the seven returns in Great Domesday. That is to put it schematically and at its simplest. In reality there may have been other stages depending on the nature and extent of written information available in the county, which, to judge from the surviving fragments, varied both in bulk and content. These fragments, the so-called Domesday 'satellites', guide our understanding of the supply of information and the procedure of the survey. The whole operation was executed to a preconceived plan. The articles of the enquiry are included in one of the satellite texts belonging to the abbey of Ely: What is the name of the manor? Who owns it now? Who owned it before 1066 in the time of King Edward the Confessor? What are its human and agrarian resources - men, ploughs, woodland, pasture and mills? What is it worth now? What was it worth in the time of King Edward and when King William gave it? Can more be got from it than it yields at present? The summary information in Great Domesday answers most or all these questions: In Gransden Earl Alfgar had eight hides of land assessed to the geld. There is land for fifteen ploughs. There are seven ploughs on the demesne, and twenty-four villeins and bordars have eight ploughs. There is a priest and a church: fifty acres of meadow; twelve acres of underwood. The pasture yields 5s. 4d. It was worth £40 in the time of King Edward, now £30. Ranulf holds it.

A great deal happened to the data between the initial sessions of the county courts and the completion of Great Domesday. The information was successively condensed: that is obvious from a comparison of the satellites and the Exon and Little Domesday with the final format of Great Domesday. In addition it was completely rearranged. Most of the initial data was ordered on a geographic basis, by hundreds or wapentakes, which were the administrative subdivisions of the county. In Great Domesday, in contrast, as in Exon and Little Domesday, it is ordered within each county by landholders. This rearrange-

Domesday Book, 1086-1986

27

ment was intended from the start and took place at an early stage, certainly no later than the county court enquiries. It was achieved very simply and elegantly by interlineating the names of landholders in documents arranged hundredally. The interlineations were then used to concoct new lists arranged by landholder. As a result the lists by landholders reveal a consistent hundredal order - an astonishing phenomenon which attracted much attention from Domesday scholars until this ingenious use of an eleventh-century database was properly understood. The survey was carried out with remarkable speed. King William issued his instructions at a council held at Gloucester at Christmas 1085. There is no real doubt that it was completed by 1 August 1086 when 'all landholding men of any account' came to Salisbury to swear fealty and perform homage to the king. One entry in both Exon and Great Domesday, concerning the bishop of Winchester's lands in Taunton, indicates beyond reasonable doubt that the breves, the Latin term used to describe the book before it was bound as a volume, were already in hand on that occasion. Thereafter the team which had conducted the Survey, which included great barons among the commissioners, must have been broken up, for the King mustered an army and crossed to France to fight his overlord, King Philip. So the chronological constraints on the Survey are severe and precise: no more than seven months. Within that short period there was not just one enquiry but two, a second set of commissioners verifying the work of the first. The two visitations led to two distinct but overlapping documents: Domesday and a geld enquiry, the record of which still survives for the south-western circuit. Great Domesday, the book, was completed with similar urgency. The Surrey section was certainly written before William the Conqueror's death in September 1087; that depends on a crucial marginal entry for the manor of Pyrford which can be related to one of William's surviving original writs; and it is probable that the whole of Great Domesday had been completed by then.1 Little Domesday was never condensed into the final format; it is usually assumed, reasonably enough, that William's death called a halt to the enterprise. Most scholars now accept that Great Domesday was written by one main scribe, with a second scribe providing amendments and additions. He was a methodical, swift, accurate worker: he would have to be! More important, behind him there was a single controlling mind at work imposing order on the whole operation and the final recension in Great Domesday. The hunt for him has become one of the sirencalls of Domesday scholarship. The editioprincepsofboth Great and Little Domesday appeared in 1783; it is still used. Since then it has been republished in various editions and translations, culminating in the magnificent facsimile of Great Domesday, bound in the original gatherings, prepared by Alec to Editions in 1986. In this light it is surprising how much Domesday scholarship is recent, some of it very recent. 1 That was my opinion in 1986. For further evidence and qualifications see the Note appended to no. 3, below, pp. 56-57.

28

Colonial England, 1066-1215

The modus operandi of the survey was not properly understood until 1942, when V.H. Galbraith gave proof of it in the greatest of all Domesday papers in which he challenged the older orthodoxies of J.H. Round and F.W. Maitland.2 The consensus that Great Domesday was the work of one main scribe was only reached in the course of the preparation of the Alecto facsimile. The hand of the scribe was first identified in other manuscripts in 1985-86.3 The method of conversion from a geographic to a tenurial base was first proposed by Dr H.B. Clarke in 1985,4 although it should have been obvious long before from a document embedded in the Book and available in the editio princeps, the so-called Yorkshire Recapitulation. The most convincing identification of the 'controlling mind' was advanced by Pierre Chaplais at the conference of the Royal Historical Society and the Institute of British Geographers held at Winchester in July 1986.5 So the celebrations of 1986 had an academic side to them. Real advances were made. There is still room for argument. In particular there is a running debate about William's purpose in ordering the survey. Here the quality of the book is itself to blame because it provides such a splendid base for the study of resources, tax assessment and land values. Historians from Round and Maitland onwards have worked marvels with the figures, demonstrating that tax yields fell between 1066 and 1086 or that assessments were awry because of the destruction of the Conquest, or the consequent dislocation in ownership, or tax evasion on the part of the new Norman magnates, and the like. All this is true. Ph.D.s are written on these and similar Domesday problems. The difficulty is that neither William nor his successors nor their government officials had much time to write Ph.D.s. Moreover, there is a logical flaw in an argument which assumes that the purpose of the Survey is to be found in its results, even more in the results which modern scholars can squeeze out of the data with calculators, computers and statistical formulae. So the notion that Domesday was, in Maitland's phrase, 'a geld book' involves some uneasy assumptions. A much better explanation, first advanced by Galbraith, is that Domesday was simply a record of settlement, delineating the landed resource of the king and his men: at one and the same time a final work of reference in matters of tenure and a guide to the management of estates if and when they came into the king's hands. This is the clear message of the manuscript, in its format and rubrication, for both together constitute a system of rapid reference to landholders, their holdings and the resources, values and tax assessments of those holdings. More consideration must also be given to the termination of the survey at Salisbury in the 2

V.H. Galbraith, The Making of Domesday Book', EHR, 57 (1942), pp. 160-77. M. Gullick and Caroline Thorn, The Scribes of Great Domesday Book', Journal of the Society of Archivists, 8 (1986), pp. 78-80; and M. Gullick, The Great and Little Domesday Manuscripts', in R.W.H. Erskine and Ann Williams, eds, Domesday Book: Studies (London, 1987), p. 102. 4 H.B. Clarke, 'The Domesday Satellites', in P. Sawyer, ed., Domesday Book: A Reassessment (London, 1985), pp. 50-70, especially 62-67. 5 Now printed as Pierre Chaplais, 'William of Saint-Calais and the Domesday Survey', inJ.C. Holt, ed., Domesday Studies (Woodbridge, 1987), pp. 65-77. 3

Domesday Book, 1086-1986

29

performance of homage and fealty. In relation to that, Domesday is nothing other than a record of enfeoffment corroborating the holdings of the new Norman settlers who were now performing homage for clearly delineated tenure. This will not be the end of the story. Centenaries leave no choice of time. This one occurred with programmes for the computerisation of Domesday in hand at three different centres: Flinders, Hull and Santa Barbara. Of these the first led to MacDonald and Snook's Domesday Economy. The second contributed two important papers to centenary publications in 1985 and 1987.6 The last is expected to be generally available on tape in the course of 1987 and will provide the index to the Alecto facsimile of Great Domesday; to put it at its very simplest it will give us for the first time a comprehensive subject-index.7 We are only beginning to exploit these new facilities; they are bound to affect Domesday studies profoundly. Some of the participants in the conferences of 1986 turned up Domesday Studies, the Royal Historical Society's celebration of the eighth centenary in 1886. Scarcely one of those essays has stood the test of time. There is something to be learned from that.

6 John Palmer, 'Domesday Book and the Computer', in Sawyer, ed., Domesday Reassessment, pp. 164-74; and idem, The Domesday Manor', in J.C. Holt, ed., Domesday Studies, pp. 139-53, which is an important revival of Maitland's view of the relationship of manor and geld. 7 The database originating in Santa Barbara awaits publication in electronic form. The Hull database was used in J.M.N. Dodgson and JJ.N. Palmer, Domesday Book: Index of Places and Persons, 2 vols (London, 1992), and indirectly therefrom byJ.D. Foy, Domesday Book: Index of Subjects (London, 1992); all these suffer from the idiosyncrasies of the Phillimore edition.

This page intentionally left blank

3

1086

On 1 August 1086 William the Conqueror held court at Salisbury and there received homage and fealty from his men. This was the central event of the year and is the pivot on which my argument will turn. To quote Stenton's translation of an oft-repeated passage from the Peterborough version of the AngloSaxon Chronicle, William came at Lammas to Salisbury, and his council came to him there, and all the landholding men of any account throughout England, whosesoever men they were, and they all bowed to him and became his men, and swore oaths of fealty to him that they would be faithful to him against all other men.1

For the present purpose many of the details of the translation do not matter: 'council' or 'counsellors', 'men' or 'vassals';2 nor is the definition of the much debated phrase 'landholding men of any account' of great moment. It perhaps matters more that on that day William accepted from his men both fealty and homage, for they swore their oath and also 'bowed down to him and became his men'. The obvious and generally accepted interpretation of the phrase is that this was an act of homage. The passage is a historiographic curiosity. It is enshrined in Stubbs's Charters? More than fifty years ago I sat at the feet successively of Edwards, Jolliffe and May McKisack: none of them suggested that the Salisbury oath had anything to do with the Domesday Survey. It came under a different heading, 'liege homage'; no one asked whether one famous event of 1086 was related in any way to the other. Years later, some time after 1974, when the present argument was beginning to take shape, I put the question direcdy to Galbraith, only to receive a downright if perhaps rueful dismissal: his two books on Domesday contain only one casual reference to the Salisbury oath.4 Galbraith was no exception. In an important and influential historical revision devoted to the Salisbury oath in History in 1934, H.A. Cronne commented: 'the fact that the Salisbury 1

F.M. Stenton, The First Century of English Feudalism, 1066-1166 (2nd edn, Oxford, 1961), p.

112.

2

Cp. EHD, ii, pp. 161-62. William Stubbs, Select Charters, p. 96. 4 V.H. Galbraith Domesday Book: Its Place in Administrative History (Oxford, 1974), p. 25. 3

32

Colonial England, 1066-1215

Council met about that time is a mere coincidence'.5 Since then in book after book on the Norman Conquest or the reign of William I historians have treated the Salisbury oath quite separately from the Domesday Survey, all of them in one way or another commenting on the oath in discussions of the superiority of the crown, or the particular characteristics of English feudalism, or the external threats to William's rule in 1085-86.6 The pedigree of these views is not in doubt. They stem from Stubbs and Freeman; they are a quaint survival from the wreck caused by Round, repaired rather than jettisoned through all the subsequent revisions in the study of the Conquest which he began.7 There was a still older view, scathingly denounced by Freeman, more judiciously criticised by Stubbs.8 This was derived from Martin Wright's Introduction to the Law of Tenures of 1730,9 and was given full expression in Blackstone's Commentaries.10 It amounted to nothing less than an argument that Domesday and the Salisbury oath marked the introduction of feudal tenures, that Domesday described the terms of such tenure and that at Salisbury homage was performed for the terms so described. Wright commented: as this general homage and fealty was done about the time that Domesday Book was finished, and not before, we may suppose that the Survey was taken upon or soon after our ancestors' consent to tenures in order to discover the quantity and to fix his homage.11

Ellis summarised Blackstone and quoted Wright in 1833, but he also noted the opposite views of Coke and Selden that such tenures were of ancient English origin. He added perceptively that 'the internal evidence of Domesday itself bears no reference whatever to any simultaneous surrender of former tenures and re-grant of the same as feudal'.12 There matters rested until Freeman launched his assault. 5

HA. Cronne, The Salisbury Oath', History, 19 (1934-35), pp. 248-52, especially 251. See, for example, F.W. Maitland, The Constitutional History of England (Cambridge, 1908), p. 161; G.B. Adams, The Origin of the English Constitution (New York, 1912),pp. 186-87; H.W.C. Davis, England under the Normans and Angevins (London, 1905), pp. 36-7; H.R. Loyn, The Norman Conquest (London, 1965), pp. 127-28; Frank Barlow, Williamland the Norman Conquest (London, 1965), pp. 110-11; R. Allen Brown, The Normans and the Norman Conquest (1969), pp. 240-41. 7 W. Stubbs, The Constitutional History of England (3 vols, Oxford, 1874^78), i, pp. 288-90; EA. Freeman, The History of the Norman Conquest (6 vols, Oxford, 1867-79); iv, pp. 694-96. For comment see Cronne, The Salisbury Oath*. 8 Freeman, History of the Norman Conquest, v, pp. 366-67; Stubbs, Constitutional History, i, p. 289nn. 9 Martin Wright, An Introduction to the Law of Tenures (1730), pp. 52-58. 10 W. Blackstone, Commentaries on the Laws of England (4 vols, Oxford, 1770), ii, pp. 49-50. 11 Wright, Introduction to Law of Tenures, p. 56. Wright continues: This supposition is the more probable because it is not likely that a work of this nature was undertaken without some immediate reason, and no better reason can be assigned why it was undertaken at this time or indeed why this Survey should have been taken at all' (ibid., pp. 56-57). See also his note: 'And this was the reason why almost all the historians of those times join the account of this Survey and of the Homage done about that time together, in such a manner that we must needs think they took them to have immediate relation one to the other' (ibid., p. 56n.). 12 Henry Ellis, A General Introduction to Domesday Book, 2 vols (London, 1833), i, pp. 15-18. 6

1086

33

No one now accepts Blackstone's argument. We seem no longer to believe in feudalism let alone the notion that it was established at a single stroke in 1086. But the case included two cogent points. First, the coincidence of the Survey and the oath cannot simply be dismissed. In 1961 Welldon Finn concluded a discussion of the date of completion of Domesday Book with the comment: The time-limit is reached, perhaps that of the Lammas gathering at Salisbury, when all the land-owning men of any account in England, no matter whose men they were, did the king homage and swore fealty to him, and at which, perhaps, there were brought to him the 'writings', with each county still separated from its fellows.13

Other scholars have followed that view, two in 1986.14 Meanwhile Sally Harvey has put forward an even closer relationship, namely that the Salisbury oath was a consequence of the Domesday Survey, a means of disciplining undertenants the importance of whom the Survey had just revealed.15 Secondly, and more important, Blackstone's case implied a question which no one has asked or answered since. At Salisbury King William's men performed homage to him. For what? No one should rule Blackstone out of court without first answering that. The chronology of 1086 depends very much on the evidence of the Peterborough chronicler. He tells of a year begun in routine fashion - a court at Winchester for Easter, another at Westminster at Whitsuntide at which William knighted his youngest son, Henry.16 Then Salisbury on 1 August marked a break, not just in the famous ceremony but also in the rhythm of the year. William left Salisbury for the Isle of Wight with the intention of going to Normandy. He crossed the Channel later in the year; the precise date is unknown. If that is correct, and there is no good reason for rejecting it, it imposes a strict consequence, namely that the Domesday Survey was completed by 1 August and the king's subsequent departure to the Isle of Wight. No king in the eleventh and twelfth centuries went to the Isle of Wight unless to cross to France. William's fortunes in France, in Maine and on the Vexin border, were deteriorat13 R. Welldon Finn, The Domesday Inquest and the Making of Domesday Book (London, 1961), p. 190; repeated in idem, An Introduction to Domesday Book (London, 1963), pp. 12, 93. 14 See, for example, D.C. Douglas: The Salisbury oath, like the Domesday Inquest of which it was in some sense the counterpart was the king's response to a challenge [the external threat of 1085-86]', William the Conqueror (London, 1964), p. 356. Compare H.B. Clarke The Domesday Satellites', Domesday Reassessment, p. 56: 'Indeed there may be a connection between the presentation of "all the writings" to King William and the famous oath of Salisbury' and Elizabeth Hallam, Domesday Book through Nine Centuries (London, 1986), p. 24: The colophon, which states that the survey of England was made by King William in 1086, implies that all the final drafts were completed in that year, perhaps even by 1 August when the king received his oaths of homage from many important under-tenants.' 15 Sally PJ. Harvey, 'Domesday Book and Anglo-Norman Governance', Transactions of the Royal Historical Society, fifth series, 25 (1975), p. 190. 16 EHD, ii, pp. 161-62; The Peterborough Chronick, 1070-1154, ed. Cecily Clark (Oxford, 1958), pp. 8-9.

34

Colonial England, 1066-1215

ing under the combined threat of King Philip, Fulk of Anjou and Robert of Flanders.17 William can scarcely have crossed without a considerable force; war in France was impending; his expedition culminated in his attack on the Vexin and the sackpf Mantes Gassicourt in the following year. There is other evidence to be brought into play in support of this account. First, the Peterborough chronicler himself tells us that after the Salisbury assembly William exacted 'a very great amount of money from his men where he had any pretext for it either justly or otherwise' - which can be read, and may be best read, as an exaction of penalties from individuals charged with the wrongful possessions which the Domesday Survey reported.18 Secondly, one of the last surviving written instruments of the reign, in favour of the abbey of St-Amand of Rouen, indicates beyond reasonable doubt that William Rufus, Robert, count of Mortain, Eudo Dapifer, Robert Dispensator, Hugh, earl of Chester, Gilbert de 1'Aigle and Walter fitz Richard were all in the king's company in Normandy in 1086-87.19 Thirdly, and more critically, in a well-known entry in both Exon and Great Domesday, it was recorded that King William had conceded lands in Taunton to St Peter and Bishop Walkelin (of Winchester) 'as he acknowledged at Salisbury in the hearing of the bishop of Durham whom he instructed that the concession should be entered in his breves1.20 The hand of this section occurs nowhere else in Exon Domesday. It is also exceptional in that it did not follow the common practice in Exon Domesday of allotting a quire or quires to each holder of land. Ker concluded that 'it was probably added in a convenient blank space after the rest of Exon Domesday had been completed'.21 If so, it can scarcely refer to other than the meeting at Salisbury on 1 August 1086, in which case it indicates not only that one item in the Survey was part of the business at Salisbury but also that the king's breveswere almost certainly in hand at that time. That is undoubtedly the best reading of the passage. All these points reinforce the simple common-sense argument that William would scarcely have interrupted the work of the Survey in order to hold the meeting at Salisbury. It follows, therefore, that it was complete by 1 August. The starting-point is even more certain: the Domesday Survey was set on foot at Gloucester, Christmas 1085. That left seven months, including the worst of the English winter, for its execution. It was a tall order. In reality the time available was even less than that. Work could have begun 17

Douglas, William the Conqueror, pp. 356-58. Peterborough Chronicle, p. 8; EHD, ii, p. 162. 19 Archives Seine-Maritime, 55H, carton 1: a confirmation and attestation by William I of the grant by Maurice, bishop of London, to St-Amand of his tithes from the forests of Aliermont and Eawy. The confirmation was also attested by William, archbishop of Rouen: hence the conclusion that it was issued in Normandy. The document is calendared with an incomplete witness-list in RRAN, i, no. 285, cp. ibid, no. xlv; and in Calendar of Documents Preserved in France, ed.J.H. Round (London, 1899), no. 94.1 am obliged to Dr David Bates for a copy of the MS. 20 Liber Exoniensis, fo. I75v; Great Domesday Book, fo. 87v. For comment see V.H. Galbraith, The Making of Domesday Book (Oxford, 1961), p. 207. 21 N.R. Ker, Medieval Manuscripts in British Libraries (3 vols, Oxford 1969-83), ii, p. 806. 18

1086

35

in Gloucestershire immediately after the Christmas festival. It could not have started in Yorkshire, Lincolnshire or Kent much before midjanuary, and county courts could scarcely have been assembled in most counties much before the end of the month. At the other end of the process the Salisbury assembly must have required approximately four weeks' notice; William must have known no later than the beginning of July that the work had been completed or would be completed during that month; the circuit commissioners must have reached that conclusion even earlier. These are not guesses. They are based on known facts easily arrayed from parliamentary writs of summons and returns of the thirteenth century.22 So from the total of seven months available for the completion of the Survey a deduction of some four to six weeks must be made to allow for administrative lag of one kind or another. The remaining period of less than six months is an even taller order. Into that we have to fit two inquiries, following the evidence of Robert Losinga, bishop of Hereford, and three if the geld inquisition is dated 1086 and distinguished from the other two. There are not many ways of reducing that overloaded schedule. One is to put one inquiry after 1 August as a kind of rounding-off exercise or to allow that some of the business of the Survey dragged on after that date: that only reduces, it does not remove the difficulties outlined above; it is an unlikely option if the king's men were mustering for an expedition overseas. A second is to put the geld inquisition or at least the geld with which the inquisition was concerned, earlier, before the Christmas council of 1085: some have argued for that,23 but it still leaves two inquiries in the first six months of 1086. A third is to assume that one of Robert Losinga's inquiries and the geld inquisition were one and the same: that is possible, even likely.24 A fourth does not recover much time, but is much more interesting: it is that the Salisbury assembly required no administrative planning in the spring of 1086 because it was predetermined, because it was known when the Survey was ordered at Christmas 1085 that it would end with the performance of homage and fealty and that all that there remained to do was perhaps to arrange or confirm a place and date, which could easily be done at the Easter court at Winchester (5 April) or at Westminster at Whitsuntide (24 May). That would imply that the Survey was conducted to a strict schedule ending with a dead-line. The generally accepted explanation of Little Domesday is that somehow it missed just such a dead-line.25 Consider the implications of such a schedule. Some are well known and require no more than a brief mention here:

22

J.C. Holt, The Prehistory of Parliament', The English Parliament in the Middle Ages, ed. R.G. Davies andJ.H. Demon (Manchester, 1981), pp. 11-13. 23 Finn, The Domesday Inquest, pp. 147-48. 24 Galbraith, The Making of Domesday Book, pp. 94-95. 25 V.H. Galbraith, Studies in the Public Records (London, 1948), p. 97; Finn, The Domesday Inquest, p. 190; the dead-line usually being assumed to be the death of William I.

36

Colonial England, 1066-1215 1

2

3

It is agreed that the Survey involved a complex intermingling of data, some of it derived from landowners, some of it derived from local courts, all of it confirmed by juries of the hundred in the county courts, data which were reduced to order by adopting at an early stage the tenurial pattern of arrangement which is the format of the final book. That depends on an analysis of the so-called satellites, some of them sources, others by-products of the Domesday data. Now there is little difficulty in relating one satellite to another in a logical pattern reflecting the accumulation of the data. But it does not follow that each logical step was also a necessary chronological stage, still less one involving the transcription and redrafting of data or the verification of information. Sally Harvey and others are surely right in seeking to reduce the successive stages to the absolute minimum.26 1086 was no time for frills or repetition. Even so, the variety of the satellites demonstrates how complex the procedure was and how tight the schedule is. It is agreed that the collection of data in the county courts could not have been done without drawing on a great deal of written information. Some of this was in the form of pre-Domesday records concerned with the assessment and collection of the geld which recorded hundred by hundred the separate assessment of each vill and landowner. Some of the written information was provided by the landowners themselves, especially by ecclesiastical landowners. This leads in turn to: The Survey could not have been carried out in the time available without the cooperation of landowners. Indeed it could scarcely have been launched at all unless William knew that it would have the support of his barons and ecclesiastical tenants, ready not only to supply information, but also to attend the county courts as the Survey proceeded. Much has occasionally been made of an outcry against the Survey. This depends on a short critical comment in the Peterborough chronicle,27 and on the lengthier statement of Robert Losinga that 'the land was vexed with many calamities arising from the collection of the royal money'.28 It is to be doubted how far the vexation and calamities extended to the king's barons except to some of those guilty of flagrant invasiones. Occasionally a holding might be concealed.29 Some men failed to answer for their

26 Sally PJ. Harvey, 'Domesday Book and its Predecessors', EHR, 86 (1971), p. 755; Finn, The Domesday Inquest, pp. 147-48,189-90. Cp. Clarke, The Domesday Satellites', which assumes a much more lavish provision both of circuits and of preliminary stages in producing the final texts. 27 There was no single hide, nor a yard of land, nor indeed (it is a shame to relate but it seemed no shame to him to do) an ox, nor one cow nor one pig was there left out and not put down in his record', £M), ii, p. 161. 28 W.H. Stevenson, 'A Contemporary Description of the Domesday Survey', EHR, 22 (1907), p. 74; Galbraith The Making of Domesday Book, pp. 51-52. 29 DB, i, 149a; DB, ii, lOla; Exon, 506a; R. Welldon Finn, Domesday Studies: The Liber Exonimsis (London, 1964), p. 72.

1086

37

manors,30 but they constitute a very small proportion of the total and there are many possible explanations of non-attendance other than wilful resistance. Against their example there is that of Robert Malet apparently referring as a defence to the day on which he was inbreviatus.31 Then we know the names of the legati of the west Midland circuit: Remigius, bishop of Lincoln, Walter Giffard, subsequently earl of Buckingham, Henry de Ferrers, ancestor of the earls of Derby, and Adam, brother of Eudo Dapifer.32 To them William of St-Calais may be added for the southwestern circuit.33 Two especially privileged tenants-in-chief, Roger of Montgomery and Hugh of Avranches, must have been responsible for mustering the county courts of their respective earldoms, Shropshire and Cheshire. Many other lords held 'private' hundreds; their stewards and bailiffs must have played their part in mustering the hundredal evidence and the hundred juries. It is unlikely that all these were pressed men. Galbraith rightly commented: The ruling class was no more than a few score of great men, each of whom was vitally interested in the results of the Survey'.34 What that interest was depends on the object of the Survey and the function of its outcome, Domesday Book. And the purpose of the Book can only be discussed on a common ground of assumption. One assumption which now seems to be generally agreed is that Great Domesday fairly reflects what King William and his men intended, in the information it contains and in its general layout and detailed format. Indeed, the strongest impression which it conveys, apart from the sheer immensity of the achievement, is that of a single controlling mind at work: unable perhaps to eliminate all the variations, for example in systems of measurement, between one circuit return and another; incapable, certainly, of reducing all the vagaries of local tenures to a single coherent scheme;35 but determined and clear-headed to an astonishing degree, sufficiently so that he arranged his data in so highly organised a fashion that it permits, even invites, sophisticated statistical analysis 900 years later.36 We can only conclude that the result was what was intended. No one now seriously accepts, with Round, that the data was rearranged from a geographic to a tenurial format almost as an afterthought.37 No one seems to have been convinced by the assertion of 30

Henry Ellis, Genial Introduction to Domesday Book, i, p. 31n. DB, ii, p. 276b. 32 Hemingi cartularium ecclesiae Wigorniensis, ed. Thomas Hearne (2 vols, Oxford, 1723), i, p. 288. 33 Galbraith, The Making Domesday Book, p. 87. 34 Ibid., p. 65. 35 Darby, Domesday England (Cambridge, 1977), pp. 9-56, 375-84, 36 See especially J. McDonald and G.D. Snooks, The Domesday Economy (Oxford, 1986). 37 See the criticism in Galbraith The Making of Domesday Book, pp. 12, 27; idem, Domesday Book: Its Place in Administrative History (Oxford, 1974), pp. 1, 17. Round's view is perhaps most clearly worked out in Feudal England (London, 1895), pp. 123-42; and in idem, 'An Early Reference to Domesday', in Domesday Studies, ed. P. Edward Dove (2 vols, London, 1888-91), ii, pp. 539-49. But the old view is best summarised in a few words by EM. Stenton, Anglo-Saxon England (Oxford, 1971), p. 655. 31

38

Colonial England, 1066-2215

Richardson and Sayles that it was all a 'vast administrative mistake'.38 What we have in Great Domesday is what was intended. Indeed, the purpose of the Book cannot seriously be discussed on any other basis, for once intention and result are detached one from the other, imagination can roam freely over all kinds of predilections. Therein lies the difficulty. For if we take the Book as an accurate reflection of its purpose it rapidly becomes obvious that it is far easier to determine what it was not than what it was. First, it was clearly not a geld book as Round and Maitland thought; that is to say, it was not designed primarily for the assessment (though it is concerned with that) still less the collection of the geld. The fundamental argument of Galbraith, that the final format is not that of a geldbook and that the final format was intended from the start and is in no way accidental, still stands. It is neither invalidated nor weakened by the argument that the crisis of 1085-86 and the consequent drafting of mercenaries into England was good cause for a heavy geld accurately assessed.39 Nor is it undermined by the survival of an occasional geld record arranged tenurially from before 1066,40 for Dr Clarke has recently pointed out what should have been obvious for years from the so-called Yorkshire recapitulation, namely that there was no difficulty at all in passing from a territorial to a tenurial order or vice-versa.41 The plain fact is that the data in Domesday Book are inconveniently arranged for geld purposes. They leave no room even for a compromise hypothesis, for example that Domesday was aimed at shifting geld liability from a geographic to a tenurial basis.42 No attempt was made to total geld liability, whether of landowners, counties, hundreds or vills. In particular, no attempt was made to total the liability of landowners within each hundred. The inquisitio geldifor the south-western counties, by contrast, provides totals of hidage,

38

H.G. Richardson and G.O. Sayles, The Governance of Medieval Englfind (Edinburgh, 1963), p. 28. For recent emphasis of this see Sally Harvey, 'Domesday Book and Anglo-Norman Governance', pp. 181-82; also Douglas, William the Conqueror, pp. 346-47. The point is at least as old as Blackstone, Commentaries (Oxford, 1770), ii, p. 49. 40 Harvey, 'Domesday Book and its Predecessors', pp. 755-59; idem, 'Recent Domesday Studies', EHR, 95 (1980), pp. 122-23. 41 Clarke, The Domesday Satellites', pp. 62-64, 67. Dr Clarke's argument establishes that the transfer from a geographic to a tenurial base was done by interlineating the landholders in documents arranged hundredally, from which were drawn up new lists arranged by landholder which necessarily retained a consistent hundredal order. He does not pursue the obvious conclusion that such a documentary change of data-base is likely to lie behind consistent hundredal order wherever it occurs throughout DB. For earlier discussion of the problem see P. Sawyer, 'The "Original Returns" and Domesday Book', EHR, 70 (1955), pp. 177-97; Galbraith, The Making of Domesday Book, pp. 15665. 42 In addition to the arguments advanced above see Tail's review of F.W. Maitland, Domesday Book and Beyond (Cambridge, 1897) in EHR, 12 (1897), pp. 768-72; and Maitland's letter in reply thereto in The Letters of F.W. Maitland, ed. C.H.S. Fifoot (Cambridge, 1965), nos 200, cp. 264. 39

1086

39

hundred by hundred, as a matter of course. It provides a summa of geld for Dorset and a summa of hides for Devon.43 Comparing the two documents, and relating them to the evidence of Robert of Hereford, it seems plain that there were two different inquiries, one concerned with the geld and the other with many other matters as well, and two different formats, one arranged geographically, hundred by hundred; the other tenurially or feudally, tenant by tenant within each county. The two formats are not totally distinct; each uses the other. But the documentary duplication for the south-western counties carries one forceful implication: there were two distinct documents with two distinct purposes; and if the geld was the objective of the inquisitio geldi, then the objective of Domesday was something else. As a result Domesday Book does not look like a geld-book. It looks like a directory with lists of tenants at the head of each county, each one numbered to correspond with his appropriate entry. But if it is not a geld-book it is equally clear that it is not simply a directory or, as Galbraith put it, a feodary. In the first place both the Survey and the Book were concerned with geld assessment; they asked and answered the question: how many hides? Now it may be that they did so because this was the only way of expressing amounts of land; that could be supported by much other documentary evidence. But that would not explain why the data contain many instances where changes in geld assessment are noted. In the second place, Domesday is concerned with very much more than geld assessment. It asks and answers questions about resources - how many ploughs? how much meadow? how much woodland? how many mills? and so on; and it assesses values, values in the time of King Edward before the Conquest of 1066, values when King William gave the land, values now. It also tells us 'whether more could be had than was being taken'. In short, the Book is very much more comprehensive in its information and very much more concerned with economic resources than we might expect if it were intended as a feodary and nothing more. We are faced, therefore, with a quandary. Domesday is not a geld-book because it is not so arranged and it looks like a feodary. But it is not a feodary either because it contains information more appropriate to a geld-book or a survey of resources. Faced with this, most historians have settled for a mixed function and purpose which we might summarise as follows: 1 2 3

43

Domesday is a description of economic resources and fiscal (geld) assessments. It is arranged in a tenurial or feudal format which identifies owners and in so doing resolves disputed ownership. In some way therefore it has to do with recording the total resources of the country, county by county, landowner by landowner.

Exon, 24a,7la.

40

Colonial England, 1066-1215

Each historian may mix these ingredients to his own particular taste, but that is always the resulting mixture, in one form or another. For various reasons this will not do, not at least without severe qualification. Domesday may contain a vast array of economic data but at the time it must have been a very poor instrument for assessing total economic resources of counties, hundreds or individual tenants. Just as the Book provides no totals of geld so it provides no totals or subtotals of any of these units of calculation. The larger or more complex the unit the more arduous it is to compute them. Where the larger tenancies-in-chief cross the boundaries of counties or of the Domesday circuits such calculations are made more difficult by the different measurements adopted in the different circuits. Doctorates are written on these matters; it took H.C. Darby and his collaborators five large volumes and many years to complete quantitative assessments of this kind. That the Book was intended to enable government officials in the eleventh century to do what Professor Darby needed years to accomplish is beyond belief. And it is easy to demonstrate that this was not the intention. The Exon Domesday does in fact contain totals of this kind: for the abbey of Glastonbury and for Ralph Mortimer, Milo Crispin, Robert fitz Gerald and Robert, count of Mortain.44 These refer to counties or groups of counties. They are almost certainly fragments of a larger endeavour rather than casual exercises. To give an example: The count of Mortain has in Wiltshire, Dorset. Devon and Cornwall 623 manors comprising 833 hides less 2!/2 virgates. There is land for 2480 ploughs; it is worth £1409 less 6s lOd. Of these hides the Count has 200 less 2 in demesne which are worth £400 and 1 mark of silver per annum; his men have 655 hides, less Vi virgate which are worth £1000 less 6s lOd.

This resulted in a total covering four counties and apparently involving all the lands of the count of Mortain in the south-western circuit.45 The most important point about it is that it was not included in the final text, Great Domesday. Nor were other similar totals which were included in Exon Domesday. Nor were the somewhat similar totals which appear in the Ely Inquest if indeed it derived its arrangement at this point from material preliminary to Domesday.46 So it is not simply that totals of this kind are difficult to calculate from the final Domesday data: they are not there in the Book: it was not intended that they should be: the opportunity to include them from preliminary material was rejected. Domesday was not to be used in this way. We may use it for what we are pleased to call macro-analysis. King William's men can have had no such intention. What then was its purpose? Or to rephrase the question significantly, how 44

Ibid., 527b-528a, 530b-531a. Ibid, 531b; Galbraith The Making of Domesday Book, p. 137. 46 Ibid., p. 137; Inquisitio comitatus Cantabrigiensis, subjidturinquisitioEliensis, ed. N.E.S.A. Hamilton (1876), pp. 121, 136. See also the lists of lands in Cambridgeshire, Norfolk and Suffolk arranged by tenants-in-chief which include total values and some totals of amounts of land, the totals usually being prefaced by 'Hoc appreciatum est' or 'Hoc totum appreciatum est', ibid., pp. 175-83. 45

1086

41

might it have been used by an official of the treasury under the Conqueror and his sons? That is to convert the problem into a practical question. It releases us at once from all those sophisticated interpretations which involve the logical error of deriving the purpose of Domesday from the perceived regularities (or irregularities) which modern analysis of its content reveals. It puts the Book on the same footing as a pipe roll, for example. And why not? At this level the answers are clear and straightforward. First, Domesday Book was concerned with the coun ty and by implication with the sheriff: that depends on the major divisions of the Book, dependent in turn on the original gatherings or groups of gatherings. Before it was bound the treasury official could move easily, as we ourselves may now move in the new facsimile, from one county to another. Secondly, within each county, the official was guided, as we are, to specific sections, to the county towns and county customs, to the king's lands, and finally- through the index of landowners at the head of each county - to tenants and under each tenant to individual manors. In this he was helped by the numbering of the index lists and texts, by the use of capitals and by the rubrication. Index lists, numbering, capitals and rubrications were all interlocked; the rubricator himself entered the numbers and usually the name of each landholder in the text. There need be no apology for recounting something so simple. These features have largely been forgotten amidst all the more sophisticated or esoteric explanations of the Book. The crucial point is that the features which I have described were not mere decoration but a frame of reference integral to the whole work and essential to its purpose. This system is obvious, more obvious once we turn from printed editions to the manuscript. In traditional Domesday scholarship it has frequently been mentioned or described, but it has not been discussed thoroughly enough to resolve all the problems it poses. These arise in the relationship of index lists to texts and in the numbering of both. Galbraith thought that the lists were entered first,47 but this is not easily reconciled with the entries for Berkshire, Dorset and Warwickshire, all of which are compressed into insufficient space.48 He also pointed to discrepancies between lists and texts: the many instances of variant personal names, the occurrence of tenants in the text but not in the lists and vice-versa, the changes in the grouping of the less important tenants and the mismatches in the numeration.49 He explained all this by arguing that the lists were based on the circuit returns; and that the rubricator who was responsible for the numeration then had to contend with the unforeseen results of edito-

47 Galbraith, The Making of Domesday Book, pp. 32,192-99, especially 193n.; idem, Domesday Book: Its Place in Administrative History, pp. 50-53. 48 Great Domesday Book, fos 56r, 75r, 238r. The Berkshire entry could simply reflect a misjudgement of the space required for the index. This is less adequate as an explanation of the other two. R.W. Eyton considered that the Dorset index was entered after the text, A Key to Domesday: An Analysis and Digest of the Dorset Survey (London, 1878), p. 74. All these instances were noted by Finn who nevertheless seems to accept Galbraith's conclusion, Finn, The Domesday Inquest, p. 169. 49 Galbraith, The Making of Domesday Book, pp. 191-92.

42

Colonial England, 1066-1215

rial amendments of the compiler or abbreviator.50 This argument is still accepted;51 if correct, it indicates that the reference system was already intended in the circuit returns as indeed the one survivor, Little Domesday Book, suggests. It matters more for the present argument that the numeration was applied systematically throughout Great Domesday by one man, the rubricator.52 He scarcely varied in his work.53 He made errors, certainly, but it is more significant that often when he did he tried to fudge, or indicate, or correct them.54 There is no real doubt that the numeration was intended for use. It can still be used. The rubrication and the use of capitals is equally plain. The rubricator's work included the county headings, the introductory notes to the lists, the heading of each tenant's entry, the numeration and the underlining of hundreds and manors and the decoration of initials. Capitals were used in the main text for hundreds, wapentakes and manors and by the rubricator for the heading of each numbered section. In all this it is easy enough to find variations and lapses; it is well known, for example, that there are no hundredal headings in the Domesday of the south-western circuit; but the purpose is clear: it was to direct our treasury official from county to landholder and within each unit of landholding to manors. That still leaps to the eye: it is a compliment to the directing genius that for this purpose, still after 900 years, the facsimile is easier to use than any subsequent edition, despite, or perhaps because of, all the typographical experiments they contain. Consider now what was not done. The rubricator, working under the direction of the compiler, could have highlighted a wide variety of matters. He could have underlined geld assessments: he did not. He could have underlined changes in geld assessment: he did not. He could have underlined values: he did not. He could have underlined changes in value: he did not. All this might have been more difficult than what he did, and for one good reason: the compiler did not have these matters highlighted either. Instead he directed his energies to providing capitals for manors, which greatly facilitated the task of the rubricator. Both could have taken other courses which would have emphasised all these fiscal and economic matters.55 Why not, for example, a simple marginal

50

Ibid., pp. 192-96. A.R. Rumble, The Palaeography of the Domesday Manuscripts', Domesday Reassessment, pp. 37-40. 52 For the sake of clarity I have distinguished here and elsewhere between the rubricator and the main scribe of Great Domesday. They could well have been one and the same person as Galbraith thought, The Making of Domesday Book, p. 193n. My argument would be reinforced if this were so. 53 Anon., Domesday Rebound (London, 1954), pp. 32-34. It should also be noted that the rubricator did not usually enter a number against the terra regis: there was no need. However, he did so for Huntingdonshire, 203, and Derbyshire, 272. Middlesex lacks the number II in the text and Somerset nos II-V. 54 Galbraith, The Making of Domesday Book, p. 194; Finn, The Domesday Inquest, p. 169. 55 Once again I distinguish between the rubricator and the main scribe. Once again the point is reinforced if they were one and the same. 51

1086

43

indication of changes in assessment or value?56 Why not a marker or rubric where more could be had? To all appearances there was red ink enough to spare. And why did the compiler not infrequently insert other information after the values which terminate the standard entry, so that the value does not always figure as the bottom line? To be sure all these matters can be excavated by a careful reading of the text, but a busy official had little time for that. He would only do it if for some special reason he had to. Then it would be a particular, perhaps a major, task. I shall turn shortly to the circumstances which might require it. Now there is plenty of evidence, quite apart from its general appearance and patent utility, to indicate that the reference system was integral to the purpose of the record. First the rubricator had difficulty in contending with exceptional information of a kind which was normally excluded from Great Domesday. At two points in the record he had to deal with lists of clamores57. On fo. 208, dealing with the Huntingdonshire clamores, he sinply gave the initials some red shading, largely coinciding with the paragraph markings. Then he gave up; there is no rubrication of the further items on fo. 208v. The same thing happened with the clamores of Yorkshire and Lincolnshire, extending over five folios, 373-7. He gave folio 373 the same treatment as folio 208, red shading of the initials, reinforced this time by underlining the headings. Then he stopped; there is no rubrication on any of the remaining folios of the clamores\ the so-called Yorkshire recapitulation which follows is

56 An exception which proves the rule is the marginal entry recording a change in the geld assessment of Pyrford in favour of Westminster Abbey, Great Domesday Book, fo. 32r. This is clearly related to and probably a consequence of a reduction in hidation authorised in the well-known writ of William I dated post descriptionem totiusAngliae. See Galbraith, The Making of Domesday Book, p. 206; Facsimiles of English Royal Writs to AD 1100, ed. T.A.M. Bishop and Pierre Chaplais (Oxford, 1957), no. 26. The writ, it should be noted, frees eight hides from all scot, custom and geld, not, as the editors suggest, the whole manor. The writ and the marginal entry are of critical importance. Taken together they establish that the Domesday Book scribe came to know of the writ subsequently to making the Pyrford entry in Great Domesday Book. Galbraith, The Making of Domesday Book, pp. 206-7, was uncharacteristically tentative in pursuing the obvious conclusion, namely that this passage in Great Domesday Book was written before William's death and, since it is likely that the writ was issued in England, before his departure for the Continent in 1086. The point is quite independent of the debate over the meaning of descriptio, in The Domesday Monachorum of Christ Church, Canterbury, ed. D.C. Douglas (1944), p. 24; Galbraith, The Making of Domesday Book, pp. 183-84. Such an early date for the completion of at least part of Great Domesday Book is inessential to, although coherent with, my argument. It is more important that, if my interpretation is right, Domesday Book must have been a powerful influence in establishing evidentiary as opposed to dispositive documentation in AngloNorman practice. 57 I assume here that the clamores in Great Domesday represent a residuum of a larger number of cases, a proportion of which had been resolved by the completion of the final record. The assumption rests on a comparison of the clamores with the text, chiefly for Huntingdonshire and Lincolnshire. See Finn, The Domesday Inquest, pp. 94-95; and below, p. 45 n. 62.

44

Colonial England, 1066-1215

a distinct section, written in yet another format and rubricated for an entirely different purpose.58 Secondly, the system of reference was applied to Little Domesday. Here the prefatory lists were based directly on the text and correspond exactly.59 Unlike Great Domesday, the headings of each tenancy had already been entered when the rubricator set to work; so there were some necessary variations in his procedure, and in general the work here was cruder and less complete. But there is the same numeration in red; the reader is guided by rubricated county headings and initials or abbreviations of the tenant at the head of the folios; the tenant-headings are underlined in the text; the hundreds are highlighted with a touch of red and occasionally with red underlining; the underlining does not extend to manors, which are emphasised simply by a capital. Mutatis mutandis, and making some allowance for a hastier job, the frame of reference corresponds closely to that of Great Domesday in both inclusions and exclusions. Its purpose was the same. Finally, from 1179 or thereabouts, less than a hundred years after Domesday, there is direct evidence from a user, no less a person than Richard fitz Neal, Henry IFs treasurer: The survey is made by counties, hundreds and hides. The king's name heads the list, followed by those of the nobles who hold of the King in chief, according to their order of dignity. The list is then numbered and the matter in the actual text of the book relating to each tenant is easily found by the corresponding number.60 That passage from Richard's Dialogus has often been quoted. It still bears repetitioni Now it should be axiomatic that no explanation of the purpose of Domesday Book is worth considering unless it is consonant with its framework and the reference procedure which fitz Neal described. And given the systematic and deliberate arrangement of the framework of reference, any explanation must not only match what is there but also accept what is not there; we cannot require 58

Great Domesday Book, fos 379r-82r. The main purpose of the rubrication here was to emphasise the interlineated names of landholders. 59 Galbraith, The Making of Domesday Book, pp. 192-93. Galbraith's argument that the lists preceded the text was criticised by Finn who suggested that the lists were done after the text, Domesday Studies: The Eastern Counties (London, 1967), pp. 65-67. Rumble, The Palaeography of the Domesday Manuscripts', p. 40, takes the same line. There are four short lists at Domesday Book, ii, 9a, I7a, 292a, 372a. On the whole the lists at 9a and I7a look like headings of gatherings or groups of gatherings. Those at 292a and 372a, in contrast, are added at the foot of preceding gatherings and are hence more consistent with Galbraith's view. However, the entry concerning St Stephen of Caen in Essex, la, is more likely to be derived from the text, I7a, rather than vice-versa. Similarly, in the Suffolk folios, the abbey of Ramsey is missing from the short list, 372a, but is included in the index and text: the short list on 372a cannot therefore have been used, as Galbraith suggests, as a guide to the order of the following entries in the text. 60 Richard fitz Nigel, Dialogus de scaccario, ed. CharlesJohnson (London, 1950); rev. F.E.L. Carter and Diana Greenway (1983), pp. 63-64.

1086

45

treasury officials of the Norman period to engage in the equivalent of Ph.Ds. In one respect fitz Neal is almost certainly misleading. 'A careful survey of the whole country was made', he tells us, 'in order that every man be content with his own rights and not encroach unpunished on those of others.'61 That imposes the common law of the twelfth century on the more primitive circumstances of the eleventh. Domesday Book was not constructed as the ultimate resolution of actions between party and party. The determining of such cases was essential to its construction, but they were listed separately as appendices to the main texts. In some counties an attempt was made to resolve such dispvites before the final book was drafted.62 The evidence of the Book is that it could fulfil three possible functions, and only three at all easily: 1

2

3

Within each county it provides a record of the customs of shire and boroughs and of the location, value, potential and assessment of the terra regis. It puts the treasury in a firm and informed position in imposing requirements on the sheriff and others who farmed and accounted for these resources. There is no need to dwell on this aspect of the Book's purpose. It gives a record, county by county, sometimes hundred by hundred, always manor by manor, of the holdings of the king's tenants. It provides a reference work for the settlement of conflicting claims. It enables the king's men to instruct the sheriff of each county to dispossess a tenant or give possession to an heir or claimant. It records, county by county, the location, extent, value and potential of particular tenements, for which the sheriff or some other nominated royal agent will be accountable if the lands come into the hands of the crown.

Hence the county arrangement of the Book refers us at once to the relationship between central and local government, between the treasury or exchequer and the sheriffs. In its major component, the lands of the king's tenants, it is concerned above all with tenure, with succession, and with royal rights of escheat and wardship. That is the plain evidence of the record. For these purposes there was no immediate need for totals of either geld assessments or values. The treasury did not need what the Book does not provide. Its concern lay 61

Ibid., p. 63. See above, p. 43 n. 57; and Finn, The Domesday Inquest, pp. 92-97. The correlation of the Domesday Book text with the lists of disputes depends in part on the point or period in the gathering of information at which the lists of disputes were compiled. A comparison of the terrae occupataem Liber Exoniensis with Little Domesday Book indicates that there were considerable differences in procedure in this matter between one circuit and another. See Finn, Liber Exoniensis, pp. 55-81; idem, The Eastern Counties, pp. 42-45. The point was reiterated by Galbraith, The Making of Domesday Book, pp. 21, 38, 60, 73-74, 83-85, 176-78. See also R.S. Hoyt, The Terrae Occupatae of Cornwall and the Exon Domesday', Trculitio, 9 (1953), pp. 155-99. 62

46

Colonial England, 1066^-1215

entirely with the county, with the sheriff, through him with the tenants, and hence with data arranged in that format. If there was a dispute over a particular manor, its tenancy or hidage, the Book would settle it at once. Instructions for the dispossession of even the greatest tenant with lands scattered over several counties could be drafted precisely, if necessary, in a few hours in writs to the sheriffs specifying his holdings (if they did not already know them from their own information). Similar action could be taken for the instatement of an heir. If land came into the king's hands on an ecclesiastical vacancy or after the death of a tenant or through more prolonged escheat or by wardship, locations and extent were equally easy to determine. It was only at this stage that it became essential to have further information: tax assessment, value and 'if more could be had'. This characteristic Domesday information was essential to, not separate from, its main feudal purpose, for without it wardships and escheats could not be assessed to farm. That is probably all that was intended in the first instance when the compiler laid down the format. Some have argued that Domesday helped in assessing reliefs or amercements, and it could certainly be used in that way; but other matters entered into these calculations, and while Domesday Book was a very precise instrument, they were somewhat blunt.63 To sum up: Domesday presents information at two different levels. As a quick work of reference it is concerned with the location and tenure of manors. At a deeper level it is concerned with resources, values, potential and geld assessment. At this level it was part of the continuing relationship of treasury and sheriff in the management of counties, boroughs and the royal demesne. For the rest, that is for the tenants-in-chief, this function could be brought into play as and when necessary for ecclesiastical vacancies and the operation of the feudal incidents of wardship and escheat. This now poses a political question. All the evidence about the procedure of the Survey indicates that it was astonishingly rapid and that it had the active support of William's tenants-in-chief. Why did they cooperate? To revert to the old 'geld-book' hypothesis for a moment, who would want to see his tax return turned into a rubricated manuscript?64 Or, on the argument outlined above, who would want to strengthen the king's control of succession, escheat and wardship? Yet it is plain that Walter Giffard and Henry de Ferrers, to name two of whom we can be certain, participated as commissioners in a survey which did just that. So why did they do it? What advantage did they see in it? For advantage there must have been. Blackstone and his predecessors are the only commentators who help in providing an answer. For the response to the question: 'For what did the landholding men of any account perform homage on 1 August 1086? must surely be: 'for the tenements recorded in the Domesday 63 In the case of amercements much depended on the seriousness of the offence. In the case of reliefs something depended on the proximity of the heir or the access of the successor to royal favour. 64 With acknowledgement and apologies for the necessary amendment to Jonathan Sale, '1086 and All That', Punch, 18 December 1985, p. 38.

1086

47

Survey',65 and that logic strengthens in its turn the argument that the oath was integral to the Survey, intended from the start at Christmas 1085, because baronial cooperation was available from the start. Hence Domesday Book seems to embody a hard-headed deal. William got a survey of his own and his tenants' resources; he was strengthened in the exercise of his feudal rights. His tenants got a record of their tenure, in effect a confirmation of their enfeoffment. In short, as regards the tenant-in-chief, Domesday Book was a vast land book which put a final seal on the Norman occupation. The Peterborough chronicler tells us that at the end of the Survey the writings were brought to the king.66 Some modern historians have suggested that the returns, even the completed Book, were presented to him at Salisbury on 1 August 1086 - rather like a prize day.67 That, indeed, is to impose a modern gloss on the words of the chronicler; the fact is that there was no one else to whom the returns might go and that is probably all he meant. But if we are to think of presentation, it is just as likely to have been in the other direction, from William to his men, for if this reconstruction is approximately right, William's homagers knew that their tenements had been recorded with the king's agreement, that the Survey was complete, that the breves and circuit returns were already to hand and that the compilation of a final record, the future Domesday Book, was going forward. So William was both receiving and presenting the consequences of the great survey in which they had all participated. That this above all was the central purpose of the Book (leaving aside of course the county and borough customs and the terra regis) is the plain message of the later evidence. It was used to determine tenure throughout the next two centuries.68 Ownership, location, amount of land (i.e. hidage), were still the main concern of the Herefordshire Domesday of 1160-70 as of all the later abbreviations.69 It was as the 'charter of the king', or 'my charter of Winchester', or 'the charter of my treasury' that the record came to be described as soon as the memory of the breveshad faded. These terms come from the royal chancery itself.70 But best of all, perhaps, they come from the earliest collateral evidence of Hemming's cartulary of 1100 or not much later: 'A descriptio of the land of the bishopric of the church of Worcester according to the carta regis which is in the king's treasury' - thus his title. And even more explicitly: the Domesday commissioners had 'this testimony written in the original (autentycd) cartulaof the king'; and again, 'In confirmation of this matter the examplar is written in

65

The point is put most succinctly by Wright, Introduction to the Law of Tenures, pp. 56-57. Peterborough Chronicle, p. 9; EHD, ii, p. 161. 67 Finn, The Domesday Inquest, p. 190; Harvey, 'Domesday Book and its Predecessors', p. 755; Clarke, The Domesday Satellites', p. 56. 68 Hallam, Domesday Book through Nine Centunes, pp. 32-51; Galbraith, Domesday Book: Its Place in Administrative History, pp. 100-22. 69 Herefordshire Domesday, ed. V.H. Galbraith and James Tait, Pipe Roll Society, new series, 25 (1950), pp. xxiv-xxxii; Hallam Domesday Book through Nine Centuries, pp. 32-51. 70 Herefordshire Domesday, pp. xxvi-xxvii. 66

48

Colonial England, 1066-1215

the original (autentica) cartula of the King which is kept in the royal treasury with the descriptiones of all England'.71 What precise advantage did the tenants-in-chief see in such a record? Hemming provides the clue to one such group which had a very special interest. Since the Conquest the church had suffered depredation at the hands of the new French nobility- and others, for churchmen preyed on churchmen. There had been some notable legal actions seeking the restoration of the lands which had been lost.72 Less obviously there was a running battle between ecclesiastical lords and laymen seeking and acquiring fiefs which bishops and abbots were forced to concede to meet the new service-quotas which the Conqueror imposed.73 So great churchmen had a very direct interest in the compilation of an exact and final record. They used the opportunity not only to record their lands but also to advance claims to the lands which they had lost over the last twenty years.74 So these landowners, a very important group to whom we owe practically all the surviving subsidiary documents of the Survey, had a very close interest in ensuring its success. This necessarily involved the laymen, for a unilateral establishment of title by the church must have been unthinkable. Hence the bishops and abbots carried the rest. Nevertheless, there were good grounds for the rest to join in. Domesday provided 'warranty' for all tenancies-in-chief- 'warranty' in the simple eleventhcentury sense which it enjoyed in Domesday, namely authority for tenure. The lay tenants needed such warranty. Moreover William needed to give them warranty, not just to achieve stability, to put a seal on the settlement, but, as I shall argue shortly, to get from them what he wanted. What warranty did they have already? It is assumed inaccurately in our text books that the Norman settlers were relatively secure in their tenures: that they had performed homage to the king for precisely defined fiefs in return for agreed military service, the arrange71

Hemming's Cartukiry,i, pp. 298,288. See alsoJ.H. Round in Domesday Studies, i, p. 546, criticised by Galbraith in Herefordshire Domesday, p. xxviii n. 1. On the dating of Hemming's cartulary see N.R. Ker, 'Hemming's Cartulary', in Studies in Medieval History Presented to EM. Powicke, ed. R.W. Hunt, WA. Pantin and R.W. Southern (Oxford, 1948), pp. 49-75; and V.H. Galbraith, 'Notes on the Career of Samson, Bishop of Worcester', EHR, 82 (1967), pp. 97-101. 72 See the varied information assembled in Finn, The Domesday Inquest, pp. 92-111, and Liber Exoniensis, pp. 69-76; also P.M. Stenton, 'St Benet of Holme and the Norman Conquest', EHR, 37 (1922), pp. 225-35; F.R.H. Du Boulay, The Lordship of Canterbury (London, 1966), pp. 36-43; Barbara Harvey, Westminster Abbey and its Estates in the Middle Ages (Oxford, 1977), pp. 71-74. For the losses of the abbey of Ely see Inquisitio comitatus Cantrabrigiensis, pp. 184-89, discussed by Finn, Eastern Counties, pp. 87-94; and by E. Miller, The Abbey and Bishopric of Ely (Cambridge, 1951), pp. 6667. For a cautionary note see David Knowles, The Monastic Older in England (Cambridge, 1963), pp. 117-18. 73 V.H. Galbraith, 'An Episcopal Land Grant of 1085', EHR, 44 (1929), pp. 353-72; Edmund King, Peterborough Abbey, 1086-1310 (Cambridge, 1973), pp. 18-23; Miller, The Abbey and Bishopric of Ely, pp. 67-70. 74 See especially the case of Ely, above, n. 72. For other examples see Finn, The Domesday Inquest, 101-9; Barbara Dodwell, The Honour of the Bishop of Thetford/Norwich in the Late Eleventh and Early Twelfth Centuries', Norfolk Archaeology, 33 (1963), pp. 185-88; Harvey, 'Recent Domesday Studies', p. 124.

1086

49

ments all concluded at one and the same time. This is demonstrably false; to choose only one example at random, Henry de Ferrers received his Domesday holdings in at least five tranches over a considerable period of time.75 The evidence of the Book itself is somewhat different from the assumptions of the text book. In most cases, where it tells us anything at all, it is simply that the land was held de rege. Sometimes it reveals that Norman tenants were put in seisin by instruction of the king, by writ, or seal (which was the same thing), by his agent or liberator, sometimes explicitly by the sheriff, or by his own word of mouth.76 Occasionally it refers to English antecessores\ in other cases, where they are not mentioned specifically, some Norman tenancies can be correlated with the holdings of English predecessors. However, much of this evidence is the product of contested possession, and it is by no means clear how far these precise procedures were themselves a consequence of dispute. The main argument that there was some formal process of enfeoffment, involving recognition of seisin, is that the settlement itself was relatively orderly. However, there is no evidence at all that William's vassals were enfeoffed by charter: no such charter survives except as a blatant forgery.77 Strictly speaking the evidence does not allow us to assume that a grant of land was accompanied by the performance of homage. Most of the companions of the Conqueror must already have been his homagers. Why bother to renew the bond when a vassal received his expected reward in the new English acquisitions? Finally, it does not allow us to assume very much, if anything at all, about terms of tenure. There is no direct evidence about title or succession. Even so, three reasonable inferences may be drawn from scattered and fragmentary evidence. First, it seems inescapable that by 1086 men had come to think that homage was part of a reciprocal act, that it was not performed in vacuo or for the promise of good lordship but in return for something material and real, actual or expected, usually in return for recognised and lawful 75

Henry de Ferrers acquired the following: probably in 1066-67, lands in Berkshire previously held by Godric the sheriff who fell in battle, probably at Hastings. b) certainly after 1068 and probably before 1070, lands in Buckinghamshire, Berkshire, Oxfordshire and Essex previously held by Bondi the staller. c) c. 1070, lands in Derbyshire, Nottinghamshire, Warwickshire, Berkshire, Essex, Gloucestershire and Lincolnshire previously held by Siward Barn. d) c. 1071, lands in Appletree wapentake, Derbyshire, previously held by twenty-four TRE landholders and contiguous lands in Staffordshire which together came to form the core of the honour of Tutbury, in which Henry was possibly preceded by Hugh of Avranches. e) other acquisitions in Derbyshire of uncertain date and apparently of miscellaneous origin. I am indebted here to RE. Golob, The Ferrers Earls of Derby: A Study of the Honour of Tutbury, 1066-1279' (unpublished Ph.D. thesis, University of Cambridge, 1985). Formore general discussion see John Le Patourel, The Norman Colonisation of Britain', IN&nnanni e la loro espansione in Europa neWalto medioevo, Settimani di studi del centro italiano di studi suiralto medioevo, 16 (1969), pp. 409-38; below, pp. 81-101. 76 Finn, The Domesday Inquest, pp. 100-9. 77 EYC, iv pp. 94-95, and frontispiece. a)

50

Colonial England, 1066-1215

seisin of land. Whatever the personal nature of the act of homage at an earlier date, men's minds must now have been conditioned by the recurrent performance of homage by heirs as they claimed the succession to estates and by bishops as they acquired the temporalities of their sees. Much more often than not, homage and tenure went hand in hand: that was the assumption of both William of Poitiers and Eadmer.78 Secondly, Domesday and other evidence indicates that exchanges of land were by no means unusual. Many were effected simply for the mutual convenience of the new Norman tenants.79 But one well-known case, the so-called exchange of Lewes, which was the most extensive transaction of this kind, demonstrates that William the Conqueror was able to dispossess one newly-endowed tenant, William de Warenne, in order to make way for another, William de Braose; and the only indication that Warenne had any kind of claim in what he lost in Sussex is that he got lands in Norfolk which were recorded as an exchange or as the exchange of Lewes.80 That is extremely frail evidence of any kind of right or title and, whether so or not, Warenne was closer to the king than the runof-the-mill tenant. The main conclusion to be drawn from this evidence is that the Conqueror could shift his original tenants around if convenience so required. Thirdly, except for the inherited possessions of the French already established in England under Edward the Confessor and for those held by the residue of the English, all the new tenancies were conquests or acquisitions. By definition there could be no confirmation of a succession until the succession occurred. Certainly such had happened by 1086. It is reasonable to assume for example that Roger de Lacy, who succeeded his father, Walter, less than a year before the Survey,81 sought the succession, performed homage and was instated in all the lands of his father, perhaps by writs directed to the appropriate sheriffs. His lands were now his patrimony, with such tide as that implied. But only a proportion of King William's tenants were in that enviable position. Just to take a few of de Lacy's immediate or not too distant neighbours, Chester, Montgomery and Ferrers were all at the stage of acquisition, warranted solely by the initial grant from the king, whatever its nature, and by the passage of years. Amidst these more general uncertainties there was also a more precise dif78 See William of Poitiers' account of Harold's alleged submission to William in which homage is distinguished from fealty and associated with William's promise to confirm his English possessions, an account all the more telling because of its quasi-fictional nature, Gesta Guillelmi ducisNormannorum et regis Anglorum, ed. Raymonde Foreville, Classiques de 1'histoire de France au moyen age (Paris, 1952), p. 104. He also links restitution of lands with the obsequiumand sacramentaofthe English lords in 1066 (ibid., p. 236). For Eadmer's association of homage with episcopal temporalities see the account of Anselm's election in which Eadmer also refers to Lanfranc, Historia novorum, ed. M. Rule, Rolls Series (London, 1884), p. 41. I am grateful to Dr Marjorie Chibnall for raising these matters with me. 79 Finn, The Domesday Inquest, pp. 20-21; idem, The Eastern Counties, pp. 31-32; Liber Exoniensis, pp. 71-72. For sales see Ellis, General Introduction to Domesday Book, i, p. 43 nn.; Freeman, Norman Conquest, v, pp. 778-85. 80 J.F.A. Mason, William the First and the Sussex Rapes (London, 1972), pp. 15-16. 81 W.E.Wightman, The Lacy Family in England and Normandy, 1066-1194 (Oxford, 1966), p. 168.

1086

51

ficulty. By 1086 a number of the original tenants-in-chief had withdrawn from England; two, Roger, earl of Hereford, and Odo of Bayeux had been dispossessed and the lands of one other, Roger of Poitou, were in the king's hands in whole or in part. How did their tenants stand? What and whose warranty did they enjoy? Are we to assume that on siding with the king against their lords they performed homage and were put into all their holdings just as an heir or someone newly enfeoffed? Dr Wightman assumes so in the case of Walter de Lacy, tenant of Roger, earl of Hereford,82 but there is no direct evidence that it happened. Indeed in some ways Domesday suggests the opposite. The extent and structure of fitz Osbern's earldom is now controversial. It is no longer possible to accept that the 1086 tenants-in-chief in Herefordshire and Shropshire had all, with certain exceptions, been his tenants.83 Nevertheless, some of the tenements of the earldom are still apparent in the Domesday record of Herefordshire and most obviously of Oxfordshire where they are listed separately and numbered both in text and index as the lands of Earl William.84 In both counties the tenants of the old earldom now held in chief either explicitly or in effect. Yet in Herefordshire Earl William still figures as warrantor. Sometimes the Survey simply records that he built a castle - Clifford or Wigmore. Sometimes it is specifically stated that Earl William established a holding for a tenant, that he had given four carrucates of waste land in the castlery of Ewyas to Walter de Lacy, for example.85 The king certainly intruded his own men into the county after the fall of Earl Roger,86 but there is singularly little evidence that he set about confirming the tenancies of the earl's tenants who remained. The record mentions the grant by King William of one and a half hides in Wolferflow which completed Walter de Lacy's holding there,87 a grant by the king to Roger de Pistes,88 a quittance of geld to King Maredudd,89 and the agreement of the king to grants by Walter de Lacy of land in Acle and Upleadon to the church of Hereford,90 but only once, in the case of Alfred of Marlborough's tenure of Ewyas castle, does it state specifically that King William confirmed a grant of fitz Osbern.91 It may well be that in particular terms the Domesday Survey itself provided the first confirmation which many of these tenants enjoyed. The same

82

Ibid., pp. 123-24. Christopher Lewis, The Norman Settlement of Herefordshire under William I', AngloNorman Studies, 7 (1984), pp. 195-213. Cp. the long-standing and still arguable view fully developed in Wightman, The Lacy Family, pp. 117-34; and idem, The Palatine Earldom of William fitz Osbern in Gloucestershire and Worcestershire, 1066-71', EHR, 77 (1962), pp. 6-17. 84 DB, i, 161a. 85 Ibid., i, 184a. 86 Lewis, The Norman Settlement of Herefordshire', p. 209. 87 DB, i, 185a. 88 Ibid., i, 186b. 89 Ibid., i, 187b. 90 Ibid., i, 184a, 184b. 91 Ibid., i, 186a. 83

52

Colonial England, 1066-1215

is even more obviously true of the Oxfordshire list which is scarcely explicable on other grounds. The fee of Odo of Bayeux, who was arrested and dispossessed in 1082, presents matters at an earlier stage of development. Odo's fee still survived as a separate unit, headed and numbered in Little Domesday.92 The effect of the record is to confirm the position of his tenants, especially of the greatest of them, Roger Bigod. In Kent the units of undertenancy do not yet appear as tenancies-in-chief, which many of them subsequently became, but the undertenants are consistently named in such a manner as to break up the regular order of lathes characteristic of the Kentish Domesday.93 Here the Domesday record could well have been the first royal warranty of Odo's enfeoffments. The lands of Roger of Poitou reflect a still earlier stage in this process. Here, between Ribble and Mersey, the present tenants are simply listed as enjoying their respective portions by gift of Roger.94 In all probability Domesday itself was the first royal confirmation of their tenancies. But in this case there was no consistency. The Yorkshire entry for Roger was merely an extract from a geld roll; no tenants were mentioned.95 In Derbyshire it was simply noted that the lands were in the king's hands; in Norfolk that the lands had been Roger's.96 In other counties he appears as a sitting tenant. It is probable that he was dispossessed in the course of the Survey.97 One further complication is perhaps worth nothing. The Chester Domesday does not follow the rules. Here the lands of Earl Hugh are listed with first the demesne holdings and then the holdings of each undertenant, none of whom appear in the prefatory list of tenants but each one of whom is emphasised by capitals, rubricated lineation and an emphatic space before each tenancy.98 It is always recorded that these men hold of Earl Hugh, but the compiler is plainly trying to tell us that they are different, and that must be because, apart from the diocesan lands, Earl Hugh held the county of the king.99 The presentation is not accidental. It is repeated only in Shropshire.100 There too Earl Roger held the county of the king, apart from the lands of the church and certain lay tenancies.101 In both these cases the treatment of undertenancies was exact and intrinsic tp the record. It contrasts sharply with the more haphazard treatment of undertenancies elsewhere.102 92

Ibid., ii, 18a-26a, 142a-143b, 373a-378b. Ibid., i, 6a-llb; Harvey 'Domesday Book and its Predecessors', p. 757. 94 DB, i, 269b, 270a. 95 Great Domesday Book, fo. 332r. The entry is not numbered either in the list or text. It is in the hand of the main scribe and rubricator, but is not rubricated in the usual detail. The heading 'Lands of Roger of Poitou' does not necessarily prove that Roger was still in possession. 96 DB, i, 273b, ii, 243a-244b. 97 Galbraith ' The Making of Domesday Book, pp. 187-88. 98 Great Domesday Book, fos 264r-70r. 99 Totam reliquam terram comitatus tenet Hugo comes de rege', GDB, fo. 262v. 100 Great Domesday Book, fos 253r-259v. 101 'Comes Rogerius quod reliquum est tenet cum suis hominibus', GDB, fo. 252r. 102 The features noted above are not readily apparent in the printed text, especially of Cheshire. 93

1086

53

How then might a lay tenant-in-chief gain advantage from the Domesday Survey? To summarise: 1 2 3 4 5

It provided a single coherent record of tenancies accumulated piecemeal over a period of up to twenty years. It provided a written record which warranted possession which in many cases had not advanced beyond the stage and status of acquisition. It recorded purchases and exchanges, many of which had presumably been made without royal confirmation. It warranted a large number of tenancies originally established by Norman lords who had withdrawn or been dispossessed. It recorded the tenancies held of the two remaining comital earldoms of Chester and Shrewsbury.

All this is quite apart from the fact that it resolved disputes from which a lay tenant might either gain or lose: some of them certainly may not have been all agog about that. Finally, Domesday was a general record. Most important of all, it was a royal record, a great evidentiary certification held in the king's treasury. It was novel. It was unique. That probably gave it all the more prestige in the eyes of men who were familiar with writs and Norman pancartes but who knew not the charter of enfeoffment. All that was worth an homage. All that was worth combining with a meticulous survey of resources conducted in the interests of the king. That was the agreement of Christmas 1085 which came to fruition at Salisbury in 1086. But if that is why king and barons did it, why did they do it precisely then? There is in fact a ready explanation. First, it has long been recognised that in 1085 William had to assemble an army of mercenaries as a defence against the combined threat of King Cnut of Denmark and Count Robert of Flanders. As Dr Harvey has emphasised that provided an impetus for the inquisitio geldi and for the survey of resources embodied in Domesday Book.103 Others have linked this crisis with the Salisbury oath.104 This seems less convincing and less again if the Survey and the oath were connected in the manner suggested above. But our doubts here, reasonable though they may seem, depend on our understanding of the oath and on William's intentions in the face of external threat. What in fact he did in the last year of his life was cross to France and deliberately attack his liege lord, King Philip of France, invading the Vexin and receiving his fatal injury as he put Mantes Gassicourt to fire and sword.105 Now it is only continued

They deserve closer attention than they have hitherto received in the discussion of the 'palatine' status of both counties. 103 Harvey, Taxation and Ploughland in Domesday Book', p. 103; idem, 'Domesday Book and Anglo-Saxon Governance', p. 181-82. 104 Stenton, The First Century of English Feudalism, pp. 113-14; H.W.C. Davis, England under the Normans and Angevins, pp. 36-37; Douglas, William the Conqueim, pp. 355-56. 106 Ibid., pp. 356-58.

54

Colonial England, 1066-1215

our insularity that blinds us to the probability that the oath sworn by William's men at Salisbury, which bound them to him 'against all other men', was most likely aimed against the king of France. Far from deterring rebellion in England it underwrote rebellion in France. It was intended not to sustain liege lordship in England but to shatter liege lordship in France. William in short was bringing all the moral and legal weight of English tenure to bear to support an intended offensive against King Philip. He was reminding his Norman followers that they now had extended resources to support their ancient duty to the dukes of Normandy. He was emphasising his requirements of those who, like the Beaumonts, held their lands immediately both of him and King Philip. He was imposing an entirely novel obligation on those of his tenants in England who, like the Breton, Count Alan, lord of Richmond, or Eustace, count of Boulogne, were not of Norman origin; in so doing he changed the balance of power in France. That interpretation may seem to involve a lot of guesswork. It certainly requires us to assume that William's campaign in France in 1087 was already in his mind at Christmas 1085. That is not impossible, for the necessary springboard had come into Norman hands in 1080-81 when the county of Meulan passed by inheritance to Robert de Beaumont.106 But it is supported by one crucial fact and one critical piece of evidence. The fact is that the English baronage somehow came to accept that they owed service for their English fees in Normandy. That was rarely questioned: it was admitted even in 1215.107 It was an unusual arrangement of which England certainly provides the most striking example.108 It demands explanation. The evidence lies in what may well be another account of the Salisbury oath, which was recognised as such by Blacks tone.109 This is cap. 2 of the so-called Ten Articles of William the Conqueror: We lay down that all free men shall affirm by fealty and oath that they are willing to be faithful to King William and preserve his lands and honour in all fidelity and defend him against his enemies both within and without England.110

That text has received some severe criticism.111 It certainly cannot be taken, as Blackstone was tempted however cautiously to take it, as a literal statement of 106

Ibid., p. 357; Complete Peerage, vii, p. 524; Orderic Vitalis, ed. Chibnall, v, p. 214. J.C. Holt, The Northerners (Oxford, 1961), pp. 88-92; idem, Magna Carta (Cambridge, 1965), p. 151. See also John Le Patourel, The Norman Empire (Oxford, 1976), pp. 201-6. 108 For continental parallels see Holt, Magna Carta, pp. 64-66. The service which German vassals of the emperor provided on occasion in Italy was, of course, of more ancient, Carolingian origin and hence antedated the tenurial arrangements characteristic of the eleventh and twelfth centuries. 109 Blackstone, Commentaries, ii, pp. 49-50. 110 'Statuimus etiam ut omnis liber homo foedere et sacramento affirmet, quod infra et extra Angliam Willelmo regi fideles esse volunt, terras et honorem illius omni fidelitate cum eo servare et ante eum contra inimicos defendere', Liebermann, Gesetze, i, p. 486; Stubbs, Select Charters, p. 98; EHD, ii, p. 399. 111 H.G. Richardson and G.O. Sayles, Law and Legislationfrom Aethelbcrt to Magna Carta (Edinburgh, 1966), pp. 46-47, 101. See also Stubbs, Constitutional History, i, p. 289n. 107

1086

55

the Salisbury oath. But it can scarcely be later than 1135;112 it has no known source; and in referring to service outside the realm it provides the first statement of an authentic peculiarity of English feudal service. If we accept that immediately or distantly it gives us the precise intent embodied in the Peterborough chronicler's phrase 'all other men' against whom the Salisbury oath was directed, much falls into place. If that was the package agreed in 1085-86, it is an intriguing one. It indicates hard bargaining between William and his men, that the 'deep speech' at Gloucester embodied a deal, but a deal firmly set in the proper context of tenure and tide such as they were in the late eleventh century. For those who find it hard to swallow there is some sugar to coat the pill. The view that Domesday was intended as a great national review of economic resources has always run into a major difficulty: nothing much happened. Explanations can be found: William and Lanfranc soon died; the data was soon out of date, and so on.113 Or perhaps something did happen and Ranulf Flambard is used as the villain of the peace,114 reassessing dues and measuring the lands of England, in Orderic's phrase, 'with a rope'.115 It is all somewhat unconvincing. If on the other hand Domesday is viewed as it is presented above little of this arises. The adjustment of farms of counties and terra regis could go on quietly in the inner sanctum of treasury or exchequer; resources could be totalled and farms set as and when other lands came into the king's hands. For the rest the Survey had already achieved its objective. Its end was largely in its beginning. It was not just a descriptio but a carta. As such it was not intended to 'do' anything. It simply 'was'. Note Since this paper was written there have been numerous studies of the Survey and the Book. Among these, Professor P. Hyams, 'No Register of Title: The Domesday Inquest and Land Adjudication', Anglo-Norman Studies, 9 (1986), pp. 127-41, by coincidence, is in broad agreement with my approach. For other examples, see A.F. Frearson, 'Domesday Book: The Evidence Reviewed', History, 71 (1986), pp. 375-92; Henry Loyn,' A General Introduction to Domesday Book', in R.W.H. Erskine and Ann Williams, eds, Domesday Book: Studies (London, 1987), pp. 1-21 ;A.R. Bradbury, 'Domesday Book: A Re-Interpretation\EHR, 105 (1990), pp. 284309; NJ. Higham, The Domesday Survey: Context and Purpose', History, 78 (1993), pp. 7-19; and David Roffe, The Making of Domesday Book Reconsidered', Haskins Society Journal 6 (1994), pp. 153-66. None of these affect my argument. 112 Liebermann, Gesetze, i, p. 486; iii, pp. 277-80. See also LJ. Downer, LegesHmriciPrimi (Oxford, 1972), p. 6. 113 Richardson and Sayles, The Governance of Medieval England, p. 28; Galbraith, The Making of Domesday Book, p. 19: The king's undertaking died with him'. Cp. ibid., pp. 29, 180, 202, 205. 114 Harvey, 'Domesday Book and Anglo-Norman Governance', pp. 188-89,190-93; R.W. Southern, Medieval Humanism and Other Studies (Oxford, 1970), pp. 190-91. 115 Orderic Vitalis, ed. Chibnall, iv, p. 172.

56

Colonial England, 1066-1215

Nor is it disturbed by the number of disputes and disputed titles recorded in the Survey, which have now been discussed by Patrick Wormald in an important paper, 'Domesday Lawsuits: A Provisional List and Preliminary Comment', in Carola Hicks, ed., England in theEleventh Century: Harlaxton Medieval Studies, 2 (Stamford, 1992), pp. 61102.1 would expect the Survey to have uncovered many more disputes, some still leaving traces, resolved and unresolved, in the record, than Mr Wormald's list contains. Disputes were integral to the Survey. They demonstrate rather than dissolve title. Three problems require special comment: 1

More is now known about the scribes of Domesday Book. It now seems certain that the main scribe and the rubricator were one and the same person. See M. Gullick and Caroline Thorn, The Scribes of Great Domesday Book: A Preliminary Account', Journal of the Society of Archivists, 8 (1986), pp. 78-80; A.R. Rumble, The Palaeography of the Domesday Manuscripts', in P.H. Sawyer, ed., Domesday Reassessment, p. 45; and M. Gullick, The Great and Little Domesday Manuscripts', in Erskine and Williams, eds, Domesday Book Studies, pp. 93-112, especially pp. 98-102. These conclusions strengthen the arguments set out above pp. 41-44.

2

I have recast a paragraph concerned with the rubrication of Great Domesday Book, in which I slipped into accepting that the present order of the book and the order of rubrication were the same (above p. 43). In fact the present order probably coincides with the order of original binding, but not with the order of composition. See Anon., Domesday Book Rebound (London, 1954), especially p. 22, for the binding. On the order of writing, see Galbraith, Making of Domesday, pp. 201-2; A.R. Rumble, 'The Palaeography of the Domesday Manuscripts', pp. 3536; and Michael Gullick. 'Great and Little Domesday Manuscripts', p. 101. All these rely on the number of lines per folio as their main guide. See also now the important paper of David Roffe, 'Domesday Book and Northern Society', EHR, 105 (1990), pp. 310-36. Dr Roffe turns to a diplomatic analysis and argues persuasively that the Yorkshire folios were written first. Without committing myself wholly to this view, I have accepted that the Huntingdonshire folios were not necessarily compiled before the Yorkshire folios, thereby amending my assumptions about the order of composition of the clamores. It should be noted, however, that the order of composition and the order or rubrication may not have been the same.

3

Further evidence has now come to light on the date of completion of the Book. Dr C.P. Lewis has demonstrated that the Sussex folios and, less certainly, the Huntingdonshire folios were written after the accession of William II and the grant of the earldom of Surrey to William de Warenne; see The Earldom of Surrey and the Date of Domesday Book', Historical Research, 63 (1990), pp. 32936. Further evidence that work on, or arising from, the Survey continued into 1088 and later has been assembled by Professor David Bates, Two Ramsey Abbey Writs and the Domesday Survey', Historical Research, 63 (1990), pp. 337-39; and Dr Pierre Chaplais has presented a powerful argument that work on the Book continued until the exile of William of Saint-Calais, bishop of Durham, in 1088, 'William of Saint-Calais and the Domesday Survey', in Holt, ed., Domesday Studies, pp. 65-77. Conversely, other evidence indicates that information which arose after 1086 might be entered by the main scribe in demonstrably later additions

1086

57

or marginalia (Gullick, 'Great and Little Domesday Manuscripts', p. 106), and certainly one interlineation, that Geoffrey the Chamberlain was chamberlain to the king's daughter (fo. 49b), must have been entered before the death of William the Conqueror. So there is no doubt that the writing of Domesday Book began in the life of the Conqueror and ended some time after his death, say in 1088. This brings us back to Dr Roffe's proposed order of writing, which puts Yorkshire first. The obvious objection is that no sane scribe would begin his mammoth task with Yorkshire, unless perhaps he were a Yorkshireman (whatever that might mean in 1086), or unless he were working under the direction of someone with a strong interest in the county, like the bishop of Durham: the 'controlling mind' proposed by Dr Chaplais.

This page intentionally left blank

4 Domesday Book and Beyond

Why still read it? Why should scholars consult it and undergraduates study it? The plain answer is that still after ninety years it remains the greatest single book on English medieval history. So it is a monument, an index of ultimate achievement, to be admired and scrutinised as one might Gibbon or Macaulay. Yet it is more than that. Domesday Book and Beyond was conceived and executed at a time of dawning opportunity when Maitland, J.H. Round and a few others were beginning to do history as it ought to be done: analytically, scientifically, with questions asked and answered. The methods Maitland followed in this book showed the way. Many, though not all, of his conclusions remain valid. Some extensive sections of the book are still the best that we have on important aspects of Norman and English government and society. No other book of such compass has so endured. The compass is immense. Maitland decided to use Domesday Book as a kind of information centre from which he explored and reconstructed the society, the law, the government, the economy, even the vocabulary and something of the imagination of early medieval England. The first essay tells us what England was like in 1066 and what had happened by 1086. It is insular; there is little of Normandy or of Norman origins. The second asks how the circumstances of 1066 had been attained; it stretches back through the Anglo-Saxon law codes and land books to the English settlement, its social structure and administrative geography. The third attacks the 'dreary old question' of the hide which is seen as 'pre-judicial' to all the great questions of early English history' (357). It focuses the arguments and conclusions of the first two essays on methods of assessment and measurement and their relationship to wealth and resources, carrying back once again the Domesday evidence for comparison with the earlier records of the burghal and tribal hidage. Less obviously attractive than the first two essays, it hides behind its arithmetic numerous insights into the coherence and systematisation of early English government. Taken together, the three linked essays are vast in range and conception. Maitland alone had the nerve, the imagination, the flair and the learning to attempt such a scheme. This is the combination which makes the book unique. And he brings it off. Naturally it shares many of its qualities with Maitland's other work.1 There is 1 For a contemporary comment on Maitland's qualities see R.L. Poole's scheme for combining him and Round in one of the volumes of the Political History of England. This, he wrote, needed 'the

60

Colonial England, 1066-1215

the same style and gentle but incisive irony, heightened here perhaps because he was less constrained by the limitations of a lecture or essay or the encyclopaedic requirements of the History of English Law. He gave himself more freedom, to rich effect. Hence, on wergelds- The sons of a villanuswho had but two oxen must have been under some temptation to wish that their father would get himself killed by a solvent thegn' (44); or, in criticising the idea of common ownership: 'Who held this manor in the past? Nine sokemen held it. Rather a large party of joint lords we say: but still families will grow* (138); or in dismissing the facile equation of the vill with the Roman villa: 'And so England is full of vills which are Roman and satraps who, no doubt, are Persian' (337). The lawyer would add, very reasonably, that the whole corpus of Maitiand's work is stamped by a legal mind and training.2 Here, to be sure, the lawyer in Maitland is often in the foreground: in the examination of 'private' jurisdiction (52, 269-92); in the beautifully subtle discussion of freedom (42ff) and very obviously in the opening sections on the borough - 'What is it that makes a borough to be a borough? That is the problem that we desire to solve. It is a legal problem' (173). But, on the whole, the lawyer is less dominant than in Maitiand's other work. Indeed here and there he is wary of the law - '[manerium] loses that meaning [i.e. a technical meaning connected with the geld] in the course of time because the danegeld gives way before newer forms of taxation. It never again acquires a technical meaning until the late days when retrospective lawyers find the essence of a manor in its court' (128). Maitland was also a mathematician.3 It is this which gives Domesday Book and Beyond its unique quality among his works. He was in the forefront not as a mathematical thinker but in the use to which he put his mathematics. He did more than engage in extended simple arithmetic, like Round: he used sampling techniques, and this long before they were in general use for social analysis. Whether this came to him by instinct, as McDonald and Snooks have suggested, or through the advisers available in the Cambridge of the 1890s, is far from clear.4 The method makes its first appearance early in the book - 'We take continued

criticism of Round joined to the constructive gift which he has not', C.H.S. Fifoot, ed., The Letters of Frederic William Maitland, Selden Society, supplementary series, 1 (London, 1965), p. 230n.; the possible outcome defies the imagination. For more general and recent comment see G.R. Elton, EW. Maitland (London, 1985). The standard biography is HAL. Fisher, Frederic William Maitland (Cambridge, 1910). See also H.E. Bell, Maitland (London, 1965). 2 See S.F.C. Milsom, reviewing Elton, F.W. Maitland, Times Literary Supplement, 28 February 1986. 3 Maitland read for the Mathematical Tripos during his first year at Cambridge (1869-70) before changing, under Henry Sidgwick's influence, to the Moral and Mental Science Tripos (Fisher, Maitland, pp. 6-10). 4 One possible source may have been William Bateson. Whether Maitland enjoyed any special contact with Bateson through his daughter Mary as early as 1894-97 is uncertain. In any case Maitland encountered Bateson in general university affairs on which they were not wholly in agreement (Fifoot, Letters of Frederic William Maitland, no. 191 and passim). Bateson was certainly using sampling methods, closely similar to Maitiand's, in biometrical studies in the summer of 1892, W. Bateson, Materials for the Study of Variation (London, 1894), pp. 40-41.

Domesday Book and Beyond

61

100 entries (four batches of 25 apiece) and see that the number of villani and bordarii has risen from 1486 to 1894, while the number of servi has fallen from 423 to 303. We make another experiment with a hundred entries. This gives the following result. . .' (35). Thereafter the language of experiment and the tactics of sampling, averaging and approximating recur throughout the work, culminating in the interplay of numbers in Essay III which led Maitland to a tentative argument quite different from Round's: 4No one can look along these lines of figures without fancying that some force, conscious or unconscious, has made for "One pound, one hide"' (465), a suggestion reinforced with characteristic humour - 'We may, if we like such excursions, fancy the conservatives arguing for the good old rule "One teamland, one hide", while a party of financial reformers has raised the cry "One pound, one hide". Then "pressure was brought to bear in influential quarters", and in favour of their own districts the witan in the moots jobbed and jerrymandered and rolled the friendly log, for all the world as if they had been mere modern politicians' (471). Hence beneficial hidation of shires as well as manors. At this point Maitland suggested that he was less than serious. In fact the work is shot through with similar comments which reveal a humorous understanding of the human condition. On gelding in the manor: 'For one reason the king can not easily tax the rich; for another he can not easily tax the poor; so he gets at the poor through the rich' (121-2). On the privileges of the church of Worcester: The bishop who fully understands the object of the inquest, does not mean to have his assessment raised' (424), a comment all the more pertinent because the bishop in question was the Englishman Wulfstan. On fold-soke: 'It is the manure that the lord wants; the demand for manure has played a large part in the history of the human race' (76). And on royal benefaction to the church: '[The king] obtains not only remission of his sins, but also the friendship and aid of bishops and clergy. And so large stretches of land are "booked" to the churches. It is to be feared that if England of the sixth century had been visited by modern Englishmen, the Saxon chieftains would have been awakened to a consciousness of their "booking" powers by offers of gin and rifles' (242). Thus easily did he use the present to interpret the past. His ability to adjust his mind to the past, to imagine and rethink it as it was, is even more remarkable. Repeatedly he dwells on language and the meaning of words: ceorl, sac and soc, manor and hall, burgh (59, 84-87,108-9,183-86). Always he moves behind the static figures of his chief record to the organic growth which deposited them. So it is a book which describes what English society was like, what government was like, how men's minds worked in organising an archaic community. He does this with marvellous dexterity and with the continued

It should also be noted that Maitland and Karl Pearson, one of the founders of modern statistics, overlapped as members of the Bar in 1882-83, after which Maitland returned to Cambridge, with Pearson going on to University College, London, in 1884 as Professor of Applied Mathematics and Mechanics. But Maitland was a Bencher of Lincoln's Inn, Pearson of the Inner Temple.

62

Colonial England, 1066-1215

most intractable material. 'Men are learning to say what they really mean', he tells us (226). Then a warning: 'Against many kinds of anachronism we now guard ourselves. We are careful of costume, of armour and architecture, of words and forms of speech. But it is far easier to be careful of these things than to prevent the intrusion of untimely ideas. In particular there lies a besetting danger for us in the barbarian's use of a language which is too good for his thought' (356). Nevertheless, Maitland tries to interpret it. The barbarian, for all his materialism, is an idealist. He is, like the child, a master in the art of make-believe. He sees things not as they are, but as they might conveniently be. Every householder has a hide; every hide has 120 acres of arable; every hide is worth one pound a year; every householder has a team, every team is of eight oxen; every team is worth one pound. If all this be not so, then it ought to be so and must be deemed to be so' (389-90). And again: The result is that every manor in a certain district has four hides and sixteen teamlands. It is very pretty; it was never (except for technical purposes) very true, and every year makes it less true' (472). That is followed immediately by a very characteristic qualification: That exactly this was done, we do not say and do not think; but something like it may have been done' (473). Maitland is not always easy to pin down. He appreciated the difficulty of his chief source; Domesday Book is 'taciturn', its language 'not very patient o f . . . analysis' (50, 67). He responded scarcely ever with vagueness, but rather with delicate qualification. But he never abandoned his logical attack on the evidence: 'If a vill consists, as in Devonshire often enough it will, of some three villani, some four bordariiznd some two servi, the "township-moot" if such a moot there be, will be a queer little assembly, the manorial court, if such a court there be, will not have much to do' (21). These qualities embolden the tendentious critic, for it is easy enough to select particular targets with no attention to their place in the whole. Maitland is much subtler, his arguments more qualified and conditioned, than his critics usually allow. For the critic, in this book above all others, Maitland presents the disconcerting habit of bouncing back - posthumously. The second essay was designed in part to challenge Seebohm's English Village Community and the doctrine which derived the English villein from the slave of the Roman villa.5 It was firmly embedded, therefore, in its contemporary setting of the conflict between Germanists and Romanists, now long subsided. Yet what Maitland has to say underlies the discussion opened by Joan Thirsk on the common fields6 and by Trevor Aston on the origins of the manor.7 It is relevant to Susan Reynolds' study of communities in western Europe;8 indeed Maitland still provides an important corrective to that work because he retains lordship in its proper proportion: 'But then we have to notice that a village which has to pay a provender rent or 5

Frederic Seebohm, The English Village Community (London, 1883). Joan Thirsk, The Common Fields', Past and Present, pp. 29 (1964), pp. 3-25. 7 T.H. Aston, 'The Origins of the Manor in England', Transactions of the Royal Historical Society, fifth series, 8 (1958), pp. 59-83. 8 Susan Reynolds, Kingdoms and Communities in Western Europe, 900-1300 (Oxford, 1984). 6

Domesday Book and Beyond

63

even a tailla or gersuma is not altogether a free village. Its communal action is called out by seignorial pressure' (147). And, most remarkably, it includes many of the essentials of the argument of Alan Macfarlane's The Origins of English Individualism? Macfarlane's generous acknowledgement of Mai tland would probably have been even more enthusiastic if he had used Domesday Book and Beyond as well as the History of English Law, for his argument is here reinforced in ringing phrases - 'the very fields themselves seemed to rebel against communities and to demand a ring-fenced severalty' (351); and also by a detailed reconstruction of the economy and organization of what he called the free villages of eastern England (352-54). That is simply one example of Mai tland revived. The work of McDonald and Snooks illustrates another, for they argue not simply that in his sampling method Maitland was far ahead of other Domesday scholars including Round, but also that his broad equation of £1 = 1 hide was statistically correct.10 And there are many phrases or short sections in which Maitland anticipates later work: Postan, for example, in his comment on the development of labour services (58) ;u Lemarignier12 in the discussion of consuctudines (78-79); Stenton in the association of the freedom of some of the eastern counties with the Scandinavian settlements13 and also R.H.C. Davis's criticism of Stenton in the reservation - 'But in truth we must be careful how we use our Dane' (139);14 and finally the continuing debate about succession and inheritance, which has exercised Thorne, Milsom and the present writer,15 in the following comment on these intricate problems, much more emphatic than anything in the History of English Lanr. 'The noble obtains a spacious territory, perhaps a county, from the king byway of "benefaction"; precarium becomes the beneficium, the beneficium becomes the feudum. The king can not prevent the beneficia, the feuda, from becoming hereditary' (301). This is not to say that the whole structure of the book still stands without need for addition or repair. The problem of assessment and taxation dealt with in Essay III now has to be read in the light of later work, especially that of

9

Alan Macfarlane, The Origins of English Individualism (Oxford, 1978). All the references to Maitland which Macfarlane indexes are to the History of English Law. 10 John McDonald and G.D. Snooks, Domesday Economy (Oxford, 1986), especially pp. 42-49. 11 M.M. Postan, The Chronology of Labour Services', Transactions of the Royal Historical Society, fourth series, 20 (1937), pp. 169-93; also E.A. Kosminsky, Studies in the Agrarian History of England in the Thirteenth Century (Oxford, 1956). 12 J.F. Lemarignier, 'La dislocation du "pagus" et le probleme des "Consuetudines" (Xe-XIe siecles)', in Melanges d'histoire du moyen age dedies a la memoire deL. Halphen (Paris, 1951), pp. 401-10. 13 F.M. Stenton, The Danes in England', Proceedings of the British Academy, 13 (1927), pp. 20346. 14 R.H.C. Davis, 'East Anglia and the Danelaw', Transactions oj the Royal Historical Sodety, fifth series, 5 (1955), pp. 23-39. 15 S.E. Thorne, 'English Feudalism and Estates in Land', Cambridge LawJournal (1959), pp. 193209; S.F.C. Milsom, The Legal Framework oj English Feudalism (Cambridge, 1976). See also below, pp. 115-26, 197-220.

64

Colonial England, 1066-1215

C.R. Hart,16 and Sally Harvey.17 H.C. Darby and his colleagues have discussed in five volumes of The Domesday Geography18 matters which Maitland compressed within less than a hundred pages throughout the book, so much of what he says of the geographic distribution of wealth and the measurement of resources has been superseded. Likewise the criticism of Anglo-Saxon charters, on which part of the second essay depends, has advanced far since Maitland's day;19 his story of the development of seignorial jurisdiction, so dependent as he saw it on the alienation of royal rights, no longer seems so clear cut. It must now be read in conjunction with the work of Helen Cam and Naomi Hurnard.20 Certain sections of the work are plainly defective. Like Round, Maitland made only occasional use of the Exon Domesday (39,120,167,479), an essential component of the survey which was given insufficient attention by all Domesday scholars until Baring's paper of 1912.21 And from a commentary on the effects of the Norman Conquest one whole area of study - the settlement of the Norman aristocracy - is almost entirely missing. Indeed, Maitland washed his hands of it: The day for an artistically proportioned picture of the growth of feudalism has not yet come; the day for a quantitative analysis of the elements of feudalism may never come' (221), thus leaving the field to a whole host of scholars among whom Stenton and Le Patourel have been the leading figures.22 From major controversies the book has emerged with varying and still changing fortunes. Within the year of publication it provoked a famous review by James Tait,23 generously acknowledged by Maitland himself.24 Tait challenged two of Maitland's arguments. The first and less important concerned the garrison theory of the origin of boroughs (172-219). Tait was right; it may be that Maitland's readiness to reconstruct systems of assessment misled him here; so the section on boroughs remains largely as a historiographical curiosity. The second was Maitland's theory that manerium had the technical meaning of a 16

C.R. Hart, The Hidation of Huntingdonshire', Proceedings of the Cambridge Antiquarian Society, 61 (1968), pp. 55-66; idem, The Hidation of Northamptonshire (Leicester, 1970). 17 Sally PJ. Harvey, 'Domesday Book and Anglo-Norman Governance', Transactions of the Royal Historical Society, fifth series, 25 (1975), pp. 175-93; idem, Taxation and the Ploughland in Domesday Book', in P. Sawyer, ed., Domesday Book: A Reassessment (London, 1985), pp. 86-103; idem, Taxation and the Economy', in J.C. Holt, ed., Domesday Studies (Woodbridge, 1987), pp. 249-64. 18 H.C. Darby et al., The Domesday Geography of England, 5 vols (Cambridge, 1954-67); supplemented by H.C. Darby, Domesday England (Cambridge, 1977), and H.C. Darby and G.R. Versey, Domesday Gazeteer (Cambridge, 1975). 19 For a useful summary of part of this see N. Brooks, 'Anglo-Saxon Charters: The Work of the Last Twenty Years', Anglo-Saxon England, 3 (1974), pp. 211-31. 20 Helen M. Cam, The Evolution of the Medieval English Franchise', Speculum, 32 (1957), pp. 427-42; idem, The "Private" Hundred before the Norman Conquest', in J. Conway Davies, ed., Studies Presented to Sir Hilary Jenkinson (London, 1957), pp. 50-60; Naomi D. Hurnard, The Anglo-Norman Franchises', EHR, 64 (1949), pp. 289-323, 433-60. 21 F.H. Baring, The Exeter Domesday', EHR, 27 (1912), pp. 309-18. 22 See especially F.M. Stenton, The First Century of English Feudalism (Oxford, 1932); and John Le Patourel, The Norman Empire (Oxford, 1976). 23 EHR, 12 (1897), pp. 768-77. 24 Fifoot, Letters of Frederic William Maitltmd, no. 200.

Domesday Book and Beyond

65

house against which geld was charged and that men gelded in the manor, not the vill (120ff). This was much more central to Maitland's general theses; it extended, for example, into his lines of social classification (24,126-27). Tait's criticism has been generally accepted, but it is by no means as convincing as his views about the borough. Some of his arguments concerning detached portions of manors are as difficult to reconcile with manorial as geld renders. Maitland was not convinced by the criticism, although he never answered it.25 It has remained for Dr J.N.N. Palmer to revive Maitland's hypothesis.26 The debate about feudalism has been much more diffuse. Here Maitland certainly deserved some of the shot he has had to take. His first attempt to evaluate the English and Norman contribution is far from satisfactory: 'in the west we have already what in substance are knights' fees. The Bishop of Worcester held 300 hides over which he had sake and soke and all customs; he was bound to put 60 milites into the field; if he failed in this duty he had to pay 40 shillings for each deficient miles. At the beginning of Henry II's reign he was charged with 60 knights' fees' (160)

That, surprisingly, contains a simple but crucial error: the 60 knights provided for Henry II did not come solely from the 300 hides of Oswaldslow but from the whole of the bishopric of Worcester. So the equation, 60 knights = 300 hides is false and the continuity is broken. The mistake condemned Maitland's argument.27 However, the critics rarely note his much fuller and subtler treatment of the same problem in Essay II. Here he examined the laen lands of Oswaldslow and concluded: These men may be bound to fight at the bishop's call, but fighting is not their main business; they are not professional warriors. They are the predecessors not of the military tenants of the twelfth century, but of the radchenistres and radmanni of Domesday Book, the rodknights of Brae ton's text, the thegns and drengs of the northern counties who puzzle the lawyers of the Angevin time' (308); 'Dependent tenure is here and, we may say, feudal tenure, and even tenure by knight's service, for though the English cniht of the tenth century differs much from the knight of the twelfth, still it is a change in military tactics rather than a change in legal ideas that is required to convert the one into the other' (309). Here, in extended form, he still maintains a challenge to our latter-day Normanists and Saxonists alike. He was concerned with the origins and nature of seignorial power. He did not now pretend, as some of his followers have done, that Norman military arrangements could be traced back to the Anglo-Saxon period.28 Indeed, he had already accepted in the History of English Lawthe conclusion of Round's great paper on the Introduction of Knight25

Ibid., and no. 264. J J.N. Palmer, The Domesday Manor', in J.C. Holt, ed., Damesday Studies (Woodbridge, 1987), pp. 139-53. 27 For a summary of the criticism see R. Allen Brown, The Origins of English Feudalism (London, 1973), pp. 61-62. 28 For a critical review of such arguments see ibid., pp. 34-43. A subsequent statement of them 26

66

Colonial England, 1066-1215

Service.29 But he could not accept that feudalism was to be denned in narrow military terms: 'when compared with seignorial justice, military tenure is a superficial matter, one out of many effects rather than a deep-seated cause. Seignorial justice is a deep-seated cause of many effects, a principle which when once introduced is capable of transfiguring a nation' (258); and here he saw real continuity across the great divide of the Norman Conquest. He deserves closer attention and better understanding than knockabout partisanship has allowed him. Curiously, the work is only occasionally determined by one of its preliminary assumptions about Domesday Book: 'One great purpose seems to mould both its form and substance; it is a geld book' (3); 'Our record is no register of title, it is no feodary, it is no custumal, it is no rent roll, it is a tax book, a geld book' (5). Maitland did not examine this contention in any depth; it was perhaps sufficient that on this he and Round were in agreement. These much quoted broad assertions are necessarily imprecise. No one is likely to dispute that Domesday is in some sense a geld-book; it records geld assessments systematically throughout. But to move beyond this to an assertion that the prime purpose of both the survey and the Book lay in the reassessment and/or collection of the geld involves a logical leap. It does not follow. It requires proof. Maitland does not provide it. It is only much later, in Essay III, and then almost incidentally, that it becomes apparent that this was indeed his view. It almost steals out: 'With an eye to future taxation, he [the king] wishes for figures expressive of the normal condition of things' (422); 'They are not asking about area; they are asking about the number of teams requisite for the tillage of the tenement. With this and its value as data, William's ministers hope to correct the antiquated assessments' (423); and most clearly of all, 'If Domesday Book is to serve its primary purpose, if it is to tell the king's officers how much geld is due, it is absolutely necessary that by some ready process they should be able to work sums in hides and acres and in carucates and acres' (475). Now it should be obvious that an idea vividly expressed on p. 5 which does not resurface until p. 422 can scarcely be taken as determining the main structure of Maitland's great work. Domesday Book and Beyond, the bulk and the best of it, is concerned with much more than this plain assumption. Yet the assumption was made, and Galbraith was surely right when he pointed out that it was from Maitland, rather than Round, that it derived its main strength.30 As Maitland came to express it, it was almost self-defeating. The king's officers were to be able to 'work sums' 'by some ready process'. At this point Maitland was concerned with the arithmetic of Domesday Book: 1 hide or carucate = 120 acres: within that his point is valid. But it is quite invalid within wider parameters. Domesday continued

is made by John Gillingham, The Introduction of Knight Service in England', Proceedings of the Battle Conference on Anglo-Norman Studies, 4 (1981), pp. 53-64. 29 Pollock and Maitland, History of English Law, i, pp. 258, 259n. 30 V.H. Galbraith, The Making of Domesday Book (Oxford, 1961), p. 13.

Domesday Book and Beyond

67

Book is a hopelessly complicated and inadequate instrument for 'working sums' 'by some ready process' concerning the geld. Such calculations are not merely difficult; the information on geld is arranged in such a way and presented in such a format that geld calculations are positively impeded. If any economic information is emphasised it is not geld assessments, still less changes therein, but values, and these only by leaving them for the most part at the end of the manorial entries. Even then no attempt was made to rubricate them and no obstacle seems to have been envisaged against adding information after them. So the geld information is obscured. Geld assessment is an essential component of Domesday but Domesday cannot be a geld-book in Maitland's sense. In this it stands in sharp contrast to the Inquisitio geldi of the south-western counties, which fulfilled just such a purpose. Perhaps Maitland was fascinated by his own arithmetic. His resulting misapprehension is not at all obvious. It is shared by others firmly committed even now to diagnosing immediate financial purposes behind the Domesday data. So let us be quite precise about what went wrong and still goes wrong. Does Domesday contain geld assessments and reassessments? Yes. Can the assessment of manors, of vills, or hundreds, or of shires, be calculated? Yes. Is the Book so arranged to present such information readily and tidily? No; only in the case of manors and there it is given no emphasis either by location in the text, or capitals or rubrication. Can the total assessment of individual tenements in hundreds, or of vills divided between tenants or of tenements in shires, be calculated? Yes; but only by inconvenient search and summing. So could Domesday tell the king's officers 'how much geld is due'? Yes; but only if they did what Maitland does. Can the Book then provide the 'ready process' which Maitland required? No. Indeed Domesday Book and Beyond itself, with all its splendid calculations occupying many pages, contradicts the purpose it attributes to Domesday. It is not that it imposes on eleventh-century officials a task beyond their competence. Their arithmetic was up to it, given time and equipment. It is that Domesday is not the ready reckoner which the argument requires. The Inquisitio geldi fills this function much better. The moral is that Domesday reveals its purpose as much in its arrangement and format as in its content.31 Behind this there lies a real weakness. No one has ever matched Maitland in his ability to conjure the society of the eleventh and earlier centuries from the aridities of Domesday. Yet Galbraith far surpasses him in his sense that the Book was an enormous artefact made for a purpose, to a plan, and for use in a particular way; that it was in its time a living, working record. Half a century lay between the two, and one highly important work intervened: Domesday Rebound published for the Public Record Office in 1954, after Galbraith's first paper but before his first book.32 Maitland, of course, was conscious enough of a plan, a system; but it was his, not King William's. It owed much to nineteenth-century statistics, practically nothing at all to the study of the manuscript. It was, as 31 32

This argument is further developed in above, pp. 38-45. V.H. Galbraith, Domesday Book Rebound (London, 1954).

68

Colonial England, 1066-1215

Galbraith put it, 'Victorian'.33 The passage of time and change in approach which lies between the two is revealed in an almost casual footnote towards the end of Essay I: The one glimpse that I have had of the manuscript suggested to me (1) that the accounts of some of the boroughs were postscripts, and (2) that space was left for accounts of London and Winchester. The anatomy of the book deserves examination by an expert' (178 n. 1). That may well astonish the reader ninety years later.34 It reflects not dereliction of duty on Maitland's part, but what has happened since, and in particular the achievement and consequences of Galbraith's inspired intervention. At this point Domesday Book and Beyond may well be compared with Round's Feudal England. To some extent the two run in parallel; indeed at several points Maitland deliberately avoids duplication of Round's work, published two years previously. It may be because of this that he is somewhat hazy about how Domesday was made. It may equally be that he was not so resolved as Round on the interrelationships of the Domesday texts. Neither of them was perhaps as firm of mind as Galbraith's critique of them suggests. However, Maitland did say something. He came to it by a roundabout route in his third essay on the hide. Here he dwells on the considerable variation in the record of teamlands, arguing that the return of 'land for x teams' in some counties or hundreds was equivalent to the statement that 'there were x teams TRE' in others (420-24). In noting this varied response, he attributes the variation in Leicestershire, where it occurs within the confines of a single county, to 'a clerk's caprice' (421 n. 2). Elsewhere, especially in the contrast in these matters between Great Domesday, where the teamlands appear, and Little Domesday, where the matter is covered by a threefold response stating the number of ploughs actually on the land, he suggests that the original threefold questionnaire, seeking information TRE, TRW and when King William gave the land, was 'unnecessarily cumbrous. The design of collecting the statistics of the past broke down. . . Some interrogatories were dropped' (421-22). This was entirely consistent with Round, who took the 'so-called second volume to be really a first attempt at the codification of the returns' on which the first volume, i.e. Exchequer Domesday, was 'a wonderful improvement'.35 But Maidand's view was subtler, for his explanation depends on variations in both the questions asked by the commissioners and the responses given by the jurors (420-23). Unlike Round, he approaches, but never really attains, the modern view that the final Book retains evidence of successive layers of data-retrieval and reduction which necessarily reveal differences in method and procedure between circuits, counties

33

Ibid., pp. 14-15. Round said something very similar in comparing Great and Little Domesday (FeudalEngland, p. 140). In assessing, both Maitland and Round it should be born in mind that the zincographic facsimile of both Great and Little Domesday was published in county volumes by the Ordnance Survey Office, 1861-64. 35 Round, Feudal England, p. 141. 34

Domesday Book and Beyond

69

and smaller units. This was genius nudging against the confines of his time. The next great leap forward was Galbraith's. How then should we assess Maitland in this work? By his range and style, certainly. By his learning and acute intelligence, equally so. Also by his moving language: the final paragraph grips the reader like the closing sentences of Wuthering Heights and is as often quoted. Perhaps above all by a single phrase which embraces his mind, method and achievement, and also the Cambridge of his day: 'We make another experiment.' It was some experiment. If only all such could be both so venturesome and so rewarding.

This page intentionally left blank

5 Feudalism Revisited

J.H. Round's great essay on the introduction of knight service into England represents his most striking historical achievement. His thesis that William the Conqueror introduced a novel and revolutionary system of military service into England is now under fire. The ranging rounds were aimed as long ago as 1948 by Miss Marjory Rollings, who presented arguments for the continuity of the old English military and tenurial system based on the survival of the five hide unit in the knights' fees of the bishopric of Worcester.1 The main barrage has come only recently in the work of Mr Eric John, who, among other matters, vigorously revived Maitland's views on the 'feudal' characteristics of the laen lands of Oswaldslow,2 and of Professor Warren Hollister who has so far contributed four papers to the discussion, in the most original of which he has argued that both the length of service of the Norman knight and the scutage payment by which this service could be commuted were based on Old English precedents.3 In these circumstances the second edition of Sir Frank Stenton's Ford lectures of 1929 (English Feudalism, 1066-1166. Oxford: at the Clarendon Press, 1961. Pp. ix + 312. £2.2s.) is especially welcome. This book has worn magnificently well since it first appeared thirty years ago. Those who read it with recent discussions in mind will be struck by the amount of evidence which Stenton himself presents to illustrate the continuity between Norman feudalism and what had gone before. Thus he finds a parallel between the Norman vavassores and the sokemen of the English Danelaw, and discusses the passage in which the author of the Leis Willelme attempted to equate the heriots of the Old English nobility with the reliefs of the new aristocracy. He notes the application of the OE heall gemotto the honorial court and the OE cnihtto the Norman knight, and discusses thejurisdictional rights which the Norman landowner derived from his English predecessor. He is as aware as Round's recent critics of English milites recorded in the Domesday and other surveys, and of the military service performed after 1 J.H. Round, The Survival of the Five Hide Unit in the Western Midlands', EHR, 63 (1948), pp. 453-487. 2 Eric John, Land Tenure in Early England (Leicester, 1960), pp. 140ff. 3 C. Warren Hollister, The Significance of Scutage Rates in Eleventh- and Twelfth-Century England', EHR, 75 (1960), pp. 577-88; idem, The Annual Term of Military Service in Medieval England', Medievalia et Humanistica, 13 (1960), pp. 40-47; idem, The Five Hide Unit and Military Obligation', Speculum, 36 (1961), pp. 61-74; idem, The Norman Conquest and the Genesis of English Feudalism', American Historical Review, 66 (1961), pp. 641-63.

72

Colonial England, 1066-1215

the Conquest by drengs. Moreover, he gives great emphasis again and again to the vagueness of English feudalism in its first century, to the primitive character of early knighthood and the relatively small status of the early knights, to the variations in the knight's service, tenure and fee, and to the slowness with which precision was achieved on these and other matters. Above all, he insists on the inadequacy of the evidence and of our resultant knowledge. What we know on many of these matters could be radically altered by the discovery of a single charter. Sir Frank clearly does not believe that the Norman Conquest represented a complete break in these matters. Equally clearly, he is aware that continental precedents were not always followed by the Conqueror and his men in England. Nevertheless, he stands firmly on the conclusions he originally advanced in his lectures, and, behind them, on the work of J.H. Round. Much of the recent criticism of Round appeared too late for it to be discussed in this work. Even so, Miss Hollings's paper of 1948 contained many points later expanded by John and Hollister. Sir Frank has acknowledged her work, but clearly considers evidence drawn mainly from the see of Worcester to be atypical. In any case, he has not attempted 'to turn a book based on a set of lectures given in 1929 into one newly written in I960'. He has added comments on some new charter material, brought the notes up to date, and has recast part of the last chapter in the light of the text of the Gesta Stephani* Otherwise the work stands as in the first edition - full of powerful scholarship, expressive, humane and elegant. But, it will be asked, are some of its main conclusions out of date? Much of the recent criticism has tended to obscure the fundamental distinction, established by Round and accepted by Stenton, between the establishment of the servitium debitum due from the English tenants-in-chief and the establishment of knights' fees, the one normally the product of a single act on the part of the Conqueror, the other occurring over a period of time and subject to varying influences in different baronies. That there was continuity in the second process is obvious; that there was continuity in the first is a very different matter. Mr John admits that he cannot see how the servitium debitum was based on Anglo-Saxon precedent; Miss Rollings produced no very convincing case for such continuity even in the case of the bishopric of Worcester; only Professor Hollister, in fact, has produced an apparently workable hypothesis of continuity, not in the servitium but in the closely allied matter of scutage. If, indeed, both the length of knights' service and the system and rate of paying scutage were derived from Anglo-Saxon precedent, then Round's argument would have to be subjected to severe qualifications. Professor Hollister's arguments depend almost entirely on the arithmetical correlation of length of service (T), scutage (S) and the daily wage of a knight (W), in the equation S = TW. The elements in the equation, however, are often insecurely or inadequately based; not only Hollister but also Round himself probably overemphasised the correlation between scutage and the knight's 4

Gesta Stephani, ed. K.R. Potter (London, 1955), p. 14.

Feudalism Revisited

73

daily wage. Under Henry II scutage was levied at three different rates - one mark, £1, two marks. The change of rate is not correlated with the general rise of prices and of the knight's wage in the late twelfth century. The two mark scutages occurred in 1159 and 1161; one mark was charged in 1162; £1 as late as 1187. These rates can, in fact, only be related to what we know of the knight's wage by artificial and insubstantial arguments that Henry II decided to reduce the requisite sum demanded; for this there is no support apart from the assumption that scutage and the daily wage were closely related as a general principle. There is no real evidence for such a principle. The commutation of castle guard, for example, was not invariably related to a daily wage of 6d. or 8d.; sometimes it was probably the product of bargaining between lord and tenants rather than any precise economic calculation.5 It is probable that political considerations also affected the general rate of scutage and after Henry's reign wages soared to a level where scutage could not follow; if the fines pro servido did something to close the gap, they certainly were not related to the daily wage or even to the servitium debitum. The primafatie case for using the equation S = TW as a general principle for the second half of the twelfth century is not particularly convincing. It loses all sense when applied to the minute amount of evidence which is all that survives for the first half of the century. Our knowledge of the length of service in this period depends almost entirely on one important charter, recording a grant to Hugh de Ralegh of Nettlecombe in Somerset for the service of one knight. The grantor wasJohn fitz Gilbert, the marshal, father of William Marshal. The relevant section is as follows: 'tali divisione quod si werra est inveniet mihi unum militem procuratum duobus mensibus et si pax est xl diebus ad tale servitium quale milites baronum terre facere debent rationabiliter'.6 Now it is true that these words carry the authority of the marshal. Even so, the charter comes from the reign of Stephen, among whose unruly subjects John fitz Gilbert was not the most pacific. He is here referring to the knight's service owed to the baron, not to the baron's service owed to the king. Moreover, his words do not imply that two months' service in time of war was either normal or abnormal; to interpret the second clause as simply commenting on the first is to stretch the sense of the Latin; the words ad tale . . . rationabiliter are best interpreted as referring to the nature and quality of the service, not to its duration. The information on the rate of scutage is only slightly less difficult. Indeed, there is no direct evidence at all that scutage was even related to the servitium debitum, let alone a daily wage. Prior to 1119, the bishop of Norwich paid £60 pro militibus\ as his servitiumwas forty knights this looks like a scutage of 30s. per fee. In a famous writ of 1127 Henry I reduced the scutage due from the bishopric of Ely from £100 to £60, and as the servitium was again forty knights, this also looks like a scutage of 30s. per fee. 5 6

S. Painter, 'Castle Guard', American Historical Review, 40 (1935), pp. 450-59. Collectanea Topographica et Genealogica (8 vols, London, 1834-43), ii, p. 163.

74

Colonial England, 1066-1215

On this very threadbare basis Professor Hollister has stated the equation S = TW as 30s. = 60 x 6d. for the reign of Henry I. Hence he considers that the crown was forced to accept a reduction in the period of service during and after the reign of Stephen, despite Henry IFs refusal to accept precedents drawn from the Anarchy, despite his urgent need for long-term forces for his continental wars, and despite direct evidence to the contrary; for in 1158 the military tenants of both Abingdon and Malmesbury abbeys were instructed to perform their service to their lords as their ancestors had done in the days of Henry I.7 Moreover, since the sixty-day period of service cannot be traced in Normandy but can be traced in pre-Conquest England, he argues that the situation under Henry was derived from pre-Conquest England for which, using the wellknown Berkshire evidence and other Domesday references, he produces the formula 20s. = 60 x 4d. QE.D. However, if such arithmetical methods are to be used at all, it is best to have the arithmetic correct. The significant sum in the Ely evidence is not the £60 to which Henry I reduced the scutage in 1127 but the £100 which was the customary rate prior to that date.8 As Sir Frank Stenton shows, there is no evidence that the church of Ely had supplied a figure of 66 2/3 fees which the equation would here require. The old enfeoffment was 561A it may be that the servitium was increased to eighty for a time by Rufus; but the equation fits neither of these figures. At this crucial point, in fact, it fails completely. We are left with nothing to fill the gap between the reign of Henry II and pre-Conquest England. It is also important that such a formula should associate like with like. Professor Hollister's equation 20s. = 60 x 4d. is derived initially from the Berkshire Domesday which states that one miles went from five hides and that 4s; were given him from each hide as food and pay for two months. He reinforces this by evidence drawn from the West Saxon boroughs among which, for example, Malmesbury sent one man or 20s. pro honore v hidarum. Now it should be apparent that these are not like but unlike. The five Berkshire hides produce one warrior and 20s.; the five Malmesbury hides produce one warrior or 20s. The 20s. in Berkshire go to the warrior; the 20s. at Malmesbury go to the king. Moreover, while there may be some distant and purely accidental parallel between the Berkshire payment and the scutage paid by knights of an honour to support some of their number in the king's service once reduced quotas were emerging at the end of the twelfth century, there is no parallel at all between the borough commutations and scutage. The earlier payments were fixed charges and the option between them and personal service probably lay with the boroughs. Scutage in contrast was permitted as an act of royal grace and the king decreed the rate at which it should be levied. The tenant had no choice in the matter; the king could say that he wanted not money but men. 7

Chronicon Monasterii de Abingdon, ed. J. Stevenson, Rolls Series (London 1858), ii, p. 225; Registrum Malmesburiense, ed.J.S. Brewer, Rolls Series (London, 1879), i, p. 335. 8 'Centum libris quas predicta ecclesia solebat dare.' E. Miller, The Abbey and Bishopric of Ely (Cambridge, 1951), p. 160.

Feudalism Revisited

75

These arithmetical methods, which prove so misleading in this instance, can and have been used more widely in discussing the general problem of continuity, by Round, for example, in discussing the constabularia in Normandy and England, by his critics in emphasising the significance of the five-hide unit. There are obvious difficulties in this, for reasonable men can make divisions into units of fives and tens without copying each other. It is hazardous to assume that like numbers are always proof and dangerously rash to assume that unlike numbers are disproof. Furthermore, the equation of fees with hidage or carucage requires careful use. Sometimes it may result from the founding of knights' fees on the pre-Conquest assessments. Sometimes, indeed more frequently, it is simply a product of the economic and financial developments of the twelfth century which encouraged and required the evaluation of fees in terms of libratae, carucates or hides, and the calculation of the last two in real rather than customary terms. The equation was made more, not less, frequently as the century advanced. Nevertheless, the work of Miss Rollings especially has been valuable in demonstrating the use of the five-hide unit as a basis for the Worcester knights' fees. The process of establishing fees was relatively slow, and where men survived previous practice was bound to have an influence. Such survival is most marked where we are dealing not with individual landowners but with institutions; it is no accident that Round's critics have drawn their evidence almost entirely from ecclesiastical lordships and from borough customs. It would be unwise, however, to assume that conditions were similar on secular baronies, or that the continuity apparent in the establishment of some fees is an argument for continuity in the establishment of the servitium debitum. Only, indeed, in the north were there barons who held their baronies either wholly or partly by old Northumbrian services, and the reason is not far to seek. Here preConquest landowners survived, and here, too, the baronies were only established slowly; many were not even the creation of the Conqueror, but of his sons or the barons of his sons. Even here, although men could talk of the tenants of St Cuthbert as omnes barones, scilicet Teines et Dreinges, aliique probi homines? it was the Norman rather than the Northumbrian influence which ultimately predominated, as it was again in Lothian. At one point the discussion turns not on numerical analysis of fees or hides, but on the interpretation of a single word - the miles of the Berkshire Domesday entry. For Sir Frank Stenton this warrior represented the ceorl, for his critics he is the thegn. The difference is crucial, for on it depends the interpretation of several passages in Domesday Book concerning pre-Conquest military service which are alleged to demonstrate that the thegn's obligation to military service was territorial and not, as Sir Frank argues, personal. Any attempt to equate the Berkshire miles with the thegn faces one crucial and decisive objection. The fyrd survived the Conquest, to be used by the Norman kings alongside their mercenary forces and the feudal host. There is no indication that it served in

9

Historiae Dunelrmnsis scriptures tres, ed.J. Raine, Surtees Society, 9 (1839), p. ccccxxx.

76

Colonial England, 1066-1215

any but the traditional manner; indeed in the incident of 1094 in which Ranulf Flambard dispersed the force summoned to Hastings there is some indication that each warrior was still maintained by local contributions as in the case of the Berkshire miles. But while the fyrd survived the Conquest, the thegn did not. Sir Frank's critics cannot have it both ways. If they accept that service to the fyrd was assessed by hides, then they cannot also argue that the characteristic warrior of this force was the thegn. If this had been so the thegn and the fyrd would have vanished from the English scene together. In any case the relationship between the English warrior and the hide was not the same as that between the knight and the fee. The hide was anterior to the Berkshire miles', the knight, in contrast, was anterior to the fee. We can have knights without fees, but we cannot have the warriors of the Berkshire Domesday without hides. Finally, it is worth remembering that the events of 1066 were described then and afterwards as a conquest. Some argued that William had imposed a tyrannous yoke on the land; others glorified Norman achievements, as in the speech which Ailred of Rievaulx placed in the mouth of Walter Espec in his account of the Battle of the Standard; some might listen to tales of Hereward and of Saxon resistance; others, indeed any secular landowner, might appeal in the courts to tenure de conquestu. A widely varied body of evidence points in the same direction, and it is towards the conclusions of Round and Stenton. They were stated earlier in the Historia Eliensis: 'Usus ergo atque leges, quas patres sui et ipse in Normannia habere solebant, in Anglia servari volens'.10 English feudalism has always been something of a special case to continental scholars, most of whom have naturally relied heavily on the work of English colleagues. Even so, it is not simply because of this that the picture built up by Round and Stenton has won acceptance with them. For Marc Bloch, for example, it fitted neady into the general European scene as he presented it in La societe feodale, now available for the first time in an English translation (Marc Bloch, Feudal Society, translated from the French by L.A. Manyon, London, Roudedge and Kegan Paul, 1961. Pp. xxi + 498, £2.10s.). One of Bloch's main contributions here was to relate social development to politics and to show from the study of genealogy, literature and war, to name but a few of the fields he covered, that organised feudalism was associated with the emergence of a new military aristocracy. Feudalism was inimical to kinship and hereditary personal status. It was the work of the fittest who had survived and indeed exploited the chaotic conditions and opportunities of the ninth, tenth and early eleventh centuries. New social relations were the work of new men. Bloch's comparative studies led him to associate Anglo-Saxon England not with what came afterwards but with Merovingian Gaul. There is some exaggeration in this, but it is well to remember how England looks to continental eyes. These two books of Stenton and Bloch were not intended as works of controversy and indeed are far bigger than that. They are both replete with suggestive ideas 10

HistoriaElimsis, ii, c. 101, quoted Miller, Ely, p. 155.

Feudalism Revisited

77

and flashes of insight- Stenton, for example, on the place-name Knighton, or the features of early twelfth century castles, or the baronial agreements of Stephen's reign; Bloch on the medieval attitude to time, or the folk memory, or the tie of liege homage. At their best, they present not only conclusions but also a technique - the recreation of a society from the apparently humdrum records in which it unconsciously displayed its organisation and sometimes its aspirations. But if they are similar in this, they are very different in other respects. Feudal Society aims to cover a far wider field than English Feudalism. Part of the difference here probably springs from a different attitude to the word 'feudalism'. But behind this there lies a difference in method and historical approach. English Feudalism is empirical; the society it presents is a direct product of the evidence; it is intentionally a partial, limited view. Feudal Society in contrast is informed by theory; history for Bloch is a science, words are a tool to investigate not society but societies, societies themselves are organisations to be compared and contrasted in any sphere from the economic to the moral, their history is to be divided into stages. The results often provide an interesting contrast. English Feudalism has little to say on matters outside the immediate range of the sources used; Feudal Society has little to say about matters unrelated to its underlying concepts. Stenton follows the lead of his evidence. Bloch discusses problems and suggests solutions because they are important to him, not because the evidence makes their solution possible. The effect of this is that while Sir Frank Stenton can amend, add to and subtract from his work after thirty years, no such revision of Feudal Society would be possible. The book is at its least effective where the historian in Bloch gave in to the sociologist. For example, one of its great contributions lay in the comparative study of feudalism throughout western Europe. Work done since Bloch's day has emphasised differences not only between different countries but within each country. Feudalism in Normandy or Flanders, for example, was different from that of Burgundy or Lorraine. Varieties, in fact, have multiplied. To state the same thing from a different point of view, the stages through which feudalism is alleged to have passed have become more numerous and complex. It would certainly be difficult now to accept Bloch's view of two stages of feudal development. It is difficult indeed to see any stages at all which are applicable to Europe as a whole. The comparative method often ends in demonstrating regional variation rather than chronological categories. Bertrand de Born might cut a figure in Poitou, but his like faced summary execution in England. There is much in Feudal Society which requires amendment and correction. For example, Bloch probably overemphasised the poverty of Anglo-Saxon England, the extent to which Philip Augustus copied Angevin administrative practices, and the efficiency of the feudal levy. He failed to recognise the early date at which cash played an important part in feudal relationships and military organisation; here there is a useful corrective in the work off. BoussardJ.O. Prest-

78

Colonial England, 1066-1215

wich and B.D. Lyon.11 He probably underestimated the administrative importance of the castle and the castlery in the emergence of feudalism as demonstrated byj. Dhondt and others.12 Modern scholars also would probably place greater emphasis on population trends and land colonization in the stabilisation of feudal states. But despite all this Feudal Society remains a truly great book, still the classic analysis of west European feudalism. Mr Manyon has produced it in a sensible and readable translation with few blemishes. He has retained and amended the original index and reproduced practically all Bloch's original bibliography. His own supplement to the bibliography is quite inadequate. Selective though it may be, no principle of selection should exclude Boussard's book on the Angevin Empire or Plucknett's work on the legislation of Edward I or fail to include the names of Southern, Miller, Naomi Hurnard, or Feuchere.13 Both these books raise the important question of price. If the present trend towards the production of works of vulgarization for an undergraduate market is to be halted, works such as Bloch's and Stenton's should be made available at prices which undergraduates can and are prepared to pay. Bloch's objective was 'never to allow the reader to forget that history has still all the excitement of an unfinished excavation'. Young students will grasp this more readily by living with such books as English Feudalism and Feudal Society. Note This bibliographical study led to an exchange with Professor Warren Hollister in Economic History Review, second series, 16 (1963), pp. 114-18. I only include here detail of the evidence which I mentioned above (p. 74 n. 7) which correlates service under Henry I and Henry II (ibid., pp. 116-17): Henricus Dei gratia rex Angliae, dux Normanniae, Aquitanniae, et comes Andegaviae, baronibus, et militibus, et omnibus tenentibus de ecclesia de Malmesburia, salutem. Praecipio vobis quod plenarie et integre et bene faciatis abbati de Malmesburia servicium quod ei debetis, sicut illud plenius fecistis, vel antecessores vestri, Rogero episcopo, vel alicui abbati, tempore H. regis, avi mei. Et nisi feceritis, justiciarii mei faciant. Teste Gregorio, apud Theok.14 Queen Eleanor also wrote to the knights of the abbey as follows: [Alienor], Regina Angliae, et ducissa Aquitanniae et Normanniae, et comitissa Andegaviae, militibus quae [sic] tenent de feodo ecclesiae de Malmesburia, salutem. Praecipio vobis quod 11

J. Boussard, 'Les mercenaires au Xlle siecle: Henri II Plantagenet et les origines de 1'armee de metier', Bibliotheque de VEcok des Chartes (1945-46), pp. 189-224;J.O. Prestwich, 'War and Finance in the Anglo-Norman State', Transactions of the Royal Historical Society, fifth series, 4 (1954), pp. 1943; B.D. Lyon, From Fief to Indenture (Cambridge, MA, 1957). 12 J. Dhondt, Etudes sur la naissance des principautes territoriales en France (Bruges, 1948). 13 P. Feuchere, 'Essai sur 1'evolution territoriale des principautes francaises, Xe-XIIIe siecles', Le moyen age, fourth series, 7 (1952), pp. 85-117. 14 Registrum Malmesburiense, ed. J.S. Brewer, Rolls Series (London 1879), i, p. 335.

Feudalism Revisited

79

faciatis abbati de Malmesburia servicium de foedis vestris integre et plenarie, sicut recognitum fuit ante regem et ante justiciaries apud Wygorniam. Teste Roberto, comite Leicestre, apud Westmonasterium. Et sicut inde ei fecistis homagium vestrum. Teste eodem.15

She also sent letters to the tenants of the abbey of Abingdon: Aleinor, regina Angliae, ducissa Normanniae et Aquitaniae, et comitissa Andegaviae, militibus et hominibus qui de abbatia de Abbendona terras et tenuras tenent, salutem. Praecipio quod juste et sine dilatione faciatis Walkelino abbati de Abbendona plenarie servitium suum, quod antecessores vestri fecerunt antecessoribus suis, tempore regis Henrici, avi domini regis; et nisi feceritis, justitia regis et mea faciat fieri. Teste Joscelino de Baillol; apud Wintoniam.16

That this, like her letters to Malmesbury, was based on a royal order is confirmed by the note - 'per breve regis de ultra mare'. The date of these writs is beyond doubt. Henry's visit to Worcester, to which Queen Eleanor's letters refer, occurred in April 1158.17 From Worcester, Henry apparently moved to Tewkesbury where he issued his letters to the tenants of Malmesbury;18 the queen's letter to Abingdon, and presumably also her letter to Malmesbury, must belong to the period August-December 1158.19 They were issued the year before the great campaign and scutage of Toulouse in which Round found his most convincing evidence for the forty-day term of service. There are no grounds for supposing that these writs are simply concerned with the number of knights owed by the vassals of these houses or for asserting that they had nothing to do with service terms or scutage; one later famous dispute between an abbot and his tenants, at Bury St Edmunds, had a great deal to do with military service.20 These documents establish three points: (a) The writ to the tenants of Malmesbury resulted from the verdict of ajury presented to the king and his justices at Worcester. They were the outcome of a formal deliberation. (b) The writs are concerned with all or any of the services owed to the abbots. They specify none and exclude none. (c) Henry II stated that the services were to be performed as under Henry I. (It is worth adding that two of the three writs are addressed, not just to knights, but to all tenants of Malmesbury, and the other to the men, as well as the knights, of Abingdon. They could concern other, besides military services.) The forty-day term of service and other aspects of feudal service in Normandy have 15 16

Ibid.

Chronicon monasterii de Abingdon, ed.J. Stevenson, Rolls Series (London, 1858), ii, p. 225. R.W. Eyton, Court, Household and Itinerary of Henry II (London, 1878), p. 35. 18 Ibid., p. 36. Eyton had some slight doubt on the authenticity of this writ, presumably because of the formula 'Henricus Dei gratia rex Angliae'. However, its close relationship to Eleanor's two writs establishes the authenticity of its content. The authenticity of Eleanor's writs is accepted by H.G. Richardson, The Letters and Charters of Eleanor of Aquitaine', EHR, 74 (1959), p. 196n. 19 R.W. Eyton, Court, Household and Itinerary of Henry II, p. 43. 20 The Chronicle ofjocelin ofBrakelond, ed. E.H. Butler (London, 1948), pp. 85-87. 17

80

Colonial England, 1066-1215

been much debated. The arguments are well elucidated by Emily ZackTabuteau, 'Definitions of Feudal Military Obligations in Eleventh-Century Normandy', in Morris S. Arnold, Thomas A. Green, Sally A. Scully, and Stephen D. White, eds, On the Laws and Customs of England: Essays in Honor ofSamuelE. Thome (Chapel Hill, NC, 1981), pp. 18-59, especially, pp. 46, 48, 53, for the forty-day terms of service. The summary discussion of scutage (pp. 73-74) should now be read in the light of Thomas K. Keefe, Feudal Assessments and the Political Community under Henry II and His Sons (Berkeley, CA, 1983), pp. 20-40.

6 The Introduction of Knight-Service in England

The cartae baronum submitted to King Henry II in 1166 are a curiously neglected source in English history. Edited by Thomas Hearne in 17281 and by Hubert Hall in 1896,2 they were one of the bases ofJ.H. Round's paper on the introduction of knight-service published in 1891.3 Yet they have not been re-examined systematically in recent discussions of Round's famous thesis. Perhaps Round has seemed to squeeze them dry. Perhaps the «zrta£have not seemed particularly relevant to the lines which have been followed by Round's critics, for they provide only ancillary evidence in the debate about local 'continuity' between thegnly and knightly estates, and, since they were composed a century after the Norman Conquest, they seem to leave room for the argument that the quotas of knight-service of the late twelfth century arose, not artificially from some act of government, but through gradual growth in the century following the Conquest. Yet the cartae, with the pipe rolls, are the fundamental source for this whole field of study. They establish the size of many quotas of military service, whether by direct statement or by permitting computation; they illustrate the attitudes both of the crown and its tenants-in-chief towards military service; and they lead to questions about military service to which there is still no satisfactory answer. It is with the cartae, therefore, that this essay is first concerned. Round's chief concern was to correlate the cartae with the information on scutage in the pipe rolls of Henry II. The present objective is to correlate the information of both the cartae and the pipe rolls with the history of feudal tenements. For it is obvious enough that the military burdens laid on the tenantsin-chief of the crown were bound to bear the mark of forfeiture, escheat, marriage, sale and all the other accidents - genealogical, economic, and political -which might befall in the descent of a great estate. Round was not concerned with this within the context of his argument, and for good reason; for he set out to refute the notion that the military service of the Norman period 'developed in unbroken continuity from Anglo-Saxon obligations' .4 To that end the establish1

Liber niger scaccarii, ed. Thomas Hearne (Oxford 1728). The Red Book of the Exchequer, ed. H. Hall, Rolls Series (London, 1896). 3 J.H. Round, The Introduction of Knight Service into England', EHR, 6 (1891), pp. 417-43, 625-45; 7 (1892), pp. 11-24; reprinted in idem, FeudalEngland (London, 1909), pp. 225-314, from which all references below are taken. 4 Ibid., p. 234. 2

82

Colonial England, 1066-1215

ment of the existence of an unrelated, artificial system of military tenure was sufficient; and he scarcely paused to consider whether that apparently artificial system might itself be a result of gradual growth, not over the years of the Conquest itself, but in the century which followed. The evidence to the contrary, which placed responsibility firmly on the shoulders of William the Conqueror, seemed to him both clear and certain. However, Round had rejected 'continuity' through the front door only to leave the back unbarred; for a historian approaching the evidence of the pipe rolls and the cartae independently will at once be struck by the number of examples which do not seem to fit Round's hypothesis, either because the service seems to be random, quite unrelated to the units of fives and tens on which Round's arguments depended, or because the tenant-in-chief does not seem to know what his quota was.5 Now it is these instances which underpin the argument that quotas emerged gradually in the century following the Conquest.6 Round noted some of them, but asserted that they 'are quite insufficient to overthrow the accumulated array of evidence on the other side and some of them are, doubtless, capable of explanation'.7 This is the starting-point of the present argument. The cartae are not easy to deploy effectively. Some of the barons who made returns seem to have been uncertain about what was required. Henry II had apparently asked three questions:8 1 2 3

How many knights were enfeoffed by the death of Henry I? (The old enfeoffment.) How many knights have been enfeoffed since the death of Henry I? (The new enfeoffment.) How many knights remain on the demesne? That is how many knights was the tenant-in-chief to supply from his own resources in order to fulfil the service quota or servitium debitum?

A number of barons confused the second and third questions by recording that the new enfeoffment had taken place on the demesne, and a few confused the old enfeoffment in 1135 with the service due. Disentangling such misunderstandings is relatively easy where the transcripts of the cartae in the Black Book and the Red Book of the Exchequer present the whole or most of the original text. In many cases, however, they simply include lists of fees under the formal headings of old enfeoffment, new enfeoffment and demesne, categories which may be misleading in view of the confusion revealed by some of the complete returns. 5 H.G. Richardson and G.O. Sayles, The Governance of Medieval England (Edinburgh, 1963), pp. 88-90. 6 In addition to Richardson and Sayles, see Frank Barlow, William I and the Norman Conquest (London, 1965), pp. 107-9; DJ.A. Matthew, The Norman Conquest (London, 1966), pp. 127-28. 7 Round, Feudal England, p. 257. 8 Ibid., pp. 236-46; F.M. Stenton, TheFirst Century of English Feudalism (2nd edn, Oxford 1961), pp. 137-39.

The Introduction of Knight-Service in England

83

There is one further essential difficulty which cannot be circumvented. The cartae only reveal the service quota if it was directly stated or if it exceeded the sum of the old and new enfeoffment; in the latter case it can be calculated by simple addition of the old and new enfeoffment and service due on the demesne. Where there is no statement of the quota, and no genuine service recorded on the demesne, then the quota cannot be computed from the cartae. In some cases it can be established from the payments of scutage recorded in the pipe rolls of 1159,1161,1162 and 1165. But if that proves impossible then it is unlikely to be established at all, since later scutages were based not on the quota, but on the new or old enfeoffment. There are nearly forty tenancies-in-chief, numbering five knights' fees or more, where the sum of the old and new enfeoffments either exceeds, or coincides with, the quota and where no service is stated to have lain on the demesne. They include the earldoms of Arundel, Cornwall, Gloucester, Buckingham, Norfolk and Clare, and the great honours of Port, Belvoir, Wallingford and Eye. In some of these cases it may be that the old or the total enfeoffment approximates to or coincides with the quota.9 There is also a handful of examples where later evidence seems to provide a hint of what the quota had been; for instance, in 1213 Earl Roger Bigod fined with King John to secure a reduction of service from 125V4 knights, a figure based on the old enfeoffment which was the usual basis of assessment for the Bigod fee, to sixty,10 which is a typical figure for the larger honours established in the generation of the Norman Conquest. But all this is inference, and in all that follows, at least as regards the analysis of quotas, almost all these baronies have been left out of account. The random service which they rendered does not disprove Round's hypothesis; nor does it give support to some alternative argument that quotas arose from the gradual accumulation of precedent. In these cases there is no evidence that quotas had ever been established; it is equally true that if they had, the information would not reveal them. The remaining baronies, totalling some 150, may be categorised in two different ways. The first may be defined by the quota, which is represented either by a small number, up to five, or by multiples of fives and tens, or by some quite random number. Now there is no need to found the decimal quotas in Round's constabularia. What matters is that such quotas must in general represent an arbitrary system of service. Precedents do not accumulate in fives and tens. Anglo-Saxon tenements of five hides certainly survived into the Norman period; some, a few, became knight's fees;11 but no convincing argument has yet been produced to equate the decimally based quotas of the Norman period with the decimally arranged hides or the duodecimally arranged carucates of the Anglo9 The knights' fees reported from the honour of Eye in 1166 totalled eighty, but there is nothing in the return to indicate that that was the servitium debitum (Red Book, p. 411). 10 Pipe Roll 13John, p. 2. 11 Marjory Rollings, The Survival of the Five-Hide Unit in the Western Midlands', EHR, 63 (1948), pp. 453-87.

84

Colonial England, 1066-1215

Scandinavian kingdom.12 There is no satisfactory escape from the conclusion that the decimal quotas were imposed. The random quotas in contrast could simply have grown. Indeed one of the objects of this essay is to show how they grew. A second, more complex set of categories is concerned with the history of the tenancies-in-chief simply as property. Some baronies enjoyed a relatively smooth passage from the Conquest to the reign of Henry II, descending directly in the male line, undisturbed by political disaster or the fortunes of civil war, unaffected by the financial consequences of excessive enthusiasm for the crusade, monastic endowment, lavish building, or the accumulation of estates. Some baronies had a more chequered history, broken by division among heiresses or the succession of a collateral, or a disputed claim sometimes coinciding with civil war. Some escheated to the crown, temporarily or permanently, because of default of heirs or the treason of the tenants. Many baronies of the twelfth century were not at first tenancies-in-chief at all; they began as undertenancies of the great lordships of the Norman period - Mortain, Bayeux, Montgomery, Lancaster - and emerged as holdings in chief only when the mesne lordship had disappeared through reversion to the crown. Others were late creations, established long after the influences which had determined the initial pattern of enfeoffment had lost impetus. Others were marcher lordships and hence had special, localised military functions. Yet others were reconstituted holdings, motley concoctions manufactured out of bits and pieces from other estates. These categories have been distinguished more sharply than a close examination of individual estates might sometimes warrant. Nevertheless, artificial though they are, they established a point of great importance. The less disturbed the descent of an estate the more likely it is to have a decimal quota. The more difficult its descent, the less likely it is to have a decimal quota. The decimal quota goes with the normal, the random with the abnormal. This contrast is by no means obvious from a casual glance at the known servitia, for the decimal quotas come from baronies which illustrate almost the whole range of'accidents' affecting the descent and homogeneity of feudal tenancies. It is only when approached through the history of the tenancies that it stands out sharply and clearly. Much the largest single group of baronies with known quotas descended relatively undisturbed as tenancies-in-chief in the male line from the time of the Domesday Survey to the reign of Henry II. Of these, nearly thirty in all, the Vere barony of Hedingham paid scutage on thirty-one fees in 1162,13 and the

12 See the difficulties encountered by Rollings, ibid., pp. 475-60 and by E.John, Land Tenure in Early England (Leicester, 1964), pp. 158-59, and the comments thereon ofJ.O. Prestwich, 'AngloNorman Feudalism and the Problem of Continuity', Past and Present, 26 (1963), pp. 45-44. Compare the comment of Helena M. Chew, The English Ecclesiastical Tenants-in-Chiefand Knight Service (Oxford, 1932), pp. 6-7, on F.W. Maitland, Domesday Book and Beyond (Cambridge, 1897), p. 160. 13 Pipe RoU 8 Henry II, p. 71. Compare Pipe Roll 14 Henry II, p. 39; Pipe Roll 6 Richard, p. 37; Pipe Roll 13John, p. 124. The total of fees returned in 1166 was 291/8 (Red Book, pp. 352-53).

The Introduction of Knight-Service in England

85

Giffard barony of Elston (Wiltshire) apparently owed a quota of nine;14 all the rest had decimal quotas.15 Of the Domesday tenancies-in-chief with known quotas which descended unimpaired in the female line, a group of fifteen, only three had random quotas in 1166; the Peche barony of Great Bealings or Great Thurlow (Suffolk) returned twelve,16 the barony of Aveley (Essex) seven and a half,17 and the Marmion barony of Tamworth probably twelve;18 all the rest had decimal quotas.19 Of the Domesday tenancies-in-chief with known quotas which descended unimpaired and without dispute through collaterals, a relatively small group of three, there was no exception to the general pattern.20 Even where rebellion led to escheat, and the temporary or final dispossession of a family, an honour was still likely to return a decimal quota so long as it was not dismembered. Hence the Lacy honour of Pontefract was assessed at sixty,21 and so also were the Mowbray honour of Thirsk22 and the great fee of Tickhill,23 all of which underwent escheat in the century after the Conquest. Holderness, confiscated in turn by the Conqueror, by Rufus and by Henry I, nevertheless retained its integrity and rendered a service of twenty knights in 1168.24 In all, more than twenty baronies out of a total of nearly thirty which escheated temporarily to the crown, or which seem to have suffered some unexplained break in descent and yet retained their unity, returned a decimal quota on performed service 14

Pipe Roll 8 Henry II, p. 14; Red Book, p. 245. Benington, Blankney, Cainhoe, Cavendish, Caxton, Crich, Eton, Hastings, Craon, Helion Bumpstead, Horsley, Kentwell, Norton, Pleshey, Poorstock, Rayne, Richard's Castle, Rothersthorpe, Stafford, Tarrington, Tattershall, Thoresway, Tutbury, Weedon, West Dean, Whitchurch, Wolverton, Wormegay. 15 For convenience of reference, here and throughout, I have used the classification of IJ. Sanders, English Baronies: A Study of Their Origin and Descent, 1086-1327 (Oxford, 1960). Limitation of space prevents me from including a full apparatus. I have restricted detailed references to particular cases or to supplementing information both on the descent of estates and military service which is summarised in Sanders. 16 Red Book, pp. 366-67. 17 Ibid., p. 353; Pipe Roll 14 Henry II, p. 38. 18 Pipe Roll 11 Henry II, p. 18. On the origins of the Marmions as lords of Tamworth see Complete Peerage, viii, pp. 505-6. The descent of Tamworth probably also involved collateral succession. 19 Blagdon, Brattelby, Chipping Warden, Dudley, Hockering, Hooton Pagnel, Kempsford, Malton/ Alnwick, Lewes, Nether Stowey, Okehampton, Trowbridge. 20 Bywell, Castle Gary (?), Monmouth. But see also Tamworth in n. 18 above. Okehampton also descended to collaterals as well as through heiresses. See Complete Peerage, iv, pp. 308-9. 21 For the breaks in the descent of Pontefract, see W.E. Wightman, The Lacy Family in England and Normandy (Oxford, 1966), pp. 66-73. For the quota see Red Book, p. 421-44. 22 1 take the sixty fees de antiquo feodo to stand for the servitium (Red Book, p. 420). On the descent of Thirsk and its associated lordships see Charters of the Honour of Mowbray, ed. Diana E. Greenway (London, 1972), pp. xxi-xxiv. 23 Pipe Roll 7 Henry II, p. 38; Pipe Roll 8 Henry II, p. 34. On the descent of the honour see The Cartulary ofBlyth Priory, ed. R.T. Timson, Thoroton Society, record series, 27, 28 (1973), pp. cxvcxl, and below, p. 155. 24 For the confiscation and descent of the lordship, see Barbara English, The Lords of Holderness, 1086-1260 (Oxford, 1979), pp. 6-16; for the servitium, see Pipe Roll 14 Henry II, p. 90; Pipe Roll 18 Henry II, p. 62. 15

86

Colonial England, 1066-1215

equivalent to a decimal quota.25 As a group they are less uniform than those which descended easily in the male line, but the decimal quota was nevertheless preponderant. The main source of random quotas lay elsewhere. Most important of all in terms of numbers are those baronies of the late twelfth century, nearly fifty in all, which had emerged from undertenancies of the great lordships of the first Norman settlement. In many instances direct continuity between the initial undertenancy and the later barony can be demonstrated without much difficulty, both as regards the holding and its tenancy. The same is sometimes true of the service, and hence this, when it emerges as a 'baronial' quota in the relative light of day of the late twelfth century, reflects the policies of the early tenants-in-chief rather than of the crown. The most obvious instance is that of the 'fees of Mortain' which were assessed by the exchequer at two-thirds of a normal fee.26 They cannot have been anything but the creation of Robert, count of Mortain, or of his son, William. Many of them, indeed the majority, carried decimal quotas.27 This is also true of some of the Shropshire baronies which were first established by Roger of Montgomery, or by his son Robert of Belleme.28 In contrast, only a few of the Kentish baronies which originated in fiefs held of Odo, bishop of Bayeux, returned a decimal quota: the Talbot/ Mountchesney barony of Swanscombe, which owed thirty,29 the baronies of Chilham and Patricksbourne, each of which owed fifteen,30 and the Arsic barony of Cogges which owed ten.31 The remaining quotas were either three, or a multiple of three, or some apparently quite random number.32 In all these cases, what determined the pattern of enfeoffment initially is now largely lost to view, but the tendency of the pattern to vary from one honour to another is 25

Returning decimal quotas in 1166 or performing decimal service apparently equivalent to the quota: Barnstaple, Bourn, Bradninch, Cogges, Curry Malet (?), Great Weldon, Great Torrington, Holderness, Little Dunmow, Little Easton, Odell, Patricksbourne, Pontefract, Shelford, Southoe, Stansted Montfichet, Stanton le Vale, Thirsk, Tickhill, Totnes (?), Winterbourne St Martin. Apparently carrying non-decimal quotas: Berkhamstead, Cottingham, Haughley, Lavendon, Salwarpe. In the case of Haughley there may have been a quota of fifty. The enfeoffment of eighty in the honour of Eye could possibly coincide with a quota. 26 Red Book, p. 232; Paul Vinogradoff, English Society in the Eleventh Century (Oxford, 1908), pp. 5254; C. Warren Hollister, The Military Organisation of Norman England (Oxford, 1965), pp. 44-46. 27 Cardinham (twenty fees of Walter Hay, Red Book, p. 261), Chiselborough ten, Hatch Beauchamp twenty, Odcombe fifteen. For further comment on the fee of the Surdeval family, see below, pp. 89-91. 28 Castle Holgate probably five (Red Book, p. 275), Cause five, Wem probably five (Book of Fees, p. 144). 29 Red Book, pp. 195-96. 30 Ibid., pp. 191-92, 197; and probably also Peverel of Dover (Pipe Roll 33 Henry II, p. 209). 31 Pipe Roll 7 Henry II, p. 26. 32 Allington probably three (Red Book, p. 197). Chatham either twelve or fourteen (Pipe Roll 11 Henry II, p. 106; Red Book, p. 191). Folkestone probably twenty-one (Red Book, pp. 615-16), Pont probably twelve (Red Book, p. 618), West Greenwich probably twenty-four (Red Book, p. 617). The above figures largely depend on the castle-guard owed at Dover. For Ros, in contrast, there is a firm statement of seven (Red Book, p. 197).

The Introduction of Knight-Service in England

87

clear and beyond question: the obvious conclusion is that the Norman tenantin-chief was himself responsible. It is not impossible, of course, that the service was adjusted when such an undertenancy became a tenancy-in-chief. But, if so, it is apparent that such an intrusion was insufficient to eradicate the characteristic features of the fees of Mortain, for example, and it is likely in general that neither William Rufus nor Henry I would want, or be in a position to reorder the service of tenants who sided with them against the great rebel lords. As a category, therefore, this group of baronies is of considerable interest. It reflects the remarkable loyalty of the undertenants to the crown and the increasing security of tenure which their loyalty achieved.33 It also suggests that these quotas at least were established at an early date: before Odo's final forfeiture of 1088 for the tenancies of his barony; before 1102 at the very latest for the baronies which stemmed from the honour of Roger of Montgomery; and before 1104, in some cases probably earlier, for the fees of Mortain.34 This group of baronies is partly overlapped by another: the marcher lordships. The records of the late twelfth and early thirteenth centuries reveal the existence along the northern and Welsh borders of estates graced with the title and status of baronies from which military service was sui generis, designed to meet the special circumstances of border warfare. The Northumbrian baronies were created in the main by Henry I; a few may have originated before 1100, either in grants from Rufus or as undertenancies of the Norman earldom of Northumbria.35 Styford and Mitford had a quota of five; Bywell returned sometimes five, sometimes thirty;36 the rest carried random quotas, all of them except the Vesci barony of Alnwick which returned twelve, of less than five.37 The pattern is similar further west in Cumbria. Here all the early holdings, whether established by the first lord of Cumbria, Ranulf Meschin, or by Henry I, were held by the render of cornage; it was only gradually in the course of the twelfth century and in the reign of John that this was converted to knightservice, and then the quotas were small and random.38 Some of the Welsh marcher lordships likewise lay outside any formal pattern. At Wigmore, in the late thirteenth century, the Mortimers claimed to hold by the service of two knights in time of war in Wales and of one knight in time of war in England.39 At Oswestry in 1166 the fitz Alans held by the service often knights in exercitu et chevaltia within the county of Shropshire and, 'as the old men testify', of five knights outside the county.40 33

For further comment see below, pp. 137-38, 141-42, 175-76. For further discussion of some of the tenancies of Mortain see below, pp. 97, 175-76. 35 William E. Kapelle, The Norman Conquest of the North (London, 1979), pp. 191-200. 36 Red Book, pp. 437-38; Pipe Roll 8 Henry II, p. 52; Pipe Roll 14 Henry II, p. 172; Book of Fees, pp.201. 37 Ibid., pp. 200-3. The figures here depend on the returns to the Inquest of 1212. 38 J.C. Holt, The Northerners (Oxford, 1961), pp. 91-92; Kapelle, TheNorman Conquest of the North, p. 212; Book of Fees, pp. 197-99. 39 Col I.P.M., iv, p. 235. 40 Red Book, p. 274. 34

88

Colonial England, 1066-1215

Baronies which were originally under-tenancies and the marcher lordships account for a very large proportion of the random quotas of service. The remainder come largely from baronies which escheated for a time to the crown and from baronies which were seriously reconstructed. True, an escheated barony might retain a decimal quota if there were no serious dilapidation. But any estate which did not descend in the direct line of adult heirs was to some degree at risk. Estates which underwent wardship or descended through heiresses or through collaterals, and even more those estates which escheated through default of heirs or treason, were vulnerable, a prey to litigation, to the greedy eye of powerful neighbours, most of all to the crown's exercise of patronage. Consider the fate of the Domesday holding of the Picard, Arnulf de Hesdin, which Round touched on in his paper on the origins of the Stewarts.41 Arnulf's sons died without heirs; one was probably hanged by Stephen at Shrewsbury in 1138; and the estates, centred on Wiltshire and Gloucestershire, were divided respectively between two daughters, Maud and Aveline. A quota, either for Maud's portion or the whole estate, was still recorded under Henry II; it was assessed at forty in 1165.42 Even so, Maud's grandson, Pain of Montdoubleau, submitted a confused carta in 1166: seven fees of the honour had got into the hands of Geoffrey de Vere during the reign of Henry I; Earl Patrick of Salisbury had received twenty fees in marriage with Pain's aunt, and had also taken a further fee by war which was now in the hands of Alfred of Lincoln.43 Indeed it would be quite impossible to construct a quota for this barony but for the assessment of 1165. For the other portion of the estate, which Aveline took by marriage to the fitz Alans of Oswestry, there is no such saving evidence. In 1166 William fitz Alan returned a carta which is among the most insistent and idiosyncratic in distinguishing between the enfeoffment and the quota. Nevertheless, in accounting for his Wiltshire fee all he could do was to list knights fees, which totalled eight and a half, and then add: 4We are not certain what service is owed to the king from this holding which belonged to Arnulf de Hesdin'.44 Yet a quota there must in all probability have been for the other division of Arnulf s Domesday holding still carried one. The obvious conclusion to be drawn from William's carta is that it was no longer possible to trace it. It is also noteworthy that William fitz Alan expected an estate which he defined by reference to its Domesday tenant, dead before the end of the eleventh century, to have carried one. Other baronies present similar stories. William son of Reginald de Ballon's was but briefly reported in his carta in language still reflecting the pathos of the dispossessed: William son of Reginald performs the service of one knight from his demesne. Hamelin de Ballon, his grandfather, was enfeoffed of the old feoffment to perform the 41

J.H. Round, Studies in Peerage and Family History (London, 1907), p. 116. Pipe Roll 11 Henry II, p. 74. 43 Red Book, pp. 297-98. 44 Ibid., p. 274. 42

The Introduction of Knight-Service in England

89

aforesaid service. He has been deprived of Great Cheverell on the order of the lord king and of the honour of Abergavenny for which he owes service, if the lord king so pleases. He has no one enfeoffed of the new enfeoffment.45

The estates of Roger of Berkeley were undergoing similar decomposition. Berkeley itself had just been lost to Robert fitz Harding. In his carta Roger reported two and a half fees of the old enfeoffment on which some of the service had been lost. He then added five fees and a number of hides from which further service had been lost; some of these were in dispute with Maurice son of Robert fitz Harding.46 From this report the Exchequer plumped for a service of seven and a half fees in 1168 and subsequent assessments.47 What it had been earlier it is impossible to say. Robert fitz Harding, in contrast, the new tenant with a clear commitment and a clean sheet, reported a quota of five for the honour of Berkeley in 1166.48 The delapidated barony is one side of the coin; the reconstituted barony the other. These two groups, especially the reconstituted baronies, are much the most illuminating of all. Up to a certain point it was possible for the Norman kings to act on a tabula rasa, to construct baronies from the lands of English tenants, from the royal demesne and from escheated estates of rebels, and to impose an artificial, decimal quota. After that point this became difficult if not impossible, for the escheats were occupied by enfeoffed knights owing specified amounts of service established by their dispossessed lords. The dividing line between the two, the point of balance, came in the reign of Rufus and Henry I, especially in the decade straddling the battle of Tinchebrai of 1106. This is well illustrated from the history of a number of Lincolnshire and Yorkshire baronies which were reaching their final form at this time. The Paynel fee, for example, was a conglomerate estate which included considerable additions to the lands which Ralph Paynell held at the time of Domesday Survey: in Lincolnshire, lands previously held by the Lacys which in 1086 had formed part of the barony of Odo of Bayeux, forfeited in 1088; and fees which Richard de Surdeval held of the honour of Robert, count of Mortain.49 Both these acquisitions probably came to Ralph by marriage.50 Despite these diverse origins, and despite the complicated division of the barony among the Paynels, it seems quite certain that the servitium came to the artificial total of forty-five in the 1160s: fifteen the responsibility of Robert de Gant;51 fifteen the responsibility of William Paynel of Hooton Pagnel;52 and fifteen apparently already divided by 1124 between 45

Ibid., p. 281. Ibid., pp. 292-93. 47 Pipe Roll 14 Henry //, p. 123; Pipe Roll 18 Henry II, p. 121. 48 He nevertheless reported that he was unable to obtain service from manors and fees still held by Roger of Berkeley (Red Book, p. 298). 49 £KC,vi, pp. 56-62. 50 Ibid., vi, p. 4. 51 Ibid., vi, pp. 33-34. 52 Ibid., vi, pp. 41-42. 46

90

Colonial England, 1066-1215

Hugh Paynel of West Rasen and Fulk Paynel of Drax.53 How was such a total figure reached? It can scarcely be accepted as an accidental accumulation. It must therefore be that either the service due on the Paynel Domesday fief, plus the service on the new acquisitions, happened to come to forty-five, which seems too much of a coincidence; or the original service due on the 1086 holding was forty-five and the later additions carried no additional service, a process of augmentation which occurred in the case of some knights' fees, but at first sight seems most improbable in this case because of the extent of the additional tenancies; or the whole service of the Paynel fee was recast after the acquisition of the Lacy and Surdeval estates and their conversion to tenancies-in-chief following the forfeiture of Odo of Bayeux and the counts of Mortain. This last hypothesis requires us to accept a projection of the atmosphere of the Conquest into the period 1088-1106, with the king still constructing great fiefs and his leading subjects still ready to accept heavy feudal service. However, this can have been achieved only so long as there was the possibility of a clean sweep, only so long as the estates in question were not already in the possession of tenants holding by precise amounts of military service. If such tenants were firmly established then the perpetuation or establishment of an artificial quota would obviously be much more difficult. Compare the Paynel fee with the barony of Bourne which was granted by Henry I to Baldwin fitz Gilbert de Clare within a decade or so of the point at which the Paynel barony was reaching its final form. Bourne included estates drawn from at least three Domesday fiefs, those of Oger the Breton, Godfrey of Cambrai and Baldwin the Fleming.54 It could only have produced a conveniently round number of knights if the lands had hitherto carried no service, or if the original service was now overriden, or if by coincidence the service came to a round figure. In fact it did not. The service of Bourne was lOVs. That Vs does not simply represent a near miss at ten. lOVs is not only the service recorded in the 1166 return for the barony but also the service for which scutage was consistently charged thereafter.55 In this instance assessment in round numbers was abandoned; instead the service was determined by the amount already due on the various fiefs which were thrown together to form the barony, plus any later enfeoffment. All this finds confirmation in the unusually diffuse and irregular carta which Hugh Wake submitted for his barony in 1166. Instead of accepting the usual distinction between old feoffment and new feoffment and service in demesne, he seems to have followed categories of his own: first the fees existing in the barony when it was created and secondly the subsequent feoffments of his predecessors, William de Rollos and Baldwin fitz Gilbert, and himself. The sum total was the service due. This is the service', 53

Ibid., vi, p. 8, 19. The Lincolnshire Domesday and theLindsey Survey, ed. C.W. Foster andT. Longley with introduction by F.M. Stenton, Lincoln Record Society, 19 (1924), pp. 162-64, 173, 196-97; Facsimilies of Early Chartersfrom Northamptonshire Collections, Northamptonshire Record Society, 4 (1930), pp. 1920. 55 Pipe Roll 14 Henry II, p. 64; Pipe Roll 2 Richard, p. 90; Pipe Roll 3 John, pp. 7, 9. 54

The Introduction of Knight-Service in England

91

he submitted, 'whereby my predecessors in their day served King Henry I who gave them the lands and I owe you, as the lord who gave them to me, the service of my body when you wish to take it.'56 This is quite different from the Paynel fee. Here the servitium is derived from the original and later constituents of a fief; there the servitium was apparently imposed on those constituents. Neither the Paynel nor Bourne fees stand as isolated examples. On the side of Paynel there is also Trussebut. This barony consisted largely of the Domesday holding of Erneis de Burun. But Geoffrey fitz Pain, the tenant in the early twelfth century, held further lands in Beckering, Riby and elsewhere in Lincolnshire, some if not all of which were derived from the estates of Roger of Poitou forfeited in 1102.57 The fee then was conglomerate in structure. Yet the servitium of the Trussebut fee was almost certainly fifteen and if not that then ten.58 Again the Brus fee comprised numerous lands in Yorkshire which were still the lands of the king in 1086, many more where Robert de Brus succeeded Robert, count of Mortain, many more still where Brus succeeded the count's tenant, Richard de Surdeval, and also the small Domesday holding of William Taillebois in Lincolnshire:59 once again a conglomerate structure, yet the service of the Brus fee was indubitably fifteen.60 Other fees followed the pattern suggested by the example of Bourne. The Ros barony of Helmsley originated in the grants which Henry I made to Walter Espec, again from the forfeited estates of Robert, count of Mortain.61 This was concurrent with the foundation of the Brus barony; indeed Robert Brus and Walter Espec were closely associated figures in the administration of the northern counties at this time. Yet while the Brus fee appears to have had an artificial servitium, the Ros fee certainly had not. The feoffment in 1166 was 52%4 de veteri and 2l/z de novo. Scutage was usually charged on 5%.62 Nor was the crown the only influence at work recasting the structure and service of tenancies-in-chief. The 1166 carta of the honour of Mowbray reveals yet another possible development. Here the knights of the honour presented the awkward number of eighty-eight enfeoffed knights, but added that sixty were de antiquofeodoo and that twenty-eight had been added by Nigel d'Albini from the demesne.63 Nigel acquired the barony in the reign of Henry I c. 1109. But for the long memory of the knights of the honour we should not know what the original service had been or that the change of ownership had been accompanied by, or had led during Nigel's lifetime to, a considerable increase in the number of fiefs. The consequences of forfeiture, accumulation and reconstitution were unpredictable. However, they contributed to many of the random servitia which 56

Red Book, pp. 378-80. £FC,x,pp. 23-27. 58 Pipe Roll 11 Henry II, p. 50; Pipe Roll 14 Henry II, p. 90; EYC, x, p. 8. 59 Ibid., ii, pp. 16-19; Lincolnshire Domesday, p. 197; Book of Fees, p. 166. 60 Pipe Roll 14 Henry II, p. 90; Red Book, p. 434. 61 EYC, x, pp. 143-45. 62 Red Book, pp. 432-33, 490; Pipe RoV ;t Richard, p. 163; Pipe Roll 8 Richard, p. 185. 63 Red Book, pp. 420, 490. 57

92

Colonial England, 1066-1215

appear in 1166. The honour of Miles of Gloucester, combining lands held in the previous generation by Walter of Gloucester and Bernard de Neufmarche, along with acquisitions from both Stephen and Matilda, returned seventeen de veteri and 334 de novo fees.64 Cottingham, the remnant of the Stuteville estates seized after the capture of Robert de Stuteville before Tinchebrai and only restored to the family under Stephen, carried a quota of eight.65 Marshwood, first established for the Mandevilles by Henry I, returned a quota of eleven plus some fractions of fees and one fee of Mortain.66 In one important case the process whereby the random quota was achieved is clearly recounted. William d'Aubigny's carta runs as follows:67 This is the holding of William d'Aubigny, butler of the lord king, of the gift of Henry [I] who gave him fifteen knights of the fee of Corbuchon. And later he gave him a fee of ten enfeoffed knights of the land of Roger Bigod with the daughter of Roger Bigod of his own hand. And later he gave him the service of Ralph son of Godric of twelve knights; and the service of Alfred de Athelburgo of two knights; and of Picot de Bavent of one knight; and the fee of Reiner Sine Averioof one knight; and the fee of William de Musterville of one knight.

These totalled forty-two fees. William then outlined his father's and his own enfeoffments. As it happened the grand total, including the ten fees held in mesne tenure of the Bigod fee, came to seventy-five. That is a useful reminder that a decimal quota could sometimes be achieved at random, for this barony clearly originated in a succession of grants of enfeoffed knights. That could scarcely produce a decimal quota other than at second-hand or by accident. Such an analysis of the relation between the quotas and the descent of baronies establishes one certain and central point: the decimal quotas were the more ancient. To dig down through the layers of evidence deposited by disturbed tenancies, to shift to one side the peculiar services of the marcher baronies, to recognise that some layers in the record are the work not of the crown but of tenants-in-chief, is to strip and lay bare the remains of a structure, much of it still surviving in 1166 in clear detail, in which quotas were based on a decimal system; that is on an artificial system, on an arbitrary system. Change was from the artificial to the random, not from the random to the artificial. Change occurred when something happened to disturb the normal descent of tenancies; and it was likely to occur only when that happened. If the hypothesis of gradual growth involves the view that there was some kind of debate about the service quotas in the course of the twelfth century, either within baronies, or between barons and the exchequer, a debate whereby in the end, through the accumulation of precedents, many quotas were settled on a decimal basis, then 64 David Walker, The "Honours" of the Earls of Hereford in the Twelfth Century', Transactions of the Bristol and Gloucestershire Archaeological Society, 79 (1960), pp. 174-211; Red Book, pp. 293-94. 65 EYC, ix, pp. 5, 70-76; Red Book, p. 429. 66 Ibid., p. 219. 67 Ibid., pp. 397-98.

The Introduction of Knight-Service in England

93

some evidence must be produced to establish a trend which seems to run clean against the demonstrable consequences of tenurial disturbance and quite contrary to the exchequer's determination, from 1166 onwards, to move away from artificial quotas towards some more realistic assessment based on the number of knights enfeoffed. The obvious conclusion to be drawn from the evidence is that the quotas were mostly known and fixed by the end of the eleventh century at the latest. There are other routes to the same end. The instances collected by Round of barons who in 1166 claimed tenure a conquestu ought perhaps to be left out of the count;68 for these assertions are balanced on the other side by statements that quotas were unknown, or were confirmed only by reference to juries of knights of the fee; and in the last resort what these soldiers said either way is not certain evidence. But the division of their estates is. Between 1145 and the accession of Henry II the lands of William Paynel of Drax were divided between the two sons of his first marriage on the one hand, and his daughter by a second marriage and her husband, Robert de Gant, on the other. Despite opposed political allegiances, which led to conflicting grants by Stephen and Henry of Anjou, and despite the fact that one half was subsequently subdivided unequally between the two sons, each half returned the same quota of fifteen in 1166.69 Sometime before 1100 Joel of Totnes was established by William Rufus in the honour of Barnstaple. When Joel died before 1130 his lands descended to a childless son and then to two daughters married respectively to Henry de Tracy and Philip de Briouze. In 1165 their sons each accounted for the service of twenty-five knights.70 William de Rames, lord of Rayne in Essex, who died before 1130, left his estates to two sons, Robert and Roger. Robert died without issue before 1142 and the lands were reunited, but when Roger died before 1159 they were split up once again. In 1159 each half of the original barony accounted for 9l/2 fees, but as each enjoyed a reduction of half a knight the quota of each half must have been ten, which one half had in fact returned in 1166.71 Hardouin de 1'Escalerie, the Domesday tenant of Caxton, Cambridgeshire, divided his lands between two sons. In 1166 his grandsons each returned a quota of fifteen.72 In all these cases only three explanations seem possible. One is that the quotas of the separate fragments of these baronies happened to coincide. A second is that equal quotas were imposed subsequently to the division of the baronies. A third is that an existing quota was divided when an estate was partitioned, and there can be little real doubt that this last is the most satisfactory explanation. If so, then the Paynel quota must have been settled before 1154; the Barnstaple quota, certainly, and the Rayne quota, probably, before 68

Round, Feudal England, pp. 295-96. £KC,vi, pp. 19,32-33. 70 Pipe Roll 31 Henry I, p. 153; Pipe Roll 11 Henry II, p. 80. 71 J.H. Round, Geoffrey deMandeville (London, 1892),pp. 399-404', Pipe Roll 5 Henry II, p. b;Liber Niger, i, p. 240, which is superior to Red Book, pp. 356-57. 72 Curia Regis Rolls, v, pp. 139-40; Red Book, pp. 367-69. 69

94

Colonial England, 1066-1215

1130; and the quota for Caxton within a few years of the end of the eleventh century. The acquisition of the lands of heiresses provides evidence less strict in its implications, but pointing to the same conclusion. The carta which William de Roumare returned in 1166 made no distinction between the honour of Bolingbroke, the lands of the heiress Lucy, on the one hand, and the patrimony of the Roumare family in Hampshire and Wiltshire on the other. These estates had long been associated for Lucy's second husband, Roger son of Gerald de Roumare, who brought them together, died c. 1097.73 Yet in 1159 and in 1162 the Roumare patrimony was assessed separately at twenty fees.74 All this seems to lead on to a ready acceptance of Round's great thesis of 1891. But it is not quite so easy, for at the very core of Round's case there lay a simplification at once unacceptable and inaccurate. It is contained in his statement: I hold that... [the tenant-in-chief s] military service was in no way derived or developed from that of the Anglo-Saxons, but was arbitrarily fixed by the king, from whom he received his fief, irrespectively both of its size and of all pre-existent arrangements.75

It is clear from the context that Round was not referring to any king, but to William the Conqueror. His statement was aimed in determined fashion against the notion of'continuity', but it also conveyed the impression that the Conqueror, at one and the same time and in one and the same act, established both a barony and its service quota, and that impression has been a pervading influence ever since Round wrote. The difficulty, of course, is that although a quota might be established in a moment the honour to which it was attached could not. One of the curiosities of the historiography of Anglo-Norman feudal tenures is that this problem was recognised or at least dimly perceived long ago. Few apparently now read Stubbs' Constitutional History on the introduction of knightservice, yet the sixth edition of volume one appeared in 1897, six years after Round's famous paper, and Stubbs overcame his episcopal responsibilities and defied the onset of academic encrustation to the extent of making three amendments to the existing text. The first referred to Round's paper and accepted his distinction between the 'exaction of military service' and the 'carving out of the land into knights' fees'. The second amounted to a cautious acceptance of Round's main point: The early date at which the due service (debitum servitium) of feudal tenants appears as fixed goes a long way to prove that it was settled in each case at the time of the royal grant.

But that concession to the new view was followed immediately by a third amendment to the old text - 'It must not however be assumed that this process was other than 73

Compkte Peerage, vii, pp. 743-44. Pipe Roll 5 Henry II, p. 47; Pipe Roll 8 Henry II, p. 36. 75 Round, Feudal England, p. 261. 74

The Introduction of Knight-Service in England

95

gradual.'76 Maitland also sank a probe into this central vagueness at the heart of Round's argument. He, like Stubbs, was convinced in principle, and agreed that 'We have good reason to believe that the Conqueror when he enfeoffed his followers with tracts of forfeited land defined the number of knights with which they were to supply him'. But he added a prescient footnote: This we regard as having been proved by Mr Round's convincing papers... Sometimes when land came to the king byway of escheat and was granted out, new terms would be imposed on the new tenant; but, in the main, the settlement made in the Conqueror's day was permanent.77

That note, properly pursued, would have started several important questions to which Round's hypothesis gives rise. For the situation which he imagined, in which an honour created in its entirety was bestowed on a baron in return for a specified amount of military service, can have occurred but rarely. The land which the Conqueror used to endow his followers came to his disposal in successive lots: in 1066 the old royal estates of the Anglo-Danish royal house, the possessions of the family of Godwin and the lands of those who had fought and lost at Hastings; in 1069-71 the midland and northern estates of Earls Edwin and Morcar and their followers; in 1075 the East Anglian properties of Ralph Guader. Even along the southern English littoral, in Sussex, where the first baronies were established between 1067 and 1070, Mr Mason has convincingly demonstrated that there was extensive subsequent readjustment.78 Further north the great baronies were built up even more obviously by miscellaneous acquisition. Count Alan of Brittany cannot have received many of his estates in Yorkshire before the fall of Earl Edwin in 1071; it is unlikely that he received the Lincolnshire and Norfolk estates of Ralph the Staller before the rebellion of Ralph Guader in 1075; the honour of Richmond was of composite origin.79 So also was the honour of Chester. As recorded in Domesday it was a hotchpotch, made up of lands which had been held not only by English predecessors, Harold, Earl Siward, Earl Edwin and Eadnoth the Staller, but also by Frenchmen, Walter of Dol, Ralph de Beaufou; and indeed in one estate Walter of Dol had been preceded in King William's time by Walter of Caen. As William Farrer noted, 'it is improbable that Earl Hugh obtained all the lands that he held in 1086 at the same time'.80 Nor did the process of accretion end in 1086. It was continued after 1088 by the escheat and redistribution of the estates of Odo of Bayeux and of some of the lands of Robert, count of Mortain; 76

W. Stubbs, Constitutional History of England (6th edn, Oxford, 1897), i, pp. 284-85. Cf. 4th edn (1883), i, pp. 284-85. 77 F. Pollock and F.W. Maitland, History of English Law (Cambridge, 1898), i, pp. 258, 259n. 78 J.F.A. Mason, William the First and the Sussex Rapes, Historical Association, Hastings and Bexhill Branch (1966), pp. 15-16. 79 EYC, iv, pp. 2, 94-95, 117; Lincolnshire Domesday, pp. 63-68; DB, ii, pp. 144, 147, 148b; VCH, Norfolk, ii, pp. 10, 17; Complete Peerage, ix, pp. 568-74; x, pp. 783-84. 80 W. Farrer, Honors and Knights'Fees, ii (London 1924), pp. 5-6.

96

Colonial England, 1066-1215

in 1095 by the escheat of the lands of Robert de Mowbray; and these were but the first of successive deprivations and redistributions continuing until after Tinchebrai. All this left its mark. The Bigod fee in Suffolk included manors which Roger Bigod had held of Odo of Bayeux at the time of Domesday.81 So also did the Lacy fee in Yorkshire and other counties.82 The Mowbray fee included over a hundred carucates in Yorkshire, Lincolnshire and Nottinghamshire which Gilbert Tison had held at the time of the Survey.83 Lands held by Erneis de Burun in 1086 are later found in the Mowbray, Trussebut and Paynel fees;84 all three lordships had decimal quotas in 1166. How were the decimal quotas established for such conglomerate baronies? There seem to be four obvious hypotheses. The first is that estates, as they were combined and redistributed during and after the Conqueror's reign, already carried service which happened in combination to produce decimal quotas. That demands excessive faith in coincidence; moreover, when such a process can be observed in the late twelfth century it clearly had the opposite effect of producing random quotas. A second possibility is that the quotas were not established until the end of the process of territorial agglomeration, and this too is unconvincing. There is clear evidence, too well known to require further discussion here, that William the Conqueror had imposed quotas on ecclesiastical tenants-in-chief by 1070/72.85 Moreover, there appears to be no obvious explanation of why we should allow Rufus and Henry I to impose arbitrary quotas, but not William. Yet again the territorial flux did not cease at any one point in time. There was no occasion when Rufus or Henry could assume that all was at last ship-shape and ready for the establishment of the neat, artificial system much of which still survived in 1166. Finally, the process led increasingly away from decimal towards random quotas; this stage had already been reached when Henry I began to enfeof William d'Aubigny, his butler.86 This leaves two possibilities. One is that nothing was done to change the quota when new estates were added to an honour. The other is that the quota was recast, either in the Conqueror's time or later, as additions or changes were made to the estates of an honour, but recast within the system of arbitrary, decimal figures. These hypotheses are not mutually exclusive. Policy must have depended on the king's relations with a particular tenant-in-chief, perhaps on the size of a barony and its existing quota, on whether the accretion of estates came suddenly or as a more continuous trickle, on whether the lands were enfeoffed or not. Hence it is reasonable to assume in any particular case that the quota may have been either maintained or recast, 81

VCH, Suffolk, i, pp. 392-93. EYC, iii, pp. 123,125; Whiteman, The Lacy Family, pp. 31-32. 83 EYC, xii, pp. 19-23. 84 Charters of the Honour of Mowbray, pp. xxii-xxiii; EYC, vi, p. 57; x, pp. 23-24. 85 The importance of the writ to vEthelwig, abbot of Evesham, which Round, Feudal England, pp. 303-4, rightly emphasised, is not seriously diminished by Richardson and Sayles, The Governance of Medieval England, pp. 64—65. 86 See above, p. 92. 82

The Introduction of Knight-Service in England

97

The evidence on this matter is very fragile. The first hint cornes from two episcopal baronies: Thetford/Norwich, which has been carefully examined by Miss Dodwell;87 and London, which was discussed by Round.88 In each case the baronies carried decimal quotas, of forty and twenty respectively; and for Norwich the quota can be shown to have been established before 1111-13. In each case the Domesday Survey listed considerable acquisitions made by divers and sometimes devious means between 1066 and 1086, one at least of which in the case of Thetford/Norwich can scarcely have been held as a tenancy-in-chief before 1075.89 In each case the acquisitions were listed separately from the other episcopal holdings. Those of Thetford/Norwich are headed terra defeudo\ those of London feudum episcopi.90 Whatever is made of this, it seems clear that the Domesday commissioners were conscious of the problem created by territorial accretion, and in these cases they seem to be settling it by stating that the acquisitions were part of, not separate from, the service quota. An alternative, less acceptable hypothesis would be that these acquisitions were the whole fee, but this does not tally with the pattern of subenfeoffment. The gains were considerable. In the case of Thetford/Norwich Miss Dodwell noted that 'they radically altered the balance of properties . ..' and since they lay in south-east Norfolk 'cannot but have facilitated the later move [of the diocesan centre] to Norwich'.91 But, whichever hypothesis is adopted, there is nothing definite to show whether territorial accretion affected the quota in these instances or not. Some further light is shed on the problem by the break-up of the lands of Robert and William of Mortain, for their enfeoffments might still survive as small 'fiefs of Mortain' in the late twelfth century. It is certain that some of their estates in the north, which never seem to have been restored after Robert's rebellion of 1088, were incorporated in other baronies without any apparent effect on the quota or the rate at which fees were charged.92 In many southern counties, in contrast, these accretions tended to figure as fees of Mortain additional to a separate quota charged at the full rate, and it may well be significant that charters to the monastery of Marmoutiers demonstrate that William of Mortain still had control of some of his Dorset estates in 1104-6.93 The different treatment of the Mortain estates may then reflect the date at which they escheated: the earlier escheats being incorporated in existing or revised servitia; the later escheats leaving the characteristic increments to servitia in the form of fees of Mortain. This problem is probably incapable of any general solution. However, it mat87

Barbara Dodwell, The Honour of the Bishop of Thetford/Norwich in the Late Eleventh and Early Twelfth Centuries', Norfolk Archaeology, 33 (1963), pp. 185-99. 88 VCH,£55**,i,p.339. 89 Eccles was recorded in Domesday Book as a tenancy held of Earl Ralph; Dodwell, The Honour of the Bishop of Thetford/Norwich', p. 186. 90 DB,ii,pp. 11,193. 91 Dodwell, The Honour of the Bishop of Thetford/Norwich', p. 187. 92 EYC, ii, pp. 11, 16, 19; vi, pp. 57-58. 93 Cal Docs. France, no. 1209.

98

Colonial England, 1066-1215

ters less than a more obvious point which Round's argument obscured. Whatever view is taken of the perpetuation or reconstruction of the original quotas, it is clear that Rufus and Henry I still had the capacity to establish considerable territorial agglomerations, either ab initio or through reconstruction, which carried decimal quotas. The last obvious examples are the Redvers barony of Plympton, which must have been established by 1107, and the Brus barony of Skelton, which originated at about the same time. The first probably and the second certainly were assessed at fifteen knights.94 Moreover, it seems probable that the main reason for moving away from such assessments toward random quotas is not to be found in any deliberate change in royal policy but in the fact that estates available for redistribution to deserving supporters were encumbered with sitting tenants. Now if this is so it necessarily reflects on the rationale behind the whole system. It has been argued with increasing force, most convincingly of all by J.O. Prestwich, that the Norman kings relied on long-term military service for pay and other rewards and that the service quotas were an unreal basis for the muster of an effective fighting force.95 What then was their function? It might easily be imagined that the Conqueror introduced the system simply because he could not conceive of land held freely on any other terms, and it is this escape-route which the present argument necessarily closes. For if, as Round argued, the quotas were established by the Conqueror, it follows that the quotas must have been reconsidered and confirmed or readjusted, and in some cases established ab initio by Rufus and Henry I. And few would be ready to accept the hazardous hypothesis that not just one, but all three, were acting in unquestioning, purposeless, absence of mind. At this point the pipe rolls provide an invaluable cross-bearing. At first sight they seem to fit into a neat scheme in which tenants-in-chief either paid scutage or not, and in which the scutage was calculated pro rata, on the quota up to 1166 and thereafter on the old or new enfeoffment. The system of the Normans was apparently still in working order. But there is a difficulty, serious enough to wreck such a neat reconstruction. The implication of the pipe roll evidence is that men paid their full scutage, or none at all, and hence that they served with their full quota or not at all. Now there are many entries of quittances on only a part of the service due, but these are usually allocated to under-tenants, who had presumably given their service; indeed some of these under-tenants were themselves tenants-in-chief and can be shown on such occasions to have received quittances in that capacity also. But, these cases apart, the pipe rolls scarcely ever reflect the ordinary accidents of military organisation and campaigning. 94

On Skelton see above, p. 91. The fifteen owed from Plympton was the recognised service and apparently included Christchurch and the Isle of Wight, Book of Fees, p. 74. At the end of the twelfth century scutage was also paid on forty-five fees which were not recognised, Pipe Roll 6 Richard, p. 171; Pipe Roll IJohn, p. 199. It may be that the break in the tenure of the honour in the reign of Stephen lies behind the discrepancy, Complete Peerage, iv, pp. 309-12. 95 J.O. Prestwich, 'War and Finance in the Anglo-Norman State', Transactions of the Royal Historical Society, fifth series, 4 (1954), pp. 19-43. Since feudal tenants often served for diverse forms of reward quite apart from their due service I have deliberately avoided the word 'mercenary'.

The Introduction of Knight-Service in England

99

The record never mentions tenants-in-chief charged just on one or two fees, representing tenants who lay sick abed or had important engagements elsewhere. Now that must mean that the impression of the pipe rolls is unreal and unacceptable. And that in turn means that service in the field must have been acquitted not necessarily by the precise quota but by a reasonable turn-out acceptable to the king and his marshal; and what that amounted to must have depended on many other factors: the interest of the tenant-in-chief in military matters, the period for which he and his men were ready to serve, and so on. In short, the earliest pipe rolls at first hide, but in fact imply, that feudal obligations were met as they were later, in the late twelfth and early thirteenth centuries, when service in the field was not precisely related to the quota or to the old or new enfeoffment. If these arguments are correct they help to explain an unusual feature of the cartae of 1166. A considerable number of tenants-in-chief expressed ignorance of their quotas or reported some difficulty in establishing what they were: that they had had to enquire, for example, from juries or from the ancient men of the fee. In some cases these difficulties can be directly related to the history of the tenancy, but not in all. Earl Patrick of Salisbury reported his enfeoffment and quota of forty knights after enquiring from 'his honest and ancient men' ;96 the 'ancient men' of the barony of Little Dunmow likewise informed Walter fitz Robert that his quota wasfifty;97and Geoffrey de Mandeville consulted with the men of his fee to establish that his quota was sixty.98 These statements, otherwise difficult to explain, fall into place if it is accepted that service need not have been performed at the level prescribed by the quota. For the common feature of the cartae quoted above is not that these men had not served; if their absence from the earlier accounts from 1159 to 1165 can be taken at its face value, they had; but that they had not paid scutage. And if the payments of sums assessed against the knight's shield was the main practice keeping the quota alive, then it is probable that these men, and others who served regularly, had not been required to remember their quotas for a very long time. There need be no wonder that the memories of the ancients had to be searched. Nothing has bedevilled the discussion of military service in Anglo-Norman England so much as the peculiar nature of the cartae and the related pipe roll evidence. In one aspect they are a living record, a revelation of a system of financial assessment still being hammered out and refined. In another aspect they are dead, an archaeological deposit left by the tenurial history of the previous century. The reality of the one should not be confused with the unreality of the other. The cartae were drawn up at a time when the pattern of enfeoffment was changing. Henry II continued old established quotas, but he never attempted to establish new baronies with large decimal quotas. Berkeley, with a quota of five, was an example from the first years of the reign and was prob96

Red Book, pp. 239-40. Ibid., p. 348. 98 Ibid., p. 347. 97

100

Colonial England, 1066-1215

ably determined by the fact that it had formed part of an already established barony. For the rest, Henry II was satisfied with small servitia which rarely exceeded one knight. Admittedly, by this time, there was less opportunity for the crown to redistribute property in the large-scale lavish manner of Rufus and Henry I. But Henry II made no obvious attempt to follow earlier practice, and the small baronies and quotas of his reign are better associated with the later enfeoffments of Richard and John. At least that was so in England. In Ireland it was different, for Henry II was to Ireland as the Conqueror had been to England, and Henry, who never established a single new barony with a quota of more than five in England, imposed a quota of a hundred on Leinster, sixty on Limerick, sixty on Cork and fifty on Meath." At the same time the Norman aristocracy, heavily engaged in conspicuous expenditure on more comfortable stone-built castles in England, reverted to raising motte and baileys by the dozen in the newly subject land across the Irish sea.100 The old system was dead, but could be resuscitated very rapidly if the circumstances required. And, as in England earlier, so in Ireland the system soon became unreal, overlain by scutage and paid service within not much more than a generation of the first settlement. What then was the system? Is system even the right word? It seems to have amounted simply to this: that when, or perhaps before, the Conqueror in England or Henry II in Ireland granted large estates to their followers they wanted some guarantee that the recipients would contribute towards the successful maintenance of the conquest. Hence the imposition of quotas. Was it 'when' or 'before'? In Ireland it was 'when'; quota and endowment coincided. In England the quota could have come first, to be matched by territorial endowment as and when it became available. That is consistent with the indications that decimal quotas survived territorial accretion. There is no need to follow Round in imposing the practices of the twelfth century, in which specified service was performed for precisely defined tenure, on the crucial years immediately after the Conquest. The English quotas were large in comparison with those in Normandy, or in the case of Ireland in comparison with Henry II's enfeoffments in England, not because William was more powerful as king than as duke, or simply because Henry II wanted to bring the Norman adventurers in Ireland to heel, but because of the greater urgency in newly conquered lands. There is little to suggest that either William I or Henry II expected their vassals to stand in serried ranks of wer fives and tens to be told off by numbers from the right; military urgencies were likely to be too immediate for such strict accounting. At this preliminary stage too, as regards the conquest of England, there can have been little serious intention (except perhaps in 1086) of using an army so raised for war on the Continent; nor was there much possibility yet of exploiting the quotas as fiscal assessments. These emphases came in later generations, the result first of the 99

J. Otway-Ruthven, 'Knight Service in Ireland', Journal of the Royal Society of Antiquaries of Ireland, 89 (1959), pp. 1-15. 100 H.G. Leask, Irish Castksand Castellated Houses (Dundalk, 1964), pp. 6-11; G.H. Orpen, Ireland under the Normans (Oxford, 1911), i, pp. 338-43; ii, map.

The Introduction of Knight-Service in England

101

wars among William's sons and then of the wide-ranging interests of the Angevin kings. If the rigid accountancy of the late twelfth-century exchequer is transferred from scutage to service, and then from late twelfth-century service to late eleventhcentury service, the system looks woefully illogical and unsatisfactory. How could men serve both as tenants-in-chief and under-tenants, for example? And how could they fight in the field and defend their castles at one and the same time? But the system in its origin was real enough if viewed as a simple, rough and ready practical response to the hazards of conquest. Note Among the many papers which have been written since 1983 on honorial origins Richard Mortimer, 'Land and Service: The Tenants of the Honour of Clare', Anglo-Norman Studies, 8 (1985), pp. 177-97, is particularly important in correlating the history of the tenements with service. See especially pp. 190-94. His paper on 'The Beginnings of the Honour of Clare', Anglo-Norman Studies, 3 (1980), pp. 119-41, discusses the development of a composite honour, but one for which there is no recorded servitium. There is further detailed investigation of the stages in the establishment of Norman lordship in England in Robin Fleming, Kings and Lords in Conquest England (Cambridge, 1991), pp. 145-214.

This page intentionally left blank

7

The Carta of Richard de La Haye, 1166: A Note on 'Continuity' in Anglo-Norman Feudalism1

In the revival of the doctrine of'continuity' between English and Norman tenurial systems considerable attention has been paid to Round's discussion of the relationship between knights' fees and hides or carucates. 'If we found', wrote Round, 'that the recognized area of a knight's fee was five hides, it would give us a link between the under-tenant (miles) and the Anglo-Saxon thegn.'2 The point has been accepted; indeed, the most striking argument for 'continuity' has been one which largely turned on the survival of the five-hide unit.3 Round himself was negative. In a brief section of his essay on the introduction of knight service, he demonstrated that the territorial assessment of the fee might range from two to ten hides or from five to fourteen carucates.4 Most of Round's examples were taken from the returns to the enquiry into feudal military service of 1166 which are entered in the Red Book of the Exchequer:5 With characteristic thoroughness he noted all the relevant instances in these returns of some kind of equation between knight service and hides or carucates. They make a very mixed bag. Most of them are drawn from ecclesiastical baronies;6 on some of these there may have been an element of 'continuity'.7 Others provide early examples of the type of correlation of knight service and territorial assessment which came into fashion in many baronies from the second half of the twelfth century.8 Others may simply represent a casual equa1

In writing the final draft of this paper I have benefited from discussions with Miss Barbara Dodwell. 2 J.H. Round, FeudalEngland (1895), p. 293. 3 Marjory Rollings, 'The Survival of the Five Hide Unit in the Western Midlands', EHR, 63 (1948), pp. 453-87. 4 Round, FeudalEngland, pp. 293-95. 5 The Red Book of the Exchequer, ed. Hubert Hall, Rolls Series, (London, 1896), i, pp. 186-445. 6 The abbeys of Westminster, Milton and Shaftesbury, and the bishopric of Salisbury (ibid., i, pp. 188, 210, 214, 237-38). 7 The crucial ratio of one knight to five hides appears twice on the Worcestershire estates of the abbey of Westminster (ibid., i, p. 188) and occasionally in the cartaof the bishop of Salisbury (ibid., i, pp. 237-38). It also appears in the carta of the abbey of Shaftesbury, but six hides to the fee was more usual here (ibid., i, p. 214). The carta of Roger of Berkeley also seems to reveal two ratios, 5:1 and 10:1 (ibid., i, pp. 292-93). 8 See the carta of the honour of Skipton (ibid., i, pp. 430-32) where, for example, Osbert the archdeacon held eleven carucates of land unde xiiij carucataefaduntfeodum militis. This characteristic form of words can be traced on many northern honours, including Lancaster (Book of Fees, pp. 206-

104

Colonial England, 1066-1215

tion which has now lost all significance. One, the carta of Richard de La Haye, stands out from all the rest: These are the knights of Richard de La Haye, and they serve in the service of the king by carucates. And five carucates make a knight, and the knights hold them, some more, some less.9

The Red Book version is only a summary of Richard's original carta. However, its idiosyncratic style points to the exceptional and archaic nature of the arrangements it describes. It is without parallel in the surviving returns to the inquest of 1166. This is frail verbal evidence, and it would be easy enough to dismiss it but for one striking coincidence: Richard de La Haye was the son of Muriel, daughter of Colswein of Lincoln, one of the two Englishmen who 'in the whole of England south of the Tees . . . were holding estates of baronial dimensions directly of the king in 1086'.10 Colswein was succeeded by his son Picot c. 1101, and Picot in turn by Robert de La Haye, husband of Muriel and father of Richard, c. 1115. The barony of de La Haye was almost entirely made up of the fee which Colswein held at the time of the Domesday Survey. This was itself a creation of the Conquest. Hence the link is not with a subject of King Edward, but with a vassal of King William, albeit, on the evidence of his name, an English one. Yet another feature of this carta requires close attention. The introductory remarks lead straight into a list of knights: Philip of Kyme andjocelyn of Ingleby owe two knights, Roger of Stixwould, one knight, Richard of Faldingworth, Robert of Hanworth and Roger of Sudbrooke, half a knight, Thomas of Pickworth andjionec of Flintham, two knights, Reginald of Newton and Alan Pescanis, one knight, continued

28), Richmond, Paynel, Warenne, Stuteville, Trussebut, Percy and Tison, EYC, pp. v-xii, subject indices and passim. All this evidence is from the second half of the twelfth century and later. The Skipton cartaonly applies the formula to the new enfeoffment. However, the carucage assessments recorded in the thirteenth and early fourteenth centuries reveal that it was a general formula for all the fees of the honour in Craven except that held by the Bulmers. All but three of the fees of the old enfeoffment lay outside Craven. One of the three was the Bulmer fee, but the other two, held by the Mauleverers and Vavassors, were rated at fourteen carucates to the fee. Since the carucage assessment of many manors of the honour remained constant from Domesday to the early fourteenth century it seems likely that the formula determined enfeoffment on the lands of the honour in Craven from the start. However, it should be noted that the honour was not created until after 1086. The problem is fully discussed by Sir Charles Clay in EYC, vii, p. 4, 42-49, 90-95, 153. 9 'Isti sunt milites Ricardi de Haia, et serviunt ad servitium Regis per carucatas. Et v carucatae faciunt j militem, et milites tenent eas, quidam plus quidam minus', Red Book of the Exchequer, i, pp. 390-91. 10 P.M. Stenton, Anglo-Saxon England (Oxford, 1943), p. 618. For a discussion of the de La Haye pedigree, see J.W.F. Hill, Medieval Lincoln (Cambridge, 1948), pp. 87-88, 95, and for Colswein's estates, ibid., pp. 48-49, 133-34.

The Carta of Richard de La Haye

105

Roger Burnel and Robert son of Julian, two knights, Gerard of Kirkby, Jordan of Ashby and Richard of Croxton, two knights, Walter de Amundeville, half a knight. These are the knights of the old enfeoffment. Richard's cartais once again unusual in recording such a proliferation of multiple tenancies. It seems likely that they arose from the distribution among several tenants of estates which were assessed at five, or multiples of five, carucates. There is an exact parallel in some of the lists of knights which appear in the Red Book of Worcester, where five-hide units were divided in the same fashion among several tenants.11 Of course, examples of fees which were split among two or more tenants are not unusual in the cartae; it may be assumed and sometimes demonstrated that many of these arose from partition among co-heiresses; but in this case six of the eight fees were held thus. There is no reason to think that the tenants of this barony had suffered an unusual loss or lack of sons. Nor is there any suggestion in the later history of these fees that they had undergone partition before 1166; on the contrary, by the thirteenth century some fees had split into separate tenancies so that no trace survived of the associations of 1166.12 Indeed there are signs of such a development by 1166 for the carta concludes: And these are of the new enfeoffment: Hugh Bardulf, two knights, Doun Bardolf, one knight, William de La Launde, one knight, And five knights are on the demesne. The conditions which produced the multiple tenancies of Richard de La Haye's de veteri fees had no effect on those established after 1135. There are reasons then for seeking an archaic territorial basis for the pattern of service revealed in 1166. But it is far from easy to relate the evidence of the carta to the earlier records of Domesday Book and the Lindsey Survey (111518) or to the later evidence of the feudal inquests of 1212 and 1242. Some lands enfeoffed to Richard de La Haye in 1166 were not held by Colswein in 1086 or not by him in chief.13 On many others no undertenant was recorded in the 11

Rollings, The Survival of the Five Hide Unit', p. 456; Red Book of Worcester, ed. Marjory Rollings, pt 4, Worcestershire Historical Society (1950), pp. 414-17, 431. 12 See, for example, the information on the Kyme and Ingelby tenancy in the record of the inquests of 1212 and 1242 (Book of Fees, pp. 192,1066,1073). 13 In 1086 Ingleby was held by Colswein and Wadard of the honour of Odo, bishop of Bayeux; Colswein can scarcely have established a tenancy-in-chief prior to Odo's rebellion in 1088, The Lincolnshire Domesday and the Lindsey Survey, ed. C.W. Foster and T. Longley, Lincoln Record Society, 19 (1924), Domesday, 4/2, cf. Survey 3/4. Colswein apparently had no interest at all in 1086 in the fief which Roger of Stixwould held of Richard de La Haye in 1166; this was the Domesday tenancy in-chief of Waldin the Breton, Lincolnshire Domesday, p. 46; cf. Book of Fees, pp. 169,179,1029,1031.

106

Colonial England, 1066-1215

Survey; apparently they were still held by Colswein in chief;14 and hence there can be no certainty that subsequent enfeoffment comprised all the Domesday holding. In other cases the difficulties are increased by sheer lack of evidence. The 1166 returns provide the solitary record of the half fee held by Walter de Amundeville, steward of the bishop of Lincoln; its location and extent has not been traced.15 Indeed, only three of the undertenancies of 1166 seem to have been established by 1086. The two fees held by Thomas of Pickworth andjionec of Flintham may be identified with lands which William held of Colswein in 1086 in Pickworth, Aisby in Haydour, Hougham and Burton Goggles;16 the two fees held by Roger Burnel and Robert son of Julian included the land which Brunei had held of Colswein in Billingborough and Ouseby in Birthorpe;17 and in the single fee in Newton, Threekingham and Dembleby held by Reginald of Newton and Alan Pescanis, Reginald was preceded by Ralph and Alan by Rainald at the time of the Survey.18 There are further difficulties in the way of a search for neat arithmetical equations. Not one of these fees can be traced intact or unchanged in the inquests of 1212 and 1242.19 In many of these cases the record is incomplete;

14 On the fee of Philip of Kyme and Jocelyn of Ingleby, Alfred held of Colswein at least one carucate of eleven bovates in Scawby and Sturton (Lincolnshire Domesday, 26/16), but no tenants were recorded in the other members of the fee in Ingleby, Faldingworth and Spridlington (ibid., 4/2, 26/12, 25). No tenants were recorded in the lands which later formed the half fee of Richard of FaldingwoVth and his co-tenants in Faldingworth and Sudbrooke although Thorold the priest was a tenant of another member of the fee in Cold Hanworth (ibid., 26/3, 4, 5, 6, 13). Reginald of Newton and Alan Pescanis had predecessors in Ralph who held 16^ bovates of Colswein in Newton and Ouseby in Birthorpe (ibid. 26/40, 41; Book of Fees, pp. 180, 1030, 1073), and in Rainaid, who had one carucate in Dembleby (Lincolnshire Domesday, 26/43; Book of Fees, pp. 180, 1029), but no Domesday tenant was recorded in Threekingham which came to form part of the fee (Lincolnshire Domesday, 26/42; Book of Fees, p. 180). Similarly Gerard of Kirkby and his co-tenants had a predecessor in Walter who held one carucate of Colswein in Marston (LincolnshireDomesday, 26/33), but none apparently in Dry Doddington and Asby de la Launde (ibid., 26/31, 32, 49). 15 Sir Charles Clay noted Walter's tenancy of 1166, but did not trace it subsequently, The Family of Amundeville', Reports and Papers of the Lincolnshire Architectural and Archaeological Society, new series, 3, pt 2 (1948), pp. 109-36. There is no evidence in the returns of 1212 or 1242. 16 Lincolnshire Domesday, 26/35, 47, 48, 53, 54. 17 Ibid., 26/36, 37; Book of Fees, pp. 180, 186, 1072; The Registrum Antiquissimum of the Cathedral Church of Lincoln, ed. C.W. Foster, Lincoln Record Society (1933), ii, p. 27. 18 Lincolnshire Domesday, 26/40-43. 19 The two fees held by Philip of Kyme and Jocelyn of Ingleby appear in 1212 as one fee held by William of Ingleby and two fees held by Simon of Kyme (Book of Fees, p. 192); the Ingleby fee was still recorded in 1242, but the Kyme fee now appeared as 1A fee in Saxilby and Spridlington and 1A fee in Faldingworth which had been subenfeoffed to John of Cadeby and Alice de Neville (ibid., pp. 1066,1073); a further 1A fee in Faldingworth which Robert de Scures held of the honour in 1242 may represent a mesne Kyme interest, then alienated or unrecorded (ibid., p. 1066). The fee of Roger of Stixwould appeared in 1212 as Vi fee in Silk Willoughby and one carucate in Stixwould of which the service was unknown (ibid., pp. 169, 179); in 1242 a 1A fee in Haceby and J/3 fee in Willoughby were recorded (ibid., pp. 1029,1031). Of the Vfcfee of Richard of Faldingworth, Robert of Hanworth and Roger of Sudbrooke, 1A> fee is probably traceable in the land in Sudbrooke

The Carta of Richard de La Haye

107

in others it may be inaccurate; in some the terms of service may have been changed. Hence the obvious line of investigation from the carta of 1166, which gives no information on the location of fees, to the inquests of 1212 and 1242, which do, and thence back to the territorial information of Domesday, provides no easy route to a satisfactory solution of the problem. The information on the size and location of fees is in many cases inadequate. On the other side of the equation the information on territorial assessment is equally insecure. Only a few of the manors which appear in both the Domesday Survey and the Lindsey Survey were rated at the same number of carucates on each occasion.20 This is critical for at the time of Domesday some of the fees named in the carta had yet to be established. Hence it would be hazardous to assume either the Domesday or the Lindsey Survey assessment as a sole basis for calculation; service may have been related to either of them or to some other assessment of which we know nothing. These are real difficulties; if on the one hand they leave small room for arithmetical optimism, it is clear on the other that 'continuity' in this case is not to be disproved by the lack of exact equations of one knight to five carucates. In fact, none of the fees of 1166, which were part of the honour in 1086, survived as units or multiples of five carucates. Yet despite the difficulties, there are fragmentary indications that this equation had determined the pattern of enfeoffment. Faldingworth was rated as two manors, one of two carucates and one of four bovates, in 1086 and as two-and-a-half carucates in 1115-18;21 in 1242, the one occasion on which its service was isolated, it appeared as two quarter

continued

held by Humphrey of Sudbrooke in 1212 (ibid., p. 192) and a further 1A and Vio fee in the lands held by Robert de Scures and Alexander Constantine in 1242 in Faldingworth and Hanworth respectively (ibid., p. 1073). The two fees of Thomas of Pickworth andjionec of Flintham appear as 1V£ fees held by Robert of Pickworth in Pickworth in 1212 (ibid., p. 180), and as Vz fee, less Vfeo, in Pickworth and M> fee in Aisby in 1242 (ibid., p. 1073). The fee of Reginald of Newton and Alan Pescanis appears as V? fee in Newton, Threekingham and elsewhere and V$ in Dembleby in both 1212 and 1242 (ibid., pp. 180,1029, 1030,1073). Of the two fees held by Roger Burnel and Robert son of Julian, two fees were recorded in 1212 in Billingborough and Parva Thorp, but only one in Billingborough in 1242 (ibid., pp. 180, 186, 1072). Of the two fees held by Gerard of Kirkby, Jordan of Ashby and Richard of Croxton, J/2 fee held by Jordan of Ashby is traceable in 1212 and IV* fees in 1242 (ibid., pp. 186,1024,1073). 20 Faldingworth (2l/2 carucates), Reepham (6 bovates) and Hackthorn (3 bovates) bore the same assessment on each occasion (Lincolnshire Domesday, 26/3, 12, 8, 15; Lindsey Survey, 2/11, 3/4). Some manors bore a reduced assessment in 1115-18: Spridlington was assessed at 2 carucates and 5V* bovates in 1086 and at 2 carucates 2 bovates in 1115-18 (LincolnshireDomesday, 26/25; Lindsey Survey, 2/11); Fillingham was assessed at 7 carucates 1 bovate in 1086 and at 6 carucates 6 bovates in 1115-18 (LincolnshireDomesday, 26/24; Lindsey Survey, 2/11); and Brattelby, the caputof the honour was assessed at 3 carucates in 1086 and at 2 carucates 2 bovates in 1115-18 (Lincolnshire Domesday, 26/1; Lindsey Survey, 3/4). Other manors carried an increase, e.g. Riseholme, IV* carucates in 1086 but 3 carucates in 1115-18 (LincolnshireDomesday, 26/2,19; Lindsey Survey, 3/4) and Barlings, 12 bovates in 1086 but 2 carucates 5 bovates in 1115-18 (Lincolnshire Domesday, 26/7, 21; Lindsey Survey, 3/4). 21 Lincolnshire Domesday, 26/3, 12; Lindsey Survey, 3/4.

108

Colonial England, 1066-1215

fees.22 Dembleby was rated as one carucate;23 it appears in both 1212 and 1242 as l/5 fee.24 Marston and Dry Doddingtori totalled two-and-a-half carucates in 1086;25 they passed to the Ashby family which owed half a knight for this fee both in 1212 and 1242.26 Most striking of all is the single knight's fee of Roger of Stixwould. This, the Domesday tenancy-in-chief of Waldin the Breton, which had been held in 1066 by Ulviet, was rated at eleven bovates in Haceby, six in Horbling, fifteen in Silk Willoughby, and one carucate in Stixwould; making a total of five carucates.27 Cold Hanworth is possibly another example; it appears as Vw fee in 1242;28 it was rated at six bovates in 1086, but as four in 1115-18.29 In other cases there is a close approximation to the formula. In 1212 Humphrey of Sudbrooke held seven bovates in Sudbrooke for Vz of a knight.30 Dembleby provided one carucate of the knight's fee of Reginald of Newton and Alan Pescanis;31 the other known members in Newton, Ouseby in Birthorpe and Threekingham totalled thirty 2/3 bovates in 1086, IVs bovates short of the appropriate total.32 The heirs of Roger Burnel held one fee in Billingborough in 1212 and 1242;33 in 1086 Brunei, Colswein's man, held two carucates and seven bovates in Billingborough and two carucates one-and-a-half bovates in Ouseby in Birthorpe.34 One tenant of the barony acknowledged the formula in the clearest terms. In granting land in Ingham, Toft and Newton to Bullington Priory sometime after 1169, Philip of Kyme provided that the convent was to perform forinsec service to the heirs of Richard de La Haye, 'namely a V$ part of a knight for a carucate of land and a fortieth part for a bovate'.35 It was still acknowledged in the inquest of 1212 which recorded that Robert de Karvesale held Hougham as four carucates 'of which five carucates make one knight's service'.36 These of course are selected examples. There is one instance, possibly a later enfeoffment, where a different ratio of service to assessment is stated,37 and 22

Book of Fees, p. 1073. Lincolnshire Domesday, 26/43. 24 Book of Fees, pp. 180, 1029. 25 Lincolnshire Domesday, 26/31, 32, 33. 26 Book of Fees, pp. 186, 1073. 27 Lincolnshire Domesday, 46. 28 Book of Fees, p. 1073. 29 Lincolnshire Domesday, 26/13; Lindsey Survey, 2/11. 30 Book of Fees, p. 192. 31 Lincolnshire Domesday, 26/43. 32 Ibid., 26/40, 41, 42. 33 Book of Fees, pp. 180, 1072. 34 Lincolnshire Domesday, 26/36, 37. 33 Documents Illustrative of the Social and Economic History of the Danelaw, ed. F.M. Stenton, British Academy, Records of Social and Economic History (1920), p. 41. Cf. C. Warren Hollister, The Military Organization of Norman England (Oxford, 1965), pp. 47-48. 36 Book of Fees, p. 186. 37 Reepham was rated at 6 bovates and Vio fee in 1212; compare 1242 (Book of Fees, pp. 192, 1073). 23

The Carta of Richard de La Haye

109

there are other fees or fragments of fees where it is impossible to trace any constant relationship at all.38 Even so, it seems likely that the words of Richard de La Haye's carta recalled a method of enfeoffment which related service to carucates. Since it affected the old enfeoffment alone,39 it must go back to the years before 1135, at the earliest to the lifetime of Colswein, at the latest to that of his son-in-law, Robert de La Haye.40 Indeed since the ratio can be traced, sometimes approximately, sometimes exactly, in the assessments of the Domesday Survey, it probably originated earlier rather than later in this period, in the time of the Conqueror rather than of his sons. It is difficult to avoid the conclusion that this is yet another example of 'continuity'. If so, it is the only known example coming from the Danelaw. Moreover, it is the only known example of such 'continuity' in the service of a secular barony, not only in the Danelaw, but in any part of the Anglo-Norman realm. Finally, it is an example which was already overlain and obscured from view by 1166. It is not therefore so dependent on late twelfth-century evidence as the much more striking example of the bishopric of Worcester.41 But 'continuity' in what sense? It has been assumed that evidence of a close relationship between territorial assessments and the knight's fee establishes a 'strong case - for the notion of institutional continuity'.42 One of the later fees 38

For example, the two fees of Thomas of Pickworth andjionec of Flintham totalled 8V& carucates in 1086 or 10V£ if a further two carucates in Hougham in the soke of Marston are included (LincolnshireDomesday, 26/34,35,47,48,53). But note that despite this the 5:1 ratio survived in the case of Hougham in 1212 (Book of Fees, p. 186). 39 The assessment of the two fees of Hugh Bardolf in Scothorne and Riseholme (Book of Fees, p. 192) cannot be established since in 1086 Scothorne was associated with other manors which did not come to form part of the Bardolf fee (Lincolnshire Domesday, 26/4, 5, 6). However, no possible combination of the Domesday assessments produces the required sum of 10 carucates. Fillingham, the fee of one knight held by Doun Bardolf (Book of Fees, pp. 192,1073), was assessed at 7 carucates 1 bovate in 1086 (Lincolnshire Domesday 26/24) and at 6 carucates 6 bovates in 1115-18 (Lindsey Survey, 2/11); it also included part of Ingham (Book of Fees, p. 1073). Ewerby Thorpe, the fee of one knight held by William de La Launde (Book of Fees, pp. 179,1072), was assessed at 3V& carucates and 1 bovate in 1086 (Lincolnshire Domesday, 26/26). 40 Robert de La Haye was still alive in 1134, RRAN, ii, no. 1895. 41 Much of the Worcester evidence is consistent not so much with the survival of the five hide unit, as Miss Hollings argues ('The Survival of the Five Hide Unit', pp. 453-87), as with an attempt on the part of the bishops of Worcester in the twelfth century to extend the ratio to all their enfeoffments and thus increase the service due to them. The most telling instances of continuity come from the late twelfth and early thirteenth century. I am quite unable to trace Miss Hollings' identification of the old enfeoffment with ilia tenura ubi quinque hide faciunt militem (ibid., p. 482). This phrase is not used and no such specific identification is made in the sections of the Red Book of Worcester to which she refers. The earliest record of the old enfeoffments does not in fact relate service to hidage (Red Book of Worcester, iv, p. 412). The whole problem of continuity is complicated in this case by the fact that much of the information on hidage in the late twelfth-century feodaries in the Red Book was drawn from Domesday Book or copies thereof. Hence genuine continuity and the revival, extension or imposition of an archaic assessment are not readily distinguishable. 42 Hollister, The Military Organization of Norman England, p. 43. Professor Hollister himself indicated some of the difficulties of this assumption: 'On the other hand there was nothing to prevent a tenant-in-chief from establishing some uniform standard of hidage for the fees on his estates. The

110

Colonial England, 1066-1215

of the barony, that of Roger of Stixwould, seems to provide emphatic evidence of such institutional continuity. This comprised four members, apparently carefully arranged so that they totalled five carucates.43 In 1066 the same estate was held by a single man, Ulviet.44 In this instance, therefore, a five-carucate unit which came to comprise a knight's fee can be traced to the time of King Edward. However, this evidence is not entirely secure since Ulviet, or others of the same name, held other estates within the county in 1066;45 hence there can be no certainty that this five-carucate unit of 1086 was not merged in a larger holding before the Conquest. Moreover, this fee was not part of Colswein's lands in 1086; it was held in chief by Waldin the Breton. It is the only such instance among the fees of the de La Haye barony. For the rest, neither the undertenancies recorded in 1086 nor the other fees recorded in 1166 bore any definable relationship to the pattern of tenure which had existed in 1066. Colswein was a great landowner in Domesday but not at the Conquest. His fee was the political pickings, it has been plausibly suggested, of a Quisling.46 He is not recorded as holding any part of his barony in 1066. He may have held some of the manors where no 1066 tenant is named, but these number only seventeen out of the fifty-four holdings which made up the barony; the remaining thirty-seven were held by some thirty landowners all of whom had been depressed or dispossessed by 1086. Thus the correlation between carucates and fees cannot have arisen from an earlier grouping of holdings, for, if this had ever existed, it was broken at the Conquest. There is a further problem. It has been suggested that pre-Conquest military service in the Danelaw was related to the characteristic six-carucate groupings of the Danelaw vills.47 This hypothesis has not been substantiated; it depends on an analogy with the five-hide unit of the Berkshire Domesday; no evidence has been found to demonstrate that a six-carucate ratio was the rule or even that there was any organised correlation of military service and territorial assessment within the Danelaw. The de La Haye evidence now points to a unit of five not six carucates. It does so most convincingly where the possibility of direct institutional continuity is strongest, in the case of the Stixwould fee. Even if this example is set aside it still remains that Colswein arranged carucates in units or fractions of five, not six, and that this arrangement survived as a consciously expressed tradition on his fee. If any arrangements for supplying a warrior from six carucates had ever existed, they were overridden. Hence 'continuity' in this case amounts to this: a tenant-in-chief, whom we CONTINUED

very fact that the Worcester fees happened to correspond in hidage to the pre-Conquest military recruitment units does not prove that they were the products of the same system' (ibid., p. 47). 43 See above, p. 108. 44 Lincolnshire Domesday, p. 46. 45 Ibid., 27/37, 38; 59/17. 46 Hill, Medieval Lincoln, p. 48. 47 Compare Hollister, The Military Organization of Norman England, pp. 15, 24, 39; and his AngloSaxon Military Institutions (Oxford, 1962), p. 51.

The Carta of Richard de La Haye

111

judge from his name to be an Englishman, related the knight-service due from his newly created barony to a system of territorial assessment developed in pre-Conquest England; he arranged it at a ratio of 5 : 1 which was current, if not in the Danelaw, then in some southern and west-midland shires of the old English realm; and he put it into effect despite the tenurial upheaval of the Conquest of which his own barony was a product. Such 'continuity' as there was in this case was not institutional but mental. Colswein did more than simply preserve pre-Conquest institutions. He used them, even developed them in new circumstances for his own purposes. In this, English though he may have been, Colswein was a good Norman. This seems the most likely reconstruction of the evidence. It is one which emphasises that English institutions might survive where there were Englishmen to preserve them. But it is also a cautionary tale. If the 5 :1 ratio determined enfeoffment on Colswein's lands despite the tenurial disturbance created by the Conquest, then the 5 : 1 ratio, wherever it occurs, can no longer be accepted by itself as evidence of tenurial continuity in which Norman knights succeeded directly to the tenements and tenurial responsibilities of Anglo-Saxon thegns. Note Ann Williams, The Knights of Shaftesbury Abbey', Anglo-Norman Studies, 8 (1985), pp. 21437, provides a useful comparison with early evidence of enfeoffment on an ecclesiastical barony.

This page intentionally left blank

8 Politics and Property in Early Medieval England

Between 1066 and the close of the thirteenth century English kings, landowners and lawyers established a sophisticated law of property. This was concerned, among other things, with title, inheritance and succession. The law came to define the relationship between patronage and inheritance, between the power of kings and lords to exercise favour to political or financial advantage, and the desire of their vassals to ensure heritable title to land, privilege and office. How far was patronage, ruthless or beneficent, limited by law? How far was law beaten into shape by political necessity? These questions have lain at the source of a curiously diffuse collection of views expressed by English and American scholars. Professor S.E. Thorne (1959), in an important essay concerned with the concepts of heritability and alienability rather than the practice of inheritance, argued for a gradual process continuing to c. 1200, whereby, through the accumulation of judgements and legal notions about homage and warranty, life interests in fees, which may in practice have passed from father to son, developed into incontrovertible heritable tenures.1 Sydney Painter (1960) suggested that there was a more sudden and decisive move towards the establishment of inheritance c. 1150.2 R.W. Southern (1962) put the critical point earlier, in the reign of Henry I.3 Professor R.H.C. Davis (1964) argued emphatically that the succession of Henry II was the decisive point: The barons.. .demanded that the King [Stephen] should recognize their hereditary right in specific and unambiguous terms. . .That was what the barons fought for in Stephen's reign, and that is what they won.'4 These views should be compared with other assessments of periods bounding these alleged developments. At the latter endjolliffe (1955) saw the crown's 1 S.E. Thorne 'English Feudalism and Estates in Land', Cambridge Law Journal, new series, 6 (1959), pp. 193-209. 2 S. Painter, The Family and the Feudal System in Twelfth-Century England', Speculum, 35 (1960), pp. 1-16; reprinted in his collected essays, Feudalism and Liberty, ed. Fred. A. Cazel, Jr (Baltimore, 1961), pp. 195-219; see especially p. 198. 3 R.W. Southern, 'The Place of Henry I's Reign in English History', Proceedings of the British Academy, 48 (1962), pp. 127-69, especially p. 145. This view is also followed by Marjorie Chibnall, The Ecclesiastical History ofOrderic Vitalis, ii (Oxford, 1969), pp. xxxvi-xxxvii. 4 R.H.C. Davis, 'What Happened in Stephen's Reign?' History, 44 (1964), pp. 1-12, especially pp. 11-12. Professor Davis restated and developed his views in King Stephen (London, 1967); see especially pp. 123-25.

114

Colonial England, 1066-1215

capacity to interfere with right, title and lawful process becoming rapidly more powerful and inimical to aristocratic interests under the Angevin kings until the culmination under John;5 his argument would scarcely substantiate any major surrender by the crown in 1154 or at any time in the twelfth century McFarlane (1965), in his turn, seemed to doubt whether any significant change had occurred even by the end of the thirteenth century, for he maintained that both Henry III and Edward I readily ignored the accepted provisions of the law in discontinuing earldoms or disposing of them in the interests of the crown.6 In this matter, McFarlane's argument implies, not much had been achieved by 1215, let alone 1100 or 1154. At the earlier bound stands Maitland: We are thus led to question whether the followers of the Conqueror who received great gifts of English land held these lands heritably. It is certain that they did;1 but this answer may require qualifications and the difficulty of the question should be seen.8

Some of Maitland's qualifications and difficulties foreshadowed recent discussion.9 Nevertheless his opinion was firm and convinced. Further comments were added by Sir Frank Stenton,10 Professor David Douglas11 and Professor V.H. Galbraith,12 but otherwise the argument remained very much as Maitland left it until the appearance of Professor Thome's work in 1959. There is another more serious difficulty. Painter made a single reference to the studies of Marc Bloch.13 Otherwise all this recent work has by-passed continental scholarship, formidable both in bulk and learning, extending from Guilhiermoz,14 through Bloch,15 to Professor Georges Duby,16 and, with particular reference to Normandy, from Genestal, Lagouelle and others to Lucien Mus5

J.E.A. Jolliffe, Angevin Kingship (London, 1955), pp. 50-86, 131-36. K.B. McFarlane, 'Had Edward I a "Policy" towards the Earls?' History, 50 (1965), pp. 145-59. 7 The italics are mine. 8 F. Pollock and F.W. Maitland, History of English Law (Cambridge, 1911), i, p. 314. 9 Ibid., i, pp. 314-6; ii, pp. 260-313. 10 F.M. Stenton, TheFirst Century of English Feudalism (2nd edn, Oxford, 1961), pp. 33-41, 1546

56.

11 D.C. Douglas, 'A Charter of Enfeoffment under William the Conqueror', EHR, 42 (1927), pp. 245-47. 12 V.H. Galbraith, 'An Episcopal Land-Grant of 1085', EHR, 44 (1929), pp. 353-72. 13 Painter, Feudalism and Liberty, p. 196. 14 A. Guilhiermoz, Essai sur les origines de la noblesse en France au moyen age (Paris, 1902), especially pp. 195-207. 15 M. Bloch, La sodete feodale: la formation des liens de dependence (Paris, 1939), especially pp. 299-304; English edition, trans. G.A. Manyon (London, 1961), pp. 194-97. 16 G. Duby, Le sodete aux Xle et Xlle siecles dans la region mdconnaise, (2nd edn, Paris, 1971), pp. 149-65, 174-201, 235-45, 317-58. See also, idem, 'La noblesse dans la France medievale', Revue historique, 226 (1961), pp. 1-22; idem, 'Au Xlle siecle: les "jeunes" dans le societe aristocratique', AnnaksESC, 19, (1964), pp. 835-46, especially 841; idem, 'Structuresdeparente et noblesse: France du nord, IXe-XIIe siecles', in Miscellanea mediaevalia in memoriamJanFrederikNiermeyer(Groiiingeii, 1967), pp. 149-65; idem, 'The Diffusion of Cultural Patterns in Feudal Society', Past and Present, 39 (1968), pp. 1-10; idem, 'Structures familiales aristocratiques en France du Xle siecle en rapport

Politics and Property in Early Medieval England

115

set, Professor Carabie and Professor Jean Yver.17 These studies have developed a single consistent argument, best summarised by Professor Duby: namely that lignage, with all its concomitants including inheritance, was established for 'the territorial princes and comital families by the middle of the tenth century.. .among the castellan families about the year 1000, and among the ordinary knights some fifty years later'.18 Whether such a timetable is universally applicable may be doubted; Picardie, for example, does not quite conform to it,19 and it is likely that Brittany also may prove exceptional. But for Normandy roughly such a pattern of development was traced independently by Professor Douglas in 1947. He showed that one of the aspects of the rise of the duchy in the eleventh century was the emergence of those aristocratic families, which, subject to the accidents of descent and political fortune, were still in possession of estate and office in 1204.20 It seems inescapable that the Norman aristocracy which settled in England after 1066 was already accustomed to inheritance. The higher the rank the more certain and secure the enjoyment of heritable family property. It might be expected in 1066 that knights were likely to be insecure in their possessions, such as they were, but not the great baronial families of the AngloNorman dominion.21 There are strong reasons for thinking that the Conqueror, however he magnified his new-found royalty and right of conquest, did not, perhaps could not, interfere to any significant degree in these general trends.22 The feudal nomenclature of Normandy and England suggests an early development of hereditary tenure in the duchy, which was transferred without apparent hesitation to the kingdom. In Normandy the word alodium, whatever its sense in other parts of the Continent, meant, not land held free of seigneurial services, but land held by hereditary right.23 Alodium, used in this sense, was a commonly recurrent word in Norman charters of the eleventh century; it was casually

continued

avec les structures de 1'etat', L'Europe aux IXe-XIe siecks, aux otigines des etats nationaux, ed T. Manteuffel and A. Gieysztor (Warsaw, 1968), pp. 57-62. 17 This work is noted in detail below. 18 Duby, The Diffusion of Cultural Patterns in Feudal Society', Past and Present, 39 (1968), p. 6. 19 R. Fossier, La terre et les hommes en Picardie (Louvain, 1968), ii, pp. 511-46. 20 D.C. Douglas 'The Rise of Normandy', Proceedings of the British Academy, (1947), pp. 115-20; idem, William the Conqueror (London, 1964), pp. 83-104. 21 The social variation of tenurial security is discussed further below, pp. 137-42. 22 The argument that William had the power to impose unusually severe terms on his vassals is balanced by the opposite case, that he could not muster adequate support through offering unattractive terms of settlement. Some 'French' settlers, notably Gherbod the Fleming and Aubrey de Coucy, apparently withdrew voluntarily from their English estates. 23 This is fully developed with extensive annotation by R. Carabie, La propriete fonciere dans le Tres ancien droit normand Xle-Xllle siecks, i, La propriete domaniak, Bibliotheque d'histoire du droit normand, second series, etudes, 5 (Caen, 1943), pp. 230-39. See also H. Lagouelle, Essaisur la conception feodak de la propriete fonciere dans le Tres ancien droit normand (Caen, 1902), pp. 118-20, 125-26, 247-49; and L. Musset 'Reflexions sur alodium et sa signification dans les textes normands', Revue historique de droit francais et etranger, fourth series, 47 (1969), p. 606.

116

Colonial England, 1066-1215

accepted by Dudo of St-Quentin, in the second decade of the century, when he stated that Rollo had received Normandy in alodo et infundo.24 Now alodium had the same sense in the eleventh century &$ feodum in the twelfth; the one term was succeeded by the other. Hence the word feodum from its first appearance in Normandy implied a hereditary tenure.25 Both alodium and feodum stood in contrast to the more precarious beneficium; indeed the emergence of the term feodum is probably to be explained by the varied, even contradictory senses which had come to be attached to beneficium.26 It follows that alodium and feodum should be given the same meaning in England. Maitland found that the alodarius of Domesday was a tenant owing seigneurial service and holding by hereditary right.27 He demonstrated that the word feodum had the same hereditary connotation28 and noted that 'we can hardly say for certain that D.B. does not use alodium and feodum as equivalents, both representing a heritable estate, as absolute an ownership of land as is conceivable'.29 He drew a sharp distinction between the hereditary feodum and the non-hereditary beneficium which the fee was supplanting.30 Maitland was cautious. He presented the emergence of the fee as a process which had already 'gone far' by the time of Domesday and noted that 'a trait of precariousness clings to the fee: it is easily forfeitable, and the lord's rights in the land appear in the shape of reliefs and wardships'.31 But essentially he reached the same conclusion for England as Lagouelle, Carabie and Musset for Normandy. The very language of feudalism, from its inception in Norman England, implied 24 DemmilwsetactisprimommNormanniaeducum, ed.J. Lair (Caen, 1865), p. 169. For the interpretation of this much discussed passage see Lagouelle, Essai sur la conception fMale, pp. 85-88; and H. Prentout, Etude critique surDudon de Saint-Qumtin (Caen, 1915), pp. 207-49. 25 Carabie, La proprietefonciere, i, pp. 242-43, 248-50. 26 Ibid., pp. 245-54. Carabie insisted on the precarious sense of beneficium. This is strongly marked in a series of charters of Richard II of Normandy confirming the privileges of Fecamp, Jumieges and St-Ouen in 1025; Recueildes actes des dues de Normandie, 911-1066, ed. Marie Fauroux, Memoires de la societe des antiquaires de Normandie, 36 (1961), nos 34, 36, 53. The significant words in the first of these are: 'we wish.. .to confirm those gifts which our faithful men, with our agreement, have made ofprecaria or benefices which belonged to us, or of the inheritances which belonged to their fathers' ('placuit.. .firmare ea que fideles nostri, nostro consensu, aut precario vel beneficiis quae nostri juris erant, vel de hereditatibus quas paterno jure possidebant, concesserunt...') (no. 34). The other two charters contain closely similar passages. Certainly two and possibly all three were issued at Fecamp. None survive as originals. There are other instances where beneficium is applied to apparently less precarious tenures (ibid., nos 21, 43, 48, 131, 145). It was also used to describe a tenancy held of a church or property belonging to a church, or simply an act of endowment. In some instances the word is capable of more than one interpretation. See the charter of Albert abbot of St-Mesmin de Micy toJumieges of 1023-26: 'I held a certain alod through maternal inheritance, not as a benefice held of anyone [?not through anyone's gift]' ('erat michi quidam alodus ex materna hereditate, non ex alicujus beneficio...') (ibid., no. 51). 27 F.W. Maitland, Domesday Book and Beyond (Cambridge, 1897), pp. 153-54. 28 Ibid., p. 152, especially n. 3. Cp. Pollock and Maitland, History of English Law, i, p. 68. 29 F.W. Maitland, Domesday Book and Beyond, p. 154 n. 1. 30 Ibid., p. 152; Pollock and Maitland, History of English Law, i, pp. 67-68. 31 Ibid., p. 68.

Politics and Property in Early Medieval England

117

inheritance. A non-hereditary fief was a contradiction in terms; it might occur in practice, but it was not what men normally intended when they gave or accepted fiefs.32 The family nomenclature of England and Normandy tells a similar story. One of the more obvious features of the new notion of lignage emerging in the tenth and eleventh centuries was the use of family names, frequently toponymic in form.33 The toponym identified the family with the chief seat of its property. Hence such toponyms provide a very rough and ready minimal measure of the development of inherited estate; indeed the Norman evidence suggests that the two emerged side by side.34 Now the majority of the great Norman families which settled in England used French toponymic names: Aubigny, Beaumont, Bully, Grandmesnil, Mandeville, Montfort, Montgomery, Montbrai, Tosny, Warenne. Patronymics were still common, some of which, like fitz Walter, developed into family names. There were also a few family surnames: Bigod, Malet, Martel, Giffard, Mauduit; and a few tenants-in-chief took English toponyms: Tonbridge, Salisbury, Gloucester, Totnes, Stafford, Berkeley, Lincoln, Essex. But French toponyms were overwhelmingly preponderant among the upper ranks of the new aristocracy. How far the English adventure, by placing the Normans in foreign context, encouraged this usage is at present a matter for conjecture. It is possible that families began to adopt French toponyms after rather than before the Conquest of England. But the most likely explanation of the evidence is that hereditary family property, which underlay such toponyms, had been securely established in Normandy among the great tenants of the duke by the third quarter of the eleventh century.35 There is also direct evidence. Inheritance was the received legal doctrine of Norman England. The Conqueror can scarcely have done other than reinforce a generally accepted convention when he instructed the Londoners that 'every child be his father's heir after his father's day'.36 None of the law books question it; on the contrary they accept it without comment as an established 32 The argument is not defeated by examples of fees which did not ultimately descend to heirs. R.W. Southern, however, has taken a different view on this basis (pp. 161-62n.). For nonhereditary tenancies in the generation following the Conquest see below, pp. 138-40,156-57. 33 Duby, 'Structures de parente et noblesse', pp. 150-59. 34 The following toponyms are attested in the acta of the Norman dukes before 1066: among the greater families, Montgomery, Belleme, Tosny, Beaumont, Ferrieres, Grandmesnil, L'Aigle, Warenne, Montfort, Mandeville, Gournay, Bully, Port, Mowbray, Tracy, Sai, St-John; and among the lesser, Drincurt, Planches, Granville, Le Mesnil, Vernon, Aunou, Limesy and Luvetot (RADN, passim). For references to hereditary tenements in Normandy before 1066 see ibid., nos 12, 34, 42, 53, 61, 93, 119, 120, 122, 125, 129, 130, 167, 173, 220. No. 200, witnessed by Duke William, 1051-66, demonstrates that Roger de Bully held Bully iure hereditario. Nos 34, 73, 74,146 refer to the hereditary possessions of the duke. Nos 147 and 233 refer to claims based on hereditary title. 35 I intend to deal elsewhere with the development of Anglo-Norman family-nomenclature baldly summarised above. See below, pp. 179—96. 36 W. Stubbs, Select Charters, 9th edn, ed. H.W.C. Davis (Oxford, 1921), p. 97; EHD, 1042-1189, p. 945. For comment see Facsimiles of English Royal Writs to AD 1100 Presented to Vivan Hunter Galbraith, ed. TA.M. Bishop and P. Chaplais (Oxford, 1957), no. 15.

118

Colonial England, 1066-1215

principle.37 Furthermore, the charter of liberties of Henry I clearly assumes the whole structure of inheritance as it is later revealed in the records of the twelfth century.38 It assumes that earls and barons are succeeded by their heirs on the payment of a relief; it assumes that heiresses inherit just as certainly as heirs; it accepts the institutions of dower, marriage portion and wardship both of heirs and their land; it emphasises that these relationships are to apply between barons and their men as between barons and the king.39 The subsequent history of the charter is as significant as its content. It lay in diocesan and monastic repositories throughout the land. It came to hand readily enough in the years before 1215 when it provided a model for Magna Carta.40 Yet apparently no one thought it worth reviving earlier in the twelfth century, even in the circumstances of the 'Anarchy'. Inheritance was not in question. The one other formal enactment which is known from the Norman period, a statutum decretum providing for partition among heiresses, was concerned, like the charter of 1100, with the definition of the hereditary principle.41 These are emphatic indications that the inheritance of feudal property was part of the natural order of things in Norrnan England. There is equally certain evidence that even the greatest in the land might suffer dispossession, and his family disinheritance. The problem is to reconcile this apparent contradiction. The first difficulty, as so often, is one. of language. The term 'law' is scarcely to be applied to this period except to denote an assemblage of customs and conventional practices which were still malleable and only slowly setting into hard and fast rules, some moulded by ancient texts or new regulations, others deriving their strength simply from inherent social needs. But if 'law' must remain vague, 'inheritance' cries out for definition. Inheritance by whom, and of what? Were all heirs, direct, collateral, immediate or distant, on the same footing? Was all property, land held in fee or otherwise, castles, office, title, equally heritable? All these questions can be answered. This itself argues for the existence of a complex system of inheritance. The critical question is inheritance by whom? Was there any law of descent and, if so, what? For Normandy the outline of an answer is clear enough. The Pays de Caux adopted absolute primogeniture; the rest of Normandy, partition among sons, who succeeded to their portion in order of seniority.42 Genestal 37 'Leges Henrici Primi', cap. 70, sections 18, 20, 20a, 20b, 21, ed. F. Liebermann, Die Gesetze tier Angelsachsen (Halle, 1903), i, p. 589; ed. and trans. LJ. Downer (Oxford, 1972), p. 225. 38 Stubbs, Select Charters (9th edn), pp. 116-19; EHD, 1042-1189, pp. 400-2. 39 Caps. 2, 3, 4. 40 For the distribution of the charter of Henry I see F. Liebermann, The Text of Henry I's Coronation Charter', Transactions of the Royal Historical Society, new series, 8 (1894), pp. 21-48. For its political importance in the late twelfth century, seeJ.C. Holt, Magna Carta (Cambridge, 1965), pp. 135-41, 150-51. 41 Stenton, TheFirst Cmtury of English Feudalism, pp. 38-41, 260-61. The 'decree' is mentioned in a charter of Roger de Valognes of c. 1145. See below, pp. 252-56. 42 R. Genestal, 'La formation due droit d'ainesse dans le coutume de Normandie', Normannia, 1 (1928), pp. 157-79; J. \Ver, 'Les caracteres originaux du groupe de coutumes de 1'ouest de la

Politics and Property in Early Medieval England

119

maintained that primogeniture was the older of the two systems, but it is more likely that they developed in parallel, each subject to different external influences from neighbouring provinces.43 Each system came to be qualified, the primogeniture of the Pays de Caux by the slow acceptance of a provision for younger sons, partition in the rest of Normandy by parage which helped .to maintain the unity of the family property.44 Practice was also determined by a constitutio, usually attributed to Henry II, which forbad partition when it led to the division of a fief; henceforth feudal property was divided among male heirs only when there was more than one fief.45 There was an even more important restriction, namely that there could be no partition of baronies, fiefs and sergeanties held of the duke. In the Tres anden coutumier vt c. 1199, this appears not as part of the constitutiobut as a separate, and almost certainly older, regulation.46 In Normandy the division of a tenancy-in-chief among sons was rare by the end of the eleventh century so that younger sons could only participate in the inheritance if there was more than one tenement.47 In England Glanvill continued

France', Revue historiquededroitfranfais et Stranger, fourth series, 30 (1952), pp. 18-79, especially pp. 41-47. 43 Genestal argued that primogeniture was earlier on two main counts: first, that it determined the early descent of the duchy; and, secondly, that the Normans introduced primogeniture into England. The first point fails to give due weight to the principle of the indivisibility of the fief which might have the same effect as primogeniture. Genestal himself noted that the custom of Caux was related to that of neighbouring provinces of northern France, whereas that of the rest of Normandy marched with the customs of western provinces. \Ver, who further developed this line of approach to the problem, was cautious about Genestal's argument (ibid., p. 46 n. 2). The second point, of course, begs a question which is part of the present argument. 44 On the provision for younger sons in the Pays de Caux see Genestal, 'La formation du droit d'ainesse', pp. 174-77. On parage see Appendix 1, below p. 148. 45 Tres anden coutumier, cap. viii, Coutumiers de Normandie, i, EJ. Tardif, Societe de 1'histoire de Normandie (1881), pp. 8-9. The attribution to Henry II has not been substantiated; it depends largely on analogy with the assize of 1185 whereby Geoffrey, duke of Brittany, confirmed primogeniture in Brittany and likewise forbad the division of fiefs: R. Genestal, La parage normand, Bibliotheque d'histoire du droit normand, second series, etudes, 1, pt 2 (Caen, 1911), pp. 1-2; Vver, 'Les caracteres originaux', p. 46 n. 2. 46 Tres anden coutumier, cap. viii, 5, ed. Tardif, p. 9. The Summa de legibus also includes comtes: cap. xxiv, I, Coutumiers de Normandie, iii, ed. EJ. Tardif, Societe de 1'histoire de Normandie (1896), p. 79. 47 The succession to Tancred de Hauteville provides the most famous early example of the maintenance of the unity of a tenancy-in-chief. However, the circumstances are far from certain. According to Orderic Vitalis the decision was Tancred's own and he implies that Tancred had a choice in the matter (Orderic Vitalis, ed. Chibnall, ii, pp. 98-100). Malaterra in contrast states that Tancred's elder sons decided to emigrate since the family estate was clearly insufficient for their father's numerous progeny, Geoffrey Malaterra, De rebus gestis Rogmi Calabriae et Sidliae camitis, ed. E. Pontieri, Rerum italicarum scriptores, v, pt 1 (Bologna, 1924), p. 9, and certainly the first Hautevilles to arrive in Italy in 1037-38 were the two eldest, William and Drogo. Geoffrey, Tancred's fourth son by his first marriage, succeeded to the Norman patrimony (Orderic Vitalis, ed. Chibnall, ii, p. 98). Robert I de Grandmesnil is said by Orderic to have divided his lands among his sons at his death in c. 1040 (ed. Chibnall, ii, p. 40). It may be that such option was still possible at this early stage, as Genestal suggested ('Droit d'ainesse', pp. 173-74). Some examples should be used warily.

120

Colonial England, 1066-1215

did not even consider multiple tenements; for him all properties held by military service were a single estate subject to primogeniture.48 Here also the subdivision of a barony among sons was unusual even in the generation immediately following the Conquest and later examples are exceptional.49 This rule was softened by another. Both in England and Normandy the law drew an apparently sharp distinction between inheritance and acquisition. The law of descent, whatever it might be, applied to inherited feudal property; acquisitions, in the form of conquests or purchases, were at their owner's disposal. In fact they were often used to endow younger sons, and very soon legal thinking hardened into the assumption that this was proper custom. Glanvill drew on this distinction,50 but on the whole in England it represents a passing phase. In Normandy it appears in a developed form in the Tres ancien coutumier and it became axiomatic in Norman law.51 Its first official appearance is much earlier,

continued

The descent of the estates of Giroie also illustrates division, but these came to Giroie in right of his betrothed through confirmation by the duke. They were therefore strictly an acquisition, not an inheritance (Orderic Vitalis, ed. Chibnall, ii, pp. 22, 28; cp. below pp. 121-24). The division of the estates of the first house of Belleme between \Ves and William Talvas involved two distinct French and Norman tenements: G.H. White, The First House of Belleme', Transactions of the Royal Historical Society, fourth series, 22 (1940), pp. 67-99; cp. Orderic Vitalis, ed. Chibnall, ii pp. 362-65. The division of the estates of the Paynel family (1151-53) also involved two separate tenements, Les Moutiers-Herbert and Hambye: EYQ vi, pp. 1-2, 97-98. 18 Glanvill, De legibus, vii, 3, ed. Hall, p. 75. 49 William de Rames divided his barony of Rayne between his two sons c. 1130, J.H. Round, Geoffrey de Mandevilk (London, 1892), pp. 399-404, and Hardwin de Scales made a similar division of his barony of Caxton after 1086, Curia Regis Rolls, v, pp. 139-40. The division of the English estates of William Paynel between his sons Fulk and Hugh, c. 1151-53, provides a late example, but it was complicated by the intrusion of Robert de Gant, husband of their half-sister (EYQ vi, pp. 7, 32-33, 97-98; on their Norman estates see above n. 47). Painter's example of the 'division' of the barony of Weldon between Geoffrey Ridel and Ralph Basset was not a division but the creation of a large mesne tenure for a younger son: S. Painter The Family and the Feudal System, p. 201; The Red Book of the Exchequer, ed. Hubert Hall, Rolls Series (London, 1896), i, pp. 329-31; Documents Illustrative of the Social and Economic History oftheDanelaw, ed. F.M. Stenton (London, 1921), no. 458. For a similar example see the agreement between Geldwin son of Savaric and Savaric, his brother, of 1156-58: Sir Christopher Hatton's Book of Seals, ed. Lewis C. Loyd and Doris M. Stenton (Oxford, 1950), no. 434. 49 Whether any of the above partitions arose from twinning it is impossible to say. Primogeniture was applicable to twins and took effect in the one well substantiated case of the succession to Robert, count of Meulan (see Appendix 2, note C, below, pp. 152-54). Approximately 0.5 per cent of male adults would have surviving male twins. This calculation is based on the figures given in L. Perirose, Outline of Human Gmetics (London, 1959), pp. 87-88, and H.H. Newman, Twns and SuperTwins (London, 1942), pp. 33-37. It may simply be coincidental that this matches the known cases of partition. 50 See his discussion of inheritance and alienability. Glanvill, De kgi,bus, does not place an absolute restriction on the alienation of the inheritance or allow absolute freedom to dispose of the acquisition, but he allows a free hand over the latter to those who hold both inheritance and acquisitions, cap. vii, i, pp. 70-71. For further discussion of this see below, pp. 146-47. 51 The first-born knight shall have the knight's fee complete, and it shall not be divided. The rest shall share the acquisitions equally' ('Miles primogenitus feodum lorice integrum habebit, et

Politics and Property in Early Medieval England

121

in the Leges Henrici Primi, where it retains its more primitive form of free parental disposition: The ancestral fee of the father is to go the first-born son; but he may give his purchases or later acquisitions to whomsoever he prefers.52

In the generation after 1066 there was no shortage of acquisitions. Few younger sons had to go a-begging. The distinction was not as sharply cut as seems at first sight. There were marginal situations where inheritance and acquisition were confused or only indistinctly separated. Land acquired by marriage, whether as the wife's portion or inheritance, could be used to endow younger sons as if it were an acquisition.53 Land obtained by collateral successions came to be regarded as acquisition in mature Norman custom.54 It is probable that these and similar problems were settled ad hoc in the late eleventh and early twelfth centuries. By the time of Glanvill the divergence of English and Norman law, in such matters as continued

non partietur, ceteri vero escaeta habebunt equaliter'), cap. viii, 2, ed. Tardif, p. 8. The sense of escaeta has been debated. In the thirteenth century and earlier it generally denoted acquisitions, including those held by military service, but it could also mean non-military tenures: the two categories frequently overlapped: Genestal, Leparage normtmd, pp. 7-8; cp. H. Navel, 'Recherches sur les institutions feodales en Normandie', Bulletin de In societe des antiquaires de Normandie, 51 (1948-51), p. 20. 52 'Primum paths feodum primogenitus filius habeat: emptiones vero vel deinceps acquisiciones suas det cui magis velit' ('Leges Henrici Primi', cap. 70, 21; ed. Liebermann, Gezetze, i, p. 589; ed. Downer, p. 224). This important passage was discussed by Pollock and Maitland (History of English Law, ii, p. 268) but has been overlooked by many subsequent scholars. 53 The allocation of Eleanor of Aquitaine's inheritance to Richard the Lion-Heart is the most obvious example. Geoffrey of Anjou is said to have made a similar distinction between his own and his wife's inheritance (see below, p. 126). Godfrey de Bouillon, second son of Count Eustace II of Boulogne and Ida, heiress of Godfrey, duke of Lower Lorraine, was heir to his mother's inheritance: S. Runciman, History of the Crusades (Cambridge, 1951), i, pp. 145-46; J.H. Round, Studies in Peerage and Family History (London, 1907), p. 152. The honour of Belvoir, which came to Roger Bigod (d. 1107), either as the inheritance or marriage-portion of his wife, Adelicia de Tosny, did not descend to his son, but was used in turn as a marriage-portion for his daughter, Cecily, who married William I de Albini Brito: EYC, i, pp. 461, 466. When William de Warenne died in 1088 his English and Norman lands went to his elder, and his Flemish lands to his younger son. Practically all his property was acquired, but the Flemish estates probably came through his marriage to Gundreda, sister of Gherbod the Fleming: ibid., viii, pp. 1-7, 44. 54 The ordinary rules of succession applied to collaterals: Summa de legibus, caps, xxiii, xxiv, ed. Tardif, pp. 72-79. For an English example in which land acquired by collateral succession was alleged to count as an acquisition see the succession to Hugh Bigod, Appendix 2, note B, below, pp. 151-52. Compare the descent of the estates of Stephen, lord of the honour of Richmond (d. 1135-36). Stephen was heir to his eldest brother Geoffrey Boterel I (d. 1093) or Geoffrey's son Conan (d. 1098) in his Breton property, and to less senior brothers, Alan Rufus and Alan Niger, in Richmond. The Breton property had belonged to his father, Eudo (d. c. 1079) but the English lands had been acquired by Alan Rufus. Stephen inherited both properties but was heir to his father in the Breton property and to his brother in the English. The Breton lands descended to his eldest son, the English to a younger: EYC, iv, pp. 84-88.

122

Colonial England, 1066-1215

indivisibility, points to the emergence of hard and fast rules which determined actions in the courts.55 But this was not necessarily so earlier. Moreover, when the distinction first appears it was permissive, not compulsory. It allowed, but did not oblige, the father to divert acquisitions away from the eldest son, and then not necessarily to his younger children. Hence it permitted a wide variety of post obitum&nd testamentary dispositions.56 Most important of all, it was limited in its application, for, whatever the precise variant, it was effective only if acquisitions and younger sons or other suitable recipients coincided. A single heir united inheritance and acquisition and once that was done the result was irreversible. The father's acquisition became the son's patrimony. Of its nature the rule could be applied only to one succession and one generation. So much for theory. The principles involved were certainly current in northern and western France at the time of the conquest of England.57 Some AngloNorman descents conformed strictly to the letter. Where there was more than one son in the generation after the Conquest, the inheritance went to the elder and the acquisition to the younger. This holds good for fitz Osbern,58 Montgomery,59 Grandmesnil,60 Clare,61 the senior branch of Beaumont,62 and Montfort.63 In these cases it followed that the Norman property remained with the senior branch and the English went to the junior. Usually the division extended only to the two elder sons; any others seem to have been expected to provide for themselves or find a career in the church.64 But not all families 55

For an example see Appendix 2, note B, below pp. 151-52. M.M. Sheehan, The Will in Medieval England, Pontifical Institute of Medieval Studies, Studies and Texts, 6 (Toronto, 1963), pp. 110-19. This work does not place the Anglo-Norman post obitum gift or bequest of land in the context of the pertaining rules of succession. However, it presents no obstacle to the present analysis; indeed it largely dovetails into it. See also below, pp. 151-52. 57 See the agreement made by Tescelinus, priest of Verri, with the monks of St-Florent-lesSaumur, arranging for his entry into the monastery. This provided for an annual payment of 10s. during life plus a post obitum gift of all his property 'without share of anyone, because I have acquired all I hold through my own labours and initiative and therefore do not recognise therein any share either of relative or friend, but leave it all complete to St-Florent and to my lord, Abbot Sigo, and the monks of that house to be held for ever' ('sine parte alicujus, quia omnia quae habeo laboravi et ex ingenio meo acquisivi, idcirco amici aut parentis in hoc partem non recognosco, sed omnia ex integro beato Florentio ac domino meo abbati Sigoni et monachis istius cenobii relinquo habenda in perpetuo') (Liber niger of St-Florent-les-Saumur, BN, MS, nouv. acq. lat., 1930, fo. 113v). The date limits are 1055-70. 58 Complete Peer-age, vi, p. 449; vii, p. 530, n. (b). 59 J.F.A. Mason, 'Roger de Montgomery and his Sons, 1067-1102', Transactions of the Royal Historical Society, fifth series, 13 (1963), pp. 13-18. 60 Orderic Vitalis, ed. Chibnall, iv, pp. 338-39. 61 Complete Peerage, iii, pp. 242-43; Orderic Vitalis, ed. Chibnall, iv, p. 211. 62 See Appendix 2, note C, below pp. 152-54. 63 The Domesday Monachorum of Christ Church, Canterbury, ed. D.C. Douglas (London, 1944), pp.-67-70. 64 Eustace of Boulogne made no provision for his third surviving son, Baldwin, subsequently count of Edessa and king of Jerusalem. On the death of Roger of Montgomery in 1095, Robert of Belleme got his Norman possessions and Hugh of Montgomery the English. Roger and Arnulf already enjoyed wide possessions in north-west England and Pembroke respectively, but, as Orderic 56

Politics and Property in Early Medieval England

123

reacted so. In a significant number of instances the Norman patrimony and the English acquisition descended as a single inheritance. In a few of these, Giffard and Chester for example, it seems that this was because there was only one son at the critical succession, but this is difficult to establish with certainty.65 In others, Malet,66 Mandeville,67 Vere,68 Boulogne69 and Bigod,70 there was more than one son at the death of the first of the Anglo-Norman line, and yet in these families the eldest succeeded to the English acquisition and in some to the lordship of the whole estate. In the case of Boulogne this in all probability represented a deliberate bid to unite cross-Channel holdings. Some families, most strikingly the Beaumonts, are sui generis, since their estates seem to have descended by special settlement.71 Hence there is no doubt that families exercised some control. The matter was not simply settled for them accidentally by the number of surviving legitimate sons at the critical succession. This creates a difficulty. In default of a full survey of the pattern of succession in the first generation after the settlement of England it is not easy to determine whether the differences illustrated above should be explained as variants within or breaches of the rule stated in the Leges Henrici Primi. If the latter, there was clearly no rule at all. However, the former is much the more likely. There are convincing reasons for thinking that most of the apparently divergent successions arose from the parental disposition which the primitive continued

Vitalis noted, no provision was made for Philip and Everard (ed. Chibnall, iv, p. 303) and Robert succeeded to Hugh's English holdings when he died without heirs (Mason, 'Roger de Montgomery and His Sons', p. 20). The Grandmesnil division was also limited to the two elder sons and excluded two other survivors, one of whom had acquired estate by marriage in Apulia (Orderic Vitalis, ed. Chibnall, iv, p. 339). For a similar limitation in the Beaumont descent see the settlement of 1107 discussed below, Appendix 2, note C, below pp. 152-54. William the Conqueror made no territorial provision for Henry, according to Orderic Vitalis simply leaving him 5000 Ibs of silver (ed. Chibnall, iv, p. 95). Geoffrey of Anjou apparently made a division between Henry and his second son, Geoffrey (see below p. 126), but nothing was done for his third surviving son, William. Henry II made no initial provision for John Lackland. In all this the notion that tenements should not be subdivided was probably a decisive factor. 65 Complete Peerage, ii, pp. 386-87; ibid., iii, pp. 164-66. In both these cases the heirs were minors. The matter is debatable because in the last resort there is no way of proving that there were no other legitimate children except where the heir was born after the father's death. 66 DNB, s.n. William Malet. 67 Compkte Peerage, v, pp. 113-14. 68 Ibid., x, pp. 193-96. 69 Eustace II of Boulogne was succeeded both in Boulogne and in his English acquisitions by his eldest son Eustace III. However, by the time of Eustace II's death in c. 1093-96, his second son, Godfrey, was already established in Lower Lorraine. See above, p. 121 n. 53. 70 Roger Bigod was succeeded in his English acquisitions by William, who died in the White Ship in 1120. William in turn was succeeded by Hugh, first earl of Norfolk. The evidence is not perhaps wholly conclusive but it is generally accepted that William and Hugh were half-brothers, William being the elder. Complete Peerage, ix, pp. 578-79; W. Dugdale, Monasticon anglicanum, new edn, ed. J. Caley et al. (London, 1817-30), v, p. 148. The genealogy given in EYC, i, p. 461, is inaccurate at this point; cp. ibid., p. 466. 71 See Appendix 2, note C, below pp. 152-54.

124

Colonial England, 1066-1215

form of the rule and the post obitum gift allowed. There is independent evidence that the father might determine the pattern of succession. Some of this depends on Orderic Vitalis,72 but Orderic is not alone. Geoffrey de Mandeville's grant of Eye (Middlesex) to Westminster Abbey stated: I have given this to God and St. Peter with the consent of my wife Lethselina and by the concession of my son William, whom I have designated as my heir.73

William succeeded to the English acquisitions of the Mandevilles; Geoffrey's charters make it clear that there was at least one other son.74 It also seems probable that families did not act at random. Where acquisition and inheritance were separated, the inheritance was often large so that the Norman patrimony still constituted a fair share for the eldest son as against the English acquisition. This was so, for example, for fitz Osbern, Montgomery, Grandmesnil and Montfort. In contrast, where the eldest son succeeded to the acquisition with or without the patrimony, the acquisition might exceed many times the old Norman patrimony. This was true, for example, of Bigod, Vere and Lacy.75 By the end of the twelfth century Norman custom allowed the eldest son to choose the acquisition rather than the inheritance if he wished. This rule is first recorded in the Tres ancien coutumierbut not as a new provision;76 it is likely that it arose from the circumstances of the Conquest in which the acquisition frequently exceeded the inheritance. It was apparently applied in the succession to Henry, earl of Warwick in 1119.77 It seems probable therefore that the Conquest, far from leading to a rejection of these notions, contributed to their later formulation. Furthermore, even where the descent of property was determined by a series of complicated settlements, the original patrimony tended to descend by primogeniture. It was still Waleran, count of Meulan, the eldest son of the eldest son of Roger de Beaumont, who held the ancestral estates of Beaumont-leRoger and Pont-Audemer in the twelfth century.78 Finally, direct contraventions of this slowly hardening system of descent seem to have been rare. Henry I intervened in the descent of the barony of Marshwood to enfeoff the son of a second marriage on the ground that 'he was a better knight', but this was done pervoluntatem regis; in 1206 the decision was reversed and the descendant of the 72 For Hauteville see Orderic Vitalis, ed. Chibnall, ii, p. 98; for Beaumont and Grandmesnil, ibid., iv, pp. 302-4, 338. 73 'Et hoc donum Deo et sancto Petro cum uxore mea Lethselina concessione filii mei Willelmi quern michi heredem facere disposui, quos eciam hujus elemosine participes fieri per omnia volo ... presentavi', Complete Peerage, v, p. 114 n. (c). 74 Ibid., v, pp. 113-14; cp. Pipe Roll Society, new series, 36 (1960), pp. 105-8. 75 The Bigods and Lacys were tenants of Odo of Bayeux, the Veres of the bishops of Coutances and probably the count of Brittany. See L.C. Loyd, The Origins of Some Anglo-Norman Families, Harleian Society, 103 (1951), s.n.; and for Vere, Complete Peerage, x, p. 193. 76 Tres ancien coutumier, ed. Tardif, cap. viii, 2, 3, p. 8. 77 See Appendix 2, note C, below, pp. 152-54. 78 Ibid.

Politics and Property in Early Medieval England

125

elder line was recognised as the lawful heir.79 In 1162-66 Henry and Sewall, grandsons of Saswalo, tenant of the Ferrers fee, decided to reverse their order of seniority by record of final concord in their lord's court; Sewall and Fulcher son of Henry confirmed the arrangement in like manner thirty years later.80 The ruling house was equally subject to these rules. A proper distinction between public and private law was still in the distant future. Even after 1199, when Roman concepts of iura coronaewere taking shape, John's succession, the casus regis, determined the competing claims of uncle and nephew in private actions for two generations.81 Earlier the options available within the distinction between inheritance and acquisition as it appears in the 'Leges Henrici Primi', could easily accommodate other ideas. The king/duke could designate his eldest son as his successor in the Carolingian style and direct to him both the inheritance and acquisition, but none of this infringed convention.82 What he could not do was to prevent his eldest son from succeeding to the patrimony. The Conqueror, or later Henry II, might threaten this, even pronounce sentence of disinheritance on a rebellious eldest son, but it never came to much. The eldest son's claim on the patrimony, and the resulting potential patronage which he could wield, were always too strong. The Leges Henrici Primi allowed that a father, if abandoned by his son in mortal need, illness or poverty, might adopt any who aided him, relative or stranger, as his heir, but the matter was not automatic, for the competing claims of the immediate heir and the newcomer had to be settled by the judgement of wise men.83 In 1087, as the Conqueror lay on his deathbed at Caen, the wise men advised in favour of Robert.84 The two successions where there was more than one surviving son, to the Conqueror and Henry II, reveal the rules operating in conventional form: the patrimony to the eldest, the acquisition to the younger.85 Normandy and England, separable as inheritance and acquisition in 1087, became a single patrimony after 1135; England/Normandy and Maine/Anjou, separable under Geoffrey of Anjou, became a single inheritance under Henry II.86 The Norman/Angevin dominions and the lands of Eleanor of Aquitaine, separable under Henry II, were treated 79

HMC, Wells, i, pp. 527-28; Pipe Roll 10John, Pipe Roll Society, new series, 23 (1945), p. 113; Stenton, The First Century of English Feudalism, pp. 37-38. 80 Ibid., pp. 52-54, 263-64. 81 T.F.T. Plucknett, A Concise History of the Common Law (4th edn, London, 1948), pp.. 678-80. 82 John Le Patourel, The Norman Succession, 996-1135', EHR, 86 (1971), pp. 225-30; idem, Normandy and England, 1066-1144, Stenton Lecture, University of Reading (1971), pp. 4-9. 83 Caps. 88,15 (ed. Liebermann, Gesetze, i, p. 604; ed. Downer, pp. 274-76). 84 The monk of Caen, in William of Jumieges, Gesta Normannorum ducum, ed. J. Marx (Rouen, 1914), pp. 145-47. 85 Le Patourel took a different view. See Appendix 2, note A, below pp. 149-51. 86 The pattern may have been broken in 1187, when, according to Gerald of Wales, Henry planned to give all his French possessions except Normandy to John, leaving Richard with the old Anglo-Norman dominion: Opera, viii, ed. G.F. Warner, Rolls Series (London, 1891), p. 232. The story may simply derive from a tactical manoeuvre by Henry or Philip Augustus. It is not supported in the Gesta Henrici. In 1183, after the Young King's death, Henry had intended Aquitaine alone for John, ibid., i, pp. 308, 311.

126

Colonial England, 1066-1215

as a single inheritance after 1189. Henry II refused to acknowledge the division which his father Geoffrey of Anjou apparently made on his death-bed,87 and after the death of the Young King in 1183 Richard the Lion Heart refused to 'move up one' and make way for John Lackland in the maternal succession,8 but otherwise the settlements of the ruling house executed the accepted division between inheritance and acquisition. The idea still survived in the thirteenth century. Gerald of Wales exhorted King John to pay due attention to Ireland so that it might provide a kingdom for a younger son.89 It was largely from acquisitions, conquests and escheats that the appanages of both the English and French royal houses were formed under Henry III and Louis VIII. The distinction between inheritance and acquisition was remarkably convenient for an aggressive and burgeoning aristocracy. At one and the same time it resolved the problems created by the accumulation of both estates and sons. For it to work effectively there had to be a steady flow of acquisitions: enough to provide for the family; not so much as to upset the balance between acquisition and ancestral lands. The conquest of England upset this balance. In 1087 the eldest son succeeded to the duchy while the younger acquired a kingdom. In the quarrel among the Conqueror's heirs families found themselves in one of three situations. Either the feoffee of the Conqueror was still alive; or he had been succeeded by a single heir, thus establishing a united Anglo-Norman patrimony; or he was dead and his land divided, the Norman patrimony usually going to the eldest son, the English acquisition to the younger. The first two circumstances encouraged rebellion. Men could not serve two conflicting lords; lay barons and great ecclesiastical lords alike had perforce to choose sides, if only because they had an obvious interest in reestablishing the single lordship which William's death had divided.90 The third circumstance was even more dangerous 87 According to William of Newburgh, Geoffrey of Anjou provided on his death-bed (1151) that Henry should hold Normandy, Maine and Anjou until England was conquered; once that was achieved the maternal inheritance of England and Normandy was to go to Henry and the paternal inheritance of Maine and Anjou to his younger brother, Geoffrey. Meanwhile Geoffrey was to have the three castles of Chinon, Loudun and Mirebeau. This reversal of the usual order, giving the marital acquisition to the elder brother, recognised the fact that Henry was already involved in the conquest of England and in 1150 had succeeded his father as duke of Normandy. Henry did not execute the settlement and apparently obtained papal dispensation from his oath to it in 1156. Chronicles of the Reigns of Stephen, Henry II and Richard I, ed. R. Hewlett, Rolls Series (London, 188490), i, pp. 112-14; J. Boussard, Le comte d'Anjou sous Henri Plantagenet et ses pis, 1151-1204 (Paris, 1938), pp. 68ff. Idem, Le gouvernement d'Henri II Plantagenet (Paris, 1956), pp. 8-11, 408-10. A fragmentary Angevin chronicle lends some support to Newburgh's account, Chroniques des comtes d'Anjou, ed. L. Halphen and R. Poupardin (Paris, 1913), pp. 251-52. It should be noted that Maine was an anomaly in this respect since it was part of both the Angevin and the Norman inheritance. 88 Gesta Henna, i, pp. 308-311. 89 Opera, ed.J.F. Dimock, Rolls Series (London, 1858), v, p. 407. 90 See the speech which Orderic puts in the mouths of some of them: 'How can we give proper service to two mutually hostile and distant lords? If we serve Duke Robert of Normandy properly we shall offend his brother William, and he will deprive us of our revenues and honours in England. On the other hand if we obey King William, Duke Robert will deprive us of our patrimonies in Normandy': Orderic Vitalis, ed. Chibnall, iv, p. 122.

Politics and Property in Early Medieval England

127

to the ruling house because it was more complicated and more enduring. To deprive a tenant for treason often did no more than drive him into the protection of a brother, uncle or cousin across the Channel. If he were replaced in his estates, he still nursed a claim to them, a claim which was carefully tended by his heirs, direct or indirect, and strengthened by the political weight of the collateral branch of the family still undisturbed in its possessions. This was the substance of the tenurial and political crisis of the Anglo-Norman period. Rules of succession were applied in political circumstances quite unsuited to them. The division between England and Normandy compelled the ruling house to intervene in title and succession. But such intervention could also be acceptable, even essential, in other circumstances. Title to an acquisition was not so strong as title to ancestral possesions. The Conqueror was able to readjust the initial acquisitions of the Normans in England without any apparent difficulty or friction; exchanges were made and compensation provided when a new tenant-in-chief was accommodated on land already occupied;91 and it was still possible under Henry I for Ranulf le Meschin to exchange one acquisition, the lordship of Carlisle, for another, the earldom of Chester.92 Moreover, succession was not always simple. There were many circumstances involving twins, or claims through heiresses, or collateral title, or claims in the half blood, in which there was no consistent practice, let alone established custom. The more distant or divided the succession the more likely it was to be settled by an ad hoc arrangement. Such arrangements required the approval of the king/duke; many he no doubt instigated.93 There were therefore many situations in which the ruling house might intrude. Instrusion for political reasons was all the easier. Such intervention had to contend with a fundamental principle of title and

91

The best example is provided by the Sussex baronies. See J.F.A. Mason, William the First and the Sussex Rapes (Hastings, 1966), especially pp. 13-17. 92 See Appendix 2, note D, below, pp. 154-55. 93 See Henry I's confirmation c. 1107 of the division between the Beaumont twins (RRAN, ii, no. 843), the agreement took effect in 1118; Henry I's grant (1121) of Sibilia, daughter of Bernard of Neufmarche, to Miles of Gloucester, along with Bernard's acquisitions: 'I give and concede this to him as Bernard's purchase which Bernard has restored to me, and this at the request of Bernard, his wife and his barons' ('Et hoc ei dono et concedo sicut emptionem Beornardi quam mini reddidit et hoc requisitione ipsius Beornardi et uxoris suae et baronum suorum'), J.H. Round, Ancient Charters Prior to AD 1200, Pipe Roll Society, 10 (1888), no. 6; Henry's confirmation (1123) of the arrangements for the succession to the lands of Geoffrey Ridel, which passed by marriage to the Bassets (Stenton, First Century ofEnglishFeudalism, pp. 34-36,259-60; RRAN, ii, no. 1389), subsequently confirmed by Matilda and Henry of Anjou (ibid., iii, nos 43, 44); Stephen's confirmation (c. 1137) of the marriage arrangements between Roger, son of Miles of Gloucester, and Cecily, daughter of Pain fitz John, which is remarkable for its careful distinction between inheritance, acquisition and maritagium and its enumeration of the last two (ibid, iii, no. 312); Henry of Anjou's confirmation of the division of the Paynel estates (1151-53), EYC, vi, pp. 96-97; letters of Robert de Caux to Henry II sent when on the point of departing for the Crusade (1188), attorning his nephew to take charge of his land and making him heir to his estate, J.H. Round, Calendar of Documents Preserved in France (London, 1899), no. 277.

128

Colonial England, 1066-1215

succession: property law was family law.94 The distinction between inheritance and acquisition was one of several means whereby the family sought both to maintain the unity of its property and to make some provision for cadet branches. Hence the family retained an interest until the relationship between collateral branches became so tenuous that it ceased to have any relevance, and this only occurred after the lapse of several generations.95 The family always sought to ensure reversion. Possession always left a residual family claim, strong or weak, depending on the length and terms of tenure and the distance of the relationship. The consequences of the Norman Conquest must be understood within this emerging system of family law. It led to an enormous increase in the number and value of acquisitions. This in turn created a large number of collateral titles to property which became more distant and more complex as each generation passed. But one family settlement, that of the ruling house, cut across all the rest. The structure of inheritance assumed dependence on a single feudal lord. In 1087 this ceased to hold good. Title and succession were now at odds with the demands of political allegiance. The great noble families tried to resolve this dilemma by rebellion, the king/duke by using disseisin and disinheritance as penalties for disloyalty. This only made matters worse, for the regrant of sequestered estates created rival family claims. This tenurial crisis had serious and permanent effects on the development of English law. First, it focused attention on the use of disseisin and disinheritance as penalties for political misbehaviour, on all those questions of process and punishment which came to a head in Magna Carta. Secondly, it probably contributed to and may explain a unique feature of Anglo-Norman custom: the wardship of heirs belonged not to the family, as in other parts of western France, but to the feudal lord; in the case of tenants-in-chief, to the king/ duke.96 In 1100 Henry I promised in his charter of liberties to accept the more general system of family wardship,97 but seigneurial wardship remained the rule. In the circumstances which followed the separation of England and Normandy in 1087 it was the only practicable system for all concerned, for the nobility as well as the ruling house. Thirdly, it imposed essential limitations on the law of succession. Direct succession of son to father was not in question. This might be denied in practice to the sons of traitors, but that arose from the development of felony and treason, not from any attempt to deny the normal rule of inheritance.98 However, collateral succession and succession through 94

The family.. .is at the same time a group of persons and a group of possessions', Genestal, Leparage noimand, p. 40; ibid., pp. 38-40. 95 Pollock and Maitland, History of English Law, ii, pp. 295-308; Plucknett, Concise History of the Common Law, pp. 680-81. Germanic law fixed surprisingly wide limits, i.e. the fifth, sixth or seventh parentela. Bracton allowed six, but in England in the thirteenth century the point was determined procedurally by the dates of limitation of legal memory. But see below, pp. 259-60. 96 J. Yver, 'Groupe de coutumes', pp. 40-41. 97 Cap. 4 (Stubbs, Select Charters, 9th edn, p. 118). 98 The punishment of such felony, like the law of succession, developed under political pres-

Politics and Property in Early Medieval England

129

married heiresses, indeed distant succession of any kind, constituted a different matter. It was likely to lead under Rufus or Henry I to a vassal of Duke Robert or an ally of William Clito, and under Henry II, his sons and grandson, to a vassal of the Capetian house. There was genuine doubt whether such an heir could lawfully perform homage and fealty to the king of England. Hence the answer to the question - 'inheritance by whom?' - was for a son or an unmarried daughter in the king/duke's custody, 'yes'; but for a married daughter, brother, uncle, cousin or relation of the half blood, 'maybe'. This was the critical point where politics and patronage intruded into custom and law. Intervention was not limited to those situations where the claimant's loyalty was immediately suspect. Indirect succession was less certain in toto and in principle. The more distant or debatable the claim, the more substantial would be the overlord's consent and the more likely that he would be concerned with the political import of the settlement. Such claims and the resulting conflicts were an important element in the tenurial crisis of the early twelfth century." The quarrel for the crown itself was simply one of many, but one which exacerbated all the rest since it provided rival sources of patronage and encouraged rival claimants in their mutally inconsistent ambitions. Henry II's accession did not put an end to this. The loyalty of the expectant heir, the 'foreign' connections of the family, clearly weighed in the recognition of William II de Forz, titular count of Aumale, as claimant to

continued

sures. It seems likely that Henry I was much less lenient than Rufus and enforced confiscation more rigorously and permanently. It may be that he deliberately set out to endow his immediate supporters with landed estates. He was also for a time in a more precarious situation than his brother. Rufus was the lawful successor to the English crown; Henry was a usurper who had overridden the claims of his elder brother and feudal lord. 99 The point may be illustrated by many well-known examples. For the pursuit of a distant claim iure uxoris see the interest of Robert de Beaumont in the lands and titles of Roger, earl of Hereford, his wife's great-uncle, deprived in 1075, which brought him into conflict with Miles of Gloucester, earl of Hereford: Complete Peerage, vii, pp. 529-30; RRAN, iii, nos 437, 438, 439. For a conflict lying between claims of a daughter and her husband against a nephew see the dispute between Hugh de Beaumont and Miles de Beauchamp over the custody of Bedford Castle, Round, Geoffrey de Mandevitte, p.. I7ln.; Stenton, First Century of English Feudalism, pp. 237-38. For a dispute between collaterals where the matter was complicated by confiscation see the quarrel between Gilbert de Lacy and Roger, earl of Hereford, the one the grandson and the other the husband of the greatgrand-daughter of Walter de Lacy, Cronne, The Reign of Stephen, pp. 156-63, which gives the best summary of the plentiful literature on this problem. For a claim arising in the half-blood through the successive marriages of an heiress see the conflict between Simon II de Senliz, son of the first marriage, and King David of Scotland, husband of the second marriage of Maud, daughter of Waltheof, Regesta regum Scottorum, i, ed. G.W.S. Barrow (Edinburgh, 1960), pp. 102-3. For a conflict arising from the claims of co-heiresses see the quarrel between the Marmions and William de Beauchamp over Tamworth, pursuing claims based on marriage to daughters of Urse d'Abetot, brother of Robert the Dispenser, Cronne, The Reign of Stephen, p. 173. All these disputes played an important part in the politics of the Anarchy. Finally, for a conflict between cousins, the one female and the other disadvantaged by title through a female and the existence of an elder brother, see the quarrel between the Empress Matilda and Stephen of Blois for the realm of England.

130

Colonial England, 1066-1215

the honours of Holderness and Skipton in 1214,100 in the differing treatment of the two halves of the honour of Leicester between 1204 and 1231 when Simon III de Montfort's claim was finally recognised,101 and in the wavering and occasional acknowledgement of the succession of the counts of Eu in the honour of Tickhill.102 The royal line itself was affected; in 1199 John's succession was assisted by the fact that the Bretons had placed Arthur in the custody of Philip Augustus of France.103 Intervention was not based solely on the need to arbitrate or ensure a loyal successor. Loyalty was relative. There was no obvious borderline between royal rejection of a claimant as an actual or potential enemy, and royal preference on personal or political grounds for one claimant over another. The crown exercised its patronage with confidence in cases where there was no danger of treason, where it was simply a matter of preferring one claimant over another because he could pay more, or because he was a favoured royal agent, or perhaps on both counts for the two not infrequently went together. On the death of William de Mandeville, earl of Essex, in 1189, the nearest heir was his aunt, Beatrice, widow of William de Say (d. 1144). Richard I first accepted a proffer of 7000 marks for the Mandeville estates and the earldom from Geoffrey de Say, Beatrice's younger son, but Geoffrey failed to meet the terms of payment and the land was taken into the crown's hand. The king then accepted a proffer of 3000 marks from Geoffrey fitz Peter, husband of Beatrice de Say, daughter of William de Say (d. 1177), Geoffrey de Say's elder brother, and grand-daughter of Beatrice in the senior line. Richard I confirmed this at Messina in January 1190 in a charter which referred to fitz Peter 'as the right and nearest heir'. The claims of Beatrice de Say's sister, Maud, wife of William de Boclarid, were totally ignored, despite the fact that the two sisters had divided the lands of their father, William. The bid of William's brother, Geoffrey, was also shelved. It was only revived in 1213-14 when, in different political circumstances, King John encouraged Geoffrey and his son to renew it; they then proffered 15,000 marks. The succession raised an uncertain point of law, namely representation, complicated in this case by the fact that the senior line was represented by two daughters. But there can be little doubt that Geoffrey fitz Peter won in 1190 100 William de Forz v:as instated in October 1214: Rot. Chartarum, p. 201b; Rot. Litt. Pat., p. 122b. He had been invited to England at the request of Robert de Ros in October 1213 (ibid., p. 104b); it seems likely that he proved his loyalty by service on the Poitevin expedition in 1214 (Rot. Lift. Claus, i, p. 200b). 101 On the death of Robert, earl of Leicester, in 1204, his property was divided between his sisters, Amice, wife of Simon I de Montfort, vassal of King Philip of France, and Margaret, who married Saer de Quenci, created earl of Winchester 1207. In 1207 Simon II de Montfort firmly committed himself to the French cause; as a result this half of the honour of Leicester was taken into the crown's hands and was only restored, with the earldom of Leicester, to Simon III in 123139. Simon III was the younger son of Simon II and had abandoned any interest in the French patrimony of his house. He was therefore free of the political ties which had led to the forfeiture. Compute Peerage, vii, pp. 537-45. 102 See Appendix 2, note E, below p. 155. 103 yj^ was m ligg Richard began to treat John as his successor almost immediately.

Politics and Property in Early Medieval England

131

because he was a trusted supporter of the crown. His winnings were put in jeopardy in 1213-14 because this was not true of his son, Geoffrey de Mandeville, son-in-law of Robert fitz Walter and one of the leaders of the rebellion of 1215.104 The relative insecurity of indirect succession had a cash value. The more distant the succession the more likely it was that the claimant would have to pay a heavy finis terrae or offer a high price for the good will or arbitration of the king in order to defeat his rivals.105 Distant and collateral succession remained relatively weak. Here as elsewhere Edward I had a shrewd eye for the exercise of royal influence. His arrangements with Aveline, countess of Aumale, were to the detriment of distant heirs going back to Stephen, count of Aumale (d. 1127). His settlement with Isabella, dowager countess of Aumale, excluded the Courtenays, her cousins, from the greater part of her inheritance. The marriage of Edward's daughter, Elizabeth, to Humphrey de Bohun was to the detriment of Humphrey's brother. Edward's settlement with Roger Bigod damaged the interests of Roger's brother and cousin. The marriage of Thomas of Lancaster to Alice de Lacy excluded Lacy collateral claims; and the marriage of Joan of Acre to Gilbert of Clare excluded Gilbert's daughters by his first wife.106 Edward was no tyrant denying right, tide and inheritance: in the true line of the Norman and Angevin kings he was simply exploiting the relative weakness of indirect and collateral succession. Hence the certainty of tide depended on the directness of the claim: on the answer to the question 'inheritance by whom?' But this was not the only variable. At any succession, whatever the precise claim of the heir, and at any time, whatever the exact title of the possessor, certainty of tenure also turned on the nature of the property, on the answer to the question 'inheritance of what?' Property was not homogeneous. In the main it lay in land, but it also included castles, office and tide. Each of these categories was subject to its own rules; each followed its own peculiar line of development within the whole. The strength of hereditary title might wax in one setting while it waned in another. Tide to castles was markedly less secure than title to land. The distinction between royal and private castles was blurred by the need to license private castle-building, by the creation of hereditary castellanships and by a prerogative right of seizure. The Norman 'Consuetudines etjusticie' of 1091 state that castles could only be built by ducal licence and that the duke might seize them

104

Complete Peerage, v, pp. 120-24; xi, pp. 465-66. See also S. Painter, The Reign of King John (Baltimore, 1949), pp. 262-63;J.C. Holt, Magn,a Carta (Cambridge, 1965), pp. 121-23. 105 It should be noted that the history of relief in the twelfth century requires reassessment in the light of the type of succession involved. There is no doubt that direct succession involved much lower reliefs than indirect. The point is difficult to express statistically since reliefs for indirect succession cannot always be distinguished from proffers for royal arbitration in legal actions or simple fines for land. 106 McFarlane, 'Had Edward I a "Policy" towards the Earls?', pp. 151-56.

132

Colonial England, 1066-1215

into his hand if he wished.107 This last prerogative was openly admitted in 1090 in the dispute over the castle of Brionne between Robert fitz Baldwin and the Beaumonts. According to Orderic, neither would admit the claim of the other, but Robert fitz Baldwin accepted the overriding right of Duke Robert even though he regarded Brionne as his hereditary property: 'If you desire to have Brionne as your father had it as his own property, I will make no difficulty in delivering it to you; but otherwise I will keep my inheritance and will not transfer it to anyone while I live.'108 The right of seizure was not abandoned. In 1154 Henry II, far from bowing to tide and inheritance in this type of property, proceeded to undermine them with persistent intent. Already before his accession he had taken over Wallingford from Brian fitz Count. He probably regarded this as the resumption of a custody, but his actions were not restricted to this. In 1153 he concluded an agreement with Bishop Jocelin of Salisbury providing for the temporary retention of Devizes which Stephen had seized from Bishop Roger in 1139.109 He did not return it; instead in 1157 he bought off Jocelin's claims by the grant of thirty librates of land from the royal demesne and various churches along with a promise of royal support in recovering the lost rights of the see of Salisbury.110 After the death of Ranulf de Gernons, earl of Chester, he retained Nottingham, Stafford, Newcastle-under-Lyme, Tickhill, Bolsover and the Peak. In 1154 he recovered Northampton during the minority of Simon de Senliz and between 1155 and 1157 he took Scarborough from the fee of Aumale, Hereford from the fee of Roger, earl of Hereford, Norwich, Pevensey, Eye, Lancaster, Lewes, Conisborough and Sandal from William, count of Boulogne, Stephen's son, and Framlingham, for a time, from Hugh Bigod, earl of Norfolk. In 1158 the Mandeville castles of Pleshey and Saffron Walden were destroyed. Similar action was taken by Henry again in 1174 and by Richard I in 1194.111 Just as Henry seized Devizes from Bishop Jocelyn of Salisbury, so Richard seized Les Andelys from the see of Rouen for the construction of Chateau Gaillard.112 John was equally demanding: when in 1213 he had to restore the traitors, Eustace de Vesci and Robert fitz Walter, he first destroyed 107 'Consuetudinesetjusticie', cap. 4; C.H. Haskins, Norman Institutions (Cambridge, MA, 1918), p. 282. For comment see J. Yver, 'Les chateaux forts en Normandie jusqu'au milieu du Xlle siecle', Bulletin de la societe des antiquaires de Normandie, 53 (1955), pp. 60-63. 108 Orderic Vitalis, ed. Chibnall, iv, pp. 204-8. The situation was confused by the fact that both parties had a hereditary claim to the castle. It had belonged to Gilbert de Brionne, Robert fitz Baldwin's grandfather; it had also been granted to Roger de Beaumont, Robert of Meulan's father, in exchange for the custody of Ivry (ibid.). See also \Ver, 'Les caracteres originaux', pp. 66-69. 109 MAN, iii, no. 796. 110 Charters and Documents Illustrating the History of the Cathedral City and Diocese of Salisbury, ed. W.D. Macray, Rolls Series (London, 1891), pp. 29-30. 111 R. Allen Brown, 'A List of Castles, 1154-1216', EHR, 74 (1959), pp. 250-55. The evidence for the seizure of the castle of William, count of Boulogne, in 1157, two years before his death, is provided by Robert of Torigny (Chronicles of the Reigns of Stephen, Henry II and Richard I, iv, pp. 19293); if correct Henry must have overridden the terms of the treaty of Westminster of 1153. 112 F.M. Powicke, TheLossofNmmandy (Manchester, 1961), pp. 115-16. Like Jocelyn of Salisbury, Archbishop Walter of Rouen received compensation, in Dieppe, Louviers and elsewhere.

Politics and Property in Early Medieval England

133

their castles.113 In 1215 the security of tenure of castles was raised in principle. The restoration of castles was included in chapter 52 of Magna Carta,114 and John had to restore fortifications which he had seized during the recent crisis.115 Furthermore, he was now faced by a number of claims to hereditary title: Robert fitz Walter to Hertford; William de Lanvallei to Colchester; William de Mowbray to York; Henry de Bohun, earl of Hereford, to Trowbridge; and Geoffrey de Mandeville to the Tower of London. The first two were admitted,116 but John contested the rest. A local enquiry was held into William de Mowbray's right to the hereditary custody of York.117 When the honour of Trowbridge was restored to Henry de Bohun on 19 June 1215 a distinction was made between theplanaeterraeznd the castle, for the restoration of which the king was allowed respite until 28 June; there is no evidence that it was restored.118 The Mandeville claim to the Tower was never accepted.119 Moreover, John clearly believed that he had rights similar to those outlined in the 'Consuetudines et justicie' of 1091. The confusion which arose over the arrangements he had made with Stephen Langton for the custody of the archbishop's castle of Rochester was the immediate cause of the renewed civil war in the autumn of 1215. In the King's eyes Langton was a 'notorious and barefaced traitor' because 'he did not surrender our castle of Rochester to us in our great need'.120 Property in office was more of a problem. The rule stated in the 'Dialogus de scaccario', that officials were liable in their bodies for the proper perform-

113 R. Allen Brown, 'A List of Castles', pp. 254-55;J.C. Holt, The Northerners (Oxford, 1961), pp. 94-95; Histoire des dues de Normandie et des rois d'Angleterre, ed. F. Michel, Societe de 1'histoire de France (1840), pp. 118-19. 114 It did not, however, figure in cap. 25 of the Articuli. 115 Richmond to Ruald fitz Alan, Fotheringhay to Earl David of Huntingdon (Rot. Liu. Pat., pp. 143b, 144,148b). 116 Hertford was restored to Robert fitz Walter 'as his right' (ibid., p. 144b). The terms of the transfer of Colchester to William de Lanvallei are not recorded (ibid., p. 151), but the Barnwell chronicle states that he had claimed the castle as of right, Memorial* Fratris Walteii de Coventria, ed. W. Stubbs, Rolls Series (London, 1872-73), i, p. 221. The Lanvallei barony was formed from land held earlier by Eudo Dapifer. Henry I granted the tower and castle of Colchester to Eudo in 1101 (RRAN,ii, no. 552). 117 For the basis of Mowbray's claim, see Complete Peerage, ix, p. 370, n. (a). For the enquiry see Rot. Litt. Pat., p. 143b, Rot. Litt. Glaus, i, p. 215, reprinted in Holt, Magna Carta, p. 346. The case is discussed by J.C. Holt in The Making of Magna Carta', EHR, 72 (1957), pp. 408-9; idem, The Northerners, p. 120; idem, Magna Carta, pp. 158, 254, 306. 118 Rot. Litt. Claus., i, p. 215, reprinted in Holt, Magna Carta, p. 346, and discussed ibid., pp. 254, 259. 119 Geoffrey de Mandeville's claim must have been based on the grants of Stephen and Matilda to Geoffrey II de Mandeville (Regesta, iii, nos 274-76) or on the fact that his father, Geoffrey fitz Peter, had custody of the Tower until his death in 1213, probably asjusticiar. Geoffrey de Mandeville was required to surrender it on succeeding to his father's lands and rights in November 1213 (Rot. Litt. Pat., p. 105b; Holt, Magna Carta, pp. 122-23). 120 V.H. Galbraith, Studies in the Public Records (London, 1948), pp. 136, 161-62; Holt, Magna Carta, pp. 255-56.

134

Colonial England, 1066-1215

ance of their duties, applied to hereditary as well as temporary officials.121 That tenure should be justified by good behaviour was an unchallengeable and enduring doctrine. But the strength of hereditary title also varied with the nature of the office. Foresterships-in-fee exemplified hereditary office at its most precarious. From Henry I onwards foresters were deprived of office or compelled to fine heavily for restoration when it was discovered that they had transgressed the somewhat lax standards set to their honesty and efficiency.122 Some of the offices of state, in contrast, the stewardships, constableships, marshalcy and chamberlainships, represented hereditary office at its most ancient and secure; some of these enjoyed a continuous history from the Conquest onwards despite the accidents of family history.123 Between these two extremes lay the major offices of local government. In Normandy the vicomtewzs already heritable by the end of the eleventh century. In England the equivalent office, the shrievalty, sometimes associated with one or more castellanships, was held by two or more generations of the same family in Gloucestershire, Worcestershire, Herefordshire, Wiltshire, Devon, Kent, Norfolk, probably in Leicestershire and possibly Warwickshire and Essex.124 Whether the office was properly hereditary may be doubted. In some counties the family lost control at an early date;125 in others its tenure of the office was discontinuous.126 Formal confirmation of hereditary title to the office was unusual, except under Stephen. The shrievalty of Worcester passed from Urse d'Abetot to his son and then to his son-in-law, Walter de Beauchamp and to Walter's son, William. But in 1131 Henry I confirmed William de Beauchamp in his father's lands and in his father's office of Dispenser, without mentioning any title to the shrievalty; this was not confirmed in feodo et hereditarie until Matilda's grant to William of 1141.127 Similarly, there is no direct evidence of a Mandeville claim to the hereditary constableship of the Tower and the shrievalties of London and Middlesex, Essex and Hertfordshire until the charters of Stephen and Matilda of 1141.128 By Stephen's time tenure of local office had become the object of aristocratic ambition. Here the issue of hereditary tide may have been one 121

Dialogus de Scaccario, ed. and trans. Charles Johnson (London, 1950), pp. 78-81. RRAN, ii, no. 1518; Holt, The Northerners, p. 161. 123 J.H. Round, The King's Sergeants and Officers of State (London, 1911), especially pp. 35-51; G.H. White, The Constables under the Norman Kings', The Genealogist, new series, 38 (1921), pp. 113-27; idem, The Household of the Norman Kings', Transactions of the Royal Historical Society, fourth series, 30 (1947), pp. 127-55; RRAN, ii, pp. xi-xvii. 124 W.A. Morris, The Medieval English Sheriff to 1300 (Manchester, 1927), pp. 50-52. For Gloucestershire and Herefordshire, see also D. Walker, 'Miles of Gloucester, Earl of Hereford', Transactions of the Bristol and Gloucester-shire Archaeological Society, 77 (1958), pp. 66-70. 125 Morris, The Medieval English Sheriff, p. 51. 126 For example the successive tenure of Wiltshire by Edward, Walter, William and Patrick of Salisbury was interrupted by William de Pont de 1'Arche in 1110 and by Warin the sheriff in 112830, ibid., p. 78; Pipe Roll 31 Henry I, Magnum Rotulum 31 Henry I, ed. J. Hunter (London, 1833), p. 12. 127 RRAN, ii, no. 1710, appendix, no. cclvi; ibid., iii, no. 68. 128 Ibid., nos 275-76; Round, Geoffrey de Mandeville, pp. 151-54. 122

Politics and Property in Early Medieval England

135

of principle between the crown and its vassals. But if so, the accession of Henry II brought defeat not victory for the hereditary principle. The sheriffs were investigated, many were removed, and the office ceased to be hereditary. Another local office to which heritable title had been claimed, the county justiciarship, was abolished. Title to local office became more precarious as title to land became more secure. By 1215 it was beyond revival.129 It may be that this also affected the earldoms. The idea that an earldom was an office was still represented vestigially by the receipt of the third penny. Under Stephen the earls had exercised real power.130 In the late twelfth century succession to the title and the associated estates did not always go together. William de Albini was styled earl of Arundel in 1189; he did not receive the honour and castle of Arundel until June 1190.131 William de Ferrers succeeded to the honour of Tutbury in 1190-91. He was styled earl of Ferrers on the Pipe Roll of 1194, but was not girded with the sword of the county of Derby until 1199.132 Geoffrey fitz Peter acquired the lands of William de Mandeville, earl of Essex, in 1190 and received the third penny from that day. He styled himself earl of Essex in a charter of 1191 but was not granted the sword of his county until 1199.133 Ranulf of Chester obtained the lands of his Roumare cousins in 1198, but was not recognised as earl of Lincoln until 1217. In this case there was an alternative claim from Gilbert de Gant, but this probably did little to deter King John after 1205 when his relations with Ranulf became settled for the rest of the reign. It is more likely that recognition of the claim was delayed because the title had lapsed after the death of William I de Roumare in 1159-61.134 These were not the only varieties of property. Mrs Wood has illustrated another in her discussion of monastic patronage. This was attached to estate in land, yet here also there was room for argument and royal intrusion. The rights of the founding family were reasserted both in 1215 and 1258.135 Nor was the crown always the aggressor. The movement towards the mediatisation of boroughs under Stephen illustrates yet another area of confused property rights; here the nobility was on the attack.136 Property varied in nature just as claims to it varied in immediacy. Security of tenure and strength of title varied with both. These conditions established both opportunities for and constraints on the 129

William de Mowbray's claim to the castle of York, the forest of Yorkshire and the manor of Pocklington is a possible example of such a claim, but .by implication only; the shrievalty was not mentioned specifically: Rot. Lift. Claus., i, p, 215; Holt, Magna Carta, pp. 346-47. 130 Davis, King Stephen, pp. 129-44. 131 Complete Peerage, i, pp. 235-36. 132 Pipe Roll 6 Richard, Pipe Roll Society, new series, 5 (1928), passim; Complete Peerage, iv, p. 194. 133 Ibid., v, pp. 122-24. The title appears in Geoffrey's charter to William Pointel, concerning the custody of one of the undertenancies of the Mandeville fee, witnessed by Reginald, bishop of Bath (d. 26 December 1191), William Salt Library, MS S.D. (Pearson), no. 248.1 am obliged to Mr F.B. Stitt for bringing this charter to my notice. 134 Holt, The Northerners, pp. 26-27; Complete Peerage, vii, pp. 667-75. 135 Susan Wood, English Monasteries and their Patrons in the Thirteenth Century (Oxford, 1955), pp. 8-28, 96-100. 136 Stenton, TheFirst Century of English Feudalism, pp. 234-35, 239-41.

136

Colonial England, 1066-1215

actions of the Norman and Angevin kings. The gap between the political interest of the crown and the extremes of baronial expectations, between one baronial claim and another, created a running argument which took place within a context of conventional practices and legal notions, some settled as hard and fast rules, others slowly crystallising through the accumulation of precedent and recurrent political pressures. It remains to ask how serious was the argument, how deep the crisis, and in what ways the context altered in the course of the twelfth century. Crises are not readily susceptible to measurement. One estimate is that, up to 1135, 53 per cent of the Anglo-Norman baronies descended undisturbed in the male line and 10 per cent through heiresses.137 This leaves 37 per cent in which the succession was broken for one reason or another. This probably .overestimates the disturbance since it seems to include discontinuity arising from the extinction of lines, voluntary abandonment or sale. Approaching the problem from the opposite logical pole it seems that deprivation and disinheritance affected the descent of twenty-four major baronies, that is 13 per cent of the total. It is arguable that even this is an overestimate. It refers to baronies, not barons, still less families. Hence it ignores, for example, the facts that Pevensey, Berkhamsted and Launceston were all held by the count of Mortain; and that Arundel, Shrewsbury, Lancaster and Pembroke were all in the hands of the sons of Roger of Montgomery. Stated in terms not of baronies but of families the proportion would be much less. Conversely it would be very much more if it were stated in terms of Domesday valets, for it would include the lands of some of the most powerful men of the realm: Odo of Bayeux, the count of Mortain, and the families of fitz Osbern, Montgomery, Mowbray and Malet. Whether such a calculation would be worthwhile may be doubted. Whatever numerical expresion it might yield to indicate disturbance in descent, there is at present no norm. Is 10 per cent per century about right as a general expression of the natural intransigence of barons and the wilfulness of kings? Or is 20 per cent or 30 per cent? And how would this compare with disturbance produced by natural causes: death, division and marriage? Until such figures are established the extent of one particular crisis cannot be assessed arithmetically. There are other more important difficulties. Forfeiture was not always complete. Sometimes baronies were transferred in marriage with an heiress of the deprived family to some reliable supporter of the crown.138 Sometimes a forfeiture was 137

Davis, 'What Happened in Stephen's Reign?', p. 9. Professor Davis's figures like my own are based on IJ. Sanders, English Baronies (Oxford, 1960). This does not trace all descents in sufficient detail for such analysis and hence sometimes presents discontinuity where in fact there was continuity, e.g. the barony of Staunton le Vale, which was the Domesday holding of Rainer de Brimou. 138 The barony of Robert de Mowbray, confiscated in 1095, was transferred to Nigel de Albini who married Robert's widow, Complete Peerage, ix, pp. 368, 705-6. The barony of Robert de Montfort, banished in 1107, was transferred to Robert de Vere, second husband of Robert de Montfort's sister: Domesday Monachorum, pp. 68, 70; Complete Peerage, x, appendix (j), pp. 111-12. The estates of Roger d'Abetot, exiled in 1114, were transferred to Walter de Beauchamp who married Roger's

Politics and Property in Early Medieval England

137

wholly or partly reversed.139 Sometimes the succession was simply transferred within the owning family.140 No doubt some of these arrangements arose from immediate political convenience, but there was also a powerful stream of thought running in favour of heritable title. The agreement of September 1146 between Robert, earl of Gloucester, and Philip, bishop of Bayeux, arranged that Robert should hold the lands which Ranulf of Chester held of the see of Bayeux 'until there is an heir whom the Duke of Normandy recognises as Ranulf s lawful heir'. This was despite the fact that the charter marked the point at which Ranulf s lands were seized by the Angevins for his adherence to the cause of Stephen.141 The crown also was affected; its exercise of patronage could be expressed in terms which acknowledged the principles into which it was intruding. Henry Fs charter granting Sybil of Neufmarche to Miles of Gloucester emphasised that Bernard her father had surrendered his lands into the king's hands so that they might be regranted.142 When Henry gave William Mauduit the lands of Michael of Hanslope with his daughter Matilda in marriage, the charter stated that Michael had made the king his heir.143 One further limitation is striking and very important. The forfeiture of many of the leading families of the Conquest often had very little effect on their immediate tenants. It is likely that one of the consequences of the Salisbury Oath of 1086 was that undertenants could expect security of tenure from the crown in return for loyalty. That they got it is demonstrated by surviving acts which converted undertenancies of rebellious lords into tenancies-in-chief of the crown,144 or excepted the estates of loyal undertenants from the grant of a rebel's estates to a new tenant-in-chief.145 This left a permanent mark on the continued

sister, RRAN, ii, p. xvi, no. 1062, appendix no. Ixxxiv. The lands of Ernulf de Hesdin, hanged at Shrewsbury in 1138, went with his sister to Alan fitz Flaad, Sanders, English Baronies, p. 124. 139 See the recurrent interest in Pevensey of the L'Aigle family which recovered seisin thrice: c. 1118,1163-65 and 1212-16, ibid., pp. 136-37. Stephen, count of Aumale, was restored to Holderness c. 1102 after forfeiture in 1095 and interim possession by Arnulf de Montgomery, lord of Pembroke: Round, Cal. Docs. France, pp. xl-xli. Robert de Stuteville lost Cottingham in 1106; his grandson was reinstated under Stephen, EYC, ix, ed. C.T. Clay (Yorkshire Archaeological Society, record series, extra series, vii, 1952), pp. 1-5. The Lacys lost Pontefract c. 1114 but were restored in or after 1135, W.E. Wightman, The Lacy Family in England and Normandy, 1066-1194 (Oxford, 1966), pp. 66-73. 140 When Roger de Lacy was banished in 1096 he was succeeded by his brother Hugh, ibid., p. 169. Geoffrey de Mandeville (d. 1144) was not succeeded by his eldest son Arnulf, who was associated with his rebellion in East Anglia, but by his younger son Geoffrey, Round, Geoffrey de Mandevilk, pp. 227-42. 141 RRAN, iii, no. 58. 142 Round, Ancient Charters, no. 6. 143 RRAN, ii, no. 1719. 144 See Henry I's confirmation to Roger de Lacy of Rowland, which Roger formerly held of Roger of Poitou, ibid., ii, no. 611. 145 Hence Stephen's grant of Hereford to Robert, earl of Leicester, excluded the fees of Hugh de Mortimer, Osbert fitz Hugh, William de Briouze and Gotso de Dinant, with the proviso that the earl could enjoy a mesne tenancy in Gotso's case by his agreement, ibid., iii, no. 437.

138

Colonial England, 1066-1215

tenurial structure of the realm. Many of the established baronies of the late twelfth century originated in enfeoffments made by the first great Norman lords: Aldington, Chilham, Folkestone, Port and Ros by Odo of Bayeux; Castle Holgate, Cause, Pulverbatch, Wem and Wigmore by Roger of Montgomery; Chiselborough, Hatch Beauchamp, Fossard and Trematon by Robert, count of Mortain;146 some of the Lancaster baronies by Roger of Poitou;147 some of the Northumbrian baronies by Robert de Mowbray.148 The forfeiture of these men made a splash in the chronicles but, where tenants remained loyal, it was only a surface disturbance which did not seriously affect the immediate management of much of the realm of England. The one obvious exception to the security of the undertenant itself proves the rule: if he owed his estate to an interloper in the tenancy-in-chief, then it might be vulnerable if the original owner was reinstated, especially after a short interval.149 Otherwise the tenant was relatively secure if he remained loyal to the crown. Liege lordship was a reality. This leads to an awkward and important problem. In most countries in western Europe security of title and succession spread downwards from the upper to the lower social strata.150 Much of the English evidence is consistent with this. Some is not. In England life-tenures survived for a time after the Conquest. There are three clear and certain examples of the grant of estates for life or a term of lives in return for military service: by Robert Losinga, bishop of Hereford, Gilbert Crispin, abbot of Westminster, and Reginald, abbot of Abingdon.151 Possibly similar tenancies, not certainly held by military service, can also be traced on the estates of the sees of Worcester and Winchester and the abbeys of Ely and Shrewsbury,152 and it may be that all undertenancies recorded as held de victu monachorumor simply de dominiowere regarded as temporary by monastic houses 146

Details of the above may be drawn from Sanders, English Baronies, s.n. Clitheroe, Hornby, and probably Manchester and Penwortham; the fee of the constable of Chester was also an early creation but not certainly of Roger: VCH, Lancashire, i, pp. 297-336; James Tait, Medieval Manchester, and the Beginnings of Lancashire (Manchester, 1904), pp. 120-30. 148 Callerton (De la Val) certainly and possibly Redesdale and Mitford, W.R Hedley, Northumberland Families, Society of Antiquaries of Newcastle upon Tyne (1968), i, pp. 18ff. 149 Hence the grants which Arnulf of Pembroke made in free alms to the monastery of La Sauve and the cell of St Martin of Sees at Pembroke from the lands of the honour of Holderness lapsed when Stephen of Aumale was restored, Round, Cal Docs. France, pp. xl-xli. 150 See, for example, the views of Professor Duby, who established this for the Maconnais, above, p. 115 n. 8. For Normandy see Carabie, Laproprietefonfierepp. 276-77. 151 V.H. Galbraith, 'An Episcopal Land Grant of 1085', EHR, 44 (1929), pp. 353-72; J. Armitage Robinson, Gilbert Crispin, Abbot of Westminster (Cambridge, 1911), p. 38; Chronicon monasterii de Abingdon, ed. J. Stevenson, Rolls Series (London, 1858), ii, pp. 34-35. In the last instance lifetenure was emphasised deliberately in the teeth of the tenants' claim to inherit. However, this case was complicated by the grant of marriage portions to the three daughters and sole heirs of the tenant and a post obitum gift involving coparceny for which the abbot claimed his approval had not been sought. 152 Galbraith, 'An Episcopal Land Grant of 1085', pp. 363-68. For Winchester, see idem, 'Royal Charters to Winchester', EHR, 35 (1920), p. 387; for Shrewsbury see H.M. Colvin, 'Holme Lacy: 147

Politics and Property in Early Medieval England

139

and cathedral chapters.153 However, practice varied. At Abingdon, tenants recorded in Domesday were succeeded by their sons; the early evidence has been claimed to reveal 'the extreme stability of the feudal arrangements on the Abingdon lands'154 At Canterbury also, the families enfeoffed by Lanfranc were often still in possession of their estates in 1171; here inheritance was a 'matter of course'.155 Moreover, even the most striking examples of life-grants were only unavailing attempts to stay the establishment of hereditary tenure. Robert Losinga's grant of Holme, albeit a life tenancy, confirmed Roger de Lacy in a holding of his father. Abbot Reginald's grant of Garsington was to the heirs of the previous tenant, Geoffrey Marshal. Both the Lacys and Geoffrey Marshal's heirs soon established hereditary control of all or part of these estates. Gilbert Crispin's grant of Totenhall' to William Baynard also descended to his successors, the fitz Walters.156 And this was typical; at Hereford grants in fee are recorded by c. 1110.157 It is hazardous therefore to argue solely on the basis of this evidence that 'heritability of fiefs was far from being generally established in the period immediately following the Norman Conquest'.158 None of the early life tenures continued

An Episcopal Manor and its Tenants in the Twelfth and Thirteenth Centuries', Medieval Studies Presented to Rose Graham, ed. Veronica Ruffer and AJ. Taylor (Oxford, 1950), p. I7n. 153 Galbraith, 'An Episcopal Land Grant of 1085', pp. 363-68. 154 D.C. Douglas, 'Some Early Surveys from the Abbey of Abingdon', EHR, 44 (1929), pp. 62022. 155 H.M. Colvin, 'A List of the Archbishop of Canterbury's Tenants by Knight-Service in the Reign of Henry II', Kent Records, 18 (1964), pp. 1-40, especially 3-4; F.R.H. Du Boulay, The Lordship of Canterbury (London, 1966), pp. 60-61. Compare also the peculiarly full statement of the arrangements made for a succession to an estate held of the abbey of Burton, 1094-1113: 'he is to come into our chapter to give for his relief such a sum as a noble man ought to give for such land, swearing an oath as his father swore, giving as his father gave, holding as his father held' ('in capitulum nostrvim veniet daturus pro relevatione ipsius terrae tantum pecuniae quantum nobilis homo dare debet pro tali terra, jurando similiter sicut pater ejus juravit, donando sicut pater ejus donavit, tenendo sicut pater ejus tenuit'), The Burton Chartulary, ed. G. Wrottesley, William Salt Society, 5, pt 1 (1884), p. 30. 156 For Holme Lacy, see Colvin, 'A List of the Archbishop of Canterbury's Tenants', pp. 15-40. The descendants of Gilbert Marshal were less successful at Garsington. William, son of Abbot Reginald of Abingdon, acquired an interest in land there for a time, and half the fee was subsequently granted by Abbot Vincent (1117-30) to Simon Dispenser (Chron. Abingdon, ii, pp. 166-67); the Marshals retained the other half (VCH, Oxfordshire, v, p. 139). For the descent of Totenhall' see Armitage Robinson, Gilbert Crispin, pp. 38-39. 157 Galbraith, 'An Episcopal Land Grant of 1085', p. 368. 158 F.L. Ganshof, Feudalism (3rd English edn, London, 1964), p. 135; a view followed by R. Allen Brown 'In the beginning the fief was not hereditary, and was scarcely to become so of right in England before 1154' The Normans and the Norman Conquest (London, 1969), pp. 241-42. Ganshof, however, thought that heritability was a characteristic feature of the English fief in the twelfth century, Feudalism, p. 135. Professor Galbraith emphasised that the evidence should not be pressed beyond the ecclesiastical estates from which it is drawn. Likewise, in considering the evidence concerning Bury, Peterborough, Abingdon and Canterbury, Professor Douglas attributed the reluctance of ecclesiastical lords to enfeoff beyond their service-quotas to the desire to avoid 'hereditary military tenures which might entail a permanent alienation of ecclesiastical land', Domesday Monachorum, p. 71.

140

Colonial England, 1066-1215

were described as fiefs unless or until they became hereditary.159 As Professor Galbraith suggested, it is likely that some conservative churchmen tried to stave off aristocratic importunity by resorting to an older form of tenure,160 and it may be no accident that much of the evidence for life tenancy comes from the west Midlands and the Marches where Wulfstan survived in the see of Worcester until 1095. But an English origin, although probable, is not certain; similar grants are also recorded in France and elsewhere on the Continent.161 There are other weighty reasons for believing that tenure and succession were less secure in the lower ranks of the feudal hierarchy. The toponymic names of English knightly families are preponderantly English; only a minority attached their lineage to French estates. At this social level the French conquerors were much less conscious of family traditions of hereditary property and perhaps much readier to accept landed endowment on less certain terms. That there was a large contingent of adventurers and mercenaries in the armies which conquered and settled England is certain;162 and it is equally certain that on some estates there were petty military tenements which were very different from the estate of several fees with which one tenant-in-chief might enfeoff another.163 It also seems very probable that the term applied to this class in the mid twelfth century - miles - had a more general meaning a century earlier, that it only acquired a restricted social sense as the men so described established their tides to heritable landed property or stemmed from families which enjoyed such title.164 Finally, the establishment of the action of mort d 'ancestor and the assumption that it would normally be used by men of small estate against the 159

For the Holme Lacy charter see Galbraith, 'An Episcopal Land Grant of 1085', p. 372; Holme Lacy was first described as a fief in the bishop's carta of 1166, ibid., p. 368; Liber nigei'scaccarii, ed. T. Hearne (Oxford, 1728), i, p. 150. For Garsington see Chron. Abingdon, ii, pp. 34-35. For Shrewsbury, see Collectanea topographica et genealogica, i (London, 1833), p. 25. For Totenhall' see Armitage Robinson, Gilbert Crispin, p. 38. 160 Galbraith, 'An Episcopal Land Grant of 1085', p. 369; cp. Wightman, The Lay Family, pp. 128-29. 161 See M. Prou, Manueldepaleographie (Paris, 1904), ii, pi. 14, for examples of 1100 and 1138; these do not specify military service, but nor do the pre-Conquest equivalents or many of the Anglo-Norman examples. The existence of precarious military tenures on the Continent has long been recognised. The work of Flach and Luchaire on this point is conveniently summarised in Feudal Documents from the Abbey of Bury St Edmunds, ed. D.C. Douglas (London, 1932), pp. ciii-civ. See also Guilhiermoz, Essai sur les migines de la noblesse, pp. 195-201. Bloch gave considerably less weight than these earlier authorities to the late examples of precarious tenure, La societefeodale, pp. 303-4; trans. Feudal Society, pp. 196-97. 162 See especiallyJ.O. Prestwich, 'War and Finance in the Anglo-Norman State', Transactions of the Royal Historical Society, fifth series, 4 (1954), pp. 19-43. 163 See especially Sally Harvey, The Knight and the Knight's Fee in England', Past and Present, 49 (1970), pp. 3-43. 164 H.G. Richardson and G.O. Sayles, The Governance of Medieval England (Edinburgh, 1963), pp. 55-61; Duby, La societedans le region mdconnaise, pp. 191-96, 'La noblesse dans la France medievale', pp. 1-22; idem, 'Les origines de la chevalerie', Otdinamenti militari in occidente nell'alto medioevo, Settimane di studio del centro italiano di studi sull'alto medioevo (Spoleto, 1968), ii, pp. 739-61, and the contribution by L. Musset, ibid., pp. 845-47.

Politics and Property in Early Medieval England

141

denial of succession by a feudal lord suggests that as late as the reign of Henry II undertenants were especially vulnerable.165 In these circumstances it may well be that development was uneven and depended on the vigour and effectiveness of the feudal courts of the great Anglo-Norman baronies.166 All this is consistent with the continental evidence. However, it is not the whole story. If the chief cause of insecurity was the division between England and Normandy, and if the greater part of the lesser nobility in England had comparatively insignificant possessions in France or none at all, then men of this class must have been relatively immune from the main source of tenurial disturbance. This presumably is why many of the tenants of Robert of Mortain, Odo of Bayeux and the rest of the great rebel lords in England remained loyal to the Crown. The same argument applies all the way down the social scale to the holder of a fraction of a fee. Indeed the central stark reality of English politics from 1087 to 1154 was that title and inheritance were really weak right at the top, with the crown itself. The next weakest were those most closely affected by the royal dispute, the tenants-in-chief. Others might avoid the conflict and backwash of insecurity which it produced. Indeed insecurity at the top encouraged security lower down. Actual or potential rebels could be undermined by securing the loyalty of their tenants.167 England may also have been peculiar in the opportunities it provided for the exercise of patronage. It is only too easy to think of the great lords, lay and ecclesiastical, as if they were hard-bitten puritans, nursing their resources, concerned above all to avoid the permanent alienation of their newly acquired or recently confirmed property. In reality their world was one where resources included expectant followings and settled tenants, only to be retained if expectancy was satisfied and tenure confirmed. The tabula rasa of the Conquest provided room enough to satisfy all without exhausting demesne resources. On any of the accepted interpretations of Anglo-Norman history the 'ready-witted, outrageous, rumbustious' Ranulf Flambard would figure as one of the chief agents of royal autocracy and predatory administration.168 Yet among his acta as bishop of Durham none of the six surviving charters of enfeoffment create life tenures; none leave the issue of inheritance in any doubt; all convey hereditary title, emphasising it with the characteristic formulae, iure hereditario, per hereditariam 165 See the very important comments of Professor S.F.C. Milsom in Pollock and Maitland,d hisHistory of English Law (Cambridge, 1968), i, pp. xxvii-xlix, especially pp. xxxvi-xxxviii. He points out that Glanvill's example for the writ is a claim for one virgate. 166 Ibid., pp. xxxiv-xxxv. 167 This technique of government is implied by the Salisbury Oath and was exploited very explicitly by King John. See, for example, his agreement with John de Lacy of 1213, Holt, The Northerners, p. 182. 168 R.W. Southern, 'Ranulf Flambard and Early Anglo-Norman Administration', Transactions of the Royal Historical Society, fourth series, 16 (1933), pp. 95-128. The quotation is taken from idem, The Place of Henry I's Reign in English History', p. 153. The earlier views of Stubbs and Freeman that Flambard inspired the system of military tenures in England were castigated by Round, Feudal England (London, 1909), pp. 228-28,

142

Colonial England, 1066-1215

successionem, or in feodo sibi et heredibus suis.169 Some of these grants were in favour of members of Flambard's family, but he was not unique in having relatives or being able to endow them.170 Coming from one so closely involved in the government of William Rufus, these enfeoffments stand in sharp contrast to the evidence from Hereford, Worcester and elsewhere. Special English circumstances also determined the general development of security of tenure and succession. It is normally assumed that on the Continent, especially in France, it was gradual and progressive. In England this was not so. It has already been suggested that in the Salisbury Oath of 1086 the crown implicitly accepted that under-tenant families enjoyed security of tenure.171 By the same token the Domesday Survey and the pattern in which its information was recorded, county by county and fief by fief, reveal that tenancies-in-chief were now firmly established as tenurial units. Domesday Book implies permanence and the corollaries of permanence are secure tenure and settled succession. To have envisaged the destruction and reconstruction of the great honours it records would have been contradictory both to its form and purpose. Yet this is what happened in many cases. If this reasoning is correct, it follows that the separation of England and Normandy in 1087, with all its consequences, marks a setback to a straightforward, even precocious development, a set-back which undermined the security of tenancies-in-chief more than under-tenancies. By the same reasoning the reunification of England and Normandy, first by Henry I in 1106-35, then by Henry II from 1154 removed the major source of instability. There remained the after effects: the prolonged legal actions among the successors of the disputants of an earlier age,172 the continuing relative weakness of collateral or distant succession, and the opportunities which this presented to the crown in the exercise of its patronage.173 The Angevins may also have recognised the problems which the Normans had created. Within the more developed legal structure of the late twelfth century the crown devised subtler means of dealing with rebellion. The destruction of castles and the payment of ransom replaced the blunt instrument of confiscation and disinheritance: 'non fiat exhaeredatio sed redemptio'.174 Moreover, when Normandy and 169 Durham Episcopal Charters, 1071-1152, ed. H.S. Offler, Surtees Society, 179 (1968), nos 11, 12, 13, 19, 22, 23.1 am grateful to Professor Offler for drawing my attention to this evidence and for making the documents available to me in proof. No. 23, which uses the formula infeudo et in hereditate, is possibly suspect in detail but not in substance; it was confirmed by Henry II in no. 23a. See ibid., pp. 101-7. 170 Round, Feudal England, pp. 301-3. 171 See above, p. 137. 172 Among the more obvious examples are Stutevillev. Mowbray, Lacy v. De la Val, Bully v. The Counts ofEu, Bohunv. The Earls of Salisbury. 173 See above, pp. 129-31. 174 Dictum de Kenilworth, cap. 12 (Stubbs, Select Charters, 9th edn, p. 409). A punitive ransom was effectively the same as disinheritance. See the deed of surrender of the estates of Robert de Ferrers, earl of Derby, in 1269, Sir Christopher Hatton 's Book of Seals, no. 411. For earlier examples of ransoming see the rebellion of 1215-17, Holt, The Northerners, pp. 246-48; and for the destruction of castles, Henry II's treatment of the rebels of 1173 and John's treatment of the traitors of 1212.

Politics and Property in Early Medieval England

143

England were separated once more in 1204 great care was taken to avoid a new crisis.175 The surviving charters of the Norman kings tell yet another story. Indeed, they provide little warranty at all for believing that the barons of the Conqueror, in Maitland's phrase, 'held their lands heritably'. Charters of enfeoffment of the eleventh century are very rare. Some, no doubt, have been lost. But it is also very likely that men gave and received tenancies without any kind of title deed. Among the earliest surviving charters, whether royal, ecclesiastical or baronial, a high proportion concerned the enfeoffment of the grantors' relatives and household-officers. Written record seems to have been sought and obtained first by those who had immediate access to a writing office, and it is no accident that the charters given to such men are among the first to resort to the formulae of the heritable grant. But this development was scarcely underway in England before the reign of Henry I. From the reigns of William I and William II there are only five surviving documents recording royal grants of land to laymen.176 None survives in the original. None implies life tenancy. One, which is a writ of seisin rather than a charter, confirms a tenant in the land of his father at the services his father used to render.177 One is a grant infeodo of a house in Hertford and the manor of Bayford to Peter de Valognes. This was subsequently confirmed to Peter by Henry I in the same terms and by Matilda in feodo et hereditate to Peter's son, Roger. Roger, or his son, lost the manor by 1154-55 when it was once more in the hands of the crown.178 Henry I's charters, as R.W. Southern has observed,179 resorted increasingly to the formulae of inheritance. But the total evidence is still slight, amounting in all to thirty or so such acts.180 Nearly two-thirds of these come from the last fifteen years of the reign. Even so, it was still possible for the king to confirm lands and rights to direct heirs without using the formulae which were coming to be considered appropriate.181 Equally he might notify his confirmation of a 175

See the arrangements for the terra Normannorumwhich were placed in a tenurial limbo until the treaty of Paris of 1259, Pollock and Maitland, History of English Law, i, p. 461. Some of those who suffered loss in Normandy were given temporary compensation in England. It is worth emphasising that it is only hindsight which allows the historian to interpret the rift of 1087 as temporary and the loss of Normandy as permanent. 176 RRAN, i, nos 226, 346, 435, 442, 456. No. 432 duplicates no. 226. 177 Ibid., no. 456. 178 RRAN, i, no. 346; ibid., ii, no. 1121; Cartae Antiquae Rolls 1-10, ed. L. Landon, Pipe Roll Society, new series, 17 (1939), nos 305, 306; Southern, The Place of Henry I in English History', p. 161n. 179 Ibid., p. 145. 180 It should be noted that the calendared versions in RRAN, ii, are inadequate for the present purpose, since the terms of the grant are not accurately noted. Among grants in inheritance not so calendared are nos 609, 793, 1087, 1121, 1395, 1524, 1723, 1730, 1760 and 1930. The same applies to ibid., i, no. 346. 181 See the grant to Miles of Gloucester, 1128, of the lands and constableship of his father, Walter: RRAN, ii, no. 1552; C.Johnson, The Constableship of Walter of Gloucester', EHR, 49 (1934), p. 34; and the grant, 1128, to William son of Robert son of Walter of Windsor of the lands of his

144

Colonial England, 1066-1215

grant in inheritance by a vassal without repeating the formulae of the original deed.182 The next reign presents clearer evidence. The surviving records of grants to laymen by Stephen, Matilda, Count Geoffrey and Henry of Anjou number nearly one hundred. Over sixty of them are grants in inheritance. Most of these resort to the language characteristic of the late twelfth century; they record grants from the donor and his heirs to the recipient and his heirs in feodo et hereditate. Whether the grantor was Stephen, Matilda, Geoffrey or Henry makes no difference; all equally made grants in inheritance. The date of the grant is immaterial; they are as characteristic of the earlier as of the later years of the reign. By 1135 practice had hardened. At one and the same time there was a sharp rise in the number of grants and in the use of the relevant formulae. This arose both from the renewed conflict over the Anglo-Norman dominion and from slower, less dramatic changes. The origin of these formulae lay far back in the development of western Francia and the emergent duchy of Normandy. From the onset of the great revival of the Norman church under Duke Richard II, grants were made to cathedral churches and abbeys of land or rights to be held iure hereditario, or inperpetuam hereditatem.183 These phrases were intended to emphasise that the church held such benefactions in perpetuity. The language chosen is at first sight strange for the church did not die and had no heirs. One source of the usage was Roman civil law.184 Another is probably to be found in the church's concern to express its everlasting tide in terms which lay benefactors could readily comprehend; if so, it assumed that they already had a securely embedded concept of what inheritance was. The church was using the idea in its own context. A grant to a monastery iure hereditario or inperpetuam hereditatem expresses an intention: the gift is to fopassed on by the monks from generation to generation, in lay terms inherited. But the words can also be used in another fashion: to express not an intention but a fact, to describe property which has been or is now being inherited. Wherever the phrase is applied to laymen in the surviving acts issued or attested by the dukes of Normandy before 1066 it is used in this second sense.185 This explains why there were no enfeoffments in inheritance in the first generation after the setcontinued

father, RRAN, ii, no. 1556; Cal. Charter Rolls, ii, p. 137. With no. 1552 compare no. 1723 a grant in inheritance to Miles of Gloucester of the land of Bicknor, Transactions of the Bristol and Gloucestershire Archaeological Society, 4 (1879-80), p. 319. 182 In 1121-22 William Peverel of Dover granted to Thurstan his steward Gedding and Daywell 'to be held of me and my heirs in fee and inheritance', Stenton, The First Century of English Feudalism, p. 274. Henry I's confirmation of this grant does not repeat the formula; RRAN, ii, no. 1295; Cartae Antitjuae Rolls 11-20, ed. J. Conway Davies, Pipe Roll Society, new series, 33 (1957), no. 476. 183 RADN, nos 7, 9, 24, 34, 43, 71, 84,101,103,134,143,147, 201, 202, 233. Compare no. 95 in which property is described as the kgitima hereditas of the abbey of St-Wandrille. 184 'Inheritance has no other sense than succession to every right which the decedent held' ('Nihil est aliud "hereditas" quam hereditas est successio in universumjus, quod defunctus habuit'), Digest, L, 16, 24 (ed. T. Mommsen, p. 910). 185 For instances see above, p. 117 n. 34.

Politics and Property in Early Medieval England

145

tlement of England. No land had been inherited. Every man's title was a conquestu. If Duke William only used these phrases in referring to estates which had passed down from ancestor to heir in Normandy, then King William could scarcely use them at all in England. The formulae of the heritable grant could not take root until inheritance itself had taken place; the documentary expression of the idea had to follow the fact. The use of the formulae to express intent only came in gradually. William Rufus's grant to Peter de Valognes infeodo of 1087-93 seems to be the first.186 Only a dozen or so of the grants of Henry I seem to create hereditary title de novo,187 and of these nearly half are grants to officers or servants of the king.188 The rest are confirmations of under tenancies in which the hereditary arrangements certainly or probably originated in the private grants which they confirmed,189 or confirmations of hereditary succession which had already taken place or was taking place at the time of the grant.190 The formulae came to be used not just because men wanted to hold land heritably, but because they already held inheritances or land to which private title deeds attached hereditary right. If this is correct there is an obvious enough explanation of the change after 1135: it was simply that the disputed succession allowed men to press claims based on tenure by ancestors and compelled Stephen and his rivals to acknowledge them. This explanation is not new. It was emphasised by Round and Stenton191 and has been confirmed by Professor Davis and Professor Cronne. It rests securely on the documentary evidence in which ancient and often distant title form the basis of the concessions of both Stephen and Matilda. By the second half of the twelfth century the formulae of inheritance had become conventional, and by Bracton's day they were essential to a heritable grant.192 The words now expressed intent; laymen had come to enjoy the formal security devised by and for the church. Since this was already envisaged in the solitary surviving enfeoffment of Rufus, in the charters of Ranulf Flambard and some of the enfeoffments of Henry I, it may seem a natural enough development.193 In reality it was one of several interrelated changes. 186

RRAN, i, no. 346. Ibid., ii, nos 793, 911, 987, 992, 1134, 1256, 1395, 1723, 1777, 1946. No. 1121, to Peter de Valognes, is a confirmation infeodo of property already in the grantor's hands. Nos 793 and 992 are grants of a fee farm. No. 1777 is a grant of the master-chamberlainship. 188 Ibid., nos 911 (William de Albini, butler), 1134 (Robert Achard, the king's master), 1395 (Walter of Gloucester, constable), 1723 (Miles of Gloucester, constable), 1777 (Aubrey de Vere, chamberlain), 1946 (two loricarii). 189 Ibid., nos 783,1268,1603,1722,1730,1758,1778,1872. Express dependence on the original private grant is clear in nos 783, 1268, 1603, 1778 and 1872. In the case of no. 1603 the original grant of Ranulf Flambard still survives. See above n. 169, Durham Episcopal Charters, nos 23, 23a. 190 Ibid., ii, nos 609,1087,1524,1668,1719,1749,1760,1835,1930. Of these nos 1719 and 1930 are concerned with the transmission of property with an heiress. 191 Round, Geoffrey deMandevilk, pp. 53,109,149; Stenton, The First Century of English Feudalism, pp. 224-26. 192 Pollock and Maitland, History of English Law, i, pp. 307-8. 193 See above, pp. 141,143-44. 187

146

Colonial England, 1066-1215

In the eleventh and early twelfth centuries secular property which had not yet been inherited was not yet an inheritance but an acquisition. To describe an initial grant as an inheritance overrode this distinction, and the point was material since the acquisition was distinguished from the inheritance not only by origin, but also by the lord's capacity to alienate.194 This was likely to last as long as acquisitions proved an effective means of providing for younger sons. But every acquisition which passed from ancestor to heir became an inheritance. The distinctive quality of the acquisition was ineluctably eroded by the years for no amount of forfeiture and regrant could stay the consequences of death and succession. Hence other means were needed to provide for junior members of the family. Part of the answer in Normandy was to refine and extend the practice of parage. In England a solution was found through the enfeoffment of younger sons during the father's life. This change was under way in Glanvill's day. Glanvill assumed that all land held by military service descended by primogeniture; he had abandoned any distinction between inheritance and acquisition at death. He allowed a military tenant to distribute his acquisitions during his life if he also held an inheritance; but denied that he could distribute them all, thereby disinheriting his heir, if he held nothing but acquisitions. Hence the alienability of the acquisition was restricted on two counts. As if in compensation, Glanvill accepted that the lord, with his heir's consent, could endow younger sons with a reasonable portion of his estates, including his inheritance, during his lifetime.195 Hence the distinction between inheritance and acquisition was being replaced by alienability applied to the whole property. In practice the division of the property as separate units among collateral branches was replaced by the subenfeoffment of junior branches, in short by the appanage, ultimately by the conditional feoffment and the fee-tail. It now became possible to make new grants in feodo et hereditate, for inheritance and acquisition were no longer potentially subject to different rules of succession.196 Indeed it was now essential to use these formulae to express intention for the object of such enfeoffment was to make a permanent provision both for the younger son and his descendants. The tenant's control of his acquisitions was not the only casualty. At exactly the same point in time land came to be excluded from testamentary bequests and post obitum gifts. With a few exceptions, of which the most important was burgage tenure, land ceased to be devisable.197 The change was not sudden. The distinction between inheritance and acquisi194

See above, pp. 121-122. Glanvill, De legibus, cap. vii, i (ed. Hall, pp. 69-71). 196 It also follows that land could be sold to be held in feodo et hereditate. For instances see the sale of Langham by Hugh Tirel to Gervase of Cornhill 1146-48 (Sir Christophei Hatton 'sBook of Seals, no. 84); and, from the reign of Henry II, Recueildes actes de Henri II, ed. L. Delisle and E. Berger, Chartes et diplomes relatifs a 1'histoire de France, 4 vols (Paris, 1909-27), nos dxlvii, dlxiv, dcvi, and Stenton, Danelaw Documents, nos 542, 551. 197 Pollock and Maitland, History of English Law, ii, pp. 325-29; Sheehan, The Will in Medieval England, pp. 269-80. 195

Politics and Property in Early Medieval England

147

tion can still be traced under Henry II.198 Likewise Glanvill presents the two related notions of heritability and alienability still only incompletely formed.199 However, the origin of the change went much further back. Many of the essentials of Glanvill's position were already implied in the provision which William de Anesye, tenant of the honour of Port, made for his younger son, Richard, c. 1120-40. He gave him a knight's fee in Sherfield, Hampshire 'which is of my acquisition', but did so with the consent both of his lord and his elder son; the acquisition was still the suitable endowment for the cadet, but the elder son already had a voice in its disposal.200 The change was not difficult to achieve for the endowment of younger sons within the family inheritance was practised by the Norman ducal house from the time of Richard I onwards. It was an older tradition into which the distinction between inheritance and acquisition may simply have been an intrusion resulting from a period of expansion. By the reign of Henry I the enfeoffment of younger sons in this way was unexceptional. The cartae of 1166 suggest that such provision was common by 1135 and it developed rapidly in the next thirty years.201 Wherever it occurred it was likely to produce a grant in inheritance and where such deeds survive, even for the reign of Henry I, they are likely to use the formulae of the heritable grant as an expression of intention. This was the change in practice on which Glanvill built his theory. The arguments which have been advanced above are necessarily schematic. Some are by intention conjectural. Some of the problems may prove susceptible to statistical analysis. Others require a thorough survey of feudal succession and the formulae of tide deeds. There is great temptation both in the call of 'new methods' and in the quieter lure to act as family solicitor to the aristocracy of Norman England. This essay has been intended to provide a context for discussion, a scheme of things, dare it be said a 'model'. That 'model' is not 198

See above p. 126, and especially the arguments on the succession to Hugh Bigod, Appendix 2, note B, below pp. 151-52. For a late example of the disposition of an acquisition see the grant of Henry II 1166-1172/3 to Hugh of Hanslope of a hundred acres of assarts in Perry 'which may be inherited as he decides', Recueil, ed. Delisle, no. ccc$xxx. Compare the purchase of two tofts in Cadwell (Lines.) by Maud, wife of Matthew of Hamby 'to sell or give to whomsoever she wishes', Stenton, Danelaw Documents, no. 478. For another grant de escaeta, see Recueil, ed. Delisle, no. ccccxcii; and for the continued distinction between inheritance and acquisition, ibid., nos cccxxv, ccccviii. The notion that it was more suitable to create enfeoffments from acquisitions than from demesne was expressed in charters of Robert, earl of Leicester, 1170-75 (Stenton, Danelaw Documents, no. 495) and Simon de Senliz, earl of Northampton, 1176-84 (Sir Christopher Hatton's Book of Seals, no. 439). 199 On the subsequent parallel development of heritability and alienability see Thorne, 'English Feudalism and Estates in Land'; Glanvill, De kgibus, pp. 184-85; and A.W.B. Simpson, An Introduction to the History of the Land Law (Oxford, 1961), pp. 44-49. I intend to discuss elsewhere the implications of my argument for Professor Thome's examination of charters of enfeoffment. 200 Sir Christopher Hatton's Book of Seals, no. 301. This is of further interest since the fee was surrendered to the chief lord, Henry de Port, who enfeoffed the younger son as his immediate vassal. However, the elder line of Anesye ultimately defeated what was presumably an attempt to avoid a mesne tenancy (ibid., pp. 307-8). 201 The only analysis of this, regrettably brief, is by Painter, Feudalism and Liberty, pp. 207-11.

148

Colonial England, 1066-1215

simple. The tenurial crisis of the Anglo-Norman period arose not because there was no law governing title and inheritance, or because kings flouted it, but because of the difficulties they encountered and created in applying it.

Appendix 1 Parage

Paragewas a refinement of partition whereby the younger sons were represented by the eldest, who was treated as the sole heir. This lasted as long as the relationship lasted, according to the Tres ancien coutumier' of c. 1199 until the fourth degree,202 and according to the 'Summa de Legibus' of 1235 until the seventh.203 It was not until this point that the junior line performed homage; until then also the senior branch recovered on the extinction of the junior.204 Genestal, following Guilhiermoz, associated the establishment of parage in its classic form with the ordinance on the indivisibility of fiefs and he attributed both to the reign of Henry II.205 However, Dr Wightman has pointed out that the Lacy fee held of the bishopric of Bayeux provides a possible example of parage in the early twelfth and perhaps the late eleventh century.206 Another example of the early use of the term in Normandy, 1070—81, was found by H. Navel.207 Maitland discussed the considerable number of examples of tenure in paragio in the Domesday of Dorset and Somerset.208 See also the many instances in Hampshire.209 Many of these examples do not coincide with later Norman tenure in paragio. However the DB clerks clearly used the word in a technical sense. Guilhiermoz's criticism of Maitland, that the term was not always applied strictly in English circumstances, is beside the point.210 It was not generally current in England, if indeed it was known at all, before the Conquest; like villanus it was used to cover a variety of circumstances. It is also worth noting that the Norman system of parage underlay the feudal arrangements between co-heiresses described by Glanvill.211

202

Tres ancien coutumier, cap. xlv, Coutumiers de Normandie, i, ed. Tardif, p. 38. Summa de legibus, cap. xxiii, 7, Coutumiers de Normandie, hi, ed. Tardif, p. 77. 204 Genestal, Le parage normand, pp. 16-30. 205 Ibid., pp. 3-4; Guilhiermoz, Essm sur les origines de la noblesse, p. 214. 206 Wightman, The Lacy Family, pp. 218-20. 207 Navel, 'Recherches sur les institutions feodales en Normandie', p. 36. 208 Maitland, Domesday Book and Beyond, p. 145; Pollock and Maitland, History of English Law, ii, p. 264. 209 DB, i, fos 38-51b, passim. 210 Guilhiermoz, Essai sur les origines de la noblesse, p. 215n. 211 Glanvill, De legibus, vii, 3, ed. and trans. Hall, p. 76; Pollock and Maitland, History of English Law, ii, p. 276. 203

Politics and Property in Early Medieval England

149

Appendix 2 A. The Succession to Normandy and England 1087 I have been indebted at many points to Professor John Le Patourel's paper on The Norman Succession, 996-1135' ,212 not least for his informed and considered views on the general relationships between England and Normandy in the eleventh and twelfth centuries. Nevertheless there are some matters, mostly of detail, where I have followed a different line. I have not been convinced by Professor Le Patourel's argument that there was 'no distinction in William's mind between propres and acquets.213 There must have been, for he allowed it to determine the pattern of descent of the lands of some of his barons. How far the distinction influenced him in settling his own succession is a different matter. If, as Professor Le Patourel contends, he planned to leave a single dominion, first to Robert and then to William, this was not inconsistent with the parental disposition which the distinction allowed at this time, or with the policy followed by some of his tenants, so long as it was accepted that Robert might be deprived on grounds of filial infidelity. However, there are serious objections to thinking that these were William's intentions either on his deathbed or at any other time: 1

212

Robert was recognised as heir to Normandy but there is no evidence that he was ever recognised as heir to England. No mention was made of England when his succession to Normandy, first recognised in 1063, was confirmed as a post obitum gift in 1079.214 Moreover, at no point in the subsequent disputes between him and William Rufus or Henry did Robert ever argue that he had been recognised as heir to the kingdom; the case stated in his favour is that he was the eldest son to whom homage had been sworn as successor to Normandy and, in 1100, that he was Henry's feudal lord.215 Neither the agreement with William Rufus in 1091 nor that with Henry in 1101 suggest that Robert had any stronger claim. Indeed the 1091 terms clearly indicate that the Conqueror had divided his possessions: 'And the king promised him [Robert] in return to reduce to obedience Maine, which his father had conquered, which had then revolted from the count [Robert] and all that his father had over there, except that whiqh he had granted to the king.'216 Robert's rights in England in contrast were determined by his present agreement with Rufus.217 The evidence on the agreement of 1101 adds little to the

J. Le Patourel, The Norman Succession, 996-1135', EHR, 86 (1971), pp. 225-50. Ibid., p. 234. 214 Orderic Vitalis, ed. Chibnall, iii, p. 112. 215 Ibid., iv, p. 122; v, p. 290. 216 Anglo-Saxon Chronicle, MS(E): Two Saxon Chronicles, ed. C. Plummer and J. Earle (Oxford, 1892), i, p. 226; trans. Dorothy Whitelock (London, 1961), p. 169. 217 Ibid. 213

150

2

Colonial England, 1066-1215 argument except in Orderic's account, which indicates that Robert's claim to England was associated with the fact that Henry had performed homage to him; he now abandoned both together: 'Duke Robert resigned to his brother the claim which he had brought to the realm of England and because of his royal dignity released him from the homage which he had once done to him.'218 Robert's readiness to compromise with Henry is all the more striking because Henry was the third surviving son of the Conqueror who had been excluded from any territorial share in the succession in 1087. The evidence on the treaties was fully examined by Freeman.219 Professor Le Patourel has kindly pointed out to me that Orderic's account of Robert's part in the negotiations between Rufus and King Malcolm of Scotland in 1091 implies that Malcolm had performed homage to Robert.220 However, the apparently inevitable conclusion that Robert must have been recognised as heir to England does not follow. Orderic's story, which is highly coloured and cast in the rhetoric of direct speech, states that the Conqueror 'commended' Malcolm to Robert as his eldest son. But there is no evidence that Robert was a party to the settlement at Abernethy in 1072, the only occasion when the two kings met. Robert and Malcolm met in 1080 when Robert commanded an expedition against the Scots. In submitting on this occasion Malcolm must have performed homage to Robert, but simply as the Conqueror's representative.221 It seems that, in recalling it, Orderic recast this incident in terms which fitted the quarrel among the Conqueror's sons for the crown of England. The impression that the Conqueror divided his lands on his deathbed was stronger than Professor Le Patourel allows. Orderic not only laid emphasis on Robert's tide to the patrimony in his famous set piece on the Conqueror's deathbed in book vii,222 but also earlier in book v in a much more circumstantial and convincing note: 'So [Robert] left his father, never to return, until his father on his deathbed sent Earl Aubrey to him in France that he might take possession of the duchy of Normandy.'223 That Rufus was designated heir of England at the same time is, as Professor Le Patourel argues, clearly implied by the grant of the regalia to him recorded by the monk of Caen.224 Further evidence is provided by the Latin account of the acts of Lanfranc added to the 'A'

218

Orderic Vitalis, ed. Chibnall, v, p. 318. E.A. Freeman, The Reign of William Rufus (Oxford, 1882), ii, appendices N and XX, pp. 522-28,688-91. 220 Orderic Vitalis, ed. Chibnall, iv, pp. 268-72. 219

221

67-68. 222

Chron. Abingdon, ii, pp. 9-10; C.W. David, Robert Cur those (Cambridge, MA, 1920), pp. 13-15,

Orderic Vitalis, ed. Chibnall, iv, pp. 78-104. Ibid., iii, p. 112. 224 Le Patourel, The Norman Succession', p. 232. 223

Politics and Property in Early Medieval England

3

151

version by the scribe of the 'F' version of the Anglo-Saxon Chronicle. 'On the death of King William overseas, Lanfranc chose his son William as king as his father had arranged. '225 This was composed at Christ Church, Canterbury. This necessarily detracts from Eadmer's statement that Lanfranc was loath to recognise Rufus's claim.226 The two accounts are roughly contemporaneous. Eadmer was biased in his attitude to Rufus by his devotion to Anselm. It is true that Robert was recognised as heir of Maine, but it does not follow that this constitutes a breach of conventional practice on the ground that Maine was as much an acquisition as England. The succession to Maine was determined by agreement between Duke William and Herbert, count of Maine, in 1063, when it was settled on Robert and Margaret, Herbert's sister. Margaret died before the marriage took place. Nevertheless, William maintained the Norman claim and Robert performed homage for Maine to Fulk, count of Anjou in 1073. The succession to Maine within the Conqeror's family was predetermined by these arrangements and there is no evidence that it was in doubt in 1087 or indeed at any other point in the Conqueror's reign.227 Both this and Professor Le Patourel's arguments assume that Maine had never been part of the Norman lands. Flodard's statement that it was ceded to the Normans in 924 is now generally dismissed,228 but for no very compelling reasons.

B. The Succession to Hugh Bigod 1177 On the death of Hugh Bigod, created earl of Norfolk in 1141, his second wife and widow, Gundreda, step-mother of Hugh's eldest son, Roger, claimed all the dead earl's acquisitions, including his earldom, for her son, Hugh: 'She said that Earl Hugh Bigod devised all his purchases and acquisitions to her son whom he begat of her' ('dicebat enim quod comes Hugo Bigot divisit filio suo quem de ea genuit omnes emptiones et perquisitiones suas'). The word' divisit1 implies a post obitum gift. Roger Bigod in contrast 'offered the king great sums to have his inheritance complete as his father held it'.229 He later obtained a charter from Richard I confirming his succession to the whole estate and title, but the matter was only settled by final concord in 1199.230 Most of the properties at issue fell into two groups: the four manors of Earsham, Halvergate, Acle 225

Two Saxon Chronicles, i, p. 290. Histojia novorum, ed. M. Rule, Rolls Series (London, 1884), p. 25. 227 Orderic Vitalis, ed. Chibnall, ii, pp. 305-11. 228 Annales, ed. P. Lauer (Paris, 1905), p. 24. 229 Gesta Hmrici, ed. W. Stubbs, Rolls Series (London, 1867), i, p. 144. 230 Cartae Antiquae Rolls 11-20, ed. J. Conway Davies, Pipe Roll Society, new series, 33 (1957), no. 554; Curia Regis Rolls, i, p. 93. 226

152

Colonial England, 1066-1215

and Walsham, confirmed to Earl Hugh by Henry II in 1155231 and probably acquired with the earldom in 1141; and Holbrook, Setterington, Hollesley and Staunton, which descended to him from his aunt, Albreda, relict of Berengar de Tosny, on the death of her husband and survivor, Robert de Lisle, c. 1130.232 Despite the fact that the lands in dispute were acquisitions, Roger Bigod was able to establish his right to them and Hugh finally settled for thirty librates of land held of Roger by the service of two knights.233 The case demonstrates the application of Glanvill's doctrine that all land held by knight-service should descend by primogeniture. Nevertheless, the matter was in dispute for twenty years, despite the fact that the arguments of Gundreda and Hugh were by then outmoded. Under Henry II Roger was not recognised as earl and held some of the disputed estates perregem.234

C. The Descent of the Lands of the Beaumont Family

Roger de Beaumont did not take part in the conquest of England in 1066, but was one of those left in charge of Normandy. His two sons, Robert, count of Meulan, and Henry, subsequently earl of Warwick, did. Each of these sons had more than one son. The descent of the family's possessions is complicated and at some points difficult. Roger de Beaumont entered the monastery of Preaux c. 1090 and died some years later. His inherited lands, Beaumont-le-Roger and Pont-Audemer, descended to his eldest son, Robert, count of Meulan. Robert also succeeded to Brionne which Roger had acquired in return for the castellanship of Ivry.235 There were other Norman lands which did not descend in the senior line. These comprised: fees held of the bishopric of Bayeux in St-Vaast and Boulon, which appear in the possession of Roger's younger son, Henry; the barony of Annebecq, which was held by Roger, subsequently by Henry's son, Robert de Neubourg, and presumably also by Henry; and Neubourg, which was a member of the honour of Beaumont and descended to Robert de Neubourg.236 This can only be brought within the current distinctions by assuming, first, that although Brionne was an acquisition it was regarded as part of the family property because Ivry had been surrendered in exchange; and, secondly, that all the Norman lands which descended to Henry were Roger's acquisitions. Neubourg is a special case. The Complete Peerage states that it was granted to Robert de Neubourg by his uncle, Robert, count of Meulan, on the ground that it was in his possession in 1118, 231

Cartae Antiquae Rolls 11-20, no. 553. EYC,i, pp. 466-67. 233 Curia Regis Rolls, i, p. 93. 234 For further discussion see S J. Bailey, The Countess Gundred's Lands', CambridgeLawJournal 10 (1950), pp. 84-103. 235 Compkte Peerage, vii, pp. 522-24. 236 Ibid., xii, pt 2, appendix A, pp. 4-5. 232

Politics and Property in Early Medieval England

153

a year before his father, Henry, died.237 Le Prevost in contrast indicated that it was not originally part of the honour of Beaumont, but was seized from the property of the abbey of Bernay; it might thus have fallen to Henry as an acquisition.238 All this is not impossible, but it would be easier to accept if these were the only peculiarities in the Beaumont succession. They are not, and the possibility that these arrangements were part of a general settlement also involving the descent of the English estates of the family cannot be excluded. Roger and his two sons all acquired property in England. Roger's lay in Dorset (Sturminster), Gloucestershire and Warwickshire (Arlscott). The land in the first two counties descended in the senior line to Robert, count of Meulan, and his eldest son, Waleran.239 Robert was also in possession of Arlscott by 1086 before his father retired to the monastery of Preaux.240 However, this passed with the rest of Robert's extensive holdings in Warwickshire to Henry on or after his creation as earl of Warwick in 1088.241 Hence Robert, the elder son, succeeded to the patrimony and also to Brionne and some of his father's acquisitions in England; Henry succeeded to some Norman estates, which may have been acquisitions, and also acquired the Warwickshire estates of his brother. The possibility that Robert and Henry were twins cannot be excluded, but it is likely that if this had been so it would have led to some comment, especially since Robert's elder sons were twins. In the senior line in the next generation the succession was more straightforward. Robert, count of Meulan and possibly earl of Leicester, d. 1118, was succeeded by his sons Waleran and Robert. They were twins, and this probably accounts for the fact that the proposed division was confirmed by Henry I c. 1107;242 the arrangement was not changed to make any provision for a third son, Hugh 'the Poor', born subsequently. As a result Waleran, the first born, succeeded to Beaumont-le-Roger, the comteof Meulan, Sturminster and other English estates which had belonged to his grandfather, Roger de Beaumont in 1086. Robert succeeded to the Leicestershire lands, most of which Robert of Meulan had acquired from Ivo de Grandmesnil, and became earl of Leicester.243 In the junior line the normal relationship of inheritance and acquisition was

237 238

Ibid.

Ordericus Vitalis, ed. Le Prevost, iv, pp. 327-28n. Complete Peerage, vii, p. 522 note f. 240 Cat. Docs. France, pp. xlix-1, no. 318. 241 Complete Peerage, xii, pt 2, p. 358; VCH, Warwickshire, i, pp. 276-77. 242 RRAN, ii, no. 843. 243 Orderic Vitalis, ed. Chibnall, vi, pp. 18-20,44-46,328-30; G.H. White, 'The Career of Waleran, Count of Meulan and Earl of Worcester, 1104-66', Transactions of the Royal Historical Society, fourth series, 17 (1934), pp. 20-21; Complete Peerage, vii, pp. 522-27; 737-40; VCH, Leicestershire, i, pp. 289-91. 239

154

Colonial England, 1066-1215

reversed. The title and English acquisitions of Henry, earl of Warwick, descended to his eldest son, Roger; his younger son, Robert de Neubourg, succeeded to the Norman possessions which Henry had inherited or received by gift from his father.244

D. The Succession to the Earldom of Chester, 1120-21

It is normally assumed that Ranulf le Meschin had to surrender the lordship of Carlisle as the price of his succession to the earldom of Chester after the death of his cousin, Earl Richard, in the White Ship, but there is in fact no evidence which establishes that the loss of Carlisle coincided with the succession to Chester.245 It is certain that he had to pay a large fine or relief for the earldom, for £1000 was still outstanding against his son in 1130.246 The usual account of Ranulf de Gernon's interest in Carlisle requires revision in the light of the distinction between inheritance and acquisition and the relative weakness of title in the latter. The following points merit consideration: 1

2

3

244

Carlisle was never the patrimony of either Ranulf le Meschin or Ranulf de Gernons,247 Ranulf de Gernons could maintain that his father had held the land, but it had not been inherited; Ranulf le Meschin was established as lord of Carlisle by Henry I;248 he surrendered it before he died and hence it did not descend to his son. Ranulf le Meschin was required to surrender his rights in the inheritance of his wife, Lucy, at the time of his accession to Chester.249 Ranulf de Gernons could claim the most direct interest in these lands which descended to his half-brother, William de Roumare. On the analogy of Carlisle, this might have brought Ranulf and William into conflict. In fact the two worked in close cooperation throughout Stephen's reign.250 The Chester interest in Lucy's lands was recognised after the extinction of the Roumare line in 1198. Ranulf le Meschin was not required to surrender his patrimony in Normandy on succeeding to Chester, viz. the vicomteofthe Bessin. Nor

Complete Peerage, xii, pt 2, pp. 358-61, appendix A, pp. 4-5. H.A. Cronne, 'Ranulf de Gernons, Earl of Chester, 1129-53', Transactions of the Royal Histoiical Society, fourth series, 20 (1937), p. 105; idem, TheReign, of Stephen (London, 1970), p. 175; R.H.C. Davis, 'King Stephen and the Earl of Chester Revised', EHR, 75 (1960), pp. 658-59. 246 Pipe Roll 31 Henry I (ed. Hunter, p. 110). 247 Cp. Cronne, 'Ranulf de Gernons', p. 112; Davis, King Stephen, pp. 49-50. 248 The Place Names of Cumberland, pt 3, ed. A.M. Armstrong et al., English Place-Name Society, 22 (1952), pp. xxiv-xxv. 249 Orderic Vitalis, ed. Chibnall, vi, p. 332. 250 Cronne, 'Ranulf de Gernons', pp. 108-9. 245

Politics and Property in Early Medieval England

4

5

155

was he denied the Chester inheritance in Normandy, viz. the vicomte of the Avranchin. The 'strategic' consequence was to unite the Bessin and the Avranchin in the hands of a single family. There is no hint of any unease between Henry I and le Meschin following the settlement of 1120. Ranulf continued to act until his death as a royal agent in the affairs of Lincolnshire.251 Ranulf de Gernons naturally objected to the intrusion of David of Scotland into Carlisle. If he nursed a grievance it had no stronger base than that his father had surrendered an acquisition in order to gain recognition of a collateral and relatively distant succession to Chester. If so, it illustrates both a wide interpretation of collateral succession and a strict view of title to acquisition. In the previous generation practice was less hard and fast.

E. The Descent of the Honour of Tickhill The death of Roger II de Bully, lord of Tickhill, after 1114, left two claimant lines descended respectively from Roger's sister, Beatrice, wife of William, count of Eu, and from Ernald de Bully, brother of Roger I and ancestor of Idonea de Bully, who married Robert de Vipont (d. 1228). Both Beatrice and Ernald predeceased Roger II de Bully. Robert of Belleme, described as Roger I de Bully's kinsman,252 intruded into Tickhill prior to 1102. After the death of Roger II the honour remained in the hands of the crown. The claims of the counts of Eu were acknowledged at two points: King Stephen granted it to John, count of Eu, who lost it sometime after his capture at Lincoln in 1141 to Ranulf of Chester;253 and King John granted it to Ralph de Lusignan in May 1214 in right of his wife, Alice, countess of Eu.254 Both Stephen's and John's concessions seem to have been bids for political support. The descendants of Ernald held six fees of the honour. They were never apparently able to press their claim in court until Idonea's marriage to Robert de Vipont. Robert, an important agent of King John, obtained the Bully undertenancy in 1213,255 and claimed the honour against Alice, countess of Eu, in 1219-20.256

251

RRAN, ii, passim. Marjorie Chibnall, 'Robert of Belleme and the Castle of Tickhill', in Droitpriveet institutions regionales: etudes historiques offertes djean Yver (Rouen, 1976), pp. 151-56. 253 RRAN, hi, no. 178. 254 Rot. Litt. Pat., p. 116. 255 Rot. Litt. Claus., i, p. 136b. 256 Curia Regis Rolls, ix, pp. 1, 153, 212-13; Rolls of the Justices in Eyre for Yorkshire, 1218-19, ed. Doris M. Stenton, Selden Society, 56 (London, 1937), pp. xxiv-xxv; Pleas before the King or hisJustices, 1198-1202, i, ed. Doris M. Stenton, Selden Society, 67 (London, 1948), p. 68; J. Hunter, South Yorkshire (London, 1828), i, pp. 225-28, 261-63. 252

156

Colonial England, 1066-1215 Note

The above paper led to a debate involving Professor Edmund King and Professor Stephen D. White in Past and Present, 65 (1974), 110-135.1 include the following from my rejoinder: Dr White has made an important contribution to this problem in drawing attention to a number of grants made to a man and his heir. He tends to think that such grants, by restricting the heir to the singular, were limited to the two lives of the grantee and his immediate successor. However, Lennard found no provision for the reversion of the estate or stock in any of the grants made to a man and his heir in phrases which implied perpetuity.257 Moreover, it would surely have been hazardous to allow the issue of perpetuity to depend on the use of the singular or plural. The pressure to convert life-tenancies into hereditary estates was strong enough for men to be aware of the problem.258 Yet they varied the number of the noun quite inexplicably. William, abbot of Holme, granted Tibenham to Roger de Girros 'and his heirs' (1127-34) ,259 but his successor Anselm confirmed it to Roger 'and his heir' 'by hereditary right in perpetuity' (1134-40).26° Abbot Richer of Holme granted Filby to Robert son of Walter 'and his heir' 'by hereditary right' (1101-11),261 but Henry I confirmed it to Robert 'in fee to him and his heirs' (11II).262 Abbot Anselm granted land in South Erpingham and elsewhere to Wither Turnel 'and his heir in fee and inheritance' and in the very same deed recorded that Wither 'and his heirs after his death' would pay the rent.263 To suggest that the singular was intended to indicate a term of lives is to ignore this casual inconsistency of usage. Stenton suggested that a grant to a man and his heir implied that the heir was present at the occasion;264 this is a possible explanation of these phrases. There are others. It may be that the men who drafted these documents sometimes described the heir in the singular because the next succession was the only one which might concern them; the rest was too far in the future, perhaps beyond their lifetime. It may have been that the grantee was known to have an heir but that the heir as yet had no direct successor. It may be that men were thinking within the developing sense of the word inheritance which seems to be indicated in ducal/royal acta and assumed that one lawful succession made an inheritance. It may simply be that by 'heir' in the singular, they meant 'any heir'.265 257

R. Lennard, Rural England, 1086-1135 (Oxford, 1959), p. 173. Ibid., p. 174 n. 3; see also above, pp. 144-45. 259 The Register of the Abbey ofStBenet of Holme, ed. J.R. West, 2 vols, Norfolk Record Society, 2-3 (1932), i, no. 127. 260 Ibid., no. 131. 261 Ibid., no. 121. 262 RRAN, ii, no. 987. 263 Register of the Abbey ofSt Benet of Holme, i, no. 137. 264 Lennard, Rural England, p. 174 n. 1, noting Stenton's suggestion. This is recorded in the case of Darlaston, one of the tenancies of the abbey of Burton, which Abbot Geoffrey (1114-50) granted to Orm and his son Robert. The agreement records that he took homage from both of them. It did not employ any of the conventional language of inheritance, but laid down that they were not to be deprived 'by French or English, or by monks or other men, now or in the future'. The Burton Chartulary, ed. G. Wrottesley, Collections for a History of Staffordshire, William Salt Archaeologial Society, 5, pt 1 (1884), p. 35. 265 This was certainly the case in the charter of Miles, earl of Hereford, which Dr White notes. This refers to succession by an heir or relative. It also permits sale. Charters of the Earldom of Hereford, 10951201, ed. David Walker, Camden Miscellany, 22, Camden Society, fourth series, 1 (1964), no. 7. It is unnecessary to pursue one further possible implication of the use of the singular: that there 258

Politics and Property in Early Medieval England

157

At all events the words might be repeated from one generation to the next, so that the apparent limitation in one grant was extended in a second, Geoffrey, abbot of Burton, granted land in Stretton 'to Orm and his heir in fee and inheritance'. He subsequently granted the same land at the same rent to Ralph son of Orm 'to him and his heir in fee and inheritance'.266 This second charter specifies that the tenancy was a fee farm and it may be significant that many of the examples cited concern land held at rent. Many of the instances at Burton are stated to be fee farms;267 some of those at Holme were made 'in fee and at farm'.268 It seems inescapable that the singular usage, appearing as it does in the acts of at least three monasteries, is connected with the creation of hereditary fee farms. For this, the language of leases drew on the language of enfeoffments. Not one of these grants provides an example of a non-hereditary fief. It would be no surprise to discover such examples for men use words in a contradictory manner to express intentions which themselves conflict. But in all these instances the wordfeodum is used to establish the hereditary element in the grant. There was no doubt about the sense of the word at Ramsey: a fief was land to which there was heritable title; a tenancy which was to revert to the church on the tenant's death was not a fief.269 My argument that feodum was hereditary is reinforced by a passage from Hemming's Cartulary, of 1090-1100, which deserves attention: And these three hundred hides [Oswaldslow] belonged to the demesne of the same church. And however or to whomsoever they were leased - the tenant could not retain the land beyond the completed term agreed between them - nor could he retain it by usurping hereditary right (iure hereditario), claiming it to be his fee (feudam suam), except with the approval (secundum voluntatem) of the bishop and by agreement (secundum conventionem) made with him. Hemingi chartularium eccksiae Wigorniensis, ed. Thomas Hearne (2 vols, Oxford, 1723), i, pp. 287-88.

The ground covered in this paper has now been explored by many scholars. It was written P.M. (pre-Milsom), and it would be an impossible task to adjust it to Milsom's many contributions to the subject and all that has followed therefrom. But it would be the terms of the argument (eg. my use of'property', 'right', and 'title') which I would want to refine rather than alter the argument itself. So here I leave on one side, if only for reasons of space, much scholarly work which followed the appearance of Milsom's The Legal Framework of English Feudalism in 1976. There is further discussion of some of the issues in nos 9,10,11,13 and 14, below. On particular points, the Beaumonts have now been fully treated by David Crouch, The Beaumont Twins (Cambridge, 1986); and the final appendix on the honour of Tickhill was revised in the light of Marjorie Chibnall, 'Robert of Belleme and the Castle of Tickhiir, in Droitprive et institutions regionales: etudes historiques offertes aJean Yver (Rouen, continued

was to be one heir and one only. However, those who believe that the singular created a term of lives are committed by the same logic to the exclusion of coparcenery. 266 Burton Chartulary, p. 31. 267 Ibid., pp. 31-38. 268 Register of the Abbey o/St Benet of Holme, i, nos 124,132. 269 Chronicon Abbatiae Rameseiensis, ed. W.D. Macray, Rolls Series (London, 1886), no. 255: a deed of 1114-30 in which Abbot Reginald conveyed Leofric's land in London to Andrew 'in fee to him and his heir'. In return Andrew granted to the monastery tenements which he retained for life and which were to revert to the monastery on his death. These are not described as a fee but simply as 'lands'.

158

Colonial England, 1066-1215

1976), pp. 151-56, when my essay was reprinted in T.H. Aston, ed., Landlords, Peasants and Politics in Medieval England (Cambridge 1977), p. 114. It is this revised version which is reprinted here. There remains the question of the royal succession of 1087. It will be clear from my comment in the Introduction (above, p. xvi) that I still hold to the views I expressed above pp. 125-26. (It is fair to say that up to his death Professor Le Patourel still held to his). The salient features still seem to me to be that Robert Curthose, William Fs eldest son, succeeded as heir to the duchy of Normandy, that he was never recognised as heir to England, and that William treated England as an acquet which he could direct to Rufus; all this being in line with the conventions of succession and the distinction between propresand acquetswhich were in use in England and Normandy in the late eleventh and twelfth centuries. This argument has now been strengthened massively by Emily Zack Tabuteau's extended examination of the role of the propre/acquet distinction in Normandy and Norman England, particularly in the succession of 1087, in 'Law in the Succession to Normandy and England, 1087', in Haskins Society Journal, 3 (1991), pp. 141-69. Coinciding with Professor Tabuteau's paper Barbara English provided a careful review of recent discussions of the question in 'William I and the Anglo-Norman Succession', in Historical Research, 64 (1991), pp. 221-36. Her further argument, which she advances tentatively, that the regalia which William I left on his death to St Stephen's, Caen, represented 'the dual role of king of England and duke of Normandy' and included a 'quasi-imperial crown' (p. 234), seems to me to be far-fetched. Such a crown would be difficult to reconcile with the view, which she seems to accept, that the Conqueror regarded kingdom and duchy as separate. And why should not the regalia given to St Stephen's simply be a set which William had used in Normandy? For the regalia redeemed by Rufus in 1096-98 was redeemed yet again by Henry I (RRAN, ii, p. 308); so Rufus also must have left them in the custody of St Stephen's. Professor Barlow suggests that the grants 'may only signify the right to borrow them for crown wearings in Normandy', William Rufus (London, 1983), p. 50n. The evidence depends on an official vidimus of 1424 of a charter of William Rufus recording the redemption of the regalia. The document is an awkward one, part narrative, part grant; the Conqueror's gift of the regalia is in the narrative, which was probably derived from the monks of St Stephen's themselves. Whatever one makes of this, (for it could tell either for or against their evidence of a gift), it is generally accepted that the regalia were at St Stephen's and that Rufus redeemed them in 1096-98 with the grant of the manor of Coker (?), Somerset. For the charter see L. Musset, Les actes de Guillaume le Conquerant et de la reine Mathilde pour les abbayes caennaises. Memoires de la societe des antiquaires de Normandie, 37, (1967), pp. 132-34; and for comment see English, 'William I and the Anglo-Norman Succession', p. 232. Perhaps the most striking feature of the story is that the regalia were still at St Stephen's in 1096-98, waiting to be claimed by Rufus. Robert Curthose had apparently made no attempt to get them. This apart there is not much left to be debated. In 1980 R.H.C. Davis demonstrated that Curthose was recognised at least as William's successor and probably as duke, during the Conqueror's lifetime, 'William of Jumieges, Robert Curthose and the Norman Succession', EHR, 95 (1980), pp. 597-606; a view now confirmed by Elisabeth Van Houts in The Gesta Normannorum Ducum of William of Jumieges (Oxford, 1992), i, pp. xxxiv, 13738. Perhaps the only matter still open is the precise form of William's arrangement for Rufus to succeed in England (English, 'William I and the Anglo-Norman Succession', pp. 228-32). The evidence of the monk of Caen on William's death-bed dispositions, which I, like many others, still used in 1972, was severely damaged by LJ. Engels 'De

Politics and Property in Early Medieval England

159

obitu Willelmi duds Normannorum regisque Anglorum: texte, modeles et origine', in Melanges ChristineMohrmann (Utrecht, 1973), pp. 209-255. See also A. Sapir and B. Speet, De obitu Willelmi, Historisch seminarium van de Universitat van Amsterdam, Werkschrift, 10 (Amsterdam 1976), and Van Houts, Gesta Normannorum Ducum, i, pp. Ixiii-lxv. I have let the reference stand, for the exclusion of Robert from the English succession does not depend on this. In any case, I was not intending to suggest, as Tabuteau supposes ('Law in the Succession, p. 155n.), that 'William could have left Normandy to Rufus, but chose at the last moment not to do so', but simply to correlate the De obitu at this point with the Leges Henrid Primi. Le Patourel's Norman Empire (Oxford, 1976), which forms the background to the whole problem, was examined critically by Professor David Bates in 'Normandy and England after 1066', EHR, 104 (1990), pp. 851-80.

This page intentionally left blank

9 Feudal Society and The Family in Early Medieval England, I: The Revolution of 1066

In 1152 King Stephen and his army began an unexpected siege of Newbury Castle. The castellan, John fitz Gilbert, warned by his constable, was able to beguile the king and put the castle in a suitable state of defence. To achieve the necessary breathing space for this he surrendered a younger son, William, as a hostage. When his trickery was discovered the boy's life was at risk; it was proposed that he should be hanged or projected over the castle walls. Stephen sent messengers to John fitz Gilbert, threatening his execution. To them John gave the daunting reply that he was not particularly worried since he had the hammer and the anvil for forging even better sons.1 In fact the boy charmed King Stephen and survived to become earl of Pembroke and regent of England. Whether the tale is anything more than a tall stpry it is impossible to say. At least it presents a problem. Family relationships mattered - or did they? In what ways and to what extent did they matter? These questions are still with us, and perhaps we have answered them in a somewhat haphazard fashion. For example, on the one hand, much has been made in the study of the development of the duchy of Normandy of the relationship of the Norman aristocracy to the ducal house, especially to Gunnor, wife of Richard I;2 again, on a smaller scale and at a later date, some importance has been attached to the family relationships between many of the Twenty-Five barons of Magna Carta;3 and in each of these cases relationships have been woven into an explanatory pattern, as if it were sufficient simply to say that men were kinsfolk and that was that. On the other hand, we seem to accept quite readily that from 1087 to 1106 the sons of the Conqueror were at each other's throats and that the eldest son died a prisoner of the youngest after twentyeight years in captivity; that England suffered nineteen long winters during which the competing claims of the Conqueror's grandson and grand-daughter were thrashed out; that Henry II's sons fought both their father and each other; that King John murdered Arthur, duke of Brittany; and that John's son kept 1

L'histoire de Guillaunw le Marechal, ed. P. Meyer, Societe de 1'Histoire de France (1891-1901), i, pp. 399-594. For the date, see Henry of Huntingdon, Historia Anglorum, ed. T. Arnold, Rolls Series (London, 1879), p. 284. 2 G.H. White, The Sisters and Nieces of Gunnor, Duchess of Normandy', The Genealogist, new series, 27 (1920), pp. 57-65,128-132. 3 P.M. Powicke, Stephen Langton (Oxford, 1928), pp. 207-12; S. Painter, The Reign of King John (Baltimore, 1952), pp. 290-93.

162

Colonial England, 1066-1215

Arthur's sister mewed up until her death at Bristol in 1241. Here again it has been sufficient to know that they were kinsfolk and that was that; the nature of their quarrel is easy to understand. But why then do we assume that in other circumstances family relationships cemented ties rather than started quarrels? What underlay a family alliance? What provoked a family conflict? Even such a cursory examination of the functioning of the family provokes a further, preliminary question. What changes did it undergo? For we should be hard put to it to detect any important differences in what has been written of the family of the Duchess Gunnor, or of the so-called house of Godwin, on the one hand, and the family ties of the house of Clare which allegedly contributed so much to the rebellion of 1215, on the other. Yet it is not on the face of it very likely jthat family structure was the same in pre-Conquest England and in Normandy, the one in Lennard's words 'an old country',4 the other a brash new viking lordship; even less that it underwent no important changes in the century and a half after the arrival of the Normans in England. The tale of John fitz Gilbert has a distinctively patrilineal flavour. It is his son who is under threat of projection over the castle walls, not his mother-in-law, nor his first cousin once removed, nor some vaguely defined kinsman. And John's answer was that he could generate new sons, not that he could turn to other kin who could succour and support him. The answer belonged very much to the twelfth century. It might not have been the same a century earlier. To go back through that century, back beyond the Conquest of 1066, is to move to more familiar ground. Most of us at some stage will have read something, or perhaps painfully written something, on kinship in Anglo-Saxon society and on the process whereby the kin's responsibility for its members was transferred to lords, or to local associations, or to local courts into which the authority of the king increasingly intruded. Indeed the relationship of kinship to lordship and the measures whereby kings brought the responsibility of the kin under various forms of legal control are two of the set pieces of the social and constitutional history of the period.5 Now, why we have looked at the family and the kin during the Anglo-Saxon period mainly in this light I am not entirely clear, but I am less concerned with the causes than with the results. We bring our students to trace the relationship between kinship and lordship and the bearing on both of the authority of the king with some zest through to the Norman Conquest, and then suddenly the topic vanishes. True, some of us, raised in the hard school of Stubbs's Charters, learned all about liege lordship and even of the frankpledge system, but we would probably agree that they have not proved to be growth-points in our understanding of the AngloNorman period in recent years. For those without this advantage, one of the main themes of the Anglo-Saxon period simply fades away, with two effects. 4

R. Lennard, Rural England, 1086-1135 (Oxford, 1959), pp. 1-21. Recent studies are discussed by H.R. Loyn, 'Kinship in Anglo-Saxon England', Anglo-Saxon England, 3 (1974), pp. 197-209. See also T.M. Charles-Edwards, 'Kinship, Status and the Origins of the Hide', Past and Presmt, 56 (1972), pp. 3-33. 5

The Revolution of 1066

163

First, although there have been many family histories for the Norman period and though much is sometimes attributed to the tie of kinship, there is very little in the way of diagnosis of what family and kin meant. To set against the extensive literature on the Anglo-Saxon kin there is indeed but one article on the Norman period, written by Sidney Painter, thirty-five years ago,6 although since then much that has been written on the Norman Conquest has reflected the influence of Georges Duby and other European scholars who have been immediately concerned with aspects of the history of the family. Secondly, and curiously, this deficiency has tended to diminish rather than exaggerate the importance of the break in 1066. In most interpretations kinship had been overtaken and overborne by lordship and the authority of the king well before 1066. The well-known argument about continuity has therefore come to turn on the comparison of the notions about lordship as practised by the AngloScandinavian and the Norman nobility in the eleventh century. As at Hastings, the conflict takes place on a very narrow front. Now this is not the only way to look at kinship, and I intend first to look at another, namely at the relationship of the family or kin to its property, an anachronistic word which I use without any implications as to the nature of title, but simply as a useful term describing all those rights in land or other benefits which members of a kin exercised in full or in part. This is really to ask how the family or kin defined and sustained itself. For the Anglo-Saxons, part of the answer is revealed on those occasions when men gave the most serious of all consideration to the disposition of their rights, namely in the post obitum gifts in which they made provision for their own deaths. Their wills establish two points. First, testamentary bequest of land was common, and, secondly, the will was used to distribute both land and other forms of wealth not only to the testator's children, but also to collaterals and to others whose relationship was undefined.7 A typical example is the will of ^Elfhelm,8 minister of King Edgar, benefactor of Ely who is described in the Liber Eliensis as 'miles divitiis abundans'.9 Its date limits are 975 to 1016, but it most probably comes from c. 990: And I grant to my son ^Elfgar the estate at Whepstead and that at Walton for his lifetime, and after his death they are to go wherever he pleases, for the souls of both of us. And I declare what I gave to my wife as a marriage*gift, namely, Baddow and Burstead and Stratford and the three hides at Enhale. And when we first came together, I gave her the two hides at Wilbraham, and Rayne and whatever pertains to it. And 6 S. Painter, The Family and the Feudal System in Twelfth-Century England', Speculum, 35 (1960),pp. 1-16; reprinted in Feudalism and Liberty, ed. Fred A. Cazel,Jr (Baltimore, 1961),pp. 195219. 7 H.D. Hazeltine in Anglo-Saxon Wills, ed. Dorothy Whitelock (Cambridge, 1930), pp. vii-xl; M.M. Sheehan, The Will in Medieval England (Toronto, 1963), especially pp. 74-76,83-99; Lorraine Lancaster, 'Kinship in Anglo-Saxon Society', British Journal of Sociology, 9 (1958), pp. 230-50, 35977, and especially 359-67. 8 Anglo-Saxon Wills, no. xiii. 9 Liber Eliensis, ed. E.G. Blake, Camden, fourth series, 92 (1962), p. 143.

164

Colonial England, 1066-1215

I grant her Carlton and I grant her the chief messuage at Gestingthorpe, and all the possessions that are on it, including produce and men; but I grant to Godric and my daughter half the woodland and open land, except that which I grant to my priest. And I grant to my wife and my daughter half the estate at Conington, to divide between them, except the four hides which I grant to ^Ethelric and ^Ifwold, and the half hide which I grant to my servant Osmaer. And I grant to ^Elfmaer and his brother ^Elfstan, to divide between them the two estates, Hatley and Potton, except what I grant to Osgar. And I grant to Godhere what I bought from Wimund. And I grant Littlebury to Leofsige after my death, on condition that the agreement which we concluded before the ealdorman shall hold good. And I grant to him and his wife the estate at Stockton for a hundred mancuses of gold, and I wish that the gold be given to my lord in payment of my heriot. And I grant to be divided among my three brothers the estate at Troston, except that I grant to ^Elfwold that which ^Ethelric had. And I grant ^Ifhelm the hide at Ickleton and the property at Maworth. And I grant Wulfmaer what I had at Barnham.

Professor Whitelock has pieced together the relationship of some of the legatees to the testator.10 The legatee called ^Elfhelm was probably the testator's brother. Athene and ^Elfwold were the sons of this second ^Elfhelm and hence the testator's nephews. A Leofsige is recorded as the testator's kinsman in the chronicle of Ramsey.11 Whether the remaining legatees, ^Elfmaer, ^Elfstan and Godhere and Wulfmaer, were kinsmen is not clear. Some other difficulties also cannot be resolved. We may wonder whether ^Elfhelm's son, ^Elfgar, was legitimate. His appearance as the first secular legatee suggests in fact that he was. We may also ask whether the will disposed of all ^Elfhelm's property and whether there were other estates descending under different rules for which the will made no provision, and this will lead us immediately to the familiar debate about bookland and folkland.12 But for the present purposes this does not matter because the message of ^Elfhem's will is perfectly clear. A testator assumed that the land and rights of which he could dispose should properly be distributed within the kin, and the kin for this purpose included not only his son and widow, but his daughter and son-in-law, brother and nephews, and other kinsmen whose relationship was not defined. In short, in each generation such property as could be bequeathed was distributed laterally through the surviving branches of the family. Only a portion of it descended vertically to the testator's immediate male descendants. How far this was done in the knowledge that there were other lands, not comprised in the will, which would descend under other rules to the testator's children we cannot know. We do know that where such rules took effect they allowed for the division of the decedent's lands among his legitimate

10

Anglo-Saxon Wills, p. 134. Cf. Liber Eliensis, p. 143. Chronicon abbatiae Rameseiensis, ed. W.D. Macray, Rolls Series (London, 1886), p. 193. 12 F.M. Stenton, Anglo-Saxon England (3rd edn, Oxford, 1971), pp. 306-12; E.John, Land Tenure in Early England (Leicester, 1964), pp. 1-63; idem, Orbis Britanniae (Leicester, 1966), pp. 64-127. 11

The Revolution of 1066

165

male offspring, whence the lands held jointly by brothers which are recorded T.R.E. in Domesday Book.13 Now, whatever view is taken of the relationship of bookland to folkland, and of the proportion of each which a man of ^Elfhelm's importance might hold, it is plain that the relationships revealed in this will were very different from those which quickly took root in the Norman colony of England after 1066. There is still much to be said for the old view that the Conquest saw the end of the testamentary devise of land. Against it Maitland was certainly correct in pointing out that the post obitum gift was known in Normandy before, and in England after, the Conquest;14 and his point has been reinforced by Professor Sheehan.15 But it is important to penetrate behind the formal act of bequest. The argument that 'the situation in England was much the same as that in Normandy' is not substantiated by testamentary bequests and post obitum gifts, whether from Normandy before, or England after, 1066, which are, in effect or in express intent, grants in alms.16 This merely obscures changes in the capacity to bequeath to laymen, in particular to kinsmen. Moreover, it is a mistake, indeed an anachronism, to conceive these changes in legal terms, as an aspect of 'freedom of alienation'. They hinged not on testamentary power, but on the canalisation of the inheritance to a restricted category of heirs; the limitation of testamentary power was a consequence, not a cause. The social conventions of the Anglo-Saxons allowed, even required, the testamentary distribution of land throughout the kin: the social conventions of the Normans did not.17 The Anglo-Saxons might bequeath land to a direct descendant:18 the Normans did 13

F.W. Maitland, Domesday Book and Beyond (Cambridge, 1897), pp. 145-46; F. Pollock and F.W. Maitland, The History of English Law (Cambridge, 1898), ii, pp. 263-64. 14 Ibid., i, pp. 343-44; ii, pp. 326-27. 15 Sheehan, The Will in Medieval England, pp. 266-78. 16 Such is the evidence advanced by Professor Sheehan (ibid., pp. 108-9) and the solitary instance mentioned by Maitland (History of English Law, ii, p. 323 n. 1). Maitland's further discussion of post obitum gifts was also restricted to grants in alms (ibid., ii, pp. 326-27). The arguments of Genestal, 'L'interdiction du legs d'immeuble', Revue historique de droit franfais et etranger, fourth series, 7 (1928), pp. 683-84, on which Professor Sheehan relies, are scarcely germane to the question since they are concerned with fifteenth-century limitations placed on testamentary dispositions of land as they had developed in the thirteenth century. As far as it is possible to judge from a short compte rendu, Genestal was not in doubt about earlier circumstances: 'Primitivement les immeubles furent indisponibles sans le consentement du seigneur et des heritiers', ibid., p. 684. 17 Once it is allowed that Norman and English practice differed in this way, the problem to which Maitland turned (History of English Law, ii, pp. 327-28) and which so exercised Professor Sheehan (The Will in Medieval England, pp. 266-76), namely that testamentary disposition of land was apparently restricted at a time when freedom of alienation was increased, is vastly reduced and is rendered in large measure fictitious. Glanvill's strictures against death-bed bequests, De legibus, ed. G.D.G. Hall (London and Edinburgh, 1965), p. 70, are still best read as a protection against ecclesiastical greed, as Maitland interpreted them, rather than as part of a more sweeping substantive and procedural change, as Professor Sheehan suggests. 18 See above, p. 163. Compare 'And Leofsige is to have the land at Weedon ... If he have a child, he is to give it to that child; if he have not, he is to give it in his lawful kindred', The Will of dEthelgifu, translated and examined b^ Dorothy Whitelock, Roxburghe Club (1968), p. 10.

166

Colonial England, 1066-1215

not need to, for with them that was how land descended, with the lord's approval, in the natural course of events. They exercised some testamentary initiative. They made post obitum gifts in alms; they could exercise some choice, to which I shall return, in bequeathing acquired land;19 but they did not, like the AngloSaxons, disperse the patrimony within the kin. Testamentary disposition of the patrimony was rare, largely restricted to the allocation of the patrimony and the acquisitions within the next generation where the acquisitions were the larger.20 Divisions of the patrimony among sons were few and far between. They required special confirmation of the king or lord; one famous instance in the Beaumont family was a provision for twins.21 Very soon primogeniture became the rule. The new arrangements are summed up in the coronation charter of Henry I where they already seem to enjoy the status of basic premises. The king's earls and barons are succeeded by their heirs and so also by implication are their tenants. It is their chattels, or more broadly their pecunia, which are subject to testamentary disposition.22 Continental scholars would have a ready explanation of these changes. It would be one which turned on a change in the structure of the family, from one conceived as a political, social and legal grouping in which responsibility was expressed through the feud and the oath, in which resources were distributed throughout the kin and in which the head was in some degree the chosen leader of a clan, to one conceived as husband, wife and children in which resources were passed on from one generation to the next, the younger children receiving pot-luck, if their father so decided or if custom required, from such acquisitions as might be available, or benefiting from some form of partition or from arrangements such as parage in Normandy, which apportioned them part of the family holding, but left them in tenurial dependence on the senior line.23 The difference is between clan and lineage, between the extended and the

19

See below, pp. 171-72. See above, p. 124. 21 Ibid., pp. 119-20, nn. 47-49; p. 153. 22 Cc. 2, 7. The importance of these provisions was succinctly stated by Plucknett: The charter of Henry I, c. 7, of 1100 accords freedom of testation of chattels only; so it would seem that land was already indevisable by that date', T.F.T. Plucknett, A Concise History of the Common Law (4th edn, London, 1947), p. 698n. 23 See the works of Georges Duby noted above, p. 114 n. 16. Some of these papers are collected in Georges Duby, Hommes et structures (Paris, 1973) and translated by Cynthia Postan in Georges Duby, The Chivalrous Society (London, 1977). Among them 'Lignage, noblesse et chevalerie au Xlle siecle dans la region maconnaise: une revision', Annales, 27 (1972), pp. 803-23; Hommes et structures, pp. 397-422, is particularly important in demonstrating that lineage did not necessarily involve primogeniture. The change I have sketched is perhaps presented most forcefully in R. Fossier, La terre et les hommes en Picardie (Louvain, 1968), ii, pp. 534-46. See also K. Schmid, 'Uber die Struktur des Adels in fruheren Mitte\zher\JahrbuchfurfrankischeLandesfoischung, 19 (1959), pp. 1-23, trans. T. Reuter, The Medieval Nobility (Amsterdam, 1979), pp. 39-59; and the cautionary remarks of Karl Leyser, The German Aristocracy in the Early Middle Ages', Past and Present, 41 (1968), p. 34. On Normandy, L. Musset 'L'aristocratic normande au Xle siecle', La noblesse au moyen age, ed. P. Contamine (Paris, 1976), pp. 71-96, supplements the earlier work of D.C. Douglas. 20

The Revolution of 1066

167

dynastic family. The pity is that continental scholars have apparently felt unable to tap the rich resources of the English evidence, the land-books, the wills, the Domesday Survey, the pipe rolls, and that English scholars have not seized the opportunity the evidence affords, for the Norman Conquest presents the change not as the relatively gradual process which bedevils much of the continental evidence, but as a sharp antithesis, the new confronting the old across the divide of 1066. If this was the case, the Norman Conquest must be seen as involving not simply the replacement of one aristocracy by another but also the replacement of one set of family relationships by another, a change not merely in personnel, not merely in all those external relations of the aristocracy which are the classic ground for the debate about continuity, but a change in internal organisation, in familial structure, in assumptions about property. Indications of some such change lie to hand. How far the Conquest led to a contraction in kinship nomenclature, especially as applied to collaterals, I leave on one side.24 More important is the manifest change which took place in family nomenclature.25 The Normans introduced hereditary surnames, above all hereditary toponymic surnames which attached the family, its history and its fortunes, to its property. This has long been recognised. It has been repeated almost ad nauseam from Bardsley's classic study of English surnames of 1873 onwards, and there is no real doubt about it. No convincing example of a hereditary Anglo-Saxon surname had ever been advanced for the preConquest period, except where an occupational surname descended with the occupation from ancestor to successor. Moreover, it is plain that the toponymic, the probable point of origin of the hereditary surname, was nothing more than a casual occurrence in old English family nomenclature; it was used where other associations, local and administrative, cut across family structure. Hence it appears in the main in attestations and warranties. In the wills it scarcely figures at all; indeed it might be seen as a positive impediment to testamentary initiative. The plain fact is that the Anglo-Saxon aristocracy did not need hereditary surnames, least of all hereditary toponymic surnames. The Normans did. With them the toponymic was not just name but title, title to be preserved by all the descendants of the founder of the family's fortunes, only to be surrendered in the junior branches if some new tide, the consequence of marriage or the accidents of succession, or the exercise of royal patronage, became available. This can be illustrated in multifarious fashion: by the principles underlying Lewis Loyd's Anglo-Norman Families, in which, time and again, names and tenure 24 The Anglo-Saxons distinguished father's brother from mother's brother, father's sister from mother's sister, and brother's son from sister's son, Lancaster, 'Kinship in Anglo-Saxon Society', pp. 232-29; Loyn, 'Kinship in Anglo-Saxon England', p. 198. In the Latin documents of the Normans all this was encompassed in a few broad Latin terms, avunculus, nepos, neptis. However, the transition from the vernacular to Latin intervenes between the two. Quite apart from the possible perpetuation of archaisms, this makes comparison hazardous. 25 For what follows see below, pp. 179-96.

168

Colonial England, 1066-1215

can be brought together to demonstrate that families did indeed originate and hold land in the places indicated by their inherited toponymic; by the readiness of the descendants of the 'new men' of Henry I to dress themselves up in old names as they acquired the property which went with them; and, occasionally, by more direct, immediate evidence. No document illustrates this new aristocratic world more convincingly than the record of the sale by Roger de Bully to the abbot of La Trinite du Mont, Rouen, of the tithe of Bully, 105166. This establishes beyond doubt that he was called de Bully, that he held land in Bully, that he held it by hereditary right and that both sale and title were acknowledged by Duke William who was a witness and signatory to the transaction.26 That epitomises the change which was about to be imposed on England after 1066. It was a change of crucial and varied importance. It was closely linked to the establishment of feudal military institutions, for these would have been impermanent had not the resources descended largely unimpaired and unfragmented from ancestors to heirs. It coincided with changes in attitudes towards marriage and the status of women. It led to a new style of monastic endowment. It contributed to a changed pattern of investment. In their wills, the AngloSaxons bequeathed rich military equipment and gold and silver to their lord and kin. ^Elfhelm indeed had a goldsmith to whom he granted land in Potton. In his world, part of a man's wealth was held in a form readily distributable to members of the kin. The Norman, in contrast, left to his children land, military resources of household and enfeoffed knights, building, which almost inevitably involved the successor in more building, and only too frequently, debt. It will not do to pretend that the Norman magnate saw himself as an entrepreneur founding a family concern for his children. But his children saw him in something like that light since, for what they enjoyed, they looked back to their father, or to their father's father, as the ultimate founder of the family's fortunes. That is reflected not simply in the perpetuation of the family name, but also in the recurrent use in succeeding generations of the founder's Christian name, in the repeated confirmations of his monastic endowments and, here and there, in direct evidence. Roger of Montgomery, earl of Shrewsbury, cut a fine figure in England, but in his foundation charter of Troarn he was but the son of 'great Roger' who had first acquired Montgomery.27 I have deliberately sharpened the contrast between the English and the Norman, the old and the new. In reality it was not quite so clear cut. Dr CharlesEdwards has rightly pointed out that 'kinship based on lineage membership' and 'kinship based on relationships between persons' are not mutually exclusive.28 Indeed, it is impossible to consider any relationship more complex than those of the nuclear family-father, mother/children -without calling on both notions. 26

RADN, no. 200. 'Ego Rogerius ex northmannis northmannus magni autem Rogerii films', BN, MS lat. 10086, fo. ii verso. 28 Charles-Edwards, 'Kinship, Status', pp. 29-30, n. 50; compare p. 21 n. 35. 27

The Revolution of 1066

169

However we adjust our terminology,29 cousins remain related through lineal descent from grandparents. The problem, therefore, is one of emphasis, and if in examining the emphasis which the Anglo-Saxons placed on their relationships it is right to make comparisons with Celtic, Germanic and Scandinavian practice, which Professor Loyn and Dr Charles-Edwards have so skilfully done,30 it is equally right, and perhaps more important, to compare the Anglo-Saxon with the Norman. This second comparison highlights succession, which is less important in the comparisons between Anglo-Saxon, Scandinavian and Celt,31 and underlines family nomenclature which has scarcely figured at all in these earlier discussions. Yet it is arguable that family nomenclature takes us closer than do the laws to the everyday mentaliteof the late Anglo-Saxon period. Quite apart from these questions of interpretation, it is likely that in time, in all probability, Anglo-Saxon England would have invented its own version of feudal institutions. Indeed it was already moving in that direction, for laen-land, either in the characteristic form revealed in the Worcester leases for three lives or in any other form which secured reversion to a lord, inhibited the fragmentation of estates; similar leases marked the spread of dependent tenures on ecclesiastical estates in northern France.32 It may also be that the late AngloScandinavian earldoms, if left in undisturbed development, would have produced a base for the establishment of secure aristocratic lineages. Certainly, in practice, the gap between English and Norman was not so wide or unfamiliar that it could not be spanned by those Normans who established themselves in England under Edward the Confessor and by those English Quislings who, like Colswein of Lincoln, acquired a barony for themselves in the years following the Conquest. But the reasons for this are not to be found solely on the English side. England was changing, but Normandy had changed even faster. One of the difficulties in the way of appreciating the importance of the family and tenurial relationships which the Normans imposed on England is that in 1066 these relationships can scarcely have been more than a generation old. This is easily demonstrated. Roger de Beaumont, Domesday lord of Sturminster Marshal, father of the sons in whose lines the earldoms of Leicester and Warwick descended, was the first of his family to succeed to Beaumont-leRoger and the first to use the name. Roger of Montgomery, earl of Shrewsbury, was the first to succeed to Montgomery and probably the first to use his name. William de Warenne, first earl of Surrey, was probably the first of his family to succeed to Varenne, although his father certainly used the surname. There is a recurrent pattern in the evidence which establishes very clearly that, except 29

See, for example, Charles-Edwards: The descent group... I shall call a lineage', ibid., p. 16. Whatever the merits of such a definition, it is far wider than the lignage of Georges Duby and others. 30 Charles-Edwards, 'Kinship, Status', and especially Loyn, 'Kinship in Anglo-Saxon England', pp. 197-98. 31 Professor Loyn discusses succession briefly, ibid., pp. 201, and there is a more extended discussion in Lancaster, 'Kinship in Anglo-Saxon Society', pp. 359-67. 32 See above, p. 140 n. 161.

170

Colonial England, 1066-1215

in a very few cases of which the Tosny family provides the leading example, the companions of the Conqueror were the first of their families to have succeeded to the landed estates from which many of them derived their titles.33 Roger de Bully described his tenure in Bully as iure hereditario. 'Heritable right' are strong words to apply to a succession which was probably no more than a generation old. Yet apply them he, or his scribe or the scribe of La Trinite du Mont, did. In so doing he converted fact into tide. And that conversion was taking place wherever a family adopted or was given a toponymic name. A generation later it seemed a simple enough story. The genealogies which Robert of Torigny added c. 1139 to the History of William of Jumieges present a picture of Norman aristocratic lineages stemming from the sisters and nieces of the Duchess Gunnor, wife of Duke Richard I, descending with inevitable certainty to Robert's own contemporaries.34 But he selected those which happened to have come through. The process left casualties, like William, count of Arques or Gilbert, count of Brionne.35 Taken on its own terms within its own generation the position of Gunnor's relatives seems simply to reflect an attempt to secure ducal control of Normandy by distributing resources and local power throughout the ducal kin. It is surely very unlikely that anyone at the time saw it as a means of initiating the lines of Montfort, Giffard, Clare, Beaumont, Montgomery, Warenne and fitz Osbern, to name only the most immediate direct and collateral descendants of Duke Richard and his wife.36 Yet that is how it soon came to appear, for by the first half of the twelfth century the milieu had changed. Compare the story of the progeny of Gunnor and her sisters with the history of the serried ranks of illegitimate children of Henry I. One, Robert earl of Gloucester, owed his territorial position to marriage with an heiress and cut a great figure. Another, Reginald, became earl of Cornwall. Another was married to the count of Perche. But no one would pretend that Henry's security on the throne of England depended on advancing his mistresses' siblings or promoting their offspring. The lines along which patronage was exercised had altered. It was no longer based on collateral lines or illegitimate relationships. Instead Henry 'raised men from the dust'. He established new lineages. Between Gunnor's relatives and Henry I's 'new men' there was a state of flux. The only certain point within it is that during William's reign as duke of Normandy the coincidence of lineage, family property and hereditary surname became more common. But in individual cases the claims of lineage might be acknowledged or rejected, sustained or forgotten. The sons and grandsons of Gilbert of Brionne nursed a claim to Brionne which, if Orderic Vitalis is to be believed, bowed to the duke's right of recovery so long as he did not alienate 33

See below, pp. 186-89. William of Jumieges, Gesta Normannorum ducum, ed. J. Marx, Societe de 1'histoire de Normandie (1914), pp. 224-29. 35 D.C. Douglas, William the Conqueror (London, 1964), pp. 40, 62-66. 36 White, The Sisters and Nieces of Gunnor, Duchess of Normandy'. 34

The Revolution of 1066

171

to another party.37 But they were not equally vigorous in pressing claims to the county of Eu which both Gilbert and his father has possessed. In any case these grievances were submerged in the good fortune of the house of Clare in England. At a later date a claim was often better remembered. The confused situation which followed the death of Roger de Bully late in Rufus's reign left Robert de Belleme, allegedly de Bully's cousin, in charge of Tickhill Castle. Ultimately the estate descended through Roger's female relative, his sister or his daughter, to the counts of Eu, but the descendants of his younger brother, themselves tenants of the honour of Tickhill in Bawtry and elsewhere, were still ready in the fourth generation to press their claim in the royal courts when the moment seemed opportune in 1219-20.38 That was tide long remembered, and the twelfth century was to yield many other examples. But it should not be thought that in these matters the Norman kings wilfully rejected lawful claims. What they did was to retain initiative in circumstances in which the claim in lineage was still by no means absolute and in which, therefore, it was still possible to choose among claimants on ground other than lineal descent, especially where there were genealogical complications. Such freedom of action blurred the contrast between old English and AngloNorman circumstances. It was further obscured by the legal corollaries of conquest. Whatever notions of hereditary succession were taking shape in the eleventh century they applied in the first place to land and rights which had already been inherited. In England, in the first generation after the Conquest, almost none had. Practically all the Anglo-Norman baronies were initially acquisitions or conquests, and these were subject to disposition by the tenant, who often sought to distribute his estates among his sons, subject to the approval of the feudal lord, the king. This left room for a limited testamentary initiative and for a freer-ranging choice of heir by the lord, which at first sight seem to mimic Anglo-Saxon practice. And, to be sure, it may help to explain how it was possible to contract the distribution of landed wealth within the Norman family within the space of two or three generations, for Norman expansion provided an enormous number of acquisitions to satisfy the ambitions of brothers and younger sons. Moreover accident took a hand. There was no neatly drawn line at the end of the first generation of Norman settlement after which acquisitions became inheritances. The consequences of the Conqueror's testament, the prolonged quarrel among his sons, the disputes of Stephen's reign, led to massive deprivations and a constant flow of new acquisitions which was scarcely staunched before the accession of Henry Plantagenet in 1154. Even after that, the claim of the younger son in the acquisition was still seriously advanced. On the death of Hugh Bigod, earl of Norfolk, in 1177, his second wife and widow took to the courts in the interest of her son, arguing that all Earl Hugh's acquisi37

Orderic Vitalis, ed. Chibnall, iv, pp. 206-9. Marjorie Chibnall, 'Robert de Belleme et le chateau de Tickhill', Droit piive et institutions regionaks: etudes historiques offertes a Jean Yver (Paris, 1976), pp. 151-56; The Cartulary ofBlyth Priory, ed. R.T. Timson, Thoroton Society, record series, 27 (1973), i, p. xix. See also above, p. 155. 38

172

Colonial England, 1066-1215

tions, which included very extensive additions to his East Anglian holdings obtained in the reign of Stephen and the title to the earldom of Norfolk itself, should all descend to her own son and not in the elder line derived from Hugh's first marriage. The lands were taken into the King's hands and the title was put in abeyance until Richard I confirmed it to the elder line in 1189.39 Some of the lands remained in dispute, compromise settlements were reached,40 but the two parties in the action took opposite sides in the civil war of 1215. Earl Roger Bigod was a leading rebel; William Bigod was a king's man.41 Here, as with many other problems, an argument which starts from the differences between Anglo-Saxon and Anglo-Norman institutions ends with the differences between Norman England and continental Europe. The proliferation of acquisitions through the first ninety years of Norman rule in England was one of the points where English feudal society parted company from the French homeland. To sum it up, England was a colonial country, a notion which Englishmen find difficult to grasp because on this occasion they were on the receiving end, and one surprisingly which has not been seized by American students of medieval England. As a result aristocratic society and the landed family did not develop quite along the lines preordained in continental Europe. England became markedly sui generis. As a Norman colony, it was a land of enterprise where society was relatively untrammelled, where young men, usually of respectable origins but frequently of moderate means, could rise high in the social hierarchy through military, political or administrative service, where, in a sense, all the invaders were equal in the act of conquest. This bears on the continental model of the new feudal family at two points; first on the emphasis given to nobility of lineage, and secondly, within that, to the importance apparently attached in some continental genealogies to ancestors in the female line as originating the accession of new noble blood. In England genealogical interest took a rather different line. My first witness is a sadly neglected source which has never really recovered from Round's attack on Freeman. In fact, it was not entirely Wace's fault that Freeman misused him, But his Roman deRou, written between 1160 and 1170, is of great interest for my present purpose; for what Wace set out to do was to attribute special prestige to particular Anglo-Norman families, not simply through any special claim to noble descent which they might have, but through their participation in the Conquest of 1066.42 Wace's information was second-hand, much of his genealogy was vague, there are details, which as a clerk of Caen, he should have known better. It was all a piece of window-dressing concocted 39

See above, pp. 151-52. Curia Regis Rolls, i, 93. 41 Rot. Litt. Pat., p. 186b; Rot. Litt. Claus., i, 262, 264b. Earl Roger and William Bigod were halfbrothers. For the relationship, see Complete Peerage, ix, p. 586. 42 LeRoman deRou de Wace, ed. AJ. Holden (Paris, 1970-73), ii, pp. 193-208. For Round's criticism of Freeman, see Feudal England, pp. 399-418. 40

The Revolution of 1066

173

for Henry II who was busy presenting himself as the heir to, and patron of, the great Norman historical tradition. But there can be no doubt about the nature of the display. Participation in the Conquest was the great accolade; the family ancestor was there rather as the Prince Regent was at Waterloo. Wace, of course is only one authority, and an interesting rather than reliable one. There are others which throw a different light: the pedigrees which Robert of Torigny interpolated in William of Jumieges reveal an interest in tracing family descent to the early generations of the ducal house;43 Orderic Vitalis noted that Eustace of Boulogne was descended from Charlemagne;44 and Geoffrey of Monmouth claimed similar descent for Waleran, count of Meulan.45 Beyond that, the Norman Scandinavians, like the Franks, absorbed or invented fictional descents which took them back to the legends of Troy.46 But Wace illustrates the legend of the Conquest well enough and it reappears in a different aspect in many other sources. At Battle it was particularly strong, for the abbey was founded to commemorate the great victory of 1066. This is how the Battle annalist saw one of Henry IPs justices, Richard de Lucy, looking back over the previous century: Lord King, this church should be extolled by you and by all us Normans, for it was there that the renowned King William, by the will of God, and with the aid and counsel of our kinsmen, overcame his foes who sought to deprive him of the realm and crown of England, and there he acquired the realm and crown for himself and for his successors. All the people of this realm rejoice that through close consanguinity to him and hereditary right you now reign from his throne, while we enjoy abundant possessions and riches through the gifts which he conferred and by succession to our kinsmen. Therefore, my lord, most excellent of kings, all this gathering of Norman nobles begs and prays that your royal sternness will protect that place, as the monument of your and our triumph, in its proper dignity and liberty against all its adversaries, and above all against the wiles of the English 47

Now some of Richard's waxing enthusiasm may be explained by the fact that the reigning abbot of Battle was his brother. But this sense of the unity of the history of the Normans in England back to the Conquest and no further must have been present in all the new foundations of the Norman period. Moreover it was reinforced repeatedly in a quite different and more practical and mundane context. In the courts of law, the most ancient tenure which might be claimed was tenure a conquestu, and it was still advanced as a rhetorical support for other evidence or as a hopeful substitute for the lack of evidence even in the late thirteenth century. Thus far and no further: 1066 was the beginning of an era. 43

William of Jumieges, Gesta Nvrmannorum ducum, pp. 324—29. Orderic Vitalis, ed. Chibnall, ii, p. 206. 45 Historia regum Britanniae, ed. A. Griscom (London, 1929), p. 220. 46 Dudo of St-Quentin, De moribus et actis primorum Normanniae ducum, ed. J. Lair (Caen, 1865), p. 130; H. Prentout, Etude critique surDudon de Saint-Quentin (Caen, 1915), pp. 44-46, 443-45. 47 The Chronicle, of Battle. Abbey, ed. Eleanor Searle (Oxford, 1980), p. 182. 44

174

Colonial England, 1066-1215

Now this was fortified, rather than diluted, by events between the Conquest and the beginning of the thirteenth century. To my suggestion that all the Normans were in a sense equal through their participation in the Conquest, one might hear the shades of Robert of Mortain, or William fitz Osbern answer - 'Ah yes, but some were more equal than others'. True, but where after the next fifty years were the heirs to Robert of Mortain and William fitz Osbern? Again, Orderic Vitalis might provide for the great house of Tosny a descent going back to the uncle of the first Norman duke, Hrolfr,48 but where, by the middle of the twelfth century, were the Tosny lands in England? Largely divided and dispersed. By this time too, the families of Mortain and Montgomery had gone, and others followed. By the later years of John's reign even the great house of Beaumont was broken, dispersed, a mere shadow of its former potency. The senior French line was represented by an exiled count of Meulan, now a mere pensioner of the king;49 in the senior English line there were no male heirs and the lands were to descend through co-heiresses.50 Nothing remained of the former pomp but the junior English line, the faineant earls of Warwick. To sum up, one of the first things to grasp about the Norman colony of England was that at the aristocratic level it was a very fluid society. By the beginning of the thirteenth century very few major families could trace their fortune back beyond the Norman Conquest to men of power and influence in ducal Normandy. Indeed of the twenty-one heads of families among the Twenty-Five barons of Magna Carta only five, Clare, Mandeville, Bigod, Vere and fitz Walter, go back to major tenants-in-chief in Domesday Book, and one of these, Geoffrey de Mandeville, was the son of a parvenu whose real origin lies hidden in a change of name. Of the remaining four, only the Clares, descendants of Gilbert, count of Brionne, had made any real mark in Normandy before 1066. Now for these reasons, the concern with noblesse, with nobility of blood and lineage, which is a feature of the continental genealogies and of the modern interpretation of them, seems less apposite to England. Norman families, and Norman writers, William of Jumieges, Orderic, Robert of Torigny, partook of the continental fashion, but it was not very relevant to circumstances in England and rapidly became out of place as the older families disappeared from the scene and new men were 'raised from the dust'. Indeed in the end in England, nobility was defined not in terms of blood and lineage, but tenurially and administratively. To be sure, the sense of lineage survived, much dependent on fiction, and it grew as time passed. Its increasing strength was marked in many ways: in the growth of heraldry; in the continuing links between the family and the monastic house or houses under its patronage; in the resistance to the newcomer, which was already strongly marked by the beginning of the thirteenth century in the outcry against disparagement; and in the increasingly secure control which families exercised over the marriage of heiresses in the course 48

William of Jumieges, Gesta Normannorum ducum, p. 157. F.M. Powicke, The Loss of Normandy (Manchester, 1961), pp. 344-45. 50 Complete Pemige, vii, p. 536. 4y

The Revolution of 1066

175

of the thirteenth century. But this was home grown. It marked the end of the colonial era of English history. There was one further marked difference between England and the continent. Georges Duby and other scholars have argued that the notion and practice of heritable tenure percolated downwards through the social scale, from nobles to knights and lesser folk, the whole process in some French provinces occupying a century or more.51 Some of the English evidence fits that picture. For example, it was not until the middle of the twelfth century that English toponymic family names took root and spread among the knightly families. But there are also good reasons for thinking that they gained some security in practice in the first two generations of the Norman settlement. It is a curious story full of cross-currents. The insecurity of the greatest families told both for and against their tenants: for them, if they remained loyal to the king as against their immediate lords, which is what the oath taken at Salisbury in 1086 required them to do; against them, if they followed their lords into rebellion; against them yet again if, though loyal to the king, they nevertheless had to contend with a new lord who might give scant attention to the enfeoffments made by a dispossessed and exiled predecessor. Yet many of these lesser families managed to win through.52 The evidence of their success lies embedded in the statements of feudal service submitted to Henry II in 1166. These recorded a system, imposed by the Conqueror, in which service quotas were allocated to ten ants-in-chief largely on a decimal or demi-decimal basis. By 1166 some of the original quotas had been disturbed by the accidents of forfeiture and the resulting fragmentation and reconstitution of tenancies-in-chief, so that the record presents the original system overlain by various later developments. One of these is germane to my present argument. A considerable number of the baronies recorded in 1166 had not originated as such at all, but as undertenancies of the great lordships of the Norman Conquest which had subsequently been fragmented and dispersed. An exact count is difficult because of the anomalous position from this point of view of the lordships of Cumbria and of the earldom of Northumbria, but on a maximum count the total number of baronies which originated in this fashion falls not far short of fifty. It includes the Shropshire baronies which were first established before 1102 by Roger of Montgomery or his son, Robert de Belleme, the Kentish baronies which originated before 1088 in Odo of Bayeux's lordship in Kent, and the 'fees of Mortain' which, assessed at two-thirds of a normal fee, run like a tracer element through the feudal service of the twelfth century, recording the original enfeoffments dating from the days of Robert, 51

G. Duby, The Diffusion of Cultural Patterns in Feudal Society', Past and Present, 39 (1968), p. 6; R. Carabie, La propnete fonciere dans le Tres ancien droit normand (Xle-XIIIe sticks), i, La propiiete domaniale (Caen, 1943), pp. 276-77. 52 For further discussion of these and other aspects of the problem, see above, pp. 137-41, and J.C. Holt, 'Politics and Property in Early Medieval England: A Rejoinder', Past and Present, (1974), pp. 130-34.

176

Colonial England, 1066-1215

count of Mortain, or his son William, certainly from before 1104, probably before 1095. In many cases direct continuity between the initial undertenancy and the later barony can be demonstrated without much difficulty. In some, Castle Holgate, Cause, Chilham, Chiselborough, Mulgrave, Ros, Wigmore,53 it can be put beyond serious doubt that the holders of the twelfth-century barony were descendants of the eleventh-century undertenants. The same point can be illustrated on a more detailed scale in the internal arrangements of particular baronies. The practice of requiring tenants-in-chief to provide decimal or demi-decimal quotas of service continued through the reigns of the Conqueror and William Rufus into the reign of Henry I. Then, in the decade or so after the battle of Tinchebrai, it was changed. New baronies were now established for which the quotas were quite random. The main reason for this is that the lands were already encumbered with enfeoffed knights. Consider the carta submitted by William d'Aubigny for his lands in Norfolk in 1166: This is the holding of William d'Aubigny, butler of the lord King, of the gift of King Henry [I] who gave him fifteen knights of the fee of Corbuchon. And later he gave him of his own hand a fee of ten enfeoffed knights of the land of Roger Bigod with the daughter of Roger Bigod; and later he gave him the service of Ralph son of Godric of twelve knights, and the service of Alfred of Athelburgh of two knights, and Picot of Bavent of one knight, and the fee of Reiner Sine Averio of one knight and the fee of William de Musterville of one knight.54

Thus were knights transferred like villeins into new lordship, apparently without any opportunity to consent or to bargain for their fealty. But they remained a separate category, distinct from the knights whom William d'Aubigny himself proceeded to enfeoff on his newly acquired demesnes. They were the original constituents of the fee. It does not follow, of course, that all enfeoffed knights maintained their tenure through such a transfer. William d'Aubigny listed the survivors; the anonymity of the fifteen knights of the fee of Corbuchon and the ten knights of the fee of Bigod may cloak insecurity and loss. How much, we cannot know. But there is one document which takes us to the heart of the question of what happened to the tenants of an honour when a new lord took over, and this comes from William d'Aubigny's brother, Nigel. Nigel, another of Henry Fs 'new men', was the founder of the second house of Mowbray. Some time between 1109 and 1114 he fell ill. He recovered and lived on until 1129, but for a time he saw the 53 IJ. Saimders, English Baronies (Oxford, 1960): for Castle Holgate and Cause, pp. 28-29; for Chilham, p. Ill; for Chiselborough, p. 34; for Mulgrave, pp. 66-67; for Ros, pp. 105-6; and for Wigmore, pp. 98-99. On Castle Holgate, see also The Cartulary of Shrewsbury Abbey, ed. Una Rees (Aberystwyth, 1975), i, pp. 4, 38. The whole problem is examined for a single county and on a wider social basis, stretching from baron to undertenant, in Early Yorkshire Families, ed. Sir Charles Clay and Diana E. Greenway, Yorkshire Archaeological Society, record series, 135 (1973). 54 Red Book of the Exchequer, ed. H. Hall, Rolls Series (London, 1896), i, pp. 397-98. For the fee of Corbuchon, who had been succeeded by his son by 1086, see DB, ii, fos 85, 258b, 425b.

The Revolution of 1066

177

Pit yawning before him and set about achieving his salvation.55 He restored lands he had seized from the churches of York and Durham and the abbey of St Mary's, York. He made new pious benefactions. And he also restored laymen whom he had deprived. I can give but a selection from the letter to his brother William in which he recorded all this, which has been so splendidly edited by Dr Greenway:56 These are the lands which I have restored to the disinherited, namely to Robert de Cambos and his heirs all his land on the understanding that Robert de Cambos or his heirs are to hold it of Robert de Wyville. And to the same Robert de Wyville I have given Langthorpe, Kirby, Milby and the service of Liulf for the land of Grafton. To William son of Warm I have given Thornton with four ploughlands and the men thereon, on account of the 2M> ploughlands which I have restored to St Peter and to Serlo the monk. To Ralph de Paveli I have restored all his land and wealth as he ever best possessed them. To Ralph de Buce I have restored his land. To Wence I have restored his land as I found him seized. To the sons of Anseis I have restored the land of their father that they should hold it under Hugh de Rampan, and to the same Hugh I have given two ploughlands in Kirby, in exchange, where he held his other land. To Butin I have restored his land in Smite and his wealth which he can demonstrate he lost.

All this had apparently already been achieved or at least arranged. Nigel went on to instruct his brother further: Thou, William my brother, return to Gerald his exchange and to Burnulf his wood of Hampton; to Humphrey Hastings his land at Nuthurst which Nigel held and to Russell of Landford one plough-land. And I want you to restore the land of Lindley which belonged to Ranulf of Landford to the heirs of Ranulf if they seek it.

And so on. It is a long and detailed record of the trail of destruction which a new lord, especially a 'new man raised from the dust', might leave. It is not difficult to read between the lines some of the detail. Robert de Cambos had been deprived to make way for Robert de Wyville, who was now to be his mesne lord. William son of Warin had been endowed with land seized from St Peter's York and now received recompense because it was to be restored to the church. Anseis or his sons had been deprived to provide land for Hugh de Rampon who was now to be recompensed for their restoration and was to remain as their mesne lord. Even some of the legal context is recoverable: Wence who is now to be restored had been found in seisin; Butin is to recover his pecunia after providing evidence of his losses; various exchanges are to be confirmed or rearranged. Restoration is to be available not only to those deprived but also 55

Charters of the Honour ofMowbray, 1107-1191, ed. Diana E. Greenway, British Academy, Records of Social and Economic History, new series, 1 (London, 1972), pp. xvii, xviii and charters nos 2-10; ibid., pp. 6-15. 56 Ibid., no. 3, pp. 7-10.

178

Colonial England, 1066-1215

to their heirs, and the heirs of Ranulf de Landford are to have their father's land in Lindley 'if they seek it'. It is remarkable that all this could still be recalled and assembled in such detail. It is likely that it was drawn from the memory or record of a baronial court. But the most remarkable feature of all is that to Nigel d'Aubigny, as he lay dying, as he thought, all these actions counted as offences. In his mind's eye he saw them as crimes, second only to his sins against the church, casting long shadows in the light of the eternal fires across his hopes of salvation. And he used one word to describe what he had done. These men were the 'disinherited'. He had 'disinherited' them: the confession of a new man, and that in a document in which he constituted his brother William as his own heir and instructed him to complete the work of atonement. 'Only God makes an heir', said Glanvill.57 On this occasion he moved in most mysterious ways to manufacture some. So in the end we are driven back to language. Of what had Nigel d'Aubigny deprived these men whom he confessed to disinheriting? What ideas, notions, assumptions, vague impressions, did the word arouse in his mind and in theirs? What did Roger de Bully and Duke William of Normandy understand when Roger conceded the tithe of his land which he held 'hire hereditario'? These are matters to which I shall return in Chapter 11.

57

Tractatus de legibus, vii, i, p. 71.

10 What's in a Name?Family Nomenclature and the Norman Conquest

The name of Sir Frank Stenton is inseparable from the study of English placenames. He never wrote at length about my present topic, but he had a deep knowledge and an instinctive appreciation of its main features. When the first introductory volume of the Place-Name Society was published in 1924 he and his fellow editor, Allen Mawer, asked James Tait to comment on the 'feudal element' in place-names.1 Thereafter, county by county, this sub-tide appeared in the introductions to succeeding volumes until 1967, and it guides us in particular to such admixtures of place- and family-names as Rotherfield Greys, Crowmarsh Giffard and Minster Lovell. No audience assembled at Whiteknights, a name first recorded in 1401-2, derived from the nickname of John, lord of Earley, who lived towards the beginning of the previous century,2 needs reminding that there is an intimate connection between place- and family-names, but such connections, in which individuals and families gave their names to places, were largely the work of the thirteenth and fourteenth centuries. My concern is with an earlier period, the eleventh and twelfth centuries, when places first supplied the names for families. This was nowhere near the limit of Stenton's interest in my topic, but it marks the extent of his organised work. In Stenton's early days J.H. Round was the acknowledged expert in the field. Then as now it extended to distant heights occupied by the College of Arms, where the concern was largely genealogical. It was surveyed in a number of standard works of which the best known is C.W. Bardsley's English Surnames, which went through three editions between 1874 and 1889; and it was further cultivated in the 1930s in a number of very important studies written largely from a linguistic standpoint by Scandinavian scholars.3 Since then Bardsley's work has been succeeded as the standard work of reference by P.H. Reaney's Dictionary of British Surnames, and the whole subject has 1

P.M. Stenton, Introduction to the Survey of English Place-Names, pt 1, English Place-Name Society, l , p t l (1924), pp. vii, 115-32. 2 E. Smith, A History oj Whiteknights (Reading, 1957), p. 6; J.C. Holt, The University of Reading: TheFirstFifty Years (Reading, 1977), pp. 138-39. 3 Of these the most important for this topic is Gosta Tengvik, Old English Bynames, Nomina Germanica, 4 (Uppsala, 1938). See also Olofvon Feilitzen, The Pre-Cowjuest Personal Names ojDomesday Book, Nomina Germanica, 3 (Uppsala, 1937); Gustav Fransson, Middle English Surnames of Occupation, 1100-1350, Lund Studies in English, 3 (Lund, 1935); and Matthias T. Lofvenberg, Studies on Middle English Local Surnames, Lund Studies in English, 11 (Lund, 1942).

180

Colonial England, 1066-1215

been placed on an entirely new footing by the establishment of the English Surnames Survey in 1965 and by the publication of the volumes of the English Surnames Series. The subject is now regarded as serious business by the social historian, with much to contribute to the Study of occupational distribution, population movement and regional definition.4 My aim is to investigate what it can tell us about the Norman Conquest. But first a warning: Here lies John Bun. He was killed with a gun. His name was not Bun, but Wood, But Wood would not rhyme with gun, but Bun would.5

In short there are names and names, some more authentic than others. The most authentic of all come from documents, mostly charters of conveyance, in which the named individual appears as the originating authority or grantor: Be it known to all faithful in God both present and future and to all my men French and English, that I Hugh of Arden and Alice my wife have granted and conceded in perpetual free alms to God and St Mary of Stoneleigh .. .6

It is impossible to improve on that. True, Hugh of Arden did not write the charter; it was drafted by a clerk; many such documents were drafted by the beneficiary's clerk. But that is equally true, mutatis mutandis, of many documents drafted today in our own names, and it is fair to assume that Hugh of Arden would no more seal, than we would sign, a legal document which got the names wrong. Nevertheless there are two snags. First, the number of such charters diminishes rapidly as we go back in time; they survive in abundance from the late twelfth century, but scarcely at all before the late eleventh century. Secondly, many, indeed most, survive as copies rather than originals, and hence may be subject to errors introduced by copyists. The second is no more than a minor irritant with which we have learned to contend. But the first is a major difficulty which inhibits many statistical approaches to the problem. That is the best kind of evidence available. It can be reinforced by that of other charters in which named individuals appear as witnesses or in incidental references, as holding property or office and so on. This is a little less direct, but it considerably expands the available material because it brings in royal and ecclesiastical grants and confirmations. Then there are all those administrative documents in which individuals are named, of which Domesday Book, because 4

For a useful guide see The Study of the Personal Names of the British Isles, ed. Herbert Void (Erlangen, 1976), especially George Redmonds, 'English Surnames Research', pp. 75-86, and RA McKinley The Survey of English Surnames', pp. 119-25. 5 The Faber Book of Comic Verse, ed. Michael Roberts (London, 1942), p. 136, unattributed. 6 Sir Christopher Hatton 's Book of Seals, ed. Lewis C. Loyd and Doris Mary Stenton (Oxford, 1950), p. 96.

Family Nomenclature and the Norman Conquest

181

of its size and early date, is the most important. Next, contemporary or nearcontemporary historians mentioned individuals or families by name. That is less simple than it seems because some wrote with the object of heightening the contributions of particular families and of emphasising the continuity of their history; in this the use of the name was an important tactic. Finally there is modern usage, in which personal and family names have often been used in a sloppily inaccurate fashion. To take a very obvious example, 'bad' King John, before he came to the throne, appears often as Prince John. He was never so called; England had no princes; John's principal tide before his accession was count of Mortain. That tells us quite a lot, which I cannot discuss now, about the Angevin royal house. 'Prince' does worse than tell us nothing. This amounts to a grading of the evidence. It helps in deciding whether, our subject really was called Bun or Wood; but it does more, for here and there it throws light on how he came to be called either. Some names were used by the named individuals, some by their descendants, some by third parties, by men outside the family resorting to convenient descriptive labels. Hence, not all names were inspired by the same purpose, and this necessarily affects the name itself and how we should interpret it. For example, there were several Anglo-French families with the name Albini established in England in the twelfth century. Of the two most important, one became earls of Arundel, the other lords of Belvoir. They were quite unconnected. The first came from St-Martin d'Aubigny near Coutances in western Normandy, the second from St-Aubin d'Aubigne in Brittany. But they could easily be confused, for the founders of the family fortunes in England were both called William, and these two Williams were drawn from that group of men from the western fringes of Normandy who secured Henry I's position on the throne and were advanced in office and landed wealth in England in the years after 1106. So one, the ancestor of the earls of Arundel, was called William de Albini Pincerna, for he was the king's butler, and the other was called William de Albini Brito, William de Albini the Breton. This did not last for very long. The first Albini Pincerna died in 1139. His son became an earl, first of Lincoln and then in 1140 of Arundel. Henceforward there was no possibility of confusion, and the distinguishing names were soon dropped. The Albinis butler took to their new title as earls. The Albinis Brito shed their soubriquet with almost equal ease. They were pernickety about names; they had to be for four of them in succession were called William and hence they had to identify themselves after the first generations as Junior, Tertius and Quartus.7 But they made very little effort to keep Brito going. The first appears frequently as Brito, both in his own charters and other records.8 The second used the soubriquet only rarely; his usual style was simply William de Albini.9 William III, who died 7

//MC, Rutland, iv, pp. 18, 103,105-7,108,131,134. Ibid., iv, pp. 158, 161; RRAN, ii, passim; iii, pp. 437, 655. Within the family and among his tenants William I was also identified as senior (7/MC, Rutland, iv, pp. 127, 140). 9 William II appears with Brito on several occasions in Exchequer records: Liber niger scaccarii, 8

182

Colonial England, 1066-1215

in 1236, and William IV, who died in 1242, never seem to have used it. So there is little ground for treating Brito as a family name; it scarcely survived the death of its first bearer; and there is no ground at all for seeing in it some fiercely loyal claim to Breton origin. It appears most frequently in royal documents where William de Albini figures as a witness, and there is no real doubt that it was planted on the Albinis as a simple administrative convenience soon forgotten when no longer necessary. Even before the elevation of the Albinis butler to the comital rank, Albini Brito was not used consistently even in government records. In the Leicestershire Survey of c. 1130, where there was no danger of confusing the two families, it does not appear at all.10 Administrative convenience took a different turn later in the family's history. When William III was a minor in royal custody in 1185 he was described, for reasons not apparent, as William of Belvoir.11 Thirty years later this same William was captured and imprisoned by King John when Rochester Castle fell at the beginning of the civil war of 1215. His wife and son, also a William, had to seek funds for his ransom and it is easy to sense the impatience of the royal clerks at having to refer to two men called William de Albini in the same documents. They resolved the difficulty by referring to the son as William of Belvoir.12 But that stuck no more than Brito. Now in that little history what happened and what did not happen are of roughly equal importance. First, despite the dangers of confusion with another family, the Albinis continued to use a name which attached them to their place of origin. Secondly, despite the obvious risks of confusion within the family, they clung with equal determination to the Christian name William, and it is not difficult to understand why: the family tradition, first recorded by Roger Wendover, who was prior of Belvoir in the days of William III, was that William de Albini Brito had led the final charge of the Breton cavalry which had brought

continued

ed. T. Hearne (Oxford, 1728) i, p. 208; The Red Book of the Exchequer, ed. Hubert Hall, Rolls Series (London, 1896), i, p. 328; Pipe Roll 8 Henry II, pp. 3, 8. The usage of William's own actais obscured by the difficulty of distinguishing between some of his and his father's charters. A notification of William de Albini Brito to Theobald, archbishop of Canterbury, concerning the concession of tithes to Belvoir Priory certainly belongs to William II, PIMC, Rutland, iv, p. 159; Avrom Saltman, Theobald, Archbishop of Canterbury (London, 1956), pp. 245-46. The witness list to the confirmation by William de Albini Brito of his chapelry to St Mary of Belvoir is more appropriate to William I (HMC, Rutland, iv, pp. 118,161;J. Nichols, The History and Antiquities of the County of Leicester, London, 1795), ii, pt 1, appendix, p. 3; Monasticon Anglicanum, iii, p. 289. The William de Albini Brito who gave a writ of protection to the monks of Sawbridgeworth (Westminster Abbey Muniments, 8580; Nichols, Leicester, p. 40) is also more probably William I. Most of William II's artolack Brito (HMC, Rutland, iv, pp. 99, 108, 134, 136; Nichols, Leicester, pp. 3-4, 8-10, 10-11). 10 The Leicestershire Survey, ed. C.F. Slade (Leicester, 1956). 11 Rotuli de dominabus, Pipe Roll Society, 35 (1913), pp. 1, 63. It should be noted that Cecily, grand-daughter of Robert de Tosny, Domesday lord of Belvoir, through marriage with whom William I de Albini acquired Belvoir, sometimes appears as Cecily of Belvoir (HMC, Rutland, iv, pp. 144, 161). 12 Rot. Litt. Pat., pp. 161b, 163b, 164, 187b.

Family Nomenclature and the Norman Conquest

183

victory to Henry I at Tinchebrai in 1106.13 Thirdly they used Brito, but did not strive to preserve the name. Their original Breton identity was submerged in the broader Anglo-Norman society of which they were now members. Finally, they showed no enthusiasm for transferring their family name to Belvoir, despite the fact that that option was used on at least two occasions in royal documents. From Aubigne they had come; to Aubigne they remained linked in generation after generation. I described this excursus as a warning. It is so in the sense that it establishes very clearly that a lot of fine tuning will be necessary before the evidence is passed through the keyboard of a computer. In particular it is apparent that the names adopted by a family and the names given by others are not on the same footing. But it is also an advertisement, for it suggests that nomenclature can take us deep into the consciousness of the Anglo-Norman aristocracy: into their family organisation, into their notions of property and title, and into their sense of history. At the time of the Norman Conquest there were four main types of surnames or by-names in use in England. A given or Christian name could be used, usually to establish a relationship most commonly of son to father in the form of a patronymic denoted by the suffix 'ing' or 'son'. Secondly, a by-name of office or occupation might be used. Thirdly, there were many nicknames; Benn, for example, is the old German Bunn 'to swell', 'to be swollen'; 'I take the meaning of the nickname' states Tengvik, writing in 1938, to be 'the plump lumpish one' ,14 Finally, a locative name or a place name might be used, usually prefaced by 'at' -^Ethelric at Bocking,15 for example - or less frequently, by 'on' or 'of. Hence convention allowed variety. No one need be in difficulty in devising names. Nevertheless, even in legal documents the system seems surprisingly imprecise. Consider, for example, the sureties given for a Devon estate in the late tenth century. They are Wulfsige, Edwy, Cytel, Denisc (the Danish man), Godwine, Hunwine, Sweta, Edwy boga (the bowed, or the boy) and the priest Brun:16 no patronymics, no toponymies, no occupational names except the priest's, only one nick-name. No further definition seemed necessary. Difficulties were resolved in the simplest fashion. Another document of about the same period concerned with an agreement between Abbot Wulfric of St Augustine's, Canterbury, and Ealdred, son of the thegn Lyfing, is witnessed by Lyfing, his father, and Siweard and his brother Sired, and Wulfstan of Saltwood 'and the other Wulfstan'.17 One might imagine simply that the scribe was at a loss were it not that such a reference to 'the other one' is by no means rare. So the Old English system seems very inexact. That impression is not entirely accurate. As I shall suggest shortly, it was adequate for its prime purposes. 13

MatthaeiParisiensis chrmica majora, ed. H.R. Luard, Rolls Series (London, 1872-83), ii, p. 132, Tengvik, Old English Bynames, p. 289. 15 Ibid., p. 31. 16 Anglo-Saxon Charters, ed. AJ. Robertson (Cambridge, 1939), pp. 98-99, 343-44. 17 Ibid., pp. 128-29,371-72. 14

184

Colonial England, 1066-1215

The Norman system of nomenclature fell into similar categories. There was little difference in the use of patronymics except of course that the French fils was used instead of the English suffixes. Many occupational names were similar; the most obvious difference here lay in those derived from the offices of the French feudal household. There was some change in the main characteristics of the nicknames, for the Scandinavian fondness for nicknames bore fruit in Normandy in some of which the derivation is still a puzzle, such as Bigod, and a whole group which portrayed evil physical or moral qualities: Maltravers, 'evil deeds', Malmaynes, 'evil hands', Mauduit, 'ill conducted', Malvoisin, 'evil neighbour', and so on. But none of these differences matter very much. It is only in fact in the toponymies, which at first sight seem so similar, that there is a fundamental contrast. In Anglo-Saxon practice the toponymic is simply used to locate the individual, to give him a distinguishing mark on a particular occasion; it almost has the sense of 'who happens to come from'. In Norman family nomenclature in contrast the toponymic conveys the sense of belonging, and that in the strongest possible manner: the individual belonged to the place because the place belonged to him. This produced two marked differences in the quality of such names. First, follow an Anglo-Saxon toponymic and it will usually lead to a village; follow a Norman toponymic and it will often lead to a lordship or castle. Secondly, and much more importantly, the Norman toponymic passed from father to son. It was hereditary. It became a family name. The Anglo-Saxon name did not. Now that is the essential central fact in my argument. Toponymies and nicknames are common enough in the Anglo-Saxon period, but no one has ever found one which became hereditary. The only names which passed from father to son were occupational or official, and occurred simply where a son succeeded to his father's office. For the great mass of our inherited surnames, therefore, we are indebted in the first place to the Normans. This is no new discovery. It has been agreed by scholars in a multiplicity of fields, by genealogists,18 by linguists,19 by historians particularly concerned with the study of surnames.20 It must be one of the few points on which J.H. Round referred to Freeman with approval, for both were at one in this.21 Indeed no one has found anything but confirmation of C.W. Bardsley's statement of 1873: The distance from Dover to Calais is not great; but were it otherwise, we should still feel bound in our notice of names of foreign introduction first of all to mention Normandy. For not merely has this country supplied us with many of our best fam-

18

Anthony Wagner, English Genealogy (Oxford, 1960), pp. 51ff. Tengvik, Old English Bynames, p. 10; G. Fransson, Middle English Surnames, pp. 33-34. 20 Richard McKinley, Norfolk and Suffolk Surnames in the Middle Ages, English Surnames Series, 2 (Leicester, 1975), p. 3; The Surnames of Oxfordshire, English Surnames Series, 3 (Leicester, 1977), p. 7. 21 J.H. Round, Peerage and Pedigree (London, 1910), ii, pp. 5, 11-13, 76-77; E.A. Freeman, The History oftheNmman Conquest of England (Oxford, 1876), v, p. 570. 19

Family Nomenclature and the Norman Conquest

185

ily names, but it enjoys the distinction of having been the first to establish an hereditary surname.22

To sum up in the words of P.H. Reaney: 'In England there were no hereditary surnames before the Conquest. They were brought by the Normans from France'.23 Now it may seem that total agreement provides infertile soil in which to plant the seeds of an argument. I attempt to do so because the implications of this simple stark fact have not been fully explored. For the main implication is that the hereditary toponymic surname goes with the hereditary tenure of land. It is not simply a name but a title. Before exploring that, it is appropriate and illuminating to look back at the use of toponymies in pre-Conquest England. First, they were comparatively rare. Tengvik collected eighty-nine with the suffix 'at', forty-four with the suffix 'on' and a mere thirteen with 'of.24 The toponymic was not all the rage. Secondly, they were adventitious. They have nothing at all to do with family structure or tradition. They are derived chiefly from witness-lists to charters and from sureties for boundaries, from situations in which the family structure was traversed by other associations, especially those of the neighbourhood and the local court. In documents which are more intimately concerned with the family, the post obitum gifts which we describe conventionally as wills, they scarcely figure at all. In Dorothy Whitelock's edition of these there are literally hundreds of named beneficiaries, let alone others mentioned incidentally. But there are only three men with toponymic names. One, in the will 955-58 of Bishop ^Elfsige of Winchester, seems distinct from the Bishop's kin.25 Of the other two, who figure in the will of the ^Etheling Aethelstan of 1015, one had been deprived by the testator and the other was a creditor.26 All three instances seem exceptional. Indeed the wills leave no other possible conclusion than that the toponymic was irrelevant and unnecessary to the Anglo-Saxon family. It could manage perfectly satisfactorily with a nomenclature based on given names, further refined where necessary by the use of patronymics and other names of relationship.27 This was an expression of a structure in which each individual member of the family stood at the centre of a set of bilateral relationships, which embraced not just ancestors and children, but all the kin, siblings, cousins, uncles and aunts, nephews and nieces, and all their spouses in any generation. Furthermore, 22

C.W. Bardsley, English Surnames: Their Sources and Significations (London, 1873, 1875, 1884; repr. 1968), pp. 150-51. 23 P.H. Reaney, The Origin of English Surnames (London, 1967), p. 300. 24 Tengvik, Old English Bynames, p. 29. 25 Anglo-Saxon Wills, ed. Dorothy White lock (Cambridge, 1930), pp. 16,114-16. 26 Ibid., p. 61. 27 Wills not included in Professor Whitelock's edition only confirm this conclusion. Compare The Crawford Collection of Early Charters and Documents, ed. A.S. Napier and W.H. Stevenson, Anecdota Oxoniensa (1895), nos ix, x; Select English Historical Documents of the Ninth and Tenth Centuries, ed. Florence E. Harmer (Cambridge, 1914), nos ii, x, xx; The WillofAethelgifu, trans, and examined by Dorothy Whitelock, Roxburghe Club (1968).

186

Colonial England, 1066-1215

it was part and parcel of a system, fully revealed in the wills, in which men disposed of both land and other forms of wealth by post obitum gift right across the whole kin.28 In these circumstances toponymies were not just unnecessary; they might be a potential hindrance to testamentary initiative. Now with the Normans the toponymic served a very different purpose. They used toponymic names more frequently. They became fashionable. The names were associated with the exercise of lordship and the title to all those rights of feudal lordship which we later would call property. The association between name and land is so clear that the reversal of the logic of the evidence permits the reconstruction of the origins of Anglo-Norman families. Again and again it can be shown that an Anglo-Norman family bearing a French toponymic originated where the name indicates, often that it continued to hold land there long after it had become established in England.29 Much of this depends on patient reconstruction from later information. Occasionally it may be established by direct evidence. Roger de Bully acquired the great lordship of Tickhill after the Norman Conquest. Sometime between 1051 and 1066 this same Roger entered upon an agreement with Rainer, abbot of Trinite du Mont in Rouen. The agreement, recorded not long afterwards in a cartulary of the late eleventh century, runs as follows: Roger de Bully has sold to the lord abbot Rainer and his monks the tithe of the same vill of Bully in so much as it belongs to him by hereditary right in return for 72/. and a horse. And that no one shall challenge or confute this sale it has been corroborated with the sign and authority of William chief of the Normans.30 Now that leaves no room for doubt. Roger de Bully was so called. He held land in Bully and he held it by hereditary right, whatever that may have meant. His name, his land, his right were all acknowledged by Duke William who was the first signatory to the transaction. Just as the looser Anglo-Saxon nomenclature fits the wider structure of the kin-group so this fits the lineal family in which both name and property have descended from ancestors to heirs. When, how and why was this Norman nomenclature established? As to 'when' there is very little doubt. In the charters of the Norman dukes and in such original documents that remain from the families themselves, toponymies, indeed any kind of by-name other than the occasional patronymic, were exceptional before the reign of Duke William. The general impression of the ducal acts is that toponymies and other hereditary names spread, and spread fast, only from the 1040s and 1050s. Some names of the more important families can be driven a little earlier. The oldest of them all is the name Tosny (Toeni) which was in use by 1014 and probably earlier;31 that is the first appearance of a toponymic 28

For further discussion see above, pp. 163-66. Lewis C. Loyd, The Origins of Some Anglo-Norman Families, Harleian Society, 103 (1951). 30 RADN, no. 200. 31 Ibid., no. 15. See also Complete Peerage, xii, pt 1, pp. 754-56; and in general D.C. Douglas, 'The 29

Family Nomenclature and the Norman Conquest

187

in any of the ducal acts. Belleme makes its first appearance in 1023-25,32 Beaumont and Montgomery in 1035-40,3SWarenne in 1037-53,34 Mortimer in 1054,35 Grandmesnil in 1055,36 Montfort in 1063.37 These dates have to be based largely on attestations of the ducal acts and other charter material; there is no other reliable evidence. That may put a particular slant on the story, but there are a number of reasons for thinking that it is not seriously misleading. First, it might be expected that in establishing hereditary surnames those which compounded a toponymic with office, namely the names of the counts would lead the way. This was not apparently so. The first mention of a Count of Evreux eo nomine comes from 1038,38 of a count of Eu from 1051,39 the latter after appearing in ducal acts simply with title and Christian name.40 Secondly, it is possible to confirm this chronology in some detail in the case of two of the most important and pregnant names, Beaumont and Montgomery. The name Beaumont, first appeared in a ducal act in 1035-40.41 The usage was not yet consistent, for the Roger de Beaumont of that act appears with his brother Robert simply as Roger and Robert sons of Humphrey in a later act of Robert of 1046-53.42 Now that Humphrey is usually described by historians as Humphrey de Vieilles, and it is known that he held Vieilles and acquired Beaumont, both of which descended to his son. He was certainly given the toponymic by twelfth-century historians.43 It also appears in the rubric to his foundation charter of the abbey of Preaux recorded in the Preaux cartulary in the early thirteenth century, but in the text of the charter he is just plain Humphrey,44 and so also he appears in some but not all, of his attestations of the ducal acts. The strictly contemporary evidence of the originals among these establishes that he might also be called Vieilles,45 how frequently we cannot say. At all events it is certain that his heir, Roger, was the first to adopt the title de Beaumont. The story of Montgomery is even more explicit. The great figure here was continued

Rise ofNormandy', Proceedings ofthe Biithh Actulemy, (1947), pp. 101-30; idem, William the Con quo-or (London, 1964), pp. 83-104. 32 /MLW, nos 33, 51. 33 Ibid., nos 94, 96. 34 Ibid., no. 135. See also L.C. Loyd, 'The Origins of the Family of Warenne', Yorkshire Archaeologicaljournal, 31 (1934), pp. 97-113. 35 Comphte Peerage, ix, 266. 36 RADN, no. 137. 37 Ibid., no. 156. 38 Ibid., no. 92. 39 Ibid., no. 123. 40 Ibid., nos 4, 18, 72,104. Count Robert appears as Robert de Eu, 1035-48 (ibid., no. 105). 41 RADN, no. 94. 42 Ibid., no. 129. 43 See below, p. 195. 44 Archives de L'Eure, H. 711, Cartulaire de St-Pierre de Preaux, fo. 97; Gallia Chmtiana, xi, instrumenta, p. 199. 45 BADN, nos 32, 55, 88. Cp. nos 29, 50, 85.

188

Colonial England, 1066-1215

Robert II de Montgomery who became earl of Shrewsbury and outlived the Conqueror to die in 1094. In a famous charter in which he refounded the Norman abbey of Troarn he described himself with a flourish as 'I Robert, a Norman of Norman stock, son of great Roger',46 and there is no doubt that he was known as Roger of Montgomery. But his father, the great Roger who had founded the family fortunes, was probably not so named. The first appearance of the name in the ducal acts in 1035-40 could perhaps apply to either.47 But it is quite clear from an original charter of 1027-35, in which Roger I reached an agreement with the abbey of Jumieges concerning the market of Montgomery, that he did not then employ the toponymic, nor do any of the witnesses to this transaction.48 Less than twenty years later, with his son, it was a different story, for when Roger II made a further grant to Jumieges in 1043-8 he appeared as Roger de Montgomery and in a form which reveals the name as if emerging from a chrysalis - 'Ego Rogerius quem dicunt de Monte Gummeri' (I Roger who is called of Montgomery).49 There is novelty in that. 'When' this new fashion began is relatively easy to answer. 'How' and 'why' are much more difficult. Though of recent origin, the current flowing towards the establishment of hereditary surnames was strong. The hereditary toponymic was not just an accident, a casual repetition by a son of his own and his father's place of origin. Other forms of by-name which became hereditary are recorded in the same period. Taisson first appears in ducal acts in 1025,50 Giffard in 1035-47,51 Malet in 1035-66,52 Marmion in 1060.53 It is to this same period, too, that the Norman habit of repeating patronymics, so infuriating to the genealogist, Richard fitz Gilbert, Gilbert fitz Richard, Richard fitz Gilbert and so on, can be traced. In short nomenclature in general was changing in the generation before the Normans set out for England. Roger de Bully's charter is perhaps the best point from which to attempt an answer. He held Bully iure hereditario.54 The most likely interpretation of those words is not that he had a heritable right in Bully, but simply that he had in fact inherited it. This suggests a three-stage process. First the land is acquired. Secondly, the son inherits it and begins to be called or begins to call himself by the appropriate toponymic. Thirdly, both land and title pass to his son so that the surname becomes hereditary, not just a personal but a family name. That certainly happened in the case of Beaumont and probably also in the case of Montgomery. Although in each of these the toponymic can be traced to the second genera46 'Ego Rogerius ex northmannis northmannus magni autem Roger! films', BN, MS lat 10086, fo. ii verso. 47 RADN, no. 94. 48 Ibid., no. 74. 49 Ibid., no. 113. 50 Ibid., nos 33, 65. 51 Ibid., no. 105. 52 Ibid., no. 173. 53 Ibid., no. 147. 54 See above, p. 186.

Family Nomenclature and the Norman Conquest

189

tion, it is only through hindsight that it can be accepted at that stage as a family surname. William, the first and only count of Arques, certainly used his territorial style before 1047,55 as did the Conqueror's half-brother, Robert, count of Mortain, by 1063.56 In the counties founded earlier, Evreux, Eu and Brionne, the territorial style seems to have come into use only in the second generation. How far other toponymies were used in the first generation of acquisition it would be hazardous to say. There are some apparently convincing examples. Orderic Vitalis recounted that Baudri the German left six sons: Nicholas de Baqueville, Fulk d'Aunou, Robert de Courcy, Richard de Neuville, Baudri de Bocquence and Viger of Apulia.57 Three of these names can be confirmed in the ducal acta: Fulk d'Aunou,58 Nicholas de Baqueville,59 and Robert de Courcy;60 but they may have inherited rather than acquired these lands. On the whole the strictly contemporary evidence is too fragmentary to permit a precise assessment of conventions which in most cases, at this period, must have been transitional. But it is likely that the passage of land from ancestor to heir, usually from father to son, preceded the appearance of a toponymic. For a toponymic to be used there must have been some sense of security that tenure of land was a reliable and permanent identifying mark. Why did all this happen? Many of the illustrations used so far have been taken from the acts of the dukes of Normandy. This has been done deliberately in order to establish an important point: there is no hint at all that the dukes resisted the development. Indeed there is room for an easy political explanation: that young Duke William ensured his succession to Normandy by surrendering to the pressure from his men to acknowledge landed title, and that half a century later old King William rounded off his conquest by conceding the same point to his men in England through the record of their tenures in Domesday Book and through the concurrent Oath of Salisbury. That is a novel picture of a concessive ruler which might well be given a place in the current fashion of historical revision. It is too facile an explanation. There were geographic variations within Normandy which may have helped to form, rather than been formed by, Duke William's policies. In the later middle ages primogeniture was more strongly rooted in the area from Rouen north to the Channel, the Pays de Caux, than in the rest of Normandy; in this the Pays de Caux marched with the littoral counties of Ponthieu and Boulogne, not with the rest of the duchy.61 There are 55

RADN, nos 107,108. Ibid., no. 156. 57 Orderic Vitalis, ed. Chibnall, ii, pp. 82-84. 58 RADN, no. 147, 231. 59 Ibid., no. 187. Nicholas also appears with a patronymic in ibid., nos 102,153. 60 Ibid., no. 189. 61 R. Genestal, 'La formation du droit d'ainesse dans le coutume de Normandie', Normannia, 1 (1928), pp. 157-79; J. \Ver, 'Les caracteres originaux du groupe de coutumes de 1'Ouest de la France', Pjavue historique de droit franfais et etranger, fourth series, 30 (1952), pp. 18-79, especially pp. 41-47; J. Yver, Egalite entre heritiers et exclusion des enfants dotes (Paris, 1966), p. 121 n. 226 and map. 56

190

Colonial England, 1066-1215

reasons, therefore, for thinking that this 'absolute' primogeniture did not originate in the political authority of the dukes, and it is worth noting that already by the middle of the eleventh century the Pays de Caux yields a rich crop of toponymic names, some going back to 1025-26,62 the reign of Duke Richard II, and some clearly belonging to the undertenants of the great lords of this region.63 It seems likely that such nomenclature and the heritable tenure which lay behind it had already taken root when the leading men of this region, the count of Eu, William de Warenne, Walter Giffard and Hugh de Gournay, won the victory at Mortemer in 1054 which finally secured William's position as duke.64 There were other variations which can only have been distantly connected with politics or social and legal change. Either Matthew Paris was engaging in a jest more suitable to Sellars and Yeatman or he was guilty of an ignorant anachronism when he derived the surname Bigod from the characteristic oaths of Hrolfr, the founder of the duchy, for the notion that a Viking warlord swore Christian oaths in English is nothing but bizarre.65 It is just possible that he was retailing a family joke of the Bigods, among whom he must have met both Roger, fourth earl of Norfolk, and his brother Hugh, the future justiciar,66 for by the time Matthew wrote, in the middle of the thirteenth century, the Bigods could have been as much in the dark as we have all been ever since as to what their name really meant. With the Crispins there is greater certainty. Their name appears very early, in 1035-48,67 and since it was then borne by two brothers it would be reasonable to conclude that it was already a family name. It was, but not for the kind of reason advanced so far: Before the Normans conquered England under Duke William, there was in Neustria, which is now called Normandy, an outstanding man called Gilbert, famous for his lineage and nobility, who is said to be the first to be known by the cognomen Crispin, because of his hair-style, for from his earliest years he had crisp hair which stood on end, erect like the branches of a pine, which always stick upwards; wherefore he was called Crispinus, a crispuspinus [a bristly pine]; and we still see the same stiffstanding hair in those who have descended from Gilbert's stock, so that they are all known by the same name by the rest of the families of Normandy.68

That is derived from a hagiographical account of the Crispins attributed to a member of the family, Milo Crispin, precentor of Bee.69 Another Crispin, also

62

RADN, no. 54. Ibid., no. 135; Loyd, The Origin of the Family of Warenne', pp. 100-1. 64 Douglas, William the Conqueror, pp. 67-70. 65 Chronica majora, i, p. 450. 66 Ibid., v, passim and p. 672. 67 RADN, no. 105. Cp. nos 110, 156. 68 'De nobile genere Crispinorum', Patrologia Latina, cols 735-44. 69 J. Armitage Robinson, Gilbert Crispin, Abbot of Westminster (Cambridge, 1911), pp. 13, 17-18. 63

Family Nomenclature and the Norman Conquest

191

called Gilbert, ruled as abbot of Westminster for more than thirty-five years from c. 1085. The supposed effigy in the abbey cloister presents him with a shock of good hair.70 Such evidence apart we can only guess why one family resorted to a toponymic, another to a nickname, while a third might simply use patronymics. Toponymies originated in inherited title. But it would be unwise to think that the aristocratic instinct for this was any weaker in the families which used other nomenclatures. Presumably the hereditary nickname, like the hereditary toponymic, bore witness to the great men of the family, the founders of fortune, even if it did no more than delineate their personal characteristics. For Gilbert Crispin did more for his family than pass on crinkly hair. He established it in the inner ring of dominant families through his marriage with a grand-niece of Gilbert, count of Brionne. One son became custodian of Tillieres, another acquired Neaufles and married a Montfort.71 The patronymic is more difficult. Generations were to pass before it crystallised into an unchanging family name; the fitz Alans of Oswestry after 117572 and the fitz Gerolds in 1177-873 were apparently the first to make this transition. At the time of the Conquest the patronymic changed from father to son according to their given names. There was no title in that other than the record that Richard was Gilbert's son, and that may have been the point. The patronymic could assert a relationship where title was uncertain or in dispute. The great aristocratic family known as the house of Clare looked back to a certain Gilbert who played a great part in the establishment of Duke William and obtained the tide of count. Later sources refer to him as Gilbert, count of Brionne, but there seems to be no strictly contemporary evidence that he used the toponymic. His father Godfrey, an illegitimate son of Duke Richard I, had been count of Eu, as also was Gilbert for a time in 1040.74 In the circumstances Gilbert might feel some embarrassment: two comtes to choose from; to choose one might imply forgetfulness of the other; to claim both would be altogether unusual. He was allowed no time to resolve the dilemma for he was murdered in that same year, 1040, and the tides went to other families.75 Gilbert's sons, Richard and Baldwin, began to attest ducal acts in the 1050s, each with the patronymic 'son of Count Gilbert'.76 They recovered some estates77 and their 70 Ibid., frontispiece; Royal Commission on Historical Monuments, London, i (1924), p. 78, pi. 202; J.P. Neale and E.W. Brayley, The History and Antiquities ofSt Peter, Westminster (London, 1818-23), ii, p. 294. 71 Armitage Robinson, Gilbert Crispin, pp. 14—15. 72

J.H. Round, Studies in Peerage and Family History (London, 1901), pp. 128-30; Pipe Roll 21 Henry II, p. 39. 73 Pipe Roll 24 Henry II, pp. xxxii, 30. 74

D.C. Douglas, The Earliest Norman Counts', EHR, 61 (1946), pp. 133-40. Idem, William the Conqueror, pp. 40-41. 76 RADN, nos 130, 179,192. 77 Baldwin received Meules and Le Sap, Richard Bienfaite and Orbec, Orderic Vitalis, ed. Chibnall, iv, p. 208. Orderic's terms reddiditznd restituit imply an act of restoration. 75

192

Colonial England, 1066-1215

sons in turn pressed claims to Brionne itself;78 one indeed, Robert son of Baldwin, had custody of the castle for a time in 1090.79 But neither sons nor grandsons called themselves de Brionne. They had lost the castle and office to which the title was attached; it was used in turn by the men who supplanted them;80 but they did not admit defeat by assuming another lesser French toponymic as a regular family surname.81 So the patronymic became their normal usage, and so it remained in England after the Norman Conquest. The reason is apparent in Domesday Book. Time and again Richard fitz Gilbert appears not simply as that but as Rzcardus filius Gilberti comitis.82 The murdered count was not to be forgotten. The enormous acquisitions which his sons achieved in England made a difference. The patronymic was still used by the sons of Baldwin,83 but Baldwin figures as Baldwin of Exeter and Baldwin the sheriff in Domesday Book.84 Richard likewise not only figures there as Richard fitz Gilbert, but also as Richard of Clare and Richard of Tonbridge.85 Tonbridge survived as a name in the family for at least another generation.86 Clare also was used occasionally by both Richard's son and grandson.87 Ultimately it overbore Tonbridge, presumably because the honour of Clare was the more important territorially.88 But in the second and third generation the patronymic was still used;89 indeed at Ely, where Richard fitz 78 Robert of Torigny recorded that Roger, son of Richard fitz Gilbert, attempted to recover Brionne by a proffer of money to Duke Robert after 1087, William of Jumieges, Gesta Normannorum ducum, ed. J. Marx (Rouen, 1914), pp. 289, 414. For Robert, son of Baldwin fitz Gilbert, see Orderic Vitalis, ed. Chibnall, iv, p. 205. 79 Orderic Vitalis, ed. Chibnall, iv, p. 204. 80 RADN, nos 131, 142. Guy, son of Reginald, count of Burgundy, also appears as Count Guy (no. 113) and as Guy of Brionne (no. 194). Nos 131 dated 1053,142 dated 1059, and 194 attributed to c. 1050-1066, all seem inconsistent with the usually accepted account of Guy's total forfeiture and banishment after Val-es-Dunes (Douglas, William the Conqueror, pp. 48-55, 87), and confirm Williams of Poitiers' statement that he remained at the Norman court, ed. Raymonde Foreville (Paris, 1952), p. 21. 81 On toponymies attributed to Count Gilbert's sons by later writers, see below, p. 196. 82 DB, i, fos 100, 126b, 189, 209, ii, fos 1, 281. In three of the four instances in vol. 1 comitisis interlineated. 83 J.H. Round, Feudal England (London, 1909), pp. 472-73. 84 DB, i, fos 75, 86, 100. 85 Ibid., i, fos 2, 14; ii, fo. 448. For charters of Richard fitz Gilbert lord of Clare see Select Document of the English Lands of the Abbey of Bee, ed. Marjorie Chibnall, Camden third series, 73 (1951) pp. 21-22. For Richard 'of Tonbridge' see also D.R. Bates, The Land Pleas of William I's Reign: Penenden Heath Revisited', Bulletin of the Institute of Historical Research, 51 (1978), p. 3. 86 WAN, p. 153. 87 For Gilbert of Clare, son of lord Richard, see Feudal Documents from the Abbey of Bury St Edmunds, ed. D.C. Douglas (Oxford, 1932), p. 153. For Richard fitz Gilbert, lord of Clare, see BL, MS Cotton, appendix 21, fo. 25. 88 Richard Mortimer, The Beginning of the Honour of Clare', Proceedings of the Battle Conference on Anglo-Norman Studies, 3 (1980), pp. 119-41. 89 For Gilbert fitz Richard see Registrum Roffense, ed. J. Thorpe (London, 1769), pp. 590-91, and Cartularium monasteriiS.JohannisBaptistede Colecestria, ed. S.A. Moore (Roxburghe Club, 1897), i, p. 141. For both Gilbert fitz Richard and Richard fitz Gilbert see Historia et cartularium monasterri

Family Nomenclature and the Norman Conquest

193

Gilbert's son, Richard, reigned as abbot from c. 1100 to 1107, they were known as the Ricardi'?Q and it was not until the fourth generation, with Gilbert fitz Richard who became an earl in 1140, that Gilbert of Clare, or Gilbert earl of Clare, or Gilbert of Clare, earl of Hertford, became the usual style. Detailed motives are lost to us but it is difficult not to see ambivalence and hesitation in that. If Count Gilbert had evaded his murderers in 1040 the odds are that the senior house of Clare would have carried the name of Brionne. The Clare claim to Brionne was long remembered. Robert of Torigny, writing c. 1139, reported that old men held that when the lowy of Tonbridge was given to Richard fitz Gilbert it was measured with a rope so that it coincided exactly in extent with Brionne.91 The Norman conquest imposed on England an entirely novel system of family nomenclature. Tengvik listed over 2000 occurrences of toponymies, referring to over 300 persons and over 200 places.92 That reflects the richness of Domesday Book and certainly cannot be used for statistical comparisons with other much more meagre sources. But comparisons are possible within the Domesday evidence itself. The names of the French tenants-in-chief listed at the beginning of each county entry may be categorised as follows: French toponymies English toponymies Earldoms or comtes Nicknames Patronymics Locatives (Britoetc.) Names of office or occupation Given names only

107 23 10 55 72 14 67 50

These figures must be treated with some caution. Some of the French toponymies may have been purely locative. Some of the English toponymies, Durand of Gloucester, Picot of Cambridge, Baldwin of Exeter, were quasi-official, for those who carried them were sheriffs. Only in a few instances is it possible to demonstrate with certainty that the French toponymies had already been inherited, that they were already in the strictest sense true family surnames. None the less, the preponderance of French toponymies is remarkable, all the more so because most of the official names, many of the patronymics and a considerable number of the nicknames belonged to the less important vassals. This may have been a direct effect of Norman settlement in England. The

continued

S. Petri Glwccstriae, ed. W.H. Hart, Rolls Series (London, 1863-67), i, pp. 106, 124; ii, pp. 73-76. For Richard fitz Gilbert, see Pipe Roll 31 Henry I, p. 53. 90 LiberEliensis, ed. E.G. Blake, Camden third series, 92 (1962), p. 226. 91 William of Jumieges, interp. Robert of Torigny, p. 289. 92 Tengvik, Old English Bynames, p. 29.

194

Colonial England, 1066-1215

toponymic served home sickness, family tradition, political domination and social snobbery at one and the same time. The change in fashion was pervasive and infectious. The new nomenclature was thrust on English families so that Thurkill, one of the two great AngloSaxon landowners south of Trent in 1086, appears in Domesday Book as Thurkill of Warwick.93 Lesser men too partook of the fashion. Many families with French toponymies or hereditary nicknames can be traced among the undertenants of the great Norman lords, and, as generation succeeded generation, those of humbler origins or older habits of nomenclature turned to the use of English toponymies to signal their arrival as securely established landed gentry.94 The Conquest also contributed to, and may well have accelerated, the transition whereby the toponymic came to record not fact but expectation. It is quite certain that Norman and other French lords used English toponymies in the generation in which they acquired their new estates. Those who took up English names, Richard of Tonbridge, Baldwin of Exeter, Joel of Totnes, Robert of Stafford or Roger of Berkeley, were surely doing more than indicate their homes. They were bringing to bear all those unrecorded assumptions which attached toponymic to tenure. They were staking out title. That is the logical implication of the invaders' readiness to adopt first-generation toponymies. The family toponymic soon became engrained in social convention. It was used by cadet as well as senior lines, presumably because their fortunes too depended on the success of the founder's line on whose lordship and patronage they depended and to whom they might succeed in default of senior heirs. But it could be changed. Both sons of Roger de Beaumont were Beaumonts, as were their sons in their turn - with one exception. Roger's grandson in the junior line inherited or acquired Neubourg and hence was called de Neubourg.95 Similar good fortune cloaks the fact that the Robert of Stafford of the Domesday Survey was the younger son of Ralph de Tosny; he still occasionally used the toponymic Tosny, but Stafford became the normal surname in his family.96 One of the clearest indications that there was a link between the family name and tenure is that it could be replaced by a stronger tenurial tie: new tenures, new names. This occurred even more obviously when an heiress intervened. The system of nomenclature was strongly patrilineal. There was only one circumstance in which a man might take a name from his mother's or his wife's line, and that was when he inherited her rights. It was through this that Eustace fitz John, sheriff and justiciar of Henry I, originated two distinctly named lines. The son of his first marriage to the heiress Beatrice de Vesci took the name William de

93

DB, i, fos 238, 240b. See in particular Richard McKinley in Norfolk and Suffolk Surnames in the Middle Ages, pp. 330; and idem, The Surnames of Oxfordshire, pp. 7-40. 95 See above, pp. 152-54. Complete Peerage, xii, pt 2, appendix, pp. 4-5. 96 J.H. Round, Feudal England, pp. 213-14; Complete Peerage, xii, pt 1, p. 168. 94

Family Nomenclature and the Norman Conquest

195

Vesci and originated the second line of Vesci.97 The son of his second marriage married a niece of Ilbert de Lacy of Pontefract. He remained plain Robert fitz Eustace and after his death his wife married two subsequent husbands. She was still alive in 1193 when her cousin, the last surviving Lacy in the direct line, died without issue. The great estates of Pontefract and Clitheroe reverted to her and her surviving heir, her grandson, who promptly took the name of de Lacy.98 So Eustace fitz John's son in one line was a Vesci, his great grandson in another was a Lacy. The most bizarre example of all is provided in the family of William d'Aubigny who originated the earls of Arundel. He had a younger brother, Nigel, and Nigel's children stemmed from his second marriage with Gundreda de Gournay. Nevertheless, they were called Mowbray and took their name from Nigel's first wife, Maud de Laigle, by whom he had no surviving children and whose only claim to the name of Mowbray was that she was the divorced wife of Robert de Mowbray, earl of Northumbria, who suffered imprisonment and forfeiture in 1095. Nigel's link with the defeated earl was further attenuated when he himself divorced Maud de Laigle in 1118.99 Nevertheless the name of Mowbray came through the sordid marital tangle. So also did the fee of Montbrai which the earl had held. The family patrimony in Aubigny had gone to Nigel's elder brother, so Montbrai was his only Norman fee until his second marriage brought him further acquisitions. Though a recent acquisition and that by marriage, it was still his earliest Norman possession and hence his children took it as their place of origin; it also gave them the lustre of a once great name. But it did not come to them through human generation. It came with the land. They had got the Mowbray place, so Mowbrays they became. Later the only residual trace of this unusual history lies in the family's Christian names. Among the recurrent Roberts, Rogers and Williams there appears the occasional Nigel. The founder's name could thus be honoured even if his family-name Aubigny had long since gone. Within two generations or so of the Conquest men came to treat lineage and tenure as if the close association of the two were part of the air they breathed. Orderic Vitalis, 1123-25, and Robert of Torigny, c. 1139, wrote of an established aristocracy, of circumstances in which the great Norman lineages could trace their descent even to the Duchess Gunnor and beyond. They pressed family names and toponymies into service, partly because they carried their assumptions back into earlier generations when their world was still being born. Both were confident that Roger de Beaumont's father was called Humphrey de Vieilles;100 Robert even referred to Humphrey's father as Turold de Pont Aude-

97

Compute Peerage, xii, pt ii, pp. 272-74. W.E. Wightman, TheLacyFamily in England antlNmmandy, 1066-1194 (Oxford, 1966), genealogical table. 99 Charters of theHonourof'Mowbray, 1107-1191, ed. Diana E. Greenaway (London, 1972), pp. xviiixix, 260-61. 100 William of Jumieges, pp. 158, 324; Orderic Vitalis, ed. Chibnall, ii, p. 13. 98

196

Colonial England, 1066-1215

mer.101 Robert did not hesitate to apply the toponymic Montgomery to both father and son; indeed he confused the two.102 Orderic knew that Richard fitz Gilbert had acquired Bienfaite and Baldwin fitz Gilbert Meules,103 and he used both these names occasionally as toponymies to describe Gilbert's sons.104 How far they were accurate in this will depend on a long, painstaking search of Norman charters and cartularies. Pending that it is unwise to rely on their unsupported evidence. They were not the first genealogists of English history, for the Anglo-Saxons had achieved some notable imaginative work in this line.105 But they were the first genealogists to be concerned with landed title. In that they reflected the world of the early twelfth century clearly and precisely. Genealogy and family nomenclature are not to everyone's taste. This may be because, until the last twenty years or so, they have been largely left to genealogists. I can only end with the most celebrated of all comment, that of Lord Crewe in the De Vere peerage case of 1625: Time hath his revolution, there must be a period and an end of all temporal things, finis rerum, an end of names and dignities, and whatsoever is terrene; and why not of De Vere? For where is Bohun? Where's Mowbray? Where's Mortimer? etc. Nay, which is more, and most of all, where is Plantagenet? They are intombed in the urnes and sepulchres of mortality.106

Famous words those and rightly so, but the metaphor is wrong. Names are not aseptic, cinerary remains. They are rich fruit, left by our ancestors. If subjected to the right pressures they will yield much rich juice to the historian.

101

William of Jumieges, p. 324. Ibid., pp. 320-22. 103 Orderic Vitalis, ed. Chibnall, iv, p. 208. 104 Ibid., ii, pp. 214, 317; iii, p. 20. 105 K. Sisam, 'Anglo-Saxon Royal Genealogies', Proceedings of the British Academy, 39 (1953), pp. 287346. 106 Les reports de Sir William Jones (London, 1675), p. 102. 102

11

Feudal Society and the Family in Early Medieval England, II: Notions of Patrimony

Notions are potent but nebulous, often direct and determining in their effect but themselves indeterminate in origin and structure. My title is designed to circumvent two lines of thought which have largely circumscribed the study of inheritance in the eleventh and twelfth centuries hitherto. First, I shall say something here and there about succession, but it will be only a subsidiary part of the argument. Heritable title was not diminished by unsettled rules of succession. On the contrary, in the eleventh century as in the thirteenth, it was emphasised and nourished by the claims and counter-claims of competitors. In such disputes the opposing arguments were couched in a common language; it is the language, therefore, that will be my first concern. Secondly, for this same reason I shall also pay scant attention to the jurisdictional aspects of inheritance. To be sure, in post-Conquest England inheritance amounted not to a tide but to a claim upon a lord; heritable title was realised when the lord admitted it; no concession by a tenant was as secure as it could be made until his lord had confirmed it. But lord and vassal also shared the same language. Each inherited; each knew that a vassal might be disinherited. Hereditas, jus hereditarium, exheredare. these and similar terms expressed and informed their attitudes, no matter whether, on a particular occasion, they were for or against hereditary practice. It is reasonable therefore to enquire what assumptions and imagery lay behind the words. It is best to begin on relatively firm ground with a case history from the late twelfth century. At the end of April 1201 an assize of mart d 'ancestor'was held before Kingjohn's justices at Tewkesbury to determine whether Adam of Cock* field, father of Margaret, was seised in demesne as of fee-farm of the manors of Cockfield, Semer and Groton in Suffolk on the day he died and whether Margaret was his nearest heir.1 The case was brought against the abbot and prior of Bury St Edmunds and was terminated in Margaret's favour. She was no more than an infant. The real victor was her guardian, Thomas de Burgh, tenant of the abbey and brother of Hubert de Burgh, the king's chamberlain.2 Powerful men were involved in the action. The justices proceeded with despatch. 1

Curia Regis Rolls, i, p. 430. Thomas de Burgh bought the wardship for 500 marks from Hubert, archbishop of Canterbury, who, in turn, had paid £100 for it to the abbot of Bury, The Chronicle ofjocelin ofBrakelond, ed. H.E. Butler (London, 1949), p. 123. 2

198

Colonial England, 1066-1215

Thomas de Burgh apparently ensured that the jury was made up of knights, tenants like him of the abbey.3 Their verdict was very much in their own general interest. The case was reported in the rolls of the court, and also in two separate passages in the chronicle of Jocelin of Brakelond.4 It is therefore possible to supplement the pleadings and the verdict with Jocelin's own account and with an appendix to the chronicle by William of Diss, monk of Bury, which reads very much like a deposition of the abbey's case for the action held at Tewkesbury. These three distinct sources, supplemented by charters in which the Bury archive was rich, allow an almost complete account, in which the views of both lord and tenant are fully represented, of the development of heritable tenure in the Cockfield fee. The first point to grasp is that this particular inheritance was a variegated accumulation made up of terms of tenure laid down over a long period punctuated by disputes and settlements. Two of the manors, Semer and Groton, were acquired by Adam's grandfather, also called Adam, in the reign of Stephen and had been held at farm ever since for three generations. But the tenure had never been strictly hereditary. The rents were increased at the first succession;5 the second Cockfield, Robert, apparently conceded in 1188 that he had no hereditary title;6 and in c. 1191 Adam of Cockfield, the decedent of 1201, was party to an agreement with the abbot and convent which laid down that at his death the manors were to revert to the convent without any claim from his heirs who would have no hereditary right therein.7 All this proved in vain. The jurors of 1201 overrode or ignored Adam's charter, which was read in court. They reviewed the long and continuous tenure, noting that both Adam and his grandfather had died seised and that Robert, Adam's father, had held throughout his life. They concluded that such long tenure constituted a fee-farm.8 Thereby they converted defacto succession into dejure inheritance. Their line of thought was by no means novel.9 The descent of the third manor, Cockfield, was a different story, for in 1201 the abbot (none other than Abbot Samson) conceded in court that Robert of Cockfield on his death-bed had said that Cockfield was his right and inheritance. The court took the abbot's statement as an admission. Cockfield was not 3

Ibid., pp. 123-24. Ibid., pp. 123-24, 138-39. 5 Ibid., p. 138. 6 Curia Regis Rolls, i, p. 430. 7 The Kalendar of Abbot Samson of Bury St Edmunds and Related Documents, ed. R.H.C. Davis, Camden third series, 84 (1954), pp. 127-28. 8 Curia Regis Rolls, i, p. 43Q; Jocelin of Brakelond, p. 124. 9 Compare the claim of Robert de Valognes, 1158-59, that he held Nor thaw wood by hereditary tenure of the abbey of St Albans. In this case, too, the wood had been held by successive generations of the Valognes family, and men equated such tenure with hereditary title despite the contrary evidence of the abbey's charters. Gesta abbatum monastmi Sancti Albani, ed. H.T. Riley, Rolls Series (London, 1867-69), i, pp. 160-62. 4

Notions of Patrimony

199

referred to the jury, but adjudged at once to Margaret, Robert's granddaughter.10 The abbot's uncertainty was shared by others of the convent. Jocelin of Brakelond felt that Cockfield may have been different in this respect because it had not been included with the other manors in the concession of hereditary title by Robert and Adam of Cockfield.11 William of Diss's more detailed account is of great interest: the Cockfield family, he stated, had no charter entitling them to the manor of Cockfield, only a writ of seisin of King Henry I, incorrectly sealed, charging Abbot Anselm to allow Adam I of Cockfield to hold it as long as he paid his rent in full; but William allowed that Robert of Cockfield had avowed in the abbot's presence that he believed Cockfield to be his hereditary right on account of long tenure in that his grandfather, Lemmer, held the manor long before his death, and Adam his son throughout his life and Robert himself throughout his life, in all nearly sixty years.12 Robert's death-bed statement was closely similar to the knights' verdict in 1201: long tenure and uninterrupted succession established hereditary tide. Cockfield was divided into two manors. The portion still in dispute in 1201 was that assigned to the sustenance of the monks.13 The other portion of the township, according to William of Diss, had been held by Lemmer, Robert's grandfather, by hereditary right and, with the assent of Abbot Anselm, had been converted from socage tenure to half a knight's fee in the reign of King Stephen.14 Moreover, this second portion was but one of several tenements held by hereditary right, for which Robert of Cockfield had the abbey's charters, which indeed William of Diss read aloud in the presence of Abbot Samson and many others. These amounted to at least one knight's fee in the time of Abbot Anselm (1121-48), to one and a half fees in 1166 and to at least two fees by 1200.15 Some of these charters survive in the Bury records. In the earliest Abbot Anselm confirmed to Adam of Cockfield and his heirs tenure jure hereditario in return for one knight's service of the lands which his father had held in Lindsey and Cockfield.16 But none of these charters indicates that Adam's father, Lemmer, held Cockfield jure hereditario as William of Diss states. The only hint that this was so lies in a distinction which William does not mention: Adam was to hold jure hereditario in Lindsey and Cockfield, but Abbot Anselm made no mention of hereditary right when he confirmed Adam in all his father's and his own purchases and acquisitions.17 10

Curia Regis Rolls, i, p. 430. Jocelin of Brakelond, p. 123. 12 Ibid., p. 139. 13 For land assigned ad victum monachorum, see V.H. Galbraith, 'An Episcopal Land-Grant of 1085', EHR, 44 (1929), pp. 363-68. }4 Jocelin of Brakelond, pp. 138-39. 15 Feudal Documents from the Abbey of Bury St Edmunds, ed. D.C. Douglas, British Academy Records of Social and Economic History, 8 (1932), pp. 120-21; Liber rubeus de scaccario, ed. H. Hall, Rolls Series (London, 1896), p. ^93; Jocelin of Brakelond, p. 123. IG Feudal Documents, ed. Douglas, pp. 120-21. 17 Ibid., p. 118-19. 11

200

Colonial England, 1066-1215

There is yet another complication. So far the story has been concerned with land, inherited or acquired, held on varying terms for diverse renders and services. Throughout, the tide ran in the tenant's favour towards heritable succession. But at one point Abbot Samson was able to stem it. When Robert of Cockfield died c. 1191 Adam laid claim to the half hundred of Cosford as his hereditary right on the ground that his father and grandfather had held it for eighty years or more. The claim was rejected, despite the fact that Adam had stong supporters including Roger Bigod, earl of Norfolk; for Abbot Samson pointed out that such hereditary tenure would make room for royal intrusion into the liberty of the abbey, if by chance Adam or his heir fell into the mercy of the king or was subject the crown's prerogative wardship.18 So while land might become hereditary, office might not. But the abbot's victory was won at a cost. This dispute too went to the king's court in an action of mort d'ancestor. It was settled by final concord in which the abbot confirmed numerous socage lands to Adam to be held hereditarily and in perpetuity in return for the renunciation of his claim to the half hundred.19 This evidence allows an almost complete reconstruction of the development of heritable tenure in the Cockfield fee. It is a little hazy at the beginning of the story. We cannot be certain on what terms Lemmer held Cockfield unless we accept William of Diss's statement of his heritable title from roughly eighty years later. But it is certain that by the 1130s Lemmer's son, Adam, held this manor, from which the family came to take its name, jure hereditario. The fee was augmented by purchase and acquisition. Descent by primogeniture was reinforced by the conversion of socage tenements, which might be partible, into knight's fees, which were not.20 Yet heritable tenure in some estates held at farm was not secured until the third generation after Adam, in the interest of the infant Margaret, and then only after action in the king's court in 1201. Moreover, the claim to heritable title in office had to be abandoned in 1191. It is plain that inheritance was a complex matter. It varied in strength and was established over different periods from one form of property to another. The lord was readier to admit some heritable tenures than others, the tenant readier to sustain claims here and abandon others there. It is worth remembering that this complex evolution, covering eighty years or more and involving at least two legal actions in the king's court, took place without any open breach between the Cockfields and their monastic lords. The family was not dissident. The first Adam of Cockfield acquired Semer and Groton in the reign of Stephen because of his ability to defend them in the abbot's interest against neighbouring castellans;21 Robert of Cockfield was one of the

is

jocelin ofBrakelond, p. 123. Feet of Fines, Henry II and Richard I, Pipe Roll Society, 17, pp. 9-11; Kalendar of Abbot Samson, pp. 71-72. 20 Glanvill, De legibus, vii. I, 3; ed. G.D.G. Hall (1965), pp. 71, 75. 2l jocdin ofBrakelond, pp. 123. 19

Notions of Patrimony

201

custodians of the abbey during the vacancy of 1180-82;22 in succeeding generations their advancing fortunes were confirmed by the abbots. Only perhaps in 1201, in the verdict at Tewkesbury, does the issue of inheritance appear as one of principle, with the abbot and convent ranged on one side and the knightly tenants on the other. For the rest the picture is one of men of business fashioning a modus vivendiwhich both resolved differences and satisfied mutual interests. The objections to accepting this as a definitive case history are obvious. Even for Bury it may be misleading. For William of Diss hereditary tenure was synonymous with tenure by charter. 'He had good charters', he wrote of Robert of Cockfield, 'for all the tenements which he held of St Edmund by hereditary right'; William even talked of 'hereditary charters', and his final word was that neither Adam nor Robert had any charters for the land in dispute in 1201 in Cockfield.23 Such coincidence of hereditary and written title belonged to the early thirteenth rather than the early twelfth century. For Lemmer, as we have seen, the charters mentioned by William of Diss have not survived. The earliest Bury deed conveying hereditary title seems to be a charter of Abbot Anselm for Maurice of Windsor, the abbey's steward, of 1114-19.24 Going back further still the well-known notification of the acceptance of King William's knight, Peter, as Abbot Baldwin's feudal man suggests that the documentation of enfeoffment in the years immediately following the Conquest was quite inchoate, if indeed any weight at all can be placed on so corrupt a text.25 It has some value, however, in at least one respect. It was produced in unusual circumstances to solve a particular problem — what to do about the services of a tenant who also owed service to the king? Some of the charters mentioned by William of Diss were also produced in special circumstances; they marked the termination of disputes, for example. Yet William assumed that charters were the norm, the necessary evidence of settled undisputed tenure. He was not the last to do so. Such charters are our main source of information on the development of hereditary tenure. These sources are reduced to a trickle and then run almost dry as we follow them back from the twelfth to the eleventh century. We tend to think that this is also true of the ideas they embodied, that hereditary notions were fresh sprung in the dawn of the Norman settlement. That this was not so, that the Normans came to England with well-developed notions about inheritance, will be one of my main arguments. In another respect the history of the Cockfield family is a sure guide. Throughout a period covering roughly eighty years the several different facets of their tenure were described in the same terminology: feodum, jus, heres, hereditas,jus hereditarium, infeodo et hereditate. Only once throughout the varied record of the family's tenure does the word patrimonium appear and only then as a 22 23

Ibid., pp. 8,10.

Ibid., pp. 138-39. 24 Feudal Documents, ed. Douglas, p. 111. 25 D.C. Douglas, 'A Charter of Enfeoffment under William the Conqueror', EHR, 42 (1927), pp. 245-47. For further comment, see below, pp. 220-21.

202

Colonial England, 1066-1215

means of distinguishing heritable land which had been inherited from heritable land which had been acquired.26 Yet it was a patrimonial system and one strongly rooted in primogeniture to which these terms were applied.27 Institutions had become far more precise than the words used to describe them. One reason for this is that, with one obvious exception, the vocabulary was ancient. The exception wasfeodum which was engendered by the development of hereditary military tenures in the late eleventh century. The rest had a long history in which the major source was the Corpus Juris Civilis,28 and such a venerable pedigree, however dimly perceived and understood, gave to the Cockfields and to all others who used them a sense of security. What better vehicle could there be for the establishment of heritable fiefs than those terms which in antiquity had sprung from heritable tide? But the consequences for the historian are important and daunting. The language may have remained the same but the notions which it expressed did not. How did the terms acquire the relatively precise usage of the second half of the twelfth century? The jurisdictional answer is that they were defined by the interplay of lord and tenant in seignorial and royal courts.29 But the words were not confined to these arenas. Nor could they be divorced from their history. In deploying them lords, tenants, justices and jurors were resorting to a language which might seem to be very restricted and very simple. Beneath the surface, however, it was rich in imagery and associations. To trace this language back into pre-Conquest Normandy is to enter a different world. In twelfth-century England inheritance has been viewed as a seignorial concession.30 In eleventh-century Normandy it was more like a primordial right over which, in time, the dukes came to exercise superiority. Inherited land and alodial land were treated as one and the same, alodium sometimes appearing simply as a colloquialism for hereditas.31 These equations were first expressed, and indeed in Normandy the whole language of inheritance was first generated, in the records of ecclesiastical and monastic endowment. Here, at first sight, they seem oddly out of place, for the language of inheritance and generation ill became the church in an age of reform which laid such emphasis on celibacy. But the convenience of terms which had a legal connotation 26

The Kakndar of Abbot Samson, pp. 17-19. My reading of this evidence is a little different from that suggested by Professor Davis, ibid., p. 17 no. 2. 27 For further comment, see below, pp. 215-20. 28 See especially Institutes, II, xiv, xix. 29 S.E. Thome, 'English Feudalism and Estates in Land', Cambridge Law Journal, new series, 6 (1959), pp. 193-209; S.F.C. Milsom, The Legal Framework of English Feudalism (Cambridge, 1976), esp. pp. 154-86. 30 Space restricts me to the baldest of summaries of the contributions of Professor Milsom and Professor Thome. 31 See the charter 1035-c. 1040 of John, abbot of Fecamp: 'Odo. . .totam hereditatem suam, quam communi voce alodum dicimus, contulit. ..', RADN, no. 93. Compare the less direct analogy drawn by Albert, one-time abbot of Jumieges in Charles de Vabbaye dejumieges, ed. J.-J. Vernier, Societe de 1'histoire de Normandie (1916), i, p. 57.

Notions of Patrimony

203

outweighed such considerations. Hence the monks of Fecamp obtained a lordship to be held jure hereditario in alodum and in alodum et hereditatem perpetuam.32 At Mont-St-Michel the monks received a benefaction to be held in omni successura generatione.33 An estate was conferred on St-Ouen perpetua hereditate; this house enjoyed its superiority over the daughter house of Sigy jure hereditario.34 Some of the charters of the founder of La Trinite du Mont also provided that the monks were to hold jure hereditario. They received further concessions in perpetua hereditateor inperpetuam hereditatem35 Such words were part of the common jargon of conveyancing. They were shared with the Anglo-Saxon landbocs and Carolingian diplomata36 It may be that these words were used in order to express monastic title in a language which the layman could grasp and understand. But this is scarcely a sufficient explanation. A monastery received gifts in hereditary right from donors who enjoyed hereditary title. The emphasis on ecclesiastical tenure jure hereditan'owas surely aimed at transferring to the beneficiary such right as the donor had in the benefaction. In extreme cases God or Christ or the saints intruded into the family succession. Land became the legitima hereditasof St-Wandrille.37 Odo son of Geoffrey granted land in perpetuam hereditatem to Fecamp because he made 'no other person but God his heir of all his land in the whole of Normandy'.38 A certain Heloise, on becoming an anchorite, made Christ the heir of her patrimony which she conceded to the abbey of Coulombs.39 This bond between inheritance and monastic endowment is equally obvious on the side of the benefactors. In one of the earliest recorded grants to 32

RADN, no. 71 (1034). Ibid., no. 132 (1054-55). 34 Ibid., nos 43 (1015-26), 103 (1037-r. 1045). 35 Ibid., nos 83 (1030-35), 84 (1030-35), 101 (1043), 123 (1051), 201 (1051-66), 202 (105166), 233 (1066). 36 F.M. Stenton, The Latin Charters of the Anglo-Saxon Period (Oxford, 1955), pp. 87-88; E.John, Land Tenure in Early England {Leicester, 1964), pp. 39-63. 37 RADN, no. 95 (1037-r. 1040). 38 Ibid., no. 93 (1035-r. 1040). 39 Ibid., no. 230 (1066). Subsequent genealogical accident sometimes upset such arrangements. Some time after 1035 a certain knight Gilbert, 'lacking an heir', conceded all his inheritance in Conde and elsewhere to St-Pierre de Preaux, the grant being agreed by Robert son of Humphrey, i.e. Robert de Beaumont. Gilbert's wife subsequently gave birth to a daughter who married Roger de Croixmarez. That the initial grant was a post obitum gift is revealed by the fact that the charter was only produced long afterwards when Gilbert died as a monk of Preaux, 1078-r. 1090. Eventually, after much discussion, Roger de Croixmarez asked Abbot William to pay the relief on (rekvaret) Gilbert's 'honour' to Roger de Beaumont dequo beneficium erat. This was agreed. Roger de Beaumont then conceded the estate to the abbey on the understanding that the monks were to have only half of Conde during Roger de Croixmarez's life, with the reversion of all he had there at his death except for thirty acres with a house, court and orchard and two knights. These, along with the other properties conceded in the original grant, were to be held by Roger's heir who would pay relief (relevaret) to the abbot and monks for them sicuti mos est terre. The effect was to confirm the original grant, but with sitting tenants, including the donor's heir (Archives de L'Eure, H.711, Cartulaire de St-Pierre de Preaux, fo. 100). 33

204

Colonial England, 1066^1215

St-Wandrille, 996-1006, the donor granted 'a certain estate conceded to me by hereditary right by my ancestors'.40 Other benefactors of the abbey during the next fifty years granted 'land of my hereditary right' or land 'pertaining to me by hereditary right from my relatives to be held by the monks forever'.41 It was a similar story at Fecamp where in 1006 Duke Richard II supplemented his father's endowment with further concessions from his 'hereditary right'.42 At St-Ouen the monks received 'part of his inheritance' from Roger son of Hilbert.43 Viscount Erchembald, on entering La Trinite du Mont, gave to the house his meadow in Sahurs and all that he held by hereditary right in Celloville.44 Such terms were not restricted to the acts of the greatest men. Milites might hold alodial property.45 Freemen might be called alodarii.46 Before 1050 William fitz Osbern confirmed to his foundation at Lyre the land of Theodoric Tirel, presumably his vassal, 'which was left to him from the inheritance of his father'.47 At almost the same time Robert de Beaumont confirmed to St-Wandrille Foucarville with its church and nine manorial tenants, hospites, with 'whatever belonged to him by hereditary right'.48 The language of inheritance was shared. The same words could be used to describe both the capacity of the donor and the tide of the beneficiary. They expressed the fact that there was a cohesive bond between inheritance and endowment. You cannot give what you have not got. To give is to hold the wherewithal to give. To give in perpetuity is to assert or imply that in some way you have had more than a life interest in what you give. That tight logical knot bound together not only donor and beneficiary, but also both to their lord. Much of the evidence used above is drawn from charters in which the dukes of Normandy confirmed or attested the grants of their men. Many reasons have been advanced in the past for the great wave of monastic endowment and benefaction set in motion by the dukes and accelerated and enlarged by their vassals in the course of the eleventh century. The donor's desire to atone or repent, on occasion in retirement in the house which he or his family had founded; the prestige which the new foundation brought to the founding family; the convenience of a house under family patronage as a receptacle for sons, daughters and sundry relatives; all these have been rightly emphasised. None of these motives, however, comes so close to explaining the close coincidence of monastic endowment and an emergent aristocracy as the simple point that a grant in perpetuity and inherited tenure were logically inseparable. To give emphasised 40 Tradidi quemdam fundum, mihi ab antecessoribus meisjure hereditario concessum' (ibid., no. 7). 41 RADN, nos 134 (1037-55), 125 (1051). 42 Ibid., no. 9. 43 Ibid., no. 42(1015-26). 44 Ibid., no. 82(1030-35). 45 Confirmation of Duke William for Fecamp, 1035-40 (ibid., no. 94). 46 Confirmation of Duke William for Jumieges, 1060-66 (ibid., no. 220). 47 Ibid., no. 120 (c. 1050). 48 Ibid., no. 129(1046-53).

Notions of Patrimony

205

that. To record the gift in a document confirmed or attested by the duke at one and the same time recognised his superiority and embodied his approval both of grant and hereditary tenure. Any man of sufficient substance to hold or acquire land in free tenure might profit from this mutual security of benefactor and beneficiary. Against his doing so there was no social barrier short of the ever steepening slopes which led to the great divide between free and unfree. But there were obstacles, ranging from procedural complications to serious jurisdictional and tenurial doubts about the donor's capacity. The many surviving charters in which lords confirmed specific grants by individual tenants, the Norman pancartes and the AngloNorman charters of confirmation in which tenants-in-chief enumerated the grants of their men, all attest that the service with which the land was burdened gave the superior lord so great a say that a grant made without his approval was infirm. The lord's charter of confirmation, his attendance at the ceremony where the grant was made, or the promulgation of the tenant's grant in the lord's court, were an important, often the essential, element in the emerging notions of warranty. But here too the logic of benefactions which were to be enjoyed in perpetuity began to seep in. Some lords, taking regard perhaps of their own administrative ease in the face of the importunities of potential benefactors and beneficiaries, began to shift ground, from confirming the particular grants of each of their tenants to enunciating general rules governing such alienations. Such licences became commonplace after the Norman Conquest. In a charter of 1080, conveying revenues and tithes in Normandy and England to St-Florent-les-Saumur, William de Briouze conceded: 'And if any of my men shall give or sell anything [to the monks] I approve.'49 If Hugh earl of Shrewsbury's confirmation of the lands and privileges of his father's foundation at Shrewsbury is to be trusted, he or his father conceded that any of his barons, burgesses or knights whosoever wished, might give to the abbey any of their lands and rights without any licence from him or his heirs so that neither donors nor recipients of such alms would perform any further service to him.50 There was yet another pattern at Chester. There Earl Hugh and his wife granted that their barons might grant to St Werburgh a hundred shillings worth of land during life and a third of their goods with their bodies at death.51 A generation later Henry, count of Eu, lord of Hastings, in a grant which foreshadowed later mortmain legislation, was even more meticulous. 'I Henry, count of Eu, concede to the church of St Martin of Battle whatever my men in England and Normandy have given or shall give to 49 Charles normandes de I'abbeye de Saint-Florent pres Saumur, ed. M.P. Marchegay, Memoires de la societe des antiquaires de Normandie, third series, 10 (1880), pp. 678-81. The passage occurs in a variant reading of a duplicate original. There is another text with a correct dating clause in Bouhier's transcripts, 'Cartulare comirum pictaviensium et engolisme', BN, MS Lat. 17089, fos 910v-ll. 50 The Cartulary of Shreivsbury Abbey, ed. Una Rees (Aberystwyth, 1975), i, p. 9. 51 The Chartulfiry or Register of the Abbey ofSt Werlntrgh, Chester, ed.J. Tait, Chetham Society, 79 (1920), p. 17.

206

Colonial England, 1066-1215

the same church in alms for the sake of their souls excepting the caput of their manors and the knight service.'52 By the middle of the twelfth century such facilities were being extended well down the social scale. In a charter of 114855, Hugh son of Richard of Hatton, tenant of ten fees of the earldom of Warwick, founder of Wroxall Priory, granted the church of Claverdon to St Mary of Monmouth, a priory of St-Florent-les-Saumur, with the provision that if any of his men wished to give to that church and the brethren there keeping watch in worship of God an acre or two, saving his service, the gift should remain firm and undisturbed.53 In all these cases the tenant's power to alienate was limited not only by the particular restrictions which his lord might apply but also very frequently by the nomination of the beneficiary. Moreover, blanket approval of alienation was not free from ambivalence. The concession of Henry, count of Eu, was one of two such charters. One was a general licence to alienate. The other differed in certain respects: it made no reservation of knight service, it included sale as well as gift, and, after giving general permission to alienate, it went on to confirm particular alienations.54 Count Henry's concession was confirmed by his successor, John, sometime before 1170,55 and was elaborated by his great granddaughter Alice, countess of Eu, between 1219 and 1246.56 She too confirmed whatever her men had given or sold in alms to Battle. In her day the reservation of military service which her ancestors had sometimes made no longer held good, but she also enumerated the grants which her tenants had made. The general licence to alienate was still not so strong that detailed confirmation would not strengthen it. The arguments so far has been aimed at establishing that monastic endow52

Huntington Library. San Marino, MS BA 42/1526. Henry succeeded in 1096 and died in 1140. The charter probably comes from earlier rather than later in this period. It is of antique script, has no seal and no indication that it ever carried one, and bears the signa of Count Henry and others. 53 Livre noir de St-Florent-les-Saumur, BN, nouv. acq. lat. 1930, fo. 34; Cal. Docs. France, no. 1146. Here Round's date of 1151-57 seems to depend entirely on the date of the confirmation of John, bishop of Worcester (ibid., no. 1147). The attestation of Robert, prior of Monmouth, indicates 1148 x 1155. For Hugh son of Richard see Red Book of the Exchequer, i, p. 325, and VCH, Warwickshire, \\\, pp. 116-17. 54 Cartulary of Battle Abbey, Lincoln's Inn, MS Hale 87, fos 52v-53. 55 London University Library, Fuller Collection, 1/28/1. Count John, like Henry, concedes 'whatever my men in England and Normandy have given or shall give. . .excepting their capital manor and knight service'. 56 Fuller Collection, 1/28/3. Alice's grant is not derived directly from the earlier concessions. She confirmed past grants and sales without reference to those which might be made in the future, and her enumeration of the grants belongs to the more usual pattern of seignorial confirmations rather than to the general licences conceded by Counts Henry and John. Her confirmation included Sl/2 marks charged on a half knight's fee in Whatlington 'by distraint on the same half fee if necessary'. All the grants were confirmed 'cum omnibus pertinencis necnon et warda castri cum omnibus que de predictis terris sive redditibus ad me vel ad heredes meos pertinent'. Even so the series indicates that seignorial control was stricter and more specific in the thirteenth century than it had been earlier.

Notions of Patrimony

207

ment strengthened inherited capacity. This was true whenever and wherever land and rights were granted in perpetuity or, as later, in free and perpetual alms. It was given its sharpest emphasis when land held in hereditary right was granted as a perpetual inheritance. However, the use of a common language had a further effect of even greater import. It not only strengthened and spread the notion of inheritance. It also contributed powerfully to transforming it. It converted inherited right into heritable claim. Donors were envisaged as giving from their inherited land because that strengthened their capacity to give. But once they were envisaged as endowing a monastery by hereditary right they were dealing not with the past but with the future, not with ancestors but with heirs, not simply with inheritance but with heritability. This too was injected into the language by logical necessity. There is no doubt that the primary sense of the hereditary right of the donor, as it figures in the Norman charters of the early eleventh century, was simply that he had inherited the land or rights in question. They were held paterno jure 57 or ab antecessoribus.58 As Baldwin Filleul put it in 1051, in conferring his alod of Glicourt on St-Wandrille, he could be seen to have possessed it 'in full ownership and freely from his father and forefathers'.59 But the act of transfer to a beneficiary added something new. Inheritance in the past underlay the donor's capacity to give. But if the land was to remain with the recipient for good the hereditary descent had to be broken, the recipient himself taking the place of the heir. Every act of perpetual endowment in short involved the claims which the donor's heir might bring. Indeed endowment was probably the main instrument whereby such heritable claim was aired and refined. So between 1035 and 1040 Duke William confirmed to Fecamp the land of Richard fitz Goscelin de Bardouville in four villages in the Pays de Caux which were to become the property of the monastery on Richard's death nullo herede calumniante vel reclamante.60 At about the same time in rather different circumstances the duke persuaded Abbot John of Fecamp to give to Hugh son of the Viscount Hugh a life interest in one of the monastery's alodial lands sine ulla calumnia sui heredis aut proheredis.61 In some documents the inheritance was allowed to run to the death of the sons of the donor. In 1015-17 Roscelin, canon of the cathedral of Rouen, bequeathed to St-Ouen all the inheritance which he possessed to go to the church 'after his death and the death of his two sons. . .who were heirs to the inheritance', the monks meanwhile to enjoy the tithe of the property. His nephew, Gonduin, made similar provisions for the inheritance of him and his son, and Roscelin ended by calling down the wrath of God on any of his heirs or any 'extraneous' person who interfered with his benefaction.62 In other documents the heir's 57

RADN, no. 34(1025). Ibid., no. 7(996-1006). 59 'Quod a patribus et proavis solidum et quietum hactenus possedisse videor' (ibid., no. 124). 60 Ibid., no. 94. 61 Ibid., no. 93. 62 Ibid., no. 21. Compare ibid., no. 19 (1006-17) in which a certain Drogo released the land of 58

208

Colonial England, 1066-1215

claim had a negotiable quality to be expressed in guarantees or cash. Between 1037 and 1048 William count of Arques, and Mauger, archbishop of Rouen, gave Perriers-sur-Andelles to the monks of St-Ouen in memory of their father, Duke Richard II, and accepted 300 livres so that 'none of their heirs, relatives or men giving themselves to the Devil, should contradict or challenge the gift'.63 Such arrangements did not always work. In 1066 it was recorded that the gift of Heloise, who had become an anchorite, made Christ her heir and ceded her lands in Autieux to the abbey of Coulombs, had been overridden by her nephew who seized the property by force. Only God's mercy and the persuasion of Duke William and his wife Matilda brought him in penitence to restore the gift.64 By the time they invaded England the Normans' notion of inheritance was expressed in terms which were both complex and sophisticated. The inheritance might descend to the holder not just from his father or from vaguely defined ancetors but also from the maternal line.65 Women might inherit and make benefactions on their own.66 Wives might give endowments from their hereditary right with the approval of their husbands.67 A husband might give land which he held by right of his wife, with her consent.68 Inheritance from collaterals, brothers and uncles, was common. And just as the inheritance could be derived from varied sources, so it might be subject to varied claims, not just from sons but from brothers and nephews.69 Even an heir still unborn might be included. Hence in the decade before the Conquest Herfred, brother of Genscelin, bequeathed his part of the land in Daubeuf to St-Martin du Bosc 'in hereditary right after his death if he did not have an heir'.70 Indeed the claim of heirs could already be envisaged as extending into the future as long as the accompanying endowment lasted. In 1063 Odo, steward of Duke William, died continued

his brother and vineyards which he had acquired in Bailleul to St-Ouen after his and his wife's death, conceding the land to the monks sine ullo herede so that they were to succeed him in eternity without any claim of heirs. He also gave two mills which he had constructed at the gate of Le Goulet with his two sons and heirs who were to pay an annual rent for them. After their death no one was to be heir or successor on their part to make claim against the monks. 63 Ibid., no. 112. 64 Ibid., no. 230. 65 Ibid., no. 51 (1023-26). 66 Ibid., nos 55, 95 (1025-26,1037-r. 1040). 67 Ibid., no. 173(1035-66). 68 Ibid., no. 167 (1035-66). 69 Ibid., nos 147 (1060), 230 (1066). Such claims could lie hidden behind a grant of an inheritance. Sometime between 1044 and 1078, probably before 1066, the knight Ansketil, son of Turulph, made a post obitum gift, with the consent of his wife and sons, to the abbey of St-Pierre de Preaux of whatever came to him by right from paternal inheritance in Tourville and Campin. At his death the inheritance was divided among his surviving brothers. The abbey's portion amounted to three tenant-holdings (hospites) and two of the brothers, Gilbert and Geoffrey, persuaded Abbot Ansfrid to grant them to them in bmeficia, with Geoffrey rendering the service. It was noted that this was done without the advice and agreement of the monks. Archives de 1'Eure, H 711, fos cviii, cxii. 70 RADN, no. 218.

Notions of Patrimony

209

at the age of twenty-six. His father Stigand, 'who loved him dearly in his life and even more dearly after his death', commemorated his death with endowments to the cathedral church of Rouen and the abbey of St-Ouen, stipulating that 'no abbot nor any man that might ever be should alienate his gifts' and, if he did, then Stigand's heir might intervene and obtain them on the same terms as the intruder.71 The inheritance was often sharply distinguished from the benefice and from precarious tenure. 'I have a certain alod', announced Albert, abbot of St-Mesmin de Micy, 'from my maternal inheritance not of the benefice of any one' in 1023-26.72 At about the same time documents in which Duke Richard confirmed grants to Fecamp, Jumieges and St-Ouen, indicated that they were made both from the various grantors' paternal inheritance and from the benefices which were ducal rights.73 Inheritances and alodial possession could be bought and sold, with transfer of full tide, both by churches and laymen.74 Other forms of tenure could be converted into alodial, perpetual or hereditary right. In 1047-63 Robert, abbot of St-Wandrille, paid 25 livres for a twelve-year lease of land in Lebecourt, purchased from Nicholas son of Baldric. Subsequently, at Nicholas's request, he paid a further 12 livres and acquired the land in perpetual possession as the alod of St-Wandrille.75 When Emma granted land to St-Ouen in 1015-26, she distinguished between her alods and the vill of Meslay which she held as a benefice of Duke Richard. It was Richard himself and he alone who could and did convert the grant of this benefice into a perpetual inheritance of the abbey.76 In one remarkable case an inheritance figured as a realty even when the tenant had been driven out by his feudal superior. A document of 1059 records that Adam de St-Brice, ejected by Duke William, was received in the city of Tours by the monks of St-Julien. He was honourably entertained and provided with a horse worth 12 livres. In return he gave the monks Roncheville 'which was his right' in Normandy, if they could obtain it from Duke William. They promptly did so. Ultimately Adam was restored and his gift was confirmed by the duke.77 This tenement was described as a 'right' and not directly as an inheritance. That it was in fact such is clear from the statement that Adam had been disinherited. This brings us face to face with the interplay between inheritance and lordship. This was a rich confusion of individual encounters any one of which might impose its own legal requirements on the common language. One such set of circumstances is revealed at an early date, 1011-33, and at a middling social 71

Ibid., no. 158. Ibid., no. 51. 73 Ibid., nos 34, 36, 53. 74 For purchases by La Trinite du Mont, ibid., nos 130 (1053), 143 (1059), by St-Wandrille, ibid., no. 154 (1047-63) and by Boscherville, ibid., no. 197 (1050-66). See ibid., no. 83, for a purchase by Goscelin, vicomte of Rouen, from Helto son of Gilbert, 1030-35. 75 Ibid., no. 153. 76 Ibid., no. 43. 77 Ibid., no. 142. 72

210

Colonial England, 1066-1215

level, by a charter of Henry, abbot of St-Ouen, which records the purchase of fifty acres of arable land in Auzouville from Gunduin, 'our knight'. The agreement laid down that none of Gunduin's successors should claim any hereditary right therein. However, to that straightforward alienation of hereditary land Gunduin added a further ten acres of his better land, as a gift, not a sale, on the understanding that 'we should agree that the beneficiumwhich he held from us should go to his sons after his death on the same terms of service whereby he served'. Gunduin's son, Hugh, attested and signed the charter and received a penny to buy a bow.78 That provided for at least one succession to a beneficium. The arrangement is not too far removed from the arguments which took place over the Cockfield fee more than a century later. In the long run hereditary right came to reside both in lord and tenant. Between 1088 and 1096 Ralph fitz Ernest, vassal of Robert de Beaufour, who was in turn vassal of William, count of Evreux, conveyed the church of St-Germain de Varaville to the abbey of Troarn. His grant was confirmed by William count of Evreux sicut rem que ad hereditatem meant pertinet and was attested by Robert de Beaufour who also confirmed ex toto in fine hereditabiliter in elemosina the grant which Ralph fitz Ernest had made hereditabiliter.79 There might be analogous ambiguities in ducal acts. Service was due to the duke from alods and inherited lands as well as from benefida. The alod was distinguished from the beneficium not by absence of service but by the long possession derived from family succession, by relative security of tenure and by easier alienability.80 Any ducal confirmation of the grant of alodial or inherited property acknowledged the duke's superiority and some degree of tenurial dependence. Richard II confirmed grants both of inheritances and beneficial1 and Duke Robert confirmed to La Trinite du Mont the grants which his faithful men had made 'from their hereditary property with our agreement'.82 On occasion ducal confirmation was expressed so strongly that the act of confirmation almost had the quality of a conveyance willy-nilly of a vassal's property. The knight Ascelin sold the land which belonged to him in Grosfy, but Duke William's confirmation was a simple conveyance of his vassal's land; only the chance survival of the deeds of sale makes clear that the land was Ascelin's 'right' to which his heirs might lay claim.83 Duke William also confirmed lands to 78

Archives de Seine-Maritime, 14H, 255. BN, MS lat. 10086, fos xlv-xlvb; R.N. Sauvage, L'abbaye de Saint-Martin de Trearn, Memoires de la societe des antiquaires de Normandie, fourth series, 4 (1911), pp. 145-46. For another charter, 1020-30, recording the hereditary title of the lord, Hugh, bishop of Bayeux, in benefaction of a tenant see Charles de Vabbaye dejumieges, i, pp. 21-22. 80 R. Carabie, La propiiete fonciere dans le Tres ancien droit normand, Xle-XIIIe siecles, i, La propriete domaniale (Caen, 1943), pp. 230-43; L. Musset 'Reflexions sur alodium et sa signification dans les textes normands', Revue histonique de droit franfais et etranger, fourth series, 47 (1969), p. 606. 81 RADN, no. 53 (1025-26). 82 'Memorati fideles ex rebus hereditariis suis et ex nostra cessione. . .deputaverant', ibid., no. 61, 1030. 83 Ibid., no. 109; F. Lot, Etudes critiques sur Vabbaye de St-Wandrille (Paris, 1913), no. 26. 79

Notions of Patrimony

211

St-Wandrille, at the request of three of his knights, in terms which did not specify whether the land was theirs or his, whether it was alodial property in which the services gave the duke an interest or beneficia in which the knights had only a life estate.84 The duke's lordship could even be used as a defence against the importunities of heirs. In 1066 itself a dispute occurred in the family of John, bishop of Avranches. The bishop wished to give half the land of Vievre, which was his inheritance, to his cathedral church. His nephew Robert, son of Richard de Beaufour, challenged this, claiming that he had received the land from his uncle, as a hereditary gift. He was bought off with the gift of 10 livres and the commendation for his lifetime of five of the bishop's knights. He was probably doing no more than assert a hereditary claim. But the bishop could call on the advice of great men including Roger de Beaumont, and he secured the transaction by surrendering the land to Duke William 'so that it should be free and quit of all claims from relatives'. The duke then conveyed the land to the cathedral church.85 In that example lordship and inheritance were in direct antithesis. Ducal power could be applied brutally. On occasion ducal superiority was asserted in sovereign terms. One of the very few occurrences of the word patrimoniumis in Duke William's confirmadon in 1050 of the foundadon of St-Evroul wherein he threatened that those who violated his charter would be guilty of lese-majeste (nostre reum esse majestatis) and would merit the ascription of all their patrimony to the ducal demesne.86 There patrimony and majesty were placed in sharp contrast, the most secure inheritance against the supreme secular authority. But in other documents the dukes seem at one with their men. Duke Robert founded Cerisy from the lands which belonged to him by hereditary right.87 His endowment of Mont-St-Michel speaks of the land he possessed 'for his own use by hereditary right'.88 Duke William also granted judicial rights which he possessed 'as his hereditary share',89 and precluded his successors from overturning his grants in perpetuity.90 The duchy itself was acknowledged to descend by right of inheritance.91 On the face of things, therefore, there is little ground for thinking that the dukes' view of inheritance was different from that of their vassals. The underlying reality was that the notion of inherited or heritable right was not dependent on ducal authority. To be sure, Duke William especially made 84

RADN,no. 126. Ibid., no. 229. 86 Ibid., no. 122. 87 Ibid., no. 64 (1032). 88 Ibid., no. 73. 89 'Per sortem hereditariam' (ibid., no. 146). 90 Ibid., no. 148. 91 See the charter of Duke Robert (1027-35) which reviews the descent of the duchy on the death of Richard II: 'Eo denique celestia, ut credimus, scandente regna, principatum ejusdem nominis Richardus filius illius obtinuit. Sed cita morte preventus, jure hereditario fratri suo Rodberto eundem reliquit', ibid., no. 74. 85

212

Colonial England, 1066-1215

grants of land which became hereditary in famous families which, like the Beaumonts, took their names from these new properties. But it would be hazardous to assume that in so doing he conveyed hereditary tide in enfeoffment as his youngest son did for his vassals half a century later as king of England. It is more probable that he conveyed possession, which was then inherited, at which point all the notions of hereditary right which I have described came into play to reinforce title in the second generation. There are some examples where a duke created perpetual title. Duke William for example conferred on St-Wandrille five churches in the Cotentin which he had previously granted to Robert de Beaumont, to be held jure perpetuo.92 The grant to St-Wandrille was made with Robert's agreement, for the salvation of both him and the duke, and the whole transaction, recorded in a single charter, seems closely similar to the AngloSaxon charters which embody a royal grant to an individual of land or rights subsequently conveyed as a benefaction.93 There are also two remarkable charters surviving only in later copies and not entirely unimpeachable, in which Osbern and Ansfrid, brothers-in-law of Duke Richard II, are recorded as granting to St-Wandrille their estates in Monterolier, that is an alod 'as it was partitioned between them and Odelarius by allotment by cord on the order and authority' of the duke, the grant being confirmed by 'the lord Richard the second, chief of the Normans, of whom we hold all these things, who had our sister Papia to wife'.94 These words may well mean that Duke Richard had indeed created this alodial property for his brothers-in-law. But such evidence appears only occasionally in the ducal acts before 1066. The documents rarely indicate how alodial or inherited tenure had originated. Instead they reveal a dual process. First, ducal authority over alodial and inherited or heritable family land was being extended and defined, its emergence marked by references to service, or to tenurial dependence on the duke, and above all by the recourse to the duke for confirmation or attestation. Secondly, with every ducal confirmation the jus hereditarium was strengthened, the alodium stood out in sharper contrast to the benefidum, and the succession to the inheritance was confirmed or canalised. It would be tempting to suggest that alodial tenure was derived from the Scandinavian settlement of Normandy, were it not that it could scarcely be defined except through the interplay between the dukes and their vassals and the interest of both in monastic foundation and endowment which are recorded in the early Norman charters. One thing is certain: the charters create no ground at all for thinking that the complex notions they record were all of ducal creation. To transfer the English experience after 1066, in which King William created fiefs for his men as feudal lord of all the land, to Normandy before 1066, in which Duke William was building up a superior92 Ibid., no. 128. The ambiguity in my translation is present in the Latin, although Mile Fauroux's punctuation attaches the perpetual right to Robert, not, as is more likely, to the monks of St-Wandrille. 93 Stenton, Latin Charters, pp. 59-60. 94 RADN, nos 46, 46 bis.

Notions of Pa trimony

213

ity over his men's inherited family holdings, which were described in a language which was older than the surviving documentation, would be quite wrong. Heritable title and feudal authority grew together. This can be demonstrated quite simply. In Norman England the heir can be regarded simply as having a prior claim on his feudal lord. He had to establish his claim and be put in by his lord after the performance of homage and fealty. But this mechanism should not be allowed to obscure the fact that, both in England after the Conquest and in Normandy before, the heir was known far more often than not. In charter after charter he attested his predecessor's grants or participated as the son in the benefactions of the father. Acts of benefaction specifically precluded subsequent contravention by the heir. Even before he succeeded his claim could be weighed against cash and bought off. In short, he had a status which could be increased by ducal and royal approval but which was antecedent thereto and independent thereof. The heir continued to enjoy this status, indeed to tend it, even when disinherited. Disinheritance cannot be understood except as its contradiction. The duke might regard the heir's claim as permissive. The vassal regarded it as title, clearly and early expressed, if we may transgress the boundaries of Normandy, by Ivo, count of Beaumont-surOise, in a benefaction to St-Wandrille of 1039: T count Ivo with my son of the same name, who is a clerk and canon, to whom jure hereditario I concede my castle of Conflent after the passing of my death. '95 Orderic Vitalis wrote of the dispositions of some of the great Norman families in closely similar terms.96 Family ties, not ducal power, defined the prior claim. The duke might override it, just as a father might deny his eldest son, but the claim was there, nevertheless, even if ignored or turned aside. The arrangements of pre-Conquest Normandy were still a far cry from the enfeoffments of Bury St Edmunds and the expanding fortunes of the Cockfields a century later. Hereditas, jus hereditariumwere words which echoed down the decades, but with some changes of meaning and discontinuously, with periods of almost total silence. One important change concerned the question - who are the heirs? The descent of the Cockfield estates went by primogeniture; it seems that collaterals to the main line were allotted under-tenancies;97 this became the predominant convention and very soon the rule within a generation or so of the Conquest. It had nowhere near so powerful a hold in Normandy before the Conquest; the further back in time, the weaker primogeniture appears. On the surface there were many exact parallels between the two periods and places; in Normandy, as at Bury, patrimoniumwith its restrictive connotation was a very rare word indeed. But behind the language the pattern of succession changed fundamentally, from one involving partition among sons and perhaps 95

Lot, St-Wandrille, no. 24. Orderic Vitalis, ed. Chibnall, ii, p. 10 (Grandmesnil); ii, p. 98 (Hauteville); hi, p. 89 (Guiscard); iv, p. 184 (Evreux). 97 For evidence of an under-tenancy by a junior line of the Cockfields see/ore/m ofBrakelond, p. 139, and The Kahndar of Abbot Samson, p. 4. 96

214

Colonial England, 1066-1215

even wider distribution within the family, in the early eleventh century, to systems of lineal descent, whether by primogeniture orparage, in the early twelfth century. The language survived these changes. Grants were made with the concurrence or consent of heirs in the early eleventh century. They were still made as concessions by the grantor and his heirs to the recipient and his heirs in the late twelfth century. But in the earlier period the heirs may appear in parallel as joint heirs; in the later they appear in series as successive heirs. Collaterals and distant heirs might figure under either system. This renders difficult, if not impossible, any statistical analysis along the lines attempted by Robert Fossier for Picardie.98 Nevertheless, there is enough casual information to indicate that division of the inheritance was quite frequent. Later sources, chiefly Orderic Vitalis, record examples of such successions from the second quarter of the eleventh century: Robert, archbishop of Rouen," Robert of Grandmesnil,100 Giroie,101 William of Belleme,102 and Ralph Tesson.103 Moreover, the language of inheritance was so deployed as to make no distinction between ancestors who were fathers and other antecedents, or between heirs who were direct descendents and collaterals. This open situation is well revealed in one of the earliest grants to St-Wandrille, 996-1006, in which a certain Albert conveyed his estate in Livry which 'had been conceded to him by his ancestors by hereditary right'. His sons, nephews and eight named blood relatives, consanguinei, whose precise relationship was not defined, figured with him as 'possessors' of the estate and fellow grantors.104 Men moved away from that primitive situation only slowly. Perhaps the main impetus to change came from the possession of castles, which could scarcely be partitioned and were not easily shared, and from a ducal interest in preserving service unfragmented. None the less on into the middle years of the century an inheritance might still be held by brothers, whether jointly or by division,105 and brothers might make or confirm benefactions in one and the same document.106 One such operation, the endowment of St-Evroul, was founded on the mutual support of the Grandmesnils and the sons of Giroie. Duke William's charter of confirmation of 1050 referred to 'the possessions which Robert and Hugh (de Grandmesnil) his brother and William (Giroie) and his brothers and sons, and other faithful Christians, had given or were to give from their inherited property, or which the same Robert had received for 98

Robert Fossier, La terre et les hommes en Picardie (Paris, 1968), pp. 262-73. Orderic Vitalis, ed. Chibnall, iii, p. 84. 100 Ibid., ii, p. 40. 101 Ibid., ii, p. 28, iv, p. 156; M. Chibnall, 'Les droits d'heritage selon Orderic Vital', Revue historique de droit franfais et etranger, fourth series, 48 (1970), p. 347. 102 Orderic Vitalis, ed. Chibnall, ii, pp. 362-65. 103 L. Musset, 'Actes inedits du Xle siecle', v, 'Autour des origines de St Etienne de Fontenay' Bulletin de la societe des antiquaires de Normandie, 56 (1961-62), 26-29. 104 RADN, no. 7. 105 Ibid., nos 19 (1006-17), 21 (1015-17), 46 (1017-26), 94 (1035-r. 1040). 106 Ibid., nos 24 (1017-23), 118 (c. 1049-1051), 119 (c. 1040-50), 147 (1060), 197 (c. 1050-66). 99

Notions of Patrimony

215

himself from his brother and his other co-heirs according to the agreed partition'107 (juxta compladtam parcium divisionem). Similar arrangements were still in vogue both before and after the English expedition. Between 1051 and 1066 Richard and Roger, sons of Herluin the Seneschal, along with their mother, granted the church of Authevernes and other properties to La Trinite du Mont. That each brother in addition granted properties e proprio jure demonstrates that they had a common interest in the joint grant.108 Something similar occurred in 1092 when Roger Tanetin and his sons and four brothers Helte, Reinfrey, Thurstan and Robert, 'firmly and hereditarily conceded and sold' to Arnulf abbot of St-Martin of Troarn and his monks part of the church and all the land which they had in Guillerville in return for 30 livres of Le Mans, apparently paid to them jointly.109 How far an ordered system of succession, giving priority to the eldest son and awarding other sons and collaterals appropriate precedence, had emerged by the time of the English expedition is a matter of guesswork. It would be convenient, but quite wrong, to take arrangements established in England after 1066 as an indication of how things stood in Normandy. Anglo-Norman and Norman succession soon parted company, the one accepting primogeniture, the other parage. Quite apart from that the Conquest added enormously to the acquisitions available to the participating families, and this must have facilitated and accelerated the establishment of lineal succession. Some of the divisions which occurred in Normandy before the Conquest can be interpreted as separations of inheritance from acquisition. The acquisition of land, whether by purchase or by grant, was common enough, but there are few references to acquisition eo nomine and none to suggest that the clear-cut distinction between inheritance and acquisition stated in the Leges Henrici Primi already had the status of a clearly defined rule.110 The rule, like the acquisitions themselves, was probably a product of the Conquest. The definition of the acquisition was merely one facet of a far larger problem. The counterpart of King William's feudal superiority over his new realm was 107

Ibid., no. 122. Cf. Orderic Vitalis, ed. Chibnall, ii, pp. 16-18, 32-40; Orderic Vitalis, ed. Le Prevost, v. 173-80. Compare the foundation of the priory of Maule, Orderic Vitalis, ed. Chibnall, hi, pp. 172-74, 182-90. 108 RADN, no. 202. 109 BN, MS lat. 10086, fos Ixxxxiiib-lxxxxiiii, R. Sauvage, L'abbaye de Saint-Martin de Troarn, p. 143. 110 In the Norman sources I have found nothing comparable to the precise contrast between inheritance and acquisition drawn in the agreement between Tescelinus, priest of Verri, and the monks of St-Florent-les-Saumur, 1055-70, see above, p. 122 n. 57. William fitz Osbern's endowment of Lyre included the tithes of all the lands and mills which he had acquired in the Bessin and 'the tithe of all things which he might acquire during the rest of his days', but the bulk of his grant was made from his inherited lands in Lyre, Breteuil and elsewhere (RADN, no. 120). Orderic Vitalis emphasised that the benefactions of Roger of Montgomery to Troarn, Seez and other houses were made from his acquisitions not from his paternal inheritance and he returned to this theme in discussing the foundation of Shrewsbury (ed. Chibnall, iii, pp. 142-144). Cf., however, below, p. 216.

216

Colonial England, 1066-1215

that initially, certainly for the first generation, no land in England could have been held by the newcomers in inheritance: all was acquisition. There is therefore a real break in continuity, and that coincides with a real gap in the evidence. We may fill that gap if we like by imagining that William's enfeoffment of his followers was recorded in charters which happen all to have been lost, but that would be to credit the eleventh century with the practices of the twelfth. The plain fact is that at the point both in time and place at which inheritance ceased to provide a point of reference the documentation changed its tone. The language of inheritance which had been used so variously and so vigorously in Normandy scarcely appears at all in England in the first years after the Conquest. No one held jure hereditario. No one made endowments to be held jure hereditario. Alo dium was now used to describe strange Anglo-Saxon forms of tenure, to be sure, these changes occurred amidst a broader terminological development. Alodium and benefidum were being replaced and to some degree amalgamated in feodum. Endowments to be held jure hereditario were being succeeded by grants in perpetuity in free and pure alms, a formula already deployed in the endowments of Richard of Clare before 1086.111 Nevertheless, the circumstances which followed the Conquest enforced change. The old security of tenure jure hereditario had gone, at least for the time being. The change is revealed most clearly in the benefactions of those families with maintained houses both in Normandy and England. The language in which Roger de Montgomery announced his foundation of the abbey of Troarn was unequivocal: 'I grant my own property of my inheritance' or 'I have determined to give a considerable part of my inheritance'.112 In 1083 this same Roger promised to found the abbey of Shrewsbury. Neither in his spurious foundation charter nor in the almost equally unsatisfactory confirmation of his son Hugh, nor in Hugh's authentic confirmation of 1094-98, is there any reference to tenure jure hereditario, although Hugh's authentic confirmation was made in pure and perpetual alms.113 Between 1051 and 1066 Roger de Bully sold to Rainer, abbot of La Trinite du Mont, the tithe of the vill of Bully in that it belonged to him by hereditary right.114 In 1088 this same Roger founded the priory of St Mary of Blyth. His various benefactions recorded in the foundation charter were made in perpetuum. There was no reference to tenure jure hereditario.115 It is plain that the English and continental possessions of these two men did not merit the same language. Even more important, it was not that the language was forgotten. It is rather that it was held in abeyance because it did not fit English circumstances.

111 Select Documents of the English Lands of the Abbey of Bee, ed. M. Chibnall, Camden third series, 73 (1951), pp. 21-22. 112 R.N. Sauvage, L'abbaye de Saint-Martin de Troarn, pp. 347, 352; /MDAT, no. 144. 113 The Cartulary of Shrewsbury Abbey, nos 2-5. 114 RADN, no. 200. 115 The Cartulary of Blyth Priory, ed. R.T. Timson, Thoroton Society, record series, 27 (1973), no. 325.

Notions of Patrimony

217

That brings us finally to the way in which those terms came to be used in the twelfth century, to the hereditary tenure of the Cockfields, to enfeoffment in feodo et hereditate, to tenure jure hereditario, and to the cartae hereditariae of William of Diss. The Cockfield evidence by itself, and many other similar collections, could easily be interpreted each as a story of hereditary title gradually emerging in a strongly seignorial context which emphasised service and feudal superiority. But long before this happened the relevant legal terms had already undergone prolonged and varied usage. That all this was forgotten, that it might have been necessary to start from scratch as it were, is improbable, indeed scarcely imaginable. There are enough indications that it was not forgotten. Orderic Vitalis has been criticised for presenting the eleventh century as seen through the spectacles of the twelfth, in exaggerating, for example, the role of primogeniture.116 But that apart, he wrote of hereditary title in the eleventh century in terms closely similar to those used to describe it in contemporary documents. Estates are held by hereditary right; inheritances pass to nephews and cousins,117 to sons-in-law jure uxoris,118 to sisters in the absence of male heirs,119 as well as to direct male descendents. Rightful heirs maybe disinherited.120 Indeed disinheritance is a denial of justice, a cause of protest121 or of feud, war and rebellion.122 We could as well charge Orderic with conserving this language and projecting these conditions into his own age. In reality he simply demonstrates that the language and the imagery were unbroken.123 116

See Chibnall in Orderic Vitalis, ii, pp. xxxvi-xxxvii; idem, 'Les droits d'heritage selon Orderic

Vital'.

117

Orderic Vitalis, ed. Chibnall, ii, pp. 80, 358; in, p. 200; iv, p. 216. Ibid., iii, p. 116; vi, pp. 42, 176. See also Orderic's statement that Duke Richard II conveyed to Giroie in hereditary tenure the lands which had been held by Heugon, father of his betrothed, even though she died before marriage (ibid., ii, p. 22). 119 Ibid.,ii,p. 116. 120 Ibid., ii, p. 282; iv, pp. 158, 302; vi, p. 33. See especially the vision of the knight William of Glos who had taken up a poor man's mill in return for a loan and disinherited the lawful heir by leaving it to his own heirs (ibid., iv, p. 244). 121 See especially the protest of Ascelin son of Arthur that the body of the Conqueror should not be 'covered with soil or laid in his inheritance' (iv, p. 106). The story was also known to William of Malmesbury, Gesta regum, ed. W. Stubbs, Rolls Series (London, 1889), p. 337. It reveals that a sense of injustice might lie hidden behind transactions which are recorded simply as sales. For the record evidence see Les actes de Guillaume k Conquerant et de la reine Mathilde pour les abbayes Caennaises, ed. L. Musset, Memoires de la societe des antiquaires de Normandie, 37 (1967), no. 14 and pp. 45-46. 122 Orderic Vitalis, ed. Chibnall, v, p. 156; vi, pp. 196, 220, 372. 123 Orderic, however, gave the language his own special nuances which are not shared by the documentary evidence. He had a very strong sense, not infrequently expressed, of 'natural' or 'genuine' heirs, by which he meant heirs of the blood or rightful heirs (ibid., ii, pp. 96, 130, 190; iv, p. 76; ii, p. 304). He also used patrimonium, which occurs only very rarely in records, quite frequently as a loose literary synonym for a paternal inheritance. Sometimes he deployed the word in a more pointed way to emphasise a claim in inheritance (ii, p. 122) or to contrast inheritance with acquisition (iii, p. 262; vi, pp. 328-30, 402). But he could also use it very loosely and inconsequentially. See his comment that the refounder of Crowland Abbey, Thurketel, possessed sixty manors 'de patrimonio parentum suorum' (ii, p. 342). In a different vein compare Robert 118

218

Colonial England, 1066-1215

That interpretation gains support from the manner in which these terms were deployed in England, in charters of endowment or enfeoffment, once inheritance came about with the passage of time and the succession of generations. It is very remarkable that the notion of inheritance appears suddenly on the scene in its fully developed form with very little indication of tentative experiment. In Normandy, over a period of half a century or more, men had worked their way by an inexorable logic from fact to expectancy, from the notion of inherited right to heritable Claim. In England from the start the terms are used to express intention, often stated very precisely. Bishop Ranulf Flambard's enfeoffment of William son of Ranulf in Houghall and elsewhere in 1114—16 comprised terras jure hereditario possidendas.124 A redrafted version of the same grant of 111 6-19 stated that the lands were granted to William and his heirs who would succeed him per hereditariam successionem.125 Almost universally this was the sense of hereditary tenure in the charters surviving in increasing numbers from the 1120s, 1130s and 1140s in which grants were made from the grantor and his heirs to the recipient and his heirs to be held infeodo et hereditate. Such tenure could be created from scratch.126 Hereditas could be an expectancy to which the heir had yet to succeed.127 Men could be deprived of their hereditary lands and yet return to claim them.128 Claim in hereditary tide could be precluded from tenements held for life. In 1136-39 Theobald, archbishop of Canterbury, stated that the manor of Stisted had been violently usurped by John son of Anfrid after the death of his predecessor, William of Corbeuil, in 1136; John's father, Anfrid, had held the land at farm in the time of Archbishop Anselm and his son, said Theobald, never had aliquam hereditatem sive aliquod feodum.129 That illustrates the violence which might lie behind the establishment of hereditary tenure. But it also contains in its denial a clear-cut notion of heritable title. Who is speaking to us in these and similar passages? Here perhaps Theobald continued

Curthose's promise to his followers of 'plurima quoque patrimoniis eorum augmenta' (ibid., iii, p. 102). 124 Durham Episcopal Charters, 1071-1152, ed. H.S. Offler, Surtees Society, 179 (1968), no. 11. 125 Ibid., no. 12. 126 Grant by William Peverel of Dover, 1121-22, to his steward, Thurstan, 'tenendas in feodo et hereditate', P.M. Stenton, First Century of English Feudalism (Oxford, 1961), p. 274; cf. RRAN, ii, no. 1295. Grant by Osbert of Arden, c. 1130, to Gerard and Nicholas, sons of Thomas 'in feudo et hereditate', Stenton, First Century, p. 280. For a variant formula of land to be held 'jure hereditario' see the grant by William, earl of Lincoln, c. 1145, to Peter of Goxhill, Documents Illustrative of the Social and Economic History of the Danelaw, ed. F.M. Stenton, British Academy (1920), no. 490. See also a grant by Peter son of William, c. 1125, to William son of Reinfrey and his heirs infeodo et hereditate, which illustrates the intimate connection between inheritance and lordship, with William son of Reinfrey and his heirs to hold in perpetuity 'quia ipse inde meus homo est', Sir Christopher Hatton's Book of Seals, ed. L.C. Loyd and D.M. Stenton, Oxford, 1950, no. 528. 127 Such is the implication of charters embodying the reddidisseformula. For examples see Stenton, First Century, p. 272. 128 Charters of the Honour of Mowbi ay, 1107-1191, ed. D.E. Greenway, British Academy, Records of Social and Economic History, new series, 1 (1972), no. 3, p. 10, discussed above, pp. 176-78. 129 A. Saltman, Theobald, Archbishop of Canterbury (London, 1956), pp. 271-72.

Notions of Patrimony

219

himself, but in the great majority of the documents we simply do not know. Very few are as specific as William de Briouze's charter of 1080 to St-Florentles-Saumur which states that it was composed and drafted by Primaldus his clerk.130 It must be allowed that the words and phrases which have made up the ammunition for my argument came immediately not from the authors of the acts from which they are drawn but from clerks, sometimes from beneficiaries, sometimes from superior authorities both lay and ecclesiastical. But that scarcely weakens the argument. Bishops and abbots were noblemen's brothers and cousins. If it is argued, for example, that beneficiaries played a large part in the drafting of these documents, then it follows that a large role must be given to all those monks of Norman and French houses who invaded England to establish the dependent cells which were the most immediate expression of the thanks which their secular patrons felt were due to God and the saints after the victory of 1066. Patron and beneficiary alike were familiar with the terms of Norman endowments. Hence it may be that these intricate concerns do not matter very much for the present argument. There is finally one document, the greatest of them all, where there is no doubt about who is speaking. The charter of liberties of Henry I has become something of a puzzle when set against all that has been written in recent years on the insecurity of title and inheritance. For just as private Anglo-Norman charters present tenure infeodo et hereditatefully grown, so Henry's charter projects a world in which all the incidents and trappings of such tenure were already in full operation. Rufus may have forced his vassals to ransom their lands, but the charter is clear evidence that there was a notion of ajust and lawful relief applicable both to the king's barons and to their men. Widows were entitled to dower and marriage portions, and children, both male and female, might be in wardship. In the end the case can be left to depend on a single word. The central sentence of chapter 3 runs: And if when a baron or other man of mine dies a daughter remains as heir, I will give her with her land by the advice of my barons.

This assumes that in the ordinary course the heir would be known. Her identity was not determined by the king's court. It was recognised by the king's court by reference to a set of social conventions which had grown both within and outside it.131But it was from her husband to whom he gave her that the king took homage. That was the final recognition and warranty of title.132 But how then could the king accept that the land might be described as hers not only 130

M.P. Marchegay, Charles normandes. . .de St-Flwent, p. 680. I do not intend to preclude here the possibility that in particular cases the king might adjudicate between the competing claims of heiresses or allocate and define the rights of seniority between co-heiresses. 132 S.F.C. Milsom, 'Inheritance by Women in the Twelfth and Early Thirteenth Centuries', On the Laws and Customs of England, ed. M.S. Arnold, T.A. Green, S.A. Scully and S.D. White (Chapel Hill, NC, 1981), pp. 60-89, esp. 62-65. 131

220

Colonial England, 1066-1215

before marriage but even before it had been decided to whom she might be married? In jurisdictional terms she had no title, yet the phrase is cum terra sua. That sua derived not from jurisdiction but from the social conventions within which the jurisdiction was set to work. These affected king and heiress alike. It is he who allows, not she who claims, that the land is 'hers'. They were at one. The first known ancestor of the Cockfields, Lemmer, was to go by his name, Leofmaer, an Englishman. How he established himself among the Norman knightly tenants of Bury St Edmunds we can only guess. One thing is certain: as his descendants established their varied heritable title in the following generations they drew not only on the developing resources of the courts of the abbot and the king but also on notions which had deep roots in the Norman past. Without them the family's title deeds might have taken a very different form: landbocs, for example. But without them there might have been no Norman Conquest.

Note

In all that has been written of the enfeoffment of the King William's knight, Peter, by Abbot Baldwin of Bury St Edmunds, it has been assumed rather too easily that the sole piece of documentary evidence, of which the earliest copy comes from the late thirteenth century, is a standard charter of enfeoffment as such were known in the late eleventh century.133 This is not so. The document has some of the characteristics of a notification, but whether it was intended as more than a memorandum for the abbot and convent is an open question. Stenton, who described it as a 'memorandum', was properly cautious of such an 'ill-drafted record' ,134 Moreover, the wording, even if assumed to be an accurate record of an authentic eleventh-century original, can scarcely bear the construction which it has been made to carry. It is difficult to follow Douglas in arguing that it reflects the supervision of local feudal institutions by the king or that the king was here endowing one of his followers.135 The document simply seeks to define the duties of a vassal of the abbot who was also in the service of the king. The interpretation frequently placed on it, that it provides evidence that feudal military service was ill-defined in the generations following the Conquest is insubstantial, for the simple reason that the document does not clearly state, as has been claimed, that Peter was to serve 'within the kingdom with three or four knights'.136 It is better construed as follows: 133 D.C. Douglas, 'A Charter of Enfeoffment under William the Conqueror', EHR, 42 (1929), p. 245; cf. Feudal Documents of Bury St Edmunds, p. 15; C.W. Hollister, The Military Organisation of NormanEngland (Oxford, 1965), pp. 50-52; RA. Brown, Origins of English Feudalism (London, 1973), pp. 90,138. 134 Stenton, First Century, p. 154, 170. 135 Douglas, 'A Charter of Enfeoffment', p. 246. 136 RA. Brown, Origins, p. 138.

Notions of Pa trimony

221

If, before he is summoned on behalf of the king, he is summoned on behalf of the abbot, Peter promises to serve within the realm in keeping guard, either before or after the royal campaign, with three or four knights at his own expense. If, as the abbot's attorney, he pleads within the realm he shall be sustained at his own expense. But if the abbot takes him with him anywhere, it shall be at the abbot's expense. In addition to this he shall provide a knight as the abbot's own, within or without the realm, where and when the abbot wishes to have one. Read thus, the service of three or four knights, which has occasioned so much discussion because of its vagueness, must be taken to refer to service in custodia not in expeditione. Moreover, the do.cument can now be seen to face up not only to the problem of arranging the feudal service of a vassal who was also a royal knight, but also to the stark fact that no feudal lord could expect personal service in custodia and in expeditione at one and the same time. The document does not define service in expeditione, presumably because Peter would then be in the king's not the abbot's service. It may be that the final sentence was aimed at repairing this deficiency. Additional Note The Norman evidence has now been further explored by Emily Zack Tabuteau, Transfers of Property in Eleventh-Century Norman Law (Chapel Hill, NC, 1988); and by David Bates, Normandy before 1066 (London, 1982), especially pp. 111-21. J.H. Hudson, Land, Law and Lordship in Anglo-Norman England (Oxford, 1994), is a particularly important investigation of the language of charters and the legal intentions which it conveyed. Stephen D. White, Custom, Kinship, and Gifts to Saints (Chapel Hill, NC, 1988), provides a comparison with a broader area of western France.

This page intentionally left blank

12 Feudal Society and the Family in Early Medieval England, III: Patronage and Politics

In a previous essay, I posed a question: whether family relations mattered. I now intend to investigate in what way they mattered. It is a complex problem. In the middle ages men were reminded more frequently than now by the varied representation of the tale of Cain and Abel that family relations are not always easy; they could be supportive, cohesive, quarrelsome or murderous. So it is reasonable to ask: why the one rather than the other? Behind that there are other questions: when did the family matter and when not? Why did it matter when it did, and why when not? A long established tradition in Anglo-Saxon scholarship has been concerned with the respective roles of kinship, lordship and local association. What can be said on the same topics for the postConquest period? For it is plain that there can be no secure conclusions about the family unless it is correctly located and properly embedded in the matrix of social and political relationships within which men lived their lives. It is a large topic. In discussing it I shall concentrate on the years between 1066 and 1216. In doing this I have a second set of objectives in mind. The continental scholars who have reshaped the history of the family in the last forty years or so have had to put up with basic material which is meagre or highly specialised or both. By comparison the English sources are replete. Two are of particular importance: the rolls of the Curia Regis and the justice in eyre, and the rolls of fines and offerings made before the king or his officers. The first require no special comment. The second have long been exploited for illustrative material on the exercise of royal patronage. They certainly provide that, but they also reflect the social and political ambitions and objectives of those making fine. They illustrate both supply and demand, and up to 1215 they reveal this interchange between king and vassals in relatively untrammelled circumstances, limited to some degree by conventional restraints but not yet subjected to the tighter restrictions imposed by the Great Charter. The proffers of the fine rolls and the procedures and pleadings in the law courts allow a far deeper and more systematic examination of the operation and importance of family ties, and that at more varied social levels, than is possible on the continent. The pipe rolls, which summarise the accounting consequences of both fines and pleadings, carry the story back, though not so clearly, to the middle of the twelfth century and to the isolated roll of 1130. Accordingly it is reasonable to ask how the continental reconstructions are to be related to the English evidence.

224

Colonial England, 106&-1215

The first requirement of those who enter these waters is that they should resist all temptation, for sirens call from every rock. Family relationships played a part in historical explanation long before their apotheosis in the pages of the Annales and the work of the Freiburg school. Consider Sidney Painter's comment, written in 1949, on the rebellion of 1215: A number of scholars have suggested the possibility that ties of blood led many men into the baronial party. There is ample evidence that the family played an important part in thirteenth century politics, but we know little of how family was defined in the minds of the men of the time. Did a man feel that he had family obligations toward his second cousin?1

Cautionary words those, but nevertheless he concluded: 'The correlation between the blood of the house of Clare and the opposition to [King] John seems too close to be mere coincidence',2 and again - 'it is clear that blood relationship may well have played an important part in forming the baronial party' .3 Painter was in good company for the pattern here had been set in an appendix on the Twenty-Five barons of Magna Carta, complete with genealogical table, which Powicke included in 1928 in his Stephen Langton. Here, too, there was a cautionary note, that of a guardian angel rather than a siren: My friend Mrs Stenton, who has sent me some valuable criticisms on this appendix, suggests that I have laid rather too much stress upon the influence of family connexions. In 1215, as at other times of civil strife, families were divided, as they must always be when the ramifications of kindred are so intricate and widespread in society; and it would indeed be unfortunate if, in the interpretation of political movements, the argument from connexions of this kind were pressed without caution. But the existence of close family ties does seem to me to have been a significant factor in forging a definite group of men personally hostile to King John.4

So in the end Powicke, too, took the lure, just as Painter did, and indeed both were following a long historical tradition. In the study of the rebellion of 1215 it went back to a short but important note on King John and Robert fitz Walter which Round published in 1904,5 and to mention Round at once broadens the field of study, for his Geoffrey de Mandevilk, published in 1892, was the first systematic application of genealogical study to political history. In all this work it is fair to say that family relationships were viewed not just as a kind of political cement, a bond taking effect once political groups were formed, but as a dynamic explanation of why groups were composed as they were and, in some degree, acted as they did. 1

Sidney Painter, The Reign of KingJohn (Baltimore, 1949), p. 290. Ibid., p. 291. 3 ibid., p. 292. 4 P.M. Powicke, Stephen Langton (Oxford, 1928), p. 212. 5 J.H. Round, 'King John and Robert fitz Walter', EHR, 19 (1904), pp. 707-11. Powicke acknowledged that this paper of Round 'led the way', Powicke, Stephen Langton,, p. 207 n. 1. 2

Patronage and Politics

225

Painter took an extreme line. He counted fourteen of the Twenty-Five barons of Magna Carta as members of what he called the Clare clan or the house of Clare. For this he traced the issue of Richard fitz Gilbert of Clare who died c. 1090.6 The method was highly questionable, for even in the direct line the resulting links were very attenuated. Richard earl of Clare in 1215 was the great great grandson of Richard fitz Gilbert; Robert fitz Walter was his great grandson.7 Outside the direct line it involved the supposition that in 1215 there was still some importance to be attached, for example, to the marriage of Alice de Clare, grand-daughter of Richard fitz Gilbert to Aubrey de Vere in the time of Henry I, and to the marriage of their daughter to Geoffrey de Mandeville, first earl of Essex who died in 1144.8 But why should not these ladies on their marriage be counted respectively as members of the houses of Vere and Mandeville? And why finally choose Clare as the base for the calculation? Richard, earl of Clare, was the one member of the Twenty-Five who descended in the direct line from one of the greatest pre-Conquest families. His family would not have known its ordinary business if over the years it had not accumulated a web of intricate relationships around it. Yet these relationships were subject to an inexorable law which widespread pedigrees necessarily ignore: the more distant the kin the more ancient the ancestry, and the more ancient the ancestry the more subject to the passage of time and hence to change: change in family structure or in the environment which shaped it. Painter's question, 'Did a man feel family obligations toward his second cousin?', brings in its wake the question: 'Was the relationship between father and children the same over three generations?' It would be difficult to point to any three generations in post-Conquest England for which the answer would be a confident 'yes'. There is yet another, more obvious difficulty. The most seductive note in the sirens' call is one which leads to a predetermined selection of the evidence. Indeed as historical explanations go, whether of events or structures, there is nothing quite so neatly predetermined as the genealogical explanation, for where it works it seems to work very well, and where not, it may be treated as aberrant or no one seems to notice. But in 1215 or shortly afterwards one man noticed, for the Barnwell chronicler, the shrewdest of all near-contemporary commentators on the rebellion against King John, pointed out that families were divided: Many joined them [the baronial party], especially the younger men, the sons and nephews of the magnates, as though they wanted to make their name through warlike deeds. Hence we soon saw houses divided within themselves, the fathers and older men adhering to the king as their lord, the sons and younger men standing on 6

Painter, KingJohn, p. 291. J.H. Round, Feudal England (London, 1895), p. 472; Michael Altschul, A Baronial Family in Medieval England: The Clares, 1217-1314 (Baltimore, 1965), table 1. 8 Complete Peerage, v, pp. 113-16; x, p. 198; Painter, KingJohn, p. 291; Powicke, Stephen Langton, pp. 210-13. 7

226

Colonial England, 1066-1215

the opposite side. We knew some who crossed to the other part for love of their sons. And there were some who did not agree with them at first, but who joined them because they were friends of fortune or because they loved new things.9

This writer did not say that the baronial party was made up of a network of interrelated families. That may have seemed so commonplace to him that it was not worth saying. But that is merely the easiest, not the best, way of reconciling his views with those of Powicke and Painter. Their arguments seem to gain strong support from more recent work of continental scholars, but this is more apparent than real. The work of Tellenbach, Schmid and others in Germany,10 Genicot and Warlop in Belgium,1T Duby12 and Fossier13 and others14 in France, not to mention the contributions to these studies by Karl Leyser and Professor Bullough,15 is of enormous importance in explaining family structures and the crucial changes which they underwent between the tenth and twelfth centuries; but the emphasis which has been placed on family solidarities and lineage is almost predetermined by the nature of the sources used. Apart from charters, which are common ground in all these studies whether concerned with the Continent or with England, much of this continental work has depended on two closely related categories of evidence: memorial books, recording the prayers due for the souls of patrons and benefactors, and family genealogies. Both are ecclesiastical in origin and both emphasise lineage and family solidarity because one of their objects was to maintain the link with the benefactor and his heirs, thereby ensuring continued possession of the benefaction.16 So the history of the family is inextricably mixed with the history of endowment. In England the libri memoriales or libri vitae have not played so important a 9

Memmialefratris Waltm de Coventria, ed. W. Stubbs, Rolls Series (2 vols, London, 1872-73), ii, p. 220. 10 See the essays conveniently collected in The Medieval Nobility, ed. T. Reuter (Amsterdam, 1978). 1 * See the essays conveniently collected in L. Genicot, La noblesse dans Vocddent medieval (London, 1982); E. Warlop, The Flemish Nobility (4 vols, Kortrijk, 1975-76). 12 See the essays conveniently collected in Georges Duby, Hommes et structures du moyen age (Paris, 1973), all the immediately relevant items wherein are translated in Georges Duby, The Chivalrous Soriety, trans. Cynthia Postan (London, 1977). See also Georges Duby, Medieval Marriage (Baltimore, 1978). 13 Robert Fossier, La terre et Us hommes en Picardie jusqu 'a la Jin du XIHe siecle (Paris, 1968). 14 See particularly La noblesse au moyen age, ed. P. Contamine (Paris, 1976). 15 KJ. Leyser, The German Aristocracy from the Ninth to the Early Twelfth Century', Past and Present, 41 (1968), pp. 25^-53, reprinted in KJ. Leyser, Medieval Germany and its Neighbours, 900-1250 (London, 1982). See also KJ. Leyser, Rule and Conflict in an Early Medieval Society (London, 1979). Cf. D.A. Bullough, 'Early Medieval Social Groupings: The Terminology of Kinship', Past and Present, 45 (1969), pp. 3-18, and KJ. Leyser, 'Maternal Kin in Early Medieval Germany', ibid., 49 (1970), pp. 126-34. 16 On the libri memoriales see especially Karl Schmid, 'Uber die Struktur des Adels im fruheren Mitte\a\ter',JahrbuchfurfrankischeLandesforschung, 19 (1959), pp. 1-23, trans, in Reuter, TheMedieval Nobility, pp. 37-59, especially pp. 42-46. For criticism see, among others, Leyser, The German Aristocracy', pp. 33-34.

Patronage and Politics

227

part in the story, but monastic family histories are common enough to be used frequently if warily by the genealogist. Two examples are sufficient to illustrate their purpose. One, a history of the Lacys, was probably composed at Whalley Abbey sometime between 1258 and 1377.17 It traces the history of the Lacys of Pontefract and Clitheroe from their establishment under the Conqueror and Rufus to a point where the tale breaks off in 1258. It contains some genealogical inaccuracies but the compiler was able to cope with a major dislocation in the descent in 1194 in which the last Lacy in the direct line was succeeded by a distant cousin, Roger Constable of Chester, whose ancestry was also included in the pedigree. In each generation the Historia re cords the major benefactions and confirmations of the Lacys both to Whalley and, less systematically, to other houses of the Lacy fee. It is a prolonged record of endowment in which continued possession was underwritten by the continued commitment of the lineage. A second example, contained in the opening chapters of Book I of the Liber memorandorumofRarnwell Priory, was written, the author tells us, to help protect the priory and the work of God from the impieties and forgetfulness of men.18 He was able to provide, among other things, a record, not always accurate, of the patrons' genealogy. This was a complex story. It began with the sheriff of Cambridgeshire, Picot, and his wife Hugolina who founded a house of canons in honour of St Giles. But it was interrupted in the next generation, for Robert, Picot's son, joined the conspiracies against Henry I. He was deprived of his barony and the canons lost their patron: the house was now poverty-stricken, spurned by men who had been friends and were now enemies. Its fortunes were revived by Pain Peverel who received Picot's barony from Henry I and refounded the house at Barnwell. The priory had a new patron and not surprisingly the author presents Pain as 'egregio militi, armis insigni, milicia pollenti, viribus potenti, et super omnes regni proceres bellico usu laudabili' and has him break out into devout rhetoric: Thirty years ago I was reborn through baptism. Thirty years hence I shall arise at a latter day. I now establish thirty canons who by God's grace will bring me to eternal life and hence just as I have succeeded to Picot in the place of an heir in possessing the inheritance so I succeed to him in endowing and extending this house.19 The author then describes Pain's endowments, moves on to those of his son (rectenephew), William, who died without issue c. 1148, and to a division of the barony first among four, then among three of Pain's daughters (recte nieces). The patronage of the Priory descended through the eldest of these and her husband, Hamo Peche, but the compiler relentlessly pursued the descent of 17

Monasticon Anglicanum, v, pp. 533-34. For discussion of the date and value of the Historia Laceiorum, see W.E. Wightman, The Lacy Family in England and Normandy, 1066-1194 (Oxford, 1966), pp. 12-14. 18 Liber memorandorum eccksicdeBnnewelle, ed.J.W. Clark (Cambridge, 1907), pp. 37-51; for the purpose of the book see p. 37. 19 Ibid., p. 41.

228

Colonial England, 1066-1215

the other two. The next sister, Rose, led through her daughter, Aubrey de Harcourt, to yet three more co-heiresses, Rose, Hilary and Agatha Trussebut. The third sister's inheritance also came to be divided between two daughters, the first of whom married twice. All that was fully recorded, as a reflection of family solidarity, certainly, but also as a record of possible warrantors of the priory's endowment. The story concludes with the death of the direct descendant of the senior heiress, Gilbert Peche, in 1291. He married twice and so loved his second wife Joanna for her gifts and beauty that he provided for their sons at the expense of the senior line whom, 4 I know not by what counsel' says the compiler, he sent away empty.20 He also surrendered his patronage of the house to the crown, a sad end which placed the house under the eye of the exchequer, but one which relaxed the need to pursue such complex genealogies in later generations. Monastic tending of the pedigree must have been matched at every stage by careful cultivation of the heir. This is less easy to study, but the following charter of 1140 x 1157 which comes from the highest ranks of the Anglo-Norman aristocracy, the lords of Tancarville, illuminates it with sharp edged brilliance: To all faithful in Christ William the Chamberlain junior greetings in the Lord. Be it known to all present and future that I William the chamberlain junior, on the fifth day after I was belted as a knight, came to St-Georges and there Abbot Louis and the monks, rejoicing with great joy, received me in a procession of honour, and there I offered up my sword on the altar of St-Georges. And then, by the counsel and advice of many of the noblemen and companions who had come with me, namely Robert des Ifs, my steward, and Roger de Cailly and John de la Londe and many others, I redeemed my sword by the gift and by the confirmation of many churches which I conceded to them on that day of my gift and confirmed them just as my grandfather, the founder of that monastery, had given them, namely the churches of Abbetot, Epretot with its tithes, St-Romain with two-thirds of its tithes, and Tibermesnil. I also confirmed the gifts of my knights and friends which they gave on the same day to the abbey in perpetual alms. Roger de Cailly gave 20s. annually, Robert de Mortemer 10s., Robert des Ifs 10s., John de la Londe, my cousin, 10s., Andrew de Bosemuncel 10s. or the tithes of one ploughland of land in Abbetot.. .and so on for another eight donors.21

20 21

Ibid., p. 50.

Rouen, Archives departementales, Seine Maritime, 13 H 15.1 have been unable to date this charter more precisely. The terminus post quern of 1140 depends on Robert of Torigny who states that William succeeded on the death of his father, Rabel, in that year, Chronique, ed. L. Delisle, Societe de 1'histoire de Normandie (2 vols, 1872-73), i, p. 218. But William was still alive in 1191 and may well have been a minor in 1140. See Complete Peerage, x, appendix F, pp. 52-54; L. Landon, Itinerary of Richard I, Pipe Roll Society, new series, 13 (1935), p. 50. A. Deville sheds no further light on the matter either in Histoire du chateau et des sires de Tancarville (Rouen, 1834), pp. 124r-27, or in Essai histoiifjue et descriptive sur Veglise et Vabbaye de Saint-Georges-de-Boschervilk (Rouen, 1827), pp. 29, 73-76, where the text is printed. Delisle, apparently referring to the same charter, gave a date 'anterieur a 1157'. SeeJ. Boussard, Legouvernment d'HenriIIPlantagenet (Paris, 1956), p. 361n. The terminus ante quern of 1157 depends on the death of Abbot Louis (Gallia Christiana, xi, p. 270).

Patronage and Politics

229

Not a bad day's work for the monks of St-Georges of Boscherville. The relationship between family benefactor and family foundation was just one manifestation of the exercise of patronage. And let there be no mistake about it, it was a powerful influence; the sense of desolation caused by a break in the connection which is manifest in the Barnwell pedigree or on larger canvas in Orderic Vitalis,22 or in the Walden chronicle,23 is evidence enough of that. But it was but one part of a larger pattern. The kin constituted a mutual benefit society of a very material kind and that over a far wider range of relationships than those likely to arise in the straightforward descent of estates. Some of the most striking examples were provided by the great administrative families, such as the Glanvilles, in which the links embraced brothers, cousins and other families brought into the fold by marriage.24 Such control of office and reward was good luck, a family windfall. On a lesser scale every family sought to make provisions which would keep its members happy. Fathers enfeoffed younger sons.25 Elder brothers enfeoffed their juniors.26 Uncles granted or confirmed lands to nephews,27 and nephews likewise to uncles.28 Godparents endowed god-children,29 brothers granted land to sisters,30 grandmothers to granddaughters,31 and landowners generally to feoffees whom they described simply as cognati?'2 Well before the end of the twelfth century men engaged in complicated negotiations which divided the inheritance and acquisitions of a family between sons.33 By 1200 the process had gone so far that a litigant, Elias Croc, could seek the judgement of the king's court whether his father could have lawfully enfe-

22 See the account of the expulsion of Abbot Robert de Grandmesnil, 1061, and the misfortunes of the founding family: 'Who can tell of the manifold tribulation that the church of St Evroul then endured?. . .During these uncertainties the church of St Evroul suffered great loss, and was robbed of many of her former possessions. . .So the church of St Evroul lost at that time many possessions that she has never, to this day, been able to recover'. (Orderic Vitalis, ed. Chibnall, ii, pp. 94-99). 23 See the effects of the intrusion of Geoffrey fitz Peter recorded in a narrative committed to the rival claimant, Geoffrey de Sai, Monasticon anglicanum, iv, pp. 145—48. 24 R. Mortimer, The Family of Rannulf de Glanville' Bulletin oj the Institute of Historical Research, 54 (1981), pp. 1-16; Leiston Abbey Cartulary and Butley Priory Charters, ed. R. Mortimer, Suffolk Record Society; Suffolk Charters, i (1979), pp. 1-9. 25 F.M. Stenton. TheFirst Century of English Feudalism (Oxford, 1961), p. 281; Sir Christopher Hatton's Book of Seals, ed. Lewis C. Loyd and Doris Mary Stenton (Oxford 1950), no. 204; EYC, xi, pp. 278-79, 327-28, 344. 26 Documents Illustrative ofthe Social and Economic History of the Danelaw, ed. F.M. Stenton (London, 1920), nos 458, 462; Facsimiles of Early Charters from Northamptonshire Collections, ed. F.M. Stenton Northamptonshire Record Society, 4 (1930), no. xxx(b); EYC, vii, p. 167; xi, p. 128. 27 Stenton, First Century, pp. 273, 276; Danelaw Documents, no. 476; EYC, xi, p. 62. 28 Danelaw Documents, no. 459; also in Book of Seals, no. 146. 29 Danelaw Documents, no. 483. 30 Ibid., nos 484, 485; Book of Seals, no. 48. 31 Danelaw Documents, nos 484, 485. 32 Ibid., nos 487, 509, 510; Book of Seals, nos 302, 330; EYC, xi, pp. 230-31. 33 Danelfiw Documents, nos 528, 529; Book of Seals, no. 434.

230

Colonial England, 1066-1215

offed his younger brother, Elias's uncle, of a fee which he claimed should descend to him as the caput of his honour and barony.34 So there was argument about it, argument most cogently expressed in Glanvill.35 But neither Glanvill's strictures against the alienation of part of an inheritance to younger sons nor the development of conditional gifts should be allowed to obscure the fact that family grants, rich in their variety, but made very frequently in fee and inheritance, were commonplace.36 Some escaped Glanvill's prohibition in that they were not held by knight service. Many were probably made from acquisitions, alienation of which, following twelfth-century practice, Glanvill allowed.37 The strength of the family tie was not expressed solely in enfeoffment. It motivated many of the fines proffered to the king: widows fining for the custody of their children,38 uncles fining for the custody and marriage of their nephews and nieces or for marriage partners for their nephews and nieces,39 co-heiresses and their husbands fining for confirmation of a partition,40 men fining on various grounds in the interest of brothers and sisters.41 Such matters might be conducted with formal precision with the crown ever watchful of its role as warrantor. When Saer de Quenci made fine of 1000 marks for the land of the dead earl of Leicester whose sister he had to wife, the agreement not only made provision for two dowers for the mother and widow of the dead earl, but also for Amice de Montfort, sister of Saer's wife and potential co-heiress: if she came to claim her portion and the king's court were to accept the claim she was to have it without any contradiction from Saer, and this he certified in letters patent to the king; but if she made good her claim in this she was to pay an appropriate portion of the fine: a businesslike conclusion, that, to an arrangement which observed all the family courtesies.42 Business was even more to the fore in the arrangements made for the disposal of the widow of Theobald Walter, the dead archbishop's brother, in 1207. Her father, Robert Vavassor, made fine in 1200 marks and two palfreys that she should have her dower and marriage portion and that he might marry her to other than the king's enemies. He did not choose well, and King John on this occasion was unusually tolerant, for Robert 34

Rot. de Oblatis etFinibus, p. 87. De legibus, vii, ed. G.D.G. Hall (1965), pp. 69-74. 36 Almost all the examples cited above were in fee and inheritance. For examples involving knight service see Facsimiles of Early Charters from Northamptonshire Collection, no. xxx(b); Stenton, First Century, p. 276; Danelaw Documents, no. 475; Book of Seals, no. 302. For examples involving homage see EYC, xii, pp. 54, 79. Glanvill's discussion of grants to younger sons assumes that they involved homage, which barred the donor and his heirs, De legibus, ed. Hall, pp. 70-73; compare his discussion of marriage portions, ibid., ed. Hall, pp. 92-94; and see especially SJ. Bailey, 'Warranties of Land in the Thirteenth Century', Cambridge Law Journal, 9 (1947), pp. 96-97. 37 De legibus, ed. Hall, p.7l. 38 Rot. de Oblatis etFinibus, pp. 8, 469, 478. 39 Ibid., pp. 2, 5, 14, 104,131, 192, 371, 485, 519. 40 Ibid., pp. 89, 108-9, 353, 507, 509. 41 Ibid., pp. 258, 293, 294, 317, 488. 42 Ibid., p. 268. 35

Patronage and Politics

231

immediately concluded an arrangement with the old marcher rebel Fulk fitz Warin, who married the widow and took over the fine, finding pledges for half of it, with Robert Vavassor and his brother William standing as guarantors both for the remainder and the whole.43 Similar examples, each with their own particular wrinkles and complexities, could be chosen almost at random year after year. The result of all such transactions was to inculcate a sense of family, an assumption that family relationships were determining across a wide range of activities. Men ask at trial by battle arising from a homicide whether a champion whom the defendant asserts is his nephew stands in near enough relationship to act for the defendant according to the custom of England.44 Time and again in all kinds of action litigants ask and the courts insist that juries should not include relatives of the parties.45 Pleadings drawn from the whole social range covered in the records of the king's court repeatedly reveal a generally shared assumption that family ties mattered. In 1201 a grand assize was initiated in Lincolnshire concerning three brothers, tenants of the abbot of Jervaulx in 180 acres in Somercotes.46 The action aborted and the lands came into the hands of the king for a year and a day because one of the brothers was appealed of robbery, convicted and hanged: the other two promptly deserted their lands which were found vacant. In 1206 Simon son of Martin was murdered at the entrance to the hall of Simon of Kyme at Bullington, Lincolnshire. One of the murderers was slain in church by Simon, clerk of Alford, the victim's kinsman. Later the victim's uncle accused Simon's son, Philip, of the murder. He alleged that he had fled the scene and failed to raise the hue and cry because Simon's servants outweighed him and he maintained that he had failed to report the matter to the coroners for fear of Philip's relatives and kin.47 In 1207 Earl Roger Bigod was involved as defendant in an assize de fossato levato in Norfolk. He challenged the names of the jurors on the ground that they all came from within the liberty of the earl of Arundel, whose sergeant the plaintiff was, and maintained that the plaintiff was seeking to attach pasture in the earl's wood to the earl of Arundel's barony. The plaintiff argued that the objection that the jurors were suitors of the earl of Arundel was insufficient. The king instructed that the assize should proceed according to the custom of England but not before Earl Roger had pointed out that some of the jurors were men of the earl of Warwick who had married the daughter of the earl of Arundel.48 There, in ascending social scale, from the freeholder of the first example to the great earl of the last, there was a common assumption about family ties which was deeply 43

Ibid., pp. 383, 405-6. Ibid., p. 87; Curia Regis Rolls, i, p. 429. 45 Curia Regis Rolls, i, pp. 246, 346, 377, 378; ii, pp. 33, 222, 291; iii, p. 21; iv, p. 194. 46 Ibid., i, p. 374. 47 The Earliest Lincolnshire Assize Rolls, 1202-9, ed. Doris M. Stenton, Lincoln Record Society, 22 (1926), nos 1451, 1475; Curia Regis Rolls, iv, p. 231; vi, p. 207. 48 Curia Regis Rolls, v, pp. 33-34. 44

232

Colonial England, 1066-1215

engrained, axiomatic, unquestioned. But the word 'assumption' may mislead because it implies something passive, a kind of datum line. On occasion the sense of family was an active, motivating force. Appeals of homicide were still made by the victim's relatives.49 When the king pardoned a homicide he could condone the felony, but not the compensation still due to the victim's family.50 The instinct of the feud with the attendant evasory compensations still ran through the social structure. Not all Europe shared Edward Plantagenet's sense of shock when Guy de Montfort murdered Henry of Almain in revenge for his father in Viterbo Cathedral in 1271.51 Why then question the unquestioned? One reason, as was suggested earlier, is that the family relationship is highly selective. Its strength cannot be measured accurately except by comparison with other possible associations. The role of Glanville's family in his ministerial relationships, for example, has to be related to his other professional connections, and the same argument applies mutatis mutandis to the great episcopal household and to the pattern of monastic endowment. Another complication arises from the particular slant which much of the evidence places on the family relationships. Patronage presents the family viewed from its fringes looking inwards towards the centre. It registers the attraction of the outermost satellites to the source of reward and power. In like manner monastic endowment reflects the work of monks exploiting every distant possible connection in pursuit of benefaction. It might be possible to resolve these two related difficulties by a careful analysis of the role of Glanville's family within his total professional relationships, but the result would probably be very impressionistic. An analysis of monastic benefaction would be easier in formal terms, but any counting of benefactions which distributed them between the founding family and other sources would run into the difficulty that these other sources might well have their own family ties. Fortunately there is a better method and that is to look at the family not from its fringes inwards but from the centre, from the head of the family, outwards. Where did he turn when he required support and what view of his responsibility was taken by his lord? One answer comes readily enough from the list of pledges and guarantors of fines entered on the dorse of the fine rolls. When Robert Vavassor and Fulk fitz Warin concluded their arrangements for the marriage of Robert's daughter, the king's approval was bought at the cost of Fulk's providing immediate guarantees for the payment of 600 marks, namely half the fine.52 It was noted at the foot of the entry on the roll that the pledges were entered on the dorse and there indeed they are, forty-four in all.53 Robert Vavassor's cousin Mauger 49

Naomi D. Hurnard, The King's Pardon far Homicide befoie AD 1307 (Oxford, 1969), pp. 171-

213.

50

See for example Rot. de Oblatis ctFinibus, p. 121. F.M. Powicke, King Henry III and the Lard Edward (Oxford, 1947), pp. 609-11. 52 Rot. de oblatis etfinibus, pp. 405-6. 53 Ibid., pp. 459-60. 51

T

233

guaranteed 40 marks of the total 600 marks. The remaining guarantors included the earls of Hereford and Salisbury and the count of Evreux, but the most prominent component was made up of barons and knights of the west midlands and the marches, where Fulk fitz Warin's power was centred: Beauchamp, Clifford, Courtenay, Basset, Lacy, St-Martin and Briouze. This particular list of pledges is a little unusual in that it includes considerable baronial 'top brass'. In many other instances it is apparent that men found their guarantors in the shire court and indeed the sheriff or other local agent of the crown might well be instructed to seek pledges there before putting the profferer in seisin of the object of his desires.54 Occasionally a member of the family played a special role in guaranteeing a proffer or a debt; in 1203 Robert fitz Roger stood as pledge for his nephew, Roger de Lacy, for the repayment of the loan of £1000 which King John advanced for the payment of his ransom after his capture at Chateau Gaillard;55 but such arrangements are not a prominent feature of the evidence. The main lesson of the lists of pledges is that men sought support locally, in the shire court or more widely in their associations in the royal court. The family had only a modest and entirely secondary role to play in these matters and it is easy to see why. Dependent members of the family, or distant members with no independent resources of their own, could scarcely act as guarantors of the debts and promises of the head. The fortunes of the head did much to determine the fortunes of the rest of the house. Yet when the head of the family was under pressure, committed like Fulk fitz Warin to some venture in the marriage market or subjected to some heavy penalty by the king, the dependent family was of little use to him. He had to turn elsewhere, to fellow landowners. The resulting experience which they shared at the receiving end of the royal prerogative became a powerful force in English politics. In more testing relations between the crown and its vassals it was a different story. Kings took hostages. In 1215-16, King John, faced with the threat or fact of rebellion, took large numbers. He did not want second cousins twice removed; there was no guarantee of loyalty in that. He wanted immediate or close heirs. In February 1216 as he marched south through Lincolnshire, rebel after rebel submitted. Eighteen of them were recorded on a single day, 28 February.56 Of these thirteen surrendered sons as hostages, one surrendered a daughter, and two surrendered nephews. The proportions both in these and other instances are such as to suggest that the king took daughters when there were no sons, and nephews when there were no direct heirs; but occasionally nephews were taken as well as sons. When greater severity seemed necessary, the king turned not to more distant relatives but to the sons and daughters and nephews of the errant baron's tenants. Gilbert fitz Reinfrey, lord of Kendal, provides a good example. A loyal and expert sheriff of the second generation of an administra54

Ibid., 338, 369-70; 372, 443-44; 384, 445-47; 425-26, 461-62. Rot. de Liberate ac de Misis et Praestitis, p. 103; J.H. Round, Geoffrey de Mandeville. (London, 1892), pp. 391-92; EYC, iii, p. 199. Roger de Lacy was the son of Robert's half-sister. 56 Rot. de Oblfitis etFinibus, pp. 579-81. 55

234

Colonial England, 1066-1215

tive family which had served both Richard and John with great devotion and skill in the northern counties, he made the mistake of joining the rebellion against King John and his son, William of Lancaster, and his steward, Henry Redmain, were captured with the rest of the garrison of Rochester castle in December 1215.57 Gilbert made his peace with the king shortly afterwards. He could not surrender his son because he was already a prisoner; indeed Gilbert was fining to secure his release. So he had to provide as hostages Benedict, son and heir of Henry Redmain; the eldest son of Roger of Kirby, who had married his daughter; the son and heir of William of Windsor, who had married his niece; the son or daughter, heir of Ralf de Aincurt; the son or daughter, heir of Roger of Burton; the daughter, heir of Adam of Yealand; the son or daughter, heir of Roger of Beetham; the son or daughter, heir of Walter of Strickland; the daughter of Richard of Copeland; and the son of Gilbert of Lancaster.58 A large group of the knights and gentry of north Lancashire, Craven and Cumbria, some of them tenants of Gilbert, some relatives by marriage, were involved in his disaster and in almost every case the terms of their subjection were quite specific; what the king sought in almost every case was the heir. The implications of his actions are plain: nothing would make rebellion less likely or the performance of agreements more certain than to have the heir in royal custody. Such a hostage could be used in varied circumstance, not only to guarantee uncertain loyalties but to secure the payment of debts or the surrender on demand of a castle.59 It was the ultimate deterrent in the relations between the king and his barons. And those who knew King John also knew that he might use it to the full. By 1215 he had already done so in the case of William de Briouze.60 It could be a matter of life and death. It is time to take stock, for the argument so far has now led to a broad distinction: on the one hand, the kin in the broadest sense, made up of all those relatives far and near who exercised and enjoyed patronage to their mutual benefit and shared in benefactions; on the other, the dynastic family or direct lineage through which the main holdings of the family descended. To be sure, the distinction does not always hold. Lack of direct heirs could lead to a distant succession. When Robert de Lacy died without direct heirs in 1193 the descent was traced to his deceased aunt, thence to that aunt's deceased daughter, and thence to that daughter's grandson. In modern parlance it led to the succession of a first cousin twice removed. Yet it took place without any apparent difficulty.61 So kinship in the wider sense must always have been reinforced by the possibility of distant succession and, where that was denied, by the preservation

57

Rot. Liu. Claus., i, 241b. Rot. de Oblatis etFinibus, pp. 570-71. 59 Rot. Chartarum, 30b; Rot. de Oblatis et Finibus, p. 275; J.C. Holt, Magna Carta (Cambridge, 1965), pp. 85-86. 60 Painter, The Reign of King John, pp. 242-50; J.C. Holt, The Northerners (Oxford, 1961), pp. 184-86. 61 Complete Peerage, vii, p. 677; Wightman, The Lay Family, pp. 85-86, 116. 58

Patronage and Politics

235

and nurturing of distant claims. But in the main the descent of the family property was likely to be the immediate concern of the lord, his children and their spouses, and his siblings and their spouses. It involved sons, daughters, sometimes uncles, more rarely cousins. It was within this group, one far narrower than the widespread kin which looked to the head of the family as the source of patronage, that really serious difficulties arose. Right at the heart of the family, in its control over the descent of the patrimony from which the family derived its name, its historical traditions and its heraldic devices, relations were potentially explosive. It is this, more than anything else, which undermines simple conclusions about the political role of the family. Here England diverged markedly from the Continent. Professor Duby has argued, tentatively but cogently, that aristocratic society in the twelfth century experienced a 'deep fissure that was to become its principal line of cleavage', namely 'the distinction between those who were elders and those who were not yet - and in most cases would never be - seniores, between the married men (conjugati) and the "bachelors", that is between elder and younger sons'.62 In short the patrimony was concentrated in the senior line; younger sons enjoyed no part of it and had to seek their fortunes elsewhere; their main hope of establishing a lineage of their own lay in marrying a heiress. It is not altogether clear, even for the Continent, how far such a 'major fault-line' reflected a general stage of development or a number of local variants. The argument relies heavily, though not solely, on one source, Lambert of Ardres' Historia comitum Ghisnensium;63 for Normandy it would be difficult to reconcile with tenure byparage, which made provision for younger sons;64 and it may not allow enough to the sound common sense, even more obvious then than now, of securing the lineage by providing a second team in case the senior line failed. Be that as it may, it is not a model to be applied to England. In England younger sons acquired estates, married, established junior lines, with relative ease. To take a random sample of a large family, of the five sons of Aubrey I de Vere, d. c. 1112, the eldest died before succeeding, the second succeeded to the patrimony, the third, Roger, established himself as a tenant of the abbey of Bury St Edmunds, whilst the fourth and fifth are known only through their attestations. Of the five sons of Aubrey II de Vere, Chamberlain to Henry I, d. 1141, the eldest succeeded to the patrimony; the second became lord of Clun iure uxoris\ the third succeeded to his father's Northamptonshire lands, and from him in turn there 62 G. Duby, Medieval Marriage (Baltimore, 1978), pp. 10-13, which is an emphatic development of his 'Les "jeunes" dans la societe aristocratique dans la France du Nord-Ouest au Xlle siecle', AnnalesESC, 19 (1964), pp. 835-46, trans, in Duby, The Chivalrous Society, pp. 112-22; reinforced by his 'Lignage, noblesse et chevalerie au Xlle siecle dans la region Maconnaise: une revision', AnnalesESC, 27 (1972), pp. 803-23, trans, in The Chivalrous Society, pp. 59-80. 63 Lamberti Ardensis historia comitum Ghisnensium, ed.J. Heller, MGH scriptores, 24 (1879), pp. 550-642. See Georges Duby 'Remarques sur la litterature genealogique en France aux Xle et Xlle siecles', Academiedes inscriptions et belles-lettres (1967), pp. 335-45, trans, in The Chivalrous Society, pp. 149-57, and Medieval Marriage, pp. 83-110. 64 See above, pp. 119, 148, 166.

236

Colonial England, 1066-1215

descended two subsidiary branches of the family; the fourth became a canon of St Paul's; and the fifth probably ended as prior of the English Hospitallers.65 That pattern, which left room in two generations in the twelfth century for the establishment of one or more cadet lines, seems typical. It may be that cadet lines in England enjoyed a relatively lavish provision because of the proliferation of acquisitions; following 1066 the whole land was an acquisition.66 At all events the distinctive circumstances in England were apparent to contemporaries. On the Continent the juvenes went tourneying to establish their reputation and fulfil the 'bachelor's' ideal. England by contrast, as William de Tancarville (the patron of St-Georges de Boscherville) pointed out to William Marshal, was not a land for tournaments: it was fit only for vavassors and stay-at-homes.67 From the stay-at-homes, none the less, were to descend the knights of the shire. This is not to imply that England lacked family strife. On the contrary, it is noteworthy that the common law, as it was taking shape in the late twelfth century, approached the family warily. Rules of seniority were carefully drawn both for heirs and heiresses, cadet lines were protected by the rule that the lord cannot also be heir, and family bickering was made less easy by the denial of the assize of mort d'ancestor to litigants of the same blood, so that contentious brothers or sisters, and usually uncles, nephews and cousins as well, were directed towards the writ or right. It was as if provision for the immediate family was concluded in an inner sanctum which only became an open stage, displayed in royal or feudal courts, if some quite exceptional settlement or argument took place. But if the law permitted a wide variety of arrangements made at the family's behest it also recognised inherent dangers. Consider what the author of the Tres ancien coutumier of Normandy, writing c. 1200, had to say about wardship:68 A fatherless heir must be in ward to someone. Who shall be his guardian? His mother? No. Why not? She will take another husband and have sons by him, and they, greedy for the heritage, will slay their first-born brother, or the step-father will slay his stepson. Who then shall be the guardian? The child's blood kinsmen? No. Why not? Lest, thirsting for his heritage, they destroy him. For the prevention of such faithless cruelty, it is established that the boy be in ward to one who was bound to his father by the tie of homage. And who is such a one? The lord of the land, who can never inherit that land in demesne. ..

65

Complete Peerage, x, appendix J, pp. 112-15. 1 have chosen an example from the early twelfth century. Professor Duby suggests that families began to make provisions for junior branches in the late twelfth century (Medieval Marriage, pp. 102-5). It may well be that the freer disposition which acquisitions allowed in England imposes a different chronology. 67 Histoire de Guillaume le Marechal, ed. P. Meyer, Societe de 1'histoire de France (3 vols, 18911901), i, lines 1536-43.1 owe 'stay-at-homes' to P.M. Powicke, The Loss of Normandy (Manchester, 1961), p. 303n. 68 Coutumiers de Normandie, ed. E.-J. Tardif, Societe de 1'histoire de Normandie (2 vols, 18811896), i, pp. 10-11. 66

Patronage and Politics

237

Maitland, whose translation I have used, may have been right to regard this as a 'quaint apology' for feudal wardship, for the writer continues with a rhapsodic passage on the advantages of raising minors in their lord's household.69 But the point expressed in the Ancien coutumierwas taken seriously in English law. If the minor was a tenant in socage then his wardship went not to his lord but to his family: to his mother's relatives if his inheritance came from the father and to his father's relatives if it came from the mother. The potential threat to the young heir from within the family was clearly recognised: 'no one shall ever remain in the wardship of a person who may be suspected of wishing to claim a right in that inheritance'.70 Relatives were all very well, but there were matters in which they could not be trusted.71 It was not simply that now and then a relative proved himself a bad lot, although no doubt that was true. It was a question of structure: of family tie, title and tenure, in particular the weight which practices, rules or laws governing succession placed on this or that relationship. In this the family was not the sole determinant. The feudal lord, above all the king, was also a powerful voice, insistent on due services and therefore on the indivisibility of the fief and tending always (at least if he was a 'good' king) towards consistency and regularity. The structure would not have stood at all if its parts had not roughly matched. Hence only very unusual circumstances were likely to set father and heir at odds. It might happen, and indeed was likely to happen, in the ruling house where the heir established his own court and could make a claim to a portion of its dispersed authority. Every royal heir who reached puberty between 1066 and 1215 made a nuisance of himself, with the solitary exception of William, son of Henry I, who may have been saved from such disloyalty by premature death, for the young men who died with him in the White Ship, many themselves heirs to great Norman lordships, look closely similar to the expectant groups which had collected around Robert Curthose earlier or were to assemble around the young King Henry later.72 But in general the relationship between father and heir worked and had to work. It was so axiomatic that the quarrels between Henry II and his sons could be viewed as divine retribution for the sins of him and his house.73 It worked by and large in 1215, for the sons of the greatest East Anglican lords were numbered with their fathers among the Twenty-Five barons of Magna Carta,74 thereby committing their succession to the cause just as King John tried to bend men in the opposite direction when he exacted hostages. 69

F. Pollock and F.W. Maitland, History of English Law (Cambridge, 1898), i, pp. 326-27. Bracton, Delegibus, fo. 87b; ed. S.E. Thome (4 vols, Cambridge MA, 1968-77), ii, p. 254. Pollock and Maitland, History of English Law, i, p. 321. 71 For comment on the problem at a later date see J.S. Roskell, The Office and Dignity of Protector of England', EHR, 68 (1953), pp. 193-233. 72 Orderic Vitalis, ed. Chibnall, vi. 304. 73 Gerald of Wales, Opera, ed. J.S. Brewer, J.F. Dimock, G.F. Warner, Rolls Series (8 vols, London 1861-91), viii, pp. 298-301; William of Newburgh in Chronicles of the Reigns of Stephen, Henry II and Richard I, ed. R. Howlett, Rolls Series (4 vols, London, 1884-89), i, p. 281. 74 Gilbert of Clare and Hugh Bigod. 70

238

Colonial England, 1066-1215

There was only one obvious exception, William Marshal junior. He may have been the example which led the Barnwell chronicler to write of families divided, young against old. It may be that he was one of this writer's 'lovers of new things' although there is not much in his later career to suggest it. The known record of his father, the old Marshal, would not preclude some collusive arrangement, for the old man had been a fine hedger of bets. There were many other relationships which were equal or almost equal strength to those between father and heir. From kings downwards marriages were used to cement alliances. The relationship between father-in-law and son-in-law or between brothers-in-law was strengthened by the marriage portion, for the recipient became the tenant of the donor.75 One explanation of the enmity between King John and Robert fitz Walter was that he had raised armed men to defend his son-in-law, Geoffrey de Mandeville.76 Such comment was commonplace in feudal society. The earl of Gloucester is much to be feared', said the countess of Leicester to her husband in 1173, 'but he is married to your sister, and not for all the wealth of France would he start any extravagant action that would cause you any trouble.'77 Again brotherhood was almost universally taken as implying unusually close relationships: monastic confraternity, bloodbrotherhood, brotherhood-in-arms were all expressions of this. And avuncular interest often seemed benign: was not Charlemagne Roland's uncle and did not uncles, both maternal and paternal in the senior line, buy custody of their nephews and nieces from kings who rarely hesitated to sell? Yet it was at this point where collateral interests were expressed, that difficulties arose. The law protected the cadet lines by the provision that a man could not be •both lord and heir. It protected the senior line by the principle of feudal wardship. But there were parts of the structure here which did not quite fit, or were as yet scarcely formed. One lay where the claims of cadets, especially of the half-blood, mingled with the rule that, though the patrimony descended to the first born, a man could dispose of his acquisition as he wished. It was easy to shift from this concession to the argument that the cadets were entitled to the acquisition, and where the acquisitions were considerable this could easily become contentious.78 And if the brother in the half-blood was an occasional threat, the paternal uncle was an ever-present danger. The influence of Charlemagne in the Song of Roland is benign, but he was a maternal uncle. The Emperor Alis, on the other hand, the paternal uncle of the hero in Chretien of Troyes' romance 75

The relationship was all the closer in that, according to Glanvill, homage was not performed for a maritagium (and services not rendered) until the third generation. The effect was that the grant was made by infeudation, but amounted to an allocation of part of the donor's inheritance, which would revert on failure of heirs of the marriage, rather than an irrecoverable alienation. See Glanvill, De legibus, ed. G.D.G. Hall, p. 106; SJ. Bailey, 'Warranties for Land', Cambridge Law Journal 9 (1947), pp. 93-95; S.F.C.Milson, The Legal Frameivoik of English Feudalism (Cambridge, 1976), pp. 142-45. 76 Histoire des dues de Normandie et des rois d'Angleterre, ed. F. Michel, Societe de 1'histoire de France (1840), pp. 116-19. 77 Jordan Fantosme's Chronicle, ed. R.C.Johnston (Oxford, 1981), p. 73. 78 See the case of the Bigod earls of Norfolk, discussed above, pp. 151-52.

Patronage and Politics

239

Cliges, is the prototype of the wicked uncle who makes his villainous way through romance from the thirteenth to the twentieth century.79 The contrast is easy enough to explain. The maternal uncle had no personal claim upon the patrimony, and it is this more than any residual trace of the matriarchal family which thrust him into the role almost of a second father. As a custodian of an heir he was safe. The cadet paternal uncle, in contrast, was close kin to the heir, perhaps the next in line of succession, his instinct to step in all the stronger if the heir was young or female, and strongest of all if the heir was merely representative, child of the uncle's deceased elder brother who had never succeeded. In 1199 John stepped in. By 1203 his nephew Arthur had been murdered. The families of Mandeville and Say were already divided against each other over the same issue.80 The house of Percy was torn between uncle and nephew who fought on different sides in 1215; and Percy was followed in 1214 by Briouze and in 1219 by Quency.81 Wherever the casus regis was replicated, uncle and nephew in knightly and lesser families battled it out in the courts.82 On these or similar issues a family might destroy itself. The Buissels, barons of Penwortham, divided between senior and cadet lines, took their case to the court ofjohn, count of Mortain, in 1194. The count's decision proved inconclusive. Between 1200 and 1202 both branches of the family committed themselves to such heavy proffers for royal arbitration that their debts and their barony were taken over by Roger de Lacy constable of Chester, and as barons of the honour of Lancaster the Buissels disappear from history.83 Within a decade of the collapse the senior line became extinct. The family had held Penwortham by the service of three knights. The junior line remained as tenants of the Lacys in Leyland and elsewhere rendering the service of a tenth of a knight.84 One may ask what they might have said about lineage and the fortunes of descent, for in challenging a title they lost the succession to it. So in the end it is necessary to strike a balance. But it will not be found by choosing a distant ancestor and tracing all his descendants, as Painter did with the house of Clare. Plainly there is no satisfactory way of measuring how important 79 Les romans de Chretien de Troyes, ii, Cliges, ed. Alexandre Micha, Classiques franf ais du moycn age (1957), where the tale is summarised, pp. iii-xi. 80 Complete Peei age, v, pp. 120-22. 81 For Percy see ibid., x, pp. 449-55; and Holt, The Northerners, p. 21. For Briouze see Complete Peerage, i, p. 22; PatmtRolls, 7276-25, p. 165; Curia Regis Rolls, viii,pp. 10-12, 27, 35;ix, pp. 9,306-7; x, pp. 134-35, 195; xi, nos 390, 1969. For Quency see Complete Pemige, xii, pt 2, pp. 749-51, which fails to note that Robert de Quency (d. 1217) had a legitimate daughter, Margaret (CompletePeerage,™, p. 679). 82 Milsom, Legal Framework, 147-49.1 hope to deal more fully with the casus regis m a forthcoming work on the succession of 1199. Although the evidence is not easy to assess, it suggests that actions of this type multiplied after John came to the throne. See below, pp. 307-26. 83 Holt, The Northerners, p. 240; VCH, Lancashire, i, pp. 335-36. The dispute lay in the first instance between uncle and nephew in the senior line in the court ofjohn, count of Mortain, and continued as litigation between first cousins in the king's court from 1200. It was not, however, an example of the casus regis, because in this case the nephew's father had in fact succeeded to the inheritance. 84 VCH, Lancashire, vi, p. lln.

240

Colonial England, 1066-1215

it was to Robert fitz Walter that he was second cousin once removed to Richard of Clare. They acted together at times, certainly, but they were neighbours, facing similar problems and common action was scarcely surprising. To seek something more concrete is to narrow the focus from the larger kin to the smaller family relationship which was close enough to involve the transfer of land and the establishment of dependent tenures. Round showed a sure touch in concentrating his attention here not on Richard of Clare but on Robert fitz Walter, the baronial leader of 1215, who earned the tribute from Matthew Paris that he 'could match any earl in England: valiant in arms, spirited and illustrious, endowed with many possessions, generous, encompassed by a multitude of powerful blood relatives and strengthened by numerous relatives by marriage'.85 It is easy to supply bits of the jigsaw which justify that comment. Robert married his daughters to Geoffrey and William de Mandeville, each in succession earl of Essex. Robert fitz Walter and William de Albini of Belvoir were cousins in the female line: de Albini's mother was fitz Walter's aunt and as a result of the alliance William held half a fee of the fitz Walter honour of Dunmow.86 William de Albini was responsible for collecting fitz Walter's ransom after the surrender of Vaudreuil in 1203;87 fitz Walter left him in charge of Rochester Castle in 1215.88 Fitz Walter and Saer de Quenci were first cousins in the half-blood in the male line.89 De Quenci, like William of Albini, was fitz Walter's tenant.90 In 1203 they conducted the defence of Vaudreuil together.91 They also helped each other in more private matters. Fitz Walter attested in 1219 de Quenci's exchange of the Scottish marriage-portion of his newlywidowed daughter, Loretta, for land in Shepshed.92 And this family link was there well before Saer had made himself a party to rebellion. Fitz Walter's treason in 1212 brought his sister Alice Peche under suspicion. She was harassed and had to surrender her daughter and other hostages. It was Saer de Quenci who held them on behalf of the king.93 Finally Robert and Saer went on the crusade to Damietta, where Saer died.94 Their long association in military endeavours suggests that they were brothers-in-arms, and this is strongly confirmed by a device on their seals: each bore the other family's shield in the field of his own equestrian figure.95 85

Gesta abbatum Monasterii S. Albani, ed. H.T. Riley, Rolls Series (London, 1867-69), i, p. 220. Rotuli de dominabus, Pipe Roll Society, 35 (1913), p. 1; Book of Fees, p. 577. 87 Rot. LitL Pat., 47b. 88 Painter, KingJohn, pp. 361-64. See also fitz Walter's letter to William in Chronica majora., ii. p. 614. 89 Compute Peerage, xii, pt 2, p. 746. 90 Red Book of the Exchequer, p. 349. 91 Chronica majora, ii, p. 482; Coggeshall, pp. 143-44. 92 Oxford, Magdalen College, Brackley Deeds, no. 12. 93 Rot. Litt. Pat., lOlb. 94 Chronica majora, iii 41; Annales monastics, ii, p. 292. 95 Walter de Gray Birch, Catalogue of Seals in the Department ofManusmpts in the British Museum, 6 vols (1887-1900), ii, no. 6016; A.B. Tonnochy, Catalogue of Seal-Dies in theBntish Museum (London, 1952), no. 332. For Saer's and other de Quenci seals see Complete Peerage, xii, pt 2, p. 750 n.(h) and ibid., appendix I, pp. 29-31; also Book of Seals, no. 418. For Robert fitz Walter's see the splendid silver die 86

Patronage and Politics

241

Three general considerations now remain. First, how in these matters is England to be related to the continent? The proliferation of acquisitions in England was one distinctive feature, and certainly a very important one. In addition to that, there were many local variations in the law of succession and in practices of wardship throughout western Europe. But, quite apart from substantial differences of this kind, the English evidence is much more systematic than the continental. It reveals relationships, both linkages and stresses, the common occurrences of aristocratic and knightly family life, which were subliminal to the literary expression of the ethic of lineage and knighthood, whether of the seniores or the bachelors, which has figured so large on the Continent. It also demonstrates that the family's grasp of its lineage was not always as firm as our reconstructions require. One litigant in 1201 lost his case in an action of right because he could not name his grandfather.96 Another in 1203 was required to defend her dower because it should have been warranted by her eldest son who had been taken into foreign parts by his uncle 'a certain abbot', and had not been heard of for the last seven years so that she did not know whether he was alive or dead.97 From all these points of view England is a starting-point, a fundamental example, not an insular freak. How typical an example much further work will be needed to reveal. It is probable that this will reduce the emphasis which some have placed on lineage, especially when presented as a political force. Continental studies will have to make room for more family feuding, internal as well as external. Secondly, how did the pattern change between the Conquest and 1215? It is tempting to assume that very little changed. In portraying family matters the early pipe rolls of the mid twelfth century are little if any different from the fine ^olls of John's reign. Twelfth-century chroniclers, like their successors, commented on family connections. Indeed Robert of Torigny provided one of the most extended of all pedigrees demonstrating the good fortune of the relatives and descendants of the Duchess Gunnor, and that carried the story back to pre-Conquest Normandy at the beginning of the eleventh century: a medieval prototype of Painter's house of Clare.98 But some differences are perhaps detectable. One clue is provided by the fact that even in 1200 family ties seem more obvious in a newly-risen family or one transferred to novel circumstances. This was true of Ranulf Glanville on his transfer to Yorkshire as sheriff;99 and even continued

in the British Museum, now generally attributed to him rather than to his grandson. On the whole question see George Henderson, 'Romance and Politics on Some Medieval English Seals', Art History, 1 (1978), pp. 26-42, where it is noted that the seal of Margaret, Saer's widow, bore both the Quenci and fitz Walter arms. Henderson does not emphasise brotherhood in arms (p. 37). For some discussion of such adoption of arms see M.H. Keen, 'Brotherhood in Arms', History, 47 (1962), pp. 1-17, especially 13-14. 96 Curia Regis Rolls, ii, p. 12. 97 Ibid., ii, p. 219. 98 See above, p. 170. 99 R. Mortimer, The Family of Rannulf de Glanville', pp. 11-12.

242

Colonial England, 1066-1215

more obviously true of the Potevins whom King John imported into England, many of whom formed an interrelated group which was contemptuously dismissed as a sequela in Magna Carta.100 So the family bond seems to have been at its strongest when it was less trammelled by other ties of local association or lordship. That implies that it would play a greater role in the earlier stages of the Norman settlement and there is something to suggest that this was so. Orderic Vitalis was peculiarly lavish not only in noting family ties but also in the varied vocabulary he used to describe them.101 That may have been an individual quirk. The coronation charter of Henry I is evidence of a quite different order. Whether it stated contemporary practice must be uncertain. It certainly stated an older tradition and a looser pattern of family organisation, for it laid down that wards were to be in the custody of their mother or near relative.102 At all events it stands as a reminder that some of the family troubles of the late twelfth century were a consequence of the strengthening of lineal succession which occurred as the twelfth century advanced. Thirdly, and finally, it is important to preserve a sense of proportion. Twelfthcentury society was not defined solely by family relationships. There were other ties, of lordship and neighbourhood, which themselves helped to determine how families behaved. To assume that the family was all important would be rather like pretending that the Song of Roland-was all about the enmity of stepfather and step-son. That certainly was the relationship of Ganelon and Roland, and there is some interest in the fact that that was chosen, but it need scarcely be said that the Song is about much else. I concluded a previous essay with some comment on a letter which Nigel de Albini wrote from his bed of sickness about 1109.103 Some indication of the importance of the family ties in the early twelfth century is that, in amending his wrongs and purging his soul, he addressed himself to his brother, William. In 1158 Warin fitz Gerold, chamberlain of Henry II, wrote a similar letter. This was addressed to the king and it brings together all the associated influences - benefaction, lordship and family tie, which I have intermingled in this argument.104 To Henry, king of the English, his dearest lord, Warin fitz Gerold his faithful man, greeting and fealty. I address you, my dearest lord, at death's door, as I am about to enter upon the way of all flesh, for I shall not see you again nor be seen by you. Hear, I pray, my final words which come not from my body, which I feel most frail, but from the anxious concerns of my soul. I have given and delivered up myself to Holy Edmund whether I live or die. In the fealty which I owed and owe to you, I counsel you that you love and honour the holy martyr and his church with all reverence, since I perceive and feel that this will be useful to you and yours both in those things which pertain 100 101

Magna Carta, cap. 50.

In addition to parentes, Orderic Vitalis uses cogn,ati, consanguinei, consobrini and both stemma and parentela, ed. Chibnall, i, index verborum. 102

Coronation Charter of Henry I, cap. 4. See above, pp. 177-78. 104 Cambridge University Library, MS Ff.2.33, fo. 51; BL, MS Harley, 638, fo. 209. 103

Patronage and Politics

243

to God and those which pertain to man. Moreover, dearest lord, I beg and beseech you that if ever I pleased you in my life, you should declare to me your grace and mercy in this my last prayer, that for the mercy of God and for the salvation and honour of both of us, you should so maintain and promote the aforesaid church and its abbot and convent who have housed my miserable body and made special intervention for my wretched soul that in this matter it should be manifest if I ever pleased you. I commit and commend to your nobility and clemency my brothers and relatives and other friends. And it is particularly necessary to me that you should know that I have given to Holy Edmund my lord, my advocate, and my host, six librates of land in Sawbridgeworth. Maintain them, dearest lord, confirm them, for the love of God and his aforesaid church, with such firmness that no improbity will be strong enough to violate or detract from my soul. Dearest lord, my soul salutes you, and in this my last salutation, wishes you perpetual prosperity. That must have been drafted by a Bury scribe. Hence Warm's dying wishes are seen through a glass distorted by rhetoric and misted by the interests of a great abbey. His concern for his family is secondary to the concern for his soul and his endowment. He commended his soul to St Edmund and his endowment and his family to his lord, the king who had made him what he was. Sawbridgeworth was one of his acquisitions. In conveying it to Bury he instructed his heirs present and future to maintain and preserve his grant against all men free of all secular service since it was made for the redemption of his soul.105 But it was the lord who had enfeoffed him there, Geoffrey de Mandeville, who was the first to attest his grant. That strikes about the right balance.

105

Cambridge University Library, MS Ff.2.33, fo. 51.

This page intentionally left blank

13 Feudal Society and the Family in Early Medieval England, IV: The Heiress and the Alien

The topic is doubly important. First, the heiress was one of the fluid elements in the social structure. By marriage, families forged links with each other, landless knights and ambitious officials made their way into the noble hierarchy, and established lineages renewed their military reputation or improved their access to government circles.1 If the bride was an heiress all this was underwritten by a territorial endowment far larger than the usual marriage-portion. She also brought to her husband title to office, all the legal and quasi-legal claims which came to be concentrated in her from her lineage, and sometimes even the family name of her forbears to be assumed, if not by him, then by the offspring of their marriage. So the heiress was special; she magnifies for us all the problems and consequences of marriage. The problems were obvious. In what way was such a system to be controlled and by whom? What say in the matter was to be enjoyed by the parties to a marriage, by their families, by their lords? Who might be regarded as an heir? Which among a man's daughters might be regarded as an heiress? Some of the answers to these questions were reinforced by stern doctrine, others were slowly pieced together from practical experience and immediate convenience. But the answer to the last question - who might be regarded as an heiress? - was different from all the rest: it was altered. Up to the second quarter of the twelfth century the inheritance descended through one daughter or sister, usually the eldest: after that point it was divided among them all. This change was the result of a specific ruling, as near an act of legislation as could be achieved in England at that time. It is the one point where anything approaching a revision of substantive law can be investigated, the one point where motives and attitudes of all concerned can be probed both in general and in depth. This is the second 1

See Georges Duby, Medieval Marriage (Baltimore, 1978); M. Sheehan, 'Choice of Marriage Partners in the Middle Ages: Development and Mode of Application of a Theory of Marriage', Studies in Medieval and Renaissance History, new series, 1 (1978), pp. 3-33; Eleanor Searle, 'Women and the Legitimisation of Succession at the Norman Conquest', California Institute of Technology, Social Science Working Paper, 328 (1980), also in Proceedings of the Battle Conference an Anglo-Norman Studies, 3 (1980), pp. 159-70; C.N.L. Brooke, 'Marriage and Society in the Central Middle Ages', Marriage and Society, ed. R.B. Outhwaite (London, 1981), pp. 17-34; Georges Duby, Le chevalier, la femme et k pretre (Paris, 1981; Eng. trans. 1983); Ragena C. DeAragon, 'In Pursuit of Aristocratic Women: A Key to Success in Norman England', Albion, 14 (1982), pp. 258-66; Jack Goody, The Development of the Family and Marriage in Europe (Cambridge, 1983).

246

Colonial England, 1066-1215

good reason for studying the topic. There was a sudden, deliberate change of policy. Women were subordinate. They were placed under the dominion of husband, head of family or lord. But the lawfully wedded wife played a crucial role: she was the sole source of legitimate heirs. In a society which, within a few generations at most, had been restructured around lineal descent, the bride was soon hedged about by a protective law which excluded illegitimate children from the succession, defined the acceptable degrees of affinity, and protected her in marriage against desertion and divorce.2 All these rules were brought into play by the twelfth century. The woman's role might be thrown in doubt by the higher flights of the more insulated theologians who presented her as an unfortunate necessity, the originator of sin, who approached nearer to the ideal if in her later years she turned from her earthly husband to the heavenly embrace of Christ, repenting her days away that she had ever borne men's children. But not all contemporary stereotypes were like that. Marriage was seen as a security for peace and legitimate succession: hence Rollo, first duke of Normandy, was made to marry Gisele, daughter of Charles the Simple; Queen Emma was said to have brought half the realm to Cnut; and Duke William was made to convey his daughter, Adeliza, to Harold with the promise of half the kingdom of England.3 Marriage and the continuation of the lineage were one and the same. A barren marriage was a tragedy, a motive though not a ground for divorce, to be explained away as the consequence of accident or more usually of sin.4 Hence the Emperor Henry V died childless peccatis exigentibus, and women who abandoned the veil remained childless all their lives and in the words of Orderic, 2 Brooke, 'Marriage and Society', pp. 23-27. See also the cases of Agnes of Ponthieu, who fled from the cruelty of her husband, Robert of Belleme (Orderic Vitalis, ed. Chibnall, iv, p. 300); Hildegarde, countess of Poitou, who appealed at the council of Rheims, 1119, against desertion by her husband, Orderic Vitalis, ed. Chibnall, vi, pp. 258-60; F. Villard, 'Guillaume IXd'Aquitaine et le concile de Reims de 1119', Cahiersde civilisation medevale, 16 (1973), pp. 295-302; and Agnes, countess of Oxford, who fought long and ultimately successfully against repudiation by her husband, Aubrey de Vere (CompletePeerage, x, pp. 206-7; app. J, pp. 116-17; Brooke, 'Marriage and Society', pp. 31-32). 3 For the fictitious marriage of Rollo see Dudo of St-Quentin, De moribus et actis primorum Normanniae ducum, ed.J. Lair (Caen, 1865), p. 169, repeated in Orderic Vitalis, ed. Chibnall, iii, p. 78; for Emma see Orderic Vitalis, ed. Chibnall, i, p. 157; for Adeliza see William of Jumieges, interp. Orderic Vitalis, Gesta Nrnmannorum ducum, ed.J. Marx (Paris, 1914), p. 191, but cf. Orderic Vitalis, ed. Chibnall, ii, p. 136. 4 See for example the sixteenth-century register of Tewkesbury Abbey which explains the descent of Robert fitz Hamo as follows: Robert, at the summit of power, enjoying great riches and the friendship of King Henry I, had everything a man could desire but issue of his marriage, a lack which would sadden the heart of any man. So he and his wife sought a remedy through alms, pilgrimage and prayer until at last they came to pray to St Benignus at Glastonbury. They promised a 100s. of land if the saint interceded to give them an heir and in a few days Robert's wife was heavy with child. However, they cheated the saint and, as a result, as long as Robert lived his wife bore female children only. Once he was dead she married at a lower status and had children of both sexes. (BL, MS Add. 36985, fo. 3v). For another example of analogous genealogical fiction see the case of Walter Espec, discussed below p. 257.

The Heiress and the Alien

247

'for a short period of worldly happiness incurred the wrath of the heavenly bridegroom'.5 Such fictions and exaggerations heightened the status and reinforced the role of the lawfully wedded wife. Inheritance in the female line was determined by these circumstances. A woman inherited not because of any title, not because of the survival of more ancient legal arrangements which might allow her a determinate share of her father's lands, but because, in the absence of male heirs in the same generation, she was the only means of continuing the lineage, the only legitimate route whereby her father's blood could be transmitted. Her children were his grandchildren just as her brother's might have been. This determined the woman's position as heir. If there were legitimate male heirs to her father then she could not expect to succeed. If there were no male heirs then the inheritance was 'hers' in the sense that it was no one else's, that the claim which she embodied was stronger than anyone else's. But it was not hers in the sense that she could succeed as a spinster. She brought her lands to her husband and ultimately to her children. Her husband had charge of them and he might be followed by her son even in her lifetime.6 It was only as a widow that she might hope to gain sole control. That was a possibility which led the cautious beneficiary to associate her with her husband in acts affecting her lands, just as her title brought her into the courts alongside her husband if her rights were subject to litigation. But settled widowhood was a status only achieved with difficulty, usually after several marriages, and it was likely to be maintained only at considerable cost. Norman and Anglo-Norman society in the eleventh and early twelfth centuries reveals these rules in operation. No spinster can be found in enjoyment of her inheritance. It was only very occasionally, after some disgrace of the male line, that a female was given precedence over a male heir.7 Widows in possession of their inherited lands were rare birds. The social cross-cut provided by the Domesday Survey is revealing. The tenants-in-chief of 1086 included one great 5 Orderic Vitalis, ed. Chibnall, v, p. 200; ii, pp. 102-4. Orderic's comment concerned Judith and Emma, sisters of Robert of Grandmesnil, abbot of St-Evroult. Judith married Roger Guiscard and had at least two daughters. See Orderic Vitalis, ed. Chibnall, ii, pp. 103-4n. 6 S.F.C. Milsom, 'Inheritance by Women in the Twelfth and Early Thirteenth Centuries', in On the Laws and Customs of England: Essays in Honor of Samuel E. Thome, eds Morris S. Arnold, Thomas A. Green, Sally A. Scully and Stephen D. White (Chapel Hill, NC, 1981), pp. 60-89. 7 No statement about rules of inheritance in this period can be hard and fast. One obvious possible exception in this case is the descent of Belleme, where Mabel and her husband Roger of Montgomery succeeded to the family lands after much confusion over the rights of her father, William Talvas, and to the exclusion of his brother, Oliver, a monk of Bee, who may have been illegitimate. See Marjorie Chibnall's discussion in Orderic Vitalis, ed. Chibnall, ii, 362-65, and ibid., i, p. 213, for a genealogical table; also G.H. White, The First House of Belleme.' Transactions of the Royal Historical Society, fourth series, 22 (1940), pp. 67-99. There are many examples of a female effectively legitimising the succession of her husband after some break in the descent through her male relations. See the case of Mowbray (Charters of the Honour of Mowbray, 1107-1191, ed. D.E. Greenway, British Academy Records of Social and Economic History), and other examples noted above, pp. 194-95.

248

Colonial England, 1066-1215

widow, the Countess Judith.8 Queen Matilda, and less than ten wives of the leading tenants of the day, held land which had presumably been given to them in dower.9 The daughters of two tenants-in-chief appear, each with a single estate which was probably her marriage-portion.10 There was a scattering of English ladies still with possessions of their own, the residue of an older society in which women had held property in their own right. This last group was minute compared with what it has been in 1066. These women saw their time out enjoying a right which the Conquest had destroyed. None of this means that women were unimportant. On the contrary, descent in the female line was common; indeed, it was the obvious course whenever the male line failed. The extent and importance of the inheritance did not affect the issue. The counties of Anjou, Maine and Evreux and the lordship of Breteuil all descended through the female line in various ways between 1060 and 1110.11 Even an unconsummated betrothal could count. Giroie, the founder of the fortunes of the patrons of St-Evroul, agreed to marry the only daughter of the knight, Heugon, and received Montreuil and Echauffour as the marriage portion. Then his betrothed died. Giroie had to seek the confirmation of Duke Richard, but he was given the land as a hereditary tenure. That story comes from the years before 1027.12 Many descents were much less direct than this. The descent of Maine involved four arrangements or claims, all in the female line, spread over a period of twenty-seven years, from 1063 to 1090. It began with an agreement concluded by Herbert II, count of Maine, with Duke William of Normandy whereby Herbert's sister Margaret was promised to William's son Robert, with the proviso that she would inherit Maine if Herbert had no children. It ended in 1090 with the succession of first one then another of Herbert's cousins, each of them the son of his aunts, sisters of his predecessor, Count Hugh IV. The first of these cousins was brought back from Italy where his father was marquis of Liguria with the argument that 'no man was closer heir than he'. The second pressed his claim on the ground that 'I no less than you am sprung from the stock of Count Herbert' (recte Count Hugh). So at least Orderic Vitalis tells us.13 His genealogy was not quite right, but it is none the less significant that he presented the succession in these terms. Such was the system which the Norman brought to England. It was transplanted 8 For Judith see Henry Ellis, A General Introduction to Domesday Book (London, 1833), i, pp. 440 41; Complete Peerage, vi, pp. 639-40. Countess Godeva, widow of Earl Leofric, was dead by 1086, Ellis, Introduction, to Domesday Book, i, pp. 370, 426. 9 Ibid., i, pp. 501-2. 10 The daughter of Ralph Taillebois (DB, i, 142b) and Roger de Rames (DB, ii, 422b). 11 For Anjou see L. Halphen. Le comte d'Anjou au Xle siecle (Paris, 1906), pp. 133ff. For Maine see Orderic Vitalis, ed. Chibnall, ii, pp. 116-18. For Evreux see Complete Peerage, vii, pp. 709, 711. For Breteuil see ibid., vii, pp. 529-30; ix, p. 574 and n. 12 Orderic Vitalis, ed. Chibnall, ii, p. 22. 13 Orderic Vitalis, ed. Chibnall, ii, pp. 116-18; iv, pp. 192-98. For corrections to Orderic's genealogy see R. Latouche, Histoire du comte du Maim pendant leXeetk Xle siecles (Paris, 1910), pp. 11315.

The Heiress and the Alien

249

unchanged and undisturbed. By 1130 more than twenty baronies had descended in the female line.14 By 1150 the number had risen to thirty.15 Some underwent more than one such succession.16 On both sides of the Channel and both before and after 1066 female succession was a source of much dispute. Some of the conflicts arose from the competing claims of sisters. Some were parentelic, between the husbands or descendants of niece and aunt. Some were quarrels in the half-blood, between the descendants of an heiress's successive marriages. The more distant successions were always more intricate and more difficult to regulate precisely.17 But all were subject to the primacy of the male line. Few maidens at their betrothal were heirs apparent. Some were heirs presumptive. All were heirs potential. In any marriage, therefore, quite apart from the marriage-portion, there were chances to be weighed. In any family a maiden unbetrothed, unless devoted to the service of Christ, was a wasted resource. Worse still she was vulnerable. So she was betrothed at any early age. In 1103 Robert, count of Meulan, pledged his daughter to Amaury, nephew of William, count of Evreux, when she was aged one.18 The same circumstances determined the position of the widow. She came complete with dowry. She might already be an heiress who had brought her inheritance to her previous husband. If not, like the maid, she was an heir potential to a succession more or less likely. The system was one which made it very difficult for a woman to enjoy estate in her own right, but used her to the full as a means of perpetuating the bond between tenure and lineage. Such arrangements were controlled in two distinct ways. First, there were conventions about who might betroth or give a bride in marriage. The corona14 Aveley, Belvoir, Blagdon, Bourne, Brattelby, Castle Combe, Chester, Folkestone, Gloucester, Hockering, Hooton Paynel, Keevil, Malton, Much Marcle, Nether Stowey, Papcastle, Salwarpe, Stogursey, Tamworth, Wallingford and Witham. 15 Bolingbroke, Bourn, Burgh by Sands, Dudley, Hanslope, Kempsford, Meschin, Okehampton, Stainton le Vale. 16 Belvoir, Folkestone and Wallingford. 17 For disputes involving distant relationships in the female line see the quarrel between Ascelin Goel and William of Breteuil for the castle of Ivry, the one possibly the great great grandson and the other certainly the great grandson of Aubree wife of Ralph, count of Ivry. Aubree reputedly built the castle. If the genealogy is correct, each party claimed through a female ancestor, grand-daughter and daughter of Aubree, i.e. a niece and aunt. The argument therefore seems to have involved seniority of line versus seniority of generation, i.e. parentelic precedence. (Orderic Vitalis, ed. Chibnall, iv, p. 199 n. 4). For a similar though milder dispute, cut short by the death of both niece in the senior and aunt in the junior line, see the case of Maine, ibid., ii, pp. 116-18, 304. For the quarrel between Robert of Balleme and Rotrou, count of Perche, depending in part on claims from the cousins, Mabel and Adeline of Belleme, see White, The First House of Belleme', p. 79; Complete Peerage, xi, appendix D, p. 122n.; Orderic Vitalis, ed. Chibnall, i, p. 213; ii, pp. 36265; iv, p. 397. For examples drawn from England see above, pp. 129-30. For descent in the female line where the division of estates was ultimately resolved see the honour of Bolingbroke (Compkte Peerage, vii, pp. 667-75; 743-46) and the lands of William of Arques, tenant of the archbishop of Canterbury, Domesday Monachorum of Christ Church, Canterbury, ed. D.C. Douglas (London, 1944), pp. 42-43; J.H. Round, Geoffrey de Mandevilk (London, 1892), pp. 180, 397-98. 18 Orderic Vitalis, ed. Chibnall, vi, p. 47.

250

Colonial England, 1066-1215

don charter of Henry I presents a strictly defined system: barons arranged the marriage of their daughters and other women folk, but had to consult the king, who would neither take payment for his permission nor prohibit the match unless to one of his enemies; if a baron died leaving a daughter as heir the king gave her in marriage by the advice of his barons; the king also promised not to give widows in marriage against their wish; similar arrangements were to pertain between the barons and their men.19 It is easy enough to illustrate all this in practice. Barons gave away daughters, sisters and nieces:20 how far they consulted the duke in the years before the Conquest is uncertain; that they did so under Henry I is occasionally made clear.21 The ruling house disposed of heiresses and widows; it also played a patronal role in other marriages.22 Real life was not as neady arranged as Henry Fs charter suggests. The ducal or royal right was not always maintained; Adeliza, daughter of the Countess Judith, niece of the Conqueror, was given in marriage to Ralph de Tosny, not by the king but by her elder sister's husband, Simon de St-Liz.23 Conversely the king's influence could extend far beyond the limits indicated in 1100; in 1121 Henry I gave Sibilla, daughter of Bernard de Neufmarche, to Miles of Gloucester during her father's lifetime, under terms which provided for the transfer to Miles of a considerable portion of Bernard's lands as a marriage-portion. This transaction was justified by the fiction that Bernard had surrendered all his acquisitions into the hands of the king, who was now granting them to Miles.24 Such a special provision may point to a breach of current conventions, but it would be wrong to conclude that practice was intended to be hard and fast. Some of the evidence is contradictory. Sometime after 1107 Hawise, daughter of Richard de Redvers, was married to William de Roumare. Orderic states that she was given in marriage by Henry I, which would be consistent with the rules of his charter.25 The lady herself, however, subsequently in her widowhood, gave part of her marriage-portion to the monks of Twynham for the souls of

19 Cc. 3, 4. It should be noted that the charter distinguishes childless widows from those with children, leaving the latter or other relatives with the custody of land and children. It also implies that the widowed mother was less liable to remarriage. 20 For daughters see the arrangements made by Heugon, Geoffrey of Mortagne and Hugh of Crepy (Orderic Vitalis, ed. Chibnall, ii, p. 22; iv, p. 160; v, p. 30). For a sister see Baudri of Bocquence (ibid., ii, p. 84). For a niece see Gilbert of Brionne (ibid., ii, p. 82). 21 See the case of Hugh de Gournay who discussed the future of his sister, Gundreda, with Henry I and, on the king's recommendation, gave her to Nigel d'Aubigny (ibid., vi, p. 192). This replicates the procedure envisaged in Henry's coronation charter. 22 For an heiress see Aubree, daughter of Guitmund of Moulins-la-Marche, given in marriage by Duke William (ibid., hi, p. 132). For a widow see Ada, widow of Herluin the elder of Heugleville, given in marriage by Duke Robert the Magnificent (ibid., iii, p. 252). For more general intervention by William I and Henry I see ibid., ii, p. 262; iv, p. 158; iii, pp. 258-60. 23 Complete Peerage, xii, pt 1, p. 762; Vita etpassio Waldevi Comitis, ed. F. Michel, Chroniques AngloNormandes, (3 vols, Paris, 1836-40), ii, p. 126. 24 Ancient Charters, ed. J.H. Round, Pipe Roll Society, 10 (1888), pp. 8-10. 25 Complete Peerage, iv, pp. 309-11; appendix I, p. 771; Orderic Vitalis, ed. Chibnall, vi, p. 380.

The Heiress and the Alien

251

several, including her dear brother Earl Baldwin, 'who gave me those lands in frank marriage when he gave me in marriage to my noble husband, William de Roumare'.26 It is tempting to say that Orderic was wrong and that Hawise was right; she after all should have known; Orderic, in contrast, got her name wrong;27 but both may reflect different aspects or stages of the same transaction. There is no need to assume that there was some kind of constitutional battle over who had the right to bestow a bride. It is more likely that there was reasonable give and take between king and baron, lord and man, so long as the result did not undermine due loyalty. A second control was stricter. Apparently, descent in the female line was always to a single heiress. In all the instances of Norman and Anglo-Norman descent mentioned by Orderic Vitalis, none involves parceny among heiresses. In England, where baronies descended in the female line, as had occurred in twenty cases by 1130, they descended to a single heiress. The coronation charter of Henry I runs: 'and if a daughter remains an heir';28 partition among several daughters was not considered. The Pipe roll of 1130 records twenty-seven fines for wardship and/or marriage. There is no hint in any of them that the inheritance was divided in the female line. The significance of this overwhelming accumulation of evidence depends, of course, on the assumption that in all these instances the heiresses were not single daughters. That can be settled beyond reasonable doubt by a few particular instances. The CountessJudith, niece of the Conqueror, widow of Earl Waltheof, left two daughters, Maud and Adeliza. Maud, the senior, married Simon de St-Liz c. 1090. Adeliza seems to have remained in the custody of her elder sister and brother-in-law until her marriage to Ralph de Tosny in 1103. It is said on the strong authority of the Complete Peerage that the two were co-heiresses, but the sole evidence is that the manor of Walthamstow, which had passed from Waltheof to the CountessJudith, ultimately came to the younger sister, Adeliza.29 That is inconclusive. The subsequent history of Judith's lands, which became the honour of Huntingdon, gives no hint of a division between the two sisters. Fees were lost from the honour but not on this account. The inheritance descended through Maud. Walthamstow was probably Adeliza's marriage-portion.30 The descent of the lands of Robert fitz Hamo provide another example. These were brought by his daughter Mabel to her husband Robert, illegitimate son of Henry I; they came to constitute the honour of the earls of Gloucester.31 The marriage took place between 1107 and 1114. At the time Mabel had at least two sisters; one became abbess of Shaftesbury and the other

26

Complete Peerage, iv, p. 311 n. 6. He misnamed her Matilda. 28 Cap. 3. 29 Complete Peer-age, xii, pt 1, pp. 761-62. 30 William Farrer. Honors and Knights'Fees (3 vols, Manchester, 1923-25), ii, pp. 296-301. Cf. J.H. Round, Geoffrey de Mandeville, pp. 191-93, 264-65. See also the Vita et passio WaMevi, which clearly refers to a marriage-portion (above, n. 23). 31 Earldom of Gloucester Charters, ed. R.B. Patterson (Oxford, 1973), p. 3. 27

252

Colonial England, 1066-1215

probably abbess of Wilton.32 Whether they had already taken the veil at the time of their sister's marriage is uncertain. Yet another example is provided by Rotrou, count of Perche, who married Maud, illegitimate daughter of Henry I. Maud died in the White Ship in 1120, leaving two daughters, Philippa and Felicia. There is no indication of when Felicia died, but there is no doubt that between 1120 and his second marriage, sometime before 1127, Rotrou treated Philippa and her husband as his sole heirs.33 Finally, there is the example of the extraordinarily complex descent of the lands of Robert de Tosny, who died in 1088, and of his son, Berengar, who died before 1116. These came, some directly, some by circuitous routes, to Alice, daughter of Robert and sister of Berengar, widow of Roger Bigod, who died in 1107. None passed by inheritance to Alice's younger sister, Agnes and her husband Hubert de Rye.34 The import of each of these instances might be challenged; it is far from easy to determine when older sisters married and when younger sisters took the veil or died. But they all point in the same direction: only one female heiress succeeded at a time. This was the rule which was altered. The change is mentioned in a charter of Roger de Valognes in favour of the priory of Binham. It refers to a statutum decretum, an established ruling,35 that 'where there is no son, the daughters divide their father's land by spindles, and the elder cannot take from the younger her half of the land without violence and injury'. Stenton dated the charter c. 1145.36 No date is given for the enactment to which it refers. The indications of the pipe roll of 1130 are that it was not yet in force. It seems unlikely that it was the work of King Stephen. A tentative date of 1130-35 would seem the best, but for the present purposes the precise dating does not matter very much. It is sufficient to say that there was a change of policy at about this time and that it was formal and deliberate. Up to this point there was no sign of parceny; after this point it became the common practice. From the 1140s onwards there is no difficulty at all in finding examples. Why was this done? We can only guess or reconstruct an answer; no contemporary provides an explanation. An essential preliminary is to understand how it was intended to take effect, and here there are too many answers. Glanvill, writing some fifty years after the statutum decretum, was in no doubt. Where there were no sons the inheritance was divided among daughters, with the proviso that the eldest daughter was to retain the chief messuage. The husband 32

The Heads of Religious Houses England and Wales, 940-1216, eds D. Knowles, C.N.L. Brooke and Vera London (Cambridge, 1972), pp. 219, 222. There is insufficient evidence of a third younger sister said to have married the count of Brittany. 33 Complete Peerage, xi, appendix D, pp. 112-13 and nn. See especially Rotrou's grant to the abbey of Tiron made 'assensu generis mei Helie filieque mee Philippe. . .Hoc concessit gener meus Helias et filia mea Philippa'. 3 \EYC, i, pp. 460-61, 466; HMC, Rutland, iv, pp. 107,144. 161; The Leicestershire Survey, ed. C.F. Slade (Leicester, 1956), p. 90. 35 After much consultation I have settled on this as the best translation, inadequate though it is. Stenton chose 'appointed law', which seems altogether too biblical in flavour, The First Century of English Feudalism (2nd edn, Oxford, 1961), p. 39. 'Enactment' is perhaps too legislative in tone. 36 Ibid., p. 261.

The Heiress and the Alien

253

of the eldest daughter performed homage for the whole fee. Husbands and descendants of the younger daughters need not perform homage until the third generation, when at last they were bound to do so. But they then performed homage and paid a reasonable relief not to the lord but to the heirs of the senior daughter. In like manner, though junior daughters and their husbands were bound to perform the due services to the lord they did it by hand of the eldest daughter.37 What Glanvill prescribed was derived from Norman parageas legal historians have long recognised.38 He has been taken seriously. In fact he leads us up the garden path. Putting it at its lowest, his scheme is a method of tax avoidance, for it deprives the lord of wardships and reliefs. Putting it at its highest, what Glanvill described certainly occurred; it engendered much litigation; but it is likely that it was still a novel procedure when Glanvill described it as part of the law of the realm;39 and it certainly was not the only way of arranging a partition. In 1236 King Henry III answered an enquiry from Ireland on these matters as follows: In our realm of England. . .it was always the law and custom in such cases that if there was any holding of us in chief, having daughters as his heirs, our ancestors and we, after the death of the father, have always had and received homage of all the daughters, and every one of them in such a case holds of us in chief; and if they happen to be under age we have always had the wardship and marriage of every one of them.

So much for tenants-in-chief. For tenants of others it was a different matter. The ruling in their case was infected to some degree by Glanvill's doctrine; the younger sisters performed service by the hand of the elder and the elder alone did homage; but the elder sister was not allowed to take homage from the younger, as Glanvill required.40 So there is a plain difference between Glanvill and the statement of 1236, in the case of tenants-in-chief a striking one. It is possible that Glanvill was right for 1189 and Henry Ill's advisers were right for 1236. But it is also possible that Glanvill is and was misleading. For his own generation and for the period stretching back to the statutum decretum, Glanvill's doctrine leads to obvious factual questions. Were the service and the performance of the service attached to eldest co-heiress or were they divided, like the land? Was relief paid to the senior line or to the lord? Was homage paid to the senior line or to the lord? In sum was there parageor complete partition? It is only in the case of homage that the answers are at all difficult and even there they can be implied. 37

Glanvill, De legibus, ed. G.D.G. Hall (London, 1965), p. 76. F. Pollock and RW. Maitland, History of English Law (Cambridge, 1898), ii, p. 276. 39 Milsom, 'Inheritance by Women', especially pp. 69-78. The earliest surviving record of such parage is in final concords concluded in the king's court. One of these is dateable to 1174^79, before Glanvill became chief justiciar (Curia Regis Rolls, xi, no. 869); another of less certain date also belongs to Henry II (ibid., viii, p. 387). 40 Statutes of the Realm, i, 5. The 1236 statement also excluded the elder from the wardship of the younger, a matter not specifically discussed by Glanvill. 38

254

Colonial England, 1066-1215

The statutum decretum is a good starting-point, for the charter which Stenton discussed is not the only evidence. The cartulary of Binham Priory, daughter house of St Albans, from which it was taken, includes some twenty documents relevant to the transaction.41 The first of these is a straightforward notification by Walter de Valognes that he and his daughter Agnes, with the consent of his wife Roesia, had granted Barney and Thursford in perpetuity to the monks of Binham.42 It is obvious that Agnes had an interest in the grant and the lack of any reference to a husband indicates that she was as yet unmarried. The second document is Stenton's charter. It is in the name of the chief lord of the fee, Roger de Valognes, probably Walter's first cousin. It makes plain that the land was granted with all solemnity, in the presence of the abbot of St Albans and of clerks and knights; Agnes and her father Walter gave seisin by placing a knife on the altar before the assembled company. Both charters, it should be noted, bear the marks of ecclesiastical draughtsmanship and were most probably the work of the beneficiaries; this accounts for the peculiar term statutum decretum.^ In this second charter Agnes remains unmarried, for there is still no reference to a husband, and she is now stated to be heir to her father in Barney according to the terms of the statutum decretum. The obvious implication is that she was the younger of two daughters who were dividing the inheritance. But the existence of an older daughter is no more than implied and the charter leaves it far from clear whether the whole, or merely half of Barney and Thursford was being conveyed; it also fails to explain why. At this point there had been some reference to Theobald, archbishop of Canterbury, by whose admonition and laudation the gift was made. That the archbishop's interest went beyond personal advice is made clear by the third document. This is a mandate of 1152-61 in which he informed Richard, bishop of London, and William, bishop of Norwich, that Richard of Colne and Geoffrey Tresgoz had contemned their agreement concluded before him concerning Barney and had withdrawn the gift of Walter de Valognes; the bishops were instructed to pronounce anathema and impose an interdict if the matter was not amended within eight days.44 So there were indeed two daughters and they were now married. The elder remains anonymous throughout; her husband, Richard of Colne, was a leading tenant of the Valognes barony.45 The younger, Agnes, who benefited from the statutum

41

BL, MS Cotton Claudius D. xiii, fos 49-59v. These are not included in Monasticon, iii, pp. 345-53, where the Valognes charters are confined to those of the senior line. 42 Binham Cartulary, fo. 53. 43 Ibid., fos 49-49v; Stenton, The First Century of English Feudalism, pp. 38-40; 260-61. The first charter (fo. 53) includes 'Concedimus igitur et statuendo sanctimus ut monarchi. . .' and 'Quicunque ergo hanc donacionem enervare presumpserit deleatur de libro vivencium et descendet ad inferni novissima; amen'. Compare the more elaborate anathema in the second (ibid., p. 261). 44 Binham Cartulary, fos 51v-52, printed in Avrom Saltman, Theobald, Archbishop of Canterbury (London, 1956), pp. 247-48. 45 He was the Richard de Calum or Calne who held 7Vfc fees of Robert de Valagnes in 1166, Red Books of the Exchequer; i, p. 360.

The Heiress and the Alien

255

decretum, married Geoffrey Tresgoz.46 The archbishop's intervention was effective. Agnes's grant was confirmed by her husband47 and subsequently by her successor, her daughter Petronilla, both with her husband and in her widowhood.48 The charter of Geoffrey Tresgoz records that an agreement had been concluded before the archbishop after which the land was divided and partitioned by measurement. The exact bounds were given both in his charter and in a subsequent confirmation of his daughter and son-in-law. The division was carried out in the presence of both Geoffrey and his wife, Agnes; his confirmation of the grant was executed in the monks' chapter at Binham, in which he was admitted to the spiritual benefits of the house.49 The grant was of half Binham, to be held by the service of a third of a knight; this service was subsequently conceded to the monks by the chief lord, Peter de Valognes.50 What interest did this leave to the older sister and her descendants? Certainly not the other part of Barney; that was a tenement which Geoffrey Tresgoz held of a different barony and conceded to the monks in his own right.51 The senior sister retained some interest in the advowson, which her daughter, Sibilla of Colne, and Sibilla's husband conceded to the monks after final concord concluded before Peter des Roches and other justices in 1214;52 but the advowson was not part of the original grant.53 Sibilla of Colne was also vouched to warranty in a grant assize between Richard, prior of Binham, and Robert of Horkesley, concerning a carucate in Barney and Thursford, which was concluded by concord in 1235.54 But that this was not part of the junior daughter's portion is demonstrated by a charter of Sibilla's grandson, which distinguished the gift of Walter de Valognes from that of Walter and Robert of Horkesley.55 This charter whereby Walter, son of Walter, son of John Bernard, great grandson of Walter de Valognes's elder daughter, abandoned all claim in Barney and Thursford wherein the monks of Binham held by gift of Walter de Valognes, Walter of Horkesley, Robert his son, or any other, is the first indication that the senior sister or her heirs had any interest whatsoever in the lands which the younger sister, Agnes, 46 Binham Cartulary, fo. 50v. The archbishop's letter, coupled with the clear implication of Roger de Valognes' charter, surely places Agnes as the younger sister. Milsom 'Inheritance by Women', p. 78, takes the other view to the point of doubting the existence of a younger sister. 47 Ibid., fo. 51. 48 Ibid., fos50v, 51-5Iv. 49 'Notifico vobis per presentia scripta in capitula fratrum ecclesie [sancte] Marie de Binham, accepta societate eorundem et beneficio loci me confirmasse predicte ecclesie de Binham dimidium partem ville que Berneya vocatur in terris, pratis, silvis, pascuis et hominibus et insuper advocacionem et dominium ecclesie ipsius ville sicut coram Theobaldo archiepiscopo inde fuit facta composicio deinde divisa et in fundo [rartefuniculo] distributionis partita', ibid., fo. 51. 50 Ibid, fo. 55. 51 Ibid., fo. 50. 52 Ibid., fos 52v, 56; Curia Regis Rolls, vii, p. 286. b3 See n. 49 above, where the grant of the advowson was made insuper. There is no specific mention of the advowson in the charters of either Walter or Roger de Valognes. 54 Binham Cartulary, fos 56-56v; Curia Regis Rolls, xv, no. 521. 55 Binham Cartulary, fo. 59v.

256

Colonial England, 1066-1215

conceded to Binham c. 1145. It dates from the late thirteenth century. Prior to that the senior sister and her heirs never intruded into the acts of the junior; the junior line confirmed Walter de Valognes's grant repeatedly without any reference to the senior; their acts were attested by different witnesses. It is difficult to avoid the conclusion that originally there had been no parage, that the division of c. 1145 had been absolute.56 The documents concerning Barney reveal nothing about the division and performance of feudal service. This is best examined at baronial level where division among heiresses began to take effect in the very same period. The Domesday holdings of Arnulf de Hesdin descended to his second son, who was probably hanged at Shrewsbury in 1138. Thereafter part descended through the marriage of one of Arnulf s daughters to the Fitzalans of Oswestry and part through the marriage of another to the Chaworths and the earls of Salisbury.57 The two parts constituted distinct baronies each accounting separately for feudal service.58 The same happened to the lands of William le Meschin when his son died without issue in 1135-40: one part of his lands descended through one daughter to the counts of Aumale, to become the honour of Skipton; another descended through a second daughter to the Courcys; yet another smaller portion went through a third daughter to the Mortimers.59 Once again the service was divided.60 The pattern was repeated in 1147-48 in the case of the lands of William Peverel of Dover and Bourn (Cambridgeshire) which were divided first among four and then among three sisters,61 and yet again in the division which followed the death of Robert Trussebut in 1193.62 In both these instances the feudal service was divided and accounted separately to the exchequer by each of the parceners, her husband and her heirs.63 When portions were subsequently divided further among a later generation of heiresses the service was further fragmented.64 And just as the service was partitioned so also were 56 This argument has to assume that Walter Bernard's charter was occasioned by circumstances not revealed in the cartulary. Another interpretation might be that the partition had reached the third generation, that the junior line performed homage to the senior and that the senior line's confirmation was now sought. Unfortunately, although it is clear that the senior line was now in the third generation, I have not been able to establish that this was true of the junior line. Moreover, the charter is a quitclaim not a confirmation and embraces other grants besides that of Walter de Valognes. In any case such an interpretation would not necessarily mean that parage had been intended ab initio. Notions derived from parage may have crept in as Glanvill's doctrine took hold. 57 J.H. Round. Studies in Peerage and Family History (London, 1907), pp. 115-29; IJ. Sanders, English Baronies (Oxford, 1961), pp. 112, 125. 58 Red Book of the Exchequer, i, pp. 274, 297-98. 59 EYC, vii, pp. 7-9. 60 Red Book of the Exchequer, i, pp. 225, 430-32; Book of Fees, pp. 97, 425. 61 W. Farrer, Feudal Cambridgeshire (Cambridge, 1920), pp. 159-166; R.W. Eyton, Antiquities of Shropshire, 12 vols (London, 1853-60), ix, pp. 64-77.

62

63

£yC,x,pp. 13-19.

For Peverel, see Red Book of the Exchequer, i, pp. 366-69, 372. For Trussebut, see Pipe Roll 6 Richard, p. 154; Pipe Roll 13 John, p. 31; Book of Fees, p. 157, 191. 64 For Peverel, see Eyton, Shropshire, ix, pp. 72-77; IJ. Sanders, Baronies, p. 19.

The Heiress and the Alien

257

the incidents. Where the payment of relief can be traced and isolated from the husband's it was paid separately by each parcener and her heirs.65 In view of this it is easy enough to say that, in the case of tenancies-in-chief, Glanvill was simply wrong, or was at least guilty of an error of omission. But it is not quite so simple. First, divisions and expectancies were biased by political interests. In 1176 for example, the earldom of Gloucester and the lion's share of the Gloucester lands were diverted to Earl William's youngest daughter, Isabella, who was to marry Henry IFs youngest son John, the future king. The arrangement was concluded by King Henry and William, earl of Gloucester, whose son had died ten years earlier. Isabella's two elder sisters and their husbands simply received £100 per annum from the estates.66 An even less equitable distribution was made between the co-heiresses and husbands of William de Say in 1185. In this case the younger daughter, Matilda, and her husband William of Buckland received the manor of Brunnington, the service of two tenants and the promise of land worth £10 from future acquisitions in return for abandoning any claim to such further acquisitions to the elder sister, Beatrice, and her husband, Geoffrey fitz Peter. Those further acquisitions would include the honour of Mandeville and the earldom of Essex. The division was confirmed before Henry II at Melksham and attested by Ranulf Glanville.67 Secondly, it is extremely difficult to penetrate behind the formal results of a partition to the decisions which led to it. The one early case where this is possible provides some basis for understanding the line which Glanvill took. This concerns Walter Espec, one of the chief agents of Henry I in northern England, a leader of the baronial army which defeated the Scots near Northallerton in 1138, who lived on to retire to the cloister in 1153 before his death two years later. Walter must in all probability have been a party to the statutum decretum; it would be astonishing if a man of his weight had not been consulted. He may have had a particular concern; a later tradition at Rievaulx provided him with a fictitious son who died in a riding accident, as a result of which he founded three monasteries and divided his lands among his three sisters.68 But what in fact happened was different, for sometime between Walter's death in 1155 and the completion of the pipe roll of 1158 Robert de Ros, son of one of the sisters, fined in 1000 marks for the inheritance of Walter Espec, and the sons of the other two sisters each fined in 100 marks for the division of the land against Robert de Ros.69 A confused entry on the charter roll of 1200 which largely

65

See in particular the relief of Gilbert Pecche of 200 marks in 1241, which included fifty marks for a third of Bourn (Excerpta e rotulisfinium, ii, p. 17). Gilbert represented the senior coparcener. Compare the direct statement concerning Agatha, youngest of the Trussebut sisters, that she held Kirk Deighton, part of the Trussebut inheritance, in chief of the crown, Cal I.P.M., i, no. 97; Yorkshire Irujuisitions, no. 11. 66 Earldom of Gloucester Charters, ed. Patterson, p. 5. 67 PRO, D L, 10/29; Pipe Roll 10 Richard, p. 139. 68 Cartularium abbathiae de Rievalle, ed. J.C. Atkinson, Surtees Society (1889), pp. 263-64. 69 Pipe Roll 2-3-4 Henry II, pp. 140, 146.

258

Colonial England, 1066-1215

repeats a charter of Henry II spells out the resulting terms.70 Robert received the whole honour of Walter Espec on the understanding that he gave £30 of land to each of his cousins, William de Bussy and Geoffrey de Trailly, and a life-interest in £50 of land to Jordan, William de Bussy's brother. The corroboration clause of this charter is a model of muddled or ambivalent thinking. Robert de Ros and his heirs were to have the whole of the honour which had belonged to Walter Espec in Northumberland and elsewhere 'except for those £30 of land which William de Bussy has thence with five enfeoffed knights and except for the other £30 of land which Geoffrey de Trailly has thence with five enfeoffed knights.' It would take a Glanvill perhaps to comment on the precise meaning of the exception and the sense of 'has thence' and that of course, more generally, is what Glanvill did. But the record of feudal service presents less ambivalence. Robert de Ros retained Wark, Helmsley and one fee in Walter Espec's patrimony of Old Wardon; Bussy and Trailly divided the rest of Old Wardon, with the service of five knights each.71 In 1161 the sheriff accounted for the scutage of the knights of the whole of Old Wardon, but William de Bussy accounted for his own portion alone in 1162 and this was also presented separately in the cartae of 1166.72 By 1185 the Bussy portion had been further divided between co-heiresses; so also was the service.73 Within the arrangements which confronted Glanvill there were embedded several general problems: the distinction between inheritance and acquisition, which may sometimes have influenced a partition;74 the possible confusion of inheritance in the female line with marriage-portions,75 and above all the need to determine aesnecia, ainesseor seniority. Now priority did not always go to the eldest heiress as the word ainesse implied and Glanville assumed.76 It might be argued that it should go to the youngest if she was still resident and unmarried in her father's household at his death;77 sometimes it went to the husband or 70

Rot. Chartarum, p. 32b. For Helmsley, see EYC, x, p. 15. For the division of Old Wardon see William Farrer, The Honour of Old Wardon', Publications of the Bedfordshire Historical Record Society, 11 (1927), pp. 6ff; IJ. Sanders, English Baronies (Oxford, 1960), p. 133. 72 Pipe Roll 7 Henry 77, p. 12; Pipe Roll 8 Henry 77, p. 42; Red Book of the Exchequer, i, pp. 335-36. 73 Rotuli de dominabus, pp. 11, 45; Sanders, English Baronies, p. 133. 74 This could well help to explain the division of Walter Espec's estates. The descendants of his two elder daughters shared his inheritance of Old Wardon. Robert de Ros, descendant of the youngest, received merely one fee of Old Wardon, but took all Walter's acquisitions. Robert probably got the lion's share on other grounds, but the distinction between inheritance and acquisition explains how the shares were defined. That Robert was descended from the youngest daughter is generally accepted. See Farrer, The Honour of Old Wardon', p. 4; EYC, x, p. 144. 75 For the relationship between partition and previous marriage-portions and actions arising therefrom see Milsom, 'Inheritance by Women', pp. 81ff. 76 Glanvill, De legibus, ed. Hall, pp. 75-76. For a straightforward example see the arrangements envisaged for the earldom Devon, 1200 (Rot. Chartarum, 52b). 77 See the case of the succession to William of Buckland, 1218, Bracton's Note Book, ed. F.W. Maitland (3 vols, London, 1887), i, no. 12, discussed with other possible cases by Milsom 'Inheritance by Women', p. 66. 71

The Heiress and the Alien

259

descendant who, like Robert de Ros, had the greatest pull.78 It might even be divided, as it was when Richard I split the Giffard inheritance between Richard de Clare and William Marshal with the proviso that Richard should have the ainesse and the caput in England and William the same in Normandy.79 But it always had to be settled in order to allocate the chief messuage, caput or castle. That was easy. It was much more difficult to determine what to do aboutjurisdiction. Glanvill's scheme provided one solution; first parage, then tenancy of the senior line. In these circumstances the original jurisdictional structure would be preserved; the acts of the junior lines would be confirmed and warranted by the senior; and this indeed is how Glanvill's system came to be reflected in the records of the king's court.80 But in these circumstances the junior heirs would be less free agents, with regard both to themselves and their beneficiaries. How could they dispose of land in perpetuity until they had performed homage for it? The alternative was to allow the partition to break the pattern of jurisdiction or at least to supplement it as more and more enactments and litigation sprang from the new tenurial conditions. If this happened it was reflected not so much in actions in the king's court, since there was far less need to argue about warranty, as in attestation of the charters of the parceners and their descendants. Sisters and their husbands may sometimes confirm each other's acts, cousins less frequently, their descendants rarely. By the same token witness-lists may overlap at first; then they too move apart. The process was determined by no rule, merely by the convenience of benefactors and beneficiaries. It signalled the reordering of jurisdiction, in all that stemmed from the granting and confirmation of charters, around lines of tenure changed by the consequences of partition. The process is admirably illustrated in the descent of the lands of Hamo Peverel, which passed with those of his brothers Pain and Robert, to three co-heiresses.81 None seems to have had seniority; Henry II addressed all three of their husbands jointly in a mandate ordering them to restore to the monks of Shrewsbury the land in Crudgington which has been given by Hamo Peverel whose heirs they were.82 The charters in which they restored the land were identical in form and shared some of the witnesses.83 It was a convincing example of the solidarity of heirs, the occasion no doubt engineered by the monks of Shrewsbury. But it soon weakened. There is no evidence of such co-operation in the next generation.84 The same is broadly true of the descendants of Walter Espec and Robert Trussebut: occasional and apparently haphazard confirma78

See above, p. 257. Cartae AntiquaeRolls 11-20, Pipe Roll Society, new series, 33 (1957), no. 564. 80 Milsom, 'Inheritance by Women', pp. 71-72. 81 Farrer, Feudal Cambridgeshire, p. 160. 82 Henry II to Hamo Pecche, Geoffrey of Waterville and Hugh of Dover: 'Precipio quod juste et sine dilatione reddatis Abbatie Salop. . .terram suam de Crugeltona qram Hamo Peverellus cujus heredes vos estis eis dedit', Eyton, Shropshire, ix, p. 67. 83 The Cartulary of Shrewsbury Abbey, ed. Una Rees (2 vols, Aberystwyth, 1975), i, nos 29-31. 84 Compare the charters of Hugh of Dover, Gilbert Pecche and Geoffrey and Asceline de Water79

260

Colonial England, 1066-1215

tion of each other's acts or shared action in the courts, usually in the first generation; thereafter separation.85 At Barney, where the consequences of partition were accentuated by the immediate alienation of the portion of Agnes de Valognes to the monks of Binham, the division seems almost tangible. There was a market place and courthouse with a garden to its south and a cross to the east before its door. These were the landmarks from which the demarcation of Agnes's lands began.86 Her portion was transferred to the monks with sake and soke and all free customs; the monks were to deal with men, lands and goods persectam ecclesie&s seemed best to them with prohibition from no one.87 In this case the linked effect of partition and alienation in alms was the fragmentation of jurisdiction. The relevance of this to the expansion of royal jurisdiction will not be lost. One effect of the statutum decretumwas to aid the work of Henry II. The reasons for Glanville's choice will now be obvious. The lands which he left to his children were largely accumulated during his own lifetime as an administrator; it is reasonable to imagine that he would seek to give them some unity.88 He knew as he wrote that they would descend to his three daughters. After his death the eldest of the daughters enjoyed seniority; she and her husband were primarily accountable for her father's debts. Occasionally the exchequer wavered and took an interest in the other heirs, but the main weight of the evidence suggests that parage was in operation among Glanville's descendants.89 Glanville had described as the custom of England what he wanted to see as the custom of England. He must have been a powerful influence in implanting such a custom. This assumes, of course, that the law book we know as Glanvill was written by Glanville. If we transfer the authorship to the next likely

continued

ville Eyton, Shropshire, ix, pp. 67, 72, 76. For analogous action in a yet further generation, between co-heiresses of Asceline de Waterville, see ibid., ix, pp. 77, 79. 85 In the case of Espec the descendants of one of the families, Bussy, now represented by heir's general in the next generation, viz. Builly and Wake, were parties to an action in 1198-1200 concerning the church of Old Wardon against the abbot of Old Wardon. One of the other original participating families, Trailly, figured in this but not as a party and the action was resolved against the heir's general of Bussy (Farrer, The Honour of Old Wardon', pp. 9-11). In the case of Trussebut, Cecily and Agatha made fine for land in Newsome in 1197 (EYC, x, pp. 48-49); all three co-heirs combined in litigation against Henry du Puiset concerning Market Weigh ton in 1204 (Curia Regis Rolls, x, p. 42); and sometime after 1227 Hilary, who was not the senior, confirmed a gift to the Temple by her dear nephew, Robert de Ros, who died in retirement as a Templar (EYC, x, p. 42). 86 The bounds appear in their earliest form in the charter of Agnes de Valognes' husband Geoffrey Tresgoz and in greater detail in the confirmations of her daughter and son-in-law, Binham Cartulary, fos 51, 50v, 51v. 87 Ibid., fos 50v, 51v. 88 R. Mortimer, The Family of Ranulf Glanville', Bulletin of the Institute of Historical Research, 54 (1981), pp. 1-16, especially 14-15. 89 SJ. Bailey, 'Ranulf de Glanville and his Children', Cambridge LawJournal (1957), pp. 163-81, especially 173-76. See alsoJ.N.L. Myres, 'Notes on the History of Butley Priory, Suffolk', Oxford Essays in Medieval Histoty Presented to H.E. Salter (Oxford, 1934), pp. 190-206, especially 190-93, which traces the descent of the patronage of Butley in the senior line.

The Heiress and the Alien

261

candidate, Geoffrey fitz Peter, the same point arises in a different but perhaps even weightier form, for on the seniority of his wife depended the honour of Mandeville and the earldom of Essex.90 This long excursus on how the statum decretum took effect helps to explain why it was enacted. There was much wider choice of action than that indicated by Glanvill. Baronies were divided into distinct entities. Feudal incidents were thereby preserved and multiplied. Lords and individual heirs might conclude arrangements which gave one of the heirs a lion's share. But heirs could go their own separate ways, thereby fragmenting the old pattern of justice of a single united fief. These effects may not all have been foreseen with equal clarity. But they provide the grounds for understanding motive. It is best to begin with the lord of a fief, who was not only a lord but also a father of sons and daughters, and one conscious that there were other lords, also fathers. So the question which he faced was not just one of whether to divide his own lands among co-heiresses, were he to be placed in that predicament. He would also be conscious of the greatly increased opportunities which the division of other baronies would create for his younger sons, for the market of actual or likely heiresses would be greatly expanded, and that at a time when his capacity to provide for younger sons was reduced as the flow of acquisitions, from which he and others had benefited, dwindled.91 And partition could do no harm so long as seniority was applied to castles, office and title. Indeed, it might well multiply the incidents of marriage, wardship and relief if each parcener and her heirs were made directly dependent on the lord. It diversified the exercise of patronage and probably added to the profits. Looked at from the point of view of the lord, especially if the lord were king, parceny was a bargain. He could scarcely lose. The same was true for potential husbands. To be sure, the lucky one who might have married an heiress sole would lose something. But all the others would gain part of an inheritance or a chance thereof, and usually this would be far more extensive than a bride's marriage-portion. And there may have been consequent advantages for others. There was less need for families to provide for younger sons if heiresses could be found to sustain them. Finally there were advantages for the potential heiresses. They were now heiresses in parallel rather than in series. Younger sisters stood a better chance of better marriages, with more hope of inheriting and less dependence on the paternal or fraternal charity of the marriage-portion. Since we have suffered so long with the Oedipus complex we might dub this the Cinderella syndrome. Who knows what ranged through the imagination of young girls and women, some of whom must have been raised on the varied sexuality of the Lais of

90

Complete Penage, v, pp. 120-22. In the event the conflicts of Stephen's reign increased the flow of acquisitions but this could not have been foreseen when the statutum denytumwas agreed, as I suggest, 1130-35. 91

262

Colonial England, 1066-1215

Marie de France, a woman poet composing for an audience of women?92 However, if the change might be expected to bring benefits for all, those who disposed of marriages and controlled succession benefited more, for it multiplied their resources; and the king benefited most of all. For him the change was one of degree: an enactment which could transform the life of the younger womenfolk was for him little more than fine tuning which improved the performance of his incidents. Kings and their officials soon took note. In 1176 an enquiry into women in the king's gift was included in the articles of the eyre.93 From 1185 the record of a similar enquiry survives as the Rotuli de dominabus et pueris et puellis.94 From 1130 onwards the Pipe rolls provide systematic information on the crown's exploitation of rights of wardship and marriage. From 1199 the fine rolls illustrate the system in even greater detail. The abundance of this evidence creates the impression that the system not merely yielded profit but was designed for financial exploitation; and that the king's right of marriage was used not simply to ensure loyal vassals but to provide parvenus of unknown, low or alien extraction with the permanent status which marriage to an heiress could provide.95 Such is the standard picture. It still leaves room for some discussion. It is common ground that the king's control of wardship and marriage established a kind of market with some very grubby characteristics. The coronation charter of Henry I implies that Rufus had charged for his approval of marriages.96 Henry's promise to abandon the practice was vain. The pipe roll of 1130 reveals a brisk trade. The exchequer already distinguished between widows who brought their husbands dower and marriage-portions and heiresses who brought 'land'.97 Through one large proffer of 1000 marks William Pont de 1'Arche obtained the daughter and office of Robert Mauduit, who had been one of the chamberlains.98 Another proffer was plainly speculative: William Maltravers offered 1000 marks and £100 for the widow of Hugh de Laval, to hold the whole of Hugh's lands (the honour of Pontefract) for fifteen years and her dower and marriage-portion thereafter.99 These were not small transac92

So Elizabeth Williams in Christine Fell, Women in Anglo-Saxon England (London, 1984), pp. 175^79, to whom I should not attribute my assessment of the audience. The Lais are ed. A. Ewert (Oxford, 1947). 93 Assize of Northampton, cap. 9. 94 Ed. J.H. Round, Pipe Roll Society, 35 (1913). The Rotuli are the returns made by the justices of the evidence presented in court. See Round, ibid., pp. xvii-xix. They are not, as they are sometimes presented, a record of exchequer admininstration, although they were used by the exchequer for the subsequent management of the estates of some heiresses. 95 A.L. Poole. Obligations of Society in the XII and XIII Centuries (Oxford, 1946), pp. 97-100; Sidney Painter, Studies in the History of the English Feudal Barony (Baltimore, 1943), pp. 66-72. 96 Cap. 3. 97 The terminology of the pipe roll and the coronation charter is identical. Only one entry on the roll, a proffer of Williams Maltravers, associates a woman with heieditas, Pipe Roll 31 Henry /, p. 87. She was his wife. 98 Ibid., p. 37. 99 Ibid., p. 34.

The Heiress and the Alien

263

tions; how the offers were assessed the roll does not reveal. The Rotuli de dominabusofllSb, however, illustrates the system of assessment in detail. The justices now took evidence of the value and annual yield of a woman's lands and the number and sex of her children. They received estimates of her age and her children's ages. Some of the returns assess her social standing and the potential of the incidents to be derived from her by indicating that she was of baronial or knightly stock.100 That was not the end of the story. By the reigns of Richard and John it is not unusual to find instances of competitive bidding or of charges placed so high that they can only be regarded as financial and political extortion.101 The seamier side of the system is even better revealed in a tale told by Adam of Eynsham.102 It starts with Agnes, wife of the Lincolnshire knight, Thomas of Saleby. She was young, he was old, and their marriage was barren. She was determined that, if he were to die, she should not come into the custody of his brother, William of Hartshill. So she pretended pregnancy, padded herself well out, took to her bed in pretended labour and, with the connivance of a woman of the neighbourhood, produced a supposititious daughter whom she had christened, with appalling candour, Grace. All details were attended to; her aged husband was well briefed and the real mother was brought into the household as nurse and foster-mother. However, her brother-in-law was not to be put off and he carried the case to the royal court whence it was claimed by Bishop Hugh of Lincoln as a matter for ecclesiastical jurisdiction; Hugh investigated the matter, pronounced on it sternly from the pulpit and threatened excommunication against the offenders, but the aged husband died before he could confess his complicity. Grace was now an heiress and she was transferred with her lands to Adam de Neville, brother of Hugh, the chief forester. Adam of Eynsham tells us that he married her though she was less than four years old. Meanwhile the bishop, reinforced by a confession of complicity from the real mother, ordered that sentence of excommunication should be pronounced every Sunday from all the neighbouring churches. In 1199 William of Hartshill launched an action in the curia regis against Adam de Neville; Adam was able by trickery to bring the case to the favourable judgement of the royal court but was snatched away by death before he could hear of his triumph. He was immediately followed by others, apparently undeterred by the accumulation of fatalities which were coming to be associated with the wretched Grace. Before the year was out she was given in marriage to a second husband, one of the king's chamberlains. He too died, and at the time at which Adam of Eynsham wrote she had been passed to a third who was worse than all the rest. Now the main points in Adam's story are undoubtedly correct.103 The only serious criti100 Poole, Obligations of Society, pp. 100-3; D.M. Stenton, The English Woman in History (London, 1957), pp. 38-40. 101 Poole, Obligations of Society, pp. 99-100; J.C. Holt, Magna Carta (Cambridge, 1965), pp. 107-8. 102 Magn,a vita Sancti Hugonis, ed. Decima L. Douie and Hugh Farmer, 2 vols (London, 1961), ii, pp. 20-27. 103 It is succinctly stated in the first record of the action in 1194, Rot. Cuiiae Regis, i, p. 78.

264

Colonial England, 1066-1215

cism is that Adam de Neville was Grace's guardian, not her husband.104 However, she had certainly been married to the second intruder in the story, Norman de Camera, at the point at which she passed to the third, Brian de Lisle, in 1205.105 So she was certainly a widow at the age of ten. The action concerning the supposititious child and the inheritance came before the royal court between 1194 and 1199.106 William of Hartshill backed his claim with an offer of 500 marks.107 Norman de Camera acquired his young wife at the cost of 200 marks and she was passed on to Brian de Lisle for 300 marks.108 Brian reinforced his proffer with some potent guarantors, including the bishops of Salisbury and London, Hugh, archdeacon of Wells, William, earl of Salisbury, William Marshal, earl of Pembroke and Saer de Quency; a wise precaution given the history of the case.109 But nothing availed against the damnation threatened by Bishop Hugh. All the marriages were barren and the lands ultimately reverted to the line of the rightful heir.110 It was not all like that. First, a fair proportion of proffers came from relatives of the heir or heiress, or from widows seeking to continue control of their children.111 Secondly, the evidence of resistance to the system requires careful handling. The murder in 1135 of William Maltravers, the royal official who intruded by the purchase of a marriage into the honour of Pontefract, was an expression of acute hostility among the tenants of the honour, but it was an isolated incident.112 The more frequent and obvious objections came not from tenants, but from widows seeking to maintain their widowhood. Sometimes these sprang from real conflict. In 1194 the stock of the widowed countess of Aumale was sold because she had refused to marry William de Forz: she married him.113 In 1205 Kingjohn seized the dower lands of Alice, widow of John Belet, because she refused to marry according to his will; a proffer for her had been made by Ralph Ridel. Ultimately her father stepped in and paid to preserve her widowhood and recover her dower.114 But these issues seem more contentious than they were in general because of the official jargon of the exchequer; 104

Ibid., i, p. 452. Rot. de Oblatis etFinibus, p. 240. 106 Rot. Curiae Regis, ed. Francis Palgrave, 2 vols (London, 1835), i, pp. 78, 452. 107 Rot. de Oblatis etFinibus, p. 20. 108 Ibid., pp. 40, 240. 109 Ibid., pp. 240-41. 110 Rotuli hundredorum, i, p. 294; Magna vita Sancti Hugonis, ii, p. 27n. I am indebted to the careful annotation of the editors for much of the detailed record of the case. 111 The exact proportion would be difficult to calculate because family relationships are not always stated in the record of the proffers on the pipe or fine roll. My impression of the fine rolls of the early years of John's reign is that roughly half of the proffers involved some family relationship. This is likely to be an underestimate. 112 W.E. Wightman, The Lacy Family in England and Normandy, 1066-1194 (Oxford, 1966), pp. 68-73. 113 Pipe Roll 6 Richard, p. 163; Stenton, TheEnglish Woman in History, p. 36. 114 Rot.de Oblatis etFinibus, pp. 226, 287; Curia Regis Rolls, iii, p. 257; Poole, Obligations of Society, p. 98. 105

The Heiress and the Alien

265

widows made fine 'that they should not be compelled', or 'that they should not be distrained to marry', and it is chiefly from this common form of words that the general impression of compulsion arises. In reality these daunting phrases must often have meant no more than that a widow had decided to buy exemption from the royal right of marriage. When she did so she was bidding not just for freedom but also for power, power over her inheritance, her dower and her marriage-portion. One of the first great ladies to do so was the Countess Lucy, thrice widowed after the death of Ranulf le Meschin, earl of Chester, before 1130.115 She offered 500 marks that she should not take a husband for five years. She also offered an additional 100 marks that she might do justice in her court amongst her men; already by 1130 there was an appeal to the king from her jurisdiction. Finally she offered a further large sum for the recovery of her patrimony which had been dispersed when her last husband became earl of Chester.116 In this too she was successful. There is a strong suspicion that the Countess Lucy fined for widowhood in order to put her house in order. So the widow's fine is ambiguous evidence. Contemporary comment is equally unhelpful. We know from the letters of Archbishop Lanfranc that some English ladies fled to monasteries after 1066 'for fear of the French'.117 Orderic commented that: Noble maidens were exposed to the insults of low-born soldiers and lamented their dishonouring by the scum of the earth. Matrons, highly born and handsome, mourned the loss of their loving husbands and almost all their friends, and preferred death to life.118 But he was not particularly concerned that a woman might be demeaned by marriage to one of lower status. He was surprised when Matilda de Grandmesnil widow of Hugh de Mont-Pincon abandoned friends and kindred in Normandy to accompany her new love, a young knight called Matthew, on the journey to Jerusalem, but he made no comment to suggest that the association was socially demeaning.119 The same is true of his famous description of the men whom King Henry I raised from the dust. They were 'of base stock' or 'low birth'; they were ennobled, 'set on the summit of power', rendered formidable to the chief men of the land; but Orderic, so conscious of their social and political promotion, did not allude to their marriages.120 Yet one of those he named, Richard Basset, and many whom he did not name, who enjoyed rapid promotion under Henry I, William and Nigel d'Aubigny, William d'Aubigne of Belvoir, Miles of Gloucester, Pain fitz John, Eustace fitz John, William Maltravers, William Pont 115 116

Complete Peei age, vii, appendix J, pp. 743-46. Pipe Roll 31 Henry I, pp. 110, 114.

117 The Letters of Lanfranc Archbishop of Canterbury, ed. H. Clover and M. Gibson (Oxford, 1979), no. 53(1077-89). 118 Orderic Vitalis, ed. Chibnall, ii, p. 268; cf. William of Poitiers, ed. Raymonde Forville, p. 232. 119 Orderic Vitalis, ed. Chibnall, iii, p. 166. 120 Ibid., vi, p. 16.

266

Colonial England, 1066-1215

de 1'Arche, owed much of their new found fortune to the marriage of a wellendowed lady, usually an heiress.121 Orderic's silence in the matter was not unusual. The vernacular literature which entertained the twelfth-century noble household contains only one clear example of an heiress demeaned in marriage. That was Gaimar's heroine Argentille, heiress to the kingdom of Lindsey, forced by her wicked uncle to marry a cook, so that the uncle could usurp the realm. The cook turned out to be Havelock, prince of Denmark, so all ended well. Gaimar wrote about 1140; the tale was repeated in the Lai d 'Haveloc of c. 1200.122 That apart, one or two tales turn on forced marriages in which the woman feels disgraced, but without any element of social disparity between husband and wife.123 The possibility of disparity was recognised. In the case of the supposititious heiress Grace, Bishop Hugh of Lincoln pointed out to one of the intending husbands that the girl was a serf and that he would not enjoy her for very long,124 but no one apart from Gaimar made much of the theme. All this has skirted round the term disparagement, and for good reason. For the word has come to play a quite inappropriate role. To put it in rough and ready terms: in 1215 King John promised in Magna Carta, following a demand in the Articles of the Barons, that heirs would be married without disparagement and after consultation with the nearest relative; this must therefore have been a ground for complaint. King John and perhaps others must therefore have disparaged heiresses; the cry of 'no disparagement' is therefore yet one more example of baronial protest. Such is the usual interpretation.125 It is quite wrong. First the derivatives of the word disparagare seem not to have been used in England before 1194. In that year two proffers were entered on the pipe roll, each concerning the marriage of a male heir on the understanding that he was not to be disparaged.126The same word was applied to both heirs and heiresses

121

For Nigel d'Aubigny, William d'Aubigny of Belvoir, Miles of Gloucester, William Maltravers and William Pont de L'Arche, see above pp. 250n., 252, 262, 264. William d'Aubigny married Maud, daughter of Roger Bigod (Complete Peerage, i, p. 233); Pain fitzjohn married Sybil, daughter of Hugh de Lacy (Wightman, The Lacy Family, pp. 175-83); Eustace fitzjohn married first Beatrice, daughter and heiress of Ivo de Vescy and, secondly, Agnes, sister and co-heiress of William fitz Nigel, constable of Chester (Complete Peerage, xii, pt 2, pp. 273-74). For general comment see R.W. Southern Medieval Humanism and Other Studies (Oxford, 1970), pp. 214-20. 122 Geffrei Gaimar, L'estoire des Engleis, ed. A. Bell, Anglo-Norman Text Society, 14-16 (1960), especially lines 96-167. For the Lai d'Haveloc, Havelok theDaneznd other derivations see L.H. Loomis, Medieval Romance in England (New York, 1963), pp. 103-14. 123 Enforced marriage and misalliance are themes in the Roman d'Eneas, Orsan de Beauvais, Raoul de Cambrai and Daniel and Beton, but without any emphasis on disparagement. In Chretien of Troyes' Yvain Laudine de Landuc is eager to secure the hand of \Vain because such a marriage will be no disgrace for the daughter of a duke (ed. T. Reid, lines 1815-18), but that scarcely presents disparagement as a burning issue. I am indebted to Dr Peter Noble for bringing these sources to my attention. 124 Magna vita S, Hugonis, ii, p. 26. 125 Poole, Obligations of Society, pp. 97-98; S. Painter, The Reign of KingJohn (Baltimore, 1949), pp. 217-19. 126 Pipe Roll 6 Richard, p. 238.

The Heiress and the Alien

267

on the roll of 1195 and the following years;127 it occurs more frequently, but still sporadically, on the fine rolls of John's reign. The evidence leaves no room for doubt: 'no disparagement' was a condition imposed by the crown; there is no evidence at all that it was a requirement of the suppliant. The first evidence comes from 1194; it provides yet one more example of a possible initiative by the new justiciar, Hubert Walter. The words seem to occur at random, a few instances each year. There is no obvious reason why they should have been attached to one proffer and not another. But coming as they do from the Exchequer there can be little doubt about the purpose: they were intended not so much to protect the heir or heiress but to state the conditions on which the proffer had been agreed and to exclude the parvenu not because he was a social threat (although he might well be that) but because his intrusion would alter the terms of the negotiation. 'No disparagement' was a financial constraint, not a social complaint. It was invented by the crown. King John lost little when it was included in Magna Carta.128 What then was disparagement? It is easier to say what it was not. Through his wife Isabel de Clare, the 'bachelor' William Marshal, fourth son of a knight, became earl of Pembroke.129 No one objected. Through marriage Geoffrey fitz Peter, knight, justice of the forest and sheriff, acquired the honour of Mandeville and the earldom of Essex. No one objected except the Walden chronicler and that not on the grounds of disparagement.130 In the next generation Geoffrey's sons married the daughters of Robert fitz Walter, after Richard of Clare the greatest of the barons of East Anglia.131 No one forced Robert fitz Walter into these alliances. Sometime before 1210 Gilbert de Gant, one of the leaders of the Lincolnshire rebellion in 1215 who laid claim to the earldom of Lincoln, married off his daughter Juliana to a tenant, Geoffrey de Armentieres, whose father held ten fees of the Gant honour. The agreement was attested by a choice selection of tenants of the Gant honour.132 They can scarcely have found the alliance unacceptable. Plainly all these marriages came within the social pale. But there was one which did not. On 1 September 1216 Baldwin de Redvers, son and heir to the earl of Devon, died aged sixteen. Within a month his widow, Margaret, who was a double heiress, both to Warin fitz Gerold and William de Courcy, was given in marriage to Fawkes de Breaute, the Norman knight who was one of Kingjohn's most vigorous commanders in the civil war.133 The reac127

Pipe Roll 7 Richard, pp. 177, 222, 246. I by-pass here the letter of 1212 of William the Lion of Scotland conceding the marriage of his son Alexander to King John 'ita quod non disparagetur' since it is of doubtful authenticity. See E.L.G. Stones, Anglo-Scottish Relations, 1174-1328 (London, 1965), pp. 12-13; Regesla regum Scottorum, ii, ed. G.W.S. Barrow (Edinburgh, 1971), no. 505. 129 Complete Peerage, x, p. 358. 130 Ibid., v, pp. 122-23, 124 note b. 131 Ibid., v, pp. 127, 132. 132 Sir Christopher Hatton's Book of Seals, ed. L.C. Lloyd and D.M. Stenton (Oxford, 1950), no. 298. 133 Complete Peerage, iv, p. 316. 128

268

Colonial England, 1066-1215

tion to this marriage determined much that followed. The scandal was threefold. First, there was a wide social gap between the status of husband and wife; the instances I have mentioned suggest that this might have been bridged. Secondly, there may have been an element of violence in the arrangement. When Margaret sought dissolution of the match in 1224 she apparently claimed that she had been seized in time of war and married against her will, or so at least we are told by Roger Wendover.134 If so, it was a matter not of disparagement but of enforced remarriage. Third, Fawkes, though of respectable Norman extraction, called down on his own head all that hostility which was directed against the officials and soldiers whom King John brought to England from the Touraine and Poitou. All those influences and instincts, political and administrative, legal, artistic, literary and linguistic which were steadily distinguishing England from France and giving definition to a sense of Englishness, came to be concentrated against him. He was seen not simply as a parvenu, an intruder, but as an alien, a foreigner. He was the first layman of Norman extraction to be damned as such. This was entirely the work of Roger Wendover and Matthew Paris. During the civil war of 1215-17 Fawkes offended St Alban by levying 'tenseries' on the abbey; in return St Alban's two historians pilloried him. At Margaret's death in 1252 Matthew could still recall: 'Noble though she was, she was married to this ignoble creature, the pious to the impious, beauty to the foul, unwilling and coerced, she was handed over by the tyrant John to whom no evil was abhorrent'.135 That was disparagement fiercely denounced. It was largely generated after the event. It stands in sharp contrast to a strictly contemporary document. Sometime, and probably quite soon after the marriage Warin fitz Gerold mortgaged the manor of Wootton Courtenay to his new son-in-law in return for 200 marks 'pro necessitate sua'. The deed acknowledges Fawkes and Margaret as man and wife and makes prospective arrangements for their heirs. It is attested by Peter des Roches, bishop of Winchester, William Marshal, earl of Pembroke, William Longsword, earl of Salisbury and others.136 For the moment, at any rate, Fawkes enjoyed the blessing of the establishment. Fawkes' case never had an end. He was driven out of England in 1224. Pope Honorius intervened on his behalf to sustain the matrimonial bond, but Archbishop Stephen Langton, who was well aware that the political situation in England made Fawkes's restoration impossible, dragged his feet. The case was resolved by Fawkes's death in 1226.137 It never had an end in yet another sense. It has cast a lurid retrospective light on the provisions of Magna Carta ever since, so that it is sometimes used to help explain the provision against disparagement of 1215 even though it occurred a year later.138 At the time it may have given the terms of Magna Carta a posthumous importance. The defeat of Fawkes 134

Chronica majora, iii, pp. 87-88. Ibid., v, p. 323. 136 HMC, Rutland, iv, pp. 55-56. 137 P.M. Powicke, KingHenry III and the Lord Edward (2 vols, Oxford, 1947), i, pp. 61-66. 138 Painter, Kingjohn, p. 218.

135

The Heiress and the Alien

269

may have made it easier for Ella, countess of Salisbury, to reject the importunities of the young knight Raymond, nephew of Hubert de Burgh, when it was thought that her husband had died on the crusade in 1226. But this too is a tale told by Wendover and repeated by Matthew Paris, and one which smacks too much of romance to ring true.139 These stories were symptoms. King John may easily have allowed the provision on disparagement in 1215 with the thought that it referred to little ill. He was probably right at the time. But within a few years those ills were being revealed as a disease.

139

Chronica majora, iii, pp. 101-2.

This page intentionally left blank

14 1153: The Treaty of Winchester

The settlement of 1153 was not part of J.H. Round's study of Geoffrey de Mandeville. He left us with what is still the best general account of King Stephen's younger son, William, count of Boulogne, and of the lands which came to him under the terms of the settlement.1 But of the settlement itself he said very little;2 and for this it was to Stubbs that Round's first readers would have had to turn and to the exact information and good sense of the Constitutional History? Round's reticence is a pity, for it detracted from his treatment of King Henry II's early charters.4 Quite apart from that, he might have helped us with insights gained from his study of the charters of King Stephen's reign. At least he would have given us some hard grist. In the event, the settlement has become something of a muddle; or rather we have got into a muddle about it. For example, my tide is not a slip. The settlement is very occasionally described as of Wallingford, sometimes of Winchester, more usually, now, of Westminster. Some have used more than one location, for good reason, or for no apparent reason at all.5 I will justify my own use of Winchester in what follows. Then, again, the document we know as the treaty of Westminster is not quite the same as the settlements described in the 1

J.H. Round, Studies in Peerage and Family History (London, 1907), pp. 147-80. Round, Geoffrey de Mandeville, pp. 176-77 and n. 3 Stubbs, The Constitutional History of England (3 vols, 6th edn, Oxford, 1897), i, pp. 359-63. 4 Round, Geoffrey de Mandeville, pp. 234-42. 5 Stubbs refers to the peace of Wallingford and Westminster, drawn up at Wallingford, adopted at Winchester, and published at Westminster. Le Patourel mentions Winchester, as does Dr Chibnall: J. Le Patourel, The Norman Empire (Oxford, 1976), p. 112; M. Chibnall, Anglo-Norman England, 1066-1166 (Oxford, 1986), p. 99. Professor Warren refers to both Winchester and Westminster, as does Professor Hyams in different studies: W.L. Warren, Henry II (London, 1973), pp. 62, 333; PR. Hyams, review in EHR, 93 (1978), p. 858; idem 'Warranty and Good Lordship in Twelfth-Century England', Law and History Review, 5 (1987), p. 497. R.H.C. Davis refers to Winchester in Gesta Stephani, ed. K.R. Potter and R.H.C. Davis (Edinburgh and London, 1976), pp. 240 n. 1, but uses both locations and elides the contents of the agreement in R.H.C. Davis, King Stephen, 1135-1154 (3rd edn, London, 1990), pp. 119, 122. Professor Biancalana uses Westminster, and is perfectly clear about the relationship of the Westminster document to the treaty: J. Biancalana, 'For Want of Justice: Legal Reforms of Henry II', Columbia Law Review, 88 (1988), pp. 434, 467n. Professor Palmer also favours Westminster, but with complications (on which see below, p. 274): R.C. Palmer, The Origins of Property in England', Law and History Review, 3 (1985), pp. 8-9.1 followed their usage in J.C. Holt, 'Magna Carta, 1215-1217: The Legal and Social Context', in E.B. King and S.J. Ridyard, eds, Law in Mediaeval Life and Thought, Sewanee Mediaeval Studies 5, (Sewanee, TEN, 1990), pp. 11-12, where I expressed an initial doubt about the accepted treatment of the settlement. 2

272

Colonial England, 1066-1215

chronicle narratives. These promise the restoration of the disinherited, and this has no place in the Westminster document. Our responses to this have varied. No one has been prepared to reject the narrative sources, or even argue that promises made at Winchester were jettisoned at Westminster. On the contrary, Professor Palmer has treated the chronicle stories and the Westminster document as complementary arrangements of equal weight. Yet there are differences revealed in curious ways. At one pole, Professor Palmer has given us 'the Compromise', decked out with a capital; at the other, John Le Patourel cannily placed the Treaty of Winchester' in inverted commas.6 Presumption has affected presentation. Alongside this we have given the settlement of 1153 diverse roles in the development of the common law. For Davis it established heritability, of the crown, of the great baronies, and hence, in time, for all the free landowning classes; it was an opinion he modified, but still held firmly in his final work.7 Professor Warren looked to the settlement for the origins of nemo tenetur, the rule that no free man need answer for his free tenement except to the king's writ.8 Professor Milsom suggested that the agreement underlay the emergence of the writ of right.9 He also hinted very tentatively that it may have led to a distinction between right and seisin, and this was developed into a full-blown theory by Professor Palmer, who argued that the arrangement made in 1153 for the descent of the crown, whereby Stephen remained in seisin but Henry was recognised as rightful heir, was used as a model for the general territorial settlement and thereby differentiated seisin and right both in real and procedural terms.10 All this is hypothetical and much of it has proved contentious.11 This is to select just a few of the many variant approaches. That there is such a choice leaves a number of impressions. One is that we have come to distance ourselves from the evidence, to give structures pride of place over documents. Another arises from the kind of structures we have built. If we turn to the documents seeking support for this or that model of the origins of the common law, we shall assuredly find it. The trouble is that we shall all find different things. The subject deserves better. We ought at least to know what the evidence allows us to say and what it does not. It is in this spirit that I turn first to the document we know as the treaty of Westminster.12 6

Palmer, 'Origins of Property', pp. 8-13; Le Patourel, Norman Empire, p. 112. R.H.C. Davis, 'What Happened in Stephen's Reign, 1135-54', History, 49 (1964), pp. 1-12, esp. 10-12; Davis, King Stephen, pp. 150-53. 8 Warren, Henry II, p. 334. 9 S.F.C. Milsom, The Legal Framework of English Feudalism (Cambridge, 1976), pp. 178-79. 10 Palmer, 'Origins of Property', pp. 8-13; foreshadowed in Palmer, The Feudal Framework of English Law', Michigan Law Review, 79 (1981), pp. 1142-49. 11 For an alternative hypothesis to Davis, see above, pp. 113-14. For an alternative to Warren, advocating the more gradual development of nemo tenetur, see Van Caenegem, Royal Writs in England from the Conquest to Glanvill, Selden Society, 77 (London, 1959), pp. 212-25, esp. 223-25. For considered criticisms of Palmer, see Biancalana, 'For Want of Justice', pp. 467-70; and Hyams, 'Warranty', pp. 497-503. 12 RRAN, iii, no. 272. For further comment on the text and on earlier editions, see EYC, viii, pp. 15-16. 7

The Treaty of Winchester, 1153

273

The first thing to be said is that it is not a treaty. We have come to think of it as such perhaps because it marked the end of the nineteen long winters of the anarchy; but it is not a treaty in the strict diplomatic sense. It is not a bipartite document; it is in the name of King Stephen, not King Stephen and Henry fitz Empressjointly. It records no exchange of sureties. There are no named guarantors; instead it is attested by great men drawn from both the opposing parties, a point of some importance to which I shall return. If not a treaty, what then is it? It states that it is a charter.13 That was perhaps the best that could be said, but it does not fit that description at all exactly. Presumably (for it survives only in a copy) it bore the seal of King Stephen, but it does not grant anything specifically and it names no beneficiaries. It is sui generis, designed for a special occasion and a particular purpose, and it has no exact parallel in any other surviving document of the twelfth century. The settlement between Henry II and his sons which we know as the treaty of Falaise of October 1174 is perhaps the nearest analogue, but that was a conventio, secured by hand of the Young King and his brothers.14 If we cast the net more widely, the Westminster document is quite unlike the various treaties and agreements concluded between the great nobles of the Anarchy or between them and King Stephen, the empress, or her son Henry. There is no affidatio in manu, no provision of guarantors, no obsidesfiducie.1^It certainly breathes of the atmosphere of the Anarchy; its insistence on the adjudication and arbitration of the great churchmen is not unlike Matilda's call on the 'Christianity of England' to back her second concession to Geoffrey de Mandeville;16 and here, as in other agreements of the Anarchy, homage is the starting-point.17 But in its construction and purpose it stands alone. 13

'per conventiones inter nos prolocutas, que in hac carta continentur'. Recueil des actes de Henri II, ed. L. Delisle and E. Berger (3 vols, Paris, 1916-27), ii, no. 468. 'Notum sit... quod pax . . . in hunc modum Deo volente reformata est'. 'Hanc conventionem firmiter tenendam ex parte sua assecuravit Henricus rex filius regis in manu domini regis patris sui.' In the particular cases of the documents of 1153 and 1174 I am taking a somewhat stronger view of the distinction between carta and conventio than that in P.R. Hyams, The Charter as a Source for the Early Common Law', Journal of Legal History, 12 (1991), pp. 173-89, esp. 174. 15 For affidatio in manu, compare the charters of Matilda and Henry for Aubrey de Vere, earl of Oxford (RRAN, iii, nos 634-35), and for both this and the abides Jiducie, see Matilda's second charter for Geoffrey de Mandeville, earl of Essex (ibid., iii, no. 275). The whole question is illuminated by Round in Geoffrey de Mandeville, pp. 176-77, 384-87. Similar, if somewhat milder, language reappears in Duke Henry's grant of Bitton to Robert fitz Harding of 1153 (ibid., iii, no. 309) and, much more understandably, in his agreement with Jocelin, bishop of Salisbury, concerning the custody of Devizes Castle, April 1153 (ibid., no. 796). It is exceptional at this late date; it does not figure, for example, in Henry's charter for Ranulf, earl of Chester, of 1153 (ibid., no. 180). It should be noted also that this section of the Bitton charter does not lie at all easily alongside Henry's grant of Berkeley (ibid., no. 310) of roughly the same date, for which Robert fitz Harding paid 500 marks in recognicione. These inconsistencies increase the probability that the Bitton charter is spurious. See below, n. 58, for further discussion and references. 16 Ibid, no. 275. 17 E. King, 'Dispute Settlement in Anglo-Norman England', Anglo-Norman Studies, 14 (1992), pp. 115-30.1 am grateful to Professor King for letting me see a copy of his paper prior to publica14

274

Colonial England, 1066-1215

The best description of it is Sir Charles Clay's: 'the King's notification of the treaty'.18 That is not quite exact: conventiones in the Latin have become 'the treaty' in the English; and perhaps 'notification' is too weak a word for a document with such a potent list of witnesses. But it contains the root of the matter. The Westminster charter is a formal promulgation of terms previously agreed. Its purpose was to bring peace to the land through the publication of agreements necessary to that end. It is an administrative document pursuant to the treaty agreed at Winchester. As for the 'treaty of Westminster', there was no such thing. This may look like splitting hairs. Yet the consequences are important. First, there is no difficulty in accepting that certain matters agreed at Winchester might not be included in the Westminster charter. The one does not invalidate or call in question the other, so long as there is sufficient explanation of the differences. Secondly, there is no ground at all for treating Westminster and Winchester as 'two components' in which 'the Treaty of Westminster was the model for the restoration of the disinherited', as agreed at Winchester.19 The two are not on a par; the one, Winchester, was a treaty, the other, Westminster, its administrative consequence; quite apart from that, Winchester came first. The Westminster charter, therefore, states matters which are past and settled, conventiones. . . prolocutas. Stephen has recognised Henry as his heir. Henry has performed homage; Stephen has given security on oath to maintain him as a son and heir. William of Blois has performed homage to Henry, and Henry has conceded the enormous territorial endowment which William received in England and Normandy. Homage has been performed by the earls and barons of the two parties, Henry's to Stephen and Stephen's to Henry; archbishops, bishops, and abbots have performed fealty to Henry. Arrangements have been made and securities taken for the custody of the critical castles: Wallingford, the Tower, Windsor, Oxford and Winchester. The charter is attested by the great men of both parties; it is subject in some of its provisions to the advice and arbitration of the archbishop and bishops, and ecclesiastica justicia is invoked against any contravention. In all this the Westminster charter almost certainly gives us terms which were first agreed at Winchester. That, at least, is our best guess. There may well have been other meetings; the assemblage of provisions may have been more miscellaneous in origin than a straightforward derivation from Winchester. Here and there, too, the Westminster charter speaks in the language of an evidentiary rather than a continued

tion. I do not, however, follow him in regarding the Westminster document as an exemplary conventio. 18 EYC, viii, p. 26. On occasion Davis used similar words: 'the terms agreed at Winchester on 6 November 1153, and subsequently promulgated in the form of a charter at Westminster' (Gesta Stephani, p. 240 n. 1); as did Biancalana: 'On Christmas at Westminster Stephen issued a charter containing the terms of the treaty' ('For Want of Justice', p. 467 n. 162). Neither, however, is consistent in his usage. 19 Palmer, 'Origins of Property', pp. 8-9.

The Treaty of Winchester, 1153

275

historic past; how else could it assert that the citizens and garrisons within the king's lordship had done homage and given security to the duke? At one point it seems to reveal the gap in time between agreement and promulgation. It was arranged that Reginald de Warenne, uncle to the wife of William de Blois, was to have custody of the castles of Bellencombre and Mortemer, if he so wished. Yet Reginald was one of the witnesses to the Westminster charter. Why should he not have come to a conclusion? The incongruity is perhaps best explained by the supposition that the Westminster charter states terms which had been agreed at an earlier stage, when Reginald was still left with an option or perhaps was not even present. But these are minor matters beside the most striking feature of the Westminster charter: it scarcely ventures at all into the future. Apart from contingencies which will require ecclesiastical intervention or arbitration, it restricts itself to three general statements: the duke's family will support the settlement; the king will deal with the affairs of the realm with the counsel of the duke; and the king will exercise his royal justice throughout the realm, in the duke's part as in his own. The last two were long-term promises. No one could know that Stephen would be dead within a year. Before that Henry had departed for France. Neither had the time or opportunity to feel encumbered. So the Westminster charter contained some extended commitments. It is all the more significant, therefore, that certain matters agreed at Winchester on 6 November were not included. For these we are entirely dependent on narrative sources. Henry of Huntingdon gives the place. Robert of Torigny supplies the date, and he and the Gesta Stephani provide detail of the content.20 These are contemporary or nearly contemporary accounts which are mutually independent.21 Torigny arid the Gesta tell of two decisions which do not figure elsewhere: first, that adulterine castles were to be destroyed: and, secondly, that lands were to be restored to their rightful owners. It is reasonably easy to guess why these matters were omitted from the Westminster charter. That was designed to bring peace to the land. At this stage, a formal denunciation of adulterine castles would have been provocative, impossible to enforce. Likewise, a promulgation that lands were to be restored might well have launched many a demandant on the path of self-help. In any case, who was to judge a rightful claim? Both provisions were full of dangers; quite apart from that, who could gauge the extent of either? It was better to omit them from any promulgation of peace. Promises for the future were best kept general and preferably obscure. On the restoration of land, the two chroniclers come closer than is normally allowed. According to Torigny, it was sworn that possessions which had been seized by intruders should be restored to the ancient and lawful possessors who

20 Henry of Huntingdon, Histvria Anglorum, ed. R. Arnold, Rolls Series (London, 1870), p. 289. Robert of Torigny, Chronicle'm Chronicle ojthe Reigns ofStephen, Henry II and Richard I, ed. R. Hewlett, Rolls Series (4 vols, London, 1884-89), p. 177. 21 Davis, King Stephen, pp. 144-46.

276

Colonial England, 1066-1215

had them in the time of good King Henry.22 The Gesta simply state that the disinherited were to be restored ad propria.23 Now propria means something more precise than 'their own'. It must be read as 'inheritances', as contrasted with 'acquisitions'. Quite apart from that being one of the prime senses of the word, this is what it clearly means in the particular context, for the agreement did not seek to give the stamp of approval to the 'acquisitions', the intrusions and enforced claims, of the Anarchy. If we read the chronicles so, they coincide; the only difference is Torigny's reference to Henry I, to which I shall return. Moreover, what they are getting at is quite clear: inheritances were to be restored; acquisitions were at risk; unlawful acquisitions would be revoked. It was a precise guide-line. Nevertheless, it brought its own particular difficulties. What, for example, was to be done when land, rights and title acquired during and through exploitation of the Anarchy had already been inherited, or at least subject to a succession, before 1153? This leads at once to the most famous case of all and the subject of Round's book, Geoffrey de Mandeville, who died in 1144. Now Round's account of the succession to Geoffrey is vitiated by his unquestioned assumption that the eldest surviving son, Arnulf, was legitimate - legitimate but disinherited. This was not so. There is overwhelming evidence that Arnulf was a bastard.24 He cannot, therefore, have affected the straightforward succession of his legitimate half-brothers, Geoffrey and, subsequently, William, to the honour of Mandeville and the earldom of Essex. Hence, Arnulf s existence has nothing to do with the manner of Geoffrey's succession, and, regretfully, the gloomy sentence with which Round concluded his study truly belongs to the Gothic fiction which it sought to emulate.25 We 22

'Juratum est etiam, quod possessiones, quae direptae erant ab invasoribus, ad antiques et legitimos possessores revocarentur, quorum fuerant tempore Henrici optim[i] regis.' 23 'exheredati ad propria revocarentur'. 24 Round, Geoffrey de Mandeville, pp. 227-33, 238. Compute Peerage, v, p. 116, suggests that Arnulf and Geoffrey may have been half-brothers, but includes them as brothers in the pedigree chart. Sidney Painter seems to be alone in stating that Arnulf was 'probably illegitimate', but he provides no evidence, Feudalism and Liberty (Baltimore, 1961), pp. 216, 217n. The evidence of Arnulf s illegitimacy is: (a) He attests charters of his father but never as son and heir, and he never occupies prime place in the lists of witnesses. The foundation charter of Walden priory is particularly significant; W. Dugdale Monasticon anglicanum, ed.J. Caley et al. (8 vols, London, 1817-30), iv, pp. 148-49. (b) He accepted enfeoffment in Kingham from both his half-brothers, thereby recognising their title - crucial evidence which Round used, but failed to appreciate. (c) He also attests charters of his half-brother Geoffrey. (d) Neither he nor his descendants ever laid claim to the Mandeville inheritance. Taken together these arguments are overwhelming. Much of the charter evidence on which they are based is collected in A. Charlton, 'A Study of the Mandeville Family and its Estates, 1066-1236' (unpublished University of Reading Ph.D. thesis, 1977). 25 'And as if by the very irony of fate, Arnulf, his disinherited son, alone continued the race, that there might not be wanting in his hapless heirs an everstanding monument to the greatness at once of the guilt and of the fall of the man whose story I have told', Round, Geoffrey de Mandeville, p. 244.

The Treaty of Winchester, 1153

277

cannot piece together the complete story of the succession, because the charter which Matilda issued between 1144 and 1147 survives only in summary form. But, to take it at its face value, it was for Geoffrey, son of Geoffrey, earl of Essex - in short, the younger Geoffrey was not recognised as earl; and it distinguished between his 'whole inheritance' and 'all tenures which she had conceded to his father'.26 The distinction on which the settlement of 1153 was to rely seems to have been made clearly enough. Now Geoffrey the younger certainly attested one of Matilda's subsequent charters as earl of Essex, but that is not clinching evidence by itself.27 Moreover, beyond the general phrases of the summary of Matilda's charter, there is nothing to show whether the younger Geoffrey obtained all that his father had gained in 1140-41. What is certain is that the whole matter was reviewed after the settlement of 1153. Henry II, as king, now made Geoffrey earl of Essex, giving him the title and the third penny of the pleas of the county, all to be held hereditarie.28 His charter uses the language of an initial creation, not, as Round thought, because of the exclusion of Arnulf, but because the terms of the settlement of 1153 put all acquisitions of the time of war in question. It was as though they had never been. Hence Geoffrey's earldom is presented as a new one.29 For the rest, Henry confirmed to Geoffrey and his heirs all the lands of his great grandfather, grandfather, and father of whomsoever held. The father, the first earl, is not mentioned without the others, either in general or with regard to the family's privileges in London, which gained special mention. It seems clear, in short, that only his inherited lands were now being confirmed to his son. The king also quitclaimed for ever the lien which Henry I had held on the manors of Walden, Sawbridgeworth and Waltham, thereby releasing the new earl from a long-standing hold which the crown had hitherto retained.30 That apart, the tide and third penny of pleas of the county of Essex were all that the new earl retained of the massive gains which his father had made in 1140-41. It is worth noting that Geoffrey de Mandeville had not been among those who attested the Westminster promulgation of the terms of 1153. Nevertheless, there is little sign that he or his descendants were disgruntled with its consequences. They clearly nursed a claim to the custody of the Tower of London which Geoffrey, the fifth earl, revived in 1215, but that probably went back beyond the days of the first earl to his father, Wil-

26 RRAN, iii, no. 277. This is a summary in BL, MS Lansdowne 229, of a copy in the lost volume of the Great Coucher of the Duchy of Lancaster. 27 RRAN, iii, no. 43, a charter of 1144 for Geoffrey Ridel, son of Richard Basset. It survives as a transcript in the Basset roll, not as an original. This detracts somewhat from its specification of Geoffrey's title. For an original charter of John, count of Mortain, attested by Geoffrey fitz Peter as earl of Essex, 1191-93, see Cat. Docs. France, no. 61. In fact Geoffrey acquired the title in 1199. 28 Round, Geoffrey de Mandeville, p. 235-36. 29 It is so understood in Complete Peei age, v, pp. 116-17. 30 For the origin and later history of the lien, see C.W. Hollister, The Misfortunes of the Mandevilles', in Anglo-Norman World, pp. 117-27.

278

Colonial England, 1066-1215

liam, and perhaps earlier.31 The rest of the settlement they seem to have accepted. The Mandevilles provide a model of the application of the terms agreed at Winchester. But it was not the only model; or, rather, the logic which it followed could be made to yield quite different results. The Mandevilles had been no great friends of the Angevin party. Miles of Gloucester, earl of Hereford, and his son Roger, on the other hand, were. No charter survives to record the succession of Roger to Miles when Miles died in 1143, but it seems clear that Roger succeeded to the title and all the lands which Miles had acquired. It is certain that this was confirmed by Henry II in 1154-55; Henry, indeed, added further grants.32 Now Henry's charter is of great interest, for a simple confirmation in fee and inheritance of the fee of his father and the fee of Bernard de Neufmarche, which Miles had acquired by marriage, was not seen to be enough. After confirming the lands which Roger inherited from his father and mother, the charter shifts from the language of confirmation, reddidisse et concessise, to the language of a fresh grant, insuper etiam dedi et concessi, and embraces not only Henry's own grant but also all the concessions which Miles had won from Matilda, along with the earldom, in 1141. These, granted infeodo et hereditate, are enumerated one by one, and, despite the language of a new grant, it is plain that Matilda's charter of 1141 was the model for Henry's of 1154-55.33 Yet he did not mention it; no one reading his charter could suspect that there had been Matilda's. Acquisitions had to be conveyed afresh. In this case it did the crown little harm. Roger took the habit in 1155, the title lapsed, and all the acquisitions were lost. Robert of Torigny noted that Roger's brother, Walter, succeeded in paternam hereditatem tantum.^ It was not until 1200 that the title was revived for Henry de Bohun, Roger's nephew. In return he had to surrender the charter of 1154-55 and all the claims which might arise therefrom. Where the original beneficiary of the Anarchy still survived under the new king, the logic of Henry's charters was not so clear cut. In 1153 Hugh Bigod's position as earl was ambiguous, though he appears as earl of Norfolk in the charters of Matilda as far back as 1141,35 The original grant of the tide does not survive, however, and there is even doubt as to who granted it, Stephen or Matilda. At all events, Henry II's charter of 1155, which made Hugh earl of Norfolk anew, is quite consistent with other charters which followed up the settlement of 1153. But it differs from them in one respect: Henry conceded to him 'all tenements of whomsoever's fee which he had reasonably acquired', and these, like the rest of the grant, were to be held by Hugh and his heirs.36 31 J.C. Holt, Magna Carta (2nd edn, Cambridge, 1992), pp. 208, 263-65. For the early association of the Mandevilles with the Tower, see Round, Geoffrey de Mandeville, pp. 38, 439. 32 Complete Peer-age, vi, p. 454; D. Walker, The "Honours" of the Earls of Hereford in the Twelfth Century', Transactions of the Bristol and Gloucestershire Archaeological Society, 79 (1960), pp. 180-82. 33 Rotuli Chartarum, p. 53. Cf. RRAN, iii, no. 393. 34 Torigny, Chronicle, p. 185. 35 Davis, King Stephen., pp. 138-39. 36 Cartae Antiquae Rolls, 11-20, ed.J. Conway Davies, Pipe Roll Society, new series, 33 (1957), no. 553.

The Treaty of Winchester, 1 153

279

We know nothing of the circumstances which lay behind this charter, which is unusual, if not unique, in leaving the acquisitions of the Anarchy unenumerated and controlled by the single word rationabiliter. In another case, such circumstances are very clear. By the accession of Henry II, Eustace fitz John had built up a vast accumulation of estates in which the main element came from his wife, Beatrice de Vescy.37 With Walter Espec, he had been the greatest of Henry I's 'new men' in northern England. He was alienated by Stephen's seizure of castles in 1 1 38 and fought on the Scottish side at Northallerton. A second marriage to the heiress of the constable of Chester then led him into the orbit of the earl of Chester. By the end of the reign of Stephen he had accumulated vast estates, including many undertenancies, the most important of which was a beneficial enfeoffment of fourteen fees which he obtained from Roger de Mowbray after the latter's capture at Lincoln in 1141.38 He now set about securing the succession of his son William. First, he tried substitution. Some time in 1 153-54, whether before or after the Winchester settlement is unclear, Duke Henry issued a charter to William, already called de Vescy, in which he confirmed the gift which his father Eustace had made to him, along with all the tenements he held on the day he had performed homage, and all his rights in England and Normandy.39 We cannot know how much that embraced. Even so, the charter is a good illustration of the tactics to which an old hand like Eustace might resort. The rest was entirely accidental. Eustace fought with the king against the Welsh in 1 157 and was killed in an ambush. His ill fate was his son's good fortune, for within days, perhaps within hours, someone secured from the king at Rhuddlan a charter in favour of William de Vescy. The king was in a giving mood. The charter confirmed to William all the acquisitions which his father had made by marriage, by royal grant and by other means. It embraced Alnwick and the honour of Yvo de Vescy, which came to Eustace by marriage with Beatrice; it specified all that Henry I had given him and included all he held of King David of Scotland and his son Henry; it recounted all the undertenancies he had obtained from the archbishops of York and the bishops of Durham, from Richmond, Mowbray, Paynel, Aumale, Clere, Mortain and Gant; it included fees simply by place and tenant; and finally, to round matters off, it concluded with all other lands and tenures of the aforesaid Eustace fitz John. It is an unparalleled record of acquisitive achievement. And all was now confirmed to William de Vescy, not only in feodo et hereditate, but under the words of conveyance normally reserved for the confirmation of an inheritance: Sciatis me reddidisse et concessisse.40 The charter was witnessed by the great men 37

For an account of Eustace, see Complete Peerage, xii, pt 2, pp. 272-74, and appendix B, pp. 7-1 1 . There is a useful summary in J.A. Green, The Government of England under Henry I (Cambridge, 1986), pp. 250-52. For a disentanglement of the Tison-Vescy lands, see EYC, xii, pp. 1-4. 38 Charters of the Hon,our of Mowbray, 1107-1191, ed. D.E. Greenway (London, 1972), p. xxvii, no. 397. 39 RRAN,iii,no.9123 . .

40 The Percy Chartulary, ed. M.T. Martin, Surtees Society, 117 (1911), pp. 291-94, summarised in Green, Henry /, pp. 251-52.

280

Colonial England, W66-1215

and household officers present in the army: William, the king's brother; Roger, earl of Clare; Geoffrey, earl of Essex; Richard du Hommet, constable [of Normandy]; Henry of Essex, constable [of England]; William de Briouze; Manasser Biset, steward; Warin fitz Gerold, chamberlain; Richard de Lucy; and others. For his family, nothing quite so became Eustace fitz John as the manner of his death. Now it may be that Henry's generosity in this case is also to be explained by the probability that most of Eustace's acquisitions were made in the reign of Henry I, for Torigny tells us that in 1153 ancient and lawful ownership was referred to the time of good King Henry. Whether this is Torigny's gloss or the true wording of the settlement it is impossible to say. None of the documents I have discussed so far uses Henry I as point of reference, although they refer not infrequently to gifts made by him.41 The fact was that reference to the time of King Henry or to the day on which King Henry was alive and dead could only complicate the settlement. Stephen, Matilda and the young Duke Henry had all used these or similar words as occasion demanded.42 They were not, therefore, a very useful tool for separating sheep from goats. Moreover, they were potentially inconsistent with the main intention of restoring the disinherited, for some had been disinherited by Henry I: the Stutevilles, for example, after Tinchebrai: the Lacys of Pontefract in 1114-18. By 1153 both had eased or thrust their way back into the reckoning, Ilbert de Lacy, reputedly through complicity in murder. His brother, Henry de Lacy was now pardoned and restored, and Robert de Stuteville obtained fees from his Mowbray rivals.43 In such cases, tempus regis Henrici was not merely useless: it was a positive impediment.44 But this was perhaps no more than an eddy in the main stream. In all probability, tenure confirmed by Henry II would have been inherited. In all probability, the ancestor of the tenant would have held under Henry I. Wherever this common situation occurred, tempus regis Henrici would very likely be called in to reinforce his grandson's confirmations. Hence it appears very frequently in Henry IFs early charters. Moreover, that much-neglected document, the coronation charter of Henry II, confirmed to church and laity all the concessions, gifts, liberties, and free customs which Henry I had granted. Witnessed by Richard de Lucy sole, it indicated that Henry I was to be the model for the government of Henry

41

E.g., the charter to Roger, earl of Hereford: Rotuli Chartarum, p. 53. For Stephen, see RRAN, iii, nos 10, 187, 255, 286, 386, 472, 678; for Matilda, ibid., nos 274, 316,343,821,897; for Henry, who mainly uses the formula tempore regis Henrici avi mei, ibid., iii, nos 130, 239f, 462, 575, 653, 704, 710, 795, 902, 997. 43 For Stuteville, see EYC, ix, pp. 1-6,116-17, 200-1; Mowbray Charters, pp. xxviii, 247. For Lacy, see W.E. Wightman, The Lacy Family in England and Normandy, 1066-1194 (Oxford, 1966), pp. 66-68, 87-88; RRAN, iii, nos 428, 429; EYC, iii, pp. 143-44. 44 For further comment on the disherisons of Henry I, see Davis, King Stephen, p. 151; R.W. Southern, 'King Henry I', in Medieval Humanism and Other Studies (Oxford, 1970), pp. 211-25; H.A. Cronne, The Reign of Stephen 1135-54: Anarchy in England (London, 1970), pp. 156-60. 42

The Treaty of Winchester, 1153

281

II.45 More gradually the main principle of the settlement of 1153 - that acquisitions made in the reign of Stephen were all subject to review - developed into the rule that no case would be accepted in the royal courts based on tenure temporis giverrae. That was firmly in place in the first surviving curia regis rolls.46 By then, 1194, rejection of acquisitions under Stephen and reliance on the day on which King Henry was alive and dead had finally come together. But earlier there were cross currents. Very soon after 1154, certainly before 1162, a statute was promulgated, one effect of which was that Engish claimants could not recover land without proof of tenure on the day on which King Henry I was alive and dead or later. Here the deadline was used, not to underwrite tenure, but to prevent the English from attempting to reverse the expropriations of the Norman Conquest. It could be used as a terminus post as well as ante quern.47 How many cases were there, how were they settled, and how soon? I have already discussed Mandeville, Hereford, Bigod and Vescy, and mentioned Lacy of Pontefract and Stuteville. To these we have to add charters for the earls of Arundel,48 Gloucester,49 Leicester,50 and Oxford,51 and for Richard de la Have,52 Fulk de Lisoures,53 Payn de Montdoubleau,54 and Warin fitz Gerold,55 to name only the more certain items. It is very likely that most of these belong to 1154 and 1155;56 the latest seem to be Aubrey de Vere's issued at Dover prior to the

45 W. Stubbs, Select Charters (9th edn, Oxford, 1913), pp. 157-58, where it is pointed out that the charter antedates the appointment of Becket as chancellor in January 1155. 46 Milsom, Legal Framework, pp. 178-79. 47 See Van Caenegem, Royal Writs, no. 169, for the prior and monks of Winchester. Velpostea cannot be dismissed as an error, for, although the phrase does not appear in a similar writ for the priory of Dover (ibid., no. 172), it does appear in the form of velpost in a writ for the prior and monks of Worcester (ibid., no. 217n.). I am proposing above a simple, straightforward explanation of these writs. Cf. Professor Hyams' comment: * [the English] alone had to plead on facts from the reign of the despised Stephen!' (Warranty', p. 500); and, for earlier views, see Van Caenegem, Royal Writs, pp. 216-17; D.M. Stenton, English Justice between the Norman Conquest and the Great Charter, 1066-1215 (London, 1965), pp. 31-32; P.R. Hyams, Kings, Lords, and Peasants in Medieval England (Oxford, 1980), pp. 251-52; Biancalana, 'For Want of Justice', p. 502. 48 For confirmation of the castle and honour of Arundel and the third penny of Sussex, see Cai Charter Rolls, iv, p. 257. 49 For general confirmation, including the privileges of Burford, see Actes de Henry II, i, no. 17. 50 For general confirmation, see L.W. Vernon Harcourt, His Grace the Steward and Trial of Peers (London, 1907), p. 60. 51 For allocation of title, see Sir Christopher Hatton's Book of Seals, ed. L.C. Loyd and D.M. Stenton (Oxford, 1950), no. 40. 52 Ancient Charters, Royal and Private, Prior to AD 1200, ed. J.H. Round, Pipe Roll Society, 10 (1888), no. 36. 53 HMC, Middleton Manuscripts, pp. 2-3. 54 Ancient Charters, no. 34. 55 H.G. Richardson and G.O. Sayles, The Governance of Medieval England (Edinburgh, 1963), p. 437. 56 The Bigod, Lisoures, and Montdubleaux charters were all issued at Northampton in January 1155; the earl of Hereford's at Nottingham in February.

282

Colonial England, 1066-1215

king's crossing to the Continent in January 1156, and William of Gloucester's, issued at the siege of Chinon in 1156. We simply do not know how large a tip of how large an iceberg these charters represent. Documentary survival was hazardous. Quite apart from that, some successions and confirmations may well have taken place without any written record. When that stalwart Angevin, Baldwin de Redvers, died in 1155, he was succeeded by his son Richard. Richard later made endowments for the souls of his father and mother and for that of the noble King Henry, 'who gave the land to my ancestors', but he made no reference to any confirmation by Henry II.57 Others, like Reginald, earl of Cornwall, simply survived (in his case until 1177), secure in the support which they had given to the Angevin cause, requiring documentary confirmation of their position only if challenged. But not all Angevin supporters felt secure. Robert fitz Harding, for example, buttressed his advance into the baronial ranks with charter after charter, some of questionable authenticity, from Henry both as duke and king.58 Disputed claims are the least documented of all. To all appearances some of the quarrels of the Anarchy simply petered out, leaving none of the charters of confirmation characteristic of the years immediately following Henry II's accession. William de Roumare continued as earl of Lincoln to die as a monk of Revesby c.1161. His rival, Gilbert de Gant, one of Stephen's supporters, died in 1156 with his title denied or ignored, only for it to be reclaimed by his nephew in 1216.59 The dispute over the associated earldoms of Huntingdon and Northampton followed a similar pattern. In 1153 these were held by Henry, son of King David of Scotland, and Simon de Senliz respectively. Both died in that year. Simon was succeeded by his son, also Simon, but the family lost the earldom on or shortly after the accession of Henry II. In 1157 it was incorporated in the earldom of Huntingdon, when that was revived for the royal house of Scotland. It was separated once more and returned to Simon de Senliz after the Scottish war of 1173-74.60 No charters survive for any of these successions or transfers. Not all disputes were a consequence of the Anarchy. The contention over the earldoms of Huntingdon-Northampton lay in the half-blood, originating in the two marriages of Maud, daughter of Earl Waltheof . Such issues were likely to occur in any family; they fed on the Anarchy, certainly, but would probably have occurred, civil war or no. The quarrel between

57

Compute Peerage, iv, pp. 312-13. R.B. Patterson, The Ducal and Royal Acta of Henry Fitz Empress in Berkeley Castle', Transactions of the Bristol and Gloucestershire Archaeological Society, 109 (1991), pp. 117-37, mounts a powerful case for the authenticity of the charters. For my suspicion of one of them, see above, n. 15. A further difficulty is that doubt about one charter affects the rest, especially those written by the same scribe. For Robert fitz Harding more generally, much the best account is now Patterson, 'Robert Fitz Harding of Bristol', Hastens Society Journal, 1 (1989), pp. 109-22. 59 Davis, King Stephen, pp. 122, 134-35; Compkte Peerage, vii, pp. 669, 672-74. 60 KJ. Stringer, 'A Cistercian Archive: The Earliest Charters of Sawtry Abbey', Journal of Society of Archivists, 6 (1980), pp. 325-34; Compkte Peerage, vi, p. 644; W. Farrer, Honors and Knights'Fees, 3 vols (London and Manchester, 1923-25), ii, p. 298. 58

The Treaty of Winchester, 1153

283

the Paynels and Robert de Gant is yet another example.61 The descent of the lands of Lacy of Weobley was equally contentious.62 Finally, within the first year of the reign, the new king had to use force or the threat of force to ensure the surrender of castles from supporters of both sides: William, count of Aumale, Roger, earl of Hereford, Hugh Mortimer, William Peverel and Henry of Blois, bishop of Winchester.63 In all these ways the settlement was very untidy, and this has to be kept in mind in considering how it was done. How it was done is the most difficult question of all. Some of the difficulties are of our making. First, the treaty of Winchester was agreed on 6 November 1153. Nobody knew that Stephen would be dead within the year. Arrangements which seem to us to be very short term were conceived and planned with no term at all in mind. So, when Stephen's men, for example, performed homage to the duke and perhaps sought some kind of confirmation of their position, they entered upon arrangements which could conceivably last for years. Secondly, territorial settlements cannot have been imposed on the great men of the land at Winchester or as an immediate consequence of the agreement, partly because neither the king nor the duke, nor both together, can have known the extent and detail of all the disputes, and partly because there was no governmental machinery capable of doing so. Only one such settlement was included in the terms of Winchester, that in favour of William of Blois, which was carefully specified, with distinctions drawn between his patrimony derived from Stephen, the lands of his wife, Isabel de Warenne, the acquisitions, the incrementum, which had come to him from Stephen, and the further gifts now made for the first time by Henry, the whole secured by the testimony of the great men of both parties and the overriding arbitration of the church. For the rest, compulsion derived solely from the homage which the duke's men now performed to the king, and the king's men to the duke. Within that it was up to each man to get the best deal he could. Now the authorities of king and duke were not on a par. True, Stephen was king and recognised as such by Henry, now become his vassal, but the Westminster charter, and probably the treaty too, accepted that the realm fell into two parts, the king's and the duke's. Moreover, Stephen was on the way out and had been for some time.64 It was the younger man and heir to whom men would look for security. Stephen had promised to govern consilio duds, and apparently did so; 611 Cf. Davis, King Stephen, p. 122. Henry's charter to Hugh Paynel of 1151-53, granting Les Moutiers Hubert and all the barony of his father in Normandy and England (EYC, vi, 96-97) is sufficient evidence of Hugh's affiliation to the Angevin cause. The main reason for thinking that Robert de Gant was of the other party seems to be that he was brother of Gilbert, earl of Lincoln. However, the opposing claims were in the half-blood, arising from the two marriages of William Paynel (d. 1145-47), and this was complicated by the marriage portion of Alice, daughter of the second marriage and wife to Robert de Gant. These were common sources of dispute. It is noteworthy that Henry II, having recognised Hugh before his accession, not only conferred part of the barony on Robert de Gant, but also divided the English holdings between Hugh and his brother Fulk: EYC, vi, pp. 18, 32-34, 96-97, 160-61. 62 Cronne, The Reign of Stephen, pp. 160-63. 63 Davis, King Stephen, p. 152; Warren, Henry II, pp. 59-61. 644 i J. Leedom, The English Settlement of 1153', History, 65 (1980), pp. 347-64.

284

Colonial England, 1066-1215

there were at least six courts, which they both attended, between Winchester on 6 November and Henry's departure for Normandy in early March 1154, so that there were opportunities for cooperation. Homages were taken at Oxford on 13 January, and a later meeting at Dunstable was taken up with the suppression of adulterine castles, in which Henry had found Stephen dilatory.65 But the concerted action which Davis saw in these meetings was limited, and on Henry's side strictly so. For example, no charter of King Stephen has yet been found attested by Duke Henry, yet after Winchester Henry was Stephen's man and indeed his heir. The simultaneous confirmations to which Davis attached importance are not quite what they seem.66 Practically all of them are consistent with two separate operations, even perhaps on two different occasions; none of them shares the same witness-list; even to share a witness is unusual. Typically a charter of the king was copied for, or by, the duke. On two occasions this was reversed, and then the king, in confirming, referred to the duke's charter.67 But Henry rarely repaid the compliment. He or his clerks were ready enough to copy the terms of confirmation by King Stephen for ecclesiastical beneficiaries, even verbatim.68 But where so, there would be no reference to King Stephen. Indeed, Henry made specific reference to Stephen only where it concerned the royal demesne or other regalian rights: a confirmation of the grant of Stanton Harcourt to Milicent, wife of Richard de Camville, made originally by Queen Adela and now confirmed petitione etprecepto of King Stephen; a confirmation of the grant by Stephen of the manor of Letcombe Regis to the monks of Cluny, which replaced a pension of a hundred marks originally granted by Henry I; and a confirmation of Stephen's grant of a die and mint to the cathedral church of Lichfield.69 That is all. None of Henry's charters confirming the lands title, or rights of tenants-in-chief, either before or after his accession, acknowledges Stephen in anyway.70 Except for William of Blois, Henry was simply not prepared 65

Henry of Huntingdon, Histmia Anglorum, pp. 289-90. For Davis's opinions, see King Stephen, pp. 121,122: The only way in which individuals could make certain of their lands which they claimed as their own, was by making treaties or marriage alliances with their rivals . . . or by getting their lands confirmed by Stephen and Duke Henry simultaneously.' 'Now that the king and duke were acting in concert. . .' He relied on RRAN, iii, nos 94-97 (all of which are brief summaries), 126-27, 215 and 239f, 457-58, 583-54, 874-75. 67 Ibid., iii, nos 126,127,130,131. 68 Ibid., nos 94-95, 96-97, 874-75, and, for a verbatim copy for Meaux Abbey, nos 583-84. 69 Ibid., nos 140,204-6,457-58. In the case of the charters for Cluny, Stephen's charter is dated 1136, seventeen years before Henry's confirmation. 70 A possible exception to this was adduced by Hugh de Gournay in the curia regis in 1212: Curia Regis Rotts, vi, pp. 272-73. This purported to be a grant of the manors of Wendover and Hough ton Regis by Henry II to Milicent, widow of Hugh de Gournay (the elder), which included 'all the new land which King Stephen gave to Hugh de Gournay ad sue hereditatis augmentum, namely Wendover and Houghton'. Henry also conceded 'whatever her husband bought or acquired in any way, as his charter testifies and confirms to her'. The open-ended nature of this second concession, the unusual and somewhat irregular wording of the charter, and the fact that Hugh's opponent in the action produced a perfectly standard charter of Henry II and a confirmation of King Richard, conveying Wendover to the ancestor of the opposing party, Faramus de Boulogne, suggest that the charter produced by Hugh de Gournay was spurious. Even if genuine, however, it should be noted 66

The Treaty of Winchester, 1153

285

to warrant the concessions of his predecessor. He was ready enough now and then to provide warranty for others, even spelling it out to include compensation, escambium, for the bishop of Lincoln, Joscelin of Louvain, and William Mauduit.71 But to Stephen's men he would not give an inch beyond the agreements of Winchester unless or until they made their particular peace with him. He had already acquired an arrogant majesty like that which had occasioned opposition to his mother in 1141. He spoke of land pertaining to the crown of the king: he did not mean Stephen.72 Indeed, in his confirmation of the foundation charter of Biddlesden Abbey (1153-April 1154), he granted quittance of suit of shire and hundred and other privileges, along with 'all customs pertaining to my crown'.73 Henry knew that Henry was boss. There is very little evidence, therefore, of joint action of the two protagonists in their meetings following the Winchester agreement. The Westminster charter stands out with its long list of witnesses of leading figures from each side. It was unique in this, and almost alone in revealing cross-witnessing of any kind. Some of Henry's men attested Stephen's confirmation to the fitz Harding house of St Augustine, Bristol,74 and Richard de Lucy figures among the witnesses to Henry's confirmation of Stephen's grant to Cluny,75 but Henry's confirmation for Meaux, which was a verbatim copy of Stephen's, had its own witnesses, Henry's men and none of Stephen's.76 We are led, therefore, in the end to a very simple model of resettlement. Take as an example Robert de Beaumont, earl of Leicester. Robert joined the Angevin cause in the spring of 1153 and quickly became a major figure in the duke's counsels. He attested the Westminster promulgation of the general settlement in December.77 His new allegiance is reflected in three charters. First, from Henry as duke he obtained a confirmation of his hereditary lands in England and Normandy and of his wife's inheritance from William fitz Osbern; Henry also conceded the recent exchanges of land which Robert had made with Roger, earl of Warwick. Then, like Eustace fitz John, the earl took steps to ensure the succession to his lands by obtaining an identical charter in favour of his son. Finally, Henry confirmed the arrangements, apparently in the very same terms, after his accession.78 That, or continued

that the reference to Stephen occurs yet again in a charter concerning the royal demesne, for both manors were terrae regis. 71 RRAN, iii, nos 306, 491, 568, 582. 72 Ibid., no. 90. Cf. The empress's reference to pleas pertaining to her crown in 1141, ibid., no. 274. 73 Ibid., no. 104. 74 Ibid., no. 127. William de Chesney; Richard de Camville. 75 Ibid., no. 206. 76 Ibid., nos 583, 584. 77 Crouch, Beaumont Twins, pp. 86-89. 78 For the two ducal charters, see RRAN, iii, nos 438, 439. The editors date the son's charter, which survives in the original, earlier than the father's, on the ground that the second gives Henry the title of duke of Aquitaine, which he adopted in June 1153. But the father's charter survives only in a copy and it may be that the title is misleading. On the other hand, as Professor King has kindly

286

Colonial England, 1066-1215

something closely akin to it, was probably the common pattern. It is repeated fulsomely in the fitz Harding charters, an example which is all the more telling if they are less than authentic. It reappears, with variations, in the cases of Geoffrey de Mandeville, Roger, earl of Hereford, and Eustace fitz John. What then happened when claims were disputed? It is well known that some were settled by marriage alliances, some by division or subenfeoffment of one of the parties by the other, some by the creation of life tenures for the weaker party.79 The solutions were miscellaneous and seem entirely ad hoc. In some cases the parties and their descendants wrangled on into the next century, leaving scattered traces of their quarrels in the plea rolls. Amongst these there are perhaps a dozen or so references to decisions or judgments by Henry II. It is a hopelessly inadequate number. Very few of these references can be dated, they cannot even be arranged sequentially, and they include much hearsay. Consider the dispute between Ruald, constable of Richmond, and Richard de Rollos which was recorded much later in an action in the curia regis in 1208.80 The facts are well known and the case has been much used by legal and other historians. In 1135 the constable's fee of the honour of Richmond was divided equally between Richard de Rollos and Ruald the Constable, possibly as a result of a division between co-heiresses late in the reign of Henry I.81 De Rollos gave service to the empress (he had considerable Norman lands) and was deprived by King Stephen, who gave his portion to Ruald. Some time after 1154, so de Rollos' descendant claimed, King Henry brought the parties to an agreement which divided the fee once again. It looks like a copy-book example of a compromise agreed in the royal court. But in 1208 Ruald's descendant would have none of it. He mentioned no agreement, he claimed the whole fee as his right, he argued that Henry II had disseised his grandfather unjustly and without judgment, and he paid £100 for an inquisition of lawful men of the neighbourhood. The inquisition took his view- Henry II had indeed disseised Ruald the grandfather unlawfully and withoutjudgment. So who was right? What had happened? Can we even accept the verdict of the inquisition? And how would the case look if we had the argument of only one party or the other? For none of the pleadings of 1208 fits the facts. De Rollos' descendant argued that Henry II had left Ruald with nothing more than a life-interest in his half of the constable's fee. Yet his son, Alan, was confirmed in his constabulary and in all the lands of his father by Conan, duke of Brittany and earl of Richmond, in 1158,82 and it is plain that Alan and his son in turn retained his half of the fee until the whole was reunited after 1204. continued

pointed out to me, the grant to the son may have come first as a particular favour during the negotiations between Henry and Robert of Leicester in 1153. Henry's charter as king survives only in summary form, but it is full enough to establish that it followed the ducal charters very closely. See Harcourt, His Grace the Steward, p. 60. 79 Davis, King Stephen, pp. 120-22; Warren, Henry II, p. 333. 80 Curia Regis Rolls, v, pp. 147-48. 81 EYC, v, pp. 85-88, 95-99. 82 Ibid., iv, pp. 48-49.

The Treaty of Winchester, 1153

287

Procedure is equally in doubt. The Rollos' claimant argued that there had been a concord in King Henry's court ('et postea venit rex Henricus et eos concordavit'). It is certain, on the other hand, that Alan, son of Ruald, was confirmed in his father's land by Duke Conan at Fougeres. If his son's argument in 1208 was right, that Henry II had acted without judgment, the matter may never have come to the curia regis. Procedure often seems hazy. It is sometimes at its clearest outside royal jurisdiction. Robert de Stuteville and Roger de Mowbray settled their differences in a conventio, probably before 1157, but it was a private agreement which never came to the king's court.83 An action between Suspirus de Bayeux and Osbert fitz Nigel, which revealed that an intruder had been given a life tenancy, derived from a settlement concluded in the court of the honour of Richmond at Boston.84 Now and then an intrusion was corrected in the court of King Henry,85 but he could equally well act by simple gift, rightly or wrongly,86 or even totally informally. Hugh de Gournay, according to a jury of 1212, was allowed to keep Wendover sine dono and was then deprived on Henry's return from Toulouse in 1159.87 Descendants of these disputants might produce Henry's charters in court: none produced a writ, and writs are not mentioned by those who recalled these actions a generation or more later. We can now bring matters to a head. First, there was no principle at work in the treaty of Winchester and its consequences which depended on the distinction between right and seisin, no '1153 issue', as it has been termed.88 It is not sustained by the evidence.89 That difficulty apart, the agreement between King Stephen and Duke Henry was no model for it. The treaty did not award seisin to the one and title to the other. Stephen recognised Henry as his heir jure hereditario, promising to sustain him as son and heir in all things and against all men. How could he do that, what was the point of doing so, unless he himself had title? How could Henry's tide possibly be based on Stephen's mere tenure? You cannot make a man your lawful heir without appearing as his lawful ancestor.90 Secondly, inheritance in 1153 meant something very precise. To assert that the treaty of Winchester established an irreversible 'hereditary principle' is to miss the point.91 By asserting inheritance the treaty questioned acquisitions, including acquisitions to which the intruder had a hereditary claim or the ten83

Chronica Rogeri deHouedene, ed. W. Stubbs, Rolls Series (4 vols, London, 1868-71), iv, pp. 118; Mowbiay Charters, no. 386; EYC, ix, pp. 5, 100, 116-17. 84 Curia Regis Rolls, v, pp. 181-82; vi, pp. 17-18. 85 Ibid., viii, pp. 18-20 (Badele v. Tattershall); even then the parties were not reconciled until the conclusion of a subsequent marriage alliance. 86 Rotuli CuriaeRegis, i, p. 93 (Raimes v. Welles}. 87 Curia Regis Rolls, vi, pp. 272-73. 88 Palmer, 'Origins of Property', p. lln. 89 See the comments of Professor Biancalana and Professor Hyams mentioned above, n. 11. It should be noted that Davis largely accepted Professor Palmer's case (Davis, King Stephen, p. 153), but without considering the arguments of Biancalana and Hyams. 90 1 leave on one side here the more general issues surrounding the concept of the lawful antecessor. These are among the matters to be discussed in a forthcoming work by Dr G.S. Garnett. 91 Davis, KingStephen, p. 121.

288

Colonial England, 1066-1215

ant a lawful title. If there was a principle, it was that of Ockham's razor, Ockham's necessitas, in this case a compound of the rules of succession, the procedures for conveyancing and more than a dash of political convenience. At one and the same time it confirmed and restricted tide. Here perhaps there were links with the legal developments of Henry II. The treaty was not confined to lay fee. When Stephen succeeded to the throne, the great wave of Cistercian expansion was getting under way, accompanied by the foundations of the Augustinian and Premonstratensian canons. Nearly 120 monastic houses were founded during the reign. All their endowments were acquisitions. All benefactions to the older monastic foundations or to the secular church subsequent to 1135 were likewise acquisitions. The same applied to the actions of churchmen as lords. All enfeoffments or leases made since 1135 were acquisitions. All intrusions into church property were acquisitions. All these varied rights of property and tenure, all actions, whether lawful or blatandy invasive, might be subject to review. The church was up to its neck in the settlement and its execution, and this in two distinct ways. First, it had suffered great loss during the Anarchy: castles, land, services, goods, tenserie. Secondly, it had made enormous gains: grants in free alms, endowments of new foundations, to salve the conscience of the robber baron and ensure salvation amidst uncertain fortunes. Perhaps, on balance, the church was the chief beneficiary of the Anarchy of Stephen. But it could not recover its losses without risking its gains. It chose to risk its gains. It was no choice: it had to do so. The church's losses have been discussed by others.92 That its gains could be subject to review is demonstrated by the treaty of Winchester, for Duke Henry confirmed King Stephen's foundation of Faversham, which had been well endowed with lands of the royal demesne in Faversham and elsewhere, along with Stephen's benefactions to other churches. This was done consilio sancte ecclesie. More than thirty years later, in a notification of 1186 or 1188, the whole question was reviewed with reference to acquisitions of the monks of Stanley in Faversham. In the time of the war of the usurper King Stephen, so the notification runs, much was lost and alienated from the demesnes of the realm, both knights' fees and church alms (both, note, now viewed equally as purprestures); in particular, the abbey of Thame had come to hold Worth in Faringdon by gift of King Henry's enemies. On his accession they had surrendered the land to the king. But it had been conferred upon a religious house; the Empress Matilda, Gilbert, abbot of Citeaux, and other abbots of the order pressed their petitions, and Henry gave the land to the abbey of Stanley, the empress's own foundation.93 Unease and uncertainty is again manifest at Kirkstall, originally 92

Stenton, First Century, pp. 244-45; E. King, 'The Anarchy of King Stephen's Reign', Transactions of the Royal Historical Society, fifth series, 34 (1984), pp. 133-42; Biancalana, 'For Want of Justice', pp. 471-73; M. Cheney, 'Inalienability in Mid-Twelfth Century England: Enforcement and Consequences', Monumentaluris Canonici, series C, subsidia 7 (1985), pp. 467-78, esp. pp. 469-71. 93 Actes de Henri II, ii, no. 682; Ancient Charters, ed. Round, no. 52. As Round pointed out, the notification refers to earlier transactions which cannot have occurred later than 1167. In all likeli-

The Treaty of Winchester, 1153

289

founded by Henry de Lacy at Barnoldswick in 1147 and transferred to its present site at Kirkstall in 1152. The tradition of the house was that Barnoldswick was a tenancy which Lacy held of Hugh Bigod, earl of Norfolk, and that the rent due from the land was in dispute. This is very likely.94 In 1153-54 Abbot Alexander had to put matters in order. As a result Henry de Lacy issued a charter confirming Barnoldswick and Kirkstall along with other acquisitions in the neighbourhood of Kirkstall.95 Hugh Bigod, earl of Norfolk, at the request of Henry, wrote to Roger, archbishop of York, notifying him that he had granted Barnoldswick to the monks at the request of Henry de Lacy.96 Henry in turn wrote to Henry II informing him that Hugh's grant had been made at his request and humbly imploring the king to issue his own confirmation.97 This King Henry duly did.98 There are four surviving documents in all, with some probably missing - quite a deal of fuss.99 The king's charter is of special importance. In addition to confirming the grants of Bigod and de Lacy, he ordered that the monks 'were not to be disseised by any writ of land for which they had charter and chirographs; nor were they to be impleaded therein except coram me. That emphasises the protection which documentary confirmations gave. It envisages that writs of seisin might be used to challenge tenure. It shows that, in the king's view, royal charters brought issues into the royal court. Henry's confirmation cannot be later than 1162100 and should probably be dated to 1155. Some of the elements of his later system of justice were already in place. continued

hood they followed soon on Henry's accession. For comment on the document, see E. King, 'King Stephen and the Anglo-Norman Aristocracy', History, 59 (1974), p. 181; M. Chibnall, The Empress Matilda (London, 1991), p. 135. 94 The Couches Book of the Cistercian Abbey of Kirkstatt, ed. W.T. Lancaster and W. P. Baildon, Thoresby Society, 8 (1896-1904), pp. ix-x. For the text of the Fundacio abbatiae de Kyrkstall, ed. E.K. Clark, see Miscellanea, Thoresby Society, 4 (1895), pp. 169-208. This refers simply to the failure of de Lacy to pay the annual rent of five marks and a hawk (at p. 174). However, this probably reflects a more serious dispute. In 1086 it was recorded that Berengar de Tosny had held twelve carucates of land in Barnoldswick which were now in castlery of Roger of Poitou (i.e. Clitheroe): DB, i, 332a. The Tosny interest descended to Bigod, EYC, i, pp. 466-67,507-8. Clitheroe was held by the Lacys from the time of William Rufus. Hence, the non-payment of rent by Henry de Lacy was probably intended as a denial of any Bigod title. According to the Fundacio, Hugh Bigod successfully claimed Barnoldswick in demesne in the king's court. He then reinstated the monks at the ancient rent. All this occurred sometime after 1152 and possibly still in Stephen's time. Henry II subsequently persuaded the earl to relax the rent and confirmed the earl's grant in free alms, Fundacio, p. 180. This last stage is confirmed by the documents, but, even so, the Fundacio makes no mention of the intervention of Henry de Lacy, and the surviving charters and letters make no reference to the rent. 95 CoucherBook of Kirkstall, pp. 50-51. 96 £KCi,pp.507-8. 97 Ibid., i, p. 508. 98 Ibid., iii, pp. 152-53. 99 There is no surviving charter for the original establishment at Barnoldswick, although a perambulation of the bounds issued by Henry de Lacy may have served the purpose, EYC, i, pp. 506-7. 100 It is attested by Thomas Becket as chancellor.

290

Colonial England, 1066-1215

Now it is true that the status of free alms gave the church a protection to title beyond anything which the laity enjoyed. Butat Winchester the status of free alms was not immediately in question. What was, was tide derived from endowments of the past nineteen years, whether free alms or not. It has long been recognised that a very large number of the acta of Henry II come from the years before 1162.101 A very high proportion of these were for ecclesiastical beneficiaries. By 1162 there was no cathedral church and few monasteries, ancient or new, which had not acquired King Henry's confirmation. The king's confirmation implied and involved the king'sjurisdiction. It was from this massive output that legal innovation emerged: the protection of seisin, and, in the reinforcement of the action of tolt, the supervision of lower by higher jurisdiction, especially that of the king. In this royal authority mingled with canonical views of the churchman's duty to preserve, and if necessary recover, the church's rights and property. It was an explosive mixture which contributed directly to the king's quarrel with Becket.102 At Winchester in 1153 all that had yet to happen. Whether the participants in the negotiations foresaw that the terms of the treaty would lead to such a demand for royal confirmations may be doubted. They certainly cannot have foreseen that it would begin within the year. And nothing in the past could have prepared them for its unprecedented scale.103 They designed the settlement in other, older terms, of homage and its consequences, of a lord's capacity to nominate an heir, above all of the difference between inheritance and acquisition. These were securely embedded in the legal thought which they inherited themselves from the reign of Henry I.104 It was to that security that they turned in 1153. The invention necessary to cope with the problems they created for themselves was for the future.

101 T.A.M. Bishop, Scriptures regis (Oxford, 1961), pp. 30-31, suggests 40 per cent, which Hyams ('Warranty', 476-77n.) regards as on the low side for the English acta. Becket's attestation as chancellor determines the date 1162. For further comment, see below, n. 103. 102 Cheney, 'Inalienability'; idem The Litigation between John Marshal and Archbishop Thomas Becket in 1164', in J.A. Guy and H.G. Beale, eds, Law and Social Change in British History (London, 1984), pp. 9-26; idem, 'A Decree of King Henry II on Defect of Justice', in D. Greenway et al., eds, Tradition and Change: Essays in Honour ofMarjotie Chibnall (Cambridge, 1985), pp. 183-93. 103 professor Hyams notes quite correctly that it was 'normal for kings to make a disproportionate number of grants at the start of the reign' ('Warranty', p. 477n.). This is true in general, but it is important in any calculation to distinguish grants from confirmations. The extensive endowments of Stephen's reign, coupled with the terms of the treaty, meant that the situation in 1154 was very different from that in 1100 and 1135. A large number of confirmations were also issued in 1189-90 and again in 1199-1200, but in each case for reasons which did not apply in 1154.1 suspect that for 1154 the estimate of the percentage of the early grants against the total acta is less significant than the comparison of English and continental acta. A high proportion of the early acts concern England alone, as indeed did the treaty of Winchester. 104 For the nomination of an heir, admittedly in circumstances where the ancestor is deserted by his own son or other relative, see Leges Hcnriti Primi, c. 88,15, ed. LJ. Downer (Oxford, 1972), pp. 274-76. For a possible example from Stephen's reign, involving Waleran of Meulan and William de Beauchamp, see Davis, King Stephen, 11 On. For inheritance and acquisition, see Leges Henrid Primi, ch. 70, 21 (ed. Downer, p. 224), and for background, see above, pp. 120-26.

15 Magna Carta, 1215-1217: The Legal and Social Context

Does the prolonged history of Magna Carta reflect something unique in the document itself and in the circumstances from which it sprang? Was it all that important? How should we assess its importance in strictly contemporary terms. These questions still occupy me. They seem to me to be essential to any serious consideration of the Charter both in general and in detail. Many of the answers lie in its subsequent history from 1217 onwards. They lie outside my present themes. Faith Thompson and many others have written of them; I have commented on them myself.1 Much more remains to be done, especially on the minority of Henry III, on which we can look forward to an important work from David Carpenter. Yet this subsequent history raises a question very close to my present purpose: was the later development of the Charter, its rapid emergence as a founding statute for example, foreseen or in anyway intended in 1215? Or in 1217? What were they after, the men of 1215, when they challenged the king at Runnymede? How did they think? How should we discuss and assess them? In attempting answers, which can only be sketchy within the limits of a short essay, I shall work from the Charter and the crisis of 1215 outward to other themes. This method underlies my differences with some recent work in which the Charter has been interpreted largely in the light of other arguments concerned with other phenomena. After being involved with it for so long, I am not very ready to make it fit anything. There is to hand a simple and straightforward explanation of the context of 1215 and of the objectives embodied in the Great Charter. It runs something like this. The Angevin kings were powerful and wilful monarchs directing an efficient administrative machine. As lords of the Plantagenet lands on the Continent they faced increasing difficulties, culminating in wars with the kings of France. They had to contend with increasing expenditure in a world of rising costs and prices. They resorted to severe financial measures, overstressed the administrative and legal machinery, relied increasingly on a select body of 1

Faith Thompson, The First Century of Magna Carta: Why it Persisted as a Document (Minneapolis, MIN, 1925); idem, Magna Carta (Minneapolis, MIN, 1948);J.G.A. Pocock, The Ancient Constitution and the Feudal Law (Cambridge, 1957; 1987); J.C. Holt, The Origins of the Constitutional Tradition in England', in Magna Carta and Medieval Government (London, 1985), pp. 1-22; idem, The Ancient Constitution in Medieval England', in Ellis Sandoz, ed., The Roots of Liberty: Magna Carta, Ancient Constitution,, and the Anglo-American Tradition of the Rule of Law, University of Missouri (Columbia, 1992), pp. 22-56.

292

Colonial England, 1066-1215

officials, at the worst on aliens, distorted the customary relationships between king and vassals, at worst resorting to deprivation, disinheritance and imprisonment without due judgment. As a result they were regarded, and in the end denounced, as tyrants. All this was especially true of Kingjohn, who, as a result, had to bear the indignity of Runnymede and the responsibility for provoking a fundamental change in the balance between royal power and the community. There are many possible variants to that: the personal qualities of John; the effects of his quarrel with the church; the extent to which he destroyed the work of his father, Henry II, who until the death of Becket enjoyed a better reputation than that which condemned him with his sons; the inevitability or otherwise of the loss of Normandy; and so on. But most of us would accept some of this summary, and some of us would accept most of it. It represents a broad stream of historical opinion flowing down through Stubbs and Powicke to Painter and to most of those who have written about John or Magna Carta in the last thirty-five years. It springs from the contemporary evidence: from the Charter itself, from the record of Angevin government revealed in the pipe rolls and elsewhere, from what is left to us of the views of those who lived at the time. It is expressed perhaps in its most sophisticated form in a justly celebrated essay by Sir Richard Southern, 'England's First Entry into Europe', which counted the cost of the Norman and Angevin foothold in Europe and contrasted the contemporary hatred of Angevin tyranny with the love of France and all things French. His story culminated with Gerald of Wales and we may agree with Sir Richard that Gerald is not a very reliable witness; nevertheless he left to us a polemical view that France was the home of liberty and England the seat of tyrants; he even composed a poem along these lines when Prince Louis of France invaded England in an attempt to claim the throne in 1216. Sir Richard, writing in 1966, was inclined to agree with him.2 Now Gerald of Wales was a disappointed clergyman, with the background of a Welsh marcher, and he leads me, for one, to seek some more objective basis for the assessment of the events of 1215. If Magna Carta was a response to tyranny and tyranny was endemic in the Angevin race of kings, as Gerald held, or in the situation they and their subjects encountered as Sir Richard, and indeed others, hold, then two points immediately arise. First, in the objectives, mechanisms and methods of government there was not a great deal of difference between the Angevins and the Capetians, between the damned and the saved. There was the same exploitation of the financial resources which stemmed from the superiority of the crown, the same development of demesne rights, the same search for administrative efficiency. Even the more extreme measures of John's autocracy, the conditional tenure of castles, the charters of fealty insuring loyalty on pain of disinheritance, the heavy financial penalties and pledges used as guarantees of good behaviour, found a parallel in the measures 2 R.W. Southern, Medieval Humanism and Other Studies (Oxford, 1970), pp. 151-57, especially 155-56. For further comment on this aspect of Gerald, see Robert Bartlett, Gerald of Wales (Oxford, 1982), especially pp. 91-100, 222-25.

Magna Carte, 1215-1217

293

used by King Philip Augustus of France to tie down the doubtful loyalties of suspect vassals.3 Indeed an objective assessor, at the other end of the spectrum from Gerald of Wales, might well conclude that John suffered the penalty of a tyrant because he failed: Philip Augustus did not because he succeeded. In France the bird came home to roost a century later, in 1315, after a similar period of monarchical failure and the accompanying political and financial stress. Secondly, tyranny meant something quite specific: the tyrant stood to a king as self-will did to law. Now if every charter of liberties is to be read as a response to tyranny it must be admitted that there was a great deal of tyranny around, so much indeed as to undermine this whole approach. For charters of liberties or similar statements of law were spawned throughout western Europe from the late twelfth to the early fourteenth century. There are several immediate analogues to Magna Carta, all coming from the same decade, close enough in time for us to dismiss any effects resulting from the passage of time. First, from 1205, there is the so-called unknown charter of Catalonia discovered and very thoroughly discussed by Professor Bisson.4 It was concerned with restricting the financial demands of Peter of Aragon and with the provision of just government by the counsel of the magnates and the administration of local knights. Like John in 1215, King Peter was to swear to observe its terms. Secondly, there is a charter of Thomas, count of Perche, for his vassals of the lordship of Belleme. This comes from February 1215.5 In it he promises that he will only make levies on his knights for his own first knighting, for his own first ransom, for the knighting of his first-born son and for the first marriage of his daughter: four occasions for the feudal aid in place of Magna Carta's three, but a close analogy all the same. Thirdly, from 1212, there is the so-called Statute of Pamiers, which is not strictly a charter of liberties, but which, like Magna Carta, sought to establish customs, consuetudines, in the lands conquered by the Albigensian crusaders in southern France.6 Some of these are concerned with the very same topics and problems as the English charters.7 In relation to these documents, wherein lay Magna Carta's special quality? They suggest surely that similar administrative pressures tend to produce similar results. But Magna Carta had a different future from the rest, indeed from all 3 J.C. Holt, The Barons and the Great Charter', Magna Carta and Medieval Government, pp. 18182; and more generallyJ.W. Baldwin, The Government of Philip Augustus (Berkeley, CA, 1986), especially pp. 266-69. 4 T.N. Bisson, 'An "Unknown Charter" for Catalonia (AD 1205)', Album Elemer Mdlyusz: Studies Presented to the International Commission for the History of Representative and Parliamentary Institutions, 56 (Brussels, 1976), pp. 63-76. For further comment see idem, Conservation of Coinage (Oxford, 1979), pp. 88-90. 5 Archives de 1'Orne, H 2164. 6 Pierre Timbal, Un con/lit d'annexion au moyen age: I'application de la coutume de Paris au pays d'Albigeois (Paris, 1949), text of the Statute at pp. 177-84. 7 See for example Statute of Pamiers, cc. 12 (alienation of land in alms), 17 (feudal service), 26 (levying of aids), 28 (process and pledging), 33 (pledging for debt).

294

Colonial England, 1066-1215

the rest whether of this decade or of others. Perhaps the documents themselves give some hint as to why. For Magna Carta has two features which lead us into two quite different legal contexts. First, it is exceptional in its concern for the freeman, liber homo. This is a complex matter to which I shall return. Secondly, unlike these other charters of liberties, it was granted in perpetuity, in perpetuum. I first want to dwell on these two words. It was no mistake. The words occur in cap. 1 of Magna Carta in the grant of freedom of election to the church. They are then repeated at the end of cap. 1 as a preface to all subsequent concessions - 'We have also granted to all the free men of our realm for ourselves and our heirs in perpetuity'. They are repeated yet again in the final clause of corroboration - 'the men of our realm shall have and hold all the aforesaid liberties, rights and concessions, well and peacefully, freely and quietly, fully and completely, for them and their heirs of us and our heirs in all things and places in perpetuity'. And the Charter of 1215 was no aberration. In perpetuum recurs in exactly the same context and with the same intent in the reissues of the Charter of 1216, 1217 and 1225. In 1217 it was transposed into the Charter of the Forest and was repeated in all subsequent reissues. The evidence allows no other conclusion than that the liberties conceded between 1215 and 1217 were to be enjoyed for evermore. It is difficult to overestimate the importance of these two words in perpetuum. From the start they established the Charter as a fundamental grant which would necessarily require interpretation as time passed and circumstances changed. That lay in the future and I have discussed it elsewhere.8 More immediately, in terms of political or legal theory, they symbolise an extraordinary step, almost a leap into the dark. True, in certain contexts men would encounter perpetual liberties - occasionally in borough charters or grants of markets and fairs, almost always in the much rarer charters of disafforestation which begin to appear in the ten years or so before 1215. But such grants were made to or for local communities, often acting through the local courts. They were analogous to grants in free and perpetual alms. Magna Carta, in contrast, conveyed perpetual liberties to the whole community of the realm. There can be no real doubt as to how this came about. Cap. 1 of the Charter of 1215 laid down first, by royal grant in perpetuity, that the English church should be free and enjoy its rights undiminished and its liberties unimpaired; the king also confirmed the grant of freedom of election 'which we shall observe and wish our heirs to observe in good faith in perpetuity'; and the origin of this was the grant of freedom of election which KingJohn conceded on 1 November 1214 - liberae sint in perpetuum electiones. On the ecclesiastical side such intentions were familiar enough. Grants were made to the church in perpetuity and had been for generations. So the clergy were now to enjoy freedom of election in free and perpetual alms, as it were. Now, like the modern sociologist, the medieval thinker was very ready to treat one mode or realm of thought as the

* See above, n. 1.

Magna Carta, 1215-1217

295

reflection of another. On the political margin of this, in 1215, men's minds moved quite easily from the liberty of the church to the liberty of the realm. Indeed, if the Barnwell chronicle has it right, the original conjuratio was 'to sustain the house of the Lord and stand fast for the liberty of the church and the realm'.9 So why should not the realm enjoy its liberties on the same terms as the church? That may have seemed an easy question; the easier the more rhetorical. But one answer to it was politically radical. The church could be viewed as an undying corporate body, ecclesia anglicana. It might not sue as such in the courts, but it fragmented into smaller corporate bodies which could and did, and could hold their lands and rights in perpetuity. But the realm? How was the realm to be construed if it was to have and hold such concessions? And how did the realm sue to protect its perpetual liberties? Simply by petitioning the king, as John and his protector, Innocent III, maintained? Or by establishing a court empowered to distrain him, as the Charter sought to do? These were fundamental practical questions which underlay those two words in perpetuum. They were part of the original design. Someone was prepared to face them. To answer them would not be easy. We are unlikely ever to know who it was: Stephen Langton, perhaps, but he was out of England when the Charters of 1216 and 1217 were drafted; much more probably one or several of the group of radical young churchmen who linked Langton's household to the chapter of St Paul's and the city of London: Simon Langton, the archbishop's brother, Gervase of Howbridge,. chancellor of St Paul's, and Master Elias of Dereham, the archbishop's steward and future designer of Salisbury cathedral;10 there were also the baronial chaplains who conducted the baronial case at the curia, John of Ferriby and John fitz Osbert. Of all these Elias of Dereham is perhaps the most likely, involved as he was with the baronial leaders in 1213 and subsequently with the execution of the Charter. But the laity should not be excluded. Robert fitz Walter was to take the title 'marshal of the Army of God and Holy Church', and four men attested the grant of freedom of election in 1214 who were later members of the court of Twenty-Five: Geoffrey de Mandeville, Saer de Quency, earl of Winchester, William of Huntingfield and Robert de Vere. So we are left with a wide choice and in the end with guesswork. It was natural enough that church and realm should intermingle. In the Statute of Pamiers they did so much more obviously and pervasively because this had a primary concern in eliminating heresy and reestablishing the church in the newly conquered lands in southern France. But in England the linkages were very specific to ecclesiastical jurisdiction. In 1213 King John had surrendered his kingdom to the papacy and become a papal vassal; this too was an

9

Memorial* fratris Walteri de Coventria, ed. W. Stubbs, Rolls Series (2 vols, London, 1872-3), ii, 217-18; J.C. Holt, Magn,a Carta (Cambridge, 1965), pp. 135-36. See also Bartlett, Gerald of Wales, pp. 198-200. 10 Holt, Magna Carta, pp. 189-90; idem, 'Origins of the Constitutional Tradition', pp. 16-17.

296

Colonial England, 1066-1215

obligation in perpetuum.11 In January 1215 both the king and his baronial opponents took their cases to the papal curia. From this point on political negotiation was also legal process, in two parallel actions in the courts of king and pope. The resulting clumsiness was accentuated by a turnaround from Rome of at least twelve weeks. Innocent III pronounced judgment in a form of peace announced in letters of 19 March. It was received in England at the beginning of May. It left much for the king and the bishops to settle, but it was manifestly on the king's side; the barons were to cease all conspiracies and perform their due services.12 Thereafter any local agreement would have to be brought within the terms of the papal/orma by presenting itself as a properly agreed compromise between the two parties. This need and the resulting terminology pervade the documents of 1215. The Articles of the Barons are articuli pads; they embody conventiones (and we may note that these terms refer not to the end of a war, though war there had been, but to a legal settlement); they end with a forma securitatis', the archbishop and bishops are to pass judgment on certain disseisins appellatione remota\ along with the nuncio Pandulf they are to give security that there will be no appeal to Rome. The Charter also is a pax:, it is guaranteed by letters testimonial of the prelates and the nuncio; comparing the Articles and the Charter it is plain that in this and other matters they had decided very carefully just what they would and would not do to secure and execute the settlement.13 It is also clear that men with a ready knowledge of the canon law and procedure of the ecclesiastical courts had influenced the negotiations throughout. They were essential. Yet this was only part of the legal context. There was also a secular mode and here I return to our three companion documents. The church and the canon law aimed at the uniform. The secular context, by contrast, was variegated. Hence to each charter its own setting. Each one was focused on a particular pattern of government and jurisdiction. The Catalan charter, for example, was concerned among other matters with currency mutation and the bovatge, and the Statute of Pamiers aimed at introducing the custom of the Parisis into southern France. Each used its own particular language. Thomas, count of Perche, referred to his levies as the tallage on fiefs, tallia defeodis, words which would have implied arbitrary illegality in England, and he implied that those which were levied lawfully were levied feudally, feodaliter.14 (Somebody should have told him: he was being tyrannised by a construct.) And each charter addressed its own particular beneficiaries. Thomas, count of Perche, made his grant to his knights of the castlery of Belleme. Peter of Aragon's charter embraced all the clergy and their men along with the magnates and 'other knights and other honest (probi) men 11

For the text see W. Stubbs, Select Charters, 9th edn, ed. H.W.C. Davis (Oxford, 1921), pp. 27981. The witnesses include Saer de Quency. 12 Holt, Magna Carta, pp. 293-95. 13 The treatment of cap. 25 of the Articles is of particular importance. See ibid., pp. 192-94. 14 Treter has tallias nee a militum feodis nee ab eorum hominibus tallias possumus feodaliter extorquere.'

Magna Carta, 1225-1217

297

of Catalonia'. The Statute of Pamiers distinguished sharply between Catholic and heretic within the native population and then between them both and the northern French crusaders; the northern barons even had their own special privilege dealing with succession, feudal incidents, feudal jurisdiction and the like.15 Magna Carta too dealt with aristocratic privilege. But in 1215 Magna Carta was granted to the freeman, simply and without qualification, in sole, not at the tail-end of some social enumeration; it was only in the proems to the charters of 1225 that the familiar social ranking first appeared.16 The liber homo distinguished Magna Carta from all other charters of liberties.17 It is important to approach these categories in the right way. The men who drafted these phrases were not thinking, as we might do, of how their societies were divided socially. In any case, their notion of social status was not ours. They were concerned rather with jurisdiction, with legal and other rights, and with procedures, legal, administrative and fiscal. Status was linked to, and often arose from, the way in which these operated. So it is only very indirectly that Magna Carta is telling us something about the rights and position of the freeman. Instead it is dealing with a mass of administrative and legal problems, some of which had to be related to the freeman if they were to be properly defined, all of which had to be granted to the freeman in the enveloping phrases of the dispositive and corroborative clauses of cc. 1 and 63 if they were to be properly conveyed. There was no other way of doing it. This was not the product of 'anti-feudal' or 'anti-magnate' instincts in those who drafted the Charter. It has long been recognised that Magna Carta is Janusfaced in its treatment of royal justice. On the one hand, it accepted practically the whole of the work of Henry II as it had been developed under his sons, the writ praedpe quod reddat being the sole exception. It even demanded more justice, petty assizes four times a year, sessions of the Common Pleas in some fixed place and so on. On the other hand, it challenged royal justice. Arbitrary action by the king was set against lawful judgment of peers; the sale of justice was forbidden (yet another derivation from canon law); restoration was promised for unlawful disseisins and unjust agreements and amercements. The usual explanation of this contrast has been that King John abused the good work of his father, in the words of McKechnie: Far from using the perfected machinery of Exchequer, Curia, and local administration in the interests of good government, John valued them merely as instruments of extortion and outrage - as minister to his lust and greed.18

But there is another explanation. The Henrician system applied almost exclusively to what we may roughly call the appellate function of the royal court- 'roughly' 15

Timbal, Un con/lit d'annexion, p. 184. It should be noted that this left an inconsistency in drafting since the original wording of 1215 remained in cap. 1. 17 For some comparisons see Holt, Magna Carta, pp. 184-85. 18 W.S. McKechnie, Magna Carta (2nd edn, Glasgow, 1914), p. 50. 16

298

Colonial England, 106&-1215

because many actions had become actions in the first instance providing direct access for the litigant. Nevertheless, much of it was a sophisticated extension of the procedure of tolt which carried an allegation of default of justice from a lower to a higher court. Now, this was inapplicable in actions between tenantin-chief and tenant-in-chief in the royal court. In these circumstances the great magnate remained a baron in the old-fashioned sense of vassal, or as Magna Carta would have it, a peer, and the king exercised justice as feudal lord. To take the simplest but perhaps most important example, the writ of right was directed to the lord of the court where the plaint lay; Glanvill's example is addressed to Earl William (of Devon); it commanded the earl to do full right; if not, the sheriff of Devon would intervene so that there should be no further complaint or default of right.19 It was plainly unthinkable that the king would direct such a writ against himself (although he might occasionally direct something similar against his agent); and the tenant-in-chief had to resort to some other procedure. Much devious work and many legal complications ensued from this and similar situations; and baronial litigants incurred great costs.20 But the plain fact was that by the beginning of the thirteenth century, in many circumstances, the rear-vassal enjoyed better legal safeguards than his lord. In 1215, therefore, the magnates who negotiated the settlement were not cdnferring benefits on the freeman as a form of baronial largesse. They were trying to bring themselves under the protective umbrella of the legal privileges which the freeman already enjoyed. In some ways it was a formidable task: the forma securitatis of the Charter and the baronial court of Twenty-Five reflect the difficulty of binding a lord who had no lord (except, of course, the pope). How had the freeman got into this advantageous situation? For it was advantageous despite all the traps and penalties which festooned legal process. A complete answer would take us back to the Norman Conquest and beyond. But nowadays many historians, especially legal historians, would not go that far. Instead they link security of tenure, and the attitude to secure rights which grows from it, with the establishment, emergence or development, however they like to call it, of the common law: especially with the settlement of 1153, which brought peace at the end of Stephen's reign, with the Assize of Northampton and the action of mort d 'ancestor, with the action of right which extended the supervisory role of royal justice, and with the rule that no one need answer for his freehold without a royal writ. This is simply to select some of the salient features of very sophisticated, subtlely presented arguments.21 19 Tractatus de legibus et consuetudinibus regni Anglie qui Glanvilla vocatur, ed. G.D.G. Hall (London, 1965), p. 137. 20 1 hope to deal with these in the forthcoming new edition of Magna Carta. 21 S.E. Thome, 'English Feudalism and Estates in Land, Cambridge Law Journal (1959), pp. 193-209; S.F.C. Milsom, The Legal Framework of English Feudalism (Cambridge, 1976); R.C. Palmer, The Feudal Framework of English Law', Michigan Law Revieiv, 79 (1981), pp. 1130-64; idem, The Origins of Property in England', Law and History Review, 3 (1985), pp. 1-50. For a comprehensive review of the problems and the arguments see Joseph Biancalana, 'For Want of Justice: Legal Reforms of Henry II', Columbia Law Review, 88 (1988), pp. 433-536. For more 'gradualist' views see

Magna Carta, 1215-1217

299

Yet we must be wary of our common law. At the time of Magna Carta, as Maitland pointed out long ago, ins commune meant the common law of the church, not the common law of England.22 G.D.G. Hall commented with some impatience: It is easy and fruitless to argue about the details of such a definition, and about the exact date at which there can first be said to be a common law: what is clear is that it is a product of the twelfth century.23

Semantic niggles are, as he said, unrewarding. But contemporary views are of some interest. Glanvill is remarkable for its lack of a historical context; it treats the reign of Henry I as no more than a base (though an important one) for subsequent rights and procedures; if the author emphasises anything as a turning point or great achievement of his time it is the grand assize.24 His contemporary, Richard fitz Neal, the treasurer, was more of a historian. In his Dialogue of the Exchequer he chose to emphasise that William the Conqueror 'decided to bring the conquered people under written law', which led him, very sensibly, to Domesday Book; he also selected for special comment from the Assize of Northampton, not anything to do with mort d 'ancestor or security of tenure, but the provision for the eyres whereby Henry II revived the great days of his grandfather.25 So their emphasis was not quite the same as ours. They were perhaps too close to the new legal procedures, especially Glanvill. Be that as it may, even legal historians, indeed especially legal historians, have to stick to the rules of evidence. It would be easier and more acceptable to build on the precise wording of cap. 4 of the assize of Northampton; whether heirs appear in the single or plural, for example; if we had more than one text of the assize and could exclude the intervention of a copyist.26 Likewise, the plain fact is that with regard to tenure, the restoration of the ante bellum status quo was not included in the text of the treaty of Westminster of 1153. Chroniclers may tell us that there was such an agreement, and indeed there had to be some continued

R.C. van Caenegem, Royal Writs in England from the Conquest to Glanville, Selden Society, 77 (London, 1959); idem, The Birth of the English Common Law (Cambridge, 1973; 1988); and the very important paper of Paul R. Hyams, 'Warranty and Good Lordship in Twelfth Century England', Law and History Review, 5 (1987), pp. 437-503. 22 F. Pollock and F.W. Maitland, The History of English Law (2 vols, Cambridge, 1898), i, pp. 17678. 23 Glanvill, De legibus, ed. Hall, xi. 24 Ibid., p. 28. 25 Richard fitz Nigel, Dialogus de scaccario, ed. Charles Johnson, corr. F.E.L. Carter and D.E. Greenway (Oxford, 1983), pp. 63-64, 77. 26 J.C. Holt, The Assizes of Henry II: The Texts', in The Study of Medieval Records, ed. DA. Builough and R.L. Storey (Oxford, 1971), pp. 85-106, especially p. 90.1 failed to note there that the last sentence of cap. 7 reinforces the argument that cap. 1 is a summary historical account, not an official text. I am now inclined to believe that cc. 2-4, as we have them, are likewise reportage rather than official instructions to the justices.

300

Colonial England, 1066-1215

such ruling, but its deliberate exclusion (and it can scarcely be anything else) from the binding terms of a treaty has to be explained; and the explanation is unlikely to be that there was some other binding agreement,/ its text now inconveniently lost; for the evidence suggests that territorial restoration may have been approached along other lines.27 One such line might run as follows. In Michaelmas 1200 King John sent an instruction to the justices of the Bench that in any matter raised before them they were to do nothing on account of charters or letters patent of his predecessors unless they were confirmed by him.28 What happened in the court as a result is by no means clear; certainly there was no general rejection of earlier royal grants. The main effect apparently was to encourage men and institutions to seek John's confirmation of their privileges; the charter rolls of his first two regnal years occupy very nearly half the printed edition of all the rolls of his reign. Let us suppose now that John's instruction was the usual policy at an accession. No such writ survives for Richard, but the chronicles make it very clear that, perhaps more than any other king of the twelfth century, he conducted an organised sale of offices and confirmations of privilege as he mustered English resources for the Third Crusade. The results are manifest in the pipe roll of 1190; and charters from his first regnal year preponderate in the total even more than in the case of John, very nearly 50 per cent.29 Henry II is much more difficult. His early pipe rolls are of little help; the work, and to some extent the expertise, of the exchequer had suffered collapse consequent on the Anarchy of Stephen. Moreover, his charters and writs carry no time-date. Nevertheless, a very large number of confirmations seem to come from the first few years of his reign. Now if Henry II attempted to do what John did in 1199 and Richard seems to have done in 1189 he was faced by a problem which his successors did not encounter. Who was his predecessor? Plainly it could not be Stephen; it could scarcely even be Matilda, for she was never a crowned monarch. He was driven inexorably to Henry I and to 1135 as the terminus of lawful government. It was this which underlay the definition of royal rights and the determination of tenure of both king and vassal. Now if that is right it matters. First, there is no need to imagine some formal agreement about the restoration of tenures in 1153. There was no need for one. Secondly, the definition of the predecessor would be likely to affect all aspects of government; it is no surprise, therefore, that the baronial cartaeof 1166, which Henry called for, distinguished between the old and the new enfeoffment on exactly the same basis of 1135. Thirdly, it links the thinking of Henry II and his sons with the antecessor, the lawful predecessor, of Domesday Book and the Norman Conquest.30 Fourthly, the definition 27

For the text of the treaty see RRAN, iii, no. 272. For comment see Hyams, 'Warranty', pp. 497-

503.

28

Curia Regis Rolls, i, p. 331. Holt, Magna Carta and Medieval Government, pp. 73-74. 30 I rely here in some degree on Dr George Garnett's study of the theory of succession under the Norman kings (Cambridge Ph.D.). 29

Magna Carta, 1215-1217

301

of the predecessor mingled with the principle that an incumbent had a duty to recover and maintain all the rights of his predecessor; this was authoritatively expressed by Pope Alexander III and was to contribute to Henry's quarrel with Becket and to much else besides.31 Why should Henry II acknowledge tide, right or lawful possession, whatever we like to call it, other than his own? The qualification surely provides the answer: his rights could not be defined except in relation to others'. What others? One certainly was the church, as Becket was to show. But as regards the laity, where were rights and lawful possession to end? With the immediate tenants of the king, for example, leaving the rear-vassals entirely to the jurisdiction of their lords? A glance at Henry I's so-called Ordinance of the Hundred will tell us that forty years or more before such a limitation was already quite impractical. And there is other evidence to hand. A century earlier the emergence of the Norman aristocracy which won England for the Conqueror had been marked by the adoption of toponymic family names. These families were dynastic in structure, committed increasingly to succession by primogeniture in absolute or qualified form.32 The implications of this for tenure and hereditary succession are clear and emphatic: Hence the Beaumonts derived their name from Beaumont-le-Roger, the name itself perpetuating the memory of the founding father, Roger de Beaumont. Likewise the Buslis came from Bully; it is recorded before the Norman Conquest that they held it by hereditary right. There are many other examples. So name, tenure, secure succession or claim to such, all went hand in hand. The same arguments can be applied to the rear-vassals. The Moultons derived their name from and held Moulton in the Lincolnshire Washlands; their neighbours, the Holbeach family, held Holbeach; the Trumpingtons came from and held Trumpington, and so on. Now in the case of the rear-vassals the cartaeof 1166, the records of feudal service which the barons submitted at Henry II's demand, supply data as valuable as Domesday Book does for their lords. The cartaename over 3000 tenants of the old enfeoffment, holding knight's fees in 1135. Of these 59 per cent had toponymic names of one kind or another. This is based on raw figures. They have to be cooked somewhat; tenants-inchief holding undertenancies can be excluded; so can names which are purely locative, Philip of Dorset and the like; and some deduction can be made for repetition. But when liberal allowance is made for these and similar reservations the proportion still does not fall below 55 per cent. This large residue of 1600 to 1700 names requires careful handling. To make the grade a toponymic family name has to be passed down from generation to generation. We could not even approach a systematic proof that by 1135 this had already happened in a significant number of these cases; the evidence one way or the other is quite inadequate. It would be much easier to demonstrate that the toponym31 Mary Cheney, 'Inalienability in Mid Twelfth-Century England: Enforcement and Consequences', Monumenta Juris Canonid, series C, subsidia 7 (1985), pp. 467-78. 32 See above, pp. 161-96.

302

Colonial England, 1066-1215

ics which appear in 1135 were retained in subsequent generations and that the family was connected with, and often still held, the manor from which its name was derived. No doubt some day this will be done. Meanwhile, here and there, it can all be seen working in detail. Compare the two Sussex lordships of Chichester and Hastings, the one held by the bishops, the other by the counts of Eu. Of the forty tenants at Chichester twenty had toponymies of one kind or another. Of the twenty listed knights at Hastings there were eighteen using such surnames.33 The difference between 50 per cent and 90 per cent suggests that practice and family nomenclature varied from fief to fief. In this case the hint is strengthened in various ways. At Chichester there were eight multiple tenancies involving thirty-seven tenants, almost certainly a vestige of an older practice of tenure and inheritance; attempts to preserve such can be traced in several dioceses. At Chichester, also, all or almost all the toponymies were English. At Hastings, in contrast, six of the eighteen were French. Two of these, St-Martin and St-Leger, can be given possible locations in the county of Eu, which lay on the eastern frontier of the Pays de Caux in Normandy. The Pays de Caux is distinguished as a land of absolute primogeniture in later Norman custom. It yields a good crop of toponymic names which antedate 1066. In England the lordship of Hastings provides a particularly good collection of charters in which the lords of the honour licensed alienation in alms by their vassals. The earliest of the series comes from before 1140.34 It also provides, as early as 1124, a possible instance of hereditary tenure confirmed by the lord of the honour, not simply of a vassal, but of a vassal's vassal.35 There are many indications in short of an early development of hereditary tenure and toponymic names in the lordship of Hastings. The counts of Eu and their tenants may have brought both with them from the Continent. It may have been precocious. By the side of Chichester it seems so; but at Chichester too toponymies were taking root. Now what did the descendants of the tenants of Hastings and their like get out of the assize of mort d 'ancestor? They had not been born into a world in which lords habitually denied seisin to their tenants' heirs; the lord appears as responsible in cap. 4 of the assize of Northampton because he alone could give seisin, and probably for no other reason. What the tenant got was a relatively cheap, summary procedure whereby he could try to avoid getting involved in a lengthy and costly action over title. Hence its popularity; hence also the not uncommon dismissal of the demandant with the instrucdon that he should seek a writ of right. Henry and his justices had provided them, not with some new security of succession (unless of course their lord had run amuck), but with a ready instrument which lay to hand for dealing with many, though not

33

The Red Book of the Exchequer, ed. Hubert Hall, Rolls Series (3 vols, London, 1896), iii, pp. 199200, 202-3. 34 See above, pp. 205-6. 35 J.C. Holt, 'More Battle.Forgeries', Reading Medieval Studies, 11 (1985), pp. 75-86.

Magna Carta, 1215-1217

303

all, of the difficulties which abounded when heir succeeded ancestor.36 They were securely established tenants, themselves succeeding and their heirs succeeding by hereditary right (I deliberately avoid the word title). The assize was icing on the cake; and to be sure many, then as now, liked icing. For the most important feature of the assize was its immediate popularity. Within thirty years it became the most used of all the possessory procedures. To this the history of final concords provides an interesting analogy. Agreements were as old as litigation. In twelfth-century England men reached concord in many settings, the lord's court, a chapter house, the shire court, before the king's justices or the king himself. The record can be followed in any adequate cartulary. But the cartulary evidence suggests that there was a change of pace. The characteristic record of an agreement concluded in the court of Henry II was a two-part indenture. On 15 July 1195, on the instruction of thejusticiar, Hubert Walter, it was converted into a tripartite indenture of which the 'foot' was filed in the treasury, whence many still survive today.37 From that point on concords proliferated, inspiring an increasing amount of collusive litigation. Just like the assize of mort d'ancestorthe final concord 'took off. There was also another parallel between the two. The assize was frequently used for small amounts of land; Glanvill's examples concern a virgate.38 The same became true of the final concord. Well within twenty years of their origin in 1195 the feet of fines testify that the final concord before the royal justices had become a standard instrument for the transfer of land among the freeholders of Norfolk, the descendants of the freemen and sokemen of Sten ton's Danelaw, smallholders conducting small transactions.39 Many of these men resorting to the court held no more than-ten or twelve acres. Godwin the tailor who brought an action of grand assize and reached final concord concerning land in Syderstone, two acres in 1198 and six acres in 1200, would be otherwise unknown to us; the same is true of the two men, Robert son of Asalac and Humphrey son of Siward, with whom respectively Godwin divided these petty holdings.40 So there was intense downward social pressure in certain functions of the legal system. Increasing governmental resources made these procedures more available. As the twelfth century advanced more men from a wider social range could afford to resort to court. Final concords agreed before the royal justices, filed as the feet of fines, provided just that mass-produced, relatively cheap form of record of their

36 The action of mort d'ancestor did not lie within the immediate family, although in its early history some litigants attempted to use it so. 37 Feet of Fines of Henry II and Richard I, Pipe Roll Society, 17 (1894), pp. 20-21. For comment see C.R. Cheney, Hubert Walter (London, 1967), pp. 95-96. 38 Glanvill, De kgibus, ed. Hall, p. 150. 39 See the comments of Barbara Dodwell in Feet of Finesfor the County of Norfolk, 1198-1202, Pipe Roll Society, new series, 27 (London, 1952), pp. xxvii-xxxix, for a very full and illuminating discussion; also/fe*/ of Finesfor the County of Lincoln, 1199-1216, ed. Margaret S. Walker, Pipe Roll Society, new series, 29 (London, 1954), p. xxxviii. 40 Feet of Fines, Norfolk, nos 178, 252.

304

Colonial England, 1066-1215

holdings which they required. Or so it seems, for we know far less of how they had dealt with such matters a century earlier. So the relationship between the source and the user of justice was symbiotic. The final concord is the classic example of the government supplying what the litigant wanted. Taking a broader view to embrace all those ways in which men from the local communities were drawn into royal government, selfgovernment at the king's command often looks much the same as royal government at the subject's behest or at least at his encouragement. Cooperation was all. But it was not all peaceful cooperation. In 1215 conflict arose because king and subject sought to use the very same institutions each to his own end. Hence the purchase of local privilege designed to give the county or other local community some control over the sheriffs and other royal agents, and the direct opposition of royal and baronial sheriffs throughout the civil war.41 Economic change played a part in this. Increasingly complex forms of justice would have been neither demanded nor supplied without the increasing resources in time, men and money which both litigant and government could deploy. Each sought to profit. It has come to be accepted in recent years that inflation stacked the cards against the government.42 In some of its operations this was certainly the case. Fixed revenues were penalised and these included the shire farms. The government imposed increments; Magna Carta in turn forbade them. Indeed inflation was the underlying reality to all arguments and claims which emphasised good and ancient custom. But that still leaves room for some reservations. First, if we take the most extreme calculation of the inflation, 300 per cent over the forty years from 1180 to 1220, that amounts to an annual rate of 2.8 per cent. Many of us would have been very content to live under that during the last forty years. Putting it another way, King John would have to find £159 in 1216 to buy what had cost him £100 in 1199. Many of us again would be very content with that. Whether such calculations matter, given the differences between the economies of the thirteenth and the twentieth centuries, is a moot point. But they leave us with one important fact: the annual rate of inflation was far less than the variations from year to year. Men at the time were conscious enough of increasing values and increasing profitability, but that there was a price revolution in progress must have been quite lost amid all the agricultural and financial hazards which beset them. This is to leave on one side local variations in prices. Secondly, to concentrate on the shire farms and other fixed sources of income, or for that matter on the customary rents due 41

Holt, Magna Carta, pp. 51-60, 251-52. RDA. Harvey, The English Inflation of 1180-1220', Past and Present, 61 (1973), pp. 1-30, which was intentionally more tentative than its subsequent reception has allowed; idem, 'The Pipe Rolls and the Adoption of Demesne Farming in England', Economic History Review, second series, 27 (1974), pp. 345-59; D.L. Farmer, 'Some Price Fluctuations in Angevin England', Economic History Review, second series, 9 (1956-57), pp. 35-43. For an inversion of the usual view which seeks to explain inflation as a consequence of legal change, in particular of the 'development of property', see R.C. Palmer, The Economic and Cultural Impact of the Origins of Property', Law and History Review, 3 (1985), pp. 375-96. 42

Magna Carta, 1215-1217

305

to landlords, is in some degree misleading. In the case of the crown, direct consumption of one's own produce was inflation proof, extraordinary taxes were adjusted to produce increasing yields, fines, offerings and amercements could be and were jacked up, the profits of justice increased with the proliferation of actions and the number of litigants. Thirdly, as we well know, the real beneficiaries of inflation are those with capital and those in debt. The king had the greatest capital resources for he was the greatest individual landowner in the realm, and he was not at all quick in paying his debts. Nor, for that matter, were the great magnates, even their debts to the king. The people who really suffered from the depreciation of money must have been the Jews. That is part of an irresistible diversion on which much more could be said. But I have said enough to indicate that Magna Carta is not to be regarded as a consequence of monetary inflation. It may seem that all the arguments in this essay tend toward a gradualist approach. This is not intended. It is a matter, not of choosing between the gradual and the cataclysmic, but of trying to assess within the total picture what was new and what of ancient origin. In my view Magna Carta had deep roots. It reflects the work of Henry II and indeed sought both to counter and to develop it. But that is far from the whole story. The Charter opens with the liberties of the church. It then proceeds to deal with feudal incidents. In this it was modelled exactly on the coronation charter of Henry I, and numerous sources, both chronicle and documentary, reveal that Henry's charter played a large part in the Great Charter's germination. It is Henry's charter rather than John's that requires a very special explanation. Magna Carta, as we have seen, was one among several, within a century one among many. Henry's charter, in 1100, was unique. But this is verging on quite another story. In conclusion it may be well to abandon the roots and look to the points where old and new mingled to produce a revolutionary mixture. One was the court of Twenty-Five; its number derived from the London council of twentyfive of 1200-1; its function to compel the king to execute the Charter's terms; its only weapon, feudal distraint.43 Another was the oath to the terms of the Charter, taken by the commune of the whole land. No one alive to the development of municipal liberties in the twelfth century could fail to miss the revolutionary implications and intent of that.44 Finally, there was something which has attracted little comment. Charters were not executive documents. They were executed by mandates to administrative agents. This was so for Magna Carta. Preliminary steps for its execution were notified in letters patent of 19 June 1215. But these were only preliminary; they provided for investigation of the king's officials and for reports to the king before further action. However, as the Charter was carried to the shire courts in the high summer of 1215, men took matters into their own hands, treating it as an executive, indeed a mandatory instrument; where necessary they appointed or accepted new baronial 43 44

Holt, Magna Carta, pp. 239-40; Palmer, The Origins of Property', pp. 390-95. Holt, Magna Carta, pp. 48-49.

306

Colonial England, 1066-1215

sheriffs. That probably happened through administrative breakdown. But at the centre of the arrangements, responsible for the delivery of four charters and twelve of the executive writs of 19 June, was Master Elias of Dereham.45 It is difficult to imagine that his eagerness for reform did not play a part in this. I have had to deal with miscellaneous matters. They are not all of my own choosing. But that this should be a miscellany, is. For Magna Carta was not a product of a coherent political theory. It was to produce one; for it became a statement of the liberties held by the community of the realm. And it became such faster than even the radicals may have expected in 1215; partly because of the political accident of the minority of Henry III, partly because the Charter generated its own new patterns of political thought. So within a very few years the commune of the land of 1215 was reemerging as the community of the realm. But in the original Charter of 1215 the political thought comprised ideas, notions, practices, drawn from many different sources: canon law procedures; perpetual alms; the communal oath; customs; liberties, ancient or new; practices and principles of feudal jurisdiction - distraint, judgment of peers; and all this intermingled with a powerful injection from royal government - its practices and routine procedures, acceptable or reprehensible. In short, like any other document of political crisis - the Declaration of Independence or the Declaration of the Rights of Man - Magna Carta drew on all the various intellectual resources which lay to hand. Magna Carta is perhaps the first political document to reveal such a process in considerable detail. It is certainly the first to demonstrate the power of such miscellanies once they are brought together in a more or less coherent whole.

45

Rot. Litt. Pat., 180b.

16 The Casus Regis: The Law and Politics of Succession in the Plantagenet Dominions, 1185-124 7

The Plantagenet succession of 1199 posed this problem: suppose a man has several sons, and suppose that at his death an elder son is already dead but himself leaves a son; who takes precedence in the inheritance, the grandson in the senior line or a junior surviving son? (Pedigree 1)? Is the grandson to sue-

father I elder son

younger son

grandson

ceed to all that his father would have inherited had he survived, thereby representing his father's title? Or does the younger son, the cadet, inherit on the ground that a son is nearer to the father than a grandson? This was the issue between King John and Arthur of Brittany which was to lead to Arthur's murder. He was not the only one to suffer. This is a crime story. Richard Lionheart died of his wounds on 6 April 1199. He left no legitimate heir and hence the succession lay within the legitimate progeny, in contemporary legal language within the parentela, of his father Henry II. Therein the eldest surviving male line was represented by Arthur, duke of Brittany, the only legitimate surviving son of Henry II's son, Geoffrey, who died in 1186. There was also John Lackland, younger brother of Geoffrey, the only legitimate son of Henry II still alive (Pedigree 2).l John succeeded, to Normandy and England in 1199, and then to the whole of the Angevin lands except for Brittany, to which Arthur was immediate heir. John's succession was acknowledged by King Philip of France, overlord of all these lands except for England and part of Gascony, in the treaty of Le Goulet of 1200. However, the succession was disputed. In Normandy and 1 See Pedigree 2. In this Arthur appears as the representative and John as the cadet claimant. Note that the relationship of cadet to representative claimant is always that of uncle to nephew or niece. In this and all subsequent pedigrees the representative descent appears in italic, and the representative's defunct father, who is always the older brother of the cadet, in bold type.

308

Colonial England, 1066-1215 2

Plantagenet

Henry II d. 1189 Henry d. 1183

Richard succ. 1189 d. 1199

Geoffrey = Constance of d. 1186 Brittany

John succ. 1199 d. 1216

Arthur b. 1187 d. 1203

England there was some debate, and in Maine, Anjou and the Touraine Arthur was first recognised as heir only to have his claim overriden by the armed intervention of John. A straightforward explanation has commonly been accepted for these events. It runs as follows. First, Brittany was not in question; there Arthur succeeded not as his father's but as his mother's heir. Secondly, in Maine, Anjou and the Touraine custom was on the side of the representative heir; hence the preliminary recognition of Arthur. Thirdly, in Normandy the reverse was the case; custom was on the side of the cadet; hence the recognition of John. Finally, in England also John was accepted immediately as heir. However, both in Normandy and in England, there was room for debate. Hubert Walter, archbishop of Canterbury, apparently supported the representative claimant; William Marshal took the opposite line in favour of John; and at John's coronation the archbishop proclaimed that kingship was elective within the royal line. That the decision went in favour of John affected the law for several generations. The 'king's claim' still stood in the way of the representative heir in the De legibus Angliae attributed to Bracton; it was he indeed who dignified it as the casus regis.2 Most of this seems to be based on strong evidence. All the chroniclers concur that there was an initial response in favour of Arthur in Maine, Anjou and the Touraine, and that is consistent with what we know of the customs of these lands later in the thirteenth century. For Normandy there seems to be the strongest confirmation of the prior claim of the younger son in the earliest Norman law book, the Tres ancien coutumierof c. 1199; indeed it has been asserted that, in his argument with Hubert Walter, William Marshal was quoting this legal source. For England the arguments on either side seem to be arranged with judicial precision by Glanvill; all the more acceptable, perhaps, because he seems so balanced, muddling through the issue in true English style. All sources, 2 Bracton de legibus et consuetudinibus Angliae, ed. G.E. Woodbine, trans, and rev. S.E. Thorne (4 vols, Cambridge, MA, 1968-77), iii, p. 284; iv, p. 46.

The Casus Regis, 1185-1247

309

chroniclers and law books alike, fall neatly into place in a coherent pattern in which variations of the law or custom of inheritance from one province of the Plantagenet realm to another seem to determine the initial choice made by each province in 1199. That was the argument presented by Roger of Howden, one of the best of contemporary chroniclers, writing in 1199 itself or shortly afterwards.3 It is equally the premise of Powicke, Painter and all other modern commentators.4 The story has scarcely changed since the event itself. Yet it is not self-evidently true. Indeed it is contradicted in other aspects of family and property law where Normandy, the apparent exception in 1199, marched with other provinces of western France. Moreover, the legal evidence, apparently so strong, allows quite another interpretation. It may be taken to reflect, not geographic variations in the law of descent, but chronological stages in the evolution of the casus regis. If so, law books and chronicles alike will reveal, not customary law determining the success or failure of the contenders, but legal doctrines and arguments concocted to give their conflicting claims respectability. Much of the evidence, in short, will appear as a put-up job. That is precisely what it is. In the case of Normandy this is relatively easy to establish. The definitive statement of the medieval custom of Normandy, the Grand coutumier of 125458, states very clearly that the ancient custom of Normandy gave precedence to the representative heir, the grandson. It also states that the opposite argument used in favour of John in 1199 was a new and perverse law, introduced not by right and custom but by the force and oppression of powerful men.5 Cap. 12 of the Tres ancien coutumier of 1199 enunciates this new rule - 'The younger son is the nearer heir of the inheritance of his father than the grandsons born of his elder brother'. That this was concerned with John's accession to the duchy is demonstrated by the later gloss: 'as happened in the case of John the English king and many others'. It then adds 'and this was the falsest of judgments'. The gloss is independent of, but coincides with, the statement in the Grand coutumier. The two together outweigh the older text. And there is a further complication which confirms this. At first sight the ruling in the Tres ancien coutumier might seem to have been concocted in order to justify the fact of John's succession: but not so, for it is followed immediately in the text by a further chapter: 'Later, during the war, when King Richard held the duchy, daughters were excluded from having any part of the inheritance of their father against grandsons born of their brother'.6 Now this rule must have been laid down between May 1194 and April 1199. The text makes it clear that it was subsequent to the rule giving precedence to the cadet and that rule cannot 3

Chronica Rogeri de Hovedene, ed. William Stubbs, Rolls Series (4 vols, London, 1868-71), iv, pp. 86-87. 4 P.M. Powicke, TheLoss of Normandy (Manchester, 1913), pp. 193-95 (revised edn, 1960), pp. 12932; S. Painter, The Reign of KingJohn (Baltimore, 1949), pp. 1-3. 5 Coutumicrs deNmmandie, ed. EJ. Tardif (2 vols, Rouen, 1896), ii, pp. 73-74, 242. 6 Ibid., i, pp. 12-13.

310

Colonial England, 1066-1215

have been introduced at a time when Arthur was recognised as heir. It must therefore belong to the years 1196-99 or to 1189-90, and there is only one likely occasion, 12 March 1190, when King Richard held a great council at Nonancourt in Normandy in order to make dispositions for the government of his dominions before departing for the crusade.7 So much for Normandy. Now for England, and at this point I make some simple chronological correlations. The succession was not a new question in 1199. It was open, apparent to all, from August 1186 when Geoffrey duke of Brittany died. Geoffrey left one surviving daughter, Eleanor, probably born in June 1184. She lived on to die at Bristol in 1241: a woman heir, true, but no Angevin could object to that. Arthur was born posthumously to Geoffrey on 29-30 March 1187. From that point on, with Henry II already at the considerable age of fifty-four and Richard heirless and about to depart on the crusade, there was every reason for men to ask: who is the rightful heir, Arthur or John? And that is relevant not only to the Tres ancien coutumierbut even more emphatically to the De legibus attributed to Glanvill, for Book 7 of the De kgibus, in which succession is discussed, was written between 29 November 1187 (after Arthur's birth) and 6 July 1189 when Henry II died.8 It is easy to undermine the legal objectivity of the Tres ancien coutumier. Glanvill is a much more complex proposition. Writing when he did, with the royal succession very much in mind, he could have concocted an argument emphatically in favour of the cadet. Instead he hedged his bets; he summarised the opposing arguments; and it is this which gives his statement the appearance of judicious objectivity. However, before that is accepted at face value, there is one further set of correlations to be made, this time concerning the descent of the honour of Mandeville and the earldom of Essex. In 1187, as Glanvill was writing his work, these honours were held by William de Mandeville, the third surviving legitimate son of Geoffrey, first earl of Essex, who died in 1144 (Pedigree 3).9 William's next of kin and heir was his aunt, Beatrice, who married William de Say (d. 1144). Beatrice, for long a widow, was now eighty years old or so and her tide was exercised by two claimant lines: first, her younger son, Geoffrey de Say; and secondly, her granddaughters in the senior line, the only surviving legitimate children of her eldest son, William, who had died in 1177. Her younger son, Geoffrey, stood to the Mandeville succession as John did to the royal tide, and the two granddaughters, like Arthur of Brittany, were representative claimants. These two were both married: the elder, Beatrice, to Geoffrey fitz Peter, the future chief justiciar of England, and 7

For the chronology of this see L. Landon, The Itinerary of Richard /, Pipe Roll Society, new series, 13 (London, 1935), pp. 196-208. 8 Tractatus de legibus et consuetudinibus regni Anglie qui Glanvilla vocatur, ed. G.D.G. Hall (London, 1965) (hereafter Glanvill, De legibus ed. Hall), pp. xxx-xxxi. 9 Most of the genealogical information which follows is to be found in Complete Peerage, under Essex, v, pp. 113-35, and Say, xi, pp. 464-74. The one important matter in this which requires correction, as also inJ.H. Round, Geoffrey de Mandeville (London, 1892), pp. 232-33, is that Arnulf, the eldest son of Geoffrey first earl of Essex, allegedly disinherited, was almost certainly illegitimate.

The Casus Regis, 1185-1247

311

3

Mandeville Beatrice de Mandeville = William de Say d. 1197 d. 1144

Geoffrey de Mandeville 1st earl of Essex d. 1144 Arnulf (illeg.)

Geoffrey 2nd earl d.s.p. 1166

Beatrice = Geoffrey fitz Peter d. 1197 4th earl, d. 1213 Geoffrey 5nd earl d.s.p. 1216

William 3rd earl ds.pm9

William de Say d. 1177

Matilda = William of Buckland d.1222

William 6th earl dsp. 1221

Geoffrey de Say d. 1214

Geoffrey de Say d. 1230

Matilda d. 1236

the second, Matilda, to William of Buckland; and it so happened that these two husbands were half-brothers: Geoffrey fitz Peter, the senior, born of his mother's first marriage; William of Buckland, the junior, of her second. The fireside family discussions among the potential successors are now lost to us, but it is reasonably easy to trace the transformation of hopeful interest into expectant and precautionary action. In 1180 William de Mandeville, the last earl in the direct male line, married Hawise, countess of Aumale. She was a famous heiress. By 1212 she had survived three husbands and then purchased widowhood from King John at the cost of 5000 marks.10 Some of her fame was not what she would have desired: Richard of Devizes, writing in King Richard's time, described her with monastic humour as 'feminam fere virum cui nihil virile defuit preter virilia'.1 * At all events she bore few children, and none at all to William, earl of Essex. Soon those with a stake were reading the signs. On 25 January 1185 Geoffrey fitz Peter and his wife, and William of Buckland and his wife, coming together in the king's court at Melksham, made final concord concerning the inheritance of William de Say. Geoffrey fitz Peter and Beatrice conveyed the manor of Brunninton and the service of two tenants to the other pair along with ten librates of land 'from the first acquisition or escheat to come by inheritance to the said Beatrice and Matilda from whomsoever it descends'. In return William of Buckland and Matilda released to Geoffrey and Beatrice 'the rest of all the inheritance of the aforesaid Beatrice and Matilda from whomsoever it descends or may descend without any retention or claim which William and Matilda or her heirs might make in the future'. By this arrangement Geoffrey 10 11

Pipe Roll 14 John, p. 37. The Chronicle of Richard of Devizes, ed. J.T. Appleby (London, 1963), p. 10.

312

Colonial England, 1066-1215

fitz Peter cleared the ground and established his wife, and himself iure uxoris, as the acknowledged representative claimant to the honor of Mandeville and the earldom of Essex. This agreement was concluded in the curia regis before a group of justices and barons headed by Ranulf Glanville. Henry II confirmed it by charter.12 It was known to the king. Ranulf Glanville, who gives his name to the De legibus, knew of it. Geoffrey fitz Peter, who is another possible author of the book, was a party of it. In any case 'Glanvill,' whoever he was, could scarcely have ignored the Mandeville succession, which affected one of the major earldoms to which a rising young curialis was staking out a claim. Our author was therefore in a quandary. He could not opt for the cadet since that Would deny the claim of Geoffrey fitz Peter to the honour and earldom of Essex. Equally, he could not opt for the representative heir since that would exclude the claims of John, count of Mortain, to the throne and all the Angevin dominions. He would have to present both cases, and that is precisely what he did. Nevertheless, it is very striking that in reviewing the arguments he came down on the whole on the side of representation.13 He introduced the topic by stating that * there is usually great doubt in law' about it. He then stated the case of the cadet: that if the eldest son did not survive until the death of his father he did not survive to inherit. In the B text this view was attributed to Ranulf Glanville. He then stated the case for the representative heir: that he should succeed to all the rights which would have been his father's had he lived. In the B text this is attributed to another justice, Richard de Lucy, but in Z, the earliest MS of the prime A text, it is attributed to both de Lucy and Glanville. It was also the author's view - ita dico. This at once strengthens the case for regarding Geoffrey fitz Peter as the author. Be that as it may, he was ominously frank - meliorest conditio possidentis: 'the man in possession has the better case'. Geoffrey fitz Peter was to demonstrate just that. By 1189, therefore, the lists were drawn, the claimants known, and the cases marshalled. What precisely were the precedents and the arguments in law? Here earlier English law books are of little help; the issue was not discussed in the relevant sections of the Leges Henrici Primi.14 There were precedents. The descent of the counties of Ponthieu and Seez on the death of William Talvas in 1171 must have been known to the contenders of 1189; it differed only in that the representative succession was formally recognised during the grandfather's lifetime.15 In this case the matter was settled by partition. For 12

PRO, Duchy of Lancaster, 10/29, from which the detail of the agreement is taken. Glanvill, De legibus, ed. Hall, p. 78. 14 Leges Henrici Primi, ed. LJ. Downer (Oxford, 1972), c. 70, 18-22, p. 225. R. Besnier argued that the use of the terms appropriate to the parentela in this section suggests that representation was already the rule, but it could equally well be argued that the author's readiness to accept succession of ascendants (c. 70, 20) indicates the opposite. See R. Besnier, La representation successorale en droit normand (Paris, 1929), pp. 132-33. 15 J.C. Holt, 'The End of the Anglo-Norman Realm', Proceedings ofthe British Acatlemy, 61 (1975), p. 41; reprinted in Magna Carta and Medieval Government (London, 1985), p. 61. 13

The Casus Regis, 1185-1247

313

England there was the less certain case of the Beauchamp estates which involved King Stephen in the siege of Bedford in 1137.16 Here the cadet line, if that is what it was, succeeded. However, the contenders in 1189 gave no sign of calling on these possible resources. Indeed it is a curious comment on the historical sense of the period that King John never exploited King Alfred, for Alfred was yet another uncle who prevailed over a nephew. Yet the contenders' claims were subject to several cross-currents. As the laws of succession took shape in the course of the twelfth century, so, in romance, did the wicked uncle emerge as the archetypal villain. Indeed, Chretien of Troyes's Cligesand the Anglo-Norman Laid'Havelocprovide literary parallels to many of the features of the casus regis; in both the nephew triumphs and all live happily ever afterwards. So the tide was running against the uncle. But he could still keep his toe in the door in certain circumstances. In the succession to a kingdom or principality he could rely on his worth as an adult as against the risks of a minority which the representative might entail. Secondly, he might stand a better chance if he were in competition not with a nephew but with a niece, for then he could call on all those instincts and rules which gave precedence to males over females - 'while the male line survives and the inheritance stems therefrom, a woman shall not inherit'.17 These considerations were much to the point in the successions which were in mind in 1189. Arthur was no more than a two-year-old babe: John was the obvious successor. The Mandeville grandchildren were heiresses: the cadet therefore enjoyed all the advantages of a male against a female claimant. In the Mandeville case yet another legal notion was at work: the distinction between inheritance and acquisition which directed the patrimony to the senior line but allowed the tenant to dispose the acquisition elsewhere. The final concord between Geoffrey fitz Peter and William of Buckland of 1185 established that they already held the Say estates which descended to their wives through William de Say, the eldest son of William de Say and Beatrice de Mandeville. That was beyond debate. However, the marriage portion which Beatrice de Mandeville brought to the Says was a different matter. This she intended to go to her younger son, Geoffrey, and that was accepted by William de Mandeville, earl of Essex. That is put beyond doubt by a charter of Earl William confirming the grant which Beatrice had made and which Geoffrey de Say had confirmed of Elsenham church to the priory of Walden.18 Moreover, both Earl William and Beatrice clearly looked upon Geoffrey de Say as the heir apparent to the Mandeville lands and tide. The earl's charters are never witnessed by the senior line, whether by the eldest son, William, or later by his sons-in-law, Geoffrey fitz Peter and William de Buckland. Geoffrey de Say, in contrast, witnesses six of the surviving deeds; in one, recording a transaction concluded in the exchequer 16 For a summary of this difficult descent see P.M. Stenton, The First Cmtury of English Feudalism (Oxford, 1961), pp. 237-38. 17 Leges Henna Ptimi, c. 70, 20b, p. 225. 18 BL, MS Harley 3697, fo. 21.

314

Colonial England, 1066-1215

before the Young King in 1170, his name figures alongside that of Ranulf Glanville.19 The views and intentions of the earl and Beatrice were shared throughout the Mandeville honour. In the eyes of the chronicler of Walden priory, the family foundation, Geoffrey de Say was the lawful heir and Geoffrey fitz Peter an intruder.20 All these cross-currents can be seen at work as the casus regis developed and precipitated similar legal actions. Richard Lionheartwas crowned on 3 September 1189. John had accompanied him to England and was present at the coronation. William de Mandeville, earl of Essex, was with the new king frequently from his arrival in England and was made jointjusticiar, with Hugh du Puiset, bishop of Durham, on 16 September. One of the five men associated with them was Geoffrey fitz Peter, who was just establishing himself as one of the justices of the curia regis. The earl died at Gisors in Normandy on or about 14 November during a mission to King Philip of France. Richard himself left England for France and the crusade on 12 December. Hence he had to make some provision for his own succession and resolve the claims to the Mandeville inheritance within the space of a few weeks. It is well known that he left a confused arrangement. John was not formally acknowledged as heir to England. On the other hand, he received such a vast territorial endowment from his brother that any other solution, in case of Richard's death, was most improbable. The matter was still open, however, when, on 12 March 1190, at a great council held at Nonancourt in Normandy, the king made final dispositions for the management of his lands in his absence. The meeting was attended by Queen Eleanor, Princess Alice of France, Walter archbishop of Rouen, Baldwin archbishop of Canterbury, and sundry bishops including Hubert Walter, Hugh du Puiset, Geoffrey Plantagenet, elect of York, and William Longchamp, bishop of Ely, who was appointed justiciar south of Humber at this very meeting. It was here that John and Geoffrey of York swore that they would not enter the realm of England for three years, an oath which was conditionally or absolutely relaxed in John's case through the intervention of Eleanor of Aquitaine. It was here, too, that recognition of John's claim to Normandy, the interference with the Norman law of succession recorded in the Norman custumals, is most likely to have occurred. At this stage there was no sign of a rift between the two royal brothers. John set off with the king on his march south. They parted company on 12 April at Gorron in Maine, over the border from John's county of Mortain. The Mandeville succession was settled at the same time and with a similar result in favour of the cadet. Geoffrey fitz Peter made the first move. He offered the king 3,000 marks for the land of Earl William, which he described as the inheritance of his wife, Beatrice de Say, 'so that it should be in the king's pleasure 19

J.H. Round, Feudal England (London, 1895), pp. 504-8. BL, MS Arundel 29, fos 52, 68. The translation of this chronicle as The Book of the Foundation of Walden Abbey', by C.H. Emson in Essex Review, 45-46 (1936-37), is unreliable, as is the version in Monasticon anglicanum on which it is based. 20

The Casus Regis, 1185-1247

315

to accept or not according to the service Geoffrey had rendered' - the classic plea of the curialis. Geoffrey de Say countered with an offer of 7000 marks.21 This was accepted, for he levied an aid on the tenants of the honor22 and issued deeds confirming the lands of Walden priory. His success, however, was to be very temporary. The Walden chronicler, no doubt with hindsight, described Geoffrey de Say as a vigorous soldier but of little wisdom or circumspection in worldly affairs.23 There was nothing more worldly than the exchequer. The chronicler recorded that Geoffrey made his first terminal payment of 700 marks but defaulted at the second; the pipe roll records no payment at all. According to the chronicle, he surrendered the estates into the hand of the chancellor, William Longchamp; the constable of the Tower, William Pantel, certainly accounted as custodian of some of them from Easter 1190.24 At that point Geoffrey fitz Peter was with the king in Normandy along with Longchamp, Hugh du Puiset, William Marshal and others associated in the new administration. It may be doubted whether Geoffrey recommended that the usual lenience of the exchequer in dealing with outstanding debts should be extended to his competitor. In a few months the matter was resolved. By 1 August 1190 Geoffrey fitz Peter held the Mandeville honor by gift of the king and was seised 'by writ of the Chancellor who was thenjusticiar'.25 The tables had been turned. The royal charter which Geoffrey had at this point has not survived. A later charter issued by King Richard at Messina on 23 January 1191 granted the lands of Earl William de Mandeville to Geoffrey fitz Peter and Beatrice de Say, Beatrice succeeding 'tanquam juste et propinquiori heridi'.26 The tables were also turned on CountJohn. On 6 October 1190, in the treaty of Messina concluded with Tancred of Sicily, Richard Lionheart agreed to a marriage alliance involving Arthur of Brittany and formally recognised Arthur as his heir. News of this drove John into incipient then open rebellion which was not suppressed until Richard's return to England in 1194.27 So far the two cases marched roughly pari passu through a zigzag of changing policies. How far there was a search for legal consistency is uncertain. Geoffrey de Say's default was a happy accident which enabled Geoffrey fitz Peter to seize his opportunity at a time when King Richard unexpectedly came to acknowledge the representative claim to the throne; fitz Peter might not have fared so well if de Say had kept up with his payments. Still consistency had been achieved by design or accident. From 1196 this no longer held good. Early in that year the Bretons denied Richard's request for the custody of Arthur and 21

For both proffers Pipe Roll 2 Richard, p. 111. Ibid., p. 2. 23 Walden Chronicle, fos 50, 58. 24 Pipe Roll 2 Richard, p. 2. 25 Rot. Curiae Regis, ed. Sir Francis Palgrave, 2 vols (London, 1835) i, p. 75. 26 J.H. Round, Ancient Charters, Pipe Roll Society, 10 (London, 1888), no. 59. 27 Landon, Itinerary of Richard I, pp. 196-208. 22

316

Colonial England, 1066-1215

placed him in the hands of King Philip of France. By October Richard was treating John as his heir once more. From this point on the two cases diverged, gradually, awkwardly and, in the end, disastrously. Legal logic would now have required the substitution of Say for fitz Peter. This never happened, but there were signs that it might. Geoffrey de Say was not in disfavour; he was one of King Richard's sureties in Germany; he witnessed royal charters there, and in England and Normandy up to 1197. But he must by now have been over fifty, and his son, also called Geoffrey, was now beginning to make the running. By 1198 Geoffrey son of Geoffrey de Say held land in Normandy of the barony of William de Mandeville; it was not held in fee, but deprestito regis Ricardi28 Somebody had got at the king, and the king was politically volatile. Meanwhile Geoffrey fitz Peter was suffering mixed fortunes. His wife died sometime before 19 April 1197; thereafter he held the Mandeville honours by courtesy of England. However, he held all the advantages accruing to a loyal and important servant of the crown. On 15 June 1198 Richard I issued a charter confirming the final concord dividing the inheritance between his wife and his sister-in-law, which had been agreed in 1185.29 Geoffrey was appointed chief justiciar on 11 July 1198. He promptly had the covenant of 1185 entered on the pipe roll.30 He did not even bother to account for the customary payment of one mark for such an enrolment. By this time it must have seemed very likely that Richard would die without a legitimate heir of the body. The new justiciar could await the outcome with confidence. But his ambition was tempered by caution. It seems very likely that he had squared the opposition. The evidence for this is delicate. Richard I died of his wounds on 6 April 1199; news of it reached England on the 17th. The curia continued in session at Westminster. With the royal succession now in the balance it became very difficult for fitz Peter to evade the claim of de Say. In two separate actions during this session tenants of members or dependencies of the great multiple manor of Sawbridgeworth vouched Geoffrey son of Geoffrey de Say to warranty.31 At some point the mesne lordship of the greater part of Sawbridgeworth, valued at at least £74 annual revenue and amounting to the service of four knights' fees or so, had passed to the de Says where it remained throughout the thirteenth century. These two actions were brought while the justiciar himself was present at Westminster and in court. The danger for him was that Geoffrey son of Geoffrey was vouched as the heir of William de Mandeville. Not unnaturally, the cases were not concluded, but Geoffrey fitz Peter did not challenge the assertion. He let it pass. Meanwhile he was making himself useful. When the archbishop of Canterbury and William Marshal arrived from Normandy to secure allegiance to John, he accompanied them to Northampton where assurances were given 28

Rotuli Noi-manniae, ed. T.D. Hardy (London, 1835), p. 104. Round, Ancient Charters, no. 66. 30 Pipe Roll 10 Richard, p. 139. 31 Rotuli CuriaeRegis, i, pp. 328,333-34. Comparison with PRO, KB 26/11 reveals that the printed text is neither complete nor accurate. 29

The Casus fogis, 1185-1247

317

to some of the magnates. At Westminster he saw to the execution of writ after writ emanating from John already entitled duke of Normandy and lord of England. John duly arrived and was crowned king on 27 May. On the same or the following day the new king belted Geoffrey fitz Peter as earl of Essex. The Walden chronicler recorded his indignation that fitz Peter had no hereditary claim.32 No one else expressed surprise. The new king and Geoffrey had helped each other into the saddle. The history of Geoffrey's new tide provides some hint that he and John had long since reached an understanding. Certainly one of his original charters survives from 1190-91 in which he appears as earl of Essex,33 and he received the third penny of the county of Essex from the account of 1191 onwards. But the tide only appears once in a document coming from the royal family. This was in October to December 1191 when Geoffrey, as earl of Essex, witnessed Countjohn's confirmation of the chapelry of Blyth to the cathedral church of Rouen.34 CountJohn was one of the witnesses to Richard's charter of June 1198 confirming the concord of 1185. In 1201 he himself as king issued a further confirmation.35 Through collusion John and Geoffrey fitz Peter provided England with a justiciar who held his lands and tide through the acknowledgment of representative succession and a king who owed his crown to a denial of that same principle. It had all been cleverly arranged. It would be less easy to escape the consequences. Like Siamese twins, the two were now tied together in a conflicting but inseparable relationship. Plainly Geoffrey could not do without John. John might be able to dispense with Geoffrey - and as time passed there were signs that the justiciar was not wholly in the king's favour - but only at the cost of depriving one who had been recognised as lawful heir both by himself and by King Richard. Knowledge of the issue was widespread, and that not just among the knightly and baronial class; it was still known sixty years later in the Yorkshire townships which, in 1212, had produced the rustic prophet, Peter of Wakefield.36 It was known in particular to litigants in analogous cases, but reference to it is always hidden, and it soon became obvious that those who mentioned it openly incurred extreme dangers. It became, as it were, the great sub-plot, the hidden text, of the reign of Kingjohn. This story of events over the fourteen years before John's accession throws light on several problems. First, it explains the new king's sensitive reaction to any challenge to his title. Hitherto historians have accepted reasonably easily thatJohn was the lawful heir to England and Normandy. That assumption is no longer possible once the objective authority of Glanvill and the Tres anden coutumieris undermined. John was in the same boat as Stephen: there was an alternative claimant with a genuine title. Hence his weakness in the face of King Philip of France and the concessions of the treaty of Le Goulet in 1200, 32

Walden Chronicle, fo. 68. Stafford, William Salt Library, S.D. (Pearson), 248. 34 Cat. Docs. France, no. 61. 35 Rotuli Chartarum, ed. T.D. Hardy (London, 1837), p. 90. 36 See the 'Invectivum contra regemjohannem', BL, MS Cotton, Vespasian E III, fos 168-175v. 33

318

Colonial England, 1066-1215

whereby he obtained Philip's agreement that he was the lawful heir of King Richard. Hence his reiteration, as the Norman war developed, that his enemies were seeking to disinherit him. Hence the murders and executions which followed: Arthur first in 1203, then Matilda de Briouze, who talked too much about Arthur's death, starved to death in a royal dungeon along with her son William in 1211; and finally, in 1213, Peter of Wakefield, who warned the king to repent of his crimes, dragged at the heels of horses from Corfe to Wareham. Secondly, it explains the uncertain treatment of representative succession in the royal courts. The king's case by itself would have encouraged the justices to take a firm line against representation. The king's and the justiciar's case together produced profound confusion - a lawyer's delight. To find uncle in conflict with nephew or niece is not unusual. There were several obvious sources of dispute; an uncle's wardship of a nephew, or the enfeoffment of a younger son who ultimately became the tenant of his nephew, were perhaps the most common. However, there seems to be no action recorded on the surviving rolls of the reign of Richard I where the uncle argues in principle that he has precedence over a son or daughter of a deceased elder brother; his case is always based on enfeoffment or final concord or the like. Indeed the courts accepted the opposite representative principle and continued to do so for a time under John. Hence in Hilary term 1199 the Bench, sitting under Hubert Walter, accepted the argument of Simon le Bret that a claimant's case should be dismissed on the ground that he had a nephew, son of an elder brother deceased, who was the nearer heir and therefore the proper person to pursue the claim.37 Again in Easter term 1200 a jury held that great-grandsons took precedence over a younger son (Pedigree 4); the possible objection that either

4 Vautort Richard de Vautort

John

Hugh Simon son

son

(issue is between John and sons of Simon)

the elder brother or his son or both had predeceased the ancestor was not advanced, but that this had happened is likely.38 And in Easter 1201 it was accepted that the existence of sons of a deceased elder brother might bar an 37 38

Curia Regis Rolls, i, p. 72. Rot. Curiae Regis, ii, p. 189.

319

The Casus Regis, 1185-1247

action of mort d'ancestor brought by the younger brother.39 The last two cases were held while Geoffrey fitz Peter was presiding. But the king already had a finger in the pie. In the second case mentioned above it was noted that the inquiry had been held 'on order of the lord King and not through the consideration of the court or according to the custom of the realm'.40 In the Cornish eyre of 1201 Hugh Sot (Pedigree 5), asserting that 5

Sot

Philip

Simon

Hugh

Claricia

Walter

(Hugh tenant as Philip's heir)

his elder brother, whose son had claimed one acre and a messuage, had predeceased his father, sought judgment of the court whether he was the next heir. The justices, headed by Simon Pattishall, adjourned the case sine diebecause judgment was pending on the will of the king.41 No judgment was ever recorded and the uncle remained in seisin. In Trinity 1200 William de Ros, nephew of Jordan de Ros (Pedigree 6), challenged his uncle's tide to two fees in Lulling-

6

Ros Geoffrey

William d. before Geoffrey William

I Jordan

(Jordan tenant as Geoffrey's heir)

stone and Farningham, Kent. Jordan's counter was a clear rejection of the representative principle. The case came before the court during a period in which Geoffrey fitz Peter was presiding. It was referred to a subsequent session at Bury St Edmunds where, before John, bishop of Norwich, and Geoffrey fitz Peter, Jordan restored the land to his nephew and William surrendered it to his uncle for life. This neat arrangement was disturbed in 1211 when the king 39

Curia Regis Rolls, i, p. 414. Ibid., ii, p. 189. The recognition seems to have been held before the parties appeared in court. 41 Pleas before the King and hisJustices, ed. Doris M. Stenton, Selden Society, 67, 68, 83, 84 (4 vols, London, 1948-67), ii, nos 484, 528. 40

320

Colonial England, 1066-1215

instated Jordan's son in the cadet line. The matter came back to the court in Michaelmas 1211. William de Ros's title was confirmed on the ground, not of representation, but of the concord reached in 1200.42 Such cases usually dragged on for years without resolution. Occasionally a party was cleverer than the rest or had a better lawyer. Take the example of the manor of Penselwood, Somerset (Pedigree 7). Here the cadet uncle, Robert 7

Dunmere Agnes = Ralph Robert

Henry William

(William tenant)

son of Ralph of Dunmere, was claimant and the nephew, William of Dunmere, tenant, in an action of mort d'ancestorfor one knight's fee. The argument turned on the fact that Henry, Williams' father and Robert's elder brother, had predeceased the ancestor. But the ancestor in this case was Agnes, mother to Henry and Robert, and she was clearly of unusual mettle. When her husband died she had allowed her eldest son to seek a wife and had arranged for the newlyweds to live with her. When the son died she provided her daughter-inlaw with dower, thereby ejecting her from the household, and herself took charge of the children of the marriage. On feeling the advance of years she decided to dispose of her inheritance and persuaded John of Montagu, chief lord of the fee, to take homage from her grandson William. She then resigned the fee to her grandson and survived in his care as an aged woman until her death. Her younger son challenged the resignation and argued that she had died seised, but this was not accepted and he was instructed to seek some other process. Old Agnes had arranged matters very well. The tale should be told by a Jane Austen rather than by the dry record of a court.43 In such difficult and uncertain legal circumstances some settled for compromise. In 1204, for example, the barony of Percy was divided between the cadet uncle, Richard de Percy, and his nephew, William (Pedigree 8). Here Kingjohn accepted the establishment of a representative heir in half the honour; the probable reason is that an influential curialis, William Briwerre, had wardship of the nephew. The division was the starting-point of recurrent litigation which ended only with the death of Richard de Percy without legitimate heirs in 1244.44 In this case uncle and nephew took opposite sides in the civil war of 121542

Curia Regis Rolls, i, p. 187; vi, pp. 134-35. Ibid., vii, pp. 117-18. 44 Cvmpkte Peerage, x, pp. 445-55. 43

321

The Casus Regis, 1185-1247 8

Percy William de Percy d. 1175 Maud, countess of Warwick

I Agnes

ds.p. 1202-4

d. 1202-4

Henry d. 1198

Richard dsp. 1244

William d. 1245 17. The issue had moved outside the courts. That it became political was the work of King John, for he, before all others, had an interest in resolving it. In 1214 the king accepted a proffer of 20,000 marks from Geoffrey de Mandeville, son of Geoffrey fitz Peter, for the hand of Isabella, countess of Gloucester. In the same year he accepted the proffer of 15,000 marks from Geoffrey son of Geoffrey de Say for the lands of William de Mandeville. Historians have long been troubled by these arrangements, attributing them with zest or puzzlement, and according to taste, to the tyranny, greed and suspicion of KingJohn. Another explanation is now possible: namely thatJohn was trying to remove all the ambiguities which dogged the casus regis by reinstating Say in the honour of Mandeville and compensating Geoffrey fitz Peter's son with the earldom of Gloucester. Geoffrey fitz Peter died on 14 October 1213. John was involved in a deepening political crisis which he partly resolved early in November by individual concessions to some of the great magnates of East Anglia. One was Robert fitz Walter, and another was Geoffrey fitz Peter's heir, Geoffrey de Mandeville, who had married fitz Walter's daughter, now deceased. By 4 November Geoffrey de Mandeville had performed homage for the honour of William de Mandeville and for all the lands which he held of the king by hereditary right. He also received fitz Peter's wardships. But there was no specific statement that he held the Mandeville lands by hereditary right, there was no reference to a relief, Geoffrey was not girt as earl, and did not receive the third penny of the shire.45 He appears as earl of Essex only very rarely in royal documents and probably then in error. On 26 January 1214 Geoffrey made fine for the hand of Isabella of Gloucester under severe terms and penalties.46 Chroniclers concur that he 45 46

Rotuli de Oblatis etFinitms, ed. T.D. Hardy (London, 1835), pp. 502-3. Ibid., pp. 520-21.

322

Colonial England, 1066-1215

did so unwillingly. He had not yet been confirmed in the lands and title of his father. He was now being forced to accept a great earldom, true, but at enormous cost and through an heiress who was too old to provide him with heirs. He was being shunted into a genealogical cul-de-sac, velvet-lined but none the less a cul-de-sac, at the end of which was a trap: if his wife died he could not continue in possession of her lands and title by courtesy of England, and waiting at the trap-door was Richard, earl of Clare, husband of Amice, Isabella's sister, to whose son, Gilbert of Clare, the earldom of Gloucester ultimately descended. Meanwhile de Mandeville's rival was pressing his case. In 1212 the king warranted that Geoffrey de Say senior (who was still alive but now of great age) had attorned one of three men, including his son, in a plea of land against Geoffrey fitz Peter.47 The plea came before the Bench at Westminster, with Geoffrey junior the attorney, in February 1214, just after the king had sailed for his campaign in Poitou. De Say claimed as heir to his mother. De Mandeville defended successfully by pleading that he had not succeeded to all the rights of Earl William; in particular he was not earl of Essex and had not obtained possession of several of the Mandeville estates, including Sawbridgeworth and Edmonton which de Say himself held. He was adjourned sine die:, Geoffrey de Say was allowed to seek another writ but was in mercy.48 But Geoffrey junior did not give up at all easily. He followed the king to Poitou (while Geoffrey de Mandeville stayed at home) and there on 11 July he made a proffer for the land of William de Mandeville of which his father had been disseised, so he alleged, 'by the will of King Richard'.49 The king instructed the new justiciar, Peter des Roches, to consult his senior colleagues and do what seemed expedient. Nothing further is recorded. What might have happened is speculative for by the autumn the ordinary functioning of government was distorted by crisis and incipient rebellion. By then Geoffrey de Say senior had died and his son succeeded to his father's lands at the very modest relief of 400 marks.50 By 1215 the Pandora's box could not be closed. Moreover it was now an issue for the opposition magnates as well as the king. For there was also the case of Briouze (Pedigree 9). Nothing is more piteous in the reign of John than the fate of Matilda de Briouze and her eldest son William, both starved to death in Windsor Castle. Their fate has long been linked with the casus regis, for Matilda died as she did because she knew and said too much about the fate of Arthur of Brittany.51 But there was another link. No one at the time was so moved as to feel that the least amends which might be made was to ensure the succession of young William's surviving son. On the contrary in 1214-15 the next brother in line stepped in and proffered 9,000 marks for the inheritance, and he was a 47

Curia Regis Rolls, vi, p. 270. Ibid., vii, pp. 110-11. 49 Rotuli Liit. Claus., ed. T.D. Hardy (2 vols, London, 1833-34), i, p. 168b; Rotuli de oblatis etfinibus, pp. 527-28. 50 Rot. de Oblatis etFinibus, p. 535. 51 F.M. Powicke, The Loss of Normandy (Manchester, 1960), pp. 315-24. 48

323

The Casus Regis, 1185-1247 9

Briouze William de Briouze d. 1211 William d. 1210 John d. 1232

Giles bishop of Hereford, d. 1215

Reginald d. 1227/8 William d. 1230

bishop - Giles of Hereford. Moreover, when he died within the year, he was succeeded by yet another brother, Reginald; and it was not until 1219-20 that young William's son, John, the representative claimant, recovered Bramber against his uncle; he was never able to recover Totnes and the Welsh estates of the family which remained with the cadet line.52 The fissures created by this and other cases snaked across the family and political relationships which lay on the surface in 1215. It will now be obvious why, in one and the same letter of 10 May 1215, just as the political crisis was about to culminate at Runnymede, Kingjohn offered the judgment of his court to Geoffrey de Mandeville and Giles of Hereford: judgment offered by one who had murdered a representative claimant, to a bishop who had by-passed a representative claimant, and to an earl whose proffer for the heiress of Gloucester offered a line of retreat from a representative claim.53 Individuals were subject to competing stresses. Consider the position of Richard, earl of Clare. He had married Isabella of Gloucester's sister, Amice. He had a claim to the earldom and indeed it was to descend to his son. So he was interested in Geoffrey de Mandeville's marriage and the terms of his proffer for Isabella. But his sister Aveline was Geoffrey fitz Peter's wife and widow, still alive. So he was interested in her dower, which should have included a third of the Mandeville lands. Meanwhile, one of his daughters, Matilda, had married the younger William de Briouze who had died in John's prison. So yet again he was interested in her dower and marriage portion, and indeed instructions for the restoration of her marriage portion, the town of Buckingham, were issued by the king from Runnymede on 21 June 1215.54 Finally, the church of Canterbury became involved. In 1214 nearly £7000 of the proffer which Geoffrey de Mandeville had made for the hand of Isabella of Gloucester was transferred to Stephen Langton, archbishop of Canterbury, in part settlement of the damages which his church 52 Complete Pea age, i, p. 22; Curia Regis Rolls, viii, pp. 10-11, 27, 35; ix, pp. 9, 306-7; Patent Rolls, 1216-1225, p. 165. 53 RotuliLitt. Pat, ed. T.D. Hardy (London, 1835), p. 141. 54 Ibid., p. 143b.

324

Colonial England, 1066-1215

had suffered during the Interdict.55 This arrangement survived the death of Geoffrey de Mandeville in 1216, of his brother William in 1227, and remained as a burden on their sister and ultimate heir, Matilda. By 1234-35 she had assigned the manors of Walden and Debden to Edmund, archbishop of Canterbury, as security for payment. Her charter was witnessed by the sons of Geoffrey de Say junior who had died in 1230.56 The Mandevilles and the Says had not come together in peace at last, for they continued to litigate on to the reign of Edward I. It was rather that the archbishop was ensuring that all possible claimants to the honour of Mandeville were involved in the arrangement. The agreement also envisaged possible interference per vim regiam. Yet the succession crisis was passing. By 1227 Matilda was the third of Geoffrey fitz Peter's children to hold the Mandeville inheritance. Ten years earlier the young Henry had been recognised as lawful king of England, crowned, confirmed by the pope, and accepted by the alternative claimant, Prince Louis of France, in the treaty of Lambeth. Lawyers were now trying to close the door which John's accession had opened. It was not easy; uncles were almost always adult and nephews and nieces might often be minors. In the courts consistency was impossible. When, in 1218, Richard de Percy claimed Wansford as son and heir of Agnes daughter of William de Percy, it was recorded that the assize did not lie because Richard had an elder brother to whose heirs the title descended. Thus the justices in eyre in Yorkshire, headed by Martin Pattishall.57 Despite that the courts continued patiently to record adjustments made by final concord to the original division between Richard and his nephew, William, and Richard retained his moiety of the Percy honour until his death in 1244. Moreover, the judgment in the Percy case in 1218 should be compared with the descent of the lands of Saer de Quency, earl of Winchester, sometime baron of the exchequer and member of the baronial court of Twenty-Five in 1215 (Pedigree 10). Saer

10 Quency Saer de Quency earl of Winchester, d. 1219 Robert d. 1217 I Margaret d. 1266 55

Roger d. 1264

Pipe Roll 9 Henry III, rot. 8, m. 2r. Cal Charter Rolls, i, pp. 196-97. 57 Rotts of Injustices inEyrefor Yorkshire, 1218-19, ed. Doris M. Stenton, Selden Society, 56 (London, 1937), nos 198, 287,1115. 56

The Casus Regis, 1185-1247

325

died at the siege of Damietta in 1219. He left a granddaughter by his elder son Robert, dead in 1217, and a younger son, Roger. It was Roger who succeeded in 1221 and this was certainly by arrangement within the family.58 The cadet could still win. The cases later collected in Bracton from the plea rolls of 1217-40 reveal a confused, unsettled situation. Where uncle and nephew were opposed the issue frequently depended on the enfeoffment of the younger by the elder brother or father, or on the younger son's acquisition of land during the minority of his nephew. Only a handful of cases involve the core of the argument about representation: namely the date of death of the elder son. In one of these the representative demandant accepted the logic of the cadet's counter and argued that his father had not predeceased his grandfather.59 In others the cadets or their descendants pleaded in a more or less standardised form that the younger child as the 'hearth-child' 'awaited the inheritance'.60 In another a wise grandfather bestowed his lands on his eldest son and his heirs irrespective of whether the son predeceased him or not.61 That at least took the trouble to counter the argument against representation. In yet another the existence of grandchildren, sons of an elder brother, released the younger brother from warranty; in short the court accepted the representative argument.62 In another again two cousins simply asserted the case for and against representative succession and asked for judgment.63 In all these cases the younger brother or his descendant was the sitting tenant. In some the issue was settled by final concord. In most it is impossible to trace a judgment of the main issue. In the last, where the arguments were clearly put, the ultimate recorded postponement carries the explanation: quia nepos et avunculus sunt.64 Despite this Bracton was far clearer than Glanvill. He was settled on principle: the representative grandchild was the nearer heir, although not the nearer in 58 Complete Peerage, xii, pp. 2, 751, which states that the elder son Robert (d.s.p. 1217), requires correction, as does Sidney Painter, The House of Quency', Medievalia et humanistica, 11 (1957), pp. 3-9; reprinted in his Feudalism and Liberty, ed. Fred A. Cazeljr (Baltimore, 1961), pp. 230-39. That the Robert who married Hawise of Chester was Saer's elder son and not his brother is put beyond doubt by the charter in which Earl Saer granted estates to Robert, his son and heir, on hi? marriage to Hawise sister of the earl of Chester. See PRO, DL 42/2, fo. 481 v, and for the corresponding charter of Ranulf of Chester see DL 25/42. For Margaret, the daughter of this marriage, see Complete Peerage, vii, p. 677. For Roger, the second son, as recognised heir in the lifetime of his father, see, for example, Liber cartarum prioratus SanctiAndree in Scotia, Bannatyne Club (Edinburgh, 1842), pp. 255-56.1 owe some of these references to the kindness of Dr Grant Simpson and Dr David Crouch. 59 Bracton's Notebook, ed. F.W. Maitland (3 vols, Cambridge, 1887), no. 892; Curia Regis Rolls, xi, no. 2033. The demandant's objective is clear, but the record confuses vigil and morrow and is thereby self-contradictory. 60 Bracton's Notebook, nos 230, 892, 951, 982, 988, 1830; Curia Regis Rolls, xvi, no. 959. 61 Bracton's Notebook, no. 968; Curia Regis Rolls, xii, nos 365, 1075, 1077. 62 Bracton's Notebook, no. 1185. 63 Ibid., no. 982. 64 Curia Regis Rolls, xii, no. 910.

326

Colonial England, 1066-1215

degree. It might be that he would find his uncle in possession, especially if the uncle was a hearth child and had lived in the ancestor's house; if so, the assize of mort d 'ancestor would not lie and an action of right would not lead to judgment. The casus regis intervened. The best the grandchild could do would be to seek a final concord.65 Bracton was in no doubt where right lay. The casus regis involved real debate in Glanvill: in Bracton it was political detritus hindering the proper operation of the law. And that seems to have been the view of the king. When, in 1247, Peter of Savoy consulted Henry III on the descent of the lands of Ruald fitz Alan, constable of Richmond, whether they should go to a representative grandson or to a cadet, the king was in no doubt: the lands were to go to the grandson - 'filio primogeniti audacter reddatis'.66 The explanation for such forceful direction is not hard to find. Six years earlier, in 1241, Arthur of Brittany's sister, Eleanor, had died at Bristol at the age offifty-seven.The danger was removed. The crown need no longer require the law to behave as an ass, though lawyers and litigants were to goad it into performing so on to the end of the century and beyond. Even Henry III advised boldness. He could afford to. He was secure on the throne. Eleanor, the representative claimant, had spent the last thirty nine years of her life in captivity - a life sentence, her only crime that she was the innocent consequence of a genealogical accident.

65 66

Glanvill, De legibus, fos 64b, 267b, trans. Thorne, ii, p. 189; iii, p. 284. Close Rolls, 1242-47, p. 524.

Index

Aaron, Jew of Lincoln, 11 Abergavenny, honour of, 89 Abernethy, treaty (1072), 150 Abe tot, family, 134; Urse de, 134 Abingdon, abbey of, 10, 11, 74, 79; tenants of, 139; Faricius, abbot of, 11; Reginald, abbot of, 138, 139 Acle (Norfolk), 51 Acquets, acquisitions, see Inheritance Acre, Joan of, 131 Adeliza, supposed daughter of William I, 246 JElfigifu (Queen Emma), 16 jElfhelm, will of, 163-65, 168 ^Ethelred II, king of England, 16 jEthelstan, aetheling, 185 Affidatio in manu, 273 and n. Africa, 2 Aigle, 1', Gilbert de, 34; Maud de, 195 Amesse, esnecia, 258-59 Albini, Aubigny, Nigel de, 91, 176-78, 195, 242, 250n., 265 266n.; William I de. (pincerna), 96, 176-78, 181, 195, 265, 266n.; William II de, 1st earl of Arundel, 281, carta of (1166), 92; William III de, 2nd earl of Arundel, 135; William IV de, 3rd earl of Arundel, 231 Albini, Aubigne (Brito), family, 181-83; William I de, 181, 182 and n., 265, 266n.; William II, 182n.; William III 182, 240; William IV (of Belvoir), 182 Alecto Editions, 27-29 Alexander III, pope, 301 Alfred, king of the West Saxons, 313; mother of, 15 Aliens, 268 Almain, Henry of, 232 Allertonshire, 24 Alnwick (Northumb.), 85n., 87, 279 Alodium, 5, 115, 202, 209, 212 Amercements, 46 American colonies, 1-2, 13

Americans, and medieval England, 172 Amundeville, Walter de, 105-6 Angevin 'tyranny', 292-93 Anglo-Saxons/Scandinavians, survival of, in Norman England, 103-11, 169, 194 Anglo-Saxon England, and the Norman conquest, xiii-xv; forms of tenure, 65, 169, 203-6, 216; kinship and lordship, 4-5, 7, 162, 185-86, 223; local institutions, 24; military organisation, 71, 75-76, 81-84; personal names, 167, 184-86; social institutions, 59-65, 168-69; testaments, 163-65, 168 Anglo-Saxon chronicle (Peterborough chronicle), 9n.-12, 31, 33-34, 36, 47, 55, 151 Anjou, county of, 248, 308; Fulk, count of 34, 151; Geoffrey, count of, 123, 125, 126, 144; Henry, count of, see Henry II Annebecq, barony of, 152 Anselm, archbishop of Canterbury, 50n., 151, 199,201,218 Antecessor, antecessores, xiv; in Domesday, 49; in 1154, 300 Apple tree wapentake (Derbys), 49n. Apulia, Viger of, 189, 123n. Aragon, Peter, king of, 293, 296 Arden, Hugh of, 180; Alice, wife of, 180 Armentieres, Geoffrey de, 267 Arques, William, count of, 170, 189; William of, tenant of the see of Canterbury, 249n. Arsic, family, 86 Articles of the Barons (1215), 266 Arundel, barony, 136; castle; 135, earldom of 83; earl of, see Albini Ascelin, son of Arthur, 217n. Ascelin Goel, 249n. Ashton, T.H., 158 Aubigne, St-Aubin de, 181, 183 Aubigny, St-Martin de, 181, 195 Aubrey, earl of Northumbria, 150

328

Colonial England, 1066-1215

Aumale, counts of, 256; Aveline, countess of, 131; Hawise, countess of, 264, 311; Isabella, dowager countess of, 131; William, count of, 132, 283; see alsoYorz Aunou, Fulkd', 189 Austen, Jane, 320 Aveley (Essex), barony of, 85 Avranches,John, bishop of, 211 Avranchin, 155 Baal, 8 Bailleul, 208n. Baldwin the Fleming, 90 Ballon, Reginald de, 88 Baqueville, Nicholas de, 189 Bardolf, Doun, 104,109n.; Hugh, 104, 109n. Bardsley, C.W., 167,179,184 Barrow, G.W.S., 1, 2n. Baring, F.H., 64 Barlow, F., 32n., 158 Barney (Norfolk), 254-56 Barnoldswick (Yorks.), 288-89 Barnstaple, honour of, 93 Barnwell priory, chronicle, 225-26, 229, 295: Liber memorandorum, 227 Baronies, classification of, 84-92; division of, 93-94; establishment of, 48-49, 95-97, 168, (in Ireland), 100; escheat, forfeiture of, 88-91 Bartlett, R., 1, 2n. Basset, Richard, 265 bastardy, xvi, 276n. Bates, David, 56 Bateson, Mary, 60n., William, 60n. Bath, 13n., abbey church, 8n. Batde, abbey, 7,173, 205-6 Baudri the German, 189 Bayeux, cathedral and see, 137,148,152; Odo, bishop of, 51, 52, 86, 87, 90,105n., 136, 138, 141,175; English estates of, 84, 95-96; Philip, bishop of, 137; Suspirus de, 287 Bayford (Herts), 143 Baynard, William, 139 Beauchamp, honour, lands of, 313; Walter de 134; William de 134, 290n. Beaufour, Richard de, 211; Robert de, 210-11 Beaumont, family/house of, 54, 122-23, 132,152,157,174,187, 212, 301; estates, 152-53,169; name, 187-88; Henry, 1st earl of Warwick, 124, 152;

Hugh de, 153; Robert I de, count of Meulan and 1st earl of Leicester, 54, 152-53, 212, 249; Robert II de, 2nd earl of Leicester, 152, 281, 285; Robert IV de, 4th earl of Leicester, 230; Roger I de, 124,152-53,169,187,194-95, 203n., 211, 301; Robert brother of, 187, 203, 204; Roger II, 2nd earl of Warwick, 154, 285; see also, Leicester; Meulan; Neubourg; Pont-Audemer; Vieilles; Warwick Beaumont-le-Roger, 124,152, 301 Beaumont-sur-Oise, Ivo, count of, 213 Beckering (Lines.), 91 Becket, Thomas, archbishop of Canterbury, 9n., 290 and n., 292, 301 Bedford, 313 Belet, John, Alice, widow of, 264 Belleme, castlery of, 296; charter for tenants of (1215), 293, 296; lordship of, 293; name of, 187; Adeline de, 249n.; Mabel de 247n., 249n.; Robert de, 86, 155, 171, 175, 246n., 249n.; William de, 214 Bellencombre, casde of, 275 Belvoir, honour of, 83, 181, 182 Bench, the, court of, stay of pleas before, 300 Beneftcium, 63, 116, 210-12 Benn, name, 183 Bendey, William, 21,23 Berkeley, honour of, 89, 273n.; Roger of 89, 103n., 194; see also fitzHarding Berkhamstead (Herts.), honour of, 136 Berkshire, 41, 49n., 74-76; five-hide unit in 75-76; Godric, sheriff of, 49n. Bernard, Walter, charter of 256n. Besnier,R., 312 Bessin, vicomteof, 154-55, 215n. Beverley (Yorks.), 15 Biancalana, J., 272n., 287n. Biddlesden (Bucks.), abbey 285 Bienfaite, 196 Bigod, family, 123-24,152,174,190; fee, 96,176; name, 190; Hugh 1,1st earl of Norfolk, 123n., 132,151-52,171, 278, 289; Gundreda, wife of, 151-52; Roger I, 52, 92, 252; William son of, 123n.; Roger II, 2nd earl of Norfolk, 83,151-52,172, 200, 231; Hugh, brother of, 151-2; Roger III, 4th earl of Norfolk, 190; Hugh brother of, 190; Roger IV, 5th earl of Norfolk, 131; William, 172 and n.

Index Binham (Norfolk), priory of, 252, 254-56; Richard, prior of, 255 Bishop, T.A.M., 290n. Bishoprics, new Norman foundations, 6-8 Bisson, T.N., 293 Bitton (Glos.),273n. Black Book of the Exchequer, 82 Blackstone, Sir William (1723-80), 32-33, 46,54 Blackstone Edge (Lanes.), 24 Bloch,Marc, 76-77, 114 Blois, Peter of, 8; William of, see Boulogne Blund,John, 15n. Blyth (Notts.), priory, 85n., 216, 317 Bocquence, Baudri de, 189, 250n. Bohun, Henry de, 3rd earl of Hereford, 133, 233; Humphrey de, 4th earl of Hereford and Essex, 131 Bolingbroke (Lines.), honour of, 249n. Bolsover (Derbys.), 132 Bondi the staller, 49n. Born, Bertrand de, 77 Boscherville, abbey of St Georges, 209n., 228-29, 236; Louis, abbot of, 228 Boston (Lines.), 287 Boulogne, county of, 189; counts of, family, 123; Eustace I, count of, 54, 173; Eustace II, count of, 123n.; Eustace III, count of, 123n.; William, count of, 132, 271, 274, 283-84; Isabel de Warenne, wife of, 283 Bourn (Cambs.), 256, 257n. Bourne (Lines.), barony of 90, 91 Boussard,J. 77, 78 Bracton, Henry of, Laws and Customs of England attributed to, 65, 145, 308, 325-26 Bradninch (Devon), honour of, Bradbury, A.R., 55 Brakelond, Jocelin of, chronicle of, 198 Bramber (Sussex), 323 Breaute, Faulkes de, 267-69 Breteuil, 215; William of, 249n. Bretons, xvi, 54, 130; language, 14 Breves, in Domesday, 34, 47 Brighouse (Yorks.), 20, 24n. Brihtric, priest of Haselbury, 15 Brionne, castle, 132, 152, 192; Gilbert, count of, 170, 174, 191,193, 250n. Briouze, family, 323; estates, 323; Giles de, bishop of Hereford, 322-23; John de, 322; Philip de, 93; Reginald de, 322; William I de, 50; William II de, 280;

329

William III de, 205, 219, 234, 318, Matilda, wife of and William son of, 318, 322,323 Bristol, 326; abbey of St. Augustine, 285, 310 Brittany, duchy of, 115, 183, 307; Alan count of, 95; Arthur, duke of, xvii, 130, 161, 239, 307-10, 313, 315, 322, 326; Conan, duke of, 286; Constance, duchess of 308; Eleanor of, 162, 310, 326; Geoffrey, duke of, 307-8, 310 Briwerre, William, 320 Brooks, N. 64n. Brown, R. Allen, 32n. Brus, fee of, 91; Robert de, 91 Buckingham, 323; earldom of, 83; see also Giffard Buckinghamshire, 49n. Buckland, William of, 130, 257, 258n., 311, 313; Matilda, his wife, 311 Building in Norman and Angevin England, 5-13; inadequacies and accidents, 8-10; as a Keynesian multiplier, 10-12; 'early English', 12-13; of castles, 6-7; of churches, 6-7, 8-10; of halls, 9-10 Bully, 216, 301; Ernald de 155; Idonea de, 155; Roger I de, 117 and n., 168, 170-71, 178, 186, 188, 216; Roger II de, 155; Beatrice his sister, 155 Bullington (Lines.), 231 Bulmer, family, 104n.; fee of, 104n. Bun (or Wood),John, 180, 181 Burg Eltz, 7 Burgh, Hubert de, 197, Raymond, nephew of, 269; Thomas de 197, 198 Burghley, Lord, 23 Burgundy, duchy of, 77 Burton (Staffs.), abbey, 156-57; Geoffrey, abbot of, 156n., 157 Burun, Erneis de, 91, 96 Bury St. Edmunds (Suffolk), abbey, 9, 79, 200, 213, 242-43, 319; charters of, 198; Baldwin, abbot of, 201, 220; Samson, abbot of 197-201 Bussy, family, 260n.; William de 258 Bywell (Northumb.), 85n., 87 Caen, 8, 11; abbey of St Etienne, 44n., 158; monk of 150, 158 Calderdale (Yorks.), 19, 24 Cam, Helen, 64 Cambos, Robert de 177 Cambrai, Godfrey de, 96

330

Colonial England, 1066-1215

Cambridgeshire, 25, 40n.; sheriff of, see Picot Camera, Norman de, 263 Canterbury, cathedral, 8, 8n., 9n., 12, 151; see, 13n., 139; Baldwin, archbishop of, 314; Edmund Rich, archbishop of, 324; Theobald, archbishop of, 218n., 254; see also Anselm; Becket; Lanfranc; Langton; Walter; abbey of St Augustine, Wulfric, abbot of 183 Capetian kings, 129 Carabie, R., 116n. Carlisle, 155; lordship of, 127, 154 Carpenter, D., 291 Cartae baronum (1166), 81-83, 92-94, 99-100, 104-5, 107, 147 Carucage, s^Hidage Castle Gary (Som.), 85n. Castleford (Yorks.), 19 Castle Holgate (Salop.), 176 Castle Rising (Norfolk), 7 Castles, and family structure, 7-8; title to 131-3; in Ireland, 100; see also Building Castle guard, 220-21 Casus Regis, xvii, 238-39; effects of, 318-26; evolution of 308-12, 314-16; historiography of, 307-8 Catalonia, 'unknown' charter of, 293, 296-97 Cause (Salop.), barony of, 176 Caux, Pays de, 119, 189-90, 207, 301 Caxton (Cambs.), 93, 94 Celts, xv, 3, 169 Ceorl, 75 Cerisy, abbey of, 211 Chaplais, P., 28, 56, 57 Charlemagne, emperor, 173, 238 Charles (the Simple), king of France, 246; Gisele, daughter of, 246 Charles-Edwards, T.M., 168, 169 Charters, characteristics of, 273-74; of confirmation, 278-79; drafting of, 158, 180, 218-19, 220-21; as evidentiary documents, 274-75; forged, 282; terminology of, 278-80; see also Henry I; Magna Carta Chaworth, family, 256 Cheshire Domesday, 52 Chester, abbey (St Werburgh), 205; constable of, 279; Roger, constable of, 227, see also Lacy; earldom and honour, 52-53, 95, 127, 154; family, 123, 155; Hugh, earl of, 34, 37, 52; Ranulf de

Blundeville, earl of, 135; Ranulf de Gernons, earl of, 132, 137, 154, 155, 273n., 279; Ranulf le Meschin, earl of 127, 154, 155, 265; see also Meschin; Lucy, wife of see Lucy; Richard, earl of, 154 Chichester, cathedral and see, 6, 8n., 9; tenants of, 302 Chilham (Kent) barony, 86, 176 Chinon, 282 Chiselborough (Som.) barony of, 176 Chretien of Troyes, 313 Church, the, under Stephen, 288; under Henry II, 288-90 Citeaux, Gilbert, abbot of, 288 Clare, see also Brionne, Eu; family, 122, 131, 162, 170, 171, 174, 191, 225, 241; earldom (Hertford), 83, 193; Alice de, 225; Baldwin, son of Gilbert (Baldwin of Exeter), 90, 191-94, 196; Gilbert de, son of Godfrey, 191; Gilbert son of Richard I, 192n.; Gilbert son of Richard II, earl of Hertford, 193; Godfrey, 191; Isabel de, 267; Richard of Clare/Tonbridge, son of Gilbert, 191-94, 196, 216, 225; Richard, son of Richard, abbot of Ely, 192-93; Richard III de, 3rd earl of Hertford, 225, 259, 267, 322, 323; Robert, son of Baldwin, 192; Roger, earl of Clare/Hertford, 280 Clark, C, 14n., 15n. Clarke, H.B., 28, 38 Claverdon (Warks.), church of, 206 Clay, C.T., 274 Clifford castle, (Heref.), 51 Cliges, 238-39, 313 Clitheroe (Lanes.), honour of, 195, 227 Clun (Salop.), 235 Cluny, abbey of, 284, 285 Cniht, cnihten, 65, 71 Cnut, king of Denmark and England, 3, 4, 16, 53, 246 Cockfield (Suff.), family, 201, 202, 213, 217, 220; manors and fee, 197-201, 210, 213, Adam of, 197-201; Margaret, daughter of, 197; Robert of, 198-201, Margaret, granddaughter of, 199, 200; Lemmer, Leofmaer, tenant of Cockfield, 199-201, 220 Cogges (Oxon.), barony, 86 Coke, Sir Edward, 32 Colchester (Essex), 133 Colne (Yorks.), valley of, 19

Index Colne, Richard of 254-55; Sibilla of, 255 Colswein of Lincoln, 104-11, 169; Picot, son of, 104; Muriel, daughter of, 104 Conde, 203 Conflent, castle, 213 Conisborough, castle, 132 Constabularia, 75, 83 'Continuity', between English and Norman institutions, xiii-xiv, 3-5, 17-18, 65-66, 74_76, 94-95, 103, 107, 109-11 Constitutio, concerning the division of fiefs attributed to Henry II, 119 Corbeuil, William de, 218 Corbuchon, fees of, 92, 176 Cork (Ireland), barony, 100 Cornwall, 40; earldom of 83; Reginald, earl of 170, 282 Cosford (Suff.),200 Cotentin, 212 Cottingham (Yorks.), 92 Coucy, Aubrey de, 115n. Coulombs, abbey of, 203, 208 County, court, 222; in Domesday Book, 41, 45; in Domesday survey, 34-37 Courcy, family, 256; Robert de, 189; William de267 Courtenay, family, 233 Cranach, Lucas, 21 Craven, 104n., 234 Crepy, Hugh de, 250n. Crewe, Lord Justice, 196 Crispin, family, 190; name, 190; Gilbert, 190, 191, Gilbert, abbot of Westminster, 138, 139, 191; Milo, precentor of Bee, 40, 190 Crok, Elias, 229 Croixmarez, Roger de, 203n. Cromwell, Oliver, 23 Cronne, H.A., 31, 145 Cross-channel tenure, 126-28, 141-42 Crowland (Lines.), abbey, 8n., 217n. Cumbria, 87,175,234 Custom, Anglo-Norman, xvi; and law, 118; in southern France, 293 Damietta, siege of, 240 Danegeld, see Geld Danelaw, 63, 71; six-carucate assessments in,110 Dapifer, Adam, 37; Eudo, 34 Darby, H.C., 40, 64 David, king of Scotland, 155, 279

331

Davis, R.H.C., 63, 113, 145, 158, 272, 284-85, 287n. Davies, R.R., 1, 2n. Debden (Essex), 324 Decapitation, 20-21; see also Halifax Derby, earls of, see Ferrers Derbyshire, 42n., 49n., 52 Dereham, Elias of, 295, 306 Devizes, castle, 132, 273n.; Richard of, 311 Devon, 39, 40, 62; sheriff of, 298; William, earl of, 298 DhondtJ., 78 Dialogus de Scaccario, see fitz Neal Disparagement, see heirs and heiresses Dispenser, Robert the, 34; office of 134 Diss, William of, 198, 199, 201, 217 Domesday Book, and Anglo-Saxon lordship, 5; as a carta, cartula, landboc, 47-49, 142; as descriptio, 47-48; as a feodary, 39; as a geld book, 38-39, 45-46, 64-67; as liberjudiciarius, 25; databases, 29; facsimiles, Alecto editions, 29, Ordnance Survey, 68; interest in, of ecclesiastics, 48, of subtenants, 50-53, of tenants-in-chief, 36-37, 48-50; purpose and use of, 28-29, 44-47, 54-57; warranty in, 48-52 Domesday Book, Great, clamores, 43, 56; date of completion, 27, 56-57; format, 25, 41-44; index lists, 41-42; order of folios, 56; rubrication, 41-47; scribes, 41-44, 56, written sources, 36, 38; Yorkshire 'recapitulation', 38, 43-44 Domesday Book, Little, date of completion, 35; facsimile 68n.; fee of Odo of Bayeux, 52; format, 25, 44; index lists, 44; relationship to Great Domesday Book, 26, 35; rubrication, 44 Domesday Survey, articles of enquiry, 26; circuits, 34-37; duration of 53-55; mode of 53-55; procedure, 26-27; 68-69; participation of churchmen, 37, of tenants-in-chief, 37 Dorset, 39-41,148; Philip of 301 Douglas, D.C., 33n., 115, 220 Dover, 22, 256, 281 Drengs, 71, 75 Duby, Georges, 114, 115, 163, 166n., 175, 235, 236n. Dunmere, family of 320; Agnes of, 320 Dunmow, Little (Essex), 99 Dunstable, 284 Durand, sheriff of Gloucestershire, 193

332

Colonial England, 1066-1215

Durham, cathedral and see, 177; tenants of, 75; Hugh du Puiset, bishop of, 314, 315; William of St. Calais, bishop of, 34, 37, 56; see also Flambard Eadmer, 50, 151 Earldoms, Anglo-Saxon, 169; Anglo-Norman, succession to, 114, 131, 135, 277 Earleyjohnof, 179 East Anglia, 25, 50n., 172 Echauffour, 248 Edgar, king of England, 163 Edinburgh, 'Scottish maiden', 20 Edward the Confessor, king of England, 3, 6,13,17, 26 (TRE), 39 (TRE), 165 (TRE), 169 Edward I, king of England, 13,17,18, 131, 232, 324; Elizabeth, daughter of, 131 Edwin, earl, 95 Eleanor, queen of England and duchess of Aquitaine, 78-79,125, 314 Ellis, H. 32, 37n. Elsenham (Essex) 313 Ely, abbey, 138; cathedral, 9; see of 9, 73, 74, bishop of see Longchamp Ely Inquest, 40 Emma, queen of England, 16, 246 Engels, F.J., 158-59 English, Barbara, 158 English language, 14-16,18,140,190, 193-94 1'Escalerie, Hardouin de, 93 Escheat, 45-46, 84, 89-90,127-79 Espec, Walter, 76, 91, 246n., 257, 258, 279 Essex, 25, 49n.; earldom of, 261, 267, 277, 310, 312, 321; third penny of, 277; shrievalty, 134; see also Mandeville, Henry of, 280 Eu, county of, 171, 302; counts of, 130,187, 190; Alice, countess of, 155, 206; Gilbert, count of, 191; Godfrey, count of, 191; Henry, count of, 205, 206; John, count of, 155,206 Evesham (Worcs.),9 Evreux, county of, 248; Amaury IV, count of, 233; William count of 210, 249; Amaury, nephew of, 249 Ewyas Harold, castlery of (Heref.), 51 Exeter, 6n., 8n., 13n.; Baldwin of, 194, and see Clare Exon Domesday, 27, 34, 40, 64 Eye (Suff.), castle of, 132; honour of 83

Eye (Middsx), 124 Eynsham, Adam of, 263-64 Eyton, R.W., 79n. Falaise, treaty of 273 Family, as a political influence, 224-26; and monastic endowment, 227-29, 238, 242-43; and pledging, 232-33; cadets; 235-36; solidarities, 226, 229-30, 231-32, 239-41; stresses within, 237-39; see also heirs and heiresses Family nomenclature, Anglo-Norman and Norman, historiography of, 179-80; evidence for 180-83; compared with pre-Conquest nomenclature, 183-86; nicknames, 184, 190-91, 193; patronymics, 191-93; toponymies, xvii, 117, 167-70, 301-2; in Normandy, 186-90; in post-Conquest England, 193-6, 242n.; collateral and marital inheritance of, 194-95 Family structure, in Normandy; 169-72; in Norman England, xiii, 161-62; in pre-Conquest England, 162-65, 167-69; pedigrees, 172-73, 195-96, 227-28, 241 Faringdon (Berks.), Worth in, 288 Faversham (Kent), abbey, 288 Fealty, 5, 31-32; see also Affidatio in manu\ Salisbury oath Fecamp, abbey of, 116n., 203-4, 207, 209; John, abbot of, 202, 207 Fee-farm, 157 Feodum, feudum, hereditary nature of, 63, 116-17,139-40,156-57, 202 Ferrers, fee of, 125, and s^Tutbury; Henry de, 37, 46, 49 and n.; William de, 135 Ferriby,Johnof, 295 Ferrybridge (Yorks.), 19 Feuchere, P., 78 Feudal incidents, see Escheat; Marriage; Relief; Wardship Feudal tenure, see Feudal incidents; Knight-service Feudal institutions, origins of, 65-66, 168-69 Filleul, Baldwin, 207 Finchale, St Godric of, 15 Final concords, 303-34, feet of fines, 303 Fines pro servicio, 73 Finn, R. Weldon, 33, 45n. fitz Alan, of Oswestry, 87, 88, 191, 256; Ruald, see Richmond, constable of fitz Baldwin, Robert, 132

Index fitz Count, Brian, 132 fitz Eustace, Robert, 195 fitz Gerald, Robert, 40 Fitz Gerold, family, 191; Warin, 242-43, 267-68, 280-81 fitz Gilbert, Baldwin, see Clare; John, 73, 161-62; Richard, see Clare fitz Hamo, Robert, 246n., 251; Mabel, daughter of 251 fitz Harding, Maurice, 89; Robert, 89, 273n., 282 and n.; charters to 273n., 282n., 286 fitz John, Eustace, 194-95, 265, 279-80, 285-86; Pain, 265 fitz Neal, Osbert, 287 fitz Neal, Richard, Dialogus de Scaccario of, 25, 44, 45, 299 fitz Osbern, William, 51, 174, 204, 215, 285; family/line of, 122, 124, 136, 170 fitz Osbert, John 295 fitz Pain, Geoffrey 91 fitz Peter, Geoffrey, justiciar and earl of Essex, 130, 133n., 135, 229n., 257, 267, 310-11, 314, 317, 319; and Beatrice (de Say), first wife, 130, 257, 260, 310-17, 319, 321-22; Matilda, daughter of, 324; Aveline, 2nd wife and widow of, 324 fitz Ranulf, William 218 fitz Reinfrey, Gilbert, 233, 234; William (of Lancaster), son of, 234 fitz Richard, Walter, 34 fitz Robert, Walter, 99 fitz Roger, Robert, 233 fitz Walter, family/house of, 139, 174; name, 117; Robert, 99, 131-33, 224-25, 238, 240, 267, 295, 321 fitz Warin, Fulk, 231-33 Five-hide unit, 71 n. Flambard, Ranulf, bishop of Durham, 13n., 55n.,76, 141-42, 145, 218 Flanders, county of, 16, 77; Robert, count of, 34, 53 Flemish, 14 Flinders, University, 29 Flodoard, 151 Forest, Charter of (1217), 294 Forests, private, wHardwick Forfeiture, 136-37 Forz, William II de, titular count of Aumale, 129, 264 Fossier, R., 115n., 214 Framlingham (Suff.), castle, 132 France, kingdom of, 54, 128, 140-41, 150,

333

172, 175, 185, 295; campaign in (1086-7), 54; Princess Alice of, 314 Francia, 144 Franks, 173 Free alms, 288-90; see also, Law, canon Freeman, E.A., 32 and n., 150, 172, 184 Freiburg school (of historians), 224 French (language), 13-18; toponymies, 140, 167-68, 193-94 Frenchmen, xv-xvi Gaelic, 14 Gaimar, 266 Galbraith, V.H., 28, 31n., 34n., 35n., 37-39, 41,42n.,44n., 67-69, 140 Gant, honour, 267; Gilbert de, 135, 267, 282-83n.; Juliana, daughter of 267; Robert de, 93, 282-83n. Garsington (Oxon.), 139 Gascony, 307 Gaskell, Mrs, 23 Gaul (Merovingian), 76 Geld, 61; gelding in the manor, 64-65 Genestal, R., 119, 148 Geoffrey the Chamberlain, 57 Germany, kingdom of, castles in, 7 Gervase of Canterbury, 9n., 10, 12n. Gesta Stephani, 275-76 Gherbod the Fleming, 115n. Giffard, barony, 259; family/line of 123, 170; name, 117, 188; Walter, earl of Buckingham, 37, 46, 190 Gilbert, knight and monk of Preaux, 203n. Gillingham, J., 66n. Giroie, 248; family of 214, 217n. Gisors, 314 Glanvill, Tractatus de legibus et consuetudinibus regni Anglie attributed to, 178, 257-58, 298-99, 303; date of composition (book 7), 310; on alienability, 146-47, 230 and n.; on the casus regis, 310-12; on inheritance and acquisition, 120-22, 146; on parage, 238n., 256n., on parceny, 148, 253-61; on primogeniture, 119-20, 146, 152; on representation, 310-12, 325-26 Glanville, Ranulf de, justiciar, 241, 257, 260, 312; his family, 229, 232 Glastonbury (Som.), abbey, 8n., 12, 40; St Benignus of, 246n. Glicourt, 207 Gloucester, courts held at, 27, 34, 55; abbey, 8n.; earldom and honour of, 83, 251,

334

Colonial England, 1066-1215

257, 321-22; Miles of, 1st earl of Hereford, 92, 137, 156n., 250, 265, 266n., 278; Robert, earl of, 137,170; William, earl of 238, 257, 281-82; Amicia, daughter of, 322-33; Isabella, daughter of, 257, 321,323 Gloucestershire, 35, 49n., 88 Godfrey of Cambrai, 90 Godwin, earl, house of 162 Godwin, the tailor, 303 Goldcliff, master Hugh of, 10 Gorron, 314 Gournay, Hugh I de, 190, 250n., 284n.; Gundreda, sister of, 195, 250n.; Hugh II de, 284-85n., 287 Goxhill, Peter of, 218n. Grace, see Saleby Grand Coutumier de Normandie, 309 Grandmesnil, family of 122, 124, 214; name, 187; Judith and Emma de, 247n.; Matilda de, 265; Robert de, 214; abbot Robert de, 229n., 247n. Grant, Lindy, 13n. Great Bealings/Thurlow (Suffolk), barony, 85 Great Cheverell (Wilts.), 89 Greenway, D., 177 Groton (Suffolk), 197,198, 200 Guader, Ralph, 95 Gullick, M., 56 Gunnor, duchess, see Richard I, duke of Normandy Halifax, parish and town (Yorks.), 19-23; guillotine (gibbet), 20-29 Hall, G.D.G., 299 Hall, Hubert, 81 Hallam, E., 33n., 47n. Hallamshire, 24 Hampshire, 148 Hanslope, Michael, of, 137; Matilda of, 137 Harcourt, Aubrey de, 228 Hardwick (Yorks.), forest of, 24 Harold, king of England, 15, 50n., 246; see also Vita Haroldi Harrison, William, 22-23 Hart, C.R., 64 Hartshill, William of, 263 Harvey, P.D.A., 304n. Harvey, Sally, 33, 36, 38n., 53, 64 Hastings, 76,163; Henry, count of Eu, lord of, see Eu; tenants of, 30 Hatton, Hugh of 206

Haveloc, Lai d\ 266, 313 Havelock, legendary prince of Denmark, 266 Haye, de la, barony of, 104-10; tenants and tenancies of, 104-8; Richard, 104-5, 108, 281; Robert, 104, 109 Hearne, Thomas, 81 Hebble Beck, Halifax, 24 Hedingham, barony, 84, castle, 7 Heirs and heiresses, 248-49, 251-55; disparagement of, 265-69; as hostages 234-45; see also Marriage; Wardship Helmsley, barony, 91, 258 Hemming's cartulary, 47-48 and n., 157 Henry I, king of England, 87, 89, 113, 129, 142-43, 257; and hereditary title and seisin, 145,147,156,199; and officials, 134; and Northumbrian and Cumbrian baronies, 87,154; and Robert fitz Hamo, 246n.; and succession to crown (1100-1106), 149-50, 181, 183; as antecessor of Henry II, 300-1; as terminus of good government (tempus regisHenrid), 275-76, 280-81, 300; control of marriage by, 250-251, 262, 265; coronation 'charter', xv, 17-18, 118, 128, 166, 219-20, 241, 250, 262; its relation to Magna Carta, 305; feudal service under, 74, 78-79, 82, 96, 98; grants of baronies by, 90-92,127, 137, 154-55,170,176,194, 227, 246; intervention in descent by, 124; OE diplomas of 13n.; ordinance of hundred, xv, 301; raises men from the dust, 265-66; Adela, wife of, 284; Maud, illegitimate, daughter of, 252 Henry II, king of England, xvi-xvii, 79,113, 129, 147n., 161,171,173, 257-58, 292, 297, 305, 307-9, 312; accession, 300-1; and earldoms, 151-52; and his family, 125-26; and hereditary office, 134-35; and recovery of castles, 132; and King Stephen, 73-74, 283-85; and territorial settlement (1154), 286-87, 279-83; as Henry fitz Empress, 144, 272-74, 280, 283-85; constitution concerning division of fiefs, attributed to, 119,148; coronation charter, 280-81; chronology of his acta, 290; feudal service under, 65, 73-74, 78, 81-82, 99-100,175; feudal service in Ireland, 100; household of 279-80; land disputes under, 282-83, 286-87; William, his brother, 280

Index Henry III, king of England, 253, 291, 306, 326 Henry, the young king, 308, 314 Henry V, emperor, 246 Henry, earl of Huntingdon, 282 Hereford, castle, 132; cathedral and see, 51, 139, 142; Robert Losinga, bishop of, 35-36, 39, 138-39; Giles de Briouze, bishop of, see Briouze; Roger, earl of, 51, 132, 278, 283, 286; Walter, brother of, 278; see also, Bohun; Gloucester, Miles of Herefordshire, 51; Herefordshire Domesday, 47 Hereward the Wake, 3, 76 Hertford, 133, 143 (house in); earl of, see Clare Hertfordshire, shrievalty of, 134 Hesdin, Arnulf, 88, 256; Maud and Aveline, daughters of, 88 Heugon, 248, 250n. Hides and hidage, carucage, 59, 61-63, 75; five-hide/carucate units, 74-75, 103-11 Holbeach (Lines.), 301; family, 301 Holderness, 85; honour of 130 Hollings, Marjory, 71-72, 75, 84n. Hollister, C. Warren, 71, 74, 78 Holme (Norf.), abbey, 139, 156-57; Anselm, abbot of 156; Richer, abbot of, 156; William abbot of, 156 Homage, 5, 31-32, 49-50, liege, 31, 53-54; see also Fealty; Salisbury oath Hommet, Richard de, 280 Honorius III, pope, 268 Hooton Pagnell (Lines.), 85n., 89; see also Paynel Hospitallers, 236 Hostages, 233-34; obsides fiducie, 273 Houghton Regis (Beds.), 284-85n. Howbridge, Gervase of, 295 Howden, Roger of, 309 Hrolfr, Rollo, duke of the Normans, 116, 174, 190, 246 Hudson, J.H., 221n. Hull, university of, 29 Humphrey, son of Siward, 303 Hunt, A., 16 Hundred and wapentake, in Domesday Book, 38-39, 42, 44; court of, 19 Huntingdon, honour of, 251; Henry of, 12, 275 Huntingdonshire, 42n., 43, 56 Huntingfield, William of, 295 Hurnard, N. 64, 78

335

Hyams, P.R., 55, 287n., 290n Hyde (I.O.W.), 8n. Infangtheof, 22-23 Inflation, price, 304—5 Inheritance, inheritances, propres, in England, 197-202, 216; in Normanday, 118-22, 202-5; and acquisition, xvi-xvii, 50, 63, 118-122, 146-47, 171-72, 197-205, 215-16, 230, 276-81, 287-88; and royal/ducal superiority, 204, 210-13; and monastic endowment, 203-5, 207-8; of castles, office, earldoms, 131-135; historiography of, 113-15; terminology of, in England, 141-45, 156-57, 197, 199-202, 216, 218-19, 277-78; in France, 144, 203-17; see also, Parceny; Succession Innocent III, pope, 295-96 Inquisitio comitatus Cantabrigiensis, 17, 35, 38-39, 53 Inquisitio Eliensis, 26, 40 Inquisitio geldi, 35, 38-39, 53, 67 Ireland, 126; conquest of, 100; feudal service in, 100; see also John, king lus commune, see law, canon Ivry, castle, 152; Ralph, count of, 249n., Aubree, wife of 249n. Jervaulx, abbot of, 231 John, king of England, 135, 304, 307-10; and aliens, 242; and claims to castles (1215), 133, 135n.; and freedom of election (1214), 294; and the government of England, 113-14, 292-93, 297; and hostages, 233-34; and marriage (disparagement) 264, 266-69; and papacy, 295-96; and renewal of charters/letters patent (1199-1200), 300; as count of Mortain, 181, 314-17; as duke of Normandy and lord of England, 317; as lord of Ireland, 126; as 'Prince', 181; fines and proffers with, 83, 230-33; government finances under, 304-5; rebellion against (1215), 182, 224, 238; succession and tide to throne, xvii, 130, 161-62, 239, 313, 316-17 John, Eric, 71, 72, 84n. JolliffeJ.EA, 113 Jones, G.P., lln. Judith, countess, 248; Adeliza, daughter of, 250

336

Colonial England, 1066-1215

Jumieges, abbey, 116n., 209; Albert, abbot of, 202; William of 158,170,173-74 Jurors, xiii, 21 Keefe, T.K, 80 Kendal, 233; see also fitz Reinfrey Kent, 35, 52; baronies in, 86,175 Ker, N.R., 34 Keynesian multiplier, 11 Kidson, Peter, 9n. King, Edmund, 273-74n. Kirkburton (Yorks.), 20 Kirkstall, abbey, 288-89 Knights' fees, and continuity with Anglo-Saxon tenures, 103-11 Knight-service, and hidage (carucage), 74-5, 103-4; institution of, 95, 220-21, in Ireland, 100; quotas, xv, 54-55, 82-94, 96-98; ignorance of, 99; term of, 74, 79-80 Knighton (place-name), 77 Knoop, D., lln. Kyme, Philip of, 104, 106n., 108; Simon of, 106n., 231; Philip, son of Simon, 231 Lacy (of Ludlow), lands of 282-83; Walter de, 50, 51; Roger de, son of Walter, 50, 139 Lacy (of Pontefract), family, 124, 195, 227, 280; estates/honour of, 85, 89-90, 96, 148, 195, 227, 262; tenants of 264; Alice de, 131; Henry de, 280, 289; Ilbert de, 195, 280; Robert de, 234; Roger de, 233 Lfl*nland, 65,169 Lambert of Ardres, 235 Lambeth, treaty of, 324 Lancashire, 24; gentry of north Lancashire, 234 Lancaster, castle, 132; honour of, 84,103n., 136; earls and dukes of, 19; Thomas, earl of 131; Alice, wife of, 131 Lancaster, William of, 234; see also fitz Reinfrey Landford, Ranulf of, 178 Lanfranc, archbishop of Canterbury, 150, 265 Langton, Stephen, archbishop of Canterbury, 133, 224, 268, 295, 323; Simon, 295 Language, xv, 13-17; Latin, 13-14, 16, 17; Middle English, xv; multi-lingualism, 16-17; Old English, survival of, 13, 15-17

Lanvallei, William de, 133 Launceston (Cornwall), barony of, 136 LaVal,Hughde, 262 Law, canon, 290, 296; as ecclesiastica justicia, 274; as ius commune, 299; influence of, on Magna Carta, 296; see also Free alms Law, common, origins of, 18; see aho Legal actions Legal actions, French v. English, xv-xvi; grand assize, 299; mort d'ancestor, 140, 197-98, 298-99, 302-3; see also Final concords Leges HenriciPrimi, 120-23, 215, 312 Legge, M. Dominica, 14n., 15n. Le Goulet, treaty of, 307, 317 Leicester, earldom of, 169; Robert Dudley, earl of, 23; see also Beaumont Leicestershire, 182 Leinster (Ireland), barony of, 100 Lemarignier,J.F., 63 Lennard,R., 156, 162 Leofmar, see Cockfield Le Patourel, J., In., 64, 149-51, 158-59, 272 LesAndelys, 132 Letcombe Regis (Berks.), 284 Lewes (Sussex), barony, 85n.; castle, 132; 'exchange' of, 50 Lewis, C.P., 51n., 56 Lichfield, cathedral and see, 284 Lincoln, cathedral church of, 6, 7, 9; Hugh of Avalon, bishop of, 263-64, 266; Remigius, bishop of, 6, 37; earls of, 135, see also, Albini; Roumare; see also Alfred; Coleswain Lincolnshire, 35, 43, 49n., 91, 233 Lindley (Yorks.), 178 Lindsey, 199 Lisle, Brian de, 264; Robert de, 152, Albreda, wife of, 152, and j^Tosny Lisoures, Fulk de, 281 Livry, 214 London, see of, 236; barony of bishop of, 97; Richard, bishop of, 97; William de Vere, canon of, 236, and see Howbridge; city, not in Domesday Book, 68; shrievalty of 134; Tower of, 133-34, 274, 277, 315; William I's writ for, 117 Lorraine, 77; Lower Lorraine, 123n. Lothian, 75 Louis, prince, later Louis VIII, king of France, 292, 324 Louvain, Jocelin de, 285 Loyd, Lewis, 167

Index Loyn, H.R., 55, 162,169 Lucy, countess, 94, 265; lands of, 154 Lucy, Richard de, 173, 280, 285, 312 Luddites, 23 Lusignan, Ralph de, 155 Lyon, Bryce, 78 Lyre, abbey of, 204, 215n. McDonald, J., 25, 29, 37n., 60, 63 Macfarlane, Alan, 63 McFarlane, KB., 114 McKechnie, W.S., 297 MagnaCarta (1215), 3, 18, 128, 133, 161, 174, 223-35, 242, 266-68; and charter of Henry I, 305; and the commune of the whole land, 305-6; and liberties of the church, 305; canon law influences in, 296; execution of 305-6; in perpetuum in, 294-96; liber homo in, 297; origins of 291-3; Twenty-five barons of, 174, 224-25, 305 Maine, county of, 33, 125, 149, 151, 248, 307; Herbert, count of, 151; Herbert II, count of 248, Margaret, sister of, 248; Hugh, count of, 248 Maitland, F.W., 28, 38, 59-69, 71, 95, 114, 116, 143, 148, 165,299 Malcolm Canmore, king of Scotland, 150 Malet, family, 123, 136; name, 117, 188; Robert, 37 Malmesbury (Wilts.), borough, 74; abbey, 78-79; William of, 8, 9n., 17, 74, 217n. Malton (Yorks.), 85n. Maltravers, William, 262, 264-66n. Mandeville, family, 123-24, 174, 239, 311, 324; claims of, 134; honour, estates, fees of, 130, 261, 267, 310, 314, 321; Arnulf de, 276n., 277; Geoffrey I de, 124; Geoffrey II de, 1st earl of Essex, 133n., 225, 273 and n., 276-77, 280, 286; Geoffrey III de, 2nd earl of Essex, 99, 243, 277, 280, 286; Geoffrey IV de, 5th earl of Essex, 131, 133, 174, 238, 240, 277, 321, 323; William I de, 277-78; William II de, 3rd, earl of Essex, 130, 135, 314, succession to, 135, 310-14, 316, 321; William III de, 6th earl of Essex, 240, 324, Matilda, sister of 324 Manor, gelding in the, 64-65 Mantes Gassicourt, 34, 53 Marches, Marchers, 14, 87 Maredudd, king, 51 Marie de France, Lais de, 261-62n.

337

Market Weighton (Yorks.), 260n. Marlborough (Wilts.), Alfred of, 51 Marmion, barony of, 85; name, 188 Marriage, 249-50; enforced, 264, 268; as a literary theme, 266 and n.; see also wardship Marshal, Geoffrey, 139; William, 4th earl of Pembroke, 73, 161, 236, 238, 259, 264, 267-68, 308, 315-16; William the younger, 238 Marshwood (Dorset), barony of, 92, 124 Martel, family name, 117 Matilda, Empress, 92, 134, 143-45, 273, 277-78, 280, 288, 300 Mauduit, name, 117, 137; Robert, 262, William 285 Mauleverer, family, 104n. Meath (Ireland), barony of, 100 Melksham (Wilts.), 257, 311 Mersey, R., 52 Meschin, Ranulf le, lord of Cumbria and subsequently earl of Chester, 87, 127, 154-55, 265; see also Chester, Lucy (his wife); William le, 256 Messina, 315 Meulan, county of, 54; exiled count of, 174; Waleran, count of 124, 153, 173, 290n.; for other counts of see Beaumont Meules, 196 Middlesex, 42n., shrievalty, 134 Midland (west) Domesday circuit, 37 Military service, abroad, 54-55, 100-101; in exerdtu et chevaltia, 87; see also Knight-service Milsom, S.F.C., 63, 157, 272 Milton Abbas (Dorset), abbey, 103 Mitford (Northumb.), barony, 87 Monmouth, 85n.; Geoffrey of, 173 Montagu, John de, 320 Montbrai, 195 Montdoubleau, Pain de, 88, 281 Monterolier, 212 Montfort, family/line of, 122, 124, 170; name of, 187; Amice de, 230; Guy de; 232; Simon Hide, 130 Montgomery, family/line of 122, 124, 136, 170, 174, 188, 196; name of, 187-88; barony of, 84; Hugh de, 2nd earl of Shrewsbury, 205, 216; Roger I de, 188; Roger II de, 1st earl of Shrewsbury, 37, 52, 86-87, 138, 168-69, 175, 188, 215n., 216, Mabel, wife of, 247n., 249n. Montreuil, 248

338

Colonial England, 1066-1215

Mont St Michel, monks of, 203; endowment of, 211 Morcar, earl, 95 Mortagne, Geoffrey de, 250n. Mortain, counts of 90, 136; family of, 174; lordship of 84, 95; fees of 86-87, 92, 97, 175; John, count of, see John, king; Robert, count of, 34, 40, 86, 91, 97,138, 141, 174, 176, 189; William, count of, 86, 97, 176 Mortemer, battle of, 190; castle, 275 Mortimer, family of, 87, 256; Hugh, 283; Ralph, 40; Richard, 101, 229n. Morville, Hugh de, 15n. Moulton (Lines.), 301, family, 301 Mountchesney, barony, 86 Mowbray, family 136; second house of, 91, 96, 176; honour of (Thirsk), 85, 86n., 91, 96; Robert de, earl of Northumbria, 96, 138; Roger de, 287; William de, 133 Mulgrave (Yorks.), 176 Nemo tenetur, 272 Nettlecombe (Som.),73 Neaufles, 191 Neubourg, 152,194; Robert de 152, 194 Neufmarche, Bernard de, 92,137, 278; Sybil, daughter of, 137, 250 Neuville, Richard de, 189 Neville, Adam de, 263; Hugh de 263 Newbury (Berks.), siege of, 161 Newcastle-under-Lyme (Staffs.), 132 'New men'/parvenus, 170, 176-78 Nonancourt, 310, 214 Norfolk, 40n., 50, 52; earldom of, 83,172; see also Bigod Norhamshire, 24 Normandy, duchy of, 8,154-55,181; expedition to (1086-87), 33-34, 54; loss of, 292; ducal superiority in, 204, 210-13, and Use majeste, 211; dynastic family in, 28,117,162,169-71, 214-15; feudal institutions in, 77; military tenure in, 79-80; hereditary tenure in, 115-16, 118-20,134,144-45, 201-5, 207-13, 216, 218; lordship in, 205; parage'm, 119, 146, 148; separation from England (1087), 142,161,171; succession to (1087), 149-51,158-59; dukes of see Hrolfr (Rollo); Richard; Robert; William Normans, and English in England, xiv-xvii, 4-5, 15; and the origins of feudal

tenure, 31-33; as a noblesse, 172-75, 184; see also Family structure; Inheritance; Normandy Norman Conquest, xiii-xiv, 2-5; tradition of, 173 Northampton, 132, 316; assize of, 298-99 Northamptonshire, 235 Northumberland, 22, 75, 87 Northumbria, 87,175; see also Mowbray Norwich, 6; bishop of, 73, barony of, 97; castle, 132; John, bishop of, 319; William, bishop of, 254 Nottingham, 19, 132 Nottinghamshire, 49n. Oakham (Rutland), hall, 12 Obsides fiduciae, 273 and n. Odo, see Bayeux; Odo, steward of Duke William 208-9, Stigand, his father, 209 Ogar the Breton, 90 Okehampton (Devon), 85n. Old Sarum, see Salisbury Old Warden (Beds.), 258, 260n. Orderic Vitalis, 123n., 124,132,150,174, 195, 213, 217 and n., 229, 242, 251, 265 OsbertofArden, 218 Oswaldslow (Worcs.), 65, 71 Oswestry (Salop), 87, 88 Oxford, 284; castle, 274; earldom, s#?Vere Oxfordshire, 49n., 51-52 Painter, Sidney, 113-14,163, 224-26, 292, 309 Palmer, J.N.N., 29n., 65 Palmer, R.C., 271-72, 304 Pamiers, Statute of (1212), 293, 295, 296-67 Pandulf, papal nuncio, 296 Pantel, William, 315 Parage, 119,148,166, 214, 253, 256 Parceny, 251-52; benefits of, 261-62; effect on feudal jurisdiction, 259-60; effect on feudal service, 256-57; variations in 257-58 Paris, Matthew, chronicler and chronicle of, 190, 268-69 Parvenus, see 'New men' Patricksbourne (Kent), barony of, 86 Patrimony, 154, 213, 235; division of, 166 Pattishall, Martin, 324; Simon, 319 Paynel, family, 282-83n.; fee, 89-91, 96, 103n.; Fulk, of Drax, 90; Hugh, of West

Index Rasen, 90; Ralph, 89; William, 89, 93; See also Hooton Pagnell Peak (Derbys.), 132 Peche, Alice, 240; Gilbert, 228, 257n., Joanna, wife of, 228; Hamo, 227, 259n.; see also Peverel of Dover Pennines, 19, 24 Penselwood (Som.), 320 Perche, la, Rotrou, count of, 170, 252, daughter of, 252; Thomas count of, 293, 296 Percy, barony of, 104n., 324, descent of, 320-21; Richard de, 320-21, 324; William II de, 321, 324, Agnes, daughter of, 324; William III de, 320-21, 324 Peter, knight of William I, 201, 220-21 Pevensey, barony of, 136; castle, 132 Peverel, William, 283 Peverel of Dover, barony, 86n., Hamo, 259; Pain, 227, 259; Robert, 259; William I, 218n.; William II, 227, 256, Rose, sister of, 227 Philip I, king of France, 27, 34, 53-54, 314 Philip II, Augustus, king of France, 77, 130, 293, 307, 314, 316, 318 Picardie, 115, 214 Picot, sheriff of Cambridgeshire, 193, 227; Hugolina, wife, of 227; Robert son of, 227 Pistes, Roger de, 51 Place-names, 179; see also Family nomenclature, toponymies Pledges, 232-33; see also Hostages Pleshey (Essex), 85n.; castle, 132 Plucknett, T.F.T., 78, 166n. Poitiers, William of, 50 Poitou, 77, 268, 322; language of, 14; Hildegarde, countess of 246n., Roger of, 51-52, 91, 138, 289n. Pont-Audemer, 124, 152, Turold of, 195-96 Pont de 1'Arche, William de, 262, 265-66 Pontefract, see Lacy Ponthieu, county of, 189, 312; Agnes of, 246 Port, honour of, 83, 147 Postan, M.M., 11,63 Postobitum gifts, 124, 146, 149, 151, 165-66 Potton (Cambs.), 168 Powicke, F.M., 224, 226, 292, 309 Preaux, abbey of, 152, 187, 203n., 208n.; Ansfrid, abbot of, 208n.; William, abbot of, 203n. Precaria, 63, 116n.

339

Prestwich, J.O., 77, 84n., 98 Propres, see Inheritance Puiset, Henry du, 260; Hugh du, see Durham Pyrford (Surrey) manor of, 27, 43n. Quarr (I.O.W.), 11 Quency, family, 324-25 and n.; honour of, 324-25; Robert de, 325 and n.; Roger de, earl of Winchester, 325 and n.; Saer, de, earl of Winchester, 230, 240, 264, 295, 324; Margaret, widow of, 241n. Radmanni, 65 Raleigh, Hug de, 73 Ralph the Staller, 95 Rames, Robert de, 93; Roger de, 93; William de, 93 Rampan, Hugh de, 177 Ramsbury (Wilts.), 8 Ramsey (Hunts.), abbey, 44n., 56, 157; Reginald, abbot of, 157n. Ranulf, earl of Chester, see Chester Ransom, 142 Rationabiliter, 279 Rayne (Essex), 93 Reaney, PH., 179, 185 Red Book of the Exchequer, 82, 103 Redmain, Henry, 234 Redvers, Baldwin I de, earl of Devon, 250n., 282; Baldwin de (d. 1216), 267; Margaret, widow of, 267-68; Richard I de, 282, Hawise daughter of, 250-51 Relief, 46 Rheims, council of, 246n. Reynolds, Susan, 62 Rhuddlan, 279 Ribble, R. 52 Riby (Lines), 91 Richard I, king of England, xvi, xvii, 126, 130, 132, 151, 172, 234, 300, 307-11, 317-18; and succession to throne (1189-99), 314-16 Richard I, duke of Normandy, 16, 161, 170, 191, Gunnor, wife of, 161-62, 170, 195, 241 Richard II, duke of Normandy, 116n., 144, 190, 204, 208-12; (or Richard III), 248 Richard Rolle of Hampole, 19 Richardson, H.G., 38 Richmond, honour of, 95, 104n., 287; Alan, count of 54; Ruald, constable of, 286; Alan, son, 286; Ruald fitz Alan, 326

340

Colonial England, 1066-1215

Richmondshire, 24 Richter, M., 14n. Ridel, Ralph, 264 Rievaulx, abbey, 257; Ailred of, 76 Ripon (Yorks.), minster, 12 Ripponden (Yorks.), 24 Robert I, duke of Normandy, 210-11, 250n. Robert II (Curthose), duke of Normandy, 125,129,149-51, 218n., 248 Robert, son of Aslac, 303 Roches, Peter des, bishop of Winchester, 255,268, 322 Rochester, castle, 133, 182, 234; cathedral, 6,13; Gundulf, bishop of, 6, 17 Roffe, David, 55-57 Roland, epic hero, 238; Song of, 238, 242 Rollo, wHrolfr Rollos, family, 286-87; Richard de, 286; William de, 90 Roman/civil law, 144 Ros, barony, 91,176; Robert de, 257-60n. Ros (of Lullingstone, Kent), family, 319-20 Rothwell, William, 13n., 15n. Rouen, cathedral and see, 132, 207, 209, 317; Mauger, archbishop of, 208; Robert, archbishop of, 214; Walter, archbishop of, 314; Roscelin, canon of, 207, Goudouin, his nephew, 207; Goscelin, vicomteof, 209n.; abbey of St-Amand, 34; abbey of St-Ouen, 116n., 203-4, 207-9; abbey of La Trinite du Mont, 170,186, 302-4, 209n., 210, 215-16; Rainer, abbot of, 186 Roumare, family, 135, Willam I de, 135 (as earl of Lincoln), 154, 250-51, 282; William II de, 94 Roumare, Roger fitz Gerald de, 94 Round, J.H., 172,179,184, 224; on Domesday, 28, 32, 37-38, 59-61, 63, 65, 68, 71, 81-83, 88, 94,103; on knight-service, 72, 75-76, 79; on Stephen's reign, 145, 271, 276 and n. Royal acts, chronology of, 290 and n., 300 Rumble, A.R., 42n., 56 Runnymede, 291-92, 323 Rye, Hubert of, 252; Agnes, wife of, 252 Saffron Walden (Essex), s^Walden St Albans, abbey, 10-11, 254; shrine of St Alban,11 St Brice, Adam de, 209 St Calais, William of, see Durham St Evroul, abbey, 211, 214, 229n.

St Florent-les-Suamur, abbey, 205-6, 215n., 219 St Leger, family, 302 St Martin, family, 302 St Mesmin de Micy, Albert, abbot of, 209 St Paul's, London, see London St Quentin, Dudo of, 116 StSwithin, 12 St Wandrille, abbey, 203-4, 207, 209, 211-14 St Wulfric of Haselbury, 15 Saleby, Thomas of, Agnes wife of, 263; Grace, suppositous daughter, 263-64 Salisbury/Old Sarum, 6, 8 and n., 27-28, 35, 47, 53-55,103n.; oath at (1086), 27, 31-35,137,142,175,189; cathedral church of, 6, 8 and n., 295; Herman, bishop of, 8; Jocelin, bishop of, 103n., 132, 273n., Osmund, bishop of, 8-9; Roger, bishop of, 8, 273n.; earls of 256; Earl Patrick of, 88, 99; William, earl of, 233, 264, 268 Sandal (Yorks.), castle, 19,132 Sandwich (Kent), 22 Santa Barbara, University of California, 29 Sarum, see Salisbury Saswalo, Henry grandson of, 125; Sewall, grandson of, 125; Fulcher, son of Henry, 125 Savoy, Peter of, 326 Sawbridgeworth (Herts.), 277, 316, 322 Sawyer, P., 28n. Say, family, 130, 311, 324; estates of, 313; Geoffrey de (senior), 130, 229, 310-11, 313-16, 322; Geoffrey de (junior), 311, 316, 322; William de, 130, 257, 310-13, Beatrice, daughter of, 257, 310, 314-15, Matilda, daughter of 257 Sayles, G.O., 38 Scandinavians, 3,169,173,184, 212; Vikings, 190 Scarborough (Yorks.), castle, 132 Sdrs, 24 Scutage, 72-73, 98-99,101 Sebricht, anchorite, 15 Seebohm, Fredereick, 62 Seez, seeof215n., 312 Selden, John (1584-1654), 32 Semer (Suff.), manor of 197-98, 200 Senliz, Simon de, 132, 251, 282; Maud, his wife, 251 Sens, William of, 9n., 10,12 Servitium debitum, 72-74, 82, 84, 91, 94, 97

Index Shaftesbury (Dorset), abbey of 103n.; Cecily, abbess of, 251 Sheehan, M.M., 122n., 165 Shepshed (Leics.), 240 Sherborne (Dorset), 8 Shrewsbury, 11, 88; abbey, 138, 205, 215n., 259; earldom of, 53, 136 and see Montgomery Shropshire, 51-52, 86-87, 168, 175 Sicily, kingdom of, xvii Sidgwick, Henry, 60 Sigy, abbey, 203 Simon le Bret, 318 Simon, son of Martin, 231 Siward Barn, 49n. Skipton (Yorks.), honour of, 104n., 130, 256 Snooks, G.D., 25, 29, 37n., 60, 63 Somercotes (Lines.), 231 Somerset, 42n., 75, 148 Sot, family, 319 Southern, R.W., 78, 113, 117n., 143, 292 South-western circuit (Domesday), 37-38 Sowerby (Yorks.), 24; forest of, 24 Sowerbyshire, 20n., 24 Soyland(Yorks.), 24 Stafford, 132; Robert of, 194 Stanley (Wilts.), abbey, 288 Stan ton Harcourt (Oxon.), 284 Statum decretum, 118, 252-55, 257; benefits of, 261 Stenton, P.M., 31, 38n., 63-64, 71-72, 74-78, 145, 156 and n., 179, 218n., 220, 252n. Stephen, king of England, 13n., 73-74, 77, 88, 92, 113, 132-35, 137, 144, 161, 172, 199-200, 271-74, 279-81, 283, 287, 298, 300, 317 Stigand, father of Odo the steward, 209 Stoneleigh (Warks.), abbey, 180 Stubbs, W., 32, 94-95, 271, 292; Stubbs's Charters, 31 Sturminster Marshal (Dorset), 169 Stuteville, family, 280; fee, 104n.; Robert de, 92, 280; Robert III de, 287 Styford (Northumb.), 87 Subtenants, undertenants, 86-87; and monastic endowment, 205-6; security of 137-41, 175-78 Succession, collateral and distant, 128-31, 214-15, 234; to crown, xvi-xvii, 149-51, 158-59, 307-17; in Anjou, Maine, Touraine, 307-8; in Brittany, 308; in

341

Normandy, 307—10; primogeniture, 119-20, 166; representative, 318-26; of women, 247 Suffolk, 40n., 44n. Surdeval, estates, 90; Richard de, 91 Surrey, 27, 56; earls of j^Warenne Sussex, 50, 56 Swanscombe (Kent), 86 Syderstone (Norfolk), 303 Tabuteau, Emily Z., 80, 158-59, 221 Taillebois, William, 91 Taisson, family, 188 Tait, James, 64 Talbot, barony, 86 Talvas, William, 247n., 312; Oliver, brother, of247n. Tamworth (Staffs.), barony, 85 and n. Tancarville, lords of, 228; William the Chamberlain junior, 228, 236 Tancred, king of Sicily, 315 Taunton (Som.), 27, 34 Tees, R., 25 Tempus guerrae, 281 Tengvik, G., 183, 185, 193 Tenure, a conquestu, 93, 145, 173; cross-channel, 126-28; see also inheritance, 141-42 Terrae normannorum, 143n. Tesson, Ralph, 214 Testaments, wills, 165-66, 168, 171 Tewkesbury, 79, 197-98, 201; abbey, 9n., 246n. Thame (Oxon.), abbey, 288 Thegns, thegnage, 75-76 Thirsk (Yorks.), wMowbray Thirsk, Joan, 62 Thompson, Faith, 291 Thorne, S.E., 63, 113-14 Thurketel, founder of Crowland, 217-18n. Triurkill (of Warwick), 194 Thursford (Norfolk), 254 Tickencote (Rutland), church of, 12 Tickhill (Yorks.), castle, 132, 171; honour of, 85-86n., 130,186, and see Bully Tillieres, castle, 191 Tinchebrai, battle of, 89, 92, 96, 176, 183 Tison, honour of, 104n.; Gilbert, 96 Tonbridge, lowy of, 192-93; see also Clare Torigny, Robert of, 170, 173-74, 195, 275-76, 280 Tosny, family, 170, 174; name, 186, 194; Berengar de, 152, 252, 289n., Albreda,

342

Colonial England, 1066-1215

wife of, 152; Ralph de, 194, 251, Adeliza, wife of, 251; Robert de, 252, Alice, daughter of, 252 Totenhall', 139 Totnes (Devon), 86n., 323; Joel of, 93, 194 Toulouse, 79, 287 Touraine, 268, 308 Tours, abbey of St-Julien, 209; city, 209 Tout, T.R, 1 Tracy, Henry de, 93 Trailly, family, 260; Geoffrey de, 258 Treaty, characteristics of, 273-74 TresAncien Cautumier, 120-21, 236-37, 308-10, 317 Tresgoz, Geoffrey, 254-55 Trowbridge (Wilts.), 85n., 133 Troarn, abbey, 168, 210, 215 and n., 216 Trumpington, 301; family, 301 Trussebut, 91, 96,104n., 260n.; Agatha, Hilary and Rose, heirs general, 228, 257n., 260n.; Robert, 256, 259 Turner, F.J., 1,2 Tutbury, honour of, 49n., 125, 135; and see Ferrers Twynham (Christchurch, Hants.), 250 Upleadon(Heref.),51 Vacancies, 46 Valognes, Peter de, 143, 145, 255; Robert de, 198n.; Roger de, 143, 252, 254; Walter, 254-56; family of, 254-56n.; Agnes, daughter of 254-56, 260 Van Houts, Elisabeth, 158-59 Varenne, 169 Vaudreuil, castle, 240 Vautort, family, 318 Vavassor, family, 104n.; Robert, 230-31; William, 231 Vavassores, 71

Vere, family, 123-24,174, 235-36; peerage case (1628), 196; Aubrey I de, 235, Roger, son of 235; Aubrey II de, 225, 235; Alice (de Clare), wife of 225; Aubrey III de, 1st earl of Oxford, 273n., 281; Agnes, wife of 246n.; Geoffrey de, 88; Robert de, 3rd, earl of Oxford, 295 Vesci, Vescy, family, 195; barony, see Alnwick, Malton; name, 195; Beatrice de, 194, 279; Eustace de, 132; William de, 194, 279, Yvo de, 279 Vexin, 33-34, 53 Victorian railways, 7

Vieilles, Humphrey de, 187, 195 Vieuxpont, Vipont, Robert de, 155 Vievre,211 VitaHaroldi, 3,15 Viterbo, cathedral, 232 Voluntas, 124 Wace, 172-73 Wager of law, 19 Wake, Hugh, 90 Wakefield, manor of, 19, 20; Peter of, 317, 318 Walden (Essex), castle, 132; priory, 313; chronicle, 267, 314-15, 317; manor, 277, 324 Waldin the Breton, 105n., 110 Wales, timber supplies from, 11 Wales, Gerald of, 14n., 126, 292 Wallingford (Berks.), castle, 274; honour of, 83,132 Walter, Hubert, archbishop of Canterbury, justiciar and chancellor, 267, 303, 308, 314, 316, 318; Theobald, 230 Waltham (Essex), 277 Walthamstow, manor of (Essex), 251 Waltheof, earl, 251; Countess Judith, wife of, 251; Adeliza, daughter of, 251; Maud, daughter of, 251, 282 Wansford(Yorks.),324 Wardship and marriage, 46, 128, 236-37; proffers for 261-65 Wareham (Dorset), 318 Warenne, family, 170; estates of 19, 24, 104n.; name, 169, 187; earls of, 19, 24; Reginald de, 275; William I de, 1st earl of Surrey, 50, 56,169,190 Wark(Northumb.),258 Warranty, by knife on altar, 254; by sword on altar, 228; see also Domesday Book Warren, W.L., 272 Warwick, earldom of, 169, 206; earls of, 174; see also Beaumont; Thurkill Warwickshire, 41, 49n. Wells, cathedral, 12; Hugh archdeacon of, 264 Welsh (language), 14 Welsh borders, marches, 87, 88 Wendover (Bucks.), manor of, 284-85n., 287 Wendover, Roger, chronicle of, 182, 268-69 Wergelds, 60 Westminster, 33, 35, 317, 322; abbey, 6,

Index 43n., 103, 124; abbot of see Crispin; hall, 9-10 Westminster, charter 'treaty' of (1153), 271-75, 299-300 Whalley (Lanes.), abbey, 227 Whepstead (Suff.), 163 White, Stephen, 156, 221n. Whiteknights (Reading), 179 Whitelock, D., 164, 185 Widows, widowhood, 247-50; fines for widowhood, 262-65 Wight, Isle of, 33 Wigmore (Heref.),87, 176 Willelmi articuli retracti, xiv-xv William, duke of Normandy, 1, 168, 178, 186, 189, 207-12, 246, 248, 250n. William I, the Conqueror, king of England, xv-xvi, 13, 25, 76; and Domesday, 27-28, 36-37, 39 (TRW), 46-47, 66, 68; and feudal service, 48, 71, 82, 94, 96, 115n., 176; and land settlement, 50-51, 68, 85, 115n., 189, 212, 215-16, 227; and overseas campaign (1086-87), 34, 53-55; and Peter, knight of, enfeoffed at Bury St Edmunds, 201, 320-21; and Salisbury oath (1086), 31-33, 46; claim and succession to England (1066), xiii-xiv, xvi, 3; death and succession to England (1087), 123n., 125, 149-51, 158-59; 'ten articles' of, 54; writs of 17, 143, 145 William II, Rufus, king of England, 9, 13n., 34, 56, 74, 85, 87, 89, 96, 98, 129, 142-43, 145, 149-51, 158, 176, 219, 227 William Clito, 129 William, knight of Glos, 217n. William Longchamp, bishop of Ely and chancellor, 314-15 William the Chamberlain, see Tancarville William the Englishman, 12 William, son of Warin, 177

343

Wilson, R.M., 14n. Wilton (Wilts.), abbess of, 252 Wiltshire, 25, 40, 88 Winchester, 12, 33, 35, 68 (see also Winchester, treaty of); castle, 274; 'charter' of (Domesday Book), 47; cathedral, 7, 9 and n.; old minster, 7, 12-13, 16; see, 27, 138; jElfsige, bishop of, 185; Henry of Blois, bishop of, 283; Peter des Roches, bishop of, 255, 268, 322; Walkelin, bishop of, 7, 34 Winchester, treaty of (1153), 271-76, 287-88; execution of, 283-85; and the church, 288-90 Windsor, castle, 322; Maurice of, 201 Wolferflow(Heref.),51 Women, descent of lands through, 248-49; legal and social position of 246-47; spinsters, 247; see also Heirs and heiresses; Widows Wood, S., 135 Wooton Courtenay (Som.), 268 Worcester, 79; cathedral, 6, 9 and n.; see, 61; feudal service of, 65; Red Book of, 109n.; tenancies of, 71-72, 75, 138, 142, 169; Wulfstan, bishop of, 61, 140 Worcestershire, shrievalty of, 134 Wormald, Patrick, 56 Wright, Martin, 32, 47n. Wright, Thomas, 21 Writs, legal, praecipe quod reddat, 297; of right, 298 Wroxall (Warks.), priory, 206 Wyville, Robert de, 177 York, assizes at, 23; ecclesiastical courts at, 20; minster, 9n.; see, 177; Geoffrey Plantagenet, elect of, 314; abbey of St Mary, 177; castle, 133 Yorkshire, 35, 38, 43, 52, 56-57, 89, 91 YverJ., 119n.