Lords of the Central Marches: English Aristocracy and Frontier Society, 1087-1265 (Oxford Historical Monographs)

LOR D S O F T H E C E N T R A L M A RC H E S OXF O R D H I S TO R I C A L M O N O G R A PH S Editors p. clavin r. j. ...

Author: Brock Holden

329 downloads 1136 Views 1MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

LOR D S O F T H E C E N T R A L M A RC H E S

OXF O R D H I S TO R I C A L M O N O G R A PH S Editors p. clavin r. j. w. evans l. goldman j. robertson r. service p. a. slack b. ward-perkins j. watts

Lords of the Central Marches English Aristocracy and Frontier Society, 1087–1265 B RO C K W. H O L D E N

1

1

Great Clarendon Street, Oxford ox2 6dp Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With ofﬁces in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York © Brock W. Holden 2008

The moral rights of the author have been asserted Database right Oxford University Press (maker) First published 2008 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose the same condition on any acquirer British Library Cataloguing in Publication Data Data available Library of Congress Cataloging in Publication Data Data available Typeset by Laserwords Private Limited, Chennai, India Printed in Great Britain on acid-free paper by Biddles Ltd., King’s Lynn, Norfolk ISBN 978–0–19–954857–6 1 3 5 7 9 10 8 6 4 2

In memory of Rees Davies

This page intentionally left blank

Preface For much of the twentieth century, the history of the March of Wales was written from the perspective of constitutional history. That approach was largely anachronistic, and was fatally undermined by a group of historians who, beginning in the 1980s, looked at England and its aristocracy within the context of the British Isles, or ‘Atlantic Archipelago’. This book is an attempt to examine the emergence of Marcher society as an organic process arising from the interaction of aristocracy, knights, Crown, and Welsh, and to trace some of the strands that went into its making by examining the ‘feudal matrix’ in which it was born. I was helped by a great many people in the course of writing this book. David Crouch and Nicholas Vincent generously shared their archival discoveries as well as their opinions. I beneﬁted immensely from discussions with John Maddicott, Paul Brand, and especially David Carpenter. Mistakes and errors of interpretation remain my own, of course. Kevin Shirley graciously made materials available when I was periodically cut off from research libraries. John Maddicott is particularly to be thanked for taking over editorial responsibilities in the last stage of the book’s preparation, a responsibility in which he showed great patience and forbearance with an author no longer in academia and well behind schedule. The staffs of the libraries and record ofﬁces that I consulted, particularly the Bodleian and British libraries as well as what was then the Public Record Ofﬁce in Kew, were uniformly of great assistance. Rosalind Caird graciously permitted me to consult the archives of the Dean and Chapter of Hereford Cathedral. The Rector and Fellows of Exeter College and the Committee for Graduate Studies in Oxford University generously provided funding for research and conference travel. In the course of bringing this book to completion, I left academia, eventually to ﬁnd myself in industry, and at the same time got married. My wife, Jody, herself a former academic, was tremendously encouraging and patient, and pushed me to ﬁnish the book when life and career brought more pressing concerns. I would not have ﬁnished without her help, not least in preparing the ﬁnal manuscript. This book is dedicated to my doctoral supervisor, the late Rees Davies. He did not live to see the completion of it, though he shepherded it

viii

Preface

most of the way. To anyone who knows Rees’s work, his inﬂuence will be apparent on every page. Rees is known as one of the great medieval historians of his generation; what those who knew him as a teacher, colleague, and friend also know is that he was a man of wide sympathies, always concerned with much more than the theses and careers of his students. He was not an academic empire-builder. His wide sense of perspective was perhaps unusual for the academy, and I for one owe him greatly for that. I hope that he would have been pleased with this book, and therefore it is dedicated to him. B.W.H.

Contents Maps List of Abbreviations 1. Introduction

xi xv 1

2. The Shaping of the Frontier, 1086–1166

12

3. Lords and Honours

46

4. The Knightly Class

88

5. The Role of Royal Government

137

6. The Region in the Reigns of Richard I and John, 1189–1216 166 7. The Region in the Reign of Henry III, 1216–1246

190

8. Conclusion

219

Appendix 1. Manors of the Magnate Families by County

230

Appendix 2. Genealogies

253

Bibliography Index

261 275

This page intentionally left blank

Maps

GW YN ED

CHESHIRE

YS

D

E

W

PSHIR

PO

SHRO

Maelienydd Wigmore Radnor

Builth

DE

HE

Elfael

UB

AR

Clifford

TH

SHIRE

HEREFORD Ewyas Lacy

Hereford

Brecon

PE

M

BR

erg

E

ny

Monmouth

en

av

OK

Ab

K

US

R

WE

GO

GLAMORGAN

NEWPORT CAERLEON Braose

Lacy

Clifford

Mortimer

Monmouth

Map 1. Lordships of the Middle and south-east March c.1200

xii

Maps

GW Y

CHESHIRE N

ED

D

YS

W

PO

R.

R. SE

SE

VE

RN

SHROPSHIRE

Builth

nyd

d R. LU G

R. TEME

Wigmore Radnor

G

DE

R. EITHON

HE

elie

N

A UB

VER

Ma

H RT

OW Elfael ARR R. HEREFORDSHIRE R. LU GG

Ewyas Lacy

Brecon

K

US

Monmouth

E

R. WY

R.

PEMBROKE

ABERGAVENNY GLAMORGAN

Map 2. The Welsh uplands above 600 feet

Maps

BUILTH

xiii

Radnor

ELFAEL

Bredwardine

Builth Painscastle

R. W

YE

Clifford Hay Bronllys

Breco

n

Penkelli

Tretower

BRECON Abergavenny R. US

Crickhowell

K

Map 3. Brecon and other Braose lordships, with sub-lordships and castles

xiv

Maps

GW

IRE

YN

ESH

D ED

CH

IRE PSH RO SH

PO W YS

Lands that did not come to the Crown through Braose expropriation

Builth

DE

HE

Elfael

Ra

dn

or

Ceredigion

UB

AR

HEREFORDSHIRE

TH

Ewyas Lacy

1 Brecon

R

WE

GO Carmarthen

GLAMORGAN

1 Blaenllyfni, granted to Peter fitz Herbert, 1211

Map 4. Land in royal hands in Wales, 1208–1215

ny

en

av

erg

Ab

List of Abbreviations ANS BBCS BF

BL Bodl. Brut c. CACW ‘Cart. Brecon’

CCR CIPM CPR CR CRR Davies, Age of Conquest Davies, Lordship DB

Anglo-Norman Studies Bulletin of the Board of Celtic Studies Liber Feodorum: The Book of Fees, commonly called Testa de Nevill, reformed from the earliest MSS, ed. H. C. Maxwell Lyte, 2 vols in 3 (1920–31), through-paginated British Library Bodleian Library Brut y Tywysogyon, or, The Chronicle of the Princes: Red Book of Hergest, Peniarth MS 20 version, ed. and trans. T. Jones (Cardiff, 1953) circa Calendar of Ancient Correspondence concerning Wales, ed. J. G. Edwards (Cardiff, 1935) ‘Cartularium Prioratus S. Johannis Evang. de Brecon’, ed. R. W. Banks, Archaeologia Cambrensis, 4th ser., 13 (1882), 275–308; 14 (1883), 18–49, 137–68, 221–36, 274–311 Calendar of the Charter Rolls preserved in the Public Record Ofﬁce, 6 vols (1903–27) Calendar of Inquisitions Post Mortem, 22 vols (1904– ) Calendar of the Patent Rolls preserved in the Public Record Ofﬁce (1891– ) Close Rolls of the Reign of Henry III preserved in the Public Record Ofﬁce, AD 1227 [–1272] (1902–38, 1975) Curia Regis Rolls preserved in the Public Record Ofﬁce, ed. C. T. Flower, 17 vols (1922–72) R. R. Davies, The Age of Conquest: Wales 1063–1415 (Oxford, 1987) R. R. Davies, Lordship and Society (Oxford, 1978) The Herefordshire Domesday Book, Alecto Historical edn, introd. Ann Williams (London, 1988)

xvi DNB Eyton, Antiquities EcHR EHR Glanvill HCA hd HDB HRO Lloyd, History of Wales Mon. Ang. P&P RBE Rot. Chart. Rot. Cur. Reg. Rot. Lib. Rot. Litt. Claus. Rot. Litt. Pat. Rot. Ob. et Fin.

List of Abbreviations Dictionary of National Biography: from the earliest times to 1900, ed. Leslie Stephen and Sidney Lee, 22 vols (Oxford, 1967–8) R.W. Eyton, Antiquities of Shropshire, 12 vols (1854–60) Economic History Review English Historical Review Tractatus de Legibus et Consuetudinibus Regni Anglie qui Glanvilla Vocatur, ed. and trans. G. D. G. Hall (1965) Hereford Cathedral Archives, Hereford hundred Herefordshire Domesday, c.1160–70, ed. V. H. Galbraith and J. Tait, Pipe Rolls Series, new ser. 25 (1950) Herefordshire Record Ofﬁce Sir John Lloyd, A History of Wales from the Earliest Times to the Edwardian Conquest, 3rd edn, 2 vols (1939), ii Monasticon Anglicanum, ed. W. Dugdale, rev. J. Caley et al., 6 vols in 8 pts (1817–30) Past & Present The Red Book of the Exchequer, ed. H. Hall, 3 vols, Rolls Series, 99 (1896), through-paginated Rotuli Chartarum 1199–1216, ed. T. D. Hardy, Record Commission, 36 (1837) Rotuli Curiae Regis 1194–9, ed. F. Palgrave, 2 vols, Record Commission, 27 (1835) Rotuli de Liberate ac de Misi et Praestitis Regnante Johanne, ed. T. D. Hardy (Record Commission, 1844) Rotuli Litterarum Clausarum in Turri Londinensi Asservati, ed. T. D. Hardy, 2 vols, Record Commission, 24 (1833–4) Rotuli Litterarum Patentium in Turri Londinensi Asservati, ed. T. D. Hardy, Record Commission, 28 (1835) Rotuli de Oblatis et Finibus . . . regis Johannis, ed. T. D. Hardy, Record Commission, 30 (1835)

List of Abbreviations RRAN TNA TRE Trans. Woolhope TRHS WHR

xvii

H. W. C. Davis, C. Johnson, H. A. Cronne, and R. H. C. Davis (eds), Regesta Regum AngloNormannorum, 1066–1154, 4 vols (1913–69) The National Archives, Kew tempore Regis Edwardi —Domesday abbreviation denoting what a piece of land was worth during the reign of the Confessor Transactions of the Woolhope Naturalists’ Field Club Transactions of the Royal Historical Society Welsh History Review

This page intentionally left blank

1 Introduction The area of the world we call Europe, and the roots of its culture, were in no small measure the creation of the High Middle Ages, the ﬁrst period of European expansion. That expansion was important, for prior to it there existed a number of different Europes: Frankish, Celtic, Slavonic, and indeed Muslim. Between 950 and 1350 the area encompassed by ‘Latin Christendom’, that Frankish part of Europe that had been shaped and inﬂuenced by the former western Roman and Carolingian empires, nearly doubled in size.¹ In the process, the culture which was Frankish or Latin came largely to deﬁne Europe, not just through conquest but also through assimilation, adoption, and attraction. Along the edges of this Frankish core, aristocratic lords, bishops, members of religious orders, and ordinary peasant settlers moved into new lands, carving out lordships, episcopal sees, monasteries, and farmsteads. But the lands into which they moved were rarely, if ever, vacant. Instead, they found peoples—Slavs, Muslims, Celts—with political and social structures that could differ vastly from those of the interlopers. Conﬂict was often the result. That conﬂict, in turn, helped to shape the character of medieval frontier societies. This book is an examination of one small part of this process of expansion and conﬂict, and one society shaped by it—that of a section of the March of Wales from 1087 to 1265. The Frankish European expansion of the High Middle Ages was marked by several salient characteristics. First, there was what Robert Bartlett has termed an ‘aristocratic diaspora’, in which nobles primarily from the north of France found themselves ﬁghting and often settling in diverse areas touching on the core of Latin Christendom, from the Crusader kingdoms of the Holy Land to Ireland.² As the Frankish aristocracy was a military caste, they brought with them military manners ¹ Robert Bartlett, The Making of Europe: Conquest, Colonization and Cultural Change 950–1350 (Princeton, 1993), 292. ² Ibid. 24–59.

2

Lords of the Central Marches

and technology marked by the use of cavalry and castles. In their wake came the economic, religious, and intellectual culture of northern France, and in turn, the Franks and their culture were inﬂuenced by what they encountered along the frontiers of Europe. In the island of Britain, the Anglo-Saxons had confronted the Welsh, or Britons, for centuries before the ﬁrst Normans arrived in England in the eleventh century as guests of Edward the Confessor. At times, the English kings asserted a general claim to rule over the entire island, but their occasional political rhetoric was not matched by any actual control of Wales. Instead, partial English expansion into Wales would occur after the Norman Conquest, as part of the broader expansionary movement of the High Middle Ages described by Bartlett. The history of attempted English expansion, sometimes successful, sometimes not, within the islands of what has been labelled the ‘Atlantic Archipelago’ or the ‘Atlantic Isles’ has recently been treated by a number of formidable historians. The overarching framework has been most ably drawn by the work of two men, Rees Davies and Robin Frame. What is attempted in this volume is to ﬁll in a small piece of a larger puzzle whose outlines have already been drawn, while perhaps seeing if that outline should be modiﬁed in places. This book thus investigates one period and one place in the history of the ‘First English Empire’.³ It is a study of English aristocratic society along the contested and rather amorphous frontier between England and Wales from the late eleventh to the mid-thirteenth centuries. The section of that frontier that concerns us ran roughly from the southern part of Shropshire through Herefordshire and the corresponding and contiguous Marcher lordships: Radnor, Builth, Elfael, Brecon, Clifford, Ewyas Lacy, Abergavenny, and Monmouth (see Map 1). Why aristocratic society? The answer is a simple one: in a hierarchical society elites matter. This work examines that aristocratic power and how it was structured, both formally and, more crucially, informally. Terminology poses a problem as it always does for the historian. Firstly, there is no good phrase for the geographic area examined in this book, for to a degree it falls in what has been labelled the ³ The title of the printed version of the 1998 Ford Lectures in the University of Oxford; R. R. Davies, The First English Empire: Power and Identities in the British Isles 1093–1343 (Oxford, 2000). Davies’s work is extensively drawn upon here; see the works cited in the Bibliography. For the work of Robin Frame, see The Political Development of the British Isles 1100–1400 (Oxford, 1990); and the essays collected in Ireland and Britain 1170–1450 (1998).

Introduction

3

‘Middle March’, while it also impinges on what can be considered south-east Wales. I ask the reader’s forgiveness for the rather awkward geographic labels occasionally deployed here. The title of this book itself acknowledges that in many respects there was no single ‘March’, but rather many ‘Marches’, changing with the tide of local power and the ﬂow of events. For much of the twentieth century modern scholarship largely ignored the March. The reason for this is simple: viewed from the centre, the March was a constitutional anomaly, one that ultimately did not contribute to the shape of the late medieval and early modern English state. Two explanations were advanced for the peculiar character of the March, both taking what can be termed a constitutional perspective. The ﬁrst explanation saw the Marcher lordships as being created by the grant of ‘palatine’ powers immediately after the Norman Conquest.⁴ The second, put forward by J. G. Edwards, saw Marcher powers as originating in the authority of Welsh rulers, with new Norman English lords assuming those powers when they conquered native political units.⁵ Neither of these explanations is entirely adequate. It was Rees Davies, in a seminal article published in 1979, who argued that the Marcher lordships had to be seen as the result of an unfolding historical process that occurred in response to the particular circumstances in which English lords and settlers found themselves in conquered Wales.⁶ Davies convincingly sketched out the elements of this historical process, and proposed that further study was needed of the ‘feudal matrix’ in which the March and Marcher powers had formed. Davies’s article followed the publication the previous year of his Lordship and Society in the March of Wales 1282–1400, which had outlined and analysed among other things how lordly power in the March operated: militarily, economically, and politically.⁷ That book in many ways forms the template for this one, as it takes the approach used in Lordship and Society and attempts to utilize it to view the development of Marcher lordship as a historical construct that evolved through time ⁴ An example of this view is W. E. Wightman, ‘The Palatine Earldom of William ﬁtz Osbern in Gloucestershire and Worcestershire, 1066–1071’, EHR 77 (1962), 6–17. ⁵ J. G. Edwards, ‘The Normans and the Welsh March’, Proceedings of the British Academy, 42 (1956), 155–77. ⁶ R. R. Davies, ‘Kings, Lords and Liberties in the March of Wales, 1066–1272’, TRHS, 5th ser., 29 (1979), 41–61. ⁷ R. R. Davies, Lordship and Society in the March of Wales 1282–1400 (Oxford, 1978).

4

Lords of the Central Marches

in response to urgent needs and practical necessities. It therefore tries to fulﬁl Davies’s call for an examination of the ﬁrst centuries of Marcher lordship and the ‘feudal matrix’ in which it formed.⁸ In attempting to fulﬁl Davies’s desiderata, this book draws on a number of critical studies, while trying to take their insights and draw a comprehensive picture of the development of Marcher society. These works have tended to examine either individual families or lordships, particularly as the lordship, or feudal honour, that social unit organized around obligations of military service owed in return for the holding of land from a lord, was the fundamental building block of Marcher society. Valuable family studies have been contributed by Michael Altschul on the Clares, I. W. Rowlands on the Braoses and Brecon, J. J. Crump on the Mortimers, and W. E. Wightman on the Lacys.⁹ The strength of some of these family approaches (notably that of Altschul) has been to locate the Welsh holdings of individual families within the wider context of their concerns in England, Normandy, and Ireland. As such, these studies remind us that the March was never wholly isolated from the wider events of the period. The drawbacks for the historian of the March are that these investigations do not concern themselves with how Marcher society formed. Of the studies that treat a Marcher lordship, the earliest and perhaps most valuable is William Rees’s ‘The Medieval Lordship of Brecon’, which charts the subinfeudation of Brecon.¹⁰ I. W. Rowlands’s ‘The Making of the March’ draws attention to the crucial role of the knights, the ‘men on the spot’ as he terms them, in the development of the March.¹¹ Frederick Suppe’s Military Institutions on the Welsh Marches takes the step of looking across more than one lordship, albeit with a limited, if important, focus on military organization.¹² ⁸ Davies, ‘Kings, Lords and Liberties in the March of Wales’, 49 and n. 38. ⁹ Michael Altschul, A Baronial Family in Medieval England: The Clares, 1217–1314 (Baltimore, 1965), is particularly good on the family’s Welsh lands; I. W. Rowlands, ‘William de Braose and the Lordship of Brecon’, BBCS 30 (1982–3), 123–33; J. J. Crump, ‘The Mortimer Family and the Making of the March’, in M. Prestwich, R. H. Britnell, and R. Frame (eds), Thirteenth Century England VI (1997), 117–26; W. E. Wightman, The Lacy Family in England and Normandy 1066–1194 (Oxford, 1966). ¹⁰ W. Rees, ‘The Medieval Lordship of Brecon’, Transactions of the Honourable Society of Cymmrodorion (1915–16), 165–224. ¹¹ I. W. Rowlands, ‘The Making of the March: Aspects of the Norman Settlement of Dyfed’, ANS 3 (1981), 142–57. ¹² Frederick C. Suppe, Military Institutions on the Welsh Marches: Shropshire, AD 1066–1300 (Woodbridge, 1994).

Introduction

5

Because Marcher society emerged over time in dialogue with the larger course of political and social events in England, Wales, and beyond, it is also necessary to draw on the work of primarily narrative histories in addition to detailed studies of families and lordships. The most indispensable of these narrative works is J. E. Lloyd’s classic A History of Wales, now supplemented (and heavily but accessibly mixed with analysis) by Davies’s The Age of Conquest.¹³ On a broader geographical scale, David Carpenter’s The Struggle for Mastery ably incorporates Wales and the March into the wider course of ‘British’ history, with due attention also paid to the impact of Continental affairs.¹⁴ Where this book differs from the family and lordship studies listed above is in its attempt to examine many different aspects of Marcher lordship as exercised by several families over time. It is not a study of a single family or lordship, nor is it a county study of Herefordshire. Instead, by looking at a group of families and lordships over some two centuries, it identiﬁes the various challenges that led to the creation of Marcher society and shows how Marcher lords dealt with those challenges. We must also engage with some of the larger questions of English social history, particularly the history of the feudal honour and the development of the common law and its impact on aristocratic society, for Marcher society developed in dialogue with these questions. To gain the entire picture of Marcher lordship we have to look at more than one family and more than one lordship, and we have to incorporate military imperatives within a larger social framework. By 1166 the lordships of the south-eastern March and the adjacent border regions of Herefordshire and southern Shropshire were already the product of a century of interaction between the English, the Welsh, and the landscape. The landscape of the region must be examined as an equal contributor to the frontier society that took shape there, for lordship of land, dominium terrae, was ineluctably shaped by the terrain in which it was set, especially in the border regions of the Welsh March. The character of this landscape as one moved from western Herefordshire into Wales largely determined where, and to what degree, extensive English lordship could be exercised. The river valleys provided the invasion routes, while the upland zones were resistant to conquest ¹³ J. E. Lloyd, A History of Wales from the Earliest Times to the Edwardian Conquest (1939); R. R. Davies, The Age of Conquest: Wales 1063–1415 (Oxford, 1987). ¹⁴ D. A. Carpenter, The Struggle for Mastery: Britain 1066–1284 (2003).

6

Lords of the Central Marches

and only ﬁtfully controlled and exploited by the English. These upland pockets of land would later be called ‘Welshries’, and were under de facto Welsh control until the thirteenth century and beyond. Anglo-Saxon expansion towards Wales had long washed against the hill country that rises on the western edges of Shropshire and Herefordshire. Indeed, it was along this natural division that the Mercian king Offa had erected the boundary marker of Offa’s Dyke.¹⁵ To the east of this line lay the fertile, though still heavily wooded, plains of Herefordshire and southern Shropshire. Compared to many counties in the south-east and eastern Midlands, Herefordshire had a high proportion of woodland when the Normans arrived.¹⁶ In terms of agriculture, it was a comparatively underdeveloped county. To the west, in the March, lay the upland commotes of Wales, partly characterized by a pastoral economy of isolated upland farmsteads, with arable farming in the lowland river valleys. The two economies were linked by the Welsh habit of transhumance. The dividing line between the two geographic zones occurs roughly at the 600-foot mark in elevation. Southern Shropshire is also dominated by high ground, reaching out from Wales. Herefordshire has isolated clusters of hills, particularly in the north-east, but the true start of high ground is on a line running from Wigmore in the north, to Clifford on the Wye, and then back south-eastwards to Ewyas Harold. Several river valleys pierce these upland areas: the Wye, the Llynﬁ, and the Usk are the most important ones. Anglo-Norman expansion followed the routes of these rivers, and settlement occurred in the valleys. For example, Bernard de Neufmarché attacked the ancient Welsh kingdom of Brycheiniog from western Herefordshire up the river Wye, building castles at Hay and Glasbury, and then, following the course of the Llynﬁ, captured Talgarth and erected a castle at Bronllys. The fate of Brycheiniog was sealed at a battle fought outside the modern town of Brecon in 1093, near the conﬂuence of the rivers Usk and Honddu.¹⁷ Rivers therefore largely determined the direction of English conquest and settlement in this section of the March. ¹⁵ What follows is based largely on H. C. Darby and I. B. Terrett (eds), The Domesday Geography of Midland England (Cambridge, 1954), chs 2 (Herefordshire, by C. W. Atkin) and 3 (Shropshire, by V. A. Saunders); Dorothy Sylvester, The Rural Landscape of the Welsh Borderland: A Study in Historical Geography (1969); and Roy Millward and Adrian Robinson, The Welsh Borders (1978). ¹⁶ Sylvester, The Rural Landscape of the Welsh Borderland, 357. ¹⁷ R. R. Davies, ‘Brecon’, in R. A. Grifﬁths (ed.), Boroughs of Medieval Wales (Cardiff, 1978), 47–8.

Introduction

7

The population ﬁgures which can be tentatively deduced from Domesday Book follow logically from this topography. In Shropshire the most densely populated areas, with ﬁve to six people per square mile, were to be found immediately to the south of Shrewsbury.¹⁸ Further south, population declined, only achieving the density seen at Shrewsbury on either side of the line of hills which separates Ludlow and Staunton Lacy from Cleobury Mortimer. In Herefordshire, Domesday population density reached from six to eight people per square mile in a broad swathe running across the county from north to south. In the west, however, this density dropped to three around Clifford and much of the surrounding area of Stretford Hundred, and one in the area bordering on the uplands of Radnor. No information was entered in the Domesday inquest for the areas of Ewyas and Archenﬁeld, newly taken from the Welsh. The county of Herefordshire was marked largely by dispersed settlement; even by the nineteenth century, there were only six nucleated settlements in the eastern half of the county.¹⁹ One sign of the militarized nature of the border with Wales is that the closer one comes to the March, the more likely it is that settlements will be nucleated, presumably for protection. This same pattern of increased nuclear settlement closer to Wales holds true for southern Shropshire. In Wales to the west of Herefordshire and Shropshire the terrain changes dramatically (see Map 2). Upland moors are periodically broken by river valleys, such as those of the Wye and the Usk. In the medieval period these river valleys were fertile but heavily forested. Beyond the valleys, regions such as Maelienydd, Elfael, and Radnor were almost entirely upland moors, many of them over 1,000 feet in elevation. The arable area was limited to the valleys of the rivers Lugg, Arrow, Eithon, Wye, and Teme. The lack of arable land in the lordship of Radnor undoubtedly contributed to the low number of knights’ fees listed there in 1211–12: ﬁve and a half.²⁰ These fees were spread over a large area, for more land was needed to support a knight’s fee in this area than in the fertile, more intensively cultivated lowlands of Brecon and Abergavenny. In Brecon the conﬂuence of the rivers Wye and Usk opened up a broad and potentially very productive valley while the ¹⁸ It must be remembered that the Domesday inquest only counted perhaps a quarter of the population. ¹⁹ Sylvester, The Rural Landscape of the Welsh Borderland, 350–1. ²⁰ RBE 603. See also Ch. 3.

8

Lords of the Central Marches

Epynt Hills, Forest Fach, and Forest Fawr provided good pastureland as well as hunting areas.²¹ Another feature, albeit man-made, that aided invaders and inﬂuenced settlement patterns was the old system of Roman roads. These roads mainly ran north to south in two parallel routes, one of which ran from Chester past Leintwardine and Wigmore to the Roman town of Magnis (just west of Credenhill) and on into Gwent. From Magnis a road ran west, following the Wye to Y Gaer (Brecon) and ending at Nedum (Neath). The Roman route along the border from the Wigmore basin to the Wye was, to judge from the location of hill forts, later used by the British after the Roman withdrawal in the ﬁfth century. The importance of this ancient road to the English is apparent from the density of castles, mottes, and nucleated settlements along the route, making this area ‘the most heavily and deeply defended stretch of the entire Borderland’.²² These same routes were still in use in the early fourteenth century.²³ The ﬁve baronial families we will examine—the Braoses, Cliffords, Lacys, Monmouths, and Mortimers—not only were the most important along this stretch of the frontier, but form a convenient group which is identiﬁable from the records of the period. When King John died in 1216, the executors of his will included the loyalist Marchers Walter de Lacy and John de Monmouth. Hugh de Mortimer also supported the king in the civil war of 1215–18. These men were among the barones de Marchia, a group which had emerged as a recognizable section of the English baronage by the beginning of the thirteenth century, similar to the ‘Northerners’ identiﬁed by Sir James Holt.²⁴ The ﬁrst occurrence of the phrase barones de Marchia, in a writ to the men of Rye in 1217, speciﬁcally names Walter de Lacy, Hugh de Mortimer, Walter and Roger de Clifford, and John de Monmouth.²⁵ Grouped together among the witnesses of the 1216 version of Magna Carta are Walter de Lacy, ²¹ Although the wealth of Brecon was not fully developed until the 14th century, its potential would not have gone unnoticed by the ﬁrst Anglo-Norman adventurers in the area and was undoubtedly under development in the period covered in this book; for Brecon at full potential, see Davies, Lordship and Society in the March of Wales, 107–29. ²² Sylvester, The Rural Landscape of the Welsh Borderland, 362. ²³ See the Gough Map in the Bodleian Library; Ordnance Survey reproduction in R. H. Hilton, A Medieval Society: The West Midlands at the End of the Thirteenth Century (1966), pl. 4. ²⁴ For the use of the term ‘Nordenses’ at the beginning of the 13th century, see J. C. Holt, The Northerners: A Study in the Reign of King John (Oxford, 1961; repr. corr., Oxford, 1992), ch. 2. ²⁵ CPR 1216–25, 108–9.

Introduction

9

Roger de Clifford, Hugh de Mortimer, John de Monmouth, William de Beauchamp, and Walter de Clifford. In 1225 we ﬁnd, likewise in order, Reginald de Braose, John de Monmouth, John ﬁtz Alan, and Hugh de Mortimer.²⁶ In their grouping together in the witness lists at the great occasions of the reissues of Magna Carta, we can picture these men standing together in a knot, perhaps a little to the side of the other barons. These were men who knew each other well, whose families were intermarried, who had at times fought alongside each other against the Welsh, and who came from a society shaped by the demands of a militarized frontier. Although lords such as Walter de Lacy and William de Braose held lands elsewhere, in England, Ireland, and before 1204 in Normandy, it was in many ways their position as lords of the March that brought them together and gave them a common identity. This identity was reinforced by feudal geography, for the lordships of the ﬁve families in the border counties and the March were contiguous. They not only shared a common mentalité, but were also neighbours. To say that these men formed an identiﬁable group should not be read to mean that they were always a uniﬁed community, particularly in the political sense. The families examined in this book all had their own individual agendas, based on family interests. Sometimes this brought them together; for example, the Braoses and the Mortimers probably cooperated in the 1140s and again in the 1190s, when the two families were attempting to expand into Elfael and Maelienydd, respectively.²⁷ Marcher lords and families could also share and take on responsibilities for each other. In the ﬁrst few years of the thirteenth century William de Braose left his Irish lands in the honour of Limerick in the care of his son-in-law Walter de Lacy.²⁸ Families could cooperate to safeguard each other’s interests, yet to expect these families always to act in unison is unrealistic. Self-interest could overpower ties of kinship as, for example, when Roger de Mortimer, whose son was married to a Braose daughter, prudently kept to one side in the feud between King John and William III de Braose that resulted in the latter’s downfall in 1210.²⁹ In this same episode we ﬁnd hints that the ties between Braose ²⁶ For the texts of the reissues, see J. C. Holt, Magna Carta, 2nd edn (Cambridge, 1992), 501–11. ²⁷ J. B. Smith, ‘The Middle March in the Thirteenth Century’, BBCS 24 (1970–2), 77–81. ²⁸ A. J. Otway-Ruthven, A History of Medieval Ireland (1968), 77. ²⁹ Roger’s son and heir Hugh was married to Annora de Braose, though the date of the marriage is uncertain.

10

Lords of the Central Marches

and Walter de Clifford were not close enough to prevent Clifford from siding with the king.³⁰ So the community of Marcher barons, despite shared concerns such as the threat posed by the Welsh, was often a latent one. It was latent in the sense that it often lay below the surface of daily routine. It took certain stimuli, either internal or external, to activate it. Thus, Welsh aggression or royal challenges to what were coming to be called ‘Marcher liberties’ would cause these lords to band together, and political cooperation was more likely within this group than outside it. The bonds of frontier society formed a framework on which lines of political and military cooperation and action could be carried out. In medieval society this framework was built on marriage alliances, landholding, patronage, and neighbourhood.³¹ By 1086 four of our ﬁve families—the Lacys, the Mortimers, the Cliffords, and the Monmouths—were established in the region. The Lacys and Mortimers were major tenants in Herefordshire and southern Shropshire. It is now apparent that Walter de Lacy, having received as many grants of land from the king as from William ﬁtz Osbern, was established in the area independently of the Conqueror’s conﬁdant, who had been sent to the region to keep order. Lacy was the successor to a number of Anglo-Saxon thegns, and this accounts for the lack of compactness of his honour compared with those further north.³² Ralph de Mortimer’s holdings, conversely, were compact ones. The Mortimer fee has been described as a ‘composite’ one, as it was composed of the lands of several antecessores but had been augmented by the king after the rebellion of Roger of Breteuil, William ﬁtz Osbern’s son, in 1075. This probably had been done to make Mortimer’s holdings resemble more closely the compact defensive fees of northern Shropshire. Mortimer was a relative of Roger de Montgomery, earl of Shrewsbury, and held land of him in Shropshire. His tenancy-in-chief, which spanned the county boundary with Herefordshire, was thus the result of cooperation between the king and the earl, being composed of lands forfeited by Breteuil and manors escheated after the death of Edward the Confessor’s ³⁰ See Ch. 6. Roger de Mortimer sent knights to serve with John’s Irish expedition in 1210. It is worth noting that Walter de Clifford’s connection to William de Braose was purely feudal, unlike those of the Lacys and Mortimers, who were married into the house of Braose. ³¹ For the importance of the ‘neighbourhood’, or visnetum, see David Crouch, ‘From Stenton to McFarlane: Models of Societies of the Twelfth and Thirteenth Centuries’, TRHS, 6th ser., 5 (1995), 193–7. ³² C. P. Lewis, ‘English and Norman Government and Lordship in the Welsh Borders, 1039–1087’, D.Phil. thesis, University of Oxford (1985), 279–81.

Introduction

11

widow, Queen Edith. Roger de Montgomery then ﬁlled the gaps with grants to be held of him.³³ Drew ﬁtz Pons, the progenitor of the Clifford line, probably was introduced to the region by the king after 1075, for he is likely the ‘Drogo’ found in Domesday holding land in the castlery from which his family was to take its name.³⁴ William ﬁtz Baderon (or Monmouth) was closely linked with the king, as his uncle Wihenoc, a Breton, had been made custodian of Monmouth Castle after the forfeiture for rebellion of Roger de Breteuil in 1075. Wihenoc became a monk and was succeeded as custodian of Monmouth by William.³⁵ Thus, the ability of the king, if not to change the character of the March, to change its contours through its personnel was vividly demonstrated after the rebellion of Roger de Breteuil. This ability was to be shown more than once over the next two centuries. The next chapter therefore looks at how the region was shaped during the reigns of William Rufus and his successors. ³³ Ibid. 288–9; Lloyd, A History of Wales, ii. 395. ³⁴ HDB 34. ³⁵ Lewis, ‘English and Norman Government and Lordship in the Welsh Borders, 1039–1087’, 289, 287; Lloyd, A History of Wales, ii. 396.

2 The Shaping of the Frontier, 1086–1166 The eighty years between Domesday Book in 1086 and the Cartae Baronum of 1166 were important in the formation of frontier society in western Herefordshire and along the central and south-eastern March of Wales. High politics played a role in this process, as both Henry I and Stephen signiﬁcantly affected the fortunes of the magnate families of the area as well as the course of English expansion along the March. Politics was less important, however, than the processes of domination, control, and exploitation set in motion by the English lords themselves and their tenants. Conquest was followed quickly by castle-building and subinfeudation, and control was deepened by settlement and colonization. The Domesday inquest of 1086 provides our ﬁrst picture of the character of the region’s society. That picture is, however, a very limited one, as the survey did not include the emerging Marcher lordships beyond some sketchy references to south-east Wales; indeed some of these lordships, such as Brecon, did not yet exist. Our next overview of the region, at least from the standpoint of central government, comes from the 1160s. In 1166 Henry II ordered a new inquest into knights’ fees, ostensibly encompassing all those in England who held land in chief from the king, and those who held from them; the result was the Cartae Baronum.¹ In the same decade a recension of the Domesday inquest for Herefordshire was made, known as the Herefordshire Domesday.² This was an exchequer document, intended to trace the current holders of Domesday fees in the county. It may represent an attempted reassertion of exchequer control over a county whose revenues had largely been lost to the Crown during Stephen’s reign. It certainly reﬂects the increased reach of royal power by which Henry II and his sons brought ¹ RBE; for discussion of the cartae, see W. L. Warren, Henry II (Berkeley, 1973), 275–81; T. K. Keefe, Feudal Assessments and the Political Community under Henry II and his Sons (Berkeley, 1983), ch. 1. ² HDB.

The Shaping of the Frontier

13

border counties such as Herefordshire more securely within the ambit of royal government—a key element in the creation of the border and south-eastern March c.1200. It is the formation of this district up to the early years of Henry II and the role of the leading aristocratic families in that process which is the theme of this chapter. The story of that formation, it has to be said, must be compiled from very scanty resources. The sources, particularly narrative ones, concerning the major families of the region for this period are often patchy; for example, Baderon de Monmouth only appears in the historical record in a few of his own charters as well as as a witness to those of Miles of Gloucester and his sons. For this reason, a chronological approach is partly necessary. We start with a rapid chronological view which weaves together the fortunes of the major families of the region as identiﬁed in the previous chapter, with the course of high politics and the interplay between Crown, nobles, and Welsh princes.³ T H E S H A PI N G O F T H E F RO N T I E R : P O L I T I C S 1086 – 1154 Although the Conqueror had once journeyed as far as St David’s, his attention was mostly turned elsewhere than Wales. The reign of his son William Rufus saw a greater degree of royal involvement in the region. This involvement was particularly important in further shaping the overall tenurial structure of Herefordshire and the border. Although rebellions involving border lords led to forfeitures in the area, they did not lead to any truly signiﬁcant overturning of tenurial arrangements. What was more important was the introduction of new men into the region, men whose aggressiveness led to increased English expansion. Foremost among these men were Bernard de Neufmarché, who with likely royal encouragement would go on to conquer the Welsh kingdom of Brycheiniog, and probably also William de Braose, who became lord of Radnor at much the same time. As mentioned above, the Mortimers, Lacys, Cliffords, and Monmouths had already been established in the area.⁴ ³ Some features of this period, particularly the role of the king in the shaping of the border region, are treated later in an analytical fashion; see Ch. 5. ⁴ William de Braose held Pembridge as part of a grant that included the crown manor of Radnor and probably the entire ﬁef in Herefordshire held in 1086 by Gruffudd ap

14

Lords of the Central Marches

The rebellions of Rufus’s reign were caused by the adherence of many lords in England to the cause of Robert Curthose, William’s brother and duke of Normandy; the way these rebellions touched on Wales demonstrates the role that Continental concerns could play in how English kings approached the March. Among the rebels of Rufus’s reign was Roger de Lacy, the son of Walter, who had inherited his family’s lands in 1086. Joining him in an uprising at Christmas 1087 were Ralph de Mortimer of Wigmore, Osbern ﬁtz Richard the lord of Richard’s Castle, and Osbern’s son-in-law Bernard de Neufmarché. Leading their men, including Welsh tenants, they had captured Hereford, burned Gloucester, and marched on Worcester before they were checked by the deﬁant stand of Bishop Wulfstan of Worcester.⁵ Rufus, however, was lenient on this occasion in punishing the rebels once the rising was put down, and those involved emerged unscathed; it may be that the king was reluctant to overturn arrangements along the March for fear of destabilizing the region. In future, the aristocratic leaders of border society would not always be so fortunate. Shortly after the rebellion of 1087 Bernard de Neufmarché embarked on the conquest of the Welsh kingdom of Brycheiniog. Bernard owed his position along the March to his marriage to Osbern ﬁtz Richard’s daughter Agnes (or Nest) and had been pardoned for his role in the 1087 revolt, possibly because he had simply been following Osbern’s lead.⁶ Bernard had been granted a portion of the Herefordshire lands of Alfred of Marlborough, a rare major Saxon survivor in Domesday, at some point after 1086, and this suggests the involvement of the king in establishing Bernard along the border. After the conquest of Brycheiniog, these Herefordshire manors were integrated into the new honour of Brecon.⁷ Maredudd. Gruffudd died in 1091 while William died in 1093/4, being succeeded by his son Philip, who was lord of Builth and Radnor by 1095; B. Copleston-Crow, ‘The Fief of Alfred of Marlborough in Herefordshire in 1086 and its Descent in the Norman Period’, Trans. Woolhope, 45/2 (1986), 386. ⁵ Lloyd, A History of Wales, ii. 397; Judith A. Green, The Aristocracy of Norman England (Cambridge, 1997), 276. ⁶ Rees, ‘The Medieval Lordship of Brecon’, 172. Frank Barlow conjectures that Bernard was pardoned owing to his relations; William Rufus (New Haven, 1983), 322. ⁷ They therefore later supplied cause for jurisdictional disputes between the lords of Brecon and the king. Bernard received Alfred’s manors of Brinsop, Burghill, Bredwardine, Stretford, and Much Cowarne; Copleston-Crow, ‘The Fief of Alfred of Marlborough’, 385–6. For the dispute of 1199 over whether Bredwardine lay in England or the March, see Ch. 3.

The Shaping of the Frontier

15

Bernard used these manors as a base for his invasion of Brycheiniog, which proceeded initially up the Wye valley. By 1088 he had seized Hay, Glasbury, and Talgarth; Bronllys Castle probably had been built as well.⁸ Thereafter he moved up the valley of the river Llyfni and so to the Usk valley. By 1091 the Anglo-Normans had advanced as far as Brecon and could reasonably assert a claim to general supremacy in the rest of south-east Wales.⁹ By 1093 Brecon Castle had been erected; that year also saw Rhys ap Tewdwr, the leading Welsh ruler in south Wales, defeated and killed in battle.¹⁰ The slow pace of Bernard’s advance into Brycheiniog indicates that he was taking time to consolidate his conquests, building castles and encouraging the movement of English settlers into the Wye valley. Major subinfeudation was carried out: it is probable that Richard ﬁtz Pons, progenitor of the Clifford line, was one of Bernard’s chief followers in the conquest of Brycheiniog, for he was granted the Welsh district of Cantref Selyf in the north of the new lordship for his efforts.¹¹ In 1095 there was another revolt, led by Robert of Mowbray, against William Rufus, sparked on this occasion by dissatisfaction with the division of territories that had followed the death of the Conqueror.¹² The rebellion aimed to replace Rufus with the Conqueror’s nephew Stephen, count of Aumale, and centred on Mowbray’s lands in the north but also attracted backers in the March, including Earl Hugh of Shrewsbury.¹³ Roger de Lacy also joined the rebellion and brought Welsh troops with him. It took Rufus himself at the head of an army to put down the rising along the border. This time the king acted against the Lacys; Roger was banished in 1096 and went to Normandy, where he attained high ofﬁce under Duke Robert. The Lacy inheritance then passed to Roger’s younger brother Hugh.¹⁴ When William Rufus died in a hunting accident on 2 August 1100, his younger brother Henry succeeded him. Under Henry I royal power ⁸ Rees, ‘The Medieval Lordship of Brecon’, 173. ⁹ This victory was the motive for Bernard’s founding of a cell of Battle Abbey at Brecon; see ‘Cart. Brecon’, 14. 141. ¹⁰ Rees, ‘The Medieval Lordship of Brecon’, 173–4. ¹¹ Cantref Selyf would serve as the base for the conquest of Cantref Bychan to the west by the Cliffords during the reign of Henry I; ibid. 173. ¹² Green, The Aristocracy of Norman England, 279. ¹³ Carpenter, The Struggle for Mastery, 131. ¹⁴ Wightman, The Lacy Family in England and Normandy, 170–2.

16

Lords of the Central Marches

was felt in the March and Wales as never before.¹⁵ Henry is famous for having used, in the oft-quoted words of Orderic Vitalis, ‘men raised from the dust’; not a few of these new men were sent to Wales and the March. Ecclesiastically, he appointed bishops to the Welsh sees of Llandaff and St David’s and exacted professions of obedience to Canterbury from the Welsh bishops. While Henry chose to pursue a policy of overlordship in north Wales rather than outright conquest, his policy was quite different in much of the south. During his reign most of lowland south Wales came under English rule, and a colony of Flemings was established in Pembroke.¹⁶ This expansion of direct English control in south Wales was due to an increased level of involvement in the March by the Crown, and to its active encouragement of aristocratic entrepreneurship. For example, Richard ﬁtz Pons (Clifford) had been invested with Cantref Bychan by Henry I prior to 1116, an investiture that was undoubtedly speculative, comparable to that referred to in a contemporary account told of Gilbert ﬁtz Richard of Clare and his activities in Ceredigion.¹⁷ In that story Henry I, in response to the continual importuning of Gilbert, ‘gave’ him the land of Cadwgan ap Bleddyn, saying, ‘Go and take possession of it.’ The king freed his lords to do as they wished in Wales; they, however, had to supply the necessary force. In the case of Richard ﬁtz Pon’s control of Cantref Bychan, it was undoubtedly little more than a loose overlordship, as the cantref (a local Welsh administrative subdivision) was left in the care of a native ruler, Maredudd ap Rhydderch ap Caradog.¹⁸ Such overlordship, involving the collection of tribute formerly due the previous Welsh chieftain, was a typical form of English domination in the Welsh uplands, and reﬂected either a reluctance or sheer inability to project direct power into difﬁcult and unrewarding terrain. Clifford overlordship of Cantref Bychan was shortlived, however, and marked the family’s furthest advance into Wales. In the eastern March itself the English intensiﬁed their lordship during these years, establishing the ‘feudal matrix’ which stamped the ¹⁵ ‘Henry I left a deeper imprint on the affairs of medieval Wales than any king of England other than Edward I’ (R. R. Davies, ‘Henry I and Wales’, in H. Mayr-Harting and R. I. Moore (eds), Studies in Medieval History presented to R. H. C. Davis (1985), 133). For the comments of a contemporary Welsh observer, see Brut, 42, s.a. 1116. ¹⁶ Davies, ‘Henry I and Wales’, 134. ¹⁷ Lloyd, A History of Wales, ii. 429; for Gilbert, see Brut, 34, s.a. 1110. ¹⁸ Even so, Maredudd proved a loyal governor, successfully defending the sole Clifford castle of Llandovery during the Welsh rising of 1116; Lloyd, A History of Wales, ii. 434.

The Shaping of the Frontier

17

character of Marcher society. Knights’ fees were created in lowland Brecon; manorialization began to take root; and a borough had been founded there by 1106.¹⁹ It was during these years that some of the more important mesne families examined later in this book were established in the area.²⁰ The Picards (or Pichards) built their castle of Tretower in the commote of Ystrad Yw Uchaf, while to their east the Turbervilles built a stronghold at Crickhowell from which to govern Ystrad Yw Isaf. To the west of the Pichards were the Baskervilles, Roger de Baskerville having been granted the portion of Brecon known as Pencelli. William Revel was granted a sub-lordship centred on Hay; this later reverted to the lord of Brecon and was absorbed into his demesne. The establishment of the Cliffords in Cantref Selyf in north-east Brecon has been mentioned above.²¹ Elsewhere little is known of the activities of the Braoses and Mortimers in Radnor, Builth, and Maelienydd during the reign of Henry I. The statement by the Welsh Brut that in 1144 Hugh de Mortimer conquered Maelienydd and rebuilt the castle of Cymaron, and that the ‘French’ (almost certainly William II de Braose) re-entered Elfael and repaired Colwyn Castle, indicates that the Mortimers and Braoses must have had some hold over these areas before the Welsh rising of Stephen’s reign in 1136.²² It is uncertain how early such inroads into the uplands began, but such an advance coincides with the high tide of aristocratic English power in Wales during Henry I’s reign. Politically, the most signiﬁcant effect of Henry’s reign on the March was his establishment of new men in the area. For example, when Hugh de Lacy died without a male heir sometime between 1115 and 1126, a large portion of his lands passed with his daughter Sybil to her husband, the trusted royal servant Payn ﬁtz John.²³ It has been suggested that out of caution Henry did not wish to see one of his ¹⁹ Davies, ‘Henry I and Wales’, 142; Lloyd, A History of Wales, ii. 436. ²⁰ See Map 3, of the lordship of Brecon, for what follows. For a partial account, see also B. Copleston-Crow, ‘Trewalkin and its Anglo-Norman Lords’, Brycheiniog, 26 (1992–3), 43–51. ²¹ Lloyd, A History of Wales, ii. 437–8; B. Copleston-Crow, ‘The Baskervilles of Herefordshire 1086–1300’, Trans. Woolhope, 43 (1979), 19. ²² Brut, 53, s.a. 1144; J. B. Smith, ‘The Middle March in the Thirteenth Century’, BBCS 24/1 (1970–2), 78. ²³ Hugh de Lacy replaced his brother Roger as lord of Weobley and Ewyas Lacy following Roger’s rebellion. Both Payn ﬁtz John and his brother Eustace did well in royal service; see Judith A. Green, The Government of England under Henry I (Cambridge, 1986), 250–3.

18

Lords of the Central Marches

rising servants become too strong and that for this reason Payn did not succeed to the entire Lacy holding. The king was also able to ‘reduce in size a dangerously large barony’; though it is equally possible that Henry wished to divide his patronage several ways.²⁴ Another portion of the Lacy honour of Weobley went to Jocelin de Dinan, who held the castle of Ludlow in Stephen’s reign and thus probably received the Lacy manors in southern Shropshire.²⁵ A third portion was given to Miles of Gloucester, a signiﬁcant move on Henry’s part. Miles of Gloucester and his family played an important role in shaping the south-eastern March during the mid-twelfth century. Miles was the son of Walter of Gloucester, sheriff of that county and constable of Gloucester Castle under both William Rufus and Henry I. In 1121 Miles was granted in marriage Sybil, the daughter of Bernard de Neufmarché. Bernard also had a son, Philip, who died before 1121, and a bastard son, Mahel.²⁶ In the marriage arrangement, probably imposed by Henry I, Miles received part of Sybil’s inheritance directly on their union and was to receive the remainder on Bernard’s death. Bernard soon obliged by dying sometime between 1123 and 1125.²⁷ From the escheated Lacy lands Miles also received manors or portions thereof in Herefordshire, mostly to the east of Leominster.²⁸ ²⁴ Wightman, The Lacy Family in England and Normandy, 175. To judge by the later carta of Hugh II de Lacy in 1166, which assessed the honour of Weobley at ﬁfty-two and a quarter fees and the statement in Herefordshire Domesday that the husband of Payn’s daughter and heir (Roger, earl of Hereford) held ten fees, excluding Ewyas Lacy, it is difﬁcult to see why a barony of some sixty-three fees was so ‘dangerously large’. Compactness and its militarized nature may have given the honour extra weight in the minds of the king and his counsellors. ²⁵ Wightman’s suggestion that Ludlow at one point was the caput of what was essentially a separate honour (or sub-lordship) is geographically convincing; The Lacy Family in England and Normandy, 134–6. The manors involved are Stanton Lacy, Plaish, Patton, Rushbury, Middlehope, Stokesay, Onibury, Aldon, Bitterly, Corton, Sutton, Wichcote, Upper and Lower Hayton, Henley, and to the east, centred on Cleobury North, Wheathill, Stanton Long, Ditton Priors, and Hopton Wafers. In 1242 these manors were assessed at a total of eleven and three-quarters fees, while Stokesay, Onibury, and Stanton Lacy along with Ludlow were demesne; BF 964. ²⁶ Ancient Charters, Royal and Private: Prior to A.D. 1200, ed. J. H. Round, Pipe Rolls Series (1888), 8, no. 6; Lloyd, A History of Wales, ii. 438; Rees, ‘The Medieval Lordship of Brecon’, 176–7. Rees helpfully deﬂates the story, related by Gerald of Wales and used by Lloyd, of Mahel’s legitimacy and disinheritance by his mother, Agnes (or Nest); Giraldus Cambrensis, Itinerarium Cambriae, ed. J. F. Dimock, in Opera, ed. J. S. Brewer, J. F. Dimock, and G. F. Warner, 8 vols, Rolls Series (1861–91), vi. 28–9. ²⁷ Rees, ‘The Medieval Lordship of Brecon’, 177. ²⁸ Humber, Brockmanton, Hanton, Gattertop, Pudleston, Salbiria, Collington; HDB 79. It must be kept in mind that other Lacy manors passed to the Gloucester family with

The Shaping of the Frontier

19

Henry I thus used his feudal rights of marriage to break up a major border honour, reward men of his own choosing, redraw the feudal map of Herefordshire, and ﬁrmly assert his control over an important section of frontier. Previous royal policy had veered between the maintenance of a balance of power between English lords and the Welsh on the one hand, and the encouragement of baronial aggression on the other. In the ﬁrst instance, the Conqueror’s visit to St David’s in 1081 almost certainly resulted in an agreement with Rhys ap Tewdwr, an agreement which may have made Rhys the king’s viceroy in Wales and effectively checked the expansionist ambitions of the Marcher lords.²⁹ In contrast to this approach, William Rufus had given his frontier barons a free rein, and the results were apparent in the south with the conquests of Brecon and Glamorgan. Henry I (aided by feudal incidents) instead placed men who were dependent upon him into positions of power along the March. With the establishment of the Clares in the person of Walter ﬁtz Richard in upper Gwent at some point prior to 1119, the marriage of Miles of Gloucester to the heir of Bernard de Neufmarché, and the installation of Payn ﬁtz John in the bulk of the Lacy inheritance, the contiguous lordships of upper Gwent, Brecon, and Ewyas Lacy, in the words of one historian, ‘formed in this district a solid nucleus of resistance to feudal unrest and disaffection’.³⁰ This was true as long as Henry occupied the throne, and he seems to have learned the lessons of the revolts of 1086 and 1095. But his reorganization of territory and personnel was not achieved without cost. The existence of an excluded claimant to the Lacy inheritance, Gilbert de Lacy, who was a nephew of both Roger de Lacy and of his brother Hugh, would add to the region’s instability during the reign of Henry’s successor, Stephen.³¹ Henry I’s accomplishments in Wales were undone during Stephen’s reign. Royal control over the borders and the March was lost as most of the area’s lords drifted into the camp of Stephen’s rival for the throne, the empress Matilda, during the civil war. Those Marcher lords who sided with the Angevins and followed Robert, earl of Gloucester, and Miles of Gloucester saw their advances against the Welsh checked to the marriage of Cecily, daughter of Payn ﬁtz John and Sybil de Lacy, to Roger, earl of Hereford (son of Miles of Gloucester). ²⁹ This same expedient would later be used by Henry II when he made the Lord Rhys his ‘justice’ in south Wales; see Ch. 5. ³⁰ Lloyd, A History of Wales, ii. 443. ³¹ Gilbert was the nephew of Hugh de Lacy; David Crouch, The Reign of King Stephen 1135–1154 (1999), 79 n. 21.

20

Lords of the Central Marches

enable Angevin resources to be diverted elsewhere. Those who did not join the empress’s party, namely Hugh de Mortimer and William II de Braose, took advantage of a weakened king and a fractured political situation to resume their ‘private’ wars with their Welsh neighbours. For those along the Welsh border, Stephen’s accession in 1135 was quickly followed by disaster.³² In 1136 the Welsh rose against the English and inﬂicted a series of defeats on them. On or around 1 January 1136 the Welsh of Cantref Bychan and western Glamorgan defeated an English force in Gower. Later that year Richard ﬁtz Gilbert of Clare was ambushed and killed while he was crossing upper Gwent.³³ In July 1137 Payn ﬁtz John was in turn killed while pursuing the Welsh. Historians have debated whether or not the Welsh uprising of 1136 was an immediate reaction to the death of Henry I, but this matters little from the perspective of those men who held land in the borders and the March. Stephen, for whatever reason, responded to these setbacks in what can only be described as a desultory fashion. Henry I would probably have come to the border quickly at the head of an army. Stephen instead concentrated on events elsewhere in England and left Marcher lords of the second rank to deal with the Welsh. Their efforts were not enough, as attempts by Baldwin ﬁtz Gilbert in Ceredigion and Robert ﬁtz Harold in Carmarthen to reverse Welsh advances achieved little. Stephen refused to use the more important men of the March who had been put in place by Henry I, at ﬁrst keeping them with him at the siege of Exeter and possibly choosing to use Miles of Gloucester as a counterweight to Henry I’s bastard son Robert, earl of Gloucester, a potential threat to Stephen’s hold on the throne. Stephen may have distrusted the more powerful Marchers and may even have seen the turmoil in Wales as an effective way of keeping potential opponents occupied.³⁴ Whatever the perhaps understandable reasons for the king’s inaction in the face of Welsh aggression, Stephen’s reign demonstrates the indispensable role royal power had come to play in stabilizing the supposed ‘private’ baronial enterprise of expansion in Wales.³⁵ ³² For much of what follows I adhere to the interpretation of Crouch, Reign of King Stephen, ch. 3: ‘Wales and Normandy’, in which he sees the king’s reluctance to address forcefully events in the two areas as the point at which his reign started to go badly wrong. ³³ Brut, 51, s.a. 1136; Gesta Stephani, 2nd edn, ed. and trans. K. R. Potter, introd. and notes R. H. C. Davis (Oxford, 1976), 16–18; Giraldus Cambrensis, Opera, vi. 47–8. ³⁴ This is the speculation of David Crouch, ‘The March and the Welsh Kings’, in Edmund King (ed.), The Anarchy of King Stephen’s Reign (Oxford, 1994), 260–4. ³⁵ As noted by Davies, The Age of Conquest, 46.

The Shaping of the Frontier

21

By the end of 1136 the Marchers, with the probable exception of Hugh de Mortimer and William de Braose, had reached a modus vivendi with the Welsh. Although this was probably a piecemeal approach on their part rather than a coordinated piece of policy, it certainly was necessitated by Stephen’s lack of support for them in the face of Welsh aggression.³⁶ Stephen could have done little more to alienate those lords who had interests in the March and border counties, especially Robert of Gloucester, and the king correspondingly sank in their estimation. With the outbreak of civil war signalled by the desertion of Robert of Gloucester in the summer of 1138, the cause of Stephen’s rival the empress Matilda was centred on the West Country and the southern and Middle March of Wales. The last thing that her chief supporter, Robert of Gloucester, needed was to have to ﬁght the Welsh while he contested the Midlands with Stephen’s forces. By 1139 Robert had been joined by Miles of Gloucester, originally a supporter of the king.³⁷ Miles, as one of the courtiers who had been prominent under Henry I, had been squeezed out of inﬂuence at court by the faction of new men, headed by Waleran of Meulan, who surrounded Stephen, and may already have been in contact with the empress before she landed in England in September 1139.³⁸ Miles of Gloucester’s defection to the Angevins contributed to the factionalizing of Herefordshire and the south-eastern March. Thus, the civil war allows us to see the political society of the area, with its stresses, fault lines, and bonds, revealed more clearly than the documentation usually allows. The ‘disinherited’ who had joined Robert of Gloucester against the king included Gilbert de Lacy, whose claim to the Lacy inheritance had been passed over by Henry I. Gilbert almost certainly joined with Robert in the hope of regaining the manors which had been divided between Payn ﬁtz John, Jocelin de Dinan, and Miles of Gloucester. In 1138 Gilbert had led an assault on the royalist stronghold of Bath.³⁹ But the presence of Miles of Gloucester in the rebel camp by 1139 posed an insurmountable problem for Gilbert and he crossed over to the royalists. ³⁶ David Crouch, ‘The Slow Death of Kingship in Glamorgan, 1067–1158’, Morgannwg, 29 (1985), 33. ³⁷ Miles was linked to Stephen’s justiciar Roger, bishop of Salisbury, and likely sensed the change in political climate to come with Roger’s arrest in 1136 and therefore crossed to the empress’s allegiance; R. H. C. Davis, King Stephen, 3rd edn (1990), 42. ³⁸ Crouch, The Reign of King Stephen, 102 and n. 54. ³⁹ Gesta Stephani, 58. The author of the Gesta, probably Bishop Robert of Bath, described Gilbert as vir prudens et in omni militari actione providus et operosus.

22

Lords of the Central Marches

Not everything that occurred in the region can be interpreted through the lens of ‘party’, for civil wars always create opportunities for the settling of local scores. It is in this context that Gilbert de Lacy and Hugh de Mortimer waged a local war against Jocelin de Dinan, lord of Ludlow, in the early 1140s.⁴⁰ Jocelin himself was a supporter of Stephen and this conﬂict shows how royalism often came second to personal aggrandizement; Mortimer and Lacy were ﬁghting primarily for their own interests.⁴¹ In another instance, according to the account in the Wigmore Chronicle, Mortimer had quarrelled with his seneschal Oliver de Merlimont, who went over to Miles of Gloucester and the Angevins. Gilbert de Lacy, as Oliver’s overlord, then seized for Mortimer the chattels of the canons whom Oliver had installed at Letton, a former Lacy manor. As the civil war continued into the late 1140s, the magnates, perhaps inspired by the bishops, began to try to limit the conﬂict and protect their interests by the formation of agreements, or conventiones, which cut across ‘party’ lines.⁴² This is what occurred in Herefordshire and the adjacent March. At some point between 1148 and 1154 William II de Braose switched to the Angevin side and a conventio was drawn up between him and Roger, earl of Hereford, Miles of Gloucester’s son and successor.⁴³ William had married Bertha, daughter of Miles of Gloucester, and was thus Roger’s brother-in-law. The date of this marriage is unknown, but it is tempting to see it as a means of rapprochement between Braose and the Gloucester family earlier in the civil war. The marriage preceded the conventio, as William is described in it as socero meo. It is probable that the offer of a handsome maritagium with Bertha’s hand clinched William de Braose’s desertion of Stephen. Also, friction seems to have developed between William and Gilbert de Lacy over the manor of Mansel Lacy, as the agreement between Braose and Roger of Hereford states that Roger would not have peace or truce ⁴⁰ J. C. Dickinson and P. T. Ricketts, ‘The Anglo-Norman Chronicle of Wigmore Abbey’, Trans. Woolhope, 39 (1969), 427–9. ⁴¹ See the grant of the borough, castle, and county of Hereford to Robert, earl of Leicester, probably in 1140, in which the fees of Hugh de Mortimer, Osbert ﬁtz Hugh, William de Braose, and Jocelin were excluded from the grant; RRAN iii. 165, no. 437. ⁴² This process is discussed in the famous chapter in Davis, King Stephen, entitled ‘The Magnates’ Peace’. ⁴³ ‘Conventio between Roger Earl of Hereford and William de Braose, his Brotherin-Law (1148–1154)’, pub. as an appendix in Z. N. Brooke and C. N. L. Brooke, ‘Hereford Cathedral Dignitaries in the Twelfth Century: Supplement’, Cambridge Historical Journal, 8 (1944–6), 185. See also the Braose genealogy in Appendix 2.

The Shaping of the Frontier

23

with Gilbert without William’s assent and the quitclaim of the manor by Gilbert to William.⁴⁴ Finally, the Angevin conquest of Normandy also played a role in William’s decision, as the conventio promises that he shall, through either Roger or his representative, hold in peace his castle of Briouze and lands in Normandy. Beyond Lacy, Mortimer, and Braose, the majority of the other families we are examining were adherents of Miles of Gloucester and afterwards his son Roger. At some point between 1147 and 1150 William, son of Robert, earl of Gloucester, and Miles of Gloucester’s son Roger, earl of Hereford, made a renewal of a treaty agreed in the early 1140s between Robert, earl of Gloucester, and Miles, earl of Hereford.⁴⁵ A number of ‘hostages’ were named in this second treaty; Roger’s included Baderon de Monmouth, Walter de Clifford, and the brothers Ralph and Robert de Baskerville.⁴⁶ Thus, by the closing stages of the war, most of aristocratic society of Herefordshire and the south-eastern March was in the Angevin camp and opposed to Stephen, Hugh de Mortimer, and Gilbert de Lacy. The witnessing of a number of charters of Miles and Roger by Baderon de Monmouth, Walter de Clifford, and several mesne tenants—men from such local knightly families as Petit, Pichard, Mucegros, Burghill, Baskerville, Waldeboef, and Bret—demonstrates the extent of what has been termed the ‘Hereford afﬁnity’.⁴⁷ ⁴⁴ Mansel Lacy must have been in the possession either of Miles or more probably of Roger and was then possibly granted as maritagium with Bertha when she married William II de Braose. At the time of Domesday it was worth 100s. but had been worth 130s. tempore Regis Edwardi and in 1086 not all of its land was under the plough. In the 1160s William de Braose still held 5 hides in the manor; HDB 72, 78. The manor later passed back to the Lacys, probably upon the marriage of Margaret de Braose, daughter of William III, to Walter II de Lacy. It passed again as maritagium when Walter’s daughter Katherine married Walter III de Clifford; the marriage contract states that the service owed to the heirs of William de Braose was to be made to Walter de Lacy and his heirs. Roger Tirel held the manor for the service of two knights. The charter dates from 1219–21; HCA, no. 3241. ⁴⁵ R. H. C. Davis, ‘Treaty between William Earl of Gloucester and Roger Earl of Hereford’, in Patricia M. Barnes and C. F. Slade (eds), Early Medieval Miscellany for Doris Mary Stenton, Pipe Rolls Series (1962), 139. Davis concludes that the treaty was probably concluded early in this period. ⁴⁶ Ibid. 144. The word used is obsides, but given the number (14) and importance of the men listed merely from Roger’s side, it is likely that they were pledges rather than truly hostages. ⁴⁷ ‘Charters of the Earldom of Hereford, 1095–1201’, ed. D. G. Walker, Camden Miscellany, 2 (1964), 1–75, nos 11, 17, 18, 27, 39 (Parvus or Petit), 40 (Pichard), 43 (Mucegros, Burghill), 48 (Baskerville, Waldeboef, Bret), 53, 55. The term ‘Hereford afﬁnity’ is Crouch’s ‘The March and the Welsh Kings’, 280 n. 67.

24

Lords of the Central Marches

What becomes apparent from this examination of Stephen’s reign is the degree to which the Marchers and their actions, despite often being founded on local and familial concerns rather than purely ties of ‘party’, were thoroughly intertwined with events in England and indeed beyond. Events elsewhere inﬂuenced what occurred in the March, just as the Marchers themselves could make their strength felt beyond the border regions of England and Wales. There was never a time when the March was isolated from the broader political world around it, even if, partly by conscious effort on the part of the Marchers, it would to a greater or lesser degree be cordoned off from certain other features of English society, such as laws and taxation. When Henry II came to the throne in 1154, the ﬁrst problem confronting him was the implementation of that part of the Treaty of Winchester which stipulated the return of inheritances to the ‘dispossessed’; many of the disputes which had resulted in dispossession reached back beyond Stephen’s reign to that of Henry I. Gilbert de Lacy, although he had gained control of most of his family’s former lands by 1146, quickly and prudently attached himself to Henry’s entourage.⁴⁸ Gilbert was aided by the slowness with which Roger, earl of Hereford (heir to Miles of Gloucester), adhered to the new king. As Henry moved to resume crown demesne and castles lost during the civil war, this resulted in the deﬁance of Roger and his one-time enemy Hugh de Mortimer.⁴⁹ Both men held royal castles, Earl Roger the castles of Gloucester and Hereford, Hugh Bridgnorth in Shropshire, and during the civil war had undoubtedly come to look on them as personal possessions. But Roger’s obstreperousness cooled, and his kinsman Gilbert Foliot, bishop of Hereford, persuaded him to hand over Hereford and Gloucester to the king. Henry, however, did not forget Roger’s opposition, and the earldom of Hereford was allowed to lapse in 1155 when Roger died without an heir of the body. Henry also took back manors which had been granted to Miles of Gloucester as terre datae. The Gloucester family lands then descended through Roger’s surviving brothers, Walter, Henry, and Mahel. When Mahel died without issue in 1165, the family’s assemblage of lands was divided among the daughters ⁴⁸ Davis, ‘Treaty between William Earl of Gloucester and Roger Earl of Hereford’, 140. ⁴⁹ This despite Henry’s exclusion of the Mortimer fee in Staffordshire from his grant to the earl of Chester, January–April 1153, at Devizes; RRAN iii. 65, no. 180.

The Shaping of the Frontier

25

of Miles of Gloucester, and it was in this way that the Braoses gained Brecon and Abergavenny.⁵⁰ At a stroke, the Braose family, who had had a foothold in the March since the 1090s in addition to their estates in Sussex, were promoted into the leading rank of baronial families in the Anglo-Welsh borders. Unlike Roger of Hereford, Hugh de Mortimer had armed his castles against the king. Henry was forced to bring up a royal army against the recalcitrant Mortimer, a baron described by Robert of Torigny as ‘a most arrogant and self-conﬁdent man’.⁵¹ Hugh’s castles of Wigmore, Cleobury, and Bridgnorth were all besieged and fell. Hugh surrendered and his submission was received in a great council at Bridgnorth on 7 July 1155; one of the most turbulent of the Marcher barons had been brought to heel by royal power.⁵² Henry had now asserted his control over the magnates along the border. It remained to be seen what policy he would adopt towards the Welsh and the March.⁵³

T E N U R I A L C H A N G E S : H E R E F O R D S H I R E TO 1 1 6 0 During Henry II’s reign the scope of royal government increased and this had an impact on those honours that lay in Herefordshire.⁵⁴ By this time, however, a number of signiﬁcant changes also had occurred in the pattern of land holding in the area. By the 1160s most lords in the region had seen their demesne lands severely diminished by further subinfeudation. Furthermore, the gradual division of those knightly estates that spanned the boundary between England and ⁵⁰ Warren, Henry II, 60; David Walker, ‘The ‘‘Honours’’ of the Earls of Hereford in the Twelfth Century’, Transactions of the Bristol and Gloucestershire Archaeological Society, 79/2 (1960), 192; Giraldus Cambrensis, Itinerarium Cambriae, i. 2. The division was effected between the three daughters of Miles of Gloucester. Margaret de Bohun and Lucy, the wife of Herbert ﬁtz Herbert, had the English manors settled upon them, but Margaret held all of them in 1165 and it was not until the reign of Richard I that Herbert had any success, and that only partial, in gaining any of his wife’s inheritance. The manors in Herefordshire that were part of the honour of Brecon went with the third sister, Bertha, to her husband, William II de Braose; Walker, ‘The ‘‘Honours’’ of the Earls of Hereford in the Twelfth Century’, 192–3. Miles of Gloucester had obtained Abergavenny from Brian ﬁtz Count in 1141–2; ‘Charters of the Earldom of Hereford’, 5. ⁵¹ Vir arrogantissimus et de se praesumens, Chronica Roberti de Torigneio, in Chronicles of Stephen, Henry II and Richard I, ed. R. Howlett, 4 vols, Rolls Series (1889), iv. 184–5. ⁵² William of Newburgh, Historia Rerum Anglicarum, in Chronicles of Stephen, Henry II and Richard I, i. 105; Warren, Henry II, 60–1. ⁵³ See Ch. 5. ⁵⁴ Examined below; see Ch. 5.

26

Lords of the Central Marches

Wales had begun. This division of estates ultimately combined with the increasingly intrusive nature of royal government to deﬁne the March against England. For this reason it is important to take another ‘snapshot’ of the tenurial structure of the area, a possibility afforded by the survival of the Herefordshire Domesday. The existence of this text, dating from between 1160 and 1170, when used in conjunction with Domesday Book and the returns of the 1166 inquest into knights’ fees, enables us to trace both the evolution of the major families’ holdings in the county, and the knightly ties of dependence to these families, over a period of eighty years.⁵⁵ Along this stretch of the Anglo-Welsh border, the Mortimers held Wigmore in northern Herefordshire (see Map 1). The Lacys held the honour of Weobley, mostly centred in Herefordshire but with a complex of manors in southern Shropshire and outlying manors as far away as Oxfordshire. The Lacys also held the Marcher lordship of Ewyas Lacy, based on the valleys to the east of the river Honddu. The Braoses had held Radnor and Builth since the end of the eleventh century and, after 1165, the important Marcher lordships of Brecon, which had outlying manors in Herefordshire, and Abergavenny. The Cliffords held the small lordship of Clifford, situated on the river Wye, and also the Brecon sub-lordship of Cantref Selyf. Finally, the Monmouths held the lordship of Monmouth, east of Abergavenny. Not all of these lordships were similarly organized. If we turn to the two major Herefordshire honours, Wigmore and Weobley, a difference is immediately apparent. Geographically, Weobley was rather diffuse, with component manors scattered throughout Herefordshire as well as in Gloucestershire, Shropshire, Staffordshire, Worcestershire, and Oxfordshire. This lack of compactness reﬂects the manner in which Weobley was formed. Walter I de Lacy (d. 1084/5) came to England around the same time as William ﬁtz Osbern, and just as ﬁtz Osbern was granted the lands of Saxon antecessores in Herefordshire in the manner reminiscent of the rest of Anglo-Norman England, so was ⁵⁵ The Herefordshire Domesday was an experimental product of Master Thomas Brown, who, though not of the Exchequer, sat at it (and indeed kept the ﬁrst of what would later be termed the memoranda rolls) and also held land in Herefordshire. The text is quite reliable and the differences between it and the Domesday of 1086, as well as scribal errors, are minor. These trivial mistakes are far outweighed by the invaluable though partial annotation of current landholders in the margins of the manuscript, as well as many updated 12th-century place names; these allow the historian to see how the tenurial patterns described in 1086 had changed some eighty years later. See the introduction to HDB by V. H. Galbraith, pp. xiii–xxxii.

The Shaping of the Frontier

27

Walter de Lacy. Lacy received grants of land from both the king and ﬁtz Osbern.⁵⁶ This accounts for the relative lack of geographic compactness of Weobley, which was assessed at ﬁfty-two and a quarter fees in the carta of 1166, with the service of three and a half fees having been lost. At the time of the Cartae Baronum, Weobley had experienced little apparent breakdown of its component fees, having only four fractional fees; subinfeudation below the level of the mesne tenant had not proceeded very far.⁵⁷ In the border region, the honour of Weobley was organized largely around two large demesne manors, Weobley in Herefordshire and Stanton Lacy in southern Shropshire.⁵⁸ Weobley was the caput of the honour and, although it is not mentioned as such in Domesday, the tight grouping of manors around Weobley suggests that it was one of the early castleries (castellaria) of the region.⁵⁹ Just south of Weobley, in Grimesworth Hundred, another cluster of manors stretched from Letton in the west to Wormsley in the east.⁶⁰ East of Hereford itself there was a further scattering of manors, the most important of which was the large demesne manor of Holme Lacy. The Lacy manors in Shropshire can be divided into two rough groupings, separated by the high ground that runs from Castle Holegate in the north to Tilleshope in the south. The manors to the west were organized around Ludlow Castle and the manor of Stanton Lacy. The eastern group of manors was centred on Cleobury North.⁶¹ Wigmore, on the other hand, was a much more compact assemblage. Its manors were rather tightly clustered around the eponymous caput, being spread across north-west Herefordshire and southern Shropshire. There were more knights’ fees in Shropshire, some eighteen, two-thirds, and a quarter in 1243, with two of those fees Mortimer demesne.⁶² In Herefordshire in 1243 there were twelve and a half and a quarter fees, ⁵⁶ For the establishing of the Lacys in Herefordshire, see C. P. Lewis, ‘The Norman Settlement of Herefordshire under William I’, ANS 7 (1985), 202–5. ⁵⁷ RBE 281–3. ⁵⁸ For a full listing of Lacy manors in the region, as well as those of the other magnate families, see Appendix 1. ⁵⁹ In the area immediately surrounding Weobley were the manors of Almely, Sarnesﬁeld, King’s Pyon, Swanston, and Birley. ⁶⁰ This group of manors included Brobury, Staunton-on-Wye, Byford, Mansell Gamage, the demesne manor of Mansel Lacy, Yazor, Yarsop, and Wormsley. ⁶¹ There were fewer Lacy manors in Gloucestershire; they accounted for thirteen and a half fees in 1211–13. ⁶² BF 963. The Shropshire demesne manors were Cleobury Mortimer, Kington, and Norton.

28

Lords of the Central Marches

with three and a half and a quarter fees held in demesne.⁶³ By 1243 two further manors in Herefordshire, Lye and Pembridge, had come under Mortimer lordship with the family’s acquisition of the lordship of Radnor by inheritance. The crucial process that stands out in the feodaries is subinfeudation. In the case of the Braoses, unfortunately, little may be garnered. The 1166 carta for William II de Braose was entered under Barnstaple, his West Country barony, and the lordships in the March—Radnor, Builth, Brecon, and Abergavenny—were not covered by the inquest. Herefordshire Domesday does, however, provide some information on the manors of the honour of Brecon in Herefordshire. Brinsop, held by Walter Torel in the 1160s, was already subinfeudated at the time of Domesday Book, while Burghill, held in 1166 by William de Miners, and Bredwardine, now held by Ralph de Baskerville, had been granted out some time after 1086. It is difﬁcult to form a precise picture of the diminution of Braose demesne in Herefordshire over the century, but some manors had evidently been granted by Bernard de Neufmarché to his followers early in the conquest of Brecon. It is impossible to tell what had occurred to the Mortimer demesne between 1086 and 1166, despite the fact that the honour was situated within the counties of Herefordshire and Shropshire at the time. Domesday Book itself lists only three manors out of a total of seventeen plus four in the marcha de Walis (sic) as having been subinfeudated.⁶⁴ Two, Kinnersley and Birley, were held by one ‘Richard’ in 1086. Whether this lack of information is due to the carelessness of the scribes or reﬂects the reality that most of Mortimer’s land was held in demesne is uncertain. If the majority of Mortimer’s land was demesne in 1086, he would have had to rely very heavily on household knights. Herefordshire Domesday is no greater help, as it only provides the single fact of Robert de Mortimer’s tenancy of Lower Lye in Herefordshire.⁶⁵ ⁶³ BF 814. The Herefordshire demesne manors lay to the immediate south of Wigmore; Kingsland, Eardisland, Burton Court, Hinton Manor, and Tuiford accounted for two, a half, and a quarter fees, while part of Orleton to the east, which was shared with Thomas de Careswell, accounted for one knight’s fee. ⁶⁴ HDB 36–9; the reference to the 9 hides in the March which were waste is on p. 38. ⁶⁵ HDB 38, 95. Galbraith cautiously supposes that this Robert was the younger son of Hugh de Mortimer of Wigmore (d. 1180/1) and that he married Margaret de Say, heir to Richard’s Castle; the editors of the Complete Peerage correctly identify the Robert who married Margaret as being of the unrelated Essex Mortimers, who held land of the honour of Peverel; George E. Cokayne, The Complete Peerage of England, Scotland, Ireland, Great Britain, and the United Kingdom, extant, extinct, or dormant, new edn

The Shaping of the Frontier

29

Hugh de Mortimer seems to have been loath in 1166 to give the government information, and one can easily imagine the recalcitrant baron of 1155 muttering about ‘new taxes’. It is not until the returns of 1242 that we get a clear picture of the tenurial structure of the honour of Wigmore. With the remaining three families we are on ﬁrmer ground. In 1086 the Cliffords had held four Herefordshire manors in demesne: Rochford, Dorstone, Myndd Brith, and Burhstansestone.⁶⁶ By 1166 Rochford was held by Osbern ﬁtz Hugh of Richard’s Castle, Dorstone by one R. de Sollers, and Burhstanestone by one ‘Fagemeneda’. Only Myndd Brith, valued at 100s. in 1086, was still held in demesne. The Monmouth demesne holdings in Herefordshire had been similarly diminished. As was the case with the Cliffords, William ﬁtz Baderon had four demesne manors in the county in 1086, accounting for 12 geldable hides of land.⁶⁷ By 1170 all but 1 half-hide had been subinfeudated, although two manors, Stretton Grandison and Ashperton, were held by Gilbert, the heir to Baderon de Monmouth (d. 1170/6), and were therefore not permanently alienated. The Monmouth barony was a small one as measured in knights’ fees (before the advent of account rolls virtually our only measure); like the Cliffords, in the thirteenth century the Monmouths would enter the king’s service as a means of tapping royal patronage and thus augmenting their resources. Our best example of the subinfeudation that occurred from 1086 to 1166 comes from the Lacy honour of Weobley. In 1086 Roger de Lacy had eighteen demesne manors in Herefordshire.⁶⁸ These manors (Gloucester, 1982–98), ix. 258 and n. This Robert de Mortimer did not marry Margaret until 1210, so it is still probable that in the 1160s it was Robert Mortimer of Wigmore who held Lye. ⁶⁶ These manors accounted in sum for 12 hides and 1 virgate of geld; Rochford, with 1 hide and 1 virgate, had 2 ploughs in demesne, while Myndd Brith (referred to as Rugenoura in the 1166 rubric), also with 1 hide, had 4 demesne ploughs. Dorstone was assessed at 7 hides’ geld and Burhstanestone three; HDB 64. ⁶⁷ Ibid. 55–6. The Monmouth demesne manors were Stretton Grandison, which had 3 1/2 geldable hides and 2 ploughs in demesne, and was worth £7 in 1086 (but £9 TRE); Whitwick, with 2 geldable hides, 2 demesne ploughs, and a value of 20s.; Ashperton, with 5 1/2 geldable hides, 4 demesne ploughs, and a value of £6; and Mainstone in Munsley, with 1/2 hide, 1 demesne plough (with room for one more), and a value which had sunk from 30s. to 15s. There was in addition a half-hide at Hundeslawa. ⁶⁸ Mansell Gamage, Bullingham, Holme Lacy, Mansel Lacy, King’s Pyon, Weobley, Fernhill, Hampton Wafer, Humber (in Leominster), Stoke Lacy, Halmond’s Frome, Ocle Pichard, Yarkhill, Weston Beggard, Tedstone Wafer, Leadon, Bishop’s Frome, and Castle Frome.

30

Lords of the Central Marches

composed 67 geldable hides with thirty-one demesne ploughs, and were worth a total of £61 5s.⁶⁹ By 1166, using the marginal annotations of Herefordshire Domesday, it is apparent that Hugh de Lacy retained only eight of the original eighteen manors in demesne.⁷⁰ These manors in Domesday ﬁgures were valued at £26 5s. Using these Domesday values, the only basis for comparison, only 43 per cent of the 1086 value of their demesne remained to them in 1166. Some of the most valuable manors had passed out of the demesne. Mansell Gamage, at £8 the second most valuable manor in 1086 along with Weobley, was in the hands of Godfrey de Gamages in 1166, and the most valuable, Stoke Lacy at £10, was held by a certain ‘Oddo’, who cannot be identiﬁed from the Lacy carta. In the case of Mansel Lacy, worth £5, as we have seen above, it had been completely lost to the Lacys as William de Braose had gained it during Stephen’s reign. Of course some, if not most, of the remaining demesne manors may well have increased not only in absolute but also in relative value, although a certain measure of inﬂation should perhaps be weighed against that. But a large amount of demesne had been granted out. Further, Hugh de Lacy’s admission that he was unable to command the service owed from two of the three fees that Elias de Say held from him points to a weakening of the honour.⁷¹ This diminution of the demesne also provides an explanation for Hugh’s eagerness to conquer new lands in Ireland when given the opportunity by Henry II. Although the Lacy demesne lands would seem to have been most affected by the need to establish tenants, demesne had also declined in the case of the Cliffords and Monmouths, and this process had myriad effects on these families. The tenurial structure of Herefordshire was thus in place by 1166, the result of subinfeudation that had been well under way in 1086 and was ⁶⁹ HDB 39–52; the valets for Holme Lacy (Hamme) were left out of the 1160–70 recension, but are entered in Domesday itself; DB, fo. 181d; Wightman, The Lacy Family in England and Normandy, 151, supplies the ﬁgure of £64 12s. 5d .; I am unable to reach this or to follow Wightman’s calculations. ⁷⁰ Holme Lacy, Weobley, Yarkhill, Weston Beggard, Bishop’s Frome, Leadon, Castle Frome, Fernhill, and Halmond’s Frome. I list the latter despite the absence of an annotation of a tenant in HDB, as the Devereux later held it and William Devereux is listed for three fees in the Lacy carta, while Roger Devereux accounts for four fees and there is no mention of Roger in the Hereford recension; HDB 99; RBE 281–3. ⁷¹ RBE 281–3.

The Shaping of the Frontier

31

largely completed by 1166. It is reasonable to assume that this process likewise had occurred in the adjacent Marcher lordships. The granting of land, of course, helped to create ties of service and dependence between lords and knights. The witness lists of the charters of Miles and Roger, the earls of Hereford and lords of Brecon, ﬁlled as they are with the names of their tenants, show this clearly. Miles of Gloucester and Roger of Hereford were able to call on their feudal tenants for support during the civil war of Stephen’s reign. But, for lords, these ties came at a price. We have seen how the demesne land of some of our magnate families, at least in Herefordshire, had been severely depleted by 1166. Nothing more clearly demonstrates the lack of land with which to entice allies, forge familial links, and reward followers than the struggle between Gilbert de Lacy and William II de Braose over the manor of Mansel Lacy during Stephen’s reign. This conﬂict was underlined by the conventio between Roger, earl of Hereford, and Braose, a major selling point of which was Roger’s backing of Braose in his claim to Mansel Lacy. This scarcity of land and the attendant pressures on remaining demesne, combined with an increasing inability to gain ground at the expense of the neighbouring Welsh, were to have a serious effect on both the magnate and knightly families examined in this book. Some, such as Hugh II de Lacy, looked to expand in Ireland; indeed, the wealth and opportunities there largely shifted the priorities of the Lacy family away from the March in the later twelfth and the thirteenth centuries.⁷² Irish revenues then helped to subsidize Lacy activities along the March.⁷³ Other families looked to royal patronage; the Cliffords sent their sons to become household knights of King John; John de Monmouth was a royal justice, keeper of the king’s forests south of the Trent, and warden of the March in the thirteenth century; while, as we shall see, the greatest example of the rewards (as well as pitfalls) of reliance on royal muniﬁcence is that of William III de Braose.⁷⁴ Finally, the shortage of land contributed to rising levels of debt in the thirteenth century for many of the important families of the region, a situation which was politically signiﬁcant and one to which we shall return later. ⁷² See Ch. 3. ⁷³ For an example of the Lacys’ economic exploitation of their Irish estates, see Ch. 3; for the assignment of Irish revenues to Craswall Priory in Herefordshire, Ch. 3. ⁷⁴ For Clifford, see Ch. 6; Monmouth, Ch. 7; Braose, Ch. 6.

32

Lords of the Central Marches

T H E S H A PI N G O F T H E M A RC H : S E T T L E M E N T A N D D O M I N AT I O N English settlement had been occurring in the lowland areas of southeast Wales since Mercian times, but the incursions after 1066 were qualitatively different. One obvious difference was the use of castles, and attendant organization of land around them, which was such a marked feature of Norman penetration. Castleries were a fairly common feature of both the Norman and, after 1066, English landscape and varied in size. The border counties of Cheshire and Shropshire were organized so that subtenants of the earls held large, compact groups of manors and this may well reﬂect the implanting of military commands meant for defence. On a somewhat smaller scale, so were the rapes of Sussex in England, organized around the south coast ports in the immediate aftermath of the Conquest. On the smallest scale were the small border castleries, such as Clifford and Ewyas Harold, found on the western border of Herefordshire.⁷⁵ Herefordshire thus had castleries, but they were very compact and limited to certain border points, while the majority of the county was a tenurial patchwork and looked much like any other English county in patterns of landholding. In the March these building blocks of military domination came into their own. In both cases, however, what made these border castleries distinctive was that they were not simply the caput of a possibly far-ﬂung honour, but instead had surrounding fees granted and organized so as to supply castle guard in the most efﬁcient manner.⁷⁶ It is debatable whether these compact groupings of manors, which were not the norm in England and required the overriding of antecessorial succession by the king, were for purely defensive purposes. Some such groupings were not situated along any border, and many were broken up fairly soon after the Conquest.⁷⁷ It has been argued that such groupings were for the purposes of short-term defence, but it has also been shown that many castleries were from the ﬁrst founded as ⁷⁵ J. Le Patourel, The Norman Empire (Oxford, 1976), 308–9. ⁷⁶ F. M. Stenton, The First Century of English Feudalism, 2nd edn (Oxford, 1961), 194–6. ⁷⁷ Le Patourel, The Norman Empire, 310. Le Patourel’s citing of Herefordshire after Roger of Breteuil’s forfeiture in 1075 is no longer convincing given the work of Lewis, ‘The Norman Settlement of Herefordshire’, 202–5.

The Shaping of the Frontier

33

administrative centres for a group of manors, rather than as defensive strong points.⁷⁸ This may have been the case for some Herefordshire castles such as Weobley, which were somewhat distant from the border and were the centre of a number of dependent manors. But those such as Clifford or Ewyas Lacy in newly conquered territory were the ﬁrst instruments of conquest, plain and simple.⁷⁹ It was necessary to offer at least the appearance of protection in order to lure the settlers necessary to establish manorial cultivation in newly won areas. For example, Gilbert ﬁtz Richard of Clare colonized Ceredigion with English settlers only after he and his men had thoroughly secured the region by building castles.⁸⁰ In Ireland, an arena of English expansion which possesses some useful similarities for the study of the Welsh March, Robert, son of Philip de Barri, conﬁrmed a grant of land ‘which is so close to my castle that it can be easily and conveniently cultivated from said castle’.⁸¹ One can imagine that the grantee, taking up his property in a wild and dangerous new land on the edge of the ‘English empire’, felt appropriately grateful. In the south-eastern March, the borough of Brecon initially existed within the bailey walls of Brecon Castle; when it later expanded beyond the castle, town walls were built and maintained throughout the medieval period.⁸² The majority of the land initially granted to the monks of Monmouth by Wihenoc de Monmouth lay just outside his castle—tres carrucatas terrae juxta castrum de Monemue.⁸³ The settlement that castles enabled helped to underpin military domination, and that domination was one route to the economic exploitation of new land. Of course, in most cases the castle served a number of functions—military, administrative, and judicial—as well as being a potent and tangible symbol of the lord’s wealth, power, and status. The castle was not simply a military construction built for the purposes of defence or offence but was the cornerstone of domination and exploitation. The monastery and priory as well as the borough often accompanied it, and ⁷⁸ Le Patourel, The Norman Empire, 310–11; Paul Dalton, Conquest, Anarchy and Lordship: Yorkshire 1066–1154 (Cambridge, 1994), 30–1. ⁷⁹ Weobley Castle is mentioned as early as 1138, when it was taken during the civil war. Being a somewhat unusual combination of ringwork and bailey, it may have been based on a pre-Conquest fortiﬁcation and have been used by Normans quite early; D. J. Cathcart King, Castellarium Anglicanum: An Index and Bibliography of the Castles in England, Wales and the Islands, 2 vols (1983), i. 212. ⁸⁰ R. R. Davies, Domination and Conquest: The Experience of Ireland, Scotland and Wales 1100–1300 (Cambridge, 1990), 42. ⁸¹ Register of St. Thomas, Dublin, ed. John T. Gilbert, Rolls Series (1889), 214. ⁸² Davies, ‘Brecon’, 50. ⁸³ Mon. Ang. iv. 596.

34

Lords of the Central Marches

wherever it was established, this combination worked to spread military power, cultural inﬂuence, and economic development and exploitation.⁸⁴ Castle, religious house, and borough were the ‘classic features of major headquarters estates’ and were common throughout Norman England.⁸⁵ The castle was thus only the ﬁrst step in establishing control of a newly conquered area and asserting lordship over it. The need to establish control was certainly present in the March of Wales. The section of the March examined here is perhaps the most densely covered in castles in the British Isles and Ireland. The lordship of Brecon (that is, that part of the honour of Brecon in Wales proper), for instance, has a very high density of fortiﬁcations, even by the standards of Wales, that land of castles par excellence. In Brecon there is one castle to every 17.9 square miles (see Map 3).⁸⁶ Of the forty identiﬁed castles in modern pre-1974 Breconshire, which included the medieval lordship of Builth, the vast majority date from the ﬁrst century of the English advance into the area. This can be determined from the number of sites that use a motte or motte-and-bailey construction, a form that had been abandoned by the thirteenth century.⁸⁷ Four of these, Brecon, Bronllys, Crickhowell, and Tretower, were all centres of either lordship (in the case of Brecon) or sub-lordships (the other three). The most important sites were situated advantageously enough to be reused over time, with new fortiﬁcations being erected on the foundations of the old. The majority of these mottes are in the Wye and Usk valleys, in the most fertile country standing close to the better agricultural land, although they usually occupy a position of local prominence.⁸⁸ Of course, a number of these Marcher castles were small, rather poor affairs, raised on small fees, sited on level ground, and no doubt often only meant to be temporary; in the words of one historian of medieval English castles, the small motte-castle of the Welsh March ‘afforded little or no splendour—protection was everything’.⁸⁹ ⁸⁴ Davies, Domination and Conquest, ch. 1. ⁸⁵ For example, in the north at the Vescy manors of Alnwick and Malton; Keith J. Stringer, ‘Nobility and Identity in Medieval Britain and Ireland: The de Vescy Family, c.1120–1314’, in Brendan Smith (ed.), Britain and Ireland 900–1300: Insular Responses to Medieval European Change (Cambridge, 1999), 216. ⁸⁶ Cathcart King, Castellarium Anglicanum, i. 15. ⁸⁷ Of the forty sites, twelve are mottes and seven are motte-and-bailey, and of the thirteen masonry castles, four were built on motte-and-bailey sites. ⁸⁸ Cathcart King, Castellarium Anglicanum, i. 15. ⁸⁹ D. J. Cathcart King, The Castle in England and Wales: An Interpretative History (1988), 5.

The Shaping of the Frontier

35

Across the border in Herefordshire, castles actually are even more abundant.⁹⁰ Here, the ratio of castles to square miles is higher than in any other English county excepting that county of the northern border Northumberland where the castles are often of a later date; there is one castle in Herefordshire for every 9 square miles.⁹¹ Once again, the majority (seventy-two) are earthwork castles, with mottes and motte-and-baileys evenly divided (thirty and twenty-seven respectively). As with Brecon, a number of masonry castles, such as Bredwardine and Clifford (now in Herefordshire though once in the March), Weobley and Wigmore, as well as Hereford Castle itself, were built on earthwork sites. Many of the smaller earthwork castles of Herefordshire are of limited defensive value, and are comparable in purpose with the border towers of the north. This comparison is apt, for the small frontier motte was never designed to delay an advancing army or withstand a prolonged siege. Rather it was intended to provide temporary shelter in the face of Welsh raiding parties, and to stamp the new lord’s authority on the area. That said, some of the fortiﬁcations in Herefordshire were much more substantial, particularly those, such as Clifford Castle, that later were converted from earthwork to masonry.⁹² Castles, imposing buildings that they are in themselves, need garrisons. The process of subinfeudation attendant on castle-building in the Welsh March can partly be demonstrated from that carried out, often by men who were from the March, in the ﬁrst generation of the English expansion into Ireland in the 1170s and 1180s. The Song of Dermot and the Earl describes the subinfeudation of Meath by Hugh II de Lacy (d. 1186): In such manner, know you all, Was the country planted (herberge) With castles and with cities, With keeps and with strongholds. Thus well rooted (ben aracinez) were The noble and renowned vassals.⁹³ ⁹⁰ The modern county of Herefordshire includes some areas that were considered to be in the March in the 12th and 13th centuries, such as Clifford, Ewyas Harold, and Ewyas Lacy. ⁹¹ Cathcart King, Castellarium Anglicanum, i. 201. ⁹² Ibid. 201, 203. Clifford is situated on a ridge overlooking the Wye, and has a 13th-century masonry ward with three round towers, the ubiquitous twin-towered gatehouse, and a hall. ⁹³ The Song of Dermot and the Earl, ed. and trans. G. H. Orpen (Oxford, 1892; repr. Felinfach, 1994), lines 3202–7.

36

Lords of the Central Marches

The poet’s metaphor was well chosen; the images of planting he used show the connection in the mind of the English between castle and manor. For the castle was not merely a base for further expansion or a defensive shelter against the Welsh, but was also the administrative and judicial centre of the lordship. Although the administrative apparatus of the Marcher lord was much smaller in the eleventh and twelfth centuries than later, even then the castle was the collection point for tribute from the Welsh of the uplands. The lord’s honour court would also be held in the castle; thus, in 1175 Seisyll ap Dyfnwal was lured by William de Braose to his court at Abergavenny Castle and killed.⁹⁴ Another crucial step in securing newly conquered territory was the founding of towns. Small urban settlements or pre-existing rural settlements which had been granted urban privileges served as a magnet for the immigration of foreign settlers, and as a means of drawing the native population into trading relationships which would begin the process of cultural assimilation.⁹⁵ Although the lord’s immediate priority was the military control of his new lordship, that control was meant to enable the economic exploitation which the processes of manorialization and urban development entailed.⁹⁶ Further, the inﬂux of settlers drawn by the promise of new lands and urban privileges was the instrument of the Anglicization of the newly conquered territories. By 1086 burgesses were to be found in the area at Wigmore, Hereford, Clifford, and Ewyas, all locations which possessed a castle.⁹⁷ Hereford of course was in existence before the arrival of the Normans, but in the March, towns were almost entirely plantations. Brecon provides a good example. Bernard de Neufmarché, the conqueror of Brycheiniog, reached Brecon by 1091 and had built a castle there by 1093. Although there may have been a small Welsh settlement already in place, the Norman town was founded c.1100, as burgages were granted in the charter which gave the church of St John the Evangelist to Battle Abbey and thus founded Brecon Priory.⁹⁸ The town of Radnor was in existence in the early twelfth century as Philip de Braose, lord of Radnor, granted a ⁹⁴ Brut, 71, s.a. 1175. ⁹⁵ For the distinction between towns as economic entities, and the town as a legal creation, see Bartlett, The Making of Europe, 167–72. ⁹⁶ See e.g. R. R. Davies’s comment on ‘the proﬁteering, market-oriented mentality’ of the new English lords in Wales and Ireland; The First English Empire, 148. ⁹⁷ Darby and Terrett (eds), Domesday Geography, 64, ﬁg. 23. Clifford and Ewyas were listed in Domesday Book under Herefordshire but were later in the March. ⁹⁸ ‘Cart. Brecon’, 14. 141–2.

The Shaping of the Frontier

37

burgage there to St Peter’s, Gloucester.⁹⁹ Builth has been roughly dated to 1095–1102 while Abergavenny dates from 1087–1100. Monmouth probably was an early urban foundation, as the castle was built in 1070 by William ﬁtz Osbern at the same time as Clifford, Ewyas Harold, Chepstow, and Wigmore. Wigmore and Clifford had chartered boroughs established by ﬁtz Osbern, so it is probable that the town of Monmouth is contemporaneous.¹⁰⁰ The granting of privileges, enshrined in charters, was a key to attracting settlers to these new urban settlements.¹⁰¹ These privileges included favourable terms for the holding of land, the right to hold a court, and borough status. The privileges of certain towns became models for those in other towns, and in the March that town was Hereford, which had in turn been granted the privileges of Breteuil in Normandy.¹⁰² The conﬁrmation charter of Brecon’s liberties, issued by its lord, Humphrey de Bohun, in 1270, states that Brecon was to enjoy omnes libertates et liberas consuetudines quas burgenses Herford’ habent.¹⁰³ The great economic advantage enjoyed by towns in the March was that of monopoly. A Marcher borough had a much larger outlying region which it dominated than did an English borough, which had to compete with ‘non-burghal market villages’ and country fairs.¹⁰⁴ This also beneﬁted the Marcher lord whose town it was, as tallages and customs payments could tap this wealth. ⁹⁹ Historia et Cartularium Monasterii Sancti Petri Gloucestriae, 3 vols, ed. W. H. Hart, Rolls Series (1863–7), i. 110; ii. 103, no. 594. The grant is undated, and is the subject of some dispute, as it is alternatively thought to date from the late 12th century, though that date is rendered problematic by the grantor being Philip de Braose; Maurice Beresford, New Towns of the Middle Ages: Town Plantation in England, Wales and Gascony (1967), 573. ¹⁰⁰ Beresford, New Towns of the Middle Ages, 536, 560. Although it should be noted that, unlike the case of the foundation of Brecon Priory, no burgage plots or rents were granted to Monmouth Priory on its foundation in 1086 and this may argue for a somewhat later foundation; Mon. Ang. iv. 596. ¹⁰¹ Bartlett, The Making of Europe, 177–82. ¹⁰² The laws and customs of Hereford and Breteuil were linked in the description of the privileges of Rhuddlan described in Domesday Book; Mary Bateson, ‘The Laws of Breteuil’, EHR 15 (1900), 302. ¹⁰³ A. Ballard and J. Tait (eds), British Borough Charters, 1216–1307 (Cambridge, 1923), 23. Davies, ‘Brecon’, 54–5, states that this was Brecon’s ﬁrst charter of liberties. It is perhaps more accurate to say that this is the ﬁrst extant charter, as the use of conﬁrmasse makes it clear that prior liberties are being conﬁrmed. Considering that Clifford was chartered at the time of Domesday and the important role of liberties in attracting burgess settlement, it is likely that the Bohun charter reﬂects long-established custom. ¹⁰⁴ Beresford, New Towns of the Middle Ages, 529.

38

Lords of the Central Marches

The third element in the classic triad of Norman English domination was the establishment of religious houses by the new lords. A new abbey or priory served a number of functions. First and foremost, the house catered for the religious needs of the lord and his followers. These needs included intercessionary prayer for the ancestors and other relatives of the lord, and burial. Patronage of a house quite often meant not only a place for possible retirement and burial for the founder and his heirs, but careers for younger sons or offspring not suited to the secular world as well. Both the fact that Norman and English Churchmen and their secular cousins disapproved of the Welsh Church, and the desire of lords to imprint their power on the physical and spiritual landscape of their new lordships, are apparent in their expansion into Wales. They did not on the whole patronize existing Welsh churches and religious houses but rather appropriated and changed them, or established new foundations. One reason for this is that the Welsh Church in the eleventh century was certain to offend the sensibilities of post-Gregorian Norman Churchmen. Although there were Welsh bishops, there was no organized system of dioceses, nor was there a metropolitan authority; this last characteristic left archbishops of Canterbury free to assert their claims to primacy. Welsh bishops did wield a broad spectrum of powers—spiritual, temporal, economic, and judicial; indeed, their authority approximated quite closely, often indistinguishably, to that of their lay counterparts. Churches were ‘monastic’, but not in any sense that a Continental Benedictine would have recognized or approved of. Under an abbot, canons lived the lives of secular clerks, marrying and passing on their property and ofﬁces to their children, founding clerical dynasties. These ‘abuses’, as they had come to be considered in the postGregorian world, could be found throughout the Church in western Europe in the eleventh and twelfth centuries, but in the political context of Norman English expansion and conquest, they provided a ready rationale for the dispossession or reform of the Welsh Church by the newcomers. This was a process with which the new lords of the March cooperated fully. Following the pattern established in dealing with the Anglo-Saxon Church after the Conquest, there was a political necessity to the subjugation of the Welsh Church that went hand in hand with the spiritual one. As one historian has written, ‘Control of the church was a natural corollary of political conquest, exploitation of its wealth but another aspect of the domination and settlement of the country.’¹⁰⁵ ¹⁰⁵ Davies, The Age of Conquest, 179.

The Shaping of the Frontier

39

The majority of religious houses in the area connected with our families date from the ﬁrst decades after the Conquest. The ﬁrst new foundations in Herefordshire and the March were all alien priories, dependent on mother houses either in Normandy or in England. Brecon, which would become the recipient of Braose patronage when that family became lords of Brecon, dates from c.1100 and was dependent on Battle Abbey. The Lacy foundation of St Guthlac, Hereford, a dependent priory of Gloucester Abbey, dates from 1107–22, while Clifford Priory, a Cluniac priory dependent on Lewes, was founded in 1129 or 1130.¹⁰⁶ Monmouth had been founded in 1086 and was a daughter house of St Florent of Saumur, while Abergavenny, established the following year by Hamelin de Balun, was later made dependent upon St Vincent, Le Mans, by Balun’s eventual successors as lords of Abergavenny, the Braoses.¹⁰⁷ These initial foundations were Benedictine houses and priories, but the twelfth century brought both new orders and changes in patronage. These foundations also further established and delineated the power of the new lord in the locality. As an adjunct to this aim, sometimes Anglo-Saxon houses already in existence in the border counties, or occasionally Welsh houses in the March, were patronized. One example of this is the Lacy patronage of St Peter’s, Gloucester. Walter, the ﬁrst of the Herefordshire branch of the Lacys, was buried at St Peter’s after he fell to his death while supervising the construction of St Guthlac in Hereford.¹⁰⁸ Lacy probably patronized St Peter’s because it was an important Anglo-Saxon house in the area of a contested border zone; it was in part a nod to local sensitivities. Previous lords, such as Bernard de Neufmarché, had granted lands and rights in the conquered Welsh borderlands to St Peter’s, helping to solidify support in their English lands.¹⁰⁹ This use of frontier resources to support religious houses elsewhere would be mirrored later when revenues from Lacy lands in Ireland were assigned to both Craswall Priory and Llanthony.¹¹⁰ St Peter’s was also patronized by the Conqueror and William Rufus, ¹⁰⁶ D. Knowles and R. N. Hadcock, Medieval Religious Houses in England and Wales (1953), s.n. ¹⁰⁷ Mon. Ang. iv. 613. ¹⁰⁸ Historia et Cartularium Monasterii Sancti Petri Gloucestriae, i. 73; Wightman, The Lacy Family in England and Normandy, 168. ¹⁰⁹ Emma Cownie, ‘Gloucester Abbey, 1066–1135: An Illustration of Religious Patronage in Anglo-Norman England’, in D. Bates and A. Curry (eds), England and Normandy in the Middle Ages (1994), 148. ¹¹⁰ HDC, no. 482, a grant by Walter II de Lacy; Mon. Ang. vi. 1. 138, conﬁrmation by Walter II.

40

Lords of the Central Marches

and thus was an even more attractive house to the local aristocracy owing to its royal connections.¹¹¹ In other instances, however, an existing Anglo-Saxon or Welsh church was taken over by the new lord in order to endow a new, ‘Norman’ priory. For example, in Henry I’s reign Wihenoc de Monmouth brought Benedictine monks from St Florence at Saumur in Anjou and installed them initially in the church of St Cadog at Monmouth.¹¹² By a process of rededication or renaming, and by the expulsion of native canons and their replacement, the new lord was able to found a ‘new’ house at little cost. Finally, the new religious foundation served as a centre for the loyalty of the lord’s tenants and afﬁnity; ‘the support of a particular religious house was frequently the expression of corporate solidarity within a feudal grouping’.¹¹³ A lord’s tenants often would contribute to the house he patronized. This tendency can be seen in the foundation charter of Brecon Priory. The new lord, Bernard de Neufmarché, granted to the priory one mill and shares in yet another, the rents of ﬁve burgesses, and land in Wales, as well as the lands and tithes pertaining to the churches of four manors with the tithes of a ﬁfth.¹¹⁴ His men then followed Bernard. Picard gave 3 carucates and the tithe of his grain and animals in Wales; Richard del Mans and ‘Harold’ gave a carucate each. Walter Cropus gave his tithe from Llansanffraid on the Usk and the church of Cleobury North in Shropshire. Ulger granted his right of tithe in Wales, while Walter de Lyonshall, Roger de Baskerville, William ﬁtz Gerald, and Robert Devereux each gave the rent of one burgess.¹¹⁵ Advowsons and tithes were popular gifts at the turn of the eleventh century, as the Hildebrandine reformers opposed their ownership by laymen.¹¹⁶ As the twelfth century progressed, the Benedictine order, although never completely eclipsed, was surpassed in popularity among the aristocracy by the Cistercians and the new orders of canons, primarily the Augustinians. The Cistercian house that had the greatest impact on the region was Dore Abbey, founded in 1147 by Robert of Ewyas ¹¹¹ Cownie, ‘Gloucester Abbey’, 147–8. ¹¹² Mon. Ang. iv. 595. ¹¹³ C. Harper-Bill, ‘The Piety of the Anglo-Norman Knightly Class’, ANS 2 (1979), 67. ¹¹⁴ Those manors were Bodenham, Burghill, Patingeham, and Hardinton’ ; tithes only were granted from Brinsop. ¹¹⁵ ‘Cart. Brecon’, 14. 141–2. ¹¹⁶ Harper-Bill, ‘The Piety of the Anglo-Norman Knightly Class’, 67; G. Tellenbach, The Church in Western Europe from the Tenth to the Early Twelfth Century, trans. T. Reuter (Cambridge, 1993), 289–91.

The Shaping of the Frontier

41

as a daughter house of Morimond in Champagne. The Augustinian houses of Llanthony Prima in Monmouthshire and Llanthony Secunda in Gloucestershire also attracted much patronage. Finally, Wigmore, another house of Augustinian canons, was initially founded between about 1131 and 1135 at Shobdon by Oliver de Merlimont, seneschal to Hugh I de Mortimer (d. 1148/50), but only settled at its ﬁnal site in 1179.¹¹⁷ The religious patronage of the area’s magnate and knightly families is examined further in Chapters 3 and 4. Our knowledge of the tenurial organization of Marcher honours is scant for this period, and at times we can only describe those manors which physically lay in Herefordshire, even though they were part of Marcher honours in a feudal sense. Our clearest view of these lordships comes from the inquests made in 1211–12 when several of them had come under royal control owing to the ﬂight of William de Braose in 1208. The honour of Radnor, although extensive in surface area, was extremely small when measured in knights’ fees, having ﬁve and a half.¹¹⁸ In Radnor, which consisted almost entirely of moor, it took more land to support a knight’s fee. Pembridge, which actually lay in Herefordshire, was perhaps the most valuable of its manors. In 1086 it had 9 hides (minus 1 virgate), although no valuation.¹¹⁹ The honour of Abergavenny, if still small, was somewhat more impressive because it had a good deal of fertile lowland, and thus supported thirteen, one-third, and a quarter fees. One point of interest in Abergavenny is the presence of Welshmen holding by knight service; the list of tenants in 1212 is headed by Iorwerth ap Madog, and the Tudur ap Iorwerth and Madog, his brother, who held 7 bovates apiece were probably his sons.¹²⁰ The honour of Brecon was by far the largest of the honours of the region, with thirty-two knights’ fees in 1212, held by thirty-six men, one woman, and one group of heirs.¹²¹ In 1212 the greatest tenant of the honour of Brecon was Walter de Clifford junior, with four fees.¹²² The Cliffords were also lords in their own right, possessing the barony ¹¹⁷ Dickinson and Ricketts, ‘The Anglo-Norman Chronicle of Wigmore Abbey’, 415–17. ¹¹⁸ RBE 603. ¹¹⁹ HDB 59. ¹²⁰ RBE 602–3. ¹²¹ RBE 601–2. The group of heirs were those of William de Waldeboef; the woman was described as the domina de Melyenoc. ¹²² Ibid. In 1212 Walter de Clifford junior is Walter III de Clifford, whose father, Walter II, lived until 1221. We have here another instance of a demise of lands by a Marcher lord to his son, like that of the Braoses c.1203; see Ch. 6.

42

Lords of the Central Marches

of Clifford, situated on the Wye and abutting the lordship of Brecon on the north-east. Clifford was not forfeited to the Crown in 1210, but later in the century the Cliffords were assessed for scutage on nine fees in Herefordshire and one in Shropshire.¹²³ The Shropshire fee was centred on the manors of Corfham, Culmington, and Clive, and the hay of Ernestou.¹²⁴ The Herefordshire manors included Dorstone in the Golden Valley and Bridge Sollers on the Wye, and a group of manors to the north of Hereford: Ullingswick, Hope-under-Dinmore, Ford, and Hamnish Clifford. Finally, the honour of Monmouth was composed of ﬁfteen knights’ fees. In 1166 ten fees were distributed among nine men and their co-parceners (et alii); one of the ten fees was held by the Knights Hospitallers, with ﬁve fees still held in demesne.¹²⁵ In 1243 the Monmouth manors in Herefordshire were Hope Mansell, Mainstone, Wolsopthorne, and Ashperton and Stretton Grandison, with these last two held in demesne.¹²⁶ The Monmouth family supplemented their holdings by the marriage of John I de Monmouth (d. 1248) to Cecily, daughter of Walter Walerand, who brought him one-third (some eight and one-third fees) of the honour of West Dean in Wiltshire.¹²⁷ By 1166 the south-eastern March was a land of two characters. In the lowlands of Brecon, Abergavenny, and Monmouth the process of subinfeudation was complete, and further subinfeudation below the level of the mesne tenants had undoubtedly occurred. Some power may have devolved into the hands of these mesne tenants, but the compact geography of the Marcher lordships and the lack of a coexisting structure of royal government and thus control ensured for the moment that these lordships were not fractured. Further, the military exigencies of life along a contested frontier zone meant that the feudal honour in the March still had a reason for its continued functional existence. This ‘feudal matrix’ had been underpinned by substantial immigration and colonization, with the development of a full-blown manorial system supplemented by small boroughs, which served as both trade entrepôts and magnets for further immigration from England and beyond. ¹²³ I. J. Sanders, English Baronies: A Study of their Origin and Descent 1086–1327 (Oxford, 1960), 36 n. 1. ¹²⁴ These manors had been the subject of a family dispute at the end of the 12th century. They were held in 1199 of the Crown for the service of one knight; Rot. Chart. 8; Rot. Cur. Reg. ii. 159; TNA, CP 25(1)/193/2, nos 3, 27. For further discussion of this dispute, see Ch. 6. ¹²⁵ RBE 280. ¹²⁶ BF 800, 808, 815. ¹²⁷ Sanders, English Baronies, 96 and n. 7.

The Shaping of the Frontier

43

In the uplands of Brecon, Radnor, Builth, and Maelienydd, however, English control was much more patchy. All that could be reasonably expected here was a loose overlordship which approximated to that of former Welsh rulers, in which tributes and renders were collected. In this division between tightly governed, manorialized lowlands and tribute-paying uplands lay the origins of the Englishries and Welshries which would mark the later social landscape of the March. These communities existed in an uneasy relationship, which was at times regulated by alliances and agreements. When these arrangements broke down, the two communities engaged in periodic outbreaks of bloody violence in which both English and Welsh drew on long memories of hostility, treachery, and atrocities. In the 1160s in Brecon, for instance, relations between English and Welsh declined precipitously as William de Braose assumed control of the lordship.¹²⁸

T H E E M E RG E N C E O F T H E M A RC H By 1166, the year of the Cartae Baronum, the concept of the March had emerged, even if the Marcher lordships had not quite reached their ﬁnal form. That would take the playing out of several processes: the further growth of royal government and ius scriptum, the separation of cross-border estates through inheritance and marriage patterns, and ﬁnally, the political events of King John’s reign. But the ﬁrst two of these processes began in the reign of Henry II. This is therefore a good point at which to examine the idea of the March. The concept of the March of Wales, ‘Marchia Walliae’, had developed over the course of the twelfth century.¹²⁹ Writers earlier in the century at various times had used the words ﬁnes, terra, and marchia with varying degrees of precision. Generally they sought to convey the idea of a borderland between England and Wales, although at other times they could mean the actual frontier or border itself, hazy as that was. The March was therefore a geographic term before it was an institutional one; the latter sense only developed along with and in opposition ¹²⁸ See Ch. 5. ¹²⁹ What follows is partly drawn from Kevin Mann, ‘The March of Wales: A Question of Terminology’, WHR 18/1 (June 1996), 1–13. It should be noted, however, that Mann deals only with the terminology of the March, not the broader historical context.

44

Lords of the Central Marches

to the growth of centralized government.¹³⁰ But by the last decade of the twelfth century the terms marchia and ‘Marchia Walliae’ had acquired a much more speciﬁc deﬁnition, that of the lordships which separated England from Wales and which were notable for their various peculiarities in law and custom as seen from the centre. Signiﬁcantly, this change in meaning is evident in royal documentation before it is found in the chronicles. The ﬁrst surviving mention in government records is in 1196 when Henry de Casteillun claimed expenses for carrying out the king’s business in both Wales and the March of Wales.¹³¹ By 1215 article 56 of Magna Carta recognized the existence of three different bodies of law: English law, leges Walliae, and leges Marchiae. By mid-thirteenth century the Tewkesbury chronicler would record that King John had had only the support of a few barons of the March (paucis baronibus Marchiae) during the civil war.¹³² The term in its best-understood meaning had by then become common coin in literary circles. The reference in the Tewkesbury Annals also shows that, like the Northerners, or Nordenses, the Marchers had developed a common identity in the eyes of contemporaries and most probably themselves. All these references are only a recognition of something that had already come to pass. New terms often take some time to become widely used and to penetrate literature. It is perhaps appropriate that it is the records of royal government that ﬁrst use ‘Marchia Walliae’, for the king’s ofﬁcials in the border area would be keenly aware of the difference between Brecon and Bedfordshire, or even Herefordshire. Indeed, it was a changing royal government that helped to deﬁne the relatively unchanging March. Those changes in the scope and power of royal government may well have begun early in the twelfth century, but by Richard I’s reign the differences between the typical English county and a Marcher lordship such as Brecon were so great that a term was needed to describe the society west of Shropshire and Herefordshire. This sharpening of terminology was the result of the increasing reach of royal government. The reorganization of the exchequer following ¹³⁰ Witness Walter Map’s description of himself as ‘a dweller on the marches of Wales’; De Nugis Curialium, ed. and trans. M. R. James, rev. C. N. L. Brooke and R. A. B. Mynors (Oxford, 1983), 194. ¹³¹ The Chancellor’s Roll for the Eighth Year of the Reign of King Richard the First: Michaelmas 1196, ed. Doris M. Stenton, Pipe Rolls Series 45 (1930), 17, 19, 20; cited in Mann, ‘The March of Wales’, 6. ¹³² Mann, ‘The March of Wales’, 10, citing Annales Monastici, ed. H. R. Luard, 5 vols, Rolls Series (1864–9), i. 62.

The Shaping of the Frontier

45

Stephen’s reign, as well as the increased number and frequency of judicial eyres, served to draw Herefordshire and the other counties bordering Wales more tightly within the embrace of Westminster. Where these mechanisms of central government did not operate helped to deﬁne the border between England and the March as much as or more than did the occasional protestations of Marcher ‘liberties’. The same processes were operating in the north; as has been noted of the northern counties, ‘It was not until the middle years of the reign of Henry II that the north began to feel the full weight of royal government.’¹³³ Henry II’s reign was for this reason highly signiﬁcant in the creation of the March, at least from external causes.¹³⁴ Before we examine the role played by royal government in this process, we must ﬁrst examine the players on the scene: the magnates and the knights of the March. ¹³³ Holt, The Northerners, 201. ¹³⁴ See Ch. 5 for campaigns and Welsh ‘policy’, and for royal justice.

3 Lords and Honours The primary creators of frontier society along the March were the lords who led the drive into Wales. With these men came the knightly followers who would eventually form the bedrock of local society in the March.¹ The great lords, of course, at times relied on both the permission of the king to lend their conquests legitimacy, and his military support against the larger Welsh resurgences.² But because English expansion into Wales in this period was not directed from the metropolitan centre, it was these barons who, more than anyone else, placed their stamp on the Middle and south-eastern March in its formative stages. The medieval aristocracy imposed their will through the exercise of lordship, and nowhere in the British Isles was aristocratic lordship stronger than in the March of Wales. Lordship was initially quite powerful in the border counties as well, but was diluted in the course of the twelfth century; this led to a widening distinction between the counties abutting Wales and the neighbouring Marcher lordships. Lordship could be exercised in a variety of ways, through several different, and not mutually exclusive, mechanisms. Along the border, and especially in the March, the ‘lordship’, or ‘honour’, was the primary unit of aristocratic control, dominance, and government.³ While some honours in England lost their cohesiveness from the second half of the twelfth century, the Marcher lordships, and even some which technically ¹ See Ch. 4; for comments on the role of the knightly ‘men on the scene’ in another region of the March, see Rowlands, ‘The Making of the March’. ² For the role of the king in the area, see Ch. 5. ³ The classic (and by now modiﬁed) portrait of the English feudal honour as ‘a feudal state in miniature’ in the 11th and 12th centuries is Stenton’s First Century of English Feudalism; other notable works in a vast literature include Sidney Painter, Studies in the History of the English Feudal Barony (Baltimore, 1943). Holt’s The Northerners stresses the strength of the honour in the north at the time of the 1215 rebellion.

Lords and Honours

47

were within England, such as Wigmore, retained their initial character.⁴ A comparison with honours in the north of England, another region that was far from the metropolitan centre and had a number of well-organized lordships, shows how tightly knit Marcher honours could be. By 1243, when on average one-third of the tenants of any Northumbrian barony held fees outside their lordship, only approximately one-seventh of the tenants of the lordship of Brecon similarly held land outside the honour.⁵ The Marcher lordships were largely self-contained units, and this tenurial integrity accounts both for the survival of the lordship as a working socio-political entity and for much of its character. For this reason our examination of lordship in the region takes the honour as its focus. The tenurial structure and geographic disposition of these honours have already been discussed.⁶ What will be examined here are the working elements of lordship, both in western Herefordshire and in the March. It is easiest to examine these elements separately: military lordship, judicial lordship, economic lordship, religious patronage, and marriage strategies. But it is important to remember that, in reality, good and effective lordship was a seamless garment; each aspect affected and was in turn reinforced by the others. Likewise, a weakness in one area could undermine the overall strength of lordship. It is not enough, however, simply to describe the character of lordship in the region. Lordship is hardly ever static; did the nature and exercise of aristocratic control in our area change over time? In the long century from 1150 to 1270, when easy territorial gains against the Welsh were no longer a possibility, we might expect to see a consolidation of lordship in our area—what one historian has termed a ‘second conquest’.⁷ Indeed, this is exactly what occurred in the neighbouring Clare lordship of Glamorgan from 1245 to 1290. In Brecon this consolidation was not accomplished permanently until the middle years of the fourteenth century under Humphrey IV de Bohun, earl of Hereford (d. 1361). We should ask, though, whether there were earlier efforts at the consolidation of lordship in the honours of the Braoses, Cliffords, Lacys, Monmouths, and Mortimers. And, if there were such attempts, how far-reaching were they, and how successful? ⁴ Although note that not all honours were in decline in the 13th century; see D. A. Carpenter, ‘The Second Century of English Feudalism’, P&P 168 (Aug. 2000), 30–71. ⁵ See Ch. 4; for Northumbria in 1242–3, Holt, The Northerners, 41. ⁶ See Chs 1 and 2. ⁷ Davies, Lordship and Society in the March of Wales, 88.

48

Lords of the Central Marches

M I L I TA RY LO R D S H I P Discussing Gerald of Wales’s description of the Welsh, Robert Bartlett commented: His analysis of Welsh military tactics can be seen as a keystone linking various other aspects of his description. On the one hand, it is clear that these tactics were a consequence of the terrain, and so Gerald’s description of warfare interlocks with his description of the geography of Wales. On the other hand, Welsh social structure was moulded into a certain shape by the requirements of military technology and organization, and this relationship comes out clearly from Gerald’s analysis. At a further level, Gerald was also aware of the way a people’s psychological characteristics were inﬂuenced by both military organization and social structure. His analysis of the Welsh is a complex in which his account of military matters has a central place.⁸

This approach works equally well for a description of Marcher society. It is a fact, and something of a cliché, that Marcher society was warlike, as it existed along a violent and militarized frontier. But nevertheless the demands of conquest, consolidation, and defence were the primary shapers of Marcher life. The importance of the castlery in the foundation and organization of the Marcher lordships and border counties has already been noted.⁹ The key role played by the castlery in turn inﬂuenced the organization of aristocratic society along the border, and everything else—judicial lordship, economic lordship, religious patronage—was intimately linked to the military nature of the border lordships. Furthermore, the military concerns of frontier society were still keenly felt in Herefordshire and the other border counties in the twelfth and thirteenth centuries, even as in other aspects (especially judicial) the boundary, both physical and mental, between them and the Marcher lordships was deﬁned more clearly. Describing the Norman English in Wales and Ireland, Gerald of Wales referred to Gallica militia.¹⁰ In one sense he was referring to the technology of war. Anglo-Norman and Angevin warfare was largely, though not entirely, based on two features: the heavily armoured horseman and the castle. But even more important was the English advantage in armour and in ‘ammunition’, as John Gillingham has put ⁸ Robert Bartlett, Gerald of Wales 1146–1223 (Oxford, 1982), 196–7. ⁹ Ch. 2. ¹⁰ Giraldus Cambrensis, Descriptio Kambriae, in Opera, vi. 220.

Lords and Honours

49

it.¹¹ Of course, Gerald was also making a distinction between the French manner of war, with its emphasis by the twelfth century on chevalrie, a code of behaviour in battle and towards captives of high rank, with what came to be seen as the barbarous manner of Celtic warfare.¹² As Gerald wrote, contrasting France with Wales and Ireland, ‘There soldiers are captured; here they are beheaded. There they are ransomed; here they are killed.’¹³ But the English way of war was of limited use in the March beyond lowland areas such as the coastal plains of Glamorgan, and in our region the major river valleys of the Llynﬁ, Wye, and Usk, owing to the terrain and to the tactics of the Welsh. The Welsh were lightly armed and armoured; they relied on speed and surprise for success, and melted back into the wooded highlands when they suffered setbacks. There was little opportunity for the English to come to grips with the Welsh on their preferred terms, the cavalry charge. Indeed, the English were highly vulnerable to ambush. It was in such an ambush in 1136 that Richard ﬁtz Gilbert de Clare was killed in Gwent.¹⁴ Henry II’s ﬁrst expedition against the Welsh, in 1157, almost ended in similar disaster for the king, when the royal party left the coastal route and cut inland into the forest and was ambushed near Hawarden. The king himself barely escaped.¹⁵ In 1174 Welsh mercenaries in the employ of Henry II ambushed a French baggage train meant for the force besieging Rouen; their mention alongside Henry’s Brabanc¸ons may be taken to mean that, like the hated routiers (or even more so), the Welsh were considered outside the chivalric pale.¹⁶ ¹¹ J. Gillingham, ‘Conquering the Barbarians: War and Chivalry in Twelfth Century Britain’, in Gillingham, The English in the Twelfth Century: Imperialism, National Identity and Political Values (Woodbridge, 2000), 49. By ‘ammunition’ Gillingham means the large quantities of arrowheads which could be produced in England, along with iron armour. Gerald of Wales, although describing the Welsh as using ‘small mail coats’ (loricis minoribus), stressed the light nature of Welsh troops, and this made them vulnerable to heavier armoured and equipped Norman forces, as well as archers, in the open ﬁeld; Descriptio Kambriae, 180. ¹² Gillingham, ‘Conquering the Barbarians’, 51–8. ¹³ Descriptio Kambriae, vi. 220. ¹⁴ Gesta Stephani, 16. Fitz Gilbert did not help matters by dismissing the escort offered him by Brian ﬁtz Count; Lloyd, A History of Wales, ii. 471. ¹⁵ Warren, Henry II, 161. See also D. J. Cathcart King, ‘Henry II and the Fight at Coleshill’, WHR 2 (1964–5), 367–75; J. G. Edwards, ‘Henry II and the Fight at Coleshill: Some Further Reﬂections’, WHR 3 (1966–7), 251–63. ¹⁶ Gesta Regis Henrici Secundi Benedicti Abbatis, ed. W. Stubbs, 2 vols, Rolls Series (1867), i. 74–5; Matthew Strickland, War and Chivalry: The Conduct and Perception of War in England and Normandy, 1066–1217 (Cambridge, 1996), ch. 11.

50

Lords of the Central Marches

A further understanding of the highly ﬂuid and mobile nature of Celtic warfare can be gleaned from descriptions of the Irish, whose tactics and methods were very similar and changed little over the course of the Middle Ages.¹⁷ Irish warfare was based not on the seizing and holding of territory, but rather on the establishment of dominance over other lords and peoples. The primary means of this was the hosting, a raid into an opponent’s lands in order to strip him of economic assets, primarily cattle (prey) but also people, who would be taken captive and enslaved. Speed and often night movement were relied upon by the Irish chevauchées. The raiding column was at its most vulnerable when it had taken prey and was slowed down, as it drove people and cattle back home. This is when most battles were fought, and thus the place of honour for any Irish chief was not in the vanguard but in the rear. Irish warfare was centred on the running battle, the raiding party attempting to evade and outrun its now faster pursuer. The pattern of warfare was quite similar in Wales. Plunder and the devastation of dwellings, as well as the taking of slaves, were a commonplace of Welsh warfare. The seizure of resources rather than territory was the objective of Welsh princes and their war bands. Plunder was particularly important as a means of income and rewarding followers. Movable wealth, rather than land, was the key to any Welsh leader’s position vis-à-vis his rivals; this was because land was relatively abundant while people and livestock were scarce. The Welsh did not necessarily change their pattern of warfare when they fought the English, although in the course of the twelfth century they did successfully adopt many features of English armies.¹⁸ John of Worcester, commenting on the Welsh irruption of 1136 that followed on Henry I’s death and the defeat of an English force in Gower, gives a vivid picture of the effects of Welsh raids.¹⁹ ‘Thereupon the Welsh invaded in force, violently destroyed churches, townships, crops and beasts far and wide, burnt down castles and other fortiﬁcations, slew, ¹⁷ For much of what follows, see Katherine Simms, ‘Warfare in the Medieval Gaelic Lordships’, Irish Sword, 12 (1975–6), 98–108; ead., ‘Gaelic Warfare in the Middle Ages’, 99–115; M. T. Flanagan, ‘Irish and Anglo-Norman Warfare in Twelfth-Century Ireland’, 52–75; Robin Frame, ‘The Defence of the English Lordship, 1250–1450’, 76–98, all in T. Bartlett and K. Jeffrey (eds), A Military History of Ireland (Cambridge, 1996). ¹⁸ For the above, see Davies, The Age of Conquest, 67–70, 74; for Welsh adaptability and the various options it gave them in battle, see Carpenter, The Struggle for Mastery, 114. ¹⁹ The Chronicle of John of Worcester, iii, ed. and trans. P. McGurk (Oxford, 1998), 218–19. For the signiﬁcance (or not) of this battle, see Crouch, ‘The March and the Welsh Kings’, 256–7.

Lords and Honours

51

scattered, and sold into captivity abroad innumerable men, both rich and poor.’ Theobald, archbishop of Canterbury, in a letter to the Pope, described the Welsh as such: ‘For they carry on a regular slave trade and sell Christians into foreign parts across the seas [likely Ireland] where they become the captives of the inﬁdels.’²⁰ As Gillingham has pointed out, Celtic warfare may be viewed as a slave-hunt.²¹ Women were most often captured, but sometimes young men as well. In order for the raiding party to escape its pursuers, the old, weak, inﬁrm, or very young would be killed. Gillingham has persuasively argued that this feature of Celtic warfare was one reason for the shift in attitudes towards the Welsh that emerges in English chronicles by the mid-twelfth century. It was also a reason for the brutal, retributive nature of warfare in Wales (and Ireland) and for the proliferation of mottes, fortiﬁed houses, and defensible church towers.²² Cattle could be freely imported into the March; new settlers could be much harder for a lord to come by. This contrast, however, between the traditional forms of Welsh warfare and the preferred manner of ﬁghting of the English newcomers should not be overdone. The Welsh did observe and learn from the English. The Welsh adapted, a process that was probably furthered by the service of a number of Welsh noblemen with Henry I and by experience as hostages at the English court. In the course of the twelfth century the Welsh learned to conduct siege warfare and also to build their own castles. Furthermore, the traditional Welsh princely military household, the teulu, was a particularly effective weapon. These households were elite bodies of permanent warriors, loyal to their princes, and moved and fought on horseback, though they were still not as heavily armed as their opponents. They were the equivalent of English household knights, and just as effective.²³ As a consequence, the Welsh could attack and respond to the invaders on differing levels as the situation dictated. The English sources at times need to be approached with some scepticism, but the mentalité of the frontier was formed in part by ²⁰ The Letters of John of Salisbury, i, ed. S. J. Millor and H. E. Butler, rev. C. N. L. Brooke (1955), 56. The Norse settlement at Dublin was a major slave market which traded as far as the Muslim lands of the Mediterranean, and lasted well into the 12th century, so Theobald’s use of the word ‘inﬁdels’ may well have been meant literally; see Paul Holm, ‘The Slave Trade of Dublin, Ninth to Twelfth Centuries’, Peritia, 5 (1986), 317–45. ²¹ Gillingham, ‘Conquering the Barbarians’, 46. ²² See e.g. the domus defensabilis at Eardisley in western Herefordshire mentioned in Domesday; HDB 47. ²³ See Crouch, The Reign of King Stephen, 55.

52

Lords of the Central Marches

this clash between the French chivalric code and the more unrestrained nature of Celtic warfare, a clash which resulted in fear and frustration on the part of the Marchers and ensured that frontier warfare would be brutal. In response to this clash the pattern of English defence and thus military lordship had been set by 1166, and this atmosphere is the essential setting for discussion of such lordship in the March and western Herefordshire. Military stalemate resulted in a number of changes in Marcher society in the twelfth and thirteenth centuries. One answer in this period to the renewed Welsh military threat was an increasing reliance on fortiﬁcations. Welsh tactics largely dictated this response. Raiding parties relied on speed and rarely took the time to reduce strongly fortiﬁed positions; counter-attacks could then be launched against the intruders, weighed down by spoils. To man these fortiﬁcations, aristocratic society along the border was organized so as to ensure castle-guard, and these obligations were still alive in the thirteenth century. As the Welsh in turn broadened their tactics and became more adept at siege operations, wooden fortiﬁcations began to be replaced with masonry ones, burdening many of the magnates and knights of the region with an increasing load of debt.²⁴ Finally, in the thirteenth century the king came to play an increasingly important and regular role along the March, as the Marcher lords proved unable to hold their own against Llywelyn ap Iorwerth. Terrain and Welsh tactics largely robbed the English lords of the March of the effective use of heavy cavalry. The Welsh often had the initiative, while the Marchers had to react to the Welsh. It was this situation that made the castle so important in both Marcher warfare and society. The threat of ambush, particularly at night, could make it dangerous to operate for extended periods in the open. Castles provided safe bases from which to operate; this was one reason why the Middle and south-eastern March of Wales was the most heavily encastellated area of the British Isles in the Middle Ages.²⁵ Castles of course served a variety of purposes: as a means of establishing control over a locality; as a centre of administration, justice, and economic exploitation; and for the ²⁴ Beginning in the 12th century, the Welsh themselves began to use castles, many of them constructed (if poorly) of masonry; Davies, The Age of Conquest, 255–6. For the shift by the Marchers to a defensive posture and the attendant castle-building, see ibid. 280–1. ²⁵ See T. E. McNeill and M. Pringle, ‘A Map of Mottes in the British Isles’, Medieval Archaeology, 41 (1997), 220–2, esp. ﬁg. 8.

Lords and Honours

53

guarding of certain strategic points, such as river crossings and mountain passes. Builth Castle, sited up a narrow valley on a crossing of the upper Wye, is a good example of this last use.²⁶ Entire lordships could be referred to by their chief castle; witness the reference in an episcopal document of 1204–14 to ‘the lordship of the castle of Brecon’.²⁷ But fortiﬁcations along the March were originally, and remained in our period, largely military in a pure sense. As Marcher expansion slowed and then halted during the twelfth century, the Marchers dug in, so to speak, for they had found an opponent whom they ultimately could not defeat, and indeed there were times when it seemed they would not even be able to hold what they had taken. That was the backdrop for the intensive fortifying of the border counties and the Marcher lordships. We should not indulge in anachronism and picture some twelfth-century Maginot Line. The motte-and-bailey castles which are scattered over western Herefordshire were ﬁrst and foremost for shelter from Welsh raiding parties. The small size of many of these sites makes it likely that they were for the protection of the lord, his family, and perhaps his immediate followers. Tenants may have been able to shelter in the castle as well, time and space permitting. The larger castles were able to do more. Economically valuable resources were protected; for example, the borough of Brecon was originally sited within the bailey of the castle.²⁸ Livestock would have been driven into the woods if such cover were available. The protection of tenants would be stressed over that of animals, as they would be harder to replace. Even if not all of a lord’s tenants could seek shelter behind the stockade of the motte-and-bailey castle, they perhaps would at least feel reassured by its presence.²⁹ The castle was as important for psychological reasons as it was for military ones. With changes in Welsh warfare and the growth of Welsh power by the late twelfth century, castles were more important than ever. The problem was that the majority of fortiﬁcations in the March were wooden motte-and-bailey castles, which were vulnerable to incendiary attacks. In 1196, for instance, the Lord Rhys had taken and burned Radnor Castle. The Welsh were also becoming more adept at siege warfare; in 1193 Maelgwn ap Rhys and his men ‘manfully breached ²⁶ See C. J. Spurgeon, ‘Builth Castle’, Brycheiniog, 18 (1978–9), 47–59. ²⁷ Episcopal Acts and Cognate Documents relating to Welsh Dioceses 1066–1272, ed. J. Conway Davies (Cardiff, 1948–53), D.406, p. 340. ²⁸ Davies, ‘Brecon’, 49. ²⁹ See the example of Robert de Barri in Ireland, who was granted land which could be safely cultivated from his lord’s castle; Ch. 2.

54

Lords of the Central Marches

the castle of Ystrad Meurig with slings and catapults’.³⁰ This series of developments was a driving force behind the reconstruction in masonry of a number of Marcher castles. Other factors also played a role in this reconstruction of castles: imitation of developments in France as well as the prestige that masonry fortiﬁcations and buildings, erected at great cost, brought to their lord.³¹ But castles in the March, more than in England, retained their military character throughout the thirteenth century and were never solely about displaying wealth and power. In the 1210s and 1220s the latest developments in fortiﬁcation were imported from France to the Welsh March; the modiﬁcations to the Marshal castle of Chepstow, with its round towers and powerful gatehouse more or less unique to the region, constitute one of the earliest and best examples of this new building. Round masonry towers sprung up at Clifford, Bronllys, and Tretower in the same period.³² It likely was in this period that Bredwardine Castle was also refortiﬁed in stone. It should be noted that neither the Cliffords, who built Clifford and Bronllys, nor the Pichards, the knightly family who built Tretower, were particularly wealthy. The need for masonry fortiﬁcations was felt down to the level of the greater gentry. Masonry fortiﬁcations also arose over the border in Herefordshire: in the course of the ﬁrst half of the thirteenth century Longtown (or Ewyas Lacy), Wigmore, Lyonshall (the caput of the Hereford branch of the Devereux), and Weobley were all constructed or rebuilt in stone.³³ Castles require men to garrison them, and castle-guard was originally a common feature of post-Conquest England. But castle-guard, where it had existed in England, gradually declined in the course of the twelfth century, with most tenants instead paying a form of scutage to their lords for the castle ward owed.³⁴ One difference in the March was the continuing presence of functioning castleries, castellaria, where fees were organized in terms of both service owed and actual geography in order to ³⁰ Brut, 75, s.a. 1193. ³¹ For the role of the castle as an indicator of aristocratic status, see David Crouch, The Image of Aristocracy in Britain 1000–1300 (1992), 260–4, and particularly the comment on the stairs and entry of Chepstow Castle, p. 265. ³² See D. F. Renn, ‘The Round Keeps of the Brecon Region’, Archaeologia Cambrensis, 110 (1961), 129–43. ³³ For castles in Herefordshire, see Cathcart King, Castellarium Anglicanum, i. 201–17. ³⁴ Sidney Painter, following J. H. Round, ﬁxes the shift from service to commutation in the reign of Henry II; ‘Castle-Guard’, in Fred A. Cazel, Jr. (ed.), Feudalism and Liberty (Baltimore, 1961), 144–56.

Lords and Honours

55

provide the garrison of a castle efﬁciently. Castellaria had been organized early in Herefordshire and along the adjoining March. Domesday Book describes them at Wigmore, Clifford, and Ewyas Harold, while the geographic clustering of manors around the Lacy caput of Weobley also suggests the probable existence of a castlery there.³⁵ In the March the commutation of service which became widespread in England in the twelfth century was likely not often allowed, the military exigencies of the region making castle-guard still necessary.³⁶ Although castle-guard undoubtedly existed from the earliest Norman attempts at expansion into Wales, documentary evidence of arrangements for it only appears in the twelfth century. An early and unusual example dates from between 1142 and 1168, in which Robert of Ewyas, lord of Ewyas Harold (the lordship which adjoined the Lacy lordship of Ewyas Lacy), granted the Wiltshire manor of Upton Scudamore to Godfrey Escudamore and his heirs for the service of one knight’s fee at the castle of Ewyas Harold.³⁷ Although there are examples of this kind of arrangement found in south-west Wales, a Wiltshire manor does not ﬁt geographically into the compact castellaria which seem to have been typical of the March; indeed, we might very well expect that service in this case was commuted. The fact, however, that the period of service was speciﬁcally laid out, starting at Candlemas and lasting to the Invention of the Holy Cross in May, argues strongly that the service may at least have been performed by proxy.³⁸ There are further examples of the continuing vitality of castle-guard in the area. Between 1185 and 1190 William de Braose (either III or IV) conﬁrmed to Walter de Clifford the castle of Glasbury, which lies on the western bank of the river Wye in the lordship of Brecon, to the southwest of Clifford itself. The charter recording the conﬁrmation states that the castle was to be held for the service of ﬁve knights performing forty days’ service each year at the Braose caput of Brecon Castle.³⁹ Fifty years ³⁵ For Wigmore, HDB 36; Clifford, HDB 34; Ewyas Harold, HDB 23, 53. ³⁶ Holt, The Northerners, 51, notes that in the north ‘in time of war castle guard was still a vital institution’, and that the rebel castles in 1215 were largely garrisoned by tenants. ³⁷ The Hungerford Cartulary, ed. J. L. Kirby (Wiltshire Record Society, 1994), 93, no. 363. I would like to thank Dr J. R. Maddicott for drawing my attention to this charter. ³⁸ Although the clause further states that ‘failing that [service] giving half a mark yearly for the royal service’. Was Robert of Ewyas uncertain of his ability to enforce service on a far-off tenant? ³⁹ TNA, DL 27/1, pub. in Rowlands, ‘William de Braose and the Lordship of Brecon’, 133.

56

Lords of the Central Marches

later, in 1243, the scutage returns for Herefordshire showed that John Waldeboef, who held Gattertop and Upton, and Matthew de Mans, who held Humber, all manors that were in Herefordshire but part of the honour of Brecon, did not owe scutage, only castle-guard at Brecon Castle.⁴⁰ In a ﬁnal concord made a few years earlier, in the honour court of Brecon in 1240, Robert le Wafer agreed that he and his heirs would discharge (aquietabunt) the service of Hugh de Radnor and Isabel, his wife, of warda sua apud Brechon’ de secta quatuor militum.⁴¹ And across the border in Herefordshire, a grant to Craswall Priory made between 1228 and 1236 by Walter II de Lacy of demesne land in the manor of Holme Lacy reserves not only the service of two knights from that manor but also the service of the two men responsible for making ‘the summons of my knights pertaining to my castle of Weobley’.⁴² Nowhere in the documentation for the region during this period is there any mention of commutation or sums owed for castle-guard; all indications are that the lord expected this service to be performed, either by the tenant or by a substitute. Although earlier charters did not routinely mention commutation of castle-guard, by the reign of Richard I documents do begin to appear that outline scutage arrangements.⁴³ One other argument can be advanced for the probable rarity or absence of scutage in the March. Article 29 of Magna Carta sets forth the rebel barons’ belief that the king could not demand both castle-guard and service in the host from the same man. While this was an assertion resisted by the Angevin kings, it is worth noting that the Marcher lords were unaccustomed to pay scutage on the fees of their baronies in the March—castle-guard obligations there would seem to have excused them of the obligation to serve in the king’s host or pay scutage in lieu of that service.⁴⁴ Indeed, arrangements for military service were zealously kept track of and maintained by lords. This can be seen in a marriage charter ⁴⁰ BF 800. ⁴¹ TNA, DL 36/1/260. I am grateful to Professor R. R. Davies for drawing this concord to my attention. ⁴² ‘qui summoniciones militum meorum pertinentium ad castrum meum de Webbelegh facere debent’ (Bodl., MS Rawlinson B329 (Hereford Cartulary), fos 172v –173). The charter is dated by the reference to the heir of Stephen Devereux (William), who was a minor from 1228 to 1236. ⁴³ Painter, ‘Castle-Guard’, 150. The castles concerned were Rockingham, Dover, Eye, Norwich, and Richmond, with rates of commutation listed. ⁴⁴ Ibid. 149–50; Keefe, Feudal Assessments and the Political Community under Henry II and his Sons, 189. Against this interpretation can be set the fact that commutation did obtain in the baronies established in Ireland; Brendan Smith, Colonisation and Conquest in Medieval Ireland: The English in Louth, 1170–1330 (Cambridge, 1991), 35.

Lords and Honours

57

by which Walter II de Lacy (d. 1241) gave his daughter Katherine to Walter de Clifford (probably Walter III) in free marriage.⁴⁵ The manor of Mansel Lacy in Herefordshire formed Katherine de Lacy’s maritagium and carried with it the service of Roger Tirel, namely two knights’ fees. The Lacys held Mansel Lacy of the Braose lordship of Radnor.⁴⁶ The charter states that Walter de Clifford was to do to Walter de Lacy’s heirs however much service Lacy himself owed to the heirs of William de Braose.⁴⁷ No service was due for a gift in maritagium until the third generation; thus, the contract had set down the service obligations which would later apply to Mansel Lacy and outlined the chain of that service—Clifford making two fees’ service to Lacy, who in turn owed it to Braose.⁴⁸ Although castle-guard was perhaps the most onerous of services owed to the Marcher lord, and its survival into the thirteenth century and beyond is one of the more telling features of the militarized society of the Welsh Marches, there was also the duty of serving in the lord’s host. If castle-guard was still exacted, then knight service certainly was. In 1242 Thomas Corbet, lord of the Marcher lordship of Caus bordering Shropshire, stated that it was the custom in the March that whenever an army was raised all tenants holding by knight’s service were to acquit it ‘in the person of a soldier’.⁴⁹ Indeed, there is no reason other than the frequent if not yearly performance of knight service, in the ﬁeld or as castle-guard, which could explain the customary pardoning of Marchers from royal scutage.⁵⁰ Welsh tenants did not escape the necessity of military service either; in the late twelfth century Meurig ﬁtz ‘Wranno’ held 3 carucates of land outside Abergavenny for 20d . rent and the provision of one serjeant for his lord’s army.⁵¹ Beyond the meagre lists ⁴⁵ HCA, no. 3241. ⁴⁶ BF 802. By 1243 the service has interestingly been reduced to one knight’s fee. For the tenurial history of Mansel Lacy, see Ch. 2. ⁴⁷ ‘faciendum heredibus meis tale servitium quando servitium inde exire debuerit quale heredes mei facere debuerint heredi Willemi de Breusa’. ⁴⁸ Glanvill, ch. VII, sect. 18, p. 92; S. F. C. Milsom, ‘Inheritance by Women in the Twelfth and Early Thirteenth Centuries’, in M. S. Arnold et al. (eds), On the Laws and Customs of England: Essays in Honour of S. E. Thorne (Chapel Hill, NC, 1981), 61. It is interesting to note that, although the marriage contract states that Katherine de Lacy was given in liberum maritagium, the conditions of service are still attached. ⁴⁹ R. W. Eyton, Antiquities of Shropshire, 12 vols (1854–60), vii. 132. ⁵⁰ Keefe, Feudal Assessments and the Political Community under Henry II and his Sons, 189. ⁵¹ ‘inveniendi unum servientem in Exercitu’ (Feet of Fines of the Reign of Henry II and . . . Richard I, Pipe Rolls Series, 17 (1894), 110, no. 124).

58

Lords of the Central Marches

of fees compiled for Herefordshire and Shropshire in 1166, 1211, and 1242–3, and for the forfeited Braose lordships in the March in 1211, it is difﬁcult to draw a further picture of military service along the border and in the Middle and south-eastern March for the late twelfth and early thirteenth centuries. Later inquisitions post mortem from the end of the thirteenth century, however, do give some further details. The inquest held in January 1299 on the death of Humphrey de Bohun, earl of Essex and Hereford and lord of Brecon, outlines the services due from Brecon tenants at the end of the thirteenth century.⁵² A full fee was expected to provide a barded (armoured) horse, while a half-fee was to provide an unarmoured horse. Service was to be discharged at Brecon Castle, which served as the point of muster.⁵³ The period of service was the traditional forty days’ feudal service and, by this time in Brecon, was to be performed at the lord’s expense. This would not always have been the case and at some point the lords of Brecon must have offered pay as a concession to their tenants. The history of relations between lord and tenants in Brecon in the thirteenth century is largely unrecoverable, but it should be noted that lordship under the Bohuns weakened, with much power devolving onto the great knightly tenants until Earl Humphrey IV reasserted it in the 1340s.⁵⁴ The Welsh of the lordship of Brecon were also expected to do their share of service in the lord’s host; in time of war 300 Welshmen were to serve in Brecon at their own cost and outside the lordship at the lord’s. This obligation derived from a native Welsh obligation that pre-dated English lordship, and would have provided the lords of Brecon with lightly armed foot and particularly with bowmen, essential for ﬁghting in the upland areas of Wales. The Marcher lord, much like the king, built his army around his familia, or household. This household military establishment served as the nucleus around which tenants and Welsh levies would form for campaigns. These households can be observed in the border counties as well. In the 1166 Lacy carta, Hugh de Lacy listed eight men who did not hold knights’ fees of the honour of Weobley. Instead, certain of them resided with Hugh and came with him when necessary, while others were in Hugh’s dwellings in Wales (in domibus meis in Wallia), probably his castles, and likewise were to travel when necessary.⁵⁵ The ⁵² CIPM iii. 426. ⁵³ For the castle as point of array, see Davies, Lordship and Society in the March of Wales, 82. ⁵⁴ For this decay of lordship in Brecon, ibid. 92–5. ⁵⁵ RBE 283.

Lords and Honours

59

lord’s familia remained important in the thirteenth century. At some point prior to 1228 Gilbert de Lacy made a grant to the prominent Herefordshire knight Stephen Devereux.⁵⁶ Gilbert was the son of Walter II de Lacy and was often in charge of the family’s lands along the Welsh border while his father was in Ireland. He gave Stephen the signiﬁcant sum of 12 virgates in the Lacy demesne manor of Staunton-on-Wye ‘for which he [Stephen] therefore should be in my familia’. The service owed was that of one knight, and, given Stephen’s position as a signiﬁcant honorial baron of Weobley, the purpose behind the grant was likely political rather than military, meant to reafﬁrm the attachment of a major knight who had tapped other sources of patronage. Yet military service still ﬁgures in the grant, rather than the purely nominal rent of cumin or pepper which one might expect. The amount of military service corresponds roughly to the norm for Herefordshire, where fees seem to have been calculated broadly on 3- and 6-hide units. One other aspect of military lordship in the March ought to be discussed: that of cost. The price of military lordship could be high; household knights expected fees and robes, and eventually land, grants that constituted an increasing drain on a lord’s demesne resources.⁵⁷ It might be necessary to make further grants in order to resecure or reinforce the loyalty of an important tenant, as with the example of Stephen Devereux above. If land could not be awarded, something akin to the money-fee was necessary—an expedient used greatly in the royal household. This would in turn pose a burden on the lord’s rent-rolls. But at best money-fees similar to those in the royal household were a short-term solution, used until wardships or grants in land became available. It was difﬁcult for the king to reward all of his knights in the manner they and society expected, and he had a much larger fund of patronage than any Marcher lord on which to draw. Although the households of our magnates were much smaller than that of the king, they had still to be a decent size, particularly for those lords such as Lacy and Braose who had concerns on both sides of the Irish Sea. Their resources being correspondingly smaller than those of the king, the ⁵⁶ HCA, no. 3235. The terminus ante quem is determined by the death of Stephen Devereux; as the charter is a copy on a 14th-century deed roll, it omits witnesses and therefore more exact dating is not possible. ⁵⁷ See e.g. the contraction through subinfeudation of the Lacy demesne in the 12th century, Ch. 2; for the similar expectations of knights in the royal household, S. D. Church, ‘The Rewards of Royal Service in the Household of King John: A Dissenting Opinion’, EHR 110 (1995), 277–302.

60

Lords of the Central Marches

necessity of the military household would have been an expensive item for the Marcher lords. For some border and Marcher lords, such as Hugh II de Lacy and his son Walter, the increasing dissipation of demesne resources in the region only helped further to turn their interests towards their new lands in Ireland. Fortiﬁcations, however, were perhaps the major item of expense as the twelfth century ended and the thirteenth began. The nature of warfare in the March had changed by the 1190s. The Welsh had not abandoned the use of guerrilla tactics, but their ability to stand in the ﬁeld against the English had increased. This change was probably one of degree more than anything else. Despite Gerald of Wales’s picture of the Welsh as lightly armed ﬁghters, attacking suddenly and quick to retreat, the defeat of a large English force in Gower in January 1136 argues for greater Welsh ability to match up in the ﬁeld. The Welsh Brut, which probably exaggerates, describes the Welsh host of that year as having not only 6,000 foot but also 2,000 mailed horsemen.⁵⁸ Later in the century, the defeat of the combined forces of Roger de Mortimer and Hugh de Say at Radnor in 1196 by the Lord Rhys undoubtedly sent shock waves through Marcher society, especially in our area.⁵⁹ With the growing power and effectiveness of the Welsh, the most expensive item of military lordship became the building of masonry fortiﬁcations. The fact that account rolls for lay estates do not appear until the mid- to late thirteenth century makes it very difﬁcult to state just how much the upgrading of castles along the March cost. It is surely no coincidence, however, that debt owed to Jewish moneylenders rises precipitously at exactly the time that major construction in stone was occurring, and for those families who were carrying out that construction. Indeed, in Herefordshire the chief debtors were mostly larger landholders rather than small to middling ones, the opposite of the pattern elsewhere in England.⁶⁰ When Walter II de Lacy died in 1241, he owed £666 in Jewish debt, while in 1233 Walter III de Clifford was pardoned a debt of £666 13s. 8d. by the king. Although it is probable that some of the debt is attributable to other causes—in the case of Lacy, for example, the cost of building at Trim Castle in Ireland or ﬁnes owed the king for his benevolence—the early thirteenth ⁵⁸ Brut, 51, s.a. 1136. ⁵⁹ Brut, 76, s.a. 1196. ⁶⁰ Joe Hillaby, ‘Hereford Gold: Irish, Welsh and English Land. The Jewish Community at Hereford and its Clients, 1179–1253. Part I’, Trans. Woolhope, 44/3 (1984), 403.

Lords and Honours

61

century also saw construction in the new masonry style at the Lacy castle of Longtown (Ewyas Lacy) and the Clifford castles of Clifford and Bronllys. This would seem to indicate that the large levels of debt to be found in the region were to a considerable degree the result of military competition along the border, a race which caused the Marcher lords to upgrade signiﬁcantly the quality of their fortiﬁcations. Stronger, more impressive castles were also a useful tool in the intensiﬁcation of lordship where a lord’s control had been contested or nominal. There were other factors which led to the accumulation of debt, of course. The need for display to maintain one’s status, to which the castle contributed, and ﬁnes, either for royal grants and privileges or to purchase the king’s benevolence, could also be major sources of debt.⁶¹ But it seems indisputable that the costs of masonry fortiﬁcations contributed prominently to aristocratic debt in Herefordshire and the adjacent March by the 1240s.⁶² One thing this indebtedness indicates is the limited territorial basis of lordship in the region: the lord’s own demesne resources, combined with rents, were rarely sufﬁcient to foot the bills. Finally, the same problems arising from feudal military service elsewhere in England—superannuated knights, disputes over the amount of service owed, or refusal to perform service—would largely have applied in the Welsh borders, especially the border counties such as Herefordshire. Indeed, this can be seen for the Herefordshire honour of Weobley in the 1166 carta of Hugh II de Lacy.⁶³ Hugh reported that he had problems extracting service from two of his knights, Elias de Say and William de Furches. Elias recognized the service of three knights, but he denied that owed from a further two fees, and Hugh stated that he was not able to have that service until it should be proved that Elias owed it. Likewise, William de Furches admitted the service of two knights, but a further fee was in dispute (calumpnia), and Hugh had never been able to extract service from it. Furthermore, Philip de Sarnesﬁeld admitted the service of a half-fee but disputed that of another half. Such problems were ideally to be resolved in the lord’s court, the ⁶¹ Walter II de Lacy was ﬁned £1, 200, an enormous sum, ‘for having the benevolence of King Richard and seisin of his land’ in 1199; Pipe Rolls, 1 John, 216–17. He incurred further ﬁnes for the conduct of his men of Meath during the invasion of Ireland by his brother Hugh, quondam earl of Ulster, being ﬁned 3,000 marks for his lands and those of his knights and free men; Rot. Litt. Claus. ii. 39b. For the large proffers by which William de Braose was indebted to King John, see Ch. 6. ⁶² As it did for some knightly families; see Ch. 4. ⁶³ RBE 281–3.

62

Lords of the Central Marches

honour court (curia honoris); but by 1166 tenants in Herefordshire had another option, namely the royal courts. In the March, where appeals to royal justices were not allowed, the exaction of service would have been easier, although it should be remembered that the power of custom and the necessity to get on with one’s tenants could be powerful brakes on the authority of the Marcher lord. We shall turn to the jurisdictional aspect of lordship in the next section. From roughly 1150 to 1250 military lordship was marked by its own process of intensiﬁcation and consolidation. The military service of tenants was scrupulously recorded and deﬁned in increasingly legalistic terms and forms. Lords sought or borrowed the ﬁnancial means to increase their military power in the face of the resurgent Welsh. Finally, in the early thirteenth century the erection at great cost of castles in stone, employing the latest advances in military architecture, planted the military basis of all lordship along the border and in the March more ﬁrmly than before. This consolidation of military lordship both was a pre-condition for the consolidation of other aspects of lordship, and went hand in hand with it. J U D I C I A L LO R D S H I P The problem of enforcing service, discussed above in the case of the 1166 Lacy return for Weobley, leads to a consideration of another facet of lordship in the region, judicial lordship. It was through the honour court that service was ultimately enforced, with the sanction of distraint as the ﬁnal guarantor. Furthermore, judicial lordship was a key tool in the consolidation of lordship, enabling the lord to increase demands upon mesne tenants or even to increase his territorial lordship outright. For a lord to use this tool, however, it was crucial for him to assert the primacy and omnipotence of his court. Our period is above all that in which Marcher lords attempted to shut off outside interference in their courts. A few initial observations must be made concerning judicial lordship in the region. The fact that court rolls did not begin to be kept until the second half of the thirteenth century means that there is an enormous lack of evidence for the operation of judicial lordship along the March. It is possible to a degree to argue back into the thirteenth century from the much fuller sources of the fourteenth century, but there are inherent dangers in such an approach. For example, sessions in eyre became a

Lords and Honours

63

frequent feature of justice in the Marcher lordships in the fourteenth century.⁶⁴ There is no evidence for such sessions in the early thirteenth century, however; it could easily be argued that such instruments of justice were dependent on a larger seigneurial administration, and a corresponding degree of specialization, than many if not most of the Marcher lords possessed in the twelfth and early thirteenth centuries.⁶⁵ Alternatively, the need for such judicial mechanisms may have been brought about by the increase in forms of action and general legal complexity in the thirteenth and fourteenth centuries. The sessions in eyre were also a conscious imitation of royal practice and followed on the adoption of royal forms of judicial action in the March, a process which began much earlier, no later than the beginning of the thirteenth century. Thus, for the thirteenth century in our region, there would have been a mix of old and new as writs selfconsciously styled on royal ones were used to initiate actions in the lord’s honour court. Judicially, the distinction between England and Wales was hardening in the course of the twelfth century, as royal lawyers increasingly deﬁned central power, and as Marcher lords set forth their own legal claims in reaction. Herefordshire was within the ambit of royal justice, and justices in eyre included the county in their circuits. In the Marcher lordships, however, the lord of the honour was the fount of justice, and the ultimate appellate authority. This did not yet mean that the Marcher lord had untrammelled authority within the boundaries of his lordship. It did, however, mean that there were at times certain fundamental differences in the execution of justice in Brecon as opposed to Herefordshire. Sometimes the jurisdictions conﬂicted. This can be seen best from certain cases that entered the curia regis which involved manors of the honour of Brecon which physically lay in Herefordshire. One such case was brought in 1200 when William de Furches brought an action of mort d’ancestor against William de Braose, and claimed that Braose had deforced him of the half-fee in Kingstone that his uncle ⁶⁴ Davies, Lordship and Society in the March of Wales, 167. For a discussion of the nature of judicial lordship in the 14th and early 15th centuries, see ibid., ch. 7. ⁶⁵ Davies notes that the personnel of such tourns were drawn from the members of the lord’s council, giving as an example the sessions in Kidwelly in 1380 which were held by the chief steward, the receiver-general, and the auditor of John of Gaunt; ibid. 168 and n. 71. Specialization was, however, occurring in the course of the 13th century—the description in 1234 of Philip de Coleville as Walter de Lacy’s ‘steward in England’ implies the existence of dual administrative organizations on both sides of the Irish Sea; CPR 1234–47, 56.

64

Lords of the Central Marches

Robert de Furches had held.⁶⁶ Kingstone was just to the east of the Golden Valley and by the standards of the time would almost certainly have been considered to be in Herefordshire.⁶⁷ Yet it was held of the honour of Brecon. There was certainly room for a conﬂict of jurisdiction here between king and Marcher lord, a conﬂict of which William de Furches was undoubtedly trying to take advantage. He could angle for the venue that was most likely to give him a favourable decision, and this was precisely the situation for which the new Angevin assizes had been developed. In this case de Furches attempted to use the royal courts to escape his lord’s discretionary power, that arbitrary power which the lord used to discipline his tenants. That power probably had been used to bar William de Furches of entry into land he regarded as his by right. William de Braose in turn was attempting to exercise his powers as a Marcher lord in Herefordshire. This case clearly demonstrates that the conﬂicting jurisdictions, no matter how similar they were in many respects, could be opposed to each other. As no result was ever recorded for the suit, it may be assumed that William de Braose either won this jurisdictional battle or quietly settled out of court. There was room for jurisdictional conﬂict because there was at the end of the twelfth century still no exact demarcation of the judicial spheres of king and lord in the region. Another case involving the honour of Brecon that came before the curia regis in 1199 demonstrates this nicely. Nesta de Baskerville, the widow of Ralph II de Baskerville (d. 1190/1) of Bredwardine, claimed Bredwardine Castle with its appurtenances as hers by right and inheritance (jus et hereditatem suam).⁶⁸ She further claimed that another Brecon tenant, Robert le Wafer, was deforcing her of the castle. The king ordered the sheriff of Herefordshire, Walter II de Clifford, to take the castle into the king’s hand, but Clifford replied that he dared not enter the castle as it was outside his bailiwick, and furthermore, the task was not within his power. William de Braose proceeded to state in ringing terms that ‘neither the king nor the justiciar nor the sheriff ought to set their hand within his liberty (in libertate)’. ⁶⁶ Rot. Ob. et Fin. 77–8. ⁶⁷ Kingstone is listed in Domesday Book as being in Herefordshire; it was originally granted by William ﬁtz Osbern to Cormeilles Abbey in Normandy. How it eventually came to form part of the honour of Brecon is a matter for conjecture, but the scutage returns for 1242–3 list it as part of the honour of Brecon; BF 812. ⁶⁸ Rot. Cur. Reg. i. 426. The case dates from early in 1199, for William de Braose himself became sheriff of Herefordshire later that year, an event that renders the wording of the case nonsensical.

Lords and Honours

65

Clifford’s statement concerning Bredwardine may have been somewhat disingenuous, and there is the possibility that the claim itself was ﬁctitious and meant to clarify title. But it may also illustrate that there was ambiguity over how to approach Bredwardine, which undoubtedly was geographically and administratively in Herefordshire.⁶⁹ In terms of jurisdiction, was it legally part of England, or did its position within the honour of Brecon make it part of the March? There are further aspects to this suit which illustrate the complex interaction between royal and seigneurial justice along the border. The sheriff of Herefordshire, Walter de Clifford, was himself the greatest honorial baron of the lordship of Brecon, holding four fees of the honour.⁷⁰ And William de Braose was one of the prime movers behind John’s securing of the throne later in 1199, a possible factor in the disappearance of this case from the record. Some things about the suit are difﬁcult to explain, for example why Nesta entered a plea of right rather than of dower, and why Robert le Wafer, the father of the future husband of Alice de Baskerville, daughter of Ralph III de Baskerville (d. 1210) and a granddaughter of Nesta, was deforcing her of the castle in the ﬁrst place.⁷¹ The question of jurisdiction was an important one for Marcher lord and tenant. One legal historian has written that ‘had lordship regularly carried most of government with it, jurisdiction would have been defended as property against centralisation’.⁷² This was precisely what happened in the March; conﬂicts of jurisdiction between Marcher lord and king tell us much about the nature of Marcher lordship. But just because Marcher lords sought to exclude royal interference did not mean that legal procedure in the March was vastly different from that in England; rather, as lords consolidated their power in the March, their legal procedures came more and more to resemble those of the king. While it must be kept in mind that the law (or laws) of the March were an amalgam of Welsh and English custom, English writs and particularly the verdict by jury were immediately popular in ⁶⁹ Bredwardine in 1086; HDB 59. ⁷⁰ RBE 601, the 1211 return, where the Brecon fees are held by Walter de Clifford junior (Walter III), presumably by demise. ⁷¹ Copleston-Crow, ‘The Baskervilles of Herefordshire’, 34 n. 62, speculates that Robert le Wafer may have been Ralph de Baskerville’s castellan at Bredwardine. Nest’s claim to possess Bredwardine Castle by hereditary right rather than dower was almost certainly designed to avoid the case ﬁrst being heard in the heir’s court, as according to Glanvill in the case of dower the writ of right must ﬁrst be directed to the heir. ⁷² S. F. C. Milsom, Historical Foundations of the Common Law, 2nd edn (1981), 20.

66

Lords of the Central Marches

the March and seem quickly to have been adopted by Marcher lords and their English vassals and tenants. One characteristic of Marcher law was that it was unwritten, a customary body of law, ‘a droit coutumier in the fullest sense’.⁷³ But this body of law adopted English practices as the problems experienced by the Marcher lords and their tenants in their courts were similar to those in England.⁷⁴ One means of this adoption was probably the experience of Marcher lords themselves, almost all of whom held land in England, and of their stewards. Another means of transmission of English practices occurred when Marcher lordships, as they at times inevitably did, came under royal wardship owing to forfeiture or the minority of an heir. Thus, it is certainly not coincidental that the ﬁrst appearance in Glamorgan of ﬁnal concords and the assize of novel disseisin occur at the beginning of the thirteenth century, when that lordship was in the custody of King John.⁷⁵ It also seems that the norms of English justice were extended to Radnor, Brecon, and Abergavenny when those lordships were in royal custody from 1208 to 1215, following the downfall of William III de Braose.⁷⁶ While it will be asserted that the extension of royal justice proved unpopular and was perhaps one reason for the support which the exiled Braoses received in 1215, it was very likely the severity with which a rapacious regime exploited that justice rather than the forms it took which proved an irritant. Thus, by 1240 we have another development of Angevin justice, a ﬁnal concord, transacted in the court of the earl of Hereford at Brecon—ﬁnalis concordia facta in curia domini comitis Hereford’ . . . apud Brechoniam—in other words the honour court of Brecon.⁷⁷ The document was drawn up in the presence of the ‘faithful men of the earl’ (aliis ﬁdelibus domini comitis) who take the place at the beginning of the concord which would otherwise be occupied by the royal justices in eyre. The great difference in the Marcher lordships was that Marcher justice was ‘truly feudal’; more than any place else, Ireland included, it approximated the ‘feudal state in miniature’, that Platonic ideal famously described by Sir Frank Stenton.⁷⁸ Although English forms might be ⁷³ R. R. Davies, ‘The Law of the March’, WHR 5 (1970–1), 6–7. ⁷⁴ ‘exemption from the judicial machinery of the kings of England was one thing; the desire or the ability to be immune from the ideas and practices of English common law quite another’ (ibid. 19). ⁷⁵ Ibid. 23. ⁷⁶ See Ch. 6. ⁷⁷ TNA, DL 36/1, no. 260. ⁷⁸ Davies, ‘The Law of the March’, 12–13; Stenton, The First Century of English Feudalism, 51. Stenton’s view of the ‘feudal state in miniature’ formed the basis for the

Lords and Honours

67

used, the Marcher lord was still the fount of justice in his lordship. He was, of course, constrained by custom and the ﬁndings of the members of his feudal curia, his knights.⁷⁹ Indeed, in the medieval period the respect for custom was usually so great as to give it the force of law, and on most occasions a lord would be well advised not to go against it. But without the option to remove cases to the royal courts, the Marcher lord had a crucial amount of discretion which his purely English counterparts certainly no longer possessed after 1176. The Marcher lord, in other words, had more room for manoeuvre when disciplining his tenants, and the weapon of that discipline was distraint.⁸⁰ It was this discretion which could be a potent instrument in the consolidation of lordship. We possess no clear examples of this process from our region (although the de Furches episode cited above is very likely one), but the neighbouring lordship of Glamorgan provides a famous example of honorial justice in the March, and the advantages it could provide to the Marcher lord. This was the famous Siward case of 1244–5.⁸¹ Richard Siward was a military tenant of Richard de Clare, earl of Gloucester and lord of Glamorgan. When he broke a truce which the earl had arranged between the Welsh chieftain Hywel ap Maredudd and another Glamorgan tenant, Gilbert de Turberville, Siward was summoned to Cardiff, seat of the curia comitatus of Glamorgan, in order to re-establish work of the later legal historians Thorne and Milsom, but has since been challenged and modiﬁed by, among others, John Hudson, Land, Law, and Lordship in Anglo-Norman England (Oxford, 1994), and Dalton, Conquest, Anarchy and Lordship, ch. 7. This new generation of scholars has challenged the view that only with the advent of the legal reforms of Henry II did the feudal honour begin to be undermined. They do this by pointing to early multiple ties of lordship and frequent royal intervention in the affairs of feudal courts. Both Dalton and especially Hudson are careful to exclude the March from their ﬁndings. As a counterpoint to their work may be taken other recent work which demonstrates the continued vitality of the English feudal honour; John Hunt, Lordship and Landscape: A Study of the Honour of Dudley 1066–1322, BAR British Series, 264 (1997), and P. E. Golob, ‘The Ferrers Earls of Derby: A Study of the Honour of Tutbury (1066–1279)’, Ph.D. thesis, University of Cambridge (1984). I owe these last two references to Dr David Carpenter, who has ably summed up the arguments of those who see continued vitality in the honour in his ‘The Second Century of English Feudalism’. For a re-examination of the ‘feudal nature of Ireland’, see Brock Holden, ‘ ‘‘Feudal Frontiers?’’ Colonial Societies in Wales and Ireland 1170–1330’, Studia Hibernica, 33 (2004–5), 61–79. ⁷⁹ Davies, ‘The Law of the March’, 10. ⁸⁰ A fruitful comparison can be made with the position of the Angevin kings themselves before Magna Carta; see J. E. A. Jolliffe, Angevin Kingship (Oxford, 1955), particularly chs. 3 and 4; Holt, Magna Carta, ch. 5. ⁸¹ For what follows, see David Crouch, ‘The Last Adventure of Richard Siward’, Morgannwg, 35 (1991), 7–30; Altschul, A Baronial Family in Medieval England, 69–74.

68

Lords of the Central Marches

the truce. Siward refused to cooperate with the proceedings and allied himself with Hywel. The two men were with some trouble defeated. Richard de Clare then brought proceedings in his court against Siward for felonious breach of the peace. Siward was offered the opportunity to wage trial by battle with a member of the earl’s household. He refused, on the grounds that the earl’s knight did not hold any land in the lordship of Glamorgan and was thus not one of Siward’s peers. Upon adjournment of the court Siward could not ﬁnd guarantors and was obliged to hand over his castle of Tal-y-fan and all his lands as pledge for his future appearance. When, according to the earl’s account of the affair, Siward failed to appear any of the allowed three times at court, his lands were seized. Siward then appealed to the royal courts, claiming default of justice. Richard de Clare not only rejected Siward’s version of events but also advanced a more general defence of Marcher liberties. Since the ﬁghting between Hywel, Siward, and the earl had occurred entirely within the boundaries of Glamorgan, only the earl had cognizance of the suit. The earl stated that ‘such was his liberty in his land of Glamorgan that none of his men or those of any other ought to plead any suit elsewhere than within his liberty’, although he offered to let a royal commission determine whether justice had been done according to the ‘customs of the region’ (secundem consuetudinem patrie).⁸² The judgment of the royal court does not survive, but it is evident that Richard de Clare was the victor in this legal contest. This case shows the proprieties—the distinction between landed tenant and member of the lord’s familia, the concept of ‘peerage’—which were important within the curia comitatus of Glamorgan, and so would also be within other honour courts of the region. The image summoned up is one of an older feudal world, one which very much still existed in the March. The earl of Gloucester had not only strenuously defended the independence of Marcher jurisdiction, but more practically had also taken advantage of Richard Siward’s fractiousness to bolster the Clare demesne lands in Glamorgan through distraint and forfeiture. Judicial instruments had been used to consolidate Clare control of their Marcher lordship, a control which had been weakened earlier in the ⁸² ‘talem libertatem in terra sua de Glamorgan quod nullus de hominibus suis nec aliis debent de aliquo placitare alibi quam infra libertatem suam’ (T. B. Pugh (ed.), Glamorgan County History, iii (1971), 51).

Lords and Honours

69

thirteenth century.⁸³ Through his court, de Clare had established his dominance over a major honorial baron, making an example to his other tenants. Indeed, one historian has argued that the earl manipulated the curia comitatus of Glamorgan in order to achieve his ends.⁸⁴ A Marcher lord would not always be able to use his judicial authority in such a manner. Although the lord himself was virtually a regal power in his lordship, much depended on his own personal qualities, how much he applied himself to the exercise of lordship, whether he could either get the suitors of the feudal curia to go along with him or overawe them into agreement. The importance of the personal exercise of lordship is one reason why men such as Walter II de Lacy and William III de Braose demised their lands in the region onto their sons when they largely became involved elsewhere, for absentee lordship was to be avoided. When the Marcher lord did not apply himself to the running of his lordship or was absent, judicial as well as other power would devolve to the honorial barons; this is what seems to have occurred in Brecon in the course of the thirteenth century, when a series of absentee lords held the lordship in custody or wardship. But if a lord was energetic, he would take up the reins of control once again, and the feudal court was the instrument of that reassertion, as was the case with Richard de Clare and Glamorgan.⁸⁵ Judicial lordship, as with other aspects of lordship in the March, was personal lordship. The nature of judicial lordship in the March may usefully be contrasted with that in Herefordshire; one need only recall Hugh II de Lacy’s complaints in his carta of 1166 that he was unable to extract knight service from some of his tenants.⁸⁶ This service had most likely been lost during the reign of Stephen, when the Lacys were not in ⁸³ For an overview of the consolidation of Clare lordship in Glamorgan in the 13th century, see Davies, Lordship and Society in the March of Wales, 86–91; J. B. Smith, ‘The Lordship of Glamorgan’, Morgannwg, 2 (1958), 9–38; Altschul, A Baronial Family in Medieval England, 241–80. ⁸⁴ Crouch, ‘The Last Adventure of Richard Siward’, 23. ⁸⁵ As it was with Brecon in the 14th century, where the lords of Bronllys and Pencelli had achieved virtual independence from the lords of Brecon. In the 1340s Humphrey IV de Bohun, earl of Hereford and lord of Brecon, used (perhaps somewhat dubious) quo warranto inquiries in the court of Brecon to bring to heel the lords of Pencelli, Bronllys, and Cantref Selyf; J. B. Smith, ‘Marcher Regality: Quo Warranto Proceedings relating to Cantrefselyf in the Lordship of Brecon, 1349’, BBCS 28 (1978–80), 267–88. As the lord of Cantref Selyf did not have his own chancery, it is quite likely that the powers he claimed had at some point devolved upon him from the lord of Brecon, as the earl’s attorney pointed out. ⁸⁶ See above, p. 61.

70

Lords of the Central Marches

control of the honour of Weobley, but the new assizes and easy access to royal courts under Henry II made it difﬁcult for the Lacys to punish recalcitrant vassals with the same vigour as a Marcher lord. Lordly discretion and seigneurial room for manoeuvre declined precipitously once the hardening legal boundary between the March and Herefordshire was crossed. Of course, the feudal curia in the March was about more than the enforcement of the obligations of a Marcher lord’s vassals. The curia was omnicompetent. The suitors to the court would be expected to provide their lord with counsel. Routine transactions between tenants were conducted there, charters and ﬁnal concords witnessed and enrolled. Examples from Brecon are numerous. Although express mention of the feudal curia of Brecon is not made in the charters preserved in the Brecon Cartulary, witness lists often show that a transaction undoubtedly occurred at such an occasion. One such example is a grant by Walter Devereux the younger of 12 acres of land to Brecon Priory.⁸⁷ The grant is described as having been made at the instigation of Walter’s lord (domini mei), William de Braose the younger, and William heads the witness list. He is followed by Roger de Baskerville, the son of Thomas de Baskerville, who held a portion of Bredwardine, Philip de Braose, a son of William the younger, and members of other important Brecon families: Miles de Mucegros, John and Miles Pichard, William de Waldeboef, William de Burghill, and Robert de Furches. What we certainly have here is a session of the honour court of Brecon witnessing a grant by one of its suitors to the house most closely connected to the lordship, Brecon Priory. Disputes and transactions (it is often difﬁcult to tell which from early ﬁnal concords) were also resolved or conducted in the curia. The earliest extant example of a ﬁnal concord from Brecon dates from 17 December 1240, after the lordship had passed to the Bohuns through marriage.⁸⁸ In this document, Robert le Wafer quitclaimed all of his issues from his portion of Bredwardine with appurtenances to Hugh of Radnor and his wife, Isabel. Hugh and Isabel gave in exchange (escambium) their manors in Pencelli, Llandefaelog, and Trosdref. Robert le Wafer gave Hugh and Isabel 30 silver marks and agreed to perform for the couple castle-guard ⁸⁷ ‘Cart. Brecon’, 14. 33. The editor has erroneously dated this charter to the lordship of William V de Braose, when the William juvenis who witnessed it is almost certainly William IV de Braose, dating the charter to before 1208; see Ch. 6. ⁸⁸ TNA, DL 36/1, no. 260. I owe the transcript of this ﬁnal concord to Professor R. R. Davies.

Lords and Honours

71

at Brecon Castle and the suit of four knights.⁸⁹ This was no doubt the service owed to the lord of Brecon for the manors which Robert had gained in the exchange; again we see the service owed by tenants being scrupulously kept track of. Lastly, it should be noted that the transaction took place in curia domini comitis Hereford’ (Humphrey V de Bohun, who in 1240 would seem to be acting in place of his son Humphrey, who was married to Eleanor de Braose and was not allowed to enter into her inheritance of Brecon until the following year) in the presence of Miles Pichard, who was then constable of Brecon, and that the initial ‘dispute’ which culminated in the ﬁnal concord was started by the lord’s writ, per brevem domini comitis. The concord otherwise resembles one which would have been concluded in any royal court in England; the only difference lay in who was the ultimate judicial guarantor of the agreement. Though court records of our honours do not survive for this period, there is every reason to believe that the pattern of courts and jurisdictions which one sees in the mid- and late thirteenth century and after in the region was already established in essence, at least in the Anglicized lowlands.⁹⁰ Judicial lordship was also proﬁtable. For example, whoever obtained the writ which led to the ﬁnal concord discussed above would have had to pay a fee to the lord of Brecon. Our earliest recorded proﬁts for Marcher courts in the region come from those of the lordships of Monmouth and Abergavenny when they were in the hands of the Lord Edward in 1256–7.⁹¹ In Monmouth pleas and perquisites of the borough of Monmouth were £6 11s. 9d., while the pleas and perquisites of the ‘Great Court’ of Monmouth were a substantial £41 18s.⁹² In the Marcher lordships, these ‘great courts’, usually biannual, were largely occasions for the imposition of heavy ﬁnes and amercements, and were concerned with criminal justice in particular.⁹³ The gross revenue for the manor of Monmouth was £388 7s. 3 3/4d.; thus issues of the court ⁸⁹ Ibid. ‘ . . . Robertus et heredes sui aquietabunt dictos Hugonem et Isabellam uxorem suam et heredes suos de warda sua apud Brechon’ de secta quatuor militum’. ⁹⁰ Davies, Lordship and Society in the March of Wales, 154–7. ⁹¹ ‘Ministers’ Accounts for the Lordships of Abergavenny, Grosmont, Skenfrith and White Castle, 1256–7’, ed. A. J. Roderick and W. Rees, South Wales and Monmouth Record Society Manuscripts, 2–4 (Cardiff, 1950–7). Abergavenny is in 2. 68–125; Grosmont and Whitecastle in 3. 22–47; and Monmouth in 4. 6–29. ⁹² The ‘Great Court’ of Monmouth may be the same as the honour court, or curia domini de Munemuya, mentioned in 1228; HCA, no. 272. ⁹³ A. J. Otway-Ruthven, ‘The Constitutional Position of the Great Lordships of South Wales’, TRHS, 5th ser., 8 (1958), 7.

72

Lords of the Central Marches

formed 12 per cent of the income for that year.⁹⁴ This indicates that the judicial structure in Monmouth was well established by the midthirteenth century. In the Anglicized lowlands of the Marcher lordships there was a pattern of manorial and borough courts much like those to be found in England, and no doubt with the same range of jurisdiction. For instance, in Monmouth manorial courts in the outlying members of the borough and manor of Monmouth, such as Dingestow, St Maughans, and Hewelsﬁeld in the Forest of Dean, also brought in revenue, albeit in much smaller amounts. In the lordship of Abergavenny the Great Court of Abergavenny brought in £113 12s. 2d., out of a gross sum of £315 5s. 1 1/4 d ., or an impressive 35 per cent of that year’s gross income.⁹⁵ Although the ofﬁcials appointed to run the Lord Edward’s Marcher lands were noted for their rapacity, records from the fourteenth century show that judicial income at that time formed a signiﬁcant proportion of overall income, and there is no reason to assume that the situation was much different in the ﬁrst half of the thirteenth century.⁹⁶

E C O N O M I C LO R D S H I P Judicial revenue leads us to a wider consideration of the management and proﬁtability of lordship in the region for our period. Demesne lordship was particularly important for, as one historian has written, ‘the broader that territorial basis [of lordship] the more secure was the authority of lordship’.⁹⁷ Land that was kept in the lord’s hands was more directly under his control, and could be exploited for proﬁt more efﬁciently. Yet in the twelfth and thirteenth centuries territorial consolidation of lordship had to be balanced against the need to provide for military tenants. In an examination of economic lordship, focusing on demesne lands, we need to ask two questions: did our lords attempt to develop and more efﬁciently exploit the lands they held, and did they increase their demesne holdings? I shall concentrate more on economic lordship in the March as opposed to Herefordshire, for Herefordshire ⁹⁴ ‘Ministers’ Accounts for the Lordships of Abergavenny, Grosmont, Skenfrith and White Castle’, 9, 13. The presence of arrears in the gross income for Monmouth for 1256–7 has inﬂated that sum. ⁹⁵ Ibid. 105. ⁹⁶ For Edward’s ofﬁcials, see Michael Prestwich, Edward I, new edn (New Haven, 1997), 20; for the importance of judicial revenue in the 14th century, Davies, Lordship and Society in the March of Wales, 179–82; id., The Revolt of Owain Glyn Dwr (Oxford, 1995), 71. ⁹⁷ Davies, Lordship and Society in the March of Wales, 86.

Lords and Honours

73

approximated to the rest of England in this period, and furthermore, the relationship between economic lordship and military and judicial lordship was perhaps closer and more important for the Marcher lords. Wales was recognized as a potentially lucrative land at a very early point by its Norman English conquerors; indeed, if that were not the case we should wonder why the English bothered, at great expense, to persevere with their attempts to expand into Wales. In the 1140s the author of the Gesta Stephani said of Wales, perhaps with some poetic licence, after its English conquerors had subdued the native Welsh, that it was ‘so productive and abounding in all kinds of resources that you would have reckoned it in no wise inferior to the most fertile part of Britain’.⁹⁸ This comment is of course one typical of conquerors and colonizers throughout history, who like to see themselves as harnessing and developing virgin territory; similar comments were indeed made of medieval Ireland as well. But to accept this at face value would be mistaken. Wales at the coming of the English was certainly an underdeveloped land in comparison to southern England, but it was not undeveloped. Evidence from the late eleventh century shows that Wales was already a land of mixed agriculture, with the cultivation of grains, predominantly oats, as important as stock-rearing. The importance of the plough and the mill was recognized throughout native Wales. Open ﬁelds, organized in long, quilleted strips, had appeared by the thirteenth century in Gwynedd.⁹⁹ The economy of native Wales was neither wholly pastoral nor unchanging. What the English lords of the March did was to intensify arable agriculture in the river valleys and to organize it along lines that maximized its value to the lord, especially in cash terms. The fact that they came from a world in which plentiful coinage played a key role directly inﬂuenced the way they developed and organized their estates; by contrast, in native Wales the commutation of food renders into cash was very slow in coming. The end result can be seen in the fourteenth century, when the potential upon which the author of the Gesta Stephani commented in the 1140s was certainly being exploited fully. The revenue of Abergavenny had more than doubled to £667 in 1348 (it had been £315 in 1257), while Builth was valued at £200 in 1398, Monmouth £514 in 1331, Radnor (and Gwerthrynion) £314 in 1339, and Brecon with Hay and Huntington £1, 543 in 1372.¹⁰⁰ These ⁹⁸ Gesta Stephani, 15. ⁹⁹ Davies, The Age of Conquest, 153–7. ¹⁰⁰ Davies, Lordship and Society in the March of Wales, 196.

74

Lords of the Central Marches

ﬁgures, coming from much later than the period under study here, are obviously of little use in gauging early thirteenth-century values. Wales in the fourteenth century was a conquered land and peace allowed a higher level of economic development in the Marcher lordships. What the fourteenth-century ﬁgures do demonstrate, though, is the potential of these lordships, which would have been recognized and as far as possible pursued by their twelfth- and thirteenth-century lords. They also demonstrate the disparity, largely owing to geography, between the economic value of our lordships. If we assume a fairly equal level of development on the part of their fourteenth-century lords, the upland areas of Builth and Radnor were obviously less valuable than Brecon with its lowland river valleys of the Wye, Llynﬁ, and Usk. By the end of the twelfth century these lowlands were already well developed, with a full-blown manorial structure. Much as the lords of the March strived to consolidate military and judicial lordship, they worked to increase the proﬁtability of their lands in the twelfth and thirteenth centuries. Indeed, the consolidation of lordship, especially military lordship, was an expensive process, and acted as a further spur to increased economic development and exploitation on the part of lords. The account rolls for 1256/7 from Abergavenny and Monmouth discussed above are eloquent testimony to this. Indeed, the fully developed structure of manor courts in itself shows the economic development of the region. Evidence from Ireland also shows how quickly English lords planted manors and how thoroughly they exploited the countryside of a conquered region. During the year 1211–12, when it was in the hands of the king following the exile of its lord, Walter II de Lacy, the Irish honour of Meath accounted for £239 13s. 8d. These issues came from twelve manors, several parcels of land identiﬁed by their former holders that may or may not have been manors, and burgage rents in Kells.¹⁰¹ It should be noted that, as some individual accounts under Meath were for only half a year, the actual annual income would have been higher. Walter de Lacy and his father, Hugh, before him had lost no time in exploiting their Irish lordship, and it would be foolish to expect any less of them and their fellow lords along the March of Wales. The contrast between upland and lowland, present to some degree in all the lordships of the March, meant that there were roughly two ¹⁰¹ ‘The Irish Pipe Roll of 14 John, 1211–1212’, ed. O. Davies and D. B. Quinn, Ulster Journal of Archaeology, 3rd ser., 4 (1941), suppl., 1–76; for Meath, see pp. 21–3.

Lords and Honours

75

economies which their lords exploited, upland pastoral and lowland arable, to draw a rough-and-ready distinction. Two forms of lordship accompanied these two economies: the upland areas were largely Welsh and thus paid tribute, although some lords had perhaps begun to exploit these areas themselves, while the lowlands, with an inﬂux of English settlement, approximated more closely to the norms of agriculture in Herefordshire, and were either more closely controlled by the lord or had been granted out as mesne holdings. The upland economy was not without value. Although stock-rearing did not really develop on a large scale in the March until the fourteenth century, there are early indications in the writs of John’s reign of not insigniﬁcant cattle-ranching in the region, a situation which must have existed for some time. In 1205 William de Braose was ordered by the king to deliver twenty cows and two bulls to the monks of Beaulieu Abbey; in the same year William de Cantilupe, who was sheriff of Hereford, was commanded to give to the prior and canons of Bradenstoke Priory forty cows and two bulls which William de Braose had given to the king.¹⁰² Whether these were from Braose’s own herds or came from Welsh tributes is uncertain, although the evidence from Ireland records large cattle-tributes rendered by the native Irish, as does the later Welsh evidence.¹⁰³ Later records from the lordship of Abergavenny show that, in accordance with Welsh custom, cattle were received by the lord for a variety of offences ranging from theft to the shedding of blood (pro sanguine), and these were probably used to stock and replenish his own herds.¹⁰⁴ Although the raising of sheep already occurred elsewhere in the March, for instance in Cydweli and Glamorgan, and although the Cistercians were already exploiting the huge sheep pasturelands of Brecon, extensive sheep-rearing on the part of our lords would seem to be a development that falls after this period. Another resource of great value to lords were forests, both for hunting and the chase, and for their resources in wood, honey, pannage for swine, and herbage for cattle.¹⁰⁵ The Great and Little Forest of Brecon ¹⁰² Rot. Litt. Claus. i. 18, 19b. ¹⁰³ ‘The Irish Pipe Roll of 14 John’, 37–9; Davies, Lordship and Society in the March of Wales, 102, 134. ¹⁰⁴ ‘Ministers’ Accounts for the Lordships of Abergavenny, Grosmont, Skenfrith and White Castle’, 115. ¹⁰⁵ Renders of honey are mentioned in Domesday Book for fourteen places in Herefordshire as well as in Archenﬁeld; Darby and Terrett (eds), The Domesday Geography of Midland England, 104–5.

76

Lords of the Central Marches

covered over 50 square miles of Cantref Mawr and were among the most valuable possessions of the lords of Brecon. Most of the March in the Middle Ages was not as manorialized as one would expect, but the major river valleys of our region were among the exceptions.¹⁰⁶ In the lowlands of the Usk, Llynﬁ, and Wye practice came more and more to resemble that of lowland England, as English colonists settled the valleys and lords originally developed home farms to feed their households and castle garrisons. Later these seigneurial manors would be shifted over to production for the market. A mixture of leasing and demesne farming was practised. The manor of Monmouth in the mid-thirteenth century may serve as an example. There the revenues, excluding judicial, were composed as follows: rents of the assize formed 8.5 per cent, the farm of the mill 15 per cent, issues of the manor 32.5 per cent, the sale of corn 9 per cent, the sale of wine 4 per cent, and ﬁnes and escheats 19 per cent.¹⁰⁷ This last category may be considered as quasi-judicial, but was a concomitant of letting out land. The largest category, issues of the manor, was composed of several items, the largest being the sale of wood followed by that of livestock. Pannage of swine, stall rentals, and the sale of honey, ale, and cider were other items.¹⁰⁸ The sale of corn was a smaller item of revenue, with rents ﬁguring more prominently. In comparison, in the revenues of the manor of Abergavenny, with its outlying members of Penrhos, Llangattock, St Michael, and Tregoythel, rents of the assize, at 8.25 per cent of the total, and the farm of mills, at 14.6 per cent, were quite similar to Monmouth. But the issues of the manor were much lower at Abergavenny, contributing only a little over 12 per cent of the year’s income. Sale of corn was higher at 16.5 per cent, and sale of wine (8 per cent) brought in double the income it did at Monmouth. Fines and reliefs, as they were labelled on the Abergavenny account, came in at only 3 per cent of income, but this of course was a highly variable source of revenue. The largest category of income for Abergavenny was a non-agricultural one: the pleas and perquisites of its Great Court provided a massive 36 per cent of its gross income for 1256/7.¹⁰⁹ If a Marcher lord could not increase agricultural proﬁts, he could always try to extract money through his courts. Nevertheless, ¹⁰⁶ Davies, Lordship and Society in the March of Wales, 107–8. ¹⁰⁷ Mills were very important sources of revenue for members of the knightly class as well and were frequently the object of litigation; see Ch. 4. ¹⁰⁸ ‘Ministers’ Accounts for the Lordships of Abergavenny, Grosmont, Skenfrith and White Castle’, 11–12. ¹⁰⁹ Ibid., pt I, 101–5.

Lords and Honours

77

the account rolls testify to both a diversiﬁed, mixed economy and its vigorous exploitation by the lord. One matter of importance to lords in the late twelfth and early thirteenth centuries was the amount of land held in demesne. One reason for this was the effect of inﬂation from c.1180 to 1220. Rising agricultural prices meant that it was to the advantage of lords to take land back into direct cultivation rather than let it out at ﬁxed rents, rents which often proved difﬁcult to adjust in order to keep pace with rising prices.¹¹⁰ Another was that land was still the most valuable gift with which a lord could attract and secure followers or endow or augment a religious house. The problem for lords was that the second of the above points worked directly counter to the ﬁrst. We have seen above how the Lacy demesne in Herefordshire was seriously depleted by the late twelfth century.¹¹¹ This was undoubtedly a major reason why, beginning with Hugh II, the Lacys concentrated more of their time and energy on their Irish lands than on those in Herefordshire or the March. Lords in Herefordshire, such as the Lacys, had to contend with the protections afforded to tenants by the king’s courts. They could not, as did Richard de Clare in Glamorgan, utilize their own courts and dubious pretexts to disseise tenants of their lands and thus rebuild the seigneurial demesne. And although Marcher lords had more latitude to increase their direct control of land, many of them had their work cut out for them. By 1200 much of the lordship of Brecon had been granted out in the form of mesne tenancies as well. Its conqueror Bernard de Neufmarché had granted Hay to William Revel (although it was resumed back into the lord’s demesne at some point in the mid-twelfth century), Bronllys to Richard ﬁtz Pons, Llantsantffraed to Walter de Cropus, Ystradyw to the Picards, and Crickhowell to Hugh de ¹¹⁰ For the ‘great English inﬂation’, see M. M. Postan, ‘Medieval Agrarian Society in its Prime: England’, in Postan (ed.), The Cambridge Economic History of Europe, 2nd edn (Cambridge, 1966), i. 590–5; Edward Miller, ‘England in the Twelfth and Thirteenth Centuries: An Economic Contrast?’, EcHR, 2nd ser., 24 (1971), 1–14; P. D. A. Harvey, ‘The English Inﬂation of 1180–1220’, P&P 61 (1973), 3–30, and id., ‘The Pipe Rolls and the Adoption of Demesne Farming in England’, EcHR, 2nd ser., 27 (1974), 345–59. Recent debate has focused on possible mechanisms for this inﬂation (Jim Bolton has suggested a small amount of currency moving at increased rates throughout the kingdom, putting pressure on the currency and generating rising values), but the orthodoxy of inﬂation itself has stood; see J. L. Bolton, ‘The English Economy in the Early Thirteenth Century’, in S. D. Church (ed.), King John: New Interpretations (1999), 27–40. ¹¹¹ See Ch. 2.

78

Lords of the Central Marches

Turberville. Bernard kept Cantref Mawr but carved out of it Pencelli and Drostre (Trosdref) for Roger de Baskerville. Furthermore, when the lordship changed hands, new men came with the new lord and were rewarded accordingly. Miles of Gloucester most likely enfeoffed the Brets and Poers, while the Havards, Muscegros, and Surdevals came with William de Braose.¹¹² The thirteenth century saw further diminution of the resources of lordship in Brecon. The loss of Blaenllyfni to Peter ﬁtz Herbert in 1211 meant that the manor of Talgarth and valuable ﬁshing in Llangorse Lake were lost, as well as service from the Pichards and Turbervilles for their lands north of the Usk. In the thirteenth century the burgess community of Brecon at some point secured privileges that would not be reined in until the mid-fourteenth century. For those lords in the March as well as Herefordshire seeking to increase direct control over land, purchase or exchange were their best options, as escheats occurred too randomly to be depended upon, and wilful, large-scale distraint could cause strains within the honorial community. One example of exchange comes from the honour of Weobley, when Walter de Lacy gave Simon de Clifford £30 in rent drawn equally from the manors of Holme Lacy in Herefordshire and the vill of Haye in Ireland in return for the manor of Yarkhill in Herefordshire.¹¹³ In 1166 Hugh de Lacy had held 2 hides in Yarkhill in demesne, and 2 hides worth 50s. are all that is described in the Domesday entry.¹¹⁴ For this exchange to make any sort of economic sense for Walter de Lacy, the value of Yarkhill must have soared by the late 1220s. Another example dates from 1196. A ﬁnal concord dated 6 April shows that William de Braose, probably William IV the younger, brought a plea of right against Meurig ﬁtz ‘Wranno’ for 3 carucates of land in Yazor.¹¹⁵ Meurig quitclaimed the land to Braose and received 3 carucates of land outside the vill of Abergavenny. Yazor lay roughly a mile to the north-east of the important Braose demesne manor of Mansel Lacy, just the other side of what is now known as Nash Wood. By means of this exchange William de Braose was rounding out his demesne holdings in Mansel Lacy. On a larger scale, in 1218 Hugh de Mortimer attempted to exchange the manor of Worthy in Hampshire ¹¹² Rees, ‘The Medieval Lordship of Brecon’, 204–5. ¹¹³ HCA, no. 483; Bodl., MS Rawlinson B329, fos 175–175v . The reference to the heir of Stephen Devereux in the charter dates it to between 1228 and 1236, when William Devereux was still a minor. ¹¹⁴ HDB 44. ¹¹⁵ Feet of Fines of the Reign of Henry II and . . . Richard I, 110, no. 124.

Lords and Honours

79

with Thomas de Erdinton in return for the Shropshire manors of Knighton and Norton.¹¹⁶ Along with purchase and exchange, new or marginal land was brought under cultivation (‘assarted’, in the period’s terminology) from an early date. Lords rarely bothered to clear land themselves; instead, they tended to impose new rents on any land assarted by their tenant peasants. A conﬁrmation charter of Geoffrey, bishop of St Davids, issued to St Peter’s, Gloucester, in 1204–14 mentions grants from the mid-twelfth century which included ‘the tithes of all assarts made and to be made in the demesne of Glasbury’ in a grant by Richard de Clifford, and a similar clause exists in a grant by Roger, earl of Hereford, for Talgarth in Brecon.¹¹⁷ But for great lords, assarting could have only a small effect on their bottom line; as K. B. McFarlane said of the later medieval aristocracy, ‘the improving landlord was only too often one whose assets were in danger of shrinking and whose other traditional outlets were barred’.¹¹⁸ Assarting might be ﬁne for the knights; greater lords needed larger amounts of new land. But as lords attempted to increase the amount of land under their direct exploitation, the pressures on that land also increased. One example of the pressures to which demesne lands were subjected can be seen in the differing uses to which the Lacy demesne manor of Holme Lacy in Herefordshire was put.¹¹⁹ The manor, originally extracted from the bishop of Herefordshire, was subsequently fragmented by a number of grants. The manor was assessed at 5 geldable hides and was worth £8 in 1086.¹²⁰ In the charter to Simon de Clifford discussed above, the £15 rent he was to receive from Holme Lacy was hedged with restrictions which show the degree to which the demesne was encumbered: the lands given to the prior of Craswall Priory, William ﬁtz Warin, Stephen Devereux, and Henry de Bradelegh were all excluded.¹²¹ Furthermore, ¹¹⁶ TNA, E210/7074; BL, Harleian MS 1240 (Cartulary Mortimer), fo. 68v ; The Victoria History of Hampshire, iv, ed. William Page (1911), 426; CPR 1216–25, 49; for more on this exchange, see Ch. 7. ¹¹⁷ Episcopal Acts and Cognate Documents relating to Welsh Dioceses, D.406, p. 340. ¹¹⁸ From his 1953 Ford Lectures, pub. in K. B. McFarlane, The Nobility of Later Medieval England (Oxford, 1973), 10. ¹¹⁹ H. M. Colvin, ‘Holme Lacy: An Episcopal Manor and its Tenants in the Twelfth and Thirteenth Centuries’, in Veronica Ruffer and A. J. Taylor (eds), Medieval Studies Presented to Rose Graham (Oxford, 1950), 15–40. ¹²⁰ HDB 25–6. ¹²¹ The grant of 12 virgates in Holme Lacy to Stephen Devereux by Gilbert de Lacy for homage and the service of half a knight pro eo quod sit in familia mea is recorded in HCA, no. 3235.

80

Lords of the Central Marches

Walter II de Lacy’s wife, Margaret, had been given dower lands in Holme Lacy.¹²² This was a very encumbered manor; nothing better illustrates the pressures on demesne lands in this period. Our lack of equivalent material for the Marcher lordships makes it difﬁcult to determine whether the same situation applied there. The charters in the Brecon Cartulary do not, however, show any signiﬁcant grants of either land or rents made under the Braoses; William III de Braose had to make do with conﬁrming the grants of churches made by his predecessors.¹²³ The balance maintained by the Monmouths in their lordship, with approximately one-third of the knight service owed for the honour coming from demesne lands, has been discussed above.¹²⁴ Even then, there were pressures on the Monmouth demesne; Gilbert de Monmouth gave his land of Hadnock near Monmouth to form the marriage portion for one of his tenants, Adam de Blakeney, upon his marriage to one Basilia.¹²⁵ The relative lack of demesne spurred the development of other resources. The Monmouths at times substituted grants of forges for land; Baderon de Monmouth had originally gained Hadnock from Monmouth Priory for three forges, and also made a grant of a forge in Monmouth to his tenant William de Marisco.¹²⁶ This exploitation of the charcoal and iron industries of the Monmouth–Forest of Dean area demonstrates the economic entrepreneurship of the barons of the region. Ultimately, the military nature of the March, whose demands in one sense spurred lords to develop their lands as proﬁtably as possible, also served to set limits to that development. Marcher lords had constantly to balance the desire, increasing in the thirteenth century, to exert direct control and exploitation of land with the need to provide for tenants holding by military service. It would be in the fourteenth century, after Wales was conquered and the royal government bore much more responsibility for its policing and defence, that lords such as Humphrey de Bohun of Brecon would be able to extract sums impressive even by English standards from their lordships in the March. In the twelfth and thirteenth centuries our lords were engaged in a balancing act: their lands were not unproﬁtable, but their expenses would seem to have outstripped their revenues. It was for this reason among others that the ¹²² Bodl., MS Rawlinson B329 (Hereford Cartulary), fo. 145, a quitclaim of Margaret de Lacy’s dower lands to Craswell Priory. ¹²³ ‘Cart. Brecon’, 14. 155. ¹²⁴ See Ch. 2. There had, however, been considerable subinfeudation from 1086 to 1166. ¹²⁵ TNA, DL 25/2069. ¹²⁶ Mon. Ang. iv. 596; TNA, DL 27/91.

Lords and Honours

81

Lacys and brieﬂy William de Braose became involved in Ireland, where opportunities for expansion in the late twelfth and early thirteenth centuries were more promising. And one need look no further for the reason behind the ﬁnancial, and ultimately political, gambles which William de Braose took in the ﬁrst decade of the thirteenth century as he expanded enormously the lands in Wales over which his family had control or custody.¹²⁷ R E L I G I O U S PAT RO N AG E As was noted earlier, religious foundations could be a serious drain on the resources of lords in our region. But it was a necessary drain, one that few lords would hesitate to make in some form. First and foremost were religious considerations. The language of charters granting land to religious houses, where gifts were made pro salute animarum of ancestors, children, the lord, and his wife, should not be dismissed for all that it seems formulaic. The spiritual beneﬁts to be gained—prayers, confraternity, and burial—were very real. One example comes from Brecon Priory.¹²⁸ William III de Braose gave his body to the priory along with a conﬁrmation of all grants made by his predecessors in the lordship (incidentally indicating the pride of place that Brecon occupied among his far-ﬂung holdings). What William was actually receiving was the right to be buried in the priory church. A similar grant was made by Roger de Clifford, the younger brother of Walter III, to Dore Abbey.¹²⁹ Religious houses thus also showed the attachment of a lord to a particular lordship, and were yet another way of demonstrating his wealth, muniﬁcence, and, by extension, power. There were, however, beneﬁts to the lord beyond those of religion. In the early days of English expansion into Wales, monasteries and priories were useful in consolidating control of the region. For example, only two lords in the Conquest generation had founded new Benedictine monasteries in England: Roger de Montgomery at Shrewsbury and Earl Hugh at Chester. It is no coincidence that both were sited along a border where the new ruling class was trying to assert control.¹³⁰ Brecon ¹²⁷ See Ch. 6. ¹²⁸ ‘Cart. Brecon’, 14. 156. ¹²⁹ Mon. Ang. v. 555. Roger granted all the land that he had of the gift of William of Ewyas. Roger was married to Sybil of Ewyas, so he may have been granting her maritagium, a common enough pattern. ¹³⁰ Green, The Aristocracy of Norman England, 399.

82

Lords of the Central Marches

Priory largely served the same purpose after its foundation by Bernard de Neufmarché.¹³¹ The new foundations were staffed by monks imported from England or France; military domination was quickly followed by spiritual domination. Monasteries and priories also could provide men trained in drawing up documents, and a place for younger sons left out of the patrimony. A monastic foundation could also serve as a focus for the loyalty of the tenants of an honour, for it gave them a means of displaying that loyalty to, and their connection with, their lord. Research has shown that just as multiple ties of lordship proliferated in the twelfth century if not sooner, so in England many tenants patronized foundations linked to different families.¹³² It is signiﬁcant to note that, in the case of Brecon Priory, there seems to be little overlap between its donors and tenants of other honours in the region. Brecon Priory was very much the foundation of the lords of Brecon and as such served as a focal point of the life of the lordship, much as the lord’s curia did.¹³³ The major lords themselves, of course, distributed their patronage more widely. Families such as the Lacys, who held land in Ireland, founded or patronized houses there, such as St Thomas, Dublin. Closer to the region, widespread patronage was useful in cultivating ties and contacts outside the lord’s immediate estates; hence the Lacy donation to Monmouth Priory or that of John de Monmouth to Clifford Priory.¹³⁴ Family circumstances could also lead to benefactions further aﬁeld, as the Clifford gift of a mill and meadow in the manor of Frampton, Gloucestershire, to Godstow Priory in Oxfordshire, which Rosamund de Clifford had entered as a nun, demonstrates.¹³⁵ The Braoses patronized an enormous number of religious houses, scattered across England and Wales: Margam, Flaxley, Sele, Battle, Lyre, Neath, the Priory of St John the Baptist at Carmarthen, Llanthony, and Aconbury, in addition to Brecon Priory. The emphasis on houses in Wales and the borders, however, is an eloquent testimony to the role that the region played in the wider Braose holdings. Most of the religious houses of the region were founded early in the post-Conquest period. Brecon Priory (Benedictine) had been founded c.1100, while the Lacy foundation of Llanthony Prima (Augustinian) ¹³¹ See Ch. 2, for the conquest of Brecon. ¹³² Emma Cownie, Religious Patronage in Anglo-Norman England 1066–1135 (1998), 175–6. ¹³³ See Ch. 2. ¹³⁴ Mon. Ang. iv. 597, v. 42. The Lacy grant was probably occasioned by the fact that Hugh II de Lacy married Rose, the widow of Baderon de Monmouth, who appears as co-grantor. ¹³⁵ Mon. Ang. iv. 358, no. 15.

Lords and Honours

83

occurred in 1103. Clifford Priory (Cluniac) was founded 1129–30, Abergavenny (Benedictine) before 1100, and Monmouth Priory (Benedictine) in 1086.¹³⁶ But patterns of religious patronage changed with both fashions and economic circumstance. The great age of Benedictine monastic foundations had long closed by 1166. In the late eleventh and early twelfth centuries the Benedictines were succeeded by newer orders—the Augustinians, Cistercians, and Premonstratensians—which were both more fashionable and also easier and less costly to establish.¹³⁷ The Lacys founded the great Augustinian house of Llanthony Prima in their lordship of Ewyas Lacy in 1103, while the Cistercian abbey of Dore (founded 1147) was patronized by the Cliffords. With their emphasis on the development of waste and marginal tracts, and their development of sheep farming, the Cistercians were particularly suited to the conditions of the borderlands and March. Dore Abbey is an example of this, and by the thirteenth century it controlled large tracts of upland pasture and forest in Cantref Selyf in Brecon and elsewhere. The religious landscape was thus already established by the midtwelfth century. Piety towards houses founded or patronized by one’s ancestors combined with diminished resources to limit new foundations, as did the prohibition of new houses by the Cistercian general chapter in 1152. It should be noted that this decline in religious benefactions and foundations was a general trend, noticeable throughout England from the mid-twelfth century, but the lack of resources was particularly acute in the March and borders.¹³⁸ Those houses which were founded in the region from the middle of the twelfth century were smaller in scope than those of the ﬁrst century of English rule in the region. Grace Dieu is one example, founded in 1226 by John de Monmouth, no doubt partly as a reﬂection of his rising status as royal forester and the king’s representative in the area. A Cistercian foundation, it most likely never had more than the obligatory thirteen monks. Aconbury, a house of Augustinian canonesses, is another example. Founded by Margaret de ¹³⁶ For dates of foundation, see the entries in Knowles and Hadcock, Medieval Religious Houses in England and Wales. ¹³⁷ For the lower cost to a founder of Cistercian houses, see Knowles, Monastic Order, 246–7; one appeal of the Augustinians was ‘the nature of the new foundations, often incorporating older communities, and not as expensive to found as Benedictine houses in proportion to their founder’s total wealth’; Green, The Aristocracy of Norman England, 399. ¹³⁸ For this trend, see Green, The Aristocracy of Norman England, 393.

84

Lords of the Central Marches

Lacy c.1215 in memory of her father, William III de Braose, her mother, Maud de Braose, and her brother William IV, the initial grant was made by King John himself in his last days and totalled 3 carucates of land for assarting in the forest of Aconbury.¹³⁹ Thus, new foundations were both few and small, and this was largely owing to diminishing resources. Walter de Lacy even had to assign burgage rents and a ninth rather than a tithe in several manors in Ireland to Craswall Priory, showing just how stretched his resources in Herefordshire and the March of Wales were.¹⁴⁰

FA M I LY LO R D S H I P Finally, the pressures which bore on lordship along a military frontier, and the move towards the consolidation of lordship, led our magnates to make hard choices about their family priorities. This is demonstrated by the marriage strategies which they pursued. Prior to his fall in 1208, William III de Braose had augmented his family’s standing by marrying his heir, William IV, to Matilda, daughter of Richard de Clare, earl of Hertford; his son Reginald was married to Graecia, a daughter of the important curialis William Brewer. After the family’s restoration in 1214, however, their star had waned and their ambitions correspondingly had contracted. Reginald’s second marriage was a political one to Gwladys, a daughter of Llywelyn ap Iorwerth, while his son, with perhaps an eye towards securing part of the eventual Marshal inheritance, married Eva Marshal. Concerns which were more local in nature had come to the fore. Likewise, the marriages of the Mortimers seem to indicate a contraction of ambitions, or at least a more determined concentration on the March. Hugh II de Mortimer (d. 1180/1) had married Matilda, daughter and co-heir of William de Meschin of Skipton-in-Craven in Yorkshire, while his second son and eventual heir Roger (d. 1214) had married Isabella de Ferrers. Mortimer interests extended well beyond the March. But then the family’s focus narrowed. Roger’s eldest son, Hugh (d. 1227), married Annora de Braose, and his second son, Ralph (d. 1246), married at some point after 1228 Reginald de Braose’s widow, Gwladys. The Mortimers, for so long stymied in their attempts ¹³⁹ Rot. Litt. Pat. 199b.

¹⁴⁰ HCA, no. 482.

Lords and Honours

85

to reconquer Maelienydd from the Welsh, used marriage to consolidate and strengthen their position in the region. Indeed, Ralph’s son Roger, by his marriage to another Braose, Matilda, would add the Marcher lordship of Radnor to the family’s possessions. The same pattern of concentration held true for the Cliffords. Walter I de Clifford (d. 1190) had married Margaret de Tony, who may have brought the barony of Clifford to the family, while his son Walter II (d. 1220) married Agnes de Cundy.¹⁴¹ Agnes brought lands in Lincolnshire held of the bishop of Lincoln, and lands in Kent.¹⁴² The Cliffords were thus broadening the scope of their possessions.¹⁴³ In the next generation, however, we see the same narrowing focus in marriage patterns, for Walter III married ﬁrstly Katherine de Lacy, and this marriage, which was dissolved, was followed by marriages to a certain Isabelle and to Margaret, a daughter of Llywelyn ap Iorwerth. His brother Roger married Sybil of Ewyas Harold, daughter of another local lord. While some families concentrated on consolidating their power in the March and the borders, others looked elsewhere. The shift in emphasis of the Lacys towards Ireland, no doubt because of the encumbrances on their lands in Herefordshire and the fact that they were no longer poised to expand in Wales, is readily apparent in their marriages. While Walter II de Lacy married Margaret, a daughter of William III de Braose, his brother Hugh, earl of Ulster, married ﬁrstly Lesceline de Verdun and secondly Emeline de Ridelsford, both of Anglo-Irish baronial families. Walter’s son and heir Gilbert was also married to a possible Marshal heiress, Isabella Bigod, though whether with an eye towards the Welsh March or Ireland it is difﬁcult to say. Walter’s daughters were used to cement alliances in both areas, with Egidia marrying the sometime Irish justiciar Richard de Burgh and Katherine marrying the Marcher lord Walter III de Clifford.¹⁴⁴ Finally, the Monmouths initially showed the same localization in marriages, but as the family began to tap into royal patronage, its marriage horizons widened. Baderon de Monmouth was married to Rose, daughter of Gilbert ﬁtz Richard de Clare. Gilbert de Monmouth’s marriage is unknown, but his son John married Cecilia, a daughter ¹⁴¹ DNB, s.v. Walter de Clifford. Ralph de Tony was lord of Clifford in Domesday; HDB 34. Margaret is assumed to be Ralph’s daughter by the author of the DNB entry. ¹⁴² Eyton, Antiquities of Shropshire, v. 157. ¹⁴³ Although Walter II’s brother Richard had married within the region, having married Letitia de Berkeley. ¹⁴⁴ Katherine was set aside at some unknown point.

86

Lords of the Central Marches

and heir of Walter de Walerand. John thus obtained one-third of the Wiltshire barony of West Dean and this represents an expansion beyond the region rather than a concentration on marriage within it.¹⁴⁵ This marriage, contracted probably before John came into the wardship of William de Braose, probably reﬂects a conscious strategy on the part of his father, Gilbert.¹⁴⁶ Lordship in western Herefordshire (although to a decreasing degree), and more particularly the March, thus existed in what by 1166 must have seemed an archaic ‘feudal matrix’.¹⁴⁷ Border lordships, they were organized ﬁrst and foremost with military concerns in mind. By the later twelfth century those concerns in turn dictated the need to preserve the honour as a working unit, and this helped to determine the nature of judicial lordship in the March. Although limited in their power by the ‘custom of the honour’ and the practicalities of local politics, the Marcher lords strove to retain and defend against royal encroachment the judicial discretion upon which depended distraint, the means of enforcing military service and potentially of enlarging the lord’s direct territorial control.¹⁴⁸ As far as they were able, living along a violent and often unsettled frontier, they developed their estates in the lowland river valleys along recognizably English lines and maximized cash income from them. This, however, was still not enough to keep the Lacys and Cliffords from falling into rather serious debt by the 1230s. This failure of revenues to keep pace with the demands of border lordship is also borne out by the diminishing patronage extended to the favoured religious houses of these families in the late twelfth and thirteenth centuries. In terms of the intensiﬁcation and consolidation of lordship, the Marcher lords of our region had made strides in their military and judicial powers, but their economic lordship perhaps was insufﬁcient to underpin these advances. None, not even the Braoses, were as effective in extending and deepening their power as were the Clare lords of neighbouring Glamorgan in the years 1240 to 1295. ¹⁴⁵ Sanders, English Baronies, 96. ¹⁴⁶ In 1201 John de Monmouth owed 120 marks and two goshawks for seisin of his share of West Dean; Rot. Ob. et Fin. 178. ¹⁴⁷ The phrase is that of Davies, ‘Kings, Lords and Liberties in the March of Wales’, 49. ¹⁴⁸ For an example of an appeal to ‘the custom of the honour of Glamorgan’, albeit during a period of royal custody in 1214, see Rot. Ob. et Fin. 535.

Lords and Honours

87

From 1166 the growing reach of Angevin government and the royal courts led to the increasing incorporation of Herefordshire within the English kingdom, separating it still further from the March.¹⁴⁹ This in turn almost certainly led our Marcher lords, through their honour courts, the arrangement of tenants’ marriages, and their powers of wardship, to inﬂuence inheritance patterns in the March, as knightly estates began to break down along honorial lines in the late twelfth and early thirteenth centuries, largely separating the tenants of Marcher lordships such as Brecon from those in Herefordshire honours.¹⁵⁰ The role of these two other elements, the knights and the king, in the shaping of the region from 1166 to 1246 are the topics of the next two chapters. ¹⁴⁹ See Ch. 5.

¹⁵⁰ See Ch. 4.

4 The Knightly Class The attempts, some successful, at the consolidation of lordship just discussed had an impact on the military tenants of lords such as Walter de Lacy and William de Braose. These tenants, prominent on a local scale but less signiﬁcant on the larger social and political one, formed what historians often call ‘the knightly class’. They were crucial to the power, status, and prestige of our lords; to be a lord without followers was not to be a lord at all. There was no automatic opposition of these two groups, lords and knights; the exercise of good lordship brought them together. There were, however, a number of tensions arising out of both the economic environment—inﬂation, the lack of available land—and the changing legal environment of the late twelfth century that could contribute to friction between lords and their men. These factors could be found throughout England, but in the border counties and the March there were additional burdens that potentially made the position of the knights even more difﬁcult. There was the not inconsiderable expense of living along a militarized frontier. Further, there were the effects of the intensiﬁcation and consolidation of lordship. For knights in the March, such consolidation meant tighter control by lords over marriage, inheritance, and the disposition of family lands. As lords held royal justice at arm’s length, and royal ofﬁces were thus unavailable, Marcher knights found themselves shut off from a key conduit of royal patronage. On the other hand, some Herefordshire knightly families used changing conditions to their advantage. It is the way in which those changes affected the knightly class of Herefordshire and the March that we must particularly examine. Who are we referring to as ‘the knightly class’? Firstly, it should be noted that, for the sake of convenience, ‘knights’ is used here as a broad, shorthand term, synonymous for our purposes with the slightly awkward term ‘knightly class’. Secondly, this usage by no means indicates that every member of the knightly class in our region carried the appellation ‘knight’, particularly in the thirteenth century, when the honoriﬁc miles

The Knightly Class

89

denoted someone who had been dubbed a knight in a ceremony, and in addition was expected to have a certain level of wealth. Instead, we are here discussing the larger body of men, holding land by military service of the magnates, from whom those who were dubbed ‘knights’ in the thirteenth century were drawn. The discrepancy between this large ‘knightly class’ and the somewhat smaller body of those who were ofﬁcially ‘knights’ is an example of both the enlarging of the aristocracy (what has been termed the ‘diffusion of nobility’) and the deﬁnition of status that occurred within the English aristocracy during the Middle Ages.¹ Originally meaning merely a soldier on horseback, miles, chevalier, or knight came to mean a member of the lower reaches of the aristocracy, as magnates assumed the cultural trappings of knighthood, and knights in turn adopted from the magnates the practice of primogeniture, toponyms based on their chief seat, and an aristocratic display of wealth, through both the building of fortiﬁcations in stone and the patronizing of monasteries. That said, the term ‘knight’ was used much more loosely in 1166 than it was in 1266. Many, if not most, of the tenants of the honour of Weobley who were listed in the carta returned by Hugh de Lacy in 1166 would have been considered knights, but not all their descendants were honoured with this appellation in the writs and charters of the middle of the thirteenth century. This trend can be seen from the witness lists of charters where, by the thirteenth century, men who had taken up the burdens of knighthood were speciﬁcally described as milites, differentiating them from other tenants who had not been dubbed, despite their otherwise similar wealth and holding of land by military service. Knighthood was more ambiguous in the twelfth century, particularly earlier in the century, than in the thirteenth. By the thirteenth century the assumption of knighthood itself can no longer be considered a sine qua non of actual knightly status, as many men who could afford to take up knighthood decided not to do so because of the cost and burdens involved. Historians have offered various deﬁnitions as to just what characteristics delineate the ‘knightly class’ of the twelfth and thirteenth centuries, and such a deﬁnition of parameters is important, if difﬁcult and arbitrary. Certain commonly agreed elements can be used: lordship of one or more manors; frequent assumption of knighthood, although this decreases in the course of the thirteenth century; the holding of local ¹ For this diffusion or enlarging of the concept of nobility, see P. R. Coss, The Origins of the English Gentry (Cambridge, 2003), 35–8.

90

Lords of the Central Marches

ofﬁce, such as coroner, sheriff or under-sheriff, forest ofﬁcial, or collector of royal aids and other subsidies.² Not all of these elements are necessary in order to be included in the group; various combinations of them are to be found depending upon the individual under examination. In many respects, these attributes point towards the gentry of the later Middle Ages and beyond.³ In the widest sense, anyone holding land by military service (although not by serjeanty) counts. Here we are discussing a broad if still recognizable class in society. Just what is the size of the group we are examining here? In Herefordshire, where, because of the records of the royal courts, the evidence for the identiﬁcation of men who were knights is much more plentiful than in the Marcher lordships, a rough idea of numbers may be reached for the thirteenth century. Examining the eyre roll of 1255, one historian placed the county in the lowest of his three categories, with only twenty-eight knights appearing in four grand assize lists.⁴ By this measure, Herefordshire possessed few knights by the middle of the thirteenth century. Judicial records have their limitations, of course. An examination of the printed Curia Regis Rolls, counting grand assize electors and jurors, as well as knights sent to view essoins of bed sickness, does not show any of the men from the level of the greater knights, such as the Pichards and Devereux. The probable reason for this is that the grand assize emphasized that the knights who were to determine right to the land in question were to be drawn from the neighbourhood, the visnetum, and this rule seems to have been very closely followed, at least in the early thirteenth century.⁵ This narrow deﬁnition of who was a ‘knight’ is not used here. Firstly, by trying to count the number of ² D. A. Carpenter, ‘Was there a Crisis of the Knightly Class in the Thirteenth Century? The Oxfordshire Evidence’, EHR 95 (1980), 722. ³ I have followed Peter Coss here in not actually using the term ‘gentry’ as a synonym for the knightly class in the period under examination. Coss points out that the crucial difference between the knightly class of the 12th and 13th centuries, and the later gentry that emerged from it, is that horizontal ties among the knights had not yet become a primary characteristic of the group, and that despite their expanded role they were still agents of the state, rather than the partners of it they became with the rise of justices of the peace; The Origins of the English Gentry, 39–43. ⁴ J. Quick, ‘The Number and Distribution of Knights in Thirteenth-Century England: The Evidence of the Grand Assize Lists’, in P. Coss and S. Lloyd (eds), Thirteenth Century England I (1986), 123. ⁵ P. R. Coss, ‘Identity and the Gentry c.1200–c.1340’, in Prestwich et al. (eds), Thirteenth Century England VI, 52–3; id., The Knight in Medieval England (Stroud, 1993), 36–8; K. Faulkner, ‘The Transformation of Knighthood in Early ThirteenthCentury England’, EHR 111 (1996), 1–23.

The Knightly Class

91

knights in this fashion, we are at the mercy of which cases are brought, for they decide geographically which knights are likely to end up on a grand assize jury. Secondly, it must be remembered that the lack of the formal appellation miles does not mean that a man was not a member of the level of society from which the knights were drawn. Once again, we will be examining a broader class of individuals than ofﬁcial designations and court documents would reveal as ‘knights’. What role did these men play in the lordships of the region? A magnate need not cultivate his feudal tenants to any great degree, and not all knights held land. For example, of the eighteen men closest to William Marshal, earl of Pembroke (d. 1219), and thus most likely to receive his patronage, twelve had no feudal ties to him.⁶ The Anglo-Scottish magnate Roger de Quincy (d. 1264) numbered only two of his ﬁve major military tenants among his familia, and they were in the ‘outer circle’.⁷ By the end of the twelfth century it was not uncommon for a lord’s familia to be extra-tenurial in composition. The March, however, does not seem to have followed this trend. An examination of the Braose acta for Brecon shows them surrounded by the military tenants of the lordship.⁸ These particular acta are admittedly small in number, but they seem to indicate that the Marcher lords, at least within their Marcher lands, recruited their familiae from among their tenants. This is partly attributable to the fact that such lords, or their sons, spent a large amount of their time in their Marcher lands. Marcher lordship by necessity was handson lordship.⁹ Another reason for this solidly tenurial composition of Marcher households and afﬁnities is the tenurial solidity of the Marcher lordships themselves. Within these lordships all land was held of one lord and, as will be shown later, by the mid-thirteenth century very few Marcher feudal tenants had tenurial ties outside their lordships.¹⁰ ⁶ David Crouch, William Marshal, 2nd edn (2002), 147–9. ⁷ G. G. Simpson, ‘The Familia of Roger de Quincy, Earl of Winchester and Constable of Scotland’, in K. J. Stringer (ed.), Essays on the Nobility of Medieval Scotland (Edinburgh, 1985), 121. ⁸ This body of acta consists of eight charters with witness lists. It should be noted that some charters printed in ‘Cart. Brecon’ are repeated in Bodl., MS Tanner 342, a 17th-century transcript, with slight differences in the witness lists. While more than one conﬁrmation may have been issued by any one Braose lord of Brecon, where there is any similarity between the witness lists I have discounted the Tanner MS copy. I have also excluded a charter of John de Braose, BL, Harleian MS 2044, fo. 50r , as John was lord of Gower, which falls out of the scope of this chapter. ⁹ See Ch. 3. ¹⁰ Brecon in 1211 accounted for thirty-two knights’ fees held by thirty-six men, one woman, and one group of heirs; by 1243 there were only ﬁve men with ties of multiple lordship. The only major holder was Roger Pichard, who held of the honours of Brecon

92

Lords of the Central Marches

Although studies of the knights in the twelfth and thirteenth centuries have usually taken the county as their unit of study, as with our examination of our magnate families, this approach is not adopted here. The wealthier and more powerful knights of this region initially had interests that crossed the boundaries between England and the March, although, as will be seen, this cross-border holding of land decreased markedly in the thirteenth century. A good late example is Roger Pichard (f l. 1230–48), who in 1243 was lord of the Brecon sub-lordship of Tretower, with his seat at Tretower Castle. Yet he also had three demesne manors and was the mesne lord of three others in Herefordshire. The county was only one arena for the advance of knightly families, and not always the most important one. In this region, the honour, and, in a looser sense, the followings gathered around area magnates, were just as important. The men (and a few women) to be studied in this chapter are for that reason the military tenants of the Braose, Clifford, Lacy, Monmouth, and Mortimer lordships of western Herefordshire and the March. Their names are drawn from the Cartae Baronum of 1166 and the feodaries of the thirteenth century. A warning: the evidence for the study of knights in this region and period, especially those in the March, is hard to come by. Limitations of evidence, particularly charter evidence, mean that on the whole the wealthier and more prominent tenants are focused upon. These greater knights were also quite often mesne lords of the lesser knights, those holding perhaps only a single manor or even a portion of one, and some attempt at generalizations about these lesser knights will be attempted. It is also much easier to discuss knights on the English side of the border in Herefordshire and southern Shropshire than it is in the March; the greater quantity of information on them is just another indication of the growing reach of royal government and thereby of documentation that affected the knights of the border counties. Three prominent knightly families have been chosen for particular attention: the Baskervilles, the Devereux, and the Pichards. At the beginning of the twelfth century these families linked the Braose and Lacy lordships of the region, holding manors of the honours of both Brecon and Weobley. By the middle of the thirteenth century, this had and Weobley; as the century progressed the Pichards would continue to divest themselves of land in Hereford. The others were Godfrey de Gamages (Brecon and Weobley), Henry of Pembridge (Brecon and Radnor), William de Blez (Brecon and Radnor), and William de Furches (Brecon and Weobley); BF 797–818 passim.

The Knightly Class

93

largely ceased to be the case, with cadet branches at best linking the two honours. This process of gradual separation on the part of those knightly families who held substantial lands in both the March and Herefordshire was crucial in the creation of this section of the March of Wales.

T H E B A S K E RV I L L E S The Baskervilles were established in Herefordshire by 1086, when Robert de Baskerville was listed in Domesday as a major tenant of Roger de Lacy in Herefordshire,¹¹ holding ﬁve manors.¹² The Baskervilles participated in the conquest of the Welsh kingdom of Brycheiniog, and Robert’s probable son Roger de Baskerville granted the rent of one burgage to Brecon Priory on its foundation by Bernard de Neufmarché.¹³ In 1121 Roger de Baskerville’s knight service was transferred by Bernard to Miles of Gloucester upon his marriage to Bernard’s daughter Sibyl.¹⁴ Roger was still alive in 1127, when he witnessed a charter of Richard ﬁtz Pons, and by this time he may have also held land in Elfael from the Tonys and then the Cliffords.¹⁵ Roger’s son was Ralph I de Baskerville. Ralph ¹¹ Copleston-Crow, ‘The Baskervilles of Herefordshire’, 18. The Baskervilles probably originated in Bacqueville, south-east of Rouen, in the department of Eure (arrondissement of Les Andelys), or possibly Bacqueville-en-Caux, south-east of Dieppe in the department of Seine-Inferieure. The ﬁrst location is more likely, as neighbouring families in Normandy included the Scohies, Mucegroses, and Tosnys, all of whom had received lands in Herefordshire, probably from William ﬁtz Osbern. Robert held Brobury, Eardisley, Stretton Sugwas, Yarsop, and Yazor in Herefordshire, all of which, with the exception of Yarsop, were still in the hands of the family in 1243. Brobury, 2 hides worth 20s. in 1086, HDB 48; Eardisley, unspeciﬁed amount in 1086 and not valued for geld (hec terra non geldat neque consuetudinem dat) and not in any hundred, HDB 47; Stretton Sugwas, 2 1/2 hides worth 50s., with a mill valued at 32d ., HDB 43; Yarsop, 1 1/2 hides worth 15s., HDB 49; Yazor, 5 hides valued at 60s., HDB 49. For holdings in 1243, BF 802 (Brobury, Stretton Sugwas, Yazor). Curiously, Eardisley is not listed in the 1243 returns. The Baskervilles of course had added other lands by 1243. ¹² For a brief version of what follows, see Brock W. Holden, ‘The Making of the Middle March of Wales, 1066–1250’, WHR 20/2 (Dec. 2000), 216–17. ¹³ ‘Cart. Brecon’, 14. 141. ¹⁴ Ancient Charters, Royal and Private, 8, no. 6. ¹⁵ Ibid., no. 23; Copleston-Crow, ‘The Baskervilles of Herefordshire’, 19–21. It is also interesting to note another Baskerville–Clifford connection. A ﬁnal concord dated 1253 between Humphrey de Bohun and Eleanor, his wife, on one side and Walter III de Baskerville shows that the Baskervilles held some 4 carucates in the Clifford manor of Bronllys in Brecon; TNA, CP 25(1)/80/12, no. 235. (Walter III demised and quitclaimed the land in question in Bronllys to Bohun.) It is not possible to determine when this land was granted out, but it is tempting to speculate that it occurred during the unsettled

94

Lords of the Central Marches

continued the family’s connection with Miles of Gloucester and his sons, gaining the Archenﬁeld manors of Orcop, Treaddow, and Westﬁeld from the earls of Hereford.¹⁶ The Baskervilles also possessed the manor of Bredwardine by the early 1140s. Ralph died about 1148/9.¹⁷ Ralph de Baskerville had at least three sons, Ralph II, Robert, and Walter. Ralph II and Robert appear in 1149 in the afﬁnity of Miles of Gloucester’s son Roger, earl of Hereford.¹⁸ Crucially, it appears that a division of the family lands between Brecon and Herefordshire holdings was effected on Ralph I de Baskerville’s death; this trend will be seen again to some degree with other knightly families in the area. Ralph II had received those lands which lay within the honour of Brecon, namely the sub-lordship of Pencelli and the detached manor of Bredwardine, while Robert received Eardisley and other manors which were held of the (at the time) former Lacy honour of Weobley. Both men were alive at the time of the compilation of the Herefordshire Domesday Book, about 1160–70, and Robert was listed in the 1166 carta of Hugh de Lacy as owing the service of eight knights, making him the middle years of King John’s reign, when Walter I de Clifford sided with the king against William de Braose and the Lacys. It should be noted that Baskerville stood aloof from the conﬂict of 1208–10 in the March and Ireland. This connection to the Cliffords may do much to explain the actions of Walter de Baskerville in the political crisis of 1207–10; see Ch. 6. ¹⁶ King Stephen having granted all of Archenﬁeld to Miles of Gloucester in 1137; RRAN iii, no. 312; Copleston-Crow, ‘The Baskervilles of Herefordshire’, 21. ¹⁷ Bredwardine, a former Domesday manor of Alfred of Marlborough, possibly came to Roger de Baskerville from Bernard de Neufmarché, who had attached Alfred’s manors to his new lordship of Brecon; Copleston-Crow, ‘The Fief of Alfred of Marlborough’, 385–6. Ralph I de Baskerville granted a virgate of land above his park there to Dore Abbey at some point after 1141. Copleston-Crow’s statement (‘The Baskervilles of Herefordshire’, 22) that Ralph I de Baskerville gave land above his park in Bredwardine to Abbey Dore upon its foundation in 1141 is not borne out by the charter he cites, Mon. Ang. v. 555, no. 10. This charter is a conﬁrmation of the abbey’s holdings by Henry III, dated 15 Dec. 1232. There is no internal evidence in the charter as transcribed by Dugdale to indicate when the grant was made, nor whether the grantor was Ralph I (d. 1148/9), II (d. 1190/1), or III (d. 1210) de Baskerville. All that can be said is that the grant dates after 1141. ¹⁸ They were listed among Roger’s hostages in his conventio with William, earl of Gloucester; Davis, ‘Treaty between William Earl of Gloucester and Roger Earl of Hereford’, 144. Ralph II de Baskerville also appears as one of Roger’s witnesses in his conventio with William de Braose (1148 × 1154), and of Henry of Hereford’s grant of the churches of Hay and Llanigon to Brecon Priory; Z. N. Brooke and C. N. L. Brooke, ‘Conventio between Roger Earl of Hereford and William de Braose’, in Brooke and Brooke, ‘Hereford Cathedral Dignitaries in the Twelfth Century’, 185. The editor’s dating of this charter means that it is a possibility that Ralph I de Baskerville was the witness rather than his son Ralph II; ‘Cart. Brecon’, 14. 151.

The Knightly Class

95

foremost honorial baron of the honour of Weobley.¹⁹ A third brother, Walter, held the royal serjeanty of Orcop of his brother Robert, lord of Eardisley.²⁰ From this point forward, the Baskervilles are split into two lines, those of Pencelli and Bredwardine, and those of Eardisley. Ralph II de Baskerville, lord of Pencelli and Bredwardine, was murdered in either 1190 or 1191.²¹ He had married Nest, a Welshwoman, and had three sons: Ralph III, who succeeded to the lordship, Robert, who received lands in Pencelli, and Thomas, who received a portion in Bredwardine and also held lands in Shropshire and Northamptonshire.²² Two daughters, Matilda and Isabel, received marriage portions worth 100s. each in Bredwardine.²³ The Baskerville line in Pencelli and Bredwardine came to an end when Ralph III died in 1210, and his lands were divided among the husbands of his three daughters.²⁴ The Eardisley branch of the Baskervilles would last intact through the Middle Ages and beyond. Robert de Baskerville of Eardisley died in 1176, and had a son, Ralph, who died in 1186. Ralph in turn was succeeded by his son Walter I, who was a minor at the time of his father’s death.²⁵ Walter had come into his majority by 1194 at the latest, when he ﬁned not to cross with the king’s army to Normandy.²⁶ Walter I de Baskerville married ﬁrst, Emma (d. c.1195), who had been the wife of ¹⁹ Robert, HDB 43, 47–9; Ralph, HDB 59; RBE 281–3. The next nearest tenants in number of fees to Robert de Baskerville each held four (Roger Devereux, Nicholas de Hungerford, and William de Bakintone). ²⁰ Walter died without issue. From 1168 Walter paid 5s. a year to the exchequer for the issues of the wood of Orcop; Pipe Roll, 14 Henry II, 114 ff. ²¹ Abbreviatio Placitorum (Record Commission, 1811), 67; CRR i. 435. Thomas does not seem to have brought his appeal until 1201 and the result of the duel is unknown, though Thomas lived for quite some time. ²² Nest agreed to recognize the right of the prior and convent of Brecon to a mill and pool in Trosdref granted by Ralph de Baskerville viri sui; ‘Cart. Brecon’, 13. 306. Robert de Baskerville made a grant ante 1175 to Brecon Priory of the land of Tudor Cymerdruc (ibid. 14. 160), which was conﬁrmed by Ralph de Baskerville (ibid. 14. 162) (the presence of William de Burghill in the witness list of this charter places it at the end of the 12th century). Thomas’s lands; BF 146, 1284; CIPM ii, no. 595. ²³ BL, Additional Charters 20408, 20410. ²⁴ See p. 112 below. ²⁵ The sheriff of Hereford rendered account of the manor of Eardisley to the exchequer in 1188 for the previous two years and three months; Pipe Roll, 34 Henry II, 214. ²⁶ Pipe Roll, 6 Richard I, 139; the previous year he was amerced for novel disseisin in Gloucestershire; Pipe Roll, 5 Richard I, 120. The royal escheator William de SaintMère-église rendered an account in Michaelmas 1195 for the farm of Easton (?) in Gloucestershire, described as being ‘of Walter de Baskerville’, for a quarter-year; Pipe Roll, 7 Richard I, 55. This would seem to place the end of Walter’s minority in very late 1194.

96

Lords of the Central Marches

Osbert Longchamp.²⁷ He married secondly Isolda Pantolf, daughter of William Pantolf, at some point during or after 1208. Isolda had already married three times, ﬁrst to Hugh Munpicun (d. 1186), then Walter de Tateshal (d. 1199), and ﬁnally Henry Biset (d. 1208).²⁸ A woman with dower lands from three previous marriages, plus her inheritance, must have been attractive to a man like Walter de Baskerville, who was already by 1205 in debt to the Crown.²⁹ Walter I died in 1213.³⁰ His heir, Walter II, was also a minor, and in this case his wardship was purchased by Andrew de Chanceaux, a curialis, kinsman of Gerard d’Athée and Engelard de Cigogneé, and ‘Poitevin’ associate of the bishop of Winchester and (in 1214) justiciar Peter des Roches.³¹ Andrew promptly married Walter to his sister Susanna. By Michaelmas 1218 Walter II was accounting at the exchequer for his father’s Jewish debt, helped by his father-in-law’s inﬂuence, which gained him a pardon of 100 marks.³² Walter II became a man of some standing in Herefordshire. In 1225 he was a collector of the ﬁfteenth granted the king in that year.³³ His importance is also evidenced by his receipt of an individual writ ordering him to join the king’s army at Montgomery in 1228.³⁴ Walter served as sheriff of Herefordshire prior to 1232.³⁵ In 1233–4 Walter supported ²⁷ Her death involved Walter in a dispute over her lands with Emma’s son by Osbert, Geoffrey de Longchamp; Pipe Roll, 7 Richard I, 181. ²⁸ Scott L. Waugh, The Lordship of England: Royal Wardships and Marriages in English Society and Politics 1217–1327 (Princeton, 1988), 23 and n. 24. Isolda would go on to marry once more after Walter de Baskerville’s death in 1213, to Amaury de St Amand. She died about 1222/3. ²⁹ This debt was the result of a ﬁne of £100 and two palfreys made in order that Walter should hold his lands in peace if a suit should be brought against him concerning the death of Walter Basad; Pipe Roll, 5 John, 57. ³⁰ He accounted for Jewish debt at the exchequer at Michaelmas 1212; Pipe Roll, 14 John, 159. The pipe roll for the following year is missing, but by Michaelmas 1214 Isolda, now Walter’s widow, owed 100 marks and one palfrey for having both her inheritance and reasonable dower, and freedom from distraint of marriage; Pipe Roll, 16 John, 113. ³¹ Rot. Ob. et Fin. 512–13; Rot. Litt. Claus. i. 162; Pipe Roll, 16 John, 136. Andrew paid 50 marks and two palfreys; Nicholas Vincent, Peter des Roches: An Alien in English Politics 1205–1238 (Cambridge, 1996), 33 and n. 78. Although the circle around Peter des Roches was referred to (disparagingly) at the time as Poitevins, Vincent has conclusively shown that they almost all came from neighbouring Touraine. ³² Pipe Roll, 2 Henry III, 91. ³³ Rot. Litt. Claus. ii. 74b. ³⁴ CR 1227–31, 115. Only nine other men in the area received these writs, including the barons Gilbert de Lacy, Walter de Beauchamp, and William de Cantilupe. ³⁵ CR 1231–34, 174 (dated Hereford, 14 Dec.). The sheriff of Herefordshire was ordered to pay to Walter his arrears from the time when he had been sheriff.

The Knightly Class

97

the king against Richard Marshal.³⁶ By 1236 he had risen to the level of prominence where he could be appointed as one of the dictatores of the truce with Llywelyn in south Wales.³⁷ He served as a collector of the thirtieth of 1237 in Herefordshire, and was summoned with men of the stature of Walter de Clifford and John de Monmouth for service in the army in 1238, and was appointed to inspect the county’s castles in 1241.³⁸ It seems likely that his father-in-law, Andrew de Chanceaux, as an associate of Peter des Roches, helped to boost Walter’s role in local government, perhaps by bringing him to the attention of the court. Walter II died in 1244.³⁹ By the middle of the thirteenth century the Baskerville family had weathered debt and had risen to become one of the most important knightly families in the region. This was largely owing to the family’s connection by marriage to the curialis Andrew de Chanceaux, demonstrating the advantages to a knightly family of making connections to the court and securing royal patronage.

T H E D EV E R E U X The Devereux were another family which had settled in the region at an early date.⁴⁰ In Bernard de Neufmarché’s foundation charter for Brecon Priory is a Robert Devereux, who granted the rent of one burgage in the town of Brecon.⁴¹ The Devereux then disappear from the historical record until the reign of Henry II; if they were supporters of Miles of Gloucester and his sons, no Devereux appears as a witness to the large number of extant charters of the earls of Hereford. From lawsuits of the early thirteenth century some genealogical reconstruction is possible. In the middle of the twelfth century there was a William Devereux, who in turn had two sons, Roger and Walter.⁴² Hugh de Lacy’s 1166 carta shows Roger Devereux as owing ³⁶ CR 1234–37, 450. He was also ordered to free a Welshman taken captive during the revolt. ³⁷ CR 1234–37, 369. Gilbert Marshal found Walter suspect, so Walter’s role was curtailed; ibid. 370. ³⁸ CR 1234–37, 551; CR 1237–42, 124, 346. ³⁹ Rotulorum Originalium in Curia Scaccarii Abbreviatio, ed. Henry Playford, 2 vols (Record Commission, 1805–10), i. 6, homage of Walter (III) de Baskerville for lands and tenements held in capite. ⁴⁰ A version of this section appeared in Holden, ‘The Making of the Middle March of Wales’. ⁴¹ ‘Cart. Brecon’, 14. 141. ⁴² CRR iv. 230 (claim of Cecily Devereux through William, her avus); CRR ii. 33 (claim of Cecily Devereux through Roger, her uncle). The only logical explanation is

98

Lords of the Central Marches

the service of four knights’ fees, while Walter Devereux owed the service of three knights’ fees.⁴³ Walter Devereux held Lyonshall, the Devereux caput in the thirteenth century, of Hugh de Lacy. Walter Devereux probably held the family lands in Brecon as well. The lack of marginal annotations in Herefordshire Domesday means that it is not possible to account for Roger Devereux.⁴⁴ The family disappears from royal records until 1188, when the land and two castles of John Devereux were in the hand of the sheriff of Hereford on John’s death.⁴⁵ The reference to the two castles in custody makes it likely that Lyonshall is referred to and that John was the son and heir of Walter Devereux.⁴⁶ John Devereux was succeeded as lord of Lyonshall by his son Stephen, who would increase substantially the fortunes of his family; his other son, Walter, would seem to have received the family lands in Brecon. Before 1197 Stephen had married Constance de Longchamp.⁴⁷ Constance died at an unknown date and Stephen subsequently married Isabelle de Cantilupe, who was very likely related to the curialis William I de Cantilupe.⁴⁸ By this marriage Stephen, much like Walter II de Baskerville, formed a link to the court.⁴⁹ At the same time that Stephen Devereux married into the Cantilupe family, he entered (if he had not already done so) into the familia of William Marshal. It is uncertain just how Stephen joined the Marshal’s that Cecily was a daughter of Walter Devereux and sister of John. In CRR v. 15 William Devereux is described as having seisin of Elstones Bridge of right in the time of Henry II. ⁴³ RBE 281–3. ⁴⁴ HDB 46. ⁴⁵ Pipe Roll, 34 Henry II, 214. ⁴⁶ Cathcart King is of this opinion; Castellarium Anglicanum, i. 209. There was deﬁnitely a castle at Lyonshall in 1209; Rot. Litt. Pat. 91. ⁴⁷ Feet of Fines 7 & 8 Richard I, no. 117 (22 Apr. 1197); David Balfour, ‘The Origins of the Longchamp Family’, Medieval Prosopography, 18 (1997), 91. The Longchamps were a prominent knightly family in Herefordshire and Gloucestershire during the reign of Richard I, and it was likely through this marriage that Stephen gained the manor of Frome Herbert (Halmond’s Frome?), granted to him by Stephen de Longchamp, his uncle; HRO, Hopton Collection, no. 107 (conﬁrmation by Walter de Lacy); Rot. Chart. 156 (cancelled conﬁrmation of King John, dated 26 July 1205). Frome had originally been granted to Stephen de Longchamp by Walter de Lacy and was in his possession as late as Michaelmas 1201; Pipe Roll, 3 John, 226. This manor, as acquisitions so often did, eventually went to form part of the dower of Stephen Devereux’s second and last wife, Isabelle de Cantilupe. ⁴⁸ Crouch, William Marshal, 225, speculates that Isabelle was a sister of William I de Cantilupe. ⁴⁹ William de Cantilupe had been a steward of the household of King John when John was still count of Mortain, and had since served as sheriff of Worcestershire from 1200 to 1204, as well as undersheriff of Hereford in 1204.

The Knightly Class

99

household, but there are four known instances on which he attested Marshal acta.⁵⁰ The Histoire de Guillaume le Maréchal tells of Stephen’s role in the crisis of 1207 between the king and the Marshal. When the Marshal was recalled from Ireland to England in that year, Stephen Devereux remained behind as one of his bailiffs. Stephen’s subsequent deﬁance of the king’s orders would seem to indicate that he either held land in Ireland, or expected to be rewarded by the Marshal.⁵¹ Stephen Devereux’s key place in the Marshal afﬁnity in the early 1200s and his position as one of the more important knights of the Lacy honour of Weobley probably served as an additional link between the Marshal and the Lacy brothers during the crisis of 1208–10. Stephen Devereux re-emerges in the records in 1211.⁵² At this time King John was trying to cultivate the Marchers, particularly Walter de Lacy, whom he had alienated or exiled during the crisis of 1208–10, in an effort to ﬁrm up his support in a region which had previously been governed by his mercenary commanders following the Braose forfeiture of 1208. Stephen Devereux was one of the targets of his blandishments. On 12 June 1214 Stephen was quit of scutage on two fees, one held of the bishop of Worcester at Crowle and another held of Peter des Roches, bishop of Winchester.⁵³ On 29 July the royal forester Hugh de Neville was ordered to have measured 40 acres at his manor of Crowle, in the royal forest of Feckenham, for assarting in accord with the licence the king had granted to Stephen.⁵⁴ Both of these writs were issued from France, where John was engaged in the southern part of his campaign to regain the lost Angevin dominions. An unclear entry on the dorse of the close rolls from just around this time, which appears to be either a scutage list or perhaps a prest or liberate roll, when combined with this patronage issuing from France, suggests that Stephen likely accompanied the king overseas. By 1215 Stephen Devereux was described in a royal writ as ‘our dear and faithful’.⁵⁵ During the civil war Stephen not surprisingly was a supporter of the king and as such was rewarded with forfeited rebel lands, both before ⁵⁰ Crouch, William Marshal, 225. ⁵¹ For the above, ibid. 101–6. The Marshal made a number of grants to his loyal followers in Ireland, including Stephen, upon his return to Ireland in 1208. ⁵² Stephen was pardoned 4 marks scutage per breve Regis in Michaelmas 1211; Pipe Roll, 13 John, 234. ⁵³ Rot. Litt. Claus. i. 167. ⁵⁴ Ibid. 168b. ⁵⁵ Ibid. 219; Rot. Litt. Pat. 147b. This writ, dated 4 July, was a quittance of an annual render of 32 gallons of honey from the men of Stephen’s manor of Ballingham to Hereford Castle.

100

Lords of the Central Marches

and after the king’s death.⁵⁶ The regent, Stephen’s old lord William Marshal, was making sure to keep Stephen on side. Stephen continued to climb in local standing during the minority of Henry III. In 1219 he was appointed a forest commissioner for the eyre in Hereford, and the following year he was made a justice of gaol delivery for Hereford.⁵⁷ By 1224 Stephen had enough standing to point out to the king’s government that the 40 acres of assart granted him at Crowle were to be placed outside the regard, and they duly were for three years from Michaelmas.⁵⁸ In 1225 he helped escort the collected ﬁfteenth for that year from Hereford to Gloucester, and in 1227 Stephen and his heirs were granted, in perpetuity, a weekly market and yearly fair at Lyonshall, the family caput.⁵⁹ The following year, 1228, Stephen Devereux died. The 1220s marked the pinnacle of Stephen Devereux’s career. The ﬂow of royal favour, which began in the previous decade, continued. Royal commissions as justice and escort for the ﬁfteenth of 1225 served to underline his signiﬁcance in the locality. He was a much-courted man in this period. Gilbert de Lacy, the son of Walter de Lacy, who by the 1220s was effectively running his father’s Marcher and Hereford holdings while his father was in Ireland, attempted to attach Stephen more closely to the Lacy family. Gilbert granted to Stephen 12 virgates of land in the manor of Staunton-on-Wye, pro homagio et servitio suo et pro eo quod sit in familia mea.⁶⁰ It was also during the 1220s that Stephen embarked on an impressive programme of construction at Lyonshall Castle.⁶¹ This construction is similar to that carried out by the Lacys at Ewyas Lacy (Longtown), and one writer has suggested a date of c.1227 for the work at Lyonshall.⁶² This suggested date coincides ⁵⁶ On 27 Jan. 1216 he received all the land of Walter de Stokes in Ballingham and Clehonger in Herefordshire, and on 30 July he was granted £20 land with appurtenances in Stanton, Worcestershire, which had belonged to Peter of Stanton; Rot. Litt. Claus. i. 246, 279. Five weeks after King John’s death on 19 October, Stephen was granted the lands of the rebel Peter of Welles at Rotherwas, which was of Stephen’s fee, if the lands should remain in royal hands for a year and a day; ibid. 294. ⁵⁷ Rot. Litt. Pat. 215, 437. ⁵⁸ Ibid. 648. ⁵⁹ Rot. Litt. Claus. ii. 74b, 189b. ⁶⁰ HCA, no. 3235 (a 14th-century copy of a Lacy deed roll). The lack of a witness list on this copy makes dating very difﬁcult. Staunton-on-Wye was a Pichard manor held of the honour of Weobley (BF 802), but the grant by Gilbert de Lacy to Stephen Devereux shows that the Lacys had held residual demesne in the manor. Also in 1242 Matthew de Mans held land in the same manor of Walter de Clifford; ibid. ⁶¹ Cathcart King, Castellarium Anglicanum, i. 209. ⁶² Joe Hillaby, ‘Hereford Gold: Irish, Welsh and English Land. Part II. The Clients of the Jewish Community at Hereford 1179–1253: Four Case Studies’, Trans. Woolhope, 45/1 (1985), 226.

The Knightly Class

101

with the grant of market and fair discussed above, and demonstrates the conscious development of the Devereux caput. It was most probably also in the 1220s that Stephen Devereux began making large grants to the canons of Wormsley Priory, which at the time was located in King’s Pyon, some 7 1/2 miles to the south-east of Lyonshall. Stephen granted his chapel at Lyonshall, mills at Lyonshall and Halmond’s Frome as well as the rents due from his mills in Hereford, pasture in his manor of Cheddrehole, and gifts of grain and 10s. per annum from his chamber for vestments.⁶³ William Devereux was a minor on his father’s death in 1228, and remained so until c.1236. He was in the custody of Gilbert de Lacy, now lord of Weobley, until Gilbert’s death in 1230.⁶⁴ In 1240 he conﬁrmed his father’s grants to Wormsley Priory.⁶⁵ But William then attempted to regain parts of his patrimony that had been granted away by his father.⁶⁶ Stephen Devereux had held six manors in Herefordshire, Shropshire, and Gloucestershire, and the manor of Wilby in Norfolk, granted him by William II Marshal.⁶⁷ By 1243 William Devereux had lost three of the Herefordshire manors (though other land was gained).⁶⁸ William’s resources were diminished, and he seems to have rapidly got himself into debt. William continued the tradition of royal service begun by his father Stephen. In 1254 he was with the king in Gascony.⁶⁹ This service would help to account for the occasional signs of royal favour which he ⁶³ These grants are to be found in BL, Harleian MS 3586 (Cartulary Wormsley), fos 81–90v . Some are printed in Mon. Ang. vi, pars i, 399–401, 403. Stephen’s grants are discussed more fully below in the section on knightly patronage of the Church. A promise and concession by Prior Ralph of Wormsley not to alienate or sell any of the lands or possessions which they held of the gift of Stephen Devereux without his assent, dated 20 Feb. 1223, indicates that quite a proportion of Stephen’s muniﬁcence dates from before this; BL, Harleian MS 3586, fo. 84v . Stephen had his grants to Wormsley conﬁrmed by both Walter and Gilbert de Lacy; ibid., fo. 89r (Walter); Mon. Ang. vi, pars I, 400 (Gilbert). As was so often the case with heirs, Stephen’s gifts to Wormsley were to be challenged by his son William Devereux, and it should be noted that Stephen did not associate his son in any of the grants which he made; see below, pp. 126–7. ⁶⁴ CRR xiii. 343, no. 1622 (Hilary 1229). Gilbert de Lacy was summoned to warrant in William’s place in a dispute between the prior of Pyon (Wormsley) and a tenant over 10s. rent in Holme Lacy. Gilbert conﬁrmed Stephen Devereux’s grants to Wormsley Priory, perhaps while guardian; Mon. Ang. vi, pars I, 400. ⁶⁵ Mon. Ang. vi, pars I, 400. ⁶⁶ For further detail, see the section on litigation below, pp. 126–7. ⁶⁷ This grant is dated by Crouch to between 1219 and 1227; William Marshal, 225. ⁶⁸ BF 805, 808. Nicholas Devereux, son of Cecily Devereux, held one (Chanstone), with the remaining two (Maund Albin, Halmond’s Frome) held in dower by William’s mother, Isabella. ⁶⁹ CR 1253–54, 230 (1 Apr.).

102

Lords of the Central Marches

received: the restoration of Wilby, a gift of ten oaks from the hay of Hereford in 1245, and a grant of free warren in his demesne lands in Gloucestershire and Herefordshire in 1251.⁷⁰ But this was not enough to keep his debts at bay and he was eventually forced to sell his lands in Stoke Lacy and Lawton to Roger de Mortimer.⁷¹ This is the situation William Devereux found himself in by the 1260s; it is little surprise to ﬁnd him in the ranks of the Montfortians by the time of Evesham.⁷²

T H E PI C H A R D S The Pichards also participated in the conquest of Brecon by Bernard de Neufmarché. The line’s eponymous founder, ‘Pichard’, was still alive in April–May 1121.⁷³ Pichard’s son was named Roger, and he was alive in the mid-twelfth century.⁷⁴ Roger Pichard held lands of the honours of both Weobley and Brecon. He had two sons, John (alive 1225) and Miles (d. c.1232), and this is where reconstruction of the family line becomes increasingly difﬁcult.⁷⁵ John Pichard was listed in 1210 as holding three fees of the honour of Brecon.⁷⁶ Miles seems to have received lands in Herefordshire, including the manors of Marston, Staunton-on-Wye, and Burleton, as well as some lands in Worcestershire. In the early 1200s the Pichards appear to have been caught up in the feud between their lord William de Braose and King John, or at least to have fallen under royal suspicion, as a writ of 1209 shows Miles Pichard receiving the wardship of Miles, the son of his brother John, along with the custody of the Devereux castle at Lyonshall. The young Miles was to be turned over to the king’s custody on demand.⁷⁷ By 1214 John Pichard himself was in the custody of the royal servant Henry ⁷⁰ CR 1242–47, 338; CCR i. 369. ⁷¹ BL, Harleian MS 1240, fo. 48. William owed Roger de Mortimer 1,000 marks in 1264; ibid., fo. 49. This could possibly have been a recognizance, a point I owe to Rees Davies. ⁷² See Ch. 8. ⁷³ When his knight service was transferred to Miles of Gloucester, who married Bernard’s daughter; Ancient Charters, Royal and Private, no. 6. ⁷⁴ Roger ﬁtz Pichard granted two parts of all his tithes to Brecon Priory; ‘Cart. Brecon’, 14. 167. ⁷⁵ See the Pichard genealogy in Appendix 2. ⁷⁶ RBE 601. He also witnessed a charter of Walter de Traveley in Brecon around the end of the 12th century; ‘Cart. Brecon’, 14. 34. ⁷⁷ Rot. Litt. Pat. 91. Interestingly, this is the enrolment of Miles Pichard’s own letter patent.

The Knightly Class

103

Braybroke.⁷⁸ It is highly likely that John Pichard had supported William de Braose in either Wales or Ireland, and had been captured by royal forces c.1210. John was still alive in 1225, when along with Walter II de Baskerville and Stephen Devereux he conveyed the ﬁfteenth of that year collected in Herefordshire to Gloucester.⁷⁹ John Pichard had two sons, Roger (d. 1247/8) and Miles (d. 1253), by his wife, Hawise. Roger would seem to have been the elder, receiving the family lands in Ystradyw (Tretower), although he still held family manors in Herefordshire as well. His brother Miles may probably be identiﬁed with the Miles ‘of Skethrog’ who died by 1253 and was succeeded by his son John.⁸⁰ Roger Pichard managed to survive the rebellion of Richard Marshal in 1233–4, when he was for a time mistakenly identiﬁed as being among the Marshal’s followers, while in 1237 Roger was a collector of the thirtieth in Herefordshire, and in May 1242 was appointed one of two knights in Herefordshire to judge and assess arms and keep the peace.⁸¹ Roger was dead by early 1248, and his son and heir was also named Roger.⁸² What is interesting about the Pichards is the disposition over time of the family lands. Pichard, the family founder, would seem to have held manors in Herefordshire of the Lacys, and then to have added to them in the conquest of Brycheiniog. By the mid-thirteenth century, the Pichards still had lands spanning the boundary between England and Wales, but their interests were much more ﬁrmly in Brecon than Herefordshire. Roger Pichard had land in nine manors in Herefordshire, but out of those nine, six had been subinfeudated at some point in the past.⁸³ Beyond this, in the course of the thirteenth century, the vestigial rights ⁷⁸ A ﬁne of three palfreys was made to free him; ibid. 125b. ⁷⁹ Rot. Litt. Claus. ii. 74b. ⁸⁰ CR 1253–54, 186 (letters of protection for ‘John Pichard, son and heir of Miles Pichard’ dated 1 Nov. 1253); TNA, CP 25(1)/80/12, no. 242 (‘John Pichard son of Miles Pichard of Skethrog’ called to warrant for land in Presteigne, 1255). Miles was in Gascony with the king in 1253; CPR 1247–58, 231. A writ was issued in March 1254 to the bailiff of Humphrey de Bohun junior, lord of Brecon, informing him of Miles Pichard’s recent death in Gascony, and instructing him to assign dower to Miles’s wife, Alice; CR 1253–54, 41–2. ⁸¹ Roger supposedly with the Marshal, CR 1231–34, 328; collector of the thirtieth, CR 1234–37, 551; peacekeeping commission, CR 1237–42, 484. ⁸² CR 1247–51, 52. ⁸³ Milton (one-ﬁfth fee subinfeudated to John de Sancto Albino), Weston Bret (2 hides, Matthew le Bret), Staunton-on-Wye (one fee consisting of 4 hides, Miles Pichard), Maund Albini (2 hides, Isabelle Devereux), Stanford Bishop (one-sixth of a fee consisting of 1 hide, Walter of Stanford), and Upton (1 1/2 hides, Thomas of Upton) were all subinfeudated. Roger Pichard kept control of Almeley, Ocle Pichard, and Thruxton. BF

104

Lords of the Central Marches

which Roger Pichard had in those Herefordshire manors were either granted or sold away.⁸⁴ By this time the centre of the Pichard holdings was in Ystradwy (now part of the ﬁtz Herbert honour of Blaenllyfni); nothing demonstrates this more clearly than the construction of a new masonry castle at Tretower, and the corresponding lack of new building at the Pichard manors of Almeley or Ocle Pichard in Herefordshire. Although the division of lands along the border was not as clear-cut in the case of the Pichards as in those of the Baskervilles or Devereux, a division, reﬂected in landholding and family priorities, had nonetheless largely taken place over time.

T H E L E S S E R K N I G H TS The knightly families we have discussed so far were prominent and wealthy ones; indeed, it is obvious that someone like Stephen Devereux was approaching the lower ranks of the baronage. The majority of ‘knights’, or at any rate those who held by military tenure, were not so grand. The nature of the evidence, however, makes it difﬁcult to offer more than generalizations about the fortunes of ‘lesser’ knightly families in the region. The best studies of knightly families further down the social scale still essentially construct family histories in order to view their situation ‘from the inside out’, and the best such efforts draw on rich cartulary evidence.⁸⁵ For Herefordshire, and even more so in the March, the historian is hampered if not crippled by the lack of evidence. Rarely can genealogies, and thus the patterns of inheritance and the descent of land, be reconstructed with anything resembling certainty; anyone who has perused V. H. Galbraith’s heroic attempts in the notes 799, 801, 802, 805, 806, 810 (subinfeudated manors); 803, 807, 812 (Almeley, Ocle Pichard, Thruxton). ⁸⁴ About 1280/90 John de St Aubin, lord of Milton in Herefordshire, granted his holding there to Reading Abbey; Reading Abbey Cartularies, ed. B. R. Kemp, Camden Society, 4th ser., 31 (1986), 284–5, no. 350. John’s grandfather had held one-ﬁfth fee in Milton of Roger Pichard in 1243; BF 799. Roger Pichard of ‘Ystradwy’ then conﬁrmed John’s grant for 20 marks of silver and 1/2 mark annual rent for all secular demands, suit of court, and royal services. He later quitclaimed to the abbey the rent; Reading Abbey Cartularies, 286, nos 352, 353. ⁸⁵ An excellent example of this approach is E. King, ‘Large and Small Landowners in Thirteenth-Century England: The Case of Peterborough Abbey’, P&P 47 (1970), 26–50. Note that King is able to draw upon the records of an abbey that was highly active in the 13th-century land market.

The Knightly Class

105

to the Herefordshire Domesday Book will appreciate the nature of the task. The lack of large religious houses in the region active on the land market in the thirteenth century, and the complete lack of family ‘estate books’, means that we are overly dependent upon the picture which the feodaries give us of landholding in the border counties and, to an even lesser extent, the March. The chief of these lists of fees are those collected for the Cartae Baronum of 1166, and those compiled in the course of 1242–3 for scutage for Henry III’s Gascon campaign. Even then, attempts to supplement the bare outlines of the feodaries of 1166 and especially 1242 meet with little success: the smaller knights of our region have left few footprints in the records, and the fact that the records are the product of bureaucracy lend an order and neatness that may not have been there in reality. As a consequence we are forced to examine smaller landholders ‘from the outside in’, as it were. We must make what limited observations and generalizations we can. As mentioned previously, the knight’s fee is certainly an imperfect measurement of social and economic standing. Further, by the thirteenth century, knight service had become somewhat divorced from knights’ fees, as measured in government inquests. The fact that in the border counties, and particularly for Marcher lordships which had manors in Herefordshire and Shropshire, knight service was still maintained as a living obligation does mean that some semblance of reality still lay beneath the assessments for scutage, and therefore the feodaries are still of some use. Making allowances for the fact that assessments, as recorded in the returns of 1242–3, were often negotiated down in the course of the thirteenth century, what is apparent is that many knightly holdings became both smaller and fragmented. The Lacy honour of Weobley demonstrates this. Making the reasonable assumption that the amount of land that composed the honour remained approximately the same between 1166 and 1242, what immediately stands out is the increase in the number of tenancies by the mid-thirteenth century. In the 1166 carta (which is somewhat misleading, as all the knights’ service of the honour, including manors in Shropshire and Gloucestershire, is listed under Herefordshire) there were twenty-four tenants holding of the honour by military service, and no fees are reported as shared.⁸⁶ By 1242 there were forty-ﬁve tenancies in Herefordshire, and eight of them were shared.⁸⁷ The number of people holding land by knight’s service had increased signiﬁcantly by 1242, and ⁸⁶ RBE 281–3.

⁸⁷ BF 816–18.

106

Lords of the Central Marches

this in turn undoubtedly points to the fact that knightly patrimonies had fragmented over time, with marriage being the most likely mechanism for the multiplication of tenures. More knights (ﬁfteen) held a single fee than any other category, but a signiﬁcant number (eleven) held half-fees, with eight holding quarter fees and nine holding ﬁfths of a fee. Also notable are the new names in the lists of 1242. Among lesser landholders of 1166 (those who held one fee), Walter del Mans may be represented in 1242 by Matthew de Mans, who by that time was a tenant of the earl of Hereford and the Cliffords rather than of Weobley; Philip of Sarnesford’s descendant, another Philip, seems to be holding steady with half a fee in Sarnesford (all that was admitted by his ancestor) and one-third of a fee in Swanston. Yet who are William de Caple, William de Cardiff, and Elias the clerk, Richard the cook, and their ‘partners’? Robert of Weston held 1 hide in Weston Beggard in 1242; in 1166 the manor had been Lacy demesne. All these men, and more, hold land by knight service of Weobley in 1242, and they are newcomers compared to families such as the Baskervilles, Devereux, Pichards, and Gamages. Indeed, those on the bottom rungs, in terms of the amount of land they held, are in no sense ‘knightly’ at all by the mid-thirteenth century. Elsewhere, knightly holdings seem rather better preserved as coherent units in the Mortimer honour of Wigmore, which spanned the boundary between Herefordshire and Shropshire, and increasingly took on the character of a Marcher lordship in the course of the thirteenth century. Interestingly, no carta was returned for Wigmore in 1166, so we are denied a point of reference. In 1210–12 the part of the honour in Herefordshire was composed of twenty-three fees.⁸⁸ By 1242 Mortimer’s honour in Herefordshire was assessed at twelve and a half and onetwentieth fees; ten fees were whole (though two fees were shared), while a further ﬁve were fractions.⁸⁹ In Shropshire subinfeudation had also resulted in the creation of fractional fees: of the total of eighteen, a half, a quarter, and one-sixth, fourteen were single fees with the remainder being fractional.⁹⁰ As for the baronies of Monmouth and Clifford, they are much more difﬁcult to comment upon, as the number of their fees was both small and partly lay in the March, placing them beyond the scope of the 1242 inquest. The problem, of course, is to what degree the assessments of 1242 indicate an actual decline in the amount of land held, or a reduction in the amount of knight service assessed on the land. The latter ⁸⁸ RBE 497.

⁸⁹ BF 814.

⁹⁰ BF 963.

The Knightly Class

107

undoubtedly was often the case, to judge by the fall in fees from 1210 to 1242. What is still important, though, is to what degree fees were divided and tenancies multiplied; this too had occurred in the honours of the region. The knightly desire to make family settlements, and quite possibly sales or subinfeudations, had resulted in some knights holding less land than would ostensibly support their original knight service. The military exigencies of the border, however, made it important in the eyes of lords, and thus in those of their tenants, to maintain what has been termed the ‘basic fee’—a tenancy of sufﬁcient size to enable the tenant to fulﬁl his obligations.⁹¹ In the honour of Weobley the near doubling of tenancies by 1242 points to a signiﬁcant whittling away of original patrimonies. Undoubtedly the Mortimers, on the other hand, had made vigorous attempts to police the break-up of knightly holdings, but this was an area in which the interests of lords and of knights might not coincide. In the border counties and the March, lords likely were active behind the scenes to prevent alienation and subinfeudation from getting out of hand. David Carpenter has pointed to the close watch kept on the process of subinfeudation, and on tenants in general, both by lay barons whose activities are recorded in the inquest of 1236 and by the contemporary Ferrer earls of Derby.⁹² The Mortimers would seem to have been rather more successful at this than were the Lacys. Some manors and thus knights’ fees in Herefordshire seem to have been based upon the standard 5 hides of land, though there are numerous exceptions, with the range generally running from 3 to 6 hides.⁹³ Examples of the 5-hide unit drawn from the demesne manors of the Lacy family in 1086 are the manors of Holme Lacy, Mansell Lacy, King’s Pyon, Stoke Lacy (10 hides), Weston Beggard, and Castle Frome. This is borne out by those places in the 1242 inquest for Herefordshire where both the knight’s fee assessment and hidage were given. With a notional 120 acres to the hide, the average military tenant of Weobley in 1242 held between approximately 300 and 600 acres. On the face of it, knightly holdings would seem to have declined in many instances. In ⁹¹ King, ‘Large and Small Landowners in Thirteenth-Century England’, 45. ⁹² Carpenter, ‘The Second Century of English Feudalism’, 38, 49–50. ⁹³ J. H. Round, in his Introduction to The Victoria History of Herefordshire, i, ed. William Page (1908), 270, stated that ‘The ancient system of assessment, based on the ﬁve-hide unit, of which Radnor is an instance, is fairly well illustrated in our county . . . ’, although he admitted that more work needed to be done. Darby and Terrett (eds), The Domesday Geography of Midland England, 66, while admitting that Round provides good examples, point out that the 5-hide unit ‘is not readily apparent in the majority of entries’.

108

Lords of the Central Marches

the honour of Weobley, Godfrey de Gamages in 1166 held 8 hides in Mansell Gamage assessed as two knights’ fees; in 1242 his descendant Godfrey owed service from one fee. Yet because we do not know how many hides Godfrey held in 1242, and because of the nature of the inquest, we cannot be certain if this marks a diminution in landholding or simply a reduction in service owed on the original holding. In other instances, the decrease in a family’s holdings is more evident. Several landholders of Weobley in 1242 obviously held subinfeudated land, and small parcels at that: Roger de Bodenham at Bodenham Rogeri, who held one-tenth of a fee of Walter de Baskerville as his mediate lord, or Walter of Brobury, who held one-ﬁfth of a fee in Brobury, likewise of Baskerville. These are tenants who had been created by subinfeudation, and the small service assessments probably accurately reﬂect the small size of their actual holdings. The large group that has been termed ‘lesser knights’ was important to the structure of society and power in the region, despite their not having left as much of a trail in the records as their betters. In so far as these families were able to maintain the ‘basic fee’, they furnished a substantial part of their lord’s military strength in what was a militarized and often violent frontier zone. As will be seen later, some of them in Herefordshire were chosen to perform the tasks of royal government for which the Angevins relied on the knights.⁹⁴ And yet these families were faced with many of the same pressures which confronted the more prominent knightly families of the region, such as the Baskervilles and the Devereux. If the ambitions of a Walter of Brobury were more constrained than those of Stephen Devereux, so were his resources. The pressures on knightly estates, both large and small, are what we must next examine.

P RO B L E M S FAC E D B Y T H E K N I G H TS The preservation of the patrimony and, if possible, its increase was the chief concern of the knights, as it was of the magnates. By the late twelfth century the relationship between lord and tenant revolved around land. The possession of land conferred wealth, power, and status. Lordship may have been in origin the bond of lord to man before that ⁹⁴ See Ch. 5.

The Knightly Class

109

of lord to land, and in the March this may still have been more true than in England, as is witnessed by the sway of the earls of Hereford at mid-century, who, albeit originally in the context of a civil war, attracted into their afﬁnity many local men who were not their tenants.⁹⁵ But personal lordship depended on the bestowal of rewards for good service, and no reward was more keenly sought than land. The problem along the March and in Herefordshire by 1166 was that the frontier to the west, which had promised fresh land for the taking earlier in the century, had been effectively closed off. A combination of Henry II’s decision to discourage private war along the March and the emergence of the power of the Lord Rhys had led to stalemate.⁹⁶ Thus, in 1166 Hugh de Lacy was supporting eight household knights, seven of whom had each received grants of only 100s. of land from Hugh’s demesne.⁹⁷ Only ﬁve and a half Lacy fees had been created since the death of Henry I.⁹⁸ As the demesnes of the great lords of the region shrank, so did their capacity to reward their men, old or new, with grants of land. This became a particular problem for the knights as the twelfth century came to a close.

ECONOMIC PRESSURES The knights of Herefordshire and the March were subjected to a number of pressures during this period. Most of these were neither new nor peculiar to this region; the need to make family settlements, and to make religious bequests and endowments, were a concomitant of knightly and indeed all aristocratic life in the Middle Ages. But there were two additional pressures which also bore upon knights c.1200, one general, one speciﬁc to Herefordshire and the March. The ﬁrst was inﬂation. The period 1180 to 1220 has been termed the ‘great English inﬂation’.⁹⁹ Prices rose enormously during this period; for ⁹⁵ See Ch. 2. For a comment on the persistence of the older character of lordship in the March, see Davies, Lordship and Society in the March of Wales, 130–48. ⁹⁶ For the orthodox interpretation of Henry II’s détente with the Welsh, see Warren, Henry II, 166–9, and Davies, The Age of Conquest, 290–2. For a recent dissenting opinion, see John Gillingham, ‘Henry II, Richard I and the Lord Rhys’, Peritia, 10 (1996), 225–36, repr. in Gillingham, The English in the Twelfth Century: Imperialism, National Identity and Political Values (Woodbridge, 2000), 59–68. ⁹⁷ With the exception of one Bartholomew, who held 120s. ⁹⁸ RBE 281–3. ⁹⁹ Some key articles are Miller, ‘England in the Twelfth and Thirteenth Centuries’; Harvey, ‘The English Inﬂation of 1180–1220’ and id., ‘The Pipe Rolls and the

110

Lords of the Central Marches

example, the daily wage paid a knight in this period increased threefold. The proﬁts of agriculture also increased markedly. But a lord could only tap this proﬁt if he had direct control of his land. If land was either held at farm or had been subinfeudated, the proﬁts of the soil accrued to the farmer or subtenant, not the mesne lord. It was in the lord’s interest to bring as much land as possible under demesne cultivation and, although challenged, this is still the orthodox view of aristocratic economic activity in the early thirteenth century.¹⁰⁰ While material only exists partially (some of the Herefordshire returns of the 1242 inquest are hidated, and subtenancies are denoted) to show how much land our families held in demesne as opposed to that which had been granted or farmed out, certain sources, particularly the records of suits in the curia regis and ﬁnal concords agreed either at Westminster or in the county court, show that these families were trying both to regain old land and sources of income and to augment their holdings with new acquisitions.¹⁰¹ This trend was not peculiar to the region. But the knightly families of western Herefordshire and the March had another pressure on their lands and incomes which families elsewhere in England did not have to deal with, and that was the expense of living along a frontier, in what was often a war zone. Because of the military character of the region’s society, there was little chance for knights to become landlords and farmers solely; they were still milites strenui.¹⁰² It is in this context that knightly expenditure on arms and horses, and, especially in the case of the more signiﬁcant knights, fortiﬁcations, must be set.

I N H E R I TA N C E A N D FA M I LY S E T T L E M E N TS The foremost pressure on knightly land and wealth was the need to make family settlements. The rise of primogeniture aside, in England it was customary to provide for younger children to some degree from the patrimony, or preferably acquisitions. This system of inheritance Adoption of Demesne Farming in England’; and Bolton, ‘The English Economy in the Early Thirteenth Century’. ¹⁰⁰ The ﬁrst statement of this view was Postan (ed.), The Cambridge Economic History of Europe, i. 590–5. ¹⁰¹ The almost complete lack of the Hundred Rolls for Herefordshire is to be much regretted; they form the backbone of several studies of other parts of England for this period, e.g. Carpenter, ‘Was there a Crisis of the Knightly Class?’ ¹⁰² See the section ‘Military Lordship’ in Ch. 3.

The Knightly Class

111

has been described as ‘diverging devolution’.¹⁰³ This was the dominant pattern in Herefordshire and the March. Although the genealogies are quite difﬁcult to trace, it is apparent from family names that some younger sons or cousins were left unprovided for, and thus, for example, went to Ireland in search of opportunity and reward.¹⁰⁴ But what is also apparent is the degree to which some form of ‘diverging devolution’ operated in the area in the late twelfth and early thirteenth centuries. One good example is that of the Baskerville family.¹⁰⁵ As late as the middle of the twelfth century the Baskerville estate in both Herefordshire and the March was in the hands of one man, Ralph I de Baskerville (d. 1148/9). But then, as outlined above, the estate was broken up, splitting along lines of tenure. Eardisley and the other manors held of the honour of Weobley went to Robert de Baskerville (d. 1176), Orcop to Walter (d.s.p.), and Bredwardine and Pencelli to Ralph II (d. 1190/1). This division still left two sons, Roger and Gilbert, and perhaps a third unnamed brother, without lands. But, a major holding had been divided to provide for a number of sons. Such a division was typical of twelfth- and thirteenth-century English practice. The difference along the border and in the March was the way in which knightly estates were divided. This division along tenurial lines likely was done in order to keep lines of military service clear-cut, and probably with the counsel and encouragement of Ralph’s lords.¹⁰⁶ This manner of division is also suggestive of the fact that men thought along tenurial and thus honorial lines. It was this process, which began around the middle of the twelfth century and can be seen not only in the case of the Baskerville estates but also in that of the Devereux (and also to a lesser degree with the Pichards), which did much to ensure the honorial integrity of the Marcher lordships. It was one of the key contributors to the making of the March. Fragmentation of the Baskerville patrimony did not stop there. Ralph II de Baskerville of Bredwardine and Pencelli had three sons and three daughters by his Welsh wife, Nest. The eldest, Ralph III, received Pencelli in Brecon and the majority of the manor of Bredwardine. The two other brothers were Thomas and Robert. Thomas’s descendants later held Weston and west Bredwardine in Herefordshire, Lawton ¹⁰³ The term is Jack Goody’s, as quoted in Waugh, The Lordship of England, 15. ¹⁰⁴ See below, pp. 133–5. ¹⁰⁵ Much of what follows relies on Copleston-Crow, ‘The Baskervilles of Herefordshire’, 18–39. ¹⁰⁶ See Ch. 3, for an example of the maintaining of military service.

112

Lords of the Central Marches

and Pickthorn in Shropshire, and Hellidon in Northamptonshire.¹⁰⁷ Robert, according to a charter in the Brecon Cartulary, held land in Pencelli of his father, Ralph II, who is described as his lord.¹⁰⁸ Ralph II de Baskerville also provided for two of his three daughters, Matilda, who married Roger Bret, and Isabel, by slicing off bits of Bredwardine as their marriage portions. Matilda was given 100s. worth of land in the manor.¹⁰⁹ Isabel’s husband is unknown, but her marriage portion was a virgate in Bredwardine.¹¹⁰ Female heirs posed another threat to the integrity of an estate, and proved a matter of concern to lords in the region. This was due to the custom of division of lands among daughters.¹¹¹ If a man had no sons and more than one daughter, the inheritance was divided evenly, with the chief messuage going to the eldest, whose husband received homage and service from the other daughters (and their husbands) and did homage to the chief lord for the entire fee. This is what happened to the Baskervilles of Bredwardine. Ralph III de Baskerville was succeeded on his death by three daughters, Agnes, Alice, and an unnamed daughter. Bredwardine was thus divided into three portions, passing to the daughters’ respective husbands, Roger Devereux, Robert le Wafer, and Roger de Radnor. In the inquest of 1242–3 it was Roger Devereux at the head of his participes who was listed as responsible for the (unspeciﬁed) service due from Bredwardine.¹¹² It should be noted, however, that fragmentation of estates through females carrying land away from the patrimony was only one-half of the equation. Marriage also could bring land into the family, and for this reason families followed strategies designed to limit losses and if possible maximize gains; this aspect will be treated below. It did not require inheritance by daughters to fragment a knightly patrimony. An example of the gradual dispersal of an estate is provided by ¹⁰⁷ BF 146, 1284; CIPM ii, no. 595. ¹⁰⁸ ‘Cart. Brecon’, 14. 160. Robert, with the consent of his wife, Elisent, gave the land held of him by Tudor Cymerdruc on his son James becoming a monk and also gave other land in exchange for a previous donation. The editor dates the charter to before 1175, when the one named witness, Jordan the archdeacon, resigned. ¹⁰⁹ BL, Additional Charter 20408. Although Matilda’s husband is not named in the charter, it is probable that he was Roger Bret, a member of another Brecon family, as 50s. of the land granted was in Woodbury, which was held by Brets late in the 13th century; Copleston-Crow, ‘The Baskervilles of Herefordshire’, 23. ¹¹⁰ BL, Additional Charter 20410. Copleston-Crow unfortunately has missed this charter in working out an otherwise excellent genealogy for the Baskervilles. ¹¹¹ Glanvill, ch. VII, sect. 3, p. 76. ¹¹² BF 811.

The Knightly Class

113

the Devereux family.¹¹³ In 1166 Roger Devereux held four knights’ fees of the Lacy honour of Weobley, while Walter Devereux held three.¹¹⁴ Walter held Lyonshall and Elstone Bridge, in the Golden Valley. This is all that can be traced from Herefordshire Domesday.¹¹⁵ An inquest of uncertain date but after 1228 showed William Devereux, son of Stephen Devereux, as still holding four and a half fees in Herefordshire along with a half-fee in Shropshire.¹¹⁶ By the 1242 inquest, William held a half-fee in Lower Heyton in Shropshire, a half-fee in Luntley, one fee in Lyonshall, a ﬁfth of a fee in Lawton, a quarter fee in Holme Lacy, and four parts of one fee in Stoke Lacy.¹¹⁷ His mother, Isabelle, held a half-fee in Hamond’s Frome as dower and the fourth part of one fee in Maund Aubin.¹¹⁸ Meanwhile, Cecily Devereux, a collateral relation, held ﬁve parts of a fee in Putley, while Nicholas, her son, held a fee in Chanstone.¹¹⁹ It is not possible to say unequivocally that William’s holdings suffered a rapid decline after his father’s death in 1228, for there is the reduction in knight’s service of the mid-thirteenth century to take into account. Yet the stability of William Devereux’s inheritance is misleading, as his holdings in Holme Lacy and Stoke Lacy were acquisitions of the thirteenth century.¹²⁰ Further, William engaged in extended litigation in order to claw back his father’s grants: the Devereux patrimony was obviously under some pressure in the mid-thirteenth century.¹²¹

R E L I G I O U S B E N E FAC T I O N S After the imperatives of marriage in this world, another pressure on knightly estates was the need to make gifts to the Church for the salvation of the family in the next world. The knights, of course, were not patrons on the scale of the great magnates, but their purposes were largely the same. Indeed, the foundation of monasteries and priories by their lords opened the opportunity for the knights to make their own benefactions. An example of this is the founding of Brecon Priory ¹¹³ See the Devereux genealogy in Appendix 2. ¹¹⁴ RBE 281–3. ¹¹⁵ HDB 42, 46. ¹¹⁶ BF 631. ¹¹⁷ BF 799 (Luntley), 803 (Lyonshall), 804 (Lawton), 805 (Bodenham), 806 (Stoke Lacy), 812 (Holme Lacy). ¹¹⁸ BF 805 (Maund Albin), 808 (Halmond’s Frome). ¹¹⁹ BF 816. ¹²⁰ Colvin, ‘Holme Lacy’, 24. ¹²¹ For William Devereux’s litigation, see below, pp. 126–7.

114

Lords of the Central Marches

by Bernard de Neufmarché c.1100, in which Roger de Baskerville and Robert Devereux participated by the gift of one burgage rent each.¹²² Land was the obvious donation to make to a religious house. For example, prior to 1175 Robert de Baskerville, on the entrance of his son James into Brecon Priory as a monk, granted to the priory the land which Tudor Cymerdruc held of him.¹²³ As land became scarce, other rights could be bundled together in a donation. Ralph de Baskerville of Bredwardine, either the second or third of that name, gave to the monks of Brecon the messuage of his gardener Semert in Bredwardine, with other lands along the Wye (at least 3 acres), the right to collect deadwood in his park and wood, and free passage over the Wye in his boat. The amount of land given in this grant was probably not a great amount but the inclusion of rights to deadwood and quittance of toll sweetened the donation.¹²⁴ As the twelfth century drew to a close, the pressures on land, heightened by inﬂation, ended the great period of monastic foundation and enrichment ushered in by the Conquest.¹²⁵ From the late twelfth century the pattern of knightly donations to religious houses changes. Land is given in smaller amounts, often combined with other perquisites. Rents are assigned rather than tenure. One thing that stands out in our region is the alienation of assets that usually were kept under seignorial control elsewhere: woods, moor, and mills.¹²⁶ Ralph de Baskerville’s grant of rights in his wood and park at Bredwardine has been noted above. Ralph III de Baskerville later granted to the monks of the Cistercian abbey of Dore all his pasture and that of his men in Bredwardine, along with the right of free entrance and exit with their carts and draft animals.¹²⁷ The Cistercians specialized in the development of waste land and pasture, so it was an appropriate donation. Stephen Devereux (d. 1228) of Lyonshall gave the mill in his manor of Halmond’s Frome, Herefordshire, to the canons of King’s Pyon.¹²⁸ The grant further speciﬁed that neither he nor his heirs would build any other mill in the manor. If the canons decided that another location was better suited (ubi magis aptum et commodum) for a mill, they ¹²² ‘Cart. Brecon’, 14. 141–2. ¹²³ Ibid. 14. 160. ¹²⁴ Ibid. 14. 163. ¹²⁵ See Ch. 3. ¹²⁶ The earl of Chester, for instance, in the late 12th century, carefully controlled and developed his woodland, meadow, and pasture; P. R. Coss, Lordship, Knighthood and Locality: A Study in English Society c.1180–c.1280 (Cambridge, 1991), 55. ¹²⁷ BL, Additional Charter 20409. ¹²⁸ BL, Harleian MS 3586, fo. 89.

The Knightly Class

115

were allowed to construct it. Suit of mill by the men of the manor accompanied the grant. Material from Devereux’s wood of Frome was allowed to the canons, by view of his foresters, for repairing the mill. As with pasture and meadow, this grant is another example of the alienation of valuable manorial resources. Landowners recognized that mills and milling were a valuable source of income, one which, if husbanded, could be made to increase its share.¹²⁹ In the twelfth century the majority of mills were held at farm, but by the beginning of the thirteenth century lords were trying to recover their mills and their revenues, much as they were attempting to recover alienated demesne arable. The bishop of Ely, for example, took his mills at farm back into his own hands early in the century, appointed his own millers, and rationalized and added to his milling operation. The mills were back at farm by the middle of the century, and indeed no pattern of demesne or lease seems to predominate for the century as a whole. Devereux’s grant was not a lease, however, but a gift in frankalmoin; his donation does not mirror seignorial activity elsewhere and would have lowered the value of his manor.¹³⁰ Of course, he may well have been in debt, and this grant may be a hidden transaction. Another form of religious benefaction by the knights was the grant of tithes and advowsons. This had been encouraged by the Church since the turn of the eleventh century, as the Hildebrandine reformers had opposed lay control of tithes. A fair number of advowsons still remained to be granted in the area by the second half of the twelfth century. In 1172 John Pichard, son of Roger Pichard, conﬁrmed his father’s grant of the advowson of their Herefordshire manor of Ocle Pichard to the Benedictine priory of St Guthlac’s, Hereford.¹³¹ Before 1202 William de Waldeboef, a Brecon tenant, had granted to St Guthlac’s his tithes in the manor of Gattertop, along with a site for the making of a messuage for their collection.¹³² Between 1228 and 1230 those tithes in turn were granted to Thomas de Waldeboef clerk, perhaps a son, for an annual rent of 2 marks.¹³³ But lawsuits in the thirteenth century show that even advowsons comprised resources that could be ﬁercely fought over, and a number of grantors tried to claw them back from the grantee. ¹²⁹ R. Holt, The Mills of Medieval England (1988), 70. ¹³⁰ Ibid. 70–1. ¹³¹ Balliol College, MS 271 (Cartulary of St Guthlac, Hereford), fo. 16r . v ¹³² Ibid., fo. 52 . The charter may be dated by the presence in the witness list of Richard, dean of Hereford, who occupied that ofﬁce from 1187 to 1202. ¹³³ Ibid. Grant made by Prior Henry with the assent of Abbot Henry of Gloucester (1228–43); Henry had been succeeded as prior of St Guthlac’s by Jocelin; ibid., fo. 58.

116

Lords of the Central Marches

F RO N T I E R R E S T R A I N TS One reason for this disposal of non-arable assets in religious benefactions was the military nature of Hereford and the March. Knight service and castle-guard were still a reality in the area in the thirteenth century.¹³⁴ Grants were speciﬁc in maintaining the military service due from land. The marriage contract between Walter de Lacy and Walter de Clifford, which dates from after 1200, grants the manor of Mansel Lacy in free marriage to Clifford with Lacy’s daughter Katherine.¹³⁵ The service of the two knight’s fees of Roger Tirel accompanies the grant. This knight service was then to be provided by Clifford’s heirs to Lacy’s, as much as Lacy was to make this service to the heir of William de Braose, from whom Lacy himself had gained the manor as maritagium with his wife, Margaret. Here we see the typical strategy, discussed below, of using land brought into the family by women to endow daughters and younger sons. Yet the military service, despite a lengthening chain of tenure, was clearly delineated and maintained. The lord of Brecon would still receive his service. To what degree this attention to service led lords in their honour courts to inﬂuence knightly alienation of land is a matter for speculation, but it must have exercised some restraining power. Military service being maintained along the frontier, tenants by knight service would have to maintain enough land—the ‘basic fee’—to fulﬁl their obligations. As noted earlier, another military drain upon family holdings was the necessity or desire to erect fortiﬁcations.¹³⁶ Clifford, Wigmore, and Richard’s Castle were all in existence at the time of Domesday Book. These castles were erected by substantial barons, but the knightly families of the area also built fortiﬁcations. The Domesday entry for the manor of Eardisley mentions the existence there of a domus defensabilis, no doubt a manor house surrounded by a ditch and wooden stockade.¹³⁷ Though crude, these fortiﬁcations were necessary, and, as described above, only became more so as the Welsh threat increased in the late twelfth century.¹³⁸ It is no coincidence that the ﬁrst decades of the thirteenth ¹³⁴ See Ch. 3. ¹³⁵ HCA, no. 3241. ¹³⁶ Herefordshire and the adjacent March formed a zone that is the most heavily encastellated in the kingdom, excepting the border county of Northumberland; see Ch. 2. Most of the fortiﬁcations in Northumberland, however, date from a later period than this one. ¹³⁷ HDB 47. ¹³⁸ Ch. 3.

The Knightly Class

117

century witnessed a dramatic increase in the building of masonry fortiﬁcations along the March; in essence, an arms race had begun. Partly out of necessity, partly no doubt from a desire to imitate the greater lords of the March and enhance their own status, some of the knightly families of the region also built in stone. Stone fortiﬁcations were erected at the Baskerville manor of Bredwardine, and possibly also at Eardisley.¹³⁹ In Brecon, Roger Pichard or his son John had built a stone shell keep at Tretower around the middle of the twelfth century. In the second quarter of the thirteenth century, Roger Pichard added extensively to Tretower’s defences, constructing a round masonry tower, of the type to be seen at Chepstow, Bronllys, and elsewhere in the March, within the shell keep. The wooden palisade of the bailey was replaced with a stone curtain wall with round towers at the angles of the bailey.¹⁴⁰ The round stone keep at the Devereux manor of Lyonshall is contemporaneous, and was surrounded by a wall that enclosed the motte and carried outside to encompass part of the bailey.¹⁴¹ It was not cheap to build in stone. Although the Baskervilles, Devereux, and Pichards invested in it, most of the other knightly families of the area stayed with their wooden motte-and-bailey fortiﬁcations, perhaps later replacing them with an encrenellated manor house late in the thirteenth or in the fourteenth century. The costs of the structures at Bredwardine, Tretower, and Lyonshall are difﬁcult to estimate. For the twelfth and thirteenth centuries, it has been estimated that any baron who wished to build a stone castle had best be prepared to spend a minimum of £350.¹⁴² A rough idea of the cost of knights’ building in stone may be gathered from the example of Sir John de Bishopsdon, a Warwickshire knight. In 1313 he built a stone gatehouse, measuring 40 by 18 feet, at his manor of Lapworth in that county. The cost was £16 13s. 4d., and Sir John had provided all the materials himself, excepting the stone. Christopher Dyer estimates the total cost at around £30, and points out that gentry with yearly incomes below £40 could only afford to build in timber, utilizing materials and labour from their own estates.¹⁴³ The cost of construction was followed by ¹³⁹ Cathcart King, Castellarium Anglicanum, i. 201. ¹⁴⁰ C. A. R. Radford and D. M. Robinson, Tretower Court and Castle, Welsh Historic Monuments pamphlet, rev. edn (1985), 7–9. ¹⁴¹ Cathcart King, Castellarium Anglicanum, i. 208. ¹⁴² Painter, Studies in the History of the English Feudal Barony, 173. ¹⁴³ Christopher Dyer, Standards of Living in the Later Middle Ages: Social Change in England c.1200–1520 (Cambridge, 1989), 81.

118

Lords of the Central Marches

the cost of maintenance and perhaps a skeletal staff; for the greater barons this cost could approach a quarter of their gross landed income, a crippling burden for those of knightly status if their costs were proportional.¹⁴⁴ It was the cost of this construction that probably pushed some of the greater knightly families of the area into debt, particularly to Jewish moneylenders. Jews had become established at Hereford by 1179.¹⁴⁵ By 1205 Walter de Baskerville of Eardisley owed £32 to Leo the Jew of Warwick. By 1210 his Jewish debt had increased to over £390.¹⁴⁶ Construction at Eardisley may not date quite this early; instead this debt may very well be related to the legal battles which Walter had fought against the prior of Llanthony and others in the ﬁrst decade of the century.¹⁴⁷ We have already seen how borrowing from the Jews and accompanying indebtedness had impacted on some of the lords of the region, exposing them to royal pressure at times of political stress.¹⁴⁸ The knightly class was subject to the same pressures as the magnates, both the necessity to spend on fortiﬁcations for defence and display, and, in their case, their concomitant exposure to debt and to whomever might buy up those debts. The Devereux provide a good example of this, as their castle at Lyonshall almost certainly contributed to the family’s debt. That money was borrowed from the Jews may reasonably be surmised. Two charters in the Mortimer Cartulary seem to indicate this. In one, William Devereux, the son of Stephen Devereux, conveyed to Roger de Mortimer pro quadam summa pecunie all his land of Stoke Lacy and Lawton; the proceeds may well have been necessary to service the family’s debt. A second states that William owed Mortimer 1,000 marks, with terms of payment of 100 marks a year to begin on 1 August 1264.¹⁴⁹ So William Devereux had sold two of his manors to raise cash and yet owed Roger de Mortimer 1,000 marks. It is tempting to suggest that the second transaction is the result of Roger de Mortimer’s having bought up Jewish bonds, a common practice up to the middle of the thirteenth century and one which ﬁgured prominently in the Petition ¹⁴⁴ N. J. G. Pounds, The Medieval Castle in England and Wales (Cambridge, 1990), 139. ¹⁴⁵ Pipe Roll, 25 Henry II, 41, when a certain ‘Mosse’ owed 5 marks for actions of debt against Henry de Miners and Hugh de Beauchamp. ¹⁴⁶ Pipe Roll, 12 John, 41. The actual amount was £390 5s. 4d . ¹⁴⁷ CRR ii. 36. For the costs of legal action, see below. ¹⁴⁸ See Chs 3 and 7. ¹⁴⁹ BL, Harleian MS 1240, fos 44–5 (new foliation).

The Knightly Class

119

of the Barons of 1258.¹⁵⁰ This possibly helps to explain why William Devereux became a Montfortian, dying at Evesham in 1265.

S T R AT E G I E S There were thus a number of pressures that bore on any knightly patrimony. But it is an exaggeration to say that there was a ‘crisis of the knightly class’ in the region; the evidence does not allow such a comprehensive statement. That said, there certainly were crises for individual families.¹⁵¹ Many of the families of the region, such as Baskerville, Devereux, and Pichard, survived into the late Middle Ages and beyond. This does not mean that there were not difﬁcult periods; the Devereux were under ﬁnancial pressure in the middle of the thirteenth century, and the participation by some families on the Montfortian side at Evesham may have reﬂected their fortunes but hardly improved them.¹⁵² But for these families to survive the centrifugal forces of marriage settlements, the endowment of younger sons, gifts to the religious, and debt, they had to have strategies for the protection and if possible enlargement of the patrimony.

Marriage Strategies Marriage was the most important tool available to the knights in the struggle to preserve their families’ fortunes, and it was also one way in which property could be diminished. As discussed above, gifts such as dower and maritagium meant the passage of land out of the immediate family line. But, not surprisingly, legal safeguards had developed by our period. Maritagium referred to that portion of land granted by the father of the bride ‘with’ her to her new husband. It was, however, an arrangement conditional upon the couple having children. Homage ¹⁵⁰ For Mortimer, see D. A. Carpenter, ‘A Noble in Politics: Roger Mortimer in the Period of Baronial Reform and Rebellion, 1258–1265’, in Anne J. Duggan (ed.), Nobles and Nobility in Medieval Europe: Concepts, Origins, Transformations (Woodbridge, 2000), 183–203. ¹⁵¹ Carpenter, ‘Was there a Crisis of the Knightly Class in the Thirteenth Century?’; King, ‘Large and Small Landowners in Thirteenth Century England’, 26–50. ¹⁵² The escheated lands of Walter Devereux and William Torel in Brinsop, Hugh del Fresne in Fresne, and John Pichard in Bredwardine were granted to Roger de Mortimer after Evesham; BL, Harleian MS 1240, fo. 37 (new foliation, 35).

120

Lords of the Central Marches

customarily was not taken for lands granted in marriage, whether frankmarriage or not, until the third heir entered the land, and there was good reason for this.¹⁵³ If homage were taken immediately for the land in question, the conditional arrangement within the family would become charged with the permanent force of feudal tenure. The granting father, for instance, would now be his son-in-law’s lord. Because of the rule which barred a lord of a holding from also being heir to it, the act of taking homage would mean that the land granted in marriage could no longer be reunited to the patrimony.¹⁵⁴ The custom of delaying the performance of homage thus served as an escape route, a means for property to revert to the family of the bride if a couple failed to produce heirs. Thus, the use of the phrase liberum maritagium in the two Baskerville marriage charters mentioned brieﬂy above meant that, though no service was due for the land in question, service would eventually be due to Ralph de Baskerville’s heirs, but if Isabel and Matilda de Baskerville or their immediate heirs died without issue, the land would return to the Baskervilles.¹⁵⁵ More important for the preservation and enlargement of the patrimony was the securing of advantageous marriages for sons. In the case of the heir, this would mean the addition to the patrimony of the marriage portion and, if the bride were a widow, the temporary addition of her dower lands. For younger sons, a good match could take pressure off the patrimony by obviating the need to make more permanent alienations in the father’s lifetime. There seem to have been two basic approaches to the marriage question: one could arrange a match within the locality, which in the March very often meant within the honour, or one could marry outside. Each approach had particular advantages. As it is, two of our best examples of marrying within the locality or honour are of instances where only daughters were left to inherit, and both come from the Marcher lordship of Brecon. The ﬁrst involves the manor of Burghill, which had been held in King John’s reign by William de Miners. William was succeeded by his only son, Henry.¹⁵⁶ When Henry died, he left three daughters, Isabel, Elizabeth, and Basilia ¹⁵³ Glanvill, ch. VII, sect. 18, p. 92. ¹⁵⁴ Milsom, ‘Inheritance by Women in the Twelfth and Early Thirteenth Centuries’, 61. ¹⁵⁵ See p. 112 above; BL, Additional Charters 20408, 20410. ¹⁵⁶ BF 1270. The descent of Burghill is related in the returns of the 1251–2 inquest made into serjeanties and terra Normannorum, as William de Miners had held land on both sides of the Channel.

The Knightly Class

121

(referred to elsewhere as Dulcia). Isabel was married to Geoffrey de Longchamp but seems not to have had an heir by him, for she later sold her portion in Burghill to Roger de Burghill, her nephew and Basilia’s son.¹⁵⁷ This indicates that Basilia had married into the important Brecon family of Burghill, which had supplied constables to the Braose lords of Brecon earlier in the century.¹⁵⁸ Her husband was Pain de Burghill, who held one knight’s fee of the honour of Brecon in 1211.¹⁵⁹ From the toponymic, it is apparent that Pain already held two-thirds of Burghill, as Henry de Miners accounted for a half-fee in Burghill.¹⁶⁰ Elizabeth, the other de Miners daughter, had married William de Gamages, who held one knight’s fee of the honour of Brecon and held the manor of Mansell Gamage of the honour of Weobley.¹⁶¹ William was not only a Brecon tenant but also a member of the familia regis from 1205 and served with King John in Ireland in 1210 against his immediate lord, William de Braose.¹⁶² The date of these marriages is difﬁcult to establish. Henry de Miners was still living in 1211, when the scutage returns for the forfeited lordship of Brecon show him holding one-half of a knight’s fee.¹⁶³ If they were arranged before Henry de Miners died and before he perhaps knew he would die without a son (and before the toppling of William de Braose), we can see Henry using two of his daughters to establish contacts with two important families within the honour. Indeed, we can expect that Henry’s lord would encourage just such matches; containing transfers of land within the lordship prevented a draining of resources away from the knightly families that were the backbone of its strength. The social and tenurial cohesiveness produced by marriages among the military tenants of a lordship was especially important when that lordship existed along a military frontier such as the Welsh March. Marrying daughters within the lordship would also be advantageous if ¹⁵⁷ For Isabel’s marriage (as well as those of her sisters), Rot. Litt. Claus. i. 345 (a writ to the sheriff of Gloucester to deliver seisin of Westbury to them and their husbands, dated 1217). Roger de Burghill was a minor in 1243; BF 802, Tulintone. ¹⁵⁸ Both William and Robert de Burghill, who were brothers, served the Braose lords of Brecon as constables; ‘Cart. Brecon’, 14. 225 (William, c.1215–22), 158 (Robert, ante 1191), Bodl., MS Tanner 342 (Brecon Cartulary extracts), fo. 170v (William, ante 1208). William also served as William de Braose’s undersheriff for Herefordshire in 1196–7; Pipe Roll, 9 Richard I, 194. ¹⁵⁹ BF 497. ¹⁶⁰ Ibid. ¹⁶¹ After he ﬁnally succeeded in his claim to it, as it had originally been held by his brother Matthew and had been forfeited as terra Normannorum. ¹⁶² Rot. Lib. 125, 212; Church, ‘Knights of the Household of King John’, 165; see Ch. 6. ¹⁶³ BF 497.

122

Lords of the Central Marches

Henry had a son, for marriage patterns of the time tended to be socially endogamous; ‘landholders tried to arrange marriages for their children within a circle of near social equals, distant cousins, or neighbors’.¹⁶⁴ This closed system of marriages meant that land circulated within the group, and Henry would eventually make good on his losses by the acquisitions which his son would gain by maritagium. Our next two examples come from two generations of the Baskerville family of Bredwardine. In addition to three sons, Ralph II de Baskerville (d. 1190/1) had three daughters, Matilda, Isabel, and Wiberga, who is known only from her attestation of Matilda’s marriage grant. Charters granting Matilda and Isabel their marriage portions survive, but unfortunately Isabel’s husband is not named.¹⁶⁵ Neither is Matilda’s, but she almost certainly married a Bret, another family of the honour of Brecon.¹⁶⁶ Here is yet another marriage contracted within the honour, serving both to keep land within a small social grouping and to strengthen the bonds between men, which was the strength of the lordship. Ralph II de Baskerville was succeeded by his son Ralph III. Ralph III de Baskerville (d. 1210) had only daughters, although how many is uncertain—there were certainly two, and likely three. Alice married Robert le Wafer, yet another Brecon tenant, while Agnes (eventually) married Roger Devereux, the son of Robert Devereux, who held one and a third fees of Brecon in 1212.¹⁶⁷ A third sister may have been married into the Radnor family, which did not hold land of the honour of Brecon, but did of the nearby Braose Marcher lordship of Radnor. By the inquest of 1242 Bredwardine was listed as being held by Roger Devereux et participes sui.¹⁶⁸ According to a historian of the Baskerville family, the manor of west Bredwardine was the portion that went to the Radnor family with the unnamed Baskerville daughter.¹⁶⁹ ¹⁶⁴ Waugh, The Lordship of England, 37. ¹⁶⁵ BL, Additional Charters 20410 (Isabel), 20408 (Matilda; Wiberga is a witness). ¹⁶⁶ Copleston-Crow, ‘The Baskervilles of Herefordshire’, 23. One-half of the 100s. grant in free marriage to Matilda was formed by 1 hide in Woodbury outside Bredwardine, which was held by the Baskervilles of the Cliffords. In the late 13th century the Brets held Woodbury; CCR ii. 407. ¹⁶⁷ Copleston-Crow, ‘The Baskervilles of Herefordshire’, 25–6; RBE 601. Robert le Wafer was a minor in 1212, his three fees being held by Reginald Croc and Ranulf Hurley. ¹⁶⁸ BF 811. ¹⁶⁹ Copleston-Crow, ‘The Baskervilles of Herefordshire’, 26. A Roger de Radnor was lord of West Bredwardine in 1286; CIPM ii, no. 595.

The Knightly Class

123

A ﬁnal concord ending an action against the abbot of Dore for a halfcarucate of land with appurtenances in Bredwardine, dated 6 October 1227, both clariﬁes and further confuses things.¹⁷⁰ The original action was brought by Robert le Wafer and Alice, his wife, Gruffudd ap Meurig and Matilda, his wife, John ﬁtz Simon and Alice, his wife, Peter de Aldermanston and Agnes, his wife, and Hugh of Radnor and Sybil, his wife; the claim was iure uxorum, as Ralph de Baskerville was the antecessor of all the women named. The suit was unsuccessful. The Baskerville–Radnor marriage is here identiﬁed, Sybil de Baskerville having been married to Hugh de Radnor. Robert le Wafer and Alice, his wife, are present. But what are we to make of Matilda de Baskerville and her Welsh husband, Grifﬁn?¹⁷¹ The one Agnes mentioned is not married to Roger Devereux but to Peter de Aldermanston. Are all these women of the same generation? Matilda may have been an aunt, the Matilda who married a member of the Bret family. Agnes almost certainly seems to have been married once before she married Roger Devereux.¹⁷² Genealogical confusion to one side, what becomes apparent is the high level of endogamy in the region’s marriage patterns. Although this conclusion admittedly is based on narrow evidence, the further descent of Bredwardine is suggestive. At some point between 1227 and 1242 Peter of Aldermaston must have died. Peter took his toponym from the manor of Aldermaston in Berkshire and was a rare outsider to the honorial group of Brecon tenants. His widow, Agnes, would then have come into the wardship of her lord, either Reginald de Braose (d. 1228), his son William V (d. 1230), or after William’s ¹⁷⁰ TNA, CP 25(1)/80/6, no.78. ¹⁷¹ At some point between 1223 and 1229 Grifﬁn, with the consent of Matilda, his wife, made a grant of land to Dore Abbey from the land granted them in Bredwardine by Ralph de Baskerville; Catalogue of Ancient Deeds, 6 vols (1890–1915), i, B404. ¹⁷² Copleston-Crow, ‘The Baskervilles of Herefordshire’, 25–6, states that the marriage of Agnes to Roger Devereux was contracted before Ralph III de Baskerville’s death in 1210, and that Roger’s father, Robert, held her share of the Baskerville estate in custody after 1210 while the couple were minors. There is no evidence for any of this. The evidence for a previous marriage in TNA, CP 25(1)/80/6, no. 78, is supported further by another action, brought against Thomas de Baskerville, the brother of Ralph III; Rot. Litt. Claus. ii. 154. The action was brought by John ﬁtz Simon and Alice, his wife, Hugh de Radnor and Isabel, his wife, and Peter de Aldermanston and Agnes, his wife, for half the manor of Lawton (in Diddlebury) in Shropshire with appurtenances. This writ is dated 7–15 July 1226. Peter of Aldermanston does not appear in any of the inquests contained in The Book of Fees, but his name suggests a Berkshire origin and thus a possible case of marriage outside of the honorial group.

124

Lords of the Central Marches

death one of the men (Hubert de Burgh, Richard of Cornwall, Gilbert Marshal) who held the lordship along with the wardship of the heir, Humphrey de Bohun. Agnes was then married to a military tenant of the honour, Roger Devereux, strengthening the honorial connections of the group of women holding Bredwardine. It would not be surprising if the tenurial separation of Welsh and English lands already noted was further reinforced by an unusually high level of endogamy at the knightly level of society.¹⁷³ Here we have a further cause of the separateness of Marcher society. This pattern in the March is somewhat different from that which may be seen in Herefordshire, where men such as Walter II de Baskerville and Stephen Devereux married into curialist families and thus broadened their sources of patronage as well as their social circles. Higher social standing, of course, could also lead to marriages outside the honour and local community, but in the cases of Baskerville and Devereux this in itself was partly a result of contacts with the king and royal government, contacts made easier by the increasing interaction of central government with the land-owning community of Herefordshire.

Purchase and Assarting Marriage was the most important means of adding to the patrimony, but there were other means by which the knights either added to their estates or increased their proﬁts. Chief among these was the development of existing holdings, either through purchase or through assarting. Purchases can be very hard to track in the records; some no doubt hide behind the language of ﬁnal concords but are not readily apparent.¹⁷⁴ One obvious example of purchase and engrossment is found in a ﬁnal concord contracted at Shrewsbury in 1236.¹⁷⁵ In this concord, Walter II de Baskerville brought an unspeciﬁed plea against forty-one separate tenants in the manor of Parva Kynlo. These tenants held from 1 to 7 acres each. All the tenants recognized Walter’s right to the land and he immediately set about a reorganization of holdings. Walter, at the petition of the tenants, ceded 100 acres to Andrew de Baskerville, who would hold of Walter for what forinsec service pertained to the land. The remaining land was redistributed among the original tenants, ¹⁷³ For tenurial division along honorial lines, see p. 111 above. ¹⁷⁴ Sales of land were most often hidden behind ‘unspeciﬁed’ pleas; introd. to Feet of Fines for the County of Norfolk, 1201–1215, and for the County of Suffolk, 1199–1214, ed. Barbara Dodwell, Pipe Rolls Series, new ser., 32 (1958), p. xx. ¹⁷⁵ TNA, CP 25(1)/80/9, no. 159.

The Knightly Class

125

with an average of 3 acres per holding. These tenants were to hold their land of Andrew for a rent of 3s. and forinsec service—scutage and reliefs in the manner of socage and feudal aids. Nothing resembling a payment is mentioned at the end of the concord, so it is possible that this deal was the result of strong-arm tactics on the part of Walter de Baskerville, rather than a purchase. Regardless of the means of acquisition, what we have here is the probable setting up of a younger son with land; Andrew was the name of Walter II de Baskerville’s brother-in-law Andrew de Chanceaux, and Andrew de Baskerville was possibly a younger son of Walter. Assarting was another route by which existing manors could be enlarged. The twelfth century was marked by large-scale assarting and reclamation of land across Europe as population increased. By the thirteenth century, reclamation on a large scale was possible only on the peripheries of the English state, and the assarting of unsuitable lands and forests was an indication of the growing hunger for land.¹⁷⁶ The clearance of land was particularly prevalent in the Welsh district of Archenﬁeld, the royal forest in the south-west of Herefordshire.¹⁷⁷ Development in Herefordshire is possibly indicated in the Michaelmas 1176 pipe roll, which recorded amercements in misericordia Regis pro foresta sua.¹⁷⁸ Thirty-seven men, including such major local ﬁgures as Gilbert de Monmouth and Roger de Chandos, lord of the honour of Snodhill, were amerced for forest offences, which may have included illegal assarting in the royal forests.¹⁷⁹ In 1214 we ﬁnd Andrew de Chanceaux, a curialis who was the guardian of Walter II de Baskerville, owing 50 marks at the exchequer for permission to assart in the wood at Orcop.¹⁸⁰ Another example of assarting is a ﬁnal concord made between Stephen, abbot of Dore, and Hugh of Radnor at Westminster in Easter term 1243.¹⁸¹ In an unspeciﬁed plea over 2 virgates of land in Bredwardine, Stephen recognized the land to be the right of Hugh. ¹⁷⁶ Postan (ed.), The Cambridge Economic History of Europe, 550–2. ¹⁷⁷ At the time of Domesday, Archenﬁeld had few plough teams, excepting the northern edge of the area, which indicates the gradual encroachment of Anglo-Norman agriculture; Sylvester, The Rural Landscape of the Welsh Borderland, 227. ¹⁷⁸ Pipe Roll, 22 Henry II, 44–5. ¹⁷⁹ The Monmouth tenants Henry de Kilpeck and Ralph le Bret are also listed, as are the Lacy tenants Robert de Baskerville and his son Ralph, Walter de Cormeilles, Herbert de Castello, and Richard de Escotot, and the Clifford tenants Hugh of Dunre and Henry de Sollers. ¹⁸⁰ Pipe Roll, 16 John, 59 (Gloucestershire nova oblata). ¹⁸¹ TNA, CP 25(1)/80/10, no. 196.

126

Lords of the Central Marches

Hugh and his heirs were allowed totam predictam terram arrare et pro voluntate sua excolare without the impediment of Stephen or his successors in perpetuity, excepting 60 acres, which Hugh granted to the abbot in free alms with the right of free passage across Hugh’s land. Here Hugh was not only bringing new land under cultivation but probably regaining part of the Baskerville patrimony which had been alienated to Dore Abbey.¹⁸²

Litigation The late twelfth and thirteenth centuries saw a marked increase in aristocratic litigation over land. The advent of new forms of action in the county and royal courts during Henry II’s reign provided the knights of Herefordshire with greater opportunities for the pursuit of claims.¹⁸³ In the March, despite the adoption by the late twelfth century of common law actions and forms, the reality was undoubtedly somewhat different.¹⁸⁴ The records of the central courts and the itinerant justices provide case studies which demonstrate the wide range of litigation engaged in by the knights of the region. One common occasion for litigation occurred on the entry of an heir into his inheritance. Heirs were naturally keen to regain parts of the patrimony which had been alienated, often without their approval. One proliﬁc example is William, son of Stephen Devereux. In 1238 he brought an assize of mort d’ancestor against Walter de la Hide for 20s. rent in Hide, while in 1241 he used the same action, this time against the prior of Wormsley for 10 marks’ rent ¹⁸² These two virgates may originally have been part of the grant of Ralph de Baskerville (II or III?) to Abbey Dore, which consisted of his parkland at Bredwardine; Mon. Ang. v. 555. ¹⁸³ The starting point for any discussion of litigation must be Sir Frederick Pollock and F. W. Maitland, The History of English Law before the Time of Edward I, 2nd edn, introd. S. F. C. Milsom, 2 vols (Cambridge, 1968); see also D. M. Stenton, English Justice between the Norman Conquest and the Great Charter 1066–1215 (1964), for the formulation of the ‘Angevin leap forward’. Maitland’s greatest challenge has come at the hands of S. F. C. Milsom, The Legal Framework of English Feudalism (Cambridge, 1976); support and explanation is provided by R. C. Palmer, ‘The Feudal Framework of English Law’, Michigan Law Review, 79 (1981), 1130–64. For a critique of Milsom, see Paul R. Hyams, review of S. F. C. Milsom, Legal Framework, EHR 93 (1978), 856–61; the newest work disagrees with Milsom’s idea, derived from Frank Stenton, of the ‘perfect feudal world’ of powerful honour courts, and is well represented by Hudson, Land, Law, and Lordship in Anglo-Norman England. ¹⁸⁴ See Ch. 3.

The Knightly Class

127

in Holme Lacy.¹⁸⁵ In the 1240s William was apparently denying the canons of Wormsley rights of estover in his woods at Lyonshall, the family seat.¹⁸⁶ In 1242 William was embroiled in a dispute with Thomas de Fauconburg, canon of Hereford, because Thomas claimed William was not upholding the agreement made between William’s father, Stephen, and Thomas’s predecessor over a pond and alder grove in Hereford.¹⁸⁷ The following year he was a party in two suits, one concerning land and another a mill.¹⁸⁸ In 1244 William had to secure a writ non obstante from the king for the restoration of his Norfolk manor of Wilby, which had been seized as terra Normannorum.¹⁸⁹ That same year the prior of the Hospitallers in England brought suit against William concerning the unjust detaining of a charter (de quadam carta injuste detenta), and one wonders if this charter concerned debts.¹⁹⁰ It was quite common for male heirs to mount legal challenges to alienations arranged by their fathers. Walter I de Baskerville, who was a minor on his father’s death and came of age no later than 1194, began to pursue his recovery of alienations shortly thereafter. An undated entry on one of the ﬁrst plea rolls from Richard I’s reign indicates that Walter had contested the right of the prior of Clifford to 40 acres in Little Cusop, near Hay.¹⁹¹ The prior defaulted and the land was taken by the sheriff of Hereford into the king’s hand; it is uncertain whether Walter ¹⁸⁵ William remitted and quitclaimed in return for Walter’s gift of a sparrowhawk; TNA, CP 25(1)/80/9, no. 171. In the Wormsley suit, William quitclaimed in return for him and his heirs being received into the blessings and prayers of the church of Wormsley; TNA, CP 25(1)/80/10, no. 176. ¹⁸⁶ TNA, CP 25(1)/80/11, no. 220. The matter was resolved in this ﬁnal concord of 1248, in which William conceded 2 acres of woods in Kingswood in frankalmoin, common pasture in his warrens at Lyonshall and his ﬁelds, and two mills in Lyonshall, for which he also granted wood for their construction and rights of ingress and egress, along with the suit his men owed to the mills. On top of this William also remitted and quitclaimed 6s. rent which he drew from the prior. In return, the prior remitted and quitclaimed any other rights of estover he might have in William’s woods, agreed to move a certain mill to a new location and quitclaim it to William, as well as renouncing some ways and paths and right of pannage for ﬁfty pigs, all in the wood of ‘Kerdesleg’. William gave the prior 2 marks. What we have here is a complicated exchange of rights but one in which we may suspect, to judge by William’s later ﬁnancial difﬁculties, he came off the worse. ¹⁸⁷ CRR xvii. 323, no. 1653 (Michaelmas term). William did not appear and was attached. ¹⁸⁸ Ibid., no. 1863 (Easter term). ¹⁸⁹ CR 1242–47, 165. ¹⁹⁰ Ibid. 238. ¹⁹¹ In concert with an Ernaldus de Baskerville, Rot. Cur. Reg., Richard I, no. 5, in Feet of Fines 10 Richard I: also a Roll of the King’s Court Richard I, Pipe Rolls Series, 24 (1900), 222. Ernaldus is otherwise unknown to the records.

128

Lords of the Central Marches

ever took possession of this land. In Michaelmas 1201 Walter was found in mercy for not having followed through on a writ which he brought against the prior of Llanthony et multos alios.¹⁹² This mention of ‘many others’ suggests that Walter was following a strategy of regaining grants and combined them into one overarching action.¹⁹³ Walter’s campaign of litigation would seem to have been unsuccessful. Not only did he not regain alienated portions of the patrimony, but he also had to defend himself against a claim by Geoffrey de Longchamp. Walter had been married to Emma de St Léger, Geoffrey’s mother. On her death Walter held onto her land, which Geoffrey claimed.¹⁹⁴ In 1195 Walter ﬁned 50 marks for seisin of his former wife’s hereditary lands, and that he be not disseised except by judgment, having been already disseised once on account of his Jewish debts. According to the exchequer, the king pardoned Walter the 50 marks because Richard I had returned the land to Geoffrey as heir in return for a 25-mark ﬁne. A ﬁnal concord dated 17 November 1195 gives further details of the arrangement between Walter and Geoffrey.¹⁹⁵ In return for Walter’s quitclaim of Emma’s lands, Geoffrey de Longchamp acquitted Walter versus Roger ‘Boin’ concerning the vill of Eardisley, which Walter himself had mortgaged (invadiaverat) to Roger. Two things become apparent: Walter was already in serious debt as early as 1195, having mortgaged the centrepiece of his inheritance, and he would seem to have lost this suit.¹⁹⁶ In the case of women, litigation provided perhaps the greatest scope for activity, and always posed a threat to the resources of the heir. One example is that of Cecily Devereux, the granddaughter of William Devereux (f l. mid-twelfth century).¹⁹⁷ As was so often the case with both men and women, a struggle developed over land given in marriage. ¹⁹² CRR ii. 36. John Pichard and Simon de Brobury were Walter’s pledges and were similarly found in mercy. ¹⁹³ Walter was almost certainly trying to regain, among other things, the advowsons of Eardisley, Stretton, and Orcop, which had been granted to Llanthony by Ralph I de Baskerville (d. 1148/9); Rot. Chart. 7 (royal conﬁrmation of Llanthony’s holdings dated 1199). ¹⁹⁴ Pipe Roll, 7 Richard I, 181 (Michaelmas 1195); Feet of Fines of the Reign of Henry II and . . . Richard I, 67, no. 81 (17 Nov. 1195). ¹⁹⁵ Feet of Fines of the Reign of Henry II and . . . Richard I, 67, no. 81. ¹⁹⁶ The ﬁnal concord lists Emma’s lands: the vills of Easton and ‘Kenpelee’, £10 land in Normandy in the vill of St Léger, and all the land with appurtenances which she had in ‘Fanigcurt’ and ‘Kinigmut’. ¹⁹⁷ See the Devereux genealogy in Appendix 2. She was the daughter of William Devereux’s daughter Orenge, who had been married to William of Lechlade. As is

The Knightly Class

129

Cecily’s campaign of litigation began in Easter term 1198, when she and Robert of Lechlade were given a licence of concord in Gloucestershire.¹⁹⁸ This was followed in Trinity term 1200 by another action, in which Cecily as plaintiff challenged Master Ralph of Lechlade and his brother William over 2 virgates and 6 acres of land in Lechlade. The plea was placed in respite the following Hilary term until Master Ralph’s lord, John de Grey, bishop of Norwich, should return from service with the king overseas.¹⁹⁹ The entry on the plea roll for the ﬁrst time sets out the details of the suit. Cecily claimed through William Devereux, her grandfather (avus), whom she asserted was seised in demesne in the time of Henry II. William of Lechlade placed himself on the grand assize. The question revolved around whether Cecily should hold the land in question or whether William of Lechlade should hold it of her along with that land given in maritagium with Cecily’s mother Orenge. Cecily was attempting to claim land that should have been held by William for life ‘by the custom of England’. Cecily lost her suit, even though two respites were granted for default of jurors, one of whom was found to have essoined falsely.²⁰⁰ This was not the only action which Cecily Devereux brought. Indeed, it seems that she seized on any claim possible. Her uncle Roger Devereux had died without issue and his four knights’ fees held of the honour of Weobley has passed to his younger brother Walter.²⁰¹ Yet, in Michaelmas 1201, Cecily obtained a recognition whether 140 acres of wood in Elstone Bridge in the Golden Valley had been held by her unfortunately so often the case, this genealogy, itself dependent upon the legal claims which Cecily brought forth, is tentative. ¹⁹⁸ CRR vii. 339. ¹⁹⁹ CRR i. 172, 401. Cecily’s opponent Ralph of Lechlade had powerful backing. John de Grey was a rising star at court. He was one of the two clerks who took over the great seal when Hubert Walter left the court in September 1199, and he probably had connections with the king when he had been count of Mortain. In the space of six months in 1200 John rose from being archdeacon of Cleveland to the post of archdeacon of Gloucester to bishop-elect of Norwich. In the words of Sidney Painter, ‘he was probably the only man whom [King] John trusted absolutely and without reservation for the entire period of their association’; The Reign of King John (Baltimore, 1949). Ralph of Lechlade almost certainly made his connection with John de Grey when John became archdeacon of Gloucester. The inﬂuence of John de Grey explains why nothing came of this suit until Michaelmas 1211 when the action was resumed; CRR vi. 144. By this point Ralph of Lechlade had died, removing John de Grey’s inﬂuence. ²⁰⁰ CRR vi. 194, 339, 367. This is an instance of what would come to be known as ‘curtesy’, a term which does not occur until Edward I’s reign; see Pollock and Maitland, The History of English Law before the Time of Edward I, ii. 414–20. ²⁰¹ CRR ii. 33.

130

Lords of the Central Marches

uncle Roger on the day he died, and whether she was his heir. The land was at that moment held by Roger’s lord, Walter de Lacy. The jurors returned a verdict in favour of Cecily and she was awarded seisin. Cecily also engaged in a dispute with the dean and canons of Hereford Cathedral, bringing an assize of darrein presentment against them over the chapel of Putley in east Herefordshire.²⁰² Once again, Cecily was pressing a claim to something granted away by her grandfather William. In Trinity 1205 Hugh, the dean of Hereford, offered the king 2 marks, and later 40s., for refuting (pro convincendo) the jurors who had made false judgment concerning ‘a certain chapel’, undoubtedly Putley.²⁰³ The chapter then protested in Michaelmas that the assize had been held in their absence, and asked that the case be placed in respite. Once again there were irregularities with the jury; a new panel of fourteen had been appointed after the refutation of the ﬁrst jury, but the sheriff did not send the names of either jury.²⁰⁴ The jurors then essoined en masse, as was the case with the suit against William of Lechlade above.²⁰⁵ Finally, in Michaelmas 1207 Cecily ﬁned 3 marks for mercy in the suit against the chapter of Hereford, indicating that she had been found in default.²⁰⁶ The aggressive legal activity of Cecily Devereux in the early thirteenth century shows how a woman of the knightly class could try, through the courts, to build upon her own position through claiming alienated parts of the patrimony. Her actions were by no means unusual. Litigation could be very expensive and often led to mounting debt. As the records of his dispute with Geoffrey de Longchamp show, Walter de Baskerville was already in debt to the Jews and had mortgaged Eardisley to Roger ‘Boin’. Where this debt originated is not known. Walter’s campaign of litigation was thus something of a necessary gamble. Regaining portions of the patrimony granted away would increase his income, enabling him eventually to get out from under his debts. Yet, litigation itself was not cheap. The amounts recorded for ﬁnes which appear in the records are only the tip of the iceberg. Perhaps the best example of the costs of litigation is the well-known Anstey case.²⁰⁷ Richard of Anstey, an Essex landowner, tried to claim ²⁰² CRR iv. 23, 230. ²⁰³ CRR iv. 23; Rot. Ob. et Fin. 297 (1205). ²⁰⁴ CRR iv. 302. ²⁰⁵ CRR v. 15 (Hilary 1207). ²⁰⁶ CRR v. 38. This was done coram rege. ²⁰⁷ See Patricia M. Barnes, ‘The Anstey Case’, in Patricia M. Barnes and C. F. Slade (eds), A Medieval Miscellany for Doris Mary Stenton, Pipe Rolls Series, new ser., 36 (1962), 1–24, and for discussion of the case, Paul Brand, The Origins of the English Legal Profession (Oxford, 1992), 1–3.

The Knightly Class

131

the inheritance of his uncle shortly after the accession of Henry II, and left records of his expenditure on litigation. His total outlay came to roughly £350. Although at one point Richard had been forced to plead his case at the curia in Rome, this was a single claim. Multiple claims and defences in the English courts could reach this level of expenditure without much effort. If Walter de Baskerville were to lose his suits, he would have no return to show for his outlay and possibly would be further in debt. This is exactly what seems to have happened; in 1205 Walter owed Leo the Jew of Warwick £32 plus interest, and by 1210 he owed over £390 at the exchequer.²⁰⁸ Walter was like a gambler, doubling down on his bets, for he had little choice once he got behind.

Royal Patronage, Multiple Lordship, and Ireland Another way of developing one’s estates was available through royal patronage, for example the right to hold markets and fairs and to enclose parkland. In Herefordshire it was those knights with some connection to the court or who served in local government who were most likely to receive these grants. In 1225 Walter II de Baskerville of Eardisley, who had served as a collector of the ﬁfteenth in that year, was allowed a market and fair at Eardisley.²⁰⁹ In 1246 his son Walter III was granted permission to impark his wood at Combe Baskerville, Gloucestershire.²¹⁰ Stephen Devereux likewise was able to expand and improve his holdings through royal patronage. In 1214 he received permission to assart 40 acres at his manor of Crowle, Worcestershire, in the royal forest of Feckenham.²¹¹ Stephen was also a collector of the ﬁfteenth of 1225, and in 1227 he was granted a weekly market and yearly fair at his chief manor of Lyonshall in Herefordshire.²¹² The proﬁts from the market and fair were possibly intended to help ﬁnance the concurrent rebuilding of Lyonshall Castle. Thus, service as ‘county knights’ fulﬁlling administrative tasks enabled both these men to round out manors, establish hunting preserves, and start lucrative local markets and fairs. Another means by which knights could increase their share of patronage was by establishing ties to more than one lord, and this ²⁰⁸ Pipe Roll, 12 John, 41. ²¹⁰ Patent Rolls 1232–47, 476. ²¹² Ibid. ii. 189b.

²⁰⁹ Rot. Litt. Claus. ii. 41b. ²¹¹ Rot. Litt. Claus. i. 168b.

132

Lords of the Central Marches

became more prevalent in the second half of the twelfth century and the early thirteenth century. As has been argued above, in the March by the early thirteenth century the consolidation of lordship and accompanying manipulation of inheritance and marriage had worked to isolate Marcher tenants tenurially; the ties of multiple lordship, at least in terms of tenure, were very few.²¹³ In Herefordshire, however, there was more room and opportunity for this strategy. This is best seen in the case of Stephen Devereux. While Stephen held land of Walter de Lacy, lord of Weobley, he was also a member of the household of William Marshal, which led to grants of land in Ireland and eventually Norfolk; thus, personal service in another lord’s household was used to supplement the beneﬁts of ties of tenurial lordship which Stephen had inherited. Stephen’s role in Ireland in 1207 has already been related, but as late as 1214 he served with William le Gros as an attorney for the Marshal in a suit against the men of the abbot of Abingdon and of Faringdon.²¹⁴ It was also during 1214 that Stephen Devereux appears as the recipient of royal favour, indicating ties to the greatest lord of all—the king.²¹⁵ Finally, Stephen’s connection to the Lacys was renewed and strengthened when, after the death of the Marshal in 1219, Gilbert de Lacy, the son of Walter, granted Stephen land in Staunton on Wye in Herefordshire pro eo quod sit in familia mea.²¹⁶ Another example of multiple lordship is that of the Pichard family, whose ties to the Lacys and Braoses were based on feudal tenure. Unlike the Baskervilles and Devereux, the Pichards moved more slowly to split their English and Welsh lands between sons.²¹⁷ As late as 1243 Roger Pichard held land in Staunton on Wye, ²¹³ See p. 91 above. ²¹⁴ For Stephen during the Irish crisis of 1207, see p. 99 above; CRR vii. 51 (Hilary 1214). ²¹⁵ Royal patronage is discussed below; see also Ch. 5. ²¹⁶ HCA, no. 3235. Because this charter is preserved on a 14th-century Lacy deed roll, there is no witness list that may be used for dating purposes. Internal evidence is also of no help. The terminus ante quem is 1228, the year of Stephen’s death, while a date post quem of 1219 has been proposed by Crouch, William Marshal, 225. This date would seem to be assigned by the need for Stephen to ﬁnd a new household on the elder Marshal’s death, and the guess is as good as any. It should be noted, however, that Stephen was the recipient of a grant of Wilby, Norfolk, by William Marshal II (1219 × 1227); ibid. Was there competition between Lacy and Marshal for Stephen’s service, or did he serve both lords? ²¹⁷ Yet the Pichards would eventually, in the course of the 13th century, divest themselves, through sale or subenfeoffment, of the majority of their lands in Herefordshire.

The Knightly Class

133

Ocle Pichard, and Almeley of the honour of Weobley, while he had subenfeoffed land in Weston Bret which he held of the honour of Brecon and was, of course, the lord of Tretower in the lordship of Brecon proper.²¹⁸ As discussed above, many knightly families of the region in the late twelfth and early thirteenth centuries divided their estates along tenurial lines, with Welsh acquisitions probably going to younger sons who started cadet branches. Even those families who retained lands on both sides of the hazy border between England and Wales would appear, however, to have switched their emphasis to their Welsh lands. The Pichards are an example. In 1243 Roger Pichard was one of the greatest subinfeudators among the knightly tenants we have examined. He was already using his English lands for purposes of patronage (if subinfeudation does not actually hide sales of land, which is possible). His son, another Roger (d. 1297), who is described as ‘of Ystradyw’, at some point in the 1280s conﬁrmed a grant to Reading Abbey which freed the abbey of all secular demands, suit of court and royal services owed to the Pichards for land the abbey held of them. Reading in turn gave Roger 20 marks and half a mark annual rent. That rent was later quitclaimed by Roger along with all his other rights.²¹⁹ This seems to indicate that the Pichards had shifted their attentions to Ystradyw in Brecon, ﬁrst subinfeudating their Hereford manors and eventually quitclaiming rents and vestigial rights. One reason for this sort of strategy might be that, following the Edwardian Conquest in 1282, Marcher lands became much more proﬁtable. This is particularly the case where Marcher lordship had weakened, allowing the great knightly families of those lordships to run their affairs largely as they pleased.²²⁰ Ireland also beckoned some members of the knightly class. The Welsh revival of the late twelfth century and Henry II’s policy of détente with the Welsh princes served to make Irish recruiting, particularly by ²¹⁸ BF 801–3. ²¹⁹ Reading Abbey Cartularies, 286, nos 352, 353. Leominster Priory in Herefordshire was dependent upon Reading Abbey. The lands in question are not delineated in the charters. ²²⁰ For the weakening of lordship and its eventual reassertion in the late 13th and early 14th centuries, see Davies, Lordship and Society in the March of Wales, ch. 4, esp. pp. 92–3 on Brecon. The replacement of such ancient Marcher families as Braose, Lacy, Marshal, and Clifford by newcomers in the middle years of the 13th century must have had much to do with the weakening of lordship in the March, which was exploited by knightly tenants.

134

Lords of the Central Marches

Diarmait Mac Murchada, attractive to Marcher families.²²¹ Although the initial group of Anglo-Normans to settle in Ireland came from south Wales, particularly Pembroke and its dependent lordships, Hugh de Lacy’s entry onto the Irish stage with the grant of Meath in 1172 provided his knights in Herefordshire and the March with access to the new frontier. Robert Bartlett has written on the colonial establishment of Hugh de Lacy in Ireland, and particularly focuses on the role of and difﬁculties for those knightly families that straddled the Irish Sea.²²² Ireland was a land of opportunity for the landless knight; Adam de Feipo, who was listed as a household knight of Hugh de Lacy in his 1166 carta, was set up as a major Irish landowner by the early 1180s.²²³ But Bartlett also points to the difﬁculties experienced by a more substantial family of knights from the honour of Weobley, who became involved in Ireland. In 1166 Richard de Escotot held three knights’ fees of Hugh de Lacy, centred on King’s Pyon near Weobley. There were junior members of the family, Roger, who held one fee, and Anketil and William, who were stipendiary knights in Hugh de Lacy’s household, but it was Richard de Escotot who received land in Ireland. The eventual problem for the Escotots, however, was that it was difﬁcult for a knightly family to manage their possessions on both sides of the Irish Sea. The Escotots would eventually either sell out or transform their relationship to their Irish estates to that of a purely rentier nature.²²⁴ It is perhaps because of this difﬁculty of administration that the knightly families examined in this book on the whole stayed out of Ireland. Stephen Devereux is the chief exception, as his service in the Marshal’s household brought him land in Ireland.²²⁵ We occasionally encounter a name which indicates an indeterminate relationship. Among those taken at Carrickfergus Castle in 1210 was Robert de Waldeboef, whose family held land in Brecon. He was probably a landless younger ²²¹ M. T. Flanagan, Irish Society, Anglo-Norman Settlers, Angevin Kingship: Interactions in Ireland in the Late Twelfth Century (Oxford, 1989), 137–45. ²²² Robert Bartlett, ‘Colonial Aristocracies of the High Middle Ages’, in Robert Bartlett and Angus Mackay (eds), Medieval Frontier Societies (Oxford, 1989), 23–47. ²²³ RBE 283; Bartlett, ‘Colonial Aristocracies of the High Middle Ages’, 37–8. Adam’s land centred on Skreen near the Boyne, and Hugh de Lacy built him a castle there. The building of castles for newly established vassals was a common occurrence of Lacy lordship in Meath. This seems to indicate the initial lack of resources possessed by the knights who came to Ireland, and thus demonstrates the type of knight most likely to make the venture. ²²⁴ Ibid. 38–40. ²²⁵ Crouch, William Marshal, 113.

The Knightly Class

135

son who took service in Ireland with the Lacys. Overall, though, it seems there were fewer ties than might be expected between the March and western Herefordshire on the one hand and Ireland on the other. In conclusion, much of what we have examined in this chapter is utterly typical of knightly society in other parts of England. The knights of western Herefordshire and the March were confronted with pressures on their patrimonies: the need to endow younger brothers and provide marriage portions for daughters, the need to make benefactions to the Church for the salvation of souls, the expenditure necessary to cut a ﬁgure in local society. Yet there was an additional pressure which was peculiar to the March, that of a terra guerre —the need for families to maintain the ‘basic fee’ well into the thirteenth century, and for many the need to build fortiﬁcations, even if at a fairly primitive level. The major knightly families, furthermore, imitated the greater lords of the March, such as Lacy and Marshal, by building in stone early in the thirteenth century. This was a very expensive undertaking, one which almost certainly added to the level of debt experienced by the Devereux and the Baskervilles of Eardisley. Family strategies for the preservation of the patrimony were also similar to those used throughout England: endogamy, the purchase and development of land, the cultivation of contacts with more than one lord, and ﬁerce litigation. In the March, however, there are two trends, one proven and the other only suspected, that should be noted, for they helped to make the March the peculiarity that it became. From the mid-twelfth century, knightly estates which had previously spanned the hazy border between England and Wales began to be split up, with the divisions made largely along honorial lines. This had the eventual effect of isolating the Marcher lordships tenurially. By the mid-thirteenth century, there were surprisingly few ties of multiple lordship entwining Marcher knights—the Marcher lordships were not only compact units, but their tenants held most, if not all, of their lands of one lord. Secondly, it is a possibility that there was a higher degree of endogamy in the Marcher lordships. Although nowhere near enough marriage contracts or genealogical details survive to establish this as undoubted fact, the examination of the marriages of the Miners and Baskervilles of Bredwardine, and more importantly the lack of multiple tenures, strongly suggest this. By the thirteenth century, to cross from the lordship of Brecon to the western Herefordshire world of a Stephen

136

Lords of the Central Marches

Devereux was in many ways to enter a much wider world. This had much to do with the further incorporation of Herefordshire into the English state that accelerated with changes in Angevin government, beginning in 1166. That incorporation and those changes are two of the topics of the next chapter.

5 The Role of Royal Government The previous two chapters examined the role played by the magnates and knights respectively in the shaping of the border region of Herefordshire and the adjacent Marcher lordships. We have seen that the late twelfth and the thirteenth centuries were a time of consolidation of lordship on the part of magnates, and that this placed on the knights the imperative to maintain the ‘basic fee’. We have not, however, examined the role of royal government in the region. How did the king’s government affect these two processes? In the ﬁrst instance, the king inﬂuenced Herefordshire and the southern March through his attitude (which periodically had a military dimension) towards the Welsh. Further, and more consistent, control in the area came through the exercise of his feudal lordship over his barons. These two aspects of royal power acted upon both Herefordshire and the adjacent Marcher lordships. In the course of the twelfth century, however, royal power and inﬂuence in Herefordshire increased greatly. Initially, royal power in the county rested upon the twin pillars of the terra regis and the ofﬁce of sheriff. Herefordshire was, of course, an English shire; but it was also unlike the other border counties of Cheshire and Shropshire in that it was not organized from the 1060s along defensive lines. From the 1160s the growth of Angevin government would serve to tighten royal control and increase crown inﬂuence in the county. At the same time, this extension of Angevin justice gave the knights of Herefordshire greater freedom from their lords, and opened new avenues of patronage and advancement to some of them. This process, occurring at the same time that cross-border holdings at the knightly level of society began to break up, helped to deﬁne and harden the boundary between western Herefordshire and the March.¹ ¹ See Ch. 4.

138

Lords of the Central Marches

It is no accident that by the thirteenth century—as royal government became increasingly intrusive in the border counties, and also during periods of political turmoil—Marcher lords such as the Mortimers began to withdraw land from the border counties into the March and thereby out of the reach of the normal judicial and ﬁnancial processes of direct royal control. In the March the king’s inﬂuence was sporadic, often dependent upon minorities or political forfeitures, but never entirely absent. Royal inﬂuence varied from reign to reign and with the demands upon the king’s attention of other parts of the Anglo-Norman realm and Angevin empire. But, as the history of occasional revolts in the Norman period as well as the importance of the area during Stephen’s reign demonstrate, no king of England could afford to ignore the March.²And, although the March of Wales, as it came to be known, was not a part of the kingdom of England, the Marcher lords were still within the ambit of the king’s power. They were feudal tenants-in-chief, which exposed them to the power of the royal prerogative; they usually held lands elsewhere in England and sometimes Normandy; they frequented the royal court; and they were often dependent on royal military power against the increasingly powerful Welsh. Indeed, tenuous claims to an English overlordship over all the British Isles pre-dated the Norman Conquest, and these claims would have new life breathed into them in the course of the thirteenth century.³ The ﬁrst half of the thirteenth century witnessed the increasingly dominant role of the king in the affairs of the March.

ROY A L P O L I C Y TOWA R D S T H E W E L S H The pattern of royal diplomatic and military policy in the AngloWelsh borders, within which the lords of the region had to operate, was actually set before the Anglo-Norman period, particularly by the actions of Harold Godwinson in the years just before the Norman Conquest. In the years preceding 1063 Gruffudd ap Llywelyn had asserted his dominance over the other princes in Wales; his growing power culminated in a series of raids into England from 1052 to 1056, which included the burning of Hereford city and cathedral in 1055. Ultimately the English response was to arrange peace with Gruffudd and to acknowledge the majority of his gains at English expense. Harold ² See Ch. 2.

³ Davies, Domination and Conquest, 66–87.

The Role of Royal Government

139

reversed this with an offensive against the Welsh in 1063, utilizing swift-moving light infantry and a ship-borne attack on south Wales, the success of which resulted in the abandonment and killing of Gruffudd by his own men.⁴ Royal policy towards the Welsh in the eleventh and twelfth centuries varied between two general approaches. One approach was to reach a modus vivendi with the Welsh princes (usually effected after punitive military expeditions to assert English overlordship, such as those in 1114, 1121, 1163, and 1165), and the concomitant reining in of the Marchers. Alternatively, the king could give to those Marchers a free hand to conquer what they could from the Welsh. The use of punitive military expeditions had a long history, beginning with Harold Godwinson. After 1066 William the Conqueror seems at ﬁrst to have given his captains along the border licence to expand into Wales with relatively little royal supervision. The revolt of Roger of Breteuil in 1075, however, made for a shift in William’s approach to Wales; ‘he seems to have chosen a policy of reaching an accommodation with some of the Welsh princes and exacting fealty and tribute from them in preference to a policy of territorial conquest’.⁵ According to English sources, the Conqueror journeyed to St David’s in 1081, at the head of an armed force.⁶ A peace was obtained with the Welsh which lasted for the remainder of William’s reign. Thus, the pattern was set by which the English king used military means to establish or reinforce his claim to overlordship and then left the Welsh princes in control, in return for the payment of a farm. William Rufus was more active militarily along the March and in Wales than was his father. He put down rebellions in 1086 and 1095 in which barons with lands in the March played a prominent role, and in 1095 led an army against the Welsh of Gwynedd, although the expedition achieved little; a further expedition in 1097 fared little better. The use of Welsh troops by the rebellious Marchers in 1086 and 1095 demonstrates how Wales could play a role in the political confrontations of the Anglo-Norman realm.⁷ But the failure of Rufus’s ⁴ Davies, The Age of Conquest, 24–7; Lloyd, A History of Wales, ii. 10–11. ⁵ Davies, The Age of Conquest, 29. ⁶ Lloyd, A History of Wales, ii. 393–4. The Welsh Brut stated, probably somewhat disingenuously, that William had come to St David’s on pilgrimage; Brut, 17. ⁷ See Ch. 2. Welsh troops were later used by Robert of Gloucester at the battle of Lincoln in 1141; Crouch, The Reign of King Stephen, 141–2; id., ‘The March and the Welsh Kings’, 277–8.

140

Lords of the Central Marches

expedition shows the limited stamina and interest of the king of England in Wales, and the obstacles which would confront the projection of royal military power there—difﬁcult terrain and limited returns. Instead, Rufus adopted the policy of encouraging Marcher enterprises in Wales, for our region none more important than Bernard de Neufmarché’s conquest of Brycheiniog.⁸ Rufus also inserted men of his own choosing into the region—including, in the south-eastern March, Hamelin de Ballon at Abergavenny. But even with the king’s attention elsewhere, a royal imprimatur was usually necessary for the Anglo-Norman barons to pursue their ambitions in Wales. Henry I was more masterful in his dealings with the Welsh. His reign demonstrated that the king of England was ultimately the dominant player in the game of power in Wales, a fact which remained true of all his successors into the thirteenth century, with the exceptions of Stephen and Richard I. Henry I certainly never aimed at the conquest of Wales; rather, he used military expeditions against Gwynedd in 1114 and 1121 to cow the Welsh rulers into accepting his overlordship.⁹ Military action was only one weapon in the king’s armoury for dealing with the situation in Wales: Henry made far more use of ﬂattery, bribery, and the fomenting of internal divisions within Welsh princely houses to achieve his aims. As for controlling the Marcher barons, the incidents of feudal overlordship and highhandedness allowed Henry I to redraw the feudal map of the March into a pattern much more pleasing to the Crown.¹⁰ But it was already becoming apparent, with the stalling of the Marcher advance into Wales by the turn of the century if not earlier, that the military power of the king of England was often necessary to push farther into Wales or to check and reverse a major Welsh resurgence. Stephen’s efforts in Wales in 1136 failed, not least owing to the inadequate commitment of resources.¹¹ Henry II’s reign was something of a watershed in medieval English history. For the border region the period of his rule was marked by two signiﬁcant events: the coming of the new legal environment, treated above, and the halt in Marcher expansion which occurred in the 1170s and 1180s. We ought to examine Henry’s actions from his accession to gauge how the Angevins exercised royal power in the March. We also have to look at the March and pura Wallia as a whole, ⁸ See Ch. 2. ¹⁰ See below.

⁹ See Davies, ‘Henry I and Wales’. ¹¹ See Ch. 2.

The Role of Royal Government

141

for the kings of England thought as much in terms of personalities as in terms of geographic and political entities which were still only half-formed. As with so much else, Henry II followed the example of his grandfather Henry I in his approach to the borders and March. The largely indirect control over the March and Wales proper which Henry I had established had been erased by the civil war of Stephen’s reign.¹² Thus, a restoration of Henry I’s mastery over both the Marchers and the Welsh was one of the many tasks which faced Henry II on his accession to the throne. Initially his method was the time-honoured one of overlordship backed by the threat of royal military expeditions. As far as priorities go, Wales was fairly low on Henry II’s list; he was only in Wales four times in the thirty-ﬁve years of his reign. Three of those times (1157, 1163, 1165) were punitive expeditions which lasted a month at most, while the fourth time occurred in late 1171 when the king traversed south Wales on his way to Ireland.¹³ Although none of Henry’s expeditions against the Welsh was particularly successful, the amount of resources devoted to them was not insigniﬁcant. For the campaign of 1163 troops were raised from the Continental parts of the Angevin empire, mercenaries were enlisted, and the feudal host was summoned.¹⁴ For the 1165 campaign, supplies and money were raised in Gloucestershire, Staffordshire, Northamptonshire, and Surrey. Over £180 was spent on the hiring of serjeants, while a further £100 was spent on knights based at Shrewsbury. The royal household formed the nucleus for the army.¹⁵ This force, while clearly not capable of effecting the conquest of a large portion of Gwynedd, still outstripped anything that an individual Marcher baron, or even a handful of them operating in concert, could muster. What do the military campaigns of Henry II tell us about the situation along the March? Although these campaigns followed the traditional pattern of an advance along the coastal plain of north Wales and were directed against Gwynedd, they reﬂect a larger reality which was true for all the March of Wales by the middle of the twelfth century. The rapid tide of Anglo-Norman expansion had been blunted and the Marcher barons increasingly lacked the military resources to ¹² See Ch. 2. ¹³ A general account of Henry’s Welsh policy is in Warren, Henry II, 153–69. ¹⁴ Lloyd, A History of Wales, ii. 515. ¹⁵ Paul Latimer, ‘Henry II’s Campaign against the Welsh in 1165’, WHR 14/4 (Dec. 1989), 533.

142

Lords of the Central Marches

continue that expansion; rather it was a period of consolidation in the south-eastern March. Indeed, at times of Welsh unity such as 1136, the Marcher lords were very hard-pressed to hold onto what they already had.¹⁶ The Mortimers had almost certainly lost control of Maelienydd in the late 1130s, while the 1140s saw war in Maelienydd and Elfael as the Mortimers and Braoses respectively pursued their families’ goals in the March.¹⁷ These conﬂicts along the border continued into Henry II’s reign and always posed the threat of disrupting the wider March, thus drawing in the king. For this reason it was important that Henry settle the political situation along the March, in order that he could concentrate his energies on the Continent, which was always a higher priority. Henry II’s three campaigns, despite their lack of outright military success, culminated in the desired, and traditional, effect with the agreement of October 1171 by which the Lord Rhys was made the king’s ‘justice of south Wales’.¹⁸ The Marchers, many of whom had become engaged in the adventures in Ireland of Richard de Clare, known as Strongbow, were in disfavour, and this was evidenced by Henry’s seizure of the baronial castles in south Wales on his way to Ireland.¹⁹ The king had correspondingly acted to check their pretensions in Wales as well as Ireland, and had done this by once again referring to the example of his grandfather. Henry I had striven to establish a balance of power along the March which relied upon a degree of cooperation among Welsh, Marchers, and the Crown.²⁰ Henry I effected this partly by placing several men of his own choosing into vacant or forfeited Marcher lordships. Henry II was aided by the royal wardship of Pembroke (1176–89) and Glamorgan (1183–9), but opportunities for promotion of his own men along the south-eastern March were few. Henry II perhaps wisely passed on his best chance of inserting new personnel with his lenient treatment of Hugh de Mortimer in 1155.²¹ ¹⁶ For the Welsh rising of 1136, see Ch. 2; Crouch, ‘The March and the Welsh Kings’, 255–74. Crouch writes that owing to the Welsh rising of that year, ‘the Marcher barons saw their weakness made manifest’; p. 273. ¹⁷ The Welsh Brut states that in 1144 Hugh de Mortimer rebuilt the castle of Cymaron and took the commote a second time; Brut, 53; Smith, ‘The Middle March in the Thirteenth Century’, 77–9. See also Ch. 2. ¹⁸ Warren, Henry II, 167. ¹⁹ Giraldus Cambrensis, Expugnatio Hibernica: The Conquest of Ireland, ed. A. B. Scott and F. X. Martin (Dublin, 1978), 90–1. ²⁰ Warren, Henry II, 160. ²¹ For Henry I’s use of ‘new men’ in the Middle March, see Ch. 2; Davies, ‘Henry I and Wales’, 145–6. For Henry II and the revolt of Hugh de Mortimer, William of

The Role of Royal Government

143

When a new generation of Marcher lords emerged in the 1170s, Henry II’s détente with the Welsh was strained. William III de Braose succeeded his father, William II, as lord of Brecon, Builth, Radnor, and Abergavenny in 1175 and promptly proceeded to kill the Welsh chieftain Seisyll ap Dyfnwal, his son, and some of the men of Gwent at Abergavenny; for this act his appointment as royal lieutenant in south Wales was cancelled.²² In 1179 the men of the young Roger de Mortimer killed the Welsh ruler of Maelienydd, Cadwallon ap Madog, when the Welsh chieftain was returning to Wales under a royal safeconduct.²³ In this instance Henry II punished the young Marcher lord, executing some of his men and putting Roger himself into prison, while conﬁscating Mortimer’s lands and the castle of Cymaron.²⁴ What these two events demonstrate is that the aggressive families of the southeastern March, Mortimer and Braose, were liable to upset Henry’s ﬁnely wrought Welsh peace in the pursuit of ancestral goals. But the king was also able and willing to chastise the Marchers if they went against the framework of Anglo-Welsh relations which he had set out. Indeed, the men of Gwent gained a measure of revenge for the massacre at Abergavenny in 1175 when, in 1182, they sacked Abergavenny Castle and then ambushed an Anglo-Norman party led by William de Braose and the sheriff of Herefordshire, Ranulf Poer, killing him and several of the party.²⁵ Although the arrival of the justiciar Ranulf de Glanvill with forces restored the situation, it was apparent that only the considerable power and personal cooperation of Henry II and the Lord Rhys served to preserve general peace along the March. It was for these reasons—among them, the increasing role of royal military power and the unreliability of certain Marcher lords, as well as the faltering of Marcher momentum—that the Crown began to assert a greater control over the conduct of military operations along the March. One form that this took was the appointment of ‘wardens’ to supervise the defence of the borders and to conduct negotiations with and military operations against the Welsh. As with so many aspects of royal ‘policy’ Newburgh, Historia Rerum Anglicarum, in Chronicles of Stephen, Henry II and Richard I, iv. 184–5; Warren, Henry II, 60–1. ²² Brut, 71; Davies, The Age of Conquest, 275. ²³ Brut, 71, s.a. 1179; Pipe Roll, 25 Henry II, 39. ²⁴ Two and a half marks were accounted for by the sheriff of Herefordshire for transporting prisoners to Windsor and Worcester to stand trial for the death of Cadwallon; Pipe Roll, 25 Henry II, 39. ²⁵ See Brut, 72, s.a. 1182; Lloyd, A History of Wales, ii. 147.

144

Lords of the Central Marches

regarding Wales, this expedient seems to have its origins under Henry I. Richard of Beaumais, based in Shropshire, was the ﬁrst man who can be credibly seen as performing this role. He was succeeded by Pain ﬁtz John, another curialis.²⁶ Henry II also appears to have used this method of asserting royal control over the March. After the campaign of 1165 Henry crossed the Channel to the Continent, where he was to stay for four years. Before he left, however, he fortiﬁed and garrisoned the March.²⁷ Furthermore, evidence preserved in exchequer records shows that Henry apparently left one of his most trusted magnates, William d’Aubigny, earl of Arundel, in charge of the March.²⁸ This is the only evidence between the reigns of Stephen and John to indicate the use of a ‘warden’ of the March. The reason for relying on one man to police the borders was simple enough: Henry’s concerns were on the Continent; he did not need the distractions that a conﬂagration in Wales would cause, and the Marchers, particularly the volatile young scions of the houses of Mortimer and Braose, certainly could not be relied upon to preserve the peace and the balance of power upon which it depended. Indeed, taking into account Henry’s expedition to Ireland in 1171–2 and his brusque treatment of Strongbow and Robert ﬁtz Stephen, the king seems determined to check the pretensions and ambitions of the Marchers wherever they might manifest themselves. John and Henry III would also on occasion appoint men to coordinate military activity along the March—for example, William, earl of Salisbury, in 1209 and John de Monmouth in 1231.²⁹ The détente between Henry II and the Lord Rhys, reached in October 1171, dependent as it was upon the good relations between the two men, was fragile, and collapsed with Henry’s death in 1189. Richard I had the same military concerns on the Continent as his father; indeed, his task was more difﬁcult than that faced by Henry. Correspondingly, the time he devoted to Wales was minimal and the Marchers were given free rein once more.³⁰ No warden or great ²⁶ For Richard de Beaumais, see Suppe, Military Institutions on the Welsh Marches, 93; for Pain, who married Sybil de Lacy and was sheriff of Shropshire, see Ch. 2; Green, The Government of England under Henry I, 252–3. ²⁷ Robert of Torigny in Chronicles of the Reigns of Stephen, Henry II and Richard I, iv. 226. ²⁸ An inquest of c.1170 states that Arundel custodivit markiam Walis, while another shows that his tenants paid scutage when he went ad servandas les Marches de Wales pluribus vicibus; RBE, pp. cclxvii, cclxx. ²⁹ Rot. Litt. Pat. 88; CR 1227–31, 601. ³⁰ Although see the dissenting opinion of Gillingham, ‘Henry II, Richard I and the Lord Rhys’, and for comment, Ch. 6.

The Role of Royal Government

145

magnate would appear to have been given responsibility for coordinating military action along the March. But the Marchers were just as vulnerable to major Welsh offensives as they had been in 1136, a fact decisively proven by the events of 1196, the year the Lord Rhys attacked Carmarthen, sacked the Braose castle at Colwyn in Elfael, and defeated the forces of Roger de Mortimer and Hugh de Say at Radnor.³¹ Finally, Rhys invested Painscastle and forced the garrison to surrender; only by reaching an agreement with the Welsh prince was William de Braose able to prevent the castle from being razed.³² As in 1136, the Marchers were overwhelmed by their Welsh foes; nothing demonstrates more the importance of the military role of the Crown along the March than the fact that it took a royal army led by the justiciar Geoffrey ﬁtz Peter in 1198 to recover the situation for the English.³³ The Marchers still had the same problems: inadequate resources and a lack of coordination. The military capabilities of the Welsh had improved as well. Only the king could provide the men, material, and leadership necessary to check a Welsh leader of the calibre of the Lord Rhys. The change in the balance of power in Wales and the March is probably one reason why King John built up William III de Braose to the heights of power which that Marcher lord reached in the early 1200s, before taking control of a large section of the March himself. The building up of Hubert de Burgh’s power along the March in the early part of Henry III’s reign should be viewed as part of this same pattern as well, rather than solely as personal aggrandizement.³⁴ True, it was easier to build personal empires in the March than elsewhere. But the maintenance of peace and stability along the March could be reinforced by the installation of a ﬁgure or ﬁgures of pre-eminent power, be he a court favourite or the king himself, with landed resources in the region which could be translated into military power. Furthermore, that man would still need the military support of the king.³⁵ ³¹ Brut, 76, s.a. 1196; Lloyd, A History of Wales, ii. 581. Forty knights and many footmen were killed at Radnor. ³² ‘concordia cum domino Willelmo de Breusa peracta, in bone pace reliquid’ (‘ ‘‘Chronica de Wallia’’ and Other Documents from Exeter Cathedral Library MS. 3514’, ed. T. Jones, BBCS 12/1–3 (Nov. 1946), 4, s.a. 1196). ³³ Brut, s.a. 1198. ³⁴ See R. F. Walker, ‘Hubert de Burgh and Wales 1218–32’, EHR 87 (1972), 465–94. ³⁵ For Richard’s and John’s reigns, see Ch. 6; for that of Henry III up to 1246, see Ch. 7.

146

Lords of the Central Marches

F E U D A L I N C I D E N TS A N D ROY A L LO R D S H I P Another important way in which the king inﬂuenced both western Herefordshire and the lordships of the south-eastern March was through the exercise of his rights as feudal overlord. The Marcher lordships were held for minimal or undeﬁned feudal obligations—Blaenllyfni would later be held in return for the performance of homage and nothing else.³⁶ The fact, however, that they were held in chief of the king still opened occasional avenues for the exercise of royal control and power. As the supreme feudal lord the king exercised a bundle of rights over his tenants-in-chief, those men who held anything from a barony to a knight’s fee of the Crown. These rights consisted of primer seisin, relief, consent to marriages and alienations, and prerogative wardship. Under the Anglo-Norman and Angevin kings these feudal rights were at times used in an arbitrary manner because of the ﬁnancial beneﬁts, patronage opportunities, and political control which they brought to the Crown. Lands held in escheat or wardship could be exploited directly or the custody sold off to the highest bidder; wardships and marriages could also be sold. Ultimately, these practices helped lead to the (brief) imposition of Magna Carta on King John in 1215 and, after it was renewed by the minority government of John’s son Henry, to the restraints it placed on the king’s exploitation of feudal incidents. Henry III respected the Charter; in the words of one historian, what was most striking about Henry was ‘the continued awareness of the principle of the Great Charter that the king must submit to the ‘‘law of the land’’ ’.³⁷ The ﬁnancial schedule for reliefs was honoured, and reliefs were no longer charged when the Crown had previously been able to exploit lands while held in wardship. Yet beyond this Henry III and Edward I forcefully resisted any further diminution of the king’s feudal rights, and used their courts to investigate vigorously and recover their prerogatives. By the end of the thirteenth century royal lordship, at least as practised by the Angevins and particularly John, had been trimmed and bounded. That said, even operating within the bounds of Magna Carta, royal lordship—especially wardship and marriage rights—remained a powerful force.³⁸ ³⁶ CIPM ii. 365, no. 606. ³⁷ Ralph V. Turner, The King and his Courts: The Role of John and Henry III in the Administration of Justice, 1199–1240 (Ithaca, NY, 1968), 277. ³⁸ For a full treatment of royal lordship under the later Angevins, see Waugh, The Lordship of England.

The Role of Royal Government

147

These rights of royal lordship had greater scope for operation in Herefordshire than in the March. Prerogative wardship, the right of the king to take into custody the lands of any tenant-in-chief, no matter how small his holding of the Crown and regardless of whom else he held land, operated in the county; it did not in the Marcher lordships.³⁹ An example of this superior royal lordship in Herefordshire in action occurred in 1244, when Henry III took the homage of Walter III de Baskerville for the lands and tenements he held of the king in capite.⁴⁰ If Walter had been a minor, all of his lands, whether held of the king or not, would have been taken into royal wardship. By manipulation of his feudal rights the king could redraw the feudal map of the March. We have already seen in Henry I’s reign the striking examples of Pain ﬁtz John, whom Henry married to Sybil, daughter and heir of Hugh de Lacy, and the marriage of Miles of Gloucester to another Sybil, the daughter of Bernard de Neufmarché, the conqueror of Brycheiniog, the arrangement being made during Bernard’s lifetime.⁴¹ Pain may indeed be regarded as having been married to Sybil so that he could function as a ‘warden’ of sorts for the king along the March. Likewise, the power of Miles of Gloucester in the region was transformed by his marriage to the heiress to Brecon. Henry I thus used his rights of wardship and marriage—and did so by anticipation—to place his own men in positions of power along the March. Henry II did not use these rights to the same degree in the region as did his grandfather. This was partly because of a lack of opportunity, as the lords of the south-eastern March all managed to produce sons to succeed them in this period. Nor did Henry strip the rebellious Hugh de Mortimer of his lands in 1155 when Hugh held out against the new king.⁴² The honours of Weobley and Ewyas Lacy did come into royal wardship from 1186, when Hugh II de Lacy was killed in Ireland, until 1189, when his son Walter ﬁned for his relief. As Walter de Lacy was a minor but did not marry Margaret de Braose until 1200, Henry did not seize the opportunity to arrange a marriage for him.⁴³ ³⁹ See ibid. 72–3. ⁴⁰ Rotulorum Originalium, i. 6. ⁴¹ See Ch. 2; Ancient Charters, Royal and Private, 8, no. 6 (Miles–Sybil marriage). ⁴² This was of a piece with Henry’s treatment of rebellious vassals throughout his reign. In 1155 Henry took Bridgnorth Castle from Hugh and turned it into a royal castle, much as he stripped private castles from those who fought against him in the war of 1173–4; Warren, Henry II, 136–42. ⁴³ For Walter’s entry into his inheritance, Pipe Roll, 1 Richard I, 145; Braose–Lacy marriage, Rot. Ob. et Fin. 81.

148

Lords of the Central Marches

Unlike their predecessor Henry I, Henry II and Richard I did not change personnel in our stretch of the March, although both kings did so when opportunities presented themselves elsewhere in Wales, in Glamorgan and Pembroke (most dramatically with the marriage of Isabelle of Striguil, claimant to the earldom of Pembroke). In Richard’s reign, when John, the heir of Gilbert de Monmouth, came into royal wardship, the custody was sold to William de Braose the younger in late 1189 or early 1190 for an offering of 1,000 marks.⁴⁴ This added further both to Braose power in the region and to Braose debt. Of course, with Richard about to embark on crusade, he was more interested in raising money than in altering the feudal geography of the March, if he ever had the desire to do so in the ﬁrst place. Neither, at least at the level of tenant-in-chief, did John use his feudal rights, but opportunities did not present themselves, and his methods of redrawing the feudal geography of the region ultimately were more dramatic.⁴⁵ It would of course be mistaken to think of the Angevin kings as having any sort of overarching policy towards the south-eastern March or the March of Wales in general. Instead they reacted to opportunities as they arose and used them to further their own aims and men as the situation warranted. The one opportunity upon which John might have seized was the marriage of Isabelle de Mortimer, whose husband, Roger, lord of Wigmore, died in 1214. But she was not distrained to marry. Of course, in 1214 John was trying to curry favour with the Marchers as the unrest that would lead in the following year to Magna Carta mounted.⁴⁶ Forcing Isabelle de Mortimer into a marriage not of her choosing would not have made him more popular in the region; neither would the intrusion of any more curiales or Poitevins. The Angevins therefore did not use their feudal prerogative to intrude new men into the region on a permanent basis, although unsettlingly King John inserted familiares such as Falkes de Breauté and Engelard de Cigogné as custodians of forfeited lands, making them powerful and unpopular agents of the king’s will. Instead, the king trusted local families to contract suitable marriages; royal approval or conﬁrmation served to reinforce those marriages. ‘The Crown used the basic feudal powers that it retained from the twelfth century to promote, as well as to supervise, the marriage alliances of the families of its tenants-inchief.’⁴⁷ Marriage, being the lifeblood of the aristocracy, was a very ⁴⁴ Pipe Roll, 2 Richard I, 48. ⁴⁵ See Ch. 6. ⁴⁷ Waugh, The Lordship of England, 71.

⁴⁶ See Ch. 6.

The Role of Royal Government

149

touchy matter for a king to interfere in. It is worth noting that although King John levied ﬁnancial exactions and most probably extended royal justice to the forfeited Braose lands in the March from 1208 to 1215, and administered them via foreign mercenary commanders, he does not appear to have interfered radically in the marriages of Braose subvassals, such as the daughters and heirs of Ralph III de Baskerville of Bredwardine.⁴⁸ The distribution of wardships and marriages to families in the region formed an important part of royal patronage. One striking example involves the Mortimers. Roger de Mortimer had been an adherent of John when, as count of Mortain, he had rebelled against his brother Richard’s chancellor William de Longchamp in 1191.⁴⁹ That loyalty was subsequently cemented by royal patronage after John’s accession to the throne in 1199. Some of this patronage was, at a price, directed to Roger’s wife, Isabelle. In 1205 she ﬁned 300 marks for the manors of Lechlade and Longborough which had been the estates of her younger brother Hugh de Ferrers.⁵⁰ These manors had apparently escheated to the Crown.⁵¹ In 1207 Isabelle received the manor of Oakham, which had been held by her elder brother Henry, who had adhered to the French king and was thus terra Normannorum, for life.⁵² In 1211 the king granted the wardship and marriage of Walter de Beauchamp to Roger and Isabelle.⁵³ When Roger died in 1214, the king respected Roger’s resignation of his lands to his son Hugh when he was taken ill. John also made sure that Isabelle received her dower lands.⁵⁴ John helped to assure the loyalty of the Mortimers by granting them lands that had escheated to the Crown, wardships, and marriages, and by acting to ensure the safe transfer of both patrimony and dower. In the case of the Mortimers, King John had indeed shown good lordship. ⁴⁸ For the imposition of royal control in the region 1208–15, see Ch. 6; for the Baskerville marriages, see Ch. 4. ⁴⁹ See Ch. 6. ⁵⁰ Rot. Litt. Claus. i. 30b. For a discussion of Isabelle, Lechlade, and later Mortimer claims to that manor, see Carpenter, ‘A Noble in Politics’, 183–203. ⁵¹ Crump, ‘The Mortimer Family and the Making of the March’, 121–2, states that Isabelle’s ﬁne was for the lands of her father, Walchelin, who had adhered to Philip Augustus following the fall of Normandy. The writ clearly states, however, that the ﬁne was for the two manors which had belonged to her brother Hugh, who presumably had inherited the title to his (and her) father’s lands. ⁵² Rot. Ob. et Fin. 398. ⁵³ Pipe Roll, 13 John, 273. ⁵⁴ Crump, ‘The Mortimer Family and the Making of the March’, 122; Pipe Roll, 16 John, 120; Rot. Litt. Claus. i. 208.

150

Lords of the Central Marches

ROY A L D E M E S N E I N H E R E F O R D S H I R E Royal power in any county normally rested on two elements: land—demesne and forest, as well as castles and boroughs—and the ofﬁce of sheriff. Compared to other counties, however, there was little royal demesne in Herefordshire. The county does not appear in the one surviving pipe roll of Henry I’s reign, but the Herefordshire Domesday, compiled c.1160–70, outlines the sheriff’s farm as it must have stood in Henry I’s reign.⁵⁵ It was composed of sums from the city of Hereford, ﬁve royal manors (Marden, Lugwardine, Stanford, Linton, and Wilton), and the Welsh district of Archenﬁeld, which was royal forest. Royal demesne in the county contributed £141 10s. to the county farm, which was set at £164 16s. 5d. The remaining £23 6s. 5d . would have come from such miscellaneous sources as sheriff’s aid. There was more royal demesne in 1086 than in the 1130s, the period which the list reﬂects, but this diminution was part of a general trend throughout the kingdom. Despite its occasional increase by forfeiture, escheat, or outright purchase, the pressures and demands of patronage soon worked to reduce the amount of royal demesne.⁵⁶ An example of this can be seen when, shortly after his accession in 1154, Henry II enfeoffed the Angevin supporters Gilbert Talbot and Hugh de Longchamp with the demesne manors of Linton and Wilton respectively.⁵⁷ The income yielded by Herefordshire decreased in the period 1066–1135 from £335 to £165, and this ﬁgure does not take into account further deductions for terrae datae and the sheriff ’s allowances.⁵⁸ There was also little royal forest in Herefordshire: the Hay of Hereford (Haywood), Archenﬁeld (which was disafforested in 1251), and part of the Forest of Dean, which stretched into Gloucestershire west of the Severn.⁵⁹ Forest law therefore had little impact on Herefordshire, unlike in the northern counties.⁶⁰ This lack of royal demesne and forest was important; comparison with the north is instructive. J. C. Holt pointed out that the retention of pleas of the Crown in the northern counties as well as the large amount ⁵⁵ HDB 75, 125. ⁵⁶ B. P. Wolffe, The Royal Demesne in English History (1971), 36. ⁵⁷ RBE 650; HDB 125. ⁵⁸ Wolffe, The Royal Demesne in English History, 33. ⁵⁹ Margaret Ley Bazeley, ‘The Extent of the English Forest in the Thirteenth Century’, TRHS, 4th ser., 4 (1921), 1400–72, esp. the map at p. 140. ⁶⁰ See Holt, The Northerners, 194–5.

The Role of Royal Government

151

of royal forest gave royal government in the region under the Angevins ‘a reasonable chance of success’.⁶¹ Even so, in the twelfth century royal government in the north was not ﬁrmly entrenched and even the sheriffs were often weak; royal power rested on the demesne and thus King John’s ‘administration was based on the great custodies rather than on the shrievalties’.⁶² The lack of a landed basis for royal power in the southern borders was thus of crucial importance. THE SHERIFFS Royal power in Herefordshire from the mid-twelfth century would therefore seem to have depended largely upon the sheriff. Particularly from the 1190s on, there were two basic choices for the Crown when appointing sheriffs: either great magnates and curiales on the one hand, or local men or professional administrators more amenable to the demands of the exchequer on the other. The needs of patronage and security therefore at times conﬂicted with ﬁscal policy.⁶³ Both magnates and curiales were harder to hold accountable for the county farm and expenditures, and reasonably viewed their appointment as a reward from the Crown—one from which they could proﬁt, and which would, if they were local magnates, enhance their power. Lesser men were much more beholden to the Crown for their appointment, but did not always have the standing in the locality to oppose the greater men of the region. It is worthwhile examining the pattern of appointments to the shrievalty of Herefordshire in the second half of the twelfth century and the ﬁrst half of the thirteenth century. Upon succeeding to the throne in 1154, Henry II not surprisingly found it necessary to court the magnates, and therefore Earl Roger of Hereford received the lands and dignities which his father, Miles of Gloucester, had held: the earldom and castle of Hereford, along with the custody of Gloucester Castle and the shrievalty of Gloucester.⁶⁴ Roger of Hereford rebelled against Henry in March of 1155, and a certain Maurice brieﬂy replaced Earl Roger as sheriff for 1154 to 1155, although, as he had served as the earl’s ⁶¹ Ibid. 199. ⁶² Ibid. 200. ⁶³ See D. A. Carpenter, ‘The Decline of the Curial Sheriff in England, 1194–1258’, EHR 91 (1976), 1–32, esp. 5–6. ⁶⁴ Rot. Chart. 53, 61. There had been tension between Henry and Roger over Gloucester Castle in April–June 1155; Robert of Torigny in Chronicles of Stephen, Henry II, and Richard I, iv. 184.

152

Lords of the Central Marches

steward, the ofﬁce was still essentially under the earl’s control.⁶⁵ Earl Roger’s younger brother Walter was sheriff from 1155 to 1159.⁶⁶ By Michaelmas 1160 William de Beauchamp had combined the shrievalty of Hereford with that of Worcestershire, the Beauchamp ‘home’ county, and the Beauchamps continued to control the shrievalty of Hereford through Michaelmas 1169.⁶⁷ The year 1170 as elsewhere in England marked a shift away from the use of baronial sheriffs. The only exception came in 1173–4, when the great rebellion against Henry II made security paramount and thus resulted in the appointment of William de Braose.⁶⁸ Instead, in some years local knights began to ﬁll the ofﬁce, men such as William de Lye (1170), who held three knights’ fees of the Lacy honour of Weobley; William Torel (1183), who held Brinsop in Herefordshire of the Braose lords of Brecon; or Miles de Mucegros (1182), who held Monnington-on-Wye of the Tosnys.⁶⁹ In the 1170s and 1180s we can thus see the beginnings of a shift towards local sheriffs who were both more amenable to exchequer control and responsive to local sentiment; this also accords with Henry II’s rapprochement with the Lord Rhys and therefore a lessening of security concerns along the border.⁷⁰ In Richard’s reign the shrievalty of Herefordshire reverted to baronial control, with the exception of the appointment of Henry de Longchamp at the beginning of the reign.⁷¹ From Michaelmas 1191 to Michaelmas 1197 William de Braose held the shrievalty. As Braose was on the Continent for large stretches of this period, he usually relied on undersheriffs ⁶⁵ Emilie Amt, The Accession of Henry II in England: Royal Government Restored 1149–1159 (Woodbridge, 1993), 115–16. ⁶⁶ The Great Roll of the Pipe 2, 3, & 4 Henry II, ed. J. Hunter (Record Commission, 1844), 50, 93, 143; Pipe Roll, 5 Henry II, 49. ⁶⁷ PRO Lists and Indexes, ix: List of Sheriffs for England and Wales (1898); see under Herefordshire. ⁶⁸ The military nature of Braose’s appointment may be inferred from the entries on the pipe roll for 1173 for the payment of Welsh troops and the installation of bretasches at Hereford Castle; Pipe Roll, 20 Henry II, 121. ⁶⁹ RBE 281–3; HDB 94, 112. ⁷⁰ It is of course debatable just how ‘independent’ of their immediate lords any of these knightly sheriffs were likely to be. ⁷¹ A piece of patronage undoubtedly advanced by the king’s chancellor and familiaris William de Longchamp, who was his brother. Following William de Longchamp’s fall from power in 1191 Henry was removed from ofﬁce; Richard R. Heiser, ‘Richard I and his Appointments to English Shrievalties’, EHR 112 (Feb. 1997), 9. Both William and Henry were sons of Hugh de Longchamp, who had been granted the royal manor of Wilton in the county in 1154; Balfour, ‘The Origins of the Longchamp Family’, 82. The family was thus established in the region, but also curialist in nature.

The Role of Royal Government

153

such as William de Burghill, a Brecon tenant and Braose constable.⁷² For the last two years of Richard’s reign Walter de Clifford held the shrievalty. This was still a baronial appointment, as Clifford had an important territorial stake in the region, though to a lesser degree than William de Braose. On the other hand, this appointment used a family which had connections to the court from Henry II’s reign. King John shufﬂed a number of men through the shrievalty—William de Braose (1199–1200), Hubert de Burgh (1200–4, obviously making use of an undersheriff), William de Cantilupe (1204), and Walter de Clifford (1205–8) consecutively up to 1208, when John’s takeover of the Braose and Lacy lordships of the Middle March led to the appointment of the king’s mercenary commanders Gerard d’Athée and then Engelard de Cigogné.⁷³ Unrest in the region in 1215 led to the reappointments of Hubert de Burgh and Walter de Clifford (brieﬂy), and then Walter de Lacy, a security appointment of the old magnate type seen in 1173–4, but one also designed to soothe local sensibilities following the rising of the county in support of the exiled Braoses.⁷⁴ Lacy held the shrievalty until late 1223, when the household steward Ralph ﬁtz Nicholas was appointed. Fitz Nicholas, who served until late 1229, was succeeded in due course by men such as John de Monmouth (Michaelmas 1231 to 7 July 1232), who had close ties to the court as well as lands in the region, and Amaury de St Amand (1234–40), another household steward.⁷⁵ Thus, from the 1220s, sheriffs of Herefordshire were appointed from the ranks of men in the royal household or those closely linked to the court, and wariness about giving the ofﬁce to greater and distrusted ﬁgures may have been one reason for this. Another reason why magnates were no longer made sheriff was undoubtedly the fact that, from the late twelfth century, the diminution of the sheriff’s power combined with the increase in his duties and responsibilities made the ofﬁce much less attractive to magnates.⁷⁶ The one exception to the appointment of ‘lesser men’ was the holding of the position by Peter des Riveaux in ⁷² For William de Burghill as constable of Brecon, see ‘Cart. Brecon’, 14. 225; Bodl., MS Tanner 342, fo. 170v . ⁷³ See Ch. 6. ⁷⁴ See Ch. 6. ⁷⁵ John de Monmouth had performed such roles as royal justice and keeper of the king’s forests south of the Trent; see Ch. 3. For Amaury, see Robert Stacey, Politics, Policy and Finance under Henry III 1216–1245 (Oxford, 1987), 40. ⁷⁶ Carpenter, ‘The Decline of the Curial Sheriff in England’, 2; for a somewhat differing view of the ofﬁce under John, see B. H. Harris, ‘King John and the Sheriff ’s Farms’, EHR 79 (1964), 532–47.

154

Lords of the Central Marches

1232, part and parcel of the enormous number of shrievalties gathered into his hands in that year. This was an appointment which, combined with the grant to him of other custodies in the region, did much to offend local opinion.⁷⁷ ROY A L J U S T I C E The impact and inﬂuence of Angevin government on Herefordshire went much further than the ofﬁce of sheriff. Although they are no longer thought to mark a break with an ‘autonomous seigniorial world’, the legal reforms of Henry II, and the attendant mechanisms for their execution, served to incorporate Herefordshire further into the ambit of central control, further differentiating the county from the adjacent Marcher lordships. Royal government under the Angevins increasingly penetrated beyond the tenants-in-chief to the knights. Herefordshire in this period was no exception. Henry II has traditionally been credited with the increased reach and multiplying forms of royal justice—the famous ‘Angevin leap forward’.⁷⁸ The grand assize determining right by means of a jury, and the petty assizes—mort d’ancestor, novel disseisin, darrein presentment, and utrum—proved highly popular with all levels of landed society, from very small freeholders to great magnates. Cases could be moved out of honour and county courts into the curia regis, and judicial eyres brought the king’s justices into the localities. What this meant for Herefordshire is that the county was gradually incorporated more tightly into the embrace of the king’s government. Further, these legal mechanisms enhanced heritability and rights to alienation over property held by the knights, and thus limited the discretion and power of lords over their tenants; ‘the centrality of lordship and the honour to social relationships tended to weaken’.⁷⁹ After these reforms, the ability of lords in England to distrain by land for services was drastically curtailed.⁸⁰ In many areas of England the Angevin legal reforms contributed to the decline of honour courts, thus ⁷⁷ See Ch. 7. Though it is worth noting that Riveaux was also a parvenu, a creation of court politics and faction, not a magnate with a solid base of power in the localities. ⁷⁸ Stenton, English Justice between the Norman Conquest and the Great Charter 1066–1215, ch. 2. ⁷⁹ John Hudson, The Formation of the English Common Law (1996), 213. ⁸⁰ Ibid. 214.

The Role of Royal Government

155

tipping the balance of power in a measure from the magnates towards the king. We may be certain that the magnates of the region took notice of the impact that these reforms, by giving legal force to their tenants’ rights to inherit and to alienate, would have on their lordship. This more than anything else is the reason why lords such as William de Braose were so strident in their exclusion of royal justice from their Marcher honours at the turn of the twelfth century. Royal justice threatened the consolidation of Marcher lordship. This weakening of the lord’s power can be seen from the 1166 carta of Hugh II de Lacy (itself a ﬁnancial document submitted to the exchequer), which shows that he was unable to regain the knight service owed by some of his tenants.⁸¹ The Angevin judicial reforms made it difﬁcult for many if not most lords to discipline their tenants through disseisin. Although the king still very occasionally claimed, and at times exercised, the right to review cases from the March, the writ of error did not run in Marcher lordships, unlike in Ireland. As royal justice did not operate in the Marcher lordships, Marcher lords still retained their full discretionary powers.⁸² Indeed, the growth of royal justice, and royal government in general, made it more important to deﬁne just where England ended and the March began. As late as the end of the twelfth century there were doubts as to where one jurisdiction ended and another began. Lords did not hesitate to extend their Marcher liberties at the expense of the Crown. Thus, the question arose in 1199 as to whether the sheriff of Herefordshire was able to intervene in a dispute over the castle of Bredwardine, which lay within the county of Herefordshire but was held as a parcel of the honour of Brecon.⁸³ Royal justice did not entirely account for the difference between England and the March—inheritance and marriage patterns were already dividing the two societies into increasingly separate units—but it did help to sharpen a deﬁnition that was already coming into being, and could make the reuniting of family holdings difﬁcult. Royal justice was extended into the region by the judicial eyre. Although cases could be, and were, heard in the curia regis at Westminster, ‘the new form of royal power was most obvious to the Angevins’ subjects when the king’s itinerant justices visited their locality’.⁸⁴ These eyres were responsible for all pleas and also played a role in levying ﬁnancial ⁸¹ RBE 281–3. For a discussion, see Ch. 3. ⁸² For an example, see the discussion of the Siward case in Ch. 3. ⁸³ See Ch. 3. ⁸⁴ Hudson, The Formation of the English Common Law, 123.

156

Lords of the Central Marches

exactions. The eyre relied upon the participation of local knights, and in this way a direct link was forged between the Crown and knights, bypassing the magnates.⁸⁵ The eyre was a forum for the projection of royal power and through the reading of writs and proclamations, the projection of royal ideology.⁸⁶ It was the great regularity with which the eyre visited Herefordshire in the later twelfth and the early thirteenth centuries which did much to impress royal justice upon the county. In the legal sphere it thus sharpened the differences in jurisdiction, and at times procedure, between Herefordshire and the Marcher lordships which helped to deﬁne the March. In Henry II’s reign, the ﬁrst eyre visited the county in 1160, followed by another in 1166, then every year from 1175 to 1180, 1183, and 1185–9. Four eyres were conducted in Herefordshire in Richard I’s reign (1190, 1192, 1194, 1199), while in 1198 the eyre was held in Shrewsbury in Shropshire. Only one normal eyre visited Herefordshire in John’s reign, in 1203, but after 1204 the county was graced by two eyres coram rege, in 1208 and 1211, while John and the curia regis were at Gloucester in 1209 and 1210.⁸⁷ This was partly because of John’s ceaseless activity about his kingdom, partly because he came to or through the region in order to deal with problems in the region in 1208 (the Braose crisis) and 1211 (the Welsh campaign) or on his way to Ireland in 1210. The personal nature of royal justice in the region during John’s reign is of a piece with his intensiﬁcation of royal authority in the area from 1208 to 1214.⁸⁸ Likewise, when the grand assize, concerning right to land, was brought, knights again were summoned. For example, in Michaelmas term 1206 in Herefordshire, Miles Pichard, Richard de Bagenden, Robert of Weston, and Warin de Grenden (Grendon Warren, in Pencombe) were required to select twelve knights for the hearing of a dispute over 12 carucates in Dormington and Bartestree. The jurors, who would be men from the visnetum, or neighbourhood, and thus presumably knew the facts, included a number of men who were tenants of the Lacy honour of Weobley, including Miles Pichard himself.⁸⁹ The extension and elaboration of royal justice also resulted in the creation of new ⁸⁵ Hudson, 124. Four knights from each hundred were to empanel another ten knights (or free and lawful men if there were not enough knights available) to answer to the articles of the eyre. ⁸⁶ Ibid. 125–6. ⁸⁷ Pleas Before the King or his Justices 1198–1212, ed. D. M. Stenton, iii, Selden Society, 83 (1967), app. I, pp. xlvii–ccxciv. ⁸⁸ See Ch. 6. ⁸⁹ CRR iv. 293.

The Role of Royal Government

157

ofﬁces, such as the coroner or keeper of crown pleas, who again was a local knight elected in the county court.⁹⁰ Thus, new avenues of crown patronage were opened up to the knights.⁹¹ Royal justice also had a ﬁnancial aspect to it. Writs had to be purchased, and proffers made to delay or expedite the hearing of cases or to secure a change of venue. For example, in Michaelmas 1176 John Pichard offered 100s. for an action of right for a knight’s fee that Ralph de Baskerville held.⁹² The increasing popularity of the new judicial forms brought considerable sums into royal coffers and, though all levels of aristocratic society used the new weapons of litigation made available to them enthusiastically, the costs to knights and magnates could add up.⁹³ In addition to these costs were amercements handed out by justices in eyre and further amercements for forest offences. In Michaelmas 1177 Ralph de Baskerville was amerced 20s. because he did not respond sufﬁciently to the royal justices then on eyre for his hanging of a man.⁹⁴ The 1177 eyre in Herefordshire also levied a number of amercements for forest offences, despite the relatively small amount of afforested land in the county; the list of those in mercy included Gilbert de Monmouth, Richard de Escotot, Ralph le Bret, Ralph de Baskerville (who was amerced 80 marks), and Robert de Baskerville.⁹⁵ The number of men amerced for forest offences in a county with relatively little forest is as much an indicator of the way the Angevin kings seized on all sources of revenue available to them as it is any lawbreaking on the part of the men of the county. At times money would be owed for no speciﬁed offence but merely ‘for having the benevolence of the king’. These quasi-judicial sums were one of the clearest statements of Angevin ‘absolutism’, and could be massive. In Michaelmas 1198 Walter de Lacy owed 3,100 marks pro habenda benevolentia Regis and for seisin of his land, a sum which he had reduced to £208 by 1208.⁹⁶ ⁹⁰ The coroner was required to inquire into any instances where a crown plea was involved, in the case of murder taking the accused into custody, and attaching others by surety to appear before the king’s justices in eyre when they should next visit. ⁹¹ See the section on royal patronage below. ⁹² Pipe Roll, 22 Henry II, 43. ⁹³ See Ch. 4. ⁹⁴ Pipe Roll, 23 Henry II, 53. ⁹⁵ Ibid. 54–5. ⁹⁶ Pipe Roll, 10 Richard I, 213–14. This ﬁne was probably the result of Walter’s support for John de Courcy in the internecine ﬁghting in Ireland in 1195; DNB, s.v. Walter de Lacy. The account for the following year, however, only shows Lacy as owing £1,200, or some 1,800 marks, indicating either a large payment or a relaxation of the debt on John’s accession; Pipe Roll, 1 John, 216–17. This debt was the reason why in the marriage contract between Walter and William de Braose by which Walter was to

158

Lords of the Central Marches

An examination of exchequer records for 1211, when King John’s efforts to raise money for the planned expedition to the Continent of 1212 were at their peak, demonstrates both how justice in the broad sense, as expressed through the eyre, contributed a signiﬁcant income to the royal coffers and just how great the burden could be to the aristocracy.⁹⁷ A number of debts were lumped together in a list of amercements levied by the earl of Winchester and his associates. John de Monmouth owed 1,000 marks, six destriers, and ten hunting-horses for having the king’s benevolence. Walter de Clifford had been amerced 23 marks, and Walter de Baskerville’s Jewish debts were scrupulously kept track of, understandable when it is considered that he owed a little over £390, a crippling sum for a knight.⁹⁸ The very level and scope of ﬁnancial exactions created opportunities for the king to dispense patronage. Stephen Devereux, alone of all the tenants of the honour of Weobley, was pardoned 4 marks for scutage, while John de Monmouth, already burdened with debt, was given the small sign of favour of having his scutage quit by royal writ.⁹⁹ And across the Wye, those holding from the former Braose lordships in the March were also having scutage assessed on them. However they were created, whether by wardship, escheat, or in this case forfeiture, direct royal ties to knights were exploited most intensively in wartime. Using these ties, as many landholders as possible were summoned for service in the royal host, and those who wished to avoid service had to pay a personal ﬁne much larger than ordinary scutage.¹⁰⁰

ROY A L PAT RO N AG E The extension of increased royal government to Herefordshire created new patronage opportunities, in addition to those provided by royal lordship. For example, Roger Pichard was a coroner prior to 1229, when the sheriff of Hereford was ordered to have a new coroner elected to replace him. In 1231 he was again coroner, but was replaced the be married to William’s daughter Margaret, Walter was forbidden to sell or alienate any of his land without William’s permission; Rot. Chart. 80. ⁹⁷ Pipe Roll, 13 John, 231–5 (Herefordshire). ⁹⁸ Ibid. 233. ⁹⁹ Ibid. 234–5. ¹⁰⁰ Hugh M. Thomas, Vassals, Heiresses, Crusaders, and Thugs: The Gentry of Angevin Yorkshire, 1154–1216 (Philadelphia, 1993), 170–1.

The Role of Royal Government

159

following year because he was a member of the familia of Hubert de Burgh (qui est de familia justiciarii). There may well have been a link between Roger Pichard’s service in Herefordshire and his move into the household of the justiciar.¹⁰¹ Of course, patronage was a reciprocal relationship. Those who received it were more likely to remain loyal to the Crown in times of political unrest, as the Mortimers did in the latter half of John’s reign.¹⁰² But, closeness to the king and service to the Crown were also a usual prerequisite of patronage. Roger de Mortimer had been an adherent of John before he was king. William III de Braose, the greatest beneﬁciary of royal patronage in the March (and Ireland) under King John, until John turned on him, was also close to John before 1199 and almost certainly played an important role in securing his succession to the throne over Arthur of Brittany.¹⁰³ Some families, usually newcomers, depended entirely on the king for their status in the region. An excellent example of the importance of royal favour is that of the Cantilupes, a family of ‘administrative’ barons ﬁrst inserted into the region by King John. William I de Cantilupe, the son of a minor landholder in Essex and Lincolnshire, had been a steward in the household of John when he was count of Mortain.¹⁰⁴ In 1205 he was undersheriff of Herefordshire, his ﬁrst appearance in the border.¹⁰⁵ He would later witness the treaty of Worcester (1218) with Llywelyn the Great, be charged with investigating encroachments upon royal forest in Herefordshire, and serve on the royal expeditions to Wales in 1228 and 1231.¹⁰⁶ Later, as a steward for Henry III, William was well placed to receive the fruits of patronage, and in 1238 his son William II de Cantilupe was himself made a steward of the royal household.¹⁰⁷ In that same year William II was granted the custody and marriage of Eva de Braose, one of the heirs of William V de Braose (d. 1230), along with the honour of Abergavenny.¹⁰⁸ Sometime between 1238 and 1245 William II married Eva to his son William III de Cantilupe, at a stroke enlarging the family holdings well beyond what they had ¹⁰¹ CR 1227–31, 143, 476; CR 1231–34, 67. ¹⁰² See above, p. 149. ¹⁰³ For royal patronage to William de Braose, see Ch. 6; for the suggestion that William was instrumental in helping John to the throne, Margam Annals in Annales Monastici, i. 24. The uncertain succession of 1199 is ably treated in Holt, ‘The Casus Regis: The Law and Politics of Succession in the Plantagenet Dominions, 1185–1247’, repr. in Holt, Colonial England 1066–1215 (1997), 307–26, esp. 307–13. ¹⁰⁴ Itinerary of Richard I, ed. Lionel Landon, Pipe Rolls Series (1935), 131. ¹⁰⁵ Pipe Roll, 7 John, 271. ¹⁰⁶ Rot. Litt. Claus. i. 379; CR 1227–31, 115, 546. ¹⁰⁷ Stacey, Politics, Policy and Finance under Henry III, 63 n. 68. ¹⁰⁸ CPR 1232–47, 228.

160

Lords of the Central Marches

been under William I. Royal service, combined with mortality in one of the established Marcher families, had led to the introduction of a new Marcher lord. Marriage was also a means of linking prominent local families to the court. As already noted, two important members of Herefordshire knightly families, Stephen Devereux and Walter II de Baskerville, married into curialist circles in John’s reign. Stephen Devereux took as his second wife Isabelle de Cantilupe, probably a sister of William I de Cantilupe.¹⁰⁹ As noted above, William de Cantilupe was an important member of the royal household and Stephen was thus making a connection to the court by his marriage to Isabelle.¹¹⁰ Walter II de Baskerville of Eardisley married Susanna de Chanceaux in 1213. In that year Walter’s father, Walter I, died leaving the under-age Walter II as his heir. The honour of Weobley, of which Eardisley was a component manor, was in royal hands at the time owing to the forfeiture of Walter de Lacy. The wardship was promptly bought by the curialis Andrew de Chanceaux, who then married the boy to his sister Susanna. In this case we may suspect the inﬂuence of Andrew de Chanceaux’s kinsmen Gerard d’Athée and Engelard de Cigogné, who were administering this region of the March for the king.¹¹¹ The award of custodies and marriages was the most dramatic form of royal favour. But there was an entire spectrum of patronage which the king used to inﬂuence and to bind men to the Crown. Temporary custody of conﬁscated land, land that could be exploited by its custodian, was one element of patronage. For example, the Herefordshire knight Matthew de Gamages had been an adherent of John in the early 1190s.¹¹² But the Gamages family also held land in Normandy, and Matthew adhered to Philip Augustus after 1204. Thus, in 1205 £10 of land which Matthew de Gamages had held in the manor of Dilwyn in Herefordshire was granted to William ﬁtz Warin, lord of the barony of Whittington in Shropshire.¹¹³ Later in 1205 Walter de Clifford ¹⁰⁹ See Ch. 4; Crouch, William Marshal, 203. ¹¹⁰ See Ch. 4. ¹¹¹ For the Baskerville–Chanceaux marriage, see Ch. 4; for d’Atheé and Cigogné in the March, Ch. 6. ¹¹² His lands in Herefordshire and Shropshire were taken into the king’s hand in the spring of 1194; they were not restored to him until mid-1197; Roll of Escheats, Pipe Roll, 6 Richard I, 5; Pipe Roll, 9 Richard I, 194. ¹¹³ Rot. Litt. Claus. i. 25. For the ﬁtz Warins, see Janet Meisel, Barons of the Welsh Frontier: The Corbet, Pantulf and Fitzwarin Families, 1066–1272 (Lincoln, Nebr.,1980), 34–54.

The Role of Royal Government

161

was granted Matthew’s house and land at Bouret in Herefordshire.¹¹⁴ The same process occurred in 1215–16 as rebel lands were parcelled out to loyal supporters. Stephen Devereux was awarded all the land in Bullinghope and Clehonger that had been held by the rebel Walter de Stokes, and also received £20 worth of land in Stanton in Worcestershire which had been held by Peter of Stanton.¹¹⁵ The custody of Walter de Stokes’s wife, with her lands, was delivered to the royalist Marcher John de Monmouth.¹¹⁶ Loyalty to the king had its rewards. An important aspect of royal patronage was the forgiveness of debts, whether owed for proffers or ﬁnes, or to the Jews. Debt was a doubleedged tool for the Angevin kings: its imposition served to punish recalcitrant vassals, while its relief or pardon could be used to reward those who had done the crown service. The king could, for example, give more lenient terms to a debtor. Walter de Lacy, who as well as being a Marcher baron was lord of Meath in Ireland, owed by 1228 a total of 7,000 marks in ﬁnes for the restoration of his Irish lands.¹¹⁷ The original terms of repayment, set in March 1228, were 500 marks payable at the Dublin exchequer at Michaelmas and Easter; by September the rate of repayment had been reduced to 400 marks each term.¹¹⁸ In 1233, when Richard Marshal’s rebellion had broken out along the March, some of Lacy’s other debts were either pardoned or assigned new terms for the repayment.¹¹⁹ Earlier in December 1233 Walter III de Clifford, who had been in rebellion and had then returned to the king’s allegiance, had his Jewish debt prorogued and then pardoned, as he was now in royal service at Hereford.¹²⁰ The large level of debt of some of the key men of Herefordshire and the March gave the king a ready handle on them in times of crisis.¹²¹ There were also lesser degrees of patronage. Customary dues owed the king could be excused, as when in July 1214 Stephen Devereux and his heirs were quit in perpetuity of their annual render of honey owed from their manor of Ballingham to Hereford Castle.¹²² Other ¹¹⁴ Rot. Litt. Claus. i. 41. ¹¹⁵ Ibid. 246, 279. ¹¹⁶ Ibid. 271. ¹¹⁷ CR 1227–31, 26–7. He ﬁned for restoration on two separate occasions, once in 1214 and again in 1225 following the rebellion of his brother Hugh de Lacy, the exiled earl of Ulster; see Ch. 7. ¹¹⁸ CR 1227–31, 80. ¹¹⁹ CR 1231–34, 351–2. ¹²⁰ Ibid. 350, 351. ¹²¹ See also Ch. 7. This debt did not exist solely at the level of the magnates of the region. Walter I de Baskerville was in debt to the Crown by 1205, and his son Walter II was accounting to the exchequer in 1218 for his father’s Jewish debts; see Ch. 4. For treatment of the Jewish debts of area notables, see Hillaby, ‘Hereford Gold’, pts i and ii. ¹²² Rot. Litt. Pat. 147b.

162

Lords of the Central Marches

economic rewards were in the king’s gift. In 1227 Stephen and his heirs were granted a weekly market and a yearly fair at the family seat of Lyonshall, while in 1246 Walter III de Baskerville was given royal licence to impark his wood of Combe Baskerville in Gloucestershire.¹²³ Important knightly families that helped to discharge royal government in the locality were rewarded, and such grants both conferred economic advantages for the recipient and displayed the standing of the grantee in the eyes of the king. And we should not overlook the numerous gifts by the king of luxury goods such as venison, wine, and timber to magnates and knights.¹²⁴ One can be certain that over dinner in his hall the Marcher lord or Herefordshire knight would be certain to point out to his guests from where he had obtained the venison or wine on the table, or the oaken beams supporting the roof over their heads. Royal patronage enhanced the social standing of its recipients in the localities. Patronage and service were inextricably linked. Thus, Stephen Devereux served with the king in Poitou in 1214 and supported John during the civil war, and later served in various local ofﬁces.¹²⁵ Both Stephen and Walter II de Baskerville helped to collect royal taxation, and Walter served as sheriff of Herefordshire, supported the king against Richard Marshal in 1233–4, and served as a negotiator with the Welsh.¹²⁶ It is no coincidence that Stephen Devereux and Walter de Baskerville were the recipients of royal patronage, for they were two of the knights most frequently called upon to carry out tasks for the king in Herefordshire. Service brought rewards, and the promise of patronage attracted men into royal service. In this way the king forged links with the knights in the localities, albeit only a small number, usually of the more inﬂuential ones. Therefore, those barons and their knights who held land in Herefordshire had to deal with an increasingly powerful, well-organized, and intrusive royal government. They were called upon to support royal military expeditions and at the same time came to depend more and ¹²³ Rot. Litt. Claus. ii. 189b; CPR 1232–47, 476. ¹²⁴ From 1223, see e.g. Rot. Litt. Claus. i. 507 (a gift of two stags to Walter de Clifford), 539b (a gift to John de Monmouth of ten ‘wind-fallen’ oaks from the Forest of Dean), and 555b (a gift to Walter de Lacy of forty cartloads of dry underwood for burning). ¹²⁵ See Ch. 4. In 1219 he was appointed a forest commissioner for the eyre in Herefordshire, and was a justice of gaol delivery in 1220; Rot. Litt. Pat. 215, 437. ¹²⁶ Rot. Litt. Claus. ii. 74b; CR 1231–34, 174; CR 1234–37, 369.

The Role of Royal Government

163

more on the Crown for defence against the Welsh. A change in the pattern of shrieval appointments from baronial sheriffs to knights and lesser curiales brought that ofﬁce more under the control of the Crown. Royal justice, extended into Herefordshire by frequent judicial eyres, sharply delineated the border with the March. Angevin government in turn forged links with the more important local knights, and royal patronage and the occasional marriage into court circles, as well as wardships and custodies, helped to make the king’s lordship felt to a higher degree in the late twelfth and early thirteenth centuries than previously. All of this was in contrast to the much more limited interaction between the Crown and the lordships of the south-eastern March.

T H E M A RC H The increasingly intrusive nature of Angevin government outlined above had a much smaller impact on the lordships of the March. Firstly, the absence of signiﬁcant royal territorial interests in Wales prevented the king from exerting a more direct control.¹²⁷ Secondly, part of the process of consolidation of lordship in the March was the assertion of Marcher ‘liberties’. As the custom of Marcher independence hardened into an unwritten legal right in the late twelfth and early thirteenth centuries, the ability of an increasingly intrusive royal government to intervene directly in the Marcher lordships was severely restricted. The king’s inﬂuence, however, was never entirely shut out and, in the case of virtually all the kings from 1166 to 1246, with the exception of Richard I, he might even decide to override those cherished Marcher customs and liberties.¹²⁸ The areas where this was most likely to occur was in the operation of the king’s feudal lordship, and in the course of royal policy towards the Welsh princes.¹²⁹ ¹²⁷ Davies, Domination and Conquest, 69. ¹²⁸ For John, see Ch. 6. For Edward’s policies in the March, see Michael Prestwich, Edward I, new edn (New Haven, 1997), 202–32; Davies, The Age of Conquest, 333–88. It should be noted, however, that Marcher lordships were such a plum of patronage, if only occasionally, that Edward was never tempted to overturn the system of Marcher lordship (indeed he created new ones, such as Bromﬁeld and Yale, attendant on the conquest); rather, he maintained his ultimate lordship by some well-chosen raps on baronial knuckles. ¹²⁹ See above, pp. 146–9.

164

Lords of the Central Marches

On his way to Ireland in 1171, Henry II appointed royal wardens to take the baronial castles of south Wales into his hands, and this served to remind the Marchers that they were ultimately subject to royal control.¹³⁰ Even Henry III, through manipulation of debt for instance, managed to keep the Marcher lords of the region largely under control during the rebellion of Richard Marshal in 1233–4.¹³¹ A direct royal challenge to Marcher liberties was rare, however. Rather, the king used other methods to control and inﬂuence the great men of the south-eastern March: his diplomacy with the Welsh and leadership of military expeditions against them; his position as feudal overlord and the attendant direct administration or disposal of escheats and wardships; and patronage, as shown by the intrusion of new men into the region or the attraction of men from the region into the circle of royal favour.¹³² All these features are brought into sharp focus in the reign of King John, to be discussed below.¹³³

C O N C LU S I O N The king played a greater role in the shaping of the March than is sometimes allowed. Although the customs of the Marcher lords were largely respected and their independence at times given free rein, most notably under Richard I, they were still tenants-in-chief and could not afford to ignore the king. The growth of Angevin government and particularly Angevin justice helped to deﬁne further the separateness of the March and its society, a separation that had largely been effected already by the division of inheritances along tenurial lines and a very high degree of intermarriage among Marcher families at the knightly level. The kings of England took varying approaches to the March. Henry I intruded new men into the region on a scale not seen again until the middle years of Henry III’s reign. Henry II halted Marcher ambitions by reaching an accommodation with the Lord Rhys, a satisfactory agreement as long as the two men lived. Richard I had too pressing concerns elsewhere, and for a space of time the ¹³⁰ Giraldus Cambrensis, Expugnatio Hibernica, 90–1. ¹³¹ See Ch. 7. ¹³² A similar process of patronage worked in Ireland; see Robin Frame, ‘Aristocracies and the Political Conﬁguration of the British Isles’, in Frame, Ireland and Britain 1170–1450 (1998), 151–70. ¹³³ See Ch. 6.

The Role of Royal Government

165

March reverted to a cockpit for baronial ambitions and vendettas. But an ever-present problem for the Norman and Angevin kings of England in controlling Wales was ‘the absence of strong royal territorial interests there’.¹³⁴ That was a situation which King John, whether by conscious design or not, was to change, and is the subject of the next chapter. ¹³⁴ Davies, Domination and Conquest, 69.

6 The Region in the Reigns of Richard I and John, 1189–1216 Between the death of Henry II in 1189 and that of his son King John in 1216 the fortunes of the Braose and Lacy families changed greatly.¹ On the other hand, the Cliffords, Monmouths, and Mortimers held their ground and, during the tumultuous events of John’s reign, positioned themselves advantageously for the future by developing closer links to the Crown. As noted above, Richard I’s reign was marked by a turning away from the policy of royal control and containment of Marcher ambitions which Henry II had pursued since 1172.² This in effect licensed the magnates of the south-eastern March to resume their ambitions for the conquest and domination of neighbouring Welsh regions. John’s reign in turn was marked by the rise of one of these magnates, his favourite William III de Braose, to even greater heights of landed power, in the most dramatic exercise of patronage by that king. The fall of Braose from royal grace, the subsequent persecution of his family, and the destruction of the fortunes of his house in many ways dominates the history of the March in this period. Outside this episode, however, few historians have, until recently, given John’s actions in Wales and the west of the British Isles the coherent attention which they deserve.³ The destruction of the house of Braose has attracted the attention of historians since the thirteenth century. One of John’s biographers, for ¹ Much of the argument and evidence in this chapter appeared previously in Brock W. Holden, ‘King John, the Braoses, and the Celtic Fringe, 1207–1216’, Albion, 33/1 (Spring 2001), 1–23. ² See Ch. 5. ³ But see the comments of Davies, The Age of Conquest, 292–3, that ‘The reign of King John was a formative period in Anglo-Welsh relations’ and that John ‘had laid the groundplans of a military conquest and settlement of the country which it only remained for Edward I to copy and put fully into operation’. See also I. W. Rowlands, ‘King John and Wales’, in Church (ed.), King John, 273–87.

The Reigns of Richard I and John

167

example, described the quarrel between John and William de Braose as ‘the greatest mistake John made during his reign’, as the king revealed to his barons once and for all his capacity for cruelty.⁴ John’s attack on the Braoses, however, was signiﬁcant not merely for the effect it had on relations with the baronage, but even more importantly because the elimination of the Braoses created a vacuum of power along the March, drawing in the Crown. This in turn led to John’s short-lived triumph of 1211 in Wales, and its subsequent reversal by Llywelyn ap Iorwerth in 1212, a reversal which was quite damaging to the king in the years immediately preceding the baronial rebellion of 1215. As a result of the removal of the Braoses in 1208, royal policy in Wales became directly and overwhelmingly interventionist; the character of the relationship between Crown, Marcher barons, and Welsh princes was utterly changed. Establishing the context of the quarrel between King John and William de Braose requires us ﬁrst to look at the March of Wales in the reigns of John’s father and brother. We have already seen how Henry II’s expeditions into Wales, and his decision to invade Ireland in 1171, established a d´etente with the paramount Welsh prince, Rhys ap Gruffudd.⁵ This d´etente in turn largely closed off the frontier along the March and on the whole effectively halted Marcher expansion. This is not to say that conﬂict between Welsh and Marchers ceased, as the massacre of a Welsh chieftain and his family by the men of William III de Braose at Abergavenny in 1175 shows.⁶ Despite ambushes and vendettas, the scale of conﬂict was much reduced in the latter part of Henry II’s reign, and Marcher lords such as Hugh II de Lacy, lord of Weobley and Ewyas Lacy, turned their attentions to Ireland. The accession of Richard I in 1189 resulted, intentionally or not, in the renewal of private baronial initiative along the March.⁷ Richard wanted to reach an accommodation with the Welsh so that he would leave a peaceful border while on crusade. The Welsh threat to England’s borders had indeed provided an excuse for Henry II to delay his ⁴ Painter, The Reign of King John, 249–50. ⁵ See Ch. 5; Warren, Henry II, 165–9; Davies, The Age of Conquest, 289–92. Indeed, on Henry’s return from Ireland in 1172 the English king made Rhys his ‘justice of south Wales’, a somewhat ambiguous title but one which W. L. Warren thought made the Welsh prince ‘a sort of royal proconsul in the south’; Henry II, 167. ⁶ Brut, 71, s.a. 1175. ⁷ John Gillingham, not surprisingly, has exculpated Richard from responsibility for the escalation of conﬂict along the March in the 1190s, and his argument is persuasive; ‘Henry II, Richard I and the Lord Rhys’, 225–36.

168

Lords of the Central Marches

departure on crusade, and on hearing the news of Henry’s death the Welsh, under the leadership of the Lord Rhys, immediately rose. In one sense, at least, Richard was successful—the Welsh did not intervene when John rebelled in 1193–4.⁸ What matters, however, is that the general d´etente of the 1170s and 1180s was at an end. The Crown lost control over the Marcher lords. In the mid-1190s two men in the region took particular advantage of this loss of control, Roger de Mortimer of Wigmore and William de Braose, lord of Radnor, Brecon, and Abergavenny in the March. Both men returned to the long-held ambitions of their families, Mortimer launching a drive into Maelienydd while Braose turned to the conquest of Elfael.⁹ This activity culminated in the Lord Rhys’s last great victory, the crushing in open battle of the army of Roger de Mortimer and Hugh de Say at Radnor in 1196. The year 1189 was important for Marcher affairs in another way, for it marked the introduction into the region of the king’s brother John, count of Mortain.¹⁰ On 29 August 1189 John was married to Isabelle of Gloucester, heir to the earldom of Gloucester and the lordship of Glamorgan; John was now a Marcher lord. John already had brief experience of Marcher affairs, having passed through south Wales on his way to Ireland in 1185, and having been sent with an army by his brother King Richard to deal with the Lord Rhys.¹¹ On this second occasion, John preferred to make a deal with the Welsh leader rather than ﬁght, and it has been suggested that the reason for John’s inactivity against the Welsh was his concern with his position in England while Richard was on crusade.¹² If this is correct, John was looking in any struggle for power to ﬁght his corner from the March as well as from his other lands. John also began to forge ties with local men, and it is possible to trace some of these connections, which are apparent later in ⁸ John Gillingham, Richard I (New Haven, 1999), 112–13. ⁹ Brut, 75, s.a. 1195; Smith, ‘The Middle March in the Thirteenth Century’, 80. ¹⁰ When Richard I left England on crusade in 1190, he attempted to secure the kingdom by means of a delicate balancing act. William de Mandeville, earl of Essex, and the bishop of Durham, Hugh du Puiset, were left as co-justiciars, while William de Longchamp, the new king’s old servant from his days in Aquitaine, was elevated to the see of Ely and a grant of legatine power was procured for him from the Pope. John was given a large landed settlement, consisting of the ﬁefs that he already held, conﬁrmation of his status as lord of Ireland, and a grant of the honour of Lancaster as well as six counties (Nottinghamshire and Derbyshire, Dorset, Somerset, Devon, and Cornwall). Richard originally meant to exclude his brother physically from the kingdom, but his mother, Eleanor, intervened and convinced him to relent. ¹¹ Annales Cambriae, ed. J. Williams ab Ithel, Rolls Series (1860), 57. ¹² Lloyd, A History of Wales, ii. 575.

The Reigns of Richard I and John

169

his own reign, to the period of crisis during Richard I’s absence in the early 1190s. These early ties are important, for they may help to explain the behaviour of local men during the crisis of 1208–10. The ﬁrst of these connections was with the Mortimers. According to Richard of Devizes, when the conﬂict between John and Richard’s chancellor William Longchamp began, Longchamp marched immediately against Wigmore Castle, the Mortimer caput.¹³ Roger de Mortimer was charged with forming a conspiracy with the Welsh against the king, and was forced into a three-year exile. Devizes named Mortimer as one of John’s accomplices, and says that he was blamed by his fellow conspirators for ‘faintness of heart’ (cordis inopia).¹⁴ Mortimer was not exiled long, however, and had been readmitted to the kingdom by early 1193. In spring 1193 Mortimer took possession of the manor of Ellesmere in Shropshire during the king’s pleasure.¹⁵ By mid-1194, however, the wife of the previous holder of Ellesmere, David ap Owen, was in possession of the manor.¹⁶ In 1194 Roger de Mortimer was with the king’s army in Normandy.¹⁷ Despite his problems with the king in this period, Roger’s links to the Crown had been forged and, perhaps more importantly, his loyalty to John consolidated. Despite the fact that his son was married to a Braose daughter, Roger de Mortimer would stand to one side in the crisis of 1208–10 between William de Braose and John, and indeed would send knights to serve with the king in Ireland.¹⁸ Matthew de Gamages, a knight of local importance in Herefordshire and Gloucestershire, was also probably in league with John.¹⁹ Matthew’s lands in Herefordshire and Shropshire were taken ¹³ The Chronicle of Richard of Devizes of the Time of King Richard the First, ed. and trans. J. T. Appleby (1963), 30–1. ¹⁴ As it was, Mortimer’s exile seems to have lasted less than three years, as the account for the defence of Swansea Castle rendered by William Marshal to the exchequer at Michaelmas 1193 has an entry for 10 marks paid to Roger per breve Regis; Pipe Roll, 5 Richard I, 148. These expenses were associated with the defence of Swansea Castle against the Lord Rhys earlier in the year. ¹⁵ Pipe Roll, 5 Richard I, 110. The manor for a half-year was worth £813s. 4d . ¹⁶ Pipe Roll, 6 Richard I, 140. ¹⁷ He was quit of scutage in Lincolnshire and Berkshire; ibid. 119, 258. ¹⁸ Roger’s son Hugh de Mortimer (d. 1227) was married to Annora de Braose, though the date is uncertain. ¹⁹ Matthew was the eldest son of Godfrey de Gamages, who in 1166 held two knights’ fees at Mansell Gamage, Herefordshire, of the Lacy honour of Weobley (RBE 281) and one fee of the Braose lordship of Brecon (RBE 601). There were also family lands in Gamaches in Normandy; in 1204 Matthew would adhere to Philip Augustus, and after a long struggle in the royal courts his brother William gained possession of the family’s English estates.

170

Lords of the Central Marches

into the king’s hands in the spring of 1194, and were not restored to him until mid-1197.²⁰ It is tempting to associate this forfeiture with John’s scheming in 1193. Matthew’s younger brother William was later a knight of King John’s household and served in Ireland in 1210. It is reasonable to suppose that the family connection dates from the support of his brother Matthew for the rebellious count of Mortain in the early 1190s.²¹ Finally, William de Braose appears in the witness list of a conﬁrmation of John’s as count of Mortain to Margam Abbey of land in Kenﬁg, dated 4 March 1193 at Cardiff.²² It is not at all certain whether this William was William III or his son, often referred to as William juvenis. This is, however, our ﬁrst indication of a link between John and the Braoses. In the 1190s John already had begun cultivating a network of friends and potential allies in the border and March. King Richard, or his chancellor, also seem to have interfered with at least one inheritance in the region, causing a situation which John capitalized upon when he came to the throne. Walter I de Clifford, lord of Clifford and Cantref Selyf in Brecon, died in 1190. His heir was Walter II, but Walter II’s brother Richard managed to gain control of the family’s Shropshire manors of Corfham, Culmington, and Clive, and the hay of Ernestou, in return for a 300-mark ﬁne.²³ It has been speculated that the reason for the acceptance of Richard de Clifford’s proffer lay in King Richard’s desire for vengeance upon the Clifford heir, for Walter and Richard de Clifford’s sister was the famous Rosamund.²⁴ Whether or not this was true, the resulting dispute between the two brothers was reopened upon John’s accession to the throne, when the new king granted the Shropshire manors to Walter.²⁵ Although Richard brought suit against Walter for the manors the following year, the two men reached a ﬁnal concord in Michaelmas 1200, with Walter keeping ²⁰ Roll of Escheats, Pipe Roll, 6 Richard I, 5; Pipe Roll, 9 Richard I, 194. ²¹ S. D. Church, ‘The Knights of the Household of King John: A Question of Numbers’, in P. R. Coss and S. D. Lloyd (eds), Thirteenth Century England IV (1992), 165. ²² Earldom of Gloucester Charters: The Charters and Scribes of the Earls and Countesses of Gloucester to A.D. 1217, ed. R. B. Patterson (Oxford, 1973), 127, no. 138. ²³ Pipe Roll, 2 Richard I, 126. See the Clifford genealogy in Appendix 2. ²⁴ Eyton, Antiquities of Shropshire, v. 151–2. The Shropshire manors were an acquisition of the brothers’ father, however, so it was within feudal custom for him to settle them upon his younger son. ²⁵ In a charter dated at Cambrai on 3 Aug. 1199; Rot. Chart. 8. This is a very curious charter as it treats Walter as the heir of Richard, as if Richard were dead. Eyton (Antiquities of Shropshire) speculated that there were two Richards, but there is no evidence to support this theory.

The Reigns of Richard I and John

171

the Shropshire manors in return for the grant of the manor of Frampton in Gloucestershire.²⁶ Walter paid 300 marks for royal conﬁrmation of his possession of the Shropshire manors.²⁷ By granting the Shropshire manors to Walter, and perhaps getting the parties to reach an agreement, King John secured Walter de Clifford’s loyalty. Walter de Clifford (or his son) would serve in Ireland, and his son Walter III would be sheriff of Hereford for a time during the civil war (1215). Walter III and his brother Roger were also knights of the royal household.²⁸ One ﬁnal event of Richard I’s reign involving a magnate in our region was to have consequences into the reign of King John. For reasons having to do with the vagaries of Irish politics, Walter II de Lacy attracted the displeasure of Richard in 1198–9. In 1199 Walter was compelled to proffer an enormous ﬁne of £1, 200 ‘for having the benevolence of King Richard and seisin of his land’.²⁹ This was a crippling amount for a baron of Walter’s standing, yet he steadily paid it off through the early years of John’s reign. By Michaelmas 1208 Walter had reduced the debt to £208. How did Walter manage to pay £1,000 into the exchequer over eight years? Some must have come from his intensive exploitation of his estates in Meath in Ireland.³⁰ But it is also probable that a good portion of the money was borrowed from the Jews. Walter de Lacy was plagued by debt throughout his life, much of it incurred for political miscalculations. When he died, he still owed some £666 to the Jews of Hereford.³¹ Ultimately, however, it was no escape to borrow from the Jews to pay the king.³² John’s later use of the lex scaccarii to persecute William de Braose must have struck uncomfortably close to Walter de Lacy, his son-in-law.³³ Upon John’s accession to the throne his greatest supporter in the March and south Wales was William III de Braose. In 1199 William was lord of the rape of Bramber in Sussex and of the Marcher lordships of Radnor, ²⁶ For the service of one knight, Rot. Cur. Reg. ii. 159, 171; ﬁnal concord, TNA, CP 25(1)/193/2, no. 27. ²⁷ Pipe Roll, 1 John, 76. ²⁸ Church, ‘The Knights of the Household of King John’, 163. ²⁹ Pipe Roll, 1 John, 216–17. ³⁰ See Ch. 3. Yet the defence of Meath, the transplanting of vassals to the colony, and the construction of Trim Castle were all serious drains on the issues of the lordship; see Hillaby, ‘Hereford Gold’, pt ii, 197; Bartlett, ‘Colonial Aristocracies of the High Middle Ages’, 31–8. ³¹ Hillaby, ‘Hereford Gold’, pt i, 405. ³² Debt to the Jews was usually secured on gauges of land. Upon the Jewish lender’s death, the king inherited his bonds, usually but not necessarily then selling them back to the heirs of the deceased. ³³ See below, pp. 174–6.

172

Lords of the Central Marches

Builth, Brecon, and Abergavenny.³⁴ The two men were both active on the Continent in the later years of Richard I’s reign, and William almost certainly played a role in John’s elevation to the throne. He had been present at Chalus when Richard I died and, together with the few other barons in attendance, may have persuaded King Richard to name John as his successor; or they may have fabricated the entire story.³⁵ William de Braose was correspondingly showered with the rewards of patronage by the new king. In 1200 his son Giles was made bishop of Hereford, with its fees and the castle of Bishop’s Castle in southern Shropshire. That same year William was given permission by the king to add as much land to his barony of Radnor as he could at the expense of the Welsh.³⁶ In January 1201, in return for a 5,000-mark ﬁne, William was granted the honour of Limerick in Ireland, excepting the city of Limerick and the land of William de Burgh. In 1202 William gained the custody of the Beauchamp barony of Elmley in Worcestershire and more signiﬁcantly, the custody of the Marcher lordships of Glamorgan and Gower.³⁷ As a result of these grants, by the end of 1202 William de Braose and his family controlled a remarkably solid and extensive block of land stretching from southern Shropshire to the Gower peninsula, a block broken only by the Marshal lands in Netherwent.³⁸ William was now also a major baron in Ireland. By 1208 the relationship between John and William had turned vicious, and ever since historians have advanced explanations for this turn of events. The ﬁrst such was Roger Wendover, writing in the thirteenth century; according to him, John’s fear of baronial rebellion after the proclamation of the interdict in 1208, and his subsequent demand for hostages, led to the rupture between the two men.³⁹ William in 1202 had captured Arthur of Brittany and turned him over ³⁴ In Normandy, Braose held a barony centred on the caput of Briouze between Falaise and Domfront on Orne, which accounted for three knights’ service in 1172; RBE 631. ³⁵ The Margam Annals indeed credit William and his accomplices with placing John on the throne; Annales Monastici, i. 24. See also Painter, The Reign of King John, 7. He points out that three other men present at Richard’s death—Thomas Basset, Peter de Stoke, and Gerard de Furnival—also did quite well under John, Basset gaining the custody of the earldom of Warwick and the marriage of the heir, Stoke being appointed a royal seneschal, and Furnival gaining a valuable heiress. ³⁶ Rot. Chart. 66. ³⁷ Pipe Roll, 4 John, 20; Rot. Litt. Pat. 19b. ³⁸ See Map 1. ³⁹ Roger of Wendover, Flores Historiarum, ed. H. G. Hewlett, 3 vols, Rolls Series (1886–9), ii. 48–9. Sir James Holt notes that Wendover’s account of a general demand for hostages seems ‘muddled’ and is supported by little evidence; Magna Carta, 82 n. 35.

The Reigns of Richard I and John

173

to the king, and likely knew of the prince’s fate.⁴⁰ Thus, in Wendover’s account William’s formidable wife, Maud, turns away John’s men who have come to take her son hostage, accusing the king of having ‘basely murdered’ his nephew Arthur.⁴¹ This is a highly romantic story, though the king’s starving to death of Maud and her son William after their capture in 1210 certainly indicates a level of personal animosity, if not fury, on the part of John against the mother and son. The idea, however, that John moved against the Braoses because Maud threatened to reveal the king’s murder of Arthur of Brittany is far-fetched. Other sources speak only, if at all, of an unspeciﬁed dispute.⁴² The conclusions to be drawn from the disappearance of Arthur were obvious; Philip Augustus met John’s request in 1204 for a truce with his own request for John to produce Arthur. Arthur’s Breton subjects believed him dead, and most of the English baronage must have assumed the same. Those barons who were ordered to turn over hostages to John in 1208 complied, including Robert de Vieuxpont, who had been keeper of Rouen when Arthur was imprisoned there.⁴³ Among modern historians, Sidney Painter saw the events of 1207 to 1210 in terms of John’s distrustful nature and sense of realpolitik. According to Painter, the build-up of Braose early in the reign made him a counterweight in the March to Ranulf, earl of Chester, who was out of favour from 1199 to 1205.⁴⁴ After Chester’s political rehabilitation in 1205 the need for Braose as a counterweight on the March was lessened.⁴⁵ Also, this strategy of balancing the two men seems to have caused the king as many problems as it solved, as the aims of Braose ⁴⁰ Ralph of Coggeshall, Chronicon Anglicanum, ed. J. Stevenson, Rolls Series (1875), 137–8. ⁴¹ Roger of Wendover, Flores Historiarum, ii. 48–9. F. M. Powicke, relying on Wendover along with the Margam Annals and French royal chronicles, believed that William de Braose’s knowledge of Arthur of Brittany’s fate, and his wife’s threat to divulge that knowledge, was the cause of the downfall of the Braoses; The Fall of Normandy (Manchester, 1913), 453–81, esp. 476–71. ⁴² ‘mota discordia inter ipsum [William] et regem’ (Annales Cambriae, 66, s.a. 1208). ⁴³ For Vieuxpont, Rot. Litt. Pat. 89b. ⁴⁴ Painter, The Reign of King John, 25–9. Ranulf was out of favour because of inheritance disputes (over the Roumar inheritance and the honour of Richmond) from 1199 until a rapprochement was effected in 1205 (one month after the death of Robert, earl of Leicester, to whom Richmond had been awarded); afterwards Ranulf was to be steadfastly loyal to the king. ⁴⁵ Richard Eales has pointed out to me that John’s policies in south Wales must have been shaped in relation to Cheshire, where it seems that 1205–15 was a ‘critical period’ in the formation of the liberties of the later palatinate.

174

Lords of the Central Marches

and Chester in Wales ran directly counter to each other.⁴⁶ According to Painter, by 1205 William de Braose was too powerful (albeit through the king’s favour), and this worked on John’s paranoid nature. W. L. Warren widened the scope of explanation by focusing on Ireland.⁴⁷ Warren stressed that the loss of Norman estates in 1204 by the Braose, Lacy, and Marshal families led to an increased exploitation of their Irish lands. This exploitation threatened royal policy in Ireland, which (according to Warren) relied on fair treatment of and cooperation with the native Irish.⁴⁸ Thus, ‘John’s expedition of 1210 had the two-fold objective of putting the barons in their place and recovering the conﬁdence of the Irish.’⁴⁹ A recent reappraisal of John’s 1210 expedition, however, has convincingly shown that John’s demand for hostages led to a quick breakdown in relations with the Irish kings.⁵⁰ In 1210 the king seems not to have treated the native Irish much better than he did the Anglo-Irish. There is, of course, one other explanation for the king’s feud with William de Braose—that made by John himself in a document of 1212.⁵¹ The king stated that the cause of the dispute was William’s tardiness in paying the ﬁne of 5,000 marks which he had made in January 1201 for Limerick.⁵² John thus portrays his actions against Braose as ⁴⁶ Chester’s expansion in north Wales was opposed by the princes of Gwynedd, while further south Braose threatened the rulers of Powys and the princelings of south Wales. Therefore, it was natural for the two men to seek alliances with the other’s enemies in an attempt to play the Welsh against each other. By 14 Dec. 1204 the king had to order a partial distraint upon Ranulf’s lands in England and the arrest of any vassals performing him service because Ranulf was allied with Gwenwynwyn of Powys, who had been engaged in a war with the Braoses since late 1202; Rot. Litt. Claus. i. 16. For a differing interpretation, see Rowlands, ‘King John and Wales’, 275–6. Rowlands sees Chester’s real interests as lying elsewhere than along the March. He does, however, brush over the royal order to distrain of 1204, mentioned below. This writ seems to indicate that Chester was up to more than mere ‘overtures to Gwenwynwyn of Powys’ in that year. ⁴⁷ W. L. Warren, ‘The Historian as Private Eye’, Historical Studies, 9 (1974), 1–18. ⁴⁸ ‘There were to be no private baronial empires’ in Ireland; ibid. 16. ⁴⁹ Ibid. 17. ⁵⁰ Sean Duffy, ‘King John’s Expedition to Ireland, 1210: The Evidence Reconsidered’, Irish Historical Studies, 30/117 (May 1996), 1–24. ⁵¹ Foedera, Conventiones, Litterae et cuiusunque generis acta publica, ed. T. Rymer, rev. A. Clarke, F. Holbrooke, and J. Caley, 4 vols in 7 pts (Record Commission, 1816–69), i. 108–9. The inclusion of this document in both the Red and Black Books of the Exchequer may be explained by the fact that John’s justiﬁcation, citing the law of the exchequer, was a precedent par excellence for future distraints for debt against barons. ⁵² William was indeed delinquent in meeting the terms of his ﬁne; the payments were set at 1,000 marks per annum. William had afterwards received the city of Limerick, which had been excluded from the original grant, for a farm of 100 marks. William had fallen well behind in accounting for the farm of the city as well.

The Reigns of Richard I and John

175

an action for debt, secundum consuetudinem regni et per legem scaccarii, and everything that unfolded followed from William’s resistance to distraint. While many of the details of John’s account for the events of 1208 and 1209 can be veriﬁed, the initial statement that William removed all of his English chattels beyond the reach of royal ofﬁcials beggars belief. Because this is the justiﬁcation for proceeding to move against William’s Welsh chattels, the dubiousness of the ﬁrst element critically undermines the veracity of what follows, in intention if not fact. It should also be noted that where John mentions the demand for two sons of William de Braose the younger and one son of his brother Reginald, he does not mention the demand for the Braose heir himself, William the younger. This is a crucial omission, as will be seen below. In the larger context created by the ﬁnancial pressures of his reign, John’s apologia also raises a few questions. J. C. Holt accepted John’s explanation; ‘the King sacriﬁced him [William] on the altar of his new ﬁnancial policies’.⁵³ The law of the exchequer, an arbitrary law, was used to strike fear into the king’s baronial debtors. If viewed from the perspective of Magna Carta with its emphasis on due process, this view makes sense, but it pre-dates John’s ﬁnancial ruthlessness. A study of proffers for heirs and heiresses, their terms, and the rates of their payment has shown that, although magnate indebtedness increased 380 per cent in the ﬁrst eight years of John’s reign, this was not indicative of John’s squeezing his barons.⁵⁴ The exchequer did not try to increase revenue by accelerating payments, yearly terms and length of pay-off increased little, and inﬂation negated the increase in levels of proffers and payments. A change in royal ﬁnancial policy towards the barons does not seem to have occurred until 1211, when ‘John foolishly turned patronage management into a weapon.’⁵⁵ This is not to say that the attack on Braose could not have been a one-off act of ﬁnancial policy, only that it does not ﬁt the pattern of the middle years of the reign.⁵⁶ If John was making a ﬁnancial ⁵³ Holt, The Northerners, 185. ⁵⁴ T. K. Keefe, ‘Proffers for Heirs and Heiresses in the Pipe Rolls: Some Observations on Indebtedness in the Years before the Magna Carta (1180–1212)’, Haskins Society Journal, 5 (1996), 103–4. ⁵⁵ Ibid. 108. Keefe’s ﬁndings accord well with the plans of 1211–12 for a new scutage, a development which Holt pointed to as a prime motivator for the abortive baronial rebellion of 1212; Holt, The Northerners, 81 and n. 1. ⁵⁶ An additional point is that John in the period 1205–8 did not squeeze the counties either, as he appointed curiales as sheriffs at the old county farms for reasons of security, abandoning the custodial experiment of the previous few years; see Carpenter, ‘The Decline of the Curial Sheriff in England’, 1–32.

176

Lords of the Central Marches

example of William de Braose, he did not take advantage of it. Indeed, the king went to much expense to eliminate the Braoses; he appears not to have been a cost-efﬁcient debt-collector. John’s explanation, issued well after the fact and at a time of political crisis, must be seen as an impressive example of early thirteenth-century political propaganda. There are important elements of truth in the explanations of Wendover, Painter, and Warren. Wendover seems unaware, however, of the territorial implications of Braose’s destruction, given the vast lands that came into royal custody as a result. Painter’s solid explanation likewise seems to ignore the results of the feud beyond the blackening of John’s reputation. Warren, focused on Ireland, ignores events in Wales from 1206 to 1209. After the fall of Normandy, King John was concerned with the structure of power in the west of the British Isles. John’s foremost goal was always the reconquest of his Continental lands; for him to do this, however, he must make certain that his house, as it were, was in order. But, after 1204, John was distrustful of the English baronage. If baronial disaffection were to become rebellion, it would be most dangerous if it arose along the western ﬂank of the British Isles, particularly in the Welsh March.⁵⁷ The reasons were simple: in the March, honours were compact blocks of land; feudal control was still strong; a military frontier meant that tenants who held by military service were still soldiers; and alliance with Welsh princes would bring troops as well. By 1206 John may very well have had reason to regret his earlier handling of affairs in the west, particularly his building up of the power of William de Braose. John did not have a coherent strategy from the beginning for cutting down to size the barons of the March and Ireland; instead, he moved cautiously, picking his opportunities. In Easter term 1206 the curialis Peter ﬁtz Herbert brought a suit coram rege against William de Braose for a third of the lordship of Brecon.⁵⁸ For the time being the king stayed the ⁵⁷ Revolts had occurred in the region in 1075 (Roger, earl of Hereford, son of William ﬁtz Osbern), 1087–8 (Roger de Montgomery, Ralph de Mortimer, Roger de Lacy, other Marchers), and 1095 (Roger de Lacy, with Welsh troops). During Stephen’s reign the Marchers and Welsh had reached a modus vivendi in 1137, and Miles of Gloucester’s defection to the Angevins in 1139 solidiﬁed control of the area for the empress; see Ch. 2; Crouch, ‘The March and the Welsh Kings’, 276–80. ⁵⁸ Although this was not made clear in the record of the suit, Peter’s claim was through his mother, who was descended from Miles of Gloucester; CRR iv. 98–9. For Peter ﬁtz Herbert’s descent from Miles of Gloucester, see Walker, ‘The ‘‘Honours’’ of the Earls of Hereford in the Twelfth Century’, 192–3.

The Reigns of Richard I and John

177

decision of the court in this action, but given the timing of the suit and the fact that it was brought by a curialis, it raises suspicion that John was involved. John at times used the curia regis to harass his magnates in just such a fashion.⁵⁹ In February 1207 Braose was stripped of the custody of Glamorgan and was replaced by John’s loyal mercenary commander Falkes de Breaute.⁶⁰ Royal harassment then moved to Ireland, as the king’s justiciar, Meiler ﬁtz Henry, waged war on the followers of the Marshal and the Lacys during 1207.⁶¹ By 1208 the campaign of harassment was turned back to the March and William de Braose. The Marshal had already been stripped of the shrievalty of Gloucestershire on 21 April 1207, and on 5 January 1208 another mercenary commander, Gerard d’Ath´ee, was placed in that ofﬁce.⁶² The demand for hostages, made in spring 1208, therefore came at a very strained time. It was a particularly sensitive demand to make of William de Braose, for Braose had demised part if not all his English and Welsh lands onto his eldest son, William the younger, by 1203 at the latest.⁶³ It was customary for the young sons of Marcher lords to assume some degree of responsibility in the March as a way of preparing them for entry into a violent inheritance. In this case, the evidence for a more formal arrangement is compelling. The demise from Braose father to son was remembered in a now lost cartulary of Neath Abbey in Glamorgan, a house patronized by the Braoses.⁶⁴ Also, the Chronica de Wallia refers to the death in 1210 of Maud de Braose and Willielmus iunior, dominus de Brecheniauc.⁶⁵ A charter of Brecon Priory refers to William the younger as the lord of the grantor (domini mei), with the grant being made at his request.⁶⁶ ⁵⁹ One example being the suit of the earl of Salisbury against the suspect Henry de Bohun, earl of Hereford, over the honour of Trowbridge in 1212; CRR vi. 270, 330. For comment, see Brock W. Holden, ‘The Balance of Patronage: King John and the Earl of Salisbury’, Haskins Society Journal, 8 (1999), 87. ⁶⁰ Rot. Litt. Pat. 68. ⁶¹ John was, however, not entirely successful in this, as Meiler was defeated and captured by Hugh de Lacy. Meiler was phased out of ofﬁce, but the terms on which the Irish barons held their lands of the Crown were redeﬁned to the advantage of the king. ⁶² Rot. Litt. Pat. 71 (castle), 75 (shrievalty) to Richard de Mucegros, 78b (transfer to Gerard d’Ath´ee). ⁶³ This was ﬁrst ably pointed out by Rowlands, ‘William de Braose and the Lordship of Brecon’, 122–33. ⁶⁴ F. R. Lewis, ‘A History of the Lordship of Gower from the Missing Cartulary of Neath’, BBCS 9 (1937–9), 151. ⁶⁵ ‘Chronica de Wallia’, 7, s.a. 1209 [1210]. ⁶⁶ ‘Cart. Brecon’, 14. 33. R. W. Banks, the editor of the Brecon Cartulary, dated this charter to either 1222 or 1230 and identiﬁed William juvenis as William V de Braose, the son of Reginald de Braose; see the Braose genealogy in Appendix 2. Banks was probably

178

Lords of the Central Marches

Further, and ﬁrmer, evidence for the demise comes from the Curia Regis Rolls. When Reginald de Braose, a son of William III, and John de Braose, a grandson, were waging a legal battle over Bramber in Sussex in 1219, John claimed right of inheritance on the grounds that William the younger, his father, had been seised of the English and Welsh lands and had performed homage to the king for them.⁶⁷ The normal form of claim would only have to trace the line of descent; the claim of receipt of homage is thus all the more signiﬁcant.⁶⁸ There is a further piece of evidence for a demise of at least the English lands. In Hilary term 1207, a tenant of the Braose honour of Bramber brought a recognition against William III for four knights’ fees. William replied that he ought not to respond, as his son held the land. A day was given for judgment but none was entered, no doubt because the crisis brought an end to Braose lordship there.⁶⁹ Ralph of Coggeshall provides one ﬁnal piece of evidence. Ralph’s history states that Giles de Braose received all the possessions of his father (William III) in custodiam with his nephew (John), donec puer ad aetatem legitimam venerit.⁷⁰ This passage is important, for Coggeshall, along with the Barnwell Chronicler, is one of our most reliable sources for John’s reign.⁷¹ Coggeshall says that this occurred in 1213, when following the history of Abergavenny Priory, which states that Reginald de Braose’s son by the daughter of William Briwerre ‘was called William junior de Braose’; Mon. Ang. iv. 615. But Banks’s ﬁrst date, 1222, is based on a misreading of the Close Rolls, for Reginald de Braose had been mistakenly reported dead; Rot. Litt. Claus. i. 500 (William de Cantilupe, John de Monmouth, and Roger de Clifford were sent by the king to take Reginald’s castles into royal custody). The date 1230 is possible, but the witness list of the charter argues against it. William de Waldeboef, a Brecon tenant, is a witness, but he died before 1211, when the scutage return for the forfeited lordship lists his three knights’ fees as being held by his ‘heir’; RBE 601. John de Waldeboef had replaced William in a charter of Robert le Wafer, another Brecon tenant, witnessed in the court of Reginald de Braose and dated 1215 × 1217; ‘Cart. Brecon’, 14. 165. The return for 1243 shows that John de Waldeboef was the holder of William de Waldeboef’s lands in Herefordshire, which were held of the honour of Brecon; BF 806. ⁶⁷ CRR viii. 10–11. ⁶⁸ I owe this observation to Dr Paul Brand. ⁶⁹ CRR v. 27. ⁷⁰ Ralph of Coggeshall, Chronicon Anglicanum, 168. Against Coggeshall’s use of custodiam must be set the use in the writ of 21 Oct. 1215 of the word ‘seisin’ rather than custody to describe the restitution; Rot. Litt. Claus. i. 235b. The writ following that to the sheriff of Surrey, addressed to the sheriff of Herefordshire, oddly speaks only of Braose lands and tenements in villa Hereford’. ⁷¹ This passage has been overlooked by those writing on the Braoses, and even by Sir James Holt in his work on the casus regis, where it would subtly alter his argument, showing how the casus was sacriﬁced in a time of political crisis (i.e. the minority); see Holt, ‘The Casus Regis’.

The Reigns of Richard I and John

179

Giles returned from France with the other exiled bishops. But his regaining of the custody of the Braose lands is best dated to 1215, as no ﬁne is mentioned until 5 March of that year, shortly before a rising in the former Braose Marcher lordships enabled Giles and his brother Reginald to regain their patrimony anyway.⁷² This rising seems to have occurred because Giles’s ﬁne of 5 March was a proffer and only gained him a hearing in the king’s court concerning his claim; a writ dated 10 May promises that Giles shall have the judgment of ‘our court’ (curia nostre [sic]).⁷³ The rising seems to have presented John with a fait accompli, as the ﬁne was accepted and writs ordering the restoration of the Braose lands to Giles went out on 21 October.⁷⁴ But the fact that Reginald remained in rebellion even after the restoration and Giles’s return to the king’s side in autumn 1215 suggests that he for one was not satisﬁed with this arrangement.⁷⁵ Giles de Braose’s custody of his nephew John, as well as William IV’s prior seisin, was referred to later by John de Braose’s son (another William) when, in 1279–80, William de Beauchamp, earl of Warwick, brought a suit for Gower.⁷⁶ The demise would have given an altogether more sinister air to John’s demand for the younger William as a hostage. There seems to have been some ambiguity as to whether the demise was actually recognized by the Crown, as is indicated by the Bramber case in the curia regis in 1207 mentioned above, as well as the later disputes over the Braose inheritance. John’s apologia of 1212 states that the distraint upon the elder William de Braose was carried out by attempted seizure of his English and Welsh chattels, and this implies either that the king did not recognize or was unaware of the demise. Yet that same document states that the king demanded as hostages the two sons of William the younger and the son of his brother Reginald, and this raises doubts as to which ⁷² Rot. Litt. Claus. i. 189b, an order to all sheriffs to take up sufﬁcient pledges to cover Giles’s ﬁne. ⁷³ Rot. Litt. Pat. 141. ⁷⁴ Ibid. 151. Writs were sent to William Marshal concerning Swansea Castle in Gower, Henry de Tracy for Barnstaple Castle, and Henry ﬁtz Count for Totnes Castle, the latter two being the centres of the Braose estates in the West Country. ⁷⁵ The recognition of John de Braose as heir points to the seisin of his father, William IV; if that had not been the case, the king, by favouring the representative claimant (John) over the cadet claimant (Reginald), would be going against that legal bugbear of his reign, the casus regis (although there was not total consistency in judgments involving the casus regis in John’s reign). For a discussion of the casus regis, see Holt, ‘The Casus Regis’. ⁷⁶ Cartae et Alia Munimenta de Glamorgancia, ed. G. T. Clark, 6 vols (Cardiff, 1910), iii. 805–8. Beauchamp, not surprisingly, denied that William IV de Braose had ever been seised of Gower.

180

Lords of the Central Marches

member or members of the family were demanded as security. Only one son of William de Braose (presumably William III) is mentioned as a hostage in 1208, when William delivered ‘his son’ to Walter de Lacy in assurance that he would follow the royal will (ad faciendum de eo voluntatem nostram).⁷⁷ The ﬁrst mention of William, son of Reginald de Braose, as a hostage is in July 1213, while William IV de Braose’s sons John and Giles are not mentioned until January 1214; both writs concerned their transfer to new custodians.⁷⁸ This does not mean that they were not taken as hostages in 1208, but they may equally have been captured in 1210. King John seems to have skirted the issue of the Braose demise in his apologia. It is also likely that he used the ambiguity of the situation, whether or not the demise was actually recognized by the Crown, as a means of keeping the Braoses off balance and striking fear into them. The question of what would happen to the Braose lands which had been demised was an open one—there was no answer in law or custom for this situation. It may well be that the land also would be taken into royal custody.⁷⁹ Thus, the demand for William the younger can be seen as a further, if politically extreme, means of trimming the power of the house of Braose. Indeed, if the Welsh chronicler of the Brut is to be believed, William de Braose and his family were banished owing to the ‘enmity and envy’ which John bore towards William de Braose the younger—a suggestive comment but one which can be followed no further.⁸⁰ Little wonder that the Braoses balked at handing the family heir over to the king. As it was, on 19 March 1208 the younger William was given up, but placed into the custody of his brother-in-law, Walter de Lacy.⁸¹ John was unlikely to gain much leverage from this situation, which looks like a compromise. The king increased the pressure, and a month later Gerard d’Ath´ee led an expedition of foot and horse into Braose lands in the March, ostensibly to distrain the elder William’s chattels but more likely to capture the family.⁸² The rest of the story is well known. William III and his wife Maud, their son William the ⁷⁷ Rot. Litt. Pat. 80b. ⁷⁸ Ibid. 102, 108b. See also the Braose genealogy in Appendix 2. ⁷⁹ Although it should be noted that, as Nicholas Vincent commented to me, had William the younger been demanded as a hostage after having taken the homage of his men, such a demand would have effectively been an order for his arrest. This of course does not ﬁt with the pattern of John’s hostage-taking, but the Braose quarrel was in many respects sui generis. ⁸⁰ Brut, 83, s.a. 1210. ⁸¹ Rot. Litt. Pat. 80. ⁸² Ibid. 81; Rot. Litt. Claus. i. 112–13; Roger of Wendover, Flores Historiarum, ii. 49.

The Reigns of Richard I and John

181

younger and his brother Reginald, ﬂed to Ireland and received shelter from the Marshal in spring 1209. From the Marshal the Braoses were passed on to the protection of Walter and Hugh de Lacy. A brief meeting before the invasion between William de Braose and the king in Wales failed to reach a peaceful solution, and John invaded Ireland in the summer of 1210, scattering all before him. The Braoses ﬂed: Maud and William the younger were captured in Scotland and starved to death in Windsor Castle (or Corfe); William III died in exile in Paris in 1211; Reginald and Giles survived.⁸³ The most important result of Braose’s downfall was that after 1208 John controlled a vast conglomeration of lands in the south March the extent of which would not be seen again until the time of the younger Despenser in the early fourteenth century (see Map 4). This was not an accident; they were not acquired in a ﬁt of absence of mind. When John lost Normandy in 1204 he faced an unstable and potentially rebellious political community. Before the king could seriously contemplate the reconquest of his continental lands he had to assure himself that England was secure. That is one reason why trusted men from the lost Angevin dominions, men such as Falkes de Br´eaut´e, Gerard d’Ath´ee and Engelard de Cigogn´e, were slotted into important shrievalties and custodianships in England. The appointment of sheriffs c.1207 was now directed as much towards security as towards ﬁnance. These appointments coincided with a hardening of the king’s handling of his barons.⁸⁴ But such trusted servants could not be inserted into the one area that was most dangerous to the king—the March. With his own Marcher experience, John would have been well aware of the potential danger that the Marcher lords posed to the king. Empires could be built quickly on the frontier; indeed John himself had done much to help build the Braose empire. The use of the March of Wales as a quick means to bolster one’s power was almost a constant throughout the Middle Ages. Thus, in the 1220s Hubert de Burgh would try to build his power base in the Middle March, while the great gains of the younger Despenser in the early 1320s and of Roger Mortimer later in the same decade also occurred in the March.⁸⁵ A particularly apt parallel is with Richard II’s policy in the north March from 1397 to 1398. Richard added Flint to Cheshire and elevated the ⁸³ Painter, The Reign of King John, 243–8. ⁸⁵ For Hubert de Burgh, see Ch. 7.

⁸⁴ Holt, The Northerners, 225.

182

Lords of the Central Marches

old palatine county to the status of a principality; he then augmented this block with the forfeited Marcher estates of the earl of Arundel.⁸⁶ Unlike Richard II, John was not attempting to rule the kingdom from the March; rather, he seems to have been determined to keep a sizeable portion of it under direct royal control in order to neutralize the threat that the March could pose to the political stability of his kingdom. William de Braose was much too powerful for John’s taste; there was little more the king could give him to keep him onside, so he was cut down to size. And there was no one to replace him along the March other than the king himself. This was precisely what John set about doing. With the exception of Blaenllyfni, which Peter ﬁtz Herbert had claimed, none of the forfeited land along the March was granted out to favourites—no attempt to woo the baronage with landed rewards was made. In Ireland in the aftermath of the 1210 campaign some baronies had been granted out to royal supporters and local men were appointed the king’s bailiffs. In the Welsh March, mercenary captains such as Gerard d’Ath´ee and Engelard de Cigogn´e administered the region. The campaigns in Ireland in 1210 and Wales in 1211 in turn were carried out at least partly in order to secure the March. As A. A. Duncan has convincingly argued, a letter in the register of Philip Augustus, dated late 1209 to early 1210, was sent by the French king to either Walter or Hugh de Lacy. The letter states that Philip had been told by a Norman intermediary, Roger des Essarts, that whichever Lacy it was addressed to promised rebellion ‘by friends and attacks in England, and by friends and defence of castles in Ireland’. Philip in turn promised French consideration of Lacy’s claims told by his ancestors, probably in Normandy.⁸⁷ The threat of some combination of William de Braose, the Lacys, and some of the Welsh princes, perhaps supported by the French, was very real in 1210. This was further borne out by the treaty of alliance agreed upon between Llywelyn ap Iorwerth and Philip Augustus of France in 1212, an alliance concluded at the height of the Welsh reaction to John’s successes of 1211.⁸⁸ Philip Augustus was determinedly ﬁshing ⁸⁶ R. R. Davies, ‘Richard II and the Principality of Chester 1397–9’, in F. R. H. Du Boulay and C. Barron (eds), The Reign of Richard II (1971), 256–79, esp. 257–60. ⁸⁷ A. A. M. Duncan, ‘King John of England and the Kings of Scots’, in Church (ed.), King John, 258–9. Letter in register of Philip Augustus preserved in the Vatican Library, pub. in Archives des missions scientiﬁques et litt´eraires, 3rd ser., 4 (1880), 344. ⁸⁸ R. F. Treharne, ‘The Franco-Welsh Treaty of Alliance in 1212’, BBCS 18 (1958), 60–75, repr. in E. B. Fryde (ed.), Simon de Montfort and Baronial Reform: Thirteenth Century Essays (1986), 345–60.

The Reigns of Richard I and John

183

in the unsettled waters of the western British Isles from at least 1210. This threat helps place the expeditions of 1210 and 1211 in context. For a king bent on recovering his continental inheritance, the Irish and Welsh campaigns were a critical waste of resources if they were not undertaken for an important reason. John’s distrust of his barons led him to perceive an imbalance of power in the west, owing to his promotion of William de Braose. John also realized that to strengthen royal power would mean an expansion and solidiﬁcation of his own authority throughout the Irish Sea arena. Once he had moved against Braose, John must have been aware of the dangerous and real possibility of an organized opposition in that part of the world. Although John’s actions preﬁgured Edward I’s conquest of Wales, 1211 was not 1282. Prior to 1208, King John had enjoyed mostly amicable relations with Llywelyn ap Iorwerth, prince of Gwynedd.⁸⁹ But the removal of William de Braose was followed by the fall of Gwenwynwyn of Powys, who had attempted to take advantage of Braose’s elimination. Only the king and Llywelyn had the resources to ﬁll the resulting vacuum of power and authority along the March. Indeed, John’s distrust of his baronage ensured that the Crown would retain control of the area. The king was now a major player in the region as he had never before been. John remained, however, on good terms with Llywelyn until January 1210. Sir John Lloyd speculated that the reason for the sudden rupture in the relationship between king and prince was the result of the Welsh leader becoming involved with the exiled William de Braose, now in Ireland. The Barnwell Chronicler states that many of the Welsh allied themselves with William de Braose and Hugh de Lacy, earl of Ulster, and launched incursions into England.⁹⁰ We do not know if the phrase Walensium nonnulli was meant to include Llywelyn, but even if the great Welsh prince himself was not involved, John’s suspicion would have been attracted and the king may very well have held Llywelyn responsible for the Welsh attacks of 1210. It is certainly easy to see why Llywelyn might have covertly supported Braose and the Lacys. Although the Marcher barons had frequently caused trouble to the Welsh, they were also in many respects easier to deal with than the English king. One could play them off against each other or seek marriage alliances, ⁸⁹ For much of what follows, see Lloyd, A History of Wales, ii. 612–23, 631–7. ⁹⁰ Walter of Coventry, Memoriale, ed. W. Stubbs, 2 vols, Rolls Series (1872–3), ii. 202, s.a. 1210.

184

Lords of the Central Marches

as Llywelyn himself later would do with the competing Braose heirs Reginald and John. One could not divide the English king against himself, and agreements struck with the Crown increasingly brought a further deﬁnition and tightening of feudal dependence for the Welsh princes.⁹¹ John ultimately posed a greater threat to Llywelyn than did William de Braose. The character of Anglo-Welsh politics had been ineradicably altered. The Welsh attacks of 1210 in support of Braose and the Lacys determined the events of the next two years. As early as 1209 John had appointed an experienced soldier, his brother William, earl of Salisbury, as royal custodian of the March.⁹² Early in 1210 Salisbury, the earl of Chester, the justiciar Geoffrey ﬁtz Peter, and Peter des Roches, bishop of Winchester, led a royal expedition into north Wales. The combined status of these individuals is perhaps the best indicator of the severity of the Welsh threat in 1210. On his return from Ireland in 1210, John seems to have been determined to deal with Llywelyn. The king’s removal of Robert ﬁtz Richard from his lordship of Haverford and his restoration of Gwenwynwyn to his lands in southern Powys further altered the balance of power in Wales, and Llywelyn’s rivals among the Welsh joined John. Having seen the return of Gwenwynwyn, those Welsh princes who had a reason to oppose Llywelyn no doubt hoped to make gains of their own by supporting the English king. They had also certainly been awed by John’s progress through south Wales on his way to and return from Ireland in 1210. The English campaign of 1211 nearly ﬁnished Llywelyn, who was saved by the intercession of his wife, Joan, John’s illegitimate daughter.⁹³ But John had overextended himself. The king’s victory had come about partly because the support of the Welsh princelings isolated Llywelyn. With a Continental campaign in the ofﬁng, John’s resources would then be diverted elsewhere and he would be even more dependent on the cooperation of the native Welsh to hold onto his gains in the country. It is at this point that John made a political miscalculation, one that was in keeping with his treatment of the forfeited Braose and ⁹¹ For ius scriptum and the intensifying of English overlordship over the Welsh, see Davies, The Age of Conquest, 293–5; Rowlands, ‘King John and Wales’, 279. ⁹² Rot. Litt. Pat. 88. Salisbury was himself recovering from a period of royal suspicion and disfavour which coincided with the ﬁrst moves against Braose; Holden, ‘The Balance of Patronage’, 85. ⁹³ For the terms to which Llywelyn had to agree in submitting to John, see J. B. Smith, ‘Magna Carta and the Charters of the Welsh Princes’, EHR 99 (1984), 344–62.

The Reigns of Richard I and John

185

Lacy lands in the March. The building of new castles at, among other places, Builth and Aberystwyth convinced the Welsh princes that it was better to side with Llywelyn, as John was now changing the rules of the game.⁹⁴ His programme of castle-building signalled in no uncertain terms that he would rule the conquered territories in Wales himself, rather than reward his Welsh allies with lands: John was going well beyond a tightened deﬁnition of overlordship. The matter of hostages also ﬁgured prominently in John’s alienation of the Welsh, as it had with the Irish in 1210.⁹⁵ Welsh misgivings were borne out when, in 1212 in the midst of the Welsh rebellion, John hanged twenty-eight hostages, including one 7-year-old boy.⁹⁶ Llywelyn’s campaign of 1212, along with the baronial conspiracy of the same year, brought a very shaky English dominance of Wales to a close. John was thus compelled to make overtures to the exiled Lacys in 1213 and to begin a policy of repairing relations with the Marchers. His attempt at a rapprochement with the Marchers was largely successful, with one exception: the Braoses and the men of the Braose lordships that were in royal hands. Although the king attempted to make a settlement with the surviving head of the family, Giles de Braose, it was not enough to keep a Braose rebellion from igniting in 1215. Although John did not lose control of the forfeited Braose lordships in the March until 1215, his authority there was not as strong as it seemed. In those lordships, the king did not do a good job of attracting the loyalty of the more important tenants, although in his defence he may not have had enough time. The beginnings had certainly been promising. In September 1208 the Braose tenants made an agreement with Gerard d’Ath´ee that they would not return to the service of their lord.⁹⁷ Indeed, prest and liberate rolls show that of the nineteen largest tenants of ⁹⁴ For Builth, erected by Engelard de Cigogn´e, see Brut, 84, s.a. 1210; Aberystwyth, Margam Annals, 31. ⁹⁵ See above. ⁹⁶ Henry II had previously hanged Welsh hostages; Brut, 64–5. For John and Welsh hostages, see the discussion in Rowlands, ‘King John and Wales’, 280–1. The hanging of the young Welsh boy at Shrewsbury is in Brut, 86. ⁹⁷ Rot. Litt. Pat. 86b (royal ratiﬁcation of the agreement dated 21 Sept.). Although this document was witnessed by Peter des Roches, bishop of Winchester, in the presence of the bishop of Bath and William Briwerre, rather than the king, John had been in the area recently. He had been at Hereford on 26 June, was at Worcester on 1–3 July, and was at Tewkesbury on 3 October followed by Shrewsbury on 8–9 October. Although his servants carried on the negotiations with the Braose tenants, John made visits to the region numerous times in 1208, eleven times in March–October 1208, for a total of twenty-three days. For these dates, see the itinerary at the beginning of Rot. Litt. Pat. 86b.

186

Lords of the Central Marches

the Braose lordship of Brecon, nine accompanied the king to Ireland in 1210.⁹⁸ They were not compelled to do so, as tenants of Radnor, Abergavenny, and the rest of Brecon ﬁned in order not to cross.⁹⁹ From 1208 to 1210 John had managed to split the Braose afﬁnity, much as he had done to the Marshal’s in 1207.¹⁰⁰ The Braose tenants may have been detached from allegiance to their lord by a number of means: royal blandishments, outright fear of the king’s power, and perhaps resentment against the consolidation of Braose lordship. By 1215, however, Walter III de Clifford, sheriff of Hereford, would write ‘that the whole county of Hereford, besides the barons and their men, was with the bishop of Hereford [Giles de Braose] against the king, and bore arms against the king or sent armed men’.¹⁰¹ In the spring of 1215 Giles de Braose and his brother Reginald had risen in rebellion, undoubtedly in response to the delay in the restoration of the Braose lands.¹⁰² The Welsh Brut relates that after the rising of the Northerners and the Welsh, Giles sent Reginald to Brecon, where he was well received by the Welsh of the lordship. Reginald then attacked down the Usk, taking Pencelli, Abergavenny, and the Three Castles (Whitecastle, Grosmont, and Skenfrith) within a few days of 1 May 1215. There was obviously little or no resistance from the Braose tenants. When Giles de Braose then advanced up the river Wye from the north, Radnor, Hay, Brecon, Builth, and Blaenllyfni castles all surrendered without resistance.¹⁰³ The disaffection of the men of the Braose lordships with royal control could not be clearer. The Braose rebellion then spilled over into Herefordshire. Walter de Clifford’s statement suggests that the majority of the men of the region sided with Giles and Reginald de Braose in 1215, not merely those ⁹⁸ Walter de Clifford junior (4 fees), William de Waldeboef (3), William de Gamages (1), Ralph Torel (1), Richard de Pauncefoot (1), William des Furches (1), Pain de Burghill (1), Walter Devereux ( 1/2), and Roger de la Zouche ( 1/2), thirteen out of thirty-two fees listed in 1211; RBE 601; Rot. Lib.177–225 and passim. ⁹⁹ Pipe Roll, 12 John, 147. ¹⁰⁰ For the splitting of the Marshal afﬁnity, see Crouch, William Marshal, 105–6. John had used the dissatisfaction of some of the Marshal’s men of Leinster to prise them away from his allegiance and set them against him; the foremost example is Meiler ﬁtz Henry, a Marshal tenant but also John’s Irish justiciar and the leader of the faction of ‘Irreis’ opposing their parvenu lord. When the Marshal, at the king’s summons, crossed to England at Michaelmas 1207 with the majority of his Irish barons, including Meiler, John dealt the earl a crushing blow by detaching a number of members of his afﬁnity, including his nephew John Marshal. ¹⁰¹ CACW 2, 1. 10 (dated probably Jan.–Apr. 1216). ¹⁰² See pp. 178–9 above. ¹⁰³ Brut, 90, s.a. 1215.

The Reigns of Richard I and John

187

of the county of Hereford. One of the three men named later in the writ above who remained in rebellion with Reginald de Braose, Robert Devereux, held a knight’s fee in Pencelli, in the lordship of Brecon.¹⁰⁴ This hints at the widespread nature of the initial rebellion in 1215, stretching across lordships and the hazy boundary of the March. New-found loyalty to the ill-treated house of Braose must be counted a cause of the rebellion, but resentment against royal lordship was equally so. For the ﬁrst time, royal administration and probably justice had been extended to the lordships of the south-east March. Although none of the Braose lordships accounted to the exchequer, Glamorgan occasionally did so, so we may infer that the majority of the occupied lands probably paid scutage and tallage into the king’s chamber. The scutage lists compiled in 1211 for Brecon, Radnor, Builth, and Abergavenny were the only ones made for those areas in the thirteenth century. The 1211 inquest into knights’ fees was the most far-reaching since the Cartae Baronum of 1166 and undoubtedly heralded new and heavier scutage. Sir James Holt saw the inquest of 1211, linked to the planned 1212 Continental campaign, as one of the root causes of the baronial plot of that latter year.¹⁰⁵ From John’s standpoint, the ﬁnancial demands of reconquering his Continental lands made such measures necessary. His barons did not see the 1211 inquest in quite the same way. New ﬁnancial exactions would have been doubly unpopular to the Marchers, who had always been excused payment of scutage and royal aids owing to their defensive role on the border.¹⁰⁶ Although there is no evidence for the operation of royal justice in the area, it was routine for such jurisdiction to operate when liberties came into royal wardship; this was indeed the case in Brecon in the 1230s, and John’s overwhelming need to raise revenues would make it strange if the Braose lordships did not feel the heavy hand of royal justice. Following the baronial plotting and Welsh offensive of 1212, John thought it wise in 1213 to recall Walter de Lacy from exile in France.¹⁰⁷ With the build-up to the Continental expedition of 1214 John needed capable commanders, and with his experience on the March and Ireland, and possibly with Simon de Montfort on the Albigensian Crusade while in exile, Lacy was valuable. The king may also have felt the need to get Walter out of France so as to deprive Philip Augustus of a valuable ¹⁰⁴ Ibid.; RBE 601. ¹⁰⁵ Holt, The Northerners, 81 and n. 1. ¹⁰⁶ Keefe, Feudal Assessments and the Political Community under Henry II and his Sons, 189; see also Ch. 3. ¹⁰⁷ Writ dated 2 June, Rot. Litt. Pat. 99b.

188

Lords of the Central Marches

tool.¹⁰⁸ Walter’s brother Hugh was also approached but refused to return to England. John took hostages in return for Walter de Lacy’s good behaviour: the sons of Miles and John Pichard, Gilbert de Lacy, and William de Furches, all Lacy tenants, along with ‘other hostages’ as Engelard de Cigogn´e felt appropriate.¹⁰⁹ With hostages having been given, Lacy’s return was accompanied by a restoration of his lands in England and the March, minus Ludlow Castle.¹¹⁰ Concessions were granted: Walter used his service with the king in Poitou in 1214 to bargain for the release of the men either captured at Carrickfergus in 1210 or held as hostages.¹¹¹ Thus, Walter rebuilt the Lacy afﬁnity, mostly based in Ireland but also including relatives of Anglo-Welsh tenants. Walter de Lacy was restored to his Irish lands on 16 March 1215, and regained Ludlow Castle on 12 April.¹¹² John then recognized the fact of Braose title to their lands later in the year. John was therefore, from 1212, able to attract the barons of the Welsh March and Ireland (with the exception of Reginald de Braose) to his side at the end of his reign, and they ultimately helped save his kingdom for his son. After Giles de Braose’s return to royal allegiance in October 1215, Herefordshire and the south-east March were largely stabilized, and Reginald de Braose seems to have had little success in his continued rebellion. When John died on 19 October 1216 at Newark, the Marcher barons Walter de Lacy, Hugh de Mortimer, and John de Monmouth were present and Walter and John were named among the executors of his will.¹¹³ Ultimately, John’s policy of an expanded royal presence in the March, with the forfeited Braose and Lacy lordships directly administered by his mercenary commanders, had proved unsuccessful. Although this policy failed, it does show awareness on the part of the king of the important and at times dangerous role that the Marchers could play in English politics. His campaigns in Ireland and Wales demonstrate ¹⁰⁸ According to the Annals of Dunstable, the baronial plotters of 1212 had elected Simon de Montfort to replace John on the throne of England after his overthrow; Annales Monastici, iii. 33. Hugh de Lacy was with Montfort on the Albigensian Crusade; DNB, s.n. ¹⁰⁹ Rot. Litt. Claus. i. 147, dated 29 July 1213. ¹¹⁰ Rot. Litt. Claus. i. 147. ¹¹¹ For Walter in France, Rot. Litt. Pat. 112, 113b. ¹¹² Ibid. 191, 132b. ¹¹³ Nine days earlier the king had granted to Margaret de Lacy, Walter’s wife, 3 carucates of land in the forest of Acornbury for the foundation of a house of nuns. This house was founded for the souls of Margaret’s father, William de Braose, her mother, Maud, and her brother William; Rot. Litt. Pat. 199b.

The Reigns of Richard I and John

189

a grasp of the interconnected nature of politics and landholding on both sides of the Irish Sea. But John’s political miscalculations in Wales and the concomitant alienation of many of the Welsh princelings, combined with the abortive baronial plot of 1212, undid his plans. John’s increased involvement in Wales, touched off by his destruction of William de Braose, brought him close to doing what his grandson would eventually accomplish, but the same approach which the king showed to the forfeited Braose and Lacy lands proved to be his undoing in Wales. What remained into the minority of Henry III and beyond was a much more heightened royal involvement in the affairs of the border region.

7 The Region in the Reign of Henry III, 1216–1246 On ﬁrst examination, the early years of the reign of Henry III seem to have been a rather inauspicious period for the aristocracy of western Herefordshire and the south-eastern March. Although this was, for at least some lordships, a period of the ‘intensiﬁcation of lordship’, little was achieved in terms of regaining lands lost to Llywelyn and the Welsh during the civil war. This is an indicator of how reliant the Marchers were becoming on the military power of the Crown; unfortunately for them, the Crown was supine for some time after John’s death. At the same time, in the words of one historian, Llywelyn’s ‘superiority within native Wales from 1218 to 1240 was geographically more extensive and more fully articulated than that enjoyed by any other Welsh prince since the coming of the Normans’.¹ One reason for this was geographical: Gwynedd, situated in the far north-west of Wales and protected by tidal estuaries and mountain ranges, was a secure base. Having narrowly escaped destruction in 1211, Llywelyn had adroitly exploited the unrest and then civil war in England from 1212 to 1217 to bring the other Welsh princes under his leadership, gain ground at the expense of the Marchers, and even exert inﬂuence on English politics. And at the root of his superiority lay a military power which was increasingly beyond the capability of the Marcher lords to resist unaided. Welsh military action was concerted as never before: although individual princes might make limited forays on their own, all major expeditions required Llywelyn’s imprimatur and were led by him.² The March, particularly the central and south-eastern areas, was a crucial area of confrontation from 1218 to Llywelyn’s death in 1240 and beyond. The region from Ceri in the north to Brecon in the ¹ Davies, The Age of Conquest, 244.

² Ibid. 244–5.

The Reign of Henry III

191

south was of strategic importance to the Welsh prince. It formed an outer line of defence for what was now an enlarged principality of Gwynedd (Powys had not been absorbed but was ruled as an appanage). The area was also important for Llywelyn’s lines of communication with his allies in upland Glamorgan and Gwent. He therefore devoted much of his effort after 1218 to extending his power in the region. Llywelyn backed the native princes of Maelienydd and Elfael against Mortimer aggression, annexed Builth by a marriage agreement with William V de Braose in 1229 and razed Builth Castle, and attracted the leaders of upland Brecon into his orbit by 1230. His control over the area was also an important barometer of his political power in native Wales.³ The English government was therefore forced to pay attention to the importance of the March. Strong castles were built at Montgomery in the north and Painscastle in the south, establishing a royal presence at the extremities of the region from the 1220s. Much of the thrust of the campaigns of 1223, 1228, and 1231 was directed towards the Middle March, but little progress in the region was made until Llywelyn’s hegemony over pura Wallia unravelled following his death in 1240. The year 1223 saw the most signiﬁcant gains: the building of Montgomery Castle at least served as a check on Llywelyn’s ambitions, while in the west the triumph of the younger Marshal in taking back Ceredigion and Carmarthen replaced the supremacy of the Welsh prince in that region. Overall, though, Marcher ambitions became subsumed in royal policy. By 1218 the Marcher barons were reliant on royal power, but that power was seriously weakened by civil war and the Crown’s consequent loss of authority and thus revenues in the ﬁrst few years of Henry’s minority.⁴ Following the Treaty of Worcester in 1218, Llywelyn occupied a position of power in Wales unrivalled by any previous Welsh ruler. Yet, the south-eastern March played an important role in politics after Henry came of age, not least in that it was one arena for Richard Marshal’s rebellion of 1233–4 against the ‘Poitevins’. By the 1240s the death of Llywelyn the Great, the acceptance of the loss of Poitou after another failed campaign on the Continent, and the end of ³ Ibid. 247. ⁴ This stress on the crippling lack of revenue experienced by the minority government in its ﬁrst years was ably brought out for the ﬁrst time in D. A. Carpenter, The Minority of Henry III (1990), esp. 112–20.

192

Lords of the Central Marches

the lines of Braose, Lacy, and Marshal meant that the Crown was poised to play a much more dominant role in the March, one that presaged the conquest of Wales under Henry’s son Edward I.⁵ The changes in the region described above, particularly in western Herefordshire, continued into the reign of Henry III.⁶ Firstly, the division of the great knightly holdings between those in Marcher lands and those in Herefordshire, which began roughly in the 1160s, was largely completed by the 1220s. Secondly, Herefordshire continued to be drawn further within the ambit of royal government, the directions issued to the sheriff of Hereford for the inquest into knights’ fees of 1242–3 being particularly instructive in this respect.⁷ Many Herefordshire tenants were claiming exemption from scutage by saying that they held their lands neither by military service in capite nor of someone who held in chief of the king. The lack of information which the central government had regarding the feudal obligations of Herefordshire resulted in a wide-ranging inquiry, ‘as much within a liberty as without’, and in many places even the hidage of holdings was recorded. Inquests meant deﬁnition, which ultimately meant control. The increasing, if still somewhat intermittent, reach of royal government also opened new opportunities for service and thus royal patronage to local men. The best example is John de Monmouth, lord of the small barony from which he took his surname. He served in a number of functions: justice, royal forester, and commander of royal forces at one stage of the Marshal rebellion of 1233–4.⁸ This in turn increased his standing in the aristocratic society of the region, a rise which is perhaps best indicated by his decision to found a new religious house at Grace Dieu in Monmouth in 1226.⁹ At the same time, the relative weakness of ⁵ The royal annexation of the earldom and county of Chester in 1237 placed the Crown on the same footing in the north March; see Davies, The Age of Conquest, 299. ⁶ See Ch. 5 for the increasing reach and impact of Angevin government on the region. ⁷ BF 797–8. ⁸ He was a royal justice for the visitation of 1218–22 which visited Herefordshire in August 1221; David Crook, Records of the General Eyre, PRO Handbook, 20 (1982), 76–7. In January 1219 he joined the sheriff of Herefordshire in viewing purprestures and levels of stock on the royal manors in Herefordshire; Rot. Litt. Claus. i. 386b. The writs directed to him as a forester are numerous; a few examples concerning the Forest of Dean are ibid. 393b (1219), 430b (1220). He was given custody of the Forest of Clarendon in December 1222; ibid. 526b. Monmouth, along with Ralph de Tony, was made commander of Henry III’s mercenary troops in the region in late November or early December 1233; Matthew Paris, Chronica Majora, ed. H. R. Luard, 7 vols, Rolls Series (1872–84), iii. 254. ⁹ See J. G. Cowley, The Monastic Order in South Wales (Cardiff, 1977), 27–8.

The Reign of Henry III

193

the Marcher lords vis-à-vis Llywelyn, the royal presence at Montgomery and Painscastle, the increasing frequency of royal military activity in the area, and mortality in Marcher families—Reginald de Braose died in 1228, his son William in 1230, Walter de Lacy in 1241—meant that although the March was not threatened with the sort of royal domination it experienced under King John, the Crown had a much-enlarged role and power in Marcher affairs. For that reason, we must again place our families in the context of the high road of Anglo-Welsh politics. When King John died on 19 October 1216, England was still divided by civil war. The men of Herefordshire and the adjacent Marcher lordships, with the exception of Reginald de Braose, were in the royalist camp. Reginald had sided with Llywelyn and the rebel English barons owing to his dissatisfaction with the settlement by which the Braoses were restored to their patrimony in 1215.¹⁰ Llywelyn had in turn attempted to cement this alliance with Reginald by giving him his daughter Gwladys in marriage in 1215.¹¹ But the loyalty of the other Marcher barons to John served to limit the activities of Llywelyn and Reginald to Wales, despite an attack, probably by one of the two men, on the Marshal’s castle of Goodrich on the very day of the young Henry’s coronation.¹² The immediate reissue of Magna Carta on 12 November 1216 and the fact that the rebel barons were now ﬁghting against a blameless 9-year-old boy served at once to take a good deal of ﬁre out of the cause of Prince Louis and the rebels. Some rebels in the region began returning to the royal side shortly after King John’s death; on 22 November 1216 Walter de Lacy, the virtual military governor of Herefordshire, was instructed to take hostages and set terms for the payment of 250 marks so that Roger of Lenham (Leinthall, Lyonshall?) might then come to the king and return to his faithful service.¹³ The battle of Lincoln on ¹⁰ A settlement which seems to have recognized his nephew John de Braose as the eventual heir; the critical piece of evidence is in Ralph of Coggeshall, Chronicon Anglicanum, 168. For a full discussion of the demise of the Braose lands in England and Wales onto William IV de Braose (‘the younger’), see Ch. 6. ¹¹ Brut, 91. The statement by the Welsh chronicler that Reginald was Giles de Braose’s ‘heir to his patrimony’ is more reﬂective of the fact that Reginald was in possession of Brecon and the other Braose lordships in the March than a clear-cut claim in the king’s court. ¹² Carpenter, The Minority of Henry III, 20. Walter de Lacy was in charge of Hereford Castle; Rot. Litt. Claus. i. 293. ¹³ Rot. Litt. Claus. i. 294. Lacy’s debt for the restoration of his Irish lands had been placed in respite ‘so long as he is in the king’s service in England and so long as he shall have custody of Hereford Castle’; ibid. 282. Lacy was custodian of the bishopric of Hereford, was reinforced by royal crossbowmen, administered forfeited lands of rebels,

194

Lords of the Central Marches

20 May 1217 then proved the turning point in the war. Reginald de Braose decided that it was time to return to the king’s allegiance and accordingly on 23 June 1217 writs went out to the sheriffs of Hereford, Worcester, Gloucester, Dorset and Somerset, and Devon informing them of Reginald’s return and restoring to him the Braose lands in their counties as his father held them. As noted above, according to Ralph of Coggeshall, Giles de Braose had been appointed custodian of his nephew John de Braose until he should come of age. This was the legal and logical course if indeed (and the evidence is convincing) a demise of the Braose lands had occurred, and thus John was the heir to the Braose estates.¹⁴ But in the writ restoring the Braose lands in England to Reginald there is no mention of his holding the lands in custody; there is no mention of John de Braose at all. John de Braose effectively had been disinherited. What the minority government under William Marshal had done was to bow to political and military necessity. Indeed, the promise of the Braose inheritance was almost certainly held out to Reginald in one of the letters sent him, beseeching him to return to the king’s side. Reginald had apparently petitioned King John to hold the Braose lands by the same agreement and ﬁne that his brother Giles had made, for a writ of 26 May 1216 recorded the king’s agreement to that petition. Peter ﬁtz Herbert, then still a royalist, and Alan Martel were sent to speak with Reginald and escort him under safe-conduct to the king to enter into his service and make security for the ﬁne.¹⁵ The negotiations, however, seem to have been difﬁcult. Two days later, on 28 May, the king wrote to Reginald ‘in the spirit of saner counsel’ (which would seem to indicate that negotiations had indeed not gone well, perhaps owing to the presence of Peter ﬁtz Herbert, who had gained Blaenllyfni in the royal courts). The negotiators were now the bishop of Coventry, Peter ﬁtz Herbert, Alan Basset, and Thomas of Erdinton. These men were also given letters of credence to the group of royalist Marcher lords who held the region for the king: Walter de Lacy, Hugh de Mortimer, John de Monmouth, and Walter de Clifford.¹⁶ But by 7 August 1216 Reginald had still not come in to make his peace.¹⁷ Given his family’s history, Reginald de Braose had good reason not to trust King John. He may also have had second thoughts about and in February 1217 had the farm of the county turned over to him to ﬁnance operations; ibid. 285, 288b (seisin of Kingsland), 298. ¹⁴ See Ch. 6. ¹⁵ Rot. Litt. Pat. 184. ¹⁶ Ibid. 184b. ¹⁷ Ibid. 192.

The Reign of Henry III

195

accepting the deal which his brother Giles had made with the king for the restoration of the Braose lands if that agreement meant that Giles was to hold in custody until his nephew John came of age. This would prevent Reginald from succeeding to the lordship. The deal would also seem to have recognized Peter ﬁtz Herbert’s possession of the Brecon sub-lordship of Blaenllyfni, which later was to be a bone of contention between Reginald and Peter. As it was, the death of King John, as with so many other things, would seem to have changed the equation. The minority government was now headed by William Marshal, a Marcher lord himself and, since 18 November 1215, the keeper of all the Braose castles.¹⁸ The Marshal had as sure a grasp of the importance of the March as did King John. Although the Welsh had not struck into England, the alliance of Reginald and Llywelyn did threaten the government in the west, and it was best to detach Reginald from Llywelyn by whatever means possible. John de Braose was still a minor and incapable of playing a political role in the Marcher lordships of his family. This decision, however, was to lead to a protracted struggle between Reginald and John in the royal courts over the Braose inheritance. Reginald’s return to royal allegiance certainly also covered the submission of those of his men who had been in rebellion with him.¹⁹ At some point after his embassy to Reginald de Braose in 1216, Peter ﬁtz Herbert, the curialis to whom the king had granted Blaenllyfni in the lordship of Brecon in 1211, had crossed to the rebel side. The date is uncertain; the arrival of Louis of France may have had an effect and Reginald de Braose probably threatened Peter’s hold on Blaenllyfni. By crossing to the rebel side, ﬁtz Herbert could hope that Llywelyn would guarantee that possession. A little over a month after Reginald’s submission, however, Peter ﬁtz Herbert came back into the king’s allegiance, and on 24 July 1217 a writ was addressed to Reginald and Walter de Lacy restoring to Peter his castle of Blaenllyfni and all his lands in the honour of Brecon.²⁰ Reginald was apparently loath to restore Blaenllyfni to Peter, for another writ ordering its return was addressed to him on 9 September 1217.²¹ To Reginald, the recognition of Peter ¹⁸ Ibid. 159. The exceptions were Bramber Castle, which had been in the custody of John de Monmouth and was to be delivered up to the king himself, and the bishopric of Hereford and the castles pertaining to it, which were put in the keeping of Walter III de Clifford. ¹⁹ Such as Robert Devereux, Walter de Stokes, and Richard Tirel. These men are mentioned in the letter of Walter de Clifford from 1215 in CACW 1, 1. 10. ²⁰ Rot. Litt. Claus. i. 316. ²¹ Ibid. 320b.

196

Lords of the Central Marches

ﬁtz Herbert’s claim to Blaenllyfni in 1211, when the Braoses had lost all their lands and were either dead or in exile, was likely intolerable, and he seems to have determined to hold onto it. Reginald and Peter would ﬁght in the curia regis in the course of 1219–20 over Blaenllyfni, a suit which was complicated by John de Braose’s claim to the Braose lands.²² This behaviour was, of course, of a pattern with the pursuit of local feuds and the unwillingness to relinquish gains that was such a feature of both the civil war and the early years of the minority.²³ Reginald also attempted to expand his local power, for he was ordered on 18 October not to send his men into either the castle of Newport or the lordship of Gwynllwg, both in the custody of the Marshal.²⁴ As with other places throughout England, the minority government also found it necessary to deal with other claims in the region. William de Gamages asserted his claim by right of his father to the manor of Dilwyn in Herefordshire, which was held in 1217 by William ﬁtz Warin. At Worcester on 27 July 1217 ﬁtz Warin, in the presence of the Marshal, Walter de Lacy, and the council, admitted that he had no right in the manor.²⁵ William de Gamages was temporarily restored to possession and then offered 80 marks for a judgment concerning Dilwyn and a ‘reasonable recognition’ of ﬁtz Warin’s admission at Worcester. The sheriff of Herefordshire was ordered to take the manor into his hand until the ﬁrst Friday after the octave of the Assumption, on which day justice would be done.²⁶ At the same time, Reginald de Braose was contesting the grant of the custody of the lands and heir of Ralph of Pembridge, a Radnor tenant, to William de Cantilupe, one of King John’s familiares and a former sheriff of Hereford.²⁷ Pembridge with its outliers was one of the chief manors of the Braose lordship of Radnor. Ralph had apparently held a tenement in capite and thus the wardship and custody fell to the Crown, but once again Reginald seems to have used his valuable position along the March to resist all encroachments on his lordship. This issue too would be contested in the curia regis.²⁸ ²² CRR viii. 2, 27; ix. 244. ²³ For one example among many, see the earl of Salisbury’s claim to Lincoln Castle and his struggle to secure the de Vesci custody in Northumberland; Carpenter, The Minority of Henry III, 66–7, 71–2, 84–5. ²⁴ Rot. Litt. Claus. i. 330b. ²⁵ Ibid. 317. ²⁶ Ibid. 320. ²⁷ Ibid. 370b. ²⁸ CRR viii. 2, 27, 86, 208, 240; ix. 17, 246, 258; x. 112. The case was complicated by the entry of Henry of Pembridge with his own claim to the custody and the wardship; the case became a three-way battle.

The Reign of Henry III

197

While the minority government tried to sort out these local matters, its chief concern in the region was to reach a peace agreement with Llywelyn and the Welsh. Reginald de Braose’s return to the royal side in 1217 had greatly angered Llywelyn. Builth and then Brecon were attacked by the Welsh, and the town of Brecon only escaped destruction by the proffer by its burgesses of 100 marks. Llywelyn then turned towards Gower, and Reginald was forced to surrender Swansea to the Welsh prince in order to avoid the further ravaging of his lands.²⁹ Nothing demonstrated more thoroughly the power of Llywelyn should he wish to use it. Thus, after the peace of Kingston ended the civil war in 1217, Welsh affairs remained on the slate of unﬁnished business for the minority government. An agreement with Llywelyn ap Iorwerth therefore was reached at Worcester in March 1218.³⁰ In many aspects this treaty recognized the de facto power and hegemony of Llywelyn over Wales. During the war, he had decisively dealt with Gwenwynwyn of Powys and had made his power felt in south Wales. The treaty can be viewed from two angles: that of the English central government and that of the Marcher lords. The minority government was concerned to reassert certain important principles: Llywelyn should accept no enemies of the Crown in Wales, and all Welsh magnates were to do homage and fealty to the king of England. The latter point was particularly important. During the rebellion Llywelyn had brought the princes of Wales under his leadership; in 1218 he probably sought to have this wartime arrangement formalized by inserting himself as the intermediary recipient of homage between the Welsh princes and the king of England. The minority government was loath to approve such a fundamental rearrangement of Anglo-Welsh political relations while the king was under age.³¹ Thus, Wales was not recognized as a principality wherein the Welsh lords did homage and fealty to Llywelyn, who in turn performed it to the English king. Llywelyn, however, would over the next few years attempt to circumvent these restrictions. More important to the Marcher barons than constitutional arrangements were the material aspects of the settlement. A number of them had lost land to Llywelyn and the Welsh during the civil war: the ²⁹ Brut, 95, s.a. 1217. ³⁰ The next two paragraphs rely heavily on Carpenter, The Minority of Henry III, 74–7. ³¹ J. B. Smith, Llywelyn ap Gruffudd, Prince of Wales (Cardiff, 1998), 20–2.

198

Lords of the Central Marches

Mortimers lost Maelienydd, while the royal ally Gwenwynwyn had lost Montgomery and control of Powys, and the Crown itself had lost Ceredigion and Carmarthen (which had been in Marshal custody). In south Wales the Marshals had lost Emlyn and Cilgerran, the ﬁtz Martins Cemais, the Brians Laugharne, the Camvilles Llansteffan, the Londres Kidwelly, and the Braoses St Clears and Gower. Llywelyn was charged by the treaty to secure the return of these lands to their English lords. He surrendered some of these gains (Ceredigion and Carmarthen) to the papal legate Guala, only to receive them back in custody until the king came of age.³² More importantly from our perspective, he retained the sometime Mortimer lands of Maelienydd in his own hands. Llywelyn admittedly had trouble forcing the Welsh princes, particularly those of Deheubarth, to comply in restoration.³³ But, the truth of the matter was that, with royal demesnes and castles in England lost, local government disrupted, and royal ﬁnances at a shocking low, the minority government in 1218 was in no position to apply force against Llywelyn in order to regain lost Marcher lordships. The Treaty of Worcester recognized a fait accompli. No mechanisms were put in place to ensure the return of lost Marcher lands and castles, and this oversight was to be a cause of friction between the government and Llywelyn.³⁴ This can be seen from the situation of the Mortimers and Maelienydd. A contested lordship, Maelienydd had been lost to the Welsh during the civil war of Stephen’s reign, and again at some point during the civil war of 1215–17.³⁵ Hugh de Mortimer, determined to reverse this loss, bolstered his position along the border by acquiring the manors of Knighton and Norton in Shropshire in an exchange with Thomas de Erdington.³⁶ This exchange created further problems along the border, as Thomas had claimed to hold the two manors of Llywelyn. The king ratiﬁed the deal in April 1218, but Llywelyn objected, stating that ³² Llywelyn also surrendered to the legate Montgomery and the other lands of Gwenwynwyn of Powys (who had been allied to John), receiving them back in custody until Gwenwynwyn’s heir came of age. ³³ Smith, Llywelyn ap Gruffudd, 22–5. ³⁴ Walker, ‘Hubert de Burgh and Wales’, 469. ³⁵ See Ch. 2. During the civil war Llywelyn numbered among his supporters the two sons of Maelgwn ap Cadwallon, a family that had long opposed the Mortimers in Maelienydd; Crump, ‘The Mortimer Family and the Making of the March’, 122. ³⁶ BL, Harleian MS 1240, fo. 68v (Thomas de Erdington’s conﬁrmation). Hugh de Mortimer exchanged the manor of Worthy (probably the present-day Worthy and Headbourne Worthy; see The Victoria History of Hampshire, iv. 489) in Hampshire for the two Shropshire manors.

The Reign of Henry III

199

Thomas had never held Knighton and Norton of him but rather that he had held them in free marriage. Mortimer never took possession of the two manors. Hugh de Mortimer, however, did not give up in his attempts to regain Maelienydd. In 1220 the minority government, in return for recognizing Llywelyn’s son Dafydd as his sole heir, appeared to have extracted an agreement which, among other things, promised the return of Maelienydd to Mortimer.³⁷ But once again the English government did not have the power to enforce the agreement, and Llywelyn negated the deal by claiming that all that was due for Maelienydd was the homage of its nobles channelled through himself as prince of north Wales. He also sternly warned the English government of the certainty of war if Hugh de Mortimer should use force to try and regain the disputed lordship.³⁸ In May 1220, with a resumption of royal demesne and castles yet to be accomplished and magnates still controlling various parts of the kingdom as if they were private ﬁefdoms, the English government had neither the resources nor the leisure to ﬁght a war on behalf of Hugh de Mortimer’s claim to Maelienydd. Llywelyn also cleverly undermined Reginald de Braose. In 1219 John de Braose came of age and brought a suit against his uncle Reginald for the Braose inheritance.³⁹ This dispute was to drag on until a ﬁnal settlement was reached in 1227, with Reginald retaining control of the family’s Welsh lands and the honour of Totnes, while John received the rape of Bramber in Sussex.⁴⁰ Llywelyn, after Reginald’s return to the royal side in 1217, had married his daughter Margaret to John de Braose in 1219.⁴¹ The Welsh prince in 1220 then delivered the former Braose lordship of Gower to his son-in-law.⁴² It was during this period of English ineffectiveness along the March that the great programme of fortiﬁcation in the region was carried out. The Marchers could not count on the support of the minority government and they did not have the force themselves to drive Llywelyn out of the lordships he had seized during the war. Indeed, the threat of ³⁷ Carpenter, The Minority of Henry III, 191–2. ³⁸ CACW 8–9, 2. 7. ³⁹ The ﬁrst notice of the suit is in CRR viii. 27 (Michaelmas 1219–20), where Reginald defended himself in the suit of Peter ﬁtz Herbert for the recovery of Blaenllyfni by stating that he had ﬁrst been impleaded by John de Braose and asked it to be considered whether he should have to answer two suits for the same land. ⁴⁰ The course of the suit may be followed in CRR ix. 244; x. 181; xi. 73, no. 390, 395, no. 1969; xii. 333, no. 1611; xiii. 30, no. 133. For commentary, see Holt, ‘The Casus Regis’, 322–3. ⁴¹ Brut, 97, s.a. 1219. ⁴² Brut, 99; Walker, ‘Hubert de Burgh and Wales’, 471.

200

Lords of the Central Marches

further Welsh gains was always a possibility. Although the construction of masonry castles with round towers in the region is difﬁcult to date, there is a good probability that they date from the ﬁrst part of the thirteenth century.⁴³ Advances in military architecture, originating in France, were spread to the British Isles by King John and certain of his barons and intimates who had served on the Continent.⁴⁴ Pembroke, a Marshal castle, is probably the earliest of the Welsh round keeps, and was followed by Usk.⁴⁵ Other lords, including the Cliffords at Bronllys and the Pichards at Tretower, then imitated the innovations of the Marshals and Hubert de Burgh in the south-eastern March. Architecturally, these advances in fortiﬁcation included ‘planned batteries’ of crossbow slits to provide cover for the curtain walls, semi-circular mural towers, and the emergence in the 1220s of the twin-towered gatehouse, the earliest and ﬁnest example (perhaps) being Montgomery.⁴⁶ This programme of fortiﬁcation helped to put many Marcher lords in debt to the Jews of Hereford and elsewhere.⁴⁷ Reginald de Braose for one may have been short of ﬁnancial resources.⁴⁸ He was also unfortunate in that he was ﬁghting several suits in the curia regis at the same time. In addition to his nephew John’s suit for the Braose inheritance, in 1219 Reginald was engaged in a dispute with William de Cantilupe over the custody of the heir and lands of Ralph of Pembridge, a tenant (as mentioned above) of the Braose Marcher lordship of Radnor, as well as a tenant in capite of the king (or so it was claimed).⁴⁹ There was ⁴³ D. F. Renn, ‘The Round Keeps of the Brecon Region’, 137, using architectural evidence, places the emergence of round keeps in the period 1185–1245. Jeremy K. Knight, The Three Castles, Welsh Historic Monuments pamphlet (Cardiff, 2000), 6–10, places their appearance in the ﬁrst three decades of the 13th century. ⁴⁴ Military architecture had advanced considerably in France under Philip Augustus, and men, such as William Marshal and Hubert de Burgh, who played a role along the March certainly were exposed to the new trends in fortiﬁcation during the campaigns in Normandy under Richard I and John. ⁴⁵ Pembroke bears close similarity to the castle of Laval in France, which was constructed slightly before that at Pembroke; likewise, the Marshal castle at Usk is similar in layout and details to that at Le Coudray-Salbart near Niort; Dublin Castle and Trim also were possibly modelled on Le Coudray-Salbart. See Jeremy K. Knight, ‘The Road to Harlech: Aspects of Some Early Thirteenth-Century Welsh Castles’, in John R. Kenyon and Richard Avent (eds), Castles in Wales and the Marches: Essays in Honour of D. J. Cathcart King (Cardiff, 1987), 75–88. ⁴⁶ Ibid. 83. ⁴⁷ See Ch. 3. For the same trend among knightly families in the region, Ch. 4. ⁴⁸ If his few dona to Brecon Priory can be taken as an indicator (which is arguable). Two charters of his survive in the Brecon Cartulary, one conﬁrming the gifts of his father and antecessores, the other a grant of 5s. rent from Brecon for the lighting of the priory church on the celebration of the Mass of the Virgin Mary; ‘Cart. Brecon’, 14. 158–9. ⁴⁹ CRR viii. 27.

The Reign of Henry III

201

also the suit of Peter ﬁtz Herbert for Blaenllyfni, mentioned above. And by 1222 Reginald was in court against William de Stuteville over another custody.⁵⁰ This litigation could only serve as a drain on Reginald’s resources. In the course of the thirteenth century the lords of Brecon would seem to have lost much of the control which they had possibly once had in the manorialized lowlands over both land and tenants; by the fourteenth century, the lord of Brecon’s ‘authority was often no more than indirect and spasmodic’.⁵¹ At what point this loss of control began is difﬁcult to pinpoint, but there is the deﬁnite possibility that it had started when the lordship was in royal hands under King John. Although the mesne tenants of the Braose lordships would seem largely to have backed Giles and Reginald de Braose in their attempt to win back their father’s lands in 1214–15, it is important to remember that King John had successfully managed to attract many of those same tenants into his service against their lord in the period 1208 to 1210.⁵² Reginald de Braose may have had to cede some of his rights and authority to his mesne tenants in order to re-establish his lordship. The renewed strength of the Welsh would also have curtailed the authority of Reginald and the other Marcher lords of the region, especially in the upland areas, where their lordship was never strong to begin with. This may help to explain a conﬁrmation of a grant of Walter de Clifford to Dore Abbey made in the 1220s. Walter II de Clifford had originally granted to the abbey Nanteglwys in Cantref Selyf at some point between 1199 and 1203.⁵³ This charter was then later conﬁrmed by Walter III de Clifford after the death of his father in 1221 and royal conﬁrmation was secured on 20 August 1227.⁵⁴ While ﬁlial piety was most likely the driving force behind the conﬁrmation, the possibility that the Cliffords may have been unable to assert their (albeit loose) lordship in the Epynt Mountains of Cantref Selyf in the face of resurgent Welsh power may well have made this act easier. ⁵⁰ CRR x. 288 (Trinity 1222). ⁵¹ Davies, Lordship and Society in the March of Wales, 92. Although Davies is referring largely to the areas beyond the manorialized lowlands, the necessity for the Bohun lords of Brecon to start quo warranto inquiries against the lords of Pencelli and Cantref Selyf points to a loss of control over military subtenants. ⁵² See Ch. 6. See also the example of the splitting of the Marshal’s afﬁnity; Crouch, William Marshal, 105–6. ⁵³ Sir Christopher Hatton’s Book of Seals, ed. Lewis C. Lloyd and D. M. Stenton (Oxford, 1950), no. 362. ⁵⁴ Mon. Ang. v. 554; CCR i. 58.

202

Lords of the Central Marches

At the same time in 1220 that Hugh de Mortimer was attempting to regain Maelienydd, Llywelyn continued to exercise his power along the border. The government was alarmed in late August when Llywelyn mustered an army, at ﬁrst thought to be directed against Reginald de Braose.⁵⁵ The sheriffs of Gloucester, Hereford, and Worcestershire were directed to ascertain whether the Welsh force had been gathered against Braose, or others of the king’s barons, and were to aid Reginald in defending his lands. As events unfolded, Llywelyn used his army against the man who was to be his chief enemy among the Marcher barons, William II de Marshal, earl of Pembroke, in an attack on Pembrokeshire. Llywelyn’s success in 1220 against the Marshal underscored Welsh strength. The initial concern of the government for Reginald de Braose’s position also shows the relative fragility of the power of the Marchers. This concern was borne out in 1223, a year which J. E. Lloyd for one stressed as a turning point in Anglo-Welsh relations.⁵⁶ In that year the earl of Pembroke landed in Pembrokeshire with an army raised in his Irish lordship of Leinster. Although unrelated, this event had followed on Llywelyn’s seizure of the castles of Kinnerley and Whittington in Shropshire. The government, in the person of the justiciar Hubert de Burgh, had come to the borders and forces had been mustered. A settlement with Llywelyn over the Shropshire manors was within reach when Pembroke landed.⁵⁷ The Marshal was highly successful in wresting back Ceredigion and Carmarthen from the Welsh, but his attack turned the border dispute over Kinnerley and Whittington into a full-scale war in Wales. Llywelyn ﬁrst chose to besiege Reginald de Braose’s castle of Builth.⁵⁸ This was an indication of the relative weakness of the barons of the March (or at least of the restored Braose lordships) as well as of the strength of a mostly united native Wales; one chronicler has Reginald imploring the king for aid, as he himself was not sufﬁciently able to raise the siege.⁵⁹Hubert de Burgh, who had been at Shrewsbury in order to reach a settlement with Llywelyn over the seized Shropshire castles, mustered a second army (one had previously been mustered after the taking of Kinnerley and Whittington) and promptly broke the siege of Builth Castle.⁶⁰ Although ⁵⁵ Rot. Litt. Claus. i. 428. ⁵⁶ Lloyd, A History of Wales, ii. 656; Walker, ‘Hubert de Burgh and Wales’, 473. ⁵⁷ Walker, ‘Hubert de Burgh and Wales’, 474. ⁵⁸ Paris, Chronica Majora, iii. 64 (wrongly attributed to 1221); Annales Monastici, iii (Dunstable Annals), 83 (the castle is not named but is Builth). ⁵⁹ Paris, Chronica Majora, iii. 64. ⁶⁰ Ibid.

The Reign of Henry III

203

1223 was a defeat for Llywelyn’s ambitions, for most of the Marchers the lesson of 1223 was that they were now more than ever dependent on the king for their ultimate survival in the face of the resurgent Welsh. The justiciar Hubert de Burgh had come to play an increasingly prominent role in Wales, particularly along the Middle and southeastern March. In John’s reign he had already had a role in Marcher affairs, for in 1201 he received in custody the Three Castles—Grosmont, Skenfrith, and Whitecastle.⁶¹ In that same year Hubert had kept watch along the March with a force of 100 knights, possibly to check a rebellion by Fulk ﬁtz Warin.⁶² Hubert would have been well aware of the potential instability of the March, an instability heightened after the civil war of 1215–17 by the strength of Llywelyn and the weakness of the minority government in England. With this in mind Hubert paid close attention to Welsh affairs after the resignation of the elder Marshal as rector in April 1219 gave him an increased share in power in the government. The reasons for this close attention were mixed. Hubert de Burgh has long been viewed as an opportunist in terms of his Welsh policy. Both Sir John Lloyd and Sir Maurice Powicke saw Hubert as following a programme of self-interest in the March. Lloyd believed that Hubert attempted to set himself up as a power in south Wales equal to Llywelyn in the north (a view with much to commend it).⁶³ Powicke wrote of Hubert’s aim of establishing ‘a kind of palatinate in Wales’.⁶⁴ Hubert was certainly not without self-interest.⁶⁵ But during Henry’s minority, nobles and sheriffs were often unaccountable and the government was quite weak. As justiciar Hubert, along with the regent William Marshal and the papal legates, was one of a few men who together essentially were the government. Thus, his increased presence in the March was partly also an increased royal presence. And it is important to remember, particularly in light of royal loss of control over the localities, that as late as 1217 Reginald de Braose was in league with Llywelyn. Hubert re-established himself in the March in 1219 when he regained possession of the Three Castles. King John had taken them from him ⁶¹ Rot. Lib. 19. ⁶² Painter, The Reign of King John, 84. ⁶³ Lloyd, A History of Wales, ii. 672. ⁶⁴ F. M. Powicke, King Henry III and the Lord Edward (Oxford, 1947), 70. ⁶⁵ A reassessment of Hubert de Burgh and Wales is found in Walker, ‘Hubert de Burgh and Wales’, 465–94, and more generally in Carpenter, The Minority of Henry III, passim.

204

Lords of the Central Marches

in 1205 and given them instead to William de Braose, at that time still in John’s favour. In 1218 Hubert demanded the castles back from Reginald de Braose, claiming them ‘as his right’. The minority council, headed by the Marshal, decided in early December 1218 in favour of Hubert owing to Reginald’s default.⁶⁶ On 26 January 1219 a writ was sent to the sheriff of Herefordshire ordering him to deliver seisin of the Three Castles to Hubert; the sheriff was to be supported by Hugh de Mortimer, Robert de Mortimer of Richard’s Castle, Walter de Clifford junior (Walter III) and Roger his brother, John de Monmouth, and others of the county with their power (posse). Resistance by Reginald de Braose to the council’s judgment was clearly seen to be a possibility. But in the list of men ordered to support the sheriff in delivering seisin we can see those in the region on whom the Crown felt it could rely. Hugh de Mortimer was dependent on the Crown for leverage in his attempts to regain his lordship of Maelienydd from Llywelyn. Walter III de Clifford and Roger de Clifford both had a background of royal service, having been household knights of King John.⁶⁷ John de Monmouth had been one of the executors of the late king’s will; in 1216 he had received a large number of custodies. The great men of the area were all to some degree now attached to, or dependent upon, the Crown.⁶⁸ This will be important to remember when examining the behaviour of the lords and knights of the region during the rebellion of Richard Marshal in 1233–4. While Reginald de Braose attempted to reassert control over his lordships in the March and stave off various legal challenges as well as the aggression of Llywelyn, Walter de Lacy increasingly turned his attentions to his lordship of Meath in Ireland. While Lacy was sheriff of Herefordshire from August 1216 to Michaelmas 1221, undersheriffs performed his tasks and accounted for him at the exchequer.⁶⁹ In 1219 he was still attempting to regain seisin of lands in Ireland of which he had been disseised by King John.⁷⁰ In 1220 Walter was in the king’s ⁶⁶ Rot. Litt. Claus. i. 404. ⁶⁷ See Ch. 6; also Church, The Household Knights of King John, 24. ⁶⁸ Carpenter, The Minority of Henry III, 149 and n. 19, points out that the respite of Walter de Lacy’s account at the exchequer of his term as sheriff of Herefordshire (Rot. Litt. Claus. i. 400b, 20 Sept. 1219) and the gift of 100 oaks to his wife, Margaret (ibid. 394b, 3 July 1219), would have been helpful for securing Lacy’s support in installing Hubert in the Three Castles. Manipulation of debt was used again in 1233–4 to keep Lacy attached to the Crown; see below, p. 212. ⁶⁹ PRO Lists and Indexes, ix: List of Sheriffs for England and Wales, see under Herefordshire. ⁷⁰ Rot. Litt. Claus. i. 408b.

The Reign of Henry III

205

service in England. On 13 August of that year another writ was sent to the justiciar of Ireland restoring to Walter his lands, excepting the castle of Drogheda, which was kept in royal custody.⁷¹ By 24 September 1220 Walter de Lacy had crossed to Ireland.⁷² The reasons for Walter de Lacy’s focus on Ireland are not hard to discern. Firstly, his Irish lordship was likely to have been disorganized after its time in royal custody, and despoliation may have occurred. Secondly, the potential for gain was greater in Ireland.⁷³ Expansion along the March was for the time being at an end, a situation compounded by the geographic situation of Lacy’s lordship of Ewyas Lacy, situated as it was between Herefordshire and the Braose lordships of Brecon and Abergavenny with no routes for expansion into pura Wallia, if such an aim had been achievable. Meath, much more easily exploited agriculturally, represented much better opportunities for Lacy. By 1222 it would appear that Walter’s son Gilbert de Lacy was in charge of the family’s possessions in Herefordshire and the March, as a royal writ informed William, earl of Salisbury, that the king had made a gift of six oaks from the Forest of Clarendon to Gilbert for building a chapel at his curia of Bredwardine.⁷⁴ Walter de Lacy was thus following the example of his father-in-law, William III de Braose, in leaving his English and Welsh lands to the care of his son while he concentrated on the family’s Irish possessions. What military action did occur along the March between the Wye and Severn in the 1220s was the result of royal, for which read Hubert de Burgh’s, initiative, not that of the Marchers. The short campaign of 1223 to relieve Builth Castle has been mentioned above, and resulted in the construction of Montgomery Castle following the breaking of the siege.⁷⁵ Montgomery was to be one of the hinges for future royal action against the Welsh, as well as forming another piece in Hubert de Burgh’s Marcher lordship.⁷⁶ The next campaign in Wales was ⁷¹ Ibid. 427b. ⁷² Ibid. 430b. The writ informs the barons of the exchequer that Lacy had appointed Warin de Grendon to act in his stead at the exchequer for the county of Hereford so long as Lacy should be in Ireland. ⁷³ For a discussion of the web of aristocratic landholding that spanned the Irish Sea, see Frame, ‘Aristocracies and the Political Conﬁguration of the British Isles’. ⁷⁴ Rot. Litt. Claus. i. 495. ⁷⁵ Paris, Chronica Majora, iii. 64. ⁷⁶ In the 14th century it was claimed that ‘the greater part of Wales was conquered from the lordship of Montgomery’; Calendar of Ancient Petitions relating to Wales, ed. W. Rees (Cardiff, 1975), nos 12075, 12780. I am grateful to Rees Davies for drawing these to my attention.

206

Lords of the Central Marches

the ill-fated Ceri campaign of 1228. The cause of the breakdown in Anglo-Welsh relations that year may have been caused by the grant of the castle and lordship of Montgomery to Hubert de Burgh, but it is also possible that the desultory course of negotiations between the Crown and Llywelyn was to blame.⁷⁷ Either way, the campaign achieved nothing, if indeed it was meant to achieve anything in the ﬁrst place.⁷⁸ But, excepting the younger Marshal’s actions of 1223, campaigns in Wales tended to be royal ones. This contrasts with the long period from 1165 to 1210 in which no royal armies were sent into Wales; royal policy was now more active and the Marchers had to follow the king’s lead. By 1228 the personnel in the region had changed. Walter II de Clifford had died in 1221 and was succeeded by his more turbulent son Walter III.⁷⁹ In 1227 Hugh de Mortimer died and was succeeded by his brother Ralph. Following the failure of the 1228 Ceri campaign, Ralph de Mortimer, doubtful of the government’s ability to help in the reconquest of Maelienydd from the Welsh, sought a closer relationship with Llywelyn. In 1229 he took as his wife Gwladys Ddu, daughter of Llywelyn and widow of Reginald de Braose, who had died in 1228. This not only brought the Mortimers into a close relationship with their great opponent, but also gained them the disputed manors of Knighton and Norton.⁸⁰ The main Braose line in Brecon was extinguished by 1230, causing further realignments in the region. In 1228 William V de Braose had succeeded his father, Reginald. William was then captured during the Ceri campaign of that year, and later met his death in 1230 when, having been caught conducting an affair with Joan, the wife of Llywelyn, he was summarily judged and hanged by the Welsh prince.⁸¹ The politics behind this execution more than explain it, for Llywelyn stood to proﬁt by Braose’s removal. The death of William V de Braose was signiﬁcant, for the main Braose line ended with him and the assemblage of Braose lordships in the March was broken up. ⁷⁷ For the ﬁrst interpretation, see Lloyd, A History of Wales, ii. 667; for the second, Walker, ‘Hubert de Burgh and Wales’, 478. ⁷⁸ See the remark of Davies, The Age of Conquest, 298, that concerning the three royal expeditions into Wales of 1223, 1228, and 1231, ‘not one effected, or was calculated to effect, a major change in the territorial settlement of 1218’. ⁷⁹ Rot. Litt. Claus. i. 446b (relief). ⁸⁰ Crump, ‘The Mortimer Family and the Making of the March’, 124. ⁸¹ Patent Rolls 1225–32, 239; Roger of Wendover, Flores Historiarum, ii. 350; Paris, Chronica Majora, iii. 194; Patent Rolls 1225–32, 336, 339.

The Reign of Henry III

207

The custody of the Braose lands, along with the marriage of the female heirs, was given to William Marshal, William de Braose’s brother-inlaw.⁸² William left four daughters as heirs, Maud, Isabel, Eleanor, and Eva. Isabel, along with the strategically placed lordship of Builth, had been promised to Dafydd, the son of Llywelyn, as one of the conditions of William’s release from captivity in 1229 following his capture in the Ceri campaign. After the marriage took place, Llywelyn razed Builth Castle to the ground.⁸³ The other three daughters were married to men of the region. Maud was married to Roger II de Mortimer, the son of Ralph de Mortimer and future companion of Edward I, taking with her the lordship of Radnor. Eleanor, whose portion of the Braose inheritance was Brecon, was married to Humphrey de Bohun. Eva married William III de Cantilupe, whose father was a royal steward, sometime between 1238 and 1245, and brought Abergavenny to the Cantilupes. Once again the accidents of death and female heirs led to a dramatic reshaping of the feudal geography of the March. The fortunes of the Mortimers and Bohuns were spectacularly improved in particular; it is interesting to note that, owing to the descent of the lands of Miles of Gloucester in the twelfth century, the Bohun earls of Hereford held at most four manors in the county before the Herefordshire manors of the lordship of Brecon passed to Humphrey.⁸⁴ The next crisis to envelop the region was the most serious since that involving the Braoses in 1208–10: the rebellion of Richard Marshal in 1233–4. William II Marshal died on 5 April 1231. His brother Richard Marshal succeeded him, though with some difﬁculty owing to Richard’s status as a vassal of the king of France. The traditional explanation for the Marshal’s revolt centres on the factionalism of English politics in the 1230s, a struggle ostensibly between Henry’s ‘foreign’ court favourites and those English magnates who opposed them. Hubert de Burgh had been removed from power in a coup in August 1232. Henry III had surrounded himself with new councillors since coming of age, and the failure of the 1231 Welsh campaign cannot have been helpful to ⁸² Patent Rolls 1225–32, 377. William V de Braose was married to Eva Marshal. ⁸³ Davies, The Age of Conquest, 247. ⁸⁴ For the descent of the lands of Miles of Gloucester, see Ch. 2. The Herefordshire manors which passed to the Bohuns with Eleanor de Braose were Tedstone Wafer, Avenbury, Venn, Little Cowarne, Bodenham, Burghill, Kingstone, Moccas, and Bredwardine.

208

Lords of the Central Marches

Hubert’s cause. The Marshal has been seen as championing Hubert’s cause, and at the same time trying to restore the position of allies of the Marshal family, the Bassets, who had been stripped by Henry of the manor of Upavon. But even contemporaries saw the root cause as a reaction against the rise of the Poitevin councillors of the king, particularly Peter des Roches, bishop of Winchester, and his nephew Peter de Riveaux. Nicholas Vincent has advanced a different explanation, one ultimately much more rooted in the local politics of the March. Vincent believes that the cause was not so much the various grievances of the English baronage in 1233—arbitrary royal government, including the seizure of manors per voluntatem regis, the levying of enormous ﬁnes as a means of political pressure, the taking of hostages—nor even the promotion of Peter des Riveaux as a power along the March, but rather the loss by the Marshal of control over his own household and afﬁnity. By taking up the grievances of these men he had effectively allowed himself to be swept into their rebellion.⁸⁵ Richard Marshal, according to this interpretation, was swept along by the actions of an afﬁnity which he could not control, actions which found some support in the March owing to the turmoil caused by the recent deaths of several Marcher tenants-in-chief, the threat of Welsh attack (realized in 1231), the Marchers’ long-standing suspicion of the king’s intentions, and their resentment against aliens who were intruded into the region as constables.⁸⁶ In the years prior to his downfall in 1232, Hubert de Burgh had managed to gain control of the Three Castles, the castle and lordship of Montgomery in hereditary right, and the honours of Carmarthen and Ceredigion, to which John de Braose’s lordship of Gower had been subordinated. His downfall brought these lordships back into direct royal control. At the same time, the death of William V de Braose in 1230 and the subsequent death in 1232 of John de Braose in a riding accident brought the entire Braose inheritance as it stood in 1210 back into royal hands.⁸⁷ Although the lands of William de Braose were initially entrusted to the keeping of William II Marshal, after the latter’s death on 7 April 1231 the king’s disposal of these estates was not designed to court the opinion of the Marchers.⁸⁸ On or about 11 April 1231 Henry III’s brother Richard of Cornwall was granted ⁸⁵ Vincent, Peter des Roches, 396–8. Their grievances admittedly still included the Basset loss of Upavon, a point stressed to me by David Carpenter. ⁸⁶ Ibid. 372–4. ⁸⁷ Brut, 101–2, s.a. 1230; 102, s.a. 1232. ⁸⁸ CR 1227–31, 353, 359.

The Reign of Henry III

209

custody of the Braose lands along the eastern March.⁸⁹ Richard of Cornwall was an outsider to the region. Owing, however, to the king’s suspicion of his brother’s close links to Richard Marshal, on 15 May 1231 Cornwall lost his custody of the Braose lands in the March to Hubert de Burgh.⁹⁰ But even worse for Marcher sensibilities would be Peter des Riveaux’s growing stature in the March. Between 18 and 22 September 1232 Hubert de Burgh’s Marcher lordships and castles were given to Riveaux’s keeping.⁹¹ In July 1232 Riveaux had already received the custody of the lands of John de Braose in Sussex and Gower with the marriage of the heir, who was to be married to his niece.⁹² Also around mid-September 1232 Riveaux was appointed bailiff for life of St Briavel’s Castle and the Forest of Dean, a grant followed by the award of the Three Castles.⁹³ So much patronage, distributed to an alien in a short span of time and a limited geographic area, was almost tailor-made to offend the sensibilities of those local magnates who might reasonably have hoped to proﬁt from the same custodies and marriages. In 1233 Richard Marshal abruptly left the royal court following the Upavon decision, and a party of the disaffected gathered around him. This group included Gilbert Basset, the Marshal associate who had been stripped of Upavon by the king. It also included Walter III de Clifford, as orders went out for the arrest of Basset and Clifford on 15 June.⁹⁴ Clifford’s reasons for allying with the Marshal are uncertain. Vincent speculates that either an attack on his vassal Hugh of Kinnersley in June 1233 or an association with Margaret, daughter of Llywelyn and the widow of John de Braose, whom Clifford was later to marry, may have been behind his disaffection.⁹⁵ A more reasonable explanation may be that as lord of Clifford and as a man who had served the Crown loyally since John’s reign, Clifford felt alienated by his complete lack of gain from the Braose estates now in royal control. It is also perhaps worth noting that Walter III de Clifford was capable of violent and overbearing action. In 1230 he and some of his men, including Hugh of Kinnersley, had wounded and imprisoned some of the men of the prior of Wenlock ⁸⁹ N. Denholm-Young, Richard of Cornwall (Oxford, 1947), 19; CR 1227–31, 585. ⁹⁰ Patent Rolls 1225–32, 434. ⁹¹ Ibid. 550–1. These lordships were the honours of Carmarthen, Ceredigion, and the custody of Glamorgan with the castles of Cardiff and Newport, which de Burgh had as guardian of Richard de Clare. ⁹² Patent Rolls 1225–32, 491; Vincent, Peter des Roches, 356–7. ⁹³ CR 1231–34, 444; Vincent, Peter des Roches, 318. ⁹⁴ Vincent, Peter des Roches, 380. ⁹⁵ Ibid. 380–1.

210

Lords of the Central Marches

in the course of a dispute in Shropshire; they had subsequently refused to appear in the county court.⁹⁶ And it was the same Walter de Clifford who supposedly in 1250 made a royal messenger swallow the royal writ addressed to him, seal and all, and return to Westminster.⁹⁷ Walter de Clifford was one of the more likely men in the March to join in a rebellion against the king. Though Richard Marshal’s rebellion, despite his own death, was largely successful in its aims, in the south-east March it was a damp squib, primarily owing to the lack of support he received from the Marchers. Walter de Clifford stayed in the rebel camp for only a brief period; Walter de Lacy and John de Monmouth were ﬁrmly on the royal side. With the exception of Clifford’s tenants and the men of Ewyas Castle, few of the knights from the region’s honours joined the Marshal’s cause. The course of the Marshal’s revolt has been plotted in detail elsewhere; rather, our aim is to outline the reasons why the rebellion in the region was not perhaps as successful as might have been expected.⁹⁸ At ﬁrst glance the March would seem to have been a propitious area from which to defy the king. The region had a long history of revolts going back to the end of the Conqueror’s reign and had played an important part in the Angevin cause during the civil war of Stephen’s reign.⁹⁹ Although localized, the rising of the Braose tenants against King John in 1215 had been successful.¹⁰⁰ But Giles and Reginald de Braose had relied on their tenants in Brecon, Abergavenny, and Radnor. Richard Marshal’s chief lordships in Wales were Pembroke, Strigoil, and Netherwent, with Pembroke the most important. His power and inﬂuence along the border with England, scene of much of the early action of the war, was much less than it was in western Wales. In the central and south-eastern March, the Marshal’s chief supporter was Walter de Clifford. Although in August 1233 on the opening of the revolt the Marshal’s supporters, most probably Clifford’s men, had taken the castles of Hay, which lay between the Clifford castles of Clifford and Glasbury, and Ewyas, which lay near the lordship of Clifford, the castles were regained by royal forces between 28 August and 7 September 1233.¹⁰¹ This quick reverse brought Walter de Clifford back into the royal fold. At some point between 17 and 25 August Clifford swore allegiance to the king, ⁹⁶ CRR xiv. 131, no. 661. ⁹⁷ Paris, Chronica Majora, v. 95. ⁹⁸ For a detailed narrative of the Marshal rebellion, see Vincent, Peter des Roches, 399–428. ⁹⁹ See Ch. 2. ¹⁰⁰ See Ch. 6. ¹⁰¹ Vincent, Peter des Roches, 389, 399.

The Reign of Henry III

211

delivering a charter for security.¹⁰² At the same time, Clifford’s knights John Devereux, Matthew de Mans, and William Dunre returned to the king’s side, making charters pledging their lands as security and delivering their heirs as hostages.¹⁰³ By 29 August the Clifford castle of Glasbury was in royal hands, the sheriff of Herefordshire, William ﬁtz Warin, being appointed as its keeper.¹⁰⁴ On 1 September the king with his army was at Hay, where Bronllys Castle was given to the keeping of Robert de Nafford; on that same day the constable and men of Ewyas Castle came over to the royal side. On 3 September members of the garrison of Clifford Castle were granted a safe-conduct, no doubt in order to treat with the king.¹⁰⁵ When the Marshal’s castle at Usk fell by 13 September (although the delay in its fall was rather humiliating for the king), the Marshal’s cause in the central and south-eastern March was effectively ﬁnished.¹⁰⁶ The rebellion in the region had lasted little more than a month, and this was largely owing to the very limited support that the Marshal attracted in the area. The rebellion was hampered in the south-eastern March because the Marshal’s cause did not attract adherents. Walter de Clifford and his men provided too narrow a base on which to rely. An examination of royal patronage shows the degree to which local men had been cultivated by the Crown. Chief among them was Walter de Lacy, who had been a loyal supporter of the Crown in the civil war of John’s reign. In the 1220s and early 1230s Lacy was in the king’s service in Ireland, and suits against him there were routinely respited. This was a valuable privilege for a lord who was most probably trying to increase control over his lordship of Meath following royal custody and the disturbances of his brother Hugh’s invasion of Ireland in 1224.¹⁰⁷ Perhaps most important was the king’s decision in 1230 that Walter had not demised to his son Gilbert, who had died that year, the lands which Gilbert held in Herefordshire and Shropshire (perhaps the entire honour of Weobley), but that Gilbert held them of his father ad sustentandum. Walter de Lacy was thus able to reclaim the family lands ¹⁰² CR 1231–34, 321. ¹⁰³ Ibid. 320. ¹⁰⁴ CPR 1232–47, 25. ¹⁰⁵ Ibid.; CR 1231–34, 257. ¹⁰⁶ CR 1231–34, 257. ¹⁰⁷ For Hugh de Lacy’s attempt in 1224 to regain control of his earldom of Ulster, see Otway-Ruthven, A History of Medieval Ireland, 90–2. For examples of respites of suits brought against Walter de Lacy in Ireland, see CR 1227–31, 535 (25 July 1231); CR 1231–34, 7 (28 Nov. 1231), 73 (16 June 1232), 182 (13 Jan. 1233), 254 (25 Aug. 1233).

212

Lords of the Central Marches

in Herefordshire and Shropshire for himself, rather than see them held in custody by the Crown, as his grandson Walter may not have been of age.¹⁰⁸ Other local men were the recipients of patronage, even after the Poitevins came to power at court. On 14 December 1232 the sheriff of Herefordshire was instructed that Walter de Baskerville, an important knight of the Lacy honour of Weobley, was to have his arrears from the time that he was sheriff of Herefordshire.¹⁰⁹ On 24 May 1233 Baskerville received a gift of ten oaks from the king.¹¹⁰ Other men of the region had little to gain from joining the Marshal. John de Monmouth had long been a royal servant; for a baron whose lordship comprised only ﬁfteen knights’ fees, royal service magniﬁed many times his standing in local society. Ralph de Mortimer came from a family that had a long history of loyalty to the king, and was as dependent as his brother Hugh had been on the Crown for help in regaining his lordship of Maelienydd. Debt also played a key role in keeping men loyal to the king. Walter de Lacy owed a large sum in both Jewish debts and a large ﬁne for having his lands in Ireland following the invasion of his brother Hugh in 1224.¹¹¹ In Michaelmas 1231 Walter was pardoned 100 marks of the 200 marks instalment owed the exchequer for that term.¹¹² After the outbreak of war Walter de Lacy received further relief of his debts. On 12 December 1233 he was pardoned the 200 marks which he owed the king for that year, and three days later new terms were set for repayment of the debt he owed to Ursell ﬁtz Hamo, Jew of Hereford, and his brothers, with interest being suspended.¹¹³ At the same time, debt was used to bring Walter de Clifford to heel. On 4 July 1233, when Clifford would seem to have drifted into the camp of the Marshal at a time of rising tension on the March, a writ was sent to the sheriffs of Oxfordshire and Herefordshire ordering that Clifford should return without delay the 1,000 marks with interest that he owed Hamo, Jew of Hereford. He was also to pay 100 marks owed Aaron of York, and 40 marks with interest which Roger, his brother, whose heir he was, owed Copin and David, Jews of Oxford.¹¹⁴ In ¹⁰⁸ CR 1227–31, 464. ¹⁰⁹ CR 1231–34, 174. ¹¹⁰ Ibid. 222. ¹¹¹ By his death in 1241 Walter de Lacy owed £666 in Jewish debt; see Ch. 3. He ﬁned 3,000 marks in 1225 for having his lands and those of his men in Meath, which had been seized by the king ‘on occasion of the war with Hugh de Lacy’; Rot. Litt. Claus. ii. 39b. ¹¹² CR 1227–31, 565. ¹¹³ CR 1231–34, 351–2. ¹¹⁴ Ibid. 314.

The Reign of Henry III

213

December 1233, after Clifford had returned to the king’s allegiance and had his lands restored excepting Clifford Castle, 60 marks debt which he owed the Jews of Canterbury was ﬁrst prorogued and then pardoned, his charters, chirographs, and tallies, as well as those of his men, being returned to him.¹¹⁵ Although this was not a great amount, it no doubt raised the hope that further debt might be either prorogued or pardoned, and indeed it does not seem that the recall of his debts ordered in July was carried out. Thus, it is misleading and inaccurate to speak of a wider Marshal ‘afﬁnity’, in which internal tensions and divisions led to the outbreak of conﬂict along the March. With the exception of Walter de Clifford and his men, the rebellion was based entirely upon the Marshal’s household and some of his tenants, particularly in Netherwent, and his allies Gilbert Basset and Richard Siward.¹¹⁶ Nicholas Vincent’s statement that ‘King John’s attack [upon William III de Braose in 1208] left deep scars, especially amongst the baronage of Ireland and the Marches, a closely interrelated group resentful of any outside interference,’ does not hold up in the context of 1232–4.¹¹⁷ True, the Marshals were related to the Braoses, and through them to the Lacys, but marriage was never the sole determinant of political allegiance in this period.¹¹⁸ In 1210 William Marshal was of little help to William de Braose when he ﬂed to Ireland. Indeed, one could just as well expect that the Lacys would have held a grudge against the Marshal family for its lack of support in 1210.¹¹⁹ Nor was there much of a connection between the Marshal’s afﬁnity and that of Walter de Lacy or indeed of any other magnate in the central and south-eastern March, including the men of the former Braose lordships in the region. Walter de Clifford had little connection to the Marshals, and of his men, only Matthew de Mans and John Devereux also held of another lordship in the area, holding Humber and part of Bodenham, respectively, of the honour of Brecon.¹²⁰ These two men were not ¹¹⁵ Ibid. 350–1. ¹¹⁶ See R. F. Walker, ‘The Supporters of Richard Marshal, Earl of Pembroke, in the Rebellion of 1233–1234’, WHR 17/1 (June 1994), 41–65. ¹¹⁷ Vincent, Peter des Roches, 372–3. ¹¹⁸ See the perceptive comment of Holt that in explaining the Northern Rebellion of 1215, ‘A family tie is in itself of little value as evidence’; The Northerners, 67. ¹¹⁹ See Ch. 6. Compare with Vincent’s statement that ‘The memory of these events was to do much to determine Marcher sympathies for many years to come . . .’; Peter des Roches, 373. ¹²⁰ CR 1231–34, 320; Walker, ‘The Supporters of Richard Marshal, Earl of Pembroke, in the Rebellion of 1233–34’, 54; BF 800, 805. Walker is incorrect in his

214

Lords of the Central Marches

signiﬁcant Brecon tenants and did not draw any other Brecon men into rebellion. Robert de Turberville was no longer connected to the Braose lands, as his holding at Crickhowell in Blaenllyfni had been split off with the rest of that sub-lordship and granted to Peter ﬁtz Herbert as far back as 1211.¹²¹ Thus, the tenurial isolation effected by inheritance and marriage patterns from the latter twelfth century could also serve as a brake on the spread of rebellion.¹²² Royal manipulation of Marcher debt also helped to contain the revolt. But, on the other hand, though the king retained or regained the loyalty of the majority of the lords and knights in the region, his inability to roll up completely the rebellion in the south-east meant there was in effect by January 1234 a stalemate with the Marshal, which, given his limited aims (not a general rebellion, but rather to bring down Peter de Roches’s regime), was all he needed. The Marshal family in the person of Richard’s brother Gilbert certainly emerged from the rebellion in a stronger position than they had had before.¹²³ For the remainder of the 1230s and the ﬁrst half of the 1240s, the region was marked by further mortality among its foremost families. Just as the main Braose line died out in 1230, the Lacys had left nothing but female heirs by 1241, when Walter de Lacy died.¹²⁴ His son Gilbert had died in 1230; his grandson by Gilbert, Walter, died without male heirs, probably in 1240. This left two granddaughters as heirs, Matilda and Margaret. By 1242 Margaret de Lacy was married to John de Verdon, a member of a prominent Anglo-Irish family. This marriage may well have been contracted before the death of Walter de Lacy and accorded with the Lacy emphasis on Ireland from the 1220s. Walter de Lacy’s grandson, also named Walter, had been married to a daughter of Theobald Butler, so a conscious strategy of marriage links in Ireland was statement that John Devereux, who held Bodenham of the honour of Brecon and followed Clifford into rebellion, also held at Bredwardine. A portion of Bredwardine was held in right of marriage by Roger Devereux at some point after 1227; see Ch. 4. ¹²¹ See Ch. 6. ¹²² See Ch. 4. ¹²³ I owe this point and much else in my discussion of the Marshal’s aims and achievements in 1233–4 to very informative debate with Professor David Carpenter. ¹²⁴ Walter de Lacy died between 19 Dec. 1240, when he was distrained for £100 owed the Jews of Oxford, and 4 Mar. 1241, when his wife, Margaret, was granted the manor of Staunton Lacy until an extent could be made of Lacy’s lands in England and Ireland and dower assigned; CR 1237–42, 258, 279.

The Reign of Henry III

215

being pursued. The other Lacy granddaughter, Matilda, was married by Easter 1244 to Peter de Geneva (she would later remarry, to Geoffrey de Geneville). The barony of Weobley and the Marcher lordship of Ewyas Lacy were both split between the two daughters and their husbands. There were delays in assigning Margaret de Lacy, Walter’s widow, her proper dower, both in England and Wales and in Ireland. Indeed, she brought a number of suits in the curia regis, attempting to recover dower lands in Herefordshire, and it is quite possible that her granddaughters’ husbands obstructed her in this.¹²⁵ At the same time, the one family which managed to avoid extinction through lack of male heirs, the Mortimers, ﬁnally regained Maelienydd. This was made possible only by the death of Llywelyn in 1240, and the problems that his son Dafydd experienced in securing his succession. As part of the price for recognizing Dafydd’s succession and accepting his homage, made on 15 May 1240, the government reached an agreement that almost certainly included the restoration of Maelienydd to Ralph de Mortimer.¹²⁶ A writ of 3 July 1240 ordered the sheriff of Herefordshire to restore seisin to Ralph de Mortimer of his land of ‘Melenich’ (Maelienydd), which he had recovered through the arbitration between the king and Dafydd conducted by the bishop of Norwich and others at Gloucester.¹²⁷ Mortimer loyalty and patience were ﬁnally rewarded. The restoration of Maelienydd in 1240 demonstrates the point made above, that the grievances and aims of individual Marcher families were, in the 1220s–1240s, being subsumed in and subordinated to the larger course of Anglo-Welsh politics, a course determined by the king and the prince of Gwynedd. The 1240s saw royal power in the Middle March, and indeed in Wales, at its highest point between the brief zenith of 1211 and the conquest of 1282–3. The reason for this is simple: the death on 11 April 1240 of Llywelyn ap Iorwerth. The hegemony which he had exercised over the princes of native Wales, despite the increasing adoption of such formal mechanisms as indentures, homage, and hostages, was still mostly of a personal nature.¹²⁸ That hegemony quickly eroded after his death, to the advantage of both the Crown and the Marchers. ¹²⁵ CRR xvii (1242–3), 135, no. 682; 138, no. 689; 148, nos 750–2; 291, no. 1498; 299, no. 1541; 300, no. 1542. ¹²⁶ Littere Wallie, preserved in Liber A in the Public Record Ofﬁce, ed. J. G. Edwards (Cardiff, 1935), 5–6. ¹²⁷ CR 1237–42, 202. ¹²⁸ Davies, The Age of Conquest, 245.

216

Lords of the Central Marches

As Llywelyn had wished, he was succeeded by his son Dafydd.¹²⁹ But power struggles within Wales soon developed, centring on Dafydd’s half-brother Gruffudd. Client rulers quickly began to desert the new Welsh prince. At Gloucester, on 15 May 1240, Dafydd did homage to Henry III for ‘his right to North Wales’, itself an acknowledgement of the English king’s suzerainty, a power Henry III had tried to claim during Llywelyn’s lifetime and especially in 1238. Dafydd also quitclaimed to Henry and his heirs, in perpetuity, the homages of all the ‘barons of Wales’, and agreed that all claims against him would be referred to a joint panel of arbitrators, under the presidency of the papal legate Otto. Following this agreement, the Marcher lords went on the offensive. The Bohuns began to reassert their control over upland Brecon, and although Maelienydd was not mentioned in the agreement reached at Gloucester, the Mortimers renewed their campaigns there and in Gwrtheyrnion. On 3 July 1240 a writ was sent to the sheriff of Herefordshire, instructing him to transfer the cantref of Maelienydd to Ralph de Mortimer per arbitrium nuper provisum . . . apud Glouc’ inter nos et David ﬁlium Leulini.¹³⁰ From 1241 on Maelienydd was controlled by Ralph de Mortimer, but the conﬂict in this section of the March then shifted to the neighbouring cantref of Ceri, to which the native dynasty of Maelienydd had ﬂed.¹³¹ The Marcher lords thus had used the weakness of Dafydd ap Llywelyn revealed in the Gloucester agreement of 1240 as the opportunity to regain the lands lost to Llywelyn earlier in the century. The royal presence in the March was also increased following Dafydd’s homage. The lordship of Builth, which Dafydd claimed as the dower of his wife, Isabella de Braose, was taken into royal hands, and a royal expedition in 1241 in the north further humbled Dafydd. In 1244 the death of Dafydd’s half-brother Gruffudd in the Tower of London, where he was ostensibly lodged as a guest of the Crown, removed the threat of Dafydd’s rival and was followed by a major uprising by the native Welsh. Resistance was at ﬁrst led by the earls of Gloucester and Hereford, the ﬁrst also lord of Glamorgan and the second the father of the new lord of Brecon. They were joined by the two wardens of the March, John ¹²⁹ For much of what follows, see Lloyd, A History of Wales, ii. 694–706; Davies, The Age of Conquest, 300–7; Smith, Llywelyn ap Gruffudd, 29–36. ¹³⁰ CR 1237–42, 202. ¹³¹ For further discussion of the Mortimer seizure of Maelienydd, see Crump, ‘The Mortimer Family and the Making of the March’, 124–6.

The Reign of Henry III

217

Lestrange in the north and John de Monmouth in the south.¹³² The role played by the latter two men as wardens is in itself an indicator of the increasing coordination and control of military operations along the March by the Crown. In 1244, in response to the rebellion, a major campaign was again launched against north Wales, and although it was inconclusive it seemed that the conquest of Gwynedd was not far off. But on 25 February 1246 Dafydd ap Llywelyn died, taking Wales temporarily off the top of the Crown’s agenda. That year also witnessed the death of Ralph de Mortimer, by 10 August at the latest.¹³³ In one sense the year 1246 marked the ﬁnal eclipse of a March dominated by the Marcher barons. This eclipse, however, had been under way for some time. The reign of King John saw a sudden (too sudden) and dramatic increase in royal involvement and control in the affairs of the March. As has been noted, John’s heavy-handed treatment of Marchers and Welsh alike, combined with unrest in England itself and Continental failure, undid his brief achievement in Wales.¹³⁴ Henry III and his government on the whole took a more gradual approach and focused on control of military and diplomatic matters, while avoiding interfering in the internal matters of the Marcher lordships. Rather than attempt to control large swathes of the March, Henry inserted his own men such as Hubert de Burgh into the region and kept certain strategic points—Montgomery, Painscastle, Builth—in the royal hand once they fell there. The one time that Henry came close to his father’s policies in the March, with his promotion of Peter des Riveaux, he provoked rebellion. Even so, Richard Marshal’s rebellion failed to attract many adherents in the south-eastern March, largely because the lords and knights of the region were by now too attached to, dependent upon, or indebted to the king: there was little or nothing for them to gain from taking up arms against Henry. It is ironic to note that the increasing cost of holding their own against the Welsh, expressed in their new masonry fortiﬁcations and the resulting debt, ultimately brought many Marcher lords more ﬁrmly under the control of the Crown. The Treaty of Woodstock, agreed on 30 April 1247, capped an impressive increase in royal power in the March and Wales as a whole through the 1240s.¹³⁵ This achievement was to last into the ¹³² Lloyd, A History of Wales, ii. 700–2. ¹³³ CR 1242–46, 450. ¹³⁵ For the treaty, Littere Wallie, 7–8.

¹³⁴ See Ch. 6.

218

Lords of the Central Marches

1250s, until unrest, the Barons’ Wars, and the rise to power of Llywelyn ap Gruffudd reversed the situation. It would take Edward I to bring all of Wales ﬁnally under the control of the English Crown, but he would draw liberally on the precedent and example of his father and grandfather.

8 Conclusion This book has examined the nature of aristocratic power along a section of the border between England and Wales during the High Middle Ages. Its focus has been the barons and their men who expanded from Herefordshire into Wales and created the Marcher lordships of Brecon, Wigmore, Monmouth, and others. The special character of the region, albeit within a polity ultimately dominated by the English king, was largely the result of the way these barons organized, administered, and exploited their lordships—militarily, judicially, and economically—how they advanced family interests through marriage, and how they related not merely to the Crown but also to each other and their knightly tenants. It was the militarized nature of this border region, with its emphasis on castleries, castle-guard, and continuing service in the ﬁeld for land held, which more than anything gave the area and its aristocracy their special character. All of this, in the Marcher lordships, meant the continued primacy of the feudal honour court. The frontier character of the region, where two peoples met, where terrain varied from lowland valleys to upland hills and moorland, and where English-type manorial institutions coexisted alongside Welsh customs and law, gave further distinctiveness to this area. Many of the characteristics of this frontier society were already in evidence by the time of Henry I. When viewed in the context of both the long-running struggle between English and Welsh, and the place of Wales within the agenda of the English kings, the shape that border society took is not surprising. The need to pacify England and to defend his duchy of Normandy meant that the Conqueror had more pressing concerns than his kingdom’s borders with Wales. William I’s focus elsewhere, however, did not mean that he granted sweeping powers to subordinates along the Welsh border such as William ﬁtz Osbern; to suppose a grant of ‘palatine’ power in Herefordshire is in any case to make use of an anachronism. Nor did the Normans assume Welsh princely powers as they moved into Wales, excepting the control they

220

Lords of the Central Marches

had over Welsh tenants. True, in the Welsh commotes and cantrefs the Normans found convenient units of political organization, but these units gave lords no special powers over their own English tenants, nor did they furnish the best basis for asserting sweeping claims against the Crown. Instead, the Norman lords who expanded from Herefordshire into Wales, encouraged but not closely supervised by the king, set up lordships resembling those to which they were accustomed in Normandy. In the duchy by the 1030s at the latest the lands of the magnates were concentrated in blocks, not interspersed among those of other magnates. In post-Conquest England, the assignment of Anglo-Saxon lands by reference to antecessores had prevented such compact blocks of land from being accumulated in most places. In Wales, however, land was not assigned by reference to antecessores as it had been on the conquest of England; Bernard de Neufmarch´e, when he conquered Brycheiniog, had the claim of sole superioritas over a large territory (although it was not much more than that outside the major river valleys). Indeed, it is interesting to note that, during the breakdown of royal authority in England during Stephen’s reign, magnates in England attempted through coercion and violence to consolidate and rationalize their holdings—‘to tidy up the socio-political map so that it resembled more closely the convenient pattern they knew in Normandy’.¹ There was never any need for the barons of the March to do this. So one reason for the eventual emergence of a Marcher society that was quite different from aristocratic society in England was the relative lack of attention shown Welsh affairs by English kings. They had of course good reason. For much of the period we have examined, the fact that the kings of England also had extensive Continental lands meant that Wales was never a ﬁrst-rank priority for them, at least not for the extended period of time necessary to effect any permanent settlement of Anglo-Welsh affairs. William Rufus might well have conquered Wales; yet his focus on wresting Normandy from his brother Robert kept him busy elsewhere during the ﬁrst years of his reign. In 1093–4, critical years which saw a widespread Welsh rising against the inroads of the Normans in Pembroke, Glamorgan, and Brecon, William was occupied ﬁrst in Scotland and then once again in Normandy. When he ﬁnally turned his attention to Wales in 1095, he was stymied by Welsh tactics, the terrain, and the leadership of Gruffudd ap Cynan. Although another ¹ Crouch, The Reign of King Stephen, 148–53, quotation at p. 150.

Conclusion

221

expedition in 1097, this time to south Wales, secured Pembroke, Rufus’s focus for the remainder of his reign was directed once again towards the Continent, especially the conquest of Maine. Henry I likewise had the ability and perhaps the opportunities to deal with Wales conclusively. But once again, the reality of a divided Anglo-Norman realm, and the desire to wrest Normandy from Robert Curthose, meant that Henry’s attention was very much directed elsewhere from 1100 to 1106. The necessity thereafter to fend off attempts by Robert’s son William Clito, in league with the king of France, to pry the duchy from Henry’s grasp continued to divert the king from purely British concerns. Henry instead was content to practise balance-ofpower politics in Wales, pitting Welsh princes against each other, while backing up his authority with the occasional (if formidable) expedition, such as that of 1115 into north Wales. Stephen’s reign, disastrous as it was for royal power in the west, has been treated at some length above; sufﬁce it to say that the Marchers truly were thrown back on their own devices in the later 1130s and in the 1140s.² Henry II’s reign is in many ways reminiscent of that of his grandfather Henry I. Though he made his power felt with the occasional royal expedition, Continental concerns remained paramount, and this accounts for the modus vivendi Henry reached with the Lord Rhys for much of the reign. The result of this intermittent attention of the English kings towards Wales was that the English barons along the border and in conquered Wales often had to fend for themselves. In this situation they fought their own tangled little wars, organized their affairs and lordships so as to at least preserve what they had, and undoubtedly came to have a viewpoint that could eventually be described as ‘Marcher’. Yet, although the king’s attention to Wales was limited during much of the period we have examined, we must remember that he was not without inﬂuence and power in the border region. Henry II disciplined and chastised the Marchers, whether they were in Wales or Ireland. Richard I’s reign, on the other hand, seems to be of signiﬁcance chieﬂy for the freedom of action against the Welsh which the Marcher lords, particularly William de Braose and Roger de Mortimer, regained. But it was the reign of King John that was pivotal, as his seizure of the extensive Braose lands in the region from 1208 to 1215, as well as those of the Lacys from 1208 to 1214, followed by his imposition of taxation on these Marcher lordships, further heightened the sense of a separate ² See Ch. 2.

222

Lords of the Central Marches

Marcher identity which was likely forming at the end of the twelfth century. And it was not just that John had brought a conglomeration of Marcher lordships under the control of the Crown, for the evidence seems to show that he had extended to them the full weight of royal government, judicial and ﬁscal. The unpopularity of these methods undoubtedly helped to fuel the rebellion of Braose tenants in 1215. By the use of forfeiture, he also, perhaps more importantly, destroyed the Marcher world of c.1200 at a stroke. The March of 1216 was in some ways very different from that of 1200; if the Marchers were never likely to be completely reined in by the king, it was not until the time of Roger Mortimer in the fourteenth century that the March again saw a ﬁgure as mighty as William de Braose, able perhaps to threaten the king himself. In the end King John, however, failed to control the March and the large portions of pura Wallia which he had conquered in 1211. His son Henry III came to exercise a large degree of inﬂuence in the region because of the necessity for royal military force to oppose Llywelyn and by his shrewd manipulation of aristocratic debt, but there would be no more attempts to take actual control of a large swathe of the March—royal authority and power had to be more circumspect. After all, it was in no small measure the Marchers who had saved Henry’s throne. Thus, by the beginning of the thirteenth century the barons of the March, at least in the eyes of outside observers, were assuming a common identity in a manner similar to that of the Holt’s ‘Northerners’. The second factor in the formation of a distinctly ‘Marcher’ society was the increasing power and organization of the Welsh. Already in the ﬁrst half of the twelfth century Welsh rulers could raise armies of several thousand foot. To do this they drew on a general obligation, recognized of all freemen, to perform military service. These Welsh armies were in turn formed around the nucleus provided by the teulu, or household unit of the Welsh prince. The teulu thus was analogous to the English king’s household knights. The Welsh made good use of the favourable terrain of their homeland, using mobile, lightly armed bowmen and spearmen and often employing ambush tactics before fading back into the forests and mountain fastnesses. Yet in the course of the twelfth century the Welsh also successfully adopted features of Anglo-Norman military practice: cavalry, armour, and castles. In the words of one historian, the Welsh ‘militarily . . . had the best of both worlds’.³ When Welsh military capabilities combined with a decline in internecine strife and ³ Carpenter, The Struggle for Mastery, 114.

Conclusion

223

the rise of strong leaders such as the Lord Rhys or Llywelyn the Great, the Marchers were faced with an ultimately unbreakable stalemate. Finally, and most importantly, Marcher society was formed in good part in reaction to the growth and potential intrusiveness of Angevin government. The inattentiveness of the English kings, and the increasingly dangerous nature of Welsh military capabilities, meant that Marcher society was often an embattled one. A crucial response to this situation was the preservation of the feudal honour. When that institution was threatened, the Marchers took steps, both through the strident assertion of near-regal claims against the Crown, and through the consolidation of lordship, to preserve it. The period from the mid-twelfth to the mid-thirteenth century witnessed the increasing expansion and elaboration of central government, particularly royal justice and ius scriptum, which acted to incorporate border counties such as Herefordshire ever further into the English kingdom. The Angevin legal reforms hardened customary property rights into law, and reduced the effectiveness of honour courts in Herefordshire. A legal boundary which denoted the respective jurisdictions of king and Marcher lords emerged. Given the importance and centrality of the honour court to a frontier lordship, it is no coincidence that defences of Marcher ‘liberties’ begin to emerge in this period. These ‘liberties’ were being deﬁned in relation to what appeared to be an increasingly powerful and intrusive Angevin royal government. Furthermore, the assertion of Marcher liberties was part of the consolidation of lordship which many Marcher lords attempted in the course of the thirteenth century. This process included the assertion of sole jurisdiction over tenants in the forum of the honour court; the enforcement, backed by threat of forfeiture, of the performance of military obligations; and at times the engrossment, through purchase or seizure, of the lord’s demesne lands. Accompanying the assertion of Marcher liberties and attempts at the consolidation of lordship was the decline of cross-border multiple knightly tenures. It is hard not to suspect the guiding hand of lords in the patterns of inheritance and marriage at the knightly level of society, for this process in itself consolidated lordship by lessening the chances of royal interference in the internal workings of the honour. In the ﬁrst half of the twelfth century, knightly estates spanned the border between Herefordshire and Wales, and often comprised manors held of more than one lord. In the second half of the century, inheritance patterns served to split these estates along tenurial lines, with Herefordshire holdings and Marcher holdings being increasingly separated and ties of

224

Lords of the Central Marches

multiple lordship decreasing markedly. This process was then reinforced by a level of intermarriage which, to judge by surviving evidence, was much more endogamous than in England, including the north. Even as knightly families such as the Baskervilles and the Devereux divided their holdings between Herefordshire and Marcher branches, the difference between England and March was further deﬁned from the centre. Processes at both the periphery and the centre thus drove the separation of the March from England. In the twelfth century there were still connections between the lordships of the eastern March and the border counties, particularly Herefordshire, and this was apparent in the cross-border holdings of the major knightly families. But by the reign of Henry II at the latest, as many of these cross-border holdings began to split, a separate Marcher identity was beginning to form. Although our only direct evidence of this identity is the references to marchiones that begin to appear in the early thirteenth-century documentation, which refer largely to Marchers at the level of the baronage, it is not unreasonable to assume that this identity had also begun to form at the level of the knights. William de Braose, for example, although the preponderance of his lands were in Wales, also was a lord in Normandy, England, and Ireland. His tenants did not move on such a wide scale, and the pressures of life along a militarized frontier almost certainly served to concentrate their attentions on the locality. By 1242 two of the most important families in our section of the March, the Braoses and the Lacys, had been removed from the scene, their male lines ended. But the dying out of the Braoses and Lacys did nothing to affect the way aristocratic society was organized in their lordships, for the king did not absorb those lordships into the English kingdom—it was much too late for that. Instead, the Braose and Lacy lordships were plums of royal patronage, passed to new men by marriage to the surviving female heirs. And although the power of the lords of Brecon waned in the thirteenth century after the death of the last Braose, that was the result of a series of wardships and absentee lords, not any fundamental change in the mechanisms of power and control available to the lords of Brecon. Three baronial families which we have studied remained: the Cliffords, the Monmouths, and the Mortimers. Of those three, however, only one was of the ﬁrst rank; while the Cliffords and Monmouths, always second-rank baronial families to begin with, would fade in

Conclusion

225

signiﬁcance in the second half of the thirteenth century, the Mortimers would come to dominate the region for the rest of the century and beyond. How they accomplished that growth in power in their lordship of Wigmore during the 1250s and 1260s makes for an instructive contrast with the situation in the former Braose lordship of Brecon after the death of William V de Braose in 1230. The Braose patrimony was divided among William’s four daughters: Builth went with Isabel de Braose to her husband, Dafydd, son of Llywelyn the Great; Abergavenny went with Eva to her husband, William de Cantilupe; Radnor with Maud to Roger de Mortimer; and Brecon, the greatest of Braose lordships, to Humphrey de Bohun by right of his wife, Eleanor de Braose. The Bohuns were earls of Hereford, though that title had never brought them a strong landed position in the county. Humphrey was a minor in 1230, and the lordship was held in ward by a succession of men. This undoubtedly allowed the major tenants of the lordship, such as the Pichards and Baskervilles, to begin to usurp the privileges that would enable them to pose as ‘lords’ in their own right later in the century. It should be noted, however, that, by Humphrey’s majority, the bonds of lordship were still able to be activated: a perusal of escheats from 1265 and 1266, after the battle of Evesham, shows that certain Bohun tenants followed the younger Humphrey in his allegiance to Simon de Montfort.⁴ The salient point about Brecon, however, is that the mechanisms and structure of lordly power were ﬁrmly in place, to be put to use again when the Bohuns applied themselves.⁵ It fell to the Marcher lords to ensure that their power did not dissipate. By the 1240s the great survivors from among our families were the Mortimers. In contrast to Brecon during Humphrey de Bohun’s minority, Roger de Mortimer made use of the power inherent in the position of Marcher lord to embark on a campaign of territorial consolidation during the 1250s, 1260s, and beyond similar to that also waged by the Clare lords of Glamorgan in the thirteenth century. As touched upon earlier, Mortimer preyed upon those of his tenants and other area gentry who were in debt, and those affected included William Devereux, Henry de Pembridge, and the Ballon family.⁶ Devereux ⁴ Close Rolls of the Reign of Henry III 1264–68, 127 (Roger Pichard); Mortimer Cartulary, fo. 37 (Walter Devereux, John Pichard, Hugh del Fresne). ⁵ See Ch. 3. ⁶ Waugh, The Lordship of England, 31–2 and n. 53 (speaking of purchase and acquisition strategies), notes that ‘The Mortimers, for example, obtained the manor of Much Marcle, head of the barony of Much Marcle in Herefordshire, from the Ballon

226

Lords of the Central Marches

in return for cash granted Mortimer all his lands in Stoke Lacy and Lawton, while Henry de Pembridge, who like Devereux was in debt to the Jews, quitclaimed all his manor of Pembridge to Mortimer in exchange for £1,000 (which may well have been a forgiveness of debt rather than cash).⁷ Mortimer was able to pursue this aggressive familial aggrandizement, which normally would have been risky within most honours, because alternative foci of lordship in the region were weak, because he continually asserted expansive Marcher liberties, and, most importantly, because his close connection to Crown and court enabled him to get away with it. Mortimer’s actions therefore may well have contributed to a number of local knights becoming involved on the Montfortian side in the civil war of 1265. William Devereux died with Montfort at Evesham; Walter Devereux, William Torel, Hugh del Fresne, and John Pichard all saw their lands escheat to Roger de Mortimer’s wife, Maud, in 1266.⁸ Mortimer’s actions, if extreme and in many ways only possible in the March, were not out of tenor with magnate–tenant relations in the latter part of Henry III’s reign, a tenor that contributed greatly to the support that the knightly class gave Simon de Montfort. The great magnates, especially those close to the king, were able to access royal favour and thus were often able to exercise a free hand against their tenants; at the same time it became increasingly difﬁcult for the knights to gain access to royal justice. Complaints about lords and their ofﬁcials rose throughout the 1240s and 1250s.⁹ At the same time, the king’s need to raise money brought increasing ﬁnancial pressure upon the knights, especially those who had Jewish debts. Henry’s government particularly squeezed the Jews for money, and in order to raise ready cash they in turn often resorted to selling their bonds cheaply. These were purchased by lords, who then in turn often gained land pledged as security for the debt.¹⁰ In essence, the political climate of the period allowed a Marcher lord such as Roger de Mortimer to use the same tactics he family after a long, arduous campaign of economic, legal, and physical harassment that began after the Barons’ War.’ For Devereux, see Ch. 4. ⁷ Devereux, BL, Harleian MS 1240, fo. 48. For Pembridge’s debt, Close Rolls of the Reign of Henry III 1254–56, 170 (a 1255 distraint); for the quitclaim, see the Mortimer Cartulary, fo. 95. ⁸ For William Devereux’s death, see Close Rolls of the Reign of Henry III 1264–68, 76. For escheats, BL, Harleian MS 1240, fo. 37. ⁹ For a summary, see Carpenter, The Struggle for Mastery, 349–53. ¹⁰ See J. R. Maddicott, Simon de Montfort (Cambridge, 1994), 126.

Conclusion

227

might use in his Marcher lordship across the border in Herefordshire. Mortimer even took things a step further, for he began to withdraw portions of Herefordshire and to declare them to be in the March. This process culminated in the 1260s when Henry granted to Roger a series of liberties and immunities.¹¹ In the situation of Roger de Mortimer, what had occurred by the 1260s was a swing in the balance of power in the region towards lords and away from knights, occasioned by the use of the tools of consolidation that were always potentially available to a magnate. Personal lordship was paramount, in one sense in the way a lord could attract the loyalty of his tenants, but also, as Roger de Mortimer demonstrated, through the attention, determination, and sheer bloody-mindedness which a magnate brought to the task of gathering and preserving power within his lordships. In the realm of Anglo-Welsh relations, the king increasingly took the lead militarily, a result of both the loss of most of his Continental lands and the inability of the Marcher lords to make headway against a revitalized and much stronger Welsh polity led by the princes of Gwynedd. But the example of Mortimer shows how the king could give a lord an additional edge in the internal politics of his lordship as well as a free hand to amass land (and indeed withdraw it into the March). The lack of a route of appeal for tenants and the absence of royal bureaucracy meant that the initiative in fashioning the political and social balance within a Marcher lordship belonged to its lord. The nature of the March by the 1260s was such that the king was in many ways supportive of the power of Marcher lords, or at least unwilling or unable to meddle with that power. Evesham in 1265 cemented this relationship between Marchers, such as Roger de Mortimer and Walter de Clifford, and the king (or more accurately in this case, the Lord Edward). This contrasts with the nature of power in the English colonial society in Ireland in the early thirteenth century.¹² The speed with which Henry II intervened in Ireland in 1171, following the success of Strongbow, is a stark indicator that the English kings had learned from history and had no intention of letting further baronial private enterprise establish areas where it was hard for them to intervene. Whereas, in the March, the opportunity and tools necessary for a consolidation of lordship were there to be ¹¹ Davies, The Age of Conquest, 276. ¹² For the next two paragraphs, see Holden, ‘Feudal Frontiers?’, and the notes contained therein.

228

Lords of the Central Marches

used, in Ireland the decisive intervention of royal, central government from Westminster and Dublin, although only partially effective, was enough to deny to Anglo-Irish lords the same tools that their brethren in the March of Wales enjoyed. In the thirteenth-century March, the honour court was paramount and virtually unchallenged, and although elements of English law such as novel disseisin were imported into the March, appeals to royal justices were barred. In Ireland mort d’ancestor and novel disseisin were in use as early as 1199, when central records become available to the historian. Furthermore, Irish tenants often appealed cases out of the honour, to either Dublin or the king himself in England; the volume of such cases was such that appeals directly to the king had to be limited. By the end of the thirteenth century the writs of error and certiorari ran, giving Irish tenants recourse against judgments by their lords. Though the Irish legal system did not have the weight that its cousin had in England, it was just effective enough to prevent most attempts at consolidation of lordship. At the same time, the king’s prerogative lordship operated in Ireland; in contrast, the Praerogativa Regis, from early in Edward I’s reign, formalized the immunity of Marcher lordships in Wales from the king’s power to exercise control over heirs. In Ireland the proliferation of multiple tenancies on the part of subtenants, many of whom also held land from the Crown, gave the king plentiful opportunities to interfere in the workings of Irish honours. Competition by lords for knightly tenants to settle their Irish lands was ﬁerce; Hugh de Lacy, for instance, went as far as to poach tenants from the lordship of Leinster during the minority that followed Richard Strongbow’s death in 1176. Many Lacy tenants in Meath also held land of the de Verdons in Louth. The king also created ties by granting land, for example in north Leinster, where King John enfeoffed many Lacy tenants on royal demesne land. This served not just to increase multiple tenure but also to make these men tenants-in-chief, exposing them to the workings of royal prerogative lordship. In the March, as we have seen, the trend was towards fewer multiple tenancies, especially those that spanned the border between Wales and England.¹³ In Ireland multiple tenancies, and a hampered ability to police service obligations within the honour, conspired to make the exercise of lordship there different from that in the March. This inability to preserve the feudal honour in Ireland as an effective mechanism of social control and military power resulted in the ¹³ See Ch. 4.

Conclusion

229

eventual shift to dependence on a lineage structure, utilizing extended family networks, rather than the honour, for military manpower. The development of these informal structures would not have been necessary had the Irish honour had anything like the effectiveness of those lordships in the Welsh March. The basic outlines of aristocratic society in the region had been set long before 1166. But it was the drawing inward, the separation of the lordships of the Middle and south-eastern March from Herefordshire that we have examined above, which served to sharpen the distinction between the March on one hand and the border counties on the other, both in feudal tenure and probably in mentalit´e as well. The Marcher lords began to emerge as a distinct aristocratic group as their ambitions began to conﬂict with Angevin policies, for instance in the reign of Henry II. The outlook of these lords would have inﬂuenced that of their knightly tenants, who themselves had nothing resembling the potential stage of a William de Braose on which to act out their ambitions. The emergence of these marchiones, ﬁrst mentioned as an identiﬁable group in the chronicles and documents of John’s reign, as opposed to the mere concept of the March itself, marks the point at which the relationship between the western border of the English kingdom and the adjacent Marcher lordships had hardened and formalized. Interference in the region by King John then only served to heighten this sense of distinctiveness. The parts played by Walter de Clifford, John de Monmouth, and Walter de Lacy in support of John during the civil war of 1215–17, and by Roger de Mortimer and Walter de Clifford against the Montfortians in 1265, underlined the signiﬁcance of these men, with both their military retinues and the ability to recruit from among the Welsh, to the Crown. Although Edward I’s conquest of Wales in many ways diminished the political signiﬁcance of the Marchers (excepting Roger Mortimer), their privileges remained an anomaly frozen in time, a testament to the way the English along the frontier with Wales practised an adaptable conservatism, preserving the feudal honour and holding at bay royal encroachments on their rights while selectively adapting elements of change such as new writs in order to buttress their own honour courts and by extension their political and social control.

APPENDIX 1

Manors of the Magnate Families by County

Family and county

Honour

Manor

Domesday value

Tonga

TRE £11, modo £6, 3 hides, 1 league woodland

Bodenham

Probably the 1 1/2 hides held by Osbern ﬁtz Richard, passing to Brecon through the marriage of Osbern’s daughter to Bernard de Neufmarché; TRE 60s., modo 48s.

Tenant

Fortiﬁcation

BRAOSE Shropshire Herefordshire

Brecon

Motte 1211–12: Walter Devereux ( 1/2 fee) 1242: Walter Devereux, 1 1/2 hides, no service amount speciﬁed, 1/2 hide subenfeoffed to Walter de Mucegros

Bredwardine

TRE waste, modo 60s., 5 hides

1242: Roger Devereux and participes Brinsop

TRE £8, modo £7, 5 hides

1166: William Torel (also held Pencombe of Ewyas Harold) 1211–12: Ralph Torel 1242: Ralph Torel (same or son), now in honour of Blaenlyffni, listed as 2 hides

Burghill

No value given; 8 geldable hides, mill 20s., 25 sticks eels, 4s. from woods

1166: William de Miners 1211–12: Henry de Miners 1242: Elisabeth and Basilia de Gamages, 2 hides, 1 fee

Sited on narrow ridge by Wye; long enclosure with foundations of oblong building (hall?) at one end. First mentioned 1199 (Rot. Cur. Reg. i. 426); as an ‘old’ castle in 1227 (CCR i. 58–9)

List of Manors of the Magnate Families

1166: Ralph I de Baskerville 1211–12: not listed in returns, held in wardship until heirs of Ralph III de Baskerville reach majority

231

a

Demesne manor.

232

Family and county

Honour

Manor

Domesday value

Burweltun 30s., 1 hide, held by Roger de Lacy; from list of tenants c.1139 (HDB 79), held of Lacy by Miles of Gloucester, later attached to Brecon

Humber

3 1/2 hides, no valuation

Kingstone Much Cowarne Sarnesford Roger Tedstone Wafer Tullington

1242: Miles Pichard, 2 hides, 1/2 fee 1211–12: heirs of William de Waldeboef, responsible for 3 fees 1242: John de Waldeboef, 1/2 fee in Gattertop, 1/2 fee in Upton 1242: Matthew de Mans, 1/2 fee 1242: William de Furches

1211–12: Pain de Burghill, 1 fee (?)

Fortiﬁcation

Appendix 1

Gattertop (with Upton)

Tenant

Kington

Avenbury Felton and Thinghill Little Cowarne Moccasa Rowden (in Edvin Ralph) Rushook Sutton and Maund

2 weak baileys, tiny mound

List of Manors of the Magnate Families

Weston Bret

1242: heir of Pain de Burghill (probably Roger de Burghill), 2 hides, 1/2 fee, now in honour of Blaenllyfni 1242: Matthew le Bret, 2 hides held of Roger Pichard

233

Gloucestershire

Manor Venn (Fenna in royal manor of Marden) Hampnett (assigned as maritagium for Annora de Braose on marriage to Hugh de Mortimer) Tetbury

Clive

Domesday value

Tenant

TRE £8, modo £6, 10 hides (originally held by Roger d’Ivry with Tetbury). 10 burgesses in Winchcombe paid 65d . (included above) TRE £33 with Tetbury Upton, which had two hides and 1 virgate; Tetbury proper had 23 hides; modo, at farm for £50. Mill, 15d ., pasture 10s., 10 acres meadow

Clifford

Fortiﬁcation

Appendix 1

CLIFFORD Shropshire

Honour

234

Family and county

Bridge Sollers (bishop of Hereford tenant) Buchstanestona (Buckton?) Clifforda

No valuation, 1 hide

TRE 100s., modo 110s., 5 hides

List of Manors of the Magnate Families

Herefordshire

Corfham Culmington and Siefton Diddlebury (Dodelebur) Hay of Ernestou Norton (with Tonge) Bradefeud (Bradford or Broadﬁeld?)

1242: Walter de Rofhey, of Geoffrey de Cruece, of Roger de Clifford, of Walter de Clifford, 1/4 fee 1166: H. de Sollers

No valuation, 3 hides Originally motteand-bailey; stone castle later erected, probably 1220s

235

236

Family and county

Honour

Manor Dinedor (and Douweswel) Dorstonea

Hanlega (Hanley’s End?) Homme (Holme Lacy?)

Tenant

Fortiﬁcation

No value given; 7 geldable hides

1166: R. de Sollers

Three different sites: two motteand-bailey, one motte only

No valuation, 1/2 hide and 1 virgate TRE 20s., modo 20s. 40d ., 1 hide

1242: William de Ford, 1/2 fee 1166: ‘Peter’ 1242: Hugh de Hamenes, 1/5 fee

1242: William de Homme, 1/2 fee (veteri feffamento)

Appendix 1

Ford (with Sarnesﬁeld) Hamnish Clifford (Hamenasse)

Domesday value

Hope-underDinmore (Hampton, Hanton)

No valuation, 1/2 hide

1166: Adam de Mappenore 1242: Robert de Mappenore, 1/2 fee

List of Manors of the Magnate Families

Gloucestershire

Mathon Rochford (near Tenbury) Sarnesﬁeld (part) Staunton-onWye (Richard’s Castle tenant) Stoke Edith Ullingswick (bishop of Hereford tenant) Dymocka (royal demesne) Easton

20s., 1 hide

1242: William de Ford, fee with Ford 1242: Matthew de Mans

1/2

3 hides, 1 fee (no mention of manor being held of Richard’s Castle)

Strong motte-andbailey

237

238

Family and county

Honour

Manor

Domesday value

Tenant

Fortiﬁcation

Frampton LACY Shropshire

Masonry castle

Small, weak motteand-bailey

Appendix 1

Butterly Cleobury North Corftona Euledon Hayton (Upper and Lower) Henley Hopton Cangeforda Hopton Wafers Longestantun

Ludlowa Middlehope Onibury Rushbury Staunton Lacya

Birley (initially Lacy, later Mortimer)

Large motteand-bailey

1242: Roger Pichard, 4 hides, 1 fee

List of Manors of the Magnate Families

Herefordshire

Stokesay Sutton Wheathill Wichecote Woodeton Almeley

Masonry

239

240

Family and county

Honour

Manor Bishop’s Frome (?) Bodenham

Brobury

Brockmanton (Brockhamptonby-Ross?) (no longer Lacy 1243)

Tenant

TRE 50s.; modo 60s. for 1 1/2 geldable hides held by Roger de Lacy (another 1 1/2 hides held by Osbern f. Richard)

1242: William de Furches, 1 hide for 1/4 fee, Isabel de Furches holds in dower of him 1242: Roger de Bodenham, mediate lord Walter de Baskerville, unspeciﬁed amount of service 1242: Walter of Brobury, 1/5 fee, mediate lord Walter de Baskerville

Fortiﬁcation

Appendix 1

Bodenham Rogeri

Domesday value

Brydenbury (?) Butterly (?) Byford

Not gelded

1166: Robert de Baskerville 1211–12: Walter I de Baskerville (d. 1212) 1242: Walter II de Baskerville

Motte-and-bailey Motte-and-bailey, with two outer enclosures enveloping the motte. Water defences. Mention, as a domus defensabilis, in DB i. 184b, 2. Mentioned as a castle in 1183 (Great Rolls of the Pipe, 29 Henry II, 111)

List of Manors of the Magnate Families

Canon’s Frome Castle Fromea Eardisley

1242: John de Kenell’, Matilda de Tregoz, Walter of Bodenham, 1 fee

Evesbatch

241

242

Family and county

Honour

Manor

Domesday value

Ewias Lacya (Longtown, lordship of Ewyas Lacy)

20s.; original grant 4 carucates waste (vastas)

Heath Holme Lacya

Tall motte with very large rectangular bailey; reconstructed in masonry, 1220s

30s., 1 hide

TRE 20s., modo 40s. TRE 40s., modo 30s., 1/2 hide

Fortiﬁcation

1166: Gilbert of Bakinton 1166: Robert le Wafer

Appendix 1

Fernhilla Gattertop Grendon (Court or Bishop?) Hamond’s Frome Hampton Court Hampton Wafer

Tenant

Humber (no longer Lacy 1243) King’s Pyona

Leadon

Ledicota Leominster Letton

1166: Richard de Escotot, 5 hides; William de Braose, 1/2 hide 1242: Robert of Auferton’ and Adam ﬁtz Maelis, mediate lord Richard of Hampton, 1 hide, 1/5 fee

Motte

Very large ﬂat motte with two baileys side-by-side, outer works

1242: Richard of Horcesley, mediate lord Walter de Baskerville, 3 hides, 1 fee

List of Manors of the Magnate Families

Lawton (in Kingsland)

£4

243

244

Family and county

Honour

Manor Little Marcle Lyde Mucegros

Lyonshalla

Mansell Gamage

Tenant

1211–12 tenant: Walter de Mucegros (?) 1242: Amabilia and Juliana de Mucegros, 1/2 hide, 1/10 fee; Richard de Feipo ibidem 1242: Robert Malhore, 1 hide, socage 1210–11: Stephen Devereux 1242: William Devereux, 1 fee with Upcot

1211–12: William de Gamages 1242: Godfrey de Gamages, 1 fee

Fortiﬁcation

Mound, outworks, keep; rebuilt to the circular plan of Longtown by Stephen Devereux c.1227 (Hillaby, ‘Hereford Gold’, pt II, 226)

Appendix 1

Lyde Saucey

Domesday value

Mansell Lacya

Maund Albin (Whitechurch?) Maund Brian (no longer Lacy 1243) Munsley Muntrath Ocle Pichard Pudleston Putley

1242: William Pichard, mediate lord Miles Pichard, 2 hides, 1/2 fee

Motte with wet moat

List of Manors of the Magnate Families

Marston (Broxash hd) Marston (Stretford hd)

1242: Katherine de Lacy, 3 hides, 1 fee, listed as Radnor fee (sub for Kington)

1242: Cecily Devereux, 1 hide, service amount unspeciﬁed

245

246

Family and county

Honour

Manor Ross Sarnesﬁeld

Swanston Wakinton (Wacton?)

Tenant

TRE waste, modo 20s., 1 1/2 hides

1166: Philip de Sarnesﬁeld (Geoffrey, marginal annotation, HDB) 1242: Philip of Sarnesﬁeld, 1/2 fee

TRE 40s.; modo 50s.; mill worth 32d .

Fortiﬁcation

Appendix 1

Stanford Bishop Stanton-onWye Stoke Lacy Street (Lestret in Kingsland) Stretton Sugwas

Domesday value

TRE 100s.; post 60s.; modo 100s.

Weston Beggarda

100s.; mill, 10s.

Wolferlow Woonton Wormsley Yarkhill Yarsop Bulley Carlswall Coates Randulf Eastleach Turville Kempley

1166: Hugh I de Lacy 1210–11: Walter de Lacy 1166: demesne 1242: Robert of Weston, 1 hide

Masonry

List of Manors of the Magnate Families

Gloucestershire

Weobleya

247

248

Family and county

Honour

Manor

MONMOUTH Herefordshire

Asperton and Stretton Hope Mansell Mainstone Wolsopthorne, in Ashperton (Radlow hd)

Tenant

Fortiﬁcation

Appendix 1

Longhope Oxenhall Siddington Stratton Wikeresindon Wormington

Domesday value

MORTIMER Shropshire

Wigmore

Burwarton Catsley Cleobury Mortimer Coreley Eudon George Hugleye Kenley La Lowe (The Low [in Wem]?)

TRE £6 10s., modo 100s., 5 hides

List of Manors of the Magnate Families

Whitwick Chesterton (?) Hope (Longhope?) Tibberton

Gloucestershire

249

250

Family and county

Wigmore

Manor Stepple Neen Savage Neenton Overton Rudge Sheinton Shipley Sudbury Aymestrey Birley Burton Covenhope

Domesday value

Tenant

Fortiﬁcation

Appendix 1

Herefordshire

Honour

Large motte-andbailey on ridge, with motte uphill. Walled with shellkeep, small barbican, many outworks, masonry of 13th and 14th cc.

List of Manors of the Magnate Families

Easton Hinton Horleson’ Kingsland Kinnardesley Leye Orleton Shobdon Staunton Wigmorea

251

252

Family and county

Manor

Radnor

Mansell Lacy (after partition) Pembridge Sarnesford Roger Stoke Blees Bisley Longborough Newenton (Newington Bagpath?)

Domesday value

Tenant

Fortiﬁcation Very low motte with wet moat Square-shaped motte on marshy site

Appendix 1

Gloucestershire

Honour

APPENDIX 2

Genealogies B A S K E RV I L L E Ralph I = dau. of Drew fitz Pons (d. 1148/9) of Clifford (?)

Ralph II [Brecon] (d. 1190/1)

Ralph III (d. 1210)

Thomas

Robert Matilda = Isabel Roger Bret

Robert [ Weobley] (d. 1176)

Wiberga

Walter [Orcop] (d.s.p.)

Ralph (d. c. June 1186)

Walter = Isolda Pantolf (d. 1212)

Agnes = (1) Peter of Aldermaston (2) Roger Devereux

Alice = Robert le Wafer

Sybil = Hugh of Radnor

Walter II = Susanna (d. 1244) de Chanceaux

Sybil = Walter III (d. 1282)

Richard

254

B R AO S E

William II = Bertha, dau. of (d. 1265?) Miles of Gloucester

Matilda de St = William III (d. 1211) Valery (d. 1210)

Margaret = john Giles (d.s.p. 1232)

Giles, bp. of Hereford (d. 1215)

Philip

Reginald = (1) Graecia Brewer (d. 1228) (2) Gwladys, dau. of Llywelyn ap Iorwerth

Walter

Roger II de = Maud Mortimer [Radnor]

John = Mabel de Limesey

Margaret = Walter de Lacy

Annora = Hugh de Mortimer

William V = Eva Marshal (d. 1230)

Isabel = Dafydd ap Llywelyn [Builth]

Eleanor = Humphrey de Bohun [Brecon, Kington]

Eve = William III de Cantilupe [Abergavenny]

Appendix 2

Matilda, dau. of = William IV juvenis Richard de Clare, (d. 1210) east of Hertford

CLIFFORD

Richard fitz Pons = Maud, dau. of Walter of Gloucester

Walter I = Margaret de Tony (d. 1190)

Richard = Letitia (d. 1213) de Berkeley

William

Lucia = (1) Hugh de Say (2) Bartholomew de Mortimer

Richard

Simon

Richard

Walter III = (1) Katherine (d. 1263) de Lacy (2?) Isabelle

Roger = Sybil, (of Tenbury) dau. of Robert of Ewyas

Giles

Rosamund

Genealogies

Agnes = Walter II (d. 1220) de Cundy

Basilia and other daughters

(3?) Margaret, dau. of Llywelyn ap lorwerth

255

Matilda

256

D EV E R E U X

William (fl. mid-12th c.)

Walter [Lyonshall]

Roger (d.s.p.?)

Cecily

Stephen = (1) Constance de Longchamp (d. 1228) (2) Isabelle de Cantilupe

William

Nicholas

Walter

Appendix 2

John (d. c.1188)

Orenge = William of Lechlade

L AC Y

Gilbert (d. after 1163) Robert (d.s.p. before 1162)

Walter II = Margaret de (d. 1241) Braose

Walter = dau. of Theobald Butler

Hugh III earl of Ulster (d. 1242)

Gilbert

Egidia = Richard de Burgh

= (2) dau. of Rory O'Connor

Robert

William

∗Katherine = Walter III de Clifford (alive 1267)

Matilda = (1) Peter of Geneva

Genealogies

Gilbert (d. 1230) = Isabella Bigod

Hugh II = (1) Rose of Monmouth (d. 1186)

Margaret = John de Verdon

(2) Geoffrey de Geneville

∗Katherine set aside, date unknown

257

258

MONMOUTH

Baderon = Rose = (2) Hugh II de Lacy (d. 1170 × 1176 )

Bertha = Gilbert

Jacob

John (d. 1257)

Albreda

Joan

Philip

Philip

Walter

William

William

Appendix 2

Cecilia, dau. = John (d. 1248) of Walter de Waleran

William

M O RT I M E R

Ralph (d. after 1104) Hugh

Roger (d.s.p. 1153)

Hugh (d. 1227) = Annora de Braose

Roger (d. 1214) = Isabella de Ferrers

Robert (?)

Philip

Ralph = Gwladys, (d. 1246) dau. of Llywelyn ap Iorwerth, widow of Reginald de Braose

Roger (d. 1282) = Matilda de Braose

Genealogies

Hugh (predeceased father)

Hugh = Matilda de Meschin (d. 1180/1)

Joan = Walter II de Beauchamp

Hugh = Agatha de Ferrers

259

This page intentionally left blank

Bibliography M A N U S C R I P T S O U RC E S Balliol College, Oxford MS 271, Cartulary of St Guthlac, Hereford Bodleian Library, Oxford MS Rawlinson B329, Hereford Cartulary MS Tanner 342, Brecon Cartulary extracts British Library Additional Charters 20408, 20410 Harleian MS 1240, Cartulary Mortimer (Black Book of Wigmore) Harleian MS 2044 Harleian MS 3586, Cartulary Wormsley Lansdowne MS 229 Hereford Cathedral Archives, Hereford Nos 272, 482, 483, 3235, 3241 Hereford Record Ofﬁce, Hereford Hopton Collection, no. 107 The National Archives C115/74/6679, Cartulary Llanthony (13th c.) CP 25(1), Feet of Fines DL 25, Deeds, series L DL 27, Deeds, series LS DL 36, Cartae Miscellaneae E159, King’s Remembrancer Memoranda Rolls E164/20, Cartulary Godstow E210, King’s Remembrancer, Ancient Deeds, series D E315/45, Augmentations Miscellaneous Book E315/55, Cartulary Aconbury E326/B, Exchequer Augmentations Ofﬁce

262

Bibliography P R I N T E D S O U RC E S

(Place of publication is London unless otherwise stated) Primary Abbreviatio Placitorum (Record Commission, 1811). Ancient Charters, Royal and Private: Prior to A.D. 1200, ed. J. H. Round, Pipe Rolls Series, 10 (1888). Annales Cambriae, ed. J. Williams ab Ithel, Rolls Series (1860). Annales Monastici, ed. H. R. Luard, 5 vols, Rolls Series (1864–9). Brut y Tywysogyon, or, The Chronicle of the Princes: Red Book of Hergest, Peniarth MS 20 version, ed. and trans. T. Jones (Cardiff, 1941) Calendar of Ancient Correspondence concerning Wales, ed. J. G. Edwards (Cardiff, 1935). Calendar of Ancient Petitions relating to Wales, ed. W. Rees (Cardiff, 1975). Calendar of the Charter Rolls preserved in the Public Record Ofﬁce, 6 vols (1903–27). Calendar of Documents preserved in France 918–1206, ed. J. H. Round (1899). Calendar of Inquisitions Miscellaneous (1916– ). Calendar of Inquisitions Post Mortem, 22 vols (1904– ). Calendar of Liberate Rolls 1226–1272 (1916–64). Calendar of the Patent Rolls preserved in the Public Record Ofﬁce (1891– ). Cartae et Alia Munimenta de Glamorgancia, ed. G. T. Clark, 6 vols (Cardiff, 1910). ‘Cartularium Prioratus S. Johannis Evang. de Brecon’, ed. R. W. Banks, Archaeologia Cambrensis, 4th ser., 13 (1882), 275–308; 14 (1883), 18–49, 137–68, 221–36, 274–311. Catalogue of Ancient Deeds, 6 vols (1890–1915). The Chancellor’s Roll for the Eighth Year of the Reign of King Richard the First: Michaelmas 1196, ed. Doris M. Stenton, Pipe Rolls Series 45 (1930). ‘Charters of the Earldom of Hereford, 1095–1201’, ed. D. G. Walker, Camden Miscellany, 2 (1964), 1–75. ‘ ‘‘Chronica de Wallia’’ and Other Documents from Exeter Cathedral Library MS. 3514’, ed. T. Jones, BBCS 12/1–3 (Nov. 1946), 27–44. The Chronicle of John of Worcester, iii, ed. and trans. P. McGurk (Oxford, 1998). The Chronicle of Richard of Devizes of the Time of King Richard the First, ed. and trans. J. T. Appleby (1963). Chronicles of the Reigns of Stephen, Henry II and Richard I, ed. R. Howlett, 4 vols, Rolls Series (1884–9). Close Rolls of the Reign of Henry III preserved in the Public Record Ofﬁce, AD 1227 [–1272] (1902–38, 1975). Curia Regis Rolls preserved in the Public Record Ofﬁce, ed. C. T. Flower, 17 vols (1922–72).

Bibliography

263

Dickinson, J. C., and Ricketts, P. T., ‘The Anglo-Norman Chronicle of Wigmore Abbey’, Trans. Woolhope, 39 (1969), 413–46. Earldom of Gloucester Charters: The Charters and Scribes of the Earls and Countesses of Gloucester to A.D. 1217, ed. R. B. Patterson (Oxford, 1973). Episcopal Acts and Cognate Documents relating to Welsh Dioceses 1066–1272, ed. J. Conway Davies (Cardiff, 1948–53). Excerpta e Rotulis Finium, i: 1216–1246 (Record Commission, 1835). Feet of Fines of the Reign of Henry II and . . . Richard I, Pipe Rolls Series, 17 (1894). Feet of Fines of the Seventh and Eighth Years of the Reign of Richard I: A.D. 1196 to A.D. 1197, Pipe Rolls Series, 20 (1896). Feet of Fines for the County of Norfolk, 1201–1215, and for the County of Suffolk, 1199–1214, ed. Barbara Dodwell, Pipe Rolls Series, new ser., 32 (1958). Feet of Fines 10 Richard I: also a Roll of the King’s Court Richard I, Pipe Rolls Series, 24 (1900). Foedera, Conventiones, Litterae et cuiusunque generis acta publica, ed. T. Rymer, 3rd edn, 10 vols (The Hague, 1739–45); rev. A. Clarke, F. Holbrooke, and J. Coley, 4 vols in 7 pts (Record Commission, 1816–69). Gervase of Canterbury, Historical Works, ed. W. Stubbs, 2 vols, Rolls Series (1879–80). Gesta Regis Henrici Secundi Benedicti Abbatis, ed. W. Stubbs, 2 vols, Rolls Series (1867). Gesta Stephani, 2nd edn, ed. and trans. K. R. Potter, introd. and notes R. H. C. Davis (Oxford, 1976). Giraldus Cambrensis, Opera, ed. J. S. Brewer, J. F. Dimock, and G. F. Warner, 8 vols, Rolls Series (1861–91). Expugnatio Hibernica: The Conquest of Ireland, ed. A. B. Scott and F. X. Martin (Dublin, 1978). The Gloucestershire Domesday, Alecto Historical edn, introd. Ann Williams (1989). The Great Rolls of the Pipe 2, 3, & 4 Henry II, ed. J. Hunter (Record Commission, 1844). Great Rolls of the Pipe, Henry II, Richard I, John and Henry III, Pipe Rolls Series, 38 vols (1884– ). Herefordshire Domesday, c.1160–70, ed. V. H. Galbraith and J. Tait, Pipe Rolls Series, new ser. 25 (1950). The Herefordshire Domesday Book, Alecto Historical edn, introd. Ann Williams (1988). Historia et Cartularium Monasterii Sancti Petri Gloucestriae, ed. W. H. Hart, 3 vols, Rolls Series (1863–7). The Hungerford Cartulary, ed. J. L. Kirby (Wiltshire Record Society, 1994). ‘The Irish Pipe Roll of 14 John, 1211–1212’, ed. O. Davies and D. B. Quinn, Ulster Journal of Archaeology, 3rd ser., 4 (1941), suppl. Itinerary of Richard I, ed. Lionel Landon, Pipe Rolls Series (1935).

264

Bibliography

The Letters of John of Salisbury, i, ed. S. J. Millor and H. E. Butler, rev. C. N. L. Brooke (1955). Liber Feodorum: The Book of Fees, commonly called Testa de Nevill, reformed from the earliest MSS, ed. H. C. Maxwell Lyte, 2 vols in 3 (1920–31). Littere Wallie, preserved in Liber A in the Public Record Ofﬁce, ed. J. G. Edwards (Cardiff, 1935). Map, Walter, De Nugis Curialium, ed. and trans. M. R. James, rev. C. N. L. Brooke and R. A. B. Mynors (Oxford, 1983). Matthew Paris, Chronica Majora, ed. H. R. Luard, 7 vols, Rolls Series (1872–84). ‘Ministers’ Accounts for the Lordships of Abergavenny, Grosmont, Skenfrith and White Castle, 1256–7’, ed. A. J. Roderick and W. Rees, South Wales and Monmouth Record Society Manuscripts, 2–4 (Cardiff, 1950–7), 2. 68–125, 3. 22–47, 4. 6–29. Monasticon Anglicanum, ed. W. Dugdale; rev. J. Caley et al., 6 vols in 8 pts (1817–30). Patent Rolls 1216–47 (1901–6). Pleas Before the King or his Justices 1198–1212, ed. D. M. Stenton, iii, Selden Society, 83 (1967). Ralph of Coggeshall, Chronicon Anglicanum, ed. J. Stevenson, Rolls Series (1875). Reading Abbey Cartularies, ed. B. R. Kemp, Camden Society, 4th ser., 31 (1986). Red Book of the Exchequer, 3 vols, ed. H. Hall, Rolls Series (1896). Regesta Regum Anglo-Normannorum 1066–1154, 4 vols (1913–69), iii, ed. H. A. Cronne and R. H. C. Davis. Register of St. Thomas, Dublin, ed. John T. Gilbert, Rolls Series (1889). Roger of Howden, Chronica, ed. W. Stubbs, 4 vols, Rolls Series (1867–71). Roger of Wendover, Flores Historiarum, ed. H. G. Hewlett, 3 vols, Rolls Series (1886–9). Rotuli Chartarum 1199–1216, ed. T. D. Hardy, Record Commission, 36 (1837). Rotuli Curiae Regis 1194–9, ed. F. Palgrave, 2 vols, Record Commission, 27 (1835). Rotuli de Liberate ac de Misi et Praestitis Regnante Johanne, ed. T. D. Hardy (Record Commission, 1844). Rotuli de Oblatis et Finibus . . . regis Johannis, ed. T. D. Hardy, Record Commission, 30 (1835). Rotuli Litterarum Clausarum in Turri Londinensi Asservati, ed. T. D. Hardy, 2 vols, Record Commission, 24 (1833–44). Rotuli Litterarum Patentium in Turri Londinensi Asservati, ed. T. D. Hardy, Record Commission, 28 (1835). Rotulorum Originalium in Curia Scaccarii Abbreviatio, ed. Henry Playford, 2 vols (Record Commission, 1805–10).

Bibliography

265

Royal and Other Historical Letters Illustrative of the Reign of Henry III, ed. W. W. Shirley, 2 vols, Rolls Series (1862–6). The Shropshire Domesday, Alecto Historical edn, introd. C. P. Lewis (1990). Sir Christopher Hatton’s Book of Seals, ed. Lewis C. Lloyd and D. M. Stenton (Oxford, 1950). The Song of Dermot and the Earl, ed. G. H. Orpen (Oxford, 1892; repr. Felinfach, 1994). Tractatus de Legibus et Consuetudinibus Regni Anglie qui Glanvilla Vocatur, ed. and trans. G. D. G. Hall (1965). Walter of Coventry, Memoriale, ed. W. Stubbs, 2 vols, Rolls Series (1872–3). Secondary Altschul, Michael, A Baronial Family in Medieval England: The Clares, 1217–1314 (Baltimore, 1965). Amt, Emilie, The Accession of Henry II in England: Royal Government Restored 1149–1159 (Woodbridge, 1993). Balfour, David, ‘The Origins of the Longchamp Family’, Medieval Prosopography, 18 (1997), 73–92. Ballard, A., and J. Tait (eds), British Borough Charters, 1216–1307 (Cambridge, 1923). Barlow, Frank, William Rufus (New Haven, 1983). Barnes, Patricia, ‘The Anstey Case’, in Patricia M. Barnes and C. F. Slade (eds), A Medieval Miscellany for Doris Mary Stenton, Pipe Rolls Series, new ser., 36 (1962), 1–24. Bartlett, Robert, Gerald of Wales 1146–1223 (Oxford, 1982). ‘Colonial Aristocracies of the High Middle Ages’, in Robert Bartlett and Angus Mackay (eds), Medieval Frontier Societies (Oxford, 1989), 23–47. The Making of Europe: Conquest, Colonization and Cultural Change 950–1350 (Princeton, 1993). Bateson, Mary, ‘The Laws of Breteuil’, EHR 15 (1900), 73–8, 302–18, 496–523, 754–7. Bazeley, Margaret Ley, ‘The Extent of the English Forest in the Thirteenth Century’, TRHS, 4th ser., 4 (1921), 140–72. Beresford, Maurice, New Towns of the Middle Ages: Town Plantation in England, Wales and Gascony (1967). Bolton, J. L., ‘The English Economy in the Early Thirteenth Century’, in S. D. Church (ed.), King John: New Interpretations (1999), 27–40. Brand, Paul, The Origins of the English Legal Profession (Oxford, 1992). Brooke, Z. N., and C. N. L. Brooke, ‘Hereford Cathedral Dignitaries in the Twelfth Century: Supplement’, Cambridge Historical Journal, 8 (1944–6), 1–21.

266

Bibliography

Cam, H. M., ‘The ‘‘Private’’ Hundred in England before the Norman Conquest’, in Cam, Law-Finders and Law-Makers in Medieval England (1962), 59–70. Carpenter, D. A., ‘The Decline of the Curial Sheriff in England, 1194–1258’, EHR 91 (1976), 1–32. ‘Was there a Crisis of the Knightly Class in the Thirteenth Century? The Oxfordshire Evidence’, EHR 95 (1980), 721–52. The Minority of Henry III (1990). ‘A Noble in Politics: Roger Mortimer in the Period of Baronial Reform and Rebellion, 1258–1265’, in Anne J. Duggan (ed.), Nobles and Nobility in Medieval Europe: Concepts, Origins, Transformations (Woodbridge, 2000), 183–203. ‘The Second Century of English Feudalism’, P & P 168 (Aug. 2000), 30–71. The Struggle for Mastery: Britain 1066–1284 (2003). Cathcart King, D. J., ‘Henry II and the Fight at Coleshill’, WHR 2 (1964–5), 367–75. Castellarium Anglicanum: An Index and Bibliography of the Castles in England, Wales and the Islands, 2 vols (1983). The Castle in England and Wales: An Interpretative History (1988). Church, S. D., ‘The Knights of the Household of King John: A Question of Numbers’, in P. R. Coss and S. D. Lloyd (eds), Thirteenth Century England IV (1992), 151–66. ‘The Rewards of Royal Service in the Household of King John: A Dissenting Opinion’, EHR 110 (1995), 277–302. The Household Knights of King John (Cambridge, 1999). Cokayne, George E., The Complete Peerage of England, Scotland, Ireland, Great Britain, and the United Kingdom, extant, extinct, or dormant, new edn, 14 vols in 7 (Gloucester, 1982–98). Colvin, H. M., ‘Holme Lacy: An Episcopal Manor and its Tenants in the Twelfth and Thirteenth Centuries’, in Veronica Ruffer and A. J. Taylor (eds), Medieval Studies Presented to Rose Graham (Oxford, 1950), 15–40. Copleston-Crow, B., ‘The Baskervilles of Herefordshire 1086–1300’, Trans. Woolhope, 43 (1979), 18–39. ‘The Fief of Alfred of Marlborough in Herefordshire in 1086 and its Descent in the Norman Period’, Trans. Woolhope, 45/2 (1986), 376–414. ‘Trewalkin and its Anglo-Norman Lords’, Brycheiniog, 26 (1992–3), 43–51. Coss, P. R., ‘Sir Geoffrey de Langley and the Crisis of the Knightly Class in Thirteenth-Century England’, P & P 68 (1975), 3–37. ‘Knighthood and the Early Thirteenth-Century County Court’, in P. R. Coss and S. Lloyd (eds), Thirteenth Century England II (1982), 45–57.

Bibliography

267

Lordship, Knighthood and Locality: A Study in English Society c.1180–c.1280 (Cambridge, 1991). The Knight in Medieval England (Stroud, 1993). ‘Identity and the Gentry c.1200–c.1340’, in M. Prestwich, R. H. Britnell, and R. Frame (eds), Thirteenth Century England VI (1997), 49–60. The Origins of the English Gentry (Cambridge, 2003). Cowley, J. G., The Monastic Order in South Wales (Cardiff, 1977). Cownie, Emma, ‘Gloucester Abbey, 1066–1135: An Illustration of Religious Patronage in Anglo-Norman England’, in D. Bates and A. Curry (eds), England and Normandy in the Middle Ages (1994), 143–58. Religious Patronage in Anglo-Norman England 1066–1135 (1998). Crook, David, Records of the General Eyre, PRO Handbook, 20 (1982). Crouch, David, ‘The Slow Death of Kingship in Glamorgan, 1067–1158’, Morgannwg, 29 (1985), 20–41. ‘The Last Adventure of Richard Siward’, Morgannwg, 35 (1991), 7–30. The Image of Aristocracy in Britain 1000–1300 (1992). ‘The March and the Welsh Kings’, in Edmund King (ed.), The Anarchy of King Stephen’s Reign (Oxford, 1994), 255–89. ‘From Stenton to McFarlane: Models of Societies of the Twelfth and Thirteenth Centuries’, TRHS, 6th ser., 5 (1995), 179–200. The Reign of King Stephen 1135–1154 (1999). William Marshal: Knighthood, War and Chivalry, 1147–1219, 2nd edn (2002). Crump, J. J., ‘The Mortimer Family and the Making of the March’, in M. Prestwich, R. H. Britnell, and R. Frame (eds), Thirteenth Century England VI (1997), 117–26. Dalton, P., Conquest, Anarchy and Lordship: Yorkshire 1066–1154 (Cambridge, 1994). Darby, H. C., and I. B. Terrett (eds), The Domesday Geography of Midland England (Cambridge, 1954). Davies, R. R., ‘Richard II and the Principality of Chester 1397–9’, in F. R. H. Du Boulay and C. Barron (eds), The Reign of Richard II (1971), 256–79. ‘The Law of the March’, WHR 5 (1970–1), 1–30. ‘Brecon’, in R. A. Grifﬁths (ed.), Boroughs of Medieval Wales (Cardiff, 1978), 46–70. Lordship and Society in the March of Wales 1282–1400 (Oxford, 1978). ‘Kings, Lords and Liberties in the March of Wales, 1066–1272’, TRHS, 5th ser., 29 (1979), 41–61. ‘Henry I and Wales’, in H. Mayr-Harting and R. I. Moore (eds), Studies in Medieval History Presented to R. H. C. Davis (1985), 133–47. The Age of Conquest: Wales 1063–1415 (Oxford, 1987; paperback, 1991). Domination and Conquest: The Experience of Ireland, Scotland and Wales 1100–1300 (Cambridge, 1990).

268

Bibliography

Davies, R. R., The Revolt of Owain Glyn Dwr (Oxford, 1995). The First English Empire: Power and Identities in the British Isles 1093–1343 (Oxford, 2000). Davis, R. H. C., ‘Treaty between William Earl of Gloucester and Roger Earl of Hereford’, in Patricia M. Barnes and C. F. Slade (eds), Early Medieval Miscellany for Doris Mary Stenton, Pipe Rolls Series (1962), 139–46. King Stephen, 3rd edn (1990). Denholm-Young, N., Richard of Cornwall (Oxford, 1947). Dictionary of National Biography: from the earliest times to 1900, ed. Leslie Stephen and Sidney Lee, 22 vols (Oxford, 1967–8). Duffy, Sean, ‘King John’s Expedition to Ireland, 1210: The Evidence Reconsidered’, Irish Historical Studies, 30/117 (May 1996), 1–24. ‘John and Ireland: The Origins of England’s Irish Problem’, in S. D. Church (ed.), King John: New Interpretations (Woodbridge, 1999), 221–45. Duncan, A. A. M., ‘King John of England and the Kings of Scots’, in S. D. Church (ed.), King John: New Interpretations (Woodbridge, 1999), 247–71. Dyer, C., Standards of Living in the Later Middle Ages: Social Change in England c.1200–1520 (Cambridge, 1989). Edwards, J. G., ‘The Normans and the Welsh March’, Proceedings of the British Academy, 42 (1956), 155–77. ‘Henry II and the Fight at Coleshill: Some Further Reﬂections’, WHR 3 (1966–7), 251–63. Eyton, R. W., Antiquities of Shropshire, 12 vols (1854–60). Faulkner, K., ‘The Transformation of Knighthood in Early Thirteenth-Century England’, EHR 111 (1996), 1–23. Flanagan, M. T., Irish Society, Anglo-Norman Settlers, Angevin Kingship: Interactions in Ireland in the Late Twelfth Century (Oxford, 1989). ‘Irish and Anglo-Norman Warfare in Twelfth-Century Ireland’, in T. Bartlett and K. Jeffrey (eds), A Military History of Ireland (Cambridge, 1996), 52–75. Fleming, D., ‘Landholding by Milites in Domesday Book: A Revision’, ANS 13 (1991), 83–98. Frame, Robin, The Political Development of the British Isles 1100–1400 (Oxford, 1990). ‘King Henry III and Ireland: The Shaping of a Peripheral Lordship’, in Frame, Ireland and Britain 1170–1450 (1998), 31–58. ‘Aristocracies and the Political Conﬁguration of the British Isles’, in Frame, Ireland and Britain 1170–1450 (1998), 151–70. ‘The Defence of the English Lordship, 1250–1450’, in T. Bartlett and K. Jeffrey (eds), A Military History of Ireland (Cambridge, 1996), 76–98.

Bibliography

269

Gillingham, John, ‘Henry II, Richard I and the Lord Rhys’, Peritia, 10 (1996), 225–36; repr. in Gillingham, The English in the Twelfth Century: Imperialism, National Identity and Political Values (Woodbridge, 2000), 59–68. Richard I (New Haven, 1999). The English in the Twelfth Century: Imperialism, National Identity and Political Values (Woodbridge, 2000). ‘Conquering the Barbarians: War and Chivalry in Twelfth Century Britain’, in Gillingham, The English in the Twelfth Century: Imperialism, National Identity and Political Values (Woodbridge, 2000), 41–58. Green, Judith A., The Government of England under Henry I (Cambridge, 1986). The Aristocracy of Norman England (Cambridge, 1997). Harper-Bill, C., ‘The Piety of the Anglo-Norman Knightly Class’, ANS 2 (1979), 63–77. Harris, B. H., ‘King John and the Sheriff ’s Farms’, EHR 79 (1964), 532–47. Harvey, P. D. A., ‘The English Inﬂation of 1180–1220’, P & P 61 (1973), 3–30. ‘The Pipe Rolls and the Adoption of Demesne Farming in England’, EcHR, 2nd ser., 27 (1974), 345–59. Heiser, Richard R., ‘Richard I and his Appointments to English Shrievalties’, EHR 112 (Feb. 1997), 1–19. Hillaby, J., ‘Hereford Gold: Irish, Welsh and English Land. The Jewish Community at Hereford and its Clients, 1179–1253. Part I’, Trans. Woolhope, 44/3 (1984), 358–419. ‘Hereford Gold: Irish, Welsh and English Land. Part II. The Clients of the Jewish Community at Hereford 1179–1253: Four Case Studies’, Trans. Woolhope, 45/1 (1985), 193–270. Hilton, R. H., A Medieval Society: The West Midlands at the End of the Thirteenth Century (1966). Holden, Brock W., ‘The Balance of Patronage: King John and the Earl of Salisbury’, Haskins Society Journal, 8 (1999), 79–89. ‘The Making of the Middle March of Wales, 1066–1250’, WHR 20/2 (Dec. 2000), 207–26. ‘King John, the Braoses, and the Celtic Fringe, 1207–1216’, Albion, 33/1 (Spring 2001), 1–23. ‘ ‘‘Feudal Frontiers?’’ Colonial Societies in Wales and Ireland 1170–1330’, Studia Hibernica, 33 (2004–5), 61–79. Holm, Paul, ‘The Slave Trade of Dublin, Ninth to Twelfth Centuries’, Peritia, 5 (1986), 317–45. Holt, J. C., The Northerners: A Study in the Reign of King John (Oxford, 1961; repr. corr., 1992). ‘The Casus Regis: The Law and Politics of Succession in the Plantagenet Dominions, 1185–1247’, in Edward B. King and Susan J. Ridyard (eds), Law in Mediaeval Life and Thought (Sewannee, Tenn., 1990), 21–42; repr. in Holt, Colonial England 1066–1215 (1997), 307–26.

270

Bibliography

Holt, J. C., Magna Carta, 2nd edn (Cambridge, 1992). Colonial England 1066–1215 (1997). Holt, R., The Mills of Medieval England (1988). Hudson, John, Land, Law, and Lordship in Anglo-Norman England (Oxford, 1994). The Formation of the English Common Law (1996). Hunt, John, Lordship and Landscape: A Study of the Honour of Dudley 1066– 1322, BAR British Series, 264 (1997). Hyams, Paul R., review of S. F. C. Milsom, The Legal Framework of English Feudalism, EHR 93 (1978), 856–61. Jolliffe, J. E. A., Angevin Kingship (Oxford, 1955). Keefe, T. K., Feudal Assessments and the Political Community under Henry II and his Sons (Berkeley, 1983). ‘Proffers for Heirs and Heiresses in the Pipe Rolls: Some Observations on Indebtedness in the Years before the Magna Carta (1180–1212)’, Haskins Society Journal, 5 (1996), 99–109. King, E., ‘Large and Small Landowners in Thirteenth-Century England: The Case of Peterborough Abbey’, P & P 47 (1970), 26–50. Knight, Jeremy K., The Three Castles, Welsh Historic Monuments pamphlet (Cardiff, 2000). ‘The Road to Harlech: Aspects of Some Early Thirteenth-Century Welsh Castles’, in John R. Kenyon and Richard Avent (eds), Castles in Wales and the Marches: Essays in Honour of D. J. Cathcart King (Cardiff, 1987), 75–88. Knowles, D., The Monastic Order in England : A History of its Development from the Times of St. Dunstan to the Fourth Lateran Council, 943–1216 (Cambridge, 1949). and R. N. Hadcock, Medieval Religious Houses in England and Wales (1953). Latimer, Paul, ‘Henry II’s Campaign against the Welsh in 1165’, WHR 14/4 (Dec. 1989), 523–45. ‘Early Thirteenth-Century Prices’, in S. D. Church (ed.), King John: New Interpretations (Woodbridge, 1999), 41–73. Le Patourel, J., The Norman Empire (Oxford, 1976). Lewis, C. P., ‘The Norman Settlement of Herefordshire under William I’, ANS 7 (1985), 195–213. Lewis, F. R., ‘A History of the Lordship of Gower from the Missing Cartulary of Neath’, BBCS 9 (1937–9), 149–54. Lloyd, J. E., A History of Wales from the Earliest Times to the Edwardian Conquest, 3rd edn, 2 vols (1939). McFarlane, K. B., The Nobility of Later Medieval England: The Ford Lectures for 1953 and Related Studies (Oxford, 1973).

Bibliography

271

‘Had Edward I a ‘‘Policy’’ towards the Earls?’, in McFarlane, The Nobility of Later Medieval England: The Ford Lectures for 1953 and Related Studies (Oxford, 1973), 248–67. McNeill, T. E., and M. Pringle, ‘A Map of Mottes in the British Isles’, Medieval Archaeology, 41 (1997), 220–2. Maddicott, J. R., Simon de Montfort (Cambridge, 1994). Mann, K., ‘The March of Wales: A Question of Terminology’, WHR 18/1 (June 1996), 1–13. Meisel, J., Barons of the Welsh Frontier: The Corbet, Pantulf and Fitzwarin Families 1066–1272 (Lincoln, Nebr., 1980). Miller, E., ‘England in the Twelfth and Thirteenth Centuries: An Economic Contrast?’, EcHR, 2nd ser., 24 (1971), 1–14. Milsom, S. F. C., The Legal Framework of English Feudalism (Cambridge, 1976). Historical Foundations of the Common Law, 2nd edn (1981). ‘Inheritance by Women in the Twelfth and Early Thirteenth Centuries’, in M. S. Arnold et al. (eds), On the Laws and Customs of England: Essays in Honour of S. E. Thorne (Chapel Hill, NC, 1981), 60–89. Millward, Roy, and Adrian Robinson, The Welsh Borders (1978). Otway-Ruthven, A. J., ‘The Constitutional Position of the Great Lordships of South Wales’, TRHS, 5th ser., 8 (1958), 1–20. A History of Medieval Ireland (1968). Painter, Sidney, Studies in the History of the English Feudal Barony (Baltimore, 1943). The Reign of King John (Baltimore, 1949). ‘Castle-Guard’, American Historical Review, 40 (1935), 450–9; repr. in Fred A. Cazel, Jr. (ed.), Feudalism and Liberty (Baltimore, 1961), 144–56. Palmer, R. C., ‘The Feudal Framework of English Law’, Michigan Law Review, 79 (1981), 1130–64. Pollock, Sir Frederick, and F. M. Maitland, The History of English Law before the Time of Edward I, 2nd edn, introd. S. F. C. Milsom, 2 vols (Cambridge, 1968). Postan, M. M., ‘Medieval Agrarian Society in its Prime: England’, in Postan (ed.), The Cambridge Economic History of Europe, 2nd edn, 8 vols (Cambridge, 1966), i. 549–659. (ed.), The Cambridge Economic History of Europe, 2nd edn (Cambridge, 1966). Pounds, N. J. G., The Medieval Castle in England and Wales (Cambridge, 1990). Powicke, F. M., The Fall of Normandy (Manchester, 1913). King Henry III and the Lord Edward (Oxford, 1947). Prestwich, Michael, Edward I, new edn (New Haven, 1997). PRO Lists and Indexes, ix: List of Sheriffs for England and Wales (1898).

272

Bibliography

Pugh, T. B. (ed.), Glamorgan County History, iii (1971). Quick, J., ‘The Number and Distribution of Knights in Thirteenth-Century England: The Evidence of the Grand Assize Lists’, in P. R. Coss and S. Lloyd (eds), Thirteenth Century England I (1986), 114–23. Radford, C. A. R., and D. M. Robinson, Tretower Court and Castle, Welsh Historic Monuments pamphlet, rev. edn (Cardiff, 1985). Rees, W., ‘The Medieval Lordship of Brecon’, Transactions of the Honourable Society of Cymmrodorion (1915–16), 165–224. Renn, D. F., ‘The Round Keeps of the Brecon Region’, Archaeologia Cambrensis, 110 (1961), 129–43. Round, J. H., Introduction to The Victoria History of Herefordshire, i, ed. William Page (1908). Rowlands, I. W., ‘The Making of the March: Aspects of the Norman Settlement of Dyfed’, ANS 3 (1981), 142–57. ‘William de Braose and the Lordship of Brecon’, BBCS 30 (1982–3), 123–33. ‘King John and Wales’, in S. D. Church (ed.), King John: New Interpretations (Woodbridge, 1999), 273–87. Sanders, I. J., English Baronies: A Study of their Origin and Descent 1086–1327 (Oxford, 1960). Simms, Katherine, ‘Warfare in the Medieval Gaelic Lordships’, Irish Sword, 12 (1975–6), 98–108. ‘Gaelic Warfare in the Middle Ages’, in T. Bartlett and K. Jeffrey (eds), A Military History of Ireland (Cambridge, 1996), 99–115. Simpson, G. G., ‘The Familia of Roger de Quincy, Earl of Winchester and Constable of Scotland’, in K. J. Stringer (ed.), Essays on the Nobility of Medieval Scotland (Edinburgh, 1985), 102–30. Smith, Brendan, Colonisation and Conquest in Medieval Ireland: The English in Louth, 1170–1330 (Cambridge, 1999). Smith, J. B., ‘The Lordship of Glamorgan’, Morgannwg, 2 (1958), 9–38. ‘The Middle March in the Thirteenth Century’, BBCS 24 (1970–2), 77–93. ‘Marcher Regality: Quo Warranto Proceedings relating to Cantrefselyf in the Lordship of Brecon, 1349’, BBCS 28 (1978–80), 267–88. ‘Magna Carta and the Charters of the Welsh Princes’, EHR 99 (1984), 344–62. Llywelyn ap Gruffudd, Prince of Wales (Cardiff, 1998). Spurgeon, C. J., ‘Builth Castle’, Brycheiniog, 18 (1978–9), 47–59. Stacey, Robert, Politics, Policy and Finance under Henry III 1216–1245 (Oxford, 1987). Stenton, D. M., English Justice between the Norman Conquest and the Great Charter 1066–1215 (1964).

Bibliography

273

Stenton, F. M., The First Century of English Feudalism, 2nd edn (Oxford, 1961). Strickland, Matthew, War and Chivalry: The Conduct and Perception of War in England and Normandy, 1066–1217 (Cambridge, 1996). Stringer, K. J., ‘Nobility and Identity in Medieval Britain and Ireland: The de Vescy Family, c.1120–1314’, in Brendan Smith (ed.), Britain and Ireland 900–1300: Insular Responses to Medieval European Change (Cambridge, 1999), 199–239. Suppe, Frederick C., Military Institutions on the Welsh Marches: Shropshire, AD 1066–1300 (Woodbridge, 1994). Sylvester, Dorothy, The Rural Landscape of the Welsh Borderland: A Study in Historical Geography (1969). Tellenbach, G., The Church in Western Europe from the Tenth to the Early Twelfth Century, trans. T. Reuter (Cambridge, 1993). Thomas, Hugh M., Vassals, Heiresses, Crusaders, and Thugs: The Gentry of Angevin Yorkshire, 1154–1216 (Philadelphia, 1993). Treharne, R. F., ‘The Franco-Welsh Treaty of Alliance in 1212’, BBCS 18 (1958), 60–75; repr. in E. B. Fryde (ed.), Simon de Montfort and Baronial Reform: Thirteenth Century Essays (1986), 345–60. Turner, Ralph V., The King and his Courts: The Role of John and Henry III in the Administration of Justice, 1199–1240 (Ithaca, NY, 1968). The Victoria History of Hampshire, iv, ed. William Page (1911). The Victoria History of Herefordshire, i, ed. William Page (1908). The Victoria History of Shropshire, i, ed. William Page (1908). Vincent, Nicholas, Peter des Roches: An Alien in English Politics 1205–1238 (Cambridge, 1996). Walker, D. G., ‘The ‘‘Honours’’ of the Earls of Hereford in the Twelfth Century’, Transactions of the Bristol and Gloucestershire Archaeological Society, 79/2 (1960), 174–211. Walker, R. F., ‘Hubert de Burgh and Wales 1218–32’, EHR 87 (1972), 465–94. ‘The Supporters of Richard Marshal, Earl of Pembroke, in the Rebellion of 1233–34’, WHR 17/1 (June 1994), 41–65. Warren, W. L., Henry II (Berkeley, 1973). ‘The Historian as Private Eye’, Historical Studies, 9 (1974), 1–18. Waugh, Scott L., The Lordship of England: Royal Wardships and Marriages in English Society and Politics 1217–1327 (Princeton, 1988). Wightman, W. E., ‘The Palatine Earldom of William ﬁtz Osbern in Gloucestershire and Worcestershire, 1066–1071’, EHR 77 (1962), 6–17. The Lacy Family in England and Normandy 1066–1194 (Oxford, 1966). Wolffe, B. P., The Royal Demesne in English History (1971).

274

Bibliography

UNPUBLISHED THESES Golob, P. E., ‘The Ferrers Earls of Derby: A Study of the Honour of Tutbury (1066–1279)’, Ph.D. thesis, University of Cambridge (1984). Lewis, C. P., ‘English and Norman Government and Lordship in the Welsh Borders, 1039–1087’, D.Phil. thesis, University of Oxford (1985).

Index Abergavenny, 2, 7, 25–6, 28, 41–2, 66, 140, 159, 172, 187, 205, 207, 225 Castle, 143, 186 economy of, 71, 73–6 Great Court of, 72 knightly tenants of, 186, 210 massacre at (1175), 36, 143, 167 Priory, 39, 83 town, 37 Aconbury Priory, 82–4 advowsons, 40, 115 Andrew de Chanceaux, 97, 125, 160 Archenﬁeld, 7, 75, 94, 125, 150 assarting, 79, 84, 99, 124–25 Augustinians, 40, 83 Baderon de Monmouth, 13, 23, 29, 80, 82, 85 Bernard de Neufmarch´e, 13, 18–19 conquest of Brycheiniog, 14, 28, 36, 102 foundation of Brecon Priory, 82, 93, 113–14 grants of land, 28, 39, 40, 77 as rebel, 14 Blaenllyfni, 78, 104, 146, 182, 186, 194–6, 199, 201, 214 Brecon Castle, 15, 33, 55–6, 58, 71 Priory, 36, 40, 70, 81–2, 93, 113–14, 177 town of, 33, 53, 97, 197 Bredwardine, 114, 123, 125 Castle, 35, 54, 64, 117, 155 curia of, 205 descent of, 14 n.7, 28, 64–5, 70, 94–5, 111–12, 122–4, 149 Bridgnorth, 24–5, 147 Bronllys, 6, 15, 34, 54, 61, 69, 77, 117, 200, 211 Brycheiniog, conquest of Welsh kingdom of, 6, 13–15, 93, 103, 140, 220 Builth, 43, 73–4, 197

Castle, 53, 185 and n.94, 186, 191, 202, 205, 207 descent of, 191, 207, 216, 225 in royal hands, 217 lordship of, 2, 17, 26, 28, 34, 143, 187 town, 37 Cantref Bychan, 15–16, 20 Cantref Selyf and Cliffords, 15, 17, 26, 201 Cistercians in, 83 Cartae Baronum, 12, 27, 43, 92, 105, 187 Castellaria, 27, 54–5 castle-guard, 52, 54–7, 70, 116, 219 Cecily Devereux, 113, 128–30 Ceri, 190, 206–7, 216 Chepstow Castle, 37, 54, 117 Cistercians, 40, 75, 83, 114 Clifford, 6–7 Castle, 32–3, 35 and n.92, 37, 54–5, 61, 116, 210–11, 213 lordship of, 2, 26, 41–2, 106 Priory, 39, 82–3 town of, 36–7 Craswall Priory, 39, 56, 79, 84 Crickhowell, 17, 34, 77, 214 Dafydd ap Llywelyn, 199, 207, 215–17, 225 demesne baronial, 25, 27–31, 42, 59–61, 72, 76–80, 110, 115, 223 royal, 24, 150–1, 199, 228 distraint, 62, 67–8, 78, 86 Domesday Book, 7, 12, 26, 28, 55, 116 Dore Abbey, 40, 81, 83, 94, 123, 126, 201 Drew ﬁtz Pons, 11 Eardisley, 51 n.22, 94–5, 111, 128, 130–1, 160 Castle, 116–18

276 economy upland, 6, 75–6 lowland, 6, 76–7 Welsh, 73 Elfael, 2, 7, 9, 17, 93, 142, 145, 168, 191 endogamy, 123–4, 135 Engelard de Cigogne´e, 96 Evesham, Battle of, 102, 119, 225–27 Ewyas Harold, 6, 32, 35, 37, 55, 85 Ewyas Lacy, 2, 19, 26, 33, 55, 83, 147, 205, 215 castle at (Longtown), 54, 61, 100 eyres, 45, 154–6, 163 Falkes de Breaut´e, 148, 177, 181 Forest of Dean, 72, 80, 150, 162, 209 Forges, 80 Gerald of Wales description of Marcher warfare, 48–9 on Celtic warfare, 49, 60 Gerard d’Ath´ee, 96, 153, 160 Gilbert de Lacy (d. post 1163), 19, 21–4, 31 Gilbert de Lacy (d. 1230), 59, 100–1, 132, 188, 205 Gilbert de Monmouth, 80, 125, 148, 157 Giles de Braose, bsp. of Hereford, 172, 178–9, 181, 185–6, 188, 194–5, 201, 210 Glamorgan, 19, 49, 168, 187, 191 and William de Braose, 172, 177 appearance of legal reforms in, 66 consolidation of lordship in, 47, 67–9, 77, 86, 225 economy of, 75 royal wardship of, 142, 148 Welsh risings in, 20, 220 Glasbury, 6, 15, 55, 210–11 Grace Dieu Priory, 83 Gwladys, dau. of Llywelyn ap Iorwerth, 84, 193, 206 Henry I, King of England and Marcher barons, 12, 15–16, 18, 20–1, 140 and ‘new men’, 19, 142, 147–8, 164, 221

Index policy towards Wales, 140–2, 144, 221 Welsh campaigns of, 140, 221 Henry II, king of England, 24, 43, 45, 151, 229 and Cartae Baronum, 12 detente with Lord Rhys, 143–4 and Ireland, 30, 227 legal reforms of, 70, 154–5 policy towards Wales, 141–4, 147–8, 164, 166, 221 Welsh campaigns of, 141–2 Henry III, king of England, 146–7, 207, 216 and Marshal rebellion, 210–14 minority of, 189–92 policy towards Wales and March, 144–5, 222 use of debt, 164, 212–13 Henry de Pembridge, 225–6 Hereford, 22 n.41, 27, 116, 133, 138 Castle, 24, 35, 151, 161 in 1215–16, 186–7 Jewish community of, 118, 171, 212 laws of, 37 sheriff of, 152, 158, 171, 192, 194, 196, 202 town, 14, 36–7, 150 Herefordshire Domesday, 12, 26, 28, 30, 94, 98, 105, 113, 150 honour courts, 68, 87, 116, 126 n.183, 154–5, 223, 229 Hubert de Burgh, 124, 153, 159, 181, 200, 202–3, 206–9, 217 Hugh II de Lacy, 78, 89, 98 household of, 58, 109 in Ireland, 134, 228 Hugh III de Lacy, e. of Ulster, 181–3 exile, 188 n.108 rebellion in Ireland, 212 n.111 Hugh de Mortimer (d. 1180/1), 17, 20–5, 29, 142, 147 Hugh de Mortimer (d. 1227), 8–9, 78, 188, 194, 202, 204 death, 206 and manors of Knighton and Norton, Salop., 198–9 Hugh de Say, 60, 145, 168 Humphrey IV de Bohun, 47, 69 n.85

Index inf lation, 30, 77, 88, 109, 114, 175 baronial response to, Ireland, 4, 9, 33, 73, 141–2, 144 Braoses in, 159, 172, 174, 176, 181, 183 castles and subinfeudation, 33–5 economy, 75, 171 effects on Welsh affairs, 167, 182, 213 honours in, 66 and n.78, 227–8 Lacys in, 30–1, 60, 61 n.61, 81–2, 85, 100, 161, 171, 182, 188, 204–5, 211–12, 214–5 legal differences, 155, 228 manorialization of, 74 military service in, 56 n.44 opportunities for knights, 111, 131–5, 170–1 revenues from used in March, 39, 84 royal patronage in, 163 n.132, 182 slave trade, 51 warfare in, 48–9, 51 Isabelle de Mortimer, 148 John de Braose, 178, 179 n.75, 193 n.10, 194–5, 199, 208–9 John Devereux (d.c. 1188, father of Stephen), 98 Jocelin de Dinan, 18, 21–2 John de Monmouth, 8–9, 82–83, 158, 161, 188, 204 as royal servant, 31, 83, 97, 144, 153, 212 during civil war, 194, 229 during Marshal rebellion, 210, 212, 217 John, king of England, 8, 153, 204 custody of Marcher lordships, 66, 148–9, 181–2, 185–6, 201 death, 193 and Giles and Reginald de Braose, 194–5 grants by, 84 and Ireland, 177, 182–3, 228 and Marchers, 44, 99, 102, 171, 187–8, 201, 204, 210, 229 and Three Castles, 203–4 Welsh policy, 148–9, 159, 165, 176–7, 188–9, 193, 217, 221–2 and William de Braose, 9, 145, 159, 167, 171–6, 177–81

277 John Pichard (f l. 1170s), 115, 157 John Pichard (f l. early 13th century), 102–3, 188 knights, 4, 92, 104–5, 108, 135, 224, 227 and Crown, 108, 131–35, 137, 152, 154, 156–8, 162–3, 226 and debt, 52, 226 deﬁnition of, 88–91 and economics, 79, 109–10, 124–26 fee, 12, 17, 31, 105–8, 137, 187, 192 household, 28, 31, 51, 58–9 litigation by, 126–31 and marriage strategies, 119–24 and Marshal rebellion, 210–11, 214, 217 and religious houses, 113–15 service of, 55 as suitors in lord’s court, 67 Llanthony Prima, 41, 82 Llanthony Secunda, 41 Llywelyn ap Gruffudd, 218 Llywelyn ap Iorwerth (Llywelyn the Great), 52, 159, 167, 182–4, 190, 223 death of, 191, 215 hanging of William V de Braose, 206 marriages of daughters of, 84–5, 225 and Treaty of Worcester, 197–8 Longtown, see Ewyas Lacy Ludlow, 7, 18, 22, 27, 188 Lyonshall, 98, 100–1, 113, 127, 131, 162 Castle, 54, 100, 102, 117–18, 131 Maelienydd, 7, 9, 17, 43, 85, 142–3, 168 during reign of Henry III, 191, 198–9, 202, 204, 206, 212, 215–16 Magna Carta, 9, 146 Mansel Lacy, 22, 30–1, 57, 78, 116 maritagium, 57, 119–20, 122 Matthew de Gamages, 160, 169 Maud de Braose (wife of William III de Braose), 84, 177, 180–1 and King John, 173

278 Miles of Gloucester, e. of Hereford, 19–25, 31, 78, 97, 151, 207 marriage of, 19, 93, 147 mills, 76, 101, 114–15 Monmouth, 42, 80, 192 Castle, 11, 37 economy of, 73–4, 76 honour of, 2, 29, 42, 80, 106, 219 Great Court of, 71–2 Priory, 39–40, 80, 82–3 town of, 37 mort d’ancestor, 63, 126, 228 Neath Abbey, 177 novel disseisin, 66, 154, 228 Payn ﬁtz John, 17, 19–21 Pencelli, 17, 69 n.85, 70, 78, 94–5, 111–12, 186–7, 201 n.51 Peter ﬁtz Herbert, 78, 176, 182, 194–5, 199 n.39, 201, 214 Peter des Riveaux, 153, 208, 217 Peter des Roches, 96–7, 99, 184, 208 Philip II Augustus, king of France, 160, 173, 182, 187, 200 prerogative lordship, 138, 146–8, 228 Radnor, 2, 196 battle at (1196), 60, 145, 168 Braoses and, 13, 17, 26, 28, 143, 171–2, 186 Castle, 53 economy of, 73–4 honour of, 41, 85, 207, 225 in royal custody, 66 scutage, 187 tenants of, 186, 210 town of, 36 uplands of, 7, 43 Ralph I de Baskerville, 93–4, 111 Ralph II de Baskerville, 64, 95, 111, 122 Ralph III de Baskerville, 65, 112, 114, 122, 149 Ralph of Coggeshall, 178, 194 Ralph de Mortimer (d. 1246), 206–7, 212, 215–17

Index Ranulf, earl of Chester, 173–4 Reginald de Braose, 9, 123, 178, 200, 204, 206, 210 death, 193 dispute with John de Braose, 199 rebellion of, 186–8, 193–4, 201, 203 return to Royalist side during civil war, 195–6 Rhys, The Lord, 53, 60, 109, 145, 168, 223 and Henry II, 142–4, 152, 164, 221 Richard I, king of England, 56, 128, 140, 168, 170 and Wales, 144, 148, 163–5, 167 and William de Braose, 172 Richard de Clare, e. of Gloucester and lord of Glamorgan, 67–9, 77 Richard de Clare, e. of Hertford, 84 Richard of Cornwall, 124, 208–9 Richard Marshal, 103, 162, 164, 204, 207, 209 Richard Siward, 67–9, 213 River Usk, 6–7, 15, 34, 49, 74, 76, 186 River Wye, 6–8, 15, 26, 34, 42, 49, 53, 55, 74, 76, 186 Robert, earl of Gloucester, 19–21, 23 Robert of Torigny, 25 Roger de Baskerville (f l. early 12th century), 17, 40, 78, 93, 114 Roger Devereux, 97–8, 112–3, 122–4, 129 Roger, Earl of Hereford, 22–4, 31, 79, 94 Roger de Mortimer (d. 1214), 60, 145, 168–9, 221 and support for King John, 9, 149, 159 Roger de Mortimer (d. 1282), 225–7, 229 and pressure on local knights, 102, 118 Roger Pichard (f l. mid-12th century), 102, 115, 117 Roger Pichard (f l. 1230–48), 92, 103–4, 117, 133, 158 and Marshal rebellion, 103 Roger Wendover, 172–3, 176

Index Scutage, 54, 56, 105, 125, 158, 175 n.55, 187, 192 Marcher exemption from, 56–7 sheriffs, 151–3, 163, 175 n.56, 181, 203 Simon de Clifford, 78–9 Simon de Montfort (elder), 187–8 and n.108 Simon de Montfort (younger), 225–6 St. Guthlac, Hereford, 39 Stephen Devereux, 59, 79, 99, 103, 104, 108, 124 death, 100 and development of Lyonshall, 100–1 grants by, 101, 114 marriage, 160 relationship with the Marshal, 98–9, 132, 134 and royal patronage, 99, 131–2, 158, 161–2 Stephen, king of England, 12, 19–23 and Marchers, 20–1 subinfeudation, 25, 27–30, 107–8, 133 and creation of the March, 4, 12, 15, 35, 42 in Shropshire, 106 terra Normannorum, 127, 149 teulu, 51, 222 Three Castles, 186, 203–4, 208–9 Treaty of Woodstock, 217 Tretower, 92, 103, 133 Castle, 17, 34, 54, 104, 117, 200 Usk Castle, 200, 211 Walter I de Baskerville, 95–6, 127 Walter II de Baskerville, 96–7, 103, 124–5, 131, 160 Walter I de Clifford (d.1190), 85, 170 Walter II de Clifford (d.1221), 64, 201, 206 Walter III de Clifford (d.1263), 60, 85, 161, 186, 201, 204, 209 Walter II de Lacy (d.1241), 8–9, 69, 171, 88, 116, 147, 194–6 death, 60, 193, 214 debts, 157, 161, 171, 212

279 exile, 187–8 grants by, 56–7, 78, 84, 132 in Ireland, 59, 74, 204–5 and King John, 99, 153, 188, 193, 229 and Marshal rebellion, 210–13 relationship with William de Braose, 23 n.44, 85, 180 ‘wardens’, royal, 143, 164, 216–17 wardship, 66, 69, 87, 146–7, 158, 187, 196 Weobley, 30 Castle, 33, 35, 54–6 honour of, 18 n.24, 26–7, 29, 58, 61, 70, 78, 89, 105–8, 211, 215 royal wardship of, 147, 160 Wigmore, 6, 8 Abbey, 41 borough of, 37 Castle, 35, 54–5, 116 honour of, 26–9, 47, 106, 219, 225 William I, king of England, 13, 15, 36, 39, 139, 219 William II (Rufus), king of England, 11, 13–15, 19, 39, 139, 220 William II de Braose, 17, 20, 22–3, 25 n.50, 28, 31 William III de Braose, 9, 41, 43, 64, 88, 143, 145, 155, 168, 221, 229 and Arthur of Brittany, 172–3 as sheriff, 152–3 crisis of 1208, 167, 169, 172–7, 182–3, 189 debts, 171 demise of lands by, 177–81 grants by, 55 in Ireland, 9, 81, 172, 174, 183, 224 relationship with King John, 65, 75, 170, 172, 203 rise to prominence, 171–2 tenants of, 102, 116, 121, 185–7 William IV de Braose (the Younger), 70, 175, 180 demise of lands to, 177–80 William V de Braose, 159, 191, 206, 208, 225 William I de Cantilupe, 98, 159–60 William III de Cantilupe, 159, 207 William Devereux (son of Stephen), 101, 102, 113, 118

280 William de Longchamp, 149, 152, 168 William Marshal, 132 and crisis of 1207–8, 99, 177, 181 familia of, 91, 98–9, 132 in Ireland, 99, 177 in the March, 194–6

Index as regent, 194–6, 203–4, 208 relationship with the Lacys, 99, 177, 181, 213 William II Marshal, 101, 207–8 William ﬁtz Osbern, 10, 26, 37, 219 Wormsley Priory, 101

Lords of the Central Marches: English Aristocracy and Frontier Society, 1087-1265 (Oxford Historical Monographs)

The Quakers in English Society, 1655-1725 (Oxford Historical Monographs)

Milton and the Ineffable (Oxford English Monographs)

On Sympathy (Oxford English Monographs)

On Sympathy (Oxford English Monographs)

On Sympathy (Oxford English Monographs)

Conrad and History (Oxford English Monographs)

Stevie Smith and Authorship (Oxford English Monographs)

The Bridges of Medieval England: Transport and Society 400-1800 (Oxford Historical Monographs)

The Suffering Traveller and the Romantic Imagination (Oxford English Monographs)

Historians and Nationalism: East-Central Europe in the Nineteenth Century (Oxford Historical Monographs)

Literature and the Irish Famine 1845-1919 (Oxford Historical Monographs)

The Arabian Frontier of the British Raj: Merchants, Rulers, and the British in the Nineteenth-Century Gulf (Oxford Historical Monographs)

The English Parliaments of Henry VII 1485-1504 (Oxford Historical Monographs)

China Marches West: The Qing Conquest of Central Eurasia

Religion, Society, and Culture in Colombia: Antioquia and Medellin 1850-1930 (Oxford Historical Monographs)

Time and Work in England 1750-1830 (Oxford Historical Monographs)

Migration and Inequality in Germany 1870-1913 (Oxford Historical Monographs)

The British Peace Movement 1870-1914 (Oxford Historical Monographs)

The Oxford English Grammar

The Bismarck Myth: Weimar Germany and the Legacy of the Iron Chancellor (Oxford Historical Monographs)

The Legend of Miaoshan (Oxford Oriental Monographs)

Lordship and Literature: John Gower and the Politics of the Great Household (Oxford English Monographs)

Rethinking Catholicism in Reformation England (Oxford Historical Monographs)

Labour and the Countryside: The Politics of Rural Britain 1918-1939 (Oxford Historical Monographs)

Sovereignty and the Sword: Harrington, Hobbes, and Mixed Government in the English Civil Wars (Oxford Historical Monographs)

Orientalism in Louis XIV's France (Oxford Historical Monographs)

The Politics of the Poor: The East End of London 1885-1914 (Oxford Historical Monographs)

The World Before Domesday: The English Aristocracy 900-1066

Oxford Dictionary of English (mobipocket)

Historical Outlines of English Accidence

Lords of the Central Marches: English Aristocracy and Frontier Society, 1087-1265 (Oxford Historical Monographs)

The Quakers in English Society, 1655-1725 (Oxford Historical Monographs)

Milton and the Ineffable (Oxford English Monographs)

On Sympathy (Oxford English Monographs)

On Sympathy (Oxford English Monographs)

On Sympathy (Oxford English Monographs)

Conrad and History (Oxford English Monographs)

Stevie Smith and Authorship (Oxford English Monographs)

The Bridges of Medieval England: Transport and Society 400-1800 (Oxford Historical Monographs)

The Suffering Traveller and the Romantic Imagination (Oxford English Monographs)

Historians and Nationalism: East-Central Europe in the Nineteenth Century (Oxford Historical Monographs)

Literature and the Irish Famine 1845-1919 (Oxford Historical Monographs)

The Arabian Frontier of the British Raj: Merchants, Rulers, and the British in the Nineteenth-Century Gulf (Oxford Historical Monographs)

The English Parliaments of Henry VII 1485-1504 (Oxford Historical Monographs)

China Marches West: The Qing Conquest of Central Eurasia

Religion, Society, and Culture in Colombia: Antioquia and Medellin 1850-1930 (Oxford Historical Monographs)

Time and Work in England 1750-1830 (Oxford Historical Monographs)

Migration and Inequality in Germany 1870-1913 (Oxford Historical Monographs)

The British Peace Movement 1870-1914 (Oxford Historical Monographs)

The Oxford English Grammar

The Bismarck Myth: Weimar Germany and the Legacy of the Iron Chancellor (Oxford Historical Monographs)

The Legend of Miaoshan (Oxford Oriental Monographs)

Lordship and Literature: John Gower and the Politics of the Great Household (Oxford English Monographs)

Rethinking Catholicism in Reformation England (Oxford Historical Monographs)

Labour and the Countryside: The Politics of Rural Britain 1918-1939 (Oxford Historical Monographs)

Sovereignty and the Sword: Harrington, Hobbes, and Mixed Government in the English Civil Wars (Oxford Historical Monographs)

Orientalism in Louis XIV's France (Oxford Historical Monographs)

The Politics of the Poor: The East End of London 1885-1914 (Oxford Historical Monographs)

The World Before Domesday: The English Aristocracy 900-1066

Oxford Dictionary of English (mobipocket)

Historical Outlines of English Accidence

Recommend Documents