Wallingford: The Castle and the Town in Context 9781407314181, 9781407323060

Table of contents :
Front Cover
Title Page
Copyright
Table of Contents
List of Figures
Note to reader
FOREWORD
PREFACE AND ACKNOWLEDGEMENTS
1. INTRODUCTION
2. THE ARCHAEOLOGY OF WALLINGFORD CASTLE: A SUMMARY OF THE CURRENT STATE OF KNOWLEDGE
3. URBAN CASTLES IN THE MIDDLE AGES: WALLINGFORD IN CONTEXT
4. AN ENGLISH LEGACY: THE LIBERTY OF THE HONOUR OF WALLINGFORD
5. ‘MOST SECURELY FORTIFIED’: WALLINGFORD CASTLE 1071–1540
6. Two Tudor Surveys of Wallingford Castle from the Rotulus de Wallingfordia
7. LANDSCAPES, LEASES AND LAWSUITS: REAPING THE REWARDS OF POST-DEMOLITION DETECTION AT WALLINGFORD CASTLE
8. A TALE OF TWO TOWNS AND TWO CASTLES: NOTTINGHAM AND WALLINGFORD COMPARED
9. Recent Work on Oxford Castle: New Finds and New Interpretations
10. OXFORD CASTLE IN 1327: INQUISITION AS TO THE STATE OF CASTLES IN OXON AND BERKS 1 EDWARD III, TNA, E143/10/2
11. THE ORIGINS OF WALLINGFORD PRIORY: THE EVIDENCE OF A MID-TWELFTH-CENTURY TERRIER
12. WALLINGFORD PRIORY
13. RENTS AS INDICATORS OF THE MEDIEVAL SETTLEMENT PATTERN IN WALLINGFORD, BERKSHIRE
14. THE IMPACT OF THE CIVIL WAR ON WALLINGFORD, CASTLE AND TOWN
15. THE LAST STAND: WALLINGFORD CASTLE AND THE CIVIL WAR IMAGES AND INSIGHTS FROM ARCHAEOLOGY
BIBLIOGRAPHY
Index

Citation preview

BAR 621 2015 KEATS-ROHAN, CHRISTIE & ROFFE (Eds)

Wallingford: The Castle and the Town in Context Edited by

K. S. B. Keats-Rohan Neil Christie David Roffe

WALLINGFORD: THE CASTLE AND THE TOWN IN CONTEXT

B A R 621 Keats-Rohan cover.indd 1

BAR British Series 621 2015 06/08/2015 11:50:44

Wallingford: The Castle and the Town in Context Edited by

K. S. B. Keats-Rohan Neil Christie David Roffe

BAR British Series 621 2015

ISBN 9781407314181 paperback ISBN 9781407323060 e-format DOI https://doi.org/10.30861/9781407314181 A catalogue record for this book is available from the British Library

BAR

PUBLISHING

Wallingford c.1300 CLAPCOT

CLAPCOT King’s Mead

North Gate Former North Gate

N

ldit

ch’

Mill

CASTLE

St Peter in the West

me s

ve rT ha Ri

Former Road

Castle Street

College of St Nicholas

Mary Grace

High Street

St Mary-the-Less

Gre

at B

St Peter’s

ridg

e

Place

Flood Gates

Site of Stephen’s Castle

East Gate The

St Martin’s

We ir

Qu een ’s A

The Great Tower

All Saints

Bullcroft

rbo ur

St Nicholas

Church & Priory of Holy Trinity West Gate

‘Wa

St John-super-Aquam St Mary-the-More

Market

Kinecroft

St Michael’s

St Ruald’s South Gate Mill

St Leonard’s

Hospital of St John

St Lucian’s

River Tham

es

Portman Field (leading west to Portman Moor)

KEY Streets Possible Streets Saxon Rampart Water Town Gate Church Castle Wall Castle Gate

0

yd 100 m 100

Castle’s outer perimeter

Contents FOREWORD���������������������������������������������������������������������������������������������������������������������������������������������������������������������� xi Robert Liddiard PREFACE AND ACKNOWLEDGEMENTS������������������������������������������������������������������������������������������������������������������ xiii 1 INTRODUCTION������������������������������������������������������������������������������������������������������������������������������������������������������������ 1 David Roffe Wallingford��������������������������������������������������������������������������������������������������������������������������������������������������������������������� 2 The castle������������������������������������������������������������������������������������������������������������������������������������������������������������������������ 3 The fabric����������������������������������������������������������������������������������������������������������������������������������������������������������������������� 4 The castle as palace, fortress, and estate centre������������������������������������������������������������������������������������������������������������� 5 Castle and town�������������������������������������������������������������������������������������������������������������������������������������������������������������� 6 Swansong����������������������������������������������������������������������������������������������������������������������������������������������������������������������� 7 Conclusion���������������������������������������������������������������������������������������������������������������������������������������������������������������������� 8 2 THE ARCHAEOLOGY OF WALLINGFORD CASTLE: A SUMMARY OF THE CURRENT STATE OF KNOWLEDGE������������������������������������������������������������������������������������������������������������������������������ 9 Oliver Creighton and Neil Christie Introduction�������������������������������������������������������������������������������������������������������������������������������������������������������������������� 9 Early Interventions������������������������������������������������������������������������������������������������������������������������������������������������������� 11 The 1960s: the North Gate site������������������������������������������������������������������������������������������������������������������������������������� 11 The 1970s: the middle bailey��������������������������������������������������������������������������������������������������������������������������������������� 12 The 1990s: developer-funded interventions����������������������������������������������������������������������������������������������������������������� 14 The 2000s: the Wallingford Burh to Borough Research Project – surveys and excavations��������������������������������������� 14 Discussion: Wallingford Castle’s archaeology in context�������������������������������������������������������������������������������������������� 18 3 URBAN CASTLES IN THE MIDDLE AGES: WALLINGFORD IN CONTEXT������������������������������������������������������� 20 Michael Fradley Introduction������������������������������������������������������������������������������������������������������������������������������������������������������������������ 20 The urban castle: definitions���������������������������������������������������������������������������������������������������������������������������������������� 20 The European context of the urban castle�������������������������������������������������������������������������������������������������������������������� 22 The Norman Conquest narrative and the interpretation of the urban castle����������������������������������������������������������������� 23 Archaeological evidence of the urban castle form������������������������������������������������������������������������������������������������������� 25 Wallingford Castle: a typical urban castle?������������������������������������������������������������������������������������������������������������������ 26 4 AN ENGLISH LEGACY: THE LIBERTY OF THE HONOUR OF WALLINGFORD������������������������������������������������ 28 David Roffe Introduction������������������������������������������������������������������������������������������������������������������������������������������������������������������ 28 The honour and honorial jurisdiction��������������������������������������������������������������������������������������������������������������������������� 28 The liberty of the honour of Wallingford��������������������������������������������������������������������������������������������������������������������� 29 Origins�������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 29 The honour and defence of the middle Thames valley������������������������������������������������������������������������������������������������ 30 The honour and the borough���������������������������������������������������������������������������������������������������������������������������������������� 32 5 ‘MOST SECURELY FORTIFIED’: WALLINGFORD CASTLE 1071–1540�������������������������������������������������������������� 34 K. S. B. Keats-Rohan Introduction������������������������������������������������������������������������������������������������������������������������������������������������������������������ 34 In the beginning: foundations of a royal castle and honour 1066–1272���������������������������������������������������������������������� 36 ‘The bones of the kingdom’: evolution of a royal stronghold 1139–1540������������������������������������������������������������������� 55 Maintaining a royal stronghold: building accounts 1284–1438����������������������������������������������������������������������������������� 67 Summary and conclusions������������������������������������������������������������������������������������������������������������������������������������������ 105 Documents List 1������������������������������������������������������������������������������������������������������������������������������������������������������� 108

i

6 TWO Tudor Surveys of Wallingford Castle from the Rotulus de Wallingfordia���������� 116 John L. Lloyd† Introduction ��������������������������������������������������������������������������������������������������������������������������������������������������������������� 116 The certificate and the keep��������������������������������������������������������������������������������������������������������������������������������������� 120 The Surveyor’s Plan �������������������������������������������������������������������������������������������������������������������������������������������������� 124 The numbers in detail������������������������������������������������������������������������������������������������������������������������������������������������� 127 Geophysical survey and the true plan������������������������������������������������������������������������������������������������������������������������ 130 Conclusion and afterword, K. S. B. Keats-Rohan������������������������������������������������������������������������������������������������������ 133 Documents List 2������������������������������������������������������������������������������������������������������������������������������������������������������� 133 Appendix 1: A codicological overview of the Wallingford Roll, Bod. MS. Top. Berks. b. 41 K. S. B. Keats-Rohan135 Appendix 2: John L. Lloyd’s account of the motte as surveyed during the late 1990s���������������������������������������������� 137 7 LANDSCAPES, LEASES AND LAWSUITS: REAPING THE REWARDS OF POST-DEMOLITION DETECTION AT WALLINGFORD CASTLE��������������������������������������������������������������������������� 138 Judy Dewey Introduction���������������������������������������������������������������������������������������������������������������������������������������������������������������� 139 Outline post-medieval history������������������������������������������������������������������������������������������������������������������������������������ 140 Boundaries������������������������������������������������������������������������������������������������������������������������������������������������������������������ 146 Baileys and buildings������������������������������������������������������������������������������������������������������������������������������������������������� 151 Getting in and keeping out – changing gateways������������������������������������������������������������������������������������������������������� 155 The river, riverside and water controls ���������������������������������������������������������������������������������������������������������������������� 167 Summary and Conclusions����������������������������������������������������������������������������������������������������������������������������������������� 176 Documents List 3������������������������������������������������������������������������������������������������������������������������������������������������������� 189 8 A TALE OF TWO TOWNS AND TWO CASTLES: NOTTINGHAM AND WALLINGFORD COMPARED��������� 194 David Roffe Introduction���������������������������������������������������������������������������������������������������������������������������������������������������������������� 194 Nottingham����������������������������������������������������������������������������������������������������������������������������������������������������������������� 194 The borough and urbanization������������������������������������������������������������������������������������������������������������������������������������ 196 Wallingford����������������������������������������������������������������������������������������������������������������������������������������������������������������� 197 A type-site for the 9th- and 10th-century burh?��������������������������������������������������������������������������������������������������������� 198 9 Recent work on Oxford Castle: new finds and new interpretations���������������������������������� 200 Andrew Norton Introduction���������������������������������������������������������������������������������������������������������������������������������������������������������������� 200 The late Saxon town, 900–1071��������������������������������������������������������������������������������������������������������������������������������� 200 The Norman and later medieval castle, 1071–1577��������������������������������������������������������������������������������������������������� 203 Conclusions���������������������������������������������������������������������������������������������������������������������������������������������������������������� 208 10 OXFORD CASTLE IN 1327: INQUISITION AS TO THE STATE OF CASTLES IN OXON AND BERKS 1 EDWARD III, TNA, E143/10/2������������������������������������������������������������������������������������������������ 210 K. S. B. Keats-Rohan 11 THE ORIGINS OF WALLINGFORD PRIORY: THE EVIDENCE OF A MID-TWELFTH-CENTURY TERRIER ����������������������������������������������������������������������������������������������������������������������� 213 K. S. B. Keats-Rohan Acknowledgements���������������������������������������������������������������������������������������������������������������������������������������������������� 219 Appendix: Translation and text of the Wallingford Priory Terrier, BL Royal 5 D xi., fol. 153v������������������������������� 220 12 WALLINGFORD PRIORY���������������������������������������������������������������������������������������������������������������������������������������� 221 David E. Pedgley From foundation to dissolution���������������������������������������������������������������������������������������������������������������������������������� 221 The priory and its precinct����������������������������������������������������������������������������������������������������������������������������������������� 222 Conclusion������������������������������������������������������������������������������������������������������������������������������������������������������������������ 225 Appendix: Documentary references to the buildings of Wallingford Priory.������������������������������������������������������������� 226 ii

13 RENTS AS INDICATORS OF THE MEDIEVAL SETTLEMENT PATTERN IN WALLINGFORD, BERKSHIRE������������������������������������������������������������������������������������������������������������������������������� 227 David E. Pedgley 14 THE IMPACT OF THE CIVIL WAR ON WALLINGFORD, CASTLE AND TOWN��������������������������������������������� 232 Judy Dewey The historical context������������������������������������������������������������������������������������������������������������������������������������������������� 232 Wallingford at war������������������������������������������������������������������������������������������������������������������������������������������������������ 234 Wallingford the town in the 17th century������������������������������������������������������������������������������������������������������������������� 234 Garrisons and defences ���������������������������������������������������������������������������������������������������������������������������������������������� 236 Fear and manoeuvres in summer and autumn 1643��������������������������������������������������������������������������������������������������� 240 War and Wallingford in 1644–46������������������������������������������������������������������������������������������������������������������������������� 240 Fairfax and the siege of Wallingford�������������������������������������������������������������������������������������������������������������������������� 242 After the surrender: demolitions, damages and disputes�������������������������������������������������������������������������������������������� 243 Conclusion������������������������������������������������������������������������������������������������������������������������������������������������������������������ 247 Appendix: Extract from the printed Articles of Agreement for the handing over of Wallingford Castle, 25 July 1646 (BRO, W/Z5). Spelling and punctuation as printed. ������������������������������������������������������������������������������������������������� 248 15 THE LAST STAND: WALLINGFORD CASTLE AND THE CIVIL WAR. IMAGES AND INSIGHTS FROM ARCHAEOLOGY�������������������������������������������������������������������������������������������������� 250 Neil Christie Introduction���������������������������������������������������������������������������������������������������������������������������������������������������������������� 250 Civil War archaeology at Wallingford ����������������������������������������������������������������������������������������������������������������������� 250 Conclusions���������������������������������������������������������������������������������������������������������������������������������������������������������������� 257 BIBLIOGRAPHY������������������������������������������������������������������������������������������������������������������������������������������������������������ 259 INDEX����������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 271

iii

List of Figures Frontispiece Wallingford c. 1300 (image: David Dewey) Figure 2.1 Wallingford Castle’s earthworks (photo: Oliver Creighton).����������������������������������������������������������������������������� 9 Figure 2.2 Map showing archaeological excavations and other interventions on the castle site (illustration: Mike Rouillard).10 Figure 2.3 Brooks’ excavations on the North Gate (photo: Nicholas Brooks). ���������������������������������������������������������������� 11 Figure 2.4 View of Carr’s excavations within the middle bailey in 1972 (photo: Robert Carr).��������������������������������������� 12 Figure 2.5 The curtain wall on the south side of the middle bailey (photo: Oliver Creighton).���������������������������������������� 13 Figure 2.6 Ground-penetrating radar survey of the inner bailey (photo: Oliver Creighton)��������������������������������������������� 14 Figure 2.7 View of Trench 1 (2008) (photo: Matt Edgeworth).����������������������������������������������������������������������������������������� 15 Figure 2.8 View of Trench 4 (2009) (photo: Oliver Creighton).��������������������������������������������������������������������������������������� 16 Figure 2.9 View of Trench 8 (2010) (photo: Oliver Creighton).��������������������������������������������������������������������������������������� 16 Figure 2.10 View of Trench 10 (2011) (photo: Neil Christie).������������������������������������������������������������������������������������������ 17 Figure 3.1 Norman urban castles in England (image © the Wallingford Burh to Borough Research Project)����������������� 21 Figure 3.2 Norman urban castles and spatial relationships to their townscapes (image © the Wallingford Burh to Borough Research Project)������������������������������������������������������������������������������������������������������������������������������������������������ 24 Figure 4.1 The honour of Wallingford c. 1129 (© Katharine Keats-Rohan, drawn by June Strong)�������������������������������� 31 Figure 4.2 The territories of the burhs of Wallingford and Sashes at the time of the Burghal Hidage (© David Roffe, drawn by June strong)�������������������������������������������������������������������������������������������������������������������������������������������������������� 32 Figure 5.1 Composite overview of geophysical survey on the Wallingford Castle site (TWHAS,Burh to Borough Project), incorporating earthwork survey by Michael Fradley (Burh to Borough Project) (Google Earth-based image © 2015 Infoterra Ltd & Bluesky, compiled by Gerard Latham)����������������������������������������������������������������������������� 35 Figure 5.2 Doorway in a surviving portion of the Priests’ Lodging in the middle bailey at Wallingford (photo: Judy Dewey).������������������������������������������������������������������������������������������������������������������������������������������������������������������������������ 36 Figure 5.3 Castles of Conquest, after History of the King’s Works, © Queen’s Printer and Controller of HMSO 1963 (redrawn by Debbie Miles-Williams).������������������������������������������������������������������������������������������������������������������������������� 37 Figure 5.4 Viewshed from Wallingford Castle site (image © Wallingford Burh to Borough Research Project).������������� 39 Figure 5.5 Norman gatehouse of Rougemont Castle, Exeter (photo: Oliver Creighton).������������������������������������������������� 40 Figure 5.6 Late Saxon St George’s Tower, Oxford (photo: Katharine Keats-Rohan). ����������������������������������������������������� 40 Figure 5.7 A bird’s-eye view of Windsor Castle in 1658, by Wenceslas Hollar, shown before Hugh May’s reconstruction of the Upper Ward (public domain).���������������������������������������������������������������������������������������������������������� 41 Figure 5.8 Stephen’s anti-castle at Crowmarsh (an oval ringwork) under excavation (photo: Judy Dewey).���������������� 42 Figure 5.9 The Great Keep at Rochester, with first-floor entrance via the forebuilding (photo: Katharine KeatsRohan).................................................................................................................................................................................43 Figure 5.10 Dover Castle, the Keep and the outer curtain wall (photo: Judy Dewey).����������������������������������������������������� 43 Figure 5.11 Artist’s reconstruction of the Great Keep and forebuilding at Wallingford Castle, looking south. (Drawing by Anthony Wilder.)������������������������������������������������������������������������������������������������������������������������������������������ 44 Figure 5.12 The walls of what had previously been seen as an enmotted shell-keep at Berkeley Castle, but now thought originally to have been a small inner ward (photo: Katharine Keats-Rohan).����������������������������������������������������� 44 Figure 5.13 Enmotted keep of Farnham Castle from the outer bailey. A tower built at ground level and subsequently enmotted was razed to current motte-top level on the orders of Henry II after 1155 (photo: Katharine Keats-Rohan).�� 45 Figure 5.14 Capped off outlet of the well on the north side of the motte at Wallingford (photo: Katharine KeatsRohan).������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 46 Figure 5.15 Sallyport at Knaresborough Castle (photo: Judy Dewey).���������������������������������������������������������������������������� 47 Figure 5.16 The view of the Tower of London from The Shard to the south west (photo: Rafa Esteve 2013, Public Domain).���������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 48 Figure 5.17 Seal of Matilda of Wallingford, attached to a 12th century document of 1122 x 1147, regranting the churches of Ogbourne St Andrew and Ogbourne St George, Wilts, to the abbey of Bec, originally granted by Miles Crispin. (Image reproduced by permission of the Dean and Canons of Windsor.)����������������������������������������������������������� 48 Figure 5.18 Oxford Castle in 2015, after conversion from HMP Oxford to Hotel Malmaison; St George’s Tower is in the background, with the motte slope in the foreground (photo: Katharine Keats-Rohan).�������������������������������������� 52 Figure 5.19 Norman chapel at Durham Castle (photo: Judy Dewey).������������������������������������������������������������������������������� 53 Figure 5.20 Lion head detail from a stone doorway in the Priests’ Lodging (photo: Judy Dewey). Male head capital re-used for a now-blocked garden entrance doorway from Castle Lane into the middle bailey. Two of the least eroded of the few surviving stone carved figures still on site (photo: Katharine Keats-Rohan).�������������������������������������� 54 Figure 5.21 Grant of the castle of Wallingford in dower to his wife Isabella by Edward II, 26 April 1317. (Reproduced by permission of the UK National Archives, E40/164.)������������������������������������������������������������������������������ 61 Figure 5.22 Account of John de Stonor for the guard of Wallingford Castle on behalf of Queen Isabella, 7–16 October 1326. (Reproduced by permission of the UK National Archives, E101/17/21.��������������������������������������������������� 63

iv

Figure 5.23 Royal castles in the reign of Henry VI (1422–61) (image redrawn by Debbie Miles-Williams after History of the King’s Works, 26, p. 239; © Queen’s Printer and Controller of HMSO 1963.)����������������������������������������� 66 Figure 5.24 Survey of Wallingford Castle, 1327. (Reproduced by permission of the UK National Archives, E143/10/2.)������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 68 Figure 5.25 Main gatehouse and barbican, with double portcullis, at Warwick Castle (photo: Judy Dewey).����������������� 71 Figure 5.26 Extract from a map of 1835 showing the still surviving ‘floodgate’ system at the West Gate at the top of modern Croft Road, then still named Lock Lane (image courtesy of Wallingford Museum).�������������������������������������� 73 Figure 5.27 The sword of Guy of Warwick at Warwick Castle (photo: Judy Dewey).����������������������������������������������������� 74 Figure 5.28 14th century verse saying that the sword of Sir Gaweyn is displayed in the Great Tower at Wallingford Castle. (Reproduced by permission of The Bodleian Library, University of Oxford, Bod. MS Auct. F. 5. 23, fol. 110v.)���������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 75 Figure 5.29 Graphite sketch by George Cumberland (1799–1848), formerly in an album, showing the Priests’ Lodgings above the wall of the middle bailey at Wallingford Castle (image © The Trustees of the British Museum)��������� 77 Figure 5.30 The stonework surviving in the NE corner of the inner bailey at Wallingford Castle, known as the Queen’s Tower (photo: Oliver Creighton).������������������������������������������������������������������������������������������������������������������������ 77 Figure 5.31 Gate of Chepstow Castle, showing wicket and lattice work (photo: Andy Dingley 2011, Creative Commons Attribution –Share Alike 3.0 Unported Licence).��������������������������������������������������������������������������������������������� 78 Figure 5.32 Constable’s Gate at Dover Castle (photo: Judy Dewey).������������������������������������������������������������������������������� 79 Figure 5.33 Drawbridge over water with stone-built bridge, castle at Sirmione, Lake Garda, Italy (photo: Judy Dewey).�� 80 Figure 5.34 Wooden hoardings at Caerphily Castle, viewed from inside and outside the castle wall (photo: Oliver Creighton).������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 80 Figure 5.35 Artist’s reconstruction of possible gatehouse arrangement on the south-east side of Wallingford Castle c. 1300. (Drawing by June Strong.)����������������������������������������������������������������������������������������������������������������������������������� 81 Figure 5.36 Great staircase at Wardour Castle, Wilts (photo: Oliver Creighton).������������������������������������������������������������� 81 Figure 5.37 Painted wall in a room in Castelvecchio, Verona (photo: Judy Dewey).������������������������������������������������������� 85 Figure 5.38 Great Hall at Berkeley Castle, minstrels gallery over the entrance formed from painted wooden panels (not in original location); four-light window with oillets above (photo: Katharine Keats-Rohan).���������������������������������� 86 Figure 5.39 Great Hall at Berkeley Castle, facing the dais end; on the right a door gives access to the great staircase leading to the great chamber (photo: Katharine Keats-Rohan).���������������������������������������������������������������������������������������� 87 Figure 5.40 Decorative capital of c.1230 (private collection)������������������������������������������������������������������������������������������� 88 Figure 5.41 Limestone relief carving of c. 1120–30, 33cm (13 inches) across, found in the later 19th century within the original confines of Wallingford Castle according to a note by J. K. Hedges shown after his death to C. E. Keysey (Keysey, 1907). The subject is the constellation Aquarius, the water-carrier, indicated by the position of its component stars, current since the 5th century BC. The relief may have formed part of a frieze or series representing the signs of the Zodiac, perhaps built into the wall of an important interior such as the castle’s hall. (Image from the Hedges Collection © Reading Museum, Reading Borough Council).������������������������������������������������������������������������������������ 88 Figure 5.42 Fine stone from the castle was widely reused by local people under the terms of the demolition order. These two examples, now in Wallingford Museum, are both 12th-century work attesting the lavish and fashionable state of castle décor at that time. The first is a beakhead stone, the most magnificent surviving local examples of which are in St Mary’s Church, Iffley (Oxon). The nailhead stone features a pyramidal motif very popular in the 12th century (photo: Katharine Keats-Rohan) ������������������������������������������������������������������������������������������������������������������ 89 Figure 5.43 Well chamber just outside the Great Hall in the keep at Dover, easily accessible from the forebuilding entrance into the keep, as seems to have been the case also at Wallingford (photo: Katharine Keats-Rohan).���������������� 90 Figure 5.44 Stone rosette boss from Hedges’ Wallingford Castle Museum, © Reading Museum (Reading Borough Council).����������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 91 Figure 5.45 Suggested layout of the inner bailey buildings in the later 14th century, overlaid onto the 1804 map of the inner bailey and guided by the geophysical survey (illustration Debbie Miles-Williams).����������������������������������������� 91 Figure 5.46 Hand mill in the kitchen at Dover Castle (photo: Judy Dewey)��������������������������������������������������������������������� 92 Figure 5.47 Quern from Hedges’ Wallingford Castle Museum. (Image © Reading Museum, Reading Borough Council)������������������������������������������������������������������������������������������������������������������������������������������������������������������������������ 92 Figure 5.48 Cut-away reconstruction drawing of Rochester Castle keep (completed c. 1140), showing access to a basement storage area under the bridge-pit of the entrance to the forebuilding (IC087_002 © Historic England).��������� 93 Figure 5.49 Evolution of hoardings (cf. Figure 34) to decorative gallery with solid machicolations seen at Berkeley Castle (photo: Katharine Keats-Rohan).���������������������������������������������������������������������������������������������������������������������������� 94 Figure 5.50 Part of the accounts of John de Clinton from 1307–9. (Reproduced by permission of the UK National Archives, TNA, E 101/489/30.)����������������������������������������������������������������������������������������������������������������������������������������� 95 Figure 5.51 Inventory of arms in the Great Tower at Wallingford by constable Miles de Beauchamp, after 25 January 1327. (Image reproduced by permission of the UK National Archives, TNA, E 101/17/28.)����������������������������� 96 Figure 5.52 Balista at Warwick Castle (photo: Katharine Keats-Rohan).������������������������������������������������������������������������� 97

v

Figure 5.53 Trebuchet at Warwick Castle (photo: Judy Dewey).�������������������������������������������������������������������������������������� 97 Figure 5.54 Places in the honour of Wallingford most frequently mentioned as source for building materials (drawn by June Strong).����������������������������������������������������������������������������������������������������������������������������������������������������������������� 99 Figure 5.55 Looking through a window arch in the Priests’ Lodging to the orchard beyond (photo: Katharine Keats-Rohan).������������������������������������������������������������������������������������������������������������������������������������������������������������������ 101 Figure 5.56 Entrance to the top of the keep at Farnham Castle; the roof visible through the archway at the top of what is now a shell-keep covers the truncated remains of a square stone tower that originally rose through the centre of the motte and was designed from the start to be enmotted (Thompson, 1960) (photo: Katharine KeatsRohan). ���������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 104 Figure 5.57 Diagram of the construction of the motte at Farnham after Thompson 1960 (redrawn by Debbie MilesWilliams).������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 104 Figure 5.58 James Blackstone’s sketch of part of an underground passage on the south side of the motte as shown in Hedges’ History of Wallingford (1881), i, 151.������������������������������������������������������������������������������������������������������������ 104 Figure 6.1 The Surveyor’s Plan, from the Wallingford Roll, Oxford, Bodleian MS Top. Berks b. 41, fol. 1. (Reproduced by permission of The Bodleian Library, University of Oxford.)�������������������������������������������������������������������������������������������� 117 Figure 6.2 The Certificate of 1555 from the Wallingford Roll, Oxford, Bodleian MS Top. Berks b. 41, fol. 25. (Reproduced by permission of The Bodleian Library, University of Oxford).������������������������������������������������������������������ 118 Figure 6.3 Muniments chest of Wallingford Corporation, late 15th-century, now in Wallingford Museum (photo: Judy Dewey).������������������������������������������������������������������������������������������������������������������������������������������������������������������� 119 Figure 6.4 Table showing possible joist values (table: Gerard Latham)������������������������������������������������������������������������� 124 Figure 6.5 The shell keep at Restormel Castle, Cornwall (image: Public Domain).������������������������������������������������������ 125 Figure 6.6 Suggested floor-plan of first and second floors of the keep at Wallingford Castle (image: Debbie MilesWilliams)�������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 125 Figure 6.7 The Surveyor’s Plan, showing the figures given on the original; lettering has been added to aid the discussion in the text. Note that the scale of the original is internally inconsistent, making exact replication of the drawing virtually impossible (image: Katharine Keats-Rohan and June Strong).����������������������������������������������������������� 126 Figure 6.8 Possible locations of the curtain walls on the western side of the inner bailey (image: author).������������������� 128 Figure 6.9 Possible angles of the modified NW tower (image: author).������������������������������������������������������������������������ 128 Figure 6.10 The Plan redrawn as a model pentagon (image: June Strong).�������������������������������������������������������������������� 129 Figure 6.11 Aerial view of the motte (Crown Copyright 1988). ������������������������������������������������������������������������������������� 130 Figure 6.12 Detail of an estate map of 1804, showing the moat around the SW tower, and the south of the motte (courtesy of The National Archives, MPEE 1/10)����������������������������������������������������������������������������������������������������������� 131 Figure 6.13 Floodwaters round the southern base of the motte in 2013, highlighting the location of the SW tower as seen in the 1804 map (photo: Katharine Keats-Rohan)����������������������������������������������������������������������������������������������� 131 Figure 6.14 The redrawn Plan (image: Debbie Miles-Williams).������������������������������������������������������������������������������������ 132 Figure 6.15 Diagram of the motte path as it was in 2000 based on survey by J. L. Lloyd (image: author).������������������� 136 Figure 7.1 Wallingford Castle c. 1300 (map: David Dewey) ����������������������������������������������������������������������������������������� 139 Figure 7.2 Detail of the map drawn up for the sale of the castle in 1817 (photo: Katharine Keats-Rohan, reproduced courtesy of The Bodleian Library, University of Oxford, MS Top. Berks. b. 10, fol. 17v-18r).������������������������������������� 141 Figure 7.3 ‘Wallingford Castle’, the Victorian mansion built for J. K. Hedges and his wife Sophie (Wallingford Museum). 142 Figure 7.4 Wallingford Castle from the west, 6th December 1965 (Copyright reserved Cambridge University Collection of Aerial Photography).���������������������������������������������������������������������������������������������������������������������������������� 143 Figure 7.5 John Kirby Hedges. Portrait by James Hayllar (1829–1920), a renowned Wallingford artist. (Image courtesy of Wallingford Town Council.)������������������������������������������������������������������������������������������������������������������������� 144 Figure 7.6 St Nicholas, High Street, Wallingford, 2015 (photo: Judy Dewey).�������������������������������������������������������������� 144 Figure 7.7 Thameside Mansion, from the riverside 2015 (photo: Judy Dewey).������������������������������������������������������������ 145 Figure 7.8 1898 Ordnance Survey map with borough boundary enhanced.�������������������������������������������������������������������� 146 Figure 7.9 Bear Lane 2014. This wall marked the Wallingford Borough boundary; the lane itself lay within the castle (photo: Judy Dewey). �������������������������������������������������������������������������������������������������������������������������������������������� 147 Figure 7.10 Gardens at the back of High Street step up towards the town’s boundary wall; the ruins of the castle’s middle wall can just be seen on the right in the distance. The ground between the two walls formed the southern outer bailey of the castle, through which the third wall and moat once ran (photo: Judy Dewey).������������������������������� 147 Figure 7.11 Combined resistivity results for Castle Gardens showing 20m grids (image: Gerard Latham). ����������������� 149 Figure 7.12 Remains of the Priests’ Lodging of the College of St Nicholas in the middle bailey, looking east. (photo: Judy Dewey).������������������������������������������������������������������������������������������������������������������������������������������������������� 152 Figure 7.13 Detail of part of the wall of the Priests’ Lodging in the middle bailey (photo: Judy Dewey).�������������������� 153 Figure 7.14 Thomas Athow’s ‘View of the Remains of Wallingford House Berkshire’, a watercolour painted c. 1809 (image © Ashmolean Museum, University of Oxford)������������������������������������������������������������������������������������������ 154 Figure 7.15 Castle entrances 18th and 19th centuries (illustration by David Dewey).��������������������������������������������������� 156

vi

Figure 7.16 Entrance gates to a scenic carriageway leading round the motte and inner bailey to the Victorian mansion (photo: Judy Dewey)����������������������������������������������������������������������������������������������������������������������������������������� 156 Figure 7.17 Looking south-east towards the middle bailey. The brick-walled Castle Lane was diverted into the old moat by John Kirby Hedges. The gate in the wall once opened on to steps into the gardens of his Victorian mansion (photo: Judy Dewey).������������������������������������������������������������������������������������������������������������������������������������������������������� 157 Figure 7.18 Detail of churchyard, cottages, gate and ‘tower’ on Castle Street. From map of c. 1715 (see Figure 7.46). (Image reproduced by kind permission of the Governing Body of Christ Church, Oxford.)������������������������������� 158 Figure 7.19 Bird’s Eye view of the castle from the north. The ‘barbican’ can be clearly seen projecting into the western end of the inner moat, while the sites of three probable towers are prominent on the outer rampart (photo: David Dewey).������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 159 Figure 7.20 Aerial view looking south. Thameside Mansion is in the foreground, with Castle Farm House behind it (slightly left) and Castle Lane House, with its expanse of lawn, to the south, beyond the courtyard (photo: David Dewey). ����������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 160 Figure 7.21 Fragments of the fallen eastern curtain wall, silhouetted against the flooded river in 2014.The bridge is visible in the background (photo: Judy Dewey).���������������������������������������������������������������������������������������������������������� 160 Figure 7.22 View of Blackstone’s track from the High Street entrance; the track continues through the gates to pass under the bridge he built to carry Castle Lane (photo: Judy Dewey).����������������������������������������������������������������������������� 161 Figure 7.23 James Blackstone’s bridge, which still carries Castle Lane over his former cart track. Note the high banks on either side where the track has cut north through the former medieval middle rampart (photo: Judy Dewey).���������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 161 Figure 7.24 Detail of Thomas Renda’s gates, avenue of trees and house. (From a map of c. 1715 in CCA, MS Estates Book 5, image reproduced by kind permission of the Governing Body of Christ Church, Oxford.)����������������� 162 Figure 7.25 Composite of castle features from maps of 1804, 1810 and 1812 overlaid on an aerial photograph of 1973 (see 7.43). (Copyright reserved Cambridge University Collection of Aerial Photography.)���������������������������������� 163 Figure 7.26 Looking south-west along the line of the middle moat below the middle wall (photo: Judy Dewey)��������� 164 Figure 7.27 Castle Lane looking north. Munt’s Mill is on the right. Notice how the lane gradually slopes upwards (photo: Judy Dewey). ������������������������������������������������������������������������������������������������������������������������������������������������������ 165 Figure 7.28 Map drawn by Edward Driver, 1810 (see Figure 7.53).������������������������������������������������������������������������������� 165 Figure 7.29 Resistivity survey of Munt’s Mill made in the 1990s (image: Gerard Latham). ����������������������������������������� 166 Figure 7.30 The widened west end of the bridge looking towards the High Street, 2015 (photo: Judy Dewey).����������� 167 Figure 7.31 A widened arch of Wallingford Bridge, viewed from the north side (photo: Judy Dewey).������������������������ 167 Figure 7.32 Detail from the c. 1835 Wallingford map showing the new gas works and water arrangements (see Figure 7.56).���������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 168 Figure 7.33 Riverside footpath round the old Oxford Canal Company wharf building (now a house), 2015 (photo: Judy Dewey).������������������������������������������������������������������������������������������������������������������������������������������������������������������� 169 Figure 7.34 The inner and middle moats appear to converge against a dam that forms part of the eastern rampart (photo: Judy Dewey).������������������������������������������������������������������������������������������������������������������������������������������������������� 169 Figure 7.35 This pond near the north-west corner of Queen’s Arbour was landscaped by the Victorians but may have related to the medieval water control system; it lies below the dam formed by the eastern castle rampart, but above the level of Queen’s Arbour (photo: Judy Dewey).����������������������������������������������������������������������������������������������� 170 Figure 7.36 Detail of 1817 castle sale map showing North Eyte (marked as 5) and other castle holdings near Benson (photo: Katharine Keats-Rohan, reproduced courtesy of The Bodleian Library, University of Oxford, MS Top. Berks. b. 10, fol. 18r).���������������������������������������������������������������������������������������������������������������������������������������������� 170 Figure 7.37 Compilation of resistivity results on south and east sides of the castle (image: Gerard Latham)���������������� 171 Figure 7.38 Excavation of the apsidal building on Queen’s Arbour, nearing completion in 2010 (photo: Judy Dewey).�� 173 Figure 7.39 Substantial stones lining the ditch along the western edge of Queen’s Arbour (photo: Nigel Ruchpaul).�� 174 Figure 7.40 Modern survival of the track from Castle Lane to the river. The car is parked in front of Munt’s Mill, close to where the outer gateway of the castle would once have stood (photo: Judy Dewey).��������������������������������������� 175 Figure 7.41 Castle Gardens, with resistivity work in progress (photo: Judy Dewey),���������������������������������������������������� 176 Figure 7.42 Bird’s eye view of the castle in 2015, looking north from the spire of St Peter’s church (image: David Dewey).���������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 178 Figure 7.42a. Resistivity surveys of Wallingford Castle super-imposed on LiDAR images. ����������������������������������������� 180 Figure 7.43 Wallingford Castle from the south, 17th October 1973. (Copyright reserved Cambridge University Collection of Aerial Photography.) ��������������������������������������������������������������������������������������������������������������������������������� 181 Figure 7.44 1548 map. Earliest known image of the castle’s inner bailey. (Reproduced by permission of The Bodleian Library, University of Oxford.) ����������������������������������������������������������������������������������������������������������������������� 181 Figure 7.45 Richard Gough’s plan of Wallingford Castle, 1768. (For Gough’s extended notes see 3.12.) (Photo: Katharine Keats-Rohan, reproduced by permission of The Bodleian Library, University of Oxford from Gough Maps I, vol. 1, fol. 43v.) �������������������������������������������������������������������������������������������������������������������������������������������������� 181

vii

Figure 7.46 The ‘Renda’ map. An early18th-century plan of the estate at Wallingford, probably drawn up by Thomas Renda as part of a dispute with Christ Church over boundaries c. 1715–19. (CCA, Maps Wallingford 1: image reproduced by kind permission of the Governing Body of Christ Church, Oxford.) ������������������������������������������ 182 Figure 7.47 Detail of Wallingford Castle, taken from ‘A Survey and Plan of the Estates of Robert Hucks Esq. of Aldenham, Herts, 1786’, showing ‘Land in the occupation of Robert Leaver. Part of a farm at Wallingford, Berks.’. (Redrawn by Berenice Pedgley from Hertfordshire Record Office, D/EAm P1.)����������������������������������������������������������� 182 Figure 7.48 Detail from the 1804 Crown Estate Map, drawn by John Fordyce. (Courtesy of The National Archives, MPEE 1/10.) ������������������������������������������������������������������������������������������������������������������������������������������������������������������ 183 Figure 7.49 Sketch of the south side of Wallingford Bridge by Joseph Farington RA, c. 1792 (Wallingford Museum). ��� 184 Figure 7.50 The north side of Wallingford Bridge drawn by Thomas Hearne F.S.A. c.1803, engraved by William Byrne F.S.A. (Wallingford Museum).������������������������������������������������������������������������������������������������������������������������������ 184 Figure 7.51 Print of bridge re-building works from the south side by William Havell, entitled ‘Wallingford Castle, taken in 1810 while the bridge was repairing’ (Wallingford Museum). ������������������������������������������������������������������������� 185 Figure 7.52 Water-colour by George Shepherd, 1815, entitled ‘The Town, and Bridge, Wallingford, Berkshire’ (image © Ashmolean Museum, University of Oxford). ������������������������������������������������������������������������������������������������������������������� 185 Figure 7.53 Sketch plan drawn in 1810 by Edward Driver of part of the Christ Church property at Wallingford Castle, showing lines of banks and ditches (CCA, MS Estates 5 document, fol. 143, by kind permission of the Governing Body of Christ Church, Oxford). ����������������������������������������������������������������������������������������������������������������� 186 Figure 7.54 1812 ‘Plan of estates situate in Wallingford Castle in the County of Berks, belonging to the Dean & Chapter of Christ Church, Oxon.’, certified by James Blackstone in 1813, who had ordered it to be made (CCA, Maps Wallingford 2; image reproduced by kind permission of the Governing Body of Christ Church, Oxford).��������� 186 Figure 7.55 Crown plan of 1817 showing ‘A Freehold Estate at Wallingford in the County of Berks belonging to His Majesty’ (photo: Katharine Keats-Rohan, reproduced courtesy of The Bodleian Library, University of Oxford, MS Top. Berks. b. 10, fol. 17v-18r.��������������������������������������������������������������������������������������������������������������������������������� 187 Figure 7.56 Map of Wallingford c. 1835. The origin of this finely executed hand-drawn map is obscure but it is the earliest known map of the town and is held by Wallingford Museum.��������������������������������������������������������������������������� 188 Figure 7.57 The riverside restaurant 2015 (photo: Judy Dewey). ���������������������������������������������������������������������������������� 188 Figure 8.1 Medieval Nottingham (© Nottingham City Museums).�������������������������������������������������������������������������������� 195 Figure 8.2 The archaeology of medieval Nottingham (© Nottingham City Museums).������������������������������������������������� 196 Figure 9.1 Areas of excavation at Oxford Castle showing features of late Saxon date. The author’s conjectural reconstruction of the street plan of the western part of the burh is shown in inset and overlay (Image © Oxford Archaeology).������������������������������������������������������������������������������������������������������������������������������������������������������������������� 201 Figure 9.2 Late Anglo-Saxon halls C2 and C3 (Image © Oxford Archaeology)����������������������������������������������������������� 202 Figure 9.3 Revealed foundations of the ten-sided keep (Image © Oxford Archaeology)����������������������������������������������� 204 Figure 9.4 Aubrey’s 1663 sketch of the keep and its crack (Image © Oxford Archaeology)������������������������������������������ 205 Figure 9.5 The revealed castle structures (Image © Oxford Archaeology)��������������������������������������������������������������������� 206 Figure 9.6 The Norman ramparts to the south of C-Wing, looking north (Image © Oxford Archaeology)�������������������� 207 Figure 10.1 Survey of Wallingford Castle, 1327. (Reproduced by permission of the UK National Archives, E143/10/2.)210 Figure 11.1 Mid-12th century terrier of Wallingford Priory (© The British Library Board, Royal 5 F X, fol. 153v).��� 214 Figure 12.1 Suggested layout of Wallingford Priory based on geophysical survey (image: author)������������������������������ 223 Figure 12.2 Results of geophysical survey in 2011 of the southern part of the Bullcroft, Wallingford (image: Gerard Latham).��������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 224 Figure 12.3 Artist’s impression of Wallingford Priory (drawing: June Strong).�������������������������������������������������������������� 224 Figure 13.1 Number of properties by street liable for quit rents in a selection of available years, the total rent paid, and the average rent per property (image: author).���������������������������������������������������������������������������������������������������������� 227 Figure 13.2 Distribution of properties paying quit rents in the 19th century (image: author)..�������������������������������������� 228 Figure 13.3 Number of properties by street liable for borough lease rents in a selection of available years.���������������� 230 Figure 13.4 Distribution of properties paying borough lease rents in the 19th century (image: author).������������������������ 230 Figure 14.1 Map of royalist winter quarters between Reading and Banbury 1642–3 (image © Judy Dewey)��������������� 233 Figure 14.2 The George Inn (photo: Judy Dewey).��������������������������������������������������������������������������������������������������������� 235 Figure 14. 3 William Cooke’s malthouse (photo: Judy Dewey).������������������������������������������������������������������������������������� 235 Figure 14.4 Colonel Thomas Blagge, Governor of Wallingford Castle throughout the Civil War (image: public domain).��������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 236 Figure 14.5 The outer works of Wallingford Castle. Civil War gun emplacements have been excavated outside the medieval ramparts in the area to the bottom right of the picture and on the ‘barbican’ - the prominent D-shaped feature, middle right (image © TWHAS).����������������������������������������������������������������������������������������������������������������������� 237 Figure 14.6 Clay base of gun platform excavated 2008 (photo: Judy Dewey).�������������������������������������������������������������� 238 Figure 14.7 St John’s Green. This was the site of the medieval Hospital of St John, the buildings of which were pulled down during the Civil War to give a clear view of the southern approach to the town (photo: Judy Dewey). ��� 239

viii

Figure 14.8 Remaining Royalist strongholds 1646 (photo: Judy Dewey).��������������������������������������������������������������������� 241 Figure 14.9 Re-enactment of the siege by the English Civil War Society, June 2015 (photo: Judy Dewey).���������������� 242 Figure 14.10 Title page of the printed arrangements for the surrender of Wallingford Castle (photo: Judy Dewey). �� 243 Figure 14.11 St Mary’s church tower, constructed with re-used castle stone in 1653 (photo: Judy Dewey).���������������� 244 Figure 14.12 Stone on the south-western buttress of St Mary’s tower (photo: Judy Dewey).��������������������������������������� 244 Figure 14.13 St Leonard’s church. The tower was added in the 19th century during refurbishment of the church. Damage in the Civil War prevented use of the church for worship for over 50 years (photo: Judy Dewey).����������������� 245 Figure 14.14 St Peter’s church, rebuilt in the 18th century on the site of the medieval church destroyed during the Civil War. The churchyard, however, remained in use throughout (photo: Judy Dewey).���������������������������������������������� 246 Figure 14.15 Wallingford Town Hall (photo: Judy Dewey).������������������������������������������������������������������������������������������� 247 Figure 15.1 Composite resistivity plots for all areas surveyed in 2008–09 in the north-west zone of Wallingford (castle inner bailey, Castle Meadows, Castle Gardens and Wallingford School playing fields) imposed over revised castle earthwork survey by Michael Fradley. (Image © the Wallingford Burh to Borough Research Project)��������������� 251 Figure 15.2 Map of the castle zone at Wallingford, identifying locations of all archaeological interventions, including the Burh to Borough Research Project trenches 1, 4, 7, 8, 10 (Image © the Wallingford Burh to Borough Research Project).������������������������������������������������������������������������������������������������������������������������������������������������������������ 252 Figure 15.3 Trench 1 – the north-western angle of the Civil War bastion feature (Image © the Wallingford Burh to Borough Research Project).��������������������������������������������������������������������������������������������������������������������������������������������� 253 Figure 15.4 Plan of Trench 1 showing plan of excavated bastion platform (Image © the Wallingford Burh to Borough Research Project). �������������������������������������������������������������������������������������������������������������������������������������������� 253 Figure 15.5 View looking east of Trench 10 after excavation, with rampart and related construction levels to left, and rubble spreads to the right (Image © the Wallingford Burh to Borough Research Project).������������������������������������ 255 Figure 15.6 Trench 10 - photograph of the exposed make-up to the Civil War platform, consisting of the angled lines of marl-clay interleaved with sandy gravel deposits extending to the medieval rampart, a trace of which is exposed in the far right of the trench (Image © the Wallingford Burh to Borough Research Project). ������������������������� 255 Figure 15.7 View from the west of the north face of the elevated ‘barbican’ zone, explored in Trench 10 in 2010 (Image © the Wallingford Burh to Borough Research Project) �������������������������������������������������������������������������������������� 256

ix

Note to reader References to texts in the three lists of Documents (Chapters 5–7) are printed in bold. Abbreviations and references can be found in the Bibliography.

x

FOREWORD Robert Liddiard Castles are evocative monuments and perhaps more than any other building capture the ideals of the Middle Ages. As icons of lordship, refuges in war, residences, judicial centres and seats of government, castles were a tangible expression of the builder’s immense landed resources, command of men and materials and, above all else, membership of an aristocratic class that saw its raison d’être as the leadership of society. As a group, the royal castles were an expression of this ethos par excellence, standing both for the dignity of the realm as well as the prestige of the monarch. The academic study of castles has a long pedigree and such has been the quantity of material written on the subject, especially in the past 20 years, that one could be forgiven for thinking that the capacity for breaking new ground was much diminished. This volume points up that this is simply not the case and shows clearly how primary research continues not just to inform, but also to change, our perception of medieval castles. Scholars have long held up the virtues of a ‘holistic’ approach to the study of castles, but only rarely can it be said that aspiration is translated into reality. The programme of research undertaken at Wallingford, the results of which can be seen in this and two complementary volumes (Keats-Rohan and Roffe 2009 at the outset of the project; Christie and Creighton 2013 as the major archaeological end result), has not only transformed our understanding of this particular castle and its environs, but also served to demonstrate what holistic means in practice. The collection of papers brought together here has as its focus Wallingford Castle itself, a fortification to which posterity has not always been kind. The slighting of the defences following the Civil War, together with an attendant borough that achieved only modest fortunes in subsequent centuries, ensured that Wallingford has been a castle that has often been overlooked in mainstream narratives. As the classic imprint of the Normans on the English urban landscape, the earthworks that sit within one corner of the town defences are familiar enough but, as this volume demonstrates, they belie an importance that cannot easily be appreciated from the mounds of earth that survive today. As the painstaking research related in the contributions here has shown beyond any doubt, Wallingford’s position on the Thames rendered it strategically important at a national level from before the Norman Conquest all the way through to the mid-17th century. If, as the 12th-century chronicler William of Newburgh put it, royal castles were ‘the bones of the kingdom’, then Wallingford was an integral part of the skeleton. This underlying significance underpins the various phases of occupation over six centuries of use as strategic fortress, royal residence and gaol. Moreover, it is a development about which scholars are now better informed thanks to the quality and sheer quantity of documentary evidence that is analysed here for the first time. In presenting the fruits of recent research, this collection also seeks to offer fresh perspectives by setting the castle in multiple contexts. Castles were not just physical buildings, but institutions that shaped medieval society in a number of ways. The intimate connections between Wallingford Castle and its landscape, its adjacent town and the close relationship with numerous religious foundations are here revealed in more detail than ever before. The Wallingford Castle that emerges from these pages is one that is altogether more complex, intriguing and yet at the same time reassuringly familiar as a monument that placed an indelible mark on the identity of the town of which it formed a part. The recognition that castles displayed distinctly local and regional characteristics in the past also acts as a reminder of what castles mean to communities in the present. It is undoubtedly instructive in the early 21st century to note that the work at Wallingford has been driven forward by partnerships between academics, based at a number of universities and from a variety of disciplines, and The Wallingford Historical and Archaeological Society. At a time when the outlook for the nation’s heritage is at best uncertain, the work presented here undoubtedly provides a model for the future, eminently showing as it does that public history and academic research are far from being mutually exclusive. This volume helps to take Wallingford Castle away from relative anonymity and place it back into the mainstream not just of castle studies, but of English history as a whole. Even today we retain the affinity between our towns and their castles; this volume will help ensure that Wallingford is now on the map in a way that it was not before. Robert Liddiard University of East Anglia, Norwich

xi

xii

PREFACE AND ACKNoWLEDGEMENTS The impetus given by the fieldwork and finds generated by the AHRC-funded Wallingford Burh to Borough Research Project (2008–10), co-ordinated between Archaeology staff from the Universities of Leicester, Exeter and oxford alongside major input from Wallingford’s own museum and Historical and Archaeological Society (TWHAS), has been substantial. Already active prior to the project, TWHAS and Wallingford Museum have gained notable prominence within the region’s counties for their vigour, outlook and outputs: a series of annual october conferences not only helped advertise the results of the Burh to Borough Research Project but also hugely stimulated local, county and wider interest in the town’s heritage and history. These conferences have drawn in a variety of experts from diverse fields to enable a fuller dialogue to be opened on aspects of Wallingford’s (and the region’s) roots, late Saxon emergence, town and castle, medieval urban population, and Civil War contribution. In particular, the discussions and debates bore first fruit in the well-received BAR volume the Origins of the Borough of Wallingford. archaeological and Historical Perspectives (Keats-Rohan and Roffe 2009), which featured notable contributions on parish origins, Wallingford in Domesday, and the genesis of the honour of Wallingford. As with all good publications, the Origins of the Borough monograph raised as many new questions as it offered answers. While the fieldwork of the Wallingford Burh to Borough Research Project was able to investigate in close detail various sectors of the town and castle and to contribute towards some of the historical, topographic and economic questions being raised, it was recognized that, in the case of the castle site especially, the excavation project could only scratch the surface of what is a multi-period, multi-layered complex. The monograph for the Burh to Borough Research Project has since (and very promptly) appeared in print (transforming townscapes – Christie and Creighton 2013) and in itself provides a vital resource for exploring and understanding the archaeology of the town through time; notably it offers a contextualization of much small-scale and unpublished archaeological work in the town and its environs. The volume exploits some of the documentary sources, but has its emphasis firmly in the archaeological data. This present BAR monograph, however, although drawing also on some of the recent archaeology, explicitly draws fully on the documentary data, which are, as will be seen, for the castle especially, hugely rich and important, and yet for far too long under-exploited. A key aim of this collection is thus to start the process of revealing the quality of this evidence for what was, for medieval England, one of the most prominent and prestigious castles. A whole host of people and organizations can be thanked for helping bring this volume to fruition: in Wallingford itself, enormous thanks go to all the members of TWHAS who have contributed in so many ways, whether in assisting at the october and other conferences, aiding in background research, manning the museum and assisting in the many events there, or simply encouraging us in our studies. The Town Council have been unfailing in their support of the work of both TWHAS and Wallingford Museum, from which a number of the illustrations have been sourced. In the county we offer thanks to Richard oram, Planning Archaeologist of oxford County Council, to staff at oxford Archaeology, the librarians of the Rare Books and Special Collections department of the Bodleian Library in the University of oxford, to the staff of the prints and images department of the Ashmolean Museum and the Bodleian Library, and to Judith Curthoys, archivist of Christ Church, oxford. Further afield, Reading Museum generously provided images of objects from the collection of John Kirby Hedges, owner of Wallingford Castle from 1858 until 1901 and historian of the town, and the Dean and Canons of Windsor have provided a fine image of the seal of Matilda of Wallingford – one of the oldest surviving intact seals of a medieval woman. Cambridge University Library has provided aerial surveys that preserve much detail now lost on the castle site. Historic England provided an image of the keep at Rochester at very short notice. In London, Katharine Keats-Rohan offers thanks to the staff of the imaging departments of both the British Museum and the British Library, and to the reading room staff of the British Library and the National Archives at Kew, and also to the imaging services at TNA. In Leicester, Debbie Miles-Williams assisted in production of some of the images reproduced in this volume, and at Exeter, Mike Rouillard contributed many of the images generated from the Wallingford Burh to Borough Research Project, and Professor oliver Creighton has also generously provided some fine photographs featuring Wallingford and other medieval castles; additional drawings and maps have been provided by David Dewey and June Strong. Professor Robert Liddiard read the whole for his Preface and Dr Edward Impey has been very generous with numerous calls on his expertise. The staff at BAR have been very helpful throughout the lengthy preparation and production of this volume, and we are delighted that this is now accepted as part of an ongoing Wallingford series. In addition, we are especially grateful to the Marc Fitch Fund, which has generously contributed to the costs of the publication, thereby once more showing the Fund’s commitment to the dissemination of high quality research. Finally, we thank all the contributors to this volume for their efforts and then their patience; we very much hope that the wait has been worthwhile. KKR, NC, DR

xiii

xiv

1 INTRODUCTION David Roffe It is probably true to say that castles have always captured the imagination. We have few accounts of their impact on the popular mind from their heyday in the Middle Ages, but they must surely have inspired awe in those who beheld them for the first time, and, even more so, fear in those who were in thrall to their lords. If Edward I’s imperial symbolism at, say, Caernavon may not have been immediately comprehensible by the common man in the furrow, the whitewashing of the Tower of London was a message that was hard to miss. More recent perceptions are better evidenced. As a distinct built form castles had become largely redundant by the 16th century, especially in urban contexts, and, as defining motifs like the tower and courtyard plan were incorporated into unambiguously domestic architecture, they merged into the larger class of grand houses. As ruins, by contrast, they were enlisted into a broader cultural scene. The Gothic revival of the second half of the 18th century saw them incorporated into landscaped parks as a backdrop to the classicism of the Georgian houses they framed. The medievalism of the following century romanticized them even further as follies were newly built and mock castles erected for fashionable display. The castle as medieval theatre remains a popular perception to the modern day. It is also a current academic pre-occupation. It was not always so. Early modern antiquaries were usually aware that stone castles were postConquest fortresses, but often enough they misinterpreted the earthworks in which they were set. As apparently earlier features on many sites, ramparts and ditches frequently appear in early histories as Anglo-Saxon or Roman structures. It was not until the later 19th century that a better understanding of these sites began to emerge. In 1895 John Horace Round originated the concept of the ‘introduction of knight service’ after the Norman Conquest and in 1912 Ella Armitage linked the concept of feudalism that it engendered to castle earthworks. At once the castle became an instrument of conquest and the defining feature of a military culture that violently superseded earlier Anglo-Saxon institutions. This was an historical paradigm that was to persist for 80 or so years. When change came it was dramatic. In the wake of the campaigns of rescue and research archaeology of the 1970s and 1980s, seemingly anomalous sites began to emerge. Goltho in Lincolnshire, for example, was revealed as a 10th-century castle in everything but name and other high-status Anglo-Saxon sites came to light beneath early Norman castles. In tandem, standing buildings were re-assessed and it became clear that iconic castles like Bodiam in Sussex had never been seriously defensible. Display was evidently their primary characteristic and raison d’être: they were houses which

were designed to impress in a culture that extolled the virtues of chivalry above and beyond the exigencies of war. In its turn this realization became the stimulus for a more general reassessment of the castle, and in the last 10 years there has emerged the concept of ‘landscapes of lordship’ in which the castle becomes the centrepiece in a social space that defines and delimits status. Although he cannot have known it, Capability Brown was not the innovator he thought he was. If this revisionism reflects modern anthropological perspectives, as the language often suggests, then it has clarified various historical problems. Foremost among these is the relationship of the castle with what went before and, by extension, the nature of the Norman Conquest and settlement. If a castle was a defended house, a mark of lordship, and an estate office, then it fits in well with an earlier tradition of Anglo-Saxon lordly residences. According to the early 11th-century text known as Geþyncðo, ‘the promotion law’, a ceorl was worthy of the rank of thegn if he had five hides, a church, a bell house, a burhgeat, and a seat in the king’s hall. The burhgeat, a defended gatehouse or tower was the most visible structural sign of lordship and as such a direct analogue of the post-Conquest castle as now redefined. The match was not exact – the motte was undoubtedly a Norman innovation, albeit often a secondary feature of a site – but modifications to the native model slotted into a common form of elite display (Creighton 2002; Liddiard 2004). From this perspective it can be seen that castle construction was often as much about legitimization of political power as domination. The early castles of the Conquest were overwhelmingly located within existing centres of lordship. In Lincolnshire, for example, 13 out of the 17 castles that came into existence within the first 50 years of the Conquest were built in manors held by pre-Conquest lords who enjoyed sake and soke, the mark of a king’s thegn and, significantly, the touchstone of tenure in barony in the 12th century (Roffe 1990; Roffe 2014). Sometimes the continuity is palpable. At Oxford St George’s Tower, an archetypical burhgeat, was incorporated into the structure of the castle (see Norton below, Chapter 9). Elsewhere mottes were piled up around existing stone and timber structures of various kinds which had probably served similar functions. It clearly made sense to command the existing infrastructure of lordship in this way. Equally, the appropriation of site demonstrated title. Outside the castleries of the south coast and the Welsh marches, antecessorial succession to the lands of English thegns had been the predominant mechanism of land transfer in the Norman settlement of England. The castle was at once a potent symbol and a display of the right of the new lord.

1

Wallingford

The new paradigm has not only placed the advent of the castle into a more credible contemporary context, it has also modified the understanding of its subsequent development. Much of castles studies in the past had seen the changing forms of castle design simply as responses to purely military problems. Timber was replaced by stone to make the fortress stronger, square towers gave way to round ones to improve stability and defensive capability, baileys were built to provide defence in depth, and keeps were added as final refuges. Other modifications, such as gun ports, were dictated by changes in military tactics and technology. There is clearly some truth in all this, but it is not the whole story. Stone walls were self-evidently more secure than a wooden palisade: any lord would have welcomed the improvement. But construction in stone was expensive and so it was at the same time also a mark of prestige. Conspicuous display was a potent factor and it had an impact on castle design. From early on the setting was carefully landscaped and the building laid out for effect; moats and water features were dug, access was contrived to show the various structures at their best, and the approach to the great hall was choreographed to enhance the prestige of its lord (Liddiard 2005). The grander the lord, the grander the effect. The rectangular tower keep, expensive and technically difficult to build, was the ultimate status symbol in the 12th century. It was entered by a forebuilding at first-floor level, but this was less for defence than impact – many also had access to the ground floor by smaller doors. Rather the visitor climbed up what was effectively a ceremonial staircase into the great hall and the presence of the lord. The type was largely confined to royal castles and those of the greatest tenants-in-chief. Lesser lords had to content themselves with more modest designs. Nevertheless, they too had to impress their own men and built accordingly as their budgets allowed. The imperative to impress never decreased. Subsequent developments, such as concentrification – the multiplication of walls and wards – were as much to do with fashion as military necessity. Castles studies have now come out of the closet of military history to take their place in a wider political, social and cultural history of England. The aim of this present volume is to examine Wallingford Castle in the historic county of Berkshire in these terms. The site is an impressive one. Bounded by the Thames and Castle Street to the east and west and the borough defences and High Street to north and south, it encompasses some 26 acres (10.5ha), almost a quarter of the area of the medieval town. Hitherto this castle has been little studied, perhaps largely because nothing of its once majestic fabric survives above ground, apart from a degraded motte and rare sections of stonework. Only its extensive earthworks are a witness to a major castle today. Now, however, a detailed account of its history and fabric is possible in the light of recent archaeological excavation and survey and the discovery and analysis of important historical sources. What emerges for the first time is that, as a stronghold, arsenal, and palace, Wallingford Castle played a key role in the governance of England from its construction in the late-11th century through into the beginning of the 16th.

Wallingford is a small town on the Thames in the historic county of Berkshire between the major urban centres of Oxford to the north-west and Reading to the south-east. It is best known for the impressive late Anglo-Saxon defensive earthworks that surround the core of the town: it is as one of King Alfred the Great’s burhs of the late 9th- or early 10thcentury Burghal Hidage that Wallingford makes its most prominent appearance in text books of English history. There is, of course, much more to the town than that. In common with many another settlement, its development in the 10th and early 11th centuries is undocumented. But by 1086 Wallingford was the county town of Berkshire. Domesday Book shows that almost all the great lords of the surrounding counties, and beyond, held land in the town and the renders that it made to the king suggest that it was one of the most significant boroughs in the kingdom (Roffe 2009). It was to play a major role in the civil war (or ‘Anarchy’) of King Stephen’s reign (1135– 54) and thereafter it emerged as a bustling town in the 13th century (Herbert 1971). At this period Wallingford’s profile was every bit as prominent as that of Oxford and was somewhat greater than Reading’s. And yet in contrast to its neighbours its history has been largely overlooked. In large measure this seems to have been a function of the town’s subsequent historical trajectory. From the late 13th century onwards there are signs that Wallingford was experiencing difficulties. The area of intramural settlement began to contract, trade contacts diminished, and by the 15th century there were signs of serious stress as parish churches became redundant and relief from taxation was sought from the crown. Wallingford had already ceased to be Berkshire’s county town and 16th- and 17th-century surveys show that it was a shadow of its former self, with fewer tenements than there were in the 11th, as discussed by David Pedgley in this volume (Chapter 13). Adjustment to changing circumstances was slow, but by the 19th century Wallingford had emerged as the small market town that it is today. As such it never attracted much scholarly interest beyond its own boundaries. In 1712 Richard Skirmer wrote a short history of the town, notable for its suspect Latinity and non sequiturs. It was never published (an edition by John Sims and Berenice Pedgley is at long last forthcoming in 2016). In 1881 John Kirby Hedges produced a twovolume history of Wallingford of much greater value. Despite his eccentric views of the origins of the town – not entirely unreasonable at the time – he brought together and preserved many useful documents and provided descriptions of upstanding monuments and buildings that are still of vital importance today. His account of the castle is invaluable. It is, though, an antiquarian work of its time. More scholarly is N. M. Herbert’s examination of the life of the town from 1155 to 1400 in an unpublished PhD thesis of 1971. It documents the burghal administration of the town and its economy. But it rarely strays beyond this narrow focus. In more general works of regional and national history, such as The Cambridge Urban History

2

of England (Palliser et al. 2000), Wallingford has been noticed only in passing. The place of Wallingford in the history of Berkshire and beyond was never seriously considered until the inception of the Wallingford Burh to Borough Research Project under the auspices of the universities of Leicester, Exeter and Oxford (with pilot work from 2001, but with the major AHRC project from 2008). The project’s immediate remit was to elucidate the origins and development of the town, but it couched its investigations in terms of a wider academic agenda. The origins of English towns had been a concern since 1897 when Frederick Maitland put forward his garrison theory. From the phenomenon of contributory manors in Domesday Book, that is rural estates that had properties in nearby towns, Maitland hypothesized that most English boroughs were primarily organized as fortresses against Danish incursions in the 9th and 10th centuries. James Tait effectively demolished this analysis in the same year and it was his view, more fully worked out in 1936, that early towns were also constituted as markets that shaped the mind-set of urban archaeology in the 1960s and 1970s. From key sites, such as Winchester, York, Lincoln, Stamford, Southampton, and Gloucester, it was argued that Roman towns were re-planned and new towns created as urban centres from the reign of King Alfred of Wessex onwards. The evidence seemed clear: urbanization was an event rather than a process. Subsequently, however, this model was challenged as new evidence became available, notably from London, and old sites such as Worcester and Shrewsbury were subject to revisionary reexamination. Wallingford was arguably the perfect site to test the new construct. The open spaces of the Kinecroft, Bullcroft, and the castle – all apparently a testament to the lack of later development of the town – promised the widespread preservation of early archaeological contexts. The Burh to Borough Project set out to test the competing hypotheses of urban origins as well as the impact of the Norman Conquest on the settlement and its subsequent development (Christie and Creighton 2013). In parallel and in conjunction with this major archaeological project, the Wallingford Historical and Archaeological Society (TWHAS) undertook an historical and topographical study of the town and its place within its region. Analysis of the early records supplemented the archaeological evidence for the distribution of settlement within the town and put it in its tenurial and administrative context. A picture emerged of a borough that was the centre for a wide swathe of land in the surrounding countryside. Initially, its territory extended across the Thames to include the Oxfordshire Chilterns, as well as western Berkshire, a complex that was substantially perpetuated in the honour of Wallingford after the Conquest. From round about the year 1000, it became the county town of historic Berkshire. Apparently with a staller in residence (a species of military governor) and a garrison of housecarls in 1066, Wallingford seems to have played a key role in the defence of the middle Thames valley (Keats-Rohan 2009; Roffe 2009). The problem of the chronology of urbanization was to remain unresolved or at least it was reformulated. As

far as can be seen the zoning of the borough – a high status aristocratic site in the north-eastern quadrant, an ecclesiastical one in the north west, an open space to the south west, and the town proper in the south east – seems to have been a very early feature of the site. Situated between the two southern zones, the market may or may not be an original feature. However, it became apparent from both textual and archaeological research that the creation of a nexus of tribute within the borough must have always brought resources in goods, treasure, and coin into Wallingford. By its nature the burghal system concentrated wealth in designated places: it was not primarily about exchange, but, as David Roffe argues below in Chapter 8, its form presupposes it from an early date. What certainly emerges loud and clear from the archaeological and historical examination of the borough and its hinterland was the pre-eminent importance of Wallingford before the Conquest. The Domesday account of the town puts flesh on a settlement that was dominated by the king and his ministers. By 1086 this concentration of power, and the network of lands which supported it, had devolved upon the castle. Its construction marked a Norman recognition of the strategic importance of Wallingford and its prodigious development thereafter attests its continuing prestige and prominence. The castle While the Wallingford Burh to Borough Research Project monograph brought together the archaeology of the castle and outlined many key elements of its then known history (see Christie and Creighton 2013, 145–218), a much fuller account of this major complex is traced in fine detail by Katharine Keats-Rohan in Chapter 5. The exact date of foundation is unknown, but Wallingford Castle was certainly in existence by 1071 when the abbot of Abingdon was imprisoned there. The Norman army had crossed the Thames at Wallingford in its advance on London after the battle of Hastings and had received the submission of the English there. The strategic importance of the site was evidently recognized from the very beginning of the Norman regime: it is not unlikely that the castle was one of the first to be built in England. In large measure its site reflects the characteristics of early urban castle elsewhere in the country as examined by Michael Fradley in Chapter 2. It was built for the king, probably by the apparently accomplished engineer Robert d’Oilly, within the defences of the borough in close proximity to the bridge over the Thames, and its main entrance opened into the town. As in many other Conquest-period towns, it may also have perpetuated an existing centre of power. While there is admittedly little archaeological evidence currently for the use of the site before the Conquest, the borough was strategically important enough to warrant a major late Saxon royal residence and the castle site is its most likely location. The church of All Saints on its western edge was certainly a major foundation, its parish formerly encompassing the extra-parochial castle site, the meadow to the east and the whole of Clapcot to the north. It was a prebend of the chapel of St Nicholas within the 3

castle from the 12th century, but both were constituted as a royal free chapel with characteristics that suggest a preConquest identity. An ancient connection with the crown can be assumed. If there was not a palace nearby, then there must have been a royal hall of some importance. Whether it was adjacent to or identical with the 15 acres where the housecarls recorded in Domesday lived is unclear. We might suppose, however, that there was an AngloSaxon royal burh within the vicinity of the castle site as in Colchester, Oxford, Southampton, Stafford, and Stamford. Continuity is certainly apparent in the land that was assigned to the defence of the castle. According to an unnamed early 13th-century constable, the honour of Wallingford had belonged to Wigod, a minister and most likely one of Edward the Confessor’s stallers, before the Conquest and had come into the hands of Robert d’Oilly and then Miles Crispin, the Domesday lord, through the marriages of his daughter and granddaughter. In reality the formation of the fee was more complex – portions were held by other tenants-in-chief in 1086 – but by 1130 or so almost all of Wigod’s lands, both patrimony and loanland, along with those of his family and their men, had come back into the honour. Heredity clearly played a part here, but the characteristics of the whole suggest that the preConquest identity of the complex was a more significant factor. Unlike most honours, at least south of the Tees, it was constituted as a liberty in which the constable of the castle was effectively sheriff. This, as argued by David Roffe in Chapter 4, was a dispensation which was more akin to the rights and duties of Anglo-Saxon earls and stallers than post-Conquest tenants-in-chief. The logic of the pre-Conquest arrangements seems to have reasserted itself. Wigod’s military role in the defence of the middle Thames valley devolved upon the castle and the lands assigned to him for his ministry passed to its constable. The castle was to remain in the hands of the king for all of its history, bar a few years in the mid-13th century. In parallel the honour was also kept under tight royal control. The terms under which it was held by Robert d’Oilly and Miles Crispin, the first two constables, are unknown. Despite the underlying right through their wives, it is likely that neither had unequivocal title: like any other minister in the 11th and early 12th centuries, constables were probably entrusted with their lands only so long as they held their office. Hereditary right was usually established subsequently, but not at Wallingford. In part this must reflect failure of its lord in the male line: Miles died without an heir in c. 1107. But policy was probably also a factor. Henry I granted both castle and honour to his close companion, Brien fitzCount, with Miles’ widow Matilda, c. 1113. Matilda, though, was almost certainly beyond child-bearing age by that time and so it seems likely that the king did not envisage permanent alienation in hereditary fee. In the event royal right prevailed: when Brien and Matilda retired from public life c. 1148 they surrendered their title to, in their eyes the rightful king, Henry of Anjou, the future Henry II. Thereafter the castle and honour of Wallingford were constituted as an appanage in the right of the heir apparent to the throne and so remained throughout the Middle Ages, apart from a

brief period between 1248 and 1272 when they were held in fee by the canny Richard, earl of Cornwall. Not surprisingly in the light of this tenurial profile, the castle played a major role in the politics of the period. Brien fitzCount was a staunch supporter of the Empress Matilda in the civil war of King Stephen’s reign and the castle and town resisted three sieges in 1139, 1146, and 1152–3, remaining a loyal redoubt for the Angevin cause throughout the period (on events and the archaeology at Wallingford linked to the Anarchy, see Christie and Creighton 2013, 202–08). In the second half of the 12th century the castle became the designated residence of the Young King Henry until his death in 1183 and was then assigned with the honour to John in the reign of Richard until his revolt against the king. It was again readied for action in the civil war of 1215–17. In the reign of Henry III, the king’s brother, Richard, earl of Cornwall, is said to have expended in excess of 10,000 marks on the castle. It was further enhanced by his son Edmund in the 1280s with the foundation of a college of priests and it became a favoured royal residence in the following century. It was the backdrop to Edward II’s fall in 1326 and Joan, the lady of Kent and wife of the Black Prince, lived there from 1361 to her death in 1385. Extensive repairs are known as late as 1424 and Edward IV seems to have kept it in good order. By the late 15th century, however, it was past its heyday. Early in his reign it was favoured by Henry VIII, but subsequently it fell into neglect after it was detached from the duchy of Cornwall in 1540. Wallingford Castle seems to have been a nodal point throughout the Middle Ages. As discussed by David Roffe in Chapter 8, its closest parallel is the castle of Nottingham. It too was in origin closely related to a pre-Conquest complex of lands charged with the defence of the area. Sited at a strategic crossing of the Trent, the borough of Nottingham had commanded the route to the north since the early 10th century. In 1066 the earl’s burh was situated there and a large number of thegns in the surrounding area owed service to it. The castle was built in 1068 just to the west of the site and all of the lands of the earl and his men was entrusted to its constable, William Peverel. Like Wallingford, the honour was constituted as a liberty and was subject to close royal supervision throughout the Middle Ages. The castle, along with honour, was entrusted to Prince John in the late 12th century, but thereafter it remained in the hands of the king as a favoured residence. Nottingham, though, was an royal outpost, albeit a key presence in the north. Keefe (1990) has shown that the core power base of the English crown in the 12th century was centred on what he called ‘an administrative enclave’ in the Thames valley and central Wessex. As a residence of trusted royal constables and then the heir to the throne, Wallingford very clearly complemented the king’s castle and palace at New Windsor at the centre of that enclave. The fabric Wallingford, then, was a major castle, not just locally in the Thames valley, but more widely in the kingdom of England. Hitherto no attempt has been made to reconstruct its fabric in any detail. In referring to ‘a third dyke’ in his 4

Itinerary, Leland hinted at a structure of extraordinary size and complexity in 1540s, but the clue was never followed up. A mid-16th-century plan and survey of the inner bailey have been known for 100 years or so. Otherwise notices of the castle in the published public records and elsewhere are few and lacking in coherent detail. The resources therefor have always been meagre: there was simply not enough evidence to make sense of the surviving earthworks. It was thus widely assumed that Wallingford Castle was a simple motte and bailey to which a shell keep and a second bailey was subsequently added. All that has now changed with the recent archaeological survey and excavations, summarized by Oliver Creighton and Neil Christie in Chapter 1, the discovery of a dozen or so building accounts from 1296 to 1426 and a comprehensive survey of 1327, analysed by Katharine Keats-Rohan in Chapter 5, and the identification of a mass of post-demolition documentation, examined by Judy Dewey in Chapter 7. Not only can the layout of the castle in its 13th- to 14th-century heyday now be reconstructed, but also much of the earlier structure too. The Domesday account of Wallingford records that eight hagae, properties, were destroyed in the building of the castle. Whether others were incorporated into the structure is unknown, although probable. The early castle is perhaps most likely to have been a single phase motte and bailey, but it is not impossible that the motte was a later insertion into a primary ringwork; there is no positive evidence either way. The gateway was probably to the west (where it was from the 12th century), but it would have led into an outer bailey rather than directly into the town. The boundary of the extra-parochial castle precinct, as mapped in 1804 and elucidated by Judy Dewey in Chapter 7, extended as far as the backlands of tenements on High Street, perhaps preserving the line of the outer defences. The main gateway into the early castle, again as later, was almost certainly close to the western end of the bridge and the East Gate of the town. The original defences would have been a timber tower and palisade, but these appear to have been replaced by stone at an early date. In 1212 a stone mason’s serjeanty attached to the castle was said to have been held de vetere, indicating that it was thought to have pre-dated the death of Henry I in 1135. Little had been expended on the castle in the reign of Henry II and already by 1139 it was described as ‘most securely fortified’. The rebuilding of the castle in stone can, then, be ascribed most likely to Brien fitzCount. The first unambiguous reference to a tower occurs in 1284, but the 16th-century survey of the inner bailey, examined by the late John Lloyd in Chapter 6, indicates that it was a rectangular tower keep with transverse internal wall of a type that is found in high-status castles in the early 12th century. The plan of the polygonal inner bailey that accompanies the 16th-century survey indicates that the angular towers were square in form and therefore probably of the same date. Brien was almost certainly also responsible for the construction of what later became known as the middle bailey wall. Excavation in 1972 to the south of the motte, discussed by Oliver Creighton and Neil Christie in Chapter 2, suggested a date in the mid12th century.

The Pipe Rolls of the later 12th century indicate that Henry II prepared the castle for use by the Young King, but, as far as is known, there were no works beyond routine repairs and maintenance until the following century. In 1220 Henry III ordered a new great hall to be built, and from 1230 Richard, earl of Cornwall, set in train an ambitious remodelling of the whole castle. The middle bailey was infilled to raise its height and a third wall was added to north, west, and south, in the process encroaching onto the north-south axial road of the borough, leading to its realignment further to the west, and the construction of a new North Gate. A complex hydraulic system controlled from the town’s West Gate supplied three encircling moats. The works were complete by 1250 or so, making Wallingford Castle one of the earlier concentric castles in England. The only major change thereafter was the re-foundation of the chapel of St Nicholas and the construction of a college for its priests in the lower middle bailey by Richard’s son Edmund in the 1280s. This was the castle that was described in the survey of 1327 which is edited and translated by Katharine Keats-Rohan in Chapter 5. In combination with building accounts and later records it provides the framework for the reconstruction of the internal layout of the castle. The main entrance was at the south-east corner of the site where there was a barbican, drawbridge, and gateway – probably known as the Constable’s Gate – close by the west end of Wallingford bridge. The outer bailey contained a pinfold but otherwise appears to have been largely empty; it was probably used for grazing and possibly lists. The middle bailey was more intensively occupied. Entered over a second drawbridge through Bydongate directly to the north of the outer gate, much of its southern area was given over to the cloister and living quarters of the college and its priests. Further to the south-west was the Great Stable, a prison known as the Blindhouse, and a building called La Studie, possibly a writing office or library. The middle bailey in its turn connected to the inner bailey by a third drawbridge and gatehouse, the Lieutenant’s Lodgings, on the west. Here were sited the castle’s main domestic buildings. As was common at the time, they were divided into discrete suites separated by palings and pentices. To the north-west was the Great Hall, the King’s Chamber block, the Great Kitchen, and various service buildings; to the north the Great Chapel, that is the chapel of St Nicholas; to the east, the Queen’s Chamber, the Red Hall, and services; to the south-west, the hall of the knights and esquires and the Great Prison. Finally, the inner bailey gave access to the Great Tower on the motte via a forebuilding. The castle as palace, fortress, and estate centre Wallingford Castle was undoubtedly built on a scale to be compared with any of the great castles of England. The multiplication of walls, its towers, and water features were clearly designed to impress. And impress they almost certainly did. From a distance the castle’s most prominent feature must have been the Great Tower – rendered, pargetted, and whitewashed, it was a potent symbol of power – and the circuitous approach from the 5

outer gate was no doubt intended to show it and its setting to their best advantage (cf Christie and Creighton 2013; Chapter 7 below, pp. 178–9 on approaching the castle). There are few details of its internal configuration, but the Great Tower was no doubt decked out as befitted its lord for largely ceremonial use. The main accommodation was in the inner bailey and the names of the Red Hall, the Green Hall, and the like hint at sumptuous decorative schemes. They may also reference a wider chivalric culture. Wallingford Castle features as a setting in a number of medieval romances and its lords clearly played on the theme, claiming the swords of King Richard (of the Romans or possibly of England), Gawain, and Guy of Warwick. The buildings may thus have consciously invoked scenes from romance and conjured up notions of an idealized Camelot. Wallingford Castle was, then, a grand house in martial style and one regaled with the traditions and legends to endorse its authenticity. It was every bit the status symbol that the revisionist school of castle studies would predict. But it was not just for show. At one and the same time the castle was also a formidable fortress. As its history shows, it was eminently defensible and the necessary infrastructure was, significantly, kept in place throughout much of the Middle Ages. From its first construction, as a royal castle it had commanded castle-guard not only from the honour of Wallingford but also from tenantsin-chief in Berkshire and southern Oxfordshire. There were also numerous serjeants who owed service of one kind or another; they were particularly numerous in and around Wallingford itself. By the 13th century obligations of this kind had generally been commuted to a monetary render; in Wallingford, by contrast, they were enforced, albeit latterly only in times of war, until well into the 15th century. Throughout the period the castle was a royal arsenal; periodic surveys of arms reveal that all sorts of military equipment, from swords and crossbows to large siege engines, were stored in the Great Tower and elsewhere, along with emergency supplies of food. The deterioration of weapons and victuals was a constant concern of its constables and royal ministers. In the later Middle Ages successive kings saw Wallingford Castle as one of their key strongholds in southern England. The castle was also the caput of the honour of Wallingford. The honorial court probably met in the Great Hall. As the honour was constituted as a liberty, its jurisdiction was more extensive than was usual. In common with most courts of its type the lord, or the constable in his place, presided in disputes between the knights of the honour. But the right to the assize of bread and ale, view of frankpledge, and the return of writs brought all manner of men of the honour into the court. One of the prisons within the castle, probably that in the inner bailey housed the prisoners of the liberty. By the mid-13th century most honorial courts were largely redundant due to the availability of common law remedies for settling disputes and changing modes of tenure (Carpenter 2000), but the court of the honour of Wallingford persisted well into 15th century because of its wider jurisdiction.

The liberty of the honour stood outside the structure of local government, but it was supplementary rather than parallel to it. Most liberties were of the nature of private property or at least became so in the course of time. Wallingford’s franchise, however, was characteristic of the prerogative that adhered to all royal and comital estates. The liberty, then, did not preclude a close association with the shire and its personnel. As often as not the constable was at one and the same time the sheriff of Berkshire and Oxfordshire (the two counties were usually administered together). Moreover, their responsibilities overlapped: the constable, for example, supervised policing within the honour, while the sheriff oversaw defensive works on the castle and the town. The castle was thus the natural centre of shrieval administration. The shire court met there, probably again in the Great Hall, and the county prison was most likely the Blindhouse in the middle bailey. Finally the castle was an estate centre and farm. As already noted, La Studie, in close proximity to the college, may have been a scriptorium and estate office. There were several demesne manors within the honour and numerous properties in the town. The lord also enjoyed dues and service from the men of the honour. All of these had to be administered. The household also had to be supplied with provisions. Wine came from the local vintners or from London and much of the food was brought in from the lord’s own estates. Some, however, was produced within and around the castle. There were numerous gardens and a vineyard within the three baileys that provided fruit for the household, the court, and for sale. Pears, a luxury fruit in the Middle Ages, were prominent. The moats provided fish, again sometimes for the king when he was elsewhere. Outside the castle there was a swannery in the vicinity of Queen’s Arbour between the eastern curtain wall and the river and the castle mills, an area reconstructed in detail by Judy Dewey in Chapter 7. The mill at the South Gate also belonged to the castle, although this may well have supplied the town rather than the household. Hay was grown for the horses of the garrison in King’s Mead to the north and Clapcot provided pasture for sheep. The castle, then, was no simple institution with a well-defined identity and role. It was clearly the central element in a landscape of lordship, a symbol of authority, and an expression of power: its architectural syntax was altogether decidedly aristocratic. As such it was in marked contrast with the more workaday Oxford, a shrieval castle discussed by Andrew Norton in Chapter 9; the parallel survey of 1327, edited and translated by Katharine KeatsRohan in Chapter 10, reveals a structure which was much less cared for and maintained. However, Wallingford Castle was not just that: it was no inward-looking royal complex. It also intersected on a number of levels with a range of different communities and interests. Not the least of these was the borough of Wallingford. Castle and town With housecarls resident there before the Conquest, the town was used to a garrison in the 11th century. Domesday Book and charter evidence indicate that ministri (royal 6

as opposed to major building projects, and were more usually labourers. If service characterizes the relations between town and castle in the 14th century, it probably represents the essence of its economy before. There is evidence for production, notably of cloth, but it was largely for local consumption. Again merchants operated from the town, but there was no great market or fair to bring in traders from further afield and so their trade was confined at best to a regional compass. Beyond the hinterland of its local market, the castle, and the Anglo-Saxon garrison that preceded it, was probably the main motor of the economy and it was this, namely its narrow base as argued by David Roffe in Chapter 8, that led to the stagnation of the town from the 13th century. The various functions of the castle continued to bring the communities associated with it into Wallingford, but its increasingly aristocratic character militated against the diversification of the town’s economy. In perpetuating the pre-Conquest garrison, the castle had undoubtedly sustained the town, but, paradoxically, its continuing prominence was its undoing. Wallingford was outstripped by the neighbouring towns of Oxford and Reading and was not destined to regain its earlier preeminence. It had probably never become truly urban like these its competitors.

officials) were as characteristic of the town before the Conquest as after: the Normans who constructed the castle and then manned it effectively slotted into the places left by their pre-Conquest counterparts. There can be no doubt that they and their successors exerted a considerable influence in the town. Norman ministri, for example, were responsible for the reform of Christchurch/Holy Trinity in Wallingford, here examined anew in the light of a recently discovered terrier of the mid-12th century by Katharine Keats-Rohan in Chapter 11. Before the Conquest the church seems to have been a college of priests with an extensive parish: the two churches of St Mary and St Martin in Wallingford had belonged to it. The process of re-foundation was a protracted one. It probably began with the grant of Holy Trinity to St Albans Abbey by Nigel de Albini before 1086. Various royal officials contributed to the endowment, but it was Geoffrey the Chamberlain who seems to have bought out the rights of the many priests and their families in the reign of William Rufus and it was he who was subsequently recognized as the founder. The priory of Holy Trinity was finally dedicated in the early years of the reign of Henry I. This, though, was no royal foundation. The priory remained a parish church throughout the Middle Ages and attracted endowments from the burgesses of Wallingford from the very start. Its history and layout is examined by David Pedgley in Chapter 12. In reality it may have been as much a popular foundation as an aristocratic one. This is significant. The town clearly had greater autonomy than is immediately apparent after the Conquest as no doubt before. In 1086 Wallingford rendered a farm of £80 by tale (a render in coin) and it is likely that it was the burgesses who were responsible for payment. The first formal charter of liberties, as far as is known, was granted by Henry II in 1155, but it seems clear that this merely confirmed privileges that they already enjoyed. Wallingford was no seigniorial borough in thrall to a lord in the 11th, 12th, and 13th centuries. As was common in towns at the time, the mayor and bailiffs were townsmen and usually English at that. From the 14th century castle officials and functionaries began to make their appearance in burghal administration, but by then their posts were largely sinecures: there was no real dominance of one over the other. As institutions, therefore, castle and town remained relatively distinct. Economically, there was more interdependence. The castle was probably always a dominant determining factor in the economic life of the town. As noted above, much produce and materials were brought in from the honour and processed within the castle itself. The townsmen largely provided services. At the top end of the scale they supplied luxury goods. The goldsmiths and vintners of Wallingford were probably dependent on the denizens of the castle for much of their trade. Lower down the social scale the townsmen and townswomen simply provided labour and then at a low level. The building accounts show that a few craftsmen employed in castle works were resident, but most were specialists brought in from elsewhere. Generally it seems locals were confined at best to routine maintenance jobs,

Swansong By the mid-16th century, the town was in an abject state. Its population had contracted to such an extent that its ten parishes churches had been reduced to three. Many tenements were empty, others derelict. The rents that were owed in respect of the borough farm fell short of the sum demanded by the Exchequer, and it is clear that the burgesses, as long before, were unable to meet the commitments imposed on them in return for their charter of liberties. Wallingford was not the only town to suffer a downturn in its fortunes at this time; many another was also in trouble. The reasons were various, but here the state of the castle was surely a major factor. The survey of 1555 reveals that lead had already been stripped from the structure: 10 tons were to be taken to Windsor Castle in 1554. Its aim was apparently to determine the value of the remaining fabric in stone and timber. The castle was no longer a royal residence and there is no evidence to suggest that it was kept in repair systematically: in 1606 the outer defences were said to be ‘altogether decayed’. Some houses within remained habitable, notably the former college which had passed to Christ Church, Oxford, as a retreat in 1548. But there was no great household to sustain the command economy that had for so long underwritten Wallingford’s fortunes. Royal interest in the castle in the later 16th century and early 17th was confined to the issues of the lands in and around the castle that had been leased out: the castle and lands were merely a source of income, first for the crown itself and then the queen consort. Despite its neglect as a building, the castle nevertheless remained sufficiently formidable to be garrisoned for the king in the Civil War as an outpost to Oxford as discussed by Judy Dewey 7

in Chapter 14. The structure was made defensible and alterations to the works were made that were appropriate to the exigencies of contemporary warfare. It was ready for action by March 1643. The details are sketchy: storehouses and other buildings were put up and two new bulwarks were constructed, probably the gun emplacements outside the northern medieval defences investigated by the Burh to Borough Research Project as summarized by Neil Christie in Chapter 15. At the same time the borough defences were put in order and various houses were demolished to provide materials for the defensive works and to afford clear sight-lines. Wallingford became the forward base for the field-army with substantial supplies of powder, bullet and match being sent by water from Oxford for safekeeping in the castle. It remained an arsenal throughout the war. Wallingford was a constant threat to parliamentary forces. The garrison raided the surrounding area and provided support troops for various royalist operations. Its strength, though, appears to have deterred a direct assault until the very end of the war. Wallingford surrendered after a 12-week siege in early 1646 when it became clear that the royalist cause had been lost. The castle passed to parliamentary control, but was apparently still considered a threat to the Commonwealth: its demolition was ordered in 1652.

in the 9th century and is one that ultimately determined the course of the history of the castle that took on its mantle. The structure was undoubtedly a symbol of power, an expression of status, and then an ostentatious palace. Nevertheless, everything that was exceptional about it – the continuity of tenure and function over the Conquest, the liberty of the honour, its extraordinary longevity as an institution, and the all-but-unique survival of its castleguard into the 15th century – points to its primary role as a fortress. It was as such that Wallingford was every bit as prominent as the royal castles of Windsor, the Tower of London, Dover, and Nottingham. It is only in its demolition in 1652 that its history departs in any marked way from these major centres of royal power. After the Restoration the crown regained its right to the much of the site. but the land was leased to tenants of one sort or another, mainly, it seems, for pasture. Christ Church also leased out the college lands, but here the Dean’s Lodging and the Priests’ Lodging therein remained as residence into the 19th century. It is these leases, and the legal wrangling that they occasioned, that provide so much of the detail of the micro-topography of the castle site recovered by Judy Dewey in Chapter 7. The two estates were finally sold off in several lots in 1817 and 1819. It was the Hedges family who reunited the whole site in single ownership in 1858 and who in 1977 granted it to the town of Wallingford. Wallingford’s Castle Meadows are now managed on the borough’s behalf by Earth Trust as a public park and open space (http://www.earthtrust.org.uk/). Apart from the surviving earthworks and fragmentary structures there is not much left to hint at what was once an institution of the greatest national importance. The aim of this volume is to undo something of the vandalism of the Civil War and restore Wallingford Castle to its rightful place at the centre of English history.

Conclusion The role of the castle in the Civil War underlines the central importance of place in its history. The remaining fortifications were clearly an important factor in the choice of the site as a forward position by the royalist forces. But it was its strategic position that was decisive in determining its re-occupation. In commanding the river crossing at a vital pinch-point Wallingford dominated the middle Thames valley and the land to north and south. This was a characteristic that goes back to the very origins of the burh

8

2 THE ARCHAEOLOGY OF WALLINGFORD CASTLE: A SUMMARY OF THE CURRENT STATE OF KNOWLEDGE Oliver Creighton and Neil Christie

Abstract Wallingford Castle survives as a complex suite of multiphase earthworks, with minimal upstanding remains, sealing a deeply stratified sequence stretching back to the late Anglo-Saxon period. Following antiquarian-style investigations in the 19th century, large-scale excavation on the North Gate site in the 1970s demonstrated how the castle expanded over the Saxo-Norman town; work in the Middle Bailey in the 1970s revealed a cob-built kitchen; and piecemeal developer-funded archaeology since the 1990s has evaluated small areas of the motte, bailey and surrounding area. In the 2000s the Wallingford Burh to Borough Research Project conducted largescale topographical and geophysical surveys and carried out targeted excavation within the inner bailey, on the ‘barbican’ area, in the Castle Meadows, and in the Queen’s Arbour. When synthesized and evaluated alongside the documentary sources, this rich archaeological record transforms our understanding of the castle’s form, and

development; its place within Wallingford’s overall urban story and in the development of the hinterland; and can help us engage with medieval experiences and perceptions of these spaces and places. Introduction Wallingford Castle’s gentle, grassed-over earthworks represent the proverbial tip of an iceberg comprising the surface and below-ground archaeology of an extensive urban fortress whose active history spans almost 600 years (Figure 2.1). Its high status and extended longevity mean a complex and deeply stratified archaeology whose full exploration would require substantial and hugely costly logistics, which in the modern economic and academic climate is far out of reach; in any case, current heritage protection would prevent any such major disruption to the site. Nonetheless, Wallingford Castle’s archaeology has been sampled on at least ten different occasions in the past

Figure 2.1 Wallingford Castle’s earthworks (photo: Oliver Creighton).

9

Figure 2.2 Map showing archaeological excavations and other interventions on the castle site (illustration: Mike Rouillard).

and present – for curiosity, for academic research and in advance of building development, sometimes realized and sometimes not (Figure 2.2). This paper presents a digest of and commentary on all the known archaeological interventions on Wallingford Castle. It forms a summary statement in support of a comprehensive monograph on the town’s archaeology that is the principal output of the Wallingford Burh to

Borough Research Project (Christie and Creighton 2013). In particular it serves to flag the rich archaeological potential of the site, to sit alongside the wealth of related documentary data that have been explored in this volume. The discussion that concludes this paper reflects upon the myriad ways in which archaeological investigation has illuminated Wallingford Castle’s landscape context.

10

Early Interventions

reports (Brooks 1965a; 1966; 1968), although the excavation has now been bought to publication within the Wallingford Burh to Borough Research Project monograph (Christie and Creighton 2013, 88–98, 114–23). Besides an ambitious cutting through part of the castle’s bailey defences to sample the burghal ramparts sealed beneath, in a remarkable exercise of archaeological ingenuity and topographical deduction, in 1965 Brooks was able to pinpoint the location of Wallingford’s lost North Gate, which had been re-located to the west when the castle’s outer defences were re-planned, thus giving the road in and out of the town on this (north) side its characteristic kink. The excavated original road running up to this gate preserved a Saxo-Norman street frontage lined with properties, all sealed beneath the castle’s outer bank. A key find from this area was the 10th-century inscribed weaving sword, now displayed at Reading Museum. The pottery assemblage from the destruction layers dated the burial of the old road and gate broadly to the 13th century; circumstantial evidence favours the tenure of Richard, earl of Cornwall, who held the fortress from 1231 until his death in 1272 and who initiated various works of repair and extension to the castle fabric, as the most likely context for transformation of the fortress into a concentric show-castle. As such, we can now identify Wallingford Castle as a prodigious exercise in concentric defensive planning, although the defences have so far only been explored on the northern side. While the late 13th-century Edwardian campaigns in North Wales are sometimes cited as the earliest concentric castle planning in Britain, we

The first known archaeological excavations on the castle are recorded in the late 19th century by John Kirby Hedges – gentleman scholar, owner of the site and historian of the town (1881, I, 149–52). While Hedges was keen to document his erroneous belief in a Roman ancestry for Wallingford, and reported the recovery of Roman objects and coins from the castle, he readily acknowledged that the archaeological features he described there were certainly medieval. He reported much earlier explorations of c. 1700 that revealed a passageway between two walls, apparently near the castle mound, as well as 19th-century investigations of a passageway with a staircase at each end (possibly the same feature) and a sluice. In addition we are told that near the base of the motte Hedges revealed the earthwork to have apparently been built upon a large saucer-like masonry foundation (although he was in all probability describing a stone-lined ditch). The 1960s: the North Gate site Of the two important excavations conducted on the castle site in the 20th century, the largest in scale were led by Nicholas Brooks in 1965, 1966 and 1968. These were intended chiefly to investigate the defences and planning of the Anglo-Saxon burh rather than the later medieval castle specifically, however (Figure 2.3). Only a short interim statement on the results had previously been published (Brooks 1965b), augmented by unpublished

Figure 2.3 Brooks’ excavations on the North Gate (photo: Nicholas Brooks).

11

should remember that the plan of Dover castle exhibited concentricity by the late 12th century; specific inspiration for the planning of Wallingford Castle may be sought at the Tower of London, where Edward I likely re-activated an initiative of Henry III in 1238 or 1239 to add an outer moat and wall to the royal castle, or even at Ascalon, on the Mediterranean coast of Israel, where Richard, earl of Cornwall, had been personally involved in the construction of a double-walled castle completed by the end of April 1241 (Pringle 1984, 143–46; Goodall 2011, 139–42, 191, 200–27).

monograph (Christie and Creighton 2013, 183–202). A further excavation funded by the Architects Benevolent Society was conducted in November 1975 but involved only machine excavation of a small trench (6 x 1m) set in the north-west part of the middle bailey, and its precise location is unknown (Thomas 1975). The 1972 excavation was limited to a 12 x 23m area within the garden of a proposed new building and a 1.50– 3.50 x 31m section across the rampart. Five clear phases were recorded: Phase 1, pre-dating the enclosure of the excavated zone within the castle’s middle bailey, produced limited evidence of 11th- and 12th-century activity, including a large (3.20m diameter and 1.48m deep) pit that forms the earliest dateable Saxo-Norman feature so far uncovered on the entire castle site. Other pits, gullies, post-holes, post-hole alignments and trample layers pre-dated the middle bailey but formed no coherent pattern.

The 1970s: the middle bailey In 1972 Robert Carr directed an excavation within the middle bailey, to the west of the standing structure related to St Nicholas’ College (Figure 2.4). The most remarkable finding – unanticipated and still unparalleled elsewhere in Britain – was a cob-built medieval kitchen. Only summary notes on the excavation were published at the time (Carr 1973b, 18; Webster and Cherry 1973, 159–161), although its importance was flagged in a short illustrated piece in Current Archaeology which described the structure as ‘One of the most remarkable buildings ever to have survived from Medieval Britain’ (Anon 1972, 318). The results are now published as fully as is practicable in the Wallingford Burh to Borough Research Project

Phase 2 saw the construction of the two cob-built structures within the newly enclosed middle bailey, which was fortified with a gravel bank and mortared chalk wall, probably in the mid-12th century. One building was excavated in totality, this having external dimensions of 8.30 x 15.80m, while only the corner of the second structure was revealed, this measuring

Figure 2.4 View of Carr’s excavations within the middle bailey in 1972 (photo: Robert Carr).

12

1.90 x 4.60m. Rising to a height of 1.80m, the cob walls of the main building had been preserved by a massive operation of earth movement related to the refortification and extension of the castle in the mid13th century. Doorways and internal hearths were identified and the complex can confidently be identified as a medieval kitchen.

is unusual for a castle site, where pig typically predominate. Horse, dog and cat bones were next in the sequence, and, although less frequently encountered, were present in similar quantities. Although all parts of the carcass were recovered for the major domestic species, when these are compared in terms of their likely survival it seems apparent that only pig remains were consistent with the deposition of relatively complete carcasses. The pig, sheep and cattle assemblages show a cull of prime meat age animals, while evidence for sheep and pig neonatal fatalities implies that they were bred on or close to the site. Indeed, there is good documentary evidence for this, as shown by Dr KeatsRohan in this volume (p. 70). Wild mammals included red, roe and fallow deer, hare and rabbit, as well as wild bird species hunted for the table, including mallard and teal, partridge, pheasant and woodcock. Oysters and fish, including cod, were also noted, but in low numbers. Medieval kitchens are notoriously ephemeral structures and the fact that most of the surviving evidence elsewhere in Britain relates to service buildings incorporated into masonry structures such as towers and donjons (as at Durham) or built against curtain walls (as at Pontefract and Berkeley) makes the existence of a freestanding example within the middle bailey at Wallingford more remarkable still (see Kenyon 1990, 138–50). The context of the Wallingford Castle cob buildings is problematic, since, while kitchens within the castle are documented from the 13th century onwards, it is not always clear whether these remained on the same site/s, and evidence for multiple kitchens within castle complexes is well known. Most

Phase 3 comprised the episode of soil dumping in the 13th century that infilled the middle bailey and raised the ground level within this zone by at least 2.00m. No structures were excavated that could be associated with this phase. It is feasible to link this work, presumably associated with construction of new, ‘modernized’ kitchens elsewhere in the castle site, perhaps in the inner bailey, with large-scale reorganization of the site during the tenure of Richard, earl of Cornwall. Phase 4 features relate to a mid-17th-century Civil War rebuild of the curtain wall, here comprising a crude arrangement of mortar-bonded chalk blocks rising to a height of 1.30m and at least 0.70m wide. Lastly, Phase 5 was directly related to the 19th-century landscaping of the outer ward. Perhaps John Kirby Hedges’ various finds came about from such works. Besides the ceramics, important information derives from the fairly rich animal bone assemblage from the Carr excavations. This zooarchaeological assemblage produced evidence for a wide range of species with sheep the most common, followed by cattle and then pig – a sequence that

Figure 2.5 The curtain wall on the south side of the middle bailey (photo: Oliver Creighton).

13

probably they were linked to an adjacent hall within the middle bailey before their destruction in the 13th century, being too distant to have served diners within the mottetop donjon, which had its own kitchens (for full discussion, see Christie and Creighton 2013, 198–202).

(Ford 1995). A related watching brief observed an area of unmortared rubble, conceivably representing part of the bailey defences and tentatively dated to the 13th/14th century on the basis of related pottery (Saunders 1995; Ford et al. 2012). Other developer-funded evaluations have sampled areas on the fringe of the castle, including the former Lamb Garage site on Castle Street, where interventions revealed evidence of late 11th-century activity seemingly corresponding to an intensification of occupation in the immediate environs of the newly established Norman fortress (Hull and Pine 2001; Mundin 2008).

The 1990s: developer-funded interventions Interventions across the 1980s were limited to only a small-scale piece of archaeological recording in 1987, although this was centred on the surveying of a segment of the 30m-long surviving stretch of curtain wall on the south-east side of the middle bailey (Figure 2.5); the work was undertaken in advance of attempts to rectify an outward lean (Durham 1987). This survey revealed that the masonry was originally squared and finished to an ashlar face, although its heavily eroded condition gives the outward impression of coursed rubble. The changing context of professional archaeology since the earlier campaigns of excavation on Wallingford Castle ensured that the next interventions into the site’s rich archaeological sequence, in the early 1990s, took a very different form – namely as small-scale evaluations funded by developers. Two such evaluations took place, both in the vicinity of the Castle Farm site to the southeast of the motte and beyond the zone of the Scheduled Ancient Monument. First, a watching brief and smallscale excavation of a 16 x 9m trench by the Oxford Archaeological Unit in advance of the insertion of a slurry tank encountered a substantial ditch, interpreted as the bailey ditch but more likely to be that of the motte, filled with layers of re-deposited building materials (OAU 1992). Subsequent building development on the site (with construction of Thameside Mansion) entailed a Thames Valley Archaeological Services field evaluation involving two machine-cut trenches, which revealed made-up ground thought to relate to 17th-century Civil War refurbishment

The 2000s: the Wallingford Burh to Borough Research Project – surveys and excavations Four excavations within or close to the castle complex have taken place as part of the fieldwork campaigns of the collaborative Wallingford Burh to Borough Research Project (2008–10) run by the Universities of Leicester, Exeter and Oxford in conjunction with The Wallingford Historical and Archaeological Society (TWHAS) (for key interim statements on the project aims and work, see Creighton et al. 2009; Speed et al. 2009; Christie et al. 2010; for the castle specifically, see Christie et al. 2010; for the Project monograph, see Christie and Creighton 2013). With the Project overall geared to exploring the origins, growth and fabric of early to late medieval Wallingford, our engagement with the archaeology of the castle was intended to elucidate this core site’s dynamic relationship within the evolving townscape, and to enhance the available archaeological image as far as possible. To this end, our fieldwork integrated several phases of geophysical and topographical survey alongside targeted excavations. As part of the Burh to Borough Research Project’s aim to generate sub-surface geophysical plots of all the large open spaces within and immediately adjacent to

Figure 2.6 Groundpenetrating radar survey of the inner bailey (photo: Oliver Creighton)

14

the historic urban core, resistivity and magnetometry surveys of accessible areas across the castle site have been conducted, as well as more limited use of GPR (groundpenetrating radar) (Figure 2.6). Covering a large swathe of the castle’s inner and outer baileys, plus Castle Meadows beyond, and alongside the river at Queen’s Arbour and King’s Mead, these data have helped pinpoint targeted excavations to answer key questions, but they also form an immensely valuable source of information in their own right about the castle and its context, to be interpreted in tandem with our extensive topographical earthwork survey. Particularly striking is the high level of stonework survival despite the site’s apparently thorough slighting, including clear traces of multiple curtain walls; a complex of internal buildings; outlying features including a curving feature projecting into the Queen’s Arbour (see below); and plentiful evidence also of post-medieval landscaping and gardening (see Figure 2.6).

by a ditch. Despite close examination of its surface, the only features cut into the platform surface were a number of small animal burrows and tree-root holes; no clear trace was found of any postholes to support any standing structure. Environmental samples from the platform surface provided additional information on the platform’s construction: here the presence of specific types of seed and fruit indicated summer or autumn as the time of year when the sediment was laid down, while the feature was shown to have been built from Glauconitic Marl that was probably imported from quarries on the east side of the Thames in the Crowmarsh Gifford area (Wilkinson et al. 2010). The balance of evidence is that the platform formed part of a mid-17th-century defensive gun emplacement – one of a set of bastions on the outer northern flank of the castle, another of which survives as a prominent earthwork immediately to the east. Support for this hypothesis also came from the numerous clay pipe fragments on the platform, which could well relate to gun crews posted here during the Civil War assaults on Wallingford (see also papers by Christie and Dewey, this volume).

Trench 1 (2008): Castle Meadows A trench of 20 x 10m was opened in the centre of Castle Meadows just outside and to the north of the castle’s outer ditch in July 2008 in order to investigate a geophysical anomaly identified as a lost tower or bastion (Figure 2.7). The earliest deposits comprised a homogenous series of dumped medieval layers from periodic cleaning or recutting of the castle ditch and containing 12th- to 14thcentury pottery as well as modest quantities of residual Anglo-Saxon sherds. These were all sealed by a substantial artificial platform, consisting of a compact but friable light grey clay laid deposit containing fragments of chalk, which lay only 0.20–0.30m below the ground surface. The total size of the U-shaped platform was 20–25m long north to south and c. 12m wide east to west, and it was surrounded

Trench 4 (2009): the inner bailey The opening of an 8 x 20m trench in the inner bailey in the summer of 2009 represented the first time that the innermost core of the castle complex had been sampled archaeologically (Figure 2.8). Positioned to sample a large rectangular feature detected through geophysical survey and thought to be a medieval building platform, the trench also examined part of the area between this putative building and the curtain wall of the inner bailey to the west. Logistical factors prevented excavation beneath a depth of 1.42m, at which depth the archaeological layers continued, although this was sufficient to confirm

Figure 2.7 View of Trench 1 (2008) (photo: Matt Edgeworth).

15

Figure 2.8 View of Trench 4 (2009) (photo: Oliver Creighton).

Figure 2.9 View of Trench 8 (2010) (photo: Oliver Creighton).

16

a sequence of occupation extending back to the first phase of the Norman castle. The earliest archaeological feature was an oval oven or dryer in the south-eastern corner of the trench; it had a heat-affected clay base and was datable to the late 11th to 12th century. Above this, and beneath a much later spread of demolition rubble relating to the castle’s post17th-century Civil War slighting, a complex suite of archaeological features including pits, gullies, and walls was cut into a series of imported silty marl layers that formed the level ground surface of the bailey interior. The core of the excavated site could be rationalized as part of an undercroft attached to a larger domestic building – perhaps a storage area associated with a kitchen, or a wine cellar. Valuably, the presence of intact stretches of (partially robbed) medieval walling alongside robber trenches demonstrates some structural survival despite the apparently heavy post-medieval slighting and later landscaping. This trench produced the largest and most important ceramic assemblage of any of the Wallingford Burh to Borough Research Project excavations within the castle. The main fabric type was Wallingford medieval fabric 37, with vessel types including sooted and burnt cooking pots with thumbed and splayed rims dated to the 11th–12th century and incised decorated and glazed pitchers and jugs of the 12th century; there were also applied decorated and rouletted band pitchers and jugs with olive green and light green glaze. Alongside cooking pots, other vessels may

have been jars for storage, industrial or even medicinal usage; skillets and pipkins (medieval frying pans and saucepans with internal glaze and external sooting) were also recovered as well as glazed and decorated jugs and pitchers for serving liquids, and tripod pitchers for liquid storage. The animal bone assemblage was perhaps relatively modest, but showed, somewhat unusually, that the Norman phase material was dominated by chicken, with sheep and pig recorded in similar quantities followed by cattle, and dog, cat, goose, duck, hare and red deer (antler only). For possible identification of this area with one of the castle’s documented halls, see Keats-Rohan below, p. 84. Trench 8 (2010): the Queen’s Arbour The area known as the Queen’s Arbour, located between the east curtain wall of the inner bailey and the River Thames, was the focus of an excavation in the summer of 2010 (Figure 2.9). A trench of 12 x 25m was positioned to reveal the total plan of a large U-shaped masonry structure initially identified through resistivity survey and sampled through a small-scale evaluation trench by members of the Pilot Project team in 2003. The main stone wall of the U-shaped structure proved to be the earliest feature on the site, underlain by/set into a thick layer of alluvial clay. Formed from a chalk wall between 1.55m and 1.8m in width, the structure itself was 16.6m across and extended westwards out of the trench towards the elevated eastern

Figure 2.10 View of Trench 10 (2011) (photo: Neil Christie).

17

slopes of the castle and to the curtain wall of the inner bailey. The chalk-built wall had been built in a crudely dug foundation slot cut into the clay, and had clearly been constructed in segments, presumably by different work gangs. The small assemblage of pottery mainly comprised medieval sherds from the Berkshire sandy ware tradition, including cooking pots and incised decorated pitchers, suggesting a likely late 12th- to 13th-century date for the structure. Seemingly undocumented, this structure cannot be easily categorized or labelled, although earthwork survey shows that it was embedded within a complex of water management features including ponds, channels, mills and, probably, a swannery. Lying beneath the documented Queen’s Tower, within which one can still observe the base mould of a large (early?) 14th-century window, the entire area was at the very least an elite and very private enclave and may have had garden-like qualities (the area is first formally documented as the Quenesherber in 1376: 1.57, p. 76 below). Alternative interpretations might see the U-shaped feature, projecting towards the river, as a quay, although it probably did not extend into the water and no clear trace of a water channel from the river to the wall was evident in the geophysical or topographic surveys. Potentially the complex is also linked to the noted busy building phase in the castle from the 1230s when goods and materials will have required offloading from river barges. For an alternative identification with a royal mill, see Dewey, Chapter 7 below.

revealing not only the plans and development of these monuments but also something of their everyday social and economic lives, will involve careful integration of different approaches and the synthesis of diverse datasets. The case study of Wallingford exemplifies the importance of drawing on and analysing the findings of modern developer-funded archaeology alongside the results of past excavations (both published and unpublished), and holds particular potential given that here geophysical and topographical survey provide us with rich contextual information alongside the documentary record (KeatsRohan, this volume). Overall, archaeology has contributed to our knowledge of Wallingford Castle’s wider context in several different ways and, given the relatively modest scale of excavations on the site to date, holds its greater potential still to deepen our understanding of the different sorts of interrelationship between the fortress and its landscape in the future. In essence, the Wallingford Burh to Borough Research Project has helped illuminate five clear facets of Wallingford Castle’s landscape context. First, our understanding of the changing topographical relationship between castle and town is now quite firmly established. The boundary between fortress and town was fluid rather than fixed as the castle was re-planned during at least two major period of expansion: when the middle bailey was created, perhaps in the mid-12th century; and when the outer (third) wall was added in the mid-13th century. This latter expansion was accompanied by a major re-casting of the northern part of the town plan that saw the road running in from the north diverted. The complex wet defences of the castle were also closely integrated with the hydrological system that fed the town ditch: the sluice which fed the fortress’s moats lay at the West Gate (KeatsRohan p. 72), which was in some senses under the castle’s jurisdiction. Yet in several senses the distinction between castle and town was not at all clear-cut: we might question, for example, whether a late medieval traveller approaching from the north would have had the sensation of entering a town, a castle or something in between. Second, the extent and complexity of an ‘elite’ landscape attached to the castle are becoming clearer. Two zones of the area surrounding the castle are important in this regard. The low-lying meadows bordering the Thames alongside the eastern limits of the castle (the Queen’s Arbour and King’s Mead) clearly formed a carefully managed enclave featuring mills, ponds, a swannery and perhaps also gardens. The gentle slopes of Clapcot to the north, meanwhile, contained zones reserved for demesne agriculture as well as areas for leisure pursuits including hunting and hawking. Somewhat unusually for a royal fortress, the deer park lay not adjacent but was detached, lying some 10 km to the north-east at Watlington, having been established by the mid-13th century (Mileson 2009, 58, 137). Third, the religious patronage of the castle lords was manifested through the foundation and financial support of a range of nearby establishments. Most obvious was the implicit twinning of the castle with a Benedictine priory from the late 11th century (as at Chester and Shrewsbury, for example), which helped transform the northern portion

Trench 10 (2011): inside the North Gate A further excavation (Trench 10; 20 x 15m) in the summer of 2011 sampled the upper levels of the prominent earthwork on the north-west corner of the castle complex (Figure 2.10). Labelled on early Ordnance Survey maps as the site of Cloere Brien, a notorious prison known to have been in existence by the mid-12th century, this feature can more probably be identified as a barbican. The geophysical survey here, combined with targeted groundpenetrating radar assessment, indicated sizeable masonry survival at depth at the lip of the ‘barbican’ earthwork, and at least one potential built structure in its southern sector. However, the excavations (Christie and Creighton 2013, 212–15; and see papers by Christie and Dewey, this volume) proved more informative about the 17thcentury re-fortification of the site than its medieval plan or function, with a massive importation of many metric tons of materials that elevated and reinforced a defensive platform that will have dominated and so helped defend the North Gate to the town and castle during the Civil War conflict. Discussion: Wallingford Castle’s archaeology in context Given the vast scale of the castle site, the multiple and largely piecemeal interventions that this paper has discussed are little more than tiny windows into an inordinately complex and deeply stratified archaeological record. As is the case for many other urban castles in Britain and Europe, the most appropriate research strategies for 18

of the town into a tangibly elite enclave. The castle also embraced within its defences (at least from the early 12th century) a collegiate church, while further afield hospitals at the South Gate and on the opposite (east) side of the bridge of the town were additional elements within the royal re-shaping of Wallingford’s urban space. Fourth, archaeology has been a little less successful in informing about lines of social and economic contact between castle and territory. Analysis of ceramic goods might tell us about the economic ‘pull’ or ‘reach’ of the castle and how this increased and decreased, but our limited dataset ensures these remain research questions for the future (see discussion by Underwood Keevill in Christie and Creighton 2013, 359–65). At Sandal (West Yorkshire) detailed studies of the pottery assemblage suggested that tenurial linkages between the widely scattered de Warenne estates had a profound influence on the array of ceramic products at the site (Moorhouse 1983; see also McCarthy and Brooks 1988, 92–3). Might such patterns be replicated at Wallingford, which similarly acted as the hub of a large and wealthy honour? Would such patterns reflect commercial trade or the physical movement of individuals such as estate officials between parts of the honour?

Archaeology has been much more successful, however, in illuminating ‘foodways’: comparison of animal bone assemblages within town and castle tells us much about the differential access of communities to foodstuffs procured from surrounding estates (Christie and Creighton 2013, 365–74). Fifth, and still more difficult to come to grips with, is the visual relationship between castle and its setting – i.e. the ways in which the fortress was perceived as approached from vantage points including the great stone bridge, for example, or via the convoluted approach from the north that by the late medieval period skirted the castle’s dramatic concentric defences. Fortunately, the rich and wide-ranging documentary study by Keats-Rohan and Dewey in this book has transformed this situation. When considered in conjunction with the documentary sources, it is clear that archaeology can feed strongly into our understanding not only of the castle’s place in Wallingford’s overall urban story and in the development of its hinterland, but can also help us engage more with medieval experiences and perceptions of these environments as they evolved and were transformed through the centuries.

19

3 URBAN CASTLES IN THE MIDDLE AGES: WALLINGFORD IN CONTEXT Michael Fradley Abstract This paper provides an outline overview of the formation, concept and design of early Norman-period castles inserted within pre-existing English townscapes, notably burhs. Wallingford forms one notable example of this ‘urban castle’ type, which relates to major statements of new royal authority in the English landscape. The paper discusses definitions, castle types, European roots and links, and then assesses how Wallingford fits into the scheme identified.

Drage’s ‘urban castle’ definition was composed in a contrasting relationship both with castles founded in a rural environment and with ‘castle-boroughs’, where a new urban settlement was implanted adjacent to a castle. While Drage’s interpretation was augmented by Creighton in his landscape-oriented synthesis on the British medieval castle (2005), no significant critiques have appeared; indeed, in some cases, the ‘urban castle’ is not assessed as a distinct classification, highlighting the peripheral relevance attributed to this research term by the wider professional field (e.g. Kenyon 1990; Pounds 1990; Thompson 1991). The widespread introduction of elite castle enclosures by members of the new elite following the Norman Conquest is a well-attested historical phenomenon (Brown 1976). Two basic models of enclosure were followed, although within these two forms there could be significant variation in design: the ‘ringwork’ consisted simply of an enclosing bank and possible ditch with one or more entrances, with the rampart often crested by a timber palisade or, less frequently, a masonry wall. In some instances, one or more adjoining enclosures were raised, and these are sometimes termed ‘ringwork-and-bailey’ castles. The second, more common type of castle was known as the ‘motte-andbailey’, formed of a large earth mound or ‘motte’, in many cases crested by some form of structure or tower, to which was attached one or more enclosures, defined again by a rampart and ditch. In the immediate post-Conquest period the majority of castle-builders in England utilized timber to create an enclosure palisade and raise internal structures. Castles were constructed for members of the medieval elite in order to achieve a range of social, political and economic objectives. Residential provisions could be created internally for the owner’s household, as well as guests and captives. A wide variety of structures could thus be built within the castle enclosure, the most common element being a hall and ancillary structures. The permanence and regularity with which these enclosures were occupied by elite groups could vary markedly, and a castle could often be inhabited by only a skeleton staff for extended periods of time. The monumental form of these enclosures clearly provided a defensive or military capacity, enabling a castle to be physically closed off to an enemy. Hugely prominent in the wider built and unbuilt landscape, the castle also became embedded within medieval elite discourse: castles provided a regulated point of social contact, communication, administration and economic activity. They were frequently linked to elite ownership of territory, either as central points within a wider estate, or as positions of control over outlying or contested elements of that estate.

Introduction The late 11th century saw the proliferation of fortified elite enclosures, termed castles, across England and areas of Wales in the aftermath of the conquest of England in 1066 by Duke William of Normandy which saw the wholesale replacement of the English by the Normans and their allies in the upper social echelons of the kingdom. Foreshadowed in the reign of Edward the Confessor, this process of castle construction included a small but significant number built within or in close correspondence to established towns, among which was Wallingford Castle. In the 1980s these were classified archaeologically as ‘urban castles’, denoting a specific group of castles built principally in the period 1066–1100 under royal command or immediate delegation, which were positioned in a direct physical relationship to an existing urban settlement as prime foci of royal and administrative control (Figure 3.1). Is Wallingford typical of castles in this wider context? This paper presents a critical discussion of the concept of the urban castle, followed by a broad analysis of how these sites have been reassessed, with a particular focus on Wallingford Castle. The urban castle: definitions In 1987 Christopher Drage published an analysis of the relationship between castles and urban settlement in which he indentified a distinct group of fortifications that he termed ‘urban castles’. The principal values attributed to these were that they were constructed in the decades immediately following the Norman Conquest in 1066 either by or on behalf of the royal executive, they were inserted into existing urban centres, and their construction frequently required clearance of sectors of these townscapes. Additional characteristics could include partial re-use of existing urban defences; direct relationship with the primary watercourse associated with the settlement; and their substantial scale. All of these ‘values’ are met in the case of Wallingford Castle.

20

H

Castles in principal pre-Conquest towns Castles in other pre-Conquest towns

1

Other important pre-Conquest towns

2

4

3 5

13

10 11

12

9

20 27

23

28

22

32

26

48

44

34

43

30

37

36

35 45

29 31

33 47

25

21

17 18

24

15

14 19

16

6

8

7

42

41

38 39

40

46

0

1 2 3 4 5 6 7 8 9 10

Newcastle Durham York Tadcaster Doncaster Lincoln Chester Newark Nottingham Stafford

11 12 13 14 15 16 17 18 19 20

Shrewsbury Tamworth Derby Leicester Stamford Hereford Worcester Gloucester Warwick Northampton

21 22 23 24 25 26 27 28 29 30

Huntingdon Bedford Cambridge Norwich Thetford Bristol Oxford Buckingham Hertford Colchester

31 32 33 34 35 36 37 38 39 40

London Wallingford Old Sarum Southampton Winchester Guildford Rochester Canterbury Dover Hastings

150km

41 42 43 44 45 46 47 48

Lewes Chichester Wareham Dorchester Exeter Totnes Barnstaple Lydford

Figure 3.1 Norman urban castles in England (image © the Wallingford Burh to Borough Research Project)

The defining elements of Drage’s characterization of the urban castle are spatial location and chronology. Spatially, these are castles built in a direct association with a pre-Conquest, urban settlement, erected either within the perimeter of the town’s defensive perimeter or detached but creating a new focus – as at Nottingham where the castle occupied an elevated site to the west; here the area between castle and burh was immediately

developed as an extension of the town. Chronologically, urban castles were established between 1066–1100, the majority in the first decade after 1066 as evidenced by documentary sources: thus the Anglo-Saxon Chronicle records the construction of castles at Nottingham, York, and Lincoln in 1068 (ASC, D, 148), and Orderic Vitalis further castles at Warwick, Cambridge and Huntingdon (Orderic ii, 218), while Domesday Book (complied after

21

1086) records examples where castle building entailed the destruction of property on the site. The documentary record is incomplete, however: Northampton Castle is an example which, despite being topographically identical to the majority of urban castles, is generally considered as an early 12th-century foundation because that is the point at which it enters the documentary record (RCHME 1985, 333). A second notable phase of castle construction in relation to existing urban settlement during the civil wars, often referred to as the Anarchy, occurred during the reign of King Stephen (1135–54). However, this group will be omitted from our study as they were raised outside the chronological definition of the post-Conquest urban castle. In fact, these foundations differ in that they were built on a smaller scale than the urban castles of the later 11th century, were developed by members of the sub-royal elite in towns that did not previously include a castle, and rarely survived in use into the later 12th century. They differ from temporary siege castles that were erected directly in response to a period of conflict where a major castle already existed, as at Exeter, Bridgnorth and Wallingford. In some instances these ‘Anarchy’ castles encroached on urban religious precincts, as recorded at Cirencester and Malmesbury (Creighton 2005, 280). The imposition of these small castles stemmed from attempts to subvert existing urban power structures, hence in some instances the encroachment on monastic establishments which were often key groups involved in the development of postConquest urban centres (Coulson 1994, 92). Also excluded are the rural castle and castle-borough, each of which were proportionally much more common than urban castles: of c. 1125 sites identified in England and Wales, 82% are defined as rural castles, 13% as castleboroughs and only 5% as urban castles (Creighton 2007, 281). The rural castle can be relatively clearly defined as those post-Conquest elite enclosures erected in the countryside as marks of new or elevated lordship and ownership. ‘Castle-borough’ refers to an urban settlement established contemporaneously with or subsequent to the foundation of a castle. This is a recurrent element in the development of towns across the later 11th to 13th centuries and well-researched examples include Ludlow (Shropshire), Pleshey (Essex) and Plympton (Devon); such sites have a more extensive history of research than the urban castle (e.g. Beresford 1967; Conzen 1969). The concept, therefore, significantly pre-dates the development of the idea of the urban castle as a particular ‘type’ and refers specifically to the process of associated settlement development. A castle-borough might, therefore, have been founded as a rural castle, with its associated settlement only developing some form of urban status subsequently. Equally, an associated urban unit may have failed to develop, or struggled to sustain its economic base and reduced to or never moved beyond a rural status.

European context. One reason for its restricted presence in north-western Europe is almost certainly the fact that the development and proliferation of castle sites from the 9th century onwards occurred contemporaneously with the expansion of urban settlement sites. As a result, where castles and towns do occur in association, such as Falaise or Caen in Normandy, the two settlement components have matured in parallel, and thus resemble what would be termed a castle-borough within the English theatre. In contrast, while England saw a similar growth in urban centres from the late 9th century onwards, castle foundation came largely in the wake of the Norman Conquest. The situation in England was not necessarily special or unique. In north-western Europe a small number of castles were constructed in a small number of instances in association with some of the earliest early medieval urban settlement developments: from c. 800 towns in northern France and Flanders had urban strongholds, often linked to the larger town fortifications or supporting weak points (Schofield and Steur 2007, 117). From the late 9th century a more localized pattern in the same region saw the secular nobility constructing new citadels on sites previously associated with the royal regime; in cities like Antwerp, Ghent and Bruges these become the focus of significant urban development (Nicholas 1997, 56; Schofield and Steur 2007, 121). Numbers of castles in rural and urban contexts increased during the 10th century, but only began to proliferate in the 11th century in north-west Europe, with the exception of the British Isles prior to 1066 and Scandinavia (Aurell 2006, 32–3). In Norway castles were rarely constructed prior to the later 12th century, and in an urban context these were constructed on the town outskirts, as at Oslo and Bergen (Ekroll 1998, 66–9). Another category is the private fortified residence, primarily in the form of a defended tower or enclosed complex, across the same region. Detailed analysis of the medieval city of Siena has shed light on the machinations of the tower house and its community, and its association with town development and inter-elite conflict in an Italian context (English 1984). Proliferating in the 13th and 14th centuries, these urban towers could form the centre of semi-independent urban districts and communities, their fortunes irrevocably linked to the fortunes of the elite grouping and the urban commune (ibid., 175). Conflict and political alliance could also occur in relation to fortified residences in the rural hinterland, and elite groups were sometimes found in control of both urban and rural residences (ibid., 189–91). This is of interest to us, since in the English context the introduction of the post-Conquest urban castle is generally taken to signal a wholly new social phenomenon distinct from the proliferation of private fortified units. In the preConquest period urban fortified defences are perceived as a public device, maintained on behalf of the crown by the wider populace for the protection of the latter. Many of the burh enclosures such as Wallingford that were created in this environment in the late 9th and 10th centuries developed into urban centres. The insertion of urban castles into this context after 1066 is therefore taken to represent a transfer of investment from the larger public burh enclosures to

The European context of the urban castle Strikingly, the urban castle as created in England is a relatively unique phenomenon in its immediate north-west 22

no surviving examples, implying that the stimulus that took hold so much more dramatically in southern Europe was a minor influence in England. In contrast, when urban tower-houses were erected in Ireland (and to a lesser extent in Scotland) in the late medieval and early post-medieval periods, these were potentially structural responses to the higher levels of contemporary internal social conflict and political instability in the country.

the private fortifications of the incoming elite (Reynolds 1999). This model is clearly simplistic and flawed; not only did private defended enclosures exist in pre-Conquest towns, as revealed beneath the castle sites at Southampton and Stafford, but in instances such as Northampton urban defences were maintained into the post-Conquest period (Oxley 1986, 47; Morris 1989, 204; Cuttler et al. 2008; Creighton and Higham 2005, 67). A starker pattern in England is that, apart from urban castles, there is little evidence of private fortified residences within towns in the post-Conquest period, as earlier preConquest enclosed residences were apparently abandoned – in contrast to the situation encountered in many other parts of Europe where private urban fortifications proliferated. The accepted explanation is the influence of a strong institutional monarchy over major urban centres, as witnessed in other regions such as Castile and Scandinavia (Aurell 2006, 34). It is generally believed that a multiplicity of fortified structures in urban contexts was not found in England, but, while archaeological indicators are lacking, there are later documentary indications of fortified structures such as licences to crenellate in locations like London, Maidstone, York and Hull (Higham and Saunders 1997, 122–6). However, contrary to the current understanding, it could be argued that private fortified or crenellated residences did occur in England, albeit at a limited level (Slater 2007, 26), by pointing out that the difference in levels of independent fortification in the mid11th century, when the process of urban castle construction began in England, would have been much lower across northern Europe. In contrast, documentary work on elite links to urban settlements in the pre-Conquest period has identified a condition very similar to that of contemporary Italy and Flanders in terms of a conspicuously high elite presence in urban centres, although this did not develop and manifest over succeeding centuries in the form of fortified elite urban enclaves (Fleming 1993, 37). Of the growing number of potentially secular towers identified in pre-Conquest urban contexts, we can note those at Tanner Street in Winchester and those of St George and St Michaelat-the-Northgate in Oxford (Renn 2003, 71–2; Norton, pp. 200–1 this volume). In addition, urban churches were core elite investments during this period. The relationship between urbanism, castles and private fortifications across Europe can therefore be seen to be relatively heterogeneous, both spatially and temporally, although themes and regional patterns exist. The scale of urban development, socio-economic conditions and societal power structures all appear to have influenced this relationship. In the English context the strength of royal power is frequently highlighted as limiting the proliferation of ‘private’ fortification (Higham and Saunders 1997, 126), although episcopal enclosures and precincts are a notable exception. The ability of the sub-royal elite to stem the construction of fortified residences in their own castleboroughs would imply that this pattern owes as much to the broader social structure as to a particular royal prerogative. The limited documentary evidence would suggest that fortified residences were constructed in at least some English urban centres, but in low numbers and so far with

The Norman Conquest narrative and the interpretation of the urban castle A key driving force behind Drage’s interpretation of the urban castle was the historical context of their construction in the aftermath of the Norman Conquest. Unsurprisingly, this perception of a foreign elite imposing its will over the conquered populace through military and monumental means has come to dominate how they are understood both archaeologically and historically (cf Ottaway 1992, 166; Schofield and Steur 2007, 123). An additional viewpoint is that their construction entailed the development of towns as military strong points which could react to regional uprisings (Oman 1926, 12; Renn 1968, 12; Rowley 1983, 36; Thompson 1991, 147). Even radical commentators on the subject imply that the image of alien domination is perhaps more appropriate to the urban castle than the majority of rural examples (Coulson 2001, 90–1; Hinton 1990, 115). The decision to construct major castles in pre-Conquest towns is similarly used to argue that these sites were central to political and economic power, the implication being that the castle oversaw this capacity – although this correlation is disputed (Griffiths 2003, 104; Hinton 1990, 116). More recent research has broadened this point, with the urban castles seen primarily as a means of seizing control of regional power bases and key nodes of communication, in addition to overseeing larger population centres and functioning as instruments and symbols of Norman royal power (Creighton 2005, 280). This overall perception has generally influenced interpretation of specific sites, particularly in relation to the selection of castle sites within the townscapes (Figure 3.2). Often the ‘peripheral’ location is seen in part as reflecting the mistrust and potential conflict between castle and town, with the castle not envisaged as providing a defensive capacity for the urban populace (Armitage 1912, 95–6; Pounds 1990, 207; Platt 1982, 1). The peripheral location is also sometimes linked to a need for access to open countryside, both for entry and escape if required (Drage 1987, 119; Hill 2000, 183; Palliser et al. 2000, 173) – as has been argued for Bedford, where the castle was sited alongside the Great River Ouse, with access to open country to the east (Baker 1973, 15). However, original posterns leading out into this ‘open country’ are rare and only Warwick and Winchester had principal gates opening outside the urban area; this limited relationship has led to exaggerated speculation that easy access was required from urban castles into the towns to suppress incipient unrest (Pounds 1990, 213–4). In the few discussions of castles that see the urban castle as a distinct phenomenon, the role of Conquest is 23

LONDON

CHICHESTER ROCHESTER

SOUTHAMPTON

WALLINGFORD

LYDFORD

HEREFORD

WORCESTER WINCHESTER COLCHESTER

EXETER

BRISTOL

GLOUCESTER

OXFORD

0

WAREHAM

CANTERBURY

1km

Figure 3.2 Norman ‘urban castles’ and spatial relationships to their townscapes (image © the Wallingford Burh to Borough Research Project)

24

given primacy, giving rise to the assertion that ‘the rapidity and coherency of this castle-building programme points towards a unified programme of Norman repression’ (Creighton 2005, 137). From a morphological perspective, English (1995) has argued that some of the earliest of these castles may have been ringworks rapidly erected under campaign conditions with many only subsequently being redeveloped with mottes as the conquest was completed. The importance of utilizing the existing administrative networks associated with urban settlements is highlighted, but again is not explored in depth and subsumed within a narrative of forcible seizure of power (Drage 1987, 127–8; Creighton 2005, 134–6; Liddiard 2005, 18–9). A central paradox is also identified, in which ‘this unique generation of early urban castles, constructed under the express orders of William [I] as tools of military conquest and in a sense the most ‘Norman’ castles of all, sometimes resemble earlier traditions of AngloSaxon fortification in terms of their social and landscape contexts’ (Creighton 2005, 137). This refers specifically to the process of 10th-century burh-building utilized as offensive garrisoned units that would dominate major settlements and communication networks in midland and eastern England, which is likened to the role of the postConquest castle. However, as the role of these burhs was a central component in the military defence and subsequent expansion of the Wessex kingdom at the expense of a ‘foreign’ invader and, ultimately, the perceived unification of England as a single political entity from which the modern polity is directly descended, their foundations are portrayed in a universally positive light. Notable are views from Liddiard who, although reluctant to question the military rationale of the immediate post-Conquest period, highlights issues with the acceptance of the exact role of the castle within those conditions (2005, 12–5).

the new elite, bishops especially, were widely involved in the reconstruction of urban churches across England, including the often major aggrandizement and, in some cases relocation, of cathedral complexes (Fernie 2002). Former cemeteries have also been observed beneath the castle sites at Norwich, Barnstaple, Hereford, Cambridge and Newark (Ayers 1985, 18–25; Drage 1987, 326, 350; Drury 1982; Liddiard 2005, 19; Miles 1986, 62–71; Marshall and Samuels 1994, 53–4; Pounds 1990, 207; Rowley 1983, RCHME 1959, 306–7). One final area of the pre-Conquest environment that has attracted attention is the re-use of elite residential sites in the construction of post-Conquest urban castles. Excavated evidence of possible high-status late Saxon halls and residences beneath castle sites has come from Southampton, Colchester and Oxford (Holdsworth 1984, 340; Oxley 1986, 47; Drury 1982, 389–90; Poore and Norton 2009). Potentially the most interesting evidence derives from the relatively short-lived castle at Stafford, where investigations showed that the castle ditch closely follows the alignment of a pre-Conquest enclosure, suggesting possible continuity between pre- and postConquest elite enclosures (Cuttler et al. 2008). Perhaps surprisingly, the archaeological examination of early urban castle phases of the later 11th and 12th centuries has been relatively limited. In part this is a consequence of the conditions of preservation: the monumental construction of castle enclosure and motte often incurred substantial earth movement, which effectively seals early deposits from the impact of subsequent development and encroachment. The digging of surrounding castle ditches was destructive of these earlier layers, while the redevelopment of castles from timber to masonry structures similarly diminishes physical data for primary phase activity. However, in some instances the historical trajectory of a castle has created conditions in which early occupation phases have survived. At Winchester an extensive phase of earth movement to raise the surface of the castle interior preserved a whole sequence of early development, including construction of a masonry chapel which probably relates to that documented in 1072; at Gloucester an early urban castle was abandoned c. 1100 in favour of an adjacent site to the north-west, and the site remained largely unoccupied for much of the medieval period so that first-phase castle deposits survived (Darvill 1988); and at Dover Castle and the Tower of London, major intrusive features such as early castle ditch alignments have endured even as the overall castle complex expanded (Parnell 1983). Castle mottes also offer some potential for preservation, depending upon the character of their re-use in subsequent periods. Sequences of pre-13thcentury buildings have been recorded on the western motte of Lewes Castle, while the raising of the castle mound at Baille Hill in York as part of post-medieval landscaping preserved much valuable data (Drewett 1992; Addyman and Priestley 1977). Much more commonplace is an absence of early evidence due to the scale of later castle remodelling. For instance, excavation at Hereford Castle (documented in c. 1050 and perhaps the only example of a pre-Conquest urban castle in England) traced no deposits prior to the

Archaeological evidence of the urban castle form One key area where archaeological investigation has proved enlightening is in the identification of pre-castle settlement deposits. Frequently urban castles were erected on the site of pre-Conquest occupation, and the monumental form of the fortification has allowed these (levelled) deposits to be sealed and often protected from subsequent development. Such demolitions are in fact recorded in Domesday Book in relation to the loss of revenue that they caused; however, excavations at sites like Oxford, Northampton and Winchester have revealed that pre-Conquest properties were encroached upon by castle building at a larger number of sites than those listed in Domesday Book (Jope 1952; Biddle 1970; Chapman 1985; Norton, pp. 202–4, this volume). In addition, castle insertion could obliterate or disrupt road systems, causing a re-routing of streets to and around the new complex. Given the size of these urban castles, their imposition could also impact significantly on the urban ecclesiastical topography, with, on occasion, up to a quarter of the old townscape usurped. Some burhs like Winchester already featured numerous churches and so some (relatively small) losses were perhaps inevitable. In any case, members of 25

13th-century reconstruction of the site (Shoesmith 1980). Similarly, investigations at Dorchester (Dorset) and Southampton have produced very limited evidence of occupation or activity at their respective urban castle sites for the late 11th or early 12th centuries (Oxley 1986, 47). In part this may signify a relative lack of activity within castle baileys; thus the north-eastern bailey of Norwich Castle appears to have been maintained as an enclosed open space prior to its abandonment in the 14th century (Ayers 1985). However, this situation was not consistent across all urban castles, as demonstrated by dense early occupation recorded at Winchester (Biddle 1970), suggesting that the enclosed space of the urban castle was open to a range of different functions. The overall form and extent of the majority of urban castle sites can be reconstructed utilizing a combination of evidence stemming from historical maps, surveys of upstanding features and analysis of major boundaries such as walls, banks and ditches identified through archaeological investigation. Despite this apparent level of survival, the early phases of castle occupation are frequently not available to investigation through the standard methodology of archaeological excavation. For example, at Wallingford Castle, preservation of both early and pre-castle deposits may be strong since our topographical survey, undertaken in tandem with the Burh to Borough Research Project, indicates that the levels of all the castle baileys were raised between the 11th and 13th centuries. However, as the Project excavations showed, the monumental scale of earth movement has created a depth of deposits that is beyond the available resources of most research-led excavations to penetrate effectively (Creighton and Christie, this volume). Architectural analysis of standing buildings is one avenue where research-led analysis has enabled the construction of innovative interpretations. A small number of these urban castles retain architectural elements of late 11th-century date: the earliest examples of masonry towers at Colchester Castle and the Tower of London may have been initiated in the 1070s, although they may not have been completed until c. 1100; and construction of the masonry gatehouse at Exeter Castle may have begun as early as 1068. The architectural study of such evidence has contributed strongly towards understanding castle functions and facilities (Marshall 2002, 27), including dismissing many ‘martial’ interpretations by highlighting the ineffectuality of otherwise highly visible military components, for instance in the provision of raised entrances and defended fore-buildings on castle ‘keeps’, while less prominent utilitarian entrances were often easily accessible away from the castle’s principal façade (Coulson 1996, 182–3). A focus on the ceremonial function of the castle keep has been prominent in recent literature, demonstrating that in both England and Normandy one can observe increasing restrictions of access to more ‘private’ halls and social spaces, which will have been available only to the upper echelons of society and their servants (Marshall 2002; Dixon and Marshall 2002, 238). Interestingly, at the Tower of London, no identifiable private domestic space

was recorded (such as smaller, heated rooms), prompting speculation as to whether the tower functioned for specialist social arenas rather than as a practical residential unit (Drury 2002, 219). However, as Dixon notes, too little is really known about the lifestyle and domestic arrangements of urban populations in the 10th and 11th centuries, even for the elite (2002, 10); furthermore, these interpretations are heavily imbedded in understandings of the evidence from an ‘elite’ perspective, in that we must recognize how these architectural designs were idealized forms created by the elite to be ‘read’ by their peers. The scale and prominence of the larger surviving masonry towers of urban castles have led to them being understood in a highly ‘symbolic’ capacity. At Colchester the keep was built conspicuously over the podium of the Roman Temple of Claudius, leading to interpretations of the direct re-use of the past through which the Norman monarchy asserted their imperial intentions (Drury 1982). Links have also been made between the development of design in Normandy and England, for instance in the similarity of the structures at Ivry-la-Bataille, begun c. 1000, and the keep of the Tower of London which may represent an anachronistic revival of the architectural form in an original English context (Impey 2002, 189, 199). In contrast, the use of typically pre-Conquest architectural motifs in the gatehouse at Exeter Castle is seen as an indication that this castle was created through a medium of pre-Norman construction and display – a facet also identified at Lincoln (Coulson 2001, 72). While only a few standing fragments of stone walls survive at Wallingford Castle and therefore much will rely on future, extensive (and expensive) archaeological investigation to reveal its array of structures, as the detailed discussions of Keats-Rohan and Dewey in this volume identify, the extant documentary sources in fact already provide a wealth of fascinating data and detail to bring many of the components of this castle alive, most notably the architectural arrangement of the keep and baileys, but also the human actors, from constables to smiths. Wallingford Castle: a typical urban castle? How, therefore, does Wallingford’s castle conform to the characteristics of ‘urban castles’ discussed above? Firstly, in terms of its physical positioning it fits the basic characteristic spatial patterning of the majority of these: it was constructed in the north-eastern corner of the pre-Conquest burh defences, utilizing that enclosure’s perimeter on its northern side and occupying a riverside position perched about the terrace of the River Thames. Noticeably, the castle was placed at the upriver end of the settlement rather than downriver – as occurs at the majority of identified urban castles. One possible explanation may be the proximity of Oxford to Wallingford, and the relevance of river traffic between the two towns, or else the access to unbuilt land (the Clapcot area) to the immediate north. The motte would have been a particularly visible component, providing the castle occupants with a wide view across the townscape and especially across the wider Thames landscape. In terms of axial roads, primary 26

crossroads, bridge, marketplace and town gates, there is no spatial suggestion of the urban castle being utilized for any form of immediate physical control over either town or populace. Wallingford Castle was apparently built in a relatively standard motte-and-bailey design, with a motte at its south end and a sub-rectangular bailey to the north. Excavation in the castle zone has so far failed to penetrate to layers from either the earliest phases of castle occupation or precastle settlement, and so it is not possible to comment archaeologically upon the nature of any settlement that the castle replaced. It has been argued that the Domesday reference to 15 acres on which the housecarls lived may refer to the castle site, suggesting perhaps a preConquest palace there (Dewey 2009). On the basis of wider evidence, though, this seems unlikely: in all those urban settlements where a pre-Conquest royal palace was situated, namely Winchester, London (Westminster) and Gloucester (Kingsholm), the post-Conquest castles were raised in an alternative position some distance away. Even in the rural example of the earlier palace at Old Windsor, the new castle was established in an entirely different location. However, a royal enclosure at Wallingford is not unlikely. The church of the extensive parish of All Saints, located on the western fringe of the castle precinct was associated with the royal free chapel of St Nicholas in the castle, which may well indicate a high status site in its vicinity before the Conquest (Roffe 2009, 36). While exhibiting most of the same characteristics as so many other urban castles, Wallingford differs from the majority in its subsequent historical trajectory, chiefly due to its strong and enduring royal connections. In some cases, an urban castle could become detached from

royal authority as part of a private lordship, as occurred for Warwick and Leicester. In other cases the position of the town itself could become diminished within the regional settlement hierarchy and this could relegate the role of an urban castle, as may have been the case at both Buckingham and Stafford. Castles attached to new towns could in parallel rise in importance and, when in royal hands, similarly operate as the centres of shire authority (as at Bridgnorth in Shropshire and Newcastle-upon-Lyme in Staffordshire). The town of Wallingford certainly suffered a decline in its importance within the county of Berkshire from the 12th century onwards. However, while investment in the majority of urban castles was rarely increased after their initial construction, even where they did continue to function as the centre of shire authority, Wallingford Castle remained a royal holding and a focus of major development, especially across the 12th and 13th centuries (Keats-Rohan, this volume). The only other castles to see expansion on a similar scale were Nottingham Castle (Roffe, this volume) and the Tower of London. Indeed, at times during this period Wallingford’s may have been the most grandiose and advanced urban castle, in terms of architecture and engineering, throughout the kingdom of England. This is, one can argue, amply demonstrated by the papers presented in this volume, and more broadly by the historical and archaeological survey work of The Wallingford Historical and Archaeological Society (TWHAS) and the AHRC-funded Wallingford Burh to Borough Research Project, which has made a major step in putting this unique and archaeologically well-preserved urban castle back at the heart of discussion in the theatres of medieval archaeology and history.

27

4 AN ENGLISH LEGACY: THE LIBERTY OF THE HONOUR OF WALLINGFORD David Roffe Abstract Wallingford is one of only a handful of honours in southern England that were constituted as liberties: by the 13th century it was effectively a separate shire. Although it might appear to be a peculiarly privileged Norman castlery, it is argued here that it owed its identity to pre-Conquest arrangements for the defence of the middle Thames valley. The honour of Wallingford was essentially a late AngloSaxon institution which survived the Conquest because of its continuing strategic importance.

The answer to that question provides not only an insight into the formation of the honour and its relationship with Wallingford itself, but also something of the nature of Norman institutions and their debt to the Anglo-Saxon past. It is argued in this paper that Wallingford belonged to a small group of honours that were also liberties. Further, it is asserted that it owed its identity to arrangements for the defence of the middle Thames valley of the 10th or early 11th century. The honour of Wallingford was essentially a late Anglo-Saxon institution.

Introduction

The honour and honorial jurisdiction

The Assize of Clarendon is not as famous as, say, Domesday Book, Magna Carta, or the Bill of Rights. But, in its own way it is every bit as important in English constitutional history. In 1166 England was suffering from an outbreak of lawlessness which threatened the peace of the realm. The king, Henry II, consulted with his barons and they agreed that local juries should be called to identify criminals, so that they could be arrested and brought to trial. The procedure was probably a temporary expedient, but was made permanent in 1176. The Assize of Clarendon was the origin of the Grand Jury which was to form the basis of peace-keeping in England until 1879 and remains a significant institution in the United States of America to the present day (Warren 1987, 108–10). What was significant about the Grand Jury was that it by-passed seigneurial privileges and local procedures. It is of some interest, then, that it is stated in two different clauses that it even applied to the honour of Wallingford (emphasis added):

We should start by defining what an honour is. That is more complex than it might at first appear. The word comes from the Latin honor, that is ‘honour’ in the every-day sense, and it is used to described estates. It is not a particularly precise term (Stenton 1932, 54–8). It may refer to anything from a single parcel of land, on the one hand, to a vast fee on the other. Common to all these usages, however, is the notion of the honour that is conferred by tenure of the land in question. The honourable relationship between lord and man is central to the concept. So it is that the most common use of the word honour is for the complex of estates that a baron held of the king in return for knight service after the Conquest. The Norman Conquest of England saw a tenurial revolution that was unprecedented in English history and has not been paralleled since. All those who fought on the English side against Duke William at the Battle of Hastings automatically forfeited their lands. In the next twenty years almost all the English of the highest ranks were similarly dispossessed and their estates transferred to the followers of William the Conqueror. Illegal seizure was one mechanism, but it was of limited extent. Most lands were granted out by the king in an ordered way. The most visible method was the transfer of blocks of land to a single lord. This was an expedient that was used in strategically sensitive areas such as the south coast and the Welsh marches. More generally, however, the estates of a single pre-Conquest lord, designated antecessor, ‘predecessor’, in Domesday parlance, were given in their entirety, regardless of their location to a Norman successor. Typically, an honour consisted of the lands of a number of such lords (Roffe 2007, 163–76). The formation of the honour and the calculation of the service due from it – its servitium debitum – was a protracted process. It is clear, however, that the Domesday inquest of 1086 marked a significant stage in its crystallization as an institution (Roffe 2007, 180–82).

9. And let there be no one within his castle or without, nor even in the honour of Wallingford, who shall forbid the sheriffs to enter into his court or his land to take the view of frankpledge and to see that all are under pledges; and let them be sent before the sheriffs under free pledge… 11. And let there be none within a city or borough or castle, or without it, nor even in the honour of Wallingford, who shall forbid the sheriffs to enter into their land or soke to arrest those who shall have been accused or are notoriously suspect of being robbers or murderers or thieves or receivers of them, or outlaws, or persons charged concerning the forest; but the king commands that they shall aid the sheriffs to capture them (EHD ii, no 24). No other honour is so distinguished. What was it about this honour that signalled it out for especial mention? 28

Changes in structure were possible thereafter and new fees might be created from it, but in its aftermath the honour had assumed an identity which, by and large, it was to retain for the next 200 years or so. The court of the honour was its main institution. There its affairs were decided and disputes between the knights and tenants who held land from its lord were determined. We are not supposed to use the f-word nowadays – I mean ‘feudalism’ of course – but, yes, we are talking about a feudal system. It is important, nevertheless, not to run away with the idea. We are dealing with a private jurisdiction but it was not an unlimited one. Apart from a few special cases, such as hand-taken felons and the like, the court of the honour was only competent to deal with the business of the honour. The king retained jurisdiction in criminal matters over its lands and men, at least in times of peace, through the public courts of hundred and shire (Green 1997, 247; Reynolds 1994, 375–6).

The franchise was not as extensive as those of the baronies in the far north of England. English kings were slow to make their authority felt beyond the River Tees and many of the fees that were created there in the 12th century inherited the extensive regalian rights of the earls of Northumbria who preceded them (Green 1997, 100–26). Nor could the honour of Wallingford compete with the ecclesiastical liberties of the great Anglo-Saxon churches of Glastonbury, Bury St Edmunds, Ramsey, and Ely: they could claim the pleas of the crown within their respective banlieus (Hurnard 1949, 316ff), and, as far as is recorded, the lord of Wallingford never claimed the same. It would, nevertheless, appear to have its own shirelike administration which was responsible directly to the Exchequer. It was, then, akin to the Rapes of Sussex, the lordships of the Welsh marches, and the Yorkshire liberties of Richmond and Holderness (Thorn 1990; Chibnall 1986, 48–53).

The liberty of the honour of Wallingford

Origins

The Assize of Clarendon was not so much aimed at seigneurial franchises than at the smooth running of these courts (Warren 1987, 108–10). Why, then, is the honour of Wallingford explicitly included in its provisions? The immediate reason may well have been some wellknown incident in which the honour had figured. No such is recorded. It must be suspected, however, that more generally it was perceived as in some way different from other honours (pace Bartlett 2000, 219). And indeed it was. The first clue comes in early 13th-century legal records: anomalously, itinerant justices in eyre held separate sessions for the honour (Crook 1982, 85, 93, 103–4, 110, 129–30, 163–4). We have to wait until 1255 to find out why. In the inquest of the rights of the king of that year the honour is described as a libertas, a ‘liberty’ (RH i, 33). The term is normally used to express the right to regalian privileges, that is jurisdiction that was normally reserved to the king. Several Berkshire juries in 1279/80 variously declared in what these privileges consisted. Edmund, earl of Cornwall, the lord at the time, enjoyed the assizes of bread and ale, view of frankpledge, pleas of vee de naam, gallows, and return of writs (RH ii, 777). The assizes of bread and ale, rooted in the right to sake and soke but regulated by statute, were by the 13th century essentially manorial rights (Cam 1963, 210–2; Davis 2004, 465–6). The liberty was unusual here only in so far as the lord enjoyed the regulation of brewing and baking throughout his honour; normally it was the sitting tenant who concerned himself with such matters. The other liberties were more substantial. View of frankpledge conferred the right to organize tithings, that is a system of communal bail that stood in for policing. Vee de naam was about the right to make distraints, the taking of chattels to enforce court orders. Gallows implied the right to capital punishment, although only in very limited circumstances. Finally, topping the lot, return of writs conferred the right to execute royal writs (Cam 1963, 205–15). In short, the sheriff, the king’s representative in the shire, was largely excluded from the honour of Wallingford (Tilley 2011, 130–207).

All of those fees were undoubtedly post-Conquest creations. The origins of the honour of Wallingford, however, were somewhat different. Remarkably, a medieval tradition survives. In 1212 the constable of Wallingford Castle was asked by the king to make an account of the number of knights’ fees in the honour of Wallingford. He began his report with this account of the honour: Wigod of Wallingford held the honour of Wallingford in the time of King Harold and afterwards in the time of King William I and he had by his wife a daughter whom he gave in marriage to Robert d’Oilly. Robert had by her a daughter, Matilda by name, who was his heir. Miles Crispin married this woman and through her had the honour of Wallingford. When Miles died, King Henry I gave the said Matilda to Brien fitzCount along with her inheritance. He had no heir by her. Both Brien and Matilda his wife gave themselves up to the religious life and Lord Henry, the son of the Empress Matilda, at that time duke of Normandy, took the honour (BF, 116). Such traditions have to be treated with care. They do not always prove to be accurate. Dr Keats-Rohan, however, has shown that this one has more than a grain of truth in it. In her comprehensive and insightful study of the 12thcentury honour she found that Wigod was indeed a common denominator. Of the 135 manors in the honour c.1129, 59% had been held by Wigod himself, his family, or his men in 1066; 16% had been in hands of a certain Beorhtric who seems to have been closely associated with him, and 6% had belonged to the king or earl. The remaining 19% had been held by others who were unrelated or, the vast majority, cannot be identified (Keats-Rohan 2009, 64). It is possible that some, or all, of these last had been men of Wigod before the Conquest or, alternatively, had been acquired by him thereafter; there is little evidence either way (Tilley 2011, 41–60). These figures are in themselves suggestive enough. What is the more remarkable, though, is that the honour 29

asserted its identity as a fee between 1066 and 1129 despite the fact that there was no simple continuity of tenure between the two dates. In 1086, the time of the Domesday inquest, just over half of the manors held by Wigod himself in 1066 were in the hands of Miles Crispin, but the rest were in the fees of Robert d’Oilly, William de Warenne, Earl Roger of Shrewsbury, and William de Braose. The majority of the remaining estates were held by Miles or Robert, but a significant minority were held by others (Keats-Rohan 2009, 64–7). What had been largely one holding before the Conquest had apparently been broken up thereafter, but had come back together by the early 12th century. What, then, constituted this identity? Clearly the rights of family were a potent factor. Wigod’s daughter, Ealdgyth by name, who married Robert d’Oilly, was an heiress as was their daughter Matilda. The rights of both daughter and granddaughter may well have informed the re-formation of the fee. Domesday Book reveals that claims of inheritance of this kind, either from family or predecessors, might determine title in 1086 (Roffe 2007, 167). However, once another tenant-in-chief acquired land, such claims were rarely made good. It is significant that neither Robert d’Oilly nor Miles Crispin is known to have ever made any at all. Forces other than inheritance seem to have been at work. It is the history and character of the caput of the honour that hints at these forces. We are fortunate to have a very detailed account of the borough of Wallingford in Domesday Book (GDB, 56–56v: DB Berks, B; Roffe 2009, 46–51). In many respects Wallingford was like many another borough in the late 11th century. Much of the land therein was held by the king, but by no means all. Just about all the great lords of the surrounding area, in both Berkshire and southeastern Oxfordshire, held tenements which were attached to manors in the countryside (Roffe 2009, 33–40). The data, in conjunction with contemporary and later evidence, can be used to construct a fairly detailed topography of late 11th-century Wallingford. Domesday Book also provides us with an account of the social structure of the borough and here we find differences. What characterizes Wallingford is the high proportion of royal ministers in the population. A large number of people were engaged in royal service and held their lands ex officio (Roffe 2009, 38–40). Foremost among these, and one of the biggest landholders in the borough, was Miles Crispin, the lord of the honour of Wallingford. He seems to have been the constable of the castle which had been built for William the Conqueror by his father-in-law Robert d’Oilly. To what extent he lived in a tied cottage, as it were, is not explicit (constables may not have had hereditary right to their office and the lands that went with them until the 12th century). However, it is clear that the bulk of his recorded land in the borough had been ministerial in 1066. Domesday Book records that ‘King Edward had 15 acres on which housecarls dwelt; Miles Crispin holds them, they do not know how; 1 of them belongs to Long Wittenham, a manor of Walter Giffard’ (GDB, 56: DB Berks, B1). Housecarls were the personal bodyguard of the king and

the reference here seems to have been to a garrison. The fact that [the burgesses of Wallingford] did not know under what terms he held suggests that Miles did not have full rights to the land. Apart from King Edward the Confessor, his predecessor in 1066 is not named. There can be no doubt, though, that it was Wigod to whom he owed most of his lands through his wife and mother-in-law. Wigod is styled ‘of Wallingford’ in the Buckinghamshire folios of Domesday Book (GDB, 150; DB Bucks, 23,7;12;33) and so he has become known to posterity. It is nevertheless clear that he was not lord of the borough before the Conquest: Wallingford was in the hands of the king as Domesday Book makes clear (GDB, 56: DB Berks, B1). Wigod, too, then, was evidently a minister. He may himself have been a housecarl. His son Toki certainly was (GDB, 129: DB Middlesex¸ 7,8). He was close to the king: in one early source he is said to have been a kinsman of Edward the Confessor and a pincerna, a butler (Waltham Abbey, 1). As Dr Keats-Rohan has suggested, Wigod was most likely a staller, that is a species of military governor who stood in for the earl (Keats-Rohan 2009, 57). As such we might expect much of his land to have been akin to what was known before the Conquest as ‘loanland’. Land of this type was an estate that was granted for a life or term of lives in return for service of one kind or another (Roffe 2007, 153, 173). Wigod’s fee was in effect appurtenant to this office. This, at least, seems to have been a characteristic that informed the subsequent history of the honour. As we have seen, in the first two generations inheritance played a part, if only a part, in the descent of the fee. But from the mid-12th century, the crown kept a close oversight. On the retirement of Matilda and Brien fitzCount, the honour was seized by Henry of Anjou, later King Henry II. Richard I gave it to his brother John, and thereafter it was regularly granted to the heir apparent of the king as an appanage (Keats-Rohan 2009). Herein, it must be suspected, lies the key to the special status of the honour of Wallingford in the 12th and 13th centuries. It was distinguished from other honours because it was always considered to be akin to a royal estate and was subject to the special procedures that they enjoyed. The honour and defence of the middle Thames valley The honour of Wallingford, then, was unusual in so far as it was an honour. That is not to say that it was unique. The pre-Conquest office of staller was regularly translated into constable after 1066 (Mack 1984). Most famously, Esgar the staller was the predecessor of Geoffrey de Mandeville and conferred on him title to his extensive estates throughout southern England. His office was based on London and, significantly, Geoffrey became constable of the Tower. However, the reservation of rights over the associated fee was more unusual. The closest parallel to Wallingford is perhaps the honour of Peverel of Nottingham. The defence of Nottingham and the surrounding area was entrusted to Earl Tostig, King Harold’s brother, and then Earls Edwin and Morcar before the Conquest. William Peverel was in possession of many of their lands in 1086 30

Figure 4.1 The honour of Wallingford c. 1129 (© Katharine Keats-Rohan, drawn by June Strong).

and those of their thegns, a substantial number of whom were still in place. Again like Wallingford, the terms of tenure did not change markedly. William was constable of Nottingham Castle, but much of his land was said to be held in custodia, that is, ‘in custody’: he did not have full rights to it. His honour passed to his son William Peverel II in 1113, but he subsequently forfeited and the honour reverted to the crown. Thereafter, it too became a royal appanage (Roffe 1997; Marshall and Foulds 1997). From at least the early 10th century Nottingham had occupied a key strategic position. Situated on the Trent and straddling the main road between the Midlands and Yorkshire, it was the fulcrum in the relations between the kingdom of the English and the bandit country beyond the Humber. Edward the Elder had fortified it in 918 and built a second borough across the Trent in 920 (ASC, 104). The defences of the English Borough were remodelled in the mid-10th century and Nottingham probably became the headquarters of the defensive alliance of the Five Boroughs of the Northern Danelaw (Roffe 1997, 25–35). Nottingham was the Balham of the North, as it were, and William Peverel’s control of it in 1086 was but the latest manifestation of provisions to garrison and defend it. The honour of Wallingford, it would appear, performed a similar function in relation to the borough of Wallingford. The strategic importance of the borough in the AngloSaxon period is well known. We first meet it in the Burghal Hidage where, rated at 2400 hides, it was, with Winchester,

the largest fortress in Wessex. The surviving earthworks remain a witness to the fact to the present day. Situated on the Thames on the border with Mercia, it played a key role in the defence of the kingdom against Viking attacks in the late 9th and early 10th centuries (Roffe 2009, 41–5). It would, of course, be fatuous to claim that the honour goes back to that period. However, it was evidently somewhat earlier than 1066. The origins of Wigod’s family do not take us far. His name is Danish, as was that of his son Toki, and so he or his father may have been one of the men of the Danish King Cnut. By contrast, his daughter’s name Ealdgyth may suggest English antecedents. Either way, there is no evidence that his family had held the land for a long time. Far more suggestive is the distribution of the lands of the honour itself (Figure 4.1). Although they were scattered over nine counties of southern England, there is a significant core – 40 manors out of 135, some 30% of the total – in Oxfordshire east of the River Thames (Keats-Rohan 2009, 60). We perhaps have here a clue to an earlier date for the honour. It is this area, the Oxfordshire Chilterns, which appears to have been part of the territory of Wallingford at the time of the Burghal Hidage (Figure 4.2). I have noted elsewhere that the 2400 hides that were assigned to Wallingford in that document cannot have been coterminous with the Berkshire of Domesday Book (Roffe 2009, 41–5). At least 1000 hides of the later county belonged to the borough of Sashes, an island in the Thames at Cookham. 31

Figure 4.2 Yhe territories of the burhs of Wallingford and Sashes at the time of the Burghal Hidage (© David Roffe, drawn by June Strong).

Wallingford must have looked across the Thames to Oxfordshire and, indeed, it is precisely this area in which the vast majority of the borough’s contributory manors – those rural estates with tenements in the borough – were situated. Does, then, the location of the core estates of the honour of Wallingford indicate that the honour pre-dates the formation of Berkshire and Oxfordshire as constituted in Domesday Book? It may well do so. Unfortunately, the chronology of shiring is not well understood. No date, as yet, can be assigned to the formation of the honour, but we must suspect that it is significantly earlier than 1066. Robert d’Oilly, and Miles Crispin, after him, was heir to much of this complex of estates through Wigod his antecessor. As at Nottingham, it seems clear that he was also heir to his office: as elsewhere the role of the staller before the Conquest translated easily into that of the constable of the castle after. Held with powers akin to those of the earl, the lands were administered separately from the shire. The liberty of the honour of Wallingford was the result.

best bet for its location is somewhere on the castle site, probably in the vicinity of the site of All Saints church or possibly the royal free chapel of St Nicholas. But, if the king had a residence, did Wigod have a separate establishment? Was there a second hall? If Wigod stood in for the earl, then the answer is probably ‘yes’: a comital manor was one of the elements that usually made up the complex of estates of which the borough was a part (Roffe 2007, 109–43). It may, then, be one of the non-customary properties in Wallingford held by the king in 1086 (GDB, 56, 56v: DB Berks, B). If so, there is a good chance that it is later represented by one of the properties leased by the borough, for the 1155 charter conferred the interests of the king in Wallingford on the burgesses. Those on the south side of the western end of High Street are possible sites if the open space of Kinecroft is indicative of a high status site (Pedgley below, p. 231 and Figure 13.4). If Wigod had a hall, Miles Crispin must have had one likewise. As constable of the castle, it might be supposed that he conducted all of his business in the great hall. But precedent is against that idea. In Nottingham the constable’s hall was not in the castle but in St James chapel within what had been Earl Tostig’s estate which became the French Borough after the Conquest (Stevenson and Stapleton 1895, 43; Roffe 1987, 170–1). Likewise in Lincoln, the constable’s honour met in Bardolfshalle in the Bail rather than the castle (Hill 1948, 105). Was there another high status site in Wallingford? Precedent would tend to suggest that we should be looking for a seigneurial

The honour and the borough Before concluding we can turn briefly to how the honour was represented in the borough of Wallingford. In the past it has been argued, rightly, that there was something like a palace in Wallingford before the Conquest (Roffe 2009, 41; Keats-Rohan 2009, 60). The borough was strategically important enough to warrant a major royal residence. The

32

centre outside the castle. What evidence we have is late and incomplete. In 1348 the constable John Anesty lived at a house, subsequently known as ‘Anastyes Place’, adjoining St Peter’s church, close to present-day Bridge House (BRO, W/TCb 11; Boarstall, no 806). Somewhere in Clapcot, immediately to the north of the castle, is also a possibility: the vill was held by Miles Crispin in 1086 and officers of the castle subsequently held extensive lands there (GDB, 61v: DB Berks, 33, 3–4; Keats-Rohan 2009, 61). Its designation as a liberty, however, appears to be late, probably no earlier than the 19th century. With the honour in the hands of the king from the mid-12th century, the honour seems to have held its court in the castle.

probably usually, were. But that is not to say that they were drawn on a new canvas. It has been recognized for many years that the first generation of Norman castles were usually built on high status English sites; archaeological excavation has demonstrated a degree of continuity of structure and function in some of them (Roffe 2014). Now it is becoming clear that the honour too might perpetuate pre-Conquest institutions. The honour of Wallingford is a good example. This should not surprise us. The Norman settlement was effected through English law (Roffe 1990). It is therefore not surprising that the newcomers adopted English solutions to the problems that confronted them where they fitted the bill. William the Conqueror and his followers were aware of that age-old wisdom: if it ain’t broke, don’t fix it.

Conclusion In conclusion, we have come rather a long way from the odd references to the honour of Wallingford in the Assize of Clarendon, but the journey has been instructive. Honours, and the castles that went with them, in Wallingford as elsewhere, have usually been seen as an exclusively Norman phenomenon. And so they often,

Acknowledgements I am grateful to Michael Fradley for drawing to my attention the reference to Wallingford in the Assize of Clarendon and David Pedgley for references to the constable’s house.

33

5 ‘MOST SECURELY FORTIFIED’: WALLINGFORD CASTLE 1071–1540 K. S. B. Keats-Rohan Abstract The surviving ramparts of Wallingford Castle betoken a once formidable castle, ‘most securely fortified’, but almost nothing of its walls remain. Recent archaeological investigation, both non-invasive and invasive, has revealed rich stratigraphy, going back in places to the late Iron Age. A two-volume history of the town that made some use of documents to produce a useful narrative of town and castle was published in 1881. Even so, it fell far short of a narrative of the castle as both a local and a national institution and the roles it played through the centuries before its demolition, and gave virtually no account of its physical evolution as a fortress and residence. This paper attempts to make good both deficits by exploiting a range of medieval documents to reconstruct the stages of the castle’s evolution and its role in national affairs, as well as in the local economy, discussed respectively in the first two and the fourth sections of the article. The third section explores for the first time unpublished material, principally annual audits and building accounts, in order to understand the castle and its constituent buildings. It incorporates the archaeological insights summarized in Chapter 2 of this volume, as well as the insights from the study of Tudor eye-witness evidence in Chapter 6, and the landscape analysis of Chapter 7.

the importance of Wallingford Castle in its day, but it has generally been reduced to a passing remark or footnote, partly, perhaps, because too little stonework survives on which to base anything rounder (Figure 5.2). Despite the extensive archaeological study of Wallingford Castle of recent years (Christie and Creighton 2013), summarized by Creighton and Christie in this volume, we still know far too little about the castle as a building or an institution, yet this need not, and should not, be the case, as this essay will demonstrate. Very little documentary evidence has been published, being limited to a single annual account for running the castle from 1296–7 (Midgley 1945, 132–5); widely scattered incidental references in chronicles and the odd laconic entry in Exchequer records such as the Pipe Rolls, or Chancery records such as the Close Rolls, are also available. There are eye-witness accounts from the 16th century attesting the extent and magnificence of the castle, even in its then near ruinous state. Documentation, including partial site plans, and the observations of later owners of the site, survives from the post-demolition period. Brown, Colvin, and Taylor’s brief entry on Wallingford in The History of the King’s Works (HKW ii, 850–2) indicated that there were additional medieval documentary sources which might shed light on the castle as a building, but they were unable to make much use of them within the scope of their own work. Taken together, a rich array of documentary evidence exists which can and must be exploited in order fully to understand the castle’s history. Both archaeological and historical approaches are essential, and whatever the difficulties, they will be mutually informative (cf. Pantin 1958). As will be seen, that is very much the case here. The aim of this and the two following papers is to bring together all this disparate material in order to recreate as far as possible the architectural history of the castle, which is inextricably linked to its evolution as a building and an institution. Taking its cue from Brown and Colvin’s remark that there is ‘no mention in the 12th century of a stone keep such as a castle of this importance might be expected to possess’ (HKW ii, 850), this paper will cover the medieval period, taking the opportunity to set the scene by comparing the evolving situation at Wallingford with the general line of castle development in England, with particular attention to stone keeps, before introducing previously unused medieval documentation. In so doing it will show that the castle played a role of national as well as local importance throughout its existence. John Lloyd’s paper looks at mid-Tudor evidence, and Judy Dewey brings the whole to completion in her scrutiny of the postdemolition period, culminating in a new proposal for the

Introduction But with no trifling thoughts I seek you heap Of ruined walls, to superficial eye Off’ring no obvious lore, but page confused, Of desolation, yet amid whose scenes, The engineer can scan the perfect site Of Castle, Chapel, Dungeon, Dike and Keep. And fancy with a gaze more keen than his, Surveys, rebuilt those walls of ancient pride, And with a motley host of soldiers, priests, And captives quick repeopled! Charles Allnatt, On Wallingford Castle, c. 1850 There are two certain dates in the history of Wallingford Castle. It already existed by 1071, when the Abingdon Chronicler tells us that his abbot was a prisoner there; and it was demolished, stone by stone, by order of Oliver Cromwell in 1652. Enough is known of the intervening period for a narrative of the castle’s history. Good evidence exists about both its importance as a fortification and its use as a prison during its existence, but little further has been known of its architectural development, its administrative role, or how it impacted on its surroundings in socio-economic terms during its lifespan. No one doubts

34

35

Figure 5.1 Composite overview of geophysical survey on the Wallingford Castle site (TWHAS, Burh to Borough Project), incorporating earthwork survey by Michael Fradley (Burh to Borough Project) (Google Earth-based image © 2015 Infoterra Ltd & Bluesky, compiled by Gerard Latham)

only in the case of urban centres, where they afforded the means both to dominate and to govern each county, providing a secure working environment for the sheriff in charge of law and order, justice and tax collection, and the garrison, overseen by the constable. Castles permitted control of the mints and the urban markets. Of the 36 castles ordered to be built by William I, 20 were placed in the leading boroughs of each county, with Wallingford being among the very first (Figure 5.3). These were the ‘castles of conquest’ and they were initially about control, of both the people and the machinery of governance and the economy (Harfield 1991). Twelve of them were located in towns with Roman roots, and eight in English burhs, including Wallingford and Oxford. In each case the new castle adapted one or more of the existing ramparts or walls in the construction of one or more of its sides, usually in a corner position close to the river and with ready access to and control over the river bridge. The evidence of Domesday Book, collected during 1086, shows that in most cases a sizeable number of properties were destroyed to make way for the castle and its perimeter, evidence that these castles were intended to establish a secure and commanding position within the towns (Pounds 1990, 57–8). William is known to have ordered the construction of castles in London – the precursor of the later Tower of London in the east, and another in the west, later replaced by the baronial castles of Baynards Castle and Montfichet (Impey 2008a, 22–6) – and Winchester in 1067 on account of their importance in the kingdom. Wallingford was another very early choice. The evidence of the Abingdon abbey Historia is that Wallingford, Oxford and Windsor, in that order, were built early in the reign, Wallingford Castle being already operational as the place of imprisonment for the abbot by 1071 (1.1). The reasons for establishing a major royal castle at Wallingford are not far to seek. The strategic importance of its location at the crossing of a major river that had once been the boundary between the rival kingdoms of Wessex and Mercia, was the reason for its creation as a central node in the burghal defence system set up in the 880s by King Alfred to keep the Danes out of Wessex. The significance of its location in the middle of the unified English kingdom was noted by the AngloSaxon Chronicler at the time of the Danish incursions in 1006, was capitalized upon by King Cnut (1016–35), and had not been lost upon William and his men in the brutal campaign that followed the battle at Hastings in October 1066. We know from the Norman historian William of Poitiers that Archbishop Stigand of Canterbury arranged for the submission of the English to William when the latter’s army reached Wallingford in November 1066. These castles along the Thames and in central Wessex, most notably Winchester, remained central to royal civil and military administration for the next two centuries (Keefe 1990; Brown 2003; HKW ii, 854–5; Keats-Rohan 2015). Stigand must have had the support of Wigod of Wallingford, an important minister and likely kinsman of Edward the Confessor, who held a role similar to that of quartermaster in relation to the English army (fyrd) up to Hastings (Keats-Rohan 2009, 56–7). As both archbishop of Canterbury and also bishop of Winchester,

Figure 5.2 Doorway in a surviving portion of the Priests’ Lodging in the middle bailey at Wallingford Castle (photo: Judy Dewey).

plan of the entire circuit of the castle precincts (see Figure 7.1). Calendars of cited documents are appended to each of these papers (Document List 1 to this one, 2 for Lloyd, and 3 for Dewey). The reference numbers in the text refer to an entry in the appended calendars. While they do not aim to be a complete set of all documentary references to the castle, those attached to this essay can be read from the beginning as an unfolding story of the castle told in the words of contemporaries. The following account consists of four sections. In the first, the foundations of a study of the castle, its role, format and institutions, are laid; the second is a historical narrative from the 11th to the 16th century; the third section presents the detailed evidence for understanding the layout of the castle and its evolution as a building and an institution, followed by a fourth section assessing what light the documentary evidence sheds on the socioeconomic history of the area. A concluding section summarizes the three core phases of castle development and assesses the castle’s national importance. In the beginning: foundations of a royal castle and honour 1066–1272 Wallingford and Conquest-period strongholds: role and organization Castles were built all over England in the wake of the Conquest. Their implantation was deliberate and controlled

36

Figure 5.3 Castles of Conquest, (after History of the King’s Works, © Queen’s Printer and Controller of HMSO 1963 redrawn by Debbie Miles-Williams).

Stigand possessed houses in Wallingford, some of them pertaining to the Winchester-owned manors of Brightwell and Sotwell, to which were attached two of Wallingford’s churches, St Lucian and St Leonard (Roffe 2009, 37–8). Although not ‘lord’ of Wallingford, Wigod will have had his own hall there, probably with an associated church. He was the man in charge of the king’s hall, and also in charge of what appears to have been the only known ‘garrison’ of royal housecarls, specifically mentioned in Domesday Book as occupying 15 acres within the town. This unique reference may furnish the strongest extant clue as to the identity and function of the town

of Wallingford, which was large enough to protect the surrounding population who maintained, and if necessary manned, its defences, and also to house a royal mint, as well as the tradesmen and farmers supplying a flourishing market. There is some uncertainty as to the exact nature of the housecarl’s role, but there is agreement that it was at least partly military and was certainly closely associated with the person of the king (Hooper 1984; Campbell 1987, 204). The Domesday evidence of a strong association of royal ministers with Wallingford, both in and after 1066, ably demonstrated by David Roffe (2009), shows that its role as royal stronghold and administrative centre

37

continued from the Old English into the Norman period. The essentially military origin of the special relationship between the king and Wallingford has been noted by Campbell (1987, 204), Astill (1991, 109), and Lawson (1993, 180), who suggested that the housecarls were stationed at Wallingford to back up the authority of the king’s officials in a key area dominated by royal holdings. This was reflected in the post-Conquest honour, held of the king in chief for the service of just over 102 knights in the king’s host or in defence of the castle. At its core was the territory originally controlled by the Saxon burh in Berkshire and Oxfordshire, with a number of important clusters in other counties, including Buckinghamshire, Wiltshire, Bedfordshire, Surrey and Middlesex reflecting the importance of emerging royal centres in Westminster/ London and Windsor (Keats-Rohan 2009, 52–63; Roffe, above, pp. 30–2). The Domesday account is also the earliest ‘official’ record of Wallingford Castle’s existence, noting that eight hagae (enclosed burgage plots) had been destroyed when it was built. This relatively low figure probably signifies the interconnected facts that the north-east quadrant of the town occupied by the castle was an inner stronghold, formerly the site of the king’s hall, possibly even a palace, and associated buildings, and that few customary (tax paying) tenements abutted its boundaries. We know from excavations carried out by Nicholas Brooks in the 1960s, Robert Carr in the 1970s and most recently by the Wallingford Burh to Borough Research Project team, that pre-1066 tenements existed on the east side of Castle Street in the area south of the North Gate, and that there was Saxo-Norman occupation of the site on the south side of the castle in the area of the later middle bailey (Christie and Creighton 2013, 88–106, 183–202, summarized this volume, pp. 11-14 above). All this suggests that much of the burh’s north-east quadrant was an area of élite occupation with dwellings attached to or serving a major royal administrative and ‘military’ centre (Creighton 2002, 140; Liddiard 2005, 19; Roffe 2009, 40–1). The relatively limited subsequent development of the area is also striking. One might wonder, especially in view of the status of the castle chapel of St Nicholas (below, p. 53) and the considerable liberties attaching to the honour, whether the area originated as an intramural burh specifically linked to the king in the Saxon period (cf. Williams 2003, 38). As we shall see (pp. 52–4), the four gates of the town were controlled by the royal castle, which invites comparison with Lincoln, where the upper city, known since at least 1163 as the Bail, was regarded as the outer bailey of the new Norman castle (Stocker and Vince 1997, 223). The best evidence for this is the extraordinary role played by Wallingford in the civil war of Stephen’s reign, when castle and town spent many years between 1139 and 1152 under increasingly close siege. The castle garrison could not have held out so considerably longer than any other castle of the time had it not both controlled the town gates and had the support of the townsmen. Certainly, the link with the enormous extramural hamlet of Clapcot (now in the parish of Brightwell-cum-Sotwell) went far beyond the normal ‘castle fee’, i.e. the urban enclave occupied

by a castle (Pounds 1990, 211). The creation of these ‘castle fees’ may underlie some of the apparently largescale destruction of burghal tenements in places such as Norwich, which lost 98 tenements. Norwich is a striking example of an early castle, built in one of the largest towns in the country. Part of the early castle was built over a Saxon cemetery, and destroyed several late Saxon buildings. It began as an earthwork enclosure but was transformed during the 1090s by William II, who had this filled and then heaped over to produce a huge flat-topped motte on which was built a massive square stone tower keep, its foundations secured in the motte with baulks of timber (Goodall 2011, 109). Unusually, Norwich does not share the profile of high status Anglo-Saxon antecedents found at many castle sites in county towns, including Winchester, Oxford, and, most recently, Northampton (excavations of April 2013, Chapman 2014, 26–9), and Wallingford. Though much of the site at Wallingford remains unexplored, evidence of occupation or activity from the late Iron Age and the mid-to-late Saxon period has been found (Christie and Creighton 2013, 196–7). It seems likely that at Wallingford the élite precinct even in 1066 occupied most of the area later defined by the 19th-century municipal boundary on the south (High St) side (see Figure 7.15), a theme explored in detail by Dewey, pp. 146–50, below. The area covered by the developed castle, and its later Civil War extensions, is estimated at 26 acres or 10.5 hectares. It can probably be identified with the reference in Domesday Book to 15 acres where the housecarls had dwelt. The castle precinct was, of course, extra-parochial but intruded into the vast parish of All Saints, which flanked it on the west and on the east. The whole parish was intimately linked to the castle, embracing the two-manor hamlet of Clapcot which stretched northwards to Shillingford Bridge and westward to the parish boundary of Sotwell. Miles Crispin, constable of the castle from c. 1084 to 1107, held the two manors of Clapcot as one, as of the honour of Wallingford in 1086. These were later known as Clapcot Manor and Rush Court and were held by men connected to castle service. Clapcot, with the part of the parish of All Hallows (as All Saints had been known from the 16th century) within the town, including the extra-parochial precincts of the castle, was reckoned as around 877 acres in 1923; including all the Wallingford parishes of that date the total was 1257 acres (VCH Berks 3 1923, 517). Flanking the castle on the east, alongside the river Thames, and effectively forming an outer bailey on its single-walled side, was Queen’s Arbour and, further north, King’s Mead; together these provided winter hay for animals, as well as an area from which the castle’s complex water systems could be managed (see below, pp. 72–3, and Frontispiece). The early castle layout. c. 1067–1135: the stone circuits and Great Tower The heart of a castle was its walled and ditched inner bailey, overlooked by the protective tower on the motte (Hulme 2007, 224). Wallingford’s inner bailey, whose north wall follows the later Anglo-Saxon rampart, was 38

Little Wittenham

Bridge End R.Thames

Warborough

Berrick Salome

Shillingford

Ewelme Brightwell cum-Sotwell North Moreton

Wallingford Crowmarsh Gifford

South Moreton Nuffield Aston Tirrold

North Stoke

Cholsey

not visible visible

0

3km

Figure 5.4 Viewshed from Wallingford Castle site. (Image © Wallingford Burh to Borough Research Project.)

extensive – an irregular polygon, roughly 480ft (146.3m) north to south and 400ft (121.9m) west to east. With a motte set in its south the fortified complex commanded views of the Thames upstream and downstream (Christie and Creighton 2013, 164) (Figure 5.4). As revealed in a plan of 1548 and discussed in detail below (see Lloyd pp. 126–30, with Figure 6.1), the southern arc of the motte formed part of the walled circuit, projecting into the waterfilled moat. It is likely that the outer perimeter in 1086 reached much further south, probably to the same line as the maximum extent occupied by the later castle, thus forming at least a second bailey. Carr’s 1972 excavation of the middle bailey and its rampart indicated that the area was cleared of earlier occupation debris sometime in the late 11th or early 12th century before the bailey rampart and wall were built, the latter occurring by or before the mid-12th century (Christie and Creighton 2013, 188–90, 194). If the perimeter were on the later parish boundary to the south c. 1066, then the site of the third curtain wall (c. 1230x50) was already adumbrated by 1100. The boundary on the western side is less certain: Brooks’ excavations in the 1960s demonstrated that the castle’s final enlargement entailed the shifting further west of the town’s North Gate and the destruction of a number of intramural properties on the street here (Christie and Creighton 2013, 88–106). At the time of the demolition of the church around 1643 (Hedges 1881, ii, 333), the churchyard of All Hallows itself lay just outside the outer west moat of the castle

and had presumably determined the furthest extent of that moat when it was built (Dewey, p. 147 below). We know that there were ultimately three baileys, the third of which divided the broader outer one proposed here. The outer perimeter would initially have been enclosed at least by a palisade because of the need to protect the main entrance, which stood close to the bridge and East Gate of the town. There are a number of documentary proofs of the location of the main entrance. A chronicle of Stephen’s reign referred to Wallingford Bridge as the ‘key’ to both castle and town in 1152 (1.5); more formally, a borough deed of 1294 gives the boundary of a messuage as lying in St Peter’s parish, 22ft (6.7m) in width and in length stretching from the high road (High St) to the outer barbican of the castle ditch (CMACC, 155; Dewey 2009, 24–5). Entrances to castles built against one of the inner ramparts of a town normally faced into the town – London, Exeter, Bristol, York, Northampton, Nottingham, Leicester and Rochester are all examples, with Winchester being a rare exception (Pounds 1990, 213–14) (Figure 5.5). Wallingford Castle’s entrance is likely to have faced into the town, although the form of the original gatehouse is completely unknown. Several castles had gates or towers adapted from or modelled upon English burhgeats, such as St George’s Tower at the west end of Oxford (Figure 5.6). Variety of form is evident from the beginning. Motte and two bailey castles from this period are also known at Huntingdon, Chester, Lincoln (which, like Lewes, had 39

Figure 5.5 Norman gatehouse of Rougemont Castle, Exeter (photo: Oliver Creighton).

Figure 5.6 Late Saxon, St George’s Tower, Oxford (photo: Katharine Keats-Rohan).

two mottes), and Windsor (Figure 5.7). Motte and one bailey castles were built at Warwick, York, Cambridge and Nottingham. London, Colchester and Canterbury had great towers. Sixteenth-century observers of Wallingford Castle attest the two outer walls and ditches (comprising the middle and outer baileys), and the presence of a great tower (‘keep’) on top of the motte, the whole built in stone (Lloyd 2.2–3).1 Many castles are thought to have

quarters. Many were later redeveloped into something more impressive by the addition of a stone revetment of the motte and the addition of flanking towers, turning them into small inner wards built on or round the motte. It has generally been assumed that given the size of Wallingford Castle and its motte, and the early date of its foundation, that it must have had a shell-keep. The documents to be considered hereafter, and the Tudor Survey presented in John Lloyd’s paper below, do not support this view, instead clearly indicating that the keep was of the quadrilateral type with crosswall favoured by the Normans until the mid-12th century. How can we square the circle? Although all too few have been thoroughly explored, it is increasingly clear that the composition of mottes is complex and frequently deceptive; English ones may well, as Davison suggested, have been the product of early post-Conquest Norman experimentation and innovation. The earliest for which there is any sort of evidence may be that at Hastings; the Bayeux Tapestry’s representation of the fortification at Hastings clearly shows a motte, probably one that is being raised against a prefabricated wooden tower (Higham and Barker 1992, 60, 154–6). Some mottes have been found to contain within them the foundations of wooden towers, such as Ascot D’Oilly, Oxon (Jope and Threfall 1959), where the motte and the tower were raised in tandem. Others were added later by piling earth around a stone tower, as happened in castles of the early 12th century at Farnham, Surrey, where the c. 15.24m square tower had a central well shaft (Thompson 1960; Riall 2003). Farnham, where the lowest part of the stone tower lacked doors or windows, is a particularly interesting case of the interrelationship between tower and protecting motte since it indicates that from the start the motte was designed to strengthen the stone tower, as may also have been the case at Mileham, Norfolk, where c. 1100 a motte was built up around a stone keep, with two baileys and a further banked enclosure (Liddiard 2000, 79–82). The square keep at Farnham was dismantled c. 1155 to motte-top level and the site converted into a shell-keep (Figures 5.56, and 5. 57, p.

  What form the keep at Wallingford took is of central importance to understanding the castle architecture and so demands an excursus here. The majority, by a ratio of 4:1, of post-Conquest castles built before 1200 were motte-and-bailey castles, rather than ringwork castles (Eales 2003, 45; Liddiard 2005, 17), but most of these had probably started as ringworks (Davison 1969; English 1995). There are no recorded instances of a motte castle being built in England after the accession of Henry II in 1154 (Pounds 1990: 21). Most mottes are artificial mounds created from the spoil thrown up by the digging of the ditch that becomes the moat encircling the motte. At the top of a motte would be a wooden palisade, sometimes enclosing a wooden tower that both acted as a watch tower and could also provide shelter if required. A freshly made motte would not be sufficiently stable to bear the weight of masonry, so any stone tower on a purely artificial motte would certainly be of later build. A stone wall round the top of the motte, replacing the original wooden palisade, which might have wooden buildings placed against it, was known as a shell-keep; these, again, are of later build, and could evolve into complex stone structures, as at Launceston in Cornwall and Lincoln (Goodall 2011, 108). Other examples are Arundel and Berkhamsted. The considerable limitations of the shell-keep were pointed out by Cathcart King (1988, 64–5) who observed that, unlike any other type of keep, a shell-keep did not house – though it may have protected – its lord’s living

1

40

Figure 5.7 A bird’s-eye view of Windsor Castle in 1658, by Wenceslas Hollar, shown before Hugh May’s reconstruction of the Upper Ward (image: public domain).

been built initially in timber, and subsequently rebuilt in stone over the next 10 years. Castle building in stone was, however, known from relatively soon after the Conquest, seen in early gatehouses such as Exeter (1068) and Bramber (1070s), and full castles from the 1080s, such as Durham, London, and Colchester. The lack of upstanding early medieval work at Wallingford presents considerable

challenges in trying to establish its early form, but we do have documentary hints to supplement the archaeological information. There are good indications that Wallingford Castle was fully in stone before the end of Henry’s reign in 1135. The Book of Fees records a number of serjeanties attached to the castle in 1212 (p. 50, see below) and, significantly, one of them was held by a mason for the service of repairing the castle walls – this a service held de vetere, meaning that it was acquired before the death of Henry I (HKW I, 27). Furthermore, writing during the reign of Stephen (1135–54), the anonymous author of the Gesta Stephani described Wallingford Castle as impregnable when it was besieged in 1139 (1.4); in truth, he described all the castles he mentioned as impregnable, but of the siege of 1152 he took care to note that at Wallingford the bridge formed the master-key to the town and to the castle on that side of the river (1.5). He also detailed the building of Stephen’s anti-castles, notably at Crowmarsh, on the other side of the river (the substantial remains of which were traced in excavation in 2011: Laban et al. 2013; discussed in Christie and Creighton 2013, 206–7), and chronicled the various sieges to which Wallingford was subjected without

104). After that date the motte at Farnham was completely revetted in stone. Such towers often comprised a single chamber on each of between three and six storeys, a layout known as a solar hall keep (Gravett 2003, 16). In other cases, the enclosed stone tower may have been of late Saxon origin, such as Sulgrave, Northants, and Groby, Leics (Creighton 1997, 22–5). These discoveries, almost all of which belong to the first half of the 12th century, prompted the label ‘fashionable enmotted stone tower’ from Coulson (2003a, 200). The exception is Lydford in Devon, where a freestanding stone tower was enmotted and converted to a two-storey keep in the 13th century, when it belonged to Richard, earl of Cornwall, (HKW 1963, 733–4; Saunders et al. 1980). It was subsequently used as the prison of the stannaries, having been incorporated into the Duchy of Cornwall in 1337. The diversity of forms a motte could take in relation to a tower was highlighted in Thompson’s (1961) typology of motte substructures, dividing them into four main types, two of which were further subdivided. Relatively unknown when Thompson discovered Farnham’s secret, enmotted keeps are now an established phenomenon in castle studies (Cathcart King 1988, 47–59). Kenyon (2005, 40–5, at 45) has pointed out that most of the identified examples of enmotted keeps are English.

41

Figure 5.8 Stephen’s anti-castle at Crowmarsh (an oval ringwork) under excavation (photo: Judy Dewey).

surrendering (Figure 5.8). In fact, alone of all the castles involved in the war Wallingford never surrendered or changed sides. This evidence lends itself to the view that the castle was in a meaningful sense ‘impregnable’, endowed with stone circuits, probably on at least two baileys as well as a mantlet wall or revetment round the motte by 1139. Certainly it is unlikely that any major works in stone could have been achieved during the years from 1139–52 when Wallingford, which depended on stone supplies no little distance away, was almost constantly under siege. The Pipe Roll for 1182–3 (PR 29 Henry II, 139) records work ‘on the bridge and motte of Wallingford Castle’ (in operatione pontis et mote castelli de Warengeford), carried out at a cost of £7 12d. The reference to a bridge may have been to the drawbridge into the inner bailey from which the motte was accessed, or perhaps to the drawbridge at the entrance to the Great Tower that rose from the motte (discussed in detail below pp. 91–5). Equally, it may have referred to a bridge over a moat, since mota means both motte and moat. Allen Brown (2003, 142–4, 153–4) pointed out that it was unusual for the king to build new castles in the 12th century, most of the sums in the Pipe Rolls representing work maintaining existing castles, some of it involving the replacement of the timber defences with stone. The sums mentioned for Wallingford are insufficient to indicate a rebuilding in stone and more likely signify

repairs and upgrading to an existing stone-clad fortress. Altogether, taking the admittedly exiguous evidence about financial investment in the castle before 1200 together with the evidence for a masonry serjeanty and the undoubted strength of Wallingford’s ‘most securely fortified’ fortress during the civil war of 1139–53, we can reasonably infer that the castle was already in stone and furnished with a masonry tower on the motte by or before the mid-12th century, although the first unambiguous mention of a great tower only comes in 1284 (below p. 92). Analysis of Carr’s excavations has shown that the wall of what was later the middle bailey was originally built in stone, some time before the mid-12th century (Christie and Creighton 2013, 194), most certainly linking the building of a stone keep directly with the king or a royal favourite (Goodall 2011, 109), Henry I’s constable Brien fitzCount (active c. 1113 to 1148) being the most obvious candidate (cf. King 1999). Such a date would also fit with Goodall’s observation that the first stone keep in a given area would have been built at a royal castle (2011, 113). Interestingly, it is in fact an isolated case regionally, since both Windsor and Oxford were shell-keeps. First generation Norman castles had few towers, confining them to the protection of curtain wall lengths and gatehouses; these early towers tended to be rectangular, probably because they fill the footprint of previous wooden towers (Goodall 2011, 42

Figure 5.9 The Great Keep at Rochester, with first-floor entrance via the forebuilding (photo: Katharine Keats-Rohan). Compare Figure 5.48.

Figure 5.10 Dover Castle, the keep and the outer curtain wall (photo: Judy Dewey).

43

Figure 5.11 Artist’s reconstruction of the Great Keep and forebuilding at Wallingford Castle, looking south. The geophysics (Figure 5.1) suggests that there may have been a paling or enclosure at the base of the motte, perhaps providing an extra storage area. Drawing by Anthony Wilder.

76). By the 13th century, the preference was for round, D-shaped or polygonal designs, although rectangular towers were re-introduced in the 14th century. The building accounts discussed below will show that since no major architectural remodelling at Wallingford occurred after the 13th century, we can accept the general outline of the Tudor plan of the inner bailey as a good indication of the original layouts of the period 1067–1135, as suggested by John Lloyd (Lloyd, pp. 132–3 below). Working from a survey of stone and timber from 1555, Lloyd surmises that the keep or great tower was a crosswalled quadrilateral, internally about 18.3 sq. m (60 x 60ft). The 1555 survey shows that the keep (like the bailey walls, built of malmstone and freestone) was entered through a forebuilding; there were at least two floors from that point, but the overall height is unknown. The contemporary bailey had four rectangular corner towers and one interval tower, all projecting into a moat. Whatever the intervening modifications, as Lloyd points out, this is certainly an early Norman layout (Goodall 2011, 108–9), showing that the earliest stone plan of the inner ward survived more or less intact into the Tudor period. The same, of course, will not always have been true of the buildings within. Despite the huge expense involved in erecting these ‘prodigy’ buildings (Goodall 2011, 115), great towers became more numerous after 1100, though only Rochester (under Henry I) (Figure 5.9) and Henry II’s keep at Dover (almost 29.9m/98ft square) came close to the palatial great towers of the sort begun at London, Colchester, and Norwich. Several great towers were built both in England and Normandy in the time of Henry I, including Bristol and Domfront, the latter having internal dimensions of 20.3m x 16.4m (66.6 x 53.8ft) (Dixon 2008, 260–65).

Figure 5.12 The walls of what had previously been seen as an enmotted shell-keep at Berkeley Castle, but now thought originally to have been a small inner ward (photo: Katharine Keats-Rohan).

44

Figure 5.13 Enmotted keep of Farnham Castle from the outer bailey. A tower built at ground level and subsequently enmotted was razed to current motte-top level on the orders of Henry II after 1155 (photo: Katharine Keats-Rohan).

The plan of most was rectangular or square, with a crosswall, pilaster buttresses and first-floor entrances (Gavett and Hook 2004, 13). In several locations great towers were added to existing royal castles, such as the ringwork castle at Carlisle, the motte castles at Newcastle, Corfe and Wareham (both Dorset); the tower at Wareham was bolstered by 1.5m (5ft) of gravel at its bottom (Renn 1960, 56). Dover (1180s) was one of the last to be built (Figure 5.10). After that, keeps were polygonal or cylindrical in form (Hislop 2010, 216–17). It is difficult to believe that such a large tower as that at Wallingford could have been associated with an artificial motte unless the foundations rested on or extended below the ground surface, rising through the motte. The well chamber certainly went through the motte. Although the keep at Norwich was built on a scale far greater than seen elsewhere, its huge size illustrates the difficulties of placing a large mass of masonry onto an artificial mound, since the tower cracked in one corner, the whole having to be rescued by the driving in of deep timber piles (Goodall 2011, 109). The Round Tower, the shell-keep of Windsor Castle, was similarly prone to subsidence (Hulme 2007, 222), and the tower at Oxford fell down in 1239 (HKW i, 77). While the tower of Wallingford was certainly smaller than that at Norwich, it was larger than most motte keeps and the motte itself considerably less substantial than would seem to be required for the support of such a large tower. It is evident from the Tudor documents that the tower was accessed via a covered and enclosed stairway which led up the motte slope to the forebuilding entrance

at first floor level; the entrance to this stairway appears to have begun at bailey level; an external stair built against a wing-wall was flanked by a sturdy tower projecting into the moat separating the inner and middle baileys (Figure 5.11). The 1555 survey shows that this forebuilding was protected by massive doors at the bottom and top of the staircase. Such an entrance is a feature of a Norman great tower, rather than a shell-keep (Cathcart King 1988, 57). Berkeley Castle, where a motte was apparently encapsulated in a stone revetment, had been thought to be the only exception, but more recent work suggests that this was not the case and that the current buildings incorporate the remains of a gatehouse great tower like that at Richmond, with the encircled mound having originally been a small inner ward (CA 305, August 2015, 38–45, at 43–4) (Figure 5.12). Could Wallingford be an unusual (because apparently larger than most), possibly experimental, case of an enmotted great tower? Without a major archaeological investigation – and all enmotted keeps have been the surprise discovery of excavation – this is a possibility worthy of consideration (Figure 5.13). There were very probably high status Saxon buildings on the site when the Normans took it over, one of which could then or slightly later have been used as a tower, possibly enmotted. If the beginnings of the tower and its motte were securely established early on, then its aggrandizement into a more major feature still, built by Richard, earl of Cornwall, becomes a possibility. The major part of the castle site was owned in the 19th century first by Dr James Blackstone, Fellow of All

45

Figure 5.14 Capped off outlet of the well on the north side of the motte at Wallingford; its current height may be an artefact of former owner John Kirby Hedges’ landscaping, but it fits the measurement of 3ft 1in that he gave the Town Council around 1890, when he said that it was 27ft deep and had contained 10ft of water in April 1860 (photo: Katharine Keats-Rohan).

Souls, and later by John Kirby Hedges, who published a two-volume history of the town in 1881. Despite some inaccuracies, Hedges’ book is a remarkably useful and judicious compendium of documentary and anecdotal information. He described his own investigations around the west side of the motte, which he believed indicated that it rested, on the three sides lapped by the inner moat, ‘upon a solid foundation of masonry several feet thick; the layers of stone sloping upwards towards the outside, and some yards beyond the keep as it now exists’. Described by Cathcart King (1983, i, 12) as ‘traces of a shell wall’), Derek Renn (1968, 139) dismissed with scorn Hedges’ ‘saucer’, as Ella Armitage (1912, 230) termed it, but he was over-hasty. Hedges owned the site and he was a very keen and competent observer. The masonry he notes may have been in part a footing for the tower; it may partly have related to a former stone mantlet wall or revetment at the base of the motte. Hedges also describes the well on the – now much eroded – north side of the motte, which ‘is constructed of worked stone, and has a smooth facing’, noting in a communication with the Town Clerk of c. 1890, the site of three other wells, including the one that was associated with the Great Hall: unfortunately, their exact location is now lost (Figure 5.14). He made three further important observations, one of them that: ‘a layer of concrete is to be found in the bottom of the moats or dykes which surrounded the Castle. This concrete formed the foundation for a thick stratum of lime, which served to keep the moats water-tight’. There is ample evidence from the constant reference to the making of mortar for guttering in the accounts to be discussed below to make this observation credible. Indeed, the moats were still water filled until Dr Blackstone traced the sluice-gates ‘on the southern side of the Moat, just below the Keep’ and drained them in the early 19th century (Allnatt 1873, 47–8; Dewey, 162 below. Hedges also wrote (ii,192):

‘We have made considerable excavations at the top of the keep, in the hope of lighting on the foundations of some of these extensive buildings [a reference to the 1555 Survey of the keep], but not the least trace of any masonry has been discovered’. This reveals that the circular flint and chalk feature at the top of the motte was another of his landscaping ventures (Armitage 1912, 230). Additionally, he and previous landowners identified an underground passage leading from the south side of the motte towards the castle ditch, as well as a passageway on the north side that may once have been associated with a postern or sallyport (Figure 5.15).2   A third important observation by Hedges is all the more telling because he wishes to disprove an opinion of Dr Blackstone. Hedges himself in 1859 traced a subterranean passage previously explored by Blackstone and pronounced ‘undoubtedly Roman’, and found that it was in fact nothing of the kind. Hedges wrote: ‘It runs from the bottom of the keep on the south side, in the valley, and takes an easterly direction, rather inclining southwards, towards the river. It is of large size, sufficiently large for a man to walk along, being about four feet high by two feet wide. [Hedges may have mistaken his figures here: a height of 4ft was no more a sufficient size for a man to walk – rather than crawl – along in 1859 than it is now]. The entire passage was effectually cleaned out and repaired up to its junction with the ditch, at or near the extremity of the Castle precincts. Beyond this point the passage could not be traced’. Dr Blackstone had written in 1820 that: ‘It appears to have no outlet from the Castle, but terminates in a staircase at each end, leading up to the buildings of the Castle; but under it there is a narrower passage, leading from either staircase at the ends, which appears like a dungeon or prison’. (Figure 5.58, p. 104) In fact, the earliest account of this feature appears to have been written c. 1712 by Richard Skirmer, a Wallingford resident, who noted that the infamous Thomas Renda (cf. Dewey, p. 162 below), ‘in clearing out rubbish, he hath likewise opened a way between two walls not far from the well, which way has been supposed to pass under the Thames & to come out at a place called the Dane Pits on Crowmarsh hill; however they had here their sally ports and this might be one.’ (Sims 2002: 7). Later in his work Hedges distinguishes between a passageway on the north side of the motte and a subterranean one on the south (1881, i, 191). Wallingford is still as full of ‘secret passage’ stories as many other old towns, but these passages are known to exist at several castles. A passage from outside leading into the king’s chamber at Nottingham Castle allowed his supporters to help young Edward III to overthrow his mother and Roger Mortimer in 1330, some seven years after enemies of 2

46

he left the English claimant, Edgar the Ætheling, behind with Archbishop Ealdred while he opened negotiations. Once the surrender had been orchestrated at Wallingford, the way was open for the final submission of Edgar at Berkhamsted (Herts), as reported in the Anglo-Saxon Chronicle. Edward Impey (2008a, 16–8) has recently made sense of the story of the surrender as told by Poitiers, who does not mention Berkhamsted but describes the postcoronation surrender of Earls Edwin and Morcar at Barking, by suggesting that the otherwise puzzling Barking episode was actually part of the Berkhamsted surrender recorded by the Anglo-Saxon Chronicle. From Wallingford, there was an easy route via the Icknield Way and Akeman Street to Great Berkhamsted (ibid, 17–9). There is not only an absence of evidence that Wallingford suffered during the Norman advance, but rather indications that it was treated with some favour. To this day a bell rings the curfew at 9.00 pm, one hour later than was permitted in any other town by the Conqueror, according to local lore (Hedges 1881, i, 197). The initial task of building a stronghold at Wallingford was entrusted to Robert d’Oilly, who soon married the daughter of Wigod. With Wallingford in hand, Robert began the castle at Oxford in 1071. The strong defensive and administrative links between Wallingford and Oxford and their shires went back to the Burghal Hidage era of the late 9th century and remains a theme common to the history of both throughout the period covered here. The motte and bailey castle at Oxford (Norton, this volume), was of a lesser order than the one at Wallingford, which achieved the status of developed castle in the 13th century, when Oxford, still partly in timber, was already becoming derelict (VCH Oxon 4 1979, 296–300). By the time of Robert’s death in the 1090s, he was also credited with the building of a great stone bridge (Grandpont) and causeway at Oxford, one of the earliest in Northern Europe (Durham 1984, 87–95; Dodd 2003, 53–4). Like other Norman barons, such as William fitzOsbern and later Robert de Bellême, Robert d’Oilly probably had an interest in building and no little skill in site selection and building design. It may be significant that his close friend Roger d’Ivry, who was his tenant in his Oxfordshire barony of Hook Norton and who himself held several manors of the king in chief, was named from Ivry-la-Bataille (Eure) in Normandy. Ivry was the location of one of the earliest Norman stone keeps, built c. 1040 under the direction of the celebrated Lanfred architectus, as we learn from the chronicler Orderic Vitalis (Orderic, iv, 290). It has been suggested as the prototype for the Tower of London (Figure 5.16) and Colchester, though unfortunately little of the original survives. Robert d’Oilly was not a baron of the first rank. Even as constable first of Wallingford and then of Oxford, he could not rival the likes of Roger de Montgomery, builder of Shrewsbury, Hen Domen and Arundel Castle, whose man he was, or William fitzOsbern, responsible for the origins of castles at Monmouth and Chepstow (Monmths), Berkeley (Gloucs), Wigmore, Hereford and Ewyas Harold (all Herefordshire). Nonetheless, he was valued for his practical skills by the king, and he was obviously seen as well-equipped to develop Wallingford as a major Norman

Figure 5.15 Sallyport at Knaresborough Castle (photo: Judy Dewey).

The early Norman constables, 1066–1148 Since Wallingford, an important Anglo-Saxon royal garrison town, was the place where two major English leaders, Archbishop Stigand and the king’s thegn Wigod, had made the first submission to William and his army, there is every reason to expect the prompt and symbolic raising here of a stone castle of some substance and grandeur. Indeed, the significance of what happened at Wallingford late in 1066 should not be underestimated: as has been shown elsewhere, it is likely that Stigand negotiated the surrender in advance of the Normans’ arrival at Wallingford, in order to spare an important town the harrying that had occurred after William’s victory at Hastings and the failure of the English to submit immediately (Keats-Rohan 2012). This initial defiance had been led by Stigand himself, as the most senior and experienced politician of the age; but after the Normans’ brutal response to their failure to take Southwark, Stigand no doubt recognized that the country’s salvation lay in joining rather than being further beaten by them. No fool, Edward II had obtained entry into Wallingford Castle itself via a postern. Worcester had a postern gate beside the Severn (Pounds 1990, 210). Postern gates, of which there were usually more than one, were small gates, protected by towers or other devices, and allowed private access for men on castle business, bypassing the portcullis and long formal procession through each of three gatehouses. They could also be used as sally-ports in the event of the garrison requiring a break-out manoeuvre. The site of Hedges’ passage was marked on the Ordnance Survey maps and provides a clue to the position to the first of at least two posterns on the east and north-east sides of the castle.

47

Figure 5.16 The view of the Tower of London from The Shard to the south west (photo: Rafa Esteve 2013, Public Domain).

royal castlery, as also of Oxford. The manorial endowment, over several counties, of the honour of Wallingford was considerable. A creation of early 11th-century military and economic planning, it became a Norman royal honour and castlery attached to a royal borough (Keats-Rohan 2009, 52–63; Roffe, above). Robert d’Oilly married Wigod’s daughter and assumed Wigod’s ministry; so long as he held the job he could enjoy the fruits of the honour. By 1086 he was holding a tenancy-in-chief in his own right, but had yielded half of what formed the honour of Wallingford to his successor as constable, his son-in-law Miles Crispin. On Robert’s death c. 1094 without male issue, what was left of the holdings acquired with his wife mostly passed to their daughter Matilda, wife of Miles, whilst his own barony passed to his brother Nigel (d. c. 1115). Miles died in 1107, leaving a childless widow who, according to an early 13th-century inquest, was remarried to a favourite of Henry I, Brien fitzCount – a curious arrangement if the details are true (chronologically, she would seem more likely to have been Miles’s daughter: Keats-Rohan 1989b, 315) because Brien would have been much younger and she is unlikely to have had much prospect of children (Stenton 1932, 233). Perhaps that was the king’s intention since Matilda’s attempts to nominate d’Oilly kinsmen as her heirs were blocked both by Henry I and later by Henry II (Bracton iii, 535–6) (Figure 5.17). This may well be linked to the special position of the honour of Wallingford and its castle, which, it will be argued below, were seen as a primary asset of the heir to the throne from at least the time of Henry II. The remarkable career of Brien fitzCount (pp. 55–6 below), and the significant development of the castle during his constableship is best understood against such a background.

Figure 5.17 Seal of Matilda of Wallingford, attached to a 12th century document of 1122 x 1147, regranting the churches of Ogbourne St Andrew and Ogbourne St George, Wilts, to the abbey of Bec, originally granted by Miles Crispin. Image reproduced by permission of the Dean and Canons of Windsor.

48

The situation at Wallingford was anomalous. In 1086 it was a substantial honour in the hands of Miles Crispin, constable of a royal urban castle. It would be further increased when some of the fees still held by his father-inlaw Robert d’Oilly de iure uxoris were reunited with the honour, in the period c. 1094 – 1113. Acquired before 1166 (less straightforwardly but by essentially the same route) was the borough of High Wycombe, Bucks. A total of over 102 knights’ fees was created. Possibly the honour was a fusion of Wigod’s tenures in his own right (and therefore heritable by his descendants) with those he held as a minister of the king, the latter being the majority (KeatsRohan 2009, 56–61; Roffe, this volume, p. 4). The two may have physically overlapped on the castle site, which, as has been suggested above, gives every appearance of being a well-developed Saxon military zone, almost a burh within a burh. Whilst hereditary constableships were tolerated in earldoms such as Chester, Warwick, Gloucester and Leicester, and occasionally in the constableships of other urban royal castles, such as Oxford, Bedford and Worcester, there generally appears to have been a clear determination to avoid it by the first three Norman kings, as was certainly the case at Wallingford. Wigod’s help in 1066 had to be rewarded – as did the sacrifice of the life of Wigod’s son Toki to save William’s in 1079 at the battle of Gerberoy – and so an apparently hereditary succession to the constableship and honour of Wallingford was tolerated for a while, misleading modern historians as to the true situation (e.g. Pounds 1990, 92). At both Wallingford and Oxford the castles were held by their constables as custodians not as lords (HKW ii, 771; Stenton 1932, 233). A major difference between them was that the same appears to be true of the honour of Wallingford. King John referred to the ‘honour of the castle of Wallingford’ (de honore castri de Walingf’). The situation may be mirrored in the later practice of the earls and dukes of Cornwall, sons of kings, in appointing constables who were also appointed steward of the honour, offices which had been held by different persons during the 12th century. At any rate, the calibre of the early constables was fundamental to the growth of one of the most powerful and architecturally impressive of all the royal castles.

includes reference to castle-guard, usually of a 40-day period for a full knight’s fee, though the terms of service vary. The inquisition post mortem of Robert Malet of mid1295 shows that he held land of the honour in Quainton, Bucks, ‘by service of 1½ knight’s fee, and of finding two men-at-arms in the castle of Wallingford if there be war, to wit one man for 40 days and another for 20 days, at his own charges’ (CIPM 3, 173–4). Robert Cervesa of Didcot was to serve his 40 days in person (RH ii, 777b). The inquest of Sir William de Bereford in 1332 revealed that he held the manor of Newnham (Murren) together with the patronage of the hospital of St Mary Magdalen of the honour for the service of a knight’s fee and finding two armed men for guarding the king’s castle of Wallingford for 40 days in time of war (CIPM 6, 471). Two entries in the Hundred Rolls specifically linked the service to a time when the castle was besieged, one providing one man with horse and arms for 40 days, the other a serjeant, at their own expense (RH ii, 777b, 778a). Castle-guard was a serious matter: in 1200 Robert Corbet brought an action against William of Cramford concerning 30 acres of woodland in Dalling, Middlesex. William did not appear in court on the appointed day, and so the woodland was taken into the king’s hands. William Briewer, one of the great men of the time, successfully testified that on that day William of Cramford had been at the castle of Wallingford in the king’s service and ought not to lose claim to the woods on that account (CRR 1226–42, 173). In many cases, even of royal castles, guard-duty was routinely commuted for a money payment from at least the time of Henry II onward (Pounds 1990, 16–18). Sometimes the commuted rents paid in lieu as late as the 15th century are the only surviving evidence of castle-guard obligations, as recorded at Oxford (VCH Oxon 4, 1979, 296). At Wallingford, however, we have a writ of King John to knights of the honour requiring castle-guard in 1215, and in March 1360 the Black Prince required that all those who owed castle-guard in time of war should be readied to respond to a summons ‘in view of the many perils which might easily befall the castle of Walyngford in these present times’ (1.49). An honour of this size could raise its garrison primarily from its own fee-holders (cf. Painter 1935, 452; Dover’s 172 fees were supplied by nine different baronies, RBE ii, 613–18). In the case of Wallingford, as would be expected for any royal rather than baronial castle, there is evidence of a wider obligation, such as the men of Berkshire being expected to repair the ditches of castle and town in 1215, at the same date that the king was summoning knights of the honour for garrison duty. The abbot of Abingdon owed castle-guard on three of his 33 knights’ fees to Wallingford Castle, though it was rarely performed; service at Windsor Castle was due on the other 30 fees (Cartularies Abingdon I, C385, 304). In 1244 the royal manor and hundred of Benson with its additional three and a half hundreds (‘the Chiltern hundreds’) were given by Henry III as a wedding present to his brother Richard, earl of Cornwall, then lord of Wallingford. But as early as c. 1153/5 the Empress Matilda gave to Gilbert Angevin half a knight’s fee in Benson (in the hamlet of Holcombe), the service to be performed at Wallingford Castle; Rainald Angevin

Castle-guard The construction of the castle was accompanied by the creation of structures for its support. First among these was the provision of service for its defence. The evidence for knight service is late, but undoubtedly points to very early arrangements. Castle-guard arrangements at royal castles can be difficult to find, though potential sources include documents such as the Book of Fees and inquisitions post mortem. One example comes from 1198, when William de Upton undertook to perform the service of castle-guard at Wallingford, with two horses, lance and shield, for threetenths of a knight’s fee at Upton, held of Samson de la Pomeray and his wife, who were to provide him with reasonable supplies of all he needed whilst in the castle (Scammell 1993, 595–6). Unsurprisingly, evidence relating to the holders of the 102 fees of the honour frequently 49

the profits of pasture within and without the castle gate by reason of his office. The 1166 list of freeholders also hints at another important service, that of watchman, when it mentions the wife (i.e. widow) of Robert the watchman, who was paid by the constable out of his fee. The 1219 extract helpfully confirms other evidence which shows that parcels of land in Clapcot were commonly assigned to castle service, both serjeanties and knights’ fees (cf. Pounds 1990, 49–51). The porter’s land was particularly extensive, occurring in Clapcot near the river at King’s Mead, next to the outer gate of the castle near the great bridge, and over the bridge in Crowmarsh, at Porters Close in Newnham Murren (Dewey, below p. 150). The tenures of the serjeanties were hereditary, but over time this system was adjusted, with the lords or constables of the honour making their own appointments to what were, after all, sensitive posts. Rights to the holdings were not lost, but many of them were bought up by wealthy burgesses or men of distinction, such as the justiciar Sir William de Bereford (d. 1327). These tenures regularly occur in the entries for forinsec rents or rents of assize in the ministers’ accounts of the 13th to 16th centuries in Wallingford.

held on the same terms in 1212 (BF, 104). The evidence of the late 13th-century Hundred Rolls spells out in the clearest terms what can be inferred from the exceptional mention of the honour in the Assize of Clarendon in 1166: the honour of Wallingford was a ‘jurisdictional peculiar’ enjoying extensive liberties which excluded the sheriff from most of his functions, effectively replacing royal government with baronial control, as seen elsewhere in the marcher palatinates such as Chester (Roffe, pp. 28–9). In Wallingford’s case, however, from 1148 the ‘baron’ in question was a member of the royal family (see Part 2 below). Its constable always accounted separately for the honour on the Pipe Rolls. Castle serjeanties The castle serjeanties were tenures that were held in return for a specified service. Again, the evidence is often late, but, as with knight service, we are looking at very early arrangements for the support of the castle. Occasionally a fraction of a knight’s fee attached to the tenure, but the service required was not usually military. There are hints of them among the ‘free tenants’ of the honour listed in 1166 in the Cartae Baronum (RBE, 310–11); these men were not knights and were often recruited from the townsmen (Pounds 1990, 49–50), as appears to be the case here. They are clearly identified in the Book of Fees in 1212, which also gave a similar list for Windsor Castle. In an account of the fees of the honour they are described as ‘serjeants of the castle of the old enfeoffment’ (i.e., they originated before 1135). Those named were Robert de Basingis who held three virgates, partly for service and partly in lordship, for service in the castle kitchen as cook; Ralph Saillard owed service as baker; Henry fitzIvo, mason, was to keep the walls of the castle and the buildings in the castle in good repair; Robert fitzWilliam was to see to castle business – a vague term clarified in 1219 when it was said that he should expect to travel up to five leagues on the king’s business (cf. the borough customs in Domesday Book, Roffe 2009, 28, 30); Ralph the arbalaster (crossbowman) was to serve with his arbalast (BF, 120); and in 1226–28 Henry Balistarius held 100 acres worth 50s a year for the service of making crossbows (ibid., 386). In 1219 the same serjeanties reappear, under a section headed escheats, all of them attached to Slotisford Hundred (ibid., 254–5). A sixth name heads the list: Robert de Porta holding 100 acres in Clapcot for his service, unspecified; this possibly links to Ralph de Porta, who held a fifth of a fee of the honour in 1212. At that date the account of the borough mentions that the prior of Wallingford held a parcel of land next to that of Hugh priest of Crowmarsh, by grant of Roger de Porta (ibid., 111). These details suggest that Robert de Porta may have owed his name to the serjeanty of porter, also known as janitor, of the castle. This person is in fact the one most often encountered in the castle accounts. His was a position of trust – he was also in charge of the gaol – and relatively well paid: in 1307, the janitor’s wages for 43 weeks and 6 days, were 51s 2d, and those of the crossbowman and his mate 39s 2d. The porter had lodgings in the castle, above the middle gate (where there was a watch tower), and took

Sheriffs and constables Wallingford Castle was not just a stronghold and the centre of the community of its honour: it was the focus of royal administration in the middle Thames valley. Occasionally constables of shire castles acted as sheriff; in fact, overall, about half of the constables of royal castles were the county sheriff at any one time (Pounds 1990, 87). Both sheriffs and constables were usually appointed for fairly short periods, from two to four years. There is no evidence that holding the castle and honour of Wallingford automatically carried with it the office of sheriff of Berkshire, though the castle was certainly available to the sheriff of the county as a base for the holding of prisoners awaiting trial at the county eyres, and for taking view of frankpledge, as was formally required by the Assize of Clarendon in 1166, when it was stated explicitly that the regulations applied even in the honour of Wallingford, then held by the king himself (Roffe, above p. 28). Organizing the men of the county to perform their obligation to effect wall repairs to castle and borough, and the repair and maintenance of the ditches, fell to the sheriff, as becomes clear from a series of writs issued by King John (1.18, 22) The sheriff also had a gaol in the castle for keeping shire prisoners. In times of stress, such as in the reigns of John and the minority of his son Henry III, the sheriff could be appointed constable or more usually custodian of the corpus of the castle, thus excluding him from any wider jurisdiction in the honour. A grant of the corpus to a constable other than the sheriff effectively excluded the sheriff from the castle. During the 12th and 13th centuries the sheriffs of Oxfordshire and Berkshire were often the same person or people (in the case of joint shrievalties), as in 1153 and 1154–55, when Henry of Oxford, an important burgess of both Oxford and Wallingford was sheriff of both counties, and so too was Adam de Catmere during 1165–69. In 50

neither case are the sheriff or sheriffs routinely associated with the keepers of Wallingford or Oxford castles, though the d’Oillys (who failed in the male line with the death of Henry II d’Oilly in 1232) usually retained their position as constables of Oxford, which certainly was a shrieval castle (Pounds 1990, 93). Robert I d’Oilly is assumed to have acted as sheriff of Oxfordshire c. 1071; a man named William held the office from about 1100 to 1110. The Robert who was sheriff of Oxfordshire in 1130 was probably Robert II d’Oilly, replacing Ralph Restold; otherwise, the only certain d’Oilly shrievalty occurs when Henry I d’Oilly was sheriff of the county from 1156 to 1160 (Green 1990, 26–7, 69–70). Again, the origins of this pairing may have predated the conquest. Parts of Berkshire lay in what was otherwise Oxfordshire, including Bampton and the great causeway of Grandpont south of Shirelake in the grounds of Christ Church, though it was within the Oxford parish of St Michael at the South Gate, a situation that facilitated Berkshire business being done in Oxford (Clanchy 1973, xxii, xl); indeed, the shire court is known to have met several times on that part of the causeway (Roffe 2009, 42–3). In the Pipe Roll of 1130 Brien fitzCount proffered a large sum of money ‘for the ministry and part of the land of Nigel d’Oilly’, Robert I d’Oilly’s brother and heir who disappears from view after 1115. This possibly relates to the custody of Oxford Castle, but the links between Wallingford and Oxford were set to continue. Soon after the death of Henry I in 1135, according to the Winchester Annals, 26 castles were strengthened against King Stephen in 1138, including those of Brien fitzCount at Oxford and Wallingford. Nicholas Riall (2003, 118) has pointed out the problems with the Annals, such as the fact that there is no other evidence that Brien had a direct role at Oxford, where Robert II d’Oilly was certainly constable in 1136 (RRAN ii, no. 948). Nonetheless, both castles saw plenty of action during Stephen’s reign, with Wallingford playing a key role in the eventual triumph of the future Henry II. There is good evidence, including the 1327 surveys (see below), that of the two castles only Oxford castle was routinely used as a shrieval base. Pounds (1990, 96) has suggested a distinction between the royal urban castles that became shrieval centres, which were often neglected and eventually fell into decay, and those which served for the protection of the realm or the residence of the king; this is supported by Rickard’s analysis of castle personnel for the later Middle Ages, which showed that after 1272 sheriffs were routinely appointed as keepers of the royal county castles (2002, 29–40). This reflects the situation at Oxford (where there was a separate royal residence in modern Beaumont Street until the 13th century) and Wallingford respectively. Pounds thought that the pairing of Oxfordshire and Berkshire as counties with Oxford the shrieval base arose because Wallingford was not retained in royal hands. Such a view is mistaken, if understandable: the ancient links between the territory of the honour, lying in west Berkshire and south Oxfordshire, including Oxford, created an area of the very royal county of Berkshire distinct from its eastern counterpart, which rapidly came to be dominated by the castlery of Windsor; this left an increasingly distinct mark in official records

such as the Pipe Rolls from the late 12th century onward, the period after which the castle of Wallingford and its liberty was almost always directly under the control of a senior member of the king’s family. Wallingford Castle’s first appearance in the records was as the prison of the abbot of Abingdon in 1071 (the earliest reference to any castle as a prison, Nevell 2014–15, 219) and the chroniclers note the prison of Brien fitzCount in the 1140s. But other county gaols lay elsewhere in Berkshire; the earliest to have made regular appearances on the Pipe Rolls was at Faringdon. In the 13th century the main prison was certainly at Wallingford, though gaols in Faringdon (town), Windsor Castle, at Bray and a private prison at Reading slightly confuse the situation. The sheriff of Oxfordshire John de Turberville was granted three oaks from the royal forest at Shotover, Oxon, for a gaol at Wallingford in 1254. In 1277 the town gaol at Windsor was granted by charter the status of chief Berkshire gaol (Pugh 1955, 4–5), though in 1315 a petition was presented to Parliament asking for the reinstatement of the gaol in the town of Wallingford. The petitioners alleged that the negligence of previous sheriffs had allowed the gaol to fall into decay, whence it was removed to Oxford Castle; by procurement of certain people it had been removed to Windsor, which was highly unsuitable for a number of reasons (TNA, SC8/257/12843). Wallingford always had one or more prisons, for both state, shire and honorial prisoners. Nonetheless, the impact of the sheriff at Wallingford was limited, due both to the extensive charter of liberties granted to the borough of Wallingford by Henry II in 1155, which gave the townsmen considerable freedom from the attentions of the sheriff and his men, and also the extensive liberties, including return of writ, attaching to the honour itself (Keats-Rohan 2009, 63; Roffe, above, pp. 28–9). A constable was appointed to a castle by royal writ. His duties were varied, ranging from ensuring that the castle was kept in good condition to meet its obligations as a base for the sheriff, for various courts, and as lodgings for the king and his retinue and any more transient guests, whether those temporarily housed by the favour of the lord, or noblemen held as prisoners of state or for ransom. He had responsibility for prisoners; being liable if any escaped, though he could charge a fee when they were formally released. He had to organize provisioning of the castle, of both victuals and munitions, and to ensure that both were kept in good condition. This last was often problematic as both food stores and arms deteriorated with the sub-optimal conditions of even the best kept castles; any surplus food about to rot was sold off, and weapons had to be refurbished or replaced. Pounds also fails to mention Wallingford amongst the castles important for the arms stores and the making and distribution of arms, dealt with in detail below. An early indication of this part of its business is the arbalast serjeanty, one of a number dating back to the reign of Henry I. The garrison, when present, had to be housed and overseen. The routine work could be shared with a janitor and watchmen, whose wages were paid by the constable out of his fee; a sum of £13 6s 8d was allocated in 1296–7 for the fee of the constable and two watchmen (Midgley 1942, 112). The constable might hope to profit from his 51

Figure 5.18 Oxford Castle in 2015, after conversion from HMP Oxford to Hotel Malmaison; St George’s Tower is in the background, with the motte slope in the foreground (photo: Katharine Keats-Rohan).

The Great Chapel of St Nicholas in the castle and other churches

office through a range of perquisites such as the income from mills and meadows, or ransoms, or various gifts or marks of favour such as robes. The downside was that he not only had to pay the wages of such as the watchmen, but he was liable to pay out on the contracts negotiated for castle maintenance, which could be expensive, since the king was a tardy paymaster and the constable could be left seriously out of pocket. If, however, he failed in his duty of maintenance then he could be called to account for the repairs even if he had left office. For example, when John Alveton, constable of Wallingford, died in 1361 without having presented his accounts for the building works then in progress, his goods were seized and sold to pay his debts to the Black Prince (Green 1998, App. 4). Wallingford’s constable was normally a member of one of the families who held of the honour; the same seems to have been true also of his deputy, the under-constable or lieutenant as he was called in the 16th century. Wallingford is unusual in that we have fairly good evidence throughout its history about the identities of the constable and his deputy, and the janitor (Pounds 1990, 90; Rickard 2002, 40–2, 90–3). At Wallingford it appears that the constable’s basic salary was £40 per annum, which would have placed him at joint second with Chester in Pounds’ list of the top 20 best paid royal constables in 1287, had Wallingford been included in that list (Pounds 1990, 87–9), above Windsor, in third place with £30. The constable also had an allowance of £10 to pay his under-constable or lieutenant. As with all royal constables, he had an allowance of 100 shillings annually for routine expenditure on the castle (HKW i, 115).

The Norman castle area at Wallingford probably contained a pre-existing church of some kind, whether a small thegnly Eigenkirche, a more important, older church, or a chapel dependent upon such a church. The latter may be more likely given the insertion of the whole castle precinct (itself extra-parochial) into the vast parish of All Saints (Roffe 2009, 36), which included Clapcot, in Slotisford Hundred, stretching way beyond the northern ramparts down to Shillingford. St Nicholas became a prestigious church, but this was not an inevitable trajectory for a castle chapel. Henry I’s constable Brien fitzCount was forced to withdraw into religion c. 1148, surrendering the honour of Wallingford Castle to Henry of Anjou before the resolution of the conflict with King Stephen in which he had played so notable a part. Brien was dead by 1153, at the date of a charter given by Henry of Anjou in association with his mother the Empress Matilda. Significantly, this charter details the provisions for building a church for regular canons in Wallingford’s southern suburb, on land between the South Gate and the ditch of the mill there, which Brien had built, up to the Winterbrook, which marked the boundary between Wallingford and Cholsey, and down to the Thames, the precise location to be chosen in consultation with the chapel clergy and other suitable persons (RRAN iii, 34). The prebends (revenues granted as clerical stipends) of the existing clergy staffing the chapel 52

of St Nicholas in the Castle, all of whom were named in the charter, were to be transferred to the new foundation through the resignation or death of their current holders. This was to be done for the benefit of the souls of Brien and his wife, in fulfilment of a work they had themselves wanted to undertake for the salvation of Henry and his mother. Apparently Henry’s full instructions were not executed since the royal free chapel, known as the Great Chapel or College of St Nicholas in the castle, and its prebends remain abundantly documented up to the time of the Dissolution. However, in 1179 the first of many such mentions is made in the Pipe Roll (PR 22 Henry II, 26) of 3s in respect of land in the royal manor of Benson held by the infirmi of Wallingford by grant of the king; this appears to be the earliest reference to the Hospital of St John, located in the area specified in Henry’s charter of 1153 (now St John’s Green abutting St John’s Road and Reading Road). The Hospital held land in Newnham Murren in Crowmarsh (in the half-hundred of Benson), and was later responsible for the small leper hospital of Mary Magadalen there, founded by the Empress Matilda but in financial trouble by 1212. The royal origin, and link with the castle, of St John’s Hospital and its link with the castle are confirmed by a later document of 1355 detailing the affronts afforded the Black Prince, then lord of Wallingford, by the burgesses, which included ‘claiming the patronage of the house of St John, Walyngford, though it used and ought to be under the protection of the lord of the castle’ (REBP iv, 173). At his death in 1327, Sir William de Bereford held the manor of Newnham Murren with the patronage of the hospital of Mary Magdalen for one fee of the honour of Wallingford (CIPM 6, 471; Christie and Creighton 2013, 288–91). The attempts by Brien and his wife to change the ecclesiastical arrangements of St Nicholas are paralleled elsewhere. His predecessor Robert d’Oilly certainly founded a college of St George at Oxford Castle, which was given by his nephew and successor Robert II d’Oilly to his new Augustinian foundation outside the castle at Oseney in 1149. Another example is St Mary in Castro at Leicester, which became an endowment of the new abbey in Leicester founded by Robert, earl of Leicester, in 1143. A castle chapel could be a very desirable asset for a religious house outside its walls, as was true at the priories of Huntingdon, Launceston, Leeds and Scarborough, among many others. At Pontefract the Lacys had a college in their castle and built a separate priory outside it (Pounds 1990, 235). St Clement’s, Pontefract was, like Wallingford, one of 124 castle-related monastic foundations of the 11th and 12th centuries, only 11 of which were secular colleges (Pounds 1990, 232–3). Although very little is known of their organization, it appears that, unlike Anglo-Saxon secular canons, these did not live as a community sharing revenues in common, but each was financed by a separate prebend which was in the gift of the lord of the castle. The canons were therefore part of the lord’s extended household and so often provided additional clerical services such as writing and accounting (Denton 1970, 2). In a royal college, appointment to one of the prebends, especially the deanship, was a mark of distinction (ibid.,

133). Only the most important royal chapels were secular colleges (ibid., 5). Castles always had chapels, at least one of which should have been spacious enough to serve the whole castle community; others were smaller and more private and for the use of the lord of the castle (Figure 5.19). Chapels in royal castles were often free chapels, i.e. they were extra parochial, but remained under the authority of a bishop. The king had the advowson in a royal castle (the right to appoint the priest). The priest of the church did not necessarily have cure of souls, but could officiate at baptisms, marriages and burials, as was the case at St Nicholas, where the cover of the font was repaired in 1296, and as late as the 1540s parishioners of All Saints were asking in their wills to be buried inside St Nicholas. The College of St Nicholas became a prestigious, wellendowed royal free chapel that bordered briefly on the status of royal peculiar, that is, a church free of both episcopal and metropolitan jurisdiction. This is shown by the case of John Bray, a Chancery clerk, who was rector of All Saints and in that capacity had to provide a chaplain for St Nicholas. He was prevented from enjoying his emoluments by the men of the castle but was unable to appeal to his bishop, which led to a parliamentary inquiry being set up in 1315 (Denton 1970, 9, 12–13, 124–5). A 12th-century inquest stated that the prebends of St Nicholas in the castle were established by Miles Crispin. They were supporting six canons in 1153 (Oseney Cartulary iv, 415). By the late 12th century three distinct prebends emerge, consisting of the churches of North Stoke, Chalgrove (with the chapels of Ipsden and

Figure 5.19 Norman chapel at Durham Castle (photo: Judy Dewey).

53

Newnham Murren), and All Saints, Wallingford, with the mill at the town’s South Gate – this possibly an addition since the mill was stated in Henry’s 1153 charter to have been built by Brien fitzCount (perhaps replacing an earlier one). The dean of St Nicholas had the advowson of the churches of St Peter, St Michael and St Mary de Stallis in Wallingford until 1374, when on account of their poverty, the parishes of St Michael and St Mary were merged with St Peter’s. The clerics appointed were often royal chaplains or chancery clerks, who occasionally held plural benefices, which could lead to inconveniences, as did the status of St Nicholas as a free chapel. An undatable charter of Thame records how at some date the dean of St Nicholas was chaplain to the abbot of Thame and was obliged to appoint a proxy for want of time (TC, 26). The prebends appear to have been foreshadowed by a section in the borough account of Wallingford in Domesday Book where the messuages held of the borough by Miles Crispin in Great Haseley, Newnham, Chalgrove and Stoke are mentioned (Roffe 2009, 31, 36). A reference to the prebendary (the dean of St Nicholas, who held the College prebend of All Saints) occurs in the borough entry on the Pipe Roll of 3 Henry II (PR 3 Henry II, 83) in 1156–7 where the sum of 30s 5d was rendered by the townsmen to him. The borough rolls for the first half of Henry III’s reign show regular four-termly payments being made by the borough to the prebendary of around 7s 8d (RCHM 1877, 577–8, 580). As for St Nicholas itself, the church, if it had existed in 1066, was probably rebuilt and re-dedicated in Miles’s time. We know from a survey of stone and timber of 1555 (detailed by Lloyd, below 120–1) that the church was 59ft (18m) long and 30ft (9.1m) wide, with a 25ft (7.6m) square belfry on the west end. The orientation of castle chapels was normally with the chancel lying between north-east and south-east (Pounds 1990, 240). Thirteenth- and 14thcentury documents indicate that it was a free-standing building in the inner bailey, surrounded by gardens, one of which was opposite the cob wall of the constable’s garden. Leland stated in his Itinerary of c. 1540 that the chapel was within the third dyke, i.e. in the inner bailey. It was re-founded in 1278 as a chantry by Edmund, earl of Cornwall, leading to major changes in the middle bailey, as discussed below. Many people nowadays wrongly assume that the collegiate buildings for the clergy established by Edmund in the middle bailey are the remnants of the Great Chapel itself, rather than the residential complex for the chapel staff, as was in fact the case; most of the surviving stonework from the castle is from these collegiate buildings (Figure 5.20). Earl Edmund’s changes brought a new distinction to both the chapel and the castle since his endowment created a staff consisting of the dean, the chaplain to the dean, six chaplains, six clerks and four choristers – altogether more than the peace-time staff of most castles (Prestwich 1982, 163–4). Their lives were well regimented: at least four masses a day were to be said, according to the rite of the church of Salisbury, and absences from the offices were punishable by deductions from their stipends. All ministers to the chapel were to have free ingress and egress of both chapel and town whenever they wanted. They had to ‘swear on the Gospels

Figure 5.20 Lion head detail from a stone doorway in the Priests’ Lodging (photo: Judy Dewey). Male head capital reused for a now-blocked garden entrance doorway from Castle Lane into the middle bailey. Two of the least eroded of the few surviving stone carved figures still on site (photo: Katharine Keats-Rohan).

that they will firmly and faithfully keep our counsel within the castle, lest any harm be done through them within the castle; and if they hear or notice anything within the castle that might lead to harm to us, they should report it at once to us or our council’ (1.51). The College had ceased to be a religious house by the early reign of Edward VI (CPR 1547–53, 392). Mention can also be made here of the creation of the priory of Holy Trinity in the north-west sector of Wallingford (Pedgley, below pp. 222–3). Élite churches inside castle complexes were common, as was an 54

association with a regular community outside. Holy Trinity was a pre-Conquest college of canons that was reformed as a house of Benedictine monks, subject to St Albans, probably c. 1086–9 (Roffe 2009, 36–7; KeatsRohan below, p. 215). Its earliest endowments include grants of the land of Henry the Larderer and the disseisin of Geoffrey the Chamberlain, both of whom bore names suggesting royal service. Henry I also resumed the patronage of the churches of St Lucian and St Leonard, which he gave to the priory of St Frideswide, Oxford, and of St Rumbald, which he gave to his Cluniac foundation of Reading Abbey, showing that royal control was being exerted over churches both south and north of the town (Roffe 2009, 37–8, and below, p. 198). Although there is no suggestion of a direct link between development of the castle and the priory, the two are undoubtedly linked by royal patronage as the two élite sectors of the town. When Richard, earl of Cornwall, established a chantry chapel after his first wife’s death in 1240 he stipulated that on the death of the chaplain (who earned 15 marks a year from the revenues of the castle mills), the seneschal (steward or major-domo) was to find a suitable replacement; if one had not been found after six months, the prior of Wallingford and the rector of Stoke were to make the appointment. The priory and several others held parts of the 88 acre fee of Le Barrse in Clapcot of the dean of St Nicholas in the mid14th century. Honorial tenants owing service at the castle and castle serjeants were frequently among the priory’s benefactors. Geoffrey Lecur (the cook) made a grant to Holy Trinity for the soul of Brien fitzCount from his arable land in Clapcot near the town’s North Gate (CMACC, 134). Even more interestingly, Alan son of Ansfrid, whose father had been steward of the honour during the siege of 1152 and later, granted land in his father’s meadow of Clapcot lying before the postern gate of the castle on the east side (below, 72). His family became lords of the manor of Aston Rowant, Oxon, a large demesne manor of Miles Crispin in 1086. By the date of the papal taxatio in 1291, the church of Aston Rowant, like that of West Hendred and the parish church of Holy Trinity itself, had been appropriated to Wallingford priory, so that the monks enjoyed the receipts of these parishes in return for providing a vicar.

as a means of prising an honour out of the hands of any potential claimant to all or part of it was very useful. In the case of Wallingford, the childlessness of Brien fitzCount and Matilda facilitated the matter, though the early 13thcentury inquest stated that when, towards the end of his life c. 1148, he withdrew into religion, Brien placed his honour into the hands of Henry – still duke, but the rightful king in Brien’s eyes (Keats-Rohan 2015). Nottingham was seized as a disciplinary escheat by Henry II (Hudson 1994, 20–1). William II Peverel had supported Stephen throughout the civil war and was still on the losing side in 1153 when Duke Henry deprived him of the honour unless he could answer for the felony of the attempted poisoning of the earl of Chester, to whom Nottingham had been given by the Angevins and who was dead by the end of the year; William’s response to this uninviting offer was to flee into a monastery. No resistance was offered when Henry, now king, marched on Nottingham in 1155, declaring Peverel and his heirs forfeit, thus firmly reinstating royal control of Nottingham castle and honour. There was a clear difference between the attitude of Henry II from the moment of his accession to the control of royal castles and those of his immediate predecessors, William II and Henry I, who were more tolerant of royal castles being in baronial hands. Henry II’s position was much closer to that of his great-grandfather, whose conquest of England had ushered in the age of royal and then baronial castle building; for both kings the castles were, in the words of William of Newburgh, ‘the bones of the kingdom’, and required constant royal vigilance. The experience of the wars of Stephen’s reign had thrown the issue into very sharp relief. It had arisen because the death of Henry I’s only male heir led to a succession crisis. Eventually, in 1127, Henry decided to nominate his widowed and childless daughter Matilda as his heiress and to oblige his barons to swear to accept her as his heir. First though, he had to arrange a second marriage for her. Henry trusted only two people with knowledge of this plan, his natural son Robert, earl of Gloucester, castellan of Bristol Castle, and Brien fitzCount, castellan of Wallingford. The two worked closely in 1128–9 on an overhaul of the treasury, and would work together again after both decided to fight for Matilda’s rights against King Stephen in the autumn of 1139. Brien was a natural son of Alan Fergant, duke of Brittany, and had been raised at Henry’s court. His abilities had been recognized by the king when he married him to Matilda of Wallingford, later granting him also the honour of Abergavenny in the Welsh marches. The holdings of the honour were augmented by the Middlesex portions that had been detached for Earl Roger before 1086, although his predecessor in several of them had been Matilda’s ancestor Wigod of Wallingford. Once again, this apparent family arrangement was in fact nothing of the sort. Wallingford always was a royal borough appurtenant to a royal fortress with a territorial hinterland spanning the Thames, and one designed to facilitate ease in waiting upon the king where ever he might be within what Thomas Keefe, (1990, 182–4), in his study of patronage under Henry II, has called ‘the royal administrative enclave’.

‘The bones of the kingdom’: evolution of a royal stronghold 1139–1540 ‘The honour of the castle of Wallingford’ There are clear analogies between the honour of Wallingford and the honour of Peverel of Nottingham (named from its constables of the Peverel family), as discussed below by David Roffe. Both functioned, to paraphrase King John, as ‘the honour of the castle of Wallingford/Nottingham’. Both were key citadels at strategic locations guarding the doors to the South Midlands and the South-West, and the North Midlands and the North, respectively. Castles and castleries of such importance demanded firm oversight by the king. At a time when the principles of honorial heritability were still being fixed and the king needed all the support he could get, being able to use a form of escheat 55

By counting the number of charters which were given at different places by Henry II, Keefe determined that the principal English centres under Henry II (1154–89) were: Westminster-London, Winchester, Woodstock, Northampton, and trailing some way behind, Nottingham and Windsor. This compares with what Stephanie Christelow (1996) found for the charters of Henry I (1100–35), where the order was Westminster-London, Winchester, Woodstock and Windsor. Keefe concluded that these figures well support the notion of a royal administrative enclave in the Thames Valley and central Wessex under Henry II. Not only that, but they bear comparison with his Angevin successors John and Henry III. If the king was favouring this area, then his favoured servants, or curiales, such as the constable of Wallingford, needed to have holdings so placed as to allow them both to attend upon the king’s court as he moved around, and to attend to their own affairs (Keefe 1990, 180, 183). As a thanks for medical care received from Abbot Faritius, Miles Crispin made a grant of Colnbrook to Abingdon, pointing out that it lay on the road to London. A grant by Brien fitzCount to Evesham abbey of tithes and land at Hillingdon in Middlesex was dated on Christmas Day, at Iver, on return from the court at London (HEA ii, 142–3; Keats-Rohan 2015, 131). Keefe (1990, 183) went on: ‘Both the Angevin and royalist factions vigorously fought for control of the area during the civil wars of Stephen’s reign. If there was a middle point to the enclave it was Wallingford Castle. This castle, resolutely defended by Brian Fitz Count for the Empress Matilda and her son, was never reduced by Stephen. Henry II understood Wallingford’s importance and rewarded its defenders when he became king. He also kept the honor and its castle in his hands for the remainder of the reign.’ Brien was not present during the final gruelling siege in 1152–3. He was probably in Wallingford Castle during the siege of 1146, but he retired into the monastery built by Henry I at Reading before June 1148, when the Empress was back in Normandy. The Empress he had so faithfully served gave the manor of Blewbury to Reading abbey ‘for the love and the lawful service of Brien fitzCount’ sometime after she arrived back at Devizes Castle, after escaping from Oxford Castle via Wallingford in 1142 and before her return to Normandy in 1148, no longer able to support herself (Keats-Rohan 2015, 134). Brien must have been either incapacitated by wounds or illness to have entered the abbey at such a critical moment. Nonetheless, his well-organized and determined garrison held out till the final relief by their lord Henry of Normandy in 1153, when peace was established following the famous parley between Henry and his rival at Wallingford (Gesta Stephani, 238– 41). Their raids for much needed supplies earned the fury of the Abingdon monks, who spent the next 200 years trying to wriggle out of the three knights’ service they owed to the king for castle-guard at Wallingford, though they were not much keener on supplying the 30 knights they owed for castle-guard at Windsor (HEA ii, 288–9, 314–17, 342–3). Nor should the role of the townspeople be forgotten: the whole town had periodically been under siege between 1139

and 1153 and they had embraced Brien’s cause as well. The town were duly rewarded with a generous borough charter at Henry’s accession. Wallingford like Nottingham had a command economy (Roffe, Chapter 8), and much of the town’s economic activity and prosperity was centred upon and depended upon the castle. This will have determined the attitude of the townspeople to moments of crisis, such as in 1066 and throughout the Anarchy. As we shall see, it would also be carried over into the turbulent 200 years that followed. Angevin stronghold, 1154–1216 Allen Brown (2003) pointed out that the Pipe Roll entries recording expenditure on royal castles tend to relate to castles visited by the king that year, the writs for work usually having been issued when the king was either at or close to the castle in question. There is little sign that Henry II himself spent much time in Wallingford, though he did visit the place to which he owed so much early in his reign. It was at Wallingford on 10 April 1155 that the notables of the kingdom recognized Henry’s two sons William and Henry as his heirs (1.6). The Pipe Roll evidence suggests that he viewed the castle as a primary residence for his heir. Little Prince William died aged two before June 1156, probably at Wallingford, and was buried at Reading at his grandfather’s feet, yielding his place in the succession and at Wallingford to Prince Henry. The earliest evidence relating to the castle in the Pipe Rolls comes from 1164, when the sheriff accounted for bringing three tuns of wine to Wallingford for the use of the king’s son, his heir the future Young King, Henry (1.7). Running costs of £67 11s 5d were recorded for 1172–3 and £31 15s 3d in 1179, while repairs of the king’s houses at Wallingford cost 13s 5d in 1174 (1.8–11). These sums may reflect both the Young King’s marriage in August 1172, and the trouble developing between the king and his sons as they emerged from youth into manhood spoiling for the fight that first erupted in 1173, probably leading Henry to reassert central control over the castle (Warren 1973, 51n). Frustrated by the restraints under which his father kept him, despite having him crowned as his heir in 1170 and again in 1172, the Young King left secretly in March 1173 and joined his brothers in France in a rebellion against their father that was soon crushed. Allied with his father against Philip of Flanders in 1181, he slipped away again in 1182, dying of dysentery seeking his father’s pardon on 11 June 1183. By the time that accounts for the running costs of the castle start to appear regularly in the Pipe Rolls, from 1172, the main outlines of the castle were well established. It appears that Wallingford may have been the designated base of the court of the Young King Henry (d. 1183), first crowned as his father’s heir in 1170, an association with the king’s heir going back to the earliest days of the new reign. When Richard I became king after Henry II’s death in 1189, in accordance with their father’s wishes, he granted various lands and honours to his younger brother and heir apparent John, including the honours of Wallingford and Nottingham, as well as the title and lands of the county of Mortain, the county of Cornwall chief 56

among them, and including also the castle and manor of Berkhamsted. Richard reserved to himself the castles of Wallingford and Nottingham (cf. Roffe, below), as well as Gloucester and Tickhill, which he entrusted to the archbishop of Rouen who was to assist the chancellor, William Longchamp, in Richard’s absence on the Third Crusade. Longchamp’s rule rapidly alienated much of the baronage, but is particularly interesting here for his policy towards castles and their constables. As Heiser has shown, he saw the importance of castles for dominating the country and took care where he could to promote his own men to constableships, rather than the local men favoured by the king in a policy popular both with the barons and the people affected by the castles. Longchamp was clearly suspicious of Count John and directed much of his energy to harassing him, mainly by strengthening castles that John did not hold, including Wallingford, Nottingham and Windsor, and by appointing his own men to constableships in these and other castles in the vicinity of John’s interests. By 1191 he had provoked a backlash by the barons, many of whom openly supported Count John. John soon took the opportunity to seize (by trickery) the castles of Wallingford and Windsor and conspired to usurp his brother’s throne. His alliance with the French king Philip to that end was a step too far. A counter-coup was mounted by the earl of Leicester and John was forced in April 1193 to surrender Wallingford and Windsor into the safe keeping of his mother Eleanor of Aquitaine. He was declared forfeit by the king at the Council of Nottingham in May 1194, but he went immediately to Normandy to make his peace with his brother, who named him as successor shortly before his death. After John became king in 1199 references in the Pipe Rolls to Wallingford Castle start to multiply. The king himself remained in Normandy until December 1203, when the tide was turning against him; he had already lost his father’s ancestral lands in Anjou and elsewhere, with the loss of Normandy itself completed in June 1204; by that time his mother’s county of Poitou was also recognizing the suzerainty of Philip of France. Once he was back in England he turned his attention to governing the country in such a way as to provoke both a papal interdict (1208) and a baronial opposition, leading to civil war (1215–17). Accounts for the maintenance, provisioning and running of Wallingford Castle occur throughout Richard’s reign, including specific mention of a chamber, a kitchen and a granary. In 1196 repairs were made to the bed chambers (thalami), and the hall was re-roofed (1.13). John clearly highly valued the castle, where he is known to have stayed on 20 April 1204, 20–21 September 1205, 25 August 1213, 31 October, 1 November 1214 (when he met some of his baronial opponents), 2–3 November 2013, 13, 14, 30 April and 1–2, 12–13 May 1215, and 5–6, 14 September 1216. In 1201 he had new iron bars fitted in the prisons, and paid about three-quarters of the allowance of the soldiers and servants who had been retained in the castle after Richard’s death (kings were notoriously slow to pay their servants’ wages) (1.14–15). Others attest the use of the castle by the king, whose visits occasioned an account for 186 pigs and 100 bushels of wheat in 1215/16, when accounts were

rendered for fortifying the castle with palisades. The same accounts mentioned ‘the expenses of William de Merc, with four gerfalcons and one falcon gentle, and five men and four serving boys and six horses and four greyhounds’ (1.20). John’s reign had soon run into trouble, dividing his baronage, many of whom took up arms against him. In 1214/15 he ordered repairs to the wall-walks (alures) behind the castle battlements, and issued a series of writs to the sheriff demanding that the men of Berkshire perform their customary duty of repairing the ditches of the town and castle (1.18). In January 1213 he appointed his clerk, John de Wikenholt, sheriff of Berkshire in 1210, custodian of the castle (1.16). The peril in which John’s reign ended, with a revolt initially contained by the signing of Magna Carta in June 1215, soon followed by the French prince Louis’s invasion (22 May 1216) after an offer of the throne from the rebel barons, occasioned the sense of high alert visible in the repair of ditches and wall-walks at Wallingford, the vital strategic importance of which John had long appreciated. In May 1216 the king sent his natural son, Richard fitzRoy of Chilham, to Wallingford, instructing the sheriff and constable that the key of the castle gate was to be delivered each night into Richard’s keeping (1.21). On 17 October 1216, two days before his sudden death, he informed his knights, serjeants and bowmen (balistarii) and everyone else in his garrison at Wallingford, that he was appointing Richard as constable of the castle. Richard fitzRoy was the nominal sheriff of Berkshire early in his brother Henry III’s reign, with Henry de Scaccario, from a local family associated with the honour, acting on his behalf. Henry served as under-sheriff for several subsequent years. The castle as royal residence and appanage, 1216–1272 Clearly Wallingford was a royal castle in the fullest sense: one that was at once one of the many residences of the itinerant kings, an administrative centre, and a stronghold. Care was taken to maintain it fit for all these purposes, but at this date we still have little detail about the castle enceinte or its buildings. With the accession of Henry III Wallingford came into the hands of a king who spent vast sums converting his favoured residences into state-of-theart palaces. Windsor and Nottingham both had fortunes lavished on them; Wallingford was not neglected either. There are signs in the rolls for 1218 and 1219, with the repair of the king’s chamber and a wardrobe, that he might have lavished significant sums of money here as well. This, though, was the period of his minority, the country initially governed in his name by a council led by William Marshal, earl of Pembroke (d. April 1219), and afterwards by the justiciar Hubert de Burgh; although his minority ended in 1222, his personal rule, unaffected by dominant figures such as de Burgh, would not start until the early 1230s. In the interval, the regency government, headed by Earl William Marshall, had to bring the civil war of the last years of John’s reign to an end by defeating Prince Louis and his adherents and reconciling opponents with the new regime. Louis was defeated in battle at Lincoln in May 1217, the peace treaty including a re-issue of Magna Carta, 57

which had been annulled by the pope shortly after it had been signed by John. Wallingford and three other major honours over which the crown henceforth maintained tight control – Nottingham, ‘Boulogne’ (Ongar) and Lancaster – were named in the document, which protected tenants of the honour from paying extortionate relief or excessive service when succeeding to fees of the honour. Facing the same task as had faced Henry II in the aftermath of Stephen’s reign, the regents also had to restore central government, which required that royal castles be brought back under royal control, with reliable sheriffs that could raise local revenues for the king, whilst delivering his peace. The regency was keen to have its own men holding the castles, but the process was very slow. There was relief when King John’s son Richard of Chilham left for the Fifth Crusade in 1218 and surrendered Wallingford to local man Henry de Scaccario, although he reserved the right to resume the constableship on his return. This did not in fact happen, but another person removed at this time was the unpopular Henry of Cornwall, natural son of Earl Reginald of Cornwall (d. 1175) who was himself a natural son of Henry I and had been a staunch supporter of the Empress Matilda and her son Henry II. Despite the reserve of its people, who still have a very strong sense of identity, this remote county was one of the most important in England on account of its stannaries, which, together with those in neighbouring Devon, were the sole source of tin in the country; it was also a valuable source of lead. Maintaining control of the county in trustworthy royal hands was clearly important. The considerable revenues of Cornwall would pass to the king’s (full) brother Richard when he was knighted and created earl of Cornwall, in May 1225, a portentous moment for Wallingford which was thereafter associated with the county of Cornwall until 1540. Between his accession and November 1222, the young Henry III’s time was largely divided between Wallingford in particular, Guildford and the palace at Havering, Essex (Carpenter 1990, 94, 241, 258–9, 305), and it is for that reason that significant expenditure on the castle was incurred. The revenues of the honour of Wallingford were important to the king at this time. The Pipe Roll of 2 Henry III (PR 2 Henry III, 35), shows that the sheriff rendered £135 17s 10d in scutage on knights’ fees, some of it owed from 1215. Matthew Brand, custodian of the honour, was allowed £4 in expenses. The borough, which struggled, and often failed, to pay its fee farm from 1129 onward, rendered £16 5s 7d in rents of assize, £4 13s 2d in tallage, £8 16d in customs, and £6 13s 8d in pleas and perquisites. The young king was living in some style. In the same year 52s was outlaid on the carriage of 100 quarters of wheat, 10 tuns of wine, 100 pounds of wax, 20 pounds of pepper, one pound of saffron, two quarters of beans and two of peas to Wallingford from Northampton. A further 48s 10½d was spent on the carriage of one quarter of wheat, 10 tuns of wine, 40 sides of pork, one load of almonds, two pounds of saffron, five pounds of cumin, 100 pounds of wax, two quarters of peas from London to Wallingford; a further six shillings were spent on carriage of the king’s robes from London to Wallingford (ibid., 43–4). In the following

year his transit from Windsor to Wallingford required the carriage of 100 pounds of wax, 20 pounds of pepper, one pound of saffron and two dozen towels (PR 3 Henry III, 74). An undated account from early in the minority period provides an increasingly rare documented instance of river transport, a ship being hired at Wallingford to take a cargo of corn to London, on which the sheriff outlaid 24s 11d (Roll of Divers Accounts, 8). The unpublished Pipe Roll for 1224–5 shows the carriage of 8 tuns of wine by water from Wallingford to Oxford. In the same year the undersheriff accounted for repair of a wall next the drawbridge, and of houses in the castle, and the escort of 15 prisoners to London; in the second half of the same year the king’s mills in the castle were repaired, and the king’s meadow mown. An entry in the 1219 Pipe Roll reflects the building of a wardrobe in the castle required for the conduct of royal business (Rot. Litt. Claus i, 393). In this context the word refers to a place set aside for the business of the office of the King’s Wardrobe which controlled expenditure and would grow in importance with the rise in military expenditure under Edward I. Like the Exchequer, which dealt with finance, including receipts of monies inwards, it was still peripatetic, with the relevant officers travelling with the king, who also was accompanied by chaplains serving his portable Chapel Royal. The 1219 Pipe Roll account speaks of repairs to the king’s houses, the term commonly used in castle accounts to refer to the various buildings or chambers within the castle used for residential or business purposes (Pounds 1990, 196). It contains the first of many references to storm damage, in this case a breached wall (1.23). The following year, 1220, Henry ordered that the king’s hall in the castle be pulled down and a new one built on the same site (1.24) Another writ to his constable John fitzHugh of Wikenholt orders that receipts from the sheriff’s aid of that year should be put to supplying timber for the building of the king’s hall (ibid., 417). The Pipe Roll for that year duly records a princely expenditure of £89 17s 11½d on this hall. Such buildings were often timber-built: the great hall at Windsor was a timber building in 1295 (Hope 1913, 39), and would anyway have required huge timbers for their roofs. A further 20 shillings were paid to servientum [sic] ville (1.25). This sum appears in the two previous rolls and refers to the serjeants who ensured the safekeeping of prisoners in the town gaol and other ‘policing’ duties on behalf of the reeves (Herbert 1971, 58–9). In 1221 the king’s young sister Joan, on her way to her marriage with Alexander II of Scotland, stayed at the castle for 16 days accompanied by her household and 16 horses. The sum of 30s 11d was paid for the upkeep of the stables, and a further 100s for work on other buildings. In the same year a new buttery (wine-cellar, or butler’s office) and storehouse were added (PR 5 Henry III, 1). A new chimney was built in the ‘king’s high chamber (alta camera nostra)’ in the following year (1.26). The meaning of this term is not without doubt, but is very probably a reference to the Great Tower. The King’s or Great Chamber, specifically a ceremonial withdrawing room leading off the Great Hall (below 87–91), first occurs in 1218 (PR 2 Henry III, 31). 58

More storm damage occurred in the winter of 1222–3, occasioning repairs to roofs on ‘houses’, a wall, battlements and a hoarding (murum et kernellos et hurdeicium). The sheriff’s accounts for the work in the unpublished Pipe Roll of 9 Henry III (1224–5) refer to the wall next the drawbridge (muri iuxta pontem torneicium), indicating that the damage occurred at one of the main drawbridges, at either the outer gate or the entry into the inner bailey in the heart of the castle. The reference to a hoarding is striking because at this date wooden hoardings around castle walls are normally associated with a need for defence and could be put up and taken down as required, their supporting stone or timber corbels remaining in place. They created overhanging galleries from which defenders could bombard attackers below. These were still the jittery final years of the minority following the civil war, and caution continued to be necessary. In 1223/4 the king’s hall and chamber were whitewashed (Rot. Litt. Claus i, 624b); the sheriff in 1225 was reimbursed for the roofing of two garderobes, and the installation of glass in the windows of the new hall (Rot. Litt. Claus ii, 68). Glazed windows were extremely expensive and were found only in royal castles or those of the very wealthiest barons. Henry III was the first to routinely install them in his favoured residences (HKW i, 122–3). The first references are to glass installed on the king’s order in two chapels at Winchester Castle in 1221 and 1222; Oxford Castle’s first known installation of glass was in 1243 and 1246, when Henry had the leaden windows of the chapel replaced (Marks 1993, 92–3). The houses in the king’s court at Wallingford – the courtyard around which the main houses were arranged in the inner bailey – were repaired and re-roofed in 1227, and by 1229 a new kitchen had been built (CLR 1226–40, 48). There is every sign, therefore, that the king viewed Wallingford as a major castle requiring the latest trappings of a royally desirable residence. But Wallingford was not to acquire a lavishly painted and decorated Rose Tower like Windsor, because in 1229 Henry gave the castle and honour to his brother Richard, earl of Cornwall, during pleasure, reserving to himself the prebends of St Nicholas. By 1231 he had been obliged to cede the whole honour and all appurtenant rights to his brother in hereditary fee (Vincent, ODNB), the first time that this had happened to the castle and honour of Wallingford, over which the crown had always maintained tight control. In August 1231 Richard was confirmed by royal charter as earl of Cornwall and lord of the honour of Wallingford (including the hamlet of Clapcot and the borough), the whole held for three knights’ fees, and of the honours of Eye and St Valery (caput at Beckley, Oxon, held in 1086 by Roger d’Ivry), in hereditary fee (CCR 1226–57, 139). The grant of Cornwall alone made Richard an extremely wealthy man. Richard was his brother’s heir apparent until the birth of the future Edward I in June 1239, but the terms of the grant of August 1231 ensured he would have the right to leave the castle and honour to his own heir, a measure of the hold he had already gained and would maintain over his weaker brother. At the time of his second marriage in 1244 to the sister of the queen,

he was further granted the royal manor and hundred of Benson and its three and a half Chiltern hundreds, together with Watlington and Henley, to be held as a bailiwick of the honour of Wallingford with the city of Chichester (Midgley 1942, xx). He thus controlled the whole of south Oxfordshire between Benson and Henley. Described by Goodall (2011, 188–90) as the greatest castle-builder of the mid-13th century, Richard of Cornwall was also a man who liked to spend money on the latest in castle fashions (and he was an outstandingly talented financier who frequently bankrolled his brother, among others). Unfortunately, the records of the earl’s spending on the castle do not survive.3 We only have the statement of Matthew Paris that after the dedication of his new abbey at Hailes, Gloucs, in 1251, which had cost 10,000 marks (£6666 13s 4d) to build, Richard remarked that he wished he had spent half as well the vast amounts he had expended on Wallingford Castle (Matt. Paris. v, 262). Some of his handiwork may have been revealed by Carr’s excavations in the middle bailey, which showed cob buildings deliberately infilled, thus raising the ground level by almost two metres (Christie and Creighton 2013, 183–202) (see Figure 2.4 above). It was during Richard’s tenure that the third wall was built, making the castle concentric on three sides with three walls. The original two walled baileys were a form of concentrification, but the raising of the level of what became the middle bailey and the rebuilding of its wall, together with the low height of the third and outermost wall, revealed in 1389 as a mere 8ft (2.4m), produced a genuine concentrification wherein the lowest outer wall could be defended from the higher inner wall. It was also the time during which a major park associated with the honour was created a few miles away at Watlington, Oxon (VCH Oxon 8, 222).4 The freeholders in the manor all had services relating to the park attached to their tenements, as shown in Rotuli Hundredorum entries for 1279 (RH ii, 33, 815– 16). This was the heyday of the castle: three walls, three moats, one partially round the motte and its Great Tower. It is impossible to say whether any of it was inspired by his time on crusade between June 1240 and July 1241, which included work on an attempted rebuilding of the fortress of Ascalon, his own account of which is preserved in the text of Matthew Paris (Matt. Paris. iv, 143), but it seems very unlikely. Richard anyway had plenty of inspiration nearer to home. The great castle at Dover had had concentric fortifications since late in the reign of Henry II, the earliest such design in England. Further embellishments at Dover castle were made by Henry III, and on his behalf by Earl   Records do survive from his work on Berkhamsted castle, where there was one bailey with a double moat (Goodall 2012, 189). He will have been able to show off his efforts to his brother during Henry’s visits to the castle. 4   This manor had been held by Robert d’Oilly in 1086 and his family proved reluctant to relinquish it until, after some vicissitudes, it was finally granted by Henry III to his brother Richard in 1231 for the service of one knight’s fee. It may never have been formally absorbed into the honour, but remained associated with it thereafter as a bailiwick of the honour. In the time of Edward I it was asserted that the manor had escheated to Henry III on the death of Sarah de Bidun, a tenant in chief, and that he had given it to his brother Richard (RH, ii. 33, 815). 3

59

armoury), residential (halls, chambers, kitchens, chapels) and administrative (wardrobe, treasury), as will be fully explored below. There is also a strong link with the politics of the age, in which all these functions are engaged, and this needs to be taken into account when trying to interpret these documents (Brown 2003, 139–40). The story of Wallingford Castle is as much part of national history as of local history. It featured in many of the struggles between the king and his barons, most notably in the time of Stephen, as discussed above, but also in the time of John, and in the barons’ wars of the middle years of Henry III’s reign. Relatively peaceful under Earl Edmund during the reign of Edward I, who devoted much of his energy to building castles in Wales and subduing the Scots, it re-emerged as an important royal fortress during the disastrous reign of Edward II. Because it exemplifies so many of the themes that have already emerged about the castle’s strategic importance, and documents them so well, the reign of Edward II will be treated here in some detail. Edmund of Cornwall’s siblings all having died, the future for his father’s line rested with his own wife Margaret de Clare. But the couple were ill-matched. By 1290 Edmund was excommunicate for refusing to live with his wife. A deeply religious man who spent a fortune on founding and endowing new and existing religious houses, the problem was clearly deep-seated. A legal separation was arranged in 1294, Margaret agreeing to live in celibacy. This meant that the castle and honour could in due course resume the role intended for it since at least the time of Henry II, as the appanage of the heir to the throne, at that date Edward of Caernavon who succeeded his father as Edward II on 7 July 1307. As early as 1296 Edward I may have been making preparations to take over the immensely valuable lands of Earl Edmund as an escheat. In that year, during which he entrusted a number of high-value Scots prisoners to Edmund’s keeping at his castles of Wallingford and Berkhamsted, he ordered the earl’s treasure at Berkhamsted to be brought to London. Edmund acted twice as regent during the reign, but like his father was mostly distinguished by his wealth and his willingness to loan large sums to the king. In taking back Wallingford Edward was maintaining a tradition: he had been very close to his uncle and godfather Earl Richard, and had entrusted his own eldest son John to Richard’s ward, though the young prince died aged just five at Wallingford in 1271.5 Apart from the brief and inglorious tenure of Edward II’s favourite, Piers de Gaveston (k. 1312), the castle thenceforth remained closely associated with the king and his family. Edward II was an unworthy successor to his father from the first, unhealthily devoted to his vain and arrogant favourite, the Gascon knight Piers de Gaveston, to whom,

Richard, acting as regent in his brother’s absence (HKW ii, 629). Wallingford’s own concentric fortifications were achieved before 1256 and had quite probably been inspired by the royal castle at Dover. His nephew Edward I would complete the design of the Tower of London around 1275 with a second curtain wall, running concentrically outside the inner curtain built by Henry III (Ashbee 2008, 147–8). Richard’s attention turned elsewhere when he was elected king of Germany in December 1256, and his work on Wallingford effectively ceased. It had played host to the king at various times, including Christmas in 1243 (Matt. Paris. iv, 283). In 1256 the queen’s sister Sanchia of Provence was married to Earl Richard on 21 November (St Cecilia’s day) at Wallingford, followed by a nuptial feast at Westminster (Luard 1869, 91). But for all the emphasis on architectural display, it remained a working castle. On 11 February 1257 Richard wrote from Wallingford to John, archbishop of Messina, announcing his election as king of the Romans. Absent for part of 1257 and 1258, he hastened back to Germany in 1262 in the vain hope of the imperial title. He left behind an unsettled situation which had arisen over several years during which the king had affronted many barons by inept and corrupt government, and over-reliance on a few favourites, many of them related to his Provencal wife. With Richard out of the country in 1257–8 and again in 1262, the barons, led by Simon de Montfort, earl of Leicester, seized their chance and directly challenged the king. They forced the king, and subsequently Richard, to agree to the so-called Provisions of Oxford, a reform agenda promulgated at a parliament in Oxford in April 1258. The situation continued to deteriorate, and between 1264 and 1267 it degenerated into civil war. Richard of Cornwall was captured at the battle of Lewes on 14 May 1264 and, after a brief stint in the Tower of London, made prisoner in his own castle at Wallingford in the custody of his sister Eleanor de Montfort, wife of the rebel barons’ leader, accompanied initially by his nephew the Lord Edward. After an unsuccessful rescue attempt, Richard was transferred to Kenilworth and kept in chains, since his nephew Edward, who had previously been sent on to Hereford Castle, had escaped from there in May 1265. The death of Simon de Montfort at the battle of Evesham led to Richard’s release by the younger Simon on 6 September 1265, after the Lord Edward had negotiated the surrender of Wallingford and Berkhamsted. Richard was back at Wallingford, to the delight of his household, on 9 September. Eleanor’s son Guy would later murder Richard’s eldest son Henry, so that when Richard died on 2 April 1272, a few months before his brother the king, his heir was his second surviving son Edmund. With Edmund, the castle re-emerges from the shadows into at least a candlelit world of documentation. The long overdue discussion of this valuable material forms Part 3 below.

  The cause of the serious quarrel between Edward and his father Edward I, which led to the first exile of Piers de Gaveston early in 1307, was that the younger Edward wished to bestow the earldom of Cornwall on his friend (ibid., 122–3). This was held to be a wholly inappropriate grant of a major asset to the crown, which Edward I may have intended to benefit his younger sons (ibid., 122–3); Wallingford Castle was provisioned against the coming of the king’s children Thomas and Edmund in October 1305. His eldest son John had died aged five in the care of his great-uncle Richard at Wallingford Castle in 1271.

5

‘Diligently and safely guarded’: Wallingford Castle and the overthrow of Edward II The accounts for castle works at Wallingford evidence the various functions of a castle, judicial (prisons, great hall as courthouse), military (gatehouses, moats, towers, 60

in the first formal act of his reign on or soon after 6 August 1307, he gave the vacant earldom of Cornwall, together with all the lands held by Earl Edmund at the time of his death, including Wallingford Castle, its honour and bailiwick (Philipps 2010, 126–7). At a tournament held at the castle on 7 December to celebrate his marriage to Edward’s niece, Margaret de Clare, Gaveston and his men humiliatingly defeated the leading nobles of the kingdom, whose enmity by then was already assured. At that time Edward was concluding his negotiations over the dower of his intended wife Isabella of France. Before he left England for his marriage on 22 January, he had made a number of arrangements for his absence. Gaveston was appointed custos regni, and the king sent a writ to the constables of the key castles of Dover, Rochester, London, Windsor, Wallingford, Corfe, Isle of Wight, and others, ‘desiring that the castles of his kingdom should be diligently and safely guarded and defended for the greater security and tranquillity of his people’(1.30). A baronial backlash led to Edward and Gaveston strengthening their castles at Windsor and Wallingford early in 1308. When his father-in-law Philip IV of France let it be known that he supported the baronial opposition, Edward had to give way, agreeing that Gaveston would go into permanent exile and the earldom of Cornwall be restored to the crown. During that exile (which lasted only until Gaveston’s reinstatement on 5 August 1309) the king held the revenues of Wallingford and appointed John de Clinton of Maxstoke as constable and seneschal of the honour. The building accounts rendered in Clinton’s name in 1309 will figure prominently in the following section. Baronial opposition to the king and resentment of his favourite continued to mount, leading eventually to the arrest of Gaveston after his surrender at Scarborough to the earl of Pembroke in May 1312. He was being escorted under guard to Wallingford to meet the king when he was seized by the earl of Warwick, and was murdered at Blacklow Hill on 19 June (Phillips 2010, 180–91). Edmund Bacun had been appointed constable of Wallingford Castle in 1311 in the anticipation of Gaveston’s return from exile and shortly after Gaveston’s death, on 28 June, he was ordered to provision the castle with victuals and other necessities and to repair the castle buildings. Ralph Restwald ‘keeper of the king’s things in the tower of that castle’, was to receive arrears of wages, and Bacun was to be allowed his wages and those of 10 men-at-arms with him from Michaelmas to St James, and the costs of Bouard de Gaveston and his men for five weeks. Bacun was given a commission of array to raise footmen in the county in August 1312 (CCR 1307–13, 428, 468, 486). In 1314, the year of the disaster at Bannockburn, the king ordered the acting treasurer and the barons of the Exchequer ‘to allow to Edmund Bacun, steward of the honor of Wallingford, in his account £131 3s 4d wherein the king is held to him for the wages of him and his three esquires in the present war of Scotland and for four large horses bought from him.’ (CFR 1307–19, 204). A month earlier Edmund had been ordered to expend £20 in repairing the houses and bridges of the castle. Further writs required repairs to the buildings and stables in 1316 and 1317, the year that the king finally

Figure 5.21 Grant of the castle of Wallingford in dower to his wife Isabella by Edward II, 26 April 1317. Reproduced by permission of the UK National Archives, E40/164.

assigned the castle and its honour as dower to his wife Isabella (Figure 5.21), a settlement detailed in letters close given on July 10 1319 at York (CCR 1313–18, 350, 357, 394; CCR 1318–23, 149–50). Notwithstanding this grant, the king began using the revenues of Wallingford for the maintenance of his heir Edward, earl of Chester, by 1314, and part of those of Cornwall by 1318 (CPR 1317–21, 154; 1.32-3). The orders continued to flow. In December 1318 the constables of Windsor and Wallingford were ordered to permit ‘Gilbert de Boghere and Thomas his brother, whom the king is sending to repair the defects of the equipments [arms] of the castle, to enter the castle, and to pay each of them 4d a day from the day when they enter the castle for so long as they shall stay therein’ (ibid., 39; 1.34). He had stayed at the castle, issuing letters patent, in January 1317, and 26 April to 2 May 1318 (CPR 1324–7, 391–2, 539, 600, 608). He issued letters close from Wallingford dated 13 March to 15 March 1321 and 1 May 1 to 4 May of the same year (CCR 1318–23, 293, 298–9, 363–4, 368, 371). 61

But for the ill-organization of the baronial opposition led by the king’s unpopular cousin Thomas, duke of Lancaster, progress might have been made after Gaveston’s death. Instead, the king’s disasters in Scotland continued, adding to his constant need to extort new monies from his hapless subjects, whilst all the while his desire to avenge Gaveston still burned. His opponents drew up a list of regulations restraining royal abuse of financial expedients which they tried many times to oblige the king to respect. Civil war was only finally averted when Lancaster and the king made peace in August 1318. Edward had already formed another strong emotional attachment to an even more rapacious favourite, Hugh Despencer the Younger, a man of unlimited ambition, and a coterie of others. The unstoppable rise of these men thoroughly alienated a growing number of barons and household knights. Matters came to a head and the king was forced to send the hated Despencers into exile in 1321. As before, he looked for an opportunity to recall his friends as soon as possible to avenge the slight to his majesty. This occurred on 1 January 1322 resulting in civil war in the course of which Thomas of Lancaster was summarily executed on 21 May 1322 after his defeat at Boroughbridge. One aspect of the conflict was the opening of a rift between Edward and the barons of the southern Welsh Marches, among whom was Roger de Mortimer of Wigmore who had joined the attacks on the Despencers that led to their brief sentence of exile in 1321. Edward engineered a resumption of hostilities and on 14 January 1322 Roger and his uncle of the same name were imprisoned in the Tower of London. Maurice de Berkeley of Berkeley Castle and Hugh Audley the Elder, were imprisoned at Wallingford Castle. This led to one of the stranger episodes in the castle’s history on 11 January 1323 when, in the words of Seymour Philipps (2010, 438–9): ‘an attempt was made to free Maurice de Berkeley and other prisoners, including Hugh Audley the Elder, from captivity in Wallingford Castle …where they had been held for almost a year. According to the Vita Edwardi Secundi, one of Berkeley’s squires, later identified as Roger Wauton, who had visited him regularly in prison, one day entered the castle with several companions, with the guard’s permission and without arousing any suspicion. That night Maurice invited the constable and all the doorkeepers and watchmen to dine with him. While they were dining the squire and his companions rose and demanded the keys of the castle; the squire then opened a postern gate [probably the Dernegate, on the northeast, with easy access to the river] and let in 20 more men. All this was done in complete silence and the conspirators might have escaped with all the prisoners, but for a young boy who realized that something was wrong and warned the mayor of the town [John Mariot] that the castle was lost. When the sheriff [Drogo de Barentin] arrived, the conspirators tried to buy time by claiming that they had entered the castle with the king’s authority. The earls of Winchester [the Elder Despencer] and Kent then came and threatened to attack the castle. On about 25 January the conspirators opened the gates and withdrew to the castle chapel in search of sanctuary: “Thus having entered, the earls found Maurice in custody as usual, and the rest in the chapel.” Of those involved, the ringleader, Roger Wauton, and two others…were taken to the

king at Pontefract, where Wauton was hanged and the others imprisoned. It had been a close call and was a foretaste of the attempts made in 1327 to release Edward II from captivity in Maurice de Berkeley’s own castle.’ By April of 1323 the king had learned that the seizure of Wallingford had been the first stage of a wider conspiracy to seize also both Windsor Castle and the Tower of London, as well as to release Roger de Mortimer; the latter escaped six months later and made his way to France where he would provide a constant source of anxiety to the king for the next three years.6 In 1324 Queen Isabella was sent to France by Edward to negotiate with her brother Charles IV with whom a dispute had arisen about the homage due to the French king for English holdings in Gascony. She agreed a truce in May 1325 but rather than appear himself, Edward agreed to send their eldest son Edward in his place. Isabella then refused to return to England, keeping her son with her, citing her fear of harm from the Despencers as her justification. During the latter part of her stay in France Isabella became the centre of a group of exiled dissidents, including Roger de Mortimer, with whom she formed a liaison. The queen, her son and supporters returned to England on 25 September 1326 with the stated intention of ending the king’s misgovernment. The king’s party collapsed rapidly and he himself can be seen to be already in flight by 7 October when he was briefly at Wallingford. Isabella followed him to Wallingford and in a proclamation issued there on 15 October she violently denounced the Despencers as evil counsellors (1.42). London rose in support on the same day. The king was captured on 16 November; the Younger Despencer was executed at Hereford on 25 November. By attempting to reach Wales Edward had technically abandoned his kingdom, which formed the pretext for making his son the nominal regent whilst a deposition process was set in train. The queen, her son and Mortimer celebrated a triumphant Christmas at Wallingford Castle, before going to London on 2 January, where they were conducted to Westminster (Stubbs 1882, i, 185). This festivity was an important opportunity for Isabella and Mortimer to consolidate alliances that would bring about the replacement of Edward by his son (Harding 1985, 53). It was in fact a council, attended also by the archbishop of Canterbury and other leading clerics, and one that made the key decision that Edward II could no longer rule (Philipps 2011, 520–4). Shortly afterwards, on 7 January 1327, parliament decided to call for the king’s removal; articles for his deposition were presented on 13 January. On or soon after 20 January he abdicated in favour of his son, who was crowned Edward III on 1 February. A danger so long as he lived, Edward II was moved to Berkeley Castle in April 1327, where his gaoler   Roger de Mortimer escaped from the Tower of London on 1 August 1323, with the assistance of the deputy constable, Gerard d’Alspaye. The constable Stephen de Segrave and the guards were drugged whilst dining, and Gerard let Mortimer out of his cell. Passing through the kitchens, they let themselves down the outer wall using a rope ladder; a boat awaited on the riverside provided by fellow conspirators. Mortimer was ferried to a mill where other conspirators provided horses for a swift escape (Philipps 2011, 440–1).

6

62

Figure 5.22 Account of John de Stonor for the guard of Wallingford Castle on behalf of Queen Isabella, 7–16 October 1326. Reproduced by permission of the UK National Archives, E101/17/21.

was Thomas de Berkeley, whose father Maurice had died in prison at Wallingford in 1326. His ultimate fate remains a mystery, though he may have survived for several years (Mortimer 2010). A number of very revealing documents about Wallingford Castle emerge from this period. Incompetent though he was, Edward was conscious of his status and highly aware of the military potential of castles. He took a close interest in some, especially Knaresborough, in Yorkshire, which was one of those he had intended for Gaveston. When trouble loomed in 1317 he ordered that a number of castles be readied against possible civil war, issuing writs detailing the level of manpower required. Wallingford and York were to have 40 men each, others between 20 and 30 (1.32–3). In both cases the orders were adhered to, but the men mostly consisted of archers rather than men-at-arms (Prestwich 1992, 163). Castles were not kept garrisoned for long on account of cost. Sixty-three castles munitioned during the civil war of 1318–22, including Wallingford, were stood down in April 1322 (CCR 1318–23, 437). The dangers of inadequate garrisoning were demonstrated by the attempt to seize Wallingford Castle in 1323. The same cost-cutting accounts for why so many royal castles were often badly in need of repairs. If they were not being used as residences there was no interest in maintaining them. Another problem with castles was that storage areas were mostly unheated and often damp. The inventory of arms held at Wallingford Castle taken in 1325–26 and again in 1326–7 (below, 95ff.) included such sorry items as broken and damaged crossbows and bolts and rusting armour, as well as 40 or 60, depending which account one uses, shields painted with the arms of Edward II and Gaveston (Prestwich 1992, 165). The same problems befell the food supplies held in a castle. They often held more than they needed for a small group of residents, or for dispatch to other royal castles; consequently, there was a great deal of waste as supplies mouldered before they could be used or sold off, or were attacked by rats and mice, as indicated in accounts from the same date and later (ibid.; TNA, E101/17/22). Prestwich (1992, 159–62) has pointed out that keeping castles in a fit state was not helped by constant changes

of constable by a king whose use of patronage was more liberal than his supply of common sense. In that respect Wallingford’s position was fairly stable in this reign. His father’s constable John de Clinton of Maxstoke died in 1310, when Edward appointed as his successor Edmund Bacon, who occurs as constable and steward of the honour until 1318, after Isabella was assigned the castle in dower. If not replaced before, he will not have survived long in office after the brief but worrying seizure of the castle by the king’s enemies in 1323, which became the subject of an inquiry led by Henry Lescrop, John de Stonor and John de Bousser (1.35–6). The king’s steward Richard Damory was briefly made keeper. John de Stonor, a justice of the King’s Bench, was made keeper in 1326, when William Marshal had been constable since at least January 1325 (1.40). John returned one of the most striking of the Wallingford Castle accounts in January 1327 (1.41). It related to the crucial 14 days from 3 to 16 October during which Isabella of France had stayed in the castle and called for the overthrow of her husband (Figure 5.22). The figures are extraordinary. John accounts first for his wages for four days service under Edward II (virtute cuiusdam commissionis eiusdem domini regis), from 3–7 October 1326, and those of four serjeants, six esquires and six grooms. Edward was certainly at the castle on 7 October, issuing from there a writ of aid for John de Stonor, who was appointed during pleasure to the custody of the corpus of the castle and of the prisoners there. The king had first ordered John to Wallingford on 26 September, when he was himself still at the Tower of London. His movements after that are uncertain; he fled London in the company of Despencer and his chancellor, who took the Great Seal with him. Letters close were issued at Acton on 3 and 4 October, at Westbury on 6 October and Wallingford on 7 October. John de Stonor’s account suggests that Edward may have himself reached Wallingford on 3 October, leaving on 7 October after committing the castle to John (cf. Tout 1928, iii, 2). Isabella was certainly there by 14 October and perhaps before. John at any rate dated his allegiance to her from 7 October, as we learn from the second section of his accounts. In this, ‘by virtue of another commission’ (virtute alterius commissionis) he accounts for his own wages and those of a knight who remained with him, 12 esquires and 8 63

crossbowmen guarding the castle by day and night, for the nine days from 7 to 16 October. On 9 October they were joined by 12 armed watchmen who were paid 6d night and day. A considerable extra force was present on 14 October, when 144 foot soldiers guarded the fortifications in both outer baileys by day and night, the number dropping to 124 on 15 October, the whole costing £66 4s 4d, a staggering sum that dwarfs many of the annual running costs of the castle. Payment was authorized by Queen Isabella, using the privy seal in her son’s name. This document sheds considerable light on the chaotic days following Isabella’s invasion and the eventual capture of the king. Isabella was close on the heels of Edward in early October, when both saw Wallingford Castle as a stronghold and a refuge. Indeed, one of the most striking things about this reign, when castles frequently played a defensive role for the king against his numerous enemies, is that Wallingford was clearly an important castle that Edward associated with the crown – himself – even as he tried to make it Gaveston’s, between 1307 and 1312, and certainly after he grudgingly conceded it to Isabella in dower. It is ironic that one of the best fortified royal castles in the country – siege-engines, crossbows and all manner of armaments are recorded in an audit of its arms store in 1326 (below, 95ff.) – should have played such a key role in the king’s downfall. Isabella soon wasted her hard won advantage in replacing her husband with her son by ill-judged decisions, extravagance and greed and her continuing liaison with the arrogant Roger de Mortimer. She took for herself many of the jewels and valuables belonging to Despencer and left in the Tower. Some she gave to her clerk Stephen de Tong when he celebrated his first mass in the Great Chapel at Wallingford in her presence at Christmas 1326 (TNA, E361/9, m.19). She appears to have replaced William Marshall with Miles de Beauchamp as constable on 14 October 1326 (1.43), obliging William to surrender for her use the provisions and arms of the castle to Swayn de Mortele, a local man and former mayor (1306, 1315, and later 1330–2: Herbert 1971, App. p. xiv). The first tentative moves towards establishing his independence saw the young king restore William to his position as constable and steward in 1329. Edward III finally threw off the tutelage of his mother and her lover in October 1330, after a coup at Nottingham Castle. These events form the backdrop to two of the accounts to be examined below, the first from 1327, ordered in the new king’s name but designed to survey her dower castles for Isabella’s benefit, the second from the earliest years of Edward’s own reign, 1331–34, clearly showing the initial wariness of the new regime. Isabella was forced to relinquish her dower, including the castle of Wallingford, on her fall in 1330, but she was compensated with a handsome annual allowance of £3000 (CPR 1330–34, 48). The settlement tacitly recognized that as wife of Edward II from 1308 she had had to put up with a great deal, including personal danger, over many years, and had done so with dignity. Her popularity among the people facilitated the success of her invasion in 1326 (Menache 1984), but her year in France, itself full of danger, had made her vengeful and that was her undoing. Wallingford was given by Edward III to John of Eltham,

his younger brother and heir apparent until 1330, who had been created earl of Cornwall, in 1328. John died unmarried on 13 September 1336. Edward then erected the earldom into a duchy and bestowed it upon his eldest son Edward, the future Black Prince, born in 1330, on 17 March 1337. This was the culmination of a process that began no later than 1154/5, both formalizing the link between the wealth of Cornwall and the Crown, and the role of Wallingford as one of three key Thames Valley castles with the Tower of London and Windsor, and the one that formed the link between the heir to the throne and the king. Thenceforth the duchy was (and is) invested in the first-born male heir from the day of his birth; the associated titles of earl of Chester and prince of Wales were conferred by special investiture by the reigning monarch. Wallingford thus formed part of the formal appanage of the king’s heir, the prince of Wales, until it was detached from the duchy in 1540 by Henry VIII. Edward III to 1540 Edward III was a frequent visitor at the castle between 1330 and 1336, staying again in 1340. Following two years of building work and strengthening, he kept Christmas and New Year there in 1333–4 (Ormrod 2011, 613, 615, 616, 630); surviving Wardrobe accounts detail clothes and arms made for the king over this period, as well as items such as 14 hobby-horses for games in the hall with his guests (ibid., 104). Letters close and patent were issued at Wallingford from 9–11 November and 8 December 1328, 29 March to 27 April 1329, 2–4 August 1329, 14 January and 27 March 1330, 8 October 1333 and from 27 December 1333 to 14 January 1334, and between 14 May and May 22 1336. For much of the reign Wallingford was held by the king’s redoubtable heir, Edward, the Black Prince, lord of Wallingford from 1337 until his death on 8 June 1376. The castle was well-maintained by the standards of the day, but its huge enceinte necessitated constant repair. In the 1350s, the Black Prince ordered remedial work on the eastern curtain wall, overlooking the river, which was causing concern (1.47) A significant rebuilding and underpinning was carried out 40 years later, suggesting that measures taken at this date were insufficient. The next wave of building repairs at Wallingford, for which the accounts span January 1363 to January 1367, was initiated less by the prince’s marriage on 10 October 1361 to his cousin Joan of Kent, who made Wallingford her chief residence, than the devastating storm of January 1362 which did widespread damage (below pp. 94–5). Further work was done and accounted for in 1375, in the final year of the prince’s life, much of which had been spent on his military campaigns in France. Joan lived until 1385, her residence at the castle maintaining royal interest there. Richard II had been associated with the castle since childhood and well understood its importance, as did his cousin and nemesis Henry of Lancaster (Rickard 2002, 38). Several writs from the reign order work on the upkeep of the castle, which underwent major repairs in 1389–91, before a flying visit by the king, and again in 1392 and 1398–99. In November 1397 Richard appointed 64

his cousin Thomas Beaufort, marquess of Dorset and Somerset, constable of Wallingford, adding several other offices including constable of Dover Castle and the Cinque Ports, in February 1398. The king’s regime was already doomed by this point. Richard briefly met his young wife Isabella at Windsor on his way to Ireland in late June 1399, giving orders that she be removed to the greater safety of Wallingford thereafter. During his absence his uncle Edmund, duke of York, was custodian of the realm. Edmund seems to have briefly envisaged Wallingford as the centre of royal administration during July 1399, spending a small fortune totalling £1296 23s 4d (Biggs 2002, 130) in munitions as the king clung desperately to his throne. Most of Richard’s supporters were with him in Ireland, leaving his position vulnerable when his rival Henry of Lancaster landed about 1 July, having left the exile imposed on him by the king in the previous year. The Wallingford garrison surrendered as soon as news of Richard’s arrival as Henry’s prisoner at Flint on 16 August had been received, York himself, and Thomas Beaufort, having already joined Henry’s forces at Berkeley on 27 July, around the time that Richard returned from Ireland. Edmund of York was the new king’s most trusted advisor until his own death on 1 August 1402 (Biggs 1994). Thomas Beaufort (d. 1410) was also close to Henry IV, who charged him with escorting the child Isabella back to France in 1401. Henry IV was briefly at Wallingford when fleeing a plot by Richard’s men to assassinate him and his family at Windsor in 1400 (to which York had alerted him), and again in May and June 1401. The new king ordered repairs to a number of castles, including Wallingford in 1403– 1404 (CPR 1401–05, 351). In 1403 his son Henry (V) was assigned the revenues of the honour as prince of Wales, having first received them on 15 October 1399 as duke of Cornwall. Building accounts survive for 1408–9 and 1411–12 (incomplete), but are modest in scale, reflecting the severely strained finances of the king at that time. The continuing but limited use of the castle thereafter does little to mask a steady decline, of both castle and town, whose fortunes were closely connected (see below 102–3). Henry V assigned the castle and honour to his wife Katherine de Valois in dower and she held it until her death in 1437. Major repairs were undertaken in 1424 (CPR 1422–29, 193), with a view to making it a suitable residence for the queen and her young son Henry VI, an infant at his accession in 1422; in 1428 the castle was formally appointed a summer residence for the king. In 1438 a new stone prison, with a hall, kitchen and two chambers for the gaoler, were built at a cost of £149 9s 9d. The accommodation included a 15ft-deep pit for the keeping of felons. The accounts for this work are very different from those already discussed. They give a considerable amount of detail about the works, including the use of malmstone and freestone, and room dimensions (see below, pp. 82–3). It was probably built a bit too late to receive the king’s stepfather, Owen Tudor, who was ordered into close custody at Wallingford following his escape from Newgate prison in London in January or early February 1438 (Rymer 5(i), 46). The constable by then was William de la Pole, earl of Suffolk,

who had become Henry’s personal favourite by the early 1440s. William had married Alice Chaucer, daughter of the king’s butler Thomas Chaucer (son of Geoffrey), by 1430. The pair were based at Alice’s family’s chief manor of Ewelme, where they later founded a hospital. Thomas probably facilitated William’s appointment as constable and steward at Wallingford, then still the home of the dowager queen, in June 1434. A member of the king’s minority council by 1431, William formed a strong bond of affection and loyalty with the king, which would lead to his undoing. He had brokered the king’s marriage to Margaret of Anjou during 1444–45, which had allowed the king to assume his majority, with the newly-appointed marquis of Suffolk as his chief councillor. The problem was that Henry VI was inadequate to the task of exercising royal authority, which left his favourite dangerously exposed in 1449–50, when serious civil disturbances in London and East Anglia, arising out of the rapidly deteriorating financial situation, brought about by the endless complications of the wars with France, led to a denunciation of Suffolk in parliament. It was alleged that he had refortified Wallingford Castle and stuffed it with guns and gunpowder with the aim of supporting a French invasion. He was arrested and charged with treason in 1450. In exchange for submitting himself to the ‘rule and governaunce of the king’, he was exonerated of all charges and exiled, thereby saving the whole regime from a public washing of its laundry (Watts 2004). On his way into exile his ship was boarded by men from a privateering vessel and he was murdered. Henry VI’s reign limped on until his overthrow in 1461 by Edward IV of York (great-grandson of Edmund of York), though his final deposition and murder did not come until 1471. His indomitable widow had headed the Lancastrian cause between 1461 to 1471, to no avail. After losing the final battle at Tewkesbury, where her only child Edward was killed, Margaret of Anjou was imprisoned at Wallingford in the charge of Alice Chaucer, dowager duchess of Suffolk until 1475, when she returned to France. Alice had held her position at Wallingford through an alliance with the Yorkists. Her son, John de la Pole, married in 1458 Elizabeth, sister of the future Edward IV, who created their eldest son John earl of Lincoln in 1463. John would fight for Richard III, brother of Edward IV, at Bosworth, and joined the rebellion in 1486 of Lambert Simnell against Henry VII, for which he suffered forfeiture. His father had remained loyal to Henry VII, and was rewarded by the regrant to him of the constableship of Wallingford Castle and stewardship of the honour, previously alienated to Francis, Lord Lovell. Lovell had been a ward of John, duke of Suffolk, and was rapidly promoted during the final year of Edward IV in 1483 and throughout the short reign of Richard III. Richard III had hoped that Lovell could strengthen his regime in Oxfordshire and Berkshire, where there had been a rebellion against him in October 1483. Lovell had already replaced Edward IV’s stepson Sir Richard Grey as constable of Wallingford by that time. Highly valued by these kings for his valour and probity, he was another of the rebels of the early years of Henry VII’s reign, a key figure 65

Figure 5.23 Royal castles in the reign of Henry VI (1422–61), after History of the King’s Works i, 26, p. 239; © Queen’s Printer and Controller of HMSO 1963.

in what has been called ‘the Thames Valley conspiracies’ against him (Luckett 1995). After Bosworth he had fled abroad and joined the conspiracy of John, earl of Lincoln, against Henry VII in 1486. Lincoln was killed in battle at Newark, from which Lovell apparently fled, though there is no certain knowledge of his fate thereafter. The Lancastrian kings maintained an interest in the castles of their northern heartlands, but with the exception of Nottingham, Windsor and Wallingford, they failed to maintain the old royal castles of the south and the midlands in a state fit for either habitation or defence. Royal castles that had controlled the main roads of England, such as Oxford, Northampton, Shrewsbury and Lincoln, were left

to decay. Under Henry VI, this failure to understand, as his predecessors had, that control of the interior meant control of the periphery, would weaken the monarchy, prolong civil unrest and lead to the deposition and murder of a third king – a stark contrast to the mastery of the kingdom achieved under William the Conqueror who had a strong castle and a loyal garrison in every county town (HKW i, 238–41) (Figure 5.23). The Yorkist kings invested more in castles, both as favoured residences (Nottingham, Fotheringhay, Windsor and Dover), and as fortresses (Tower of London). Ministers’ accounts show that Wallingford was kept in good order under Edward IV, when its revenues were used for his heir (TNA, SC6/1302/1). Allowing royal castles to 66

Maintaining a royal stronghold: building accounts 1284–1438

fall into ruin risked weakening monarchy and undermining government because they performed a complex but coherent set of interlocking functions in administering, regulating and safe-guarding the towns and regions of the kingdom. But castles demanded huge sums in maintenance, especially the oldest ones. With significant exceptions such as Windsor and Dover, new forms of buildings and institutions started to take the place of the old stone castles. The process accelerated under the Tudors. By the later 15th century the heyday of Wallingford Castle was well behind it. Apart from accounts for the building of a new prison and accommodation for the gaoler in 1438–40, no full building accounts have been traced further than 1411. Sundry ministers’ and receivers’ accounts for the 15th and early 16th centuries attest ongoing routine maintenance work into the reign of Henry VIII, but there is no firm evidence for any major works, the sole survivor apparently a warrant for repairs ordered by Arthur, prince of Wales, sometime between 1490 and his death in 1502 (TNA, SC1/44/85). His brother Henry became the new Duke of Cornwall in the following year, but after his accession as king in 1509 there was no Prince of Wales until 1542, two years after Henry detached Wallingford from the Duchy in 1540. Henry’s relationship with Wallingford began well enough. It seems to have been one of his residences in July 1518, when the court removed to Bisham following a local outbreak ‘not only of the small pokkes and mezils, but also of the great sickness’ (Hedges 1881, ii, 83–4); and in August 1523 Wallingford and Windsor were both excused the collection of a four-year subsidy recently granted to the king by the Commons (ibid., 85). We also have a letter of the last dean of St Nicholas stating that Henry had extensively rebuilt the collegiate buildings in the middle bailey (ibid., 296–8). His disenchantment with the place may have been connected to his tempestuous private life. Richard III’s man Lord Lovell, a constable of the castle, was succeeded by his sister Frideswide who married Sir Edward Norris. Their son Henry Norris became a favourite of the king, who made him a life grant of the keepership of the parks of Watlington and Ewelme in the honour of Wallingford in 1523. On 29 November 1535 Norris was appointed constable of Wallingford Castle. But in May 1536 he became implicated in the fall of Henry’s second wife, Anne Boleyn, being arrested with her brother, Lord Rochford, tortured, and executed on 17 May. No successor as constable was appointed for several years. The aim of these first two sections has been to provide an overview of what we might call the institutional genesis of the castle and its functions, and to establish the key features of the role it played in both English national history up to 1540, and also its role in local government. It remains now to seek elucidation of the architectural evolution of the castle from what accounts survive. The documents we shall be using are very difficult to interpret, but the task will be made easier by the clear historical narrative that has been established here.

The King’s Works at Wallingford: the documentary base We have already seen that a certain amount about the castle’s development can be gleaned from occasional accounts summarized in the annual Pipe Rolls, sometimes amplified by surviving letters close or patent. Periods of direct royal control are most likely to yield references in Exchequer or Chancery sources to the castle. Great magnates like the earls of Cornwall had already developed offices mimicking the offices of state, so they had their own chanceries recording their decisions as to estate management or building maintenance, and so on. But their records have not necessarily survived as well as royal ones. We have very little direct information from the time of Earl Richard (1230–72), but we have three important texts from the time of Earl Edmund (1272–1300), comprising two socalled ministers’ and receivers’ accounts for the running costs of the castle, and his re-foundation charter for St Nicholas. As a builder, Earl Edmund is mostly remembered for his religious foundations at the priory of Ashridge, Herts, and Rewley Abbey in Oxford, and it was in this way that he made his mark at Wallingford, where he effectively converted the middle bailey into an ecclesiastical enclave. The routine ministers’ and receivers’ accounts surviving from 1284 and 1296 from the earldom of Cornwall are welcome though isolated gems, as will appear hereafter. These annual audits of expenditure and revenues in the honour of Wallingford include a section on the castle. They represent a huge advance on the information available from the 12th and earlier 13th century documentation, but they still leave us a long way short of fully understanding the castle layout. Fortunately, there is another class of documents that can add considerably to our knowledge of the castle topography, its function, and its role in the local economy. These are the accounts of works done that had to be presented when a campaign of repairs or new building had been instigated. Wallingford was a royal castle so anything done on the king’s account appears in a number of Exchequer records, such as Pipe Rolls, Close and Patent Rolls. Even during the times when the castle was in the hands of a member of the royal family, such as Henry III’s brother Richard of Cornwall or Edward III’s son, the Black Prince, the king could and did order that the castle be kept or put on a military footing and prepared for defence. Several accounts survive from the 14th century when work was done in the king’s name, and two from the early 15th. Now kept in The National Archives at Kew, they have never been edited. The work detailed in the accounts was initiated by a writ stipulating what was required. The constable then had to organize a series of contracts (indentures) for aspects of the work, such as hire of a master mason and his associates, carpenters, labourers, etc. and both parties would agree to a fixed rate (tascha) for the job. In the mid12th century a system of viewing was introduced to ensure that work ordered and paid for was carried out, the sheriff accounting for the expenditure overseen by two reeves 67

from the town. By the end of the century the constable rather than the sheriff (though sometimes the same person) was responsible for arranging, overseeing, paying and accounting for the works. In the 13th century a system of viewers who handled and audited the monies involved was developed, which during the 14th century became the Office of the Clerks of the King’s Works (HKW i, 52–4). The resulting accounts are the work of the ultimate bureaucratic bean-counter. Every last nail is accounted for, every single artisan or labourer’s wage, down to the last farthing. They also account for the pay of the clerk who wrote down the accounts and the parchment on which they were written. These documents have nothing to do with architecture, and not much to do with construction methods. The work of accountants, they present formidable challenges of interpretation. They divide a building campaign into carpentry, masonry, leadwork, tiling, and so on, so the same particular job will be divided up between all categories. Trying to put it together is like assembling a jigsaw puzzle out of slices of onion. There is

very little attempt to be precise about where a particular job took place, though topographic references relating one building to another do occur and may provide some valuable insights into layout. The text is generally heavily abbreviated, to the point that considerable doubt arises as to the syntax in places, including essentials such as singulars and plurals, punctuation is erratic or non-existent, and the vocabulary is, to say the least, esoteric. They are written in Latin, but the influence of the French vernacular begins early. Increasingly, throughout the 14th century English words also occur, so that Rubea Camera, the Red Chamber, ends up as ‘le Rede-hall’ (cf. Quiney 1999). There are several such accounts for Wallingford from the 14th century which are rich in detail (comprehensible or not) about the castle buildings, and rather fewer from the 15th century, where the building detail is significantly less, but where the socio-economic detail is often much richer. This follows the national trend noted by HKW (i, 976). The surviving accounts are from 1307–9, 1327, 1331–34, 1363–66, 1369, 1375, 1389–91, 1408 and 1411.

Figure 5.24 Survey of Wallingford Castle, 1327. Reproduced by permission of the UK National Archives, E143/10/2.

68

Their evidence is usefully supplemented by Ministers’ and Receivers’ Accounts surviving from the reign of Edward II to the end of Henry VIII’s, though none of them is as detailed as those surviving from 1284 and 1296. The earliest such account is 1284–85 and it gives a taste of the challenges ahead. Unfortunately, it is written on a single very large membrane considerably damaged first by folding, and later by singeing and water. Not a single one of its very long lines survives intact, yet the basic script is very legible. This is a considerable loss, but nevertheless the document has a lot to offer, some of it confirming or shedding further light on the later 1296–97 account, the only one so far to have been published (Midgley 1942). Poignantly, it belongs to the period of the only known pregnancy of Edmund’s wife Margaret, with whom he soon afterwards refused to live, thereby inviting ecclesiastical censure.

[and] the Barbicans and the towers on all sides of the said castle with Gaweynes herber. They say that the bridges of the said castle, viz, the interior, middle and exterior bridges require repairs up to the sum of 100 shillings. Total: £165 The Middle Bailey: They say that the middle bailey requires repair in divers places, needing up to the sum of £13 6s 8d for building stone, lime, sand, tiles and workmen’s’ wages for the workers repairing the walls and towers of the said bailey. They also say that in the same place there is a certain ruined stable of beechwood and they need for its repair up to the sum of 40s in timber, tiles and other necessities and the labourers’ wages.

The 1327 Survey of Wallingford Castle

They also say that the prison house of the said bailey and the chambers and the chamber above the bridge need repair and amounts up to 30s are required to supply building stone, lime, sand, tiles and other necessities, together with the workers’ wages. They say also that the chamber called La Studie with other chambers and the bridge both above and below require repair up to the sum of 20s.

Before attempting to explore the unedited building accounts, it will be helpful to try to relate an understanding of what they reveal about the castle’s layout in its late13th to 14th-century heyday to a single fixed point. Such a thing exists in the form of a survey made of the castle in 1327, one of a series of surveys of the castles then still held in dower by Isabella of France, mother of Edward III, in whose name the survey was ordered. (Figure 5.24) There has been confusion about the layout of the castle in terms of number of baileys and moats since the castle was first discussed (VCH Berks 3, 517–31), although Midgley’s edition of the 1296 accounts was a major advance. The issues are put beyond doubt by the 1327 Survey of Wallingford Castle, reproduced here in full for the first time in print (1.45). It provides a concise and clear account of the castle layout at that date. The translation reads as follows:

Total: £17 16s 8d The Inner Bailey: They also say that the inner bailey of the said castle, viz. the Great Tower (i.e. keep) with stair turret with the towers and turrets of the surrounding walls and the chambers are generally in good condition and newly repaired by William Marshall lately constable of the said castle. However, repairs are needed up to the sum of 100s. They also say that there are in the High Tower four hand-mills which need repair up to the sum of 13s 4d. They also say that small defects, viz in the foundation and the corbels above the [Great] Tower and other houses, could be repaired with 40s for lead, nails and the plumber/ leadworker’s wages. They also say that the Great Hall of the Lord King and the Chamber of the Lady the Queen, the Green Chamber, the Red Chamber, the Queen’s Wardrobe and the chambers of the knights and esquires want repair in respect of tiles and other small necessities for roofing up to the amount of 20s. And they say that they need up to 60s for shingles for repairing the roofing of the Great Chapel at the said castle, with the wages of the labourers. And they say that they need up to £7 for repairs to the Great Kitchen, the Bakery, the Brewhouse with the small chambers adjacent and three furnaces as well as the defects of the great prison of the said castle.

TNA, E143/10/2 1 Edward III, 1327 Inquest as to the state of all existing castles whether in the hands of the Lord King or the Lady Queen in the counties of Oxfordshire and Berkshire, taken in the presence of A[mbrose] de Newburgh at Wallingford on the Wednesday after the feast of St Matthew the Apostle in the first year of the reign of King Edward third after the Conquest, on the oath of Richard de Louthes, Nicholas of Turneston, Andrew of Sutton, Henry of Melburn, William Arnyet7, William Broum, John Stacy, William fitzPeter, Richard Mason, John Carpenter, Richard Gratard and Walter Crispe. The Castle of Wallingford The Outer Bailey: They say on their oath that the outer bailey of the said castle is wholly old and ruinous so that they need repair costs up to £160 on the walls

Total: £18 13s 4d Grand total £201 10s They say also that all defects in the said castle are not caused by the negligence of the recent constable William Marshal, etc.

  William Arnyat a vintner represented Wallingford in Parliament twice during the 1330s (Herbert thesis p. 111); Richard Gratard was an MP in 1319 (ibid., App. p. xviii).

7

69

From this point on the discussion of the various buildings and features of the castle will follow the order established by the 1327 Survey, working from the outside to the buildings of the inner bailey. The commentary builds towards a suggested layout of the buildings in the inner bailey – the heart of the castle – as they might have been in the mid-14th century.

the serjeanty of castle porter or janitor was probably in this outer bailey. Other amenities were a vineyard and orchards (Midgley 1942, 134). The 1308 accounts include 10d for a messenger sent several times to the queen with cherries. Although the castle was, unusually, extensive enough to enclose a number of gardens and kitchen gardens within its walls, it also had the resources of the several hundred acres of arable land lying in the attached hamlet of Clapcot on its northern circuit. There is no sign of any use of the area as a park during the occupation of the castle, but there is evidence for its use for grazing sheep and hay-making. A record of the tax of a ninth part of moveable goods in 1341 records 3 flocks of a total of 320 sheep in the area (Non. Inq., 9). Weaving and the woollen trade were still important in the town, which had by then begun a period of precipitous decline (Kightly 1993). At the time this third wall was built, in the mid-13th century, the towers would normally have been cylindrical or drum-shaped, the rectangular towers of the Norman period having been progressively abandoned during the 12th century. There are references to barbicans on the outer walls in other documents. Barbicans were often used to protect gateways or major towers, and could be diverse in form, ranging from major masonry structures flanking gatehouses to large earthworks doubling the outer gate by providing an obstacle against attack (Cathcart King 1988, 121). Most castles had a single main gateway, which would be a well-protected but grand formal entrance, as well as one or more discreetly placed and well-protected posterns, as was the case at Wallingford. As noted above, a charter of 1294 gives the boundary of a tenement as lying in St Peter’s parish and stretching from the high road (High St) to the outer barbican of the castle ditch (CMACC, 155). A set of two accounts from 1307 to 1309 presented by John de Clinton, seneschal of the honour, mentions, under 1307, the repointing of the outer wall of the castle on the south side between the two barbicans. Although we know nothing of the form of these two southern barbicans, there is reason to believe that one at least was an outer work controlling access to the main entrance, which lay beyond the outer moat. As was often the case, there was a palisade outside the outer gate, mentioned as repaired in the 1308 and 1331 accounts. That main outer gatehouse (Constable’s Gate) was probably substantial, attached to the curtain wall by two flanking towers constituting a barbican, (Figure 5.25). This area lay closest to the bridge entrance into the town, itself controlled by the East Gate on the end of the bridge. It was also the area through which the mill under the castle was accessed. The observation of the Gesta Stephani that the bridge over the river was the key to castle and town (1.5) points to the relationship of the town’s East Gate to the castle outer gatehouse. Tight control over the town gates by the lord of the castle – the king or his close kin – was clearly exercised, as the castle barbican controlling access via the North Gate, and the king’s mill on the South Gate show. The West Gate was equally important as it was there that the sluice-gate system that controlled the water in the castle moats was located (see next section). The East Gate’s intimate relationship with the castle is probably reflected

Baileys, bridges, gates and posterns This survey is of major importance for the understanding of a castle of which precious little survives. It establishes beyond doubt the existence of three wards, with three bridges, all of which will have crossed moats. The gatehouses attached to the bridges are not mentioned, but can be inferred, and are certainly evidenced in other accounts. It also coincides with geophysical survey which has confirmed the three ward layout, each with its own wall and moat (see Figures 5.1 and 7.1). We shall trace the buildings mentioned in other documents shortly, but it is also worth noting here that the surveys survive also for Wallingford’s nearest neighbours, Windsor Castle and Oxford Castle. The survey of Windsor is essentially that of a magnificent but ruinously expensive palace that efforts were continually made to keep up to scratch. It was published in full by Sir William St John Hope in 1913. The survey of Oxford Castle is different again (below, p. 210– 12). It bears out the idea that the castles of neighbouring counties such as Berkshire and Oxfordshire had contrasting roles, one a fortress and royal residence, wellmaintained, and the other a shrieval castle that was usually neglected and eventually allowed to decay (Pounds 1990, 96). They form quite a contrast. The survey of Oxford is about a third longer than the survey of Wallingford, and it tells the sorry tale of extensive problems requiring over £800 worth of repairs, an enormous sum of money for the time. Very little was done then or later and by 1388 the building was described as ruinous. The repairs suggested for Wallingford totalled a comparatively modest £201 10s. The Wallingford survey confirms the view of John Leland c. 1540 that ‘Al the goodly building with the tourres and dungeon be withyn the third dyke’, and Camden’s observation in 1586 that ‘in the middle stands a tower, rais’d upon a very high mount’ (Lloyd 133–4; 2.2–3). The heart of this castle was undoubtedly the inner bailey with its motte and Great Tower. The 1327 survey also confirms the existence of the third wall enclosing the outer bailey, a wall with towers and barbicans, and makes clear that there were no significant buildings in the outer bailey, which is borne out by other accounts. This does not mean that it was empty or unused. Areas such as this could form lists, that is, places where jousting could take place, as certainly happened at Wallingford, most notoriously in 1307 (above, p. 61, and below, p. 73–4). It may also have housed the pinfold a wooden enclosure, usually in the outer bailey under the charge of the porter, where stray animals were impounded until their owners claimed them (Pounds 1990, 203); other references occur in 1308–9 and 1483. The pasture within the precincts which was one of the perquisites pertaining to 70

respect of the inevitable need for maintenance of failing stone and timbers. The church itself remained in the inner bailey, and has long since completely disappeared. So have all traces of gates or gatehouses. This reference is the first real documentary clue to the whereabouts of any of the castle gates beyond the main entrance. Given the known location of the surviving St Nicholas complex, Bydongate must lie on the south-east side of the castle, and as such on a line with the main gatehouse entrance on the south-east corner (above p. 39), in close relation to the East Gate of the town, at the foot of the great bridge over the Thames. Another reference to the Bydongate occurs in the surviving Ministers’ Accounts relating to Wallingford from 1296–97 (Midgley 1942, 135). This source mentions the building of a sluice outside the Bydongate (in exclusa facienda extra portam Bydon), apparently made from wattle, the leftovers from the process being taken back for storage in an undercroft. A possible further reference to Bydongate is found in the damaged membrane for 1284– 85, which refers to work on the alure (either an internal or other covered passage, or a wall-walk) between the gate and the house of Lord Roger. This was the Dean’s Lodging (Dewey pp. 141–2 below), since the man in question was Roger de Drayton, the earl’s treasurer who had been rector of Harwell and dean of St Nicholas, and was addressed as such in Edmund’s charter. Whilst accompanying the earl to parliament in 1292, he was murdered in London by three brothers from Berkhamsted, said to have acted in revenge for his treatment of their mother, whom he had put in the stocks at Berkhamsted (Midgley 1942, xxxiii). Bydongate last occurs under that name (instead of just ‘the middle gate’) in the 1307–8 accounts when a roofer worked ultra Bydonsgate. The outer walls on the north side were also repointed in 1307 and timber obtained for a lintel (i.e. a wall-tie); we also learn that a man was paid to clear fallen stone to make access easier for the carters bringing new materials. The barbicans on the north, Clapcot, side are also mentioned in a document of 1367 for Boarstall Priory, which refers to ‘buildings in Clopcote between le Barbicanes of Wallingford Castle and the land of John Colles, one head abutting upon the high road towards the west’ (Boarstall, 244). Those on the north would have related to the North Gate of the town, the location of which was shifted when the castle was extended on that side for the creation of the third wall in the mid-13th century. Leland (2.2) put it, c. 1540, ‘The castelle yoinith to the north gate of the toune’, but such evidence as there is does not indicate that this northern entrance to town and castle, with large and doubtless strongly defended barbicans inside and outside the gate, was seen as one of the main gatehouse entrances. There will have been service routes through the baileys, perhaps even a minor service entrance somewhere along Castle Street towards the town centre, but if so there is no documentary evidence. From at least the mid-13th century onward the only major gate oriented west-east was the inner drawbridge and gatehouse leading into the inner bailey, accessible only from within the castle precincts.

Figure 5.25 Main gatehouse and barbican, with double portcullis, at Warwick Castle (photo: Judy Dewey).

in the view of a Tudor surveyor of 1552 who noted that: ‘There is within the said town 4 bridges called the South Gate Bridge & the West Gate Bridge & the North Gate Bridge, two of them being built of stone and one of timber’. The East Gate is unnamed because it was reckoned as part of the castle; a similar situation pertained at Warwick, of which John Leland (ii, 41) wrote: ‘The easte gate and the west yet remayne. The northe gate is downe. The strengthe of the bridge by the castle stondithe for the southe gate’. It is also clear that the second major gatehouse, leading into the middle bailey, was also on the south-east and hence aligned with the town’s East Gate. When Earl Edmund refounded the Great Chapel or College of St Nicholas as a chantry, he was very specific in his charter about the place where accommodation for his enlarged body of clergy was to be built. He prescribed that the ‘dean, chaplains and clerks serving in perpetuity in the chapel should be able to live and dwell within our castle, viz in that open space within the middle bailey that extends in length and breadth from the head of our great stable towards the east up to the gate [called] Bydongate’ (portam de Bydonesgate) (1.28). This is crucially important information for two reasons. In the first place, part of the complex built to house the clergy of the new chantry still survives in the former middle bailey of the castle and therefore provides a much needed anchor point for discussion of the castle layout from this time. Secondly, it is the last evidence we have of a major transformation of the castle layout, the moat, wall and bridge circuits having already achieved their final developed form, which henceforth would only change in 71

The northern barbicans may relate also to a postern called Dernegate, which was on the north-east side of the castle, somewhere near the water outlet known as Walditch running from the northern wall of the castle towards the Thames, at the southern end of King’s Mead. This was probably the postern that was first mentioned in a grant to Holy Trinity Priory c. 1180 by Alan fitzAnsfrid of his father’s meadow land in Clapcot, before the postern on the east side of the castle (pratum meum quod fuit patris mei quod est in Clopcote quod jacet in orientali parte castelli ante posternam ejusdem castelli: Mon. Ang. iii, 280a). Perhaps on account of the sort of storm damage that caused breaches in the walls towards Clapcot (above, 58), the 1307–9 building accounts refer to the repair, over nine days, of a wall contingent with Dernegate that was falling down towards the Thames. There follows a reference unique in the building accounts to the transport of stone for this site by water, though we know from other sources that the river was used for transport of materials and personnel. The Dernegate postern occurs again in the 1389–90 account which mentioned cleaning of the foundations in a section of wall from the Dernegate to the interior wall. It appears in the 1284–85 accounts as the postern towards Clapcote (posternam uersus Clopcote); at the same date a carpenter was employed to make a hanging lamp for the chancel [of St Nicholas] and a bridge towards Clapcote, the latter presumably associated with the Dernegate postern. The name of this gate, which may derive from Middle English meaning ‘secluded, solitary (place); secret (passage, etc.)’, is often found in association with castle posterns or minor gates in town walls, as for example at Berkhamsted Castle, Winchester, Coventry, Northampton (Derngate, now a shopping mall), though at Clare Castle it was one of three imposing gates, Nethergate, Redgate and Derneygate. A study by Richard Coates has shown that most of them had north-east orientations, were always independent of gates related to compass points, and all show a concern for privacy (Coates 2010). Postern gates in urban castles usually accessed extra-mural areas away from the gaze of the town, their chief protection being narrowness and difficulty of access (Pounds 1990, 214). It may be wondered if this was a convenient entry point for the lord of the castle himself, since it is clear that the Black Prince was accustomed to arrive by a ferry barge taken at Shillingford, as did Edward, Prince of Wales, in 1482 (1.70). In 1389 Richard Bargeman of Shillingford was paid 3s to bring the prior of St Frideswide’s and fellow overseer of the castle works by barge to the castle. There was at least one other major postern, twice referred to towards the end of the record for 1308–9, which accounted for a hinge and a key ad primam posternam. This may have related to the entrance to the Great Tower area in the inner bailey. Given the numerous moats and water features in the vicinity, such a postern could probably only be accessed from within the castle, having first crossed the outer and middle bridges. A postern near the motte would have been useful and is found elsewhere; Nottingham Castle, for example, was provided with a postern giving access to the motte (PR 6 Richard I: 80). It may well have been the underground passage, or sallyport, described by

Hedges in 1881 discussed above, pp. 46–7. Alternatively, this first postern may have related more directly to the long gatehouse entry leading from the outer gate to the middle gate. There are several references to various posterns (including the plural, posternas) in the 1284–85 earldom of Cornwall accounts, some of which probably describe little more than small doors into the several gardens in the inner bailey. Men were employed carrying material to the wall/s and postern of the herber behind the hall and to the postern at the long trellis and the wall at the same place for repairs (cariag’ ad mur’ et posternam herbarii retro aulam et ad posternam ad longam trayllam et murum ibidem emendandum). Moats and water The reference to the sluice at the Bydongate in 1296 is a reminder that in this castle by the late-13th century there were three wet moats, each of which was crossed by a drawbridge leading to a gateway. Castle moats were often dry and this arrangement makes Wallingford very unusual. All the moats were thickly lined with lime, which was kept in good repair; like all the ‘guttering’ around the castle, the moats were cleaned out thoroughly at least once a year, and the leads of the gutters or water channels repaired as necessary. Guttering or water channels are frequently mentioned in all accounts, referring both to roof gutters and to lower or ground level channels or drains. The word goter is always used and it is impossible to be sure how what was a sophisticated system of water management worked; only in the 1375 account is the word pipa used as well. Elaborate arrangements were made for the water to circulate round the moats, through a lock system devised outside the town’s West Gate (Figure 5.26). Here water from streams from North Moreton and the surrounding area was channelled into the town’s encircling ditch. The building account of 1307 mentions the making of a water enclosure between the ditches at the West Gate and a key made for the same, the enclosure of akenate, and cleaning of the first channel or gutter where the water entered the castle. The accounts from 1389–90 describe the building of the floodgates belonging to the castle, built in the ditch outside the West Gate of the town, on its south side. This was a renewal of a lock-operated water engineering system which ensured that water filled, and could be made to flow round, the castle moats. It was taken round the town ditch section known as Black Ditch, on the north end of the Bullcroft, as confirmed in a Minister’s and Receiver’s Account of 1481–3 which accounted for the scouring of the Black Ditch neighbouring the priory (pro escur’ cuiusdam fossate nigre iuxta prioratum), which the same system is also likely to have served; the same ditch was referred to in 1425 as ‘the water of the ditch called Constable’s Ditch’ (Herbert 1971, 7–8). The priory rental of 1552 (Oxford, Bodleian, Top. Berks. b. 41, fol. 24v), repeated by the later antiquarian Richard Skirmer, adds a note on the floodgate system at Wallingford describing ‘a certain sluice or lock, built of timber, to turn the water into the castle ditches every Saturday at noon until even-song time the Sunday following; the which bridges and sluice or 72

Figure 5.26 Extract from a map of 1835 showing the still surviving ‘floodgate’ system at the West Gate at the top of modern Croft Road, then still named Lock Lane (image courtesy of Wallingford Museum).

lock were to be repaired from time to time at the only costs and charges of the king and queen’s majesty’. Repairs to sluices and water channels are frequently mentioned in the later building accounts (see below). A barbican at the West Gate is mentioned in 1314 (CMACC, 149). The moats provided an essential water supply to castle services – supplemented by rainwater cisterns and drinking water taken from wells – sanitation (removal of latrine waste), and allowed fish such as pike, perch, tench and roach to live in the moats, as well as in the fish pool associated with the mill outside the castle on the Queen’s Arbour site (Midgley 1942, 113). The fish were eaten in the castle, sometimes sent to the king at other places such as Windsor, or sold. In 1363 the sale of eels made 14s 8d, two tench 3s 6d, three pikes 7s 6d, a pike captured in the outer ditch of the castle fetched 10s, six pickerels 5s 5d, and two pickerels, two roach and two small roaches made 5s 8d (TNA, E101/545/27). There were sluices controlling the moat system on the castle site itself, as appears from references in the 1284–85 and 1307 accounts. There was certainly one at the Bydongate, on the south-east, which, whilst it could derive from a surname, was possibly related to the sanitary arrangements since bidōn is a Middle English verb meaning to befoul (a place); defile (oneself); subject (sb.) to vile odour (the road outside the manor of Gainsborough, Lincolnshire, was called Shitgate). Another sluice was probably located on the north east, at Walditch. The beauty and sophistication of these landscaped water defences were not an unmixed blessing: at least one carter, bringing a cart laden with the earl of Cornwall’s kitchen utensils, is known to have drowned when he fell into the castle moat (Herbert 1971, 62). A charter of the Black Prince, lord of Wallingford from 1337 until his death in 1375, establishes the boundaries of his authority as lying between Walditch in the north down

to the abbot of Reading’s in the south (REBP iv, 86). This last was Winterbrook, now called Bradford’s Brook, which marks the boundary between Wallingford and Cholsey. (Henry I had given the church and other benefits in Cholsey to Reading, RAC ii, 72.) A further rather obscure reference to castle ditches occurs in the 1389 account concerning rope for the bridge (le Brygg) at port’ eyt, probably referring to Porter’s Eyte or Eyot or even Acre. The porter had one acre of arable land in Clapcot called Porter’s Acre in 1622, and one and a half acres of meadowland called Porter’s Ditch in King’s Mead. It occurs as two greens next the Thames in a charter of Edmund, earl of Cornwall, granting it to Thomas de la Russhe. The various accounts of the porter’s fee suggest that the ditch ran north-south between Clapcot field and King’s Mead, and that the Walditch was the ditch that is still visible running eastwards from the northern ramparts into the river (see Dewey below 170). Gaweynes Herber/Queen’s Arbour and the riverside: Wallingford in romance It also appears to be the case that the outer bailey had appended to it Gaweynes herber, which we can take to mean the riverside area that has been known for some time as Queen’s Arbour, occurring in a survey of the town conducted in 1548 as Queneherber. The name form is difficult to see as a corruption of Middle English quēn(e), and indeed there is strong reason to believe that its name really was Gaweynes Herber in 1327. Edward III himself was obsessed with the Arthur legend: he is recorded in an armorial roll of 1334 as having fought at a tournament bearing the arms of the Arthurian knight Sir Gawain (Goodall 2011, 283). On 1 January 1332, after a tour of Arthurian sites at Cadbury and Glastonbury in December 1331, Queen Philippa gave the king a silver enamel bowl

73

engraved on the outside with castles and ships and divers beasts, and inside a grand castle, pennants flying, with the king in the middle, with a banner showing the three lions passants guardants of the English royal coat of arms. A ewer which accompanied the bowl was adorned with images of Julius Caesar, Judas Maccabeus, Kings Charlemagne and Arthur, Roland, Oliver, Gawain and Lancelot of the Lake (Ormrod 2011, 98–9). The name, of course, is that of Sir Gawain, the heroic nephew of King Arthur, as he figured in the extremely popular and influential mid-12th-century work of Geoffrey of Monmouth, as well as in a number of romances written in Anglo-Norman or Middle English that mention Wallingford Castle. Geoffrey was a canon of St George’s in Oxford, then part of Oseney Abbey, and (falsely) claimed to have derived his information from a book owned by his colleague Walter the archdeacon, who briefly held land in Clapcot in the 1150s (Howlett 1995, 65–6). Two early successes of Chrétien de Troyes, Erec et Enide and Cliges, both feature King Arthur’s spotless knight and nephew Sir Gawain, whose sister Soredamour was mother of Cliges. Of the two, Cliges is particularly interesting because its details about the Thames Valley from London to Oxford, including a siege of Windsor Castle and a tournament at Wallingford, indicate that Chrétien had visited these places in person (Bullock-Davies 1981). His description of Windsor Castle is authentic in its detail and he could have seen repair work there between 1168 and 1174. It was besieged briefly in 1193 when Prince John held it against the regents of his brother Richard I. Chrétien also describes the bed of the Thames as being dry, a phenomenon wellattested in the 12th century between 1102 and 1149, and again from 1157. Finally, the ‘Wallingford tournament’, which required a large expanse of flat terrain, was held in the countryside near the town and ‘below Oxford’. Constance Bullock-Davies’ discussion of this topographical knowledge points to a writ of Richard I of 1194 permitting five tournaments, one of them to be held ‘between Stanford and Wallingford’, referring to Stanfordin-the-Vale (Bullock-Davies 1981, 19–32). Little is known about Chrétien, but it seems clear that he had some sort of connection with the court of Henry II and his wife Eleanor of Aquitaine and their children. Joseph Duggan has suggested that any overlap between Chrétien’s work and that of Geoffrey of Monmouth was due to both having used Celtic sources independently (Duggan 2001, 196–206). The fact that Erec et Enide includes a coronation scene set at Christmas in Nantes in Brittany ‘unmotivated by anything in the rest of the plot’, suggests a link between the poem’s origin and the Christmas court that Henry II summoned at Nantes in 1169. The purpose of the court was to require the Breton barons to render homage to his third son Geoffrey, then aged 12, who had been betrothed to Constance, heiress of Brittany, in 1166. They married in 1181, and their son, born posthumously in 1187, was named Arthur. For Duggan (2001, 96), Constance (d. 1201), perhaps together with Geoffrey, could have been the patron behind the work, though the Young King is also a candidate for both works (ibid., 64). In one couplet of Erec a character called Brianz

Figure 5.27 The sword of Guy of Warwick at Warwick Castle (photo: Judy Dewey).

des Isles [i.e. Britain] presents Arthur and Guinevere with two ivory faldstools carved with images of crocodiles and leopards, the latter being the heraldic animal of the kings of England (ibid., 68–9). However, Geoffrey was killed in a tournament in 1186 in Paris and Cliges probably refers to real events at Windsor and Wallingford in 1193 and 1194. The first mention of Wallingford in an AngloNorman prose chronicle occurs in the introduction to Wace’s Roman du Rou, and describes Matilda’s escape from Oxford to the safety of Brien fitzCount’s Wallingford in 1142. The notion that family patronage or ‘ancestral romance’ underlies some of these literary texts is generally unconvincing (Dannenbaum 1982) and is often based on very unsound history, but there does seem to be good reason to associate the inspiring career of Brien (great-great-uncle of Duchess Constance) and his defence of Wallingford with the otherwise improbable role that Wallingford played in some of these texts (Weston 1925). One should recall that Geoffrey of Monmouth was a one-time canon of St George’s College before it became the abbey of Oseney in 1142 (Barron 2002), and that the earliest written copy of the Chanson de Roland was copied at Oxford. Oxford 74

who had escorted Gaveston to his death in 1312 and himself a distinguished soldier under Edward III, named his sons Guy, Thomas and Reynbrun. His will left ‘the coat of mail sometime belonging to that famous Guy of Warwick’ to his son Thomas, whose own will of 1400 left to his heir Richard an item ‘wrought with the arms and story of Guy of Warwick, and the sword and coat of mail, which was that worthy Knight’s’ (TV I, 79, 154).8 To this day a 5ft (1.5m) long sword is kept at Warwick Castle as that of Guy of Warwick (Figure 5.27). The idea that the career of Brien fitzCount may have been the inspiration for this hero also is not improbable, but to see a connection with Wigod, suggested as the basis of the name of Guy – Wido or Guido – and the d’Oillys seems farfetched (Mason 1984). There was no blood relationship between the d’Oillys and the lords of Wallingford after c. 1094, and the heirs of Brien fitzCount were the Plantagenets. All the more interesting then, is the occurrence of some jottings in a manuscript compilation of the 14th century, made for Coventry Priory (Oxford, Bod. F. Auct. 5), in which someone noted that the sword of Sir Gawain was on display in the Great Tower of Wallingford Castle and was engraved with a four-line octosyllabic verse (Figure 5.28).9 In a very serious audit of the munitions at Wallingford ordered when Edward III succeeded his father in January 1327, discussed in detail below, mention is made of three swords, those of (?) King Richard, Gawain and Guy of Warwick. Traditions about both Guy of Warwick and Sir Gawain, whose sword Clarent was originally his uncle Arthur’s sword of peace, appear to have started by this date at Wallingford. The 14th-century Middle English Alliterative Morte Arthure (v, 4202–3) is the only text to mention this sword, which Arthur kept in his wardrobe at Wallingford, along with locked coffers of regalia that belonged to the crown, in the keeping of Guinevere, now revealed as a traitor to her husband (Armstrong 2008). Food for thought, then, for commentators on a poem believed to date no earlier than 1365 and probably around 1400 (Finlayson 1967; Toohey 2000). The name read unsatisfactorily under UVL as Regis Ricardi in the inventory of arms does make some sense. Goodall (2011, 189) has pointed to the ephemeral castle at Tintagel built by Richard of Cornwall, king of Germany, which he characterizes as a ‘chivalric folly, constructed on the supposed site of King Arthur’s conception’. Richard has been suggested as the model for a character in the AngloNorman translation of the Letter of Prester John (Legge 1963, 201–3), and, rather less flatteringly, as the villain Earl Godrich of Cornwall in Havelock (Weiss 2012). Nor should we overlook King Richard I the Lionheart (1189–

Figure 5.28 14th century verse saying that the sword of Sir Gaweyn is displayed in the Great Tower at Wallingford Castle. (Reproduced by permission of The Bodleian Library, University of Oxford, Bod. MS Auct. F. 5. 23, fol. 110v.)

had both a royal palace and a shrieval castle, and was home to intellectuals in the 12th century, culminating in the foundation of the university in the early 13th century. It was close to other residences at Woodstock, Windsor, and Reading, all favoured by royals young and old alike. The texts mentioned so far were written before 1200. An even more striking appearance for Wallingford comes in a text of the 13th century which was translated into Middle English in the 14th century, and was probably the most successful of all, the Romance of Guy of Warwick. Guy’s home was Wallingford; his son Reinbrun was raised there also. His sole link with Warwick was as the socially inferior admirer of the earl’s daughter, Felice, whom he eventually married. Having endured numerous adventures to prove his prowess and worthiness against Arthur’s knights among others, he eventually becomes a hermit and ends his life a saint as well as a hero. This romance is often described as an ancestral romance associated with the Beauchamp earls of Warwick, one of whom had married the heiress of Henry II d’Oilly (d. 1232). Difficult to credit as an origin, there is no doubt that the later Beauchamp earls of Warwick identified strongly with it (Mason 1984; Liu 2005). Thomas de Beauchamp, son of the Earl Guy

8   Earl Guy de Beauchamp of Warwick’s Will (TV i, 53–4), left to his son Thomas ‘my best coat of mail, helmet, and suit of harness, with all that belonging thereto; to John, my son, my second coat of mail, helmet and harness; and I will that all the rest of my armour, bows, and other warlike implements, shall remain in Warwick Castle for my heir.’. . 9   The text appears to read: En la grante tour de Walingford est l’espee mounsire Gawayn./Enquei [= still today ?] cestes es paroles sunt escriptes/que hic sequntur: Sage felon deit home duter/E fol felon eschiver, /Fol debonere deporter/ E sage debonere amer.

75

99), hero of the romance Rycharde cuer du lyon of c. 1300 (Weber 1810, ii, 1–128). Whatever the origin of the name Gaweynes Herber, the area in question was a flood-prone meadow, lying between the Thames and the moat at the foot of the inner bailey curtain wall on the east side of the castle. It possibly appears under its more familiar name in the 1331–4 accounts, when one man worked for six days on the bridge and the queen’s herber (pontis et herbari Regine), though it is more likely that the latter was one of several gardens in the inner bailey; the occurrence in the 1284 accounts of the pontem comitisse uersus fontem comitisse may hint at a pleasure garden in the bailey, perhaps the ‘herber behind the hall’ of the same account. The Queen’s Arbour area was probably regarded as an integral part of the castle defences, although external to the walls and lying within the parish of All Saints, unlike the area within the walls all of which was extra parochial, since it lay between the eastern moat and the river. It contained a cornmill, often referred to as the ‘mill under the castle’ or the East Mill, with its granaries, ponds and mill-leats, fish pond (stuwe) and probably also the swannery referred to in the 1284 and 1296–7 accounts (Midgley 1942, 134–5). Such layouts occur at other castles, including Kenilworth, Hereford and Gloucester (on gardens and water landscapes see Creighton 2002, 73-84). The water-powered mills on the east side of the castle, by the Thames, and two at the South Gate, are frequently mentioned in the accounts, as their relatively flimsy wooden superstructures were constantly in need of repair – a running theme at many castles (Pounds 1990, 193–4). They were described as decrepit in 1359 and 1391. Later they were often let at farm, the lessee being allowed to make what profit he could, provided he kept them in good repair. They are not mentioned in the 1327 survey, which restricts itself to stone or other substantial buildings. In 1307–8 the mills had a comprehensive refit, including the rafters of the South Mill and two granaries at the head of the East Mill. A plumber and his mate were set to soldering and filtrating the water channels of the South Mill. The rupture at the stank of the East Mill was repaired, and the stank between the South Mill and the West Gate was cleaned. Two hundred beeches from Wyfold were used to repair the great and upper enclosures, which also required 55 cartloads of stone in one season and 55 the next. Work was also done on an enclosure of water between the ditches at the West Gate [Figure 5.26 above], with reference also to the enclos/ure/ing of akenate, and the cleaning of the channels where water entered the castle. The word akenate is obscure, but is perhaps reflected in the grant in 1314 for life to Ralph Restwold of meadow and weirs at Sakenate and order to put a watermill there. This was a small islet at Clapcot on the east side of the castle, opposite Benson Lock and was afterwards known as the North Mill at Northeyte. The great and small enclosures were repaired in 1483, and the outer stank next Clapcot (North Mill) was dredged, by order of the council of the child Prince Edward of Wales; at the same date a new barge was ordered for the ferry at Shillingford to bring the prince to the castle (TNA, SC6/1302/1).

Several accounts refer to a ‘herber’, which was a herb or a pleasure garden, or garden (gardin), even to a rose garden (rosarium) (Midgley 1942, 132). There was also a constable’s garden and a garden round the chapel in the extensive inner bailey, as we learn from the 1296–7 (ibid., 134), 1308 and 1389 accounts. In 1331, when the castle was revamped for a visit by the queen, Philippa of Hainault, and the earl of Chester (formally invested as earl on 18 March 1333), her infant son, born 15 June 1330, the future Edward the Black Prince, an enclosed herber was made for the little earl, with reference to turves cut for the same. This might indicate the making of a turf seat, as was fashionable in gardens at this time, or simply the laying of a lawn. There are also references in 1284, 1308 and 1331–4, to La Trailye, which might refer to a private garden of the type sometimes referred to as a ‘bower’. The name relates to the word for trellis and may indicate trailing plants, such as in the herb garden at Gloucester Castle (Pounds 1990, 198). In 1284 repairs were made at the postern of the long trellis and the wall there (posternam ad longam trayllam et murum ibidem). The 1308 account refers to carpentry on the entrance to the outer bower (introitum traile exterioris), and the 1331–4 account to mending the lock of the door (hostium de La Trayle) of the bower. However, all these references are probably linked to the cultivation of vines in the castle’s well-evidenced vineyard (Midgley 1942, 132, 134), described in some detail in 1296. At the time of the death of Joan of Kent, widow of the Black Prince, in 1385, ‘the garden called the Vineyard’ was part of the estate she had held in dower and remained linked with her for some time; in 1441 it was specified as part of the farm of the banks of the middle and outer ditches, which continued to yield revenues in fish caught long after the vines trained upon the slopes had ceased to be productive (TNA, SC6/1090/19) (Figure 5.29). There is unlikely to have been a garden on the water meadow now called Queen’s Arbour, since it was a busy area and anyway prone to flooding. The first formal use of the name Quenesherber comes in 1376, when it was specified as part of the dower of Joan of Kent, Princess of Wales. Possibly the queen’s private apartments at this time partly occupied one of the towers, perhaps the most southerly of two towers on the east side of the castle, overlooking Queen’s Arbour, below the surviving stone fragment of the curtain wall known as Queen’s Tower today (Figure 5.30). Such a location would have afforded views over the water features on the Queen’s Arbour and the wider landscape of the Thames. This busy area was also part of a deliberately designed water landscape. The windows installed in the curtain wall on that side in 1389 also worked to that end. The queen’s chamber may have contained the chamber over the water (camera super aquam) where partitions were made for a garderobe for Queen Philippa’s young son in 1331–4 (discussed under Red Hall, below). This could have been either of the wall towers projecting into the moat on the east side of the castle, but it is most likely to have been the south-east tower which overlooked the riverside moat and the water pooled in the sluice system around the Bydongate (Dewey, below, pp. 169–72). At any rate it is possible that the location of 76

Figure 5.29 Graphite sketch by George Cumberland (1799–1848), formerly in an album, showing the Priests’ Lodgings above the wall of the middle bailey at Wallingford (see Figure 5.2 above). It shows that more of the building and the curtain wall was still standing at that date, and confirms that the heightening of the tower on the NE was an alteration by a later owner, probably J. K. Hedges who housed his museum there. Note the signs of trellis work on the bank below the wall; this was where the vines of the medieval castle vineyard were trained. (Image © The Trustees of the British Museum.)

what emerges as the queen’s lodgings in the 14th century may have given rise to the name of the adjoining meadow, replacing an earlier but similar-sounding name perhaps as early as 1334. Unfortunately, the occurrence in the 1331–4 accounts is ambiguous. Although these accounts show the nervousness of the new regime, the reference to six days work on emandacione pontis & herbarij regine comes in an account of small expenses after the section dealing with defences and thus is probably a reference to a bailey garden, rather than concern with security on the castle’s eastern flank. The importance of Gaweynes or Queen’s Arbour lay primarily in its economic functions, including the porterage of goods and personnel by water, as well as very valuable meadowland providing winter feed for animals. The principal source of hay for the castle’s needs was 40 acres of meadow in Clapcot known as King’s Mead. The whole area between the East Gate mills and the Shillingford ferry was part of the honour of the castle and was exploited for what was later called the ‘gainery’ [i.e. farm] of the castle. As appears in the surviving Ministers Accounts from 1308 to 1526, this consisted of a customary due called ‘cartingsilver’, a sum of 6s paid by the free tenants of Clapcot in lieu of a service of carting hay, the revenues of the mill and fishery at Northeyte, and the fishery at the lock which operated the East Gate mill, the latter associated with 4 acres of meadow on Queen’s Arbour. In addition, there was the fishing in the castle ditches of the middle and outer baileys, the vineyard, and the produce of castle orchards, cherries, apples and pears (TNA, SC6/1090/19). Farmed separately were the two mills at the South Gate, the ferry at Shillingford and the St Mary Magdalen day fair at Crowmarsh.

Figure 5.30 The stonework surviving in the NE corner of the inner bailey at Wallingford Castle, known as the Queen’s Tower (photo: Oliver Creighton).

77

Great Stable Hay was stored in the hay barn (domus feni), mentioned in 1331, which would have been located in the middle bailey, near the Great Stable. In the time of Richard and Edmund of Cornwall, this was known as the long stable. In all periods, its location next to the prison sometime called the Blindhouse is frequently noted. Enormous amounts of hay, and other fodder, much of it supplied from the castle meadows, were required for the king’s horses. These included cart-horses (fed with hay from 40 acres of castle meadow), horses used for breeding (special stalls for foals are mentioned in 1296, when a louvre was installed in the long stable: Midgley 1942, 133), riding horses for the king’s servants, as well as the king’s bay destrier (1307–8). Looking after the bay destrier for 97 days cost 8s 1d in 1307; a local marshal hired for its care earned 2s. The total spent on the care of this animal was 52s 4d. In the same year the expenditure on four carters, their four assistants and 20 cart horses was 58s. In 1308 three men earned 9d between them over three days for turning over hay lest it moulder. In that year £51 19s 8d was spent on the king’s carts, carters, cart horses and other horses, including three infirm horses for whom separate supplies of bran and vinegar were provided. Of this, £11 3s 6d was spent on oats for 20 horses. The figure also included care of the horse harnessings and their shoes, plus the expenses of men on castle business using the horses. The final total included 22s charged for the provision of horses for Gilbert de Mandeville and his servant who presented his view of the accounts to the Exchequer in London. In February 1316, at a time when a nervous Edward II was spending quite a bit on Wallingford Castle, he ordered constable Edmund Bacon to expend up to £20 on the repair of the Great Stable, and a few months later he allotted another £20 to the same end. The Great Stable was re-roofed and its foundations repaired for a total of £4 16s in 1331. A total of 16s 1d in materials and manpower was spent on cleaning of the ditch around the Great Stable and repairs in 1361. In 1389–90 there was extensive renewal of wattle and daubing in the castle, which included the partitioning of a chamber within the Great Stable to make stalls for the king’s horses; the roof was repaired and new staples made for the door. These considerable sums emphasize the importance of the horse to a military nobility and to the economy in general.

Figure 5.31 Gate of Chepstow Castle, showing wicket and lattice work (photo: Andy Dingley 2011, Creative Commons Attribution –Share Alike 3.0 Unported Licence).

repaired with fresh timber, some of its planks being reused for repairs to the middle and outer bridges. The gatehouses are less often mentioned. We can piece together some idea about the appearance of the main gatehouse, which a deed of 1294 confirms was on the south-east, between two barbicans (above, p. 70). The 1296–7 accounts mention a latch provided for the small wicket gate set in the outer gate of the castle (Midgley 1942, 133). The wicket was a small door that was commonly inserted into one of the two large doors that formed the gate so that people could easily move in and out without the need for both larger doors to be fully opened. Such great castle doors were often made from oak and on their outer face were usually sheathed with wrought iron cladding and iron straps; the inner face of the door was decoratively strengthened by the use of sophisticated lattice work (Burton 2010–11, 248). Decoration of this kind was referred to in the first main section of the 1296–97 account where the master carpenter made a lattice on the inner face of the door at the constable’s entrance (facienti unum laticium infra portam ad introitum constabularii). Taken together with the reference to the wicket gate of the outer gate, this is the earliest reference in any of the available texts to the door furniture of the outer castle gate ( Figure 5.31). The next reference comes in 1331, when the timber portcullis that stood in front of the double doors, with its iron fittings, was renewed, as was the wicket gate.

The castle bridges and gatehouses Attached to the 1327 account of the outer bailey is the statement that the three castle bridges, outer, middle and inner, all needed repair. We can understand these as the bridges that led into each of the three baileys, crossing a moat each time. At the end of each bridge there was a gatehouse. Repairs to the boards of the bridges are frequently mentioned in the accounts, wood being a relatively perishable material. Damage to the middle bridge caused by the cart of the Countess of Gloucester, Edward I’s widowed daughter, had to be repaired in 1296 (Midgley 1942, 133). In 1389 the inner bridge was 78

Figure 5.32 Constable’s Gate at Dover Castle (photo: Judy Dewey).

The documents that have been used here very rarely give proper names to features mentioned and very rarely use the word ‘entrance’ (introitum). It seems reasonable to infer that this principal outer gatehouse was called Constable’s Gate, like the main gatehouse at Dover (Figure 5.32). At Dover and elsewhere the constable’s principal lodging was in or near the main gatehouse, facilitating his primary duty of controlling access to the castle (Kenyon 2005, 135). There was a major overhaul of the outer bridge in 1307, when three planks were replaced, two new winches of wood made, and iron for spykenails purchased. Tallow was also purchased for oiling the bridge axles, and for preserving the ropes from winter rain. The following year saw carpentry repairs on all three bridges, with a new palisade being built outside the outer gate – possibly a porcupine formation of pointed stakes driven into the moat. The account of 1331 starts with work on the portcullis, requiring 24 timber piles, iron spikes and nails for the same, an iron chain for the same, and the windlass. Hempen rope for each of the drawbridges (pontibus tractandis), was renewed. The 1307 accounts open with the repointing and refurbishment of the outer walls on the south, between the two barbicans. We can conclude, then, that a palisaded moat, two barbicans (perhaps the towers of a twin-towered gatehouse) and a drawbridge and portcullis defended the main entrance to the castle on the south-east. The probable site of the gate can be shown in relation to the East Gate (on the bridge) of the town, and lands held by

the castle porter (Dewey, below p. 150). The strengthening of gatehouses with portcullis and drawbridge was one of the innovations of the 150 years after the accession of Henry II in 1154, along with crenellations and over-sailing hoardings or brattices, all of them found at Wallingford (Pounds 1990, 102). The windlasses of the bridges are first mentioned in 1296 (Midgley 1942, 133) (Figure 5.33). The accounts of 1331–4 are very much about defence. Hoardings and a palisade or garillum were erected around the outer walls, and a watch-tower was built at the middle gate (Figure 5.34). We learn in 1307 of the janitor (porter)’s lodgings above the middle gate (camere janitoris ultra mediam portam); this may have been the ‘chamber above the bridge’ (camera super pontem) mentioned in the 1327 survey’s account of the middle bailey. In 1363 carpentry work was carried out on the bridge outside the castle (i.e. crossing the moat before the outer gatehouse) (see Figure 5.35). At that date a certain John Tyler was paid 40s to roof the ‘house’, later in the same account called ‘the great chamber’, known as the Wardrobe; he was paid the same sum for work on what we can infer was ‘another house’ (a plural cannot be ruled out), this one within (infra) the castle (inner bailey) against or above (super) the third bridge. It is unclear whether this work might have related either to the gatehouse there or the flanking tower or bastion indicated at the drawbridge on the plan drawn in the 1540s, or simply to a nearby building. The gatehouses were clearly substantial and well-defended (Figure 5.35). 79

Figure 5.33 Drawbridge over water with stone-built bridge, castle at Sirmione, Lake Garda, Italy (photo: Judy Dewey).

Figure 5.34 Wooden hoardings at Caerphilly Castle, viewed from inside and outside the castle wall (photos: Oliver Creighton).

The accounts relating to the 1389–90 works carried out in anticipation of a visit by Richard II are particularly rich, but particularly challenging. Unusually, the abbreviation is relatively light and the accounts are mostly easily legible, but punctuation is minimal. The problem here is that the richness of the detail is expressed with extreme brevity. Some important information can nevertheless be extracted. At that date a section of wall 108ft (33m) long at the outer

gate (main entrance) was rebuilt to a height of 8ft (2.4m). The relatively low height for this third wall is typical of concentric castle architecture. The gatetower in the south corner (apparently the same outer gate) was repaired or renewed, with six new windows; an adjacent section of wall 118ft (36m) long was also repaired. The repaired wall at the outer gate acquired a new window. At the same date modifications to the inner bailey included a wall 80

Figure 5.35 Artist’s reconstruction of possible gatehouse arrangement on the south-east side of Wallingford Castle c. 1300. Drawing by June Strong.

Figure 5.36 Great staircase at Wardour Castle, Wilts (photo: Oliver Creighton).

repair at the point where the wall carrying the inner end of the drawbridge abutted the inner gatehouse, specifically from the water in the moat to the summit of the wall. The gatehouse was to have a new window dividing its façade into two parts. The account also mentions a chamber of the inner gatehouse, the name of which chamber appears to have been le Enese. It also mentions two carpenters hanging a bell in the tower above the inner gate, these being the only clues as to the configuration of the inner

gatehouse in any of the accounts. At the same time a decayed wall carrying the outer end of the drawbridge was repaired, as was the way or road (camino) next to it (either the road leading from the middle bridge, through the Bydongate, to the inner gatehouse, or one coming from the inner barbican which protected an entrance route via the North Gate). This appears to refer to the stone footings of the drawbridge. Bridges were usually made of timber (Rigold 1975), but some major drawbridges were partly

81

supported by masonry piers, as appears to be the case here. Interestingly, signs of this masonry are visible in the geophysical survey of the castle (see Figure 5.1). The same accounts are the only ones to mention the almonry, a feature common in major castles and perhaps related to the treasure house mentioned in earlier accounts. Figure 5.35. Bridges were damaged by frequent use, or overloaded carts, as in 1296, and wood was anyway a perishable material. In 1331 major work on the inner bridge was required, and 52 timbers were purchased for making a watch tower at the middle gate and repair of the middle bridge. Iron work on the inner gate was renewed and on the wicket gate and other gates. In 1389–90 there were major repairs to the inner bridge called, with stunning originality, the Drawbridge (pontem vocatur le Drawebrugge) [a possible indication of the name of the inner gatehouse?]. A wooden stair or staircase was repaired in the gatehouse there, and work was done at the same date on or near another wooden staircase, which extended for 24ft (7.3m) from the King’s (or Great) Chamber in the direction of Queen’s Arbour. This might be the Great Stair mentioned in 1408, a common feature of major castles, usually leading from the Great Hall to the Great Chamber (Goodall 2011, 36–7) (Figure 5.36). The site of the Drawbridge is known from the Surveyor’s Plan of the inner bailey from the 1540s (Lloyd, below), so this information ought to be of some assistance in placing some of the buildings in the bailey at this date. Also in 1389–90 carpenters were employed on an instrument for casting rowemortar (rendering) on the Great Tower and other walls, making a gutter next the Norise Chamber and renewing and affixing a wallplate over a wall of a chamber next the buttery.

prison was fairly common and this may be the case here (Pounds 1990, 99–100). The only other reference to Benets Tower comes in 1308, when it received a new lock and key, immediately preceded in the account by repairs to the fetters and padlocks of the prisoners (incarceratis). Possibly these were the south-west corner tower and the small rectangular stone building shown on the plan of 1548 (Figure 6.1), lying in the shadow of the Great Tower. Both timber and stone were used in construction. Escapes from flimsily built common gaols were not infrequent. Ironically, in Wallingford the common gaol in the middle bailey was located next to the Great Stables, as we learn from the 1408 and 1411 (magni stabuli iuxta gaolam) accounts. There were at least these two prisons at any one time. There was another outside the castle, sometime called the king’s prison without the castle, or the prison or gaol of Wallingford, which the king ordered the sheriff to have repaired for no more than 100s in 1254, and again in 1265, the king himself frequently supplying the timber for such repairs. Sheriff Henry de Scaccario accounted £20 for a gaol (re-)built in villa de Walengeford in 1229. This presumably was the common prison, which was the subject of a parliamentary inquiry in 1315 because the king had removed it to Windsor (above, 51), where it was complained that facilities were inadequate and escapes frequent, so that the king was asked to restore the gaol at Wallingford. Continuing references to prisons imply that the common gaol at Wallingford was indeed reinstated, its other gaols having remained in use. Berkshire was unusually well supplied with holding prisons, partly as a consequence of lacking a clearly identifiable county town (Pugh 1955, 5). In 1496 a jury of the county of Berkshire was summoned to inquire into the escape of prisoners from Wallingford Castle. This gaol housed prisoners awaiting hearings at the shire eyres, some sessions of which were heard at Wallingford; felons from all over the honour could expect imprisonment and trial at the castle. A gaol delivery at Wallingford Castle is recorded in 1540 (LPFD 15, 831), but the honorial courts had ceased by c.1500 (Hedges 1881, ii, 83, 90). Pleas relating to the borough were usually heard at the Motehall, which was located near one of the two Clapcot manors, at Rush Court, some distance from both town and castle, although certain exceptional cases were heard in the guildhall (Herbert 1971, 67–8). The borough had its own prison, rebuilt in the time of Edward II. It is clear from the various accounts (1308, 1331, 1359, when it was rebuilt, 1363 and 1389) that the Motehall was the same as the hall of pleas (aula placitorum) outside the castle gate, which occurs in a number of accounts. That the lord of the castle took responsibility for the borough courthouse shows the firm hand that the royal masters of Wallingford kept on the reins at what was otherwise a widely enfranchised borough following the generous charter granted in 1155 by a grateful Henry II. Injunctions to build a new prison, as in 1359 and 1438, seem as likely as not to be de novo constructions, as in certainly true of the building of a prison house (domus prisone) and pit (puteus) in 1438–40, for the keeping of prisoners and the confinement of felons (una prisona

Prisons, Hall of Pleas In the middle bailey was the gaol or prison house (domus prisone), sometimes referred to as the Blindhouse, repaired in 1201, 1224, 1253–4, and 1363, which probably held (suspected) felons from the shire awaiting trial. The Great Prison, perhaps for honorial prisoners, was in the inner bailey, probably near or associated with the Great Tower. The Great Tower itself was used for the most important prisoners of state. In 1346 the Black Prince ordered his serjeant-at-arms, William de Welham, to join with Sir Edmund de Kendale in keeping good guard over the body of the Chamberlain de Tancarville, the prince’s prisoner, ‘not suffering him to go anywhere outside Wallingford Castle, and seeing to it that he sleeps every night henceforth in the tower of the castle, and he with him. The prince has ordered the constable to prepare the chambers and cause keys and other necessary fastenings to be made’ (REBP i, 33). Valuable prisoners for whom ransoms might be sought were gently treated overall. For the rest, a constantly renewed supply of fetters, chains, bars and locks awaited, necessary because many goals holding criminals awaiting trial were relatively flimsy wooden constructions (Pugh 1955, 14). The walls under the Great Prison were repaired in 1296–7, as were those of Benets Tower (Midgley 1942, 134). The use of a tower for or in association with a 82

pro prisonariorum custodiendo. unus puteus pro felonis imponendo et una domus cum diuersis cameris… pro custodis eiusdem prisone inhabitando). The richly detailed account of the building of this two-storey battlemented prison house suggests that what was a substantial building was built on ground that had not been dug before, and that it may have lasted longer than most of the castle buildings (Dewey, pp. 154–5, and Figure 7.14). A considerable quantity of malmstone and freestone was used to build the prison house, as well as nine-inch (0.2m) thick wooden flooring of the prison itself. The accounts reveal that the prison was on the lower floor of the gaoler’s two-storey battlemented house, which included a hall and kitchen, with buttery and larder on the ground floor, as well as the prison, and two chambers above, reached by a flight of 15 stairs, 3ft 6in (1.1m) wide. The gaoler’s house was spacious and well lit. The prison, on the ground floor of the building, had three doors and three windows, perhaps indicating division into cells, or a security arrangement (Nevell 2014–15, 209). The overall dimensions of the house were about 70ft by 42ft (21 x 13m). Next to this building was a subterranean prison, called the pit (puteus), with foundations 15ft (4.6m) deep and serviced by a trap door, within a small building that was very close to or attached to the main building. The main building and the prison were well supplied with windows (unlike the earlier Blindhouse?). The wood flooring and walling specified for the prison was for walls 19ft (5.8m) long (on all sides, presumably) and 10ft (3.0m) high.

of the clerical lodgings, all in timber, according to a letter to the Lord Privy Seal by the egregious Dr London, last Dean of the college (Hedges 1881, ii, 296–7). Work was also carried out in advance of a brief visit by Elizabeth I in 1570 (Bodleian, MS Rawlinson C 25, fols 267–8). A striking feature that occurs only in the 1327 survey is the building called La Studie, though it may have been ‘the earl’s new chamber’ mentioned in the 1296–7 accounts. This may be a type of prestige building such as the study (studium) built at Eltham for Henry IV at the turn of the 15th century, described c. 1540 by John Leland as a reading room with chests of books and jointed boards that could be pulled open to make a reading surface on which to lay the books (Goodall 2011, 23). It was in the middle bailey, so it may have been linked to the College clergy, who would have performed some of the writing offices for the lord of the castle, keeping records of decisions of court sessions and the like. A castle of this importance would have had its own archive, including records generated by the sheriff’s court, in the custody of the chaplains (Pounds 1990, 98). The name La Studie is more suggestive of culture than mere record-keeping, however, and is just as likely to be a study or library as an archive. Earl Edmund bequeathed an illuminated manuscript of the Historia scholastica of Peter Comestor to his priory of the Bonhommes at Ashridge. Now British Library MS Royal 3 D VI, it was probably written for Earl Edmund between 1283 and 1300. It bears his arms and those of Edward I and Edward (II) of Caernavon on fol. 234 (Schulz 1938).

Houses of the chapel clergy and the studium in the middle bailey

Great Chapel and the east curtain wall When Earl Edmund refounded the castle’s Great Chapel or College of St Nicholas as a chantry he changed the nature of the church and created a permanently inhabited ecclesiastical enclave in the middle bailey. Elite castle chapels had certain trajectories of development, often being initially built on the foundations of an earlier Saxon church and subsequently granted to a regular order of Benedictines or Augustinians (Denton 1970; see further above, p. 71). A few remained as colleges of canons. Refoundation as a chantry was part of the same trajectory and was happening in baronial castles from the later 12th century (Crouch 2001). A chantry was created when the donor’s requirement for daily masses for his and his family’s soul was such that it would impede the function of a religious house designed to minister to its own congregation, so a priest or priests were attached instead to a separate foundation to fulfil the patron’s rather narrower mission. The move had started with Earl Richard’s endowment of a small chantry chapel of 15 marks annually from the revenues of the castle mills for masses for the souls of Richard and his first wife Isabella of Gloucester in perpetuity. Apparently extending the dedication to the Holy Trinity, St Mary and St Nicholas, Edmund reorganized the college as a chantry, specifically to pray for his parents, Earl Richard and Sanchia de Provence, his ancestors and his posterity, expanding the clergy to a dean, six chaplains, six clerks and four choristers. Their lives were well regimented: at least four masses a day were to be

A notable omission in the 1327 account is reference to the houses of the college clergy in the middle bailey, where Earl Edmund had established them in 1282, though they were probably subsumed in the phrase ‘with other chambers and the bridge above and below’, the bridge being the middle bridge and its gatehouse, the Bydongate, which fixed the location of the clergy’s lodgings in Edmund’s foundation charter (above p. 53). They are evidenced in other accounts, when repaired, re-plastered, or re-roofed in 1296–7, 1307–8, and 1331, when the clerks’ house was thatched and a stair in the house of the beneficed priests was repaired, along with the gutters. Full scale repairs to these houses, and to the College itself, were ordered by the Black Prince in 1362 following storm damage. In 1389 two tilers worked for 50 and 52 days respectively on the roofs of the Great Hall, Great Stable, chapel, part of a chamber over the buttery, and the roofs of all the houses, whether of the king’s lodging or the houses of the college priests and clergy. Repairs to the walls of the cloister of the college were ordered in 1482–3, on behalf of Edward IV’s young heir (TNA SC6/1302/1). The lord of the castle was responsible for the maintenance of the whole estate, but a clear sense of the distinction between the royal lodgings and the clerics’ occurs in the 1389 accounts for repairs to houses in the castle tam in manso regio quam in omnibus aliis domibus collegiis presbitorum et clericorum. The only known work from the reign of Henry VIII was rebuilding 83

said, according to the rite of the church of Salisbury, and absences from the offices were punishable by deductions from their stipends. All ministers to the chapel were to have free ingress and egress of both chapel and town whenever they wanted. They had to ‘swear on the Gospels that they will firmly and faithfully keep our counsel within the castle, lest any harm be done through them within the castle; and if they hear or notice anything within the castle that might lead to harm to us, they should report it at once to us or our council’ (1.51). Edmund endowed his chantry with an income of £40 per annum from the rents of Warborough and Shillingford (members of the manor of Benson), which he confirmed to the dean, Roger de Drayton, in his foundation charter. Of particular interest in the 1389–90 accounts is the major repair to a section of wall on the ‘east part of the college of the said castle extending from the north towards the south and containing in length 188ft [57.3m] and in height from the ground 14ft [4.3m] and in thickness 6ft [1.8m] with eight new windows, which parcel of the said wall contain in the foundations of the same beneath the ground 12 arches on account of instability in the foundations and the same foundations with arches reach 10ft [3.0m] under the ground’, for which six master masons were paid 6d a day and six other masons 4d a day (HKW ii, 851). The need for this repair had been identified nearly 40 years earlier (1.47), before the havoc wrought by the great storm of 1362. However great the need for repair, the fact that the wall had survived all this indicates that the build quality was fairly high. The importance of the description of this stretch of wall is that it enables us to locate the Great Chapel or College of St Nicholas, which was certainly situated within the inner bailey, as shown by the 1327 survey. For such an enormously long wall section to be located with reference to the 30ft (9.1m) width of the college east end, as recorded in the Tudor survey (above, p.54), must mean that the section in question ran from a point parallel with the college east end from the north towards the south for 188ft (57.3m). Logically, this places St Nicholas along (but not necessarily attached to) the north curtain wall of the inner bailey (which used the original Saxon ramparts). It is likely to be the rectangle drawn in that location on the Tudor plan of 1548, which is further confirmed by the near coincidence of the dimensions suggested by both the plan and the written account of the 1555 surveyors (Lloyd 134, 2.1), and by recent analysis of the geophysical survey (Christie and Creighton 2013, 168). One is tempted to infer that the reference to the college east end without reference to any other building in a long 188ft (57.3m) stretch is an indication that the main halls and chambers of the bailey were on the opposite, western side, which has some support from the excavations of 2010 (ibid, 173– 82). There were, however, at least three major halls, as we shall see, as well as a number of minor chambers or halls. What we can say is that the account of 1308 shows a man working for 13 days making the garden around the Great Chapel, confirming that it was a free standing building. This suggestion accords with the evidence of the 1548 plan (Figure 6.1), the geophysics (Figure 5.1), and the 1804 estate map (Figure 7.48).

The 1389 account refers to the refurbishment of the timber belfry for the church. It had also been repaired, using recycled timber, in 1308, though fresh beechwood was obtained for hanging the bells. In 1307, when 2100 shingles were laid on its roof, roof boards were laid on the oratory on the north side of chapel and on an oratory and lavatory on the south side of the chapel. One of the oratories was possibly the small chantry chapel endowed by Earl Richard on the death of his wife Isabella of Gloucester (1240), some 40 years before his son’s refoundation of the chapel itself as a chantry. Richard had allotted 15 marks of annual revenue from the castle mills for the sustenance of a chantry priest (St John’s College, Oxford, Muniment III. 1, fol. 5v-6r). A good deal of work was done on the chapel, including repairs to the ironwork of the windows in 1363–6. In 1411 Thomas the plumber of Abingdon was paid for various jobs including a repair to the window pane above the chancel of the Great (actually ‘greater’, maj[or]’) Chapel; and John Culham stone-cutter, was paid for the repair of a window and a step before the altar of the aforesaid chapel. This description of St Nicholas as ‘greater’ rather than ‘great’ chapel may be significant, since by the mid-14th century at the latest there was a chapel of no mean size in the Great Tower, on which work was recorded in 1363–66, 1369 and probably 1375, discussed below, as well as smaller chapels in other chambers. Chamber of the Knights and Esquires, the Green Chamber and Hubert’s Hall Among the many buildings or houses in the inner bailey was the Chamber of the Knights and Esquires, first mentioned in 1284–5, and again in 1296–7 with reference to re-tiling the roof of its garderobe; refurbishments were carried out again in 1331,when a key was purchased for the knights’ great chamber (magnam cameram militum) and the locks repaired. Since in 1284 carpenters repaired the louvre and gradum of this chamber, it is likely to have been a groundfloor hall; the gradum may have been a reference to its high (dais) end, or to a staircase. Its benches, locks and other furnishings were refurbished in 1331. These may be fairly rare references to the accommodation of honorial tenants or their proxies who performed castle-guard, or the knights and esquires may have been the household knights of the royal lord of the castle. It is clear, however, that this was one of at least three major halls in the inner bailey and it is the one most likely to have been on the south southwest curtain that was excavated during 2010 (see Figure 2.8 above). The Green and Red Chambers probably derived their names from some form of decoration; at this castle, with its connections with Cliges, Sir Gawain and Guy of Warwick, the décor of the chambers probably reflected themes in courtly romances. The Green Chamber only occurs under that name in 1307, when it was provided with a key, and in 1327. The 1308–9 account refers to repairs of the locks of the coffers in the Great Tower and its door, and repairs to the lock of the chamber next the Thames (camere iuxta Tamis), the King’s Chamber, the outer gate, the Constable’s Chamber and the first postern. If the chamber next the 84

Figure 5.37 Painted wall in a room in Castelvecchio, Verona (photo: Judy Dewey).

in the Great Hall, and four iron braces for the countess’s table. The countess’s chapel was glazed and her wardrobe re-tiled. One line accounts for timber bought for the countess’s bridge toward her well or fountain (ad pontem comitisse versus fontem comitisse), and 700 tiles bought for the countess’s chamber. The clear distinction between the countess’s accommodation and the earl’s suggests that the move towards separate residential suites around the bailey that is a common feature of the later Middle Ages began in the time of Earl Edmund. In 1307 1000 tiles were laid on the pentice between the Red Hall and the King’s Chamber (super penticium inter rubeam aulam & cameram regis); possibly this occurs in the 1284 accounts as the long alure to the new chamber (longam aleam uersus nouam cameram). The following year two louvres were installed (a good indication of the size and importance of this hall), and the guttering above was re-leaded. In 1331–32 its lock and key were mended, and screens and partitions renewed, and window bars were renewed. The screens passage (alurarum) leading towards it from the buttery, bakehouse, larder, second kitchen (secunde coquine) and scullery was re-tiled. An important second hall of this date would have had its own kitchen, as the Red Hall clearly did, with its own services, separate from those of the Great Hall. In this account it is clearly connected with the visit of Queen Philippa and her young son, the three-year old earl of Chester. This suggests it may have been the hall block serving the Queen’s Chamber. Reference to the repair of an outer chamber next to it (cameram forinsecam iuxta rubeam cameram) in 1296– 7 is a possible reference to a latrine, but in 1331 a privy was made for the young earl in a parclose in the chamber over the water (cameram super aquam), which suggests a

Thames was not a sallyport in this area, then one might speculate that it was the same as the camera super aquam associated with the queen’s visit in 1334, and was located in the south-east tower. Perhaps the Thameside chamber was the one also called the Green Chamber, echoing the Green Chamber of Henry III’s Westminster palace from which he could view his garden (HKW ii, 505). Also hard to identify is Hubertshall, mentioned in the accounts of 1307–8 as the hall once of John Hubert, a burgess and former reeve. The Green Chamber and Hubert’s Hall do not occur under these names in the later accounts (Figure 5.37). Red Hall, Queen’s Chamber The interchangeability of the words hall (aula) and chamber (camera) is occasionally confusing (Quiney 1999; Blair 2003b), but the importance of the rubea camera is not in doubt. It helps that the one building of which the descriptor never varies is the Great Hall, magna aula. The Red Chamber was an important building, and recurs again and again in the accounts. It clearly became the second hall of the castle during the 14th century, when domestic plans were moving away from the great hall as a fixed point and starting to break down into smaller discrete suites of lodgings (Goodall 2012, 25). It is first mentioned by name in the 1296 accounts, which postdated the breakdown of Earl Edmund’s marriage. It possibly originated in the suite of rooms attributed to the countess in the 1284 accounts, which distinguish between earl and countess’s accommodation throughout. The benches (formas) of the garderobes of both the earl and countess were replaced, as were the benches, trestles and windows 85

(iuxta cameram Sen’), stools were repaired and benches made for both it and the chamber of the knights; the latter was certainly in the inner bailey. The constable’s lodging was one of the most important and was often located in the main gatehouse, where it would give the constable control over admissions to the castle. It was usually close to the porter’s lodgings (Kenyon 1990, 135). As we have seen, the outer gatehouse at Wallingford was referred to as the constable’s entrance in 1296, recalling the Constable’s Gate at Dover, where it remains the lodgings of the governor of the castle or his deputy to this day. Given the importance in Wallingford of the inner bailey, with its access to all the principal houses and the Great Tower, office or chamber accommodation for the constable in the inner bailey, next his garden, would make sense, as would accommodation or offices in the inner gatehouse. The survey of 1555 described the gatehouse as a long entry, called the Lieutenant’s Lodging, a timber building some 40ft (12.2m) long and 16ft (4.9m) wide (Lloyd 134 and 2.1 below). In the 14th century the

latrine chute in a tower, most likely the south-east tower at the foot of the motte, discussed with the Great Tower below. The accounts for the period 1363–69 reinforce the view that the Red Hall became associated with the queen, perhaps during the tenure of Queen Isabella. These accounts relate to the major works following the great storm of 1362, but also coincide with the marriage on 10 October 1361 of the Black Prince to Joan of Kent, who made Wallingford her chief residence until her death there in 1385. Most of the information is attached to the names of those contracted to carry out the works. John Tyler was contracted to re-roof the Wardrobe and other chambers in the inner bailey (super terciam pontem), including a hall, the Queen’s Chamber and another chamber at the end of the hall (aule, camere regine & alie camere ad finem aule), for £9 3s 4d. Master carpenter William Shalstone worked for eight weeks on new projecting moulding for the Red Hall (novum supercilium unius aule voc’ la reedhalle), for a total of 24s. The work required 100 oaks felled at Nettlebed. It is clear from this, given the way that accounts deal with the same project under different categories – stone, timber, gutter etc. – that the hall on which John Tyler worked was the Red Hall. In 1369 the alure abutting it was re-tiled using 8300 lathnails. John Tyler worked for four weeks and three days tiling the roof of the Red Hall, as well as nine weeks and a day on other tiling works, over the Great Hall, the Queen’s Hall (aula Regine), the Great Kitchen and the priest’s house. In this case the Queen’s Hall is almost certainly to be identified with the Queen’s Chamber of the earlier account. The two accounts are closely linked, both in time and in the main focus of the works. It helps to remember that in 1327, when the survey on which our detailed knowledge is based was made, Queen Isabella held the castle in dower. Thus in the list of buildings in the inner bailey we should understand ‘Queen’s Chamber’, coming after the king’s Great Hall, as ‘Great or King’s Chamber’, and the Queen’s Wardrobe as Great Wardrobe. It is possible that by the 1360s the second hall complex housed the Red Hall and Queen’s Chamber, the latter possibly a new name for the former Green Chamber, unless that were the ‘other chamber at the end of the hall’ occurring in 1363. It was probably also linked to the ‘chamber over the water’, probably in the south-east tower. Constable’s Chamber, Constable’s Gate, Lieutenant’s Lodging The constable’s house or chamber, repaired in 1296 and 1308, may have stood in the inner bailey near its garden, mentioned in 1296–97 when a master carpenter spent 12½ days repairing the constable’s dresser and making four perches of cob walling for his garden (Midgley 1942, 133– 4). The house had an adjacent stable, repaired in 1308. The leading of the guttering over it was repaired at the same date, when it also received the attentions of a tiler and a stone mason. By this time the role of constable was merged with that of honorial steward, so we can associate the seneschal’s chamber with him, or perhaps with his proxy, the under-constable. In 1331 a wall was built besides it

Figure 5.38 Great Hall at Berkeley Castle, minstrels gallery over the entrance formed from painted wooden panels (not in original location); four-light window with oillets above (photo: Katharine Keats-Rohan).

86

constable and his under-constable (lieutenant) would have been a real presence in the castle by contrast to the 16th century, when the post of constable was largely a sinecure and the castle little used. The Tudor evidence is difficult to interpret; the Surveyor’s Plan shows a bastion rather than the gatehouse tower at the entry into the bailey indicated by the 1389 accounts; the 1555 survey describes a substantial timber gatehouse and does not mention a bastion or barbican. This inner gatehouse is only once described as a tower, when a new bell was hung there in 1389, when the façade of the building was of stone. It is difficult to believe that the timber-built Lieutenant’s Lodging was the inner gatehouse, rather than an important element of one of the chamber blocks associated with the Great Hall. Although it is possible that this inner gatehouse was built on a small scale and that the Lieutenant’s Lodging was built onto it during the 15th century, it seems highly unlikely since such a long building would have considerably impeded the cartage of goods in and out of the bailey, which housed all the castle’s main stores.

of glass. At the same date there were three windows with two lights and three oillets in the lord’s Great Chamber, containing 68ft (20.7m) of glass. (An oillet is a small circular opening, with a ring of mouldings surrounding it, that was used in window tracery in Gothic architecture.) In that year the Lord’s or Great Chamber was a primary focus of the various works. In separate passages we learn that iron casements for four windows in the Lord’s Chamber and its chapel were repaired, and a new door was fitted on the Lord’s Chamber and a lock of the door of its chapel. The Great Chamber occurs under a number of names, depending on who was in charge of the castle at the time: King’s Chamber, Queen’s Chamber, Lord or Earl’s Great Chamber. For much of the 14th century at least, the private apartments of the lord were in the Great Tower. For the lady or queen, they appear to have been associated with the Red Hall. The services of the Great Hall are enumerated in the 1327 survey as kitchen, bakery, brewhouse, associated smaller rooms and three ovens. Even where a great tower

Great Hall, kitchen and services, King’s or Great Chamber, Treasury and Wardrobe, and Queen’s Tower The Great Hall will have been where the manorial court met, the manor of Wallingford being coterminous with the castle precincts and the meadows of Clapcot; the phrase ‘castle and manor’ occurs in the documents of many castles almost as synonyms (Pounds 1990, 201). A castle’s great hall was one of the major buildings of the castle complex, traditionally the social and administrative hub of the castle community, as well as a courthouse for its external business, whether for honour or shire. It is invariably called magna aula in the accounts, except for one of four references to a repair of its west gable in the 1389 accounts when it is called camere principis (sic). It was also used as a communal dining room, presided over by the lord sitting on a raised dais at one end (the high table). Service areas connecting to kitchen and pantry could be accessed by screen passages leading off the hall. There might be an upper gallery, for entertainers (Figure 5.38). At the opposite end to the services would be the chamber block, which contained audience and bedchambers. The hall and associated chamber block, whether attached or separate, could have one or more floors. As time went on there was a general move towards building smaller hall and chamber blocks for principal officials as well as the lord, reserving the Great Hall for meetings of the honorial court or ceremonial occasions. We know that in 1220 Henry III ordered the demolition of the existing building and the building of a new one on the same site (1.24). We have no idea what form it took then in relation to service and chamber ends, and it is extremely difficult to state with any certainty what the various stages of its evolution might have been from what clues the building accounts provide. It was first glazed in 1223, since it was accounted for by the sheriff in the Pipe Roll of 8 Henry III. In 1375 we learn that there were then three windows with two lights and two oillets in the Great Hall requiring 75ft (22.8m)

Figure 5.39 Great Hall at Berkeley Castle, facing the dais end; on the right a door gives access to the great staircase leading to the great chamber (photo: Katharine Keats-Rohan).

87

Figure 5.40 Decorative capital of c.1230. The decoration takes the form of foliage, with large clasped trefoils at the top and simpler ones rising from the necking below. Note the palmette superimposed on the lobe of one of the larger trefoils, an unusual feature. The capital is carved from chalk, showing that it comes from an internal architectural feature, probably a doorway or possibly a window embrasure. It is shaped to sit at the top of at least two engaged colonnettes flanking the opening, and to serve as the springer for the arch above it. 30 x 14.5 x 12cm. Private collection.

Figure 5.41 Limestone relief carving of c. 1120–30, 33cm (13in) across, found in the later 19th century within the original confines of Wallingford Castle according to a note by J. K. Hedges shown after his death to C. E. Keysey (Keysey, 1907). The subject is the constellation Aquarius, the water-carrier, indicated by the position of its component stars, current since the 5th century BC. The relief may have formed part of a frieze or series representing the signs of the Zodiac, perhaps built into the wall of an important interior such as the castle’s hall. Representations of the twelve signs of the Zodiac are common in medieval art, sometimes associated with occupations or ‘labours’ of the month: as Aquarius is matched with January, and so with feasting – a subject highly suitable for a hall - the sculpture may in this case have been alone. The depiction of the water carrier is, however, very unusual. Instead of a youth pouring water from a jar, here the constellation is represented by a dog-headed figure or ‘cynocephalus’, with hoofs not feet, carrying water in skin bags slung over a shoulder-mounted pole. (Image from the Hedges Collection © Reading Museum Reading Borough Council.)

provided a separate suite of accommodation, as was the case at Wallingford, the buildings in the bailey would include a great hall in close relationship to a king’s or great chamber. There are several references to both in the 1284– 5 accounts, including work on the (entrance) porch of the hall – porch’ aule – and whitewashing of a building outside the hall door where spices are kept(?) (per unam domum dealband’ extra host’ aule ubi speties iacent). In 1307 a wall between the Great Hall and the King’s Chamber was re-built and coped (in factur’ et tabulat’ cuiusd’ muri inter magnam Aulam & Cameram Regis), the job requiring 14½ cartloads of lime and 14 of stone; and part of a chimney in the King’s Chamber was repaired. The foundation of the entrance going towards the king’s chamber was repaired in 1308 (Item in fundament’ introit’ versus cameram R. emendand’ ad tasc’ viij d.). The walls were re-plastered in 1366–7. There is a lot of detail about repairs to the partitions in the screens passage in the 1331 account, as well as in 1375. In the latter year, when wainscotting and three windows with three oillets for the Great Hall were also mentioned, carpenter John Bate was paid £6 13s 4d for work on the panelling or ceiling of the lord’s Great Chamber at the high (dais) end (superiorem partem) of the Great Hall and for sawing boards for the same. The same accounts state that this chamber was next the Queen Tower (iuxta le Qwenetour), which was therefore the north-west tower of the inner bailey, on the opposite side to the stone fragment to which the name adheres today. At that date there had been no direct association of the castle with a queen for many years, suggesting that the name reflected earlier domestic arrangements. The damaged 1284 accounts record the cleaning of ashes from the kitchen, its andiron, and the repair and

cleaning of its guttering. Both this and the 1296 accounts give rare insights into domestic conditions. In 1296–97 the earl’s new chamber (first mentioned in 1284–85), and his salem (sic), are mentioned, the latter probably referring to a first-floor chamber known as a solar (Midgley 1942, 135; Goodall 2011, 28; Cobb 1883). There is also reference to tiling and pointing the ymam cameram (Midgley 1942, 133). A ‘winter chamber’ was one suitable for use during the cold of winter; alternatively, the word was imam, meaning lowest or innermost. It is also mentioned in the 1284–85 account, followed by the plural et ymas but the following word is lost, along with the chance for enlightenment. It is known that withdrawing chambers in the hall block over the services could be used in winter, when they benefited from warmth generated by the kitchen (Goodall 2010, 38). A castle in winter offered many challenges: the 1284 accounts itemize the clearing of snow from the hall and other buildings. Many building accounts show concern for protecting masonry from winter frosts with bundles of brushwood and other materials, especially when building works were unfinished. The 1295–96 account refers to tiling and pointing over the nutricium and the kitchen. The former, from a word relating to nursing or nourishment, occurs also in 1284– 88

account mentions buttery, bakehouse, saucery, scullery and the second kitchen, which appear to be serving the Red Hall; some of the walls were stone, but the partition walls were of wattle and daub construction, which appears to be true of the wall of the small (parua) kitchen; a granary is also mentioned, and a chandlery, and the cleaning of water channels next the bakehouse to improve the water flow there, as well as the purchase of a new key. In 1356 a new kitchen was to be built and there is reference in 1365 and 1369–70 to re-tiling of the Great Kitchen. This service block was probably two-storeyed; a chamber over the buttery was re-roofed in 1389. At the same date repairs were made to rafters and boards in the screens passage between the hall and the chambers going to the buttery. A pastry kitchen was also repaired (Pasthous, la Pasterye). In 1388 the west gable of the hall was rendered and part of the alure or screens passage extending from the hall towards the buttery was repaired; the walls of the pastry kitchen (La Pasterie) were repaired and a wall plate provided for the chamber next the wine cellar. The account of 1363 suggests that the Queen’s Chamber and another chamber were at one end of the hall, but in this case, as argued above, the hall in question was the Red Hall. In 1408 the spiral staircase serving the Great Chamber next the hall was repaired (in emendacione cuiusdam vice infra Castrum deservient’ magne camere ibidem iuxta aulam), as well as another ‘great stair’ within the castle (Et solut’ cuidam carpentar’ pro emendacione cuisdam magne scale existent’ infra Castrum ibidem). This may have been a grand entrance to what was at this date a first-floor great hall (Quiney 1999), or perhaps the grand staircase leading to the Great Chamber from the dais end of the Great Hall (Goodall 2011, 36–8). The 1389–91 accounts refer to a staircase extending 24ft (7.3m) from the King’s Chamber towards the Queen’s Herber (quod quidem grete [sic] restat et extendit a Camera Regis versus le Queneherber’ in longitudine xxiiij pedes). It also mentions repairs to a great paling, 30ft (9.1m) long, between the door of the King’s Chamber and a wall opposite (unum magnum Paleys & altum iam existentem inter hostium Camere Reg’… & quendam murum exopposito dicte Camere per longitudinem xxx pedum). This could indicate a floor below the King’s Chamber containing a stair up to it and a door which opened into a northern court dominated by the Great Chapel. The general impression is that by 1363 at least, the Great Hall and the King’s or Great Chamber had separate entrances, but were internally linked via staircases. The chamber block and, at the opposite end of the hall, the service block, were probably both two storeyed, projecting beyond the line of the hall into the bailey, but the hall itself is likely always to have been a single storeyed room with arcading and a high vaulted roof. The Great Hall certainly had a screens passage and integral services in the late 14th century, but the Great Kitchen may have remained a separate building, as was often the case due to the risk of fire. The chamber block probably abutted the hall and projected beyond it, using the line of the curtain wall and tower (Thompson 1991, 93–4; cf Goodall 2011, 36–8). Given that, as discussed above, we know the location

Figure 5.42 Fine stone from the castle was widely reused by local people under the terms of the demolition order. These two examples, now in Wallingford Museum, are both 12th-century work attesting the lavish and fashionable state of castle décor at that time. The first is a beakhead stone, the most magnificent surviving local examples of which are in St Mary’s Church, Iffley (Oxon). They became popular in English Romanesque architecture under the patronage of Henry I, founder of Reading Abbey where they were used extensively, appearing also at Norwich Castle and in the buildings of Bishop Roger of Salisbury in Old Sarum (Wilts). Evidently they were also adopted by Henry’s friend and supporter Brien fitzCount of Wallingford, who died a monk of Reading. The nailhead stone features a pyramidal motif very popular in the 12th century, and surviving in several local churches, including those at Buckland and Upton (Berks).

85, and recurs in the accounts of 1307–10, when 19 carts of stone were required for repairs to the foundations of La Noricerie; it does not occur again except in an account from 1389–90, when the ‘chamber next the chamber called Norise Chamber’ was given a gutter (or water channel). It was presumably related to the catering function of the kitchen and associated buildings, though the word noricerīe in Middle English means nursery for children. The 1331 89

Figure 5.43 Well chamber just outside the Great Hall in the keep at Dover, easily accessible from the forebuilding entrance into the keep, as seems to have been the case also at Wallingford (photo: Katharine Keats-Rohan).

Jewish community in England, would have required a safe treasury. In 1308 repairs were made to the locks of the coffers in the Great Tower, as well as carpentry on the gable of the Treasury next (iuxta) the Great Hall. The Treasury was probably associated with the work of the Great Hall, which served as both honorial and shire courthouse; in addition to the collection of dues owed by honorial tenants, most of the penalties inflicted by the court would have been fines. Unlike the Treasury, evidenced between 1296 and 1308, the Wardrobe as an administrative department is mentioned throughout the period from 1219 to 1411.A wardrobe was ordered to be built in 1219, at or around the same time as the new great hall was commissioned for Henry III. Both treasury and wardrobe are mentioned in the 1296 accounts. The Wardrobe, like the Treasury, was a separate building, In 1363 ‘a house called Wardrobe’ was tiled, and in 1369 there was tiling work on the Great Hall, the Red Hall and Wardrobe. The accounts from 1389–90 show extensive carpentry works, using the wood from 21 oaks which the king’s forester, Fabian, was ordered to select from the royal forest at Pamber, Hants. It was used for the renewal of the floodgates (the sluice system turning the water in the moats), repairs to the inner bridge, the apparatus of the draw-well outside the hall, the new belfry, and the passage extending from the Great Hall between the chambers towards the wine cellar (quadam alura extendente a dicta aula inter cameras versus celarum vini) as well as the paling outside the King’s Chamber; at the same place a

of the Great Chapel, reference to the repair of the west gable of the Great Hall indicates that it stood in one of three possible locations. Least likely is in the middle of the bailey, parallel with the east-west aligned Great Tower on the motte and the Great Chapel along the north wall, since this would make latrine arrangements very awkward. So, using the Surveyor’s Plan of 1548 as a base, (see Figure 6.1), it lay either on the curtain wall in a southwest to south orientation, or against the curtain wall on the opposite side of the drawbridge entrance, in a west to north-west orientation. The latter option would bring it most closely into the normally desirable relationship with both the gatehouse entrance and the Great Chapel (Pounds 1990, 240; Thompson 1991, 94). It would also mean that the chamber block was on the north facing south, the aspect normally favoured (Thompson 1991, 111). The Great Kitchen could have been in the middle of the bailey (ibid., 93–4). A draw-well stood at the door to the hall (Figure 5.43), and the Treasury, a separate building, stood close to its west gable. The Treasury is mentioned in 1296 and in 1307–9, once as camere Thesaurarii and three times as domus scaccarii. It has not be traced later using this terminology. Could it have reappeared in 1331 when the door of the ‘Scotfine’ was repaired? ‘Scot’ refers to a tax, so this building could be related to collection of money or even imprisonment for default. An almonry is mentioned in 1389. A royal castle and one associated with the fabulously wealthy Richard of Cornwall, who had oversight of the 90

carpenter who earned 5d a day, the other, possibly his apprentice, earning 3d a day. Between Pentecost and 28 November 1390 seven carpenters in rotation were used on the works. The reference to a great and high paling is likely to indicate that the inner bailey, which was anyway unusually large, was divided into different courtyard complexes, a not uncommon arrangement. The ‘king’s court’ will have been the one containing the Great Hall, Great Chamber, Queen’s Tower, and private access to the Great Chapel. The Red Hall suite may have formed a separate court on the south-east, and the Chamber of the Knights and Esquires another on the south-west. A pentice linked the Great Chamber and the Red Hall. It may also have created another court, containing the Great Chapel and the gardens surrounding it, since it appears that access may have been controlled by the location of a doorway beneath the King’s Chamber, as discussed above. A further enclosure of the sensitive area accessing the forebuilding to the Great Tower and associated bailey-level storage areas is also likely, as may be suggested by the geophysics (Figure 5.1 above). At this point it is possible to advance what can only remain a sketch of the possible layout of the bailey buildings given in Figure 5.45.

Figure 5.44 Stone rosette boss from Hedges’ Wallingford Castle Museum, © Reading Museum (Reading Borough Council).

rose (a wood-carved boss for a roof vault) was made for the Muggetour. (Figure 5.44) Later some timber was left in store in the Muggetour in the care of the chaplain, perhaps indicating that this was the north-east tower. Between Christmas 1389 and Pentecost 1390, 18 jobbing carpenters were employed on these works at 3d a day for 18 days; two others worked for a 104 days, one of them a master

KEY 1 Great Chapel of St Nicholas 2 Queen’s Tower 3 King’s Chamber block 3a internal great stair to KC from dais end of Great Hall 3b doorway into internal court 4 Great Hall 4a porch 4b spicerie and well 5 service block; includes buttery/ wine cellar, pastry kitchen, reached by screens passage 6 Treasury 7 Great Wardrobe 8 Great Kitchen and bakehouse 9 Drawbridge entrance and tower; Lieutenant’s Lodging 10 Hall of Knights and Esquires 11 Benets Tower and Great Prison 12 Great Tower 12a secured storage area and forebuilding entrance to Great Tower 13 Water Tower 14 Queen’s Chamber 15 Red Hall and services 15a pentice to King’s Chamber 15b door to garden 16 Garden and herber court 16a long trellis/bower 17 Muggetour

Great Tower The Great Tower on the motte appears to have been first noted in the 1220s (king’s alta camera), but had probably existed for a good deal earlier, as argued above. It was unambiguously mentioned in 1284, 1296, and in the 1307

Figure 5.45 Suggested layout of the inner bailey buildings in the later 14th century, overlaid onto the 1804 map of the inner bailey (Figure 7.53) and guided by the geophysical survey (Figure 5.1) (illustration Debbie Miles-Williams).

91

Figure 5.46 Hand mill in the kitchen at Dover Castle (photo: Judy Dewey)

Figure 5.47 Quern from Hedges’ Wallingford Castle Museum, © Reading Museum, Reading Borough Council; there were four hand mills in the Great Tower at Wallingford in 1327.

accounts when roof boards were used on a turret of the Tower, and various locks were repaired. Normally it was described as the Great Tower, but in the 1307–9 accounts, when the lead flashing on its roof guttering was renewed, it is simply the Tower, and also in 1331–4. The reference to the four hand-mills in the latter puts the matter beyond doubt, since these are mentioned in the 1327 account as being in the High Tower (alta turris), which is one of the normal terms for Great Tower – magna turris, along with turris and donjon: the word ‘keep’ is late medieval and not found in the documents used here (cf. Thompson 1960, 91). Their appearance confirms that one of the several roles of the Great Tower was as a residence. These hand-mills would have been essential had the castle been besieged and access to the riverside mills denied, but were probably just used for domestic convenience (Figures 5.46, 5.47). A similar arrangement is recorded at Windsor (Pounds 1990, 193–5). It is also abbreviated to ‘the tower’ in those accounts in which it occurs as the Great Tower. The frequent reference to the extraction of large casks from the tower of arms or victuals for dispatch to other royal residences, or to storage of materials, suggests that the tower had considerable storage space available. In the 1307–8 accounts two cartloads of lead were brought from Oxford Castle to Wallingford, where a good deal of lead was being used for repairs to roof guttering in the mills, the Red Hall, the Constable’s Chamber and the Great Tower. Two cartloads of lead were carried into the tower and one tun (content unspecified) was taken out at a cost of 12d. Ten tuns of flour were taken from the tower and carted to Windsor; the heading in the accounts, as for the following year, is custus celar’. In all, 17s 10d were spent on carpenters and a cooper repairing tuns, and for two men helping over four days to turn the tuns, presumably to maintain the quality of the contents. Candles for oversight of the wine and the mending of tuns in the period came to 8d. Difficult though it is to imagine that all the moving of casks was done by

running up and down the steep stairs to the forebuilding, accounts show that this was quite normal in many castles. The forebuilding is primarily evidenced in the 1555 survey of the ‘keep’, as the Great Tower was then called, discussed by Lloyd below, but an important early reference to the entrance into the tower comes in the damaged 1284–5 accounts, when timber and large nails were bought for the drawbridge of the tower and the platform (wooden step) going into the tower (in meremio et clauis grossis emptis ad pontem trahic’ turr’ et ad gradum turr’). Storage capacity under this forebuilding entrance into the tower at first floor level is quite possible. There was a similar arrangement to the forebuilding at Norwich, also on a motte (Goodall 2010, 113), and at Rochester, which was a great tower (Figure 5.48). Mention of old timber stored in the Great Tower in the 1296 accounts reads: infra introitum magne turris (Midgley 1942, 135). Although infra is more likely to mean ‘within’ than ‘below’ in these documents, it conceivably refers to a storage area or chamber accessed below the stair of the forebuilding (Cathcart King 1988, 67), the bulk of which was the entire ‘ground’ floor underneath the first floor entrance into the tower and its main accommodation. Lloyd’s discussion of the forebuilding shows that it contained a very long, substantial staircase, with heavy oak doors top and bottom. This perhaps occurs in 1369, when a good deal of work was carried out on the Great Tower; Robert the mason of Henley worked for 4s a week for three weeks on ‘divers works on the Great Tower and other houses, viz the Aleyne annexed to the Great Tower..’ (divers oper’ magn’ Turr’ & al’ dom’ viz Aleyne dicti Turre annex’). The forebuilding should be envisaged as a structure that began at bailey level and was built up one face of the motte and then attached to the north face of the tower (see Figure 5.11). This in turn suggests that the motte slope on that – the north – face was steeper than on the south, where it sloped down into the moat that surrounded it on all but the north side. This accords with 92

Figure 5.48 Cut-away reconstruction drawing of Rochester Castle keep (completed c. 1140), showing access to a basement storage area under the bridge-pit of the entrance to the forebuilding (IC087_002 © Historic England). Here access to the well is provided on all floors through the central spine wall.

The 1327 survey account of it is slightly ambiguous, though it is certain that the Great Tower had a stair turret, and an internal stair, as well as an enclosed staircase entrance from the bailey leading to a porched forebuilding, as mentioned above. Externally it was usual for pilasters at each corner to rise to a turreted top. The base of these pilasters were splayed to increase stability. These were the ‘foundations’ for which, along with corbels, modest repair costs were sought in the 1327 survey. Orders were given to prepare the donjon against a visit by Richard II in 1389–90; it was certainly, if rarely, being used as the lord’s private residence at that date, following the evolving trend towards more and more privacy in lordly apartments, but it boasted at least two ovens, mentioned in the accounts of 1369. The tower had a chapel, possibly first mentioned in the account of 1284, when repairs were made to the ‘glass window in the countess’s chapel, the glass window in the Great Chapel and the tower’. The first unambiguous reference to the chapel in the Great Tower comes in 1364, when Richard

the shape of the motte as it survives and with the otherwise rather unusual fact that there is no evidence that there was ever a moat, dry or wet, on the bailey side (cf. Aarts 1999, 21 n. 63). A separate stair ran from the bailey floor along the south-east curtain wall, where it was protected by the southeast tower. The south-east tower was especially important because it guarded the approach to the Great Tower, and perhaps also a postern. It overlooked the moat on the south of the motte and the Bydongate to the middle bailey and its water defences. It is this tower which is most likely to have housed ‘the chamber over the water’ (above, pp. 84– 5). Further storage areas would have been possible both at the foot of the forebuilding and under the wing-wall stair. Certainly, the heavy engines of the armoury (below, p. 97) cannot have been stored within the tower without having first been dismantled. There is a faint line on the geophysics on the northern edge of the motte which might be the foundation of the forebuilding, or of a courtyard containing it and the entry to the bailey level stores (see Figure 5.1). 93

the smith of Benson was paid 3s 6d for making an iron casement for the window in the chapel of the Great Tower; he earned a further 4s for an iron casement for a window in the Great Tower itself. Much more work was done in 1369 when the same Richard was paid a total of 21s 11d for the making of six iron casements for the windows of the chapel of the Great Tower, and making hooks and hinges for hanging the doors and windows in the Great Tower and the chapel in the Great Tower; he also repaired a further 20 old hooks and hinges, while Robert Smith of Wallingford made a vessel for founding lead, and ‘an instrument for piercing the walls in order to hang the hinges’. In the same year, 1369, John Plumber of Berkhamsted was paid 40s for roofing the chapel of the Great Tower and also a small turret on the latter. He and his mate earned a further 17s, at 9d a day for four weeks, for removing the old lead on the Great Tower and on the alure outside the Red Hall, and for re-covering the roofs and soldering their lead flashing. A clearer understanding of the chapel in the Tower is hard won by battling through the detailed but elliptical accounts of 1363, 1369 and 1375. Ironwork and glass in a window in the Tower chapel damaged by a storm were repaired in 1363. In 1369, Robert Brompton, glazier, provided 40ft (12.2m) of glass for two windows in the tower chapel for 43s 4d. Carriage of the glass from London cost 20d. In 1375 68ft (20.7m) of glass was used for a two-light window in the lord’s Great Chamber (i.e. Edward the Black Prince), with two oillets; three windows with three oillets in the Great Hall took 75ft (22.8m) of glass, as well as 40ft (12.2m) of glass for two windows of four lights and two oillets in the chapel and the chamber of the junior lord (i.e. the future Richard II). Hinges and other ironworks were provided for four windows in the lord’s chamber and chapel; a lock was provided for the chapel of the lord’s chamber. These last refer to the Great Chamber and associated chapel, but the figures for the chapel window glass suggest that the ‘chambers and chapel of the junior lord’ of the 1375 account were those in the Great Tower mentioned in 1369; the Great Tower is not explicitly mentioned in the 1375 account. Together these references to large glazed windows indicate that the most prestigious rooms in the whole castle were the Great Hall and its associated Great Chamber, and the Great Chamber and chapel in the Great Tower (Lloyd, pp. 121–4 and Figure 6.6). As part of the extensive works of 1363–5, master carpenter William Shalneston renewed the projecting mouldings (supercilium) on the Queen’s Chamber, and repaired the two inner bridges. He was also paid 30s to renew the mouldings of the great chamber called the Wardrobe, and for first taking down the mouldings against the outer walls of the Great Tower and raising a small crane or windlass against them for the convenience of masons working on them, as well as laying two floors where the same masons were repairing and renewing inner walls. Three men laid rubble stone, brushwood or straw over the new leading and masonry against the rains and frost of winter. In the following year William reaffixed the moulding to the outer walls, earning 6d a day for 4 days. 32 cartloads of timber, comprising 200 oaks, were brought in from Earley for these works. No less than 100 oak trees are said to have been

Figure 5.49 Evolution of hoardings (cf. Figure 5.34) to decorative gallery with solid machicolations seen at Berkeley Castle (photo: Katharine Keats-Rohan).

felled at Nettlebed for the mouldings of the Red Hall alone. These mouldings were otherwise known as hoardings; their original purpose was as easily mountable structures attached to corbels that enabled defenders to assail attackers below. They became semi-permanent decorative features in some castles (Figure 5.49). This, though, was still a castle with a defensive purpose, as the order for garrisoning in 1360 (1.48) and the vast amount spent on defence in mid-1399 shows (Biggs 2002). The amount spent in 1363–5, over £416, was the largest of all the accounts surveyed here, followed by the £206 spent in 1375 when further upgrades involving more window glass and the installation of now fashionable wainscotting were carried out. These extensive works were not the product of a major architectural change or a sign of gross neglect, they were primarily due to an act of God. The 1367 view of the accounts by Henry de Aldrington for work carried out between 1363 and 1366 records that £28 6s 8d was raised by the sale of 1700 beech trees thrown down at Watlington after a recent storm. The accounts themselves mention the pay of a man selling the wood at Watlington and two others acquiring some of it for the works. John Tyler of Wallingford and John Tyler of Roke and their crews repaired halls, chambers and other houses, and the glass, broken by the great wind (aulas 94

cameras & alios domos defractos [sic] cum vitro magno ventu) This was the infamous and widely chronicled storm known as the ‘great drowning of men’ (Grote Mandrenke) of 15–16 January 1362, one of the most destructive storms ever recorded in northern Europe, that resulted in at least 25,000 deaths.10 The tower on its eminence was particularly vulnerable – even today, standing on top of the bare motte in a strong wind is an uncomfortable experience. The collapse of the spire of Norwich cathedral is attributed to this event. The build quality of the Great Tower of Wallingford Castle emerges from this episode remarkably well. The Great Tower was re-rendered, cleaned and made fit for a visit by Richard II during 1389–90; the west gable of the Great Hall, other buildings, a south corner tower and a 188ft long repair to the East curtain wall were all rendered. A small crane or similar device – instrumentum – was made to assist with this rendering or resurfacing (cuiusdam instrumenti pro magn’ turr’ cum rowe mort’ iactando). At the same date the west gable of the Great Hall was rendered, and various internal walls replastered. These works involved lime plaster or gypsum, both durable and of fine quality, and white in colour. Some of the additives required to bind the plaster, such as sand, could discolour it, so such surfaces were often whitewashed – the Latin dealbo refers both to daub (plaster) and whitewash (Gapper 1998). We can infer from this that the render of the Great Tower, as well as other major buildings in the castle, would have been whitewashed, like its better known counterpart at the Tower of London. The accounts for 1389–90 are the only ones to refer to the Great Tower as the donjon, a French term that befitted a Francophile king. His child bride Isabella of France was living at Wallingford when Richard II was overthrown and imprisoned by Henry IV in 1399. The ‘great tower called le Donjon’ was to be cleaned against the king’s arrival. With the onset of winter, unused material was taken to the store-house (domus stauri); some ladders and timber were stored in the Muggetour (the north-east corner tower?) in the custody of the chaplain. Three locksmiths were bought in from Oxford to renew the locks, earning 6s 8d between them, and a woman was brought in to clean the hall and chambers (of the inner bailey), and to cut back nettles and weeds and remove them from the castle, for 8d. Two carpenters were hired to hang the bell in the tower over the inner gate and to make trestles for other jobs. Oil was bought to make the wheels of the wheelbarrows run more smoothly for those engaged in moving stone, sand and flints around the various sites.

as an arsenal until at least the reign of Richard II. It was his fortification and munitioning of Wallingford that led to the downfall of Henry VI’s constable, William de la Pole, in 1450. The most detailed evidence comes from Edward II’s reign, during which it certainly was an operating arsenal and one that was turned against him (above 62) in one of the most ironic episodes in English history. The reign had been characterized by baronial unrest arising out of the king’s reliance on a small coterie of favourites, and in early 1323 the castle had been taken over by the king’s enemies. Forces under Robert Damory were dispatched to besiege it and bring it back under the king’s control, as was swiftly achieved without a full siege being required. Robert remained there with a guard varying between seven and 23 men-at-arms for the next three weeks. The kingdom remained unsettled and audits of arms in several castles were ordered in 1325–26. The result has been described as a picture of decay and neglect by Prestwich (1992, 164), although the perishing of much that was held in castles stores, including foodstuffs, probably has as much to do with unheated stone or even timber buildings as with venal incompetence. Important background to the audits of 1325 and 1326, it harks back to 1307–9 when 5s were spent on the burnishing of arms in storage, and 1296, when the sum had been 6s (Figure 5.50). In 1307–9, 8d was paid for extracting caskets filled with quarrels, that is crossbow bolts, and stirrups for crossbows, as well as crossbows,

Arms and armour The care and storage of arms and armaments was always important in a royal castle and Wallingford certainly served   ‘..circa horam vesperam dicti diei, inceperunt tempestates horribiles, nusquam alias viae vel auditae, et ventorum turbines in Anglia, adeo quod domus et aedificia pro magna parte corruerunt ad terram, et quaedam alia discooperta deformiter per flatum ventorum hujusmodi remanserunt, arboresque fructiferae in gardinis et in locis aliis et arbores aliae in nemoribus et alibi existenes cum magno sonitu a terra radicitus evulsae fuerunt, ac si [dies] judicii adveniret.’ (J. Tait, ed., 1914, Chronica Johannis de Reading et Anonymi Cantuariensis, 1346–1367, 213). 10

Figure 5.50 Part of the accounts of John de Clinton from 1307– 9. (Reproduced by permission of the UK National Archives, TNA, E 101/489/30.)

95

Figure 5.51 Inventory of arms in the Great Tower at Wallingford by constable Miles de Beauchamp, after 25 January 1327. (Reproduced by permission of the UK National Archives, TNA, E 101/17/28.)

from the Great Tower for transfer to Knaresborough, another of the castles formerly held by Edmund of Cornwall that was given by Edward to Gaveston. Nine bushels of charcoal were purchased for rectifying twisted bows by heating. These were wooden bows for crossbows which were prone to warp, leading to the development of the composite bow, made from horn or whalebone, yew and tendon (Payne-Gallwey 1903, 62). Composite bows were also held in quantity at Wallingford, though are not mentioned in 1307–9. Crossbows and their bolts and lighter bows and their arrows were being made and fletched on site. A horse was hired for 4d to take the arbalaster to Abingdon with a writ from the king requesting the abbot to supply wood for making arrows and crossbow tillers. Also made at the same date were nine keys for crossbows, seven belts for bending them, and seven triggers, while seven foot stirrups were cleaned. Six hundred goose feathers were bought in 1307–8, and 700 in 1308–9, when 3000 quarrels were fletched or bevelled; two men earned 9d working for three days placing quarrels in buckets. Three tuns were purchased from which a cooper made 18 buckets for sheaves of quarrels. One steel mallet and two steel wedges for trussing the bolts to the stock were made. The arbalaster and his mate earned 5d a day for 105 days. This pell-mell list indicates a range of weapons, from the ‘sporting’ crossbow used for hunting, for which the bolts

or arrows would be fletched with goose feathers, and the larger crossbow or balista which required a stirrup for one or both of the bowman’s feet; the bolts for these larger bows were often ‘fletched’ with thin sheets of metal, or were metal tipped. The larger crossbows, which could be very large indeed and took their place alongside the siege engines of the day, were for military use. A separate section in these accounts indicates that the armaments did indeed include these larger balistas, since a plumber soldered lead over them (ultra balistas), which probably refers to something like the weighting of the counterpoise of ‘engines’ such as a trebuchet (Pounds 1990, 127). The audits of armour on 20 January 1326 and January 1327, ordered respectively before and after the fall of Edward II, show the full range of weaponry in the castle store and give a more accurate idea of numbers than annual building accounts, which mention only items on which work was done during the period of the accounts. The audits were part of surveys ordered of provisions and armaments in a number of castle stores, including also the Tower of London, Windsor and Odiham. The first was written in French and the return for Wallingford was made by Ralph Restwald, under-constable to William Marshal (TNA, E101/17/11). It is a terse account, in short entries with marginal headings. It is rubbed and of variable legibility (much of the accompanying membrane, which accounts for victuals, is even harder to read). 96

Although different in layout and content to the second, slightly later account, returned by the constable Miles de Bellocampo, the two recognizably discuss the same arms store (TNA, E101/17/28). In the second, written in Latin, the armaments are explicitly stated to be kept in two (at least) rooms in the (Great) Tower of the castle (Figure 5.51). The bottom lines of the membrane are rubbed and difficult to read. Inevitably, these lines contain matter of the greatest interest, much of it apparently missing in the earlier account, but we can still make out the salient points. Both audits describe stores of body armour — hauberks, haubergerons (short mail shirt), bassinets, cuisses, mailed gauntlets, armour for horses, much of it in poor condition, including 40 shields bearing the arms of the king (Edward II) and Piers Gaveston. Both mention two Irish axes. Miles’ account includes 16 lances for war (pro guerra), and five for jousting or other lance-games (pro hastiludio). The first lists 35 one-foot wooden crossbows in various conditions, and five of horn, with 32 two-foot crossbows of horn, and tuns full of great and smaller bolts. The size of the quarrels depended upon the type of crossbow for which they were used. Both accounts mention the windlasses and cords of the larger crossbows, including those that were probably the catapult-type of balista, a serious war-machine (Figure 5.52). Both mention another such machine, a springald, a tension-driven device similar to a balista, though reference to a trebuchet is only legible in Miles’ account, which lists two great iron kevels for the engine (ingenio) and

two for the springald and trebuchet. He refers several times to ‘engines’, including 32 pieces of timber wood from which one was to be made (or possibly repaired) (xxxii pec’ maerem’ pro uno ingenio et uno springald). It is very difficult to see how these massive siege-engines could have been stored in the Great Tower (Figure 5.53). Other castles, such as Dover, built special store-houses for these machines. That may have been the case here also – a domus stauri where ladders were stored over winter is mentioned in 1389; in 1411 John Smyth of Wallingford worked on fixings for a timber pile in the Great Wardrobe, which at that date could refer to a specific room for holding bulky stores. At the Tower of London the wardrobe was specifically associated with armaments and, as normal when an individual castle store was meant, became known as the garderoba armorum. Indeed, a national system of arms stores in various castle wardrobes evolved by the mid-13th century into the institution known as the Great Wardrobe, of which the Tower became the HQ only during the 14th century (Storey 1998, 176–80; cf. Tout 1911; Tout 1928, iv). Though Miles’ account specifies a camera infra Turrim eiusdem, there is no clear sense of a division of material between one or more chambers, apart from the fact that the majority of personal arms are mentioned before the ‘other chamber’. What is striking is that in the midst of the casks of quarrels and arrows, the crossbows and the huge engines, comes the incongruous mention of the

Figure 5.53 Trebuchet at Warwick Castle (photo: Judy Dewey).

Figure 5.52 Balista at Warwick Castle (photo: Katharine Keats-Rohan).

97

at a fixed rate for the job (ad tascham), or by a daily wage fixed in advance. Almost all were men, apart from the cleaner in 1390 (plus ça change), though women might figure amongst the suppliers of materials. In 1389 2d was allowed for the carriage of lead from the house of Mariote Litle to the castle. There is a clear difference before and after 1375; before that date, only the master craftsmen are mentioned by name, though the wages of his assistants and labourers are itemized; afterwards, the accounts are mainly in the form of a list of names under headings such as carpenters, masons, labourers etc., followed by a brief description of the daily rate and total amount paid, sometimes with a brief indication of the work involved (HKW ii, 976). A hybrid of the two types would have made the business of interpreting these documents very much easier, but the amount of light they shed on this long-lost building demands gratitude simply that they survive at all.11 We learn a great deal about the sourcing of materials from the building accounts. Almost all of it was sourced from within the resources of the honour of Wallingford, supplemented by the extensive additional estates that made up the bailiwick of Wallingford in Benson and the Chiltern Hundreds, and Henley, part of the earldom or dukedom of Cornwall, which also included the honour of St Valery in Oxfordshire (Figure 5.54). Occasional grants of wood from the royal forests were made for the benefit of the sheriff in the execution of his duties, notably the building and maintenance of prisons, as in 1225, when wood from the royal forest of Shotover, near Oxford, was supplied for this purpose. Timber from Pamber, part of Windsor Great Forest, is frequently mentioned in repairs of the main castle buildings or facilities ordered by the king, as in 1307 and 1389. Much of it was even more locally sourced, like many of the workers and the overseers of the works, who were recruited from local people, whether leading townsmen or officials of the honour. Twelve beams were bought in Wallingford in 1331 for the building of a watch tower at Bydongate and the renewal middle and outer bridges, as well as the portcullis of the outer, Constable’s, gate. A further five timbers were bought from John Mariot of Wallingford for repairs to the inner bridge. He had been one of the overseers of the 1307 accounts of John de Clinton, and was a former town reeve, during 1299–1300, and 1303–4, and several-times mayor, his longest stint falling between 1304 and 1314 (Herbert 1971, App. pp. xiv-v). In 1307 beeches for mill enclosures came from Wyfold, a manor attached to Checkendon, in Benson Hundred; carpenters cut wood at Whitchurch. In 1308 withies were bought at Bray; three horses were provided (in rotation) for regular trips by the master carpenter to select six trees for felling in the forest of Pamber to be used in repairing the king’s mills; four carpenters there shaped the selected timber, which was then taken to the home in

three swords of those heroes of romance, (probably) King Richard, (Sir) Gawayn and Guy of Warwick. As we have seen (above, 75), there is independent evidence that these were on display in the Great Tower at Wallingford. Perhaps they were kept in the ‘plate room’ (plata camera), which may have been the place where the personal arms of the earl of Cornwall were kept since on its first occurrence in the 1284–85 account it is platam cameram domini comitis. This chamber is mentioned again in the 1296–97 account (Midgley 1942, 134), which recorded casting lead on (the roof of) the plate room. Clearly, Wallingford was a major arms depot and manufacturer. Pounds’ discussion of how castle warfare was organized and supplied is, of course, lacking mention of Wallingford, but the evidence slots into what we can now say is its usual place, working in conjunction with the Tower of London and Windsor to form a supply chain along the Thames and from Wallingford across country in all direction to other such bases (Pounds 1990, 106– 113), as can be seen from audits of arms and the itinerant attiliatores whom the king sent round his castles to make specialist repairs to weaponry (1.34, 37–8). On many occasions Wallingford Castle can be seen as part of royal calculations for military security. In December 1307 Edward II ordered the securing of several castles against his going overseas. They are named in two lists, the first of which reads: Rochester, London, Windsor, Wallingford, Corfe, Isle of Wight (Carisbrooke). In 1317, when the castle was part of his wife Isabella’s dower, he ordered the constable ‘to put 40 fencible men, crossbowmen and others, in that castle at the king’s wages, for the safe custody thereof until further notice’ (1.33). Occasionally plans backfired, as they shortly afterwards did for Edward II. In 1399, when for a short time, about 12 July to 19 August, the (relatively) well-maintained and formidable castle was briefly the centre of government for those loyal to the doomed Richard II, whose childbride was resident there, no less than £1296 12s 4d was disbursed for its defence on 12 July 1399 (Biggs 2002, 130). By 29 September it was all over and Richard was coerced into abdicating. In 1450 the parliamentary speaker of the Commons appeared before the Lords expostulating against the treason of William de la Pole, duke of Suffolk, to whom Henry VI had entrusted the castle, only to find that ‘he hath fortified it and repaired it, and also stuffed it with gunnes, gunepowder and other habilimentez of werre, and with sufficiant vitaill, as it is said, to th’intent that if this puyssance [France] come into this realme, they may have there a place of refute and of socour’ (Rot. Parl. iv, 177a). At the speaker’s request, the duke was committed to the Tower of London. His luckless master Henry VI died a prisoner in the Tower in 1471, probably murdered on his successor Edward IV’s orders. Socio-economics: sourcing of materials, workers and wages

  As the work on this paper was drawing to a close high-quality images of many of the accounts used herein began to appear on the website of the Anglo-American Legal Tradition project, where they may be accessed under the heading Exchequer: Various Accounts: Works, at http://aalt. law.uh.edu/E101nos458_504.htm.

11

Of considerable interest in the various building accounts are the insights afforded into the rates of pay for different types of artisan or labourer, who were usually paid either 98

Figure 5.54 Places in the honour of Wallingford most frequently mentioned as source for building materials (drawn by June Strong).

the forest of William Woodward who turned it into laths and boards. The carpenters made the rafters for the vaulted roof of the mills from timber in the park at Watlington. Another carpenter worked for one day stripping out the old wood in the mills. Timber for the mill wheels came from Queen Wood (Queneholt) in Watlington. The same works required timber for the wheel and cogs of the south mill, where the old roof was stripped and new tiles, including corner tiles, laid. Repairs, using clay and earth, were made to the three mill-dams, as was an enclosure behind the East mill damaged by water. A mill leat serving the castle mill near the East Gate was widened to improve flow to the mill wheel; 200 beeches from Wyfold were chosen for making of the great enclosure and upper enclosure of the east mill; and two carpenters made chevrons (ridged timbers for rafters) at Whitchurch. In the same year 135 beechwood timbers from Swyncombe were used for hanging bells in the belfry. Straw was bought at Reading to protect tiles etc. being transported to Wallingford. Wooden tiles were bought from a (saw)mill in Reading: the earliest medieval sawmills are thought to date from c. 1280. In 1363 tiles of various sorts were sourced from Nettlebed (Salzman 1952, 230: Mellor 1994, 33, 80, 149, 194, 201) and the park of Watlington; 100 oaks from Nettlebed were used for the timber mouldings of the Red Hall, and 200 oaks from Earley, near Reading, for the Great Tower and other buildings. The extensive works of 1389–90, sourced from timber at Pamber chosen by Master Carpenter Fabian have already been mentioned. The resources of Nettlebed, notably timber and tiles, were especially important.

Thomas Hatfield, keeper (custos) of Nettlebed in 1383 was paid 2d per day (TNA, SC6/1096/7) Stone sources were also varied. In 1363 nine men collected flint for three days, requiring 10 carts to carry them all to Wallingford. 133ft (10m) of ragstone (very hard limestone, often laid as uncoursed work) were quarried at Wheatley. In 1375 there was quarrying at Clapcot for small stones to be made into mortar (a frequent occurrence in the accounts), and 460ft (140m) of freestone (stone fine enough to worked with a chisel) and ragstone were bought from Nicholas Harald at Wheatley for quoins and ashlars, and a further 20ft (6.1m) of gabel or gable stone (a limestone of suitable quality for the gables). 26s 3d was paid (presumably to Harald) for the cutting of 212ft (64.6m) of freestone. 140 carts sourced in divers vills were required to carry the stone from Wheatley to Wallingford, at a total cost of £7. Between 1363–5, 212 cartloads of Chiltern flints were taken to the castle, 414 cartloads of freestone were brought from Wheatley; 132 cartloads of timber from Earley, 115 barrels of lime were brought from Nettlebed, and 26,000 tiles. In the final year of these works no less than 288 carts of freestone were brought from Wheatley. In 1386 flint was sourced from the Chilterns, as it had been in 1308. Nearer the time of the Conquest, fine stone may have been acquired from the quarry at Taynton, Oxon, (mentioned in Domesday Book), found in several churches of this date in the vicinity of Wallingford (Secker 2006); for higher-end needs, stone was probably imported from Caen, but no direct evidence of this is known. In 1438, when a substantial new prison house was 99

built, 128 cartloads of malmstone was purchased from John Warfold and bought from the quarry in Clapcot to the castle. Subsequently four labourers worked for 62 days on malmstone taken from ditches outside the castle, producing corner tiles, crest tiles and quoins, and then carting them to where they were needed. The latter material was probably the local chalk known as clunch, an easily worked but not particularly fine material. The Clapcot quarry occurs in the 1389 accounts, when fallen stone from tumbled walls was gathered up and recycled in time-honoured castle fashion. It was probably located some way north of the castle, in the region of what is now called Shillingford Hill (overlooking the Thames near Shillingford bridge), where outcrops of Glauconitic Marl occur (cf. Christie and Creighton 2013, 20–22, 238–41), easily recognizable in a ‘particular’ of tenants in the manors of Clapcot and Rush Court of c. 1663 that refers to the ‘East Field malme pits and Windmill Hill Furlong’ (BRO, D/ER E38/1). Furnaces and kilns were usually built close to the site of major operations. Furnaces for casting lead or alloys for soldering guttering are frequently mentioned. A new furnace has its own section in the 1307 account, requiring 21 cartloads of flint; a tree had to be felled and its roots removed for the building of the furnace, which also had to be desiccated by the laying of straw and its hearth by laying down brushwood. For the 1363 to 1366 works, a lime kiln named Le Ode was built in the park at Watlington; it was used again in 1375. This was probably used for the production of lime plaster, a process that could be a nuisance in terms of dirt and smells, hence its location at some remove from the castle (Gapper 1998). John Tyler built the kiln; for that and for burning three hundredweight of lime he was paid £3. John Drayton was paid 8s 8d for squaring 103ft (31.4m) of ragstone at Wheatley at 1d per foot. A great deal of smithing was carried out on site, for the repair of iron fittings of varying sorts such as window casements, iron bars, locks, keys and so on. Smiths and sometimes masons were also paid to repair the tools used by other workers, or, with carpenters, to build the tools required, such as scaffolding and laddering. Some tools were purchased fairly frequently, such as sieves, buckets, picks, and wheelbarrows. Most of the goods sourced externally were brought in by cart, such as lead from Oxford Castle’s stores in 1307. Some was brought by water, as when stone was needed for the repair of Dernegate, which was near the Thames, in the same year. In fact, this is an isolated reference to local water transport in the available accounts, though presumably the Thames was used more often for local materials coming from the Oxford area. Some of the high-end imports of the later period travelled considerable distances, because, for example, medieval lead could only be found in a few places like Cornwall and Derbyshire – all a long way from Wallingford. In 1375 three fothers of lead were purchased in London from Robert Beauchamp, then carted from Billingsgate to Baynards Castle, from there to Henley by water and thence overland to Wallingford. In 1369–70 40ft (12.2m) of glass for the chapel in the Great Tower was bought in London. Like the estrichbord, a high quality wood panelling material imported from Scandinavia

or Estonia, it was also transported from London, at Queenshithe, by water to Henley and then overland to Wallingford. The importance of Henley as an entrepôt at this date is now well evidenced (Tiller and Darkes 2011). In 1375 estrichbord and other speciality wood, some of it for a windbarge for the window in the lord’s Great Chamber was also bought in London and transported the same way. With the odd exception of gifts from the king’s woods at Pamber or Shotover, all the places mentioned lay within the earldom of Cornwall, mostly within the extended honour of Wallingford. Queenshithe became part of the honour as a gift of Queen Isabella to her son Richard, earl of Cornwall. The later accounts name all the various artisans and labourers and record their wage rates and pay. Some have already been mentioned. Among the names are those of men who were clearly local, or from other parts of the honour in Buckinghamshire or Oxfordshire, some of whom recur frequently, like Richard the mason of Henley, so were obviously good at their jobs. Robert the smith of Wallingford occurs several times, once repairing the tools of masons working on spiral staircases in 1363. Thomas Carpenter of Warborough and John of Warborough, a quarryman, were both named in 1363–66, as were Richard Couper of Dorchester, and labourers Jocelyn Crowmarsh and Ivan Goring. The Black Prince retained master carpenter Richard of Wallingford to carry out all his carpentry works, and even endowed him with authority to impress workmen not otherwise employed on the king or queen’s works, for several years from 1351 (HKW i, 471). Some probably depended upon the work. In separate accounts for 1408 and 1411 the same names occur. There are also signs of labourers who had travelled some distance for the work in their chosen speciality. Two Welsh names stand out in 1363, John ab Zenan and David ab Grefeth, and probably a third, John Walisse, all earning 3d a day, John for 24 weeks and the others for nine weeks and three days. Another example might be William North, also a labourer at 3d a day. The master mason on this building campaign was Robert Yvele, brother of the king’s master mason Henry (HKW i, 209–10). Henry Yvele was one of the foremost master masons of his day, being appointed ‘disposer of the king’s works pertaining to the art of masonry at the Palace of Westminster and the Tower of London’ in 1360 (ibid., 177), a position renewed and enhanced under Richard II. Robert Yvele constructed the vaulting of the Bloody Tower gateway, with its unusual bosses in the form of lions’ masks, during 1360–62, when he was warden of the Tower masons (ibid., 726). John Plumber of Berkhamsted was engaged for roofing Wallingford’s Great Tower, its turret and chapel in 1369– 70, for which he was paid 40s. At the same date, John Tyler who worked for seven weeks and a day tiling the Great Hall, the Queen’s Hall and the Great Kitchen, at 5d a day, earned in all 22s 11d. Stone layers worked for 5d a day and labourers for 3d. The work was seasonal, since major external building works once autumn set in were stopped on account of frost damage. There are many references to buying in straw or brushwood to cover partly repaired or recently rendered walls and roofs. Rain was also a problem. In the hugely 100

ambitious and expensive 1389–90 renovation campaign it appears that some tasks were reassigned on rainy days to prevent loss of time over a short and concentrated season. There are plenty of references to materials and tools; sieves, shovels, picks, and wheelbarrows (on which see Matthies 1991) are just a few of many. Storms appear to have been a considerable problem, as we have already seen in relation to glass windows at both Wallingford and Windsor Castles, where storm and wind damage occurred several times during the minority of Henry III. The hugely destructive storm of 15–16 January 1362 has been shown to have occasioned much of the extensive repairs carried out between then and 1369. The castle buildings accounts are richly informative about materials and wages, but other types of accounts can also shed valuable light of the socio-economic background to castle life, from the humdrum of day-to-day living to wider trends in the environment. As outlined above, the castle was part of a great estate, partly self-sufficient in growing fruits and herbs, and also in fish (including elvers), of which enough was available to supply the castle staff, to send to the king, and for any surplus to be sold. The castle had its own vineyard and its own orchards. (Figure 5.55) Enough was produced to be sent elsewhere, as in 1284–85, when the pears favoured by the countess of Cornwall were sent to her at Eltham, and nuts and figs were sent to the earl at Berkhamsted. In 1306–7 apples raised 12d but there were no surplus cherries. In 1315– 16, Edmund Bacun accounted £16 for the profits of the four water mills, 50s for the fisheries, 3s for apples and cherries, 10s for the sale of herbage from the meadow at the north mill. £8 2s 6d was raised from the sale of hay from 32½ acres of meadow. In the following year profit from the mills was £8 and 25s from the fisheries. At a cost of 30s a horse had had to be purchased for carrying corn to the mills; a further 8s was expended on timber for

strengthening the mill enclosures and the mill houses. In 1363 74s 10d was received from the keeper of the fee of sheep provender (TNA, E101/545/27). Annual allowances were made out of these profits to the Dean of the Chapel for the sanctuary light, and grants in perpetual alms to the prioress of Goring, and Dorchester Abbey. After 1317 references to fruit produce from the castle cease, although the name Vineyard remained attached to the formerly fruitful areas. This was the period when the beginnings of climate change, manifested as a seven-year run of cold winters and wet summers starting in 1315, led to the devastating great famine of 1315–17. The first sign in surviving Ministers’ Accounts comes in 1316–17, when it was reported that no money was raised by sale of hay that year because part of the meadows had been flooded by the Thames (TNA, SC6/1096/3), although flooding in these meadows would not have been too rare an event. Nonetheless, herbage on Northeyte, the island of meadow attached to the north mill, yielded 10s. In good years, such as 1384–86, herbage from the Northeyte was worth 20s and from Queen’s Arbour 23s (TNA, SC6/1096/7). The revenues from the leasing out of the elements of this ‘gainery’ or farm continued to be important into the 17th century, attested by Ministers and Receivers’ accounts, the Wallingford Roll (discussed in Lloyd below), and various later leases (Dewey, below). Cartingsilver (above, 77) continues to appear in Tudor accounts, but it thereafter disappears. Indeed, these documents reflect the way that the use of both castle and meadows changed considerably following the dissolution of monastic institutions and the move away from castle-based residence during the 16th century, with milling becoming the dominant economic activity. The men who took up the leases during the 16th century either were or became the ancestors of the leading families of the town thenceforth, many of them still local to this day.

Figure 5.55 Looking through a window arch in the Priests’ Lodging to the orchard beyond (photo: Katharine Keats-Rohan).

101

There are no known building accounts surviving for Wallingford Castle after 1411, apart from the rebuilding of the prison house in 1438–40, though the castle continued in intermittent use as a residence, and more regularly as a prison. Surviving Ministers’ Accounts from 1307 to 1540 show that routine maintenance was carried out until the 16th century as part of the estate of the heir to the throne; a warrant for repairs by Arthur, Prince of Wales (1490–99), is still extant. Female tenants played their part: Joan of Kent instigated work at the castle at her own cost in 1378, though no accounts survive (1.58). The 1411 accounts have the distinction of having been presented by John Coterell, a crown servant who served as an alderman of the borough and who was elected MP for Wallingford from 1393–97 and again in 1410. He was a tax collector in Berkshire and, more importantly, served as receiver of the honour of Wallingford for at least 17 years. In all, Coterell served as a royal official in Wallingford for 30 consecutive years, during which time he sat in the Commons nine times, and was also appointed to royal commissions of inquiry (Kightly 1993). He was controller and surveyor of the 1389–90 accounts, in which he also occurs as a supplier of chalk and other building materials for the works. Three other men paid for work or supplies in the 1411 accounts also represented the town in parliament, elected by the burgesses, Richard Horlock, who was paid, with others, 5s in the same account for the provision of wood panels, John Culham, stone cutter, paid 18d for work in the Great Chapel, and John Derby who furnished 4000 nails amongst other things worth about 23s (Kightly 1993; Jeavons 2013). Other names found in contemporary borough records that occur in the accounts of 1408 or 1411 are John Pope (carter, 1d each for six cartloads of sand), John Best (carter), and John Hunt, apparently a labourer who in 1408 dug sand and cleared out the ditch around the Great Stable earning 6d over two days, and felled wood for a sum now illegible in 1411. A routine annual account of Hugh, a serjeant of the honour of Wallingford and a townsman, from 18 Henry III (1244) is full of interest in its own right, showing that, as we should expect, sourcing from local tradesmen and craftsmen was routine; this includes the hire of local ploughmen, labourers for work on the dovecote, the hire of Edward the carpenter on a mill repair, and the purchase of a mill stone from alderman Richard Blawe, town reeve in 1228–29 and 1233–34. Richard was a vintner, whose trade was the most dominant in the town of the time (Herbert 1971, 109), most of the vintners being large propertyowners in the borough. Richard was assessed at 20s in the tallage of 1235–36, one of the highest sums (ibid.). The serjeant’s roll shows that 30s were paid out to Richard for various services. Clement the clerk was paid 50s for unspecified but clearly important services. Clement was a town reeve around 1225 and an alderman in 1247; he also served as coroner, and mayor c. 1250 (Herbert 1971, App 3, pp. xxvi, xvii, xxi, xxvi; BRO, RTb 13). That labourers and suppliers should be linked with the town, albeit not exclusively, is exactly what we should expect. Throughout the period surveyed here the names associated with the presentation of the accounts have

shown links with both honour and town. Some of the later aldermen and reeves like Gilbert de Mandeville, also known as Le Messagier, who was one of the overseers of the 1307–8 accounts, were members of junior lines of the tenant families of the honour. John de Clinton, the steward and constable who presented the same accounts was likewise a member of such a family, and William le Marshal, the constable appointed by Edward II and later re-instated by Edward III, was another, far lower in status than his famous namesake who led the first minority government of Henry III. As time went on and the role of castles changed, becoming less important militarily and less used as residences, so too the ill-defined but wideranging role of the constable changed. In the later period, from the end of the 14th century onward, he was no longer an officer appointed at pleasure for usually only a few years, he was someone of higher status who had worked closely with king or lord and expected to delegate much of his duties to deputies (Bond 1967). Similarly, men who commended themselves to the service of the king or the Black Prince could be rewarded by a post relating to the castle honour and subsequently become embedded in the local gentry, such as Sir William de Bereford, who acquired considerable property in Clapcot relating to castle tenures. Apart from maintenance and military functions, the townspeople also furnished victuals and high-end goods to the castle tenants. Royal interest in the town and its market is evident in Pipe Roll entries such as that of 1207 (PR 9 John, 186), when the king’s writ paid 8s 6d for repairs to the market stalls; a number of entries in the borough rolls show three visits by Henry III to the serjeants of the market during 1232, 1235 and 1236, and one in 1252 (RCHM 1877, 579–80).12 The garrison was only present for short periods at times of stress, and lords such as earls and dukes of Cornwall were never resident for long at any one time, so the primary users of services were the resident clergy and the household of the queens and dowagers for whom it became a principal residence. The importance of this relationship begins with the first signs of the town’s decline from its 11th-century apogee, when constable Brien fitzCount successfully pleaded poverty on the burgesses’ behalf in the Pipe Roll account for 1129–30. They proved chronically unable to raise the fee-farm of £80 demanded by the Exchequer. It was eventually reduced to £42, but even this proved too much. Failure was constant and on more than one occasion under King John and Henry III, the town was taken over by the sheriff – who had been excluded under the very generous terms of the borough   RHCM 1877, 579: ‘At the close of this roll there are eight lines in Latin, of which the following is a translation: Delivered by Simon Gurmund, of the rent of Wallingeford, at the first visit of the Lord the King, on the day of St. Denis, to the Serjeants of the Market, in hay and oats 11d. At the second visit of the Lord the King, for the feast of St. Edmund, to the Serjeants of the market, in hay and oats, 11d.; parchment 1d. To Henry the Clerk, for his service, at Christmas Term, 15d. To the Prebendary (prebennario) 7s. 81s 4d. At the visit of the Lord the King before Our Lord’s Ascension, to the Serjeant of the market, in hay and oats 10d. To Henry, Clerk of St. Mary’s 15d. To Alexander Dublet, for his journey to London 18d. At the last visit of the Lord the King, before the feast of St. Michael, to the serjeants of the market 6d. To the Prebendary, for the Term of St. Michael 7s. 8d. For the expenses of the . . . of the prebendary, Ralph Culebule 4d. To William Pret, for his horse for London 8d.’ 12

102

charter granted in 1155 by Henry II. A very eloquent statement comes from the Black Prince’s Register, in an order to his purveyor dated June 4 1351: Order to the prince’s purveyors and request to all others – inasmuch as the maintenance of the estate and livelihood of the burgesses and tenants of Wallingford depend for the most part on trade, and if they are in any way disturbed in such trade they cannot pay the prince his fee-farm or do what they ought and are bound to do for him in other respects, and moreover, unless they are reasonably treated and supported, there is reason to fear that owing to their poverty the said town, which is already partly decayed, will shortly be quite ruined, to the great damage of the king and the prince – to take no crops, beasts or other victuals or wares, wherever they may be found, whether in fairs or markets, or on land or sea, from John atte Russhe or any other burgesses and tenants of the town, and to refrain from interfering with them or the carriage of their wares, during the prince’s pleasure. (REBP iv, 17). Still the defaults continued. Joan of Kent, who lived at Wallingford for many years in the mid-14th century, successfully interceded on the town’s behalf with her son Richard II in 1383, earning them a reduction in their fee farm from £42 to £20 for seven years (CPR 1382–5, 449). By this point the town was in crisis. Apart from the effects felt in all communities countrywide of the adverse change in climate conditions and associated population decline from the time of the great famine of 1316–17, it suffered badly from successive waves of plague, which had both depopulated the town and deterred newcomers. The market for luxury goods produced by such as the town’s goldsmiths was contracting. A further plea for relief to Richard II in 1395 is eloquent; trade was almost at a standstill following the death of Queen Anne of Bohemia, the king’s first wife, who with her retinue at the castle had provided a market and a stream of work (CPR 1391–6, 720). By 1438 only four of the 12 parishes remained in use, and two of them were so depopulated as to be unable to pay their tithes. Only 44 householders remained, as was determined by an inquest held by constable William, earl of Suffolk (CPR 1436–41, 317–18). The difficulties are also evident in the way that formerly specialist merchants and artisans were having to diversify. John Derby, MP in 1386 and 1388, and Richard Horlock, MP in 1391, both traded in grocery, fish, meat and ironmongery (Kightly 1996). There were other factors at work, chief among them being the ruinous state of the great bridge, which had led to the diversion of the route to the wool towns of Gloucestershire from Wallingford to the new bridge at Abingdon from 1416. Wallingford itself had been a contributor to this trade as well as a conduit, with sheep flocks maintained in Clapcot and elsewhere, and weavers and blanket makers amongst its tradesmen in the 13th century (Herbert 1971, 119). The great bridge was itself a perennial problem for the townsmen. Two bridge-wardens responsible for maintaining the bridge are traceable from 1258, assisted by revenues from certain tenements in the town, occasionally supplemented by grants of other revenues, such as the proceeds of fines

levied in the borough court (Herbert 1971, 10–11). In 1344 Edward III had made a grant of 12 years-worth of pontage to the burgesses for the repair of the bridge ‘which threatens to become a ruin’ (CPR 1443–45, 421); a grant of pontage for two years was made in 1429, and again in 1433 for five years (1.66). The bridge was of major importance to the castle, since the East Gate of the town at the foot of the bridge and the main gate of the castle were closely linked. The bridge carried the main west-east highway out of the town on the road leading direct to Henley and on to London. The over-reliance on the castle for Wallingford’s economy emerges clearly. Contraction of towns in the later middle ages is a well-attested phenomenon, with recurrent plague being one factor common to many cases, as well as the prolonged economic downturn of the mid-15th century. Other castle towns felt the same breeze as Wallingford, including Winchester, which began to decline at the end of the 13th century. But Wallingford had no bishopric or university or vital access to export markets on which to fall back. The increasing dominance of its parliamentary representation by royal servants and protégés of the constables, themselves now royal protégés recruited from outside the old honorial families, has been seen as a factor in Wallingford’s rapid economic decline – between 1422 and 1450 11 out of the 18 elected Members were nonresidents (Kightly 1993). Nonetheless, one might wonder whether the fact that Wallingford survived at all was due to its royal status and usefulness in providing sinecures for royal favourites, a trend that continued throughout the 15th and 16th centuries. The castle would never be important again as either a stronghold or a royal residence, apart from the brief interlude of the 1640s, so the town was forced to look for alternative means of survival. Somehow, it succeeded. But the legacy of the centuries of Wallingford’s identity as and with a stronghold, from its foundation in the ninth century, still abides in the modern town, which uniquely preserves a medieval townscape that includes Saxon ramparts and castle earthworks, making it attractive to archaeologists and historians. It is also vital to the future prosperity of a small town that needs to attract tourists in order to flourish. Safeguarding its heritage – the Alfredian ramparts and the castle earthworks – is for today’s residents more important than ever. After all the picking through the complex and often frustratingly vague detail of these accounts, it is good to end this section on a human note. The 1389–90 account is the only surviving one for what was a two-year campaign, and it is rich in detail. It does not end on the usual dry note of accounting for the cost of the parchment on which it was written, though it does mention purchase of parchment and paper for making records of the amounts being expended over the year. It describes repairs to the wheelbarrows that were used to take ready mixed mortar to where it was needed, and to the building of a sort of cement mixer. They also mixed resin and wax to form a binding agent with crushed stone to patch up the small tower on the south of the castle. Towards the end, it accounts 3d for pots of ale provided to those who erected the device for casting mortar 103

Figure 5.56 Entrance to the top of the keep at Farnham Castle; the roof visible through the archway at the top of what is now a shell-keep covers the truncated remains of a square stone tower that originally rose through the centre of the motte and was designed from the start to be enmotted (Thompson, 1960) (photo: Katharine Keats-Rohan).

Figure 5.57 Diagram of the construction of the motte at Farnham after Thompson 1960 (illustration by Debbie MilesWilliams).

Figure 5.58 James Blackstone’s sketch of part of an underground passage on the south side of the motte as shown in Hedges’ History of Wallingford (1881), i, 151.

104

(rendering) on the Great Tower and cleaned one gutter, and 10d for pots of ale and nuncheon (Noenshynches), a noonday meal provided for the stone masons. Nor is it the sole surviving glimpse of human relationships between great and small: the surviving original writ by which Richard II bestowed a life pension on his Wallingford laundress Emma Stokkere (TNA, SC6/1282/12, no. 81) is just one example from a mass of still unedited documents relating to the king and his residences. Especially rich are the Wardrobe and Household accounts, which detail all the clothing of the royal household and their possessions, such as the suit of red velvet made for the three-year-old earl of Chester in 1333, and the gold-painted cradle made for his infant sister Isabella. An indenture for Wardrobe accounts for the first half of 1334 includes items commissioned by Edward III for seasonal games at Christmas and New Year 1333–4, spent at Wallingford Castle. Among them are 14 hobby-horses. Kings and courtiers could occasionally let their hair down without requiring the services of an executioner! Summary and conclusions What has preceded has been an attempt at establishing the stages of Wallingford Castle’s architectural and institutional evolution in the first 400 years of its existence. The process has been lengthy because with a largely nonextant building we have had to proceed by analogy, as well as by employing a number of diverse and rich but very challenging unpublished documents. Most of the evidence from analogy is archaeologically based; archaeological evidence relating to the castle, recently summarized by Christie and Creighton (2013, and above), has also been used. The primary emphasis of the analysis, however, has been based on documents. The interpretation of documents, especially those written in Latin, and heavily abbreviated Latin into the bargain, is no less challenging, and hence open to challenge, than the evidence of archaeology. Incontrovertible, though, is that in the case of Wallingford some long overlooked written evidence alone holds the keys to the castle. What follows now is a summary of the phasing of the castle’s development from 1066 to 1300 as outlined in detail above, and some concluding remarks. Broadly, there is agreement with the phasing recently suggested by Christie and Creighton (2013, 218), but the use of medieval documents has permitted the addition of significant detail. First phase, 1066–1113 Wallingford Castle was one of the urban castles of conquest deliberately planted in the key administrative centres of each English county by William the Conqueror. They were all built early in the reign, as were the castles defending the English coastline and the western marches. An emphasis on the Thames Valley and the late Old English royal centres of Westminster/London and Winchester is also apparent. By 1071 castles had been or were being built in London, Winchester, Wallingford (already active as a royal prison), Windsor and Oxford. Wallingford’s importance from the end of the ninth century was as a ‘stronghold on the upper reaches of the Thames, astride the main lines

of communication in the south of England’ (Warren 1973, 24). All these castles would remain important until at least the end of the 13th century. Most of the urban royal castles slotted into existing urban fortifications. At Wallingford it used the northern rampart of the Saxon burh, situated in the north-east quadrant of the town, an area that gives every indication of having been an élite site associated with a royal hall and a ‘garrison’ of the armed retainers of both Edward the Confessor and Harold Godwinson, earl of Wessex and king of England. As such, it may have contained buildings, including a church, that could be incorporated or adapted into the new castle, or anyway affect the layout of the new castle. The town played an important role in the Conquest by being the place where the first negotiations for the submission of the English to William took place. The creation of a Norman urban castle will have been almost immediate, and easily achieved because the site was already established as a military and administrative centre, separate from but closely linked with the urban and agricultural functions of the rest of the burh. This site may have been far more extensive than was usually the case. The castle in its final extent certainly encroached on the town on the west side, but for the rest, its enceinte seems to have followed what was described as the municipal boundary by the 18th century, and shown in 19th-century maps. This means that the unusually large inner bailey, as it became, was surrounded by a very much larger outer bailey. This outer bailey may initially have been defended as in the Anglo-Saxon period, by a ringwork and palisade. The form taken by the outer gate, then as later on the south-east side in close relation to the town’s East Gate, at the foot of the great bridge, is unknown. Second phase c. 1100–1139 Because Wallingford has an early castle and a motte it is generally assumed to have had a shell-keep. The documentary analysis here has failed to find any evidence for that view, which would require indications that the great tower – magna turris was used by contemporaries to describe both shell-keeps and great towers – was round and small, neither of which appears to have been true. The conversion of a timber castle to stone was very expensive. Only the king or the greatest magnates could afford it. Evidence of the transition at Wallingford is lacking. In the reign of Henry II, when royal records become more prolific, there is evidence for huge financial investment in stone castles, including shell-keep castles such as Windsor, but relatively little is spent at Wallingford, where a single reference to ‘bridge and motte’ in 1182 is the only reference to the arx – citadel – from the entire reign. The same is true up to the time that Henry III conceded the castle and honour to his brother Richard, earl of Cornwall, in 1231. This must mean that the conversion to stone and the building of the great tower took place early on. The serjeanty de vetere of keeping the masonry in good repair that antedated the death of Henry I in 1135 is further evidence for this (HKW i, 27). The eye-witness evidence from the 16th century that the great tower was on the motte is perplexing given that the motte is wholly artificial. Further evidence from the 105

same period that the great tower was both quadrilateral and sizeable, and was entered through a forebuilding, has required a radical rethink of the notion of a shell-keep at Wallingford. It is suggested here that the great tower was built on the initiative of Henry I, overseen by constable Brien fitzCount, between 1113 and 1135. It may have been one of a number of enmotted keeps built between 1100 and 1150, and like many of them could have started as a small pre-Conquest stone tower or a noble residence that was further enhanced and fortified by the addition of a motte, a move that on completion would have fixed the final layout of the inner bailey (discussed in Lloyd, below). Further evidence for the sophistication and strength of Wallingford Castle by 1139 comes from the considerable evidence relating to the castle during the civil war of Stephen’s reign, in which it played a crucial role. Only a very strongly defended castle with complete control of the town – effectively a huge further bailey – could have withstood the almost constant siege warfare of the period 1139 to 1152. If the garrison came close to surrender towards the end, the reason, as so often, was not the weakness of the walls but the threat of starvation. Archaeological investigation has shown that the creation of the second walled bailey, incorporating the barbican that regulated access into the town or castle through the North Gate, probably also predates the civil war, when it added to the considerable strength of the castle. The author of the Gesta Stephani described the bridge as the key to the castle and the town in 1152 (1.5). By that date there will have been a more substantial outer gatehouse than the Conquest-period original. This was located on the south-east of the castle in direct relationship with the East Gate and bridge (Figure 7.1, detail in Dewey below). Again, the lack of evidence of substantial spending on the castle under the Angevin kings suggests that any changes to the outer gatehouse(s) belongs to the time of Henry I. Whilst both Richard I and John were clearly concerned to keep the castle as strongly fortified and well-maintained as possible, there is no sign of major building works or architectural innovation. Third phase 1231–1300 Architecturally, the castle probably did not change a great deal until the reign of Henry III. Although work continued to be done on it, these were maintenance works that kept the fabric as sound as possible, making repairs as required. Since many of the bailey buildings were timber built they may have had to be replaced periodically. In the case of the great hall, as was often the case, Henry III ordered that it be taken down and a new one built in the same spot. The reason was that there were expected layouts of castle interiors that related directly to both function and aesthetic imperatives (Goodall 2011, 24, 28). If a great hall was ideally placed to serve this purpose when first built, there was no point in moving it elsewhere. Once a castle was in stone, the room for manoeuvre in development was limited. At Wallingford, when the castle was granted to the king’s brother Richard, earl of Cornwall, in 1229, the room for manoeuvre was discovered.

Richard of Cornwall has been described as the foremost castle builder of his day (Goodall 2011, 188–90). His brother, the king, tried to run him a close second, spending prodigious sums on Windsor and elsewhere, but Richard had the deeper purse. He spent lavishly on Wallingford, perhaps equalling the 10,000 marks (£6666 13s 4d) spent on his new monastery at Hailes (Matt. Par. v, 47). The abbey, founded in 1246, fulfilled a vow he had made on his return from a disastrous campaign in Gascony in 1242, following his return from a crusade to the Holy Land in 1240. His remark to Matthew Paris in 1251 about his great expenditure at Wallingford probably implies that his concentrification of the castle occurred principally (if not completely) between his return in 1242 and 1251, with work possibly continuing to 1256, when he left the country upon his election as king of Germany. In the interval, apparently in 1254, he had overseen changes at his castle at Berkhamsted, which now has concentric ditch fortifications, but only one walled bailey. There he built a shell-keep on the motte containing a three-storey tower; an inquest of 1338, after the death of John, earl of Cornwall, in 1336 found it to be cracked in two places and in imminent danger of collapse (Cobb 1883, 106–113). Although Richard was fond of Berkhamsted, his work there was insignificant beside his achievements at Wallingford. The survey of 1327 presented above shows that all three walls were in stone and had towers along their lengths. The creation of the third wall led to an extension of the castle outwards on its western side, and the relocation of the North Gate of the town. This was the most significant rebuilding of the castle since it was first built, because of its incursion into borough space. More strikingly still, each of the walls was surrounded by a water-filled moat, as is evidenced both from medieval documents and the notes kept by post-demolition owners of the site who traced the sluices and drained them (Dewey, 162). We also have evidence that most major buildings and the walls had wooden hoardings, by the 14th century perhaps intended as decorative rather than martial features, and that the Great Tower was whitewashed. The result must have been stunning, everything that the aesthetics of lordship could wish and more. Of this magnificence nothing but shadows remains. The major change under Earl Edmund (1272–1300) was the transformation of the lower middle bailey into an ecclesiastical enclave centred on the cloister of the clergy of the College of St Nicholas and their lodgings. He was probably also responsible for the Red Chamber in the inner bailey which evolved into a major second hall complex during the 14th century, accommodating the lady of the time. Royal stronghold and royal appanage 1066–1509 What is clear from the first is that Wallingford was very much a royal castle. The borough by definition was royal, and the extensive honour was the ‘honour of Wallingford Castle’, one that held liberties going back into the Saxon period. This was a role and an importance it never lost. The king had many castles and, peripatetic as kingship 106

then was, he had a large choice of where to stay. Some kings such as John, more peripatetic than most, seem to have particularly favoured it. Because of its important strategic location, it was always prominent in royal calculations whenever trouble loomed, even if more congenial residences at Windsor and the palace in Oxford’s Beaumont Street were preferred. Part of the reason for this is probably that Wallingford lacked the forests or parks required for the hunting so necessary for royal relaxation, though falconry seems to have had its place here. Its heyday as a royal residence was the 13th century, but it seems to have been intended as a primary residence for the heirs to throne as early as 1155. Before his death in 1189 Henry II directed that the castles and honours of Wallingford and Nottingham be handed to John, then his brother Richard I’s heir apparent. Richard had good reason not to trust John and retained the castles in his own hands. When John’s back as king was finally against the wall, two days before his death he entrusted the castle to his natural son, his legitimate heir being then still a child. The castle was one of the principal bases of Henry III during his minority (as it would be for Henry VI in the 15th century), and he stayed there frequently after he had relinquished it to his brother Richard, earl of Cornwall. But for Henry’s reliance on his abler and very much richer brother, Wallingford would have stayed firmly in the king’s hands, in trust for his own heir. When Richard’s son Edmund died childless, the castle could resume its intended purpose. That did not allow for the incompetence and ill-judgement of Edward II, in whose final undoing Wallingford Castle, under the control of his wife, Isabella of France, played a central role. It was given to his brother and heir apparent by Edward III, and on his death, to Edward’s son and heir the Black Prince. By this time the honour and castle of Wallingford was an integral part of the earldom of Cornwall, which since its first formal grant had been held by a son or nephew of the king. Its importance lay in the revenues generated by its stannaries, providing immense wealth for the earls who both financed the king and lavished money on their castles, including Wallingford. In 1337 the earldom was raised to a duchy and became the birthright of the heir to the throne. It last figures as a key player in the medieval period in the dying days of Richard II’s reign in July 1399, fittingly, since Richard had also spent a fortune on it and housed both his wives, Anne of Bohemia and Isabella of France, there. It continued thereafter as part of the appanage of the princes of Wales, but was often used as a residence for dowager queens, with ill-recorded changes being made to make the accommodation more suitable for female occupation. Even if not often used by kings, it is clear from surviving Ministers’ Accounts and other documents that royal heirs learned estate management in charge of the castle and its honour. But by the late-15th century its prisons are the most evident of its remaining functions. Its final humiliation, after Henry VIII turned his back on it, was the stripping of its lead for Windsor Castle.

Conclusion As the relatively few published cases have shown, detailed examination of building accounts can be very revealing (Simpson 1960, Colvin 1971), and not just in relation to grand buildings such as castles but also for medieval urban buildings (Rimmer 2006). There is a huge amount of interest in the Wallingford accounts, which one can only hope will one day be properly edited and published in full. What has been gleaned from them here is but a fraction of what is possible. They go a long way towards giving us back the vanished past of what must have been a truly magnificent building. Moreover, it was a very important building whose history has for far too long been neglected. Whether or not some of the speculations about the early form of the castle in the earlier part of this essay could ever be proved or disproved by archaeological investigation, this rummage through what is actually fairly abundant documentation has undoubtedly justified the exercise. Archaeological enquiry might consider core sampling of the motte in order to resolve its enigmas; further geophysical survey, with more powerful equipment currently in development, may also provide further clarification in the future. Greater precision about the site of the gatehouses and some of the buildings (see further in Dewey below) gleaned from the documents might also guide small-scale excavation. Properly targeted excavation, great or small in scale, ought to yield significant results. All the evidence points to a considerable stratigraphic depth and richness surviving on the castle site, with Carr’s results so far still the most spectacular. Absolutely crucial, therefore, is that every effort is made on all sides to preserve this important site, which has suffered considerably from benign mismanagement in recent years. [As we go to press, Dr Jim Leary and colleagues on the ‘Extending Histories: from Medieval Mottes to Prehistoric Round Mounds Project’ are hoping to investigate the Wallingford motte as another potential prehistoric mound, such as the Marlborough Castle motte (Leary 2013), using coring and surveying techniques; such an origin could help explain the anomalies discussed in this paper (cf. Creighton 2002, 69–70).] Strikingly, Wallingford has been shown at every stage to exhibit the same features of major royal castles elsewhere. It was always amongst the front-runners wherever castle fashions led. Its concentrification was avant-garde. It may even be that some other features, such as the studium, are exhibited earlier here than can be shown for other castles. There is plenty of interest on the socio-economic side, which has only been touched on here. Much more work on the still under-studied borough documents, and comparison with accounts from other southern castles, is still needed before this aspect of these accounts could be fully exploited. Wallingford Castle was of considerable importance because of its strategic location on the Thames as one of the crucial royal ‘castles of the interior’. In addition to the national importance lent it by its location, it was the place where the importance of that location was learned at first hand by the heir to the throne, who was often involved in the running of its estate as well as an occasional resident 107

there during his childhood. Those that were, such as Edward III and his grandson Richard II, retained an interest in and affection for the castle until they made way for their own heirs; in the childless Richard’s case this manifested itself in his confirmation in their posts of many of his staff on his accession, and the numerous pensions he awarded to the servants of his father there. A political as well as architectural trend-setter, Wallingford Castle was also the place where, over the Christmas holiday of 1326–7, the legitimate overthrow and deposition of a crowned king of England was plotted for the first time. This paved the way for the removal of Richard II in 1399, Henry VI in 1461 and 1471, both of them eventually murdered in prison on the orders of their successors, and culminating in the execution of Charles I after a trial engineered by the Rump Parliament in 1649. Bravely withstanding a lengthy siege on that king’s behalf in 1646, Wallingford Castle did not long outlive him, being demolished on Cromwell’s orders in 1652. Acknowledgements I am very grateful to Professor Michael C. E. Jones, Dr Edward Impey, Professor Robert Liddiard, Mr Julian Munby and Dr David Roffe, for advice on earlier drafts of this paper, and to Judy Dewey for the innumerable hours (indeed, years) of discussion and walking over the castle site that have helped to shape our ideas and the conclusions presented in this volume. My thanks also to Professor Ian Short for advice on medieval French, and for a stimulating if ultimately unfruitful discussion of the possible meaning of Muggetour, and to Dr Malcolm Vale for permission to use his transcript of Oxford, St John’s College, Muniment III, unpublished cartulary of the College of St Nicholas. All opinions expressed and errors committed are my own. A considerable debt must also be acknowledged to Elizabeth Will, but for whose meticulous transcriptions of the unedited documents that form the basis of this paper I should not have learned to read them well enough to exploit them. Special thanks go also to Gerard Latham, whose geophysical expertise and unstinting support with mapping have been invaluable to me and Judy Dewey as we have sought to recreate the castle enceinte. The industry of TWHAS members, led by David Pedgley, over many years in collecting references to Wallingford in various printed sources, has been invaluable as a starting point for the short Calendar appended hereto. Thanks also to June Strong, who produced the map showing sources of building materials, and the reconstruction of the castle entrance, and to Anthony Wilder for the drawing of the motte and Great Tower. A much fuller version of the Calendar will be made available on the Wallingford Museum website in due course. I hope also that this essay will stand as a tribute to the memory of my late husband, Dr John L. Lloyd, a physicist with an inexhaustible enthusiasm for all medieval castles, and the challenges of Wallingford’s in particular.

Documents list 1 The items below are cited in the main article as (1.1), (1.6), etc. The text is normally given here in translation; where the argument in the article has relied on my translation of an unpublished document, the Latin is given here. Two such documents, which have never been printed, are given in full. This list in in chronological order. Inserted into it are

references to the manuscript building accounts that form the basis of the article and are cited in the text by their dates. 1. 1066 November. Gesta Guillelmi, 146–7 The duke, advancing wherever he wished, then crossed the Thames by both a ford and a bridge and came to the town of Wallingford. Stigand, the archbishop, coming to him there, did homage to him, confirmed his fealty with an oath, and renounced the ætheling, whom he had elected without due consideration. As soon as William, advancing from there, came in sight of London, the chief men of the city came out to meet him; they submitted themselves and the whole city to him just as the men of Canterbury had done previously. 2. 1071 HEA i, 226–7 Imprisonment on King William’s orders, of Abbot Ealdred of Abingdon at Wallingford Castle (apud castellum Walingafordense in captione poneretur). 3. 1071–83 HEA ii, 5–6 In the first days of his abbacy, Abbot Adelelm went nowhere unless surrounded by a band of armed knights. Indeed, this was a matter of necessity, for at that time many and widespread rumours of conspiracies against the king and his kingdom boiled up, forcing everyone to defend themselves. Then castles were erected at Wallingford, Oxford, Windsor and other places for the defence of the realm. So it came about that this abbey was ordered by royal command to provide knights for guard duty at Windsor Castle. 4. 1139 Gesta Stephani, 91 At that time there was a certain Brien fitzCount, a man of distinguished birth and splendid position, who was extremely delighted at their [Matilda and Earl Robert of Gloucester’s] arrival and after strengthening an impregnable castle that he had at Wallingford rebelled against the king with spirit and great resolution, assisted by a very large body of soldiers... But King Stephen ... came to Wallingford with an innumerable host, and was minded to shut them in persistently with a ring of besiegers that could not be broken, when he was deterred by better advice from his barons and gave up his intention for the time being. For they said (what was the truth) that that castle was most securely fortified by impregnable walls, that supplies had been put into it in very great abundance, enough to last for a great number of years, that the garrison consisted of a very strong force of invincible warriors, and that he could not linger there any longer without the greatest injury to himself and his men... 5. 1152 Gesta Stephani, 226–7 For in front of the castle of Wallingford... he seized by force of arms the bridge that ran up to the entrance of the town, and was the master-key not only of the town but of the castle on that side. 6. 1155 April 10 Torigny, 184 On April 10, King Henry, at Wallingford, made the barons of the English kingdom swear allegiance to his eldest son 108

William concerning [his right to inherit] the kingdom, and if the boy should die young, then to Henry his brother.

wall walks behind the battlements (alure) at our castle of Wallingford.

7. 1164–5 Pipe Roll 11 Henry II, 32 Sheriff accounts for carriage of 3 tuns of wine to Wallingford for the use of the king’s son

18. 1215 May 11 Fragments of Close Roll of 16 and 17 John (PR. Soc., NS 31), 136, No. 49 The king to the sheriff of Berkshire, greetings. We order that without delay you summon everyone from your county to come and repair the ditches of our castle and town of Wallingford as they are anciently accustomed to do.

8. 1172–3 Pipe Roll 19 Henry II, 77 Thomas Basset accounts for honour of Wallingford, running costs of castle £67 11s/5d. 9. 1173–4 Pipe Roll 20 Henry II, 88 Running costs of castle of £12 7s. 10. 1175–6 Pipe Roll 22 Henry II, 135–6 And for lead and timber that Gervaise de Cornhill had taken from Wallingford to Woodstock, 28s 7d by the king’s writ. And for repair of the king’s houses at Wallingford, 13s 5d...and for repair of king’s houses in Wallingford, 11s 4d...and for running the king’s houses in the castle of Wallingford, £19 10s ...and for repair of the aforesaid houses [in the castle], 64s 5d... 11. 1178–9 Pipe Roll 25 H II, 100 Thomas Basset accounts £31 15s 3d for the running of the houses and castle of the honour of Wallingford. 12. 1190/1 Pipe Roll 2 Richard I, 36 And for the running of the castle of Wallingford and the mills, £35 20d. And for providing shields and lances and many arms 13. 1196 Pipe Roll 8 Richard I, Roll of Escheats, 205 And for repair of the private (bed)rooms and for roofing the hall in the castle of Wallingford, £6 5/2d by writ of H[ubert] archbishop of Canterbury.

19. 1215 May 13 Rot. Litt. Pat., 135b The king to all his knights of the honour of Wallingford Castle, greetings. We order you without delay to have the houses in our castle of Wallingford to be made ready for dwelling in, as you have been accustomed to since ancient times. Witnessed by me myself at Wallingford... 20. 1215 Pipe Roll 17 John, 42, 43–4 And for the provisioning of Wallingford Castle, purchase of 186 pigs, £18 and 17s, by the king’s writ for which the same sheriff must answer; and £10 for the provision of 100 bushels of wheat for the same... And for care of 13 cart horses and 6 men [carters, according to another copy] sojourning at Wallingford for 12 days, 44s 6d by the king’s writ. And for care of 1 kennelman with 24 dogs from the eve of SS Simon and Jude [28 Oct] up to the feast of St Thomas Martyr [29 Dec], 70s 2d by the same writ. And for carriage of the utensils of the king’s kitchen to Worcester, 11d by the same writ. And for the expenses of William de Merc, with 4 gerfalcons and one falcon gentle, and 5 men and 4 serving boys and 6 horses and 4 greyhounds etc., etc… And for the cartage of 10 tuns of wine from Wallingford to Oxford, 13s 2d by the same brief. And for maintaining, fortifying with hoardings and guarding the castle of Wallingford [blank]

15. 1201 Pipe Roll 3 John, 202 £30 for the allowances of the knights and servants who were retained in Wallingford Castle after the death of King Richard. And £11 2s 3d is owing. For running Wallingford Castle, £4 10s.

21. 1216 Jun 3 Rot. Litt. Pat. i, 185b The king to his dear and faithful J[ohn] the clerk sheriff of Berkshire, Know that we wish that one of your men, together with one man of our son Richard and one man of Walter Foliot, should be on guard during the day at the gate of our castle of Wallingford, so that none shall have entry into the castle except by the common consent of you and the aforesaid men of Richard and Walter. At night we wish that the keys of the gate be committed to our son Richard. But you, sheriff, shall continue as formerly to have charge of the county. Concerning the honour of Wallingford, you shall have entry as before.

16. 1213 Jan 30 Rot. Chart., 110 The king to Peter fitzHerbert, etc. Know that we have committed to John de Wikenholt, sheriff of Berkshire, the castle of Wallingford with all its appurtenances to hold in custody as long as pleases us, wherefore we order you to deliver the said castle of Wallingford and all appurtenances without delay.

22. 1216 Jun 4 Rot. Litt. Pat. i, 142b Mandate to John de Wigenholt, sheriff of Berks, and the constable of Wallingford, that they receive Philip de Wygorn [Worcester] into Wallingford Castle and act upon his advice for the strengthening of the castle and to have faith in what he tells them on the king’s behalf to the profit and honour of the king.

17. 1214/15 Feb 13 Rot. Litt. Claus i, 187b Paid to the sheriff of Berkshire the account submitted by sight and testimony of 12 legal men for repairs to the

23. 1219 Pipe Roll 3 H III, 158 For the expenses of the king at Wallingford...£20. And for livery, armour and carriage of two approvers, 15s 2d.

14. 1201 Pipe Roll 3 John, 195 And for the repair of a chamber in the castle of Wallingford, 9d by the king’s writ. And for ironworks for the prisons, 3s by the same writ... And for iron bars in the windows in a chamber of the castle, 3s by the same writ.

109

And for the repair of buildings [domorum] in the castle of Wallingford, £6. And the same has in surplus for the running of the wardrobe in the castle, and for the repair of a breach [in a wall] which fell down in a storm, £22 6s 4d blanch.

31. 1307 July 8 -1310 July 7 1–3 Edward II, TNA, E101/489/30 Very detailed accounts by John de Clinton, seneschal of the honour of Wallingford, of extensive programme of works.

24. 1220 Apr 20 Rott. Litt. Claus.1, 415b King to the sheriff of Berks, greetings. We order you by the view of John de Wigenholt and other lawful men that the same John chooses to act with him, to pull down our hall in the castle of Wallingford and to start building a fitting replacement on the same site.…

32. 1317 Nov 1 CCR 1313–18, 505 To Edmund Bacun, constable of Wallingford, order to cause that the castle be kept safely by ministers and other sufficient men as he ought to have therein, according to the fee that he receives yearly from the king. The like to the following constables, of the Tower, Dover, Windsor, Cockermouth and Pevensey.

25. 1220 Pipe Roll 4 H III, 197 And for the stipends of the servants from the town, 20s. And for the building of the king’s hall at Wallingford, £89 17s 11d 26. 1222 Apr 4 Rott. Litt. Claus. 1, 491b Sheriff ordered to make a chimney in our upper (alta – high) chamber in Wallingford Castle... 27. 1224 Pipe Roll 8 Henry III, 222 And for whitewashing (dealbandis) the hall and chamber of the king in the castle of Wallingford 72s 2d… and for upkeep of the houses in the castle 100s. Ibid., 234 (Residual account from the time of King John by John de Wikenholt) For repair of the alures in the castle of Wallingford £72 2s 6d. 28. 1282 Feb 10 St John’s College, Oxford, Muniment III, fol 3v (Earl Edmund) Volumus eciam et concedimus pro nobis et heredibus nostris quod predictus Rogerus, decanus, et successores sui decani et capellani ac eciam clerici, Deo in predicta capella imperpetuum servituri, habitare possint et commorari infra castrum nostrum predictum, videlicet in illa placea infra medium ballium que se extendit in longitudine et latitudine a capite magni stabuli nostri versus orientem usque portam de Bydonesgate. 29. 1284 Nov 20 – 1286 Nov 19 TNA, E101/489/29 Badly damaged single membrane of accounts on works at Wallingford Castle under Earl Edmund. 30. 1307 Dec 19 CCR 1307–13, 50 To the keeper of the king’s castle of Dover. Order to safely and securely keep and defend the said castle so that no damage nor changes happen to the same, the king, who intends shortly to set out for parts beyond the sea, desiring that the castles of his kingdom should be diligently and safely guarded and defended for the greater security and tranquillity of his people. The like to the keepers of the castles of [two lists; in the first, in order, Rochester, London, Windsor, Wallingford, Corfe, Isle of Wight]

33. 1317 Nov 1 CCR 1313–18, 505 To Edmund Bacun, constable of Wallingford in the hands of Queen Isabella, Order to put 40 fencible men, crossbowmen and others, in that castle at the king’s wages. For the safe custody thereof until further notice. 34. 1318 Dec 3 CCR 1318–23, 39 To Edmund Bacun, constable of Wallingford Castle. Order to permit the king’s serjeants, Gilbert de Boghere and Thomas his brother, whom the king is sending to repair the defects of the equipments (attilii) of the castle, to enter the castle, and to pay each of them 4d a day from the day when they enter the castle for so long as they shall stay therein ....[Duplicate of same instructions to Warin de Insula, constable of Windsor] 35. 1323 Feb 5 CPR 1321–24, 257 Commission of oyer and terminer [enquiry to hear and decide] to John de Stonore and John de Bousser touching the persons who seditiously entered the castle of Wallingford, co. Berks, wherein Maurice de Berkeleye and Hugh Daudele the elder and other prisoners were detained, and held it against the king; as the king now understands that the said Maurice and Hugh and the other prisoners consented thereto, and kept the castle against the king jointly with said persons. 36. 1323 Apr 7 CPR 1321–24, 314 Commission to Henry le Scrop, John de Stonore and John de Bousser to enquire touching the persons who planned the seizure of the castle of Wallingford, the Tower of London, the castle of Windsor and other of the king’s castles. 37. 1325 Jan 20-Feb TNA, E358/3 Roll 12, Drogo de Barentin, sheriff of Oxon and Berks, accounts for provisions of Wallingford Castle; Roll 13, John de Coton and Gilbert de Mordon, sheriffs of London, account for victuals for the Tower of London and the castles of Wallingford and Windsor [in that order], William le Mareschal constable viewing the accounts for Wallingford.

110

38. 1326 Jan 22 TNA, E101/17/11, membrane 3 Part of an inquest of arms and stores in various castles incl. Windsor, Tonbridge, Leeds, Wallingford, Odiham, Conway, Carnarvon, Beaumaris and Harlech. William the Marshal, constable, answers for the arms and equipment in the castle, listed, much of it in poor condition. Ralph Restwald, under-constable of Wallingford Castle, answers for household victuals, vitailles souscrite de mesnie le chastel. In French. 39. 1326 Sep 30 CFR 1319–27, 417 Commitment to John de Stonore of the keeping of the castle of Wallingford, so that he answer at the Exchequer for the issues thereof. Order to William le Mareschal to deliver the same to him, with all things therein in his keeping, by indenture to be made between them. Order to the said John to go to the said castle forthwith without excuse, and to assume that keeping and to stay there with his whole posse, notwithstanding any office to which he has been assigned by the king, and to apply such diligence to the keeping of the castle and of the bodies of Henry de Bello Monte [recently constable of the castle at Somerton, Lincs] and other prisoners therein that he can answer for the same at his peril. Given at the Tower of London. 40. 1326 Oct 7 CPR 1323–27, 331 Given at Wallingford. Appointment during pleasure of John de Stonore to the custody of the corpus of the castle of Wallingford, and of the prisoners there, but he is not by pretext of this grant to be charged with the victuals in the castle or with the lands, or other foreign appurtenances of the same. Mandate to William le Mareschal, constable of the said castle, to deliver to him the said corpus. Writ of aid for him directed to the sheriff of Oxon and Berks and others of those counties, all sheriffs and others and the tenants of the castle and honour of Wallingford. 41. 1326 Oct 3 to Oct 16, accounts of John de Stonor keeper of the castle, TNA, E101/17/21 Compotus Johannis de Stonore custod’ castri de Walyngford a tercio die Octobris anno regni E. patris regis nunc vicesimo virtute cuiusdam commissionis eiusdem domini Regis usque septem diem eiusdem mensem anno supradicto. In primis de victualibus nec de exitibus terrarum et ten’ honoris de Walyngford ?nichil se onerat recepisse per tempus predictum eo quod Willelmus le Mareschal constabularius castri predicti et senescallus predicti honoris de victualibus et de exitibus terrarum et ten’ predictorum per totum illud tempus se ut prius intromisit et nichil inde predicto Johanni liberavit. Idem Johannes comp’ in vadiis suis in eodem castro commorant’ a tertio die Octobris usque septimum diem eiusdem mensis anno supradicto per quatuor dies viii. s. capiend’ per diem .ij. s. Summa –p[robet]. Item comp’in vadiis sex armigerorum secum commorancium in eodem castro per totum tempus predictum .xxiiij. s. cap’ quibus eorum per diem xii. d.

Summa – p[robet] Item comp’ in vadiis quatuor hominum de ministerio per tempus predictum .viij. s. capiend’ quilibet eorum per diem .vj. d. Summa – p[robet] Item comp’ in vadiis sex garcionum per tempus predictum. iiij. s. cap’ quibus eorum per diem .ij. d. Summa – p[robet] Summa expens’ per tempus predictum xliiij .s. Compotus predicti Johannis de Stonore custod’ corporis castri de Walyngford virtute alterius commissionis a septimo die octobre usque sextemdecimum diem eiusdem mensis anno supradicto. In primis onerat se recepisse de thes’ et camerariis de servicio domini Regis predicti centum marcas pro expensis castri supradicti per breve domini Regis sub priuato sigillo predictis thes’ et camerariis directo. Summa – p[robet] De quibus comp’ se liberasse domine Isabelle regine apud Walyngford sexagintes et quinque librarum quindecim solid’ et duos den’ pro ut in litteris eiusdem domine Regine eidem Johanni inde confectis plenius continetur. Summa – p[robet] Item comp’ in vadiis suis et expensis unius militis secum commorant’ ibidem a septimo die October anno supradicto usque sextadecim diem eiusdem mensis per novem dies .xxxvj .s. capiend’ quibus eorum per diem .ij. s. Item comp’in vadiis duodecim armigerorum commorantium in dicto castro tempore predicto per novem dies .Cviiij .s. capiend’ quilibet eorum per diem .xi j. d. Item comp’ in vadiis octo arbalistorum commorantium in dicto castro tempore predicto per novem dies .xxxvj .s. capiend’ quilibet eorum per diem .vj. d. Item comp’ in vadiis duodecim hominum armatorum vigilancium in dicto castro tempore predicto per septem noctes .xlij. s. capiend’ quilibet eorum per diem et noctem .vj. d. Item comp’ in .cxviij. hominum pedit’ pro municione castri predicti vigilant’ in duobus forinsecis balliis .xiiij. die Octobr’ per diem et noctem .xvii. s. x. d. capiend’ quilibet eorum per diem et noctem ii d. [corr. suprs. xix s. viij. d.] Item comp’ in vjxx iiijor hominum pedit’ pro municione castri predicti vigilant’ ut supra quintodecimo die Octobr’ xviij. s. .x. d. [corr. supras. xx. s. viiij d.] capiend’ per diem et noctem .ij. d. Summa recept’ lxvi. li. xiij. s. iiij. d. Summa liberat’ lxv. li. xv. s. ii. d. Summa expens’ xij. li. xviiij. s. viij. d. 42. 1326 Oct 15 at Wallingford Rymer, Foedera II, 645–6 Isabelle, par la grace de Dieu, Reyne d’Engleterre, dame d’Irlande, contesse de Pountiff; & nous Edward, aisne fiz au noble Roy d’Engleterre, duces de Guyenn, counte de Cestre, de Pountyf, & Monstroyl; & nous Edmund, fiz au

111

noble Roy d’Engleterre, counte de Kent; a toux iceaux, que cestes lettres vendrient, salutz. Por ceo, que conue chose est notoirement, que l’estat de seint eglise & del roialme d’Engleterre est en moult des maneres durement blemy, & abesce par mavoys consail & abet Hugh le Despenser, que pur orgoil & coveytise de seigneurier & mestrier sur toux altres, ad purpris roial poair countre droit & reson, & sa ligeaunce, & en tele manere le ad use par le mavoys consail R. de Baldock, & autres adherdans a ly, que seint eglise est de lour biens, countre Dieu & dreiture, despoilez, & en trops des maneres ledenges & dishonurez: Et le coronne d’Engleterre destrue en divers maneres, en disheritaunce de nostre seigneur le Roy, & de ces heirs: Les graunts du roialme, par envie & mavys crualte de dit Hugh, plusours saunz coupe, & saunz cause, a hountouse mort liverez; les uns desheritez, les altres imprisonez, banniez, & exilez: Veues, orphanyns de lour droit a tort forjugiez: Et le poeple de la terre, par diverses taillages, & noun dues exactions, trop sovent reynez, & par divers oppressions, saunz nulle mercy, grevetz; Per queaux mesprises le dit Hugh se monstre apert tyrant & enemy de Dieu & de seint eglise, de nostre treschier seigneur le Roy, & de tout le roialme. Et nous, & plusours autres, que sount ovesque nous e nostre compaigne, que loungement avoms este aloignez de la bone voillaunce nostre dit seigneur le Roy, par la fause suggestion & mavoys procurement des avaunt dits Hugh & Robert, & lour adherdaunts, sumes venuz en ceste terre, pur lever l’estat de seint eglise & del roialme, & le people de la terre, del dits meschiefs & grevous oppressions, garder & mayntener a nostre poair, l’onur & profit de seint eglise & nostre dit seigneur le Roy, & de tout le roialme, sicome est desutz dit; Pur quoi nous vos mandoms & prioms, pur commun profit de vous toux & chescun de vous, endroit de sey, nous soietz aidaunt, bien & leaument, toutz les foitz, que vous verretz lieu & temps, & par toutes les voies que vous saveretz & pourretz, a ceo que les choses desutz dites puissent hastivement venir a effect, & a bon fin; car soietz certaynes que nous toux, & toux iceauz que sunt en nostre compaignie, ne en pensoms a faire chose, que ne serra pur l’onur & le profit de seint eglise, & de toute le roialme, sicome la verretz & troveretz par temps, si Dieu plest. Dat’ a Wallingford, le xv. jour d’Octobre, l’an del regne nostre treschier seigneur ly Roy vintysme. 43. 1326 Oct 14 – 1327 Jan 27 20 Edward II, TNA, E101/17/22 Particulars of an account of victuals for Wallingford Castle. Damaged. Series of entries on one membrane. Section headed 14 October 1326 appears to be when Isabella replaced William Marshal as constable with Miles de Bellocampo. Their accounts settled. Considerable damage done to supplies of winter oats by rats and mice. Reference to the war, tempore guerre.

44. 1326 December TNA, E101/17/28 Account of victuals in Wallingford Castle in December 1326. Sweyn de Mortele, appointee of Queen Isabella, answers for the provisions, and Milo de Bellocampo for the armaments. In Latin, poor condition. 45. 1327 soon after Feb 24 TNA, E143/10/2 Inquisitio capta coram A de Novo Burg assignato ad omnes defectus omnium castrorum existent’ tam in manibus domini Regis quam in manibus domine Regine in Com’ Oxon’ & Berk’ apud Walingford die Mercur’ proxima post festum Sancti Mathei apostoli Anno Regni Regis Edwardi tercij post conquestum primo, per sacramentum Ricardi de Louches, Nicholai de Turneston’, Andre de Sutton’, Henrici de Melburn’, Willelmi Arnyet, Willelmi Broum, Johannis Stacy, Willelmi filii Petri, Ricardi le Mason, Johannis le Carpenter, Ricardi Gratard & Walteri le Cripse. Castrum Walingford Balliva exterior Qui dicunt per sacramentum suum quod balliva exterior dicti castri est tota vetus & ruinosa ut in muris, barbican’, & turrellis circumquaque dictum castrum, cum Gaweynes herber, indigent reperacione usque ad summam CLX libri. Dicunt etiam quod pontes dicti castri, videlicet interior, medius & exterior, indigent reparacione usque ad summam C s. Summa C LXV libri Balliva media Dicunt etiam quod balliva media indiget reperacione in diversis locis ut in petra calce Sabulonis & tegulis & stipendiis operariorum emend’ mur’ & turrell’ dicto ballive usque ad Summam xiij li vj s viij d. Dicunt etiam quod est ibidem quoddam stabulum de fago ruinosum, & indiget reparacione ut in meremio, tegulis & aliis necessariis, cum stipendiis operariorum, usque ad Summam xl s. Et dicunt quod domus prisone dicte ballive & camerarii & camera super pontem indigent reperacione ut in petra, calce, sabulone, meremio, tegulis & aliis necessariis, cum stipendiis operariorum, usque ad summam xxx s. Dicunt etiam quod camera qua vocatur La Studie cum aliis cameris, & ponte subtus & desuper, indigent reparacione usque ad summam xx s Summa xvij li xvj s viij d Balliva Interior Dicunt etiam quod balliva interior dicti Castro, videlicet magnam Turrem cum Turrell’ Grecia, cum turribus & turrellis murum circumquaque, & camere, sunt generabiliter in bono statu & de novo reparatis per Willelmum le Mareschall nuper constabularium dicti castri; Set tamen indigent reparatione usque ad summam C s. Dicunt etiam quod sunt in alt’ Turr’ quatuor molendina manuales qua indigent reparacione usque ad summam xiij s iiij d. Dicunt etiam quod minuti defectus, videlicet in fundamentis & corbellis desuper Turris & aliis domibus, ut in plumbis, clavis, stipendiis plumbatorum & aliorum operariorum, possunt emendari cum xl s. Dicunt etiam quod magna Aula domini Regis & Camera Domine Regine, Viridis camera, Rubea camera, Garderoba Regine,

112

Camera Militum & Armigerorum, ut in tegulis & aliis minutis necessariis cooperiend’, indigent reperacione usque ad summam xx s. Et dicunt quod magna Capella dicto castro indiget reparacione ut in scindulis pro cooperatur’ cum stipendiis operantium usque ad summam LX s. Et dicunt quod magna coquina, pistrina, bracina, cum parvis cameris adiacentibus & tribus fornacis in eisdem, cum defectibus magne gaole dicti castri, indigent [re]paracione usque ad summam vij li Summa xviij li xiij s iiij d Summa Total’ defect’ ?predict’ ?comp’ CC j li x s 46. 1331 Jan 25 – 1334 24 Jan 5–7 Edward III, TNA, E101/489/28 Accounts of Gilbert le Messager and William Clerk, supervisors of the works at Wallingford Castle. [Gilbert occurs on the Clinton 1307 roll receiving 2s 6d for work as supervisor]. Total £65 8s 5d. 47. 1352/3 Mar 8 REBP iv, 82. Order to John de Alveton, lieutenant of Sir Bartholomew de Burgherssh ... to repair all the defects of the walls of the castle of Wallingford toward the river from the foundations upwards, and to remake those parts of the wall which have fallen down in the same fashion as the remainder. These are to be the first masonry works undertaken at the castle. Order, also, to prepare quickly the quarry of Whetelee and make it ready for digging stone against next summer. He is to pay the moneys required for the above works out of the issues of the farm to John de Wendout, chaplain, who it to pay the workmen as he has heretofore done for the prince’s other works of the said castle. Order, also, to instruct Richard, the prince’s master-carpenter, to carry out the essential timber works of the castle until someone of the prince’s council comes there to survey the matter and give further order as to what is necessary. Also on 31 March the said Sir John had a commission to take workmen for opening up the quarry of Whetelee and digging stone for the repair of the castle. 48. 1355 Nov 25 REBP iv, 173. Order to John de Alveton, the prince’s yeoman and lieutenant of the steward of Walyngford, to look into various alleged misdemeanours of the mayor and burgesses of Walyngford: 1. seizure of various tenements in the borough that ought to have escheated to the prince; 2. taking chattels of felons and fugitives owing to the prince; 3. entering the prince’s mills under the castle and the mills of Suthgate claiming they are within the bounds of the franchise of the borough; 4. occupying the fishery in the stank of the prince’s mills of Suthgate; 5. not providing the constable of the castle every weekday 5½ gallons of the best ale brewed in the borough, he paying 2d; 6. keeping a great number of unguarded swine which do great damage and trample down the castle dykes;

7. claiming the right to feed their mares, oxen and sheep in the prince’s demesne meadows of Clopcote on Good Friday and Easter every year; 8. claiming the patronage of the house of St John, Walyngford, though it used and ought to be under the protection of the lord of the castle. 49. 1360 Mar 26 REBP iv, 346 Order and strict injunction to John de Alveton … in view of the many perils which might easily befall the castle of Wallingford in these present times, to cause the castle to be safeguarded to the best of his knowledge and ability, and to warn all those who are bound to do guard in the castle in time of war that without delay they arm and array themselves to come to the castle and do guard according to their tenure when warned by him by letter, and to repair the bridges of the castle at as small a cost as possible, so that they may be drawn up if necessary. 50. 1362 23 Mar REBP iv, 426 Order to Sir Bartholomew de Burgherssh – on information that the roofing of the chapel of St Nicholas in the castle of Wallingford, and the houses of the chaplains and clerks thereof, is broken and spoilt by the late storm and greatly needs repair – to remake and repair the same, as well as the other defects which need repair and can be repaired at small cost, by the advice of Sir William de Clobhoo, chaplain of the said chapel, making an indenture with him touching the costs incurred. 51. 1363 Feb 10 St John’s College, Oxford, Muniment III, fol. 15r-v [Black Prince] Statuimus et ordinamus quod omnes ministri nostre capelle cum suis servientibus habeant liberum ingressum et egressum tam ad capellam quam ad villam tociens quociens fuerit eis oportunum. Item, Statuimus et ordinamus quod omnes ministri in suis primis receptionibus jurent super Sancti Dei Evangelia quod consilium nostrum infra castrum firmiter ac/[fo. 15v] et fideliter conservent, nec aliquod dampnum per eos infra castrum fiat. E[t] si aliquid infra castrum audierint vel perceperint quod possit nobis ad dampnum cedere vel devenire, nobis vel nostro consilio quam citius poterint indicabunt. Et quia propter transgressores posita est lex. Ideo Statuimus et ordinamus ad evitandum negligenciam ministrorum penam sequentem. 52. 1366 April 5 CPR 1364–7, 235 Appointment of Henry of Aldrynton, Simon de Lagenhull and John de Cotesmor, bailiff, to take carpenters, masons and other artificers in the counties of Oxon, Berks, and Bucks for repairing the castle of Edward, prince of Aquitaine and Wales, of Wallingford and the houses and buildings therein, to put them in the works at the prince’s wages, to stay there as long as shall be necessary to provide carriage for timber, stone and other necessaries for the said works, for the princes’ moneys and to arrest and to commit to prison all that hinder them therein.

113

53. 1367 Jan 25 - 1369 Jan 24, TNA, E101/545/27 View of the accounts of Henry of Aldrington. 54. 1369 25 Jan - 1371 Jan 24 43 Edward III, TNA, E101/545/28 Account of Henry de Aldryngton. Total £76 9s 8d 55. 1374 May 12 CPR 1370–74, 436 On the supplication of the parson of the church of St Peter, Wallingford, and certain of the parishioners of the churches of St Mary the Less and St Michael of the same town, which are of the patronage of the dean of the free chapel within the castle of Wallingford, praying that, whereas the bishop of Salisbury, the diocesan, in compassion for the poverty of the church of St Peter, the profits of which do not suffice to maintain a parson, and considering that the churches of St Mary and St Michael are as it were desolate and destroyed for lack of rule because no-one has cared to undertake the cures thereof for a great time owing to their excessive poverty, purposes to unite the two lastmentioned churches to the first, he will give his royal licence for this; the king has granted such licence. 56. 1375 49 Edward III, TNA, E101/490/3 Payments made by Walter Yonge, receiver of the Black Prince, for works at Wallingford overseen by Henry de Aldryngton during 1375 (49 Edward III). Total spent on the works, £206 19½ d. 57. 1376 Oct 13 CCR 1374–77, 408 To Nicholas Somerton, escheator in Oxfordshire and Berkshire. Order to deliver in dower to Joan, who was the wife of Edward prince of Wales etc., Walyngford castle and 70l 8s 9½d of rent arising of the lordship thereof in divers parcels, namely of a fishery in the river Thames 46s 8d, of the mill there with the fishery of eels at ‘Floodgates’ 8l 10s, of the island of meadow called Northacche 20s, of the 4 acres of meadow called Quenesherber 24s, of 36 acres of meadow called Kyngesmede 7l 4d, of the fee farm of the manor of Bensyngton 39l 9s 7d, and of the perquisites of court and of the view of that manor 10l 18s 2½d, also the manor of Netilbed with the wood there to the value of 60s. 58. 1378 July 10 CPR 1377–81, 258 Appointment of Aubrey de Veer, constable of Wallingford Castle, Walter Yonge steward there, and Richard Branche, bailiff of the honour of Wallingford, to take labourers, artificers and workmen wherever found (except in the fee of the church, and those employed elsewhere in the king’s service), who are to be put upon the works at that castle at the charges of the king’s mother. 59. 1389 Nov 9 CPR 1388–91, 146 Appointment of the sheriffs of Oxford and Berks and the mayors and bailiffs of Oxford and Wallingford to arrest masons, carpenters and other workmen needed for the repair of Wallingford Castle, at the king’s charges, to be paid by John, prior of St Frideswide’s, Oxford, governor and disposer of the said work.

60.1389 Jun 22 - 1391 Jun 21 13 and 14 Richard II, TNA, E101/490/4 Richly detailed accounts by John prior of St Frideswide’s, Oxford, overseer, from major renovations carried out for what proved a very short visit by the king. Total £125. 61. 1404 Nov 11 CPR 1401–05, 476 Appointment of the king’s clerk, Henry Merston, as the clerk of the works at the palace of Westminster, the Tower of London and the castle of Wallingford [with 6 others] ... with power to take stone-cutters, carpenters and other workmen and labourers and stone, timber, tiles, shingles, glass, iron, lead and other necessaries ... and sell boughs, bark and other residues of trees. 62. 1408 Sept 30 – 1409 Sept 29 10 Henry IV, TNA, E101/490/5 Account of John Coterell, recorder, for the repair of houses and walls within Wallingford Castle. Total 113s 3d 63. 1411 Sept 30 - 1412 Sept 29 13 Henry IV, TNA, E101/490/6 Incomplete. John Coterell’s accounts. 64. 1414 June 30 Rymer, Foedera ix Henry V grants licence to Queen Joan, the king’s mother, to reside during his absence at any of the castles of Windsor, Wallingford, Berkhamsted, and Hertford. 65. 1428 May 8 Proceedings and Ordinances of the Privy Council iii, 294 Guardians of young Henry VI appointed the castles of Wallingford and Hertford as the king’s summer residences, and those of Windsor and Berkhamsted as winter residences. 66. 1433 Dec 10 CPR 1429–36, 330 Grant to the bailiffs and good men of Wallingford of pontage for 5 years to be taken by the view and testimony of Thomas Chaucer esq., Thomas Haseley esq. and John Warfeld. Passengers, horses and carts have lately been injured through the bridge being out of repair. 67. 1438 Jun 26 Extracts from the Issue Rolls of the Exchequer, Issue Roll 16 Hen VI [1437–8], 436 To John Ardern, clerk of king’s works. In money paid to him, by assignment made this day, by the hands of William Marmyon, out of the issues of the lordship of Wallingford, in advance for repairs and for newly making and building the king’s prison there, for the safe and secure custody of the king’s prisoners within his castle of Wallingford. By writ, etc. £40. 68. 1438–41 Sept 1 1438–31 Aug 1441, Henry VI, TNA, E101/479/7 William Marmion and deputy John Arden account for building the prison house for a total of £142 9s 9d.

114

69. 1440–42 TNA, SC6/1096/19 Receiver’s [Geoffrey Kidewelly] accounts of castle, borough and honour of Wallingford 19 Henry VI. Gainery of the castle described based on the dower of the late Joan, princess of Wales (d. 1385). Includes meadows called Northayte and Queen’s herber, a fishery in the Thames next Kingsmead, and the farm of two ramparts, viz the outer and middle rampart with the vineyard which is trained along the sides and banks of the ramparts (super litora et ripas dictorum fossatorum). 70. 1481–3 Ministers’ and Receivers’ Accounts, Edward IV, TNA, SC6/1302/1 Accounts of receiver for Edward, prince of Wales, whose guardian is Anthony, Earl Rivers, his uncle. Scouring of moats, repair of main mill stank and dredging of the stank of the mill in Clapcot (i.e. North Mill). New boat for the Shillingford ferryman who brings the prince to the castle. 71. 1485–7 1–2 Henry VII TNA, SC6/HENVII/1845 Receiver’s accounts, including castle, borough and honour of Wallingford. Repairs to houses in the castle, the outer bridge and the Great Stable, and new millstones for the mills, materials costing £27 10s 10d 72. 1490–1499 TNA, SC1/44/85 Arthur, prince of Wales, to Richard Ludlow: warrant for repairs to Wallingford Castle, Berkshire. Signet. Dated at Chester.

73. 1524 Oct 28 Will of William Aldworth, TNA, PROB 11/21/276 In the name of God amen The yere of our lord God M D and xxiiij Saint Symonde and Jude day I William Aldworth of Clopcote in the Countie of Berks beyng in hole mynde and fitt of body ordeyne and make my testament and last will in this wise ffirst I geve and bequeth my soule unto almighty god to our lady Saint Mary and to all the holy company in hevyn my body to be buried in the Churche of saint Nicolas within the Castell of Walingford. 74. 1535 21 Nov LPFD 11, 864 Sir Walter Stonore to [Thomas] Cromwell. This present Sunday last past, the mayor of Wallingford has sent me the two constables, and his examination of John Perse, surgeon, for words spoken by him in the house of John Juell, constable, in the presence of Sir John Mychell, priest, touching the king. I send you the examination and the priest, and Perse I have sent to the castle at Wallingford. The constables and others taken to record by the priest are honest poor men ... 75. 1537 LPFD 12, 796 Sir Walter Stonore to Cromwell. Begs remembrance of his fee for the half-year ended on Lady Day in Lent last, for the lieutenantship of the honor of Walyngford, which he had of the king when his Grace gave the constablewick to Harry Nores, deceased. Was charged at his own risk with the gaol till the day of deliverance, which was about the middle of August ensuing.

115

6 TWO Tudor Surveys of Wallingford Castle from the Rotulus de Wallingfordia John L. Lloyd†

Completed and revised by K. S. B. Keats-Rohan

Abstract Various eye-witness accounts of Wallingford Castle survive from the Tudor period. Two are of especial importance, each produced as part of a quantity surveying inquiry between c. 1548 and 1555, relating to the inner bailey of the castle, neither of which has been fully discussed before. They are found in a Rotulus de Wallingfordia (Wallingford Roll), now Oxford, Bodleian Library MS Top. Berks. b. 41, the work of royal surveyor Roger Amyce and others between 1548 and 1561. The analysis begins with the Certificate of 1555 describing the quadrilateral keep. Attention then turns to the Surveyor’s Plan. Though versions of it have been printed several times, no serious attempt has ever been made to decipher the annotations and other marks made by the surveyor. Once that is done, some problems emerge since the plan as drawn does not accurately show the stated measurements, most notably in relation to the southern, motte, end of the bailey. A final section compares the information derived from the plan with more recent evidence derived from geophysical and earthwork survey and finds that the geophysics helps to resolve the problem with the plan, and that the corrected plan provides information no longer available for archaeological investigation. It is suggested that the earliest layout of the bailey can be detected from the Plan and that it remained remarkably stable throughout its history. Introduction The Icknield Way and the Ridgeway survive from prehistoric times: travelling salesmen of the day found there a dry route, potable water and customers. Wallingford lies at, or very near, their junction. The Thames was easily fordable here (and both upstream and downstream, as can be seen from place-names) and the approach from the Chilterns was fairly good. Modern dredging has increased the depth to seven feet at Wallingford. King Alfred established a defensive burh here around 880, with bermless ditch and bank, much of it visible today. Two centuries later, following the battle of Hastings, William the Conqueror moved towards London, but met with fierce resistance, preventing him from crossing the Thames. He therefore marched upstream to Wallingford, crossing the Thames by bridge and ford, and there began negotiations for the surrender of the English, including Wigod of Wallingford (p. 47 above). Here was a fortified town with a friendly thegn, at a strategically excellent place to plant a castle. And one was here planted, within the defences of the Saxon town, as was done elsewhere (Keats-Rohan, p. 36). It is this early castle that will concern us, and

specifically the inner bailey, its motte and its keep, as drawn in outline in a Tudor plan and described in writing by quantity surveyors in 1555 (Figures 6.1 and 6.2, and List 2.1). These eye-witness accounts have never received the full discussion they so clearly merit. The north-east corner of the burh was not much settled when the Normans arrived, and the town ditch and the defences on the northern rampart section were made part of the castle boundaries. A moat was dug that with them enclosed a large rectangular area. In the southern part of this area sits the motte. The moat (now much modified by later changes to the fortress) was fed from the town ditch, which itself was fed by a stream. The beds of the ditch and of the moat lie well above today’s river level. The area so enclosed was somewhat greater than that of the defensive core of the castle, i.e., the inner bailey, and I shall suggest that the ground-plan of that inner bailey was marked out before the motte was raised. The parts outside this inner core could have constituted an early outer bailey, easily fortified using the ample supplies of beech and oak, both hardwoods, on the Chiltern and White Horse hills. This early outer bailey was subsequently fashioned into a middle and outer bailey (Keats-Rohan, p. 39 above). John Leland (c. 1540) described these two outer wards, saying that they had moats and stone walls; of the inner ward he noted the towers rather than the wall (2.2). Camden (1586) also wrote an informative note on the town and castle (2.3). The length of the inner moat (excluding the town ditch section) was probably and still is about 1080ft (329m). We can compare its volume with that of the motte. On the western side, which may be the least modified part, the cross-sectional area of the excavation can be guessed at, assuming that although its shape will have changed, its area will not. An estimate, based on measurement, is 1000ft2 (93m2), though this could be inaccurate by 30 per cent or more. That leads to an excavated volume of some 1,080,000ft3 (30,582m3). Some of the spoil could have been used in a moat-side bank. The land enclosed may have been somewhat higher than its surroundings. There seems to be a slight fall to the north, beyond the town ditch. To the south the topography has been greatly modified, first by the later fortifications constituting the two outer baileys, and then by subsequent levelling in the post-demolition period (and probably also by dumping of dredging spoil), but the gardens of the houses south of these later fortifications lie some ten feet lower than the meadow that was the inner bailey. The volume of the motte, a truncated cone with a base radius of 100ft (30.5m), a top radius of 50ft (15.3m) and a height of 36ft (11m), would have been 660,000ft3 (18,689m3). The assumptions that the cross sectional 116

Figure 6.1 The Surveyor’s Plan, from the Wallingford Roll, Oxford, Bodleian MS Top. Berks b. 41, fol. 1. By permission of The Bodleian Library, University of Oxford. The watermark is visible towards the bottom of the sheet, just above the shelfmark.

area of the moat has not changed much — we know that it was dredged, later — and that the measured section is representative of the whole are difficult to defend, but the rough similarity between the volume of the motte and that of the excavated moat may suggest that the moat made the motte. For further discussion, see the conclusion below. An outline plan of the inner bailey of the castle, shown in Figure 6.1, the Surveyor’s Plan, was almost certainly made under the supervision of Roger Amyce, the king’s surveyor hereabouts. It survives and is now bound up with related documents in Oxford, Bodleian Library MS Berks Top. b. 41. There are in all 26 paper folia, some blank, which were produced by or for Roger Amyce between 1548 and 1561. Codicological examination shows that all the folia were kept together, and were at some point c. 1561 stitched together and given a membrane cover in which they were kept as a roll. The now-illegible title on the dorso of the membrane was copied with the rest of the text in the late 19th century into another manuscript, Bod. Top. Berks. c. 33, where the then still legible endorsement was recorded in Latin as: Wallingford Roll. Survey of

Wallingford made by Roger Amyce surveyor in the same place in the [third] year of the reign of our queen Elizabeth [1561], by the grace of God Queen of England, France and Ireland etc. The additional material in MS Top. Berks. b. 41 includes a lengthy written survey of the town and borough by Roger Amyce dated 1548 (fols 3–14), with interlined additions and revisions made in 1561, and a rental of the former priory lands made by him in 1552 on fol. 18. Whilst there is no formal proof that Roger Amyce was the instigator of the Surveyor’s Plan, on fol. 1, or that it was produced in the same year as the Survey of 1548, both are reasonable working assumptions. Amyce’s career as a surveyor lasted from 1547 to 1567, spanning the reigns of Edward VI, Mary I and Elizabeth I. Originally from Essex, he was a servant of Thomas Cromwell in 1538, but survived Cromwell’s fall to become the receiver of the dissolved estates of Glastonbury Abbey and Reading Abbey during 1540 to 1547. He was a founder of Christ Church School, Abingdon, in 1553, of which he was master from 1566 until his death in 1574. He was thrice 117

Figure 6.2 The Certificate of 1555 from the Wallingford Roll, Oxford, Bodleian MS Top. Berks b. 41, fol. 25. By permission of The Bodleian Library, University of Oxford

returned as an MP, for Reading in 1545, and New Windsor in 1553 and 1559 (Fuidge 1981). His survey of the town of Wallingford in 1548 is now also bound up with another written survey of part of the castle dated 1555, discussed here as the Certificate.

A good deal of the information in the original Wallingford Roll relates to revenues arising from Crown property, notably in the castle estate and its adjuncts. It appears to have been kept by borough officers responsible for overseeing the collection and recording of revenues,

118

Figure 6.3 Muniments chest of Wallingford Corporation, late 15th-century, now in Wallingford Museum (photo Judy Dewey).

some of which were paid to the borough as well as by it, and stored with various leases and other documents in the Corporation chest (now on display in Wallingford Museum, Figure 6.3). Rentals continued to be produced from time to time, now found in the Berkshire Record Office (Pedgley, below pp.227–31), but this one at some point passed into private hands, the most likely date being the time of the sale of the Crown’s estate in 1817. The new proprietor of much of the estate was Dr James Blackstone, son of judge Sir William Blackstone. If he acquired the Wallingford Roll at the time he acquired the estate, then it may subsequently have been sold by, or on the death of, his son William Seymour Blackstone, a former borough MP, briefly jailed for debt in 1854, who died in 1881. It was eventually bought by Mr John Mitchell Marshall of Wallingford from Sotheby’s in 1907, according to J. Field, who published its main text as Survey of Wallingford 1550 [sic] (Field 1917, 83). He died soon afterwards and Alfred C. Marshall subsequently donated both the original Roll and the late 19th-century transcript to the Bodleian Library in 1924 and 1928 respectively (Summary Catalogue II, 1205, nos 53722–3). It should be noted that John Kirby Hedges, who owned Wallingford Castle from 1858 to 1901 and wrote an important two-volume History of the town, was clearly unaware of either manuscript’s existence. His work cites only the Certificate, discussed below, and its appended account of the commons and town gates, indicating antiquarian Thomas Hearne as his source (Hedges 1881, ii, 96–7, 189–90; Keats-Rohan, above 70– 1). The same extracts had previously been quoted in the still unpublished Antiquities of Wallingford by Richard Skirmer (d. 1716), from ‘a Roll in the hands of Mr Stonor Crouch, which he did me the favor to communicate’, which was certainly a reference to the original Roll. Crouch was a burgess of the town and an attorney, but was controversially also a recusant. Nonetheless, he was noted as a constable of the borough in the visitation of 1665, and worked for the borough and in private practice as a lawyer until his death in 1723.

There are plenty of exquisite examples of threedimensional drawings of castles from this period (HKW, iv, plates 5, 48, 52), so the untitled plan drawing on fol. 1 of the roll must have a distinct purpose. Although there is no specific context for the plan in any document in the roll, the motive behind it was surely the same as that behind the written Certificate of 1555, discussed in detail below, which was the result of an inquest to determine the extent of remaining timber and stone in the inner bailey of the castle. Each was an exercise in quantity surveying. The Certificate detailed the type and quality of remaining stone, and counted and measured timber joists. Most Tudor surveyors after the Dissolution of the Monasteries began in 1536 found themselves engaged in quantity surveying. Wallingford Castle was no longer valued as a residence, but as a source of building materials. Much of its lead (10 tons and 58 pounds) was removed to Windsor Castle for the building of a conduit in 1554 (Tighe and Davis I, 165). That said, in the Surveyor’s Plan, only wall lengths are shown, with no indication of height and thickness. Those figures may have been recorded separately. The figures noted on the internal survey lines (dotted), some of them lengths and some degrees (as we shall see), may have formed the basis for calculation of the volume of stone; however, there is insufficient information for calculations, one of a number of signs that this plan was a first, or incomplete, draft. Nonetheless, an attempt to understand what the various figures can reveal about the layout of the inner bailey may yield something of interest to us, 450 hundred years after the castle’s demolition, above and beyond whatever they revealed to the surveyor. That is the task before us here. In approaching the Surveyor’s Plan we should be prepared to allow for a margin of error. The job of measuring the wall circuit of a dilapidated bailey surrounded by a wide and deep moat, which was waterfilled until the 18th century and even later, cannot have been easy. The Surveyor’s Plan was reproduced in the august History of the King’s Works (HKW iv, plate 20) and has 119

been printed shorn of its detail a number of times, most recently in 2013 (Christie and Creighton, 152), though a sensible summary in words was given VCH Berks 3, 531. The Certificate of 1555 is also very challenging. It too has both words and numbers, rather more of the former, but no drawn plan showing what was meant. Nevertheless, it is a slightly easier place to start for the reader. The importance of both these documents is that they are directly based on eyewitness evidence as to the layout of the inner bailey of Wallingford Castle, in the case of the Plan, and of the keep in the case of the Certificate, in the mid-Tudor period. This is some 500 years after the castle was first built and so we should not expect that it shows the ‘original’ castle. However, there is reason to believe that the major structural changes to the castle layout after the Norman period (1154 onward) mainly affected the two outer baileys and their gatehouses, as discussed by Keats-Rohan in this volume. At the very least, the Plan shows the ghost of the Norman castle. The certificate and the keep Written on paper with now damaged and uneven edges at bottom and left side, 266 x 303 mm, the Certificate of 1555 describes the keep in some detail (Figure 6.2). Although nearly two hundred years had elapsed since the Great Tower had last been described in any detail in one of the castle building accounts (Keats-Rohan 91ff.), the Certificate clearly refers to the same building and provides important information about it. It is not straightforward to read and its interpretation is challenging. We can only test its meaning by relating the contents to what is known of the layout of extant examples of such buildings. The Certificate was the work of three inhabitants of Wallingford. John Purdon was an alderman and had been mayor in 1552/3 (Wallingford court roll, BRO, W/JBc/25, f. 2). He was also a lessee of the priory by a grant of Henry VIII in 1538 (LPFD 13 i, 1309:16). Nicholas Payne, gentleman, was a burgess and a lessee of a royal watermill at the South Gate, by grant of Edward VI in 1553 (TNA, E310/8, no. 8), and also became a lessee of the nearby former hospital of St John’s in 1557 (deeds of St John’s Farm, BRO, D/EB T43). By 1568 he had become Clerk of the Scullery of Elizabeth I when his mill lease was renewed (CPR 1566–69, 170), and by 1588 Clerk Comptroller of the Green Cloth in the Exchequer (Sainty 1999). Thomas Hurst was also a burgess, and described as a carpenter in the accounts (1544–1548) for repairing the Great Bridge over the Thames (TNA, E101/490/8). All three are named in Amyce’s survey of 1548 in the same manuscript. The account of the keep reads as follows (the full text appears in the appended calendar as 2.1): Allso entering up the steres into the kepe a greate dore wth barres of Iron, and on the same steres the leade being taken awaye there remayneth of tymbre good and bad of Joyse lx conteyneinge in lenght viij foote, and ix inches square the one waye and vj the other. Allso at the upper ende of the same steres one othe[r] greate doore wth barres of Iron and a porche

adioyneinge unto the same haveing a flower on it of ix Joyese cont in lenght viij foote and vj and viij Inches square oke. Allso in the kechyn in the same kepe ij wyndowes with barres of Iron, and ij roufes with a chaimbr on the same, ij dores with Iron hynges and v Iron barred wyndowes, and of Joyses, in both flowers xlvij of x foote in lenght, and square vj and viij Inches. Allso ij donngons or prysons within the same kepe haveinge iij wyndowes wth barres of Iron iij great doores barred wth Iron and one locke, iiij greate haspes of Iron and stapells to the same haveinge a flower with lx Joyses of tymbr cont in lenght xix foote, and square viij Inches the one waye and x the other . . / Allso the greate chambr and pryve chambr with iiij verye great Iron barred wyndowes, haveing one flat rouf and the same of tymbr the Joyses wherof be lix of xix foote longe viij and x Inches square on both sydes. The first part of this description relates to the entrance stairs, which were carried externally to at least the first floor, inside a porch or forebuilding whose roof beams or joists were eight feet long (i.e. the staircase was about 8ft (2.4m) wide). Here we should point out that the stairway described is not the stairway shown on the Surveyor’s Plan, to be discussed hereafter, shown as starting at bailey level and rising along a curtain wall. The Certificate uses the words ‘there remayneth’, so there had possibly been more joists originally. More likely, given the regular pattern in the rest of the account, it simply means that the lead has been removed (to Windsor Castle), but the timber remains. The stairs would have been of stone on a stone base, with a landing at the top. Sixty-nine joists were used to roof this structure – 60 over the stairs, which must therefore have been at least 30ft (9.1m) long, and most likely 65ft (19.8m), and able to be taken to a good height. Similarly the landing must have been at least 4.5ft (1.4m) long, as it had a floor on it of nine joists, not called a roof. If this was in fact a floor, there may have been a room or a fighting platform above the landing. The joists over the stairs, though more numerous than those in the keep, were external to it (they were lead-covered) and almost certainly packed much more closely together. If the stairs were merely steps on the slope of the motte, duplicating the ascent of the steps along the curtain wall, they would not have been roofed and the timbers would not have been described as joists by a carpenter. At the foot of these stairs was a great door with bars of iron, which suggests by its great physical strength that it is guarding an entrance to the keep – they are ‘the steres into the kepe’, not ‘the steres to the kepe’ nor ‘the steres in the kepe’ – and at the top was ‘one othe greate doore’, all of which emphasizes that the stairs were an external structure. Certainly what is described is a forebuilding (see Figure 5.14). By the time the Certificate was made, the allocation of floor space inside the keep may have changed very much since the early days and the three men may have been mistaken about the rooms’ uses. Their intention was not to describe the keep but to assess the quantities of timber and 120

stone there. The preamble to their report uses the words ‘not yet defassed nor pulled downe’, strongly suggesting that the castle was in active decline in 1555. The earlier, Norman, keeps (and this may well have been one) were commonly square or rectangular, with an internal crosswall dividing the internal space into one or more rooms of unequal size. The purpose of this wall was to permit the use of shorter joists and to produce rooms of the shape and disposition that society desired. It is often assumed that the larger room functioned as a hall, with a small chamber adjoining, though little is known for certain about how the rooms were used. One constant is that the entrance via the forebuilding was always into the larger room (Gravett 2003, 12). By the late Middle Ages, such a building would contain the lord’s private residence, withdrawn from the bustle of the bailey below. The Certificate itemizes a number of rooms, whose functions we cannot assume the assessors truly understood, as there is little evidence that the building had been much used as a residence for well over 50 years by this time, though there are indications that it had been used as a prison in 1532 (1.76) and perhaps later. There are references to 60 joists, 19ft (5.8m) long and 8 x 10in (0.20 x 0.25m) square, and 47 joists 10ft (3.05m) long, 6 x 8in (0,15 x 0.20m). There is also a porch of 9 joists 8ft (2.44m) long. The joists spoken of throughout the Certificate form groups with regular dimensions, thus they all relate to floors or ceilings within the keep and not to roof structures, which would have required considerably more variety in timber lengths. At first sight, the description of joists is suggestive of a quadrilateral building with an off-centre crosswall. One side is narrower than the other in a ratio of roughly 1:2. The narrower side has fewer joists across its span. It would be natural for the three men involved to count the joists from below but to measure them from above. If there was a lower storage or dungeon level (cf. Keats-Rohan above 92), it was not explicitly mentioned, but if the measurements for both ‘floors’ were measured in this way, then a floor below what was normally a first floor forebuilding entrance into the keep becomes implicit. There is certainly abundant evidence that major stores, including armaments, were housed in the keep (KeatsRohan above pp. 95–7). It appears that three main areas are being described, the first of which is a two-storey forebuilding, as discussed above, the other two being the rooms on two floors. There were two principal rooms on the entrance floor, one of which is described as a kitchen and the other as a prison. The latter is the larger of the two, which normally would indicate that this was the room into which the visitor first stepped on entering the keep; if so, at some point this room is more likely to have been a hall. The kitchen is described as having two floors, that is, storeys, and two roofs, each with a chamber above. The word ‘roof’ can refer to both a ceiling and the area amongst rafters. If it is not simply duplicating the idea of two storeys, which is possible, this may indicate that the kitchen had an internal wall, presumably of light wattle and daub construction, or it could simply refer to internal divisions of the kitchen space, possibly on different levels (mezzanines), such as

in the considerably larger example at Dover. The upper storey may have been similarly divided. If the joists were of the same separation on both sides of the cross wall, then this side was shorter than the other, perhaps indicating internal divisions of stone rather than timber. On the other side of the cross wall, at entry level the surveyors report two prisons, and on the upper storey the great chamber and privy chamber. Three windows and doors are mentioned for the lower floor; on the upper floor there were four ‘very great’ iron barred windows, which might indicate that this space included the former chapel. The roof on this side is expressly described as flat, in contrast to the two roofs mentioned for the other side. Since the description cannot refer to the roof of the whole tower, this should perhaps be understood as an unusual feature of this upper floor. A possible explanation is that the non-flat area accommodated the higher-pitched roof line of the chapel. Evidence from the castle accounts shows that the Great Tower (keep) had a chapel, associated with a chamber, large enough to take 40ft (12.2m) of glass (Keats-Rohan above, pp. 93–4). Interpretation of the written account requires us to consider normal layouts of this type of building, a foursided tower rising from a motte top. The dimensions of the rooms described depend upon the separation between the joists, whether they were set into their wall sockets on their broader or narrower sides, and whether there were double or triple centre beams or the joists abutted walls at either end. The table below (Figure 6.4) gives the various possibilities. It immediately becomes obvious that the joists described must span both of the rooms (kitchen 1 and kitchen 2 on one side of the cross wall, prison 1 and prison 2 on the other side) on the entry floor. The joists described are massive; such timber, from tall mature trees would have been very valuable. Allowing for a central beam and assuming a joist separation of 24in (0.6m), we arrive at an internal measurement of 59.3 x 39ft (18.1 x 11.9m) for the prison side, and 46 x 21ft (14 x 6.4m) for the kitchen side. The shorter side would accommodate the chapel, which is not expressly mentioned in the Certificate. If we further assume a 5ft (1.5m) thickness for the cross wall and 10ft (3m) thick walls all round, we arrive at external measurements of 79 x 85ft (24 x 26m). A joist separation of 20in (0.5m) would give overall internal measurements of 50 x 60ft (15 x 18m). These figures are speculation of course, but they are fair; they suggest that the building was more square than rectangular, but it is impossible to be certain. Such an interpretation makes the cross wall run in an east to west direction, as shown in Figure 6.6 below. Whichever possibility is nearest the actual size of this long vanished building, the dimensions and descriptions given by the surveyors exclude the notion that Wallingford had a shell keep (Figure 6.5), as has normally been assumed. The top of the keep’s well, capped in Victorian times, survives on the much eroded north-east face of the motte. It is now right on the edge of the top, the well shaft in fact jutting out somewhat, showing that the northern flank of the motte has been lost (see Figure 6.15 below). This is the bailey side, and it is possible that as the keep was dismantled in 1654 its stones, some from a great height, 121

122

123

were simply tumbled down to carts waiting there, breaking up the structure. Camden claimed to have seen this well on going up the tower in 1586 (2.3). Originally internal, it may have been accessible close to the point where the forebuilding entered the keep, as at Dover. An internal stair, presumably a spiral staircase, occupying the stair turret mentioned in the 1327 survey of the castle (KeatsRohan, above p. 69), would have been remote from this entrance, though not necessarily on the other side of the keep. The stair turret is referred to in the Certificate when it says that: ‘the keep, with the long and winding stairs, are of malme and freestone’ (2.1 below). Another, intramural, internal spiral staircase would have been needed to provide access to the well from the kitchen side. The description of the four very great barred windows on the prison/great chamber side fits well with what is known about domestic architecture of great residences at this time. On the narrower side there is a space about 21 x 14ft (6.4 x 4.3m) that is unaccounted for by the surveyors. It quite possibly fixes the location of the chapel known to have been in the keep, which probably rose through both the first and second floors. Could there have been an apsidal end on this side? Colchester and the Tower of London are just the most august examples of projecting chapels, a common feature in great towers of which this could be a scaled-down version. Whether this was true or not we simply cannot know because we have no information about wall thicknesses or other variables. We can only note that the chapel is well evidenced from earlier documents, and would normally have been accessible to visitors entering from the forebuilding into the hall, and also from the second-floor great chamber, on the further side of the cross-wall. All this helps to determine the orientation and layout of the keep, which was the east to west orientation common to great towers, as suggested in Figure 6.6. The Surveyor’s Plan

Figure 6.4 Table showing possible joist values

The plan (Figure 6.1) is drawn in ink on paper, some 8in or 9in across. The paper has in the past been separated into two complete sheets of half the thickness and a leaf of plastic has been sandwiched between them, a conservation process now deplored. The plan shows two enclosures, a circular one that can be interpreted as the region of the motte and keep, and a polygonal part that is surely the inner bailey. Walls, though they had thickness, are shown as lines, and at their corners are towers. Except in their back walls, the corner towers are drawn as squares. At certain angles are pin-holes (five at most curtain towers) and the ink has collected round some of the holes and dried there. The pin-holes suggest that a trial plan was drawn on an over-lying sheet and the information very carefully transferred to that below. Some dotted lines have had similar treatment, the line having been scored by an instrument, presumably applied to the sheet above, the very faint furrow in the lower sheet then being delineated by peck marks of ink. This is no rough sketch dashed off by a casual traveller but a serious attempt by an expert observer to record what was there. Moreover, although it 124

Figure 6.5 The shell keep at Restormel Castle, Cornwall. (Image: Public Domain)

Figure 6.6 Suggested floor-plan of first and second floors of the keep at Wallingford Castle (drawing: Debbie Miles-Williams)

125

Figure 6.7 The Surveyor’s Plan, showing the figures given on the original; lettering has been added to aid the discussion in the text. Note that the scale of the original is internally inconsistent, making exact replication of the drawing virtually impossible.

may have been made as a record of quantities and value, it is not a purely schematic diagram. To assist the reader with the interpretation Figure 6.7 gives all the figures and the lines to which they refer. At this time, the mid-16th century, the ordinary surveyor’s principal instrument was a measuring rod or a knotted rope. The theodolite had recently been invented and used abroad (Skelton 1970: 81). Its precursor, the cross staff, was well-known, but probably appeared more often in the text-books of the day than in the field. Some cross staffs could be read in degrees, and had inter-changeable cross pieces to cover different angular ranges. The surveyor’s cross, used since Roman times, was a simple

cross used to lay off right angles, but more up-market models consisted of a hollow metal cylinder with sighting slits at 90 degree intervals, sometimes at 45 degrees. This was a time of transition and great change in surveying (Derek 1955; Skelton 1970; Dudley 2006). Measurements were most easily made on level ground, but the keep sat on a hill. It does not appear on the plan, but is described in the 1555 survey, as was discussed above. Distances are shown on the plan in unspecified units, mostly, but two are given in Latin in poles and feet as follows: (a) 14 pol 8 ped

126

which amounts to 239ft and is repeated nearby as 240, and (here, for convenience, with a modern ampersand): (b) 16 pol ½ 4 ped & qt or 276ft and 3in and a further quarter foot (the qt), repeated along the wall as 113 + 16 + 148, totalling 277. Clearly the units are feet, and for that reason feet and inches are used in this present account. There had been a competing unit, of 13.22 modern inches, but that was suppressed by statute in 1439 (Zupko 1975). There was another foot, a short foot of 11.61 modern inches, that is said to have been expressed in some medieval English buildings. The discrepancy between 239 and 240 is not important, given the irregularities to be expected in such a structure, but the reason for it is not obvious. Conversion from feet to poles requires division by 16.5, and that is trickier than the inverse operation, so perhaps the actual measurement was 240ft, made with a rope. The discrepancy between 277 (which is the sum of three integers, each the result of rounding) and 276.25 is understandable. This last is the only measurement that is not in round feet. Again, if the actual measurement had been made with a rod, that final quarter foot would have been discarded. The fact that a half-foot was kept may mean that the surveyor was unable to decide whether to round up or round down. The three integers could well have been obtained in a separate measurement exercise; that might show that although there is a small discrepancy, there was no gross error. On the plan the motte circle is touched by three dotted survey lines, all of which arise from measurements within the bailey. Strictly, these lines do not explicitly show that a circular wall stood at the foot of the motte, at bailey level, but they come very close to doing so, and two particular points on them do suggest that. These points are the junctions where the long line is met by the 45-degree lines (FGZ, HIJ), and they lie no more than eight or nine feet from the wall (as seen in plan view). Banks are rarely steeper than 45 degrees, so the motte slope did not extend far outside the circle, and perhaps not at all. Accurate measurement required that the two points lay on level ground. Moreover there is another line, of length 68ft (FD) that runs to a point on the circle, and there is another measurement line running from that same point (DG). That point must have been at bailey level, not on its flank or on top, or the measurements would have been impossible. The tangential lines show that the circle represents the stone outer face of whatever surrounded the motte ‘circle’ at its base, or perhaps it simply represents the limits of an area dominated by motte and keep. We can further extrapolate from this that the purpose of the sketch was to indicate stone buildings, whether walls, corner towers or buildings. Only two buildings are shown. The one on the north wall of the bailey can be identified as the Great Chapel of St Nicholas from the written account of it in the Certificate already discussed. The other, smaller building in the southwest corner is unidentified. It is too small to have been any of the three known halls (Keats-Rohan above, pp. 84–5). There is no attempt to represent the stone keep described in detail in the Certificate, which may be an indication that the sketch was not completed. Building accounts from the 13th to the 16th century show that a great deal of stone as

well as timber was used in all the main buildings of the bailey. Very difficult to detect on any reproduction of the original today, but still visible on the original, the surveyor indicated that the whole bailey was surrounded by a wet moat, using the time-honoured symbol of the wavy line. This confirms what we should infer, that measurements on the curtains cannot have been made from outside the walls, because they reach to the re-entrant corners of the towers, but they could have been made along the wall walk, and that would have permitted access to the outer face of the walls, or almost so, though somehow the bailey measurements would have had to have been tied in. We have to take the features as lines on a plan, but if the measurements were made along the wall walk (and any lean-to buildings would have made this necessary) then probably the wall walk ran between the towers, the towers were backed, and they rose higher than the curtain. It may also be noted that no separate figures are given for the dimensions of the towers, except for one, which possibly had been rebuilt (see below); that measurement gives us, roughly, the dimensions of the towers. Their floor-plan would seem to have been about 25 x 25ft (7.6m). We know from the building accounts for 1389–90 that the east curtain wall was 14ft (4.3m) high (Keats-Rohan, p. 84), so the towers were anything upwards of 20ft (6.1m) high. The numbers in detail Now let us look at the numbers in more detail. With the circle at the bottom (and this orientation will be maintained in this description) to its right there is a rightangled triangle with sides 88, 88 and 128 (HIJ). If the 88, 88 are correct, then the hypotenuse should be 124.45, and this would have been glaringly obvious to the surveyor, as such geometry was basic to his training (cf. Shelby 1972). Luckily, he seems to have written down what was in his notes, without trying to improve them. From the general scale of the plan it can be found that the short sides of this triangle have been drawn about right, and the drawn hypotenuse length is about 123.8. The scale of the plan can be found from the wall lengths, which give individual scales agreeing to about 2 per cent. (On one photocopy of the plan the scale – based on stated lengths of lines – varies from 27.6 to 28.4 feet to 1 cm). The radius of the circle can then be measured to be 120.6. But that is quite wrong, as will now be shown. Consider the line of length 276.5 that runs tangentially past the circle (FGYHI). The length of the centre section (GYH) can be found by subtraction to be 84.5, give or take 1, and the circle fits snugly against that and the two lines at 45 degrees (HJ, GZ), as it must. The circle that so fits has a radius of 102.00, and this value accords with features on the ground today. The radius is given by Radius = D/(2.tan(22.5 degrees)), or D/0.82843, where D is here 84.5 ft. The problem lies in the number 460 attached to the circle, which simple measurement on the plan (perhaps using a length of string) shows was intended by the draftsman to apply to the longer arc between the two 127

Figure 6.8 Possible locations of the curtain walls on the western side of the inner bailey.

points where the curtain joins the chemise. This number is in error, and the whole plan is distorted by it. In fact the whole circumference is truly about 640, and the most common mistake made in copying numbers is said to be the transposing of adjacent digits; but 460 having been written, it could then only be applied to the arc. (Applied to the whole circumference, the circle would not fit snugly against the two 45 degree lines that should fix its position; it would not even reach them. The over-sized circle ought not fit either, but the drawing has been fudged by drawing the central 84.5ft section as 97.7ft long.) To the left of the circle is another right-angled triangle, with sides 104, 104, and the hypotenuse (apparently not measured) of 147.1 (GFZ). The shortest distance from its right-angled apex to the circle is 76.25, but a measurement of 68 is shown on the plan. This is another serious error, occurring as it does in such a short distance. If the 68 is correct, that triangle must be much smaller. In fact, the 68 would sit easily within the 88, 88 triangle on the other side (HIJ). It looks as if the surveyor got his notes muddled. Allowing for these errors (that is, discarding the 128, setting the radius at 102.00 and for the present disregarding the 68), it may be possible to draw the plan that the measurements imply, after dealing with two items. First, the number 140 that appears at the centre of the circle. It is not a distance: it connects nothing. Nor is it a height: the motte today is some 35ft (10.6m) high and, because its top would be too small, never was much higher. A tower of 140ft (42.7m) would probably have sunk straight through it; and the surveyor measured no other heights. It is an angle, in degrees, the angle subtended at the centre by the shorter arc between the points where the curtain reaches the chemise (abc), and it is so drawn, quite accurately. It would have been deduced from measurements of the lengths of the shorter and the longer arcs, made around the wall walk of the chemise. If the

Figure 6.9 Possible angles of the modified NW tower.

surveyor used a long pole, that would have been awkward, but he probably used his rope. It was not measured with some angle-measuring device: from the middle of the keep roof (which was not necessarily the centre of the chemise circle) the chemise wall would not have been visible. Secondly, the length of new wall at the presumably rebuilt north-west corner, which has been read as 18 and as 25. Experienced readers all express uncertainty, but the consensus is that the first number is a 2. The deciding detail is the seriph base. On the 1 in the circle it is drawn from left to right without lifting the pen after the down stroke, while on the 2 in ½, and on the uncertain digit it seems to run independently from well to the left of the down stroke. (The draftsman was quite obviously right-handed). The second digit is clearly a 5 and not an 8 (of which elsewhere there are twelve examples), but unfortunately there is no other 5 on the plan. That length will be taken to be 25ft. As such it accords well with the scale of the plan. The plan can be drawn satisfactorily along the top, north, and on the right, east, side and the circle can be placed without much uncertainty. The point on the left of the circle where the very short curtain begins can be fixed by making use of the 140 degree angle (bE). But the three curtains on the left (aE, EU, SR) can be drawn in an infinity of positions, as seen in Figure 6.8. Except for one possible anchoring line, discussed later, the measurements were too few, and it is necessary to plump for a value of one more quantity. The most convenient quantity to fix is the angle between the curtains at the presumed re-built tower, at the top left (RA). (The curtain lines must of course be projected to 128

meet, as they do at the other towers.) An extreme value would be 90 degrees. An acute angle would be absurd on this site and in fact the angle is clearly shown to be obtuse; but 90 degrees is the lower limit. Setting that angle to 90 degrees leads to one of the dashed lines in Figure 6.9. The other extreme is an angle of 116.3 degrees, because the kink at the end of the very short curtain then disappears: the line becomes straight. This would make the tower (EC) there redundant, and in any case a marked corner is indicated. But this is the other extreme case, shown by the other dashed line. It is worth saying again that these extremes are absurd extremes, and not credible. The true line lies somewhere between them, but where? We should look imaginatively at the plan, and notice that the (virtual) corner at the top left (north-west tower) and that at top right (north-east tower) have been drawn with very similar angles. The right has 107.8 degrees, found from trigonometry, and the left seems to have an additional degree (108.8o). So let us take the unknown left angle to be 108 degrees too. The result is the full line. The plan is now not very different from that drawn four and a half centuries ago, though it is more soundly based. But there is something wrong. It is lop-sided. We can look again at the unresolved faults. First the disagreement with Pythagoras, where the 88, 88 triangle (HIJ) cannot have a hypotenuse of 128. Secondly the length 68 (FD), which in that location must be (taking the other figures at face value) at least 76.25. That is weakly dependent upon the value chosen for the circle radius, but a credible lower limit is 75.2. The 68 value would fit

into the 88, 88 triangle (HIJ), and that would resolve this difficulty. But then either the 68 length must be transferred to the right of the plan or the 88, 88 triangle and the 104, 104 triangle (GFZ) must be transposed. A choice must be made, and the action must be taken. This is inescapable. The first choice is not very interesting. As Figure 6.10 shows, nothing much changes, but another distance, not yet mentioned, seems to serve no purpose. The second has a significant effect: the circle is shifted 16ft to the left and now lies quite centrally – just 15 inches awry – relative to the top curtain (when that curtain is extended to the virtual corner, BA10), and a degree of symmetry seems to be appearing. Part of that symmetry comes from choosing the top left angle to be the same as the top right, but it was drawn like that on the original. But the shifting of the circle has altered other things too: the two points where the curtain meets the chemise circle (ac) are now roughly equidistant from the top curtain (AB), and the angle between the two sections of the east curtain (BCD), at the south-east tower, is now about 108.2 degrees. The angle at upper right (north-east tower, CBA), it may be remembered, is 107.8 degrees. This near equality is quite remarkable and perhaps fortuitous: the surveyor could not have measured more accurately. At this point it should be recalled that a regular pentagon is a figure with five equal sides and five equal angles, and those angles are each 108 degrees; and now on the modified plan there are two essentially equal sides and three angles of about 108 degrees. Can the plan of the castle have been based somehow upon a regular pentagon? Or is all this merely an artefact of an inaccurate attempt to plan field measurements?

Figure 6.10 The Plan redrawn as a model pentagon (drawing by June Strong).

129

Geophysical survey and the true plan At this point we should stop and consider a wholly different plan, this one derived from recent archaeological survey of the site. Described by modern archaeologists as ‘sub-rectangular’, the bailey outline clearly has five sides of unequal length (cf. Figure 2.2 above). The shape of this irregular pentagon is similar on the north and west sides to the Surveyor’s Plan, but very different on the eastern side. In fact, investigating the figures on the Plan as if they were intended to help place a circle within and upon a regular pentagon demonstrates how the surveyor went wrong in producing his plan, as Figure 6.10 shows. Recent earthwork and geophysical surveys have given us a clearer picture of the configuration of the bailey walls. Good aerial surveys are available from a time when the motte site was in far better condition than now (see Figure 6.11). What all these new sources show is that the outsize circle of the plan never was an accurate approximation of the motte shape, and that the lines drawn to the southeast tower and from there to the motte circle are wholly inaccurate (see Figure 6.7, NLc). At the north-east tower the surveyor has drawn a survey line at right angles to it. If that line is increased to 95 degrees and extended towards the bottom of the figure, it will follow the line of the east curtain wall suggested by the resistivity survey (Figure 5.1 above). It also coincides far better with what the resistivity suggests could be the original south-east tower, and provides a credible location for the staircase that started at bailey-level and rose along the curtain wall (not to be confused with the stairs to the forebuilding). Perhaps the

biggest single stumbling-block to an understanding of the Plan as drawn is the impossible location of those baileylevel stairs. The scale suggests that they would have extended for about 60ft (18.3m), part of which would have risen against the slope of the motte. Another check on the reliability of the Plan is provided by the estate map of 1804, so illuminatingly discussed by Judy Dewey below (pp. 152–5, Figure 7.48). This map shows that the moat surrounded the southern half of the motte, running more deeply into the bailey on the eastern side, something confirmed by flooding to this day (Figures 6.12, and 6.13). It also shows that the sub-circular area of the motte is nothing like as dominant as the surveyor’s motte, occupying no more than a third of the bailey in its northern extent, and a half in its east-west extent. We have to rethink the Plan. An interesting picture (Figure 6.14) emerges if one first makes a copy of the Plan and continues the south-west line marked as 60ft through the bottom of the circle (AB). At this point, cut out the whole figure, and then cut out the motte circle, together with its moat. Taking care that the moat section resembles the moat line shown in the 1804 map, place the circle so that its southern arc touches the new south-east to south-west line and that it forms a tangent on the east with the survey line from the north-east tower (DE). The survey line will be seen to pass through the bailey-most side of the stair as drawn on the Plan. From the end of the curtain-wall part of the stair draw a line to the north-east tower. The line of the east curtain wall and the location of the south-east tower are now in more credible positions when compared to the resistivity survey. The new line is about 240ft — the same distance

Figure 6.11 Aerial view of the motte (Crown Copyright 1988).

130

Figure 6.12 Detail of an estate map of 1804, showing the moat around the SW tower, and the south of the motte (courtesy of The National Archives, MPEE 1/10)

Figure 6.13 Floodwaters round the southern base of the motte in 2013, highlighting the location of the SW tower as seen in the 1804 map (photo: Katharine Keats-Rohan)

131

Figure 6.14 The redrawn Plan (image Debbie Miles-Williams).

from the north curtain to the circle tangent shown on the Plan — and agrees roughly with the resistivity. The new position of the south-east tower suggests that the gatehouse leading into the bailey from the drawbridge should be set at right angles to the opening indicated on the plan, which would reflect the layout of many surviving castles. The line of short curtain going to the south-west tower is also shown to have been more than twice the marked length of 60ft. Although only a crude check can be made, it also seems that the angle of tower and wall at this point, 110 degrees, is in line with the 1804 map. Essentially the same result occurs if the circle is redrawn with the correct radius of 102ft, maintaining the 140 degree angle subtended at the centre by the shorter arc between the points where the south-west and south-east curtains touch the circle (abc), as seen in Figure 6.14. What emerges very clearly is the importance of the south-east tower as a bastion or barbican guarding the entrances to the keep, and overseeing the middle bailey and its gatehouse (Dewey below, pp. 158– 9). Very likely the curtain wall staircase concealed the entrance to the sallyport tunnel, running under the moat, and emerging in the middle bailey. This makes the best sense out of the location of the tunnel described by 19thcentury owners of the site and marked upon early OS maps (Keats-Rohan p. 46, and Figure 7.8), and further illustrates the importance of this tower. What can we conclude from this? Actually, comparison of the Surveyor’s Plan with the redrawn plan, the 1804

map, the resistivity, and the final conclusions about layout reached by Keats-Rohan and Dewey and shown in Figure 7.1, clearly reveals a close relationship between them all. Remarkably, most of the surveyor’s wall lengths broadly conform to the resistivity. Field survey and observation indicate that the perfect compass-drawn circle was never going to be an accurate representation of the motte, but it was not completely inaccurate. It was drawn as a means of representing both the area dominated by the motte and its enclosure as it then was and a convenient way to represent the surveying data. As such, it left something to be desired: as drawn, the moat on the eastern side flows down a slope and the south-east tower has tumbled down with it (though it has something to tell us about water flow on the site, as revealed by Dewey, pp. 167–72). The fact that so little of the stone building in the bailey, or the keep and its forebuilding, is shown may perhaps indicate that the surveyor, or perhaps more likely the assistant (probably his ‘aide in London’ John Denison; see the Appendix below) who was trying to decipher his superior’s field notes, realized that bad mistakes had been made and simply abandoned the drawing. Presumably he made another, because the job was not being undertaken for recreation purposes, but if so, it has not come down to us. The simple enclosure that is the bailey is what may be expected of a very early Norman castle. At that time perhaps the perceived danger was a secret night rising, simultaneous and country-wide, and the defence was castles: small

132

fortified villages, each walled and guarded. Even the loss of the horses would have been a severe blow. Anyone doubting the practicality of such a rising should ponder the (Indian) Great Mutiny of 1857. But the Saxon leadership had been destroyed in battle and the outcome of the contest expressed the will of God and that rising never came. Later, of course, the new rulers built castles not against the Saxons but against each other. Many of those castles present a higgledypiggledy chaos of long-continued accretion, but the simplicity and coherent uniformity of design at Wallingford suggest that the inner ward of the castle that existed in the sixteenth century had been built as a whole at one time. The rectangular corner towers are typical of both Norman and 14th-century work. The relatively abundant documentation from the 14th century suggests maintenance work rather than rebuilding. Although the castle was of course modified continuously as long as it was occupied regularly, up to the end of the 16th century, there is no documentary indication of major rebuilding of the walled circuit of the inner bailey from the 12th century onward. There may have been changes during the time the castle was occupied in the 1640s, but we know nothing for certain. The potential for archaeological survey is severely limited on the southern side of the bailey because of major landscaping by post-demolition users and owners of the site (Dewey, below). That we know what to look for when future opportunities arise will in large part be due to the Tudor surveyor and his fudged attempt to place a sub-circle inside an irregular polygon. Conclusion and afterword K. S. B. Keats-Rohan The foregoing discussion was principally the work of my late husband, Dr John L. Lloyd, a particle physicist who died in September 2004, having studied the Tudor plans for many years, primarily by site survey, using a theodolite and compass, and by mathematical analysis. He is the only person ever to have made a serious study of this material. His edition and analysis of the Certificate has been key to my attempt, in the previous paper, to interpret aspects of the medieval building accounts, requiring as it did a complete reassessment of what had been assumed about the great tower or keep at Wallingford. His perception that the Plan revealed the castle to have been laid out as a regular pentagon did not sit too comfortably with his view, surely correct, that the Plan preserves major elements of the original Norman castle, since so regular a plan would be quite unparalleled at such a date. He knew nothing, of course, of the great discoveries made in recent years by the Burh to Borough Project and the historical work that has accompanied it. This valuable new evidence has led me to conclude that the regular pentagon he discerned was one concocted by the man who was struggling to draw the Plan with what appears to have been defective data, rather than the design of those who built the early castle, even though they too would have thought in geometrical terms. He knew nothing of enmotted keeps, which provide the only satisfactory solution to the problems presented by the data relating to the keep described in the Certificate of 1555 and the documents discussed in my paper above. The building has joists which could only have fitted in

a quadrilateral building and indicate that this was not a shell keep. Yet, as he pointed out, a tower of any great height – and Wallingford’s was at least two storeys high and almost certainly three, of solid masonry – would have sunk straight through the motte. But it did not. So what prevented it? His estimation from the moat lengths and depths of the available spoil was only a little over half of what would have been required if the motte had been a compacted mound (even allowing for the well chamber which ran through it). A masonry core, an enmotted keep, would account for this discrepancy. A final unanswered question remains: what sort of stone circuit was intended by the surveyor in his drawing of the motte. Was it a chemise wall, or rather, as I have suggested above, a mantlet wall containing the stone or timber revetted motte slopes enclosing a tower? Acknowledgements John Roche, Jim Bennett, Charles Coulson, Chris Lewis, the R.A.F. at Benson, Mr Ryall the farmer, (especially) my wife Katharine Keats-Rohan, Clare Brown and her colleagues at the Bodleian, and others have helped me in this study, some unwittingly, and I thank them (JLL). My thanks to Colin Harris and Bruce Barker Benfield of the Bodleian Library for advice on the manuscript, to Gerard Latham for the calculation of joist values, to Gerard Latham, David Pedgley, David Roffe and Anthony and June Strong for invaluable comments during the difficult process of revising this paper for publication after John Lloyd’s death, and to June Strong and Debbie Miles-Williams for help in drawing the figures (KKR).

Documents list 2 2.1. 1555 The Certificate, Bodleian MS. Top. Berks b. 41, transcr. J. L. Lloyd. Lost readings at the end have been supplied from the version printed in Hedges 1881, ii, 97. A certyfycate of soche tymbr and stone, as remayneth of any of the howses wthin the Castell of Wallingford as yet not defassed nor pulled downe made by us John Purdon Nicholas Payne and Thomas Hurst carpenter this xv th of Auguste Anno Secundo et Tertio Regni Regis et Regine Phillipi et Marie, et Anno Dni 1555. as followeth That is to saye Fyrste a longe enterye or gatehowse going into the iner warde called the leftenantes loginge buyllte of tymbr and covered wth tyles conteyneinge in lenght xl foote, and in breadth xvj foote. and one flower in the same haveinge xxxvj greate Joyses of xvj foote longe and of vj Inches square on the one side and ix Inches on the other. Allso the bodye of the colleg churche there conteyneing in lenght lix foote and in breadeth xxx foote, haveing a stepe rouf of tymbr covd wth tyles is buyllt wth mallme and free stones and hath ij wyndowes yet lefte wth Barres. of Iron Allso there is a fayre square tower or stepell adioyneing unto the said churche where in the bells dyd hange conteyneinge square on every syde xxv foote buylte wth

133

verye good free stone, haveing wthin the same iij flowers and one flatte rowfe in the toppe of good oke conteyneing square xix foote, not as yet defassed nor pulled downe, and iiij wyndowes wth barres of Iron. Allso entering up the steres into the kepe a greate dore wth barres of Iron, and on the same steres the leade being taken awaye there remayneth of tymbre good and bad of Joyse lx conteyneinge in lenght viij foote, and ix inches square the one waye and vj the other. Allso at the upper ende of the same steres one othe[r] greate doore wth barres of Iron and a porche adioyneinge unto the same haveing a flower on it of ix Joyese cont in lenght viij foote and vj and viij Inches square oke. Allso in the kechyn in the same kepe ij wyndowes with barres of Iron, and ij roufes with a chaimbr on the same, ij dores with Iron hynges and v Iron barred wyndowes, and of Joyses, in both flowers xlvij of x foote in lenght, and square vj and viij Inches. Allso ij donngons or prysons wthin the same kepe haveinge iij wyndowes wth barres of Iron iij great doores barred wth Iron and one locke, iiij greate haspes of Iron and stapells to the same haveinge a flower with lx Joyses of tymbr cont in lenght xix foote, and square viij Inches the one waye and x the other . . / Allso the greate chambr and pryve chambr wth iiij verye great Iron barred wyndowes, haveing one flat rouf and the same of tymbr the Joyses wherof be lix of xix foote longe viij and x Inches square on both sydes. And the kepe itsellf wth the longe and wyndynge steres the walls of ij wardes and ij gatehowses ar buyllt of mallme and free stone, and for the [build]ings the kyn[ge] quenes maiesties, moche [good] stone [will be of the same.] 2.2. c. 1540 Leland’s Itinerary, i 118–19 The toun of Walingeforde hath beene a very notable thing and welle wallid. The ditch of the toun and the crest whereon the walles stoode be yet manifestely perceyvid, and begin from the castelle going in cumpace a good mile and more, and so cummith to Walingford bridg a large thing of stone over the Tamise. There remayne yet the names of these streates emong other: Tamise-streat, Fische-streate, Bred-streat, Woodstreat, Goldsmithes-row. And by the patentes and donations of Edmunde Erle of Cornewaul and Lord of the Honor of Wallingeford [it appeareth] that ther wer 14. paroch chirchis in Walingford. And ther be men yet alyve that can shew the places and cemiteries wher yn the al stoode. At this tyme there be but 3. poore paroch chirchis in the town. Ther was a priory of blake monkes, a celle to S. Alban, suppressid by Thomas Woulsey cardinale, standing hard withyn the west gate of Wallingford. The toun and the castelle was sore defacid by the Danes warres. Yet they meatly reflorichid in the tyme of Richard King of Romaines and Erle of Cornewaulle, brother to King Henry the 3.

This Richard did much cost on the castelle. The castelle yoinith to the north gate of the toune, and hath 3. dikis, large and deap, and welle waterid. About ech of the 2. first dikis, as upon the crestes of the ground cast out of, rennith an embatelid waulle now sore yn ruine, and for the most part defaced. Al the goodly building with the tourres and dungeon be withyn the 3. dike. There is also a collegiate chapel emong the buildinges withyn the 3. dike. Edmund Erle of Cornewale, sunne to Richard King of the Romains, was the first founder and endower of this college. Prince Edwarde, as one told me, the Blak, augmentid this college. There is a decane, 4. prestes, 6. clerkes and 4. choristers. . . . . . . the late decane afore Dr. London that now is buildid a fair steple of stone at the weste ende of the collegiate chapelle, to making whereof he defacid, as it is said, withoute licens a peace of the kinges lodginge, joyning on the est ende of the chapelle. The decane hath a fair lodging of tymbre withyn the castelle: and to it is yoinid a place for the ministers of the chapelle. 2.3 1590 Camden’s Britannia, description of Wallingford, (Frankfurt), p 207 Moenibus olim obuallatum erat, quae ut ex eorum tractu uidere est, mille passus ambitu collegerunt, Castrum habet ad flumen amplum sane, et adeo munitum, ut inexuperabilis munimenti spes quosdam ferociores fecerit. Cum enim communi quasi bellorum incendio deflagrauerit Anglia, subinde inani obsidione a Stephano Rege incinctum legimus. Amplitudinem eius et magnificentiam cum illic Oxonia secederemus demirabamur (est enim iam secessus studiosis ex aede Christi Oxon.) duplici murorum ambitu, duplici item uallo circumdatur, in medio, moli in magnam altitudinem aeditae arx imponitur, in cuius accliui per gradus ascensu fontem immensae profunditatis uidimus. It [Wallingford] was formerly walled about, and, as may be seen by the tract, was a mile in compass; it hath a castle seated upon the river, very large, and so well fortify’d in former times, that the hopes of its being impregnable hath made some persons over resolute. For when the flames of Civil War, had, as it were, set all England on fire, we read that King Steven ever now and then attempted it by siege, but still in vain. We much wondered at its greatness and magnificence, when we were boys and retir’d thither from Oxford (for it is now a retiring place for the Students of Christ-Church at Oxford) it being double-walled and surrounded with two ditches. In the middle stands a tower, rais’d upon a very high mount; in the steep ascent whereof, which one climbs by stairs, I saw a well of an exceeding depth.

134

Appendix 1: A codicological overview of the Wallingford Roll, Bod. MS. Top. Berks. b. 41 K. S. B. Keats-Rohan A further word on the manuscript seems in order. When it was acquired in 1924 the Bodleian Quarterly Record vol. 4, p. 122, described it as ‘a notable acquisition’. This article cited the shelfmark as MS. Top. Berks. c. 28 (R[oll]). In the Library’s handlist recording MS. Top. Berks. c. 28 (R) the shelfmark is crossed out and on the opposite page there is a note: ‘transferred to MS. Top. Berks. b. 41, 3 Apr. 1974, c. 28 VACANT’. (The History of the King’s Works, published from 1963 onward, referred to this manuscript with its original shelfmark.) In December 1973 the MS had gone into Judy Segal’s paper repair workshop as a roll and came out again on 3 April 1974 bound as a book. Conservation records provide a bare minimum of information: the item was described in its ‘before’ state as ‘Paper Roll old repair across fold paper very weak mould damage in this area’. The card briefly records ph levels and treatment with enzymes, but provides no information at all on individual folios or their arrangement before or after treatment. As seen today there is no outward sign that the texts were originally rolled up inside the membrane cover. But the clues are there. On the last (blank) recto a Bodleian archivist has written ‘23 ult’, i.e., folio 23, last in the roll. The number is an error, since there are, and always were, 26 folia. We can be sure of this from a study of the folia that make up the book. Along what is now the top of each folio are stitchmarks showing how the gathered folia were stitched into their membrane cover, onto which was stitched a seal, now missing. Each recto is marked with a pencil arabic numeral; each verso carries the same number written in violet ink. As seen today, however, the text on the verso is inverted so that one must upend the volume in order to read both the text and the violet numeral. This is related to a conservation technique no longer used that involved splicing the paper sheet and inserting a sheet of plastic between the two halves, as related to J. L. Lloyd in the 1990s but of which no record has been found in existing conservation archives. Although it seems unlikely that the text would have been written after the sheets were stitched together, it is helpful to think of the folia as like a reporter’s notebook, with the metal coils of the notebook representing the stitch marks. After each page had been completed the copyist turned it over, so that what had been the bottom of the sheet became the top, and started writing again on the reverse, using the same style of text frame throughout the record (e.g. the 1548 survey, fols 3–12, is all written in a left-hand column; the rental of St John’s, fols 13–14, is written in a central column). The paper is fragile and has been damaged in places around the edges; checking the mirroring of the damaged areas, as well as the red staining probably due to hot wax that runs throughout all the folia, and the stitch holes, confirms that this was how the texts were written. An additional check is that the watermark which appears on each sheet can be seen on both recto and verso when backlit. There are two variants of the watermark, a cuffed wrist and hand with projecting star detail, which is a fairly common one found in paper produced in Normandy c. 1548–1560 (Briquet, 1907, no.

11337; Heawood 1950, no. 2511). Even the texts on folia 16 and 17, which are not the work of Amyce and seem to be written on a heavier paper and in a different ink, have the same watermark, as is true also of folios 24 and 25, though here the paper is of inferior quality. The Bodleian pencil numbering is, by its own conventions, oddly placed, normally towards the middle of the recto on its right margin, but the violet number on the verso is clearly placed on the top right in the same direction as the text. It too, can only be read with the MS upside down. When viewed so that the text can be read the stitch marks are always at the top of the recto and the bottom of the verso: they have been arranged in the conserved and rebound MS so that the stitch marks are always at the top (the sole exception is folio 13, where the text reads normally on both sides). The sheets were probably stitched when all the components of the document had been assembled, c. 1555, since this was clearly intended as a working copy of specific texts relating to the rents and obligations arising out of borough properties many of which related to the crown’s castle estate and the former priory, as well as rents due from manors of the former honours of Wallingford (or Ewelme) and St Valery in Berkshire. These texts should also have been enrolled at the Exchequer, as many like them had been previously and would be in the future, but so far no trace has been found in the National Archives. They were important enough to be meticulously updated in 1561 by Roger Amyce, who was responsible for the majority of the texts, but thereafter the enrolled sheets were a collection of record among the muniments of the borough. Alternatively, they may have been stitched together when updated in 1561, or as a consequence of that, since this appears to have been the moment when the sheets were further stitched into a membrane cover and rolled up to become the Rotulus de Wallingfordia (and here one must observe that in the original texts the spelling is always Wallyngforde, systematically written as Wallingforde by the copyist of the later MS). It certainly seems to be the case that the Plan of the inner bailey on folio 1r always was the first item in the roll, judging by the greasy rubbed appearance of a first folio that would have been covered by the membrane. The over-large circle representing the motte shows through the verso. In the same area are a number of ink blobs which clearly match others on the opposite folio 2r, indicating a direct relationship between these two sheets. That contains just the following, in an irregular hand not readily identifiable with any other in the manuscript: ‘Right worshipful and well beloved I am at this time in greate wantes and mesesitise therfore if that yow will be please to send me that small recninge which is betwene yow and I I shall be very much ingaiged to yor wor[shi]p and will alwais rest your worship(s) aide at London’, signed John Denison. Nothing more has been discovered of him. The membrane cover is now kept inside the back cover of the bound manuscript. It has an inscription which is illegible apart from the word Wallyngforde. As related above, the unknown copyist of the late 19th century who made a meticulous copy of the MS preserved for us the title of the Rotulus de Wallingfordia. As also mentioned, 135

it is confirmed as a roll in the muniments of the borough of Wallingford by Richard Skirmer (d. 1716). This is the last certain mention of the Roll before it was purchased in an auction at Sotheby’s in 1907 by local solicitor John Mitchell Marshall, according to J. E. Field. He conjectured that it had belonged to Sir William Blackstone, but if it was disposed of legitimately, as distinct from purloined from the borough chest, it is more likely to have been acquired by James Blackstone, who achieved a major stake in the castle estate when it was sold by the crown in 1817. After his death the bulk of his estate passed to his son William Seymour Blackstone, who was imprisoned for debt in 1853 (Johnson 2010, 64–5). If the Roll was then a Blackstone possession this may have been the moment it left Wallingford. Certainly, John Kirby Hedges who acquired the whole castle estate in 1858 and wrote his history in 1881, had never heard of it. This makes the very faithful copy of the Roll that is Bod. Top. Berks c. 33 all the more interesting. It was clearly copied by or at the behest of the owner of the Roll and kept with it. Both were given to the Bodleian by J. M. Marshall’s heir Alfred C. Marshall four years apart, in 1924 and 1928, but he seems to have had little interest in or understanding of the contents, though he clearly recognized that they would be of interest to the Bodleian. The copyist of Top. Berks. c. 33 noted all the changes and additions in the original in a violet coloured ink contrasting with the black of the main text. It was he (or

she) who introduced the violet numbering system on the verso. These were the numbers he introduced into his copy to indicate where he was in the original, the verso then still being the right way up. Preserving the now illegible title on the covering membrane of its parent, Top. Berks. b. 41, is the main value of this 80-leaved copy, but it does have its own intriguing detail. At the end, scrawled on a blank verso is a pencilled note in a different hand: Joseph Mayer Esq, Bebington, Birkenhead. The note has nothing of an ex libris about it, more the appearance of an aide-mémoire, possibly relating to the auction of 1907 or earlier. Joseph Mayer (1803–86) was a very distinguished Fellow of the Society of Antiquaries closely associated with Liverpool, where he was a founder member of the Historic Society of Lancashire and Cheshire. Unmarried, he went to live at Bebington in 1860 and died there in 1886. He amassed a vast collection of manuscripts, archaeological objects and art works, much of which he made available at museums in Liverpool, one of which he founded to house his Egyptian collections (Sutton, ODNB). Another vast collection of drawings, prints and autograph letters was intended to lead to a history of art in England, but was never finished. His collections were dispersed after his death in auctions at Sotheby’s and Leete in London in July 1887, and the Hanover Gallery, Liverpool, in July 1887 (Tythacott 2011, 133). Could he have been the owner at some point of the Wallingford Roll? There is no way of knowing at present,

Figure 6.15 Diagram of the motte path as it was in 2000 based on survey by J. L. Lloyd, showing the centre point and location of the capped well-head (W).

136

but he would certainly have recognized the importance of the Bodleian Library’s ‘ notable acquisition’. Appendix 2 John L. Lloyd’s account of the motte as surveyed during the late 1990s The motte survives, laced with paths to the top, which distort its shape a little. On the top is the keep’s well – there were others in the baileys, of course. It was noted by Camden as being ‘of an immense depth’ (2.3) and would have been constructed as the motte was raised. It is now right on the edge of the top, the well shaft in fact jutting out somewhat, showing that the northern flank of the motte has been lost. This is the bailey side, and it is possible that as the keep was dismantled its stones, some from a great height, were simply tumbled down to carts waiting there, breaking up the structure. The surrounding fence, which may roughly represent the line of the chemise, is there somewhat farther out from the slope than elsewhere, and there is nothing so permanent as a boundary. To the south the motte dips smoothly into the moat. There is no real change in slope, nor any wide flat part between. This fixes the position of the chemise, if its foot was at bailey level, and it may have left some trace, and indeed there is a path here, horizontal and about three feet wide, cut into the slope. On the uphill side there is a little revetment of flints; this land was in private hands for 400 years and has seen many tourists. My own measurement of the motte is shown in Figure 6.15. The southern path extends round the west and towards the east. Near its eastern end there is a very slight concave kink in the revetment, just visible also on the Ordnance Survey map. The path turns outward here, about 10 degrees. Does this kink represent the beginning of the curtain? If so, one might expect a kink of 36 degrees. The end of the path, a few feet, was encased in brambles when viewed, and then immediately met a concreted farm-yard, now gone. On the western side the path rises significantly, but its position is tightly constrained by the steep bank on its outer side. The path makes an excellent semi-circle, as can be seen, but I wonder whether, at the north-western end of that semicircle, where there is a junction, the path could lie a foot or two within the circle of the chemise, because it has there risen a few feet above the level of the bailey meadow. The purpose of the path is to lead to the top, and to do this it must turn inside any remains of the wall, and the contours nearby could have been adjusted by the Victorians. The radius of the chemise, derived from the reconstructed plan, would seem to be about 100ft (30.5m). The model shape – a circle placed symmetrically upon a regular pentagon and passing through one of its points, as suggested above in Figure 6.10, taken together with other measurements – implies a relationship between the radius of the chemise and the length of the sides. The first value,

Rgap, comes from the gap between the two 45-degree triangles, which gap starts life as 84.5ft, the second, R169, from the residual length, 169ft, of the SE curtain, after subtracting the chord (whose length is 2R.sin(36 degrees)). The third, Rtop-tail, comes from the distance from the N curtain, next to the small tower, to the pentagon point that lies on the circle and the fourth, R68, from the 68ft line that runs to the circle. The values that emerge depend upon the value adopted for the curtain length. Here they are: Curtain length 292 290 288 286 284

Rgap

R169

Rtop-tail

R68

101.20 99.54 97.12 94.71 92.30

104.63 102.93 101.23 99.53 97.82

104.67 103.13 101.59 100.05 98.52

100.80 99.80 98.82 97.87 96.94

The differences, for a chosen pentagon side length, reflect the fact that the model is not closely followed. Another value, from the 60ft length of the short curtain, is not reliable. It will be seen that it is difficult to argue for a radius much less than 100ft, since the measured sides are 288ft and (when extended) 291.95ft. I have of course made a computerized ‘best fit’ to the measurements, but it adds little and the machinations are opaque to the reader. The method does not permit of great accuracy. There is an uncertainty of a couple of feet, and the geometry of the path is not incompatible with the plan. Nevertheless, the outer edge of the path follows a very good circle of radius 95ft (98 ‘short feet’ of 11.61 modern inches). The cross marked on Figure 6.15, showing the centre of the path, is additionally equidistant from the boundary fence to the north and the path to the south. That does not mean that it marks the centre of the top. The top extends on the west and north-west to the unfenced path there, and on the east the fence is very necessary, where the path beneath may well have removed a few feet of material. To the south, where there is a short fence of four posts, one can just see, from the slope of the motte between the two southern paths, that the flat top must once have extended outward some 8ft beyond that short fence. On the northern side the slope is very steep and the present edge is marked by the path and fence. As mentioned, the well is on the very edge, an impossible position, showing that the northern face has been lost. Hedges dug the top and found nothing. The top could have been oval, the long axis lying roughly east-west, to match a rectangular keep, and the north face could have been steeper than the south (compare the aerial view in Figure 6.11). [Note: this has been included without alteration because it predates what is now the rapidly deteriorating condition of the motte.]

137

7 LANDSCAPES, LEASES AND LAWSUITS: REAPING THE REWARDS OF POST-DEMOLITION DETECTION AT WALLINGFORD CASTLE Judy Dewey Abstract The imposing grassy ramparts of Wallingford Castle disguise a landscape that has been much altered since the castle’s defences were destroyed after the 17th-century Civil War. Under the continued ownership of the Crown to 1817, numerous leaseholders adapted the site for their own ends, planting trees, constructing buildings, digging out ramparts and making new entrance-ways. Once the castle was sold, private owners made further changes but fortunately protected the integrity of most of the site. This paper presents the findings of a multi-disciplinary study which has sought to understand the impact of these changes on the landscape, with a view to re-discovering the late medieval layout of the castle. It has involved a close study of the landscape itself, an investigation of post-demolition documents – leases, lawsuits, minute books, maps etc. – and an appraisal of archaeological reports from the mid-20th century to the most recent work by the Wallingford Burh to Borough Research Project, and including additional resistivity work by The Wallingford Historical and Archaeological Society (TWHAS). Running parallel to this has been Katharine Keats-Rohan’s work on the medieval documents; the sharing of unfolding information has been mutually most valuable. It is hoped that the resulting interpretation of the layout of the castle walls, baileys, gateways and water-systems, offers a well-evidenced starting-point for future research. Introduction It is 363 years since Cromwell’s Council of State ordered ‘Wallingford Castle to be demolished, and the works effectually slighted’ (CSPD iii 1651–52, 496). The order, made on 17 November 1652, was carried out with considerable energy and thoroughness, leaving the site of this vast medieval edifice as a pile of rubble, while its good stone was sold off or robbed out for building material. Yet in 1688 when the future William III, Prince of Orange, stopped briefly in Wallingford in some fear of the army of the deposed King James II, the Duke of Schomberg was reported to have declared ‘that he knew no place in the world better suited than Wallingford for a defence and that in twenty-four hours he could make the place so strong as to hold out against a good force for some time, and that in three weeks he could render it altogether impregnable’ (cited in Rannie 1898, 283). The declaration was never put to the test as ‘they afterwards found there was no manner of occasion for any such attempt’ (ibid.). Sadly the truth of this account cannot be verified but the earthworks and scattered stone remnants of Wallingford Castle still surprise the visitor with their awesome scale, evoking something of the castle’s past glories.

Equally impressive are the resistivity surveys, generated largely by TWHAS as part of the Wallingford Burh to Borough Research Project, which reveal the mass of solid material that still lies scattered beneath the present grassy mounds (see Figure 5.1). Excavations carried out by the project further demonstrated the complexity of the medieval and late Saxon layers and confirmed that the re-fortifications of the 17thcentury Civil War had made significant changes and additions to the surviving medieval castle before its final demolition (Christie and Creighton 2013; see also contributions in this volume). The project’s detailed and wide-scale coverage has enabled fresh discussion of many aspects of Wallingford Castle’s layout and function, moving well beyond some of the long-held conceptions of its development, to both question and confirm them. Meanwhile, the detailed examination of the medieval and Tudor documentary evidence presented in the two previous papers (Keats-Rohan and Lloyd, above) now adds significantly to this material, further confirming the importance of a full multi-disciplinary re-assessment of this major royal castle. It is the aim of this paper to add one further element to the discussion, namely an awareness of the impact made on the landscape of the site by its history and evolution in the three and a half centuries since the demolition of the castle. Following an outline history of the post-medieval castle, a detailed assessment will be provided of the castle’s boundaries, baileys and buildings, gates and track-ways and its relationship with the Great Bridge and riverside, including mills and water defences. In each case changes will be assessed and the landscape of the castle site linked, where possible, to its medieval heyday. At the very least, this should help to eliminate some obvious but long-standing errors of interpretation; at best, it will shed more light on our understanding of the layout of the medieval castle. The research has clarified that certain areas of the castle’s landscape have been largely unaffected by time, but that others have been almost totally destroyed by the events of later centuries. The worst affected areas are the outer bailey, the north-west of the middle bailey, and the whole southeastern corner of the castle; yet even in the most damaged areas, careful piecing together of the evidence yields some interesting results. The research has been largely founded on the meticulous work of members of the TWHAS Documents Group, to whom a large debt of gratitude is due; they have spent many years seeking out and gathering together documentary evidence from diverse sources throughout the country’s archives. These documents have been analysed against a background of diverse archaeological reports (including the most recent work of the Wallingford Burh to Borough Research Project), 138

Wallingford Castle c. 1300 King’s Mead

Porter’s Land (1.5 acres)

Former North Gate

1 2

Mill

iley

’s A een

rT

College of St Nicholas

1

ve

2

Ri

Former Road

S

iley

D

High Street

S

D

Site of Stephen’s Castle

Mary Grace

5 St Peter’s

We ir

ha

Oute r

Castle Street

3

4

Qu

Middl e

5 Ba

Ba

rbo

Inner Bailey All Saints

ch’

S

St Nicholas

GD

ldit

ur

Barbican

‘Wa

3

me s

4

The

Gre

at B

ridg

e

KEY GD - The Great Drawbridge

1

Constable’s Gate

1 - Queen’s Tower

2

Bydongate

2 - Muggetour

D - Drawbridge

3

Dernegate

3 - Water Tower

S - Possible sluice site

4

Town North Gate

4 - The Great Tower

- Conjectural towers

5

Town East Gate

5 - Benet’s Tower

- Church

Figure 7.1 Wallingford Castle c 1300 (map David Dewey)

correlation of features in a series of surveys, maps and images, and numerous hours of walking the castle site and its environs. References have been extracted concerning key landmarks, such as boundaries, buildings, defences, access points, routes, water systems and mills, together with comments, however trivial, on work done and changes made. Digital overlaying of relevant maps and plans has been informative, but when the lack of an accurate scale has made this impossible, noting

the general relationship of features has proved illuminating (see Figure 7.25). Keats-Rohan’s work on the medieval documents (above p. 39) has shown with clarity that the main entrance to the castle lay not on the west side, as was generally assumed, but somewhere in the south-east corner, close to Wallingford Bridge and the town’s East Gate. This finding has encouraged a particular focus on the southern side of the castle where little

139

attempt has pareviously been made to ascertain the possible layout of defences and gateways; even the existence of an outer wall in this vicinity has previously been questioned (Christie and Creighton 2013, 216). The proximity of the High Street properties and gardens has clearly made archaeological investigation and survey work difficult in this area (see Figure 7.4), but much potential still exists. The documentary research, however, has enabled other fundamental castle features to be confirmed, most importantly the presence of three baileys – inner, middle and outer – together with the location of the buildings of the College of St Nicholas within the middle bailey (1.28); this in turn has enabled the likely site of the main gateways to be established. Lloyd’s work on the 1548 plan of the castle (above pp. 124ff.), when considered with the geophysics results, has also led to a more accurate mapping of the motte, keep, walls and towers of the inner bailey than has previously been possible. An outline post-medieval history The documentary evidence indicates that by the time the honour of Wallingford was transferred to Ewelme, in 1540, the castle had been abandoned for its military value and was being treated as a material asset. The Tudor survey of stone and timber (2.1) reflects this attitude. Nevertheless, the only actual disposal of castle land by the Crown before the 19th century was that of the dissolved College of St Nicholas in 1548, granted (for a fee) to Michael Stanhope and John Bellowe (3.4). The collegiate buildings in the middle bailey, comprising the lodgings of the dean, the priests and the clerks with their associated land, were sold on to Christ Church College, Oxford in August 1548 for £105 6s 8d; they were to be used as a bolt-hole for college students when plague threatened Oxford (CCA, MS Book of Evidences, 903–907). In the meantime, these castle buildings were leased out to various tenants and continue to be clearly traceable in the Christ Church records until their final sale in 1819. The rest of the castle estate was retained by the Crown until 1817, with the single exception of the years 1646–60, when it was sequestered by Parliament. By the 17th century, and probably much earlier, large swathes of the rest of the castle lands were being leased out by the Crown to bring in convenient revenue. The Civil War came as a rude intrusion to such leaseholds. The Royalist garrison commandeered the whole castle by 1642, including the Christ Church land, and the castle was a military zone from then until its destruction in 1652/3 (Dewey pp. 236–43 and Christie pp. 251–7, below). The surrender of the Royalist garrison after the final siege in 1646 brought the castle’s considerable land assets to Parliament, which was quick to assess the extent and value of the former royal possessions at Wallingford. Arthur Evelyn, parliamentary governor of the castle from its surrender in 1646 until its demolition, purchased the manor and castle of Wallingford (valued at £10 12s 4½d per annum) and quickly sold it on to Edmund Dunch of Little Wittenham, Cromwell’s cousin; the rest of the honour and manor of Ewelme (valued at £74 17s 9½d) went in 1655 to John Hemsdell of London and William Cleeve of Inglesham, Berkshire (TNA, E304/1 B20).

The physical destruction of Wallingford Castle, ordered to be carried out by Major Evelyn, Edward Jennings and John Rusden (3.8), was a major undertaking which must have posed enormous logistical problems in the organization and discipline of the workforce and the disposal of valuable materials. Unsurprisingly, there were some unfortunate repercussions for the leaseholders of the site. One such example is that of Thomas Freeman, mentioned as a leaseholder from Queen Henrietta Maria as early as 1644 (CSPD ii 1644–5, 214) and whose lease of the inner ward was subsequently passed on to his son John. When Charles II was restored to the throne in 1660 the ownership of the castle reverted directly to the Crown and claims for compensation swiftly followed. John Freeman was quick to make it known that the destruction of the buildings had included the walls and caused costly damage to his gardens (see p. 153 below). When John Freeman died in 1684 his inventory recorded two leases of Wallingford Castle, worth £1300 of his total possessions of £1347 (TNA, PROB 5/3845). These leases were for the inner ward (from his father Thomas) and the herbage and pasture of the outer banks, due to expire in 1687. Freeman had apparently had some cash flow problems, perhaps partly because of his losses during the demolition, so in 1681 he had mortgaged his property for £850 and had subsequently borrowed £1500 from a lady called Ann Lawton on security of the castle land. Unfortunately for his successors, widow Ann Freeman and Richard Pont (named as an appraiser of his inventory), the Lawton family foreclosed the mortgage in 1687 and the case was brought to Chancery. Thomas Renda of London, solicitor and agent for Pont, agreed to pay £1900 – the sum now required to clear the mortgaged Freeman property, £1200 of which related to the Wallingford Castle estate, which was then to be assigned to Renda; the settlement was finally made in January 1699/1700. Notably Renda did not pay the outstanding £700 (relating to lands in Essex), maintaining that the title to them was not good, despite being pursued in Chancery for the money (TNA, C9/461/96). Thomas Renda rapidly emerges as a man undaunted by litigation, prepared to act first and deal with the consequences later. This has been to the benefit of historians who have been left numerous legal records to consult which shed some light on the changes he made to the castle site. These cases during his leasehold (1699 until his death in 1722) included a successful action by the Mayor and Borough Council against him for encroachment on land near Wallingford Bridge in 1705 (BRO, W/ Ac/1/1/2 ff. 209–15) and an accusation of ‘divers ill practices’ during his election as MP for Wallingford in 1709, eventually dismissed after a House of Commons enquiry (Journal of the House of Commons, 16, Dec 15 1709, 242). There was a further Chancery proceeding, brought against him by the Dean and Chapter of Christ Church College for encroachment on a half acre of their land to create a wide driveway to his new house in the inner bailey (TNA, C11/1973/10). The latter trouble arose in 1715 after Renda had acquired the lease of the site of the Dean’s Lodging (destroyed during or after the Civil War); 140

he had (probably) conveniently removed the boundary (‘meer’) stones – something which he alternately claimed was done by agreement and then denied doing. Christ Church was unable to prove title and the case languished in Chancery until Renda died. No final resolution has been discovered but there were still resonances a hundred years later, as will be seen. At this stage, however, the dispute produced many interesting and informative witness statements and a map (see Figure 7.46) which will be considered in more detail below. Finally, in the period from 1719 to 1721 Renda was taken to court by the Borough of Wallingford for long outstanding non-payment of rent on more than one property in the town; he was served with an ‘ejectment’ notice for one of these (BRO, W/Ac/1/1/2, f 272v). There can be no doubt that Thomas Renda was a man to be reckoned with! After his death, the problems over Renda’s estate had to be settled by further court cases involving his widow Sarah, who was his second wife, and other members of his family by his first marriage. It took some eight years

for these to be resolved and during that period Renda’s leases on the castle property lapsed. In 1722 the newly elected MP for Wallingford, William Hucks of St Gilesin-the-Field, negotiated a Crown lease of the castle’s herbage and pasture, including the porter’s land and the Prison House, and in 1726, of the inner ward and also the Thames fishery and eyots. Sarah Renda, whose home was her late husband’s house in the inner bailey, disputed William Hucks’ right to any lease and was vehement in her wishes not to allow him to acquire the property (TNA, 1729 Exchequer Proceedings, E133/39/49). Nevertheless, William Hucks and his descendants were clearly in possession of the castle from this period through until 1806, including, from 1728, a lease from Christ Church of the Priests’ Lodging and the site of the old Dean’s Lodging in the middle bailey (CCA, MS Estates 5, fol. 123, and BRO, D/EH T5; see Figure 7.47). All these lands provided a source of income by being sub-let to tenants. With the Hucks castle lease due to expire in 1806, the Surveyor General of His Majesty’s Land Revenue, drew

Figure 7.2 Detail of the map drawn up for the sale of the castle in 1817 (photo: Katharine Keats-Rohan, reproduced courtesy of The Bodleian Library, University of Oxford, MS Top. Berks. b. 10, fol. 17v-18r).

141

up a new plan of the Crown Estates at Wallingford (see Figure 7.48); particulars and a valuation were also listed. However, an attempted renewal of the expired 41-year lease made by the representative of William Hucks’ son Robert (by then designated a ‘lunatic’) called into question the boundaries of the Christ Church lands shown on the plan. Thus yet another dispute was provoked, this time between the Crown and Christ Church College, as to the exact land boundaries and liabilities. The confusion almost certainly stemmed from Thomas Renda’s activities a century before and once again the dean was asked to clarify his claims. James Blackstone (son of the late renowned Judge William Blackstone) was engaged as an arbiter between the college and the crown but it took a further twelve years for the matter to be settled. As will be seen below, the documentation and maps produced in the process have proved to be yet another useful source for the historian in understanding the finer details of the changing landscape of the castle. In 1817 the Crown finally sold the honour of Wallingford and Ewelme and the Crown Lands of Wallingford Castle. Re-using a form of the 1804 plan of the estate as a base map, numbered lots were delineated for sale (Figure 7.55). The auction was held at the Lamb Hotel in the High Street on 12 August 1817: ‘c. 220 acres including the site of Wallingford Castle with the gardens and moats thereto belonging; also two dwelling houses, several aytes and a valuable fishery; the King’s Mead and other lands adjoining the Town’ (Bodleian, MS. Top. Berks. b.10, 14–20, which includes the plan with names of purchasers) (Figure 7.2). The castle lots were purchased by a number of different people but James Blackstone bought Lots 1, 2 and 4 which constituted the original Inner Bailey, all the outer banks to the north and west and the south-eastern corner of the outer bailey, bordered to the north by the Christ Church

land on which the old Priests’ House still stood (by then in use as a malt-house). Lot 3, part of the outer bailey to the south and west of the Christ Church property, was purchased by Edward Wells who subsequently sold most of it to the leaseholder of the neighbouring Clerks’ Lodging, the Rev. Isaac Arding. In the following year, 1818, James Blackstone finally arbitrated a settlement of the land dispute between Christ Church and the Crown, enabling the former to offer its property for sale. Blackstone himself paid the £324 necessary to redeem the land tax on the Christ Church land and in return received all the land and buildings in the middle bailey for himself, including the ‘mansion house...commonly called the Clerks’ Lodging’, the ‘Priests’ Orchard’, the ‘College House’ (Priests’ House) and the approximately half acre of ground to the east, site of the original Dean’s Lodging, which had been the main source of the longstanding disputes (BRO, D/EH T1(1). In 1819 Blackstone transferred the reversion of the Clerks’ Lodging (i.e. the freehold after the expiry of the existing lease) to the Rev. Isaac Arding who thus became the owner of that part of the property. James Blackstone kept the rest of the former Christ Church land for himself, which, together with the land he had already acquired from the Crown, gave him control of a large proportion of the castle site by 1819 (BRO, D/EH T1, 2). During his ownership he was to be responsible for several notable changes to the castle landscape and access, which will be discussed in detail below. When James Blackstone died in 1831 he left his property to his 21-year-old son William Seymour Blackstone, a very wealthy young man who was elected as MP for Wallingford in 1832 and held the office for 20 years. There is clear evidence that he spent money on his Castle Priory property in Thames Street and was generally profligate with his fortune, but there are no indications that he made any changes to the castle site. Instead, he

Figure 7.3 ‘Wallingford Castle’, the Victorian mansion built for J. K. Hedges and his wife Sophie (image courtesy of Wallingford Museum).

142

embarked on a series of mortgages for the acquisition of various properties in the town (including premises in the High Street with the Town Arms on the east) and spent lavishly on the building of Howbery Park in Crowmarsh. He was incapable of maintaining his mortgages and by 1852 he owed £37,000 and had judgement debts for a further £10,000. Necessity led him to sell his property, including the castle, to his cousin the Reverend Harry Lee who took over the mortgages and debt liabilities and paid off some of the judgement debts. In 1852 Lee transferred the mortgages (now reduced to £31,000) to James Morrison, of Basildon, who thus became the new owner of the castle lands. Morrison died within a few months and in 1858 his executors finally put the castle lands up for sale. They were purchased by John Kirby Hedges, solicitor and town clerk, whose family was to continue to own them until 1977. By 1858 there had also been considerable changes on other parts of the castle site. Several small exchanges of land had followed the initial sale of the castle, but the largest holding was that of the Rev. Isaac Arding, who had purchased the Clerks’ Lodging in 1819. On his death in 1829 his property was inherited by his three sons, but by a series of complex negotiations, John Allnatt Hedges gradually acquired these inheritances by purchase. He also negotiated rights of way with James Blackstone and bought from him a piece of land including a wheelwright’s shop – almost certainly part of the original Prison House area (below pp. 150ff.). By 1837 J. A. Hedges and, more particularly, his son John Kirby Hedges were actively involved in the planning and building of a new mansion house in the middle bailey of the castle. The architect/

builder was John Plowman of Oxford and the plans involved the final demolition of the Clerks’ Lodging and adjoining buildings (BRO, D/EH E6). The house was completed by 1840 (Figure 7.3). In April that year J. A. Allnatt Hedges made a marriage settlement of the whole property on John Kirby Hedges and his wife Sophie Cassandra, who were cousins by birth (BRO, D/EH T1,1/2). Thus when J. K. Hedges subsequently purchased the Morrison castle land in 1858, he effectively reunited the main castle site for the first time since the 16th century. He built a new gateway into the property off Castle Street, removing some old cottages, building a lodge, landscaping the grounds and planting trees (Figure 7.4). This work effectively obscured the remaining traces of the outer ditch in the vicinity of what is now the outer area of Castle Gardens. John Kirby Hedges (Figure 7.5) was a keen antiquarian who carefully researched the history of his property, privately publishing a two-volume work, The History of Wallingford, in 1881. This was followed by a shorter version in 1893. He had consulted many original sources and, although some of his interpretations are questionable, his books have proved to be a very valuable starting point for local historians. His contemporary observations are a primary source for the castle in the Victorian period and his appreciation of landscaping is reflected in the numerous exotic trees and some floral species which have survived to the present. The ruins of the Priests’ House became a summerhouse and a large greenhouse abutted the surviving wall on the south side. Hedges had a small garden building in which he collected historic artefacts to form a private museum, occasionally inviting the Berkshire Archaeological Society to visit the castle. On the eastern

Figure 7.4 Wallingford Castle from the west, 6th December 1965. In the foreground is Castle Street with the entrance to J. K. Hedges’ Victorian mansion, known as ‘Wallingford Castle’. The house dominates the middle bailey, with the motte and inner bailey to the north. To the east, beyond Queen’s Arbour, lies Wallingford Bridge and the River Thames (top of picture). A stretch of original castle wall can been seen on the eastern rampart; only a small section of this wall survives today, the rest having fallen down in the mid 1970s. To the south of the house (right) lie the gardens filling the outer bailey, enclosed by the curving castle/town boundary wall. Outside the boundary are the houses lining High Street (Copyright reserved Cambridge University Collection of Aerial Photography).

143

Figure 7.5 John Kirby Hedges. Portrait by James Hayllar (1829–1920), a renowned Wallingford artist. (Photo courtesy of Wallingford Town Council.)

side of the site, abutting Queen’s Arbour, was a rustic pond area adorned with planted flowers and also a boathouse linked by an inlet to the Thames (see Figure 7.35). John Kirby and Sophie Hedges had four children but their only son died in 1878 at the age of 35 and none of their daughters married. When their father died in 1901 the two daughters Sophia and Julia, the ‘Misses Hedges’, continued to live in Wallingford Castle (as the mansion was then known), dying within one day of each other in 1921. The youngest daughter, Ellen, then moved into the castle and was resident till her death at the age of 90 in 1936. The inheritance passed thereafter to Major Francis Reade Hedges, the great-nephew of John Kirby Hedges, and subsequently, in 1952, to his son John Francis Hedges (knighted in 1962). Sir John and Lady Hedges resided in a large stone-built house on the High Street, known as St Nicholas, slightly set back and still having extensive grounds in the southern outer bailey of the castle (Figure 7.6). They leased the Victorian mansion to The Distressed Gentlefolk’s Aid Association, while the northern outer banks were leased to a farmer whose farmyard and buildings lay to the southeast of the motte, off the present Castle Lane which leads from the High Street near Wallingford Bridge through to Castle Street. Sir John and his wife enjoyed and cared for the rest of the castle grounds, but eventually the upkeep became impossible and by the early 1970s the Victorian mansion had become very dilapidated. In 1972 the decision was made to demolish it, with a view to the middle bailey

Figure 7.6 St Nicholas, High Street, Wallingford, 2015 (photo: Judy Dewey).

144

Figure 7.7 Thameside Mansion, looking west, from the riverside 2015 (photo: Judy Dewey).

being developed as a retirement home for The Architects’ Benevolent Society. Archaeological excavations took place there in the same year, discovering some well preserved medieval cob buildings (Christie and Creighton 2013, 183–202). Planning permission, for both the new retirement home and a modest house for Sir John and Lady Hedges on the site, was rejected by the Secretary of State in October 1976 after a public inquiry (in which TWHAS was heavily involved) had deemed the site to be of historic importance. The following year Sir John and Lady Hedges generously donated the entire castle site to the Wallingford Town Council in return for the presentation of a flowering shrub each year on Lady Day. They spent their latter years living in the Coach House, set within the castle grounds off the Castle Street entrance. To assist the cost of upkeep of the new Castle Gardens, including the motte, with its new timber access bridge, the formal gardens in the western outer bailey and the upper middle bailey, Wallingford Town Council continued to lease, and eventually sold, the northern and western outer banks and the inner bailey (excluding the motte). They were bought by the tenant farmer, Edward Ryall, who in 1997 sold the land to a developer. The castle site therefore came under threat once again. Permission was granted by South Oxfordshire District Council for the building of Thameside Mansion overlooking the Thames in the southeast corner of the castle lands, on the grounds that the area had previously been disturbed by farm buildings (Figure 7.7); only limited archaeological investigation was done in advance of the building (Ford 1995; Saunders 1997).

The District Council then purchased the remaining castle land (the present Castle Meadows) which is currently managed on their behalf by Earth Trust (formerly The Northmoor Trust). Today a modern gateway and the bridge over Castle Lane link the two areas of ownership; there is unlimited public access to Castle Meadows and daily access for several hours to Castle Gardens. The documentary sources used to unravel the history of the post-demolition castle site have also provided a basis for the more challenging task of mapping changes made to the landscape during the last three and a half centuries. No images of the castle in its heyday have survived to support the documentary and archaeological sources, apart from the 1548 outline plan of the castle’s inner bailey and moat, fully discussed by Lloyd (above pp. 124ff. and Figure 6.1). The earliest drawing to feature the whole castle is that held in the Christ Church archives, featuring the estate at the time of the disputes with Thomas Renda c. 1715 (see Figure 7.46). Frustratingly, it lacks a scale, and at first sight it is a rather odd looking illustration, but it is of critical importance for many reasons which will become apparent. Apart from sketch plans of the Hucks estate in 1786 (see Figure 7.47), the next known map of the castle is the detailed 1804 Crown Estates plan (see Figure 7.48), generally well-executed and later reused as the base map for the drawing up of lots for sale by the Crown in 1817 (see Figure 7.55). By this period, other helpful, if sometimes problematic, images of the castle are beginning to appear – artistic sketches, views, and plans associated with legal wrangles. An important survival is a unique hand-drawn 145

map of Wallingford c. 1835 now in Wallingford Museum (see Figure 7.56) which is the earliest to show the whole town and castle. This map, together with the parish tithe maps, provides a most valuable source of information before the later 19th century OS maps contribute solid mapping evidence. The older maps and images are complemented by the surviving features of the landscape which have recently been surveyed by Michael Fradley, where accessible, as part of the Wallingford Burh to Borough Research Project (Christie and Creighton 2013, 163). To this is added the variety of archaeological reports from recent and past excavations. The rest of this paper presents and analyses the material evidence that has emerged from these varied sources to enable a fresh interpretation of the layout of the medieval castle precinct. Boundaries It is fundamental to begin by establishing the boundaries of the castle and their relationship to the town layout. The origins of Wallingford’s town and parish boundaries have been already been considered in this and an earlier volume (Keats-Rohan above p. 38; Dewey 2009, 17–26), where it was suggested that the castle site had a royal Saxon antecedent. The form of that early stronghold cannot be

defined with any certainty but the medieval castle that was its successor lay within a defined extra-parochial area that most probably related to a pre-Conquest church or chapel, later the royal free chapel of St Nicholas (Roffe 2009, 36). The boundaries were maintained or extended over the whole period of the castle’s existence. Even after the demolition, the castle’s extra-parochial boundaries continued to define the municipal boundary and are a survival of an earlier landscape that is worth considering carefully. Extra-parochial boundaries Bearing in mind Wallingford’s sad lack of early maps and illustrations, it is particularly useful that there is a detailed written description of a perambulation of the municipal bounds in 1707, recorded in the town’s Statute Book. There are several key points on this perambulation, which can still largely be followed on the OS map of 1899 (Figure 7.8). The castle section of the boundary walk approached the site from the west following the outer rim of the town’s Saxon ditch (the ‘black ditch’) to the north of the Bullcroft, crossed the main road at the former North Gate of the town, known then as ‘Clabcott Gate’ (Clapcot Gate, at the north end of Castle Street), then proceeded southwards along the line of the first (outer) moat:

Figure 7.8 1898 Ordnance Survey Map with borough boundary enhanced. Note that the College of St Nicholas is erroneously marked as the remains of St Nicholas Church, which was actually in the inner bailey.

146

‘...and from Clabcott gate downe unto the first Moate and soe along in the Moate to the Churchyard [All Hallows] takeing in all Castle Street and alsoe the Churchyard; and from the east Corner of the Churchyard next to Mr Rusdens garden Wall [the southern side of the present Bear Lane] strait through the gardens there to the gates of Thomas Renda esq standing at the Southend of the walke that goeth upp to the Castle’ (BRO, W/AC/1/1/2 folio 223v). The perambulation was made just 55 years after the castle demolition and the clear implication from this section is that in 1707 the outer moat of the castle was still welldefined, continuing all the way down the eastern side of Castle Street at least as far as All Hallows churchyard, which in 1707 obviously lay within the town boundary, not the castle precinct. (To prevent any confusion, it should be noted that the name ‘All Hallows’, in common

local use for this Wallingford church from the 16th century, is synonymous with the medieval ‘All Saints’.) The northern end of the outer moat is still clear today but the more southerly stretch was so much altered in the 19th century, when J. K. Hedges built the Lodge and entrance to his mansion off Castle Street (see Figure 7.4), that the moat line is now lost. No mention is made of gardens along Castle Street in 1707 and none appear on the c. 1715 Renda map (see Figure 7.46) which shows the boundary fenced or walled as far as the churchyard area and ignores landscape features other than trees and buildings; there are just three small isolated cottages within the northern section of the churchyard area. A lack of properties on Castle Street at this time is unsurprising since the area was devastated in the Civil War. Notably, in 1806 a similar perambulation mentions both houses and gardens (3.13). Since the moat seems to have delineated the borough boundary, the implication is that it ran through

Figure 7.9 Bear Lane 2014. This wall marked the Wallingford Borough boundary; the lane itself lay within the castle (photo: Judy Dewey).

Figure 7.10 Gardens at the back of High Street step up towards the town’s boundary wall; the ruins of the castle’s middle wall can just be seen on the right in the distance. The ground between the two walls formed the southern outer bailey of the castle, through which the third wall and moat once ran (photo: Judy Dewey).

147

what is now part of the outer Castle Gardens and that it could be expected to continue to follow the inner line of the town boundary along the back of the High Street. This is reinforced by Richard Gough’s valuable record of his visit to the castle in 1768 (3.12 and see Figure 7.45) which describes the western and southern castle ramparts as ‘exceedingly bold and fresh,’ confirming that there were still two well-distinguishable lines of outer fortifications existing at the time. Since the inner bailey walls had been demolished after the Civil War (3.8 and 3.9), it is clear that Gough’s ‘inner S one’ refers to the rampart of the middle bailey on which a section of wall still stands today (see Figure 7.26); the ‘outer’ one must therefore be describing a second defensive bank to the south. It is also interesting to note on the 1804 castle plan a large stretch of water surviving at just the place where the ditch associated with the outer rampart might be expected to run, suggesting that it did indeed continue to follow the boundary line. The presence of ditches and ramparts on the south side (and elsewhere on the site) is reinforced by the use of diagonal pen-strokes and an orange wash on the plan, apparently marking the edges of downward slopes (see Figure 7.48). The importance of the precise location of these features will be considered in more detail later. The 1707 perambulators had ‘Mr Rusden’s wall’ as their marker for the town boundary at the back of the High Street. The northern face of this flint based, tile-capped clunch wall is the modern southern boundary of Bear Lane, still surviving as a fine example of construction prevalent locally in the 17th century; the use of dressed stone as capping in the lower part of the wall suggests it may just post-date the destruction of the castle, when such stone would have been readily available (Figure 7.9); in 1804 the proprietor of The Bear, a public house off the High Street, had to lease this stretch of castle land for £10 a year to keep his access (TNA, CRES39, 18). Most of the gardens through which the perambulation passed along the back of the High Street (marked on the Renda map as ‘Backsides and Gardens’ (Figure 7.46) still show well-defined boundaries. A slight rise in ground level towards their northern boundary wall can still be observed in several of these High Street gardens, an indication of an earlier bank defining the limits of the royal property (Figure 7.10). Interestingly, a borough deed of 1294 makes reference to a High Street tenement stretching to ‘the outer barbican of the castle ditch’ (Keats-Rohan 39) and deeds dating from 1572 to 1822 for the former ‘Antelope’ pub, situated just west of the present 90 High Street, refer to ‘a ditch of Wallingford Castle called the Black Ditch’ at its northern extent (BRO, D/EH T2-16). It is also worth noting that neither the 1707 perambulation, the Renda Map, nor the 1804 plan give any indication of the deep carriageway that today leads into the castle off the High Street (opposite Thames Street) marked by a stone-pillared gateway; this must therefore be a later insertion (see below). In 1707 the castle’s southern boundary wall reached the riverside close to the large gates clearly shown on the Renda map (see Figure 7.46), which sealed off the southern end of what is now Castle Lane. This was at a point adjoining the property today called ‘Munt’s Mill’;

the small building which stood there in 1707 was bounded on the east by the stream flowing along the western edge of Queen’s Arbour. Having passed Renda’s gate, the walk continued ‘from the said gate round the old Winch Eyott and all the lockwater and Eyott thereunto belonging and then over the River…’. The Winch Eyott was one of two islands once situated in an inlet of the river just to the north of Wallingford Bridge. A lease of the ‘bridge water and eyots’ in 1533 mentions a ‘little round eyot next the chapel eyot’ (BRO, W/Ac 1/1/1, fol. 33r) and the islands are also mentioned, but not named, in the 1606 town survey (TNA, E315/369/3, fol. 101–123). An 1803 engraving from a Thomas Hearne drawing shows the inlet, including a winch in the far right of the picture (see Figure 7.50); such a winch would have been used to help pull barges through the narrow arches of the old bridge. At some stage the island on which it stood became part of the town wharf. (This area is north of the present restaurant which stands partly on the other reclaimed island known as Chapel Eyot, see below p. 168.) The old Winch Eyott anciently paid tithe to the parish of Brightwell, not to Wallingford: the Brightwell Glebe Terrier of 1634 records ‘the Tythe of an Aite lyinge neare unto Wallingford Bridge nowe in the occupacion of Mathew Mascall a Carryer of Wallingford aforesaid’ (cited in Mortimer 1995). It is not therefore surprising that in 1707 the town boundary perambulation went ‘round’ the eyot to exclude it, probably crossing the stream by a small bridge and following the defined right of way that took them round the edge of the old island to reach Queen’s Arbour – still the main path today. How far the perambulation went before it crossed the Thames is unclear, but on a similar occasion in 1806 the walkers followed the line of the ditch/stream skirting the western edge of Queen’s Arbour (named as Bishop’s Ditch) until it reached the boundary ditch at the end of Queen’s Arbour, running east towards the Thames (3.13). At this point it crossed the river to the Crowmarsh side. The same was probably true in 1707. The complexity of this riverside area is considered in more detail below. The tithe maps for the parishes of All Hallows (1841) and St Peter’s (1851) confirm that the boundaries described above were largely unchanged by the mid-19th century: the main castle precincts were still extra-parochial, implanted into an established parish landscape, bounded to the north, west and east by All Hallows (including Queen’s Arbour) and on the south by St Peter’s. The same northern boundaries are clearly shown on the Renda map (c. 1715) apart from the addition of a large rectangular extension on the north-east corner (see Figure 7.46). The most likely explanation for the extended boundary is that Renda had acquired a lease of part of the porter’s land (lying in All Hallows), one and a half acres of which are described as being in just this location (see Figure 7.1); the map was drawn to show the property he held, not the ancient boundaries. The northern boundaries remained unchanged until 1894 when the Wallingford Rural and Urban District Councils were separately established. Further major boundary changes followed in 1974 with the Local Government Act.

148

Figure 7.11 Combined resistivity results for Castle Gardens showing 20 metre grids (image: Gerard Latham).

The question that must now be tackled is whether the boundaries of 1707 were those of the original castle precinct. It is only on the western side that the castle, once established, can be demonstrated to have encroached significantly on the town. Nicholas Brooks’ excavations in the 1960s concluded that the alignment of Castle Street had

been twice disturbed. Firstly, houses on the eastern side of the street, occupied from the early 11th century, together with the North Gate of the late Saxon town, were destroyed by the cutting of the middle bailey ditch of the castle; the date of this cut was not definitive but Brooks provisionally assigned it to the mid-12th century. Subsequently, in the 149

13th century, a third castle wall overlaid the rest of the houses and destroyed the re-built North Gate of the town, shifting the line of Castle Street to its modern alignment several yards to the west (Brooks 1965a; Christie and Creighton 2013, 88ff, 114ff). A more recent excavation, in preparation for an extension at the back of 24 Castle Street (immediately north of the entrance to Castle Lane), revealed the edge of the outer ditch running north-south. It suggested that the ditch widened slightly at the northern end, which accords with the surviving features (Hammond 2006). When projected southwards (as indicated by the 1707 perambulation), the alignment of the ditch observed would bring it through the western part of the present Castle Gardens, just to the east of All Hallows churchyard. (It is worth noting that Gough’s observation in 1768 that the churchyard of All Hallows was ‘within the W ditch’ (3.12) was probably referring to an observation of the sunken line of the original Castle Street, still to be seen running through the churchyard, rather than the outer ditch of the castle which obliterated the rest of the old road – see Christie and Creighton 2013, 167.) The 13th-century outer wall would have run parallel to the outer ditch, enclosing an extremely narrow outer bailey (see Figure 7.1). A probable section of this wall was revealed by a resistivity survey in Castle Gardens (Figure 7.11). Notably, it is in alignment with a section of solid masonry observed in the present Coach House to the north (Christie and Creighton 2013, 167). The survey also revealed low resistance on the west, in the area where the moat followed in the 1707 perambulation ran parallel to the apparent wall section. An extension of the resistivity survey, made by TWHAS in 2014 in the southernmost part of the outer Castle Gardens, revealed not only a continuation of the wall and moat but also a mass of demolition debris which appears to have largely filled it in. The 1327 castle survey records the outer bailey as having walls, barbicans and towers ‘on all sides’ of the castle (Keats-Rohan, 69) and no other buildings have been recorded in this vicinity since the castle’s demolition. The substantial debris therefore seems likely to be that of the destroyed outer wall and perhaps even a corner tower where the wall and moat turned eastwards along the back of the properties on High Street (see Figure 7.1). The original geophysics also showed an area of high resistance immediately north of All Hallows churchyard which could indicate the site of the church itself. The building would have been on the western side of the original Castle Street but after the expansion of the castle and westward shift of the road in the 13th century, it would have stood on the east side of the new street. From that time onwards the eastern extent of the churchyard of All Hallows marked the town and parish boundary with the castle (as described in 1707). The implication of these encroachments is that the western boundary of the earliest fortified enclosure lay immediately east of the original alignment of Castle Street and on the southern side, the raised bank at the back of the High Street properties may mark the survival of the original outer palisade of the earliest extra-parochial

fortification, much later capped with the stone wall that still surmounts it to mark the castle/borough boundary. Castle leaseholds The boundaries so far discussed encompassed the extraparochial precincts of the castle, the main defensive working area before the demolition. However, as KeatsRohan has demonstrated, the castle’s sphere of authority extended far beyond these confines, being ‘intimately linked’ with the Liberty of Clapcot within the parish of All Hallows (Keats-Rohan p. 38). These links are reflected in post-medieval leases which are worth considering in more detail. In 1600 Queen Elizabeth I leased out 33 acres of King’s Mead, lying next to the Thames to the north-east of the castle defences (see Figure 7.55), for the use of the mayor and burgesses of Wallingford who then sub-let the lands in individual leases, usually on an annual basis (BRO, W/TLc2). A survey of the borough of Wallingford in August 1606 (TNA, E315/369/3, fol 101-23) records a further seven acres of the ‘banks and ditches within the Castle’ leased for ‘herbage’ (cutting or grazing) for £1 per annum; it also mentions a ruinous ‘Gayle [gaol] House’ (later ‘Prison House’), with land adjoining it, lived in by Thomas Williams, who held his lease from Lord Knowles, the appointed constable of the castle. By the early 18th century these and other leases can be more clearly defined: firstly King’s Mead continues to be leased in parts to diverse people; secondly there is a further lease of castle meadow land called ‘north eight’ which lay to the north of King’s Mead and included rights to fishing in the Thames and small islands along the banks of King’s Mead (3.11 and Figure 7.55); thirdly there are the properties and land described as ‘formerly pertaining to the office of Porter or Janitor of the Castle of Wallingford’. The porter’s lands comprised two tenements in the High Street near Wallingford Bridge (see Figure 7.47), three acres of meadow land in and adjoining King’s Mead (see Figure 7.1), an acre of arable headland in Clapcot field and two acres of pasture in Newnham Murren, on the Crowmarsh side of the river (3.7). Together these castle leases reflect how in its heyday the castle would have been able to control key areas. Possession of the Clapcot meadowlands and riverside would have facilitated dominance over the usage of the river, while the porter’s field and meadows lay comfortably close to the northern approaches of the castle; his lands in Crowmarsh enabled a presence at the Oxfordshire end of the Great Bridge while his properties immediately inside the town’s East Gate would have ensured dominance of this crucial entrance to the town, adjacent to the castle’s main gateways. Other exertions of castle influence can be detected on all boundaries and approaches to the privileged chartered town of Wallingford: royal mills and the patronage of the ‘House of St John’ at the South Gate (Frontispiece); water control systems near the West Gate (Keats-Rohan 72); extensions near the North Gate, requiring the realignment of the main north route out of the town (Christie and Creighton 2013, 95). Each of 150

these areas will be revisited in due course but first the management of areas within the castle precinct itself needs to be considered. The details of three long term leases of castle lands from the mid-17th century provide a valuable guide to the later divisions of the castle precinct, particularly because they continued in much the same form until the sale of the castle in 1817. All had been subject to requisition by the Crown during the Civil War and subsequently by Parliament. First there was the inner bailey area, including the motte, described in a 1650 survey as ‘the inner ward of the Castle of Wallingford and the bank near encompassing the said castle wall with the moat and fishing thereunto adjoining’. This was the land leased to Thomas Freeman (and later his son John) but not surveyed at the time because the Parliamentary garrison was in occupation (TNA, E317/Berks/35). A second lease, held by John Freeman in 1656 for an annual rent of £3, comprised the use of the outer banks and ditches of the castle to the north, west and south for grazing (excluding the Christ Church holdings), described as ‘all the herbage feeding and pasture of the whole of the scite and precinct of the Castle of Wallingford now altogether decayed’ (TNA, E367/840). These lands had been given by Charles I to his Queen, Henrietta Maria, as a source of income; they were subsequently to pass to Charles II’s wife, Catherine of Braganza. The third lease was that of ‘The Prison House and backside’, mentioned above, which continued into the 18th century (TNA, E367/6786); it was another area demised to the queen and will be visited in some detail later. Quite separate from these three leases was the land and property in the southern middle bailey which had belonged to Christ Church College, Oxford, since 1548 and which continued to be sub-let in different parts until its sale in 1819. These leaseholds have been listed in some detail as they are the basis for understanding the complex history of the castle holdings from its demolition in 1652 to its sale in 1817 and, ultimately, the interpretation of surviving original features. Various leaseholders are known to have made changes to the buildings and landscape of the site and their activities provide the key to a better understanding of the original layout of the castle. Baileys and buildings Although much reference has already been made to ‘the inner’ and ‘the middle’ baileys and to the ‘outer banks’ defining the bounds of these spaces is inevitably difficult in a constantly evolving castle landscape. Keats-Rohan’s work on the castle survey of 1327, which verifies the presence of three baileys, has provided a valuable fixed point from which to begin (Keats-Rohan p. 69). The Inner Bailey The inner bailey is clearly the focal point of the castle, with its Great Tower, Great Hall, Great Chapel (St Nicholas), Great Prison and numerous other chambers and facilities. The sheer language of the descriptions leaves no doubt as to its importance and as late as c. 1540 John

Leland confirmed that the inner bailey still contained ‘goodly buildings’ (2.2). The 1548 castle plan (Figure 6.1) fortunately shows its entranceway, the site of the Great Drawbridge, so the area can be mapped with some confidence (Lloyd, above). This plan also shows, sketchily but unquestionably, a moat surrounding the whole circuit. Today this moat is still clearly traceable on the north, west and south (where it still holds some water); the land on the east drops dramatically to the ditch at the edge of Queen’s Arbour, where the moat must have been. The arrangement will be considered in more detail later. Within the inner bailey nothing but the imposing motte and two fragments of stonework have survived the ravages of time and Cromwell’s demolition gang, but an auditor’s report in 1656 of the inner bailey leasehold continues to make reference to ‘the bank next encompassing the castle wall’, suggesting that the area is still easily defined even though the walls are known to have been slighted by then; he also includes the moat and its fishing rights. No reference, however, is made to any buildings in the area (TNA, E367/2202). Lease descriptions tended to be recopied in full on renewal even when things had changed, as long as no rights had been lost, so the lease suggests that the moat continued to supply fishing. In this context it is notable that a new lease in 1726 states that the ‘said Inward ward and premises do now consist of a dwelling house, a garden, a small bowling green, and two orchards’ (TNA, E367/6898). This is the house depicted on the Renda plan of c. 1715 (Figure 7.46) and is an aid to placing other features on that rather enigmatic drawing (see below). The house was relatively short-lived, certainly gone by 1786 when it does not appear on the Hucks plan (Figure 7.47). The Outer Bailey The outer bailey in 1327 was clearly walled, with ‘barbicans and towers on all sides of the said castle’ even though they were in poor repair. Gaweynes Herber (Queen’s Arbour – see Keats-Rohan pp. 73ff.) is included ‘with’ the outer bailey on the east side. The clear implication is that the outer bailey encompasses the whole castle. As seen above, John Freeman’s lease in 1656 comprised ‘herbage feeding and pasture’ resonating with the 1327 outer bailey description where there is no mention of any building other than the defences; it was probably used for grazing stock animals throughout its history, as part of it still is today. The Middle Bailey The 1327 survey mentions several buildings in the middle bailey: a prison, a stable, and several other chambers, including La Studie (Keats-Rohan p. 83); there is also a bridge. Some of these buildings housed the lodgings of the clergy serving the Great Chapel of St Nicholas, known from the endowment by Edmund, earl of Cornwall, in 1282. Edmund specified that they should be given ‘that open space within the middle bailey that extends in length and breadth from the head of our great stable towards the east up to the gate of [called] Bydongate’ (1.28). Not only does this description contain a rare mention of the location 151

Figure 7.12 Remains of the Priests’ Lodging of the College of St Nicholas in the middle bailey, looking east. After the dissolution of the College in 1548 the buildings were variously used as accommodation, a malt-house and later a Victorian summerhouse. The surviving walls still incorporate medieval features and masonry. The sunken track beside the site is a remnant of the original route through the castle from the Bydongate on the east through the middle bailey to reach the Great Drawbridge and the inner bailey (photo: Judy Dewey).

of a gateway (discussed below) but it is a vital piece of evidence linking original medieval buildings with known successors that survived the Civil War destruction. Nothing has emerged from the medieval or Tudor documents to suggest that the location of these residences of the College of St Nicholas changed before they were acquired by Christ Church in 1548, and from that time they can be confidently traced to the ruins in the middle bailey today. Indeed, more is known of the history of this particular area than any other in the castle precincts. Excavations by Robert Carr in 1972 on part of this land (confusingly referred to by Carr in his brief report as ‘the outer bailey’) suggested that the site had been continuously occupied from the 11th to the 17th century. However, major landscape changes were detected: a 12th-century cob building, preserved to a height of 1.80m, complete with lime-washed walls, hearths and internal divisions, was revealed to have been totally and deliberately buried by soil sometime in the 13th century, probably in the time of Richard, earl of Cornwall, when the outer defences were being extended. (Carr 1973b, 18; see also Christie and Creighton 2013, 183–198 for full analysis and discussion). Although dating evidence is not robust for this re-modelling of the landscape of the middle bailey, such an infilling would have created a clear stretch of ground, a likely ‘open space’ for the endowment of the college after 1278 (Keats-Rohan p. 71); certainly the college houses were built at the new, higher, level. Today part of the middle bailey land in question is the upper level of Castle Gardens, accessed by steep steps from the outer bailey and bounded on the south and west by the imposing middle rampart that faces the visitor entering from Castle Street (Figure 7.12). Location of the college buildings is the key to understanding the layout of the whole middle bailey and other important features of the castle. It is remarkably fortunate that a full description was included in the survey made by the Court of Augmentations in 1548 and this deserves detailed attention (3.5). As has been noted,

there were three main buildings: the Dean’s Lodging, the Priests’ Lodging and the Clerks’ Lodging. No map was appended to this document, but the1804 plan (see Figure 7.48) and the more detailed 1812 plan of the Christ Church properties (Figure 7.54), commissioned by James Blackstone during a law suit (see below), show buildings which clearly link to the 1548 descriptions and to the features surviving today. By 1804 only the former Priests’ Lodging and Clerks’ Lodging could be identified, the Dean’s Lodging, having been ‘utterly demolished’ in the Civil War ‘and afterwards the ground whereon it stood plow’d up by Mr Freeman, and things so very much alter’d that the bounds of the College ground could not be found’ (CCA, Book of Evidences, 904). It was labelled as ‘grass ground’, ‘part of a garden’ and ‘part of an orchard’ on the 1812 plan. Today, the site of the Dean’s Lodging lies outside the eastern extent of Castle Gardens, an area which will be discussed in more detail later. The Dean’s Lodging was a substantial property with several rooms listed that are familiar in many late medieval buildings, notably the hall, parlours and chambers and the ‘gallery’, a popular Tudor addition. It may have been a stone building but could equally well have been of timber construction. Remains of the Priests’ Lodging can still be seen in the eastern part of the upper Castle Gardens with a surviving section of wall marking the southern boundary and part of the western stone wall of the lodging running north from it. (Figure 7.12). The latter still has a c. 14th-century stone doorway in its probable original position with a much worn threshold stone, two eroded lion-head decorations adorning it on either side and set into an adjoining stretch of wall that displays many large dressed stone blocks and features of windows (Figure 7.13 and Figure 5.20 lion’s head). Although the wall has clearly been much patched and altered, its continuous use over several centuries suggests that the Priests’ Lodging was enclosed with a stone wall even though its inner buildings may have been timber. Lying immediately to the west of the Dean’s Lodging, it 152

Figure 7.13 Detail of part of the wall of the Priests’ Lodging in the middle bailey (photo: Judy Dewey).

took the form of a group of buildings surrounding a small cloister quadrangle, the general shape of which can clearly be seen on the plans of 1804 and 1812 and is still just traceable on the ground today; part of the eastern boundary wall of Castle Gardens aligns with the eastern extent of the Priests’ Lodging. The general arrangement of the building has been likened by Colvin to the 15th-century Bede-house (almshouses) at Ewelme (HKW i, 300). College accounts record work being done for a visit by Queen Elizabeth in 1570; it seems that the Queen slept in one or other of the college buildings, so they must have been considered royally acceptable at that date (3.6). The Clerks’ Lodging was the most westerly of the College buildings, separated from the Priests’ Lodging by an orchard (under which lay the earlier cob buildings) and with its own stable and larder. It appears on the early 19th-century plans but only fragments of it survived after the building of the Victorian house in 1838. Some of the materials from the old coach house and stable of the Clerks’ Lodging were reused in the new mansion to save money, together with a ‘wind-up jack’ from the old kitchen and a bell for the ‘top of the house’ (BRO, D/EH6 Hedges papers); the bell was perhaps even the one referred to in the 1548 survey by the Court of Augmentations (3.5). It is also likely that the well, described by J. K. Hedges as lying ‘near the south-west corner of the scullery door’, formerly served the Clerks’ Lodging; it was 29ft 6in (9.0m). deep and still holding 12ft 6 in (3.8m) of water in 1890 (BRO, D/EH E26, 66). The only named properties within the castle for which any form of compensation was given after the Civil War were those held by John Freeman and Henry Knapp. (Michael Mollyns also received compensation, but his house and land lay outside the main castle precincts: Dewey pp. 246–7, below). John Freeman complained only of damage done to his gardens and orchards when the walls were destroyed, and although he claims the substantial sum of £300 in compensation, no reference is made to a

house. Henry Knapp, however, was paid compensation for the use of his house from 1646 for six and a quarter years by the governor, Major Evelyn, during the Parliamentary occupation which preceded the demolition of the castle in 1652. The location of this house cannot be certain but since no mention is made of its loss, it suggests it was a building that survived the demolition. The simplest explanation is that Knapp held the lease of one of the college buildings at the time. Knapp’s compensation of £125 was agreed in 1652 but not paid until 1654 when it was to come ‘from the revenue in Ireland, whence he is going’ (CSPD iii 1651, 17; 1654, 316). However, in 1657 Knapp is listed as leaseholder of the Clerks’ Lodging (CCA, Book of Evidences, 907) which seems to be the most likely building in question. These details shed some light on what happened during the demolition of the castle. It is easy to condemn it as random destruction but the fact that only limited compensations were demanded suggests that it may have been more carefully handled. The evidence strongly suggests that the inner bailey, the hub of the castle, was totally denuded of its buildings, together with all the storehouses and additional buildings constructed by the royalist garrison (3.9). The castle walls were also destroyed, which caused the damage to John Freeman’s property and was almost certainly the reason why the Dean’s lodging, lying nearest the castle gate (see below), would have suffered. All this destruction was deemed necessary to avoid any future use of the site as a military base. However, the land of the middle and outer baileys, sequestered from the Crown, were of real value, and needed to be protected as a potential financial asset to the Commonwealth. Evidence of sales and castle leases begin to emerge from 1652. The Priests’ Lodging and Clerks’ Lodging, belonging to Christ Church and not a threat to military re-occupation, survived as leaseholds: the Priests’ Lodging became a malt-house by 1669 and continued in use for many years (see Figure 7.54), while the Clerks’ 153

Lodging remained as a separate residence. The latter was improved over the years from the original ‘little house and kitchen’ (3.5) to a desirable ‘Mansion House… with kitchen, cellars, solars and chamber’ by 1804, retaining the old orchard and stable until it finally succumbed to the Victorian mansion (BRO, D/EH T1(2)). The middle bailey was much more extensive than just the college area. As has been seen, it was bounded by the rampart which is still well-defined today, bordering the college lands on the south, then turning northwards, passing through what is now Castle Gardens and then, somewhat hidden amongst the trees, continuing through the private land parallel to Castle Street. The deepcut Castle Lane, the modern bridge over it and the gate between Castle Gardens and Castle Meadows all divide the middle bailey today, but none of these features existed before the 19th century, and so the middle bailey would have linked seamlessly to the present plateau of ground between the rampart to the west and the well-defined inner moat to the east. This is the area where the Great Stable must have stood, according to the 1282 reference which placed the college land eastwards from ‘the head of the Great Stable’ (1.28). Fourteenth-century accounts refer to the gaol building lying next to this stable, one of several prisons in the castle (Keats-Rohan, p. 82); this structure was sometimes referred to as ‘The Blindhouse’, a name probably derived from a lack of windows for security purposes. Repairs to prison buildings often appear in the castle accounts, but in 1438–40 a substantial new prison house was built, with its layout and dimensions fully recorded – about 70 x 42ft (21 x 12.8m) overall (Keats-Rohan, pp.

82–3). The exact location was not given for this new solid stone two-storey building enclosing secure rooms and accommodation for the gaoler, but later leases of a ‘Prison House’ suggest it may have had an extended history, perhaps surviving the demolition of the castle because it lay outside the main fortress of the inner bailey and still held potential as a valuable lease. The ruinous ‘Gayle House’ (also called ‘Prison House’) has already been noted in the 1606 town survey (TNA, E315/369/3); after the Civil War it emerged in Chancery proceedings in 1686 concerning the estate of the late John Freeman, described as ‘all that house and backside called the Prison House standing in and being upon the walls of the late Castle of Wallingford’ (TNA, C6/260/111). Leases of ‘The Prison House’ and its lands continued unbroken throughout the 18th century, first found in a further Chancery proceeding concerning Thomas Renda in 1700 (TNA, C109/433) and from 1722, in a series of renewed Crown leases to William Hucks and his successors up to at least 1775. A report on 20 September 1725 includes reference to the demolition of the Prison House but states that there remained an ‘old barn called the Gaol Barn’ (Calendar of Treasury Papers, K553,1720–28, 365:43). No other documents give any indication of this apparent demolition, but a ‘Gaol Barn’ is certainly referred to and appears on the Renda map, c. 1715 with a dove-cote adjacent to it and a dotted line indicating a trackway leading from Castle Street past the barn and on towards the college land (see Figure 7.46). A further mention refers to the track as leading ‘from Castle Street to the Gaole Barn’ and thence onwards (CCA, MS Estates 5, fol. 60). Gough, visiting the castle in 1768, confirms the location of this building, stating that ‘towards the W end

Figure 7.14 Thomas Athow’s ‘View of the Remains of Wallingford House Berkshire’, a watercolour painted c. 1809 (image © Ashmolean Museum, University of Oxford).

154

of the N bank within is an old stone building now a barn formerly part of the base court’. A question is raised by these varied accounts, as to whether the ‘old barn’ was in fact the ruinous survival of the 15th-century prison building. Gough’s words suggest that the ‘barn’ was of some antiquity and the known paucity of stone in Wallingford makes it unlikely that any such great stone structure would have been built from scratch during the Civil War. This large rectangular barn, with dimensions potentially similar to the 15th-century prison, was still in use in 1804 as a ‘wheelers shop and smith’s forge’ in the same position as Renda’s Gaol Barn, but with a small, newer building, adjacent. (see Figure 7.48). A painting by Thomas Athow entitled ‘Wallingford House’, considered to date before1809 and recognizably on the castle site, also depicts a large rectangular stone building in poor repair (Figure 7.14). It sits on a moated rampart with the motte on the left in the distance beyond it; the artist appears to have been looking south-east from the north-western outer bank of the castle. There is no other candidate for this building other than the Gaol Barn – a very substantial ruined stone building, in an area where stone is a rare material, standing in what was by then an open rural landscape. Athow does not show the small newer building (perhaps it was not there when he drew the picture, or maybe he chose to omit it), but in 1829 both buildings passed from James Blackstone to John Allnatt Hedges, (BRO, D/EH T1,1/2). Described as a ‘barn and building standing on part of a triangular piece of land’, they were demolished soon afterwards; they would have been an eyesore to the new Hedges residence. (The wheelwright’s was later rebuilt in brick tucked into the outer castle bank to the north, the structure surviving into the late 20th century: see Figure 7.43.) Interestingly, the association of the site with a prison lived on: the only obvious ‘triangular piece of land’ is noted on the 19th-century OS maps as ‘site of Cloere Brien’ (see Figure 7.8) – a reference to the notorious 12thcentury prison in which William Martel was apparently incarcerated by Brien fitzCount; the narrow prison was jokingly said to have been made as a cloaca (bag) in which to put the martellus (hammer), punning on the prisoner’s name (Matt. Paris II, 174). J. K. Hedges (who must have had a hand in the ‘Cloere Brien’ nomenclature), recorded in his history of the castle: ‘The last trace of a dungeon… with huge iron rings fixed in the walls, disappeared about sixty or seventy years ago. It is probable some portion of the buried remains may yet be found between the first and second moat on the west, where a hollow in the ground seems to mark the spot’ (Hedges 1881 i, 245). Substantial stonework was identified by resistivity in this area but lay just outside the limits of the 2011 excavations and so it could not be investigated (Christie and Creighton 2013, 212). Getting in and keeping out – changing gateways The usage of the castle site after the Civil War led to many alterations in its landscape, mostly relating to access for its leaseholders to their property. Understanding the impact of these changes presents a challenge, but it is one

that must be faced if there is to be a proper attempt at an interpretation of the medieval castle layout. The method used has been to commence with the most recent changes and work chronologically backwards. Modern and Victorian Today there is public access on all sides of the castle grounds: north, through a gate near the cemetery; east, through another gate from the riverside path which leads in through King’s Mead; west, off Castle Street along the western end of Bear Lane, skirting All Hallows churchyard and entering through a small stone gateway into Castle Gardens; south, off the High Street near Wallingford Bridge up Castle Lane, giving access to the riverside walk, Thameside Mansion, the upper part of Castle Gardens (through gates that are usually kept closed) and diverting to cut right through the castle site until turning a corner to emerge on to Castle Street. There are also private entrances: Bear Lane provides access to properties with grounds behind the High Street, there is an entrance to the Victorian Lodge off Castle Street and, off the High Street, an imposing stone pillared gateway leads to a deep lane and more gates, into Castle Lane House. The first two public entrances (north and east) are new, created when South Oxfordshire District Council opened the Castle Meadows for public use in 2001, but the history of the other access points provides the insight needed to unlock the evolution of the castle landscape. These will be approached geographically, beginning with the western side, then assessing the more complex southern and eastern approaches. (Figure 7. 15) Castle Street runs north along the western boundaries of the castle site and the creation of the western entrance to Hedges’ Victorian house required the demolition of some older cottages on the east side of the street (see Figure 7.18). The drive led through a gateway and up a slope to the house; there it met the more circuitous and scenic carriageway which entered off the High Street through the pillared gates (Figure 7.16) and circumnavigated the inner and middle bailey to approach the house over a newly constructed bridge across the diverted Castle Lane. The 1835 map shows the former route of Castle Lane, as it was before the demolition of the Clerks’ Lodging and the building of the Victorian mansion; it is marked as a ‘road’ on the Christ Church plan of 1812 (see Figures 7.56 and 7.54). It was actually a cart-way from the High Street (near the bridge) leading into the middle bailey and giving access to the malt-house (former Priests’ Lodging) and the Clerks’ Lodging before narrowing into a footpath which finally turned sharply west to link with Castle Street; carts could not get right through to Castle Street, having to turn and go back the way they had come. The origins of this link to Castle Street derive from James Blackstone’s re-landscaping on the southern side of the castle after his purchase of the site in 1817/19. Blackstone enclosed his new castle property with wooden palings, closing up what had evolved into more or less free access across the site since previous fences, erected by Thomas Renda in the early 18th century (see below), had fallen 155

Castle Entrances - 18th & 19th Centuries

KEY Renda’s driveway c.1700 Possible site of Renda’s house Approximate route from Castle Street c.1715 (Renda map) Original access to Christ Church land Blackstone’s farm lane c.1828 J.K. Hedges’ carriage-way post 1858 J.K. Hedges’ entrance from Castle Street Diversion of Castle Lane post 1858

Figure 7.15 Castle entrances 18th and 19th centuries [no caption on page, built into map] (illustration by David Dewey).

Figure 7.16 Entrance gates to a scenic carriageway leading round the motte and inner bailey to the Victorian mansion (photo: Judy Dewey)

156

Figure 7.17 Looking southeast towards the middle bailey. The brick-walled Castle Lane was diverted into the old moat by John Kirby Hedges. The gate in the wall once opened on to steps into the gardens of his Victorian mansion (photo: Judy Dewey).

into disrepair. To maintain the perceived public right of way through the castle site, Blackstone generously added the narrow footway link from the middle bailey to Castle Street, which emerged between two properties (116 and 117 on the tithe map of 1841: BRO, D/D1/136/1). At the eastern end he gated the entrance to the middle bailey (now the locked gates to the upper Castle Gardens) but allowed free public access through them. The route was little used, but later its close proximity to the grand new mansion of J. K. Hedges became an irritation, so in 1858 Hedges tried to have the right of access closed. In his case to counsel, which described the recent history of the lane, he wrote of the footway at the Castle Street end that it was ‘a narrow dirty crooked lane, with a high wall on one side and a paling on the other’. He said it was not much used ‘except by bad characters, it being just without the Town’, presumably meaning it was on castle land (BRO, D/EH E8). The attempt to close the right of way was disputed, so, at his own expense, Hedges diverted the mid-section of the old road away from the middle bailey slightly to the north and partly into the drained inner moat, to make the new rather muddy and high walled present route of Castle Lane (Figure 7.17). It thus removed it from his house and garden and linked it on the east to the old way to Wallingford Bridge. Hedges had two entrances off the lane into his property, one (still existing) leading into the lodge area near the Castle Street end, the other a stone doorway in the high wall which led by steep steps into his garden. The latter gateway is still visible in the wall but is now totally blocked off and the walls have undergone much renovation. Hedges also created a bridge over the lane for the grand carriageway described above; today a reinforced bridge carries the gateway linking Castle Gardens to Castle Meadows.

18th-century impact Now that these more modern changes to the western castle landscape and the one surviving route through the castle have been established, the earlier maps of the castle become more approachable. The 1804 Crown plan (Figure 7.48) and the 1786 Hucks estate map (Figure 7.47) show a landscape partly dominated by trees, planted, it would seem, both for show and for value. The western ramparts are lined with an avenue of trees and there is no obvious entry to the castle from the west. The earlier Renda map, c. 1715 (Figure 7.46) shows a tree-covered motte and southern outer bailey, with more trees alongside the churchyard on the west and in the outer ward to the north. The western ramparts, however, are clear of trees; the land is enclosed by a fence and a small stretch of wall abutting All Hallows churchyard but with two entrances through the western boundary. Firstly, there is a way through the wall south of the churchyard, approximately where the present public entry to Castle Gardens lies; secondly, lying north of the churchyard and three cottages (probably those later demolished by Hedges) a barred gate is shown off Castle Street which abuts an isolated short stretch of wall with a turreted tower attached to the south of it (Figure 7.18). The latter feature is particularly interesting as no other reference to it has been found and, although the artist’s general perspective and scale leave much to be desired, this feature seems to be a deliberate depiction of something unique on the plan. Could it be a remnant of castle fortification, or, perhaps more likely, a last survival from the church of All Hallows, destroyed in the Civil War? Whatever it was, it would have been removed during the Victorian development, if not before. The gate off Castle Street is referred to in the lawsuit brought by Christ Church College against Thomas Renda 157

Figure 7.18 Detail of churchyard, cottages, gate and ‘tower’ on Castle Street. From map of c. 1715 (see Figure 7.46). Image reproduced by kind permission of the Governing Body of Christ Church, Oxford.

c. 1715. Renda was said to have closed up an ‘ancient’ access to the college and a witness testified that it had been ‘a large cartway with a spar’d gate from the castle street over the Bank up to the Gaol Barne, and from thence over other banks under a gatehouse which stood near the said College’ (3.10 [66]). The Renda map (Figure 7.46) clearly shows the gateways in question linked by the dotted line of the carriageway. More than one complaint was made that Renda had blocked off this track to carriages, carts and those on foot by ploughing up the ground and sowing it with corn and that, with the connivance of Anthony Leaver, tenant of the Clerks’ Lodging, he had pulled down the earlier gateway to the college lands. This interesting former college entrance in the middle bailey was once a substantial two-storey structure, ‘a gateway or grainery, under which carts and carriages passed’ (3.10 [59]), standing at the corner of the stables which were part of the leasehold of the former Clerks’ Lodging. Renda put in its place ‘another gate which is most times locked’ and ‘upon which gate the defendant did set spikes [or ‘bussels’, according to another deponent] whereby foot people were also debarred of their passage there’. Further witness reports refer to his removal of the old footings (‘ground pinnings’), so it is clear that he totally destroyed it (3.10 [66–67]). To be effective, this old gatehouse, with its upper storey used as a grain store, must have linked to some sort of protective boundary fence around the college property, possibly set up when Christ Church purchased the land in 1548. It could even have been part of a medieval enclosure of the college buildings, but it seems unlikely to have been an integral part of the castle defences if it survived the destruction of the Civil War. It was clearly no longer an effective barrier when Renda removed it; his new gate,

shown on his map, was a smaller affair, but its vicious spikes obviously did a better job of keeping people out. The route of the track from the west, which Renda ploughed up, had been dictated by the landscape. The medieval documents make no reference to a western entrance off Castle Street, where the outer moat would have been an impassable barrier. The implication must be that part of the outer ditch had been filled or bridged to allow the carriageway to be created, but, as the marked route of the track and the witness statements testify, having crossed the ditch the track followed the crest of the ramparts. The origins of the western gate and section of the track cannot be traced precisely, but it seems to be a post-medieval creation for the convenience of those doing business with the surviving houses in the middle bailey. It is most likely to have been part of the reconstruction of a devastated Castle Street area after the 17th-century Civil War (Dewey 245 below). The section of the track from the gaol barn to the college, however, is almost certainly a remnant of the main medieval route through the middle bailey. Assessing the evidence for the medieval defences There is no specific reference in the medieval or later records to a gateway into the castle on the north-west, but an excavation report of work in this area in 1966 states: ‘At the extreme southern limit of the site a cobbled roadway seating two rubbish pits of the 12th century was excavated. The cobbles were set in a solid mortar of chalk and greensand. The alignment of the roadway established that this was an entrance to the castle, in all probability the main entrance from the 12th century until the royalist defence of the castle in the Civil War’ (Brooks

158

Figure 7.19 Bird’s Eye view of the castle from the north. The ‘barbican’ can be clearly seen projecting into the western end of the inner moat, while the sites of three probable towers are prominent on the outer rampart (photo: David Dewey).

1966; Christie and Creighton 2013, 120–2). The medieval documents make it quite clear that this was not the castle’s main entrance, but a way into the castle near the North Gate from within the town would have been highly convenient for bringing in goods and visitors from the Oxford direction. What seems most likely is that this cobbled way led into a protected D-shaped barbican, the distinctive shape of which is clearly discernible in aerial photographs (Figure 7.19); the presence of enclosing walls can also be detected on the resistivity survey (see Figure 5.1). Similar examples are known at Goodrich Castle and the Tower of London (Goodall 2011, 210, 201), and at Wallingford the barbican also provided protection for the entrance to the inner bailey across the Great Drawbridge. Excavations on the ‘barbican’ in 2011 did not reach the medieval levels as they were found to be overlaid by a large 17th-century clay platform, almost certainly a gun emplacement similar to the one found on the northern castle perimeter in 2008. Medieval material had been thrown up against the original rampart to create this Civil War defence and had disturbed the profile of the neighbouring landscape; any medieval stonework of a barbican in situ remains hidden beneath the Civil War material (Christie and Creighton 2013, 209–15). Nevertheless, as Keats-Rohan has demonstrated, whilst there are no specific references to a north gateway, there are several references to barbicans, one of which, in 1367, specifically relates to this north-west area (Keats-Rohan, 71). The medieval sources have also shown clearly that the south-eastern entrances to the castle were its most

important. The southern part of the castle, and in particular its south-east corner, presents the greatest challenge in understanding the original landscape for it has been greatly altered during the intervening centuries. The 20th century saw extensive changes: the building of two modern properties off Castle Lane – Castle Lane House (1950s) and Castle Farm House (1970s) – the demolition of the large Victorian house in the middle bailey (1972), and the removal of farm buildings prior to the erection of the present Thameside Mansion overlooking the river (1997/9) (Figure 7.20). Archaeological excavations associated with the developments have been illuminating, if limited. Prior to and during the building of Thameside Mansion it was revealed that there had been earlier levelling of ground for the footings of one of the original farm sheds (Ford 1995, Trench 2). Similar observations had been made in a 1992 investigation, at Castle Farm when a farm shed was rebuilt, but that excavation had also recorded a probable bailey ditch (OAU 1992). Work in 1997 revealed further tip-lines of crushed mortar which were interpreted as ‘almost certainly’ evidence of ‘bank or rampart material forming part of the outer bailey of the castle’ (Saunders 1997, 4), though it is worth noting that the name ‘outer bailey’ is loosely used in such reports. Documentary evidence indicates a substantial medieval wall on the eastern edge of the castle which would have surmounted this eastern rampart (Keats-Rohan, p. 84). The fragment of surviving wall to the north-east of Thameside Mansion is all that remains of it today. Until the early 1970s this was a much longer stretch of wall (see Figure 7.4) but it

159

Figure 7.20 Aerial view looking south. Thameside Mansion is in the foreground, with Castle Farm House behind it (slightly left) and Castle Lane House, with its expanse of lawn, to the south, beyond the courtyard (photo: David Dewey).

Figure 7.21 Fragments of the fallen eastern curtain wall, silhouetted against the flooded river in 2014.The bridge is visible in the background (photo: Judy Dewey).

160

was by then somewhat hazardous and it collapsed one day, depositing its scattered remains on the eastern slope of the rampart (Figure 7.21). Another section of ditch was observed at Castle Lane House during the building of a garage on the north side of the house in 1990 (see Figure 7.20); the report of this clay-lined ditch-fill suggested that a curtain wall ran ‘at a higher level built on the crown of this dumped material’ (CBA 1990, 85–6). This was most probably part of the castle’s middle wall, depicted in a sketch map and description of 1810 drawn and noted by Edward Driver (Figure 7.53 and 3.14) which shows the line of a wall/moat edge running approximately on the observed alignment. Driver’s map, by his own admission, is only a sketch, but the noted line is valuable evidence. On the Christ Church map (see Figure 7.54) the southern boundary of the college property is not marked as a wall to the east of the buildings, but its alignment closely follows that on the Driver map; indeed it is possible that the Christ Church

boundary was fixed by the course of the original middle wall (see discussion below). The results of geophysics in the garden of Castle Lane House have suggested the presence of another wall, south of the house and crossing the garden west to east (see Figure 7.37). This is also indicated on the 1810 plan and seems likely to be part of the 13th-century southern outer wall of the castle (see below). Clearly, this south-eastern corner of the castle has been an area of considerable re-landscaping over the centuries and it remains to unpick the documentary and archaeological evidence more thoroughly to demonstrate what has changed. The 1810 plan (see Figure 7.53) shows the land on which Castle Lane House and its garden now stand to have been partly a large raised bank and partly a ditch, but there is no sign at that date of the wide entrance lane with the stone-pillared gateway leading off the High Street opposite Thames Street; this is clearly a later route into the castle and can be identified as the work of James

Figure 7.22 View of Blackstone’s track from the High Street entrance; the track continues through the gates to pass under the bridge he built to carry Castle Lane. Note the puddles gathering in an area which would once have been the outer moat of the castle (photo: Judy Dewey).

Figure 7.23 James Blackstone’s bridge, which still carries Castle Lane over his former cart track. Note the high banks on either side where the track has cut north through the former medieval middle rampart (photo: Judy Dewey).

161

Blackstone who, as noted earlier, purchased and enclosed much of the castle in 1817/9; his residence, however, was at Castle Priory, sited in Thames Street, not on the castle site. In 1828 Blackstone bought an additional strip of castle land ‘east of the malt-house’ (the old Priests’ Lodging, BRO, D/EH T5 12) and this was the land used to create a new driveway into the castle estate. It provided a convenient link from his residence in Thames Street and was a private route to his new farm – now the site of Thameside Mansion. Map evidence confirms that Castle Farm dates to about this time: it first appears on the 1835 map (Figure 7.56), but was not in existence in 1817 when the castle land was sold by the Crown. The two grand stone pillars flanking the entrance to this cartway off the High Street (Figure 7.16) may have been Blackstone’s work, but they more likely belong to the post-1858 period when J. K. Hedges used Blackstone’s sunken lane to create part of his new imposing driveway round the inner bailey and then over Castle Lane to his mansion (above, p. 155 and see Figure 7.15). The creation of Blackstone’s new entrance would have involved substantial excavation of soil and also the building of the bridge which still takes Castle Lane (a public route) across Blackstone’s carriageway (Figure 7.22 and Figure 7.23). Supporting the indications that these major access changes were James Blackstone’s initiative are references to his activities on the site: the works would have required the draining of the inner moat around the motte (through which the new lane partly ran) which Blackstone is reported to have achieved by locating and opening a sluice on the ‘southern side of the moat, just below the Keep [motte]’ (Allnatt 1873, 48). The route might also have been expected to disturb the remains of a ‘passage’ discovered by Thomas Renda in about 1700 and subsequently described by John Kirby Hedges as running eastward ‘from the bottom of the keep on the south side, in the valley’, so it is interesting to note that in 1820 Blackstone recorded his own exploration of this feature (Hedges 1881, i, 151; see also Keats-Rohan, note

2, p. 46–7 for a full discussion of this possible medieval postern gate). As has been seen, Blackstone’s dramatic alterations to the landscape were not the first to have been associated with private entrances to the castle grounds. Thomas Renda’s 18th-century activities on the western approach to the castle have already been described, but on the south side he was also responsible for intrusive landscaping. The Renda map (see Figure 7.46 and Figure 7.24) shows a formal tree-lined carriageway leading from a spiked iron gate close to the bridge to a second, more ornate iron gate, flanked by walls, beyond which lies an apparently early 18th-century style house within a walled inner bailey. The avenue is bounded by a fence on the west and by a wall on the east; north of his inner gate the eastern boundary has ornate railings. The carriageway, with its avenue of walnut trees, was the main cause of an accusation of encroachment by Christ Church College in 1715/9. As was normal procedure in such disputes, several elderly witnesses were called to testify their knowledge of the ancient boundaries (3.10). It is apparent from the accounts of these deponents that Renda had considerably altered the landscape in creating his imposing entranceway and in so doing had most probably encroached on the half acre of college ground on which the Dean’s Lodging had once stood, shown as ‘grass ground’ on the Christ Church plan of 1812 (see Figure 7.54). The root problem was the lack of definition of the boundaries and it took more than a century to get final agreement, meanwhile creating much legal documentation which has proved very useful to an understanding of the landscape uses and changes. As has been seen, there was no map of the original sale of land by the Crown in 1548, but it seems highly unlikely that any defensive wall would have been sold at that time. Although by 1555 assessments of the value of timber and stone suggest the castle was becoming more a financial asset than a stronghold, accounts from the early Tudor period certainly indicate that the castle was still

Figure 7.24 Detail of Thomas Renda’s gates, avenue of trees and house. From a map of c. 1715 in CCA, MS Estates Book 5, image reproduced by kind permission of the Governing Body of Christ Church, Oxford.

162

Composite of 19th-Century Maps and Key Locations

Outworks King’s Mead

Barbican 4

Middle Bailey

Inner Bailey

Castle Street

Outer Bailey

7

Queen’s Arbour

Motte 2

6

Middle Bailey

8 1 3

Oute

r Ba

10

9

iley

5

11

High Street Brid ge

To Crowmarsh

KEY Christ Church property, 1812 Buildings on 1804 map Additional buildings, 1812 Water, 1804 Victorian Pond 1804 shoreline

Chapel Eyot, 1804

1 Site of Priests’ lodging

7 Victorian Coach House

2 Site of Clerks’ lodging

8 Thameside Mansion (site of farm)

3 Site of Dean’s lodging

9 Castle Lane House

4 Gaol Barn

10 Castle Farm House

5 Munt’s Mill

11 Site of Winch Eyot

Site of town East Gate to 1809 6 Site of Victorian mansion (1837 - 1972) Line of ramparts, 1804

Line of ramparts, 1810

Municipal Boundary, 1878

Figure 7.25 Composite of castle features from maps of 1804 (see Figure 7.48),1810 (see Figure 7.53) and 1812 (see Figure 7.54) overlaid on an aerial photograph of 1973 (see 7.43). (Copyright reserved Cambridge University Collection of Aerial Photography.)

163

being maintained into the early 16th century (1.71–3). The standing walls would, however, have formed convenient markers to delineate the area for purchase. It has already been suggested that the substantial college gateway on the west (replaced by Renda) was probably part of a fenced enclosure of the college lands on a stretch of boundary which was unlikely to have had a stone wall to define it in 1548, whilst on the south the castle’s middle wall would have provided an obvious boundary. At least part of this wall survived the ravages of the Civil War and it is worth considering whether the southern Christ Church boundary might be an indication of the line of the original medieval middle wall. Along the northern side of the Christ Church land the moat edge formed a potential boundary, but, to be sure of clarity, this and the eastern boundary (which was to be the main cause of contention) were certainly marked by the ‘meer stones’ which Renda was later accused of moving to his advantage (see Figure 7.46). By the 19th century, walls and meer stones were mostly long gone and the debated boundaries had to be settled in court. When the college drew up the 1812 map, seeking a resolution to the boundary dispute, nearly a hundred years after it had begun, it claimed the land according to boundaries described by witnesses in the 18th-century hearings. These testimonies therefore give valuable insight into the earlier landscape and deserve careful scrutiny (3.10). Complementing these written statements are features shown on the early 19th century maps; overlaying these plans on a modern photograph has provided a revealing and helpful composite (Figure 7.25). Most interesting of the testimonies is Robert Merwick’s assertion that the college land lay between two moats, north and south, (almost certainly still water-filled at the

time) and from the malt-house at the west end ‘to a place where a drawbridge formerly was at the east end and so along to another place where a large stone formerly lay by the castle gate’ Considered in conjunction with Renda’s map of c. 1715 (Figure 7.46), the Christ Church plan of 1812 (Figure 7.54) and Edward Driver’s sketch of 1810 (Figure 7.53), several interesting facts emerge from this witness. The moats referred to must be the water-filled inner moat, surrounding part of the motte on the north side of the college land and the middle castle moat, surrounding the college holdings on the south – still discernible today below the ruins of the wall and Priests’ Lodging (Figure 7.26), though now much in-filled towards the east. Merwick’s mention of the site of the former drawbridge at the east end of the college land accords with the medieval evidence that the Bydongate (the middle gate) lay east of the college buildings (1.28). The drawbridge there would have spanned the middle moat described. More enigmatic is the phrase ‘and so along to another place’ for the position of the stone which had once stood ‘by the castle gate’, but he seems to be referring to the middle gateway itself, associated with the drawbridge. Merwick was 78 when he made his testimony, so would have been 11 or 12 years old when the castle was destroyed. His childhood would have been spent during the Civil War and as a boy he must have been aware of the layout of the castle, so his is a rare and important eye-witness account. The medieval documents indicate that the area in question was part of the main entrance complex of the castle, sited close to Wallingford Bridge, with an outer gateway (the Constable’s Gate) and a middle gateway (Bydongate) in close proximity (Keats-Rohan, pp. 79–81; Figure 5.35). It is likely that the ‘narrow footway’ mentioned by Merwick

Figure 7.26 Looking south-west along the line of the middle moat below the middle wall (photograph: Judy Dewey)

164

Figure 7.27 Castle Lane looking north. Munt’s Mill is on the right. Notice how the lane gradually slopes upwards (photo: Judy Dewey).

as lying to the east side of the two rows of walnut trees lining Renda’s entrance was a remnant of the original entry route. Further testimony by a Mr Hobbs reinforces the location of the track and the existence of a former college boundary marker stone as ‘beyond the lowest rank or row of walnut trees now planted’. The overlaying of the various early maps on to the present map clearly suggests

that the present Castle Lane leading from the end of the bridge towards the recently built Thameside Mansion is the modern survival of the southern part of Renda’s drive (Figure 7.27). The making of Renda’s new driveway had required major earthworks. Hobbs spoke of it being ‘digged down levelled and laid into a way or walk’ and that this had altered the eastern part of the college ground. William Maxey said much the same, making it clear that Renda’s digging had been to the west side of the previous trackway. The testimony of Owen Tuder adds another important piece of information – that the gate and stile entrance to the lane before the alterations stood ‘between the highway or street and the banks’, which suggests there were remains of castle ramparts still in place that would have needed to be levelled to create the new drive. This perhaps explains why the garden of the present Castle House stands at a higher level than the lane; it also accords with the resistivity survey result in Castle House garden which showed some indication that the third, outer, wall may have run across it on the crest of a rampart (see Figure 7.37). Edward Driver’s description and map (3.14 and Figure 7.28 and Figure 7.53) says this area ‘between the two moats’ (inner and middle) was ‘distinguished by its resemblance to a Hog’s Back’, noting that it was ‘formerly a necessary fortification to the place’. All this clearly

Figure 7.28 Map drawn by Edward Driver, 1810 (see 7.53).

165

Figure 7.29 Resistivity survey of Munt’s Mill made in the 1990s (image: Gerard Latham).

suggests that part of the ‘grass ground’ of the college once formed a defensive rampart. It would certainly make sense that this rampart carried the middle wall of the castle towards the middle gate and drawbridge, somewhere near the present Castle Farm House, while the outer wall, moat and gateway lay to the south, beyond the college property, on the land between the town boundary at the back of the High Street properties and the middle moat, which would include part of the gardens of both Castle Lane House and Munt’s Mill, the property on the east of the present Castle Lane (see Figure 7.25). Appraisal of the southern gateways After long and careful consideration of the land, maps and overlays of maps, it seems likely that the southern Christ Church boundary marks the alignment of the castle’s middle wall, terminating at the site of the Bydongate, the middle gate of the medieval castle, which lay beyond the eastern boundary of the Christ Church land. In the making of his new driveway, Renda would have had to cut out any remnants of the rampart to ease the upward slope of the track; it still rises noticeably to the gates of the present Thameside Mansion before levelling out into what would have been part of a middle bailey courtyard inside the (middle) Bydongate; the gateway itself most probably lies under Castle Farm House where no archaeological

investigation has yet been recorded. The archaeological indications from the several small excavations in the south-east corner of the castle (beneath the old farmyard buildings, the modern Thameside Mansion and in the grounds of Castle Lane House) are largely inconclusive, but the noted positions of moat edges, ditches and slips of masonry contain nothing to broadly contradict this conclusion; they rather tend to reinforce it. The documentary testimonies quoted above, have also led to the conclusion that the outer castle gate (the ‘Constable’s Gate’ Keats-Rohan, 79–81), lying just inside the extra-parochial bounds of the castle, must have stood in the area that is now Munt’s Mill garden. As noted above in the discussion of the boundary perambulation of 1707, the line of the town boundary passes just to the south of Munt’s Mill, close to the site of Renda’s outer gateway in 1707. A small resistivity survey was made in the garden of Munt’s Mill some years ago, hoping to locate a mill site; it revealed considerable areas of high resistance which now seem far more likely to be associated with the outer castle gateway (Figure 7.29); it should be recorded that this survey did not come to light until after the position of the gate had been proposed – so it was most welcome supportive evidence. It is also worth noting that during a conversation with the current owner of Munt’s Mill, it was discovered that builders working on a small extension to the house on its northern side had been unable to find 166

any solid footings at all, a fact which perhaps supports the supposition that the house itself sits over part of the outer moat of the castle. Interestingly, this location aligns westwards with a distinct drop in level at the southern end of the garden of Castle Lane House (also just inside the castle boundary) which might also be a vestige of the moat. The northern edge of Munt’s Mill garden also drops away, at a point where the middle moat might be expected to cut west to east to meet the riverside water system. These are tenuous suggestions which need proper archaeological investigation, but in the wider picture they do begin to make sense. Indeed, one of the most remarkable aspects of this whole project has been the unusual, unexpected and almost uncanny way that evidence has linked seamlessly into place, not a common experience in historical research. Any opportunities for further archaeological investigation in the area need to be carefully monitored.

Figure 7.30 The widened west end of the bridge looking towards the High Street, 2015 (photo: Judy Dewey).

The river, riverside and water controls With many strands of evidence pointing to the main castle gates being in the south-eastern corner, near the great bridge over the Thames, consideration must be given to other fundamental questions. How would the moat outfalls have been controlled? How would the gateways relate defensively to the bridge, the entrance to the town and the river? Where were the king’s mills ‘under the castle’ and their related waterways? What part did the river itself have to contribute? Again, the documentary and landscape evidence can assist with possible answers. The Great Bridge It has already been noted that the among the holdings of the porter of the castle were the properties on either side of the High Street closest to the end of the bridge; these gave him the ability, if required, to wield direct control of the area immediately inside the town gate, through which anyone approaching the castle gates would have to pass. The Gesta Stephani described the bridge at Wallingford in 1139 as ‘the master key not only to the town but to the castle on that side’ (1.5), which makes good sense if the outer castle gate lay just inside the town, close to the end of the bridge. It certainly played a key role in the 12th-century civil war; it was not necessarily a stone structure at this time, but contemporary chroniclers record King Stephen building a wooden tower ‘at the entrance to the bridge’ in an attempt to control it. The tower was subsequently taken by the young Duke Henry of Normandy resulting in the capture of 20 soldiers and the beheading of 60 of the king’s archers (Christie and Creighton 2013, 204). A recent study of the structure of the present bridge suggests that its stone origins may be as early as the 12th century; more certainly there are features of the 13th-14th century surviving (Christie and Creighton 2013, 224–9). There are no structural remains of the medieval gateway, but some details are revealed in the 1508 accounts of work done ‘upon the cawsey under the Mary of Grace’ (3.2) and in the town surveys of 1548 and 1606 (Bodleian, MS Top Berks b. 41; TNA, E315/369/3, fols 101-23): the chapel known

Figure 7.31 A widened arch of Wallingford Bridge, viewed from the north side (photo: Judy Dewey).

as ‘the Mary of Grace at the foot of the bridge’ (1548), a former possession of the College of St Nicholas, was ‘scituate upon the Eastgate’ (1606) and closely associated with the porter’s properties ‘next to the town gate there called The East Gate, both totally ruined’ (1548). As was seen above, despite their ruinous state in 1548, these porter’s properties, continue to appear in many later castle leases. The descriptions suggest that the Mary of Grace was a gate chapel, built over the town’s East Gate, with the road from the bridge passing beneath it into an area dominated by properties in the charge of the castle’s gatekeeper. There are surviving examples of such gate chapels, most notably at Warwick, where both the east and west gates of the town lie beneath churches; it is possible that St Peter-in-the-West in Wallingford was a similar gate chapel. The gate chapel was apparently damaged or destroyed during the 17th century Civil War, for in January 1662/3 Moses Slade was given the lease of the bridge tolls on condition that he built a house on the site of the Mary Grace within three years (BRO, W/TLd 1). The arrangement of the renewed town gateway can best be seen in a painting from the south side of Wallingford Bridge by William Havell in 1810, depicting the work of reconstruction

167

after the great flood of 1809 (see Figure 7.51). It has been noted that the 1804 plan of the castle does not mark this town gateway (Christie and Creighton 2013, 224), but perhaps the cartographer was interested only in castlerelated details; the accuracy of Havell’s detail is certainly reinforced by other contemporary illustrations of the bridge arrangement (see below). The gateway was finally demolished, as part of the rebuilding arrangement, to create the wider approach to the western end of the bridge which exists today. (Figure 7. 30 and Figure 7.31) Responsibility for maintenance of the bridge lay with the borough, probably from its origins, until Oxfordshire County Council somewhat reluctantly took charge in 1933. Upkeep of the bridge was an onerous but vital task, to ensure not only the security of the eastern gateway, but also to maintain a constant flow of trade in and out of the town. The earliest copy of the borough seal is attached to the grant of a property in 1299 for the maintenance of Wallingford Bridge (BRO, W/TCD1). There are several surviving examples of such grants and gifts in the 14th and 15th centuries including mention of two ‘caretakers and wardens of the bridge’ in a bequest of 1377 (3.1). By the 16th century a number of properties in the town constituted a ‘Bridge Estate’ administered by specially appointed bridge wardens whose number was increased to six to deal with much needed renovations in 1513 (3.3). (The modern survival of the Bridge Estate is the Bridge Charity administered by trustees and separate from the amalgamated town charities). Further funds were raised by tolls charged for using the bridge to cross the river and, from 1571, for the passage of boats under it. The latter privilege was confirmed by a charter granted by Elizabeth I, a year after she had visited the castle (BRO,

W/1C3). During the 17th-century Civil War, the insertion of drawbridges into four of the arches of the medieval bridge (arches 2, 6, 11 and 16) must have weakened it severely. These were reinstated with brick in the mid-18th century and the structure lasted until the great flood of 1809 terminally damaged the arches over the main river (Christie and Creighton 2013, 224–9). The rebuilding that followed was to transform the western approaches to the bridge, masking anything remaining of the medieval landscape in the vicinity of the castle gates, so it is essential to assess what is known of the earlier arrangement of the riverside in this critical area. At the beginning of the 18th century tolls were being collected by the lessee of a newly built property called Bridgehouse, situated on the north side of the bridge at its western end, adjacent to the main gate of the town; he paid the ‘ancient yearly rent of £8’ for the privilege. This building, and the associated town gateway, can be seen on two engravings of the north side of the bridge, one after Thomas Hearne (see Figure 7.50) and one by Tagg after J. M. W. Turner (dated 1795, Tate Gallery, T05913). The Hearne engraving shows foliage and land around the foot of the building, which is probably the remains of Chapel Eyot, also clearly depicted on the 1804 map (Figure 7.48); in 1701 Thomas Bishop, (d.1698) was described as having had a way ‘from the Bridge House to his eyot’ (BRO, W/ RTc 1, 22r). The Tagg engraving shows no sign of the island, but this appears to be artistic licence, since Turner’s original sketches in 1792/3 (Tate Gallery, D00182 Turner Bequest XVII G) on which the engraving is based, showed foliage and a light outline of the island, as did a later Turner sketch in 1805 (Tate Gallery, D05943 Turner Bequest XCV 39). A stable was added to Bridgehouse c. 1768 and

Figure 7.32 Detail from the c. 1835 Wallingford map showing the new gas works and water arrangements (see 7.56).

168

Figure 7.33 Riverside footpath round the old Oxford Canal Company wharf building (now a house), 2015 (photo: Judy Dewey).

at about the same time the property became an inn, as well as a place for toll collection; soon after this a small toll-hut on the bridge itself was brought into use. The inn was first known as The Wallingford Arms, but it rapidly became the Town Arms, a name which survived to be associated with the property on the corner of High Street and Castle Lane after the original inn was demolished in 1811 as part of the bridge-work following the 1809 flood. A painting by G. Shepherd dated 1815 shows the arrangement of the landscape soon after the completion of the bridge-work (see Figure 7.52). The wooden panelling on the far right of the picture is the camp-shedding which by then protected the wharf and its towpath, once the site of the old Winch Eyot mentioned in the perambulation of the bounds in 1707 (above 148) and shown in Hearne’s

engraving (see Figure 7.50; on wharves, see also Christie and Creighton 2013, 376–8). Interestingly, the Thames Commissioners’ Towing Path Rental books for 1797–1813 refer to this section of the towpath as Winch Mead (BRO D/TC 21, 274). By c. 1835 the area between the bridge and the wharf had been largely reclaimed to create a platform for the new town gas works, but water channels remained on either side (Figure 7.32). Today the whole area is infilled and occupied by the riverside restaurant, yet the public path around the wharf remains in use (Figure 7.33). Streams, moats and water control A key feature in this area was the outflow of water from the castle moats and mills, a practical aspect of control which

Figure 7.34 The inner and middle moats appear to converge against a dam that forms part of the eastern rampart (photo: Judy Dewey).

169

needs to be better understood in relation to the castle’s defences and one which was still a matter of concern as late as the 19th century. The castle moats lay above the level of the river, supplied by strong streams brought in eastwards down the Millbrook from the Moreton area and controlled by ‘floodgates’ south of the town’s West Gate (Keats-Rohan, 72 and Christie and Creighton, 2013, 96–7). The eastern castle rampart appears to form a high-level dam against which the northerly moats converged (Figure 7.34). Interestingly, however, the 1548 castle plan indicates water surrounding the whole inner bailey, including the eastern side (Figure 6.1) which implies that such an eastern moat must either have been supplied by water sluiced down from the castle moat system or from water flowing in from the river at the lower level. It is suggested here that both these sources may have been used to create an eastern water defence, and indeed a mill stream, the final outflow of which was controlled near the bridge. It is clear from the 18th- and early 19th-century plans of the castle that there was a stream flowing east-west from the river marking the boundary between King’s Mead and Queen’s Arbour, and turning south to flow along the western edge of Queen’s Arbour (see Figures 7.46, 7.47 and 7.48). The east-west section, referred to in 1355 as Walditch, marked the northern boundary of the medieval ‘water of Wallingford’ (REBP 4, 186 May 24) and was still the municipal boundary in 1898 (see Figure 7.8); the southerly section, known as Bishop’s Ditch in 1806 (3.13, as discussed above 148), marked the eastern boundary of the extra-parochial castle. Furthermore, in 1786 and 1804, there are clearly wide expanses of water at the base of the castle’s eastern rampart, separate from the stream. Interestingly, there is still a pond in this vicinity today (Figure 7.35) which in October 1858 J. K. Hedges recorded to be 1ft 8in (0.5m) above river level (BRO, D/ EH E26, 24). The levels and flow of water would normally have been well controlled by sluices, dependent on many factors: the height and state of water in the moats (including the need for drainage of effluent), the height and flow of the river and the need for power to drive the castle mills (see below); even so, flooding of the meadows

must have been a common hazard (see Figure 7.21). The moats would have needed sluices and overflow points: one is mentioned in the 1296–7 Ministers’ Accounts as being near the Bydongate (Midgley 1942, 135), a name which may have foul water connotations (Keats-Rohan, 73); another was discovered c. 1828 by James Blackstone (above 162), near the base of the motte on the southern side, from which ‘the water was gradually drawn off from the Moat into the river; a communication with that stream having been effected by the original engineer’ (Allnatt 1873, 48). Porter’s Ditch was another well documented medieval feature. Beginning at the north-east extremity of the castle, at the point where the Walditch stream turned south, it can be traced northwards, with the porter’s two meadows lying on either side, delineating the Clapcot boundary with the castle-held King’s Mead and linking with the most northern holdings of the castle. These features are reflected on the 1817 plan drawn up when the castle was sold by the Crown in 1817 (see Figure 7.55); the lots included riverside lands and rights which had been reflected in leases over the centuries. The most northern area, Lot 10, included a weir and mill, with land on either side of the Benson mill race and a narrow string of riverside land leading north as far as the present Benson pound lock. It also included a meadow known as North Eyte (Figure 7.36), the island

Figure 7.35 This pond near the north west corner of Queen’s Arbour, was landscaped by the Victorians but may have related to the medieval water control system; it lies below the dam formed by the eastern castle rampart, but above the level of Queen’s Arbour (photo: Judy Dewey).

Figure 7.36 Detail of 1817 castle sale map (see Figure 7.55) showing North Eyte (marked as 5) and other castle holdings near Benson. Photo: Katharine Keats-Rohan, reproduced courtesy of The Bodleian Library, University of Oxford, MS Top. Berks. b. 10, fol. 18r.

170

name deriving from at least the 14th century when it was described as ‘an island of meadow called northacce’ (1.57). By 1817 only the north-easterly edges were still bounded by water but the alignment of the field boundary with the western edge of North Eyte meadow, the former island, strongly suggests that Porter’s Ditch was once another water channel flowing from the river at North Eyte to link with the Walditch and to supply additional water for an eastern moat for the castle and possibly, as will be seen below, for the castle mill (Thacker 1968, II, 189. See also Christie and Creighton 2013, 248–51 for discussion of features on Queen’s Arbour). Any controls of a confluence of water in the vicinity of the Walditch would have been well placed for protection by the north-eastern castle defences. There are medieval references to Dernegate, a probable north-eastern postern (Keats-Rohan, 172), which may have led out to the riverside near this point and to a ‘porter’s bridge’ in this vicinity, both of which suggest that it was potentially an area with necessary access from the castle (see Figure 7.1). It is difficult to define the original landscape after so much infilling by demolition and Victorian landscaping, but this area is likely to have been highly secured, a suitable place for sluices (mentioned as being on the castle’s eastern side in both medieval and later documents) to control the water defences. Today the original water system lying between the base of the castle’s eastern rampart and the western edge of Queen’s Arbour has been reduced to a pond and a drainage ditch, while the Walditch was converted by the Hedges family in the 19th century to form a waterway to a new boathouse (see Figure 7.8); the boathouse is now demolished and the ditch drained. There was considerable concern in the 19th century about the ancient southern outfall of water from the

castle’s eastern ditch. The 1835 town map and 1881 OS map show the stream running between the western edge of Queen’s Arbour and the castle until it becomes the eastern boundary of the property that is now called Munt’s Mill, already demonstrated to be the probable site of the castle’s outer gateway, and owned from 1858 by J. K. Hedges. At its most southerly point the stream was channelled under a small bridge carrying the well-established path that gave public access from Castle Lane, across the stream’s outflow to the Town Wharf and Queen’s Arbour; this bridge once displayed a marker for the town boundary (BRO, D/ EH E15 14 June 1894). The stream flowed out into the river via the small inlet on the north side of the gasworks (see Figure 7.32). In a series of letters concerning many water problems associated with the Munt’s Mill area, J. K. Hedges wrote about ‘his ditch’, stating that ‘the waterway conveys into the river the water from the castle moats’ (BRO, D/EH E12, 1 March 1884). Changes to the stream’s outfall are traceable in these letters: the ditch had been cleaned out several times by Hedges’ predecessors and he himself had dammed it up soon after he purchased his castle property in 1858. He adds that ‘it is clear that this bank was a considerable size along the ditch on this side, and was formed originally when the ditch was dug’ (BRO, D/EH E26, 29, 17 Jul 1875). Another known change in this southern drainage area was the cutting of a new relief channel eastwards through to the river at the southern end of Queen’s Arbour, sometime between 1804 and 1835 (there is no sign of it on earlier maps). This channel appears on the St Peter’s tithe map (1851) but ceased to be a wet ditch by the last quarter of the 19th century. The perceived need for it was probably reduced by other changes; these included the reclamation of more land, filling in the original outfall of the stream

Figure 7.37 Compilation of resistivity results on south and east sides of the castle (image: Gerard Latham)

171

near the gasworks to form an enlarged dry area, where a boathouse and yard (now the riverside restaurant, illustrated in Figure 7.57) eventually replaced the gasworks, which were moved in 1875 to a new site west of the town. The water from the castle’s eastern ditch would have been considerably lessened by this time because of the changes made, but any outflow surviving was channelled to disperse into the river. Hedges complained of the inadequacy of the depth and size of drainage at the time, and there is a wealth of detail concerning these water problems in the Hedges papers. The low-lying somewhat marshy meadow area to the east of Munt’s Mill had not been acquired by Hedges in 1858, but he offered to purchase it in the late 19th century in hopes of being able to drain it properly and improve the riverside. The lengthy correspondence on this (BRO, D/EH E15 Wharf House) reveals that the old Winch Eyot was part of this re-claimed land on which the Oxford Canal Company Wharf (later Town Wharf) stood by the 19th century. Mrs Sheppard, who had inherited the land, tried unsuccessfully to drive a hard bargain with Hedges over its sale. It was obviously a very soggy area, much prone to flooding, with ancient boundaries and rights which were unclear even to Hedges, so the exact relationship of the former Winch Eyot and the outfall of the moats sadly remains obscure. A recent geophysics survey to the east of Munt’s Mill revealed some areas of very low resistance (Figure 7.37), suggesting, as might be expected, that this was a wet area, but the infilling of the island and the building of the later wharf have greatly disturbed the ground; invasive archaeology is needed to gain a clearer understanding of the area. Mary Broster who lived in Munt’s Mill in the 1960s remembers a water diviner locating a channel beneath the garden on the alignment of the watercourse shown on the early maps; she also mentioned that a drainage pipe was found when the restaurant was extended and that Edward Ryall, farmer of the castle land at the time, had filled in ‘various ditches’ on Queen’s Arbour. The original ditch on the west of Queen’s Arbour and the meadow itself is now in the care of SODC and steps have been taken, under the stewardship of Earth Trust, to increase the water again on the meadows to rejuvenate wildlife. Such relatively modern changes are a salutary caution to an historian seeking to reveal a medieval landscape, but it has become increasingly clear in these studies that water is fundamental in the defensive plan of the medieval castle and that a key point in the control of the whole system is its outfall to the river on the eastern side in close proximity to the main gateways near the bridge. Fishing The castle’s control of the river, which it dominated from the bridge to Shillingford, was essential not only for defence but as a means of transport and communication, and as a source of food, revenue and power. Two important aspects of this remain to be considered in more detail: fisheries and mills, both of which are related to the vicinity of the castle gates. Fishing rights were closely guarded since fish were an important part of the medieval diet and the river was a rich

source. The medieval accounts show that the castle moats were fished (Keats-Rohan, 73) and, as has been seen above, the 17th-century leases indicate that the inner moat still held fish. Outside the walls there appear to have been two separate fisheries associated with the river. The first, noted earlier (3.11), pertained to river fishing rights alongside King’s Mead, including its adjacent islands, and the lease of the meadows of North Eyte. An inquisition taken after the death of Edmund, earl of Cornwall, in September 1300, described these fishing rights as running ‘from the king’s mill under the castle and the lock [weir] of the Bishop of Winchester to the stream which comes from Sildenebrugge [somewhere between Benson and Shillingford] and falls into the Thames’ (CIPM 3, 465). A second fishery appears as a ‘fishery of eels’ at the ‘floodgates’ (1.57); in the postdemolition period it appears to have survived as a lease of ‘certain water in the River Thames called Wallingford Locke’ (BRO, W/Ac/1/1/2, 1662, 22r), later more precisely defined as ‘near the Old Lock called Old Locke water in the Parish of St Peter and fishing and eyots adjoining on the east side’ (BRO, W/Ac/1/1/2, 1779, 41v). The clear division of these two areas of fishing was a weir in the river, its location so well known in the past that it needed no further explanation. Traces of this weir, and the mills with which it was associated, have long since disappeared, leaving their exact position elusive. However, various avenues of recent research are worth summarizing. Weirs and mills The origin of the bishop of Winchester’s association with a weir near the castle was probably ancient: in 1086 the bishop held 27 properties in Wallingford, some of which abutted the castle on the north-east side of the High Street (Roffe 2009, 31); he also held the manor of Brightwell which, as has been discussed, as late as 1634 had the anomaly of the tithe of a small island near Wallingford Bridge which continued into 19th century by which time the island (Winch Eyot) had become part of the town wharf. It is possible that this island had some link to the Bishop’s Weir, or more likely to control of the water for mills driven by the weir-stream, since a weir that close to the bridge would have been extremely hazardous. Nowhere in the copious medieval documentation, however, is anything but royal control associated with the mills, which might even suggest that the bishop’s involvement pre-dated the building of the Norman castle. The bishop’s association with this area is long-lasting: a presentment in the reign of Richard II describes the bishop of Winchester’s ‘lock at Wallingford Bridge’ as being ‘raised so high that the overflow thereof floods the meadows on either side both in winter and summer’ (Flower 1923, 125). Such flooding would be caused by holding water back to create a greater flow to the mill stream or for building up a flash of water to allow boats to pass the weir. The ditch on the west of Queen’s Arbour was known as Bishop’s Ditch in 1806 and on the Crowmarsh side ‘meadows and eyots’ were still held under the bishop of Winchester (3.13). So the question remains: where was the ‘Bishop’s Weir’ and where were the mills ‘under the castle’? There 172

are references to remains of the ‘hazardous piles’ of a weir being removed with difficulty from the river in 1791 and their remnants still being visible in the early 20th century (Thacker 1968, ii, 189–190). Thacker placed the weir somewhat south of the northern end of Queen’s Arbour (the Walditch); interestingly, when the river was dredged in the early 1990s, substantial pieces of stone were observed to have been removed somewhere in this area (Stephen Capel-Davies, pers. comm.). Thacker also recorded observing a dry channel in 1920 running diagonally across the meadow. Geophysics has revealed diagonal lines of apparent stonework beneath Queen’s Arbour which could indicate lined mill-streams (see Figure 7.37) and there are notable high resistance areas at the southern end of Queen’s Arbour which could be mill foundations. These interpretations work well with Fradley’s survey here (Christie and Creighton 2013, 248, 250) which revealed a complex pattern of channels and possible ponds, both north and south of the substantial 13th-century apsidal building excavated in 2010 (Figure 7.38). A later channel, interpreted as a mill leat, was roughly cut through the remains of this 13th-century structure and the conclusion drawn was that the castle mills lay to the south of Queen’s Arbour (Christie and Creighton 2013, 250–6). The use of the 13th-century building, however, remains an important unresolved question. One suggestion is that

the structure was part of a ‘semi-ornamental walkway’ linking to an eastern entrance to the castle (Christie and Creighton 2013, 56; a possible quay is also proposed), but Keats-Rohan’s researches (above) provide no evidence to support this idea of a grand riverside entry. Good evidence has emerged for the location of the castle gateways and many of the key buildings, but the only candidate for the medieval structure on Queen’s Arbour is the king’s mill ‘under the castle’. It therefore seems necessary to reconsider what is known of the mills and water controls on Queen’s Arbour in the light of the new research. The medieval accounts reveal continuing costs for the upkeep of the castle mills – both under the castle and at the South Gate with the latter location, near St John’s Green, well documented from the 12th to 20th centuries. The mill under the castle was a corn mill, essential for supplying the needs of the castle household. There is documentary evidence of considerable use of timber and stone in repairs to this mill in the early 14th century, with further large quantities of similar materials also being used to make ‘great’ and ‘upper’ enclosures associated with the mill or, perhaps, with the castle swannery. The enclosures were still being maintained as late as 1483 (Keats-Rohan, 76). Interestingly, the building excavated in 2010 had stone foundations ‘broad and substantial enough to envisage a relatively high wall either of stone, or a low stone wall

Figure 7.38 Excavation of the apsidal building on Queen’s Arbour, nearing completion in 2010 (photo: Judy Dewey).

173

Figure 7.39 Substantial stones lining the ditch along the western edge of Queen’s Arbour (photo: Nigel Ruchpaul).

supporting a timber structure’ (Christie and Creighton 2013, 251). Resistivity revealed that the U-shaped complex had solid western foundations lying parallel to the northsouth stream along the western edge of Queen’s Arbour which, as has been seen above, carried a controlled flow of water from the river; the surviving ditch has been observed recently to have some sort of stone lining, much obscured by vegetation, which merits proper archaeological investigation (Figure 7.39). The most logical interpretation is that the U-shaped building was actually the medieval mill, probably operating an undershot wheel located at its western end and using the controlled water flow from the Thames which had its final outfall to the river near the Great Bridge. This stream may be referred to in an entry in the castle accounts of 1307–09 (1.31; Keats-Rohan, 99) which describes improvements to a damaged water flow in the stretch from the East Gate to the castle mill; the same accounts also mention repair to an enclosure ‘behind the castle mill’. The Bishop’s Weir, if situated where Thacker suggests (above), would have spanned the Thames, enabling control of the water-flow to the Walditch stream and Bishop’s Ditch, creating a suitable millstream to drive the wheel. This would also account for the accusations of flooding on both sides of the river being caused on occasion by the Bishop’s Weir being ‘raised so high’, as mentioned above. The ‘great and upper enclosures’ would perhaps have been an improved water system for the mill, possibly also related to the known swannery. When the outer stone face of the potential mill building was excavated it was noted to have been comprised of dressed blocks (Christie and Creighton 2013, 254), presenting a good appearance in what must have been an attractive and impressive watery landscape. If this was indeed the medieval mill, how did it relate to later known mills? In the 16th century the first references appear to a fulling mill in Queen’s Arbour. In 1553 a record in the Borough Minute Book of the renewal for 21 years of ‘a fishery of the bridge water, with the eyots belonging to

the same’ reserves ‘a sufficient way over part of the same water and eyot into the fowlling [fulling] mill there’; the lease also made a proviso for the mayor and burgesses to be allowed to use their rods for ‘fishing off the shelf next unto the Mary Grace within the same water’ (BRO, WAc1/1/1). In 1574–77 this Tudor fulling mill became the subject of a legal dispute between Nicholas Payne (farmer of the South Gate mills) and Ralph Pollington (TNA, E112/2 no. 9; BRO, D/Ex6 T1; cf Christie and Creighton 2013, 381). Pollington, having leased meadow land in Queen’s Arbour, was accused of erecting two fulling mills there for his cloth trade and then, adjoining them, two corn mills ‘under one roof’. The latter were said to have encroached the ancient rights that all the corn of the town should be ground at the South Gate mills, then in Payne’s possession and declared by him to be the only mills ever to have served the town. Pollington claimed, with some justification, that his corn mills were built ‘where there have been corn mills from ancient times’. Payne successfully disputed this; he also accused Pollington of having extended the weir, altering the flow of the river to force water down the millstream. It is a fascinating and acrimonious row between two prominent Elizabethan burgesses but is of particular interest here in showing that the medieval weir was being usefully adapted. It seems possible that this is the time when new millstreams were dug, channelling the water across Queen’s Arbour, cutting through the old mill building to a new mill site further south on Queen’s Arbour; it is this reworked landscape which has survived to be reflected in recent surveys (Christie and Creighton 2013, 248, 250). A lease in 1553 also refers to the long-standing right of way into Queen’s Arbour to give access to the mills (Hedges 1881 ii, 255), another medieval survival which, as has been noted, was still perpetuated when the water near the bridge was eventually in-filled and built on; indeed, it remains a public right of way today. Ralph Pollington’s son appealed against the injunction imposed on his father’s mills; the final outcome is unknown, but the Pollington family kept 174

Figure 7.40 Modern survival of the track from Castle Lane to the river. The car is parked in front of Munt’s Mill, close to where the outer gateway of the castle would once have stood (photo: Judy Dewey).

the lease into the 17th century; the mills eventually fade from the records. The Borough Minutes (BRO, W/Ac 1/1/1) reveal that several attempts were made in the 18th century to encourage leaseholders to reconstruct the mills, but they came to nothing and, as noted, it was the South Gate mill that remained in business until the 20th century. Appraisal of the riverside arrangements Surveys, archaeology, geophysics and documents combine therefore to reveal a very complex history of water controls in the area from Wallingford Bridge to the northern boundary of Queen’s Arbour, a landscape which would have been critical to the defence and visual impact of the castle in its heyday but which has evolved in later centuries. Further archaeological investigation and research are needed to clarify the arrangements but the following summary reflects present thinking. The evidence seems to point to the castle’s ‘mill under the castle’ being on the western edge of Queen’s Arbour, most likely an under-shot wheel supplied by a mill-stream derived from the Bishop’s Weir on the river (which also provided an eel fishery). The east-west section of this stream was the Walditch, which converged with a further controlled flow of water brought in from North Eyte (close to the north mill at Clapcot). The mill was associated with enclosures of water at the northern end of the meadow which may also have been part of a swannery. Precisely how the water system was controlled is a matter for further research, but the extensive water source also allowed the creation of an outer eastern moat for the castle which in turn carried away sluiced water from the inner moats, in more than one place. (The most southerly sluice, near the

Bydongate may have been a foul drain from which the gate derived its name: Keats-Rohan 73.) This water-based engineering would have provided a spectacular view to anyone approaching the castle across Wallingford Bridge, with the outer wall of the castle and the Great Tower behind it rising from the watery landscape. The mill tail and other channels seem to have converged to an outfall on the river near the site of the main castle gates close to the property now called Munt’s Mill. Associated with this area was the small island, referred to in the 18th century as ‘the old Wynch Eyot’, today part of the wharf. This was also the area in which the outer and middle moats of the castle’s southern defences met the riverside. Some sort of enclosure no doubt existed here to allow the massive gateways to be protected by this significant body of water (Figure 5.35). The presence of the well-documented drawbridges of the outer Constable’s Gate and beyond it the Bydongate, testify to the watery landscape, while in the post-demolition period control of water was a recurring problem for several centuries. The exact arrangement can only be conjectured but there is an interesting parallel at the Tower of London with an enclosure of water at Edward I’s Middle Tower and Byward Tower (c.1270), also adjacent to the Thames (Goodall 2011, 201). Access to the mills on Queen’s Arbour, would have been round the enclosure, via the small island. Between Wallingford Bridge and the approaches to the massive castle gates lay Chapel Eyot and a shallow inlet of the Thames. Little is known of the usage of this waterfront but it was a potential area for the offloading of goods close to the castle gates; for example, the trader Thomas West landed goods at Wallingford Bridge in 1572 (Prior 1981, 87), though this could conceivably have been south of the bridge. The whole area was closely 175

protected by the bridge itself, the town’s East Gate and the porter’s properties adjacent to it. Today the entrance to Castle Lane and the site of the medieval castle gateways leads from the High Street, close to the eastern end of Wallingford Bridge. The lane to the riverside path forms a narrow track off Castle Lane, passing the front of Munt’s Mill and running between the wall surrounding the wharf property and the northern side of the large riverside restaurant to reach the riverside (Figure 7.40). This track is following the route described in the 1707 beating the bounds, which at that date carried it via a small bridge over the stream adjacent to Munt’s Mill (now piped underground) on to the edge of the old Wynch Eyot. All the land on which the restaurant now stands has been reclaimed since 1809 but in 1707 it was still a water-filled inlet of the river, an arrangement dating back to medieval times. Summary and conclusions This study of the post-medieval evidence for the castle site has identified several features that can be clearly assigned as ‘new’ rather than part of the original medieval landscape. Once these features have been summarized and eliminated from the picture, some attempt can be made to assess what, if anything, is left of the layout of the castle in its heyday around 1300, taking into consideration the wealth of new evidence from the medieval documents and the archaeological findings (Figure 7.1) Several buildings have been imposed on the castle site since the 17th-century destruction, the latest being Thameside Mansion on the riverside. This overlies early 19th-century farm buildings, one of which survives on

the southern side of the newly laid out courtyard. There are also two other neighbouring 20th-century houses (see Figure 7.20). The lane that leads to the Mansion dates from the early 18th century when it was dug out to form the grandiose entrance to Thomas Renda’s new house in the inner bailey, partly encroaching on land that belonged to Christ Church College, Oxford, destroying the older lane leading into the castle and obscuring boundaries with the planting of trees. This area had once been the main castle gateway. The pillared entry to the castle grounds off the High Street was first created as an early 19th-century farm entrance, the construction of which must have required digging through the remains of the castle’s outer rampart, probably using the earth removed to fill in the remains of the outer moat. The farm track was enhanced after 1858 to form a grand carriageway leading to the inner bailey, then circumnavigating it to approach the Victorian mansion house in the middle bailey, its route partly marked out by iron railings, a few of which still survive; the public way through the castle at that time was diverted to form the present footpath extension of Castle Lane to Castle Street. Not long before the Victorian mansion was built for John Kirby Hedges and his bride, the ruinous ‘Prison House’, or ‘Gayle barn’, in the middle bailey, was demolished, as was the old Clerks’ Lodging. The mansion demanded landscaping which ‘tidied’ the middle castle rampart and embellished early stonework with added features, including some heightening of the surviving wall in the middle bailey, and repairs to the old wall of the former Priests’ Lodging. The gardens were planted with exotic trees and enhanced with a small ‘grotto’, still featured in Castle Gardens (Figure 7.41). The Victorian Lodge and entranceway to the mansion off Castle Street destroyed the

Figure 7.41 Castle Gardens, with resistivity work in progress (photo: Judy Dewey),

176

earlier buildings in the vicinity, but the former churchyard of All Hallows survived; part of it, however, became a carpark c. 1970. In the southern outer bailey, however, there has been no major development since the demolition of the castle in the 17th century; other than minor changes of access via Bear Lane, the enclosure of St Anthony’s Close and the construction of a private tennis court, the land appears to have been used continuously as garden ground. The well-defined extra-parochial boundaries of the medieval castle still existed in the 19th century, as the sale documents of 1817 confirm when they quote the lack of tithe obligations as a selling point (Bodleian, MS. Top. Berks. b.10, 14–20). This status was invoked as late as 1845 when a destitute occupier of one of the castle cottages was legally denied assistance by the guardians of the workhouse because his home was extra-parochial (Dewey and Beasley 1989, 91). Such problems were finally resolved by the Local Government Act of 1894, after which the castle precincts were united (or more probably re-united) with the old parish of All Hallows with Clapcot, the new parish being known simply as All Hallows. This ended almost a thousand years of extraparochial status for this area. The late Saxon antecedents of the site have been well-discussed (Keats-Rohan and Roffe 2009, 25, 36, 52ff), concluding that the Norman castle was successor to a pre-Conquest royal holding implanted into an existing Saxon defensive and ecclesiastical landscape, the boundaries of which survived relatively unchanged, apart from a slight western extension, until the late 19th century. This precinct was surrounded on the west, north and east by the parish of All Hallows (formerly known as All Saints) and on the south by St Peter’s parish. All Hallows, incorporating the extensive lands of Clapcot, stretched north to Shillingford and included the meadowlands of Queen’s Arbour and King’s Mead on the east, areas ‘outside’ the castle but dominated entirely by it and entrusted only to the care of loyal servants; they formed an integral part of the castle’s defences and environs. It remains to highlight features of the post-medieval period that can reasonably be considered to relate to the medieval landscape. When the Norman castle was first constructed its position seems to have been influenced by a Saxon stronghold in the north-east quarter of the town, features of which were incorporated in the castle, most notably the northern burh rampart and ditch, the water supply to the moats and possibly some form of central tower (Keats-Rohan, 106). The water in the Saxon ditch must have had a controlled outfall to the riverside on the east but its arrangements are uncertain. One possibility is that the medieval Walditch dividing Queen’s Arbour and King’s Mead may have earlier origins; it marked the boundary of ‘the water of Wallingford’ in 1355 and was the point at which the town’s well-attested boundary crossed the river to take in the eastern bridgehead. Other features of the late Saxon site (an eastern rampart perhaps?) could have been incorporated by the Normans but solid evidence is so far lacking. It is not until the 13th century that documentary accounts provide more useful details of defensive arrangements.

The town of Wallingford was not apparently perceived as a great threat by William the Conqueror after 1066 (Keats-Rohan, 37–8), but the safe construction of the castle’s inner bailey stronghold would have necessitated some sort of outer boundary defence; it is suggested that this was approximately on the line of the castle precincts discussed above. The form of this boundary in the 11th– 12th centuries is unknown, but at the least it would have been a palisade with a secure entranceway, possibly even a stone gateway as at Exeter (Goodall 2011, 60). This was a predecessor to the later more substantial outer gate, which the evidence places in the vicinity of the garden of the present Munt’s Mill at the southern end of Castle Lane. The dominance of the castle is clear at the eastern entrance to the town: a town gateway on the eastern end of the Great Bridge over the Thames, the castle porter’s holdings on both sides of the High Street at the end of the Great Bridge and the outer defensive boundary and gateway to the castle close by. The chronicler’s comment c.1139 (1.5) that the bridge was the key to the town and castle reinforces the suggestion that the town’s eastern entrance was, from the beginning, an area of coordinated defence. The well-fortified Norman inner bailey, protected by the large outer bailey, probably provided sufficient security until the 12th century, when the need for greater defence was achieved by the sub-division of the outer bailey with a second wall, ditch and gateway – the Bydongate (KeatsRohan, 39). Although previously this insertion has been assigned to the early years of the Stephen and Matilda conflict, Keats-Rohan suggests that it was more likely to have been a prestigious upgrading by Brien fitzCount (Keats-Rohan, 41–2). None of the various archaeological investigations has produced a definitive date that would make this conclusion untenable and the chroniclers are quite certain of the impregnable strength of the castle in the early stages of the war (1.4). The Pipe Roll accounts of the following reigns of Henry II and his sons indicate upkeep rather than construction of new defences at Wallingford (Keats-Rohan, 42), so it seems likely that the middle wall, complete with water-filled moats and Bydongate was in place, or at least largely completed, by the onset of the conflict between Stephen and Matilda in 1138. The barbican in the north-west middle bailey may also have been part of this work. By the late 13th century the impressiveness of the castle had been enhanced by the addition of a third wall and moat encircling it in the fashion of concentric castles on all but the riverside, its moat almost certainly linked into the existing water supply system. The height of this wall, a mere 8ft (2.4m), compared with the 14ft (4.3m) high and 8ft (2.4m) thick wall on the eastern side of the inner bailey (Keats-Rohan, 84), reinforces the notion that it was more a showpiece than a serious defensive addition, although it would still have been a substantial deterrent to attackers, enabling simultaneous attack from both the middle and outer walls. It was most probably the work of Richard, earl of Cornwall, and seems to be the final phase of major new castle fortification at Wallingford. It lay within the original precinct on the south, seemingly sub-dividing the existing bailey but retaining the outermost boundary palisade. The 177

new rampart and moat ran north-south approximately parallel to Castle Street, passing through the present outer castle gardens and creating a very narrow walled outer bailey. The wall and moat then turned east and ran a few metres north of the town boundary until it culminated in the formidable outer gate and drawbridge under which the moat would have flowed. A deepening of the middle bailey ditch in this same period was demonstrated in Carr’s excavations in 1972, together with a raising of the ground level within that bailey; this was the area where land was made available in 1282 (1.28) for the new buildings of the College of St Nicholas, fragments of which can still be seen today, most notably a stone doorway and part of the wall of the Priests’ Lodging (see Figures 5.2 and 7.13). The earl of Cornwall’s landscaping involved a considerable use of water. The three wet moats surrounding the castle walls, well supplied from the town’s moat, survived as features long after the stonework had crumbled to a ruinous state, still providing a formidable defence in the 17th-century Civil War (see Dewey p. 236). Several sluices within the castle controlled the water levels with outfalls to the riverside, such as the one at the Bydongate (Keats-Rohan, p. 72) and others to the north. The medieval Queen’s Arbour with its adjacent stream, moat, mill, islands and enclosures was a major feature of the eastern approach to the castle. The post-medieval documents have revealed the likely position of the main castle gates on the south-east, the Bydongate and the outer Constable’s

Gate, which were adjacent to the riverside, gateways which featured state-of-the-art defences including large moveable drawbridges to span the two southern moats (Keats-Rohan, p. 79). There must also have been an outer enclosure to control the water and its eventual outflow to the river, but its precise layout (in the area which later became the town wharf) is unknown. The landscape in this crucial vicinity has been much changed over the centuries, but its importance is indisputable (Figure 7.42). In the early 1300s, a visitor to Wallingford Castle would have followed an awe-inspiring journey to reach his final destination. Having crossed Wallingford Bridge, with its panoramic view of the watery landscape of moat, mill, ponds and swannery in Queen’s Arbour, dominated by the high eastern wall of the castle and the great tower behind, he would have passed under the well-guarded East Gate near the western end of the bridge and into the bustle of the lower end of the High Street, dominated by properties of the castle porter. Turning immediately right, he would pass the busy waterfront to approach the drawbridge leading to the formidable Constable’s Gate, surrounded by water. Once through that gateway he entered an enclosed passage to the next drawbridge and the Bydongate, surmounted by the porter’s lodging. Safely through, he would come into a courtyard surrounded by walls and by the broad moat at the foot of the vast stone-clad motte and adjoining wall. (This area would be approximately where the present courtyard and garden of Thameside Mansion stands

Figure 7.42 Bird’s eye view of the castle in 2015, looking north from the spire of St Peter’s church. To the right of the picture Castle Lane runs north, (turning off the High Street) with Munt’s Mill, site of the outer gateway, glowing in the sun to the north of the riverside restaurant (once the site of Chapel Eyot). Castle Farm House, probable site of the Bydongate, lies behind the garden of Munt’s Mill (with Thameside Mansion in the distance beyond). The line of the middle wall of the castle stands out clearly in the centre of the scene. The castle boundaries to the north are marked by the distant tree line, whilst its extensive riverside lands Queen’s Arbour to the south, and King’s Mead beyond (lighter in colour) – curve away into the far distance towards Benson (image: David Dewey).

178

today.) The shining white-washed Great Tower, rising from the motte, would have dominated the skyline above. The route then followed to the heart of the castle would have been an imposing one, following a track through the middle bailey with the motte and Great Tower looming above to the right and the buildings of the College of St Nicholas to the left – the Dean’s Lodging, the Priests’ Lodging with adjoining orchard and the more modest Clerks’ Lodging – all sheltered by the high middle castle wall. Also in this bailey was La Studie (Keats-Rohan, above, p. 83) and probably other smaller buildings. To the west of the college complex the route followed the outer edge of the inner moat, with the inner wall of the castle on the far side of it linking to the motte and the Great Tower. The visitor would now have been approaching the Great Stable (where more lowly visitors might have left their horses), passing the Blindhouse prison as the road approached the end of the Great Drawbridge, protected by the north-western barbican (through which northern travellers would have entered the castle). Finally, he would have turned to cross the bridge over the inner moat and pass the long entry and the under-constable’s lodging, into the huge heart of the castle, the inner bailey. Encased in its curtain wall and dominated by the motte and the Great Tower, the huge sub-divided courtyard would nevertheless have appeared spacious, housing (according to KeatsRohan’s reconstruction, 91 and Figure 5.45) on the northwest (to his left) the Treasury, Great Hall and adjoining Great Chamber with kitchen and well close by. On the north was the imposing Great Chapel of St Nicholas, while ahead of him, on the east, there were enclosed gardens, the Red Hall, and apartments of the Queen, which overlooked the river. The Hall of the Knights and Esquires lay to his right, with stables and other services much in evidence throughout the bailey. Finally, the visitor would have been escorted across the courtyard to the south-east to ascend the steep covered stairway, past the well and rising up through the fore-building of the Great Tower, to be received into the private apartments of the lord, where he might be permitted to give thanks for his safe arrival in the lord’s own chapel. Such a description might be considered fanciful for a castle which has no complete upstanding buildings, but it is an informed attempt to capture the sense of awe that Wallingford Castle has evoked for this author as details of its buildings and layout have emerged through the research for this trio of papers. The combined work has brought together the most comprehensive assemblage of documentary evidence for Wallingford Castle ever gathered. The newly accessible material, in association with recent archaeological work, has allowed a fresh assessment to be made of some of the fundamental

questions about the layout of the lost castle – its boundaries, defences, buildings, entrances and water systems. What has become increasingly clear is that this was an exceptional royal castle. There are still many unanswered questions and there is much potential for further research: more geophysical survey work, particularly in the southern outer bailey area and the properties off Castle Lane could help pinpoint more precisely the lines of moats, outer walls and gates; more test pits in gardens on the back of Castle Street and the High Street might shed more light on the boundaries; and the mills, the landscape of the riverside and the water systems merit detailed examination. What is hoped is that a solid base has been laid that will encourage further research and much scholarly debate, at last raising Wallingford Castle from the footnotes into the main body of castle research where it rightfully belongs. Much of the post-medieval documentation has derived from legal issues. On reflection this seems highly appropriate, for the whole process of examining the material has been a little like preparing a difficult court case: all the available evidence has been considered and weighed in the balance but in some parts is still missing key elements which might deter a successful verdict. The case has nevertheless been pursued in the hope that, even if it fails to convince the jury at this stage, fresh evidence might soon emerge from further investigations to bring in a positive verdict on appeal. Acknowledgements Some acknowledgement has already been made to the work of the TWHAS Documents Group but I am particularly grateful to David and Berenice Pedgley, whose meticulous labours in Record Offices and Libraries have made more accessible a series of key Wallingford sources, invaluable in these researches; thanks also go to Daphne Baker and John Sims, whose initial work on the Christ Church archives and the Hedges papers gave many useful pointers. On the subject of water controls, I have much appreciated the professional river engineering expertise of Stephen Capel-Davies, whose suggestion that the building discovered on the riverside might be the mill, has led to a fresh appraisal of the castle’s water defences, weirs and mills. Similarly, Gerard Latham’s additional geophysics surveys, have provided crucial new evidence on the castle’s probable layout. Sincere thanks also go to my family, Becky and David Dewey, who have supplied the moral and technical support needed for me to complete this work after the loss of their father; in particular several aerial views and plans have been created by my son. Lastly, but most importantly, I have been much encouraged and assisted throughout by Dr Katharine Keats-Rohan, a great scholar and friend, who has come to share my life-long passion for understanding Wallingford Castle; working with her in the preparation of this trilogy of papers has been a most rewarding and enjoyable privilege.

179

7.42a. Resistivity surveys of Wallingford Castle super-imposed on LiDAR images. The orientation is north to the top of the page, with the River Thames cutting across the south-east corner of the image. The raised motte is evident towards the centre, with the angled line of the middle rampart to the south-west clearly showing as a continuation of the middle bailey wall which is revealed on the geophysics. An outer moat (with adjacent wall) can be seen in the westerly resistivity (Castle Gardens) and is traceable to the north on the LiDAR close to where the walls of the D-shaped ‘barbican’ turn southwards. On the southern side, ‘modern’ work clearly intrudes: the ‘sunken lane’ cut through from High Street to the drained southern moat of the motte by James Blackstone in the early 19th century (later used as a carriageway to the Victorian Mansion) and the more easterly Castle Lane, widened by Thomas Renda in the early18th century, close to the suggested site of the castle gates, the latter being partly revealed in the small resistivity survey (at the back of Munt’s Mill) to the east of the lane. On the eastern side of the castle, the ditch of the old stream from Benson can be seen; its flow formerly helped drive the Castle Mill on Queen’s Arbour (probably the U-shaped building on the resistivity). Also discernible is the ancient ‘Walditch’ (lying at right angles to the Benson ditch) which marked the northern boundary of the medieval ‘water of Wallingford’ and the Borough boundary. (LiDAR reproduced courtesy of the Environment Agency; resistivity super-imposed by Gerard Latham)

180

Figure 7.43 Wallingford Castle from the south, 17th October 1973. The motte can be seen prominently in the centre, capped by trees, with the sheds of Castle Farm catching the light on the south east. The Borough boundary wall curves along the back of the High Street properties, defining the southern limits of the castle grounds. To the north, lies the D-shaped barbican and the surviving ramparts of the outer walls. Copyright reserved Cambridge University Collection of Aerial Photography.

Figure 7.44 1548 map. Earliest known image of the castle’s inner bailey. Note how the moat surrounds it on all sides (cf. Figure 6.1 and Lloyd p. 124 ff.). Oxford, Bodleian MS Top. Berks b. 41, fol. 1, by permission of The Bodleian Library, University of Oxford.

181

Figure 7.45 Richard Gough’s plan of Wallingford Castle, 1768. (For Gough’s extended notes see 3.12.) Photo: Katharine KeatsRohan, reproduced by permission of The Bodleian Library, University of Oxford from Gough Maps I, vol. 1, fol. 43v.

Figure 7.46 The ‘Renda’ map. An early18th-century plan of the estate at Wallingford, probably drawn up by Thomas Renda as part of a dispute with Christ Church over boundaries c. 1715–19. It shows the tree-lined drive to Renda’s new house, marking the Christ Church boundary with a dotted line and squares. The latter almost certainly depict the ‘meer’ stones which Renda was accused of having moved during building works. The plan is 19inches x 28inches, drawn in ink with considerable detail, but with no scale. CCA, Maps Wallingford 1: image reproduced by kind permission of the Governing Body of Christ Church, Oxford.

Figure 7.47 Detail of Wallingford Castle, taken from ‘A Survey and Plan of the Estates of Robert Hucks Esq. of Aldenham, Herts, 1786’, showing ‘Land in the occupation of Robert Leaver. Part of a farm at Wallingford, Berks.’. (Note: numbers in pencil are additions to the original map.) The properties marked ‘A’, shown in red on the High Street, together with the triangular-shaped property block opposite, may have derived from the original holdings of the castle porter. Redrawn by Berenice Pedgley from Hertfordshire Record Office, D/EAm P1.

182

Figure 7.48 Detail from the 1804 Crown Estate Map, drawn by John Fordyce. (Note: the Queen’s Arbour ‘College Land’, is a reference to Pembroke College, Oxford, which held the tithes of All Hallows parish, including Queen’s Arbour, in 1804.) Courtesy of The National Archives, MPEE 1/10.

183

Figure 7.49 Sketch of the south side of Wallingford Bridge by Joseph Farington RA, c. 1792 (Wallingford Museum).

Figure 7.50 The north side of Wallingford Bridge drawn by Thomas Hearne F.S.A. c.1803, engraved by William Byrne F.S.A. Note the winch on the far right of the picture, used to help pull boats up through the narrow bridge and giving the name ‘Winch Eyte’ to the island on which it was sited (Wallingford Museum).

184

Figure 7.51 Print of bridge re-building works from the south side by William Havell, entitled ‘Wallingford Castle, taken in 1810 while the bridge was repairing’. Note the Town [East] Gate on the western end of the bridge and the ruins of the castle’s middle bailey wall in the middle background. On the far right is the temporary wooden bridge set up to enable people to cross the river. The building on the extreme right is probably what is now called Munt’s Mill (Wallingford Museum).

Figure 7.52 Water-colour by George Shepherd, 1815, entitled ‘The Town, and Bridge, Wallingford, Berkshire’. This splendid painting, made shortly after the completion of the major bridgeworks recorded by Havell (Figure 7.51) makes a fascinating comparison with Thomas Hearne’s depiction of the same area (Figure 7.50) a few years before the great flood of 1809 severely damaged the medieval bridge. The town gateway and associated buildings have gone; the Chapel Eyot has been much reduced, while the Winch Eyot is just visible, now incorporated into a town wharf with wooden camp-shedding (image © Ashmolean Museum, University of Oxford).

185

Figure 7.53 Sketch plan drawn in 1810 by Edward Driver of part of the Christ Church property at Wallingford Castle, showing lines of banks and ditches (CCA, MS Estates 5 document, fol. 143, by kind permission of the Governing Body of Christ Church, Oxford).

Figure 7.54 1812 ‘Plan of estates situate in Wallingford Castle in the County of Berks, belonging to the Dean & Chapter of Christ Church, Oxon.’, certified by James Blackstone in 1813, who had ordered it to be made (see CCA, MS Estates 5 document, fol. 156). The buildings shown include: No. 1, a malthouse with cottages and outhouses (the former Priests’ Lodging) and ‘grass ground’, once the site of the Dean’s House; No.2, the ‘Dwelling House,’ was the former Clerks’ Lodging. Also clearly marked as No. 3 is the road which gave access to the buildings. The plan measures 11 inches by 15 inches (28 x 38 cm) and has a scale of one inch to the chain (22 yards). CCA, Maps Wallingford 2; image reproduced by kind permission of the Governing Body of Christ Church, Oxford.

186

Figure 7.55 Crown plan of 1817 showing ‘A Freehold Estate at Wallingford in the County of Berks belonging to His Majesty’. The map is based on the 1804 Estate map (Figure 7.48) but includes the Lot Numbers by which the estate was to be sold. It has proved a useful guide in understanding the various ancient leaseholds. It is interesting to note that the area now recognised as the site of the main castle gateways and water control in the south-east corner is clearly defined as Lot No 5 (photo: Katharine Keats-Rohan, reproduced courtesy of The Bodleian Library, University of Oxford, MS Top. Berks. b. 10, fol. 17v-18r.)

187

Figure 7.56 Map of Wallingford c. 1835. The origin of this finely executed hand-drawn map is obscure but it is the earliest known map of the town and is held by Wallingford Museum. The earliest date possible for it is 1835, based on the presence of the Gas Works by the bridge, which opened in that year, However the absence of the house built for J. K. Hedges in 1838– 40 gives a potentially narrow window for dating.

Figure 7.57 The riverside restaurant 2015. This building is sited on reclaimed land, but the pathway to its right follows the ancient right of way round the Winch Eyot to Queen’s Arbour, beginning at Castle Lane, In the medieval period this scene would be dominated not by the spire of St Peter’s church (18th century) but by the town gateway on the bridge (left), the riverside wharf (centre) and the palisade and drawbridge of the outer castle gate (right) (photo: Judy Dewey).

188

Documents list 3 1. 1377 May 25 CIPM 14, no. 332, p. 326, William Goldsmyth of Walyngford. A messuage, held of the bishop of Winchester in fee by service of 7½d yearly. He bequeathed it to Roger Cole and Richard Stokebrugg, then caretakers and wardens of the bridge of Wallingford, and their successors, for the repair, amendment and maintenance of the bridge. He died about 21 Edw III [1347], day not known. After his death, Christiana his wife had possession of the messuage according to the custom of the borough of Wallingford for 21 years, when she died. After her death the said wardens had possession for a year and then enfeoffed Thomas Reginald of Wallingford, to hold to him and his heirs. Thomas has had possession ever since. William has no heir to the knowledge of the jurors. 2. 1508 May 6 23 Hen VII, Borough Minute Book, BRO, W/Ac 1/1/1, 5v Cometh to the Gilde Hall Geofferey Baynton, Robert Holon’ and William Hyll dyer, and payeth to be burgers and admitted and sworn. John Humson, being Beyly, paid and delivered to William Norborow, being one of the proctors of the Grt Brygge for a … that the foresaid John Humson received of Thomas Whythorn, the which he owed to the Gret Brygge of W 19s 6d Also that William Norborow received of Peter, parson of St Peters, for the repairing of the causey 4d ‘This ben the costs done upon the cawsey under the Mary of Grace be the oversight of William Norborow with the money that he recevyd of John Humson abovesaid’ to Bowdon for carayge of 8 lode of stone, the lode 5d 3s for 12 lode of white erthe a lode of sonde 3d to a mason for 6½ days 3s 3d to Thomas Hunte for 3½ days 21d to his man for 1½ days 6d to Blocksam for 3½ days 14d for 2½ qrs slecked lyme 20d for 9 bu of stone lyme 8d for 5 posts 2s to John Blake 2½ days 10d to William Bowe 2 days 12d for 2 raylys 3. 1513 Oct 11 5 Hen VIII Borough Minute Book, BRO, W/Ac 1/1/1, 8r Thomas Polenton mayor Richard Kelyng Thomas More Thomas Dogett Thomas Hevyn William Hyll William Baker of Mongewell chosen proctors for the Grete Bryge of W, and one or two of them that come into the Gyldehall every court day while the reparacions be adoing and there to receive money of Mr Mayor or his assigns to buy such stuff as shall be necessary for the bridge and those that shall receive this money shall bring a bill to Mr Mayor every week and so to give account of the moneys spent and the names of the labourers in the same bill and what wages they do take.

4. 1548 Aug 18 CPR 1547–53, 392 For £1477 15s 8d paid in the Court of Augmentations by the king’s servant Michael Stanhope kt, first gentleman of the Privy Chamber, and the king’s servant John Bellowe esq., grant of [among many others] the site of the late college of St Nicholas within Wallyngforde castle Berks and all houses, buildings, structures, yards, orchards and gardens commonly called ‘the Deanes lodgyng, the Preestes lodgyng and the clerkes lodgyng’ and all other houses etc within that site. 5. 1548 Survey by the Court of Augmentations, Oxford, CCA, MS. Estates 5, fol. 3 Firste the howses and lodginge called the Deane’s Lodginge parcell of the saide Scite and beinge within the Castell, standinge and adioyninge to the Hall of the same Castell, extendeth in length frome the howse called the preests lodginges in the west ende, unto the further ende of the gallarie of the said howse conteyninge therein, A hall, two parlors, diverse Chambers, Gallaries, a Buttery, a kechin, and other howses of office, Togither with a small garden or Orcharde severall and enclosed, with a pale adioyninge to the saide deane’s lodginge. The howses called the preests lodging, beinge a severall quadraunt Lodginge, extendinge from the west parte of the deane’s lodginge aforesaide unto the Est side of an Orcharde theire called the preests Orcharde latelie annexed to the lodginge called the Clerkes Lodginges, where in conteyneth a hall, a Buttery, pantery, kechinge, and certaine Chambers, and other necessary howses, with a Cloyster, and a square plote of grounde, within the saide Cloister. The little howse and kechyn called the Clerkes lodginge, with a plote of grounde aboute the same and a litill Stable thereunto adioyninge togeither with the Orcharde called the preests orcharde, and the litill olde larder house, and Well thereunto adioyning, now in the tenure of Thomas Parry, Esquier. Md. one side of the Cloyster is covered with leade. Also there is yeat remayninge, within the saide preests lodginge, one litill Cloke, and Bell thereunto belonginge, whiche is charged in the inventory amoung th’other goodes to the saide Colledge late apperteyninge. 6. 1570 Charges done at Wallingford in the Colledge for Queen Elizabeth’s progress, Bodleian, MS. Rawlinson A. 195c, fols 2678 [calendared extracts only; text modernized and accounts abridged in square brackets] Masons occupied in taking down and overthrowing certain high walls above the gate and other necessaries [3 masons for 3 days @ 3d a day] Carpenters occupied not only in mending of the floors, windows, stairs and doors in the Queen’s lodging but also in making a pair of stairs for the maids of honour chamber and other necessaries. [2 carpenters for 6 days @ 6d a day] Labourers occupied in laying of the stone paving in the way and making clean the [buildings?].[3 labourers for 4 days @ 2d a day, 1 for 7 days 7d]

189

Payments for plane-, elm- and plank-board, nails, making 2 double casements of wood, locks, bolts, iron hooks, and transport of the tent and other necessaries from Rycote to Wallingford. 7. 1622/3 Feb 25 20 James I Extract from Indenture, TNA, E317/Berks/33 i. Sir Henry Hobarte bart, chief justice of HM Court of Common Pleas ..., Thomas Murray esq, Sir James Fullerton kt ..., Sir John Walter kt ..., and Sir Thomas Trevor kt; ii. Thomas Payne of Wallingford gent. Whereas i are possessed of the following lands, parcel of the manor of Wallingford, for the use of Charles prince of Wales; and whereas the late queen Elizabeth by Lrs Pat 24 Nov 32 Eliz [1589] did demise these lands to Nicholas Payne clerk, Controller of HM Household, for 21 years at a yearly rent of 40s; and whereas the late queen Elizabeth by Lrs Pat 17 Feb 32 Eliz [1589/90] did demise in reversion the several parcels hereafter demised unto Grigory Lovell esq, Cofferer of HM Household, for 21 years at a yearly rent of 40s; and the same have come to Thomas Payne; now this Indenture witnesseth that i, in consideration of £22 10s paid to the Receiver General as follows - £11 5s on delivery hereof and £11 5s on 12 August next - have let to Thomas Payne (a) the herbage, feeding and pasturage of the whole site and precinct of the castle of Wallingford now altogether decayed, and all the herbage, feeding and pasturage of all the ditches, banks and valleys of the said castle called the Castle Banks; (b) 1½ ac meadow with a certain ditch called the Porters Ditch lying in a meadow called the Kings Mead near the said castle; (c) 1 ac of arable land called the Porters Acre lying in a certain field called the Clapcote [Field] now or late in the tenure or occupation of Nicholas Payne or his assigns; (d) all that tenement called the Dyehouse lying in the High Street near the Eastgate ... in the parish of St Peter, with the toft opposite to the said tenement now in the tenure or occupation of Albane Spencer dyer or his assigns; (e) all that close lying in Newenham Morren co. Oxon called the Porters Close containing by estimation 1½ ac of pasture late in the tenure or occupation of Ralph Pollington deceased, with buildings lately erected; which were the lands and tenements of old belonging the office of porter of the said castle and parcel of the manor of Wallingford, parcel of the possessions of the honor of Ewelme (the inner ward of the castle and the bank next encompassing the said castle with the moat and fishing thereunto next adjoining, and also all great trees, timber trees, woods ... excepted), from feast of St Michael last past before the date hereof for 31 years at 30s p.a. ... [one condition being to plant each year six trees of oak, elm or ash and to preserve them to the end of the lease; another condition being not to plough up any meadow

ground or pasture ground or sheep walks, parcel of the demised premises]. Signed John Walter, Thomas Trevor 8. 1652 Nov 18 Wallingford Castle Demolition Order, transcribed in 1977 from the original demolition order for Wallingford Castle loaned by Sir John Hedges – still in private hands. The order followed the promulgation the previous day in the Council of State: Calendar State Papers Domestic: Interregnum 1651–2, 496, 17 Nov 1652) Gentlemen, The Councell haveing taken into consideracion the present state of affaires, and thereupon finding it necessary, that as many of the Inland Guarrisons as can conveniently be spared, be forthwith demolished, as alsoe the request which hath beene made to the Councell on the behalf of severall of the Inhabitants of the County of Berks, for the demolishing of Wallingford Castle, have thought fit that the said Castle should be forthwith demolished, & the works thereto belonging effectually slighted. And that the execution thereof should be committed to yor especiall care. You are therefore, upon receipt hereof, desired & Authorised to proceed to the demolishing of the said Castle of Wallingford, and the works and fortification thereto belonging. And for the defraying of the charge wch shall accrue upon the doeing of what is herein required, You are to make sale of the Materialls belonging to the said Castle & Fortifications, as alsoe of the Hutts, Magazine House, or any other buildings which were newly erected within the Fortifications of the said Castle, whilst it was held as a Guarrison for the late King, and with the proceeds thereof pay off the Labourers and other persons who shall be employed upon this service, and in case that after the defraying of this charg, there shall be any remaynder of money left upon the proceed of the above said materials & houses, you are to dispose it as you in yor judgmts shall thinke fit for the reliefe of such of the poor people dwelling thereabouts as suffered by that Guarisson whilst it was held by the late King. Signed in ye Name and by order of the Councell of State appointed by the Authority of Parlamt. Whitehall, E: Jo: Thurlow, Clerk of the Councell. Wm Masham, Presidt 18 Novembr 1652 To Maior Evelyn, Mr Edw. Jennings of Long Wittenham in ye County of Berks, & Mr John Rusden 9. 1660 June Papers relating to the Act of Indemnity, RCHM 1879, 96a. Petition of John Freeman gent. Petitioner and his predecessors have been ancient tenants to the Queen, of the castle of Wallingford, which was in 1641, by order of his late Majesty, made a garrison, and large storehouses and other buildings for the service of the garrison erected; in 1652, by colour of a pretended warrant, Major Arthur Evelin and others pulled down not only the storehouses and buildings but also the ancient walls, and in doing so destroyed the petitioner’s gardens, orchards, etc. so that it cost him £300 to repair the same; prays that the persons above named, and also William Cooke, who without any pretended order carried away a great part of the stone 190

and other materials, may be excepted out of the Act of Indemnity. 10. 1719 June 4 Extracts from witness statements from the boundary dispute between Thomas Renda and Christ Church College (taken at the house of Thomas Maxey called The Black Horse Wallingford), CCA, MS Estates 5, fols 51 onwards, starting at foot of fol. 53. [53] Robert Merwick: Containing half an acre in this defendant’s judgement or rather more for that the land doth extend from the mote lying on the south side up to the mote lying on the north and [54] from the malthouse at the west end to a place where a drawbridge formally was at the east end and so along to another place where a large stone formally lay by the castle gate but how the same is now bounded this defendant cannot depose, the afore said ancient bounds being removed and carried away and this defendant further saith that a great alteration hath been made therein for that the land hath been digged and cutt and a coachway made of some part thereof about fifteen or sixteen years [55] since, and saith that the way or walks from Wallingford Bridge to the castle by the west side of the said parcel of ground is made wider than the same formally was by eight yards or thereabouts the same being formally about a yard and lying on the east side of the two severall rows of walnut lately planted and now standing, but whether the said way or walk is made on the castle ground or on the College ground this defendant cannot depose … [58] To the sixth interrogatory this defendant saith that he did very well know the antient ways or passages on the west side of the house called the College [inter] which ways and passages have lately stopt up, and taken away, by the defendant or on his order, and the gate leading from the Castle Street to the said College, have been locked up by the defendant on his order soe that noe carts or carriages could pass and saith that the way leading from the said gate to the said College have been plowed up and sown with corn and this deponent further saith that by the stable [59] belonging to the house in the possession of Mr Longland parcell of the premises belonging to the said College, there was a gateway or grainery, under which carts and carriages passed, with large gates to shut up the same, to which such tenants, belonging to the said College had a key to lock or unlock the same, at their pleasure, and this defendant saith that the same Gateway or Grainery is now taken down, and hath heard and believes that the same was taken down by the defendant and one Anthony [60] Leaver who was a tenant to the said College and the timber and materials thereof were parted and divided between them, since which this defendant saith that the said defendant, Renda, hath set up an other gate which is most times locked for that carts or carriages can’t pass to the house belonging to the said college upon which gate the defendant did set spikes whereby foot people were also debarred of their passage there......[64, middle of deposition of Owen Tuder of Wallingford, labourer aged 73 years]....To the fifth interrogatory the defendant saith that he did know the ancient way or passage leading from Wallingford bridge inter’ and the gate and stile belonging

to the same which stood between the highway or street and the bank/s, and saith that [65] the same are much altered by the defendant Renda for that the the wicker gate and style are taken down and a large pair of gates set up near the same place or somewhat more westward …To the sixth interrogatory the defendant saith that the ancient way or passage on the west side of the south inter was a large cartway with a spar’d gate from the castle street over the Bank up to the Gaol Barne, and from thence over other banks under a gatehouse which stood near [66] the said College, which spar’d gate had been long since removed and carried away, but by whom this defendant knows not, saith that an other gate had been made and put up and kept locked and bussels placed thereon by the defendant or his order so that not only carts and carriages but foot people have been prevented and stopped from passing, and this deponent also saith that the said cartway has been plowed up by the said defendant and corn and grain planted thereon, and further saith that he [67] hath heard and believes that the said gatehouse was pulled down by one Anthony Leaver or the defendant or one of them, but particularly remembered that the said defendant took away the ground pinnings thereof, saith that where the said Gatehouse stood the defendant Renda or some person by his order hath put a gate with spikes which was kept locked whereby the ancient way was stopped and prevented. [70] William Maxey: some alteration hath been made by the defendant for that he hath greatly enlarged the way or walks from Wallingford Bridge, to the Castle, by digging down the west side of the said parcel of ground and this deponent saith how much [71] the same way, or walk, is enlarged, this deponent can’t set forth, but believing great parts thereof is made on the ground belonging to the said college… [80] Mr Hobbs and others: declared, that the same [property boundary] extended from the end of the wall next the malthouse, now in this deponent’s possession, down to a stone which this deponent apprehends and believes was beyond the lowest rank or row of walnut trees now planted and standing upon or near the said parcel of ground, and from thence [81] to the end of the Castle wall on the east side of the outhouse situate on the east part of the court or yard belonging to the messuage or tenement, now in the possession of the said defendant Renda and the deponent further saith that [he] remembers some alteration hath been made on the east part of the said parcel of ground and some considerable part thereof digged down levelled and laid into a way or walk, but when the same was done or how much in quantity, the same [82] this deponent can’t depose, but saith that the old way or walks from Wallingford Bridge to the castle was thereby made much wider. 11. 1726 Extract of property definitions from a Crown lease to William Hucks, TNA, E367/6898 Parcel of the possessions of the manor and castle of Wallingford within the Honor of Ewelme and late in joynture to her Majestie Catherine … The farme of all that meadow called or known by the name of North Eights [Eyots] … in Wallingford and

191

extendeth in length from the Thames where the king’s fulling mill was sometimes scituate and built and so extendeth along the west side of the Kings Meadow there and all that piscary and fishing upon the river of Thames near to the meadow called the Kings Meadow late in the tenure or occupation of Robert Arnold …and all those small eights or peeces of land containing about half an acre … lying within the banks of the said river .. where the said fishing was against the Kings Meadow and the Little Meadow.... which said premises were demised .. by her late Majestie … by indenture 28 Mar 1682 to John Arnold of Bensington [on lives of his sons Nathaniel, John and Joseph] paying £2 1s yearly. The farme of all those 23 acres of meadow, parcel of 36 acres called Kings Mead near the Castle of Wallingford, which said premises were demised by the late Princess Queen Henrietta Maria [widow of Charles I, died 1669] by indenture 7 Jly 1631 to John Gregory the elder and John Banister for 60 years [on lives of John Gregory junior, Richard son of Henry Kersall and Paul son of John Banister]; and by indenture 6 Nov 1678 to John Freeman esq and Moses Slade gent both of Wallingford in trust for Stoner Crouch of Wallingford, Anthony Lever and John Lever of Wallingford, and afterwards to the said Henry Kersall, John Gregory and John Banister for remainder of 60 years if Edward son of Robert Lever of Wallingford and Joseph son of Paul Banister so long live, at rent of £4 17s 6d yearly and £2 heriot at either death. The farme of all those 10 acres of meadow, parcel of 36 acres of Kings Mead near the Castle of Wallingford, now or late in the occupation of Thomas Leaver of Crowmarsh yeoman and one other acre now or late in the tenure or occupation of Thomas Bishop of Wallingford farrier which premises were amongst others demised by ... Henrietta Maria by indenture 7 Jly 1631 to Henry Kersall, John Gregory the elder and John Banister for 60 years [on lives of John son of John Gregory, Richard son of Henry Kersall and Paul son of John Banister]; later demised by Catherine … by indenture 26 Nov 1668 to John Ince of Cliffords Inn London gent for 60 years [on lives of Frances daughter of John Ince and John son of Paul Banister] for yearly rent of £2 4s 6d The farme of all that Inward Ward of the Castle of Wallingford and the Banks next encompassing the said Castle Wall with the moate and the fishing thereunto next adjoyning … which said Inward ward and premises do now consist of a dwelling house, a garden, a small bowling green, and two orchards, and are parcel of the manor of Wallingford and likewise parcel of the Honor of Ewelme which said premises the late Catherine … demised by indenture 11 Aug 1699 to Ralph Lawton of the town of Northton esq for the yearly rent of 10s for 55 years [on the lives of Ralph third son of William Lawton of Lawton Ches dec’d, William eldest son of [blank] Trafford of Swythomley Staffs esq, and James son of James Chetham of Mellor Derbys gent] 12. 1768 Bodleian MS Top. Gen. e. 16, fols 175–6, the original note on which Richard Gough based the description published in his edition of Camden’s Britannia

(I, 158) which Hedges quotes (II, 191); it includes some details not in the published version. The castle situate at ye NE end of the town is fortified with a treble vallum and ditch; ye outer extent about 300 paces from – to – by 400 from – to – . The outer W rampart planted with trees & ye outer & inner S one both in garden grounds are exceedingly bold and fresh. The outer N one is in corn fields; & the E one is single being secured by the river. This last is broke into by two bastions, to which two others correspond on the N side. On the S inner bank are some fragments of stone walls in houses & in another building a wooden door case as of a religious house. The principal gate is on the SE corner of the inner bank: part of the E pier remains & just within it is herring bone work: further on appears a pointed window as of a chapel. A large piece of ground has been walled in for an orchard at the NE corner; & towards the W end of the N bank within is an old stone building now a barn formerly part of the base court. The keep, of considerable height, falls into the SE corner of the inner works. The area may contain about – acres. The yard of St – church [All Hallows] demolisht in the civil war is in ye W ditch. [Then quotes Leland] Opposite to the south side of St Peter’s, now rebuilding, is an old chapel, now a dissenting meetinghouse. Some banks are discernable just without the town on ye left hand going to Oxford. 13. 1806 July 8 Borough Minute Book, BRO, W/Ac 1/1/1, 105r Begin at the north end of Castle St at a place there opposite to the Black Ditch called Clabcot Gate; and from thence down to the first moat and so along the said moat by the east side of the gardens belonging to the houses in Castle St and on the east of the church yard there (taking in all Castle St and the church yard) to the garden wall late of Mr Charles Toovey and now of Mr James Wells; from thence straight by the garden walls there to the gate of Robert Hucks esq at the south end of the wall that goeth into the castle; from the said gate along the ditch called the Bishop’s Ditch and so following the whole course of the ditch to the River Thames; then cross the said river (taking in the whole of the river, and also the eyots belonging to the Corp, the meadow and eyots belonging to William Nedham esq held under the Bishop of Winchester, and the Bridge meadow belonging to the Corp) to the east end of the Great Bridge; from thence (taking in the Bridge Pools on the south side of the said bridge, and also a meadow called the Island containing about two acres belonging to Henry Blackstone esq) on the east side of the river and also the whole of the river southwards as far as Chalmore Ditch where again cross the river; and so along the said ditch to Winterbrook Bridge; and so up Bradford’s Brook (only leaving out a small piece of ground belonging to Mr Thomas Greenwood held of the manor of Cholsey) to the Old Way Bridge; and so along the middle of the Old Way (taking in the north side of the Way) unto the west end of Wallingford Moor; and from thence (taking in the whole of the said Moor) to Lock Bridge leading to South Moreton; and from the said Lock Bridge down the Mill Stream to a mear baulk on the west side of the culver [sic], otherwise

192

Brook Piece; along the said mear baulk to a place near the Turnpike called Handycross; then cross the road (taking in the Turnpike House and garden) and proceed on the outside of the North Hedge belonging to a piece of land formerly called Priory but now Picked Piece, straight by the said hedge to the north corner of a piece of ground called Georgia, formerly an orchard in the possession of Mr Charles Fludger (taking in all Priory or Picked Piece); and from the corner of the said orchard along Black Ditch in the Bull Croft to the said place called Clabcot gate’. 14. 1810 Jan 26 Letter accompanying the sketch plan drawn by Edward Driver of part of the Christ Church property at Wallingford Castle, CCA, MS Estates 5 document, fols 142–3. Sir, Having received from your Office a Plan and description of the Claims of Christ Church College, Oxford to a certain part of the Estate at Wallingford belongingto the Crown, I have in consequence thereof taken another view of the Premises with their Plan in my hand and had much wished the Surveyor who made the same could have done me the favor of a meeting on the ground but which I learn he was prevented from doing being upon a Journey into Hertfordshire, however on this sheet I annex a Sketch of the Scite of Ground claimed by them and which is coloured Green, distinguish [sic] thereon the remains of the antient Moats etc which most

evidently formerly enclosed the Castle and Lands and which from the remarkable declivity of those parts marked A.A were made for the Security of the Castle etc, and upon the top of which are standing in a regular line, a number of very large and old Elm trees, which certainly may fairly be presumed to have been planted by the Crown and not claimed by the College and which is marked B. and the space between the two Moats and which is not unusually distinguished by its resemblance to a Hogs Back is only about one third the width of the whole and was no doubt formerly a necessary fortification to the Place. Therefore under all these very strong circumstances it appears to me very improbable the Land required by the College could possibly be in the Situation so ingenuously designated by them upon the Plan; so as to comprise [described] precisely the dimensions described in their description as they thereby descend the Bank of the Moat which is nearly perpendicular to the Edge of the water or Moat at the place marked C. and include by that line about 14 of the large Elm Trees and on the Southern side their line extends to include half the old Moat but which has been reduced in the depths of its banks and since cultivated as a Garden and at this time is extremely Steep - There does not appear upon the Ground any trace or remains of a division of the Land claimed by the College, nor are there any Meer Stones to the Same - I am Sir Your very obedt. & humble Servt.Edwd. Driver. Kent Road Jan: 26th 1810.

193

8 A TALE OF TWO TOWNS AND TWO CASTLES: NOTTINGHAM AND WALLINGFORD COMPARED David Roffe Abstract It is a commonplace of English historiography that royal power was conducive to economic growth. It is the more strange, then, that Wallingford should appear to show signs of a faltering economy in the 13th century despite a 300-year history as a major centre of power. It is argued here that this is to misunderstand the mainspring of the borough’s economy in the medieval period. A comparison of the experience of Nottingham with Wallingford shows that vitally strategic sites, by the very virtue for their military importance, developed a command economy that effectively stifled growth. Wallingford, like Nottingham, was essentially a fortress and never developed a vibrant urban economy. As such, its characteristics provide a model for pre-urban burghal communities elsewhere in Wessex and Mercia. Introduction It may seem somewhat perverse that I should propose to compare Nottingham and Wallingford. Today the towns are like chalk and cheese. From a southern perspective, I suspect, Nottingham is seen as very much a post-industrial northern town. There are affluent areas, although they tend to be in the suburbs, but its culture is flinty working class. No longer Saturday Night and Sunday Morning, perhaps, but still decidedly ‘up-north’. By contrast, from a northern perspective Wallingford is a much softer place: genteel, middle class, quintessentially English, Causton in reality as well as in Midsomer Murders. We all have our preconceptions and prejudices. Nevertheless, I think it is true to say that a world has always separated Nottingham and Wallingford. I suspect that there was never a time at which the history of one directly touched the other. However, having said all that, there are some remarkable parallels between the medieval histories of the two towns that are worth closer examination. I have touched on something of the parallels in the discussion of the liberty of the honour of Wallingford (Roffe, Chapter 4 above). There I argue, following Katharine Keats-Rohan (2009), that Robert d’Oilly inherited not only the lands of his father-in-law, Wigod of Wallingford, but also his office. I concluded that the liberty of the honour of Wallingford was essentially a preConquest institution and cited Nottingham as an example of a place where a constable also inherited pre-Conquest lands and functions. That, however, does not exhaust the similarities between the two towns. Here I want to explore the parallels in more detail to see if they can suggest a model for the development of Wallingford and the relationship of castle and town.

Nottingham The history of Nottingham, as with so many towns, starts with a river crossing. The River Trent plays much the same role in the East Midlands as the Thames does in the south (Dixon, Knight, and Firman 1997). Draining into the Humber estuary, it was the main line of communication between the North Sea and the heartlands of Mercia. It was, then, a vital strategic link in the East Midlands and Nottingham dominated it. The town is sited atop a cliff on the north bank at its lowest fordable point where the main road between the south and the north crossed the flood plain of the river. Throughout the medieval period and beyond Nottingham was the key to the control of the North (Roffe 1987; Roffe 1997). According to Asser writing in the late 9th century, it was called Tigguacubauc in the Middle Saxon period, a British name meaning ‘house of caves’ (Asser, 24 and note). Nottingham is still famous for its caves. The Old English name, Snotengaham, Snotingeham, ‘the ham of the people of Snot’ where ham means something like ‘estate’ (Gover, Mawer, and Stenton 1940, 13), suggests a wider territory, perhaps encompassing much of the land around the present town on both sides of the Trent (Roffe 1997, 25–6). The Danes over-wintered there in 868 and it was probably their defences that were repaired by Edward the Elder, king of Wessex, in 918. Two years later Edward returned and built a second borough south of the river – its site is unknown – and a bridge (ASC, 104; Roffe 1997, 27– 9; Haslam 1987). Thereafter it remained English with only a brief interlude of hostile Danish occupation between 939 and 942 (ASC, 110; Downham 2009, 147–9). Although Nottingham had a defined territory in the reign of Athelstan – it took in southern Nottinghamshire and south Derbyshire (Roffe 1986, 112) – there does not seems to have been a coordinated burghal system like that of Wessex at this time. Or at least we don’t know about one. We only have evidence for later in the century. Lincoln, Leicester, Nottingham, Derby, and Stamford have usually made their outing in English historiography as an alliance of Danish boroughs (Lapidge 1999, 186). However, in reality they never seem to have acted together before the mid-10th century. In fact, quite the contrary: Leicester and Northampton had a temporary alliance between 913 and 917, while Lincoln remained allied with York until 927 or later (ASC, 96–8; Roffe 1987, 236; Smyth 1987, ii, 8–9). The Five Boroughs of the Northern Danelaw, as described in the late 10th-century Wantage Code (EHD I, no 43, c.1), incorporates late 10th-century legal ideas and is clearly a later English institution (Roffe 1992). Like Alfred’s burghal system, it was defensive in nature, its target here 194

Figure 8.1 Medieval Nottingham (© Nottingham City Museums).

being Yorkshire and the North which remained Danish until 954 and thereafter bandit country well into the 11th century. Nottingham may well have been its administrative centre. By 1016 the Five Boroughs had given way to the shires of the East Midlands and Nottingham had become the county town of Nottinghamshire (ASC, 148–9). On the eve of the Conquest, there was a massive concentration of royal and comital estates in the shire which is unparalleled elsewhere in the East Midlands (GDB, 280–93: DB Notts). Nottingham itself was above all a royal settlement. Domesday Book records no lords in the borough other than the king and earl. Ecclesiastical structure attests the essential accuracy of this picture. There were only ever three churches in medieval Nottingham and in 1066 there was probably just the one. St Mary’s church, on top of the hill within the English Borough (Figure 8.1), was a major royal foundation which signals the dominance of the king (Roffe 1997, 35–6). Nottingham was no less important thereafter. The castle, built by William the Conqueror in 1068 to secure the route to the north, was one of the first to be built in the East Midlands (Marshall and Foulds 1997, 43). It was entrusted to William Peverel and he was granted an extensive honour in Nottinghamshire, Derbyshire and Northamptonshire for the support of his office. The Nottinghamshire element of the fee looks like a post-Conquest castlery. And, indeed, it has usually been treated as such (eg Fleming 1991, 148– 51). In reality, however, the block of land had a decidedly pre-Conquest identity in much the same way as did the

honour of Wallingford. In 1086 William apparently held Earl Tostig’s estate to the west of the English Borough in Nottingham where the honour met throughout the Middle Ages (GDB, 280: DB Notts, B2; Foulds 1988, 42–3; Stevenson and Stapleton 1895, 43). In addition he held the comital manor of Clifton to the south of the borough (GDB, 287: DB Notts, 10,5–6). His remaining estates, almost all the lands around Nottingham, had been held by the earl’s thegns, many of whom continued to hold at the time of Domesday under much the same terms as they had in 1066 GDB, 287–8: DB Notts, 10; Roffe 1997, 36–7). William Peverel seems to have assumed the military functions of the pre-Conquest earldom and its apparatus. He did so, however, as a minister. William was never the lord of Nottingham and he held the castle in his capacity as constable of the king. Much of his lands may also have been held ex officio. A block of the Derbyshire lands of his fee and a number of Nottinghamshire manors are explicitly said to be held in custody in Domesday Book (GDB, 273, 287v: DB Derby, 1,29; 32;35–6; DB Notts, 10,23–4;32). Nevertheless, the honour passed to his son William II Peverel in 1114 who held the castle and town for King Stephen during the Anarchy. For his pains, he forfeited the honour in 1153. The fee was briefly held by Ranulf of Chester, but, on his death in the same year, it escheated to the crown (Sanders 1960, 136; Marshall and Foulds, 47–8); COEL). Thereafter, it remained royal demesne throughout much of the Middle Ages. Like the honour of Wallingford, it was a royal appanage and was considered to be a liberty (PQW, 638b). 195

The castle was one of the major fortresses in the North. Prince John held it against King Richard and it was besieged in 1194 (Foulds 1991). In the civil war of 1215– 16 it became the headquarters of the king’s government north of the Trent. Henry III spent vast sums of money strengthening and extending the defences, more than on almost all the other castles in the realm. In the 13th century it was a garrison and arsenal, being used for the storage of siege engines and the like. It was, though, not just a fortress. In the 1250s Henry III ordered a new donjon to be built and specified that it should have windows overlooking both the town and the Trent. He minutely supervised the furnishing of the rooms. The castle, then, was also a favoured residence and it was to be appreciated as such by Henry’s successors. By the 14th century it had become more a palace than a fortress and so it was used throughout the Middle Ages. Royal interest in the site continued until the reign of Elizabeth (Marshall and Foulds 1997). The history of Nottingham and its castle in the medieval period is characterized above all by unremitting royal oversight and control. The parallels with Wallingford are resounding. Wallingford’s pivotal role in the defence of Wessex in the 9th and 10th centuries and the control of the middle Thames valley thereafter is well known. Here too after the Conquest was one of the earliest and biggest castles in the country and the provisions for its defence also drew on Anglo-Saxon arrangements. It was to remain a royal stronghold and palace for much of the Middle Ages (KeatsRohan 2009; Roffe 2009). Nottingham and Wallingford are remarkably similar. What, then, can the experience of the one tell us about the other? I think quite a lot.

The borough and urbanization One of the things we ‘know’ about the major boroughs of the later Anglo-Saxon period is that mercantile activity went hand in hand with fortification. Boroughs were as much about secure trading as defence (Biddle and Hill 1971). Nowadays we are, perhaps, less likely to think in terms of a single town plan (Edgeworth 2009, 79–80), but it cannot be doubted, it is argued, that markets and merchants tenements were an integral part of the preConquest townscape. Domesday shows us vibrant urban communities with structural links to the countryside – the contributory manors – that were all about trade (Fleming 1993). The construction of castles after the Conquest further stimulated growth (Creighton 2002, 133–74). This is a picture that clearly had some reality in the Danelaw. Cities and towns like York, Lincoln, and Stamford were powerhouses of economic growth and trade from at least the early 10th century (Hadley 2006, 247– 54). Their phenomenal growth has come to characterize all the boroughs of the Danelaw. In the past it has been assumed that Nottingham was no different. However, archaeological excavation in the 1970s and early 1980s, still unfortunately unpublished, has changed that view. There is, in fact, little evidence for a Danish borough in Nottingham after the, as yet, unlocated winter camp of 868. The boroughs of the southern Danelaw took the brunt of the campaigns of English conquest in the late 9th and early 10th centuries and Nottingham probably remained in the sphere of York and was left largely undefended. It was re-fortified and garrisoned with English and Danes by

Figure 8.2 The archaeology of medieval Nottingham (© Nottingham City Museums).

196

Edward the Elder in 918 in what can be seen as a peace treaty, but it did not become a frontline borough until 920 when the Danish regime at York was superseded by the vikings of Dublin (Roffe 1997, 25–9). It was probably from that date that Nottingham was fortified with a massive timber rampart and ditch encompassing an area 500 x 600m which came to define the English borough for the first time (Figure 8.2). The defences were re-modelled in the mid-10th century, but there is little evidence of what we might call urban activity. No vestiges of industry have been found, apart from a small pottery kiln in Halifax Place that seems to have produced wares only for the high status site on High Pavement, the main street of the settlement, with which it was associated. There is some evidence for planning in the later 10th century in the east of the borough – terracing and the like were picked up in an excavation between Woolpack Lane and Barker Gate – but plots were large and the building on them small. There was always a great deal of space in the borough (Young 1985; Roffe 1997, 31–5). Economic activity is, of course, difficult to measure. But if the output of mints is some sort of proxy within the context of the vibrant economy of the Danelaw, then it was always modest before the Conquest. Neighbouring boroughs were much more productive. In particular, considerably more coin was produced in Derby some twenty kilometres to the west (Stafford 1985, 45). The two boroughs were, in fact, closely related. The tolls of Derby belonged to Nottingham until the 12th century and the sheriff of Nottingham administered both boroughs (Roffe 1986, 111–13). Derby itself was probably never defended in any substantial way – the defences noticed in 917 were apparently those of nearby Little Chester (ASC, 101; Birss and Wheeler 1985) – and it is very likely that it was a mercantile suburb, as it were, of Nottingham (Roffe 1997, 38). Before the Conquest Nottingham itself seems to have been primarily a fortress. In conventional historiographical terms, this conclusion seems bizarre, indeed most unlikely. But it is a pattern of development that is well attested in the Middle Saxon period. The manifold functions that we associate with towns – market, courts, churches etc. – were at that time often dispersed among a number of settlements. This is best exemplified with the pairing of wics, markets, with burhs. So, Hamwih was twinned with Winchester, Lundenwic with London, Fordwich with Canterbury, etc. (Biddle. 1976, 461–2; Hobley 1988; Tatton-Brown 1988, 214–15). I am suggesting here that it is precisely this pattern that persisted into the late Saxon period in Nottingham. As a regional centre of royal power, the borough commanded food rents from the surrounding royal estates and they supplied most of its needs. It was the very strength of royal power and the resources that it could command that made mercantile activity, if not otiose, then less essential. The construction of the castle saw significant changes in the topography and society of Nottingham, but in the long run did not fundamentally affect this tributary economy. The castle itself was built on what appears to have been an unoccupied site on Castle Rock some 500m to the west of the borough. However, a French borough was founded

sometime between 1068 and 1086 in the space between, perhaps with the specific aim of servicing the stronghold: its three axial streets radiate out from the east gate of the castle (Marshall and Foulds 1997, 44–6; Mastoris 1981). We have no information on the relationship of the castle with either the French or English borough in the first hundred years of its existence. Thereafter, though, it was little more than a grand backdrop to the town of little economic value. Although an arsenal and favoured residence, the castle was rarely occupied. Henry II visited only seven times between 1153 and 1185 and Henry III 13 times between 1226 and 1264. From 1226 to 1405 three-, five- and seven-year gaps between royal visits were the norm. None lasted more than a few weeks at most. Much of the time there was only a skeleton staff in place, probably not more than a handful of people. Building campaigns increased the numbers, especially in the reigns of Henry II, Henry III, and Edward III. Local labour was used from time to time but the King’s Works supplied almost all of the craftsmen. Much of the materials used was also sourced outside Nottingham, usually from royal manors in the region (Marshall and Foulds 1997, 48–54). The castle was largely irrelevant to the development of the town. The French Borough was a stimulus of kinds. As far as can be determined all of the holdings recorded in Domesday Book were located in the new borough (Roffe 1997, 36–9). It would therefore seem that the institution introduced a degree of tenurial heterogeneity that was theretofore unprecedented. Significantly, St Nicholas’ church was situated within the new borough and the third church, St Peter’s, nearby. It is certainly true that, with the construction of the Saturday Market between the English and French Boroughs, perhaps in the 12th century, the centre of gravity of Nottingham shifted decisively to the west. Nevertheless, Nottingham was always a small town, ranking well below most towns in the East Midlands in the 14th and 15th centuries (Foulds 1997). Although it received its first charter of liberties c. 1155, it could never compete with the likes of Lincoln and Stamford. By then it had missed the boat. Wallingford Nottingham, then, was a major royal centre both before and after the Conquest, but the borough never developed a particularly vibrant economy. Did Wallingford have a similar trajectory in the medieval period? The similarities in the history of the two settlements suggest to me that we should take the possibility seriously. Hitherto, I think, we have tended to assume that since Wallingford was one of the most important royal centres both before and after the Conquest, it must have also been on a par with, say, Oxford and Winchester with a flourishing mercantile economy. We have been left with the considerable reality of a ‘decline’ in the 13th century that defies explanation. Why should Wallingford not have been as successful as its neighbours? If royal control stifled development as it seems to have done in Nottingham, then perhaps we shall have an answer. Wallingford too had always been a fortress. 197

The power of the crown in Wallingford is clear from the Domesday account of the borough. King Edward the Confessor and William the Conqueror after him were lords of the town. That, of course, is true of all the more important boroughs in England at the time. What distinguishes Wallingford, however, is the large number of ministri, royal servant, who also held land there. Foremost amongst these in 1066 was a garrison of housecarls, that is household troops (Roffe 2009, 33–42). Other such garrisons may well have been found at other places, but this remains the only reference to such. It suggests that the king attached special importance to Wallingford. Equally significant is the absence of the earl. After his conquest of England in 1017, King Cnut had kept Berkshire in his own hands. Subsequently, the earldom, and the control of the fyrd that it suggested, passed to the Godwine family (Hooper 1988, 7–9). The few manors in the county held by Earl Harold in 1066 may have been appurtenant to the office. There is no record of a comital fee in Wallingford, however. Defence of the borough, and presumably oversight of the army, was assigned to a minister, Wigod of Wallingford, probably as a staller. The social structure of the borough reflects that degree of control. A large number of properties were attached to rural manors in the surrounding areas. Elsewhere such links are usually taken to indicate mercantile activity. Most of these in Wallingford, by contrast, would appear to be primarily related to military obligations. Thus there is a concentration of the manors in southern Oxfordshire east of the River Thame, the probable territory of Wallingford beyond the Thames before the foundation of Berkshire in the late tenth or early eleventh century. Although their lords probably enjoyed some of the king’s customs, many dues were reserved. Few, if any, of the fees had their own courts in the later Middle Ages. Above all, they generally do not seem to have had ecclesiastical rights. Of the eleven medieval churches in the town only St Mary’s and possibly Holy Trinity, is known to have belonged to one of the manors. The remainder were the king’s or subsequently came into his hands (Roffe 2009). Wallingford’s credentials as a major royal centre before the Conquest are as strong as those of Nottingham. Of its economy we know little. There is no evidence of significant industry, although this is perhaps not surprising for a Wessex borough at this time. The output from the mint was not especially exceptional until the mid-11th century, and then perhaps reflecting its political rather than economic prominence (William 2012; for the role of minting in tributary relations, see Carver 2010). Above all, it is clear is that the extent of development was always limited. The south-east quadrant of the borough was the only area of the town that was intensively settled (Creighton et al 2009). Wallingford was no Lincoln or York which both experienced rapid growth at this time. I have already suggested that Nottingham’s needs were probably satisfied by food rents and renders from surrounding royal estates. Wallingford provides eloquent evidence of the process. In the Domesday account of the borough it is stated that ‘they who dwelt there did service for the king with horses or by water as far as Blewbury, Reading, Sutton Courtenay [and] Benson [Oxon.]’

(GDB, 56: DB Berks, B1). The passage surely indicates the provisioning of the garrison and town from the surrounding royal manors. Do we see here the mainspring of Wallingford’s economy? If we are looking for a mercantile twin, then Oxford is the obvious candidate. Its links with Wallingford are not as pronounced as Nottingham’s with Derby, but are nevertheless there. As far as I know, there is no evidence of overlapping toll boundaries. But both boroughs were subject to the same administration from the time of the Conquest. Robert d’Oilly built both castles and was presumably constable of both until Miles Crispin, his sonin-law, was appointed to Wallingford in the 1080s (KeatsRohan 2009). In the 12th and 13th centuries Oxfordshire and Berkshire shared a sheriff and the county court met on the boundary between the two at the southern end of the causeway at Oxford (Roffe 2009, 42–3). Of earlier links there is no explicit evidence. The construction of the castle can have had little impact on this putative tributary economy. From 1067 to 1153 when it was surrendered to Henry II, it may have been relatively intensively used by its lords, but none would have been in permanent residence. Thereafter, royal visitors were of necessity infrequent. All the factors that divorced Nottingham Castle from the borough of Nottingham appear to have applied with equal force to Wallingford. Analysis of the building accounts of from the early 14th century onwards show that expertise was organized by the King’s Works and materials were largely sourced outside the town, mainly from the honour of Wallingford, the duchy of Cornwall, and the king’s demesne. From time to time Wallingford Castle was provisioned by townsmen, with vintners figuring prominently and it provided jobs for labourers. But it was essentially a service economy, dependent on the whims and accidents of aristocratic residence, that it fostered (Keats-Rohan above, p.102–3). I would suggest, then, that our initial conundrum – the apparent rapid decline of Wallingford as it emerges into the light of record – is an illusion. The level of economic activity of the 13th and 14th centuries was probably not much different in kind from that of the first 300 years of its history. Wallingford was not an urban community if we mean by that a society driven by a self-sustaining industrial and mercantile economy. Its mainspring was otherwise. It was the demands of its military role that dictated its fortunes and, paradoxically, in those terms it may well have been the construction of the castle that impinged upon it most adversely. The defence of the burh brought all manner of free men into Wallingford before the Conquest. Castle-guard, by contrast, was confined to a military elite. It is probably no coincidence that the urban appurtenances of rural manors became little more than a source of income in the 100 year after the Conquest. It seems clear to me that the fact of neither a borough nor a castle guaranteed urban development. A type-site for the 9th- and 10th-century burh? Royal power was at once the making of Wallingford and its unmaking What, then, can it tell us about the origins of 198

the burh as an institution and its relationship to processes of urbanization? Probably quite a lot. Richard Holt has recently argued that in origin all boroughs were fortresses largely occupied by an aristocratic elite and it was not until the late 10th or early 11th century that they became towns (Holt 2009). By ‘all’ I take him to mean the boroughs of Wessex and western Mercia – the Danelaw towns were, of course, thriving mercantile communities by 900 or so – and there it might be doubted that urbanization is so late in some of them. The evidence for late 9th-century planning at Winchester remains robust (Biddle and Hill 1971) and, lest that be seen as the precocious development of a capital, the late development of Holt’s type-site of Worcester is open to question. There it has been argued that the primary planning of the hagae on the east side of High Street postdated the infilling of the burh ditch to the east, indicating that urbanization was a secondary phase of the late 10th century at the earliest. However, this conclusion, drawn from questionable topographical evidence (Haslam 2012), is belied by an authentic charter of 904 granting a haga within the burh to Ealdorman Æthelred and Æthelflæda, Lady of the Mercians (Sawyer, no 1280). Planning of some kind was clearly more or less contemporary with the documented foundation of the burh in the last decade of the 9th century. If Wallingford remained essentially a fortress, then it has a better claim than Worcester to being a type-site for the burh of the 9th century. Its experience might suggest that there was no simple dichotomy of fortification and urbanization in the late 9th century. Zoning was apparently an early feature of the site (Roffe 2009; Creighton et al. 2009). The north-east quadrant of the borough, and probably also the north-west, were high status areas from the beginning. The continuing authority associated with them discouraged intensive development and ensured that they have survived until today as more or less open spaces. By analogy, the south-west quadrant, the Kinecroft area, could also have been set aside for aristocratic use: it is perhaps one of the likelier sites for Wigod’s hall and the earl’s manor before him (Roffe above, p.32). Alternatively, it may have originally been planned as an open area for impounding livestock and the like (Creighton et al 2009, 74). The fourth quadrant, by contrast, seems to have been assigned to the borough proper as it were. Here were to be found the non-customary ‘contributory’ tenements that belonged to rural manors in the surrounding area. Here too were the majority of the custom-paying burgesses, as eloquently illustrated by the distribution of landgabel and quitrents (Pedgley below, p. 227–9). Incorporating the probably pre-burghal settlement around the South Gate, this mix looks neither accidental nor organic. The borough with its market has every appearance of being an original element in the layout of the burh.

Zoning of a similar kind has been suggested for both London and Stafford at much the same time (Vince 1990, 20; Creighton et al. 2009, 74–5; Carver 2010). Wallingford may well represent the best model of early burh development. The detail is telling. The axial roads – High Street, Castle Street, Market Place – are as much boundaries as means of communication. Likewise, Cheapside in London and High Street in Stafford demarked different zones of activity. It follows, then, that the sampling of evidence on one side of a road, as in the High Street in Worcester, will not necessarily characterize the other side, much less the whole of the borough. Clearly, the burh was a centre of authority and at its foundation aristocratic concerns have may predominated. But that did not preclude more localized industry and trade. Athelstan stipulated in the early 10th century that all buying and selling must be conducted in a borough (EHD I, no 35, c,12–14). There must have been markets there for this legislation, even if only an aspiration, to have meaning. It is, then, the more significant that the only detailed description of the construction of a burh, the foundation charter of Worcester, notices the division of market rights (Sawyer, no 223). Whether this indicates an existing market on the site, an old one moved into the burh, or an entirely new one is unclear. But evidently the parties to the agreement, Ealdorman Æthelred and Æthelflaeda on the one part and Bishop Wærferth of Worcester on the other, were as alive to the advantages of protecting exchange within a fortification as Athelstan after them. Theirs was probably no conscious programme of urbanization; presumably the immediate aim was merely to secure royal interests and assets at a time of uncertainty. More widely, however, the funnelling of the service of all free men into the burh for the first time must have increased the returns and necessitated arrangements for their collection (Abels 1988, 58–78). The provision for exchange within the burh can be seen as yet another corollary of the transformation of royal rights at the time under the pressure of invasion. The opposition of trade to fortification is a meaningless dichotomy in the context of the late 9th and early 10th centuries. Far from precluding a market, the command economy of Wallingford, as probably that of Nottingham, presupposed one. Fortresses needed to exchange, the more so since they embodied new relations between king and subject in the late 9th century. The construction of the burh thus provided the infrastructure for urbanization. Elsewhere the opportunity fostered growth: the progressive monetarization of society in the 10th century was both cause and effect. Wallingford cannot have been immune to these new forces in society, but tribute continued to be the mainspring of its economy. By the 13th century it was beginning to look like an anachronism and it paid the price.

199

9 Recent work on Oxford Castle: new finds and new interpretations Andrew Norton

Abstract Between 1999 and 2009 Oxford Archaeology (OA) carried out a programme of archaeological evaluation, excavation and watching brief work at the site of Oxford Castle and the Paradise Street Business Centre (centred NGR SP 510 061). The work was carried out in advance of the development of the two sites by Oxford Castle Ltd and Oxfordshire County Council, and by St Peter’s College, and also included works on the castle motte following a landslip in 2007 due to heavy rain. As part of a more general research objective and so as to present a coherent picture of the castle mound to the wider public, Oxford Preservation Trust (OPT) also commissioned a geophysical survey and excavation at the summit of the motte in the summer of 2009. The main areas of excavation were located north and south of the retained prison buildings (see Figure 9.1 Areas A, B and C), but many of the most significant discoveries were made during a long-running programme of watching brief work on service trenches and groundworks. The work revealed evidence for occupation of the site during the late Saxon period, comprising timber-framed halls, cellar pits and possible street surfaces. Significant remains of the Norman castle were also unearthed, dating from the time of the castle’s foundation in 1071 through to the end of the medieval period. The following summary is based on the paper given at the Wallingford Castle in Context conference, and a full discussion of the results will be published in due course. Introduction The medieval town of Oxford was built on a spur of the Summertown-Radley gravel terrace between the river Thames to the west and its tributary the river Cherwell to the east. Oxford Castle was built at the western end of the medieval town, overlooking the Thames floodplain to the west and south. The castle site slopes gently from north to south; at the north the ground level is at c. 61.7m OD, and in the south it is at c. 60.5m OD. Paradise St, to the south, lies at c. 57m OD. The investigations hoped to establish the survival of late Saxon occupation evidence and clarify the layout and extent of the western end of the late Saxon town (or burh), thought to have been constructed around 900. The burh’s western limits were unknown, having been remodelled and obscured following the construction of the motte and bailey castle by Robert d’Oilly in 1071. The construction of Oxford Prison at the end of the 18th century led to the demolition of most of the surviving castle buildings, and the recent investigations gave scope to refine our

understanding of the layout, history and function of the castle. The late Saxon town, 900–1071 Burghal defences Until the closure of Oxford Prison in 1997, opportunities for excavation at the west end of the burh had been very limited, and the form and location of any late Saxon defences here remained unknown. The castle excavations confirmed that the late Saxon defences extended to this edge of the burh, and added to evidence recovered in previous work for occupation of the area and for the presence of possible street surfaces. Large earthen ramparts were observed in at least three locations along the southern part of the castle site measuring up to 1.2m high and c. 15m wide. The defences were constructed directly upon natural gravel implying that the line of the defences must have been deturfed prior to their construction, possibly to define the construction area for the earthworks and provide turves for their construction. The ramparts (Figure 9.1) were constructed from redeposited gravel, brickearth, topsoil and turves. The primary deposits were layered, with turf lines evident; micromorphological analysis of the deposits indicated that the turves were rich in dung, and deposits in keeping with a dung-rich trackway were evident. The turves were removed from pastoral land, possibly the area immediately to the south of a mid-AngloSaxon hall or barn which was also partially revealed during the works. The upper deposits were later in date and formed from bulk soil dumps with no evidence of turves. This material may have been upcast from re-digging of the ditch following the construction of a ragstone retaining wall, or used to strengthen the ramparts in the 11th century. The ragstone retaining wall was constructed against the rampart’s southern (outer) face; its core was formed of rubble, with facing stones on its southern/outer side. The presence of three sherds of medieval Oxford ware (OXY) in the primary fill of the construction cut suggests that this pottery was in use rather earlier than its conventional start date of c. 1075, and possibly from around the middle of the 11th century. St George’s Tower and a cemetery St George’s Tower is located at the western edge of the castle site, overlooking the Castle Mill stream beyond. Although it has always been assumed to be of SaxoNorman date, its origins have always been uncertain. St

200

Figure 9.1 Areas of excavation at Oxford Castle showing features of late Saxon date. The author’s conjectural reconstruction of the street plan of the western part of the burh is shown in inset and overlay.

George’s Tower differs markedly from the tower of St Michael’s at the Northgate, which John Blair dates to the first half of the 11th century (Blair 2003a). It was thought that St George’s Tower may have formed part of the original castle constructed in 1071, as the first structure to be built in stone at Norman castles was often a tower

at ground level due to the instability of the motte (Renn 1968, 10). It is also feasible that the tower was constructed by Norman engineers prior to the Conquest. Late AngloSaxon earls had halls in certain towns, which acted as a base of administration and a place to hold their treasury (Baxter 2007, 102). Comital manors or complexes were

201

Street 8185

Structure C2 Structure C3

Figure 9.2 Late Anglo-Saxon halls C2 and C3

set aside by the king and could be transferred between earls with relative ease (ibid., 13, 141). King Edward assigned control of Oxfordshire to his nephew, Earl Ralph, in the 1050s and it is possible that Ralph built a comital complex around St George’s Tower, which was subsequently taken over by Robert d’Oilly (ibid., 68 and 102). A Saxon date for the tower is supported by the presence of two late Anglo-Saxon burials found within a service trench on the southern side of St George’s Tower (Figure 1). The skeletons were cut into the late Anglo-Saxon rampart and were radiocarbon dated to the period cal AD 983–1021 and 992–1036 (NZA 32288 and NZA 32287, 68% confidence level). Quite possibly these burials lay within a graveyard located close to a church or chapel, and it is reasonable to assume that the church of St George in the Castle was located on the site of a late Anglo-Saxon predecessor or added to an existing structure.

of topographical factors as Area C is low-lying and any deep pits would have filled with water. The most complete structure (C2) seems to have formed the western end of a large hall (Figure 9.2), with an annexe to the west. A set of postholes was located in its western end and was indicative of an upper storey or hayloft. Postholes at the south-east corner of the structure were rich in chaff and culm nodes, and straw must have been present within the building. The straw may support the evidence for a hayloft, but alternatively may have been used as flooring or bedding within the structure. An extension or phase of rebuilding was evident on the north-eastern side of structure C3 and was cut through a metalled surface (8185). Only a limited area of the surface survived, and it may have been part of a yard, although it was of identical construction to the first late Saxon street surfaces widely observed elsewhere within the burh.

Post-built halls

Cellar pits

At least three 10th- or 11th-century post-built structures fronting a possible street were revealed within the southern part of the castle site (Area C), directly to the north of the rampart. The structures were constructed from earthfast posts and probable beamslots, although it is possible that some of the beamslots were trenches for closely set posts. There is limited evidence of non-cellared structures within Oxford, and the few examples that are known were constructed from post-and-mud or cob walls, and are generally thought to date to the 10th century (Dodd 2003, 40). Potentially such structures represent an earlier form of urban building at Oxford, prior to the general introduction of cellars, which are widespread by the early 11th century. However, the absence of cellars might also be the result

No evidence for above ground structures was revealed within the northern part of the site, but three flat-bottomed vertically-sided pits were revealed within Area A, and three similar pits were observed in the eastern part of the site (Figure 9.1). The pits measured up to 1.4m deep, 3.4m long and structural features (beamslots and postholes), and lime or crushed chalk surfaces were evident at their bases. The pits formed 10th- and 11th-century cellars that once lay below timber framed structures. All the pits had been rapidly infilled, unlike the cess/refuse pits that were revealed to the rear of the cellars, and although some cellars may have been infilled in advance of altering or rebuilding existing properties, any open cellars would have been infilled in advance of the construction of the 202

castle. In earlier excavations, Tom Hassall also recorded a sunken hut to the south-east of the Area A cellar pits (see Figure 9.1), and an associated pit contained a coin of Eadred (948–954; Hassall 1976, 249). Cellar pits seem to be a short-lived characteristic of urban building tradition in 10th- and 11th-century contexts throughout the country (Dodd 2003, 35). Ranging in size from a few metres square to substantial rooms comparable to later medieval undercrofts, they were lined with wattle and daub or walls constructed from planks and posts (Dodd 2003, 35; Sturdy and Munby 1985, 92–4). In general the pits are likely to have formed storage cellars below larger structures, with the smaller pits forming temporary accommodation, workshops or storage areas for perishable foods (Horsman 1988, 69–70; Sturdy and Munby 1985, 93). Material evidence in the settlement There was little evidence of industry amongst the material remains, but fragments of ovens or kilns and annular loomweights give us some insight into life in the area. There was evidence for spatial separation of industry, with the majority of fired clay oven furniture recovered from the northern part of the site, and the majority of loomweights recovered from posthole fills and soil layers within the southern part of the site. The medieval pottery assemblage is one of the largest ever excavated in Oxford. The 10thand 11th-century pottery shows an initial dominance of St Neot’s-type ware, giving way in the 11th century to Cotswolds-type ware, which came into use from the late 10th century. The 10th-century pottery assemblage is also notable for the presence of a group of Stamford ware, some of which dates from the earliest years of the burh at Oxford, in the late 9th or early 10th century. Conversely, there was relatively little pottery in late Saxon Shelly ware (OXB), which is the dominant pottery tradition of this period at sites in the centre and southern part of the burh. Maureen Mellor previously observed this pattern at the castle, and noted that 11th-century sites in the central streets of the burh tended to favour either St. Neot’s-type or Cotswolds wares (Mellor 2003, 342–3). She suggested that this could reflect the preferences of Danelaw settlers in the town, who used St Neot’s-type vessels and were first living on the periphery, subsequently moving into more central sites perhaps after the accession of the Dane Cnut to the throne of England in 1016. In his analysis of the pottery from the recent Oxford Castle excavations, Paul Blinkhorn (pers comm.) suggests that the new evidence supports this view of cultural differences, but that the notable links with the East Midlands in these assemblages may reflect the presence of Anglo-Saxons of Mercian origin rather than a community of Danes. The 11th-century pottery assemblage at the castle was dominated by Cotswold-type wares, with a much reduced proportion of St Neot’s-type ware. This might point towards a more cosmopolitan western quarter by that time. The town plan The castle excavations have added considerably to our understanding of the western quarter of Anglo-Saxon

Oxford, although the areas excavated remain small in proportion to the overall extent of the former prison, much of which has been retained in the recent redevelopment. Moreover, the obliteration of the late Saxon topography of the area by the castle means that there is not the continuity of streets or building plots that is demonstrable elsewhere in the city centre to provide a contextual framework for the interpretation of excavated results of this period. However, some evidence for the possible position of late Saxon streets has been recovered, and the inset to Figure 9.1 shows the author’s suggested integration of these into the known late Saxon street network of the burh. The metalled surface observed in Area C (8185, see above) was likely to have been up to 6m wide, and was constructed of a dense layer of small rounded stones, identical in form to the earliest street surfaces of Oxford observed elsewhere in the city centre. Unlike these latter, however, there was no evidence for later resurfacing at the castle. This raises the possibility that surface 8185 may have been part of a street running along the rear of the rampart, but which was less heavily used than streets in the centre of the burh. The location of another possible street might be inferred in the north of the site (Area A), based on the positions and alignment of the cess/refuse pits and cellar pits. This coincided with a row of 18th-century gravel pits, which, unsurprisingly, were targeted on areas where the natural gravel remained undisturbed, and avoided areas already disturbed by Anglo-Saxon and medieval pitting. A street in this location could have formed part of a southwest to north-east aligned route linking New Inn Hall St with the western edge of the town; interestingly this would represent a continuation of the line of Shoe Lane (Saewig’s Lane), which may have formed part of the burh’s original street system (Blair 1994, 155–56), although this has not been confirmed by excavation. No Anglo-Saxon remains survived in the centre of Area A, which had been truncated by a much larger quarry, which was less discriminatory in seeking areas of undisturbed gravel. Queen Street/Castle Street was the main axial eastwest street through the western half of the burh, and if it originally extended as far as the western rampart on the same alignment, it would have passed to the south of St George’s Tower and its cemetery. However, a gate at St George’s Tower seems more likely and Castle Street may have curved to the north, mirroring the southern curve of the High Street in the east. If the north-western AngloSaxon town limits were on the same alignment as the postConquest town limits, then St George’s Tower would have been located at the mid-point of the town’s western limits. The gate probably lay to the north of the tower, which was then rebuilt as the castle’s c. 13th-century (possibly earlier) Oseney Gate. The Norman and later medieval castle, 1071–1577 Following the Norman Conquest, the need to control and dominate Anglo-Saxon England led to the construction of earth and timber castles throughout England and Wales. Over 80 such structures were documented prior to 1086 203

(Higham and Barker 1992, 41, Figure 2.1), but is has been conjectured that there may have been as many as 600 castles constructed in the half century following the Conquest (Eales 2003, 47). The defensive complex at Oxford was an early creation, contemporary with nearby Wallingford (Luard 1869, 9): the south-western corner of the burh, located at the edge of the gravel terrace, provided an ideal site for the Norman motte and bailey castle. The motte was constructed immediately adjacent to St George’s Tower, and would have dominated the skyline in the western quarter of the town and provided a demonstration of Norman might to travellers by road and river. The Anglo-Saxon structures across the western part of Oxford, and particularly those in Area A, will have been demolished in 1071 once the site of the motte had been decided. It is likely that at first Oxford Castle comprised only the motte and motte-ditch, and the bailey area would have been largely undefended.

The castle motte The 2008 and 2009 investigations on the castle motte provided an invaluable insight into the mound and its keep. Not including dumped material on the top of the motte, which was added in the 17th century as part of the Civil War refortifications, the volume of the motte has been calculated as c. 27,000m³. Assuming that a person could move 1-2m³ of material a day, a team of 200 people would have completed Oxford Castle in 10 to 20 weeks. If the work was carried out by specialist labourers, such as the latter-day Victorian navvies, the work may have been carried out in only a month. Oxford’s motte was constructed by means of dumped layers of gravel and stabilizing clay deposits. Steps were cut into the deposits to enable the ‘keying in’ of subsequent dumps of material, particularly a thick deposit of Oxford Clay that sealed all the motte deposits. The

Figure 9.3 Revealed foundations of the ten-sided keep

204

Figure 9.4 Aubrey’s 1663 sketch of the keep and its crack

stepped construction techniques employed at Oxford are uncommon; more typical is Norwich Castle, where the motte was constructed from horizontal dumps of material and layers tipping at 45º into the centre of the motte (Shepherd Popescu 2009, 446).

was constructed within an existing stone tower a central location would make sense, and as such it is unlikely that the ten-sided stone keep was constructed before 1174.

The ten-sided stone keep and ‘High Tower’

The defensive ditch around the motte was revealed within the Area A excavations (Figure 9.5) and like the motte it was also seen to be stepped, presumably to avoid the slumping of the gravel during construction. The base of the motte ditch/bailey moat lay at c. 5.5m below ground level and the width of the channels was at least 30m. It is unclear whether the motte ditch fully encircled the motte as a gatehouse was constructed over its assumed line to the south (see below and Figure 9.5). The recent excavations and borehole data demonstrate that the motte must have formed a penannular ditch at the very least, and environmental evidence indicates that any water was slow moving or still. There was no evidence for scouring out or recutting of the primary fills of the motte ditch, and the earliest artefactual evidence indicates that the ditch was used as a dumping ground from the 11th century onwards. The motte ditch would have provided material for the construction of the motte, and would have served as an additional defence whilst the ditch and ramparts around the bailey were constructed; interestingly, a build-up of a late 11th-century soil over the Anglo-Saxon occupation evidence in the south of the site hinted at a significant lag between the levelling of Saxon structures and the construction of and infilling of the castle bailey.

The castle’s wooden fortifications were rebuilt in stone during the 12th century and an assumed wooden structure on the motte was replaced by a ten-sided stone keep. Three sides of the ten-sided stone keep were revealed (Figure 9.3), which closely matched an 18th-century plan of the remains of the tower excavated by Daniel Harris (King 1796, Pl. CXXVI F.2). A large void was revealed on the south-eastern side of the foundations, which was probably the base of a large crack depicted on one side of the tower in a sketch by John Aubrey (Figure 9.4), drawn from memory in 1663 after the tower had been demolished in the Civil War. Harris seems also to have found an inner decagonal tower, as described by King in a somewhat confused manner (Figure 9.3), and recent excavations have revealed a probable wall robber trench to the north of the southern section of keep wall. Although not depicted on any historical drawings, it seems possible that an inner ‘High Tower’ was constructed within a ten-sided perimeter wall. This could have been similar to the circular high tower within a shell-keep at Launceston Castle, Cornwall, which was a late 13th-century addition, and assumed to be constructed as an expression of enhanced status (Saunders 2006, 457). In whatever form it was built, the shell-keep was very likely the ‘Great Tower’ described as ‘almost completely ruinous’ by 1327 with the usual exaggeration of such reports (below, 211; cf. HKW ii, 774). A stone well-chamber (still surviving today) lay beneath the stone keep and is thought to date from 1173–4 (ibid., 772). The well-chamber was seen to be off-set from the centre of the keep and it is most likely that the chamber was added to an earlier wooden keep and replaced an earlier well accessed from the top of the motte. If the chamber

The motte ditch

The castle’s curtain wall and northern tower At around the time of the construction of the stone keep, the castle curtain wall, towers and gates would also have been refortified. Based on cartographic and documentary evidence Oxford castle is generally conjectured to have three square towers along its northern extent, a round tower to the south, the extant St George’s Tower to the west, and fortified gates and barbicans to the west and east. Part 205

Figure 9.5 The revealed castle structures

of the northern circuit of curtain wall leading down into the motte ditch was revealed within Area A, and a square stone tower base (6304) abutted its inner face (Figure 9.5). For the most part the curtain wall and tower bases were removed during post-medieval robbing and quarrying, as they were at ground level or carried by the earth ramparts around the circuit of the castle. The revealed tower base was at least 4 x 2m wide and is likely to have originally straddled the curtain wall. It is possible that originally the tower only abutted the inner face of the curtain wall, much like the west gate (see below), but a width of 2m seems remarkably narrow for such a structure. The remainder of the northern circuit of the castle wall and its towers remains conjectural, aside from Agas’ plan and references to the ‘bastions’ being discovered during the construction of New Road (Munby and Walton 1990, 123–130).

Ramparts The earthen ramparts circuiting the bailey covered an area measuring c. 120m north to south and 90m east to west (Figure 9.4). The excavations demonstrated that, like the motte, the ramparts were also constructed by means of stepping dumped clay, gravel and silts arising from the excavation of the bailey ditch. The ramparts were seen in their entirety within Areas B and C, in the southern part of the site, and were up to 25m wide and 2.3m high at this point (Figure 9.6). Unlike ramparts seen at Launceston (Saunders 2006, 86) and Therfield Castle in Hertfordshire (Kenyon 1990, 31–2), no evidence of the revetment of the rampart’s face was seen. Within Area B the clay capping was seen to be at least 1.5m thick and the bonding nature of the Oxford Clay 206

Figure 9.6 The Norman ramparts to the south of C-Wing, looking north

may have made additional revetting unnecessary. Within many parts of the rampart the clay had the appearance of a single deposit, although it is likely to have been laid by the shovel or cart load. Prior to the formation of a topsoil the clay rampart would have formed a steep slippery surface, posing an additional problem to any would-be attackers.

part of the medieval tower’s foundations (Figure 9.5). Its foundations to the south were most likely above the level of the investigation area, within the upper part of the southern rampart.

East Gate, bridge and round tower

The curtain wall (5745) was also recorded within the Tidmarsh Lane excavations, between St George’s Tower and the motte, and measured over 12m in length. The northern limits of the curtain wall were cut into the base of the motte, and Agas’ 16th-century plan of the site shows the wall abutting the ten-sided stone keep on top of the motte. The foundations and upper 2m of two square (c. 5m²) gate houses (5613 and 5746) abutted the inner face of the curtain wall, and most likely formed the western ‘Osney’ Gate. The gap between the two structures was 2.5m and a rebate for a draw bar was revealed in the southern section of curtain wall; a rebate would suggest that the gate comprised swinging doors rather than a portcullis (Renn 1968, 72). The lower part of the gate must have been obscured by the curtain wall, and any wall walk stepped back to access the tower. The gate had gone out of use by the 17th century and a robber trench and newly constructed wall and road ran over the curtain wall at this point. An entrance only 2.5m wide does seem a little narrow compared to modern entrance gates, but is comparable to that at Arundel (see below); the width of the pier base and bridge abutment forming part of the castle’s eastern gate also indicate an entrance

The East Gate and bridge are shown in some detail on the Christ Church plan of 1615, and the gate is described by King as originally being the castle’s postern and only becoming the principal entrance following the decay of the Osney Gate to the west (King 1796, 27). The western limit of a bridge abutment (9182) was revealed at the top of the bailey moat within the prison’s B-Wing site, and a single stone pier base (9218) was revealed below the B-Wing foundations (Figure 9.5). Although much of the pier base (9218) remained below the level of excavation, it can be assumed that the stone structure was supported by sill beams or timber. The bridge abutment can be seen on Pridden’s 1785 south-east view of the castle and a series of six piers are shown on the Christ Church plan of 1615. It was previously thought that the round tower at the southern end of C-Wing was the re-built medieval round tower, but excavations against the tower’s base on the southern side showed it to be a 19th-century construction with no re-used medieval foundations. A test pit (6) within the tower revealed a layer of limestone and coral ragstone rubble beneath the tower’s floor, which may have formed

The west gate

207

of approximately 2m width. The dating evidence for the construction of the castle walls and western gatehouse was inconclusive, but Edward King produced a conjectural plan of the castle and describes the tower at Osney Bridge and Gate as ‘a square Tower nearly of the same construction with the tower of entrance that still remains at Arundel Castle’ (King 1796, 27). There are two types of gatehouse at Arundel, and King could be referring to either the 13thor 14th-century barbican or else a Norman gate tower. King also describes a covered way and series of steps leading from the gate tower to the keep tower, the entrance of which was a narrow door in ‘nearly in the same manner as at Arundel’ (King 1796, 27). He describes the door as probably to the east of the southern gatehouse and it is clear that he is not describing an intact structure. Arundel’s keep tower is linked to the Norman gateway, rather than to the barbican, and if King’s ‘tower of entrance’ and ‘gate tower’ are the same structure, then the Osney Gate must had closer parallels with Arundel’s Norman gateway. However, King may not have visited Arundel at all and his observations may be based on a 1789 survey of Arundel, published seven years prior to King’s Vestiges of Oxford Castle. Arundel’s elevation and plan were some of the earliest accurate drawings of a castle, and King would have been keen to reference these drawings, if only to demonstrate his knowledge of the latest castle publications. There is very limited cartographic evidence for a western gatehouse. Several artists depict a simple arched entrance within the curtain wall at the foot of St George’s Tower but do not show a gatehouse proper. A 1615 Christ Church plan does depict a two-storey structure at the foot of the motte, next to the arched entrance and, although unclear, it appears to be closer in form to Arundel’s Norman gatehouse. Other Castle Buildings The bailey was probably crowded with buildings in similar fashion to other sites such as Hen Domen, but, largely because of the post-medieval quarrying, only two halls and the church of St George in the castle survived. The footprint of a rectangular north-west to south-east aligned stone structure, measuring over 13m wide, was recorded during a watching brief immediately to the west of County Hall. For the most part excavation took place to the base of the relevant ground works, and unfortunately the function of the structure (or structures) revealed could not be fully understood. However, the exposed walls were centrally placed within the medieval castle and dated by the recovered pottery to the early post-Conquest period. The date, size and location of the walls strongly indicate an important medieval hall, most likely the Shire Hall/ Sessions House (Figure 9.5). A stone-built cellar was recorded within the south of the site, and was most likely located beneath one of the castle’s halls or chambers; the cellar had been infilled by the 14th century. The foundations of the church of St George in the castle were revealed beneath the prison’s D-Wing in the west of the site (Figure 9.5). The chapel or church of St George was founded on the east side of the castle or tower in 1074,

where it was made a collegiate church with four or six secular canons (VCH Oxon 5, 19, 268–9). The graveyard associated with the church of St George was located along the northern and southern sides of the church, and formed a continuation of the pre-Conquest graveyard (see above). The remains recorded during the recent excavations represent a small sample of the individuals buried in this area of the site, and the small size of the watching brief trenches makes it difficult to make assumptions about the graveyard as a whole. Up to 12 graves were recorded either side of D-Wing. In the late 18th century, Edward King also described a vast number of human skeletons buried beneath the church and two stone coffins buried to the south of the church (King 1796, 18–19). Radiocarbon dating of four of the recently discovered individuals all produced date ranges from the early 11th century to the mid-12th century. The date ranges show that the graveyard continued in use after the construction of the castle but may have gone out of use by 1160, or at the very least the area adjacent to the church was full by this time. The graves appeared to be placed in straight rows, and laid shoulder to shoulder, but the prevalence of juvenile burials on both sides of the church has distorted that arrangement somewhat. In Anglo-Saxon Christian cemeteries children were buried under the eaves of churches so that the water running off the ‘holy’ roof would bless them, or double bless them (Daniell 1997, 128). Conclusions The excavations at Oxford Castle have revealed the best preserved and largest area of Anglo-Saxon occupation yet studied in Oxford. While there are still many unanswered questions, it is clear that the site was home to an agricultural community prior to the foundation of the burh, probably forming part of a wider settlement around the minster to the east in the 7th or 8th centuries. The fields around the settlement were incorporated into a defended town, most likely constructed at the turn of the 10th century. The excavations have proved that the late Saxon rampart at the western edge of the burh had two distinct phases of construction. The results of the excavations have also strengthened the case for dating St George’s Tower to the late Saxon period. Interesting questions about the cultural background of the burh’s 10th-century inhabitants are raised by contrasts in the pottery assemblages from the castle and from other areas of the city centre. The construction of the castle in 1071 swept away most elements of the western Saxon townscape. Despite postmedieval quarrying of the bailey area, the results of the excavations have provided clear and valuable evidence for the layout of the stone castle, with the locations of all but two assumed towers now known. The castle was largely utilitarian in function during its early years, but the archaeological evidence demonstrates that it became a thriving and active site. Traders and artisans would have operated within the bailey, while the custodian and his family lived and worked within the castle halls and held feasts for important visitors. 208

The castle structures were maintained until the 15th century but gradually fell into disrepair, and it appears that the site was largely abandoned in 1577 following an outbreak of gaol fever (VCH Oxon 4, 296–300, note 31).

Acknowledgements Many thanks are extended to Anne Dodd for editing this paper and to Sarah Lucas who produced the figures. Great thanks are also extended to all the specialists involved with the excavation projects, whose reports will be produced in full in the forthcoming monograph from Oxford Archaeology.

209

10 OXFORD CASTLE IN 1327: INQUISITION AS TO THE STATE OF CASTLES IN OXON AND BERKS 1 EDWARD III, TNA, E143/10/2 K. S. B. Keats-Rohan

Figure 10.1 Survey of Oxford Castle, 1327. (Reproduced by permission of the UK National Archives, E143/10/2.)

210

Inquest taken at Oxford on Monday on the Feast of St Matthew Apostle [24 February] in the first year of the reign of King Edward the Third after the Conquest, in the presence of Ambrose de Newburgh assignee, as to all defects of all castles now in the hands of the lady Queen or of the lord King in the counties of Oxfordshire and Berkshire, by oath of John de Gonwardly, Stephen de Wolleford, Walter de Fynestoke, Richard de Briggewater, Thomas the mason, Thomas de Curtlyngton, Laurence the mason, Geoffrey atte Cuchene, Adam the plumber, Henry the slater, Michael of Wolleford and Henry the plumber. They say on their oath that the mantlet wall before the great gate needs to be wholly repaired from scratch and they estimate the cost of the same, with purchase of stone, chalk and sand, and labourers’ wages and all other necessities attaching to the same, at £400. Item, they say that that £4 is required for stone, chalk and sand to repair the turret over the main gate, and other necessities, plus wages for the workmen, required for repair of the stairs. And they say also that upwards of 20s for stone, chalk and sand, plus labourers’ wages are needed for repair of the wall before the Jews’ Tower. And they say that £6 is needed for stone, chalk and sand, with carriage and labourers’ wages, to repair the wall of the Chamber in the same place. And they say that 46s 8d is needed to make two news doors for the hall of the said castle, and for repair of the wall of the hall, in respect of stone, chalk and sand and other necessities and the labourers’ wages. And they say that the west gate of the said castle is wholly ruinous and wants for repair in stone, chalk and sand and all other necessities and workers’ wages, £300. They also say that the Great Tower of the same castle is almost totally ruinous and wants for repairs up to 100 marks in stone, chalk and sand and all other necessities and workers’ wages. And they say that the wall beyond the great gate and another wall next the same needs 10 marks for repairs in sand, stone and chalk and other necessities and carriage of the same and labourers’ wages. They also say that a stone buttress supporting the wall ought to be renewed and two other buttresses repaired, and the cost of the making and repair of the said buttresses in stone, chalk and sand with the labourers’ wages is 4 marks. And they say that the stable and the bakehouse of the same castle want repairs in divers places costing 60s in stone, chalk and sand and labourers’ wages. And they say the ?tablement of the wall (gradus mur’) of the castle requires repairs costing 13s 4d in stone, chalk and sand and labourers’ wages. And they say the great bridge of the said castle needs to be wholly repaired from scratch costing 200 marks in timber, nails and other necessities and workmen’s wages. And they say that the east gate wants for repairs in timber, iron and other necessities and workmen’s wages up to the sum of 40s. And they say that the stable of the same castle wants for repairs, in timber, nails and other necessities and carpenters’ wages, up to the sum of 5 marks. They say also that hall of the same castle wants for repairs in timber, nails, and other necessities and carpenters’ wages up to the sum of 10 marks. And they say that the Great Chamber of the same castle needs 60 marks

for repairs in timber, nails and carpenters’ wages. They also say that two other chambers next the same Chamber want for repair in timber, nails and carpenters’ wages up to the sum of 6 marks. And they say that the west gate needs, for carpentry, nails and other necessities, with the purchase of timber, up to the sum of 60 marks. They say also that the Jews’ Tower wants for repairs timber, nails and other necessities with the workmen’s wages, up to the sum of 100s. And they say that the Jews’ Tower needs for repair as to the covering of lead and other necessaries up to the sum of 30s. They also say that the gutters of the hall, Chambers and Kitchen need repair to the leading with workmen’s wages, up to the sum of 20s. And they say that the East Gate needs for repairs to the lead coverings with workmen’s wages up to the sum of 20s. They also say that the new hall of the said castle wants for repair in respect of roofing, slates, lathes, tilepins, chalk, sand and other necessities and the worker’s wages, up to the sum of 40s. And they say that four chambers next the said Hall want for repairs to roofing with slates, chalk, sand, tilepins and other necessities with the labourers’ wages up to the sum of 40s. They say that the stable of the same castle wants for repairs to the roofing with roofing stone, tilepins, chalk, sand and other necessities with the workers’ wages up to the sum of 20s, And they say that other houses of the same castle need repair in respect of roofing with slates up to the sum of 40s. They say that the Bridge towards the king’s Hall which is now the hands of the Brothers of Mount Carmel in Oxford at the head of the ditch of the said castle requires to be wholly repaired and renewed and the cost of the repair of the same in stone, chalk, timber, nails and all other necessities, with workers’ wages, is 100s. Sum total of all the aforesaid repairs [amount blank] They say also that all the aforesaid defects found in the said castle are not due to any defect or negligence of any Custodian since they have been notified many times by petition and inquest to the Treasurer and the barons of the Exchequer, and they cannot warrant more fully. Inquisitio capta apud Oxon’ die Lune in festo Sancti Mathei Apostoli anno regni Regis Edward tercij post conquestum primo coram Ambrosio de Novo Burgo assignato ad omnes defectus omnium Castrorum existentium tam in manibus domine Regine quam in manibus domini Regis in Comitates Oxon’ & Berk’ per sacramentum Johannis de Gonwardly Stephani de Wolleford’ Walteri de Fynestoke Ricardi de Briggewater Thome le Mason Thome de Curtlyngton Laurencij le Mason Galfridi atte Cuchene Ade le Plomer Henrici le Sclattere Michaelis de Wolleford & Henrici le Plomer. Castrum Oxonie Qui dicunt per sacramentum suum quod mantellum ante magnam portam oportet omnino de novo reperari & estimatur reparacio eiusdem cum empcione petri calce sabulonis stipendiis operantium & omnibus aliis neccessariis ad eundem spectantibus ad cccc li. Item dicunt quod turellum ultra capitalem portam indiget reparacione videlicet in petra calce sabulone & aliis neccessariis pro gradibus emendandis cum stipendiis operantium de iiij li. Dicunt etiam quod murus ante Turrem Judeorum indiget reperacione ut in petra calce sabulone cum stipendiis operantium de xx s. Et dicunt quod murus Camere ibidem indiget reperacione ut in petra calce sabulone arena cum

211

cariagio & stipendiis operantium de vj li. Dicunt etiam quod aula dicti castri indiget reperacione pro duobus hostiis de novo fac’ ?et muro dicte aule reperand’ ut in petra calce sabulone et omnibus aliis neccessariis ad idem cum stipendiis operantium de xlvj s viij d. Et dicunt quod porta occidentalis eiusdem Castri est tota ruinosa & indiget reperacione ut in petra calce sabulone & omnibus aliis neccessariis ad idem cum stipendiis operantium de ccc li. Dicunt etiam quod magna Turre eiusdem Castri est fere tota ruinosa & indiget reperacione ut in petra calce sabulone & omnibus aliis neccessariis ad idem cum stipendiis operantium de c marcis. Et dicunt quod murus ultra capitalem portam & alius murus iuxta eandem indigent reperacione ut in petra calce sabulone & omnibus aliis neccessariis cum cariagio eorundem & stipendiis operantium de x marcis. Dicunt etiam quod oportet de novo faceri unum boteracium petri ad sustinendum muri eiusdem Castri & ad emendandum ij aliorum boteraciosorum eiusdem Castri & valent ?factura & emendacione dictorum boteraciorum ut in petra calce sabulone cum stipendiis operantium iiij marcas. Et dicunt quod Stabulum & Pistrina eiusdem Castri indigent reperacione in diversis locis ut in petra calce sabulone cum stipendiis operantium de lx s. Dicunt etiam quod gradus mur’ dicti Castri indigent reperacione ut in petra calce sabulone cum stipendiis operantium de xiij s iiij d. Et dicunt quod magnus pons eiusdem Castri oportet omnino de novo reperari ut meremio clavis & omnibus aliis neccessariis cum stipendiis carpentariorum cc marc’. Dicunt etiam quod porta orientalis indiget reperacione ut in meremio ferro & aliis neccessariis cum stipendiis operantium usque ad summam xl s. Et dicunt quod stabulum eiusdem Castri indiget reperacione ut in meremio clavis & aliis necessariis cum stipendiis carpentariorum usque ad summam v marcarum. Dicunt etiam quod Aula eiusdem castri indiget reperacione ut in maeremio & aliis necessariis cum stipendiis carpentariorum usque ad summam x marcarum. Et dicunt quod magna Camera eiusdem Castri indiget reperacione ut in maeremio & aliis necessariis cum stipendiis carpentariorum lx marcas. Dicunt eciam quod due alie Camere iuxta eandem Cameram indigent

reperacione ut in meremio clavis & aliis neccessariis usque ad summam vj marcarum. Et dicunt quod porta occidentalis indiget ut in carpentaria clavis & aliis neccessariis cum empcione meremii usque ad summam lx marcarum. Dicunt etiam quod Turellus Judeorum indiget reperacione ut in meremio clavis & aliis neccessariis cum stipendiis operantium usque ad summam c s. Et dicunt quod turellus Judeorum indiget reperacione ut in coopertura plumbi & aliis neccessariis usque ad summam xxx s. Dicunt etiam quod Gutter’ aule Camerarum & Coquine indigent reperacione ut in plumbo cum stipendiis operantium usque ad summam xx s. Et dicunt quod porta orientalis indiget reperacione ut in coopertura plumbi cum stipendiis operantium usque ad summam xx s. Dicunt etiam quod nova aula eiusdem Castri indiget reperacione ut in coopertura sclatt’ latth’ tylpynnes calce sabulone & aliis neccessariis cum stipendiis operantium usque ad summam xl s. Et dicunt quod iiijor camere iuxta dictam Aulam indigent reperacione ut in coopertura cum sclatt’ calce sabulone tylpynnes & aliis neccessariis cum stipendiis operantium usque ad summam xl s. Dicunt etiam quod stabulum eiusdem Castri indiget reperacione ut in coopertura cum petra de sclat’ tylpynnes calce sabulone & aliis neccessariis cum stipendiis operantium usque ad summam xx s. Et dicunt quod alij domus eiusdem castri indigent reperacione ut in coopertura cum sclat’ usque summam xl s. Dicunt etiam quod Pons versus aulam Regis que nunc est in manu Fratrum de Monte Carmeli Oxon’ ad capud fossati dicti Castri oportet omnino de novo reperari & valet reperacione eiusdem ut in petra calce meremio clavis & omnibus aliis neccessariis cum stipendiis operantium c s. Summa totalis omnium defectuum predictorum [blank] Dicunt etiam quod omnes predicti defectus in dicto castro inuenti non sunt pro defectu seu negligentia alicuius custodis quia per petitionem et inquisitionem coram Thesaurio et Baronibus de Scaccario returnabil’ sepius significauerunt et Warrantum ad plenum habere non potuerunt.

212

11 THE ORIGINS OF WALLINGFORD PRIORY: THE EVIDENCE OF A MID-TWELFTH-CENTURY TERRIER K. S. B. Keats-Rohan Abstract The chance discovery of a mid-12th century manuscript belonging to Holy Trinity Priory, Wallingford, containing a 17-line terrier on its final verso, permits a re-evaluation of the traditions concerning its origins as a cell of the abbey of St Albans. The text is completely independent of any of the chronicles and other material from St Albans. The priory was clearly a Benedictine reformation of an existing pre-Conquest English secular college, a process possibly begun c. 1086, but mainly pursued during the period 1087–1100, with final consecration and dedication early in the reign of Henry I. The St Albans traditions are largely vindicated, and the work is shown to have been essentially a collaboration between royal officials and the townspeople. The chance discovery, during this book’s lengthy gestation, of an online image from a mid-12th century manuscript, Royal 5 F x, now in the British Library, London, shines new light on the hitherto obscure origins of Holy Trinity Priory in Wallingford. It contains a collection of sermons by Caesarius of Arles and Eusebius, and sundry other texts, including Guitmund of Aversa’s Confessio de Sancte Trinitate. Written in single columns in a competent protogothic littera textualis, displaying classic characteristics of a mid-12th-century date – the e caudata, long final ‘s’ and straight ascender on ‘d’ – it was written by two scribes, the second of whom was responsible for the texts on fol. 153v (Figure 11.1). This unattractive final verso was damaged and subjected to an insensitively stitched repair, but it is nonetheless possible to discern two texts, the first a 17-line terrier of the priory’s property, the second a deed of its prior, Nicholas. The material is unrelated to the content of the manuscript up to and including fol. 153 recto. There is no heading or any gloss on the content of fol. 153 verso. None of the properties associated with the priory in the terrier is linked to a benefactor. The deed of Prior Nicholas appears to make a life grant of land in the parish of St Mary le More, with reversion to the (unknown) beneficiary’s son James, for 3s a year. There can be no doubt that the manuscript belonged to Holy Trinity, Wallingford; although a west country provenance cannot wholly be excluded, it is most likely to have been written at the priory itself. Internal evidence in the terrier text confirms the palaeographical dating to the mid-12th century. It is stated that the priory held the church of Bucklebury, Berks, with its hide and the tithes of the vill payable by both monks and villagers, indicating at least two stages in the grants composing this holding. The interest here is that the church

was granted by St Albans and its priory of Wallingford to Reading Abbey in a deed dated May 1151 x 7 December 1154 (i.e. between the accession of Abbot Robert of St Albans and the death of Abbot Edward of Reading); in return, Reading surrendered its church of Aston’s claim to the parish of Shephall, Herts, the church of which became the right of St Albans, and agreed to pay two marks per annum to Wallingford (RAC ii, 26–9). 1154 would seem to be, therefore, the terminus ante quem for the list, which is thus the earliest extant property list or terrier for the priory. The agreement between St Albans and Reading Abbey about the parish of Shephall is recorded in three separate charters of the same date. In one of them, Prior Nicholas of Wallingford grants the church of Bucklebury to Reading with the assent of the abbot and convent of St Albans (RAC ii, 28–9). This is an important reminder of the status of Wallingford Priory. It was a cell of St Albans and its prior was appointed by the abbot, who could recall him at any time. From the time of its foundation, in the reign of William II, until the early 14th century, any decision in relation to its spiritual or territorial endowment made by the prior and convent had to have the approval of the abbot of St Albans. Nonetheless, it was a distinguished community. Because the number of fully professed monks seems not to have exceeded six, it was semi-conventual, but, even so, it was of sufficient status to have its own great seal (Heale 2004, 83–90). The quality of the religious was such that four of them were elected abbot of the mother house, starting with John of Wallingford (1195–1214). Prior Nicholas himself was elected abbot of Malmesbury in 1183, an office he resigned in 1187; he died in 1205, probably at an advanced age. He appears to have been prior from early in the 1150s until 1183, and to be identical with Master Nicholas of St Albans, one of the abbey’s leading theologians, author of a controversial treatise on the Immaculate Conception, and a Passio Sancti Edmundi, both in a distinctive rhyming prose (Thomas 2002, 66– 7). St Albans was an institution of immense prestige, housing as it did the relics of the country’s first saint, the protomartyr Alban. St Albans’ property and privileges undoubtedly stretched far back into the Anglo-Saxon past, whatever the true status of the documents that purportedly demonstrated grants by King Offa and others (Taylor 1995: Keynes 1993). More importantly, it persuaded the papacy to back its pretensions and by the mid-12th-century it had been extended the protection of the papacy and St Peter, creating the sizeable liberty and archdeaconry of St Albans in Hertfordshire, free of episcopal jurisdiction (Sayers 1971; Crick 2001). With a bull of Alexander III in 1170, St Albans’ exemption was extended to its cells, which were

213

Figure 11.1 Mid-12th-century terrier of Wallingford Priory (© The British Library Board, Royal 5 F X, fol. 153v).

214

to be free of episcopal visitation and correction, with the prior obliged to profess canonical obedience to the bishop only in respect of appropriated churches. This prerogative is not helpful to the historian, since it deprives us of a useful source of information about the priory. There are others, of course, including surviving annual accounts from the priory, as well as the entries in St Albans’ late14th to late-15th-century Catalogus Benefactorum (BL, Cotton Nero D vii), and the earlier and slightly fuller version of its monk, John de Trokelowe (Trokelowe), and its various chronicles. The first part of the Catalogus Benefactorum compiled by Thomas of Walsingham in the late 14th century (completed in 1380), and printed in Dugdale’s Monasticon Anglicanum, will be frequently cited in what follows, occasionally supplemented by the version printed by Riley in 1867 (GASA). The terrier is the sole authentic account of the priory’s early endowment, as will appear, allowing us to re-address some of the problems with early St Albans traditions about its cells and their origins and holdings identified by Mark Hagger. All accounts of the evolution of the priory and its endowment published to date are in error in one way or another, primarily on account of a failure to distinguish between three categories of grant: a grant made to St Albans Abbey, a grant made to Holy Trinity Priory, and a grant or endowment made by St Albans to Holy Trinity. The impartial evidence of the terrier, an aide-mémoire produced c. 1150 at a turning point in the history of Wallingford and perhaps of the priory itself (the appointment of Prior Nicholas?), is therefore worth a full analysis. With all this in mind, let us look in detail at the terrier in Royal 5 F. x, which, unfortunately, is not completely legible. A full transcript and translation of the document is given in the Appendix below. The elements named in the terrier have been numbered in the translation in order to help the reader. Those numbers appear in bold in what follows. The opening statement is important evidence as to the status and nature of the church of Holy Trinity. The statement that the churches of St Mary and St Martin pertained to the church of Holy Trinity (1) confirms the hypothesis of David Roffe (2009, 37) that Holy Trinity was a pre-Conquest secular college reformed by the introduction of Benedictine monks. As such, it evidenced both the concern to maintain continuity of religious life and a concern with pastoral care which have been identified as the principal motivations for the foundation of the English (i.e. non-alien) dependent priories (Heale 2004, 30–9). It was clearly, as would be expected, well endowed with land (1a) within the borough; the land outside (1b) may refer to Clapcot, which abutted the castle and was certainly an area where the priory acquired a good deal of land (Mon. Ang. iii, 282). Acra (Acre) probably formed the core of the precinct and its farm, identified as the priory’s manor of Wallingford, including the prior’s house and the houses, barns and orchards within the precinct, as described in Henry VIII’s grant of the dissolved house to Cardinal Wolsey; Bodecroft (long since known as Bullcroft) was described as a close adjacent to

the same, with smaller parcels of land (pightellis) (Mon. Ang. iii, 282). Other parcels will have formed part of its pre-Conquest endowment. Post-Conquest, we have the evidence of the St Albans Catalogus Benefactorum that Walter Giffard of Crowmarsh gave to St Albans an acre of land and houses that he had in Wallingford which rendered 9s annually (GASA, 450). A grant to St Albans, rather than to the priory itself, indicates that the grantor was Walter Giffard I, who died before 1088 when his son and heir Walter II was created earl of Buckingham by William II. According to a charter forged in the early-13th century, probably at Wallingford Priory, another tenant-in-chief who granted land in the borough to St Albans was Nigel I de Albini of Cainhoe (d. 1095x1107), who is said to have given the church of Holy Trinity and half the church of St Mary le More, and thirty solidates of land (Salter 1929, 41). The St Albans tradition remembered him and his wife Amice (de Ferrers) as grantors to the abbey of the Wallingford churches, thirty acres of land outside the town, and the manor of West Hendred (3) (Mon. Ang. ii, 220a, 220b). Thirty solidates of land in Wallingford were amongst the grants made under Abbot Richard de Albini (1097–1119) as listed in the Gesta Abbatum Sancti Albani (GASA i, 67). Nigel had certainly made his grant of West Hendred to the abbey by 1086, when it was recorded as the abbey’s only manor in Berkshire (DB Berks 12.1). An unfortunately obscure reference in the account of the borough of Wallingford shows it holding one site there, the subject of a dispute (DB Berks B.6; Roffe 2009, 48– 9); Nigel de Albini held one site from his father-in-law, Henry de Ferrers (DB Berks B.1: Roffe 2009, 46), who had been granted the lands formerly held by Godric, sheriff of Berkshire. This association with the king’s local government is significant. According to the Gesta Abbatum, Abbot Paul sent monks to Wallingford to reform Holy Trinity with the support of Archbishop Lanfranc (d. 1089). In fact, it claims that all the abbey’s early cells were the work of Abbot Paul. As Mark Hagger has shown, this tradition arose from hostile claims made by Ely Abbey and the bishops of Lincoln, the latter stemming from Abbot Richard’s consecration at the hands of Bishop Robert Bloet (Hagger 2008, 385–8). Hagger convincingly shows that the large claims made for Abbot Paul are in the main unjustified, but this need not mean that he did nothing to kick-start some of the cells. Initiation between 1086 and Paul’s death in 1093 of the lengthy reform process that ended with the consecration of Holy Trinity Priory is perfectly plausible and may even be foreshadowed in the Domesday entries. Such delays were not uncommon, whether a community was a completely new foundation, such as the St Albans cells of Binham, Norfolk, Belvoir, Leicestershire, and Wymondham, Norfolk, or a reformation of an earlier religious community, with attendant complexities arising from existing proprietary rights pertaining to several persons, as at Wallingford. Nevertheless, in Wallingford as in the other cells, by far the most pro-active lead in bringing projects to completion was taken by Abbot Richard de Albini. None of the churches of the cells, including Wallingford, can be shown to have been consecrated before the time of Henry 215

I, though Tynemouth (Northumberland) may have been (slightly) earlier. Wallingford was certainly consecrated after 1100 since an inquest of 1212 recorded that William de Albini made a grant of land in Wallingford’s corn market when the new church was dedicated. William can only have been either Nigel of Cainhoe’s younger son of that name, or his nephew, the king’s butler and founder of Wymondham (BF, 113). When the initial endowment of the newly established Wallingford Priory was first determined by St Albans, probably at the time of its consecration, under Abbot Richard, the church and manor of West Hendred (2) formed part of it (GASA i, 55). Significantly, this valuable asset had been given to the abbey before 1086 by Nigel de Albini of Cainhoe, perhaps the abbot’s brother, as we have seen. He was the man credited with giving the church of Holy Trinity itself and half that of St Mary le More, and hence we might expect him to be remembered as the priory’s founder, as was the case with Peter de Valognes and Binham Priory, and William de Albini pincerna and Wymondham Priory. But Nigel de Albini of Cainhoe’s grants were made directly to St Albans; they were not made as preparation for founding a family church as in the other cases. The Catalogus Benefactorum lists the founders of each of its eight cells (Mon. Ang. ii, 223a). For Wallingford, fifth in the list, the man named is Geoffrey the Chamberlain, ‘an illustrious man devoted to God, who gave the cell to St Albans’. According to the early 13h-century inquest in the Book of Fees, Geoffrey gave several parcels of land in Wallingford to the priory around the time of its foundation, some of which he had first purchased (BF, 111, 113). In so doing he was probably acquiring the separate holdings of the English canons, effectively buying up their rights for transfer to the new foundation. The land before the door of the prior’s court was such a parcel, and may have been the land more elaborately referred to in the forgery claiming that Nigel de Albini was the donor (Salter 1929, 41). Nothing is known of the origins of Geoffrey the Chamberlain, whose association with the priory was important but short-lived, so what was distinctive about the foundation of Wallingford Priory attributed to him? There are three strands to the answer. Besides the grants by Geoffrey, grants of land in the borough to the priory between the dedication of the new church and the inquest of 1212 are otherwise attributed mainly to English townsmen, often when their sons took the habit, such as Sewold Graiessune and Godwin Child (BF, 111, 112). Three were priests, perhaps representatives of the earlier canons (ibid., 113–4). Only one grant is firmly dated, that of the son of Godwin Polsore whose grant was made in the time of William Rufus (ibid., 111). A potentially significant acquisition was that of five acres of common pasture which Edward monetarius had been allowed by the townsmen for pasturing his palfrey. Against borough custom he willed the pasture to the priory. The only known post-Conquest moneyer named Edward at the Wallingford mint struck two surviving coins of the cross pattée and fleury type of William II, produced from 1095 to 1098 (Williams 2012, Table 4b). Although William II is not usually associated with active monastic patronage, the

reformation of a secular college in an important royal town is wholly consistent with his idea of kingship (Barlow 1983, 178), the more so in light of the friendship he and his brother and successor Henry allegedly enjoyed with Abbot Richard (GASA i, 66). Whatever the truth of that, he could and did rely on the abbot’s family, entrusting Nigel I de Albini of Cainhoe with a campaign in Northumberland in 1095. In the next generation, Nigel’s nephews William de Albini Pincerna and Nigel were valued royal servants of Henry I. The younger Nigel, who was given the vast honour of Mowbray, granted Easole Street (Estwella) in Nonington, Kent, to St Albans, where he arranged to be buried (Greenway 1972, 6–7). The grant occurs in a list of donations received during the abbacy of Richard de Albini (GASA i, 67), following mention of 30 solidates in Wallingford. The later Catalogus Benefactorum conflated the two grants, attributing both to Nigel I of Cainhoe (Mon. Ang. iii, 220a). The most formal evidence that the priory was essentially a creation of the reign of William II is the writ of Henry I ordering Hugh of Buckland to see that the monks of Wallingford had their tithes of Moulsford (26) and of the land of Henry the Larderer as well as they had in the time of his brother, William, and on the day that Geoffrey the Chamberlain was disseised of the land. Like Geoffrey, Henry the Larderer was clearly a royal servant holding a serjeanty, occurring as son of Hugh the Larderer in the Winton Survey (Biddle 1976, 36, 43, 161, 390, 399; Round 1911, 233–40). The same source shows Geoffrey the Chamberlain in the service of Henry I from early in the reign until c. 1110 (Biddle 1976, 64, 66, 129, 157, 390). He might well therefore be the chamberlain of the king’s daughter mentioned in Domesday Book for Hampshire (Roffe 2009, 37). He was succeeded by his son Herbert in his Winchester properties before 1144 (Biddle 1976, 125, 161). Everything discussed so far indicates a strong connection with royal service, and hence that the foundation of the priory was an initiative more directly connected with royal administration than with baronial sentiment. It was also clearly one that had the active support of the townspeople, many of whom were parishioners of the priory and its churches. Their grants figure prominently among the grants listed in the Book of Fees (1212), and in the unpublished Berkshire charters in the Bodleian Library, Oxford (Coxe 1878, 4–23), but not among the mainly extra-burghal properties named in the terrier. The grants listed in the Book of Fees relate to the Borough of Wallingford and range from the very earliest – Geoffrey the Chamberlain is mentioned – to grants much closer to the inquest date of 1212, though few can be dated with precision. The work of royal servants and the townsmen in the formative stages of the priory’s growth in Wallingford are the first strand in its creation. The second strand is the subsequent addition of properties outside Wallingford, granted as an initial endowment by Abbot Richard, and the third the subsequent grant of those in or within the ambit of the honour, again associated with royal service. By the early 13th century growth was largely yielding to consolidation. 216

Unsurprisingly, several of the properties mentioned in the terrier and discussed below were held of the honour of Wallingford, the unusual status of which as a royal honour and castlery only ostensibly held as a barony has been discussed above (Roffe, p. 30, Keats-Rohan p. 48–9). Again, things were not what they might seem. The first constable of the castle was Robert d’Oilly (d. c. 1091). He is recorded by the St Albans Catalogus Benefactorum as the grantor of a fine bowl or chalice made of horn (Mon. Ang. iii, 221a). Clearly, he was not involved in the foundation of the priory. His son-in-law and successor Miles Crispin is not mentioned at all. Miles was a great benefactor, as were all his family, of the Norman abbey of Bec. He made extensive grants of tithes from the demesnes of the honour to Bec (Salter 1925), and is also known to have made grants to Abingdon Abbey. There is nothing to connect him directly with the creation or endowment of the priory. Whilst it is clear that St Albans regarded Geoffrey the Chamberlain as founder of Wallingford Priory, the role of Nigel de Albini and his family as patrons of St Albans was considerable. Nigel’s own grants, and those of his sons, to St Albans, provided the enterprising Abbot Richard de Albini with the primary endowments of the priory. The Wallingford forgery of Nigel’s grant to St Albans of West Hendred alleged that he was brother of Abbot Richard. That the relationship was close need not be doubted. Nigel was the Domesday lord of Cainhoe in Bedfordshire and the first lord of that barony. There is no certain record of him after his campaign in Northumberland in 1095–6, though his obit was preserved on 3 May in the necrology of another of St Albans cells, Belvoir Priory (BL, Add. 4936, f. 23r). Some time before Abbot Richard’s death in 1119, Nigel’s sons Henry, William and Nigel granted the church of Clophill (20) with half a hide and its tithes, with the tithes of Cotes and Cainhoe to St Albans, confirming the grant of their father Nigel of the church of Holy Trinity, Wallingford, and the manor of Hendred (Mon. Ang. iii, 220b). Tithes in these three places, all part of the manor of Clophill, Bedfordshire, occur in the Wallingford terrier. The assumption may be made that most of this also formed part of Abbot Richard’s endowment of the priory. Another part of Clophill occurs in a charter of Henry de Albini’s son Robert, c. 1140–46, as a grant to yet another of the cells, the nunnery of Beadlow or Beaulieu, in Bedfordshire, the core grant of which had been made by the first Nigel de Albini’s great friend and kinsman Nigel del Wast to St Albans (Mon. Ang. iii, 276–7; VCH Beds 2, 320–5). Also in Bedfordshire and apparently part of the initial endowment of Wallingford was Tempsford (10), where the terrier records a holding of two hides. This lay in the manor of Brayes, held in 1086 by Richard Puignant (VCH Beds 2, 251–5); it was given to St Albans between then and 1093 (GASA i, 56). In the 1291 Taxatio the priory is seen to have held a pension from the church of Tempsford for this holding. Before 1119 Nigel’s nephew William, butler to the king, founded the cell of St Albans at Wymondham (Mon. Ang. iii, 330). William’s brother-in-law, an otherwise unrelated William de Albini Brito, was patron of another of the cells, Belvoir, to which Wallingford later leased its

holdings in Clophill (20) (Mon. Ang. iii, 281). The butler’s brother, Nigel, also a favourite of Henry I, and father of Roger de Mowbray, was patron of Tynemouth. The church of West Hendred (2) and its ten hides were the key holding of the priory from its beginning. Between 1151 and 1166, Abbot Robert (de Gorron) of St Albans confirmed the grant by the prior (Nicholas) and convent of Wallingford of their tithes of Mapledurham in Oxfordshire to their sacrist and those of West Hendred to their almoner (Milne 1940, 53). Subsequently, Abbot Simon permitted the use of the revenues of the church and tithes of West Hendred for the upkeep of the convent kitchen and for the purposes of hospitality, one of the main obligations of Benedictine houses (Milne 1940, 53–4). It is clear from the wording of these grants that, contrary to Milne’s view that they constitute first-time grants by the respective abbots, they are in fact the approvals that had to be sought when a dependent cell wished to change the way it allocated the revenues or rights deriving from its properties. These charters, and a substantial archive of material relating to West Hendred surviving at Corpus Christi College, Oxford, make it the best evidenced of the priory properties. At first St Albans appointed the rectors of the church, but after a dispute with the bishop of Salisbury on the death of the incumbent Philip de Calna, the priory was eventually, c. 1184, accepted as rector (Milne 1940, 53–4; Kemp 1999, 115–16, no. 148); West Hendred church is confirmed as an appropriation of Wallingford Priory in the 1291–92 Taxatio. The next most important church listed in the terrier is that of Aston Rowant (3), the seat of a rural deanery and probably a church of some antiquity. The church was appropriated by the priory and a vicarage ordained by 1219, when it had a dependent chapelry at Stokenchurch (VCH Oxon 8, 16–43). This exceptionally large parish of over 7000 acres had an interesting history. According to the Abingdon Historia it had been a manor of Archbishop Stigand, possibly as bishop of Winchester. In 1055 he had offered it to Abingdon in compensation for their manor of Cerney, Gloucs, which he had appropriated (Hudson 2007, clxvi, 196–7). The grant was never made and in 1086 it was a demesne manor of the honour of Wallingford (DB Oxon, 35,3). Stigand, who was deposed in April 1070 as a pluralist, held the abbacy of St Albans among several others (cf. Hagger 2008, 380, 382, 384–5, 392, 395). Very influential in the affairs of Wallingford in the Conquest period (Keats-Rohan 2012, 172–3), he may have been instrumental in forging the link between St Albans and the borough in 1086 (Roffe 2009, 48–9). At that date the only other English abbey with a presence in Wallingford was Abingdon, whose unhappy relationship with the town included the imprisonment of its English abbot in the new castle in 1071 (Keats-Rohan above 51). Soon after the Domesday inquest it appears that Chalford in the parish of Aston Rowant became a separate manor. According to confused accounts in the 13th-century Hundred Rolls (RH ii, 786), the hamlet of Chalford with 4 hides of land were given to the priory by Brien fitzCount and Matilda the Empress, his wife. This either refers to Brien and his wife Matilda of Wallingford, or to the Empress. The Empress died in 1167 and Brien soon after 1148, before the terrier 217

was written. At the time of his withdrawal as a monk to Reading Abbey, Brien had committed the castle and honour of Wallingford to Henry of Normandy, who made good his claim at the relief of the third and final siege of the castle in 1153, when King Stephen accepted him as his heir. Henry was the first to enfeoff a mesne tenant in the manor, Peter Boterel, who was dead by 1154 (BF, 111). Chalford remained a priory holding until the Dissolution. It ought to be identifiable in the terrier because Brien was involved in the donation to the priory; the only candidate would appear to be Helesford (7), where the priory held three hides. The church of Bucklebury (4), Berks, listed in the terrier after Aston Rowant, was exchanged with Reading c. 1151x54, as we have seen. In 1086 it was in the king’s demesne. Reading claimed that the manor had been given to the new abbey by its founder Henry I. No documentary proof survives, but it is quite likely to be true (RAC i, 16). The priory held the advowson of the church of Garsington (5), with its appurtenant hide. Garsington was a large parish, the greater part of which was held by Abingdon Abbey. the hide was held in 1086 from Miles Crispin by Toli, who also held land in Cowley from Miles (VCH Oxon 8, 134–56). This property passed to a family named de Chausi who gave their name to Mapledurham Chazey (16). Mapledurham was a demesne manor of the honour of Wallingford in 1086. The Garsington hide was held by the priory from the Chazey family and had been sublet to Alan son of Siward for 20s per annum by 1170; subsequently it was purchased by Henry III from Alan’s heiress, Alice de Colleville, to endow a chapel of St Mary at St John’s Hospital, Oxford, with 20s to be paid annually to the priory (RH ii, 39). The earliest extant mention of the Chazey family and the priory comes in a confirmation of the priory’s possessions, particularly those in Berkshire, by Bishop Jocelin of Salisbury, which includes the grant of three parts of the demesne tithes in Moulsford (26) by Emma de Chausi. This may be the origin of the portion in the church of Cholsey held in 1291 by Wallingford Priory (Taxatio). The document is perhaps inauthentic in its surviving form but based upon a genuine original (Kemp 1999, 114–15, no. 147), and likely to be related to the dispute concerning presentation to West Hendred (2), resolved c. 1184. Again, the origin of the priory’s possession of the church of Chinnor (6), in Aston deanery, is not known (VCH Oxon 8, 55–80). The priory fee included land in Chinnor, Henton and Wainhill (11). The terrier gives two hides at Chinnor with the church and a separate two hides at Wainhill. Chinnor was held by Leofwin in 1086, but had been acquired by the de Vernon family before 1130, when Richard son of Hugh de Vernon accounted for his father’s lands in Oxfordshire. Wainhill had been held by Reginald Canute of the honour of Wallingford in 1086. The second tithes of Ralph Chanu, his descendant, in Rodbourne Cheney (23), in Wiltshire, are listed in a confirmation charter of Bishop Robert of Salisbury in 1243 (Mon. Ang. iii, 280b). The same charter confirms the grant of the tithe of hay in the demesne of Miles fitzTurold (fl. 1196) in Aston Tirrold (14), the demesne tithes of Manasses de

Sanderville in South Moreton, and the tithes of Geoffrey the Chamberlain in North Moreton (8), the whole demesne tithes of Carswell (18), Donington and Earley (29), the second tithes of Samson Foliot in Draycot Foliat in Wiltshire, and the demesne tithes of the Pipard fee in Swindon (24), where the priory had a pension in 1291. These are mostly places mentioned in the terrier, the two last being part of the honour. The damaged name apparently reading...hechota (25) may refer to Nethercot in Henton, part of the priory’s estate in Chinnor (6) where it is known to have held Stephen’s Mill (VCH Oxon 8, 65–80). Both North and South Moreton were close to Wallingford but not part of the honour; the priory held portions in the churches of both North and South Moreton in 1291. The apparent association of Geoffrey the Chamberlain with the grant of North Moreton, in the fief of William Corbucion in 1086 (Roffe 2009, 37), may indicate that it was the origin of the terrier’s reference to Moreton. A Geoffrey held Charlton (17) in Wantage Hundred of William Corbucion in 1086 (DB Berks, 27,3), probably where the virgate mentioned in the terrier lay. South Moreton was held by William Luvet in 1086 (ibid., 26,3). His holdings in Berkshire, which included Donnington, were held by his successors or subtenants of the Sanderville family, the whole fief passing to the earls of Albemarle (VCH Berks 3, 498–504). He was remembered as the grantor of Flittwick (Beds) and South Moreton to St Albans (Mon. Ang. iii, 220b). Donnington, which is possibly the illegible place-name (13) between Eythrope and Aston Tirrold in the terrier, was later joined with the parish of Shaw (VCH Berks 4, 87–97). The priory had a pension in the church of Shaw in 1291; the rector of Shaw took portions from Eddington (possibly the illegible penultimate name (32) in the terrier) in the priory accounts of 1385–6, while the rector of Eddington took tithes in Carswell (18) in 1484 (TNA, SC6-752–1). Carswell was part of Buckland parish; in 1291 Wallingford Priory had a portion in the church of Buckland. The grant of tithes in Carswell had been made by Amalric son of Ralph, whose successor William of Sulham may have been the William son of Turold, nephew of Wigod of Wallingford, tenant of Britwell Salome in 1086 (DB Oxon, 35,23). The half hide was allocated to Wallingford’s sacrist, with the assent of Abbot Robert (1151–66) (Milne 1940, 53, no. 2). Its virgate in North Marston (31), in Waddesdon Hundred, Buckinghamshire, appears to derive from a Domesday holding of Bishop Odo of Bayeux, three hides of which were held from him by Robert (d’Oilly?) and one of which passed to the honour of Wallingford, where Bernard held a further hide under Miles Crispin in 1086 (VCH Bucks 4, 76–80). The three-fee manor of the honour of Wallingford in Waddesdon was known as Eythrope manor (ibid., 107– 118). Land there (12) was first granted by Henry, duke of Normandy, to Henry of Oxford, the burgess of Wallingford and Oxford who had helped defend Wallingford against King Stephen from the 1140s and whom Henry made his sheriff of Berkshire in the early 1150s (Keats-Rohan 1989a, 307–08). East Garston (30), Berks, is so named as a manor of Geoffrey de Mandeville in 1086; he gave the land and all tithes to his priory of Hurley. Possibly the priory’s half 218

hide originated in the royal manor of Lambourn, or was the part of East Garston held as a serjeanty by Maurice de London of Kidwelly, who was obliged to make a grant of tithes to Hurley in accordance with the founder’s precept (Wethered 1898, 94; Mon. Ang. iii, 433). The remaining holdings were all in Oxfordshire. Watlington (28) was a large manor held in 1086 by Robert d’Oilly that came late to the honour of Wallingford (VCH Oxon 8, 210–52). The grantor of the priory’s two hides there is not known, but its estates at Maunsells, Cartersland and Herewardsland in Watlington are mentioned in its accounts for 1455 (TNA, SC6-752–2). The terrier also mentions land in North Stoke (27), where the priory had a mill and a fishery until its dissolution (Mon. Ang iii, 231). North Stoke (27), like Chalgrove (9), was a prebend of St Nicholas in the Castle, perhaps established by Miles Crispin who had given tithes there to Bec when the manor was still held in demesne (Salter 1925, 75). The hide in Stanlake in Berkshire (15) was given by Manasses I Arsic, baron of Cogges, who had been given the land formerly held by Odo of Bayeux’s man Wadard by Henry I. It was confirmed by his son Robert who succeeded him and died c. 1161, when his son Manasses II succeeded. The charter as given in Milne (1940, no. 24) must, on the authority of the terrier, now be re-dated to c. 1151, the first occurrence of Robert Foliot, archdeacon of Oxford. Ferlei (22) has not been identified. Coming before two Wiltshire holdings in the list, it was possibly Farley Hill, a hamlet on the east of the Berkshire manor of Swallowfield, that was itself in Wiltshire until 1884 (VCH Berks 3, 267). Earley (29) was part of the king’s demesne (DB Berks 1,21;42), and a part of the prebend of Sonning, a demesne manor of the bishops of Salisbury (DB Berks, 3,1; VCH Berks 3, 210–25). The terrier shows that the grant of tithes there to the priory was made before 1151x54. In 1086 the manor of Sonning included East Ilsley and a church at Wallingford. At the date of the Taxatio, Earley as a prebend of Salisbury had two chapels. None of the various chapelries among the prebends related to Wallingford. David Roffe (2009, 37) has suggested that the charter of Jocelin, bishop of Salisbury, of c. 1160, regarded as problematic in its preserved form (Mon. Ang. iv, 280a), was in effect a quitclaim of the episcopal rights over the priory’s churches, further suggesting that the church in question was St Martin’s, but noting the possibility of St John instead. The latter is probably correct. It may have been a grant, made when surrendering the last of the diocese’s Wallingford parishes to the priory. One of the difficulties with both episcopal charters, Jocelin’s and the later confirmation of Bishop Robert de Chesney, is that they both refer to the church of St John, as well as to Holy Trinity, Mary le More and St Martin. None of the surviving priory documents consulted refer to a church of St John. This would have been the church of St John-super-Aquam in Thames Street, which occurs in borough documents of the 13th century as a parish, but one that had failed by 3 March 1349 (‘former parish’, BRO, RTb 88), and had been united with the parish of St Mary le More before 5 March 1420 (Boarstall, 280, no. 840; cf. VCH Berks 3, 539–46 and notes 97–8).

The last item to be mentioned is the land at Holy Sepulchre without Newgate in London (33). This church is first recorded in 1137, when it was given by Roger, bishop of Salisbury, to the Prior and Canons of St. Bartholomew. They held the advowson until the Dissolution of the monasteries. St Albans held quite a bit of property in London, mainly churches and tenements. Several of those listed in its Catalogue of Benefactors made these grants (Mon. Ang iii, 220). It may be assumed that such property was useful because the records of St Albans itself and the surviving accounts of Wallingford Priory, indicate that travel to London on business was a matter of routine. In the case of Wallingford Priory at least, there is no reason not to accept the version of its origins sketched in the Gesta Abbatum. Apart from a couple of names that cannot be read and one that has not been identified, all the properties listed in the terrier were among those still held by the priory at the time of its suppression. They have marked features in common. Many were demesne manors of the king or the honour of Wallingford that were first enfeoffed during the civil war against Stephen by Brien fitzCount, or by Henry of Normandy who took control after Brien’s retirement in 1148. Several of those infeudations were preliminary to a subsequent resumption and re-grant by the then king Henry II. Others were royal demesne manors that had been used for royal servants. There was, surprisingly or not, no direct involvement by the constables of Wallingford with the exception of Chalford in Aston Rowant – and that is not certainly mentioned in the terrier and was not certainly a grant of Brien and his wife, but possibly a grant by Henry’s mother the Empress. The formation of the initial endowment points to a close involvement and doubtless co-operation between the king (William II and Henry I), his ministri and the townsmen, on the one hand, and on the other Abbot Richard de Albini, subsequently boosted by similar patronage by Henry I’s men, and then by his grandson Henry. It is further evidence of the importance of the royal borough and honour of Wallingford and the close interest taken in its development by the post-Conquest kings. Widening the inquiry out from the terrier has also demonstrated the close involvement of the townsmen in the lengthy foundation process and their support for it thereafter. The terrier itself dates to a pivotal moment in the history of both priory and town, the appointment of Prior Nicholas and the tightening of the Angevin grip in this strategically important town, soon to culminate in the accession of Henry II. Acknowledgements My thanks to Dr Tessa Webber for advice on the manuscript, and to Dr David Roffe for valuable comments on the first draft. Thanks also to Elizabeth Will for transcribing the late medieval accounts of the priory in TNA which have provided useful checks on the material presented here, to David Pedgley for photographing those accounts and whose interest in them has stimulated my own, and to the archivist of Corpus Christi College, Oxford, who allowed David Pedgley and me to photograph accounts relating to West Hendred.

219

Appendix: Translation and text of the Wallingford Priory Terrier, BL Royal 5 D xi., fol. 153v First and foremost the (1) church of Holy Trinity in Wallingford and two other churches pertaining to the aforesaid [church], that is, the churches of St Mary and St Martin; and the land which is called (1a) Acre and Bodecroft [Bullcroft]; and many other parcels (1b) of land within the borough and without; (2) the church of [West] Hendred [Berks], and 10 hides in the same; (3) the church of Aston [Rowant, Oxon], and 11 hides in the same and the tithe of the vill; (4) the church of Bucklebury [Berks] and its hide and tithe of the vill, both of the monks as well as those of the villagers; (5) the church of Garsington [Oxon] and its hide; (6) the church of Chinnor [Oxon] and two hides; (7) three hides in Helesford; (8) one hide in [South] Moreton [Berks]; (9) 2 hides in Chalgrove [Berks]; (10) two hides in Tempsford [Beds]; (11) two hides in Wainhill [in Chinnor, Oxon]; (12) half a hide in Eythrope [in Waddesdon, Bucks]; (13) half a hide in D?... and (14) one virgate in Aston [Tirrold, Berks]; (15) a hide in Stanlake [Berks]; (16) [half] a hide in Mapledurham [Oxon]; (17) a virgate in Charlton [Northants]; (18) a virgate in Carswell [in Buckland, Berks] and tithe of the vill; (19) two parts of the tithe in [...]; (20) the tithe of Cotes, Cainhoe and Clophill [Beds]; (21) the tithe of T[...wes...]; (22) the tithe of Ferlei; (23) the tithe of Rodbourne [Wilts]; (24) the tithe of Swindon [Wilts]; (25) the tithe of [...]hechota; (26) the tithe of Moulsford [Berks]; (27) a virgate in [North] Stoke [Oxon] and the mill and a fishery; (28) two hides in Watlington [Oxon]; (29) tithe of Earley [in Sonning, Berks]; (30) tithe of East Garston [Berks]; (31) a virgate in Marston [North, Bucks]; (32) a virgate in [illeg] and the tithe; (33) land of Holy Sepulchre without Newgate, London.

In primis ecclesia Sancte Trinitatis in uilla Walengefordie et alie due ecclesie pertinentes ad supradictam scilicet ecclesie Sancte Marie et Sancti Martini. et terra que dicitur acra et Bodecroft. et alie multe terre infra burgum et extra. ecclesia de Henred. et x. hide in ipsa. ecclesia de Estona. et vi. hide in ipsa et decima uille. Ecclesia de Burgildeburia et hida eius et decima uille tam de monachis quam de uillanis. Ecclesia de Gersidona et hida eius . Ecclesia de Cennora et .ii. hide. Tres hide in Helesford. Vna hida in Mortuna. Due hide de Chelgraua. Due hide de Tamiseford. Due hide in Winehale. Dimidia hida in Aetrope et una uirgata in Estona. Vna hida in Stanlac. Dimidia hida in Mapuldraham. Vna uirgata in Cerlitona. Vna uirgata de la W[...]. Vna uirgata in Chersuwella et decima uille. Due partes decime in (?) D[...]. Decima de Chotes et de Chahesso et de Clopehulla. Decima de ??Tiwes... Decima de Ferlei. Decima de Redburna. Decima de Swindona. Decima de [...]hechota. Decima de Mulesford. Vna uirgata in Stoches et molendinum et piscarium. Due hide in Watlitona. Decima de Herleihe. Decima de Hesegarestune. Vna uirgata in Merxtona. Vna uirgata in C[...] et decima. Terra Sancti Sepulchri iuxta nouam portam London[...]

220

12 WALLINGFORD PRIORY David E. Pedgley Abstract Apart from a royal castle, the most prominent inhabited building in medieval Wallingford was a small, selfsupporting Benedictine monastery. It was the Priory of Holy Trinity, one of eleven priories and nunneries that were cells, or dependents, of the great abbey of St Albans in Hertfordshire. The priory existed for nearly four and a half centuries, from the late 11th to the early 16th. In contrast to the well-known banks and ditches of the former castle, and the mound upon which the keep stood, no obvious remains of the priory can be seen today. However, something of its structure and function can be deduced from two principal sources of evidence: written records and archaeological survey and excavation. The aim of this paper is to present the results of an analysis of this evidence. From foundation to dissolution Founded as a cell of St Albans Abbey, Wallingford Priory was to remain a dependent throughout its history. Its endowment was modest. As was typical of daughter houses, its largest possessions were granted early on in its history. West Hendred was given to St Albans by Nigel de Albini before 1086 (VCH Berks 4, 302), and Chalford (or Chalkford) by Brien fitzCount and his wife Matilda of Wallingford or the Empress Matilda in the early 12th century (RH ii, 786). Income from these lands, however, was supplemented by a considerable endowment in spiritualities. Probably from its foundation Holy Trinity was in possession of the advowsons of the Wallingford churches of Holy Trinity, St Martin, St John super Aquam and St Mary the More, as confirmed in the late 12th century by the bishop of Salisbury (Coxe 1878, 14). Outside the borough it also had the advowsons of West Hendred, Chinnor, Garsington, Aston Rowant (with Stokenchurch chapel) and Shabbington (Bucks). In total these churches were valued at £72 5s in 1291 (Taxatio), although from some the priory was only entitled to a pension. The priory also enjoyed tithes from a further 14 parishes in five counties. Aristocratic patronage was probably confined to the first 50 years or so of the priory’s existence. It certainly did not extend into the 13th century. By contrast, grants by local inhabitants in small parcels of land and rents continued until the late 14th century (CPR 1441–46, 460). With changing patterns of piety, however, supply progressively diminished and the priory resorted to the land market. In 1334 King Edward III had granted a licence for the priory to acquire lands and rents to the value of £10 (CPR 1330–34, 494), although £7 was still remaining in 1392 (CPR 1391–96, 110). In 1465 the priory acquired much more land by a 99-year lease of the nearby manors

of Rush Court and Clapcot from Sir Edmund Rede, but at a cost of 40 marks (£26 27s 4d) yearly (Coxe 1878, 12). In the absence of early accounts, it is not known how the priory managed its assets in the first 200 years of its existence. If it was at all typical, however, the priory would have initially leased its lands. Demesne farming became increasingly profitable from the mid-12th century and West Hendred and Chalford may thus have been directly exploited. From the late 13th century leasing of both lands and tithes again became the strategy of choice. The papal taxation of 1291 and extant accounts from 1385–92 and 1454–56 reveal that income remained static or fell to the mid-15th century, reflecting the impact of climatic change and the plague in the 14th century (Taxatio; TNA, SC6/752/3; TNA, SC6/752/1). Thereafter, however, it seems that there was a modest recovery in the priory’s fortunes (Corpus Christi, Oxford, MS Z/2/1/1; TNA, SC6/ HENVIII/106). Of the religious life of the priory relatively little is known. Founded within an urban parish church and responsible for the cure of souls throughout much of its history, it can never have been an entirely closed community. There is no evidence that Holy Trinity housed a school for the townspeople or, indeed, provided medical expertise, but it is not unlikely that it did so, as so many other religious communities did. Without doubt it had a wider educational role within the monastic community of St Albans. Many daughter houses were as much estate offices as religious communities and were effectively training grounds for high office in the parent house and beyond. There are indications that Wallingford Priory may have had something of this character in the 12th century: Prior Nicholas was elected abbot of Malmesbury in 1182 and Prior John de Wallingford abbot of St Albans in 1195 (Thompson 2002, 66–7). The priory may have also fed novices into the wider monastic community; the Book of Fees indicates that some grants of land were made to fund novitiates (BF, 109–16). There are no records of the number of monks until the late 14th century. According to the accounts, there were, in addition to the prior, six monks in 1385–92 (TNA, SC6/752/3). In 1445 King Henry VI granted £8 yearly, to be taken out of the borough’s fee-farm paid to the crown, in order that the prior and convent ‘may pray for the good estate of the king and his consort, Margaret, and for their souls after death, and in consideration of the kindness which they have shown the king, and that the number of monks there may increase and serve more devotedly the Holy Trinity by night and day’ (CPR 1441–46, 362). The intended increase in numbers apparently never materialized. There were only four monks in 1481–82, three in 1484–85, and three in 1521–22 (Corpus Christi,

221

Oxford, MS Z/2/1/1; TNA, SC6/HENVIII/106). In common with many religious houses, it seems that Holy Trinity experienced a marked decline in vocations, and the grants of land and rents that made them possible, in the late 14th and 15th centuries. At the same time there was a change in the religious life of the priory. The common table disappeared in the 1380s and the monks were each allotted 26s 8d for their upkeep (TNA, SC6/752/3). A century later the dole had risen to 40s where it remained until the suppression (TNA, SC6/HenVIII/106). Furthermore, in 1385–86 the dormitory had been partitioned by order of the prior at a price of 60s to afford more privacy (TNA, SC6/752/3). It would be anachronistic to characterize these changes as laxity and decline. Shifts in modes of piety were certainly a factor, but these were as much cause as effect. In fact, patterns of endowment suggest that the priory remained a significant feature in the spiritual map of Wallingford itself almost to the end of the 14th century, albeit with differing expectations. Early on gifts were usually for specific purposes, such as for the health of the soul of the donor, or supporting a particular monk’s duties, or providing candles ‘to burn before the image of St Mary in the church of Holy Trinity’ (Coxe 1878, 17, ch. 43). Later ones, by contrast, were unconditional. A parish altar was maintained in the priory church until the mid-15th century when the decline in population made it redundant. It seems clear, however, that by the early 16th century the priory had ceased to be the vital institution it had once been. Wallingford itself was undergoing change as it sought a new role as a market town. At the same time the castle had ceased to be the royal palace that it had once been. It is general decay in the town that forms the backdrop to the suppression of the priory. In June 1524 Pope Clement VII gave Cardinal Wolsey permission to suppress Wallingford, together with 20 other small religious houses, with the aim of providing funds for the foundation of a new college in Oxford (Mon. Ang. iii, 282a). In September of that year King Henry VIII issued a licence for the priory to be used for the planned Cardinal College (LPFD 4, no 650). In January 1525 Wolsey commissioned Thomas Cromwell and two others to survey the priory and its possessions (ibid., no 989), and in April it was surrendered by Geoffrey, the then prior, to the king (ibid., no 1137:19). The rectories and churches of the priory were added in July (ibid., no 1468). Cromwell visited the priory in April 1528 and found that its goods had been removed; he also sent the ‘evidences’ (deeds) to the dean of Cardinal College – these are the charters (or some of them) now held in the Bodleian Library (ibid., no 4135). In the following July the king at last granted Wolsey the priory and all its possessions – manors, advowsons, tithes and rents (ibid., no 4471) – whereupon the dean immediately appointed attorneys to take possession (ibid., no 4494). Demolition was speedily commenced. By October 1529, lead was being taken away for use on the college (ibid., no 6023) and £3 13s 4d was paid by the Corporation for taking down the church, from which 120 cart loads of stone went to repair the bridge over the river Thames (BRO, W/Ac1/1/1, f.49r). The fate of the prior and monks is unknown, but the last steward,

William Belson, demanded a 40s pension for life (ibid., no 1138:12). The priory and its precinct The 1525 survey is not known to have survived. Other records of the suppression, however, indicate that the priory was considered to consist of two components: (a) a capital messuage in which the last prior dwelt, along with the site of the priory itself, and (b) a capital messuage and site of the farm which is described as the ‘manor’ of Wallingford (Mon. Ang., iii, 282; CPR 1550–53, 119). Both appear to have been situated in the southern part of the Bullcroft. Writing around 1540, John Leland recorded that the priory stood ‘hard withyn the Westgate’ (Leland i, 119). Numerous post-medieval references place it to the north of tenements on the High Street. Geophysical survey (soil resistivity and magnetometry), undertaken in 2008 as part of the Wallingford Burh to Borough Research Project, revealed evidence of extensive building activity there, which had also been seen previously in test pits. A further pit and small trench in the back garden of 61 High Street subsequently uncovered demolition layers with fragments of floor tiles over a mortar floor. Hedges (1881, ii, 358) records that tiles found there in the 1870s were used to floor a summer-house in the Bullcroft. According to a report in Jackson’s Oxford Journal for 10 March 1894, this was in the north-western corner, where OS maps indicated a shelter on the town rampart. Today there is no sign of it. The northern part of the Bullcroft, by contrast, revealed medieval ploughing, later confirmed in excavation, with no sign of extensive construction (on the archaeological survey and excavations in the Bullcroft, see Christie and Creighton 2013, 262-75). Of the priory itself, references in medieval and early modern sources have been found to a variety of component buildings, listed in the appendix. It must be supposed that there were also cloisters, an infirmary, and parlour. Something of the layout of these structures is suggested by the results of a further soil resistivity survey of around three hectares of the southern part of the Bullcroft undertaken by TWHAS in 2011. This imaged the demolition layer in some detail and in particular revealed a rectangular feature ABCD (Figure 12.2). A pseudo-vertical section across the line AB suggests that this is a wall footing at a depth of between 1m and 2m. At approximately 15 x 20m the feature probably represents the outline of the cloisters. If correctly identified, it suggests the function of other structures in its vicinity. First, AB is parallel to the extant wall EF about 14m to the south and forming the northern boundary of properties nos. 50 to 53 along the north side of High Street. According to detailed borough surveys of 1548 and 1606, properties here had belonged to the priory but lay outside the priory precinct, so it is likely that this wall preserves the line of the southern boundary of the priory (Bodleian Library, MS Top Berks b. 41; TNA, E315/369/3, 101–23). Indeed, it may even incorporate fragments of the medieval wall because, although mostly of recent brick, it rests on a clunch (hard chalk) footing for at least 35m. 222

Figure 12.1 Suggested layout of Wallingford Priory based on geophysical survey (image: author)

The area between the feature ABCD and this boundary is perhaps best interpreted as the priory church. Just such a position would provide easy access for parishioners from the town and is, moreover, consistent with John Leland’s description of the priory in around 1540. Its cemetery was certainly close by. In 1833, a stone coffin large enough only for a child was discovered in the garden of the present 61 High Street during excavations and later used as a cattle trough (Hedges 1881, ii, 358). In the 1860s, two or three skeletons were discovered under the parlour at the back, facing east but without coffins (Allnatt 1873, 64). Human remains have been found since at 56 and 64–65 High Street during building work. The position of the church marked on 19th-century OS maps further east on the western side of the present entrance to the Bullcroft seems to have been based on supposition rather than physical evidence. The footings of the wall EF may, then, not only preserve the southern boundary of the precinct, but also part of the south wall of the church itself. A further fragment of the structure may also survive. It is noteworthy that E closely continues the line DA. From lay-outs of other priories, E would be close to where a southern transept joined the nave. It is at this very spot that pieces of moulded stonework were found in 2010, having been re-used in a wall (Christie and Creighton 2013, 261). Point A would then in its turn be close to where a corresponding north transept joined the nave. It would be to the north of this point, on the eastern side of the cloister, that the sacristy and chapter house were most likely situated.

Eastern walls of the transepts may be represented by the linear feature GH. Other buildings lay in an area up to 30m to the east of this line partly obscured by modern structures. IJ is somewhat longer and seems to represent a significant boundary. A pseudo-vertical section across its line located a wall foundation. It is possible that this linear feature represents the precinct outer wall, lying roughly west–east across the Bullcroft. K is at the centre of a rectangular close similar to ABCD may be an inner court, containing such buildings as the guest house. The outer court would then comprise the whole or part of the remainder of the precinct, containing most of the farm. In contrast to the conventual buildings which were probably demolished at the same time as the church, at least some of these structures seem to have remained upstanding after the suppression of the house, no doubt because the farm continued to function into the mid-19th century. The ‘priory great barn… backside and heyse [probably a small enclosure] thereto adjoining’, was leased in 1709 ‘with liberty to water cattle… in the pond in the Bullcroft near unto the said barn and backside; … heretofore part and parcel of the late dissolved Priory of Wallingford’ (HRO, D/EAm/T34). The pond is shown on a plan of 1786 (HRO, D/EAm/P1), the tithe map of St Mary’s parish of 1847, and on subsequent OS maps. As for the barn, both the plan and the tithe map show a building close by on a site described as ‘house, outbuildings and farmyard’ (TNA, IR29/2/136) extending northwards

223

Figure 12.2 Results of geophysical survey in 2011 of the southern part of the Bullcroft, Wallingford (image: Gerard Latham).

Figure 12.3 Artist’s impression of Wallingford Priory looking southwards (image: June Strong).

224

from the house often known as The Priory, now 61 High Street. It had dimensions around 35 x 10m. Moulded stone indicative of a high-status building, recovered from a demolition layer found in a test pit excavated in 2011 in the garden of adjoining 62 High Street, seems likely to have come from this building when it was taken down to make way for the formal garden following the sale of the farm in 1858 and subsequent conversion of the farm house into a private residence. Priory barns of a similar size have survived elsewhere to modern times, as at Watton (East Yorkshire) and Bradenstoke, Lyneham (Wiltshire). Geophysical survey, combined with historical evidence, seems to show a typical priory layout with a cloister lying to the north of an east-west church with a farm to the east. Figure 12.1 shows this suggested interpretation. It must be emphasized that the reconstruction needs to be confirmed by more detailed survey and preferably also by excavation. However, the extent of the priory so indicated is consistent with medieval evidence. Hedges (1881, ii, 349) reproduced an indenture of 1425 which outlined a dispute with the town whereby the priory made good its claim to the Bodecroft, as the Bullcroft was then known, and which gave its area as 7 acres (2.8ha). The present-day Bullcroft, including the site of the priory (in the southern part), is around 13 acres (5.2ha), implying that the priory itself occupied around 6 acres (2.4ha). This figure accords well with the area of the southern Bullcroft delimited by the suggested northern precinct wall. The same dispute also provides valuable evidence of an access route to the priory church. At that time the Bodecroft was said to have extended from the West Gate of Wallingford to near the North Gate, implying that it included a strip of land between the western precinct wall and the town rampart north of the West Gate. This implication is supported by the discovery of a paved, north-south track associated with medieval pottery during building excavation in the garden of The Keep, 50 High Street, according to a previous owner. The track would have provided access not only for the priory to the strip of Bullcroft but also for the public to the west end of the church, if in fact that lay where the geophysical survey suggests. The priory church gate, noted in 1388 and 1516/17, may have been located on this track (Figure 12.3). There is no obvious sign of a stream or spring within the precinct and so water must have come from the town supply. The Mill Brook, fed by springs in the Moretons and Brightwell, flowed eastwards into the town ditch which passed close to the priory. How water was brought into the precinct and there managed is unknown, although it might be supposed that there was a conduit through the town defences (see also Christie and Creighton 2013, 2654, 273). Fish, an important commodity in a monastery diet, required a source of flowing water, and so by necessity the

priory ponds had to be located elsewhere. The priory rental of 1516–17 includes among its properties ‘Colverhowse Close [presumably indicating a priory pigeon house there] with ponds, lying outside the West Gate of Wallingford’ (TNA, SC11/70/5); and a borough survey of 1606 lists two fish ponds outside the West Gate on former priory land (TNA, E315/369/3). A perambulation of the borough boundary in 1806 refers to ‘the Culver, otherwise Brook, Piece’ showing that it lay next to the Mill Brook (BRO, W/Ac11/1/3, f.105r). It is unlikely that these ponds would have made the priory self-sufficient in fish, and indeed surviving accounts refer to purchases of fish, both fresh and salt. In addition, there was access to river fisheries on the Thames – to the south at North Stoke and to the north at Huddesbutts in Shillingford (parish of Warborough). Conclusion Wallingford priory was always a modest foundation and its history is not particularly exceptional, apart, that is, from the peculiar circumstances of its suppression by Cardinal Wolsey. If its initial prosperity was checked in the 14th century and its communal religious life underwent fundamental change, then it was in the company of many another religious house of the time. Its strategies in estate management were again the norm. That, however, is not to diminish its importance to Wallingford. Throughout its history the priory was a significant element in the spiritual, social, and economic life of the town. Apparently founded by a group of royal ministri or officers, it was in a sense a counterpart to the castle to its east. But there was not much that was particularly aristocratic about it. The priory was never a private chapel like St Nicholas that exclusively served the garrison. As a parish church, it was always open to the townspeople of Wallingford and it appears to have been an important focus of local piety, even at a time when Benedictine foundations were falling out of favour with pious benefactors. The priory, it would seem, was fully integrated into the urban scene. It was a significant landowner and rentier in Wallingford itself until its suppression and, as an estate centre for a wider scatter of lands, rents, and tithes, it brought wealth into the town. The priory precinct not only loomed large in the townscape of Wallingford, but also in its life. Acknowledgements I am much indebted to David Roffe for guidance in interpretation of the medieval records used in this study, including those sought out and abstracted by members of TWHAS. I thank other members, led by Gerard Latham, who undertook the geophysical survey of the priory site to help generate results, and discussion.

225

Appendix. Documentary references to the buildings of Wallingford Priory. Structure Lady Chapel chapter house sacristy

dormitory latrines refectory/hall kitchen cellar/store brewhouse bakehouse prior’s lodging chamberlain’s lodging guest room cemetery great gate gate of priory church farm

Date 1485 1427 1503 1212 1342 1355 1386 1482 1386 1389 1482 1522 1386 1522 1482 1482 1547 1386 1389 1212 1260 1517 1522 1388 1340 1347 1501

Source CCC, MS Z/2/1/2. CCR 1422–29, 383 Coxe 1878, 22, ch. 152. BF, 109. NI, 9. Coxe 1878, 19, ch. 95. TNA, SC6/752/3 CCC, MS Z/2/1/1. TNA, SC6/752/3 TNA, SC6/752/3 CCC, MS Z/2/1/1. TNA, SC6/HenVIII/106 TNA, SC6/752/3 TNA, SC6/HenVIII/106 CCC, MS Z/2/1/1. CCC, MS Z/2/1/1 Mon. Ang., iii, 283a TNA, SC6/752/3 TNA, SC6/752/3 BF, 109. Coxe 1878, 10, ch. 117*(e). TNA, SC11/70. TNA, SC6/HenVIII/106 Coxe 1878, 20, ch. 126. CCR 1339–41, 436 CCR 1346–49, 268 Coxe 1878, 22, ch. 152.

226

13 RENTS AS INDICATORS OF THE MEDIEVAL SETTLEMENT PATTERN IN WALLINGFORD, BERKSHIRE David E. Pedgley Abstract Quit rents, available from surveys and borough rentals, and contributing to the fee-farm of Wallingford, provide an indication of the medieval settlement pattern in the town and its subsequent decline through death and emigration. Because quit rents remained largely constant, 19th-century records pinpointing property positions enable them to be mapped. Absence of quit rents in some streets is consistent with population decline shown by medieval parish absorption and by reduction in the feefarm payable. Borough rents, from the increasing number of borough leases (after the late 16th century) of lands that are likely to have formerly paid quit rents, provided added contributions to the fee-farm. Among the inhabitants of Wallingford in the late 12th century, following the granting to the borough of its royal charter in 1155, were many who paid quit rents, thereby releasing them from any services that might otherwise be due, in this case to the Crown. These rents were contributions to the borough fee-farm (Anglo-French feeferme), a perpetual fixed tax of £42 yearly (in effect, the price paid for the charter privileges) that may well have been a continuation of the Saxon landgabel or land tax. The earliest known list of inhabitants paying quit rents, arranged by street, comes from 1548, when a detailed survey of rents in the town was made by the royal surveyor Roger Amyce (Bodleian, MS Top Berks b 41). A similar survey was made in 1606 (TNA, E315/369/3). These two surveys, together with numerous borough rentals (annual records of rents paid by individual burgesses, collected by the borough bailiffs for the year ending Michaelmas, and surviving for many years from the mid-17th to the late 19th centuries; BRO, W/FRa) reveal changes in both the number of properties liable to payment and the total sums

Properties in: Fish Street High Street St Martin’s Street Castle Street Thames Street Goldsmiths Lane Total properties Total quit rent Average quit rent

1548 survey

1585 rental

33 30 8 5 10 1 87 264s 11d 3s 0d

27 24 7 7 5 1 71 206s 9d 2s 11d

paid. Figure 13.1 illustrates these changes using a selection of available years. Almost all these rents were payable on properties situated in streets wholly within the late Saxon town rampart (but in one or two cases associated with land outside this); there were only two rents wholly outside the rampart, most notably the 74s 8d for St John’s Farm (for locations of streets, see Figure 13.2). Almost all rents were paid on dwelling houses, suggesting that the few that were open spaces once had houses. For example, the only rent in Goldsmiths Lane was for the open space called Beansheaves, which perhaps had been occupied by the 14th-century mayor Thomas Benshef. Similarly, the close called Choles, between Wood and Thames Streets, may have been occupied by the 13th-century mayor Thomas de Cholsey. Another space between Wood and Thames Streets called F[r]obbins from at least the 15th century onwards may have been occupied previously by some unknown person with a surname like Fobbin. Such place-names would have perpetuated distant memories, like a more definite recent one – Bodington’s Place at the south end of Wood Street, a dwelling so called in the 16th and 17th centuries (although not subject to a quit rent) and no doubt named after its 15th-century occupier, mayor William Bodington. Quit rents were paid by the owners of a scatter of properties. Because the rents were largely fixed (although there are a few irregularities that seem to be associated with amalgamation and re-division of neighbouring properties), the pattern of their distribution was correspondingly fixed – even into the early 20th century, when such rents were abolished. Hence the 19th-century distribution of properties liable to quit rents should reflect the 12thcentury settlement pattern, once allowance is made for any loss of properties in the meantime. The rentals list owner-occupiers who paid the quit rents, or owners and 1606 survey 27 23 7 7 5 1 70 200s 3d 2s 10d

1678 rental 24 18 9 7 6 1 65 190s 3d 2s 11d

1733 rental 29 19 8 7 7 1 71 201s 9d 2s 10d

1799 rental 29 19 7 7 6 1 69 214s 1d 3s 0d

Figure 13.1 Number of properties by street liable for quit rents in a selection of available years, the total rent paid, and the average rent per property.

227

Figure 13.2 Distribution of properties paying quit rents in the 19th century. The dashed line shows the late Saxon rampart. Some dots represent more than one property. Diamonds are properties paying more than 7s. A St Peter’s Street B St Leonard’s Lane (image: author)

228

(different) occupiers, but they are divided at best only by street. However, it is possible to pinpoint each property within streets by using some 19th-century lists of occupiers that provide exact positions: the 10-yearly censuses from 1841; the tithe apportionments and maps of the 1840s (TNA, IR29/2/37, 135–7)); and a list of 1846 drawn up by (or for) an inhabitant named George Banbury (Wallingford Museum, 01–31-14–15). The national valuation of 1910 (ORO, DV XI/3) is a confirmatory later list. Among the difficulties that had to be overcome in this research were: the absence of house numbers until 1900, lack of strict geographical layout of households by some census enumerators, demolition of properties between censuses, and multiple occupancy. Figure 13.2 shows the distribution of properties paying quit rents in the 19th century. Fish Street (now St Mary’s Street) had the most whereas High Street had fewer, but in both streets the trends were similar to the trend of the total. In contrast, St Martin’s Street and Castle Street showed little change over time. There were no quit rents from St Leonards’ Lane, St Peter’s Street and Wood Street – this is consistent with a note made in the 1606 survey for Wood Street that the ‘buildings [were] now wholly decayed’, implying that previously there had been properties there perhaps liable to quit rents, and that this street had been contributing to the above-mentioned exodus relatively more recently than the other two streets. Such buildings are likely to have been on the east side of Wood Street because those on the west would have been considered to be in the rear of Fish Street properties – Wood Street was indeed called a ‘back lane’ at one time. A similar, but partial, decay seems to have been continuing in Thames Street, where the number of rents halved later in the 16th century; but decay seems already to have been advanced in St Leonard’s Lane, for no quit rents are recorded there. However, a possibility is that this part of the town was never subject to rents payable to the Crown because it had been part of the king’s estate in the town before the granting of the charter. Figure 13.1 reveals that in the mid-16th century the total number of quit rents was decreasing. This is consistent with appeals by the borough, repeatedly over the centuries, for a reduction in the fee-farm. It had been set originally at £42 yearly, but already by 1396 a petition from the borough to Richard II resulted in a reduction to a half; and in 1439 an inquisition found that there were then only 44 householders and that ‘not more than £10 can be paid as farm without final destruction and exile of the whole population’ (CPR 1436–41, 317). A memorandum accompanying a valuation in 1527 (TNA, SC12/5/66) states ‘by process of time the town had fallen in decay, decreased in wealth and prosperity’ such that in 1507 Henry VII had reduced the fee-farm to £15. By 1527 it was claimed the ‘town has fallen into such and so great a desolation, ruin and decay – as well in the ‘edificing’ and buildings as in rent and revenues – that much pity it is to behold and consider’. Following a further appeal, in 1562 a pardon for 40 years was granted by Elizabeth I to ‘the mayor, burgesses and commonalty of Wallingford

of £27 yearly, parcel of £42 yearly of the fee-farm of the said borough, from Michaelmas last, ... also pardon and release of arrears ... ; [for] it is reported that many of the burgesses and inhabitants are leaving the town because of the great burden of the fee-farm and that not more than £15 yearly thereof can be levied without final destruction of the borough’ (CPR 1560–63, 355). But Figure 13.1 shows that by the end of the 16th century numbers paying quit rent had stabilized at around 70. Even so, the heading to the 1606 survey shows continuing difficulty ‘whereof for the extreme ruin, poverty, spoil and waste of the mansion houses and inhabitants, the mayor, burgesses and commonalty have been released and discharged of £27 of the fee-farm of £42’. The decline may well have been exaggerated; nevertheless, it was real. This severe decline in the town’s fortunes from the 14th century or earlier, and the consequent reduction in ability to pay the fee-farm, is confirmed by the 14th-century absorption of some medieval parishes into neighbours because there were insufficient parishioners to support the clergy – notably St Michael and St John-super-Aquam in the south-east of the town (probably comprising largely Thames Street, Wood Street and St Peter’s Street) and St Ruald (or Rumbold) in the south-west (probably largely Goldsmiths Lane). St Michael’s parish united with St Peter’s in 1374 because it was ‘as it were desolate and destroyed for lack of rule because no-one has cared to undertake the cures thereof for a great time owing to its excessive poverty’ (CPR 1370–74, 436). St Ruald’s last known incumbent had been presented in 1353 (index at Wiltshire & Swindon History Centre). Poverty leading to this decline was attributed by John Leland in his Itinerary (Leland, i, 306) to the great pestilence in the time of Edward II – i.e, the Black Death – but it almost certainly involved other events, particularly the progressive decline in royal use of the castle (see KeatsRohan above, p. 101). The reduction of the fee-farm in 1507 to about one third of the original £42 implies a loss of about two thirds of the properties liable to pay any kind of rent – not just quit rents. If we assume that the proportion of quit-rent properties lost was much the same as for all properties, then we may ask where were these lost properties and what happened to their sites? It is possible that they were scattered among those properties continuing to pay quit rents but became open ground as dwellings decayed and no-one paid the quit rents. Other properties belonged to the borough. Although these were not subject to quit rents payable to the Crown they were subject to lease rents payable to the borough. Figure 13.3 shows the trend in the number of town properties liable to pay lease rents, using the same selection of available years as in Figure 13.1. Not included are small rents (typically 4d yearly) for porches built out from substantial houses into the street, and thereby encroaching on to town property. A few errors and ambiguities in the rentals (e.g., entries in wrong streets) cause some unavoidable lack of precision in the numbers, but the trend is clear: the town leases increased rapidly from none in the mid-16th century to around 50 by the mid-17th century, when they stabilized.

229

Properties in: Fish Street High Street St Martins’ Street Castle Street Thames Street Goldsmiths Lane St Leonard’s Lane Lock Lane Old Moor Lane Total properties

1548 survey

1585 rental

1606 survey

1678 rental

1733 rental

1799 rental

0 0 0 0 0 0 0 0 0 2

1 6 2 2 2 2 0 0 0 15

8 8 2 4 3 6 0 0 0 31

11 11 1 3 1 14 5 3 3 52

9 5 1 4 1 9 9 7 3 48

11 7 3 4 0 10 11 6 3 55

Figure 13.3 Number of properties by street liable for borough lease rents in a selection of available years.

Figure 13.4 Distribution of properties paying borough lease rents in the 19th century. The dashed line shows the Saxon rampart. Some dots represent more than one property. Squares represent the two private properties that were in the otherwise borough-owned strip of land outside the Saxon rampart and ditch (image: author).

230

Figure 13.4 shows the distribution of properties paying borough lease rents in the 19th century. There is a striking difference compared with the distribution of quit rents: notably the concentrations in the south-west quarter of the town (along Goldsmiths Lane), the south side of St Leonard’s Lane, and the narrow strip of land just outside the defences from the West Gate to the Thames. These are places that might be expected to be formerly part of the king’s estate and kept as open land (for ready access to the defences from both inside and outside) long after they were acquired by the borough at the granting of the charter. Because the heading of the 1585 rental states that lease rents also formed part of the borough fee-farm, it seems that the town decided, around the time of the successful appeal in 1562 for a reduction of the fee-farm to £15 yearly, to increase income by leasing more of its lands. Apart from filling some gaps in the principal streets (Fish Street and High Street), increasing numbers of leases were granted for land in St Leonard’s Lane and Goldsmiths Lane – in just those south-eastern and south-western parts of the town that had suffered most depopulation. Although no property in the 1548 survey is listed as being town-owned, it is likely that the then apparently private owners of properties subsequently known to be town-owned in 1606 were in fact town lessees in 1548 without being so described. This suggestion is supported by a few references to leases in the first council Minute Book, 1507–1649, the earliest noted being from 1526 (BRO, W/Ac1/1/1, f. 30r). Moreover, there is no evidence of what would be quite exceptional subsequent transfers of ownership to the town by sale or gift between the survey of 1548 and the survey of 1606. It is unfortunate that the heading of the 1548 survey does not indicate whether lease rents are included, but we may assume that there were in fact some (perhaps many) leased properties then and that the ‘0’ entered in Figure 13.3 is therefore suspect. Unfortunately, the borough lease book (BRO, W/ RTc 1) does not record the granting of leases before the

mid-17th century, but many of the properties in the strip just outside the defences had dwellings on them when first leased. The surrenders of any previous leases are not noted, as was commonly the case with later renewals, suggesting that the dwellings had been erected some time during the 17th century with borough consent and that formal leases were eventually agreed. One of the few remaining old buildings in this strip is the former ‘Row Barge’ beer-house, at the east end of St Leonard’s Lane, next to the churchyard. The oldest part dates apparently from the 17th century (David Clark, pers. comm.) – a onebay, two-storey, timber-framed structure – thus supporting the idea of 17th-century development of the strip, perhaps following the Civil War. Also within this strip are two properties that were privately owned (shown as squares in Figure 13.4) – the house opposite the mill on the corner of Fish Street and Old Moor Lane (now St John’s Road) near the South Gate (on the site of a property formerly belonging to the nearby medieval Hospital of St John the Baptist), and the ‘Cross Keys’ at the West Gate (on the site of a property formerly belonging to the nearby medieval Priory of Holy Trinity). In conclusion, a striking difference has been demonstrated between the distributions of properties liable to quit rents payable directly to the Crown and others liable to lease rents payable to the borough, although both formed part of the fee-farm. Whereas quit-rent properties are concentrated along the principal streets and probably reflect the medieval settlement pattern of customary lands, lease-rent properties are concentrated in areas that would long have been open land (required in part for ready access to defences; cf. Roffe, p. 199 above), although including a scattering elsewhere, and probably reflect the medieval holding of non-customary lands by the king before the granting of the charter. Five notable gaps on the north side of the High Street were occupied by properties paying rents to the bishop of Winchester (three), the College of St Nicholas in the castle (one), and the Bridge Estate (one).

231

14 THE IMPACT OF THE CIVIL WAR ON WALLINGFORD, CASTLE AND TOWN Judy Dewey Abstract In the 17th-century Civil War Wallingford Castle was renowned as a royalist stronghold, its last stand in 1646 defying the forces of Thomas Fairfax after all other major royalist strongholds had succumbed; ultimately, the ongoing threat posed by its proven strength was to lead to its total demolition. Substantial evidence of the impact of this conflict on the castle was revealed in the course of archaeological investigations undertaken by the Wallingford Burh to Borough Research Project (see Christie, Chapter 15 below); the aim of this paper is to complement the archaeological picture by an analysis of the varied contemporary documentary sources. In the context of the national conflict, it will reflect on the strategic role played by the castle and town, together with the immediate and long term effects of ‘the unhappy wars’ for the people of Wallingford. The historical context When King Charles I raised his standard at Nottingham on 22 August 1642 he was declaring war on Parliament and all who had challenged his God-given right to govern. The gentry of England could no longer choose to ignore the constitutional and religious convictions that had led to that critical moment. It was now inevitable that they would eventually have to take sides and face the consequences of their situation, wherever they lived and whatever others in their families might think; it was not an easy choice that faced many men of conscience. As with most wars, the causes of the English Civil War were complex. The roots on one side lay in a perceived threat to royal prerogative, both in politics and religion, and on the other in the fear that the lawful rights and liberties of the king’s subjects, enshrined in the power and privileges of Parliament, might be fatally undermined. The seeds were planted in the reign of James I and a brief background will set the scene (see also Davies 1937, for a general background to the period 1603–60). When James I accepted the throne in 1603 he linked his Scottish crown with that of England, becoming also Supreme Governor of the Anglican Church, despite his Scottish Calvinistic upbringing. It is important to remember that religion figured very highly in everyone’s life at this time; it was less than 40 years since Queen Elizabeth had executed James I’s Catholic mother, Mary Queen of Scots, and fought off the Catholic Spanish threat of the Armada. At the other end of the scale, Puritanism was growing in England, but James was not inclined towards its more extreme forms. While some Puritans

felt the need to seek religious freedom in the New World, many others were content to adopt their own simple way of worship, with a lifestyle that rejected frivolous worldly pursuits such as the traditional ‘holiday’ entertainments of dancing and drinking. James I’s successor, his younger son Charles, had a less tolerant attitude towards religion. Although tutored by a Scottish Calvinist, he was instead inclined towards the so-called ‘Arminian’ or ‘high church’ form of Anglican worship which favoured ritual rather than long, worthy sermons. His choice of Henrietta Maria, sister of Louis XIII of France, as his wife led to a growing fear of a renewed Catholic influence in England, particularly when she was permitted to practise her religion openly. Charles I’s appointment of the high-church Archbishop Laud led to further dismay among Puritans, especially in the Scottish Church (Gardiner 1962, xxvii). Such religious matters needed more careful handling and were a fundamental cause of division within society. Politically, as with James I before him, Charles found himself in constant need of money, but the solution depended on parliamentary co-operation in approval of taxation. The status of the House of Commons had developed under Elizabeth and its members were selected representatives from the shires and boroughs, many of whom had their finger firmly on the pulse of trade and the economy; without their backing, the collection of taxation could be problematic. Charles I saw himself as a divinely appointed king who knew what was best for his subjects. He resented the strictures of Parliament as set out in a Petition of Rights in 1628 which condemned valuable ancient practices such as billeting of soldiers, forced loans and enforced knighthood (Gardiner 1962, 66–70). It was in the king’s power to call and to dismiss Parliament, and when its members objected to his demands in 1629 he sent them away and ruled for 11 years without them. His subsequent means of raising money included the expansion of the Ship Money tax, originally raised from coastal towns to pay for ship-building for the defence of the realm but extended by Charles to become a tax on all counties, gathered by local officials. It was a difficult tax to enforce and led to further divisions between those who were obliged to collect the tax and those who felt that the principle behind it was flawed (Gardiner 1962, xxv; Dewey 1996, 10). Already, therefore, matters of conscience, personal beliefs and political lines were being drawn. Matters came to a head in 1640. Charles had gone to war with the Scots, whose National Covenant defied his attempts to enforce Laud’s High Anglican service in the Scottish Church. This so-called ‘Bishops’ War’ had proved

232

Figure 14.1 Map of royalist winter quarters between Reading and Banbury 1642-3 (image © Judy Dewey)

disastrous and with coffers empty Charles was driven to recall Parliament and demand support. In the ensuing Short and Long Parliaments, Charles’s chief advisor, the Earl of Strafford, became a scapegoat for the king’s misguided actions, being impeached and finally beheaded for treason. By then the lines of conflict with Parliament were being firmly drawn. Two main issues drove things to the final crisis: Parliament’s insistence that it should control the militia (the only permanently armed force in the country apart from the Royal Guard), and the king’s refusal to compromise on his high-church attitude, which many interpreted as a dangerous step towards a Catholic revival. In January 1642 Charles took his guards to Parliament with the intention of arresting five leading members, but they had been warned; ‘the birds had flown’ and war became inevitable. The king left London and both sides prepared their armies in earnest (Gardiner 1962, xxxvi–viii).

The first, indecisive, battle was fought at Edgehill, just south of Warwick, in October 1642 (Figure 14.1). The king attempted then to attack London but was thwarted at Turnham Green when the London Trained Bands – the well-drilled citizens’ militia – came out in force against him. His decision was to retreat and establish his court and administrative capital in Oxford, just 15 miles to the north of Wallingford. He entered Oxford on 29th October 1642, setting up court in Christ Church College. Queen Henrietta Maria joined him the following year, having first travelled abroad to raise money and equipment; she became established in Merton College (Eddershaw 1995, 45–72). Oxford was swiftly and effectively fortified, but it relied for its wider safety on a strong ring of royalist garrisons based in towns and fortified manors in Oxfordshire and Berkshire. It was in this capacity – as a reliable southern bulwark to Oxford – that Wallingford was to play its last major role as a royal stronghold (Figure 14.1). 233

Wallingford at war Apart from the considerable size of its ancient royal castle, Wallingford had available open spaces, barns and houses within the town that would prove to be an invaluable asset for accommodating troops and horses. The moated AngloSaxon ramparts surrounding the town on the north-west, west and south flanks, together with the Great Bridge over the Thames on the east side, all had additional defensive value against any Parliamentary attack. It was 500 years since the only other major English civil war had been fought. In that 12th-century conflict between King Stephen and the Empress Matilda, Wallingford, both castle and town, had played a significant role in the eventual success of Matilda’s son Henry II (Christie and Creighton 2013, 202–08). The town’s loyalty during the action had led to the granting of Wallingford’s charter in 1155, incorporating its ancient market rights and new privileges. To the men and women living in Wallingford in the 17th century, however, war was something distant, fought on foreign shores by mercenary soldiers. Lacking modern media coverage, there could have been no concept for the ordinary citizen of the reality of the death and destruction of warfare that threatened. People were used to the fear of death – as recently as 1631 16 people had died of plague in Crowmarsh and the Bridge Wardens had closed off the bridge to protect Wallingford (BRO, W/Ac 1/1/1, 114v 7 Dec 1631) – but the Civil War was to be a far greater shock to the townspeople than anything they had previously experienced. Statistics for the Civil War are hard to verify, but the casualties have been estimated as 85,000 direct deaths by military action and a further 100,000 deaths attributable to disease; that equates to over 4% of the British population of around four and a half million – at least twice the percentage of the UK population killed in World War I (War Office 1922, 237). It is also estimated that just under a quarter of all adult males served in the army at some time in the 17th-century Civil War (Porter 1997, 14). Just as in the 12th century, the people of Wallingford found themselves deeply embroiled in the conflict, but surviving contemporary documentary and archaeological evidence has enabled a more informed understanding of the impact of war on the population than for the earlier conflict. The documentary sources range from State Papers and Committee Proceedings from both the Royalist and Parliamentary records, to the more local official town records; however, the most illuminating survivals are the reports of spies and occasional surviving letters and diaries, most notably the journals and letter books of Sir Samuel Luke, the Parliamentary spy master. Wallingford the town in the 17th century In 1642 Wallingford was a modest market town with a population of around 600. A survey of the town in 1606 (TNA, E315/369/3, fols 101–23) shows that the main areas of occupation were the central streets – High Street, Fish Street (now St Mary’s St), St Martyns Lane (now St Martin’s Street) and Castle Street. In the backstreets,

Wood Street had buildings on the east side which were ‘nowe decayed wholye’, but there were a few tenements in Goldsmiths Lane. Thames Street had one substantial property and also a dyehouse and a tanhouse. The town had a guildhall, three hostelries – The Elephant, The George, and The Bell – all in the High Street (Figure 14.2), and five ‘shopps’. There were three surviving medieval churches serving the community: St Mary’s in the Market Place, St Peter’s near the bridge and St Leonard’s in the south-east corner of the town. All Saints (or All Hallows) in Castle Street had been out of use for over fifty years, since the dissolution of the College of St Nicholas in the castle which had formerly provided its priests. Also within the town walls were many pieces of open ground, gardens, orchards, and several barns. Only two gateways are named – the East Gate and the West Gate – but by the south entrance there were mills and beyond them, in Chalmore, the ‘capital messuage’ of ‘St Jones’ with a number of associated buildings – the former medieval Hospital of St. John. The town’s livelihood depended largely on its market, in which local produce from surrounding villages was sold and craftsmen, such as basket-makers, carpenters and smiths, plied their wares. Evidence from early 17thcentury wills and inventories suggests that there was still a small cloth-making industry in the town, of which the dyehouse in Thames Street was presumably part. However, the cloth trade had been badly affected when, for political reasons, the sale of English coloured cloth to Flanders was banned in 1612. Reading in Berkshire, just 15 miles south of Wallingford, showed a reduction in the export of cloths from 150 a week to just 40 because of the ban (Disbury 1998, 4). Wallingford’s output is likely to have been less than this but it is nevertheless one of the towns described as ‘brought to want’ by the ban (Newcombe 1940, 279). Wallingford had some trading links with London and Oxford by river. Malt (processed barley for the brewing industry) and many other goods were carried by barge. Navigation to Oxford had improved by the 1630s with the building of new pound locks at Iffley and Sandford (Philip 1937, 153–4). However, the narrowness of Wallingford’s medieval bridge arches prevented the passage of barges wider than 16ft 4in (5.0m), and so goods had to be transhipped to smaller vessels to go further upriver (see Figure 7.49). In 1633 some of the bargemen operating smaller boats above the bridge were described as ‘poor’ (CSPD ii 1633/4, 168). Charges were high for transport upriver of Wallingford and slow because there were insufficient bargemen. In all, therefore, in 1642 Wallingford was not a wealthy town and its assessment for the Ship Money tax demanded by Charles I in 1636 testifies to this rather impoverished status compared with other Berkshire towns: Wallingford was assessed at only £20, whereas Reading was £220, Newbury £120, and Windsor and Abingdon £100 each (Disbury 1998, 6). The choice of Wallingford as a key royalist stronghold in late 1642 meant that the townspeople, wherever their true sympathies lay, found themselves living under the flag of the king. Not all would have been happy at this turn of events. 234

Figure 14.2 The George Inn (photo: Judy Dewey).

Figure 14. 3 William Cooke’s malthouse. (photo: Judy Dewey)

For example, in August 1642 two Wallingford aldermen, John Bannaster (a former mayor) and John Day, had gone so far as to submit a complaint to the House of Commons about Richard Pauling, the rector of St Mary’s, for ‘very reproachful and scandalous Words spoken by him against the Parliament’ (Journal of the House of Commons, Vol 2, 743). John’s son Paul can also be demonstrated to have had sympathies with Parliament and the same is true of several other prominent burgesses. The earlier years of Charles I’s reign had already caused difficulties for Wallingford and potential differences of attitude are discernible from entries recorded in the town Minute Book (BRO, W/Ac 1/1/1), an important source for this period. The record for

1626 described it as a ‘troublesome year’; forced billeting of 48 soldiers in the town for 16½ weeks eventually led to a gibbet being set up in the town centre as a threat against disorder – a device that fortunately ‘bred a great deal of peace’. Furthermore, forced loans of up to 40 shillings, demanded from certain burgesses by the king in the same year, were paid by most, but notably refused by a few, including Ambrose Cottrell, Thomas Savage and Henry Kersall, who were subsequently fined the greater sum of £5. Yet all these objectors had held the elected office of mayor of Wallingford in the years before the Civil War, and Henry Kersall was actually in office when Wallingford became a royalist garrison. (Interestingly, his next appointment as

235

mayor came in 1647 after Wallingford had surrendered to Parliament and William Loader had been removed from the mayoral office for his royalist sympathies.) Another notable local man was the successful London brewer Walter Bigg. He had been born in Crowmarsh and had relations who ran The Bell public house in Wallingford. He presumably bought the malt for his brewery from Wallingford; this can certainly be demonstrated in the 1650s when he traded with William Cooke, maltster of Wallingford (Figure 14.3). But during the war, Walter Bigg fought against the king, rising to the rank of major in the parliamentary forces. River trade, on which the livelihood of men like William Cooke would have depended, was much disrupted by the war. There was to be little commercial trade on the Thames in those years and even that was hindered by searches for goods and ammunition that might help the enemy (Philip 1937, 155). The military of both sides, however, made use of the river as far as possible, especially for moving heavy armaments. Some notice exists in parliamentary spy reports of provisions coming from London as far as Henley, then being offloaded and taken to Oxford by road to supply the garrison. That may have been the case for Wallingford too, but the royalists worked hard to prevent supplies travelling downstream to supply parliamentary London (Disbury 1998, 157ff). Such details are an indication that not all Wallingford citizens would have been overjoyed at the arrival of the royal garrison; discretion must have become a watchword. Garrisons and defences The royalist military garrison was established in Wallingford by November 1642 under the governorship of Colonel Thomas Blagge (alternatively written as Blague or Blake) who was to serve throughout the war (Figure 14.4); his regiment of foot-soldiers had been raised in his home county of Suffolk. He appears to have taken command of both the castle and the town since, although a mayor of Wallingford was elected in 1642 and 1643, all other official records of town business disappear at this time. References to Wallingford Castle in the early 17th century reflect that it, like the town, was much impoverished by then. As early as 1540 it had been described as ‘sore in ruins’ (Leland, i, 119). The outer banks and ditches in the 1606 survey were also described as ‘altogether decayed’ and yet there were still some habitable buildings within the castle, probably some of those described by Leland as ‘goodly buildings’ within the inner bailey. There was also accommodation in buildings which had been the lodgings of the dean, priests and clerks of the former College of St Nicholas in the middle bailey. The latter now belonged to Christ Church, Oxford, but were leased to tenants. Military works began immediately and King Charles himself visited the town on 27th November 1642 (RCHM 1879, 443a). Storehouses and other buildings were constructed and munitions – powder kegs, bullets and match – were shipped in by river from Oxford (Dewey 1996, 17). By the spring of 1643 it is clear that a great deal of building work had been taking place during the winter

Figure 14.4 Colonel Thomas Blagge, Governor of Wallingford Castle throughout the Civil War (image: public domain).

months to refortify the castle. Although documentary records of such work are limited, there are certain clear statements within reports sent to Sir Samuel Luke, the parliamentary spy master who co-ordinated a network of men to gather details for him. These accounts would have been as accurate as the informants could make them and, when comparable to royalist sources of similar information, they are demonstrated to have been reasonably reliable. The recent excavations of the Wallingford Burh to Borough Research Project have significantly complemented the written sources by presenting rare archaeological evidence of the materials used, the location, scale and type of the building work and the effects of the final devastation on the surviving landscape (Christie, Chapter 15 below; Christie and Creighton 2013, 208–15). What follows gives a documented context to this work. Spy reports suggest that the initial defences were complete by early March 1643: on March 4th it was reported that ‘they have fortified the castle very strongly with double bulwarks’ and on March 8th that they ‘have now finished their works’ (JSL i, 21, 24). Recent excavations have shown that additional defences were constructed beyond the outer ditch of the medieval castle on the north side. (Figure 14.5) Perhaps these were the ‘double bulwarks’. Reports also suggest that the outer of two moats was water-filled while the inner one was apparently dry (JSL i, 87). The building materials required for such works appear to have been locally sourced: there is a record of the destruction of the vacant church of All Hallows (which lay in Castle Street, thus conveniently close to the castle on the west) to provide stonework for re-building (Richard Symond’s Notebook 236

Figure 14.5 The outer works of Wallingford Castle. Civil War gun emplacements have been excavated outside the medieval ramparts in the area to the bottom right of the picture and on the ‘barbican’ - the prominent D-shaped feature, middle right (image © TWHAS).

BL, Harl. MS. 965, f. 294) and local clay was used to form gun emplacements; the relatively crude quality of associated shelters for the defenders has illustrated the difficulties of rushed building work with poor materials (see Christie, below). Another report on 5th May 1643 revealed that ‘they have taken downe a great part of the castle and made it lower, and planted their ordnance on the toppe’ (JSL i, 69). Such work would have had the twofold advantage of providing gun platforms and releasing building materials for work elsewhere (Figure 14.6). Large storehouses and garrison buildings were also constructed (HMC 7th report: House of Lords, 96a) and, as noted earlier, the inner and middle baileys had enough good surviving or refurbished buildings and space to house a substantial garrison, ensure a secure prison, and also provide suitable and safe accommodation for royal and other high status visitors and their entourage. The castle was protected by ‘a doble drawbridge at the entrance into the castle and two drakes planted there upon one carriage’ (JSL i, 75). This is probably a reference to the well-documented drawbridges of the castle’s outer and middle gateways near Wallingford Bridge (see Keats-Rohan and Dewey, above) but it could be describing the long drawbridge that gave access to the main buildings of the inner bailey, which would have been protected by the gun platform discovered in 2011 (see Christie, p. 254 below).

The fortification work at Wallingford extended to the whole town, with the surviving late Saxon and medieval earthworks and ditches brought back into active use. The improved defences would almost certainly have entailed unpaid local labour, as was the case in Oxford where, as well as townspeople who proved highly reluctant to assist (Eddershaw 1995, 52), residents in halls and colleges between the ages of 16 and 60 had to work one day a week on the fortifications or pay a fine of 12d for each default (Carter and Stevenson 1974, 8). Troops likewise would have been set to work and oversee. The cost of the work, however, was not laid on the town alone since correspondence between Colonel Blagge and the mayor of Reading in June 1643 demonstrates that Reading and its neighbouring hundreds had been charged with raising contributions for the defences at Wallingford (RCHM 1888, 188, 215, 219). Building material for the town defences, such as wood for palisades, would have been hard to acquire and so houses (a mix of timber-framed, chalk and perhaps some brick construction) in Wallingford were destroyed as part of the process. Indeed 20 houses were pulled down ‘over against the castle to make the bulwarks, and to strengthen themselves’ (JSL i, 75, May 1643). From the description, some of these are likely to have been in 237

Figure 14.6 Clay base of gun platform excavated 2008.

Castle Street where, in 1606, there were 13 properties (two of which on the east side were ‘late buylded in the streete for the poore’; others could perhaps have been at the eastern end of the High Street on the north side where there were some 10 properties between the East Gate and The George. Based on the number of properties listed in the 1606 survey (which might have increased in number by 1643), this would have constituted around 13% of the habitable buildings in the town. Wherever they were, these were the residences of local people who would have been made homeless by the destruction. Their fate goes largely unrecorded but, before any military action had been experienced, they were already casualties of war. This destruction was only one impact on the local population. In early March 1643 a spy report details 1500 of the king’s horse and foot soldiers in and about Wallingford, with a further 600 of Prince Rupert’s royal troops in Cholsey, Hagbourne and Moreton (JSL i, 21 and 34). Wallingford, with its compact population, thus had to make room for more than double its number in troops who required accommodation either in tents, barns or billets. The horses, which required stabling, would have been a further problem. Both men and horses needed food which came at a considerable cost: the official estimate for the weekly outlay on a horse was seven shillings, and for a man three shillings and sixpence (Carter and Stevenson 1974, 10). The constant need for money, food, weapons and transport (notably horses and carts) for the army led to much scavenging in the local area. This saw groups of cavalry riding out to requisition goods as needed. Such expeditions were not without danger, both for the soldiers, who might encounter the enemy on similar forays, and for those from whom both sides were seeking supplies. People living in

villages and farms towards the Chilterns, beyond which lay Parliament territory, were particularly badly affected, experiencing raids from both sides in the conflict; we hear of men being imprisoned for refusing to supply goods because the losses threatened the livelihood of both them and their families. In Wallingford and the surrounding area houses were searched for weapons – pikes, halberds, muskets and even ‘birding pieces’ – which were taken away and brought into Wallingford Castle to arm the garrison, reportedly lacking in weaponry (JSL i, 66). To feed the army and to stock the castle in case of siege, all local ovens were being used to bake ‘biskett’ (a basic ration). If proper payment was made, then the local bakers may have done quite well from this. Similarly there was potentially good business for local smiths to mend weapons, and certainly for farriers and wheelwrights in the maintenance of horses and carts; the barber’s shop was also busy, a good place for spies to listen to gossip (JSL I, 76). Not all services were immediately paid for, however, and many probably never were; the sudden exit of a troop of cavalry or a regiment could easily mean debts were left behind and never honoured. Wallingford was not alone in preparing for war between November 1642 and April 1643; strong fortifications were constructed at Oxford to protect the king’s new court, and a number of towns, including Abingdon, Reading, Banbury and Faringdon, joined various great houses in fortifying as best they could in support of the king. Wallingford, with its castle, was a vital element in this ring of protection for Oxford. Indeed, when parliamentary forces, led by the earl of Essex, attacked and besieged Reading in April 1643, a relief force of musketeers from Abingdon and Wallingford was sent to assist the town. Wallingford became the forward base for the field-army with substantial supplies of powder, 238

bullet and match being sent by water from Oxford for safekeeping in the castle (ROP, i, 173). Wallingford Castle’s role as an arsenal would continue throughout the war. The relief of Reading proved to be too late and the articles of surrender were signed on 27th April (Disbury 1998, 46). There was then a strong expectation that the next assault would be directed at Wallingford. Accordingly, further supplies of ammunition were dispatched by water from Oxford, and convoys of powder, match and bullets sent by road under guard from Abingdon. Sir Samuel Luke’s spy reports in May describe scouting parties from the castle raiding the countryside to ‘bring in all manner of provision’ (JSL i, 71). Local village constables were ordered to ‘bring in victual and other provision because they know not how soon the town may be besieged’, whilst warrants were sent from Colonel Blagge as far as Newbury to demand that money previously paid into Reading should be diverted to support Wallingford (JSL i, 75). A report of 30th May said that ‘all the contry within seven miles of Wallingford are in great feare of the Cavallyers coming to plunder them whoe having already taken away their goods, and imprisoned their persons … they greatly desire that our forces would advance that way to secure them’. The supplies collected were considered by Luke’s spies to be sufficient for twelve months (JSL i, 76). More people and troops arrived in Wallingford, including many ‘papists and rich men’ who took refuge in the castle ‘and carried their goods, money and plate along with them’ (JSL i, 81). It is clear that Wallingford Castle was viewed as a safe haven. Troops were housed in four barns around the town and the townspeople were instructed that if an attack occurred, then men who could bear arms must go into the castle. The governor was obviously concerned that the town might not withstand an assault, despite the refortification works undertaken. By this time the castle had two pieces of ordnance (mortars) and two drakes (cannon)

on the east side, set to protect Crowmarsh Hill and the eastern approaches to the town. Wallingford’s bridge was made more secure by replacing four arches with wooden drawbridges which could be raised against the enemy – a substantial piece of engineering (Edgeworth and Christie 2011, 234–235). These measures also ensured that men of the garrison could be prevented from absconding easily, a growing problem as lack of pay and shortage of supplies led to discontent (JSL i, 73 and 76). On the north side, the castle works already noted were sufficient defence for the town gateway: cannon set on the platform there could defend both the gateway and the road to Oxford, while the house demolition close to the castle ensured an uncluttered view. Beyond the town’s West Gate more properties had been pulled down to allow a clear view of any approaching enemy and there is reference to great baskets filled with earth being set up on the ‘Brightwell side’ as a further defence, which would have been a barrier to attackers and good cover for the defenders (JSL i, 87). The southern approach to the town was a vulnerable area, having no surviving gateway (1606 Survey, TNA, E315/369, 101123), and yet this was the most likely point of attack for an army coming towards Oxford from Reading. Loads of lyme (probably limestone) were brought into the town to build a drawbridge over the town ditch by the mill at the site of the old South Gate, which was then defended with two ‘pieces of ordnance’ (JSL i, 73). (That this was a drawbridge is shown by a reference in the bailiff’s accounts of 1651 to mending ‘the drawbridge by the mill’: BRO, W/FAb1). To prevent the enemy having any shelter or cover in these southern approaches more buildings were destroyed; these most likely included the ‘capital messuage of St John’s’ (1606 Survey), its barns and possibly other buildings. The way was cleared by troops ‘that they can discern 2 miles forward’ (JSL i, 87–8) (Figure 14.7).

Figure 14.7 St John’s Green. This was the site of the medieval Hospital of St John, the buildings of which were pulled down during the Civil War to give a clear view of the southern approach to the town (photo: Judy Dewey).

239

Fear and manoeuvres in summer and autumn 1643 These southern preparations would have completed the circle of defence for Wallingford. All was then ready for the expected attack. The soldiers were reported by a Parliament spy to ‘pray for our comming thither, being (as they conceive) soe well provided and having soe many of the kings forces round about to assist them’ (JSL i, 74). Wallingford waited with growing impatience while rations diminished. There were rumours of the approach of Essex and even the spies could not get near the town: ‘the soldiers there intend to keepe close, and not to stirr out of the castle for feare his Excellency will march to them, having for 3 dayes together expected his comming, and that all the coast thereabouts is soe cleare that one cannot perceive soe much as a passinger on the way eyther on foote or horsebacke’ (JSL i, 91). Despite all the preparations, the anticipated attack never came. Many of the earl of Essex’s men at Reading had fallen sick, and it is clear that both sides were suffering a similar problem by the early summer of 1643, which affected the ability to wage war; indeed, this was a particularly virulent disease which is well recorded in both Berkshire and Oxfordshire, often associated with the army and with overcrowded towns. Studies of parish registers in 22 Oxfordshire parishes show the annual average numbers of burials rising by 217% in 1643, with figures for towns such as Thame and Henley being considerably higher than that. A contemporary journal by Thomas Wyatt, rector of Ducklington near Witney, notes that ‘There was a grievous sickness began in the autumn 1643 spreading; many died in Oxford, Witney and other places. Some called it the new disease [typhus] others morbus castrensis’ (Eddershaw 1995, 95–7). Nor was the sickness short-lived, beginning about June but continuing into the winter in some areas. In Oxford a real fear of disease, even within the queen’s residence, can be detected in the reporting, adding to the many anxieties of war and insecurity (Eddershaw 1995, 60). No records of this kind survive for Wallingford, which is unlikely to have escaped unscathed, though it is possible that the siege-like conditions may have helped protect the town from infection in the early stages. In June 1643 Essex finally began to advance towards Oxford, but he chose to avoid Wallingford, moving instead towards Thame and Islip, which led Prince Rupert to counter-attack with a dawn raid on Chinnor where some of the parliamentary soldiers were resting. The ensuing action led to the battle at Chalgrove on 18th June, just a few miles north-east of Wallingford, in which about 1000 of Prince Rupert’s cavalry engaged with some 500 parliamentary cavalry and many more dragoons, led by Major John Gunter. It was a royalist victory, most remembered for the mortal wounding of the renowned parliamentary leader, John Hampden, who died of his wounds six days later in Thame (Stevenson and Carter 1973, 346–56). Prince Rupert withdrew safely to Oxford; the immediate pressure on Wallingford had been relieved. In the summer and autumn the number of troops stationed in Wallingford fluctuated as they were called to assist campaigns against Bristol and Gloucester where the main action was now

focused. A garrison of 200–400 men is recorded (JSL i, 127, 130, 136, 154), but when the army was on the move there were often many more soldiers passing through or resting en route. 1643 had been a good year for King Charles and he visited Wallingford on several occasions during it. A royal visit had been a rarity since medieval times, and the anticipated arrival of the king in April had seen country people waiting around for three days, just to catch a glimpse of their monarch (JSL i, 56). On that occasion he made an inspection of the new defences, dined with Michael Molyns, owner of the manor of Clapcot and Rush Court, which lay to the north of the castle, and stayed overnight in the governor’s quarters in the castle. In early October 1643 Charles, Queen Henrietta Maria, Prince Rupert and Prince Maurice all dined at Wallingford Castle, celebrating the christening of one of Colonel Blagge’s children, for whom the Queen and Prince Rupert were ‘gossips’, that is godparents (JSL i, 162, 166). Such an august gathering testifies to the perceived safety of Wallingford Castle and to the existence of satisfactory royal accommodation within its walls. It also opens a window on the life of a garrison where the governor’s whole family was clearly in residence. War and Wallingford in 1644–46 Times were not always so amicable between King Charles and Colonel Blagge; an incident early in 1644, in which Wallingford was required to supply 200 soldiers to the king for the campaign against Gloucester, highlights the problem of retaining a loyal garrison without proper resources. Blagge, called before the king, expressed much concern for the safety of Wallingford if robbed of its garrison, stating that there had been no money for pay and many soldiers had already deserted. The king was furious at the response ‘and struck him twice over the head with his cane’. It was a low point for Blagge, compounded a few days later when he inspected his precious siege stores in Wallingford and discovered that 40 hogshead (large casks) of beef were rotten ‘and stunke and were not fit to be eaten’. The garrison was then forced to search the neighbourhood for stores of bacon as a replacement. They were also encouraged to find cattle to bring into the castle in lieu of their pay (JSL i, 255). Meanwhile, Blagge forced the townsmen to do unpaid duty as watchmen at night to ensure the safety of the castle and town. It would have been in their interest to comply. Despite occasional difficulties in maintaining the appearance of a secure garrison to the prying eyes of enemy spies, Wallingford continued to be perceived as a major stronghold, its medieval ramparts and added cannon emplacements presenting a daunting prospect to any attacker. It was clearly considered to be a secure prison; there are numerous examples of the incarceration of those refusing to supply goods for the king, including William Brackston, mayor of Reading in 1644 (RCHM 1879, 219–20). Others were held after capture in action whilst awaiting an exchange, such as Lieutenant Evans in 1644, who reported his experience (LSL, 381 no. 915); in 240

Figure 14.8 Remaining Royalist strongholds 1646 (image © Judy Dewey).

the same year 60 men of Sir William Waller’s parliamentary force were taken by Lord Hopton and imprisoned here (LSL, 648 no. 102). This custodial role of the castle was to continue even when it finally came into Parliament’s possession and followed a long medieval tradition of use of the castle as a prison (see Keats-Rohan, above , pp. 51, 82). The presence of the Wallingford garrison remained a constant threat to movements of parliamentary troops throughout the war; spies report the activities of raiding parties, such as one on Henley in November 1644 which returned to Wallingford bringing ‘some horse into the town and four carts laden with plundered goods’ (LSL, 676, no. 193). Raids were widespread: an account from a Buckinghamshire householder described how ‘nine or ten soldiers from Wallingford came to her house and after remaining an hour seized her husband William Lydgall, and five horses and three more from John Foord a neighbour of hers with pistols, swords and a fowling piece...all of which they carried away to Wallingford where her husband remains a prisoner till the arrears of all such taxes they pretend to be due from that parish be paid and £10 more to the party that fetched him away’ (Bodleian MS. Nalson 5, 16 Dec. 1645). Both sides kept up

a continuous flow of similar actions as pressure increased on Oxford. Abingdon was abandoned by the royalists in May 1644, allowing Parliament a good base from which to harry Wallingford and Oxford (Eddershaw 1995, 116). Major-General Thomas Crawford made a daring raid on Wallingford’s fresh supplies by stealing 1000 sheep from the fields of Clapcot, close to the castle (Whitelocke 1732, 115–6). Colonel Blagge retaliated with a group of 120 men, leading a personal vendetta against Crawford, even fighting him in hand-to-hand combat. He appears to have come off badly in this encounter, escaping with his life but being wounded and losing most of his men, injured, killed or taken prisoner (Whitelocke 1732, 126). The governor proved notable for his involvement in such skirmishes. Anthony Wood, dining as a young man in Crendon near Thame in January 1644, had witnessed an earlier action: Colonel Blagge, with some 70 or 80 Wallingford horse, had been ‘roving about the countryside’ when he encountered 200 ‘rebels’ from Aylesbury’s parliamentary garrison. Blagge and his troop were greatly outnumbered, but the colonel and his captain, Walters, fought a valiant rearguard action which allowed all but 18 men to escape, two being killed and 16 taken prisoner. Wood describes 241

the 50 or so retreating horsemen thundering past the house where he was dining, ‘Blagge with a bloody face’ being hotly pursued by Colonel Crafford, ‘discharging his pistol at some of the fag-end of Blagg’s horse’ (Clark 1891, 114). Much of the local action in the war seems to have been waged in close-quarter skirmishing and raids, reported only incidentally by observation and spy activity. Nevertheless, it is clear that the castle remained a focal stronghold for the king and an invaluable staging post in many of the more serious campaigns. In May 1644 the King Charles, Prince Rupert and Lord Hopton were at Wallingford Castle from which they rode out to view the royal army of some 15,000 horse and foot en route towards the slighting and abandonment of Reading (Disbury 1998, 63). At that time it was again feared that Wallingford might be besieged and the town had only a small garrison to protect it. The attack was expected from eastern approaches, so precautionary measures were taken, such as positioning carts across the street in Crowmarsh, together with felled trees, hoping to hinder enemy cavalry by blocking the road. But, once again, the threat was never realized. In September 1644 Wallingford was the rallying point from which a successful campaign was launched to relieve Basing House, Hampshire. A month later, it became a refuge for wounded and retreating troops, after the second Battle of Newbury in October 1644. By then Abingdon, Henley and Reading were all in parliamentary control and Oxford was under considerable pressure. Wallingford remained critical as the prime southern defence to Oxford. (A very useful compilation of the complex activities in Berkshire at this time can be found in Disbury 1998.)

In the following year, 1645, the efficiency of Oliver Cromwell’s New Model Army, under the command of General Sir Thomas Fairfax, began to have an impact on the war. The campaigns that year avoided Wallingford but included the first siege of Oxford and the fiercely-resisted attacks on Faringdon and Boarstall. The Battle of Naseby on 14th June tipped the balance of power against the royalists, with long defended strongholds such as Bristol, Winchester and Basing House finally succumbing by the autumn. In November the king returned to the security of Oxford, now protected by a diminishing ring of outposts of which the castles of Banbury to the north, and Wallingford to the south were the most resilient (Figure 14.8). Hope that reinforcements, in the form of new recruits from Wales and the west of England, would reach the king were dashed in March the next year by a defeat at Stow-on-theWold. On 27th April 1646, King Charles escaped from Oxford, disguised as a servant, just days before Fairfax laid siege to the city on 2nd May. Oxford capitulated only after lengthy negotiations had achieved reasonable terms, and the royal garrison began to leave on 24th June. Colonel Fairfax ensured that order was maintained and that important buildings of the university and town were unmolested (Eddershaw 1995, 148–53). Fairfax and the siege of Wallingford Fairfax turned on Wallingford as soon as the Oxford siege began, mustering all the strength he could afford, despatching Colonel Weldon on 4th May to lay siege to Wallingford on the north, west and south of the town, and

Figure 14.9 Re-enactment of the siege by the English Civil War Society, June 2015.

242

ordering the garrison at Henley to close up the Crowmarsh side on the east. Captain Gibbons of Henley eventually responded, engaging with a small group of the Wallingford garrison on his unexpected arrival in Crowmarsh, but, although a few men were taken prisoner, the rest escaped back across the bridge, where the drawbridges must have been hastily raised behind them. So began the final siege of Wallingford, which was to drag on for twelve long weeks. (Figure 14.9) There are surprisingly few descriptions of the action, but there can be little doubt that there was plenty of it. The influential parliamentarian, Bulstrode Whitelocke of Henley, described in his diary the fear that he and a visitor experienced at Wallingford during the siege ‘att the Great Gunnes which the castle lett flye att them when they passed by’, but which, thankfully ‘did no hurt to them’ (Spalding 1990, 188). If the siege at Wallingford was anything like the earlier sieges at Banbury and Faringdon (Eddershaw 1995, 136–7, 146) there would have been many such attacks from both sides. Nonetheless, Wallingford’s fortifications were a formidable challenge to Fairfax and Cromwell, and Colonel Blagge was not a man to give up easily. Firstly, he refused to consider surrender until he had received express permission from the king himself to do so and even then he insisted on generous terms being offered. Blagge’s correspondence suggests that he was fully aware of the vulnerability of Wallingford’s town defences in the event of a full-scale attack; he had set guards within the town but sent word to Fairfax that if the Parliamentarian army tried to enter the town he would set fire to it and withdraw to the castle (Rushworth 1722, 276–98). This was surely a very tense time for the inhabitants of the town, uncertain of their fate and that of their properties. On 10th June the king sent a direct order, signed at Newcastle, to Blagge (and other commanders): ‘Having resolved to comply with the Desires of our Parliament in every thing which may be for the Good of our Subjects, and leave no Means unessay’d for removing all Differences amongst us: Therefore we have thought fit, the more to evidence the Reality of our Intentions of settling an happy and firm Peace, to require you, upon Honourable Terms, to quit those Towns, Castles, and Forts intrusted to you by us, and to disband all the Forces under your several Commands’ (Rushworth 1722, 249–76). On 27th June, probably soon after receipt of this order, Blagge finally agreed to open negotiations to hand over ‘the strongest fort in England’ (Rushworth 1722, 276– 98). The ‘Honourable Terms’ were eventually laid out in printed ‘Articles of Agreement’ (BRO, W/Z5), which had required much negotiation before Blagge would accept them (Figure 14.10 and Appendix). By then his garrison in Wallingford Castle was at breaking point – the war was effectively over, they wanted the siege to end and they had not been paid; they would have been well aware that once the castle was surrendered there was no likelihood of receiving any arrears. It was later reported that: ‘Colonel Blagge sold the corn and provisions there, and put the money in his own purse, without giving one penny of it to the soldiers, though much pay was in arrears to them, and they were so much discontented at the greediness of

Figure 14.10 Title page of the printed arrangements for the surrender of Wallingford Castle.

their governor, that they fell into a mutiny’ (Whitelocke 1732, 217). The truth of this cannot be verified, but things were certainly so tense that the agreed date of surrender, 29th July, was brought forward to the 27th at Blagge’s request and a parliamentary regiment occupied the town to ensure the royalist garrison made an orderly departure (Rushworth 1722, 276–98). For the townspeople the sight of the governor, his officers and the soldiers of the garrison marching out from the castle ‘with flying colours, trumpets sounding and drums beating’ (BRO, W/Z5) must have been a moment of mixed relief and fear; they cannot have known for certain that the agreed terms would protect them or their town from destruction. A thousand foot soldiers ‘besides horse’ departed from the castle, leaving behind ‘fourteen pieces of ordnance, store of arms, ammunition and some provisions’ (Whitelocke 1732, 217). The records are silent as to how well the takeover by the new parliamentary garrison was managed. After the surrender: demolitions, damages and disputes Twelve weeks of siege must have taken its toll on the town, but the surrender was by no means the end of Wallingford’s traumas, since military occupation continued under the new parliamentary governor, Major Arthur Evelyn, until the final demolition of the castle in 1652. The decision that Wallingford Castle ‘shall be slighted and dismantled’ was 243

made as early as February 1647 (Journal of the House of Commons, February 1646/7, 96), and yet for six years the castle continued in use as a prison and a busy munitions centre. The demolition order was finally issued on 18th November 1652, with Major Arthur Evelyn, Mr Edward Jennings of Long Wittenham and Mr John Rusden, given responsibility for carrying out the work (Dewey and Dewey 1996, 39). It was thoroughly done, with only a few buildings in the middle bailey, most of which belonged to Christ Church, surviving. The repercussions of this destruction were to be felt for centuries (see Dewey, Chapter 7 above). The effects of parliamentary control after 1646 are evident in the town records. To ensure the removal of strongly royalist personnel, the mayor, William Loader, as noted above, was dismissed from office and replaced by Henry Cursell whose sympathies were more amenable to the new government. The churches, too, were affected: Robert Fortie, rector of St Mary’s, was ejected from his living, together with Daniel Letsham of St Peter’s, and replaced by others who were presumably of Puritan conviction. It is interesting that John Astell (Austell) of St Leonard’s survived in post until 1652, when Thomas Pinkney took over as rector of the newly united parishes of St Mary, St Leonard (with the chapel of Sotwell) and St Peter (BRO, Borough Statute Book, W/AC1/1/2, 1650). By 1656 Cromwell’s decrees had endorsed the combination of yet more parishes with St Mary’s: All Hallows, St Nicholas (the church in the castle precinct) and Crowmarsh Gifford (CSPD iii May 1656, 311). Under Cromwell church marriages were banned, and so, from 1654–58, all Wallingford marriages recorded were civil ones. Furthermore, Puritan views were regularly endorsed on Fridays by public lectures by well-known speakers in the Market Place (BRO, Borough Statute Book, W/ AC1/1/2, 1650). All these changes were to be reversed within two years of the Restoration of the monarchy in 1660. It must have been an extremely divisive time for the people of Wallingford, coming to terms with true loyalties, royalist or parliamentarian, finally exposed by changing circumstances. The mayor of Wallingford during the Commonwealth period under Lord Protector Oliver Cromwell was William Cooke, the maltster whose trade with Walter Bigg in London would have been severely disrupted by the war and whose sympathies certainly lay with the new regime. Several other burgesses can be identified with similar leanings; these are the men who, after 1660, found themselves being forced to sign an oath of allegiance to Charles II if they were to continue in office. Many refused, revealing more clearly how loyalties in the town must have been silently divided during the years of the war. Some of these men, such as William Cooke, never returned to public office. Others, and their sons, do drift back into the town records as time passes, laws are eased and the memories of past troubles begin to fade. Assessing the extent of damage caused to property within the town during the war is difficult. Recovery was probably achieved only gradually, especially given that Wallingford was not a prosperous and populous town in

Figure 14.11 St Mary’s church tower, constructed with re-used castle stone in 1653 (photo: Judy Dewey).

Figure 14.12 Stone on the south-western buttress of St Mary’s tower (photo: Judy Dewey).

the 17th century; much of the Civil War damage must lie hidden behind the later facades of more prosperous times in the 18th century. Three of Wallingford’s four parish churches had been destroyed or severely damaged 244

during the war, only St Mary’s seemingly surviving unharmed. This church was actually enhanced in 1653 when stone from the demolished castle was purchased to help rebuild the church tower which had been struck by lightning in 1638, before the war began. Interestingly, a commemorative stone on the south-west buttress of the tower is carved ‘Will Loader 1653’, the former royalist mayor, a gentleman who was presumably in some way instrumental in funding the rebuilding (Figures 14.11 and 14.12). All Hallows had gone by 1643, destroyed at the hands of the royalists: a description of Wallingford churches in 1644 by Richard Symonds gives the date for its destruction as 1643, adding that the church had been ‘unused these 60 years’. Significantly, he notes no damage to any of the other churches at that date (Richard Symonds’s Notebook, BL, Harl. MS. 965). The implication has to be that the severe damage to St Leonard’s and St Peter’s, recorded after the war, must been caused by the final siege in 1646 or by subsequent military occupation. In fact, the churchwardens of St Leonard’s, writing to the bishop of Salisbury in 1662 explained that: ‘our church by the wars was much defaced, the seats and our three little bells were carried to St Mary’s aforesaid, and the said bells amongst others were melted or cast into six bells now hung up there. And afterwards our church was turned into a stable house’ (WRO, D1/54/1/5,1662). Skirmer, writing in his History and Antiquities of Wallingford, c. 1712 (BRO, D/EN/Z8), adds that: ‘it seems to have been formerly a larger Church and the colour of the stone pillars plainly shew it to have been burnt which pillars also shew that it hath been much wider’ (Figure 14.13).

Similarly, for St Peter’s, the churchwardens in 1671 reported that ‘Our church was pulled down and destroyed in the time of the late wars’ (BRO, Churchwardens’ presentments to the bishop of Salisbury, D1/54/4/2, 1671). While its churchyard saw some continued use, the parishioners had to go to St Mary’s to worship; there was resentment lasting into the 18th century that church taxes had to be paid by parishioners who had no church. It was 50 years before the over-crowding of the single surviving church of St Mary’s was eased by the re-opening of St Leonard’s in 1704 after significant repairs, but it was not until 1769 that St Peter’s was finally rebuilt (Figure 14.14). Records survive of a few citizens receiving compensation for their personal losses by making claims to the Commonwealth and later to Charles II. Henry Knapp, whose house in the castle had been used for over six years by the parliamentary governor, from 1646–52, was awarded £125 by order of the Council of State in January 1652/3 (though it was not actually paid until August 1654); the property itself survived the demolition (Dewey above). John Freeman, who was a tenant of the castle when the war began, claimed £300 for war damages in 1660, complaining that his gardens, orchards and ‘ancient walls’ had been destroyed during the castle’s demolition. He also revealed strong animosity towards the mayor, William Cooke, requesting that he be ‘excepted out of the Act of Indemnity’ because he had ‘without any pretended order carried away a great part of the stone and other materials’ (RCHM 1879, 96a, no. 10). In 1663 Robert Browne claimed compensation for the destruction of his house near the castle, worth £100, which had been demolished by

Figure 14.13 St Leonard’s church. The tower was added in the 19th century during refurbishment of the church. Damage in the Civil War prevented use of the church for worship for over 50 years (photo: Judy Dewey).

245

Figure 14.14 St Peter’s church, rebuilt in the 18th century on the site of the medieval church destroyed during the Civil War. The churchyard, however, remained in use throughout (photo: Judy Dewey).

Colonel Blagge for materials to re-fortify the castle; these materials were then sold ‘when the castle was demolished in Oliver’s time’ without any compensation being given to Browne (CSPD iv April 1663, 116). Rather different was the 1663 petition of widow Dorothy Thompson of Wallingford ‘for recompense for her furnishing the late king’s army with beds in Wallingford Castle and keeping the soldiers half a year at her own cost’. Widow Thompson was promised that she should ‘have the very next almsroom that becomes void at Ewelme’ – most likely a great relief for her (CSPD iv May 1663, 148). The most interesting claim, and by far the most protracted, was made by Michael Molins (or Moleyns, or Mullins) of Clapcot. The Molins family had been in possession of the manors of Clapcot and Rushe, to the

north of Wallingford Castle, since the late 16th century. Lack of funds had forced Michael Molins to mortgage his property in 1635 for £2000 to Nathaniel Brent, warden of Merton College Oxford. When Brent tried to foreclose the mortgage in 1640, Molins refused to give up his property and Nathaniel Brent started proceedings in Chancery against him. The exact location of Molins’ manor house in Clapcot is unclear, yet it was probably there that he entertained the king in 1643 (see above). In 1644 Richard Symonds described it: ‘Upon the North side of the Castle is a pretty fair howse called Clapcott the seat of the Molyns, it was in the p(ar)ish of Allhallows, now St Peter’s parish. ‘Tis called Villa de Clapcott.’ The parish location (especially the St Peter’s parish reference) is particularly difficult to pin down but it must have been close to the castle. The date of this reference reinforces the view that the property itself survived the initial destruction of houses for materials (though Molins may have owned other houses in All Hallows), but he later claimed to have lost timber valued at £450 during the re-fortifications. His losses as a result of the war escalated soon after when his house was destroyed by fire, probably in an unrecorded enemy action, a disastrous event for a man already facing Chancery proceedings for debt. Molins petitioned the king early in 1644 and was apparently promised other land as compensation. This promise had not been fulfilled when Molins’ petition was reinforced by testimony from the castle governor, Colonel Blagge, and the mayor of Wallingford, John Smythe, who stated that the ‘Petitioner has suffered in his houses and lands at Clapcot, Berks, to the value of £3,000, at least, by reason of the Wallingford garrison, for which his Majesty promised satisfaction in lands in his forest of Shotover, Oxfordshire. He prays warrant to the Lord Treasurer for perfecting the exchange.’ They added that: ‘he has been damaged £1,500 since his first petition and that upon the view of carpenters, masons, farmers and others we find the petitioner’s damages amount to £4,026’ (CSPD ii 1644, 50). Despite this support, the compensation remained unpaid, the king was defeated, and by 1648 Molins was being hounded for his debts. He took refuge in the castle, held then by Major Evelyn for Parliament; Evelyn was ordered to give him up to justice (CSPD iii June 1649, 241). It is unclear what happened next, but in 1652/3 it was agreed that the Molins case should be re-examined and that if it was justified, he should be given compensation from the proceeds of demolishing and selling off stones and materials from Wallingford Castle. Once again, nothing happened, but Molins continued to petition. Major Evelyn was ordered to account for the monies raised in destroying the castle and a report in 1657 agreed that Molins should be paid £450 from the demolition money (CSPD iii December 1657, 207); the accounts however had still not been produced. When they finally were, the money raised by sales of stone etc. was stated to have been £516.17s.11d, the expenses £450.5s.8d, and the money left for Molins a mere £66.12.3d (CSPD iii March 1658, 356). This was ordered to be given to him, but it was not paid before the Restoration of the monarchy in 1660 once more transformed the situation. 246

Molins now had to start the petitioning process all over again. His determination (or was it simply desperation?) must be admired. Finally, in December 1660, the Treasury accounts record a payment of £100 to ‘Michael Molins, gent, for his great sufferings and the damage to his estate about Wallingford, whilst that was a garrison’ (Calendar of Treasury Books, December 1660, 106). This was presumably in lieu of the £66.12s 3d with a bit extra, but it was not the end. Molins still pursued the greater sums: the surviving papers are endorsed as ‘Received 5 Dec 1661, Mr Michael Molyn’s account certified to be in all £4,000, for which he desires a Privy Seal’. Two warrants for payment by the Treasury of £2000 were at last issued: on 29th January 1661/2 ‘as a gift in consideration of great losses’ and on 6th February 1661/2 ‘without account’ (CSPD iv February 1662, 257 and 267). The Treasury records, however, show only a single payment of £2000 on 22th February 1661/2 ‘to Michael Molins for his damages etc. by occupation of the garrison kept at Wallingford during the late distraction’ (Calendar of Treasury Books, February 1661/2, 366). It was too little, too late. Sadly, despite all his efforts, Molins had been unable to pay off his mortgage and the Brent family had finally acquired the Clapcot lands in 1660. We can only wonder how many others in Wallingford suffered a similar fate, unrecorded, uncompensated and probably on a much smaller scale, but no less devastating, especially to the poorer townspeople. Conclusion The Civil War and its aftermath marked the end of an era for Wallingford. It had been a time of considerable hardship for those living under military regimes for some 18 years. It had caused divisions in families, divided loyalties in the community, brought destruction of property and disruption of trade. For a few individuals, supplying the needs of the

garrison and billeted soldiers may have brought some profits and for those elevated to power under the new regime, like William Cooke, there may have been some moments of recompense, but this was soon replaced by reciprocal animosity and demotion. As a royalist bastion Wallingford had proved its worth. The war seems to have engendered a revival of the perception of Wallingford as an impregnable royal stronghold, even though in truth it was actually extremely vulnerable at times. It could perhaps be argued that Wallingford Castle’s past history as a royal seat, and its town as a privileged royal borough, had acted as a deterrent to attack during the war. Despite the vigorous assaults on other local towns and strongholds, such as Reading, Abingdon and Newbury, it was only when all other such places had succumbed, including the king’s seat of government at Oxford, that the parliamentary army felt able properly to tackle Wallingford. Even then, the true vulnerability of the garrison, whose commander was facing a near mutiny from his unpaid troops just before the final surrender, seems not to have been recognized. The total demolition of the castle brought a final end to the town’s historic role as a royal stronghold which had its origins in the 9th or 10th century. Although the Crown retained possession of the castle site until 1817, its later history, though fascinating, is not of national significance (see Dewey Chapter 7 above). Charles II renewed the town charter in 1663, but Wallingford was by then a simple market town struggling to rebuild. The town documents of the later 17th century include references to ‘those late unhappy wars’ which thankfully never returned. The present Town Hall – newly built in 1670, funded with difficulty by loans from the burgesses, timber-framed but coated in stucco and moulded to look like the fine stone they could not afford – is a remarkable and lasting symbol of Wallingford’s determination to recover and survive. (Figure 14.15)

Figure 14.15 Wallingford Town Hall (photo: Judy Dewey).

247

Appendix: Extract from the printed Articles of Agreement for the handing over of Wallingford Castle, 25 July 1646 (BRO, W/Z5). Spelling and punctuation as printed. Articles of Agreement concluded and agreed by his Excellency Sir Thomas Fairfax Knight, Generall of the Forces raised by the Parliament, on the one part, and Colonell Thomas Blagge Governour of Wallingford, on the other part. For, and concerning the rendring of the Garrison of Wallingford Castle and Towne, (viz.) 1 That the Castle and Town of Wallingford, with all the Ordnance, Arms Ammunition, Stores and Provisions of Warre thereunto belonging, shall be delivered up without wilfull spoile or imbezzlement, unto his Excellency Sir Thomas Fairfax, or such as he shall appoint to receive the same, upon Wednesday the 29 July 1646. in such manner, and with such exceptions as are contained in the ensuing Articles. 2 That on the said 29 day of July, the Governour, and all Officers and Souldiers of the Garrison, with all other persons therein (that will) shall march out of Wallingford, with their Horses and Arms properly belonging to them (and proportionable to their present or past Commands or Imployments) with flying Colours, Trumpets sounding, Drums beating, Match lighted at both ends, Bullets in their mouths, and every Souldier twelve charges of Powder, Match and Bullet proportionable, with one Peece of Ordnance with equipage, and with bagge and baggage, to any place within ten miles of Wallingford, which the Governour shall chuse, where, in regard His Majesty hath no Garrison open, nor Army neare, all their Horses and Arms, except what are allowed in the insuing Articles, are to be delivered up to such as His Excellency Sir Thomas Fairfax shall appoint; all the Souldiers shall be disbanded, and all such, both Officers, Souldiers, and others, as shall ingage themselves never to beare Armes against the Parliament, nor to do any thing wilfully to the prejudice of their affaires, during their abode in the Parliaments Quarters, shall have the benefit of the ensuing Articles; That is to say, 3 That all such as shall desire to go to their homes or private friends, (who shall not be prejudiced by receiving them) shall have the Generals Passe and Protection, for their peaceable repaire to, and abode at the severall places they shall desire for to go unto, the Governour, Officers, and Gentlemen, to passe with equipage of Horses and Armes, answerable to their present and past commands or qualities; and all, both Officers, Souldiers, and others, to passe with bagge and baggage. And the Troopers onely with their Swords, and their bagge and baggage. 4 That all such if there be any) who shall desire to take entertainment for forraigne service, shall have Passes for their Officers (not exceeding foure) with their Horses and two servants apeece, to go to London to treat with any Forraigne Ambassadour or a Gentl. for entertainment, and all of them shall have liberty and Passes to march, the Officers with equipage with Horses and Arms properly their owne, and answerable to their present or past Commands, the Common Souldiers with their swords,

and all with Bagge and Baggage to the Quarters neer unto Harwich or Southampton, or to any Port between them, to be transported, where they shall be assisted in the procuring of Vessels and shipping for their transportation, at the usuall rate accustomed for fraught by the Goverour of the next Garrison or Port Towne, who shall also take care for their safety and accommodation with Quarters untill shipping be provided and weather reasonable, they paying for the same after 28 daies from the Render. 5 That all the Persons now in Wallingford, not being of the Souldiery of the Garrison, shall have liberty for their persons and goods to stay in the Towne of Wallingford one Moneth after the rendring, if they desire it, and then to have the Generals Pass and Protection as others going out at the rendring, upon the like ingagement; And that any person whatsoever, who being sick and wounded, cannot at present remove, shall have liberty to stay till they be recovered or able to go away, and shall have fit accommodation and subsistance provided for them during such their stay, and then shall injoy the benefit of these Articles, 6 That no person whatsoever comprized in the Capitulation shall be reproached, reviled, affronted, plundered, or injured in their march, Rendezvouz, or quarters, journies, or places of abode by these Articles allowed, nor shall be compelled to beare Armes, nor be imprisoned, restrained, sued, molested, damnified for any matter whatsoever, of publike or private concernment, relating to this present War, the matter of grounds arising before the rendering of the Garrison, during the space of six Moneths after the rendering therof, nor be compelled to take any Oathes, or ingagements then what is mentioned in the second Article during the time of six Moneths, and to have liberty during the said time to travell about their lawfull affaires. 7 That all Horses, and other Goods now in Wallingford, taken as lawfull prize of Warre, or properly belonging to the Governour, or Officers of the Garrison before, or during the siege, shall be continued in the possession of the present possessors, except such as are to be delivered by the returne of these Articles. 8 That such Houshold-stuffe and Goods now in Wallingford, as shall appeare to the Generall, or such as he shall depute for that purpose, to have been borrowed by any Officer or Gentleman in the Garrison, for their use and accommodation in the Garrison, shall be restored back to the Owners. 9 That if any person or persons shall wilfully violate these Articles in any part, the guilt thereof shall be imputed to such person or persons only, and shall not prejudice any other not acting nor consenting in the same. 10 That the Governour and three more such Officers and Gentlemen as he shall name, shall have Passes from the Generall for themselves, with two servants a peece, their Horses, Swords, Pistols, and necessaries to goe to the King, to give him an accompt of the said Garrison, and to returne to their homes or friends, and that five weeks shall be allowed for this journey, which shall not be reckoned any of the six months mentioned in the sixt Article, but he

248

and they shall be allowed six Moneths after the end of the said five weekes. 11 That no Officer, Souldier, or other person, who by the Articles are to march out of the Castle or Town of Wallingford, or shall march in, shall plunder, spoyle, or injure any Inhabitant or other person therein, in their Persons, Goods, or Estates, or carry away any thing that is properly belonging to any of them. 12 That all persons comprised within these Articles shall peaceably and quietly injoy all their Goods, Debts, and moveables, during the space of six Months after the rendering, and shall have liberty within the said space (if they shall resolve to go beyond the seas) to dispose of their said Goods, Debts, and moveables, and to depart the Kingdome with the same as they shall thinke fit and desire, and shall have the Passes for their Transportation accordingly from his Excellency.

13 That all persons comprised within these Articles, shall upon request have a Certificate under the hand of the Generall, or such as he shall appoint, that such persons were in the Castle or Towne of Wallingford at the time of the rendering, and are to have the benefit of these Articles. 14 That the Townes-men and the Inhabitants of the Towne of Wallingford shall not bee troubled, or questioned for any thing said or written by any of them, nor the Corporation thereof prejudiced for anything done by any of them by express Command since it was a Garrison, and that they shall have the benefit of these Articles that may concerne them. 15 That Mr. John Chamberlin of Sherborne shall have the benefit of these Articles. Signed Tho. Blagge FINIS

249

15 THE LAST STAND: WALLINGFORD CASTLE AND THE CIVIL WAR. IMAGES AND INSIGHTS FROM ARCHAEOLOGY Neil Christie Abstract Although the aims of the Wallingford Burh to Borough Research Project were centred on the early medieval to late medieval townscape, excavation trenches at the castle in particular revealed many traces of post-medieval, Civil War-period activity, notably two likely cannon/ gun emplacement platforms, each signifying extensive logistical operations. These works have partly masked the medieval archaeology but are a very important element of the last phases of use of the castle, with repercussions for its structural fate. This paper outlines the archaeology revealed in the project and finds from earlier interventions. Introduction The prominence and extent of the castle earthworks at Wallingford give rise to two main impressions: firstly, the bulk and scale of medieval input here, reflecting the investment of royalty and high elites in the defensive and courtly structures that dominated the town and its immediate hinterland across the full medieval period; and secondly, the dramatic and extensive repercussions of the mid-17th-century Civil War, which saw the definitive slighting of the castle, leaving the complex almost totally shorn of standing masonry remains. Other papers in this volume have explored facets of the medieval castle and its history and archaeology, but in this and the previous contribution (by Judy Dewey) the castle’s final burst of notoriety and military activity in the Civil War will be assessed, again seeking to draw together the material and textual sources. As will be seen, both sets of data offer excellent scope to ‘read’ much more of the value of Wallingford Castle in this post-medieval context. Civil War archaeology at Wallingford The Wallingford Burh to Borough Research Project, while focussed on teasing out the archaeology of Wallingford across the early medieval to late medieval periods, inevitably encountered a variety of features and changes of post-medieval to modern date that have masked, damaged or even removed the earlier deposits. Thus in the Bullcroft, not just the impact of the suppression/dissolution but also subsequent land and property ownership – with structures reused, materials robbed, units demolished, open spaces imposed – have largely removed all visible trace of the former religious complex of the Holy Trinity Priory; and yet extensive geophysical survey has enabled sub-surface mapping of the complex and precinct, signifying its presence in the south of the Bullcroft zone, although thick rubble/demolition deposits and spreads have prevented

close pinpointing of the church and cloister, even if testpitting has started to provide a few small windows on the ‘lost’ remains (see Pedgley, this volume, and discussion in Christie and Creighton 2013, chapter 7). Comparable is the image on some levels for the castle complex. Its own ‘dissolution’ relates, as noted above, to the fall-out after Wallingford’s extended resistance in the Civil War and to a determined effort to cancel out any future military revamping of the former royal site. Here, the extensive earthworks of the castle motte, bailey, and defences as well as spaces beyond the castle zone to the north, have been denuded of visible stonework, their heights presumably levelled or ‘smoothed down’ in many instances. The documented sale of materials, the subsequent sale of the grounds, creation of a private mansion, then 19th-century landscaping – such as with tree-lined avenues (see Figure 7.48) – combined with the much more recent division of the castle complex into the open Castle Meadows space (now managed by Earth Trust, formerly The Northmoor Trust, on behalf of the South Oxfordshire District Council (SODC) – www.earthtrust.org.uk/Places/ Communitymeadows/Wallingford-Castle.aspx) and the Town Council-owned Castle Gardens, have all impacted on and effectively ‘blurred’ many parts of the buried archaeology (see full discussion by Dewey, Chapter 7 above). Geophysics at the castle site The broad open spaces of the castle have afforded scope for an almost unique exploration – in an English context – through sub-surface survey of the whole of an urban castle complex. A substantial and highly rewarding programme of geophysical survey was thus undertaken as part of the Wallingford Burh to Borough Research Project in the Easter seasons of 2008 and 2009 in the Castle Meadows, in ‘suburban’ space to the north-west, and in the Castle Gardens to the south-west, mapping a total area of c. 3 hectares. Further mapping through resistivity was then completed by TWHAS in 2009 and 2010 under Gerard Latham, extending coverage down to the junction with the Queen’s Arbour in the east, close to the riverside (giving in total nearly 5ha of coverage). The Queen’s Arbour and, to the north, the King’s Mead had been explored by resistivity survey as part of the Pilot Project in 2003, the key result being the presumed medieval quay feature or formal walkway (see Christie and Creighton 2013, chapter 6). A final survey component was Ground-Penetrating Radar (GPR), carried out in April 2010 on a sector of the inner bailey and on the presumed ‘barbican’ to its west, indicating deep deposits of robbed or in situ masonry features on the line of the medieval defences (see below). 250

Although now a largely open space, tree cover and especially the high and steep slopes of rampart ditches made the programme of total geophysical (resistivity and magnetometer) survey an often challenging task for the teams. However, the resultant plots (with an underlying grid of 20 x 20m squares, with readings taken every metre) are excellent and enable, alongside the detailed topographic earthwork survey work led by Michael Fradley (Chapter 3 above), a more secure formulation of the castle’s phased development, including its post-medieval uses. Key has been recognition of the arrangement of defensive and defining ditches and banks and a questioning of how far we can trace in these both pre-castle burh rampart lines (i.e. how directly did the castle bailey northern line reuse the late Saxon barrier) and post-Civil War efforts to delete

the defensive apparatus. Additional hopes were to identify areas of potential survival of buried stone features within the bailey spaces and to correlate some of these with documented castle components. What stands out from the plots (see Figure 15.1) is the legibility of the curtain wall lines, suggesting that masonry elements are still extant below the present pasture land; this interpretation was, as noted above, reinforced by the targeted GPR assessment. Of particular note in our interpretation of the plots was the substantial pair of projecting towers or bastions on the northernmost rampart line, each accompanied by a surrounding ditch and with a further likely outer ditch detected as a curving anomaly to the north. The scale of these outworks suggested something larger than medieval towers, with the assumption drawn

Figure 15.1 Composite resistivity plots for all areas surveyed in 2008–09 in the north-west zone of Wallingford (castle inner bailey, Castle Meadows, Castle Gardens and Wallingford School playing fields) imposed over revised castle earthwork survey by Michael Fradley. (Note: Image excludes 2003 resistivity surveys in Queen’s Arbour and King’s Mead by the riverside) (Image: TWHAS, Wallingford Burh to Borough Project.)

251

that they might relate to Civil War period reinforcements and extensions to older features here. As discussed below, the Project’s Trench 1 targeted the archaeology of the westerly tower. Trench 1 – Castle Meadows An excavation trench of 20 x 10m was thus opened in July 2008 at the front end of one of the large projecting tower spaces of the castle north circuit in the centre of Castle

Meadows (see Figure 15.2). Excavation rapidly confirmed the presence of a large north-facing artificial platform, surrounded by a substantial ditch (Figures 15.3 and 15.4). The earliest deposits comprised a homogenous series of medieval layers and dumped upcast from periodic cleaning or re-cutting of the castle ditch, these containing 12th- to 14th-century pottery; these deposits were sealed by the later platform but also cut by the ditch set around this to west, north and east. The platform itself was identified just 0.20–0.30m below the ground surface. Of 0.50–0.60m

WS EADO M E TL S CA

Trench 1 (2008)

OU

Trench Trench 77 (2010) (2010)

TER WARD masonry fragment

Trench 10 (2011)

INNER BAILE Y

Brooks Brooks excavation excavation (1965-68) (1965-68) IIL BA DL E DD MIID M

ABS ABS (unlocated) (unlocated) (1973) (1973)

LEE YY

Trench 4 (2009)

Trench Trench 88 (2003, 2010) (2003, 2010)

Queen’s tower

QUEEN’S ARBOUR MOTTE

Castle Farm Castle Farm (TVAS 1995) (TVAS 1995)

M

E

S

Carr Carr excavation excavation (1972) (1972)

0

H T .

St Nicholas’s College (site of)

R

curtain wall (fragment)

A

Castle Farm Castle Farm OAU (1992) (1992) OAU

100m

Figure 15.2 Map of the castle zone at Wallingford, identifying locations of all archaeological interventions, including the Burh to Borough Research Project trenches 1, 4, 7, 8, 10 (Image © the Wallingford Burh to Borough Research Project).

252

Extension B

0

3m

[1033]

1002

Trench 1

1014 1024

1002

Extension A

[1003]

1014 1002 1002

[1017] 1014 1002

[1015]

unexcavated 1000

1002

1014

[1022] 1021

1021

unexcavated 1001

Figure 15.3 Trench 1 – the north-western angle of the Civil War bastion feature (Image © the Wallingford Burh to Borough Research Project). E 1002

in thickness, the platform comprised a compacted, but friable light grey clay laid deposit containing fragments of chalk. Within and below the clay deposit were numerous consistent small dumps of sand and gravel; at the northern end of the trench the gravel-sand deposit formed a clear preparation and levelling deposit up to 0.4m thick. A long north to south section, 0.75m wide, was cut through the platform along the east edge of the trench to provide a cross-section through the alternate layers of orange gravely sand and grey clay used in its construction. Variations in the patterns of levelling and laying down of preparation material indicated either different construction phases or, more likely, the operation of different work teams. The platform was defined on its west and north sides by the surrounding ditch; the presumed return of the ditch on the east side lay outside the trench, and the castle’s outer ditch marked the south edge of the platform (Figures 15.3, 15.4). This angled ditch was up to 2m deep but was of unknown width, although on the basis of the geophysical plot the width may be in the order of 5m. Several segments were excavated through the inner part of the ditch fills which revealed a lower fill of sandy clay similar to the grey clay of the platform itself, and clearly derived from its slumping; early post-medieval finds were recovered from this as well as some residual medieval pottery. Given that the upper ditch fills contained 18th- to 19th-century finds, we should presume that the ditch will have remained visible as an earthwork feature until quite recently (it is now barely visible on the surface). The total size of the platform,

[1003]

1014 W

1010 1013

1024 [1003]

0

3m

Figure 15.4 Plan of Trench 1 showing plan of excavated bastion platform (Image © the Wallingford Burh to Borough Research Project).

which had a U-shaped plan, was thus 20–25m long from north to south and c. 12m wide from east to west. Despite close examination of its surface, the only features cut into the platform surface were a number of small animal burrows and tree-root holes; no clear trace was found of any postholes to support any standing structure. In terms of ceramics, almost 550 sherds (total weight 4.01 kg) came from Trench 1, the majority being medieval and post-medieval in date. The noted platform overlay sealed medieval deposits with materials from the 10th to 13th centuries, while finds from the platform itself and its slumping intriguingly featured a few residual finds (a 5th- to 6th-century organic tempered ware sherd, 10th/11th-century St Neots type ware and 10th- to 11th-century limestone sandy ware, plus cooking and jug sherds of the 12th and 13th centuries of local and Oxford production). Otherwise the platform and related contexts yielded early post-medieval material including 16th- to 17th253

century German stoneware and 17th- to 18th-century red earthenware. While the latter quantities were limited, this relative dearth is perhaps not surprising if this is recognized as a defensive and military feature of occasional usage and not a fixed point of habitation generating long-term rubbish deposits. The evidence overall supports the hypothesis that the platform formed part of a defensive gun emplacement. In and below subsoil and over and on the surface of the clay rampart was a high number of clay pipe fragments (520), including bowls, that would fit a mid- to later 17thcentury date; in total c. 1.6 kg of clay pipe fragments was recovered, the bulk from the uppermost (topsoil and subsoil) contexts. Almost no finds derived from the clay fill itself or the lower sand and gravel make-up, suggesting that the building operations were fairly rapid. While some post-medieval ceramics were found these were limited, as were more modern sherds; as noted, the general lack of other finds over the area support the idea that this was a short-lived structure, certainly not one that saw any lengthy occupation. What is important to recognize is that the construction of the platform, requiring the importation of clays, sand, gravel from the river area or other nearby sources on a substantial scale, testifies to considerable effort by a large number of people, and it would make sense if it was a military force that built it. It seems in fact to have been one of an array of bastions on the outer northern flank of the castle, designed to help protect the castle from a northerly enemy approach. Whether this excavated platform/bastion actually saw much in the way of military action is questionable, and the absence of many Civil War musket balls and other munitions is surprising in that such were recovered in a few other trenches opened by the Wallingford Burh to Borough Research Project, including the Queen’s Arbour by the riverside. The clay pipe debris could most likely be interpreted as anticipatory smokes, by soldiers tensely waiting to see if conflict would call them into action. It is likely that it was the North Gate that was the core focus of assault and these supporting bastions may have been as much for show as actual use – indeed, it is important to recall how the extended Civil War siege of Wallingford was far more of a blockade than a pounding of artillery from both sides (see Dewey, Chapter 14 this volume). A final element to note from the Trench 1 excavation was the collection by Leicester University geologists of samples of material from the platform surface for microfossil examination (published in Wilkinson et al. 2010). In brief, these samples provided two valuable pieces of extra information about the platform’s construction: firstly, the presence of specific types of seeds and fruits suggests a deposition of the materials in either summer or autumn (e.g. hazel nut, black nightshade, poppy); secondly, the samples contained a rich microfossil assemblage of ostracods and foraminifera, which allowed the platform material to be provenanced: it can be argued that the royalist garrison had imported Glauconitic Marl from contemporary and active quarries on the east side of the Thames in the Crowmarsh Gifford area where it occurs but is obscured beneath more recent superficial deposits, or

else from fields to the north of the castle – as identified in project excavations in Wallingford School Playing Fields (see Christie and Creighton 2013, 240). It was also shown by the geologists that, when compacted, this deposit will have formed a durable, almost road-like base, eminently suitable as a gun platform. Defending the North Gate: results from Trench 10 A further excavation (Trench 10) under the auspices of the Wallingford Burh to Borough Research Project was undertaken in July/August 2011 to examine the prominent earthwork on the north-west corner of the castle site, a space tentatively associated with the ‘barbican’ or the fortified formal entrance into the inner bailey from the North Gate – a proposal based on location, height and prominence (see Figures 15.2 and 15.7). The geophysical survey here, combined with GPR assessment, had indicated likely masonry survival at depth at the lip of the ‘barbican’ earthwork, and at least one potential built structure in its southern sector. Trench 10’s location was thus sited to test some of the archaeology in the eastern half of the space and to pinpoint part of the defensive component and of the built unit. While a trench of 20 x 15m was planned, this was reduced on both north and east once we recognized that proximity to the slopes on each side would weaken the stability of the exposed trench edges; this unfortunately curtailed chances of tracing directly any of the postulated medieval circuit masonry. The results of the excavation (see Figure 15.5) were different from what was anticipated, adding less to our understanding of the medieval castle but more to our understanding of the mid-17th-century Civil War changes wrought on the castle spaces. Initial machine clearance included supervised slot cuttings down to recognizable archaeology in the northern half of the trench; much shallower clearance was required in the southern half, here coming down to scattered rubble deposits and a possible building. The latter survived as a set of crude walls, reusing older material, both chalk and sandstone, some well cut and others roughly worked only, and with limited bonding traces; in addition, clear traces of any internal surface or floor were lacking. The walls here overlay relatively shallow rubble and mortar spreads, whereas a more consistent and large rubble spread was traced to the north and north-west of the ‘building’. Most probably the rubble relates to a phase of demolition or levelling, which was subsequently covered by imported sandy silt deposits, which featured noticeably mixed ceramic finds of 11ththrough to 15th-century date. One possibility is that the material had been re-deposited here deliberately from clearance of the inner bailey ditch. The archaeology of the northern half of the trench revealed an even more substantial and diverse remodelling of the site. Two main deposit types were identified in the various north-south excavation slots, generally running as alternate angled bands laid from south to north: orange brown sandy gravel bands set over or overlain by firm, green-grey or brown-grey marl deposits (see Figure 15.6). The thicknesses of these deposits varied – from substantial 254

Figure 15.5 View looking east of Trench 10 after excavation, with rampart and related construction levels to left, and rubble spreads to the right (Image © the Wallingford Burh to Borough Research Project).

Figure 15.6 Trench 10 - photograph of the exposed make-up to the Civil War platform, consisting of the angled lines of marl-clay interleaved with sandy gravel deposits extending to the medieval rampart, a trace of which is exposed in the far right of the trench (Image © the Wallingford Burh to Borough Research Project).

255

Figure 15.7 View from the west of the north face of the elevated ‘barbican’ zone, explored in Trench 10 in 2010 (Image © the Wallingford Burh to Borough Research Project)

sand-gravel fills at the north-west end of the trench over 1m deep to thinner angled deposits of 20–30cm thickness. The slot sections all indicated that these were careful episodes of alternative laying of preparation and construction deposits (quite possibly by different work gangs, as suggested above also for Trench 1), angled to run up to the north end of the ‘barbican’ prominence. There were minimal finds from within these deposits, indicating a rapid process of preparation and construction. The sources for the materials are to be determined, but each could be local – whether from river gravels or from local fields. Excavation at the north-east corner of the trench in fact uncovered the angled (sloping southwards) remnants of a rampart formed of dark reddish brown silt, comparable to the build of the late Saxon rampart excavated by Brooks and in Trench 7 of the Burh to Borough Research Project. Only 1.2m length and 0.6m depth of this construction was revealed since the trench depth at this corner and the relative weakness of the trench edges (due to the presence of the fairly loose sand-gravel layers) made it dangerous to continue without shoring. Most probably this rampart trace forms part of the line of the late Saxon and medieval defences, quite possibly with a stone wall front added and maintained from the 10th/11th century. The later sand-gravel and marl deposits were thus designed to run up to the back of this rampart and wall in a systematic process of reinforcement and raising of the ground level behind. The rubble spreads and soil deposits in the south half of the trench could then

be seen to represent a rearward securing and stabilizing of the ground. The noted built structure appears to overlie this levelling work but it was not possible to determine its date or function; certainly it was not something of quality, beauty or durability, however, and perhaps it was storage space or a shelter contemporary with the marl-clay platform to its north. Problematically, no secure surface survived to cap the raised levels on the ‘barbican’, except for a thin trace in the northern end section; nonetheless, we can draw on the levelled platform surface identified in Trench 1 to the north as a direct comparison. A few fragments of clay pipe and occasional post-medieval pot and some bone were recovered from the uppermost marl-clay deposits, though in no way do these match the levels of finds made in Trench 1. The interpretation proposed is that Trench 10 revealed major re-working of the presumed castle barbican structure, with a massive importation of materials to elevate and reinforce what was perceived as a major defensive position. The sand/gravel and marl/clunch deposits in fact match the Trench 1 Civil War bastion construction methods to the north of Trench 10 in the Castle Meadows (see above), but have here been carried out on a much bigger scale, with many metric tons of imported building materials. Trench 1 was more securely dated by the high number of clay pipes from the top of the platform, and the relative absence of such material and comparable surface at Trench 10 can easily be explained by a deliberate slighting and clearance 256

of the bastion, pulling down the remnant facing stone wall and dragging down the top skin and flanks of the imported platform and materials. Potentially the debris from this is reflected in the stepped character of the earthwork on its western flank (Figure 15.7). Nonetheless, some clay pipe was recovered (30 fragments), many of which appear likely 17th-century in date, but with later finds too. One can stress the very elevated setting of this barbican site (Figure 15.7) – deliberately raised higher for the Civil War conflict – and its highly strategic position in sight of the North Gate and controlling what we should assume was still the main entrance-way into the actual castle core (where, we can assume, was housed the entrenched garrison). It is valuable to re-quote the reference made in the documents of 1643 to the newly established defensive dispositions: ‘They have made very strong workes about the castle, and a double drawbridge at the entrance to the castle and two drakes [guns] planted there upon one carriage’. Potentially also, the substantial infill of the inner bailey ditch to the south-east of the barbican earthwork belongs to the Civil War works, perhaps undertaken in order to facilitate the drawing up of cannon or ‘drakes’ to this site. If correct, we have here notable archaeological testimony to the substantial works undertaken by the garrison and other labour to re-fortify Wallingford Castle; it offers a clearer recognition of not just strategy but, as importantly, the physical logistics to build and equip these positions. At the same time, the limited material culture we have gathered up puts us into some contact with the human presence that manned these stations; however, the restricted character of this material meanwhile helps us visualize far less a town, castle and garrison under constant siege but one which may have seen a much patchier series of confrontations – verbal as much as military. Other archaeological findings Briefly, one can note a small number of other excavations on the castle site which provide further insight into the nature and extent of its 17th-century re-fortification: these include Carr’s 1972 intervention in the middle bailey, which was shown to have been (re-)enclosed with a stone wall at this time, and a Thames Valley Archaeological Services (TVAS) evaluation south-east of the motte in 1995, which revealed built-up ground from the same period (see Christie and Creighton 2013, 173, 195, 215; Ford 1995). In the 1960s Nicholas Brooks undertook excavations at the site of the town’s former North Gate, destroyed when the castle’s outer rampart was created in the 13th century. In his trenching (Trenches D2 and D3) he recognized deep, tipped deposits of brown-orange gravel with clay which lay close to the turf line and which contained some pieces of clay pipe, tile and postmedieval pottery (see Christie and Creighton 2013, 98). These deposits filled part of the old ditch here and should, as recognized in our project trenches, relate to a levelling and raising of the ground to create a platform. Brooks also traced two 30cm square postholes cut to a depth of at least 1.5m, which he suggested formed part of the setting for an observation tower behind the gate/rampart. An alternative

reading of the postholes is that they might have anchored ‘cannon baskets’ or ‘gabions’ (cf Harrington 2004, 34–37). Finally, the nearby Trench V cut by Brooks’ team in 1968 down the slope of the castle/barbican bank to the east of the North Gate trenches offered interesting data for this final castle phase: Brooks’ unpublished notes claim that he could see a demolition of the residual castle wall at this point prior to the sieges; he argued that the Royalist garrison probably pulled down the already ruinous malmstone (chert) walls as they would be an inadequate defence against Cromwellian artillery; instead they then built up new earthen ramparts (see Christie and Creighton 2013, 98). Loss, damage and potential A final point to draw from the discussion of the two main Project trenches (1, 10) and from other noted work is that we should now recognize that, at least in the case of the ‘barbican’ site and potentially in other sectors of the castle site, the defensive occupation and re-workings by the Civil War garrison can be viewed as substantial in terms of the importation of building materials – sands, gravels, clay, timber, turves – drawn from the vicinity. This on one level means that the medieval, full castle deposits can lie deeply buried beneath these 17th-century works, perhaps between 1.0 and 2.0m down; on another level this has protected some of this earlier archaeology, although we should expect that, as hinted at in the rubble deposits in Trench 10, on occasions removal and demolition of some medieval components may have been part of the strategy of construction and entrenchment. At the same time we can see also now that post-Civil War slighting of the defensive bastions has ‘blurred’ our reading of the underlying earthworks and created a different artificiality to the whole. Nonetheless, we might expect that the garrison’s strategy was often to select to reinforce set or key points in the older castle apparatus – i.e. pre-existing tower stations and gateways – meaning that this Civil War ‘masking’ may not be everywhere. Even so, the physical retributions in the aftermath of the siege and conflict did more than enough to cancel out many other sectors of the former castle site. Conclusions Wallingford Castle can now feature prominently in the list of sites where archaeology has illustrated physically how medieval fortresses were re-fortified ‘roofless bulwarks’ during the mid-17th-century English Civil War (Harrington 2004, 36). Other important case-studies include (in Scotland) Huntly and Tantallon, (in Wales) Caerphilly and Montgomery, and (in England) Corfe, Beeston, Dudley and Sandal (ibid., 36–59; Crossley 1990, 113–117). A significant point of difference with the refortification of Wallingford Castle in the conflict is the absence of characteristically arrow-headed bastions (those on the north side of the castle being U-shaped), which awaits explanation. It should not escape our attention that the two short, sharp bursts of violent military activity that punctuate the 257

long and largely peaceful history of Wallingford Castle – in the mid-12th and mid-17th centuries respectively – have some salient features in common. Both sieges occurred within the context of prolonged civil wars, with the castle held in both cases as an outpost to a much wider loyalist heartland. Both were protracted events with little evidence that the fortress was directly and seriously assaulted – a fact which tells us as much about the essential psychology of medieval and early modern warfare as it does about the castle’s physical strength. And both periods of resistance resulted in major episodes of slighting – in the mid-12th century of multiple siege-castles that were erased from

the landscape, and in the mid-17th century of Wallingford Castle itself. In the case of the latter, while parliamentarian slightings are usually viewed as fiscally and militarily inspired initiatives, one should also bear in mind the often under-estimated political and symbolic undertones of these operations (Rakoczy 2007; see also Thompson 1987, 138– 57). Certainly, however, the Civil War slightings denuded Wallingford Castle of the majority of the standing fabric which for so long had given this complex such prominence in the English medieval landscape. Potentially, of course, future research may add further to our understanding of Wallingford’s Civil War roles and its castle’s ultimate fate.

258

BIBLIOGRAPHY

Aarts, B. 1996. Early castles of the Meuse-Rhine border region and some parallels in Western Europe c. 1000: a comparative approach. In Études de castellologie médiévale 1994. Actes du colloque international de Abergavenny II–23, Caen. Abels, R. P. 1988. Lordship and Military Obligation in Anglo-Saxon England. London, British Museum. Addyman, P. 1972. The Anglo-Saxon house: a new review. Anglo-Saxon England 1, 273–307. Addyman, P. V. and Priestley, J. 1977. Baile Hill, York: a report on the Institute’s excavations. Archaeological Journal 134, 113–156. Allnatt, W. 1873. Rambles in the Neighbourhood of Wallingford. Wallingford, S. Bradford. Anon 1972. A medieval ‘cob’ building. Current Archaeology 35, 318. Armitage, E. S. 1912. The Early Norman Castles of the British Isles. London, John Murray. Armstrong, D. 2008. Rewriting the chronicle tradition: the alliterative Morte Arthure and Arthur’s Sword of Peace. Parergon 25:1, 81–101. ASC. The Anglo-Saxon Chronicles, ed. M. Swanton. London, Phoenix Press, 1996. Ashbee, J. 2008. The structure and function of the White Tower 1150–1485. In E. Impey (ed.), The White Tower, 140–159. London and New Haven, Yale University Press. Asser. Asser’s Life of King Alfred, together with the Annals of St Neots, Erroneously Ascribed to Asser, ed. W. H. Stevenson. Oxford, Oxford University Press, 1959. Astill, G. 1984. The towns of Berkshire. In J. Haslam (ed.), Anglo-Saxon Towns in Southern England, 53–86. Chichester, Phillimore. Astill, G. 1991. Towns and town hierarchies in Saxon England. Oxford Journal of Archaeology 10:1, 95–116. Atkin, M. 1991. Gloucester’s Norman castle rediscovered. Fortress 9, 20–23. Aurell, M. 2006. Society. In D. Power (ed.), The Central Middle Ages, 950–1320. Oxford, Oxford University Press, 28–56. Austin, D. 1984. The castle and the landscape. Landscape History 6, 70–81. Ayers, B. 1985. Excavations in the North-East Bailey of Norwich Castle, 1979. East Anglian Archaeology Report 28. Dereham, East Anglian Archaeology. Baker, D. 1973. Bedford Castle: some preliminary results from rescue excavations. Chateau Gaillard 6, 15–22. Barlow, F. 1983. William Rufus. London, Methuen. Barron, J, 2002. The Augustinian Canons and the University of Oxford: the Lost College of St George. In C. M. Barron and J. Stratford (eds), The Church and Learning in Later Medieval Society: Essays in Honour

of R.B Dobson. Proceedings of the 1999 Harlaxton Symposium, 228–254. London. Bartlett, R. 2000. England under the Norman and Angevin Kings, 1075–1225. Oxford, Oxford University Press. Basset, S. 2009. Anglo-Saxon Warwick. Midland History 34:2, 123–155. Baxter, S. 2007. The Earls of Mercia. Oxford, Oxford University Press. Beresford, G. 1987. Goltho: The Development of an Early Medieval Manor c 850–1150. London, English Heritage. Beresford, M. W. 1967. New Towns of the Middle Ages: Town Plantation in England, Wales and Gascony. London, Lutterworth Press. BF. The Book of Fees, commonly called Testa de Neville, 2 vols in 3. London, HMSO, 1920–31. Biddle, M. 1969. Excavations at Winchester, 1968; seventh interim report. Antiquaries Journal 49, 295–329. Biddle, M. 1970. Excavations at Winchester, 1969: eighth interim report. Antiquities Journal 50, 277–326. Biddle, M. 1974. The development of the Anglo-Saxon town. In G. Fasoli (ed.), Topografia urbana e vita cittadina nell’alto medioevo in Occidente I, 203–232. Spoleto, CISAM. Biddle, M. (ed.) 1976. Winchester in the Early Middle Ages. Oxford, Oxford University Press. Biddle, M. 1987. Early Norman Winchester. In J. Holt (ed.), Domesday Studies, 311–331. Woodbridge, Boydell Press. Biddle, M. and Hill, D. 1971. Late Saxon planned towns. Antiquaries Journal 51, 70–85. Biggs, D. 2002. ‘To aid the Custodian and Council’: Edmund of Langley and the defense of the realm, June-July 1399, Journal of Medieval Military History 1, 125–144. Birss, R. and Wheeler, H. 1985. Roman Derby: excavations 1968–83. Derbyshire Archaeological Journal 105, 340–343. BL. British Library, London. Blair, J. 1994. Anglo Saxon Oxfordshire. Oxford, Oxfordshire Books. Blair, J. 2003a. St. Michael at the Northgate tower survey: date and context. In A. Dodd (ed.), Oxford before the University. The Late Saxon and Norman Archaeology of the Thames Crossing, the Defences and the Town, 162–163. Thames Valley Landscapes Monograph 17. Oxford, Oxford Archaeology. Blair, J. 2003b. Hall and chamber: English domestic planning, 1000–1250. In R. Liddiard (ed.), AngloNorman Castles, 307–328. Woodbridge, Boydell Press, (First published in G. Meirion-Jones and M. Jones (eds), Manorial Domestic Buildings in England and

259

Northern France, Society of Antiquaries Occasional Paper 15. London, 1993). Boarstall. The Boarstall Cartulary, ed. H. E. Salter, Oxford Historical Society 88. Oxford, Clarendon Press, 1930. Bodleian. Bodleian Library, Oxford. Bond, S. 1967. The medieval constable of Windsor Castle. English Historical Review 82, 225–249. Boüard, M. de, 1973. De l’aula au donjon: les fouilles de la motte de la Chapelle à Doué-la-Fontaine (Xe–XIe siècle). Archéologie Médiévale 3, 5–110. Bracton. Bracton’s Notebook, ed. F. W. Maitland. Cambridge, Cambridge University Press, 1887. Bradley, J. and Gaimster, M. (eds) 2004. Medieval Britain and Ireland in 2003. Medieval Archaeology 48, 287– 288. Brand’honneur, M. 2004. Manoirs et châteaux dans le comté de Rennes. Habitat à motte et société chevaleresque (XIe–XIIe siècle). Rennes, Presses Universitaires. Brewster, T. C. M. and Brewster, A. 1969. Tote Copse Castle, Aldingbourne, Sussex. Sussex Archaeological Collections 107, 141–179. Briquet, C.-M. 1907. Les Filigranes Dictionnaire historique des marques du papier, dès leur apparition vers 1282 jusqu’en 1600. Paris, Alphonse Picard et fils. BRO. Berkshire Record Office. Brooks, N. P. 1965a. Wallingford Castle excavations from the 9th to 31st August, 1965. Unpublished notes. Wallingford, Wallingford Museum, Brooks archive. Brooks, N. P. 1965b. Excavations at Wallingford Castle 1965: an interim report. Berkshire Archaeological Journal 62, 17–21. Brooks, N. P. 1966. Wallingford Castle excavations, August 1st to 27th 1966. Unpublished notes, Wallingford, Wallingford Museum, Brooks archive. Brooks, N. P. 1968. Wallingford Castle excavations 1968, August 13th to September 7th. Unpublished notes, Wallingford, Wallingford Museum, Brooks archive. Brown, I. 2009. Beacons in the Landscape. Macclesfield, Windgather Press. Brown, R. A. 1976. English Castles. London, Batsford. Brown, R. A. 2003. Royal castle-building in England 1154–1216. In R. Liddiard (ed.), Anglo-Norman Castles, 133–178. Woodbridge, Boydell Press (first published in English Historical Review 70, 353–398). Bullock-Davies, C. 1981. Chrétien de Troyes and England. Arthurian Literature 1, 1–61. Burton, P. A. 2010–11. Original castle gates and doors – a survey. The Castle Studies Group Journal 24, 246–259. Cal. Pat. Rolls. Calendar of Patent Rolls (1216–1582). London, HMSO, 1891–1986. Calendar of Treasury Books, ed. W. Shaw. 31 vols. London, HMSO, 1904–57. Calendar of Treasury Papers, ed. J. Redington. 6 vols. London, Longmans, Green, Reader, and Dyer, 1868– 89. Cam, M. M. 1963. The Hundred and the Hundred Rolls. London, Merlin Press.

Camden, W. 1586. Britannia. Siue Florentissimorum Regnorum, Angliæ, Scotiæ, Hiberniæ, et Insularum Adiacentium ex Intima Antiquitate Chorographica Descriptio. London, Randulph Newbery. Campbell, J. 1987. Some agents and agencies of the late Anglo-Saxon state. In J. C. Holt (ed.), Domesday Studies, 201–218. Woodbridge, Boydell Press. Carpenter, D. 1990. The Minority of Henry III. London, Methuen. Carpenter, D. 2000. The second century of English feudalism. Past & Present 168, 30–71. Carr, R. D. 1973a. Excavations at Wallingford Castle 1972: An interim report. Unpublished Report. Swindon, National Monuments Record. Carr, R. D. 1973b. Wallingford Castle. Council for British Archaeology Group 9 Newsletter 3, 18. Carter, A and Stevenson, J. 1974. The Oxfordshire Area in the Civil War. Oxford, BBC Radio Oxford Publications. Cartularies Abingdon. Two Cartularies of Abingdon Abbey, ed. C. F. Slade and G. Lambrick, Oxford Historical Society new series 32–33. Oxford, 1990–92. Carver, M. 2010. The Birth of a Borough: An Archaeological Study of Anglo-Saxon Stafford. Woodbridge, Boydell Press. Cathcart King, D. J. 1983. Castellarium Anglicanum. An Index and Bibliography of the Castles in England, Wales, and the Islands. 2 vols. Millwood, NY: Kraus International Publications. Cathcart King, D. J. 1988. The Castle in England and Wales. London and Sydney, Croom Helm. CCA. Oxford, Christ Church Archives. CCC. Corpus Christi College, Oxford CCR. Calendar of the Charter Rolls Preserved in the Public Record Office. 8 vols. London, HMSO, 1903– 27. CFR. Calendar of Fine Rolls Preserved in the Public Record Office. 22 vols. London, HMSO. 1911–62. Chapman, A. 1985. Northampton Castle: a review of the evidence. Unpublished report held in National Monuments Record Centre Swindon. Chapman, A. 2014. In search of Northampton Castle. British Archaeology 135, 26–31. Chibnall, M. 1986. Anglo-Norman England, 1066–1166. Woodbridge, Boydell Press. Christelow, S. M. 1996. A moveable feast? Itineration and the centralization of government under Henry II of England. Albion 28, 187–228. Christie, N. and Creighton, O., with Edgeworth, M. and Hamerow, H. 2013, Transforming Townscapes. From Burh to Borough: The Archaeology of Wallingford, AD 800–1400. The Society for Medieval Archaeology Monograph 35. London, Society for Medieval Archaeology. Christie, N., Creighton, O., Edgeworth, M. and Fradley, M. 2010. ‘Have you found anything interesting?’ Exploring late-Saxon and medieval urbanism at Wallingford: sources, results and questions. Oxoniensia 75, 35–47. Christie, N., Edgeworth, M., Taylor, J., Hyam, A., Baker, P., Creighton, O. H., Fradley, M. and Latham, G. 2010.

260

Mapping Wallingford castle. Medieval Archaeology 54, 416–420. CIPM. Calendar of Inquisitions Post Mortem and other Analogous Documents Preserved in the Public Record Office. 22 vols. London, HMSO, 1904. Clanchy, M. T. (ed.) 1973. The Roll and Writ File of the Berkshire Eyre of 1248. Selden Society 90. London, Selden Society. Clark, A. 1891, The Life and Times of Anthony Wood, Antiquary of Oxford, 1632–1695, Described by Himself. Vol I: 1632–1663. Oxford Historical Society 19. Oxford, Oxford Historical Society. Clark, G. T. 1889. Contribution towards a complete list of moated mounds or burhs. The Archaeological Journal 46, 197–217. Clark, G. T. 1894. Mediaeval Military Architecture in England. 2 vols. London, Wyman. CLR. Calendar of Liberate Rolls Preserved in the Public Record Office: Henry III. 6 vols. 1226–40. London, HMSO, 1916–64. CMACC. Cartulary of Medieval Archives at Christ Church, ed. N. Denholm Young. Oxford, Clarendon Press, 1931. Coad, J. G. 1995. The Book of Dover Castle and the Defences of Dover. London, Batsford. Coad J. G. and Streeton, A. 1982. Excavations at Castle Acre castle, Norfolk, 1972–7. Archaeological Journal 139, 138–301. Coad, J. G., Streeton, A. D. F. and Warmington, R. 1987. Excavations at Castle Acre castle, Norfolk, 1975– 1982: the bridges, lime kilns, and eastern gatehouse. Archaeological Journal 144, 256–307. Coates, R. 2010. Hidden gates. Nomina 33, 139–168. Cobb, J. W. 1883. Two Lectures on the History and Antiquities of Berkhamsted. London, Nelson. COEL. K. S. B. Keats-Rohan, The Continental Origins of English Landholders 1066–1166 Database and the COEL Database System on CDROM. Coel Enterprises Ltd, 2002. Colvin, H. M. 1971. Building Accounts of King Henry III. Oxford, Clarendon Press. Conzen, M. R. G. 1969. Alnwick, Northumberland: A Study in Town-Plan Analysis. London, British Institute of Geographers. Coulson, C. 1979. Structural symbolism in medieval castle architecture. Journal of the British Archaeological Association 132, 73–90. Coulson, C. 1992. Some analysis of the castle of Bodiam. Medieval Knighthood 4, 51–107. Coulson, C. 1994. The castles of the Anarchy. In E. King (ed.), The Anarchy of King Stephen’s Reign, 67–92. Oxford, Clarendon. (Reprinted in R. Liddiard (ed.), Anglo-Norman Castles, 179–202. Woodbridge, Boydell Press.) Coulson, C. 1996. Cultural realities and reappraisals in English castle-study. Journal of Medieval History 22(2), 171–208. Coulson, C. 2001. Peaceable power in English castles. Anglo-Norman Studies 23, 69–95.

Coxe, 1878. W. H. Turner (ed.), under direction of Rev. H. O. Coxe, Calendar of Charters and Rolls Preserved in the Bodleian Library. Oxford, Clarendon Press. CPR. Calendar of the Patent Rolls Preserved in the Public Record Office. Multiple vols cited by monarch and date range. London, HMSO. Creighton, O. H. 1997. Early Leicestershire castles: archaeology and landscape history. Transactions of the Leicestershire Archaeological and Historical Society 71, 21–36. Creighton, O. H. 2002. Castles and Landscapes: Power, Community and Fortification in Medieval England. Sheffield, Equinox. Creighton, O. H. 2005. Castles and Landscapes: Power, Community and Fortification in Medieval England. 2nd edition. London, Equinox. Creighton, O. H. 2007. Town defences and the making of urban landscapes. In M. Gardiner and S. Rippon (eds), Medieval Landscapes, 43–56. Macclesfield, Windgather. Creighton, O. H., Christie, N., Edgeworth, M. and Hamerow, H. 2009. New directions in tracing the origins and development of Wallingford: targets and results of the Wallingford Burh to Borough Research Project. In K. S. B. Keats-Rohan and D. R. Roffe (eds), the Origins of the Borough of Wallingford: archaeological and Historical Perspectives, 68–76. British Archaeological Reports British Series 494. oxford, BAR Publishing. Creighton, O. H. and Higham, R. 2005. Medieval Town Walls: An Archaeology and Social History of Urban Defence. Stroud, Tempus. Crick, J. 2001. Liberty and fraternity: creating and defending the liberty of St Albans. In A. Musson (ed.), Expectations of the Law in the Middle Ages, 91–104. Woodbridge, Boydell Press. Crook, D. 1982. Records of the General Eyre. Public Record Office Handbook 20. London, HMSO. Crossley, D. 1990. Post-Medieval Archaeology in Britain. Leicester, Leicester University Press. Crouch, D. 2001. The origin of chantries: some further Anglo-Norman evidence. Journal of Medieval History 27, 159–180. CRR. Curia Regis Rolls of the Reigns of Richard I and John. 20 vols. London, HMSO, 1922–2006. CSPD I. Calendar of State Papers Domestic. Edward VI, Mary, Elizabeth and James I, ed. R. Lemon. 12 vols. London, Longman, Brown, Green and Longmans, 1856. Online at http://www.british-history.ac.uk/calstate-papers/domestic/edw-eliz/1547–80 (accessed 2 May 2015). CSPD II. Calendar of State Papers, Domestic Series, of the Reign of Charles I, ed. J. Bruce, W. D. Hamilton, S. C. Lomas. 6 vols. London, Longman, Brown, Green and Longmans, 1858–97. Online at http://www. british-history.ac.uk/cal-state-papers/domestic/chas1/ addenda/1625, 17–2649 (accessed 2 May 2015). CSPD III. Calendar of State Papers, Domestic, 1649– 1660, ed. M. A. E. Green. 13 vols. London, HMSO,

261

1875. Online at http://www.british-history.ac.uk/calstate-papers/domestic/interregnum/ (accessed 2 May 2015). CSPD IV. Calendar of State Papers, Domestic Series, of the Reign of Charles II, ed. M. A. Green et al. 28 vols. London, Longman, Green, Longman, and Roberts, 1860–1938. Cuttler, R., Hunt, J. and Rátkai, S. 2008. Saxon burh and royal castle: re-thinking early urban space in Stafford. Transactions of the Staffordshire Archaeological and Historical Society 43, 39–85. Dannenbaum, S. 1982. Anglo-Norman romances of English Heroes: ‘Ancestral Romance’? Romance Philology 35:4, 601–608. Daniell, C. 1997. Death and Burial in Medieval England. London, Routledge. Darvill, T. 1988. Excavations on the site of the early Norman castle at Gloucester, 1983–4. Medieval Archaeology 32, 1–49. Davies, G. 1937, The Early Stuarts, 1603–1660. Oxford, Clarendon Press. Davis, J. 2004. Baking for the common good: a reassessment of the assize of bread in medieval England. Economic History Review 57, 465–502. Davison, B. K. 1969. Early earthwork castles: a new model. Château-Gaillard 3, 37–47. DB Berks. Domesday Book: Berkshire, ed. P. Morgan. Chichester, Phillimore, 1979. DB Bucks. Domesday Book: Buckinghamshire, ed. J. Morris. Chichester, Phillimore, 1978. DB Derby. Domesday Book: Derbyshire, ed. P. Morgan. Chichester, Phillimore, 1978. DB Middlesex. Domesday Book: Middlesex, ed. J. Morris. Chichester, Phillimore, 1975. DB Notts. Domesday Book: Nottinghamshire, ed. J. Morris. Chichester, Phillimore, 1977. Denton, J. H. 1970. Royal Free Chapels 1100–1300: A Constitutional Study. Manchester, Manchester University Press. Derek, J. 1955. Medieval land surveying and topographical maps. The Geographical Journal 121:1, 1–7. Dewey, J. 2009. The origins of Wallingford: topography, boundaries and parishes. In K. S. B. Rohan and D. R. Roffe (eds), the Origins of the Borough of Wallingford: archaeological and Historical Perspectives, 17–26. British Archaeological Reports British Series 494. oxford, BAR Publishing. Dewey, J. and Dewey, S. 1977. The Book of Wallingford. Chesham, Barracuda Books. Dewey, J. and Dewey, S. 1996. Wallingford and the Civil War. Cholsey, Pie Powder Press. Dewey, J. Dewey, S. and Beasley, D. 1989. Window on Wallingford 1837–1914. Cholsey, Pie Powder Press. Disbury, D. 1998, Beef, Bacon and Bag Pudding, Old Berkshire in the Civil War. Reepham, Timescape. Dixon, P. 2002. The myth of the keep. In G. MeirionJones, E. Impey and M. Jones (eds), The Seigniorial Residence in Western Europe, AD 800–1600, 9–13.

British Archaeological Reports International Series 1088. oxford, BAR Publishing. Dixon, P. 2008. The influence of the White Tower on the great towers of the twelfth century. In E. Impey (ed.), The White Tower, 243–275. London and New Haven, Yale University Press. Dixon, P., Knight, D. and Firman, R. 1997. The origins of Nottingham. In J. V. Beckett (ed.), A Centenary History of Nottingham, 24–42. Manchester, Manchester University Press. Dixon, P. and Marshall, P. 2002. Norwich Castle and its analogues. In G. Meirion-Jones, E. Impey and M. Jones (eds), the Seigniorial Residence in Western europe, aD 800–1600, 235–243. British Archaeological Reports International Series 1088. oxford, BAR Publishing. Dodd, A. (ed.), 2003. Oxford before the University. The Late Saxon and Norman Archaeology of the Thames Crossing, the Defences and the Town. Thames Valley Landscapes Monograph 17. Oxford, Oxford Archaeology. Downham, C. 2009. ‘Hiberno-Norse’ and ‘Anglo-Danes’: stereotyping ethnicity in Viking Age England. Medieval Scandinavia 19, 139–169. Drage, C. 1987. Urban castles. In J. Schofield and R. Leech (eds), Urban Archaeology in Britain, 117–132. CBA Research Report 61. London, CBA. Drewett, P. J. 1992. Excavations at Lewes Castle 1985– 1988. Sussex Archaeological Collections 130, 69–106. Drury, P. J. 1982. Aspects of the origins and development of Colchester Castle. Archaeological Journal 139, 302–349. Drury, P. J. 2002, Norwich Castle keep. In G. MeirionJones, E. Impey, E. and M. Jones (eds), the Seigneurial Residence in europe aD c.800–1600, 211–234. British Archaeological Reports International Series 1088. oxford, BAR Publishing. Dudley, C. J. 2006. Measure and the medieval architect, Peterborough Cathedral – The Complete Geometry 1100–1500. http://medievalarchitecturalgeometry. com/Index.htm (accessed 15 May 2012). Duggan, J. 2001. The Romances of Chrétien de Troyes. New Haven, Yale University Press. Durham, B. 1984. The Thames crossing at Oxford: archaeological studies 1979–82. Oxoniensia 49, 57– 100. Durham, B. G. 1987. Wallingford: castle, south curtain wall and St. Nicholas’ College. South Midlands Archaeology 17, 99. Durham, B. 1990. Wallingford Castle, Castle Lane House. South Midland Archaeology 20, 85–86. Eales, R. 2003. Royal power and castles in Norman England. In R. Liddiard (ed.), Anglo-Norman Castles, 41–67. Woodbridge, Boydell Press. Early, R. 1999a. Les origines du château de Mayenne: apports archéologiques. In Actes du VIIe Congrès internat. d’Archéologie médiévale. Le-Mans-Mayenne. Early, R. 1999b. Une étude unique en Europe: le château de Mayenne. Archeologia 358, 2–54.

262

Eddershaw, D. 1995. The Civil War in Oxfordshire. Stroud, Alan Sutton Publishing Edgeworth, M. 2009. Comparing burhs: a WallingfordBedford case study. In K. S. B. Keats-Rohan and D. R. Roffe (eds), the Origins of the Borough of Wallingford: archaeological and Historical Perspectives, 77–85. British Archaeological Reports British Series 494. oxford, BAR Publishing. Edgeworth, M. and Christie, N. 2011. An archaeology of crossing-places: the bridge and ford at Wallingford on the River Thames. In T. Pflederer and C. S. Sommer (eds), Archäologie der Bruken: Vorgeschichte, Antike, Mittelalter, Neuzeit, 232–239. Regensburg, Verlag Friedrich Pustet. EHD i. English Historical Documents, c. 500–1042, ed. D. Whitelock. London, Eyre and Spottiswoode, 1955. EHD ii. English Historical Documents, 1042–1189, eds D. C. Douglas and G. W. Greenaway. London, Eyre and Spottiswoode, 1953. Ekroll, Ø. 1998. Norwegian medieval castles: building on the edge of Europe. Chateau Gaillard 18, 65–74. English, B. 1995. Towns, mottes and ringworks of the Conquest In A. Ayton and J. L. Pryce (eds), The Medieval Military Revolution: State, Society and Military Change in Medieval and Early Modern Europe, 43–62. London, Tauris. English, E. D. 1984. Urban castles in medieval Siena: the sources and images of power. In K. Reyerson and F. Powe (eds), The Medieval Castle: Romance and Reality, 175–198. Medieval Studies at Minnesota 1. Everson, P. 1988. What’s in a name? “Goltho”, Goltho and Bullington. Lincolnshire History and Archaeology 23, 93–99. Extracts from the Issue Rolls of the Exchequer. Issues of the Exchequer: being a collection of payments made out of His Majesty’s revenue, from King Henry III. to King Henry VI. inclusive; with an appendix, extracted and translated from the original rolls of the ancient Pell Office, now remaining in the custody of the Right Honourable Sir John Newport, Bart., ControllerGeneral of His Majesty’s Exchequer, ed. F. Devon. London, John Murray, 1837. Fernie, E. 2002. The Architecture of Norman England. Oxford, Oxford University Press. Field, J. E. 1915–17. Survey of Wallingford, 1550. Berkshire, Buckinghamshire and Oxfordshire Archaeological Journal 21, 111–13; 22, 21–5, 46–9, 82–4; 23, 26–9, 66–73. Finlayson, J. 1967. Morte Arthure: the date and a source for the contemporary reference. Speculum 42:4, 624– 638. Fleming, R. 1991. Kings and Lords in Conquest England. Cambridge, Cambridge University Press. Fleming, R. 1993. Rural elites and urban communities in late Anglo-Saxon England. Past and Present 141, 3–37. Flower, C. T. (ed.) 1923. Public Works in Mediaeval Law. Vol. II. Selden Society Publications, 40. London, Selden Society.

Ford, S. 1995. Castle Farm, Wallingford, Oxon: An archaeological watching brief, 95/17. Thames Valley Archaeological Services, unpublished report. Reading. Ford, S., Hammon, S., Lowe, J., Mundin, A., Jenkins, P., Pitt, J., Preston, S., Saunders, M. J. and Wallis, S. 2012. Investigations on the lines of the Saxon burh defences and outer Norman castle defences. In S. Preston (ed.), Archaeological Investigations in Wallingford, Oxfordshire, 1992–2010, 76–102. Thames Valley Archaeological Monograph 10. Reading, Thames Valley Archaeological Services. Forde-Johnston, J. 1979. Great Medieval Castles of Britain. London, BCA. Foulds, T. 1988. The foundation of Lenton Priory and the reconstruction of its lost cartulary. Transactions of the Thoroton Society 92, 34–42. Foulds, T. 1991. The siege of Nottingham Castle in 1194. Transactions of the Thoroton Society 95, 20–28. Foulds, T. 1997. The medieval town. In J. V. Beckett (ed.), A Centenary History of Nottingham, 72–83. Manchester, Manchester University Press. Fragments of the Close Rolls of 16 and 17 John, ed. R. A. Brown. Pipe Roll Society new series 31. London, 1957. Fuidge, N. M. 1981. Roger Amyce. In P. W. Hasler (ed.), The History of Parliament: The House of Commons 1558–1603. London, HMSO. Online at http://www. historyofparliamentonline.org/volume/1558–1603/ member/amyce-roger-1515–74, (accessed 12 January 2015). Gapper, C. 1998. Plasterers and Plasterwork in City, Court and Country c.1530–c.1660. Unpublished PhD thesis, Courtauld Institute of Art. Online at http://clairegapper. info/introduction-british-renaissance-plasterwork.html (accessed 2 May 2015). Gardiner, S. R. 1899. The Constitutional Documents of the Puritan Revolution 1625–1660. Oxford, Clarendon Press. GASA. Gesta Abbatum Monasterii Sancti Albani, a Thoma Walsingham, Regnante Ricardo Secundo, ejusdem Ecclesiæ Præcentore, Compilata, ed. H. T. Riley. Rolls Series, London, 1867. GDB. Great Domesday, ed. R. W. H. Erskine. London, Alecto Historical Editions, 1986. Gelling, M. 1974. The Place-Names of Berkshire. English Place-Name Society 49. Cambridge, Cambridge University Press. Gelling, M. 1992. The West Midlands in the Early Middle Ages. Leicester, Leicester University Press. Gemmill, E. 2013. The Nobility and Ecclesiastical Patronage in Thirteenth-Century England. Woodbridge, Boydell Press. Gesta Stephani, ed. K. R. Potter, R. H. C. Davis, 2nd edn. Oxford, Clarendon Press, 1976. Gesta Willelmi. The Gesta Guillelmi of William of Poitiers, ed. R. H. C. Davis and M. Chibnall. Oxford, Clarendon Press, 1998. Gibson, E. 1695. Camden’s Britannia, Newly Translated into English. London, Andrew Swalle.

263

Goodall, J. 2011. The English Castle, 1066–1650. London, Yale University Press. Gover, J. E. B., Mawer, A. and Stenton, F. M. 1940. The Place-Names of Nottinghamshire. English PlaceNames Society 18. Cambridge, Cambridge University Press. Gravett, C. 2003. Norman Stone Castles (1). The British Isles 1066–1216. London, Osprey Publishing. Green, D. 1998. The Household and Military Retinue of Edward the Black Prince. Unpublished PhD thesis, University of Nottingham. Green, J. A. 1990. English Sheriffs to 1154. London, Public Record Office. Green, J. A. 1997. The Aristocracy of Norman England. Cambridge, Cambridge University Press. Green, J. A. 2006. Henry I. Cambridge, Cambridge University Press. Greenway, D. E. 1972. Charters of the Honour of Mowbray, 1107–1191, Records of Social and Economic History new series 1. London, British Academy. Griffiths, D. 2003. Exchange, trade and urbanisation. In W. Davies (ed.), From the Vikings to the Normans, 73–104. Oxford, Oxford University Press. Hadley, D. M. 2006. The Vikings in England: Settlement, Society and Culture. Manchester, Manchester University Press. Hagger, M. 2008. The Gesta Abbatum Monasterii Sancti Albani: litigation and history at St Albans. Historical Research 81, 373–398. Hamilton, J. S. 1991. Piers Gaveston and the royal treasure. Albion 23:2, 201–220. Hammond, S. 2006. 24 Castle Street Wallingford 06/3. Thames Valley Archaeological Services, unpublished report. Reading. Harding, D. A. 1985. The Regime of Isabella and Mortimer 1326–1330. Unpublished MA thesis, Durham University. Online at http://etheses.dur.ac.uk/7483/ (accessed 2 May 2015). Harfield, C. G. 1991. A hand-list of castles recorded in the Domesday Book. English Historical Review 106, 371–391. Harrington, P. 2004. English Civil War Archaeology. London, Batsford. Haslam, J. 1987. The second burh of Nottingham. Landscape History 9, 45–51. Haslam, J. 2012. Planning in late Saxon Worcester. http:// jeremyhaslam.files.wordpress.com/2009/12/planningin-late-saxon-worcester.pdf (accessed 2 May 2015). Hassall, T. G. 1976. Excavations at Oxford Castle, 1965– 1973. Oxoniensia 41, 232–308. HEA. Historia ecclesie Abbendonensis: The History of the Church of Abingdon, ed. J. Hudson. Oxford Medieval Texts. 2 vols. Oxford, Clarendon Press, 2002–07. Heale, M. 2004. The Dependent Priories of Medieval English Monasteries. Woodbridge, Boydell Press. Heawood, E. 1950. Watermarks, mainly of the 17th and 18th centuries. Hilversum, Holland, The Paper Publications Society.

Hedges, J. K. 1881. The History of Wallingford, in the County of Berks, From the Invasion of Julius Caesar to the Present Time. 2 vols. London, William Clowes and Sons. Hedges, J. K. 1893. A Short History of Wallingford, Ancient, Mediaeval and Modern, to which is added Rambles in the Neighbourhood. Wallingford, John Bradford. Heiser, R. R. 2000. Castles, constables, and politics in late twelfth-century English governance. Albion 32/1, 19–36. Herbert, N. M. 1971. The Borough of Wallingford 1155– 1400. Unpublished PhD thesis, University of Reading. Higham, R. 2003. Timber castles – a reassessment. In R. Liddiard (ed.), Anglo-Norman Castles, 105–118. Woodbridge, Boydell Press (first published in Fortress 1, 1989, 50–60). Higham, R. and Barker, P. 1992. Timber Castles. Exeter, Exeter University Press. Higham, R. and Barker, P. 2000. Hen Domen, Montgomery: A Timber Castle on the English-Welsh Border. A Final Report. Exeter, Exeter University Press. Higham, R. and Saunders, A. 1997. Public and private defences in British medieval towns. IBI Bulletin 50, 117–128. Hill, D. 2000. Athelstan’s urban reforms. Anglo-Saxon Studies in Archaeology and History 11, 173–185. Hill, J. W. F. 1948. Medieval Lincoln. Cambridge, Cambridge University Press. Hinton, D. 1990. Archaeology, Economy and Society: England from the Fifth to the Fifteenth Century. London, Seaby. Hislop, M. 2010. A missing link: a reappraisal of the date, architectural context and significance of the great tower of Dudley Castle. The Antiquaries Journal 90, 211–233. HKW. H. M. Colvin et al., History of the King’s Works. 6 vols, London, HMSO, 1963–82. Hobley, B. 1988. Saxon London: Lundenwic and Lundenburh, two cities rediscovered. In R. Hodges and B. Hobley (eds), The Rebirth of Towns in the West AD 700–1050, 69–82. CBA Report 68. London, CBA. Holdsworth, P. 1984. Saxon Southampton. In J. Haslam (ed.), Anglo-Saxon Towns in Southern England, 331– 344. Chichester, Phillimore. Holt, R. 2009. The urban transformation in England, 900– 1100. Anglo-Norman Studies 32, 57–78. Hooper, N. 1984. The housecarls in England in the eleventh century. Anglo-Norman Studies 7, 161–176. Hooper, N. 1988. An introduction to the Berkshire Domesday. In A. Williams and R. H. W. Erskine (eds), The Berkshire Domesday, 1–28. London, Alecto Historical Editions. Hope, W. St John 1913. Windsor Castle: An Architectural History, 2 vols. London, Country Life. Horsman, V. 1988. The timber buildings. In V. Horsman, C. Milne and G. Milne (eds), Aspects of Saxo-Norman London I, Building and Street Development, near

264

Billingsgate and Cheapside. London and Middlesex Archaeological Society Special Paper 11. London. Howlett, D. R. 1995. The literary context of Geoffrey of Monmouth: an essay on the fabrication of sources. Arthuriana 5:3, 25–69. HRO. Hertfordshire Record Office. Hudson, J. 1994. Land, Law, and Lordship in AngloNorman England. Oxford, Clarendon Press. Hull, G. and Pine, J. 2001. The Lamb Garage Site, Castle Street, Wallingford, Oxfordshire. Unpublished Report for Thames Valley Archaeological Services, Reading. Hulme, R. 2007. Twelfth century great towers: the case for the defence. The Castle Studies Group Journal 21, 210–239. Hurnard, N. D. 1949. The Anglo-Norman franchises. English Historical Review 64, 289–327, 433–460. Hurst, H. 1984. The archaeology of Gloucester Castle: an introduction. Transactions of the Bristol and Gloucestershire Archaeological Society 102, 73–128. Impey, E. 2002. The Turris Famosa at Ivry-la-Bataille, Normandy. In G. Meirion-Jones, E. Impey and M. Jones (eds), the Seigniorial Residence in Western europe, aD 800–1600, 189–210. British Archaeological Reports International Series 1088. oxford, BAR Publishing. Impey, E. 2008a. London’s early castles and the context of their creation. In E. Impey (ed.), The White Tower, 13–26. London and New Haven, Yale University Press. Impey, E. 2008b. The ancestry of the White Tower. In E. Impey (ed.), The White Tower, 227–241. London and New Haven, Yale University Press. Johnson, P. 2010. Making the Market: Victorian Origins of Corporate Capitalism. Cambridge, Cambridge University Press. Jope, E. M. 1952. Late Saxon pits under Oxford Castle mound: excavations in 1952. Oxoniensia 17/18, 77– 111. Jope, E. M. and Threfall, R. I. 1959. The twelfth-century castle at Ascot Doilly, Oxfordshire: its history and excavation. The Antiquaries Journal 39, 219–273. Journal of the House of Commons: Volume 2, 1640–1643. HMSO, London, HMSO, 1802. Online at http://www. british-history.ac.uk/commons-jrnl/vol2 (accessed 2 May 2015). JSL. Journal of Sir Samuel Luke, Scoutmaster General to the Earl of Essex 1643–44, ed. by I. G. Philip. 3 vols. Oxfordshire Record Society, 29, 31, 33. Oxford, 1950–52. Keats-Rohan, K. S. B. 1989a. The making of Henry of Oxford: Englishmen in a Norman world. Oxoniensia 54, 287–310. Keats-Rohan, K. S. B. 1989b. The devolution of the honour of Wallingford 1066–1154, Oxoniensia 54, 311–318. Keats-Rohan, K. S. B. 2009. The genesis of the honour of Wallingford. In K. S. B. Keats-Rohan and D. R. Roffe (eds), the Origins of the Borough of Wallingford: archaeological and Historical Perspectives, 52–67. British Archaeological Reports British Series 494. oxford, BAR Publishing.

Keats-Rohan, K. S. B. 2012. Through the eye of the needle: Stigand, the Bayeux Tapestry and the beginnings of the Historia Anglorum. In D. Roffe (ed.), The English and Their Legacy, 900–1200. Essays in Honour of Ann Williams, 159–174. Woodbridge, Boydell Press. Keats-Rohan, K. S. B. 2015. Fortunes of war: safeguarding Wallingford Castle and Honour 1135–60. In P. Dalton and D. Luscombe (eds), Rulership and Rebellion in the Anglo-Norman World, c. 1066–c. 1216. Essays in Honour of Professor Edmund King, 125–139. Aldershot, Ashgate. Keats-Rohan, K. S. B. and Roffe D. R. (eds) 2009. The Origins of the Borough of Wallingford: archaeological and Historical Perspectives. British Archaeological Reports British Series 494. oxford, BAR Publishing. Keefe, T. K. 1990. Place-date distribution of royal charters and the historical geography of patronage strategies at the court of King Henry II Plantagenet. Haskins Society Journal 2, 179–188. Kemp, B. (ed.) 1999. Twelfth-Century English Archidiaconal and Vice-Archidiaconal Acta. Canterbury and York Society. Woodbridge, Boydell. Kenyon, J. R. 1990. Medieval Fortifications. Leicester, Leicester University Press. Kershaw, M. J. 1998. Knaresborough Castle. Harrogate, Harrogate Borough Council. Keynes, S. 1993. A lost cartulary of St Albans Abbey. Anglo-Saxon England 22, 253–279. Keyser, C. E. 1907. Note on a carved Norman stone from the museum of Wallingford Castle. Proceedings of the Society of Antiquaries, 2nd series 21, 118–123. Kightly, C. 1993. The Borough of Wallingford. In J. S. Roskell, L. Clark, and C. Rawcliffe (eds), The History of Parliament: The House of Commons 1386–1421. Stroud, Alan Sutton. Online at www. historyofparliamentonline.org/volume/1386–1421/ constituencies/wallingford (accessed 7 November 2012). King, E. 1796. Vestiges of Oxford Castle. London, George Nichol. King, E. 1999. The memory of Brian fitz Count. Haskins Society Journal 13, 75–98. Laban, G, with contributions by L. Blackmore, B. Richardson, J. Morris and K. Stewart, 2011. Evidence for a siege castle at the Lister Wilder site, The Street, Crowmarsh Gifford. Oxoniensia 78, 189–211. Lapidge, M. (ed.) 1999. The Blackwell Encyclopaedia of Anglo-Saxon England. Oxford, Blackwell. Lawson, M. K. 1993. Cnut. The Danes in England in the Early Eleventh Century. London and New York, Longman. Leary, J., Canti, M., Field, D., Fowler, P., Marshall, P. and Campbell, G. 2013a. The Marlborough Mound, Wiltshire. A Further Neolithic Monumental Mound by the River Kennet. Proceedings of the Prehistoric Society 79, 1–27. Legge, M. D. 1963. Anglo-Norman Literature and its Background. Oxford, Clarendon Press.

265

Leland. Leland, J. The Itinerary of John Leland in or about the Years 1535–1543, ed. L. T. Smith. 5 vols. London, George Bell and Sons, 1908–10. Liddiard, R. 2000. Landscapes of Lordship. British Archaeological Reports British Series 309. oxford, BAR Publishing. Liddiard, R. (ed.) 2003. Anglo-Norman Castles. Woodbridge, Boydell. Liddiard, R. 2005. Castles in Context. Power, Symbolism and Landscape, 1066 to 1500. Macclesfield, Windgather Press. Liu, Y. 2005. Richard de Beauchamp and the uses of romance. Medium Aevum 74, 271–287. LPFD. Letters, Papers, Foreign and Domestic, Henry VIII, vol. 4, ed. J. S. Brewer, London, 1867; vol. 13 part I, ed. J. Gairdner, London, 1892. LSL. The Letter Books of Sir Samuel Luke 1644–45, ed. H. G. Tibbutt. London, HMSO, 1963. Luard, H. R. (ed.) 1869. Annales Monastici, iv, Annales Monasterii de Oseneia, Chronicon vulgo dicto Chronicon Thomae Wykes, Annales Prioratus de Wigornia. Rolls Series 36. London. Luckett, D. A. 1995. The Thames Valley conspiracies against Henry VII. Historical Research 68, 164–172. Mack, K. 1984. Changing thegns: Cnut’s conquest and the English aristocracy. Albion 16, 375–387. Mahany, C., Burchard, A. and Simpson, G. 1982. Excavations at Stamford, Lincolnshire, 1963–9. Society for Medieval Archaeology Monograph 9. Leeds, Maney. Mahany, C. and Roffe, D. 1983. Stamford: the development of an Anglo-Scandinavian borough, Anglo-Norman Studies 5, 197–219. Maitland, F. W. 1897. Domesday Book and Beyond. Cambridge, Cambridge University Press. Marks, R. 1993. Stained Glass in England during the Middle Ages. London, Routledge. Marshall, P. 2002. The great tower as residence. In G. Meirion-Jones, E. Impey and M. Jones (eds), the Seigniorial Residence in Western europe, aD 800–1600, 27–44. British Archaeological Reports International Series 1088. oxford, BAR Publishing. Marshall, P. and Foulds, T. 1997. The royal castle. In J. V. Beckett (ed.), A Centenary History of Nottingham, 43–55. Manchester, Manchester University Press. Marshall, P. and Samuels, J. 1994. Recent excavations at Newark Castle, Nottinghamshire. Transactions of the Thoroton Society of Nottinghamshire 98, 49–57. Mason, E. 1984. Legends of the Beauchamps ancestors: the use of baronial propaganda in medieval England. Journal of Medieval History 10, 25–40. Mastoris, S. N. 1981. The boundary between the English and French boroughs of mediaeval Nottingham. Transactions of the Thoroton Society 81, 68–74. Matt. Paris. Matthaei Parisiensis: Monachi Sancti Albani, Chronica Majora, ed. H. R. Luard. 7 vols. London, Longman, 1872–83. Matthies, A. L. 1991. The medieval wheelbarrow. Technology and Culture 32, 356–364.

McCarthy, M. R. and Brooks, C. M. 1988. Medieval Pottery in Britain, AD 900–1600. Leicester, Leicester University Press. Mellor, M. 1994. A synthesis of Middle and Late Saxon, medieval and early post-medieval pottery in the Oxford region. Oxoniensia 59, 17–218. Mellor, M. 2003. The Saxon and medieval ceramic finds from the town sites. In A. Dodd (ed.), Oxford before the University. The Late Saxon and Norman Archaeology of the Thames Crossing, the Defences and the Town, 326–345. Thames Valley Landscapes Monograph 17. Oxford, Oxford Archaeology. Menache, S. 1984. Isabelle of France, queen of England – a reconsideration. Journal of Medieval Studies 10, 107–124. Midgley, M. L. 1942. Ministers’ Accounts of the Earldom of Cornwall 1296–1297, Vol. II. Camden Society 3rd series 67. London. Miles, T. J. 1986. The excavation of a Saxon cemetery and part of the Norman castle at North Walk, Barnstaple. Journal of the Devon Archaeological Association 44, 59–84. Mileson, S. A. 2009. Parks in Medieval England. Oxford, Oxford University Press. Milne, J. G. 1940. Muniments of Holy Trinity Priory, Wallingford. Oxoniensia 5, 50–77. Mon. Angl. Monasticon Anglicanum, ed. W. Dugdale; rev. ed. by J. Caley, H. Ellis and B. Bandinel. 6 vols. London, 1817–30 (repr. 1846). Moore, J. S. 2000. Anglo-Norman garrisons. AngloNorman Studies 22, 205–260. Moorhouse, S. A. 1983. Documentary evidence and its potential for understanding the inland movement of medieval pottery. Medieval Ceramics 7, 45–87. Morris, R. 1989. Churches in the Landscape. London, Phoenix. Morte Arthure, or, The death of Arthur, ed. E. Brock, Early English Text Society old series 8. London, Truebner, 1871. Mortimer, I. (ed.) 1995. Berkshire Glebe Terriers 1634. Berkshire Record Society 2. Reading. Mortimer, I. 2010. Medieval Intrigue. Decoding Royal Conspiracies, London, Continuum. Munby, J. and Walton, H. 1990. The building of New Road. Oxoniensia 55, 123–130. Mundin, A. 2008. Former Lamb Garage site, Castle Street, Wallingford, Oxfordshire. An Archaeological Watching Brief for Greyswood Limited. Unpublished report, Thames Valley Archaeological Services, Reading. Naismith, R. 2014. The social significance of monetization in the early middle ages. Past and Present 223, 3–39. Nevell, R. 2014/15. Castles as prisons, The Castle Studies Group Journal 28, 203–224. Newcombe, A. P. 1940. Calendar of the Manuscripts of Major-General Lord Sackville K.B.E., C.B., C.M.G., [formerly] Preserved at Knowle, Sevenoaks, Kent, Volume I Cranfield Papers 1551–1612. Historical Manuscripts Commission 80. London, HMSO.

266

Nicholas, D. 1997. The Growth of the Medieval City: From Late Antiquity to the Early Fourteenth Century. London, Longman. Non. Inq., Nonarum Inquisitiones in Curia Scaccarii Temp. Regis Edwardi III, ed. G. Vanderzee, Record Commission. London, Eyre and Strahan, 1807. OAU. Oxford Archaeological Unit. ODNB 2004. Oxford Dictionary of National Biography, 60 vols. Oxford, Oxford University Press. Available online at www.oxforddnb.com/ (accessed 2 May 2015). Oman, C. 1926. Castles. London, Great Western Railway. Orderic. The Ecclesiastical History of Orderic Vitalis, ed. and trans. M. Chibnall. 6 vols. Oxford, Oxford University Press, 1969–80. Ormrod, W. M. 2011. Edward III. New Haven, Yale University Press. ORO. Oxfordshire Record Office. Oseney. The Oseney Cartulary, ed. H. E. Salter. 6 vols. Oxford Historical Society 89–91, 97, 98, 101. Oxford, 1929–36. Ottaway, P. 1992. Archaeology in British Towns: From the Emperor Claudius to the Black Death. London, Routledge. Oxford Archaeological Unit 1992. Castle Farm, Wallingford, Oxon: An archaeological watching brief/ excavation Feb 1992. Oxford Archaeological Unit, unpublished report. Oxford. Oxley, J. (ed.) 1986. Excavations at Southampton Castle. Southampton Archaeology Monograph 3. Southampton, Southampton City Museums. Painter, S. 1935. Castle-guard. American Historical Review 40, 450–459. Palliser, D. M., Slater, T. R., and Dennison, E. P. 2000. The topography of towns 600–1300. In D. M. Palliser (ed.), The Cambridge Urban History of Britain: Volume I, 600–1540, 153–186. Cambridge, Cambridge University Press. Pantin, W. 1958. Monuments or muniments. The interrelation of material remains and documentary sources. Medieval Archaeology 2, 158–168. Parnell, G. 1983. The western defences of the innermost ward, Tower of London. Transactions of the London and Middlesex Archaeological Society 34, 107–150. Payne-Gallwey, R. 1903. The Crossbow. London, Longmans, Green and Co. Pettifer, A. 1995. English Castles. A Guide by Counties. Woodbridge, Boydell Press. Philip, I. G. 1937. River navigation at Oxford during the Civil War and Commonwealth. Oxoniensia 2, 152–165. Philip, I. G. (ed.) 1950, The Journal of Sir Samuel Luke. Oxfordshire Record Society 29, 31, and 33. Oxford. Philipps, S. 2010. Edward II. New Haven, Yale University Press. Platt, C. 1982. The Castle in Medieval England and Wales. London, Vintage. Poore, D. and Norton, A. 2009. Excavations at Oxford Castle: Oxford’s western quarter from the Mid-Saxon period to the late eighteenth century. Oxoniensia 74, 1–18.

Porter, S. 1997. Destruction in the English Civil Wars. Stroud, Sutton Publishing. Pounds, N. J. G. 1990. The Medieval Castle in England and Wales. A Social and Political History. Cambridge, Cambridge University Press. PQW. Placita de Quo Warranto, ed. W. Illingworth, Record Commission. London, 1818. PR. Pipe Roll, Great Roll of the Pipe, Pipe Roll Society, London, 1884–in progress. Prestwich, M. 1992. English castles in the reign of Edward II. Journal of Medieval History 8, 159–178. Pringle, D. 1984. King Richard I and the walls of Ascalon. Palestine Exploration Quarterly 116, 133–147. Prior, M. 1981. The accounts of Thomas West of Wallingford, a sixteenth-century trader on the Thames. Oxoniensia 46, 73–93. Proceedings and Ordinances of the Privy Council, vol. iii, ed. H. Nicholas, London, Eyre and Spottiswood, 1837. Pugh, R. B. 1955. The King’s prisons before 1250. Transactions of the Royal Historical Society 5th, series 5, 1–22. Quiney, A. 1999. Hall or chamber? That is the question. The use of rooms in post-Conquest houses. Architectural History 42, 24–46. RAC. Reading Abbey Cartularies, ed. B. Kemp. 2 vols. Camden Society, 4th ser. 31 and 33. London, 1986–88. Rakoczy, L. 2007. A Re-interpretation of Castle Slightings in the English Civil War. Unpublished DPhil thesis, University of York. Rannie, D. W. (ed.) 1898. Remarks and Collections of Thomas Hearne: Vol. IV (15 December 1712–30 November 1714). Oxford, Clarendon Press. RBE. Red Book of the Exchequer, ed. H. Hall. 3 vols. Rolls Series 99. London, 1896. RCHM 1877. Sixth Report of the Royal Commission on Historical Manuscripts. Part I. Report and Appendix. Royal Commission on Historical Manuscripts 6. London, HMSO. RCHM 1879. Seventh Report of the Royal Commission on Historical Manuscripts Part I. Report and Appendix. Royal Commission on Historical Manuscripts 7. London, HMSO. RCHM 1888. Supplementary Report on the Manuscripts of the Duke of Leeds, the Bridgewater Trust, Reading Corporation, the Inner Temple, etc. (11th Rep., app. vii). Royal Commission on Historical Manuscripts 22. London, HMSO. RCHM 1940. Calendar of the manuscripts of MajorGeneral Lord Sackville, K.B.E., C.B., C.M.G., [formerly] preserved at Knole, Sevenoaks, Kent. Vol. I: Cranfield papers, 1551–1612. Royal Commission on Historical Manuscripts 80. London, HMSO. RCHME 1959. An Inventory of the Historical Monuments in the City of Cambridge: Part II. London, HMSO. RCHME 1985. An Inventory of Archaeological Sites and Churches in Northampton. London, HMSO. REBP. Register of Edward, the Black Prince preserved in the Public Record Office, ed. M. C. B. Dawes. 4 vols. London, HMSO, 1930–33.

267

Renn, D. F. 1960. The keep of Wareham Castle. Medieval Archaeology 4, 56–68. Renn, D. F. 1968. Norman Castles in Britain. London, John Baker. Renn, D. F. 2003. Burgeat and Gonfanon: two sidelights from the Bayeux Tapestry. In R. Liddiard (ed.), AngloNorman Castles, 69–90. Woodbridge, Boydell. Renoux, A. 2000. Mayenne: de la villa au castrum (VIle– XIIIe s.). Comptes-rendus des Séances de l’Académie des Inscriptions et Belles-Lettres 144:1, 211–231. Reynolds, A. 1999. Late Anglo-Saxon England: Life and Landscape. Stroud, Tempus. Reynolds, N. 1975. Investigations in the Observatory Tower, Lincoln Castle. Medieval Archaeology 19, 201–205. Reynolds, S. 1994. Fiefs and Vassals: The Medieval Evidence Reconsidered. Oxford, Oxford University Press. RH. Rotuli hundredorum temp. Hen. III et Edw. I, ed. W. Illingworth. 2 vols. Record Commission. London, 1812 and 1818. Riall, N. 2003. The new castles of Henry de Blois as bishop of Winchester: the case against Farnham, Surrey. Medieval Archaeology 47, 115–129. Rickard, J. 2002. The Castle Community: The Personnel of English and Welsh Castles 1272–1422. Woodbridge, Boydell Press. Rigold, S. E. 1954. Totnes Castle: recent excavations. Transactions of the Devonshire Association 86, 228– 256. Rigold, S. E. 1975. Structural aspects of medieval timber bridges. Medieval Archaeology 19, 49–91. Rimmer, J. 2006. A re-assessment of the use of building accounts for the study of medieval urban houses. Proceedings of the Second International Congress on Construction History 3, 2599–2612. Rodwell, K. A. 1976. Excavations on the site of Banbury Castle, 1973–4. Oxoniensia 51, 98–148. Rodwell W., Davison, B. K., Flook, R. and Bedwin, O. 1993. The Origins and Early Development of Witham, Essex: A Study in Settlement and Fortification, Prehistoric to Medieval. Oxford, Oxbow Books. Roffe, D. R. 1986. The origins of Derbyshire. Derbyshire Archaeological Journal 106, 102–122. Roffe, D. R. 1987. Nottinghamshire and the North: A Domesday Study. Unpublished PhD thesis, University of Nottingham. Online at http://www.roffe.co.uk/ phdframe.htm (accessed 2 May 2015). Roffe, D. R. 1990. From thegnage to barony: sake and soke, title, and tenants-in-chief. Anglo-Norman Studies 12, 157–176. Roffe, D. R. 1992. Hundreds and wapentakes. In A. Williams and G. H. Martin (eds), The Lincolnshire Domesday, 32–42. London, Alecto Historical Editions. Roffe, D. R. 1997. Anglo-Saxon Nottingham and the Norman Conquest. In J. V. Beckett (ed.), A Centenary History of Nottingham, 24–42. Manchester, Manchester University Press.

Roffe, D. R. 2007. Decoding Domesday. Woodbridge, Boydell. Roffe, D. R. 2009. Wallingford in Domesday and beyond. In K. S. B. Keats-Rohan and D. R. Roffe (eds), the Origins of the Borough of Wallingford: archaeological and Historical Perspectives, 27–51. British Archaeological Reports British Series 494. oxford, BAR Publishing. Roffe D. R. 2014. An Englishman’s home. Online at http://www.roffe.co.uk/horncastle.htm (accessed 8 December 2014). Roll of Divers Accounts for the Early Years of Henry III, ed. F. Cazel, Pipe Roll Society new series 44, 1982. ROP The Royalist Ordnance Papers, 1642–1646, ed. I. Roy, 3 vols, Oxfordshire Record Society 29, 31, 33. Oxford, 1929–36. Rot. Chart. Rotuli Chartarum Rotuli chartarum in turri Londinensi asservati, 1199–1216, ed. T. D. Hardy, Record Commission, 1837. Rot. Litt. Claus. Rotuli Litterarum Clausarum in Turri Londinensi Asservati, ed. T. D. Hardy. 2 vols. Record Commission, London, 1833–44. Rot. Litt. Pat. Rotuli Litterarum Patentium in Turri Londinensi Asservati, ed. T. D. Hardy. Record Commission, London, 1835. Rot. Parl. Rotuli Parliamentorum : ut et Petitiones et Placita in Parliamento, 1272–1533, 6 vols. London, 1783; new edition, The Parliament Rolls of Medieval England, Gen. ed. C. Given-Wilson. 16 vols. Woodbridge, Boydell and Brewer for National Archives, 2005; electronic version PROME, CDROM. Leicester, Scholarly Digital Editions, 2005. Round, J. H. 1911. The King’s Serjeants and Officers of State, with their Coronation Services. London, J. Nisbet. Rowley, T. 1983. The Norman Heritage, 1055–1200. London, Routledge and Kegan Paul. RRAN. Regesta Regum Anglo-Normannorum, i, ed. H. W. C. Davis; ii, ed. C. Johnson and H. A Cronne; iii, ed. H. A. Cronne and R. H. C. Davis. 3 vols. Oxford, Clarendon Press, 1913–68. Rushworth, J. 1722. Historical Collections of Private Passages of State, March-May 1646, Volume 6, 1645– 47. London, D Browne. Online at http://www.britishhistory.ac.uk/rushworth-papers/vol6/pp249–276; also May-August 1646, pp 276–298 (accessed 2 May 2015). Rymer, Foedera. Rymer’s Foedera. 17 vols. London, John Neulme, 1739–45. Sawyer. P. H. Sawyer, Anglo-Saxon Charters: an Annotated List and Bibliography, Royal Historical Society Guides and Handbooks 8. London, Royal Historical Society. 1968. Updated as The Electronic Sawyer at http:// www.esawyer.org.uk/searchfiles/index.html, accessed 27 August 2014. Sainty, J. C. 1999. Officers of the Green Cloth: clerks and clerk comptrollers. Provisional lists 1485–1646. Institute of Historical Research, University of London. Online at http://www.history.ac.uk/resources/office/ greencloth_clerk (accessed 2 May 2015).

268

Salter, H. E. 1925. Two deeds about the Abbey of Bec. English Historical Review xl, 73–76. Salter, H. E. 1929. Facsimiles of Early Charters in Oxford Muniment Rooms. Oxford, Clarendon Press. Salter, H. E. 1936. Medieval Oxford. Oxford Historical Society 100. Oxford. Salzman, L. F. 1952. Buildings in England down to 1540. Oxford, Clarendon Press. Sanders, I. J. 1960. English Baronies: A Study of their Origins and Descent, 1086–1327. Oxford, Oxford University Press. Saunders, A. D. 1980. Lydford Castle, Devon. Medieval Archaeology 24, 123–186. Saunders, A. D. 2006. Excavations at Launceston Castle, Cornwall. The Society for Medieval Archaeology Monograph 24. Leeds, Maney. Saunders, M. J. 1995. Castle Farm, Wallingford, Oxon. An archaeological watching brief 95/17b. Thames Valley Archaeological Services, unpublished report. Reading. Sayers, J. 1971. Papal privileges for St Albans Abbey and its dependencies. In D. Bullough and R. Storey (eds), The Study of Medieval Records: Essays in Honour of Kathleen Major, 57–84. Oxford, Clarendon Press. Scammell, J. 1993. The formation of the English social structure: freedom, knights, and gentry, 1066–1300. Speculum 68:3, 591–618. Schofield, J. and Steur, H. 2007. Urban settlement. In J. Graham-Campbell (ed.), The Archaeology of Medieval Europe: Vol. 1, Eighth to Twelfth Centuries AD, 111– 153. Aarhus, Aarhus University Press. Schulz, H. C. 1938. The monastic library and scriptorium at Ashridge. Huntingdon Library Quarterly 1:3, 305– 311. Secker, D. 2006. Four Saxo-Norman churches near Wallingford. Oxoniensia 71, 31–68. Shelby, L. 1964. The role of the master mason in medieval English building. Speculum 39:3, 387–403. Shelby, L. 1965. Medieval masons’ tools. II. Compass and square. Technology and Culture 6:2, 236–248. Shelby, L. 1972. The geometrical knowledge of mediaeval master masons. Speculum 47:3, 395–421. Shepherd Popescu, E. 2009. Norwich Castle: Excavations and Historical Survey, 1987–98. Part I: Anglo-Saxon to c. 1345; Part II c.1345 to Modern. East Anglian Archaeology Report 132. East Dereham. Shoesmith, R. 1980. Hereford City Excavations, Volume 1: Excavations at Castle Green. CBA Research Report 36. London, CBA. Simpson, D. 1960. The Building Accounts of Tattershall Castle, 1434–72. Lincoln Record Society 55. Lincoln, Lincoln Record Society. Skelton, R. A. 1970. The military surveyor’s contribution to British cartography in the 16th century. Imago Mundi 24, 77–83. Skirmer. Reading Public Library, BU/D/L6867. Slater, T. R. 2007. The landscapes of Anglo-Norman towns: Anglo-European comparisons. In M. Gardiner and S. Rippon (eds), Medieval Landscapes, 11–26. Macclesfield, Windgather.

Smith, D. M. (ed.) 1980. English Episcopal Acta I: Lincoln 1067–1185. Oxford, British Academy London Smyth, A. P. 1987. Scandinavian York and Dublin. Dublin, Irish Academic Press. Soden, I. 2006. Ranulf de Blondeville. Stroud, Amberley Publishing. Spalding, R. (ed.) 1990. The Diary of Bulstrode Whitelocke 1605–1675. Oxford, British Academy. Speed, G., Christie, N., Creighton, O. H. and Edgeworth, M. 2009. Charting Saxon and medieval urban growth and decay at Wallingford. Medieval Archaeology 53, 355–363. Stafford, P. 1985. The East Midlands in the Early Middle Ages. Leicester, Leicester University Press. Stalls, C. 1995. Possessing the Land: Aragon’s Expansion into Islam’s Ebra Frontier under Alfonso the Battler 1104–1134. New York, Brill. Steane, J. 2008. The Abingdon monks’ map. Oxoniensia 73, 17–32. Stenton, F. M. 1932. The First Century of English Feudalism, 1066–1166. Oxford, Clarendon Press. Stevenson, J. and Carter, A. 1973. The Raid on Chinnor and the Fight at Chalgrove Field. Oxoniensia 38, 346– 356. Stevenson, W. H. and Stapleton, A. 1895. The Religious Foundations of Old Nottingham. Nottingham. Stocker, D. and Vince, A. 1997. The early Norman castle at Lincoln and a re-evaluation of the original west tower of Lincoln cathedral. Medieval Archaeology 41, 223–233. R. Storey. 1998. The Tower of London and the Garderobae Armorum, Royal Armouries Yearbook 3, 176–183. Stubbs, W. 1882. Chronicles of the Reigns of Edward I and Edward II, Vol. 1, Annales Paulini de Tempore Edwardi Secundi. Rolls Series 76. London, Longman and Co. Sturdy, D. and Munby, J. 1985. Excavations in Cornmarket and Queen Street 1959–62. Oxoniensia 50, 47–94. Summary Catalogue. Summary Catalogue of PostMedieval Western Manuscripts in the Bodleian Library Oxford, Acquisitions 1916–1975, M. Clapinson and T. D. Rogers, 3 vols. Oxford, Clarendon Press, 1991. Tait, J. 1897. Review of Domesday Book and Beyond. English Historical Review 12, 768–77. Tait, J. 1936. The Medieval English Borough. Manchester, Manchester University Press. Tatton-Brown, T. 1988. The Anglo-Saxon towns of Kent. In D. Hooke (ed.), Anglo-Saxon. Settlements, 213–232. Oxford, Blackwell. Taxatio. Taxatio Ecclesiastica Angliae et Walliae Auctoritate Papae Nicholai IV. Record Commission, London, 1802: all references are to http://www. hrionline.ac.uk/taxatio/ Taylor, P. 1995. The early St Albans endowment and its chronicler. Historical Research 68, 119–142. Thacker, F. S. 1968. The Thames Highway. 2 vols. Newton Abbot, David and Charles. Thomas, R. 1975. Wallingford Castle ABS Site. Unpublished report, Oxfordshire Historic Environment Record. Oxford.

269

Thomas, R. W. 2002. Manuscripts from St. Albans Abbey, 1066–1235: Texts v.1. Woodbridge, Boydell Press. Thompson, E. M. (ed.) 1889. Chronicon Galfridi le Baker de Swynebroke. Oxford, Clarendon Press. Thompson, M. W. 1960. Recent excavations in the keep of Farnham Castle, Surrey. Medieval Archaeology 4, 81–94. Thompson, M. W. 1961. Motte substructures. Medieval Archaeology 5, 305–306. Thompson, M. W. 1987. The Decline of the Castle. Cambridge, Cambridge University Press. Thompson, M. W. 1991. The Rise of the Castle. Cambridge, Cambridge University Press. Thompson, M. W. 1995. The Medieval Hall. The Basis of Secular Domestic Life, 600–1600 AD. Aldershot, Scholar Press. Thorn, F. R. 1990. Hundreds and wapentakes. In A. Williams and R. W. H. Esrkine (eds), The Sussex Domesday, 29–33. London, Alecto Historical Editions. Tighe, R. R. and Davis, J. E. 1858. The Annals of Windsor being a History of the Castle and Town with some Account of Eton, 2 vols. London, Longman. Tiller, K. and Darkes, G. (eds) 2011. An Historical Atlas of Oxfordshire. Oxfordshire Record Society 67. Chipping Norton, Oxfordshire Record Society. Tilley, C. 2011. The Honour of Wallingford 1066–1300. Unpublished PhD thesis, King’s College, London. TNA. The National Archives. Tomkinson, A. 1959. Retinues at the tournament of Dunstable, 1309. English Historical Review 74, 70–89. Toohey, K. 2000. King Arthur’s swords. In B. PostonAnderson (ed.), The Grail Quest Papers 26–38. Sydney, University of Technology. Torigny. The Chronicle of Robert of Torigny, Abbot of Mont St Michel, ed. R. Howlett in Chronicles of the Reign of Stephen, Henry II and Richard I, vol. iv, Rolls Series 82, London 1886. Tout, T. F. 1911. Firearms in England in the Fourteenth Century, English Historical Review 26, 666–702. Tout, T. F. 1928. Chapters in the Administrative History of England, 4 vols. Manchester, Manchester University Press. Trokelowe. Johannis de Trokelowe et Henrici de Blaneforde Chronica et Annales, ed. H. T. Riley, Rolls Series 26. London, 1866. TV. Testamenta Vetusta: being Illustrations from Wills, of Manners, Customs, &c. as well as of the Descents and Possessions of Many Distinguished Families. From the Reign of Henry the Second to the Accession of Queen Elizabeth, ed. N. Harris. 2 vols. London, Nichols and Son, 1826. TWHAS. The Wallingford Historical and Archaeological Society. VCH The Victoria County History, founded 1899 and ongoing. Cited by county abbreviation, followed by volume number, and page reference. All are available online at www.british-history.ac.uk. Vince, A. 1990. Saxon London: An Archaeological Investigation. London, Batsford.

Walker, D. 1991. Gloucestershire castles. Transactions of the Bristol and Gloucestershire Archaeological Society 94, 5–23. Waltham Abbey. The Early Charters of Waltham Abbey, 1062–1230, ed. R. Ransford. Woodbridge, Boydell, 1989. War Office, 1922. Statistics of the Military Effort of the British Empire during the Great War 1914–1920. London, HMSO. Warren, W. L. 1973. Henry II. London, Methuen. Warren, W. L. 1987. The Governance of Norman and Angevin England, 1086–1189. London, Arnold. Watts, J. 2004. Pole, William de la, first duke of Suffolk (1396–1450). Oxford University Press, online edition, Sept. 2012 http://ezproxy.ouls.ox.ac.uk:2204/view/ article/22461, accessed 3 Aug 2014. Weber, H. R. 1810. Metrical Romances of the Thirteenth, Fourteenth, and Fifteenth Centuries. Edinburgh, A. Constable and co. Webster, L. E. and Cherry, J. 1973. Medieval Britain in 1972. Medieval Archaeology 17, 138–188. Weiss, J. 2012. Mordred. In N. Cartlidge (ed.), Heroes and Anti-Heroes in Medieval Romance, 81–98. Woodbridge, D. S. Brewer. Weston, J. L. 1925. Who was Brian des Iles? Modern Philology 22:5, 405–411. Wethered, F. T. 1898. Hurley Charters and Deeds. London, Bedford Press. Whitelocke, B. 1732. Memorials of the English Affairs: or, An Historical Acount of what passed from the Beginning of the Reign of King Charles the First to Charles the Second His Happy Restauration. London, J. Tonson. Wilkinson, I. P. Tasker, A., Gouldwell, A., Williams, M., Edgeworth, M., Zalasiewicz, J. and Christie, N. 2010. Micropalaeontology reveals the source of building materials for a defensive earthwork (English Civil War?) at Wallingford Castle, Oxfordshire. Journal of Micropalaeontology 29, 87–92. Williams, A. 2003. A bell-house and a burh-geat: lordly residences in England before the Norman Conquest. Medieval Knighthood 4, 221–240 (Reprinted in R. Liddiard (ed.), Anglo-Norman Castles, 23–40. Woodbridge, Boydell Press). Williams, T. J. T. 2012. The mint at Wallingford: an introduction to the corpus. Yorkshire Numismatist 4, 141–158. WRO. Wiltshire Record Office. Wykes, T. Chronicon vulgo dicto Chronicon Thomae Wykes. In Annales Monastici iv, ed. H. R. Luard. 5 vols. London, Longman, Roberts and Green, 1864–69. Woodburn, B. and Guy, N. 2005–6. Arundel Castle. Castle Studies Group Journal 19, 9–24. Young, C. S. B. 1985. Archaeology in Nottingham: the preConquest borough. In S. N. Mastoris (ed.), History in the Making, 1–4. Nottingham, Nottingham Museums. Zupko, R. E. 1985. A Dictionary of Weights and Measures for the British Isles: The Middle Ages to the Twentieth Century. Memoirs of the American Philosophical Society 168. Philadelphia, American Philosophical Society. 270

INDEX

Index

All references are to Wallingford unless otherwise indicated; place references are All references are to Wallingford unless otherwise indicated; place references are to pre-1974 county boundaries. to pre-1974 county boundaries. Before 1538 names start with first name; after 1538, names start with surname. Before 1538 names start with first name; after, start with surname.

Abingdon (oxon) abbey abbot bridge Christ Church School Abergavenny (Wales) Acton, nr London (Middx) Adam de Catmere, sheriff Adam the plumber Adelelm, abbot of Abingdon Æthelflæda, lady of the Mercians Æthelred, Ealdorman akenate Alan Fergant, duke of Brittany Alan son of Ansfrid Alan son of Siward Alexander II, king of Scots (121449) Alexander III, pope Alexander Dublet Alfred, king (871-99) Alice Chaucer, duchess of Suffolk Alice de Colleville All Hallows: see All Saints All Saints, church and parish

alure (passage-way or wall-walk) Amalric son of Ralph Ambrose de Newburgh Amice, wife of Nigel d’Albini Amyce, Roger Anastyes Place Andrew of Sutton Anne Boleyn, queen, wife of

96, 103, 238, 239, 241, 242, 247 36, 56, 217, 218, 234 3, 34, 49, 51, 108 103 117 55 63 50 211 108 199 199 72, 76 55 55, 72 218 58

Henry vIII Anne of Bohemia, queen, wife of Richard II Anthony, Earl Rivers Arding, Isaac, Rev. armaments Armitage, Ella armoury Arnold, John, of Bensington Arnold, Robert Arthur, king Arthur, prince of Wales Arthurian romances: see also Chrétien de Troyes Arundel Castle (Sussex) Ascalon, Crusader castle, Israel Ascot d’oilly Castle (oxon) Ashridge Priory (Herts) assize: see also Clarendon Asser Astell (Austell), John Aston Rowant (oxon), church

213 102 2-3, 36, 116, 194 65 218

Aston Tirrold (oxon) Athelstan, king (924-39) Athow, Thomas Aubrey, John Aubrey de veer, constable Augmentations, Court of aula: see Great Hall, Red Hall Aylesbury (Bucks)

3, 27, 32, 38-9, 52-4, 76, 1478, 150, 155, 157, 177, 183, 192, 234, 236, 244-6 57, 71, 85-6, 89, 94, 109, 110 218 211 215 116, 117, 120, 135, 227 33 69, 112 67

bailey: inner (or inward ward)

271

103, 107 115 142-3 64, 95-8, 112, 121, 236 1, 46 60, 63, 75, 93, 95 192 192 73-5 67, 102, 115 73-6 40, 47, 207-8 12, 59 40 67, 83 6, 28-9, 50, 58 194 244 55, 217-20, 221 39, 218, 220 194, 199 154, 155 205 114 152, 153, 189 241

5, 6, 9, 15-18, 38-9, 42, 44-5, 54, 59, 69-71, 76, 80, 82-95, 105-6, 116-17, 119-34, 13943, 145-6, 148, 151-6, 158, 162-3, 170, 176-7, 179,

middle

outer

bailiff bailiwick bakehouse balista Bampton (oxon) Banbury, George Banbury, George Banister, John Paul and son Joseph Bannaster, John, and son Paul Bannockburn, battle of barbican

barge, bargemen Barking, nr London (Essex) Barnstaple Castle (Devon) Bartholomew de Burgherssh Basing House (Hants) Bayeux Tapestry Baynards Castle (London) Beadlow, Beaulieu (Beds), nunnery beams, beam-slots

181, 236-7, 250-2, 254, 257 5, 6, 9, 12-15, 18, 38, 39, 40, 42, 45, 54, 59, 64, 67, 69, 71, 77-9, 82-3, 106, 116, 13845, 148-9, 1515, 157-9, 163, 166, 176-80, 185, 236-7, 244, 252, 257 5, 38, 40, 64, 69, 70, 73, 778, 105, 116, 138-40, 142-5, 147, 151-3, 157, 159, 163, 177-9, 252 7, 113, 114, 227, 239 59, 61, 98 85, 89, 91, 211, 226 96, 97 51 238, 242-3 229 192 192 235 61 5, 9, 18, 39, 69, 70-3, 78, 79, 81, 87, 106, 112, 132, 139, 148, 150, 151, 159, 163, 177, 179-81, 205, 208, 237, 250, 254, 2568 18, 72, 76, 148, 234 47 21, 25 113 242 40 36, 100 217

Beansheaves Bear Lane

227 147-8, 155, 177 Bec Abbey (France) 48, 217, 219 Beckley (oxon) 59 Beeston Castle (Chesh) 257 bell(s) 1, 47, 81, 84, 87, 95, 99, 134, 153, 189, 245 Bell, The, inn 234, 236 Bellowe, John 140, 189 Belson, William 222 Belvoir Priory (Leics) 215, 217 Benets Tower 82, 91, 139 Benshef, Thomas 227 Bensington (Benson), oxon 49, 53, 59, 76, 84, 94, 98, 133, 170, 172, 178, 180, 192, 198 Beorhtric 29 Berkeley Castle (Gloucs) 13, 44, 45, 47, 62, 65, 86, 87, 94 Berkhamsted (Herts) 47, 71, 94, 100, 101 castle 40, 57, 59, 60, 72, 106, 114 Berkshire, sheriff of: see Drogo de Barentin, Henry of oxford, John fitzHugh of Wikenholt, Thomas Basset Bigg, Walter, brewer 236, 244 Billingsgate: see also London 100 Binham Priory (Norfolk) 215, 216 Bisham (Berks) 67 Bishop, Thomas, farrier 168, 192 Black Death 229 Black Ditch 72, 146, 148, 192-3 Black Prince: see Edward Blackstone, Henry 192 James 45-6, 104, 119, 136, 142, 143, 152, 155-7, 161-2, 170, 180, 186 William, Sir 119, 136, 142 William Seymour 119, 136, 142 Blagge, Thomas (Col.) 236-7, 239-43, 246, 248-9 Blake, John 189 Blewbury (oxon) 56, 198 Blindhouse: see prisons

98, 120-1, 202, 207

272

Boarstall (oxon) boathouse Bodecroft: see Bullcroft Bodiam Castle (Sussex) Bonhommes, at Ashridge Priory Boroughbridge (Yorks), battle of Bosworth (Leics), battle of Boulogne, or ongar, honour of (Essex) Brackston, William, mayor of Reading Bradenstoke Priory (Wilts), barn Bramber Castle (Sussex) brattice Bray (Berks) Brayes (Beds) brewer, brewing trade brewhouse (castle) brewhouse (priory) Brien fitzCount

Brent, Nathaniel, warden of Merton College oxford bridge, bridge-head, bridge-house, bridge-wardens, bridge-work: see also drawbridge

Bridge House Bridgnorth Castle (Shrops) Brightwell (oxon) Bristol Castle (Gloucs) Brittany (France) Britwell Salome (oxon) Browne, Robert Buckingham Castle (Bucks) Buckland (Berks), church of

71, 242 144, 171, 172

Bucklebury (Berks), church Bullcroft

1 83 62 65-6 58

bulwark burh: see also oxford

240 Burghal hidage burgheat Burgildesburia: see Bucklebury butler

225 41 79 51, 98 217 29, 69, 113, 234, 236 69, 87 226 4-5, 29-30, 42, 48, 51-6, 74-5, 89, 102, 106, 108, 155, 177, 217-19, 221 246

buttery buttress (also boteraciu/-iorum) Bydongate (middle gatehouse)

Byrne, William

3, 5, 19, 27, 36, 39, 41-2, 50, 61, 69, 703, 76, 79-82, 85, 90, 94, 98, 103, 105-6, 108, 113-16, 120, 138-40, 143-5, 148, 150-1, 154-5, 157, 160, 1615, 167-9, 1702, 174-9, 1845, 188-9, 1912, 222, 231, 234, 237, 239, 243, 33, 168 22, 27 37-9, 148, 172, 225, 239 21, 24, 39, 44, 55, 240, 242 55, 74 218 245-6 21, 27 89, 218, 220

Cadbury Castle (Somerset) Caen, France Caerphilly Castle (Wales) Caernavon Castle (Wales) Cainhoe (Beds), tithes Canterbury (Kent) Calvinists camera: See chamber Cambridge Castle (Cambs) Camden, William

213, 218, 220 3, 72, 146, 215, 220, 2225, 250 8, 233, 236, 237, 257 2, 4, 8, 11, 202, 25-6, 32, 36, 38, 49, 105, 116, 177, 197, 198-9, 200, 251 2, 31-2, 47 1, 39 30, 58, 65, 216, 217 58, 82, 83, 85, 89, 91,189 211-12, 244, 245 5, 71-3, 76, 81, 83, 93, 98, 110, 139, 1512, 164, 166, 170, 175, 1778 184 73 22, 99 80, 257 1, 60, 83 215-17, 220 21, 24, 36, 40, 62, 108-9, 197 232 21, 25, 40 70, 116, 124, 134, 137, 192 98 45 218, 220

Carisbrooke Castle (IoW) Carlisle Castle (Cumb) Carswell (in Buckland, Berks), Chersuwella cartingsilver 77, 101 Catherine of Braganza, queen, 151, 191-2 wife of Charles II castle: see Arundel, Ascalon, Ascot d’oilly, Barnstaple, Baynards, Beeston, Caernavon, Cambridge, Carisbrooke, Carlisle, Chester, Clare, Colchester, Cockermouth, Conway, Corfe, Devizes, Domfront, Dover, Dorchester, Durham, Exeter, Farnham, Falaise, Fotheringhay, Goodrich, Groby, Guildford, Hereford, Hertford, Huntingdon,

273

Kenilworth, Launceston, Leeds, Leicester, Lewes, Ludlow, Lydford, Mileham, Montgomery, Newark, Norwich, Nottingham, odiham, oxford, Pontefract, Rochester, Sandal, Shrewsbury, Somerton, Stafford, Therfield, Tickhill, Tintagel, Wareham, Warwick, Windsor, York castle-guard 6, 49, 56, 84, 198 Castle Farm 14, 159, 162, 181, 252 Castle Farm House 159-60, 163, 166, 178 Castle Gardens 143, 145, 14850, 152-5, 157, 176, 178, 180, 250-1 Castle Lane 54, 144-5, 148, 150, 154-7, 159, 161-2, 165-6, 169, 171, 175-80, 188 Castle Lane House 155, 159-61, 163, 166-7 Castle Priory 142, 162 Castle Street 2, 14, 38, 71, 139, 143-7, 149-50, 152, 154-8, 163, 176, 178-79, 191, 199, 227, 229-30, 234, 236, 238 ceramics: see pottery Cerney (Gloucs) 217 Chalford, in Aston Rowant (oxon) 217-19, 221 Chalgrove (oxon), Chelgraua, 53-4, 219-20, church 240 Chalkford: see Chalford Chalmore 192, 234 chamber: see Great, Green, Knights’ and Esquires’, Noricerie, chamber over the water, next the 76, 84-6, 93 Thames Chamberlain de Tancarville 82 chapel: see also Great Chapel, Great Tower, Houses of the chapel clergy chapel of the countess 85, 93 Chapel Eyot 148, 163, 168, 175, 178, 185 Charles I, king (1603-49) 108, 151, 192, 232-6, 240, 242 Charles II, king (1660-85) 140, 152, 190, 244-5, 247 Charles Iv, king of France 62

Charlton (Berks), church Chazey family (Mapledurham) Cheapside (London) Checkendon (oxon) Chester (Chesh.)

Chetham, James, of Mellor (Derbs) Chichester (Sussex) Chilterns

218, 220 218 199 98 18, 21, 39, 4950, 52, 55, 61, 64, 76, 85, 105, 115, 195, 197 192

21, 24, 59 3, 31, 99, 116, 238 Chinnor (oxon) 240 church 218-20, 221 Choles 227 Cholsey (oxon) 39, 52 , 73, 192, 218, 227, 238 Chrétien de Troyes 74 churches and parishes: see All Hallows/Saints, Holy Trinity, St Leonard, St John the Baptist, St Johnsuper-Aquam, St Lucian, St Mary-le-More, St Mary-the-Less, St Martin, St Peter-in-the-East, St Peter-in-the-West, St Rumbold Clapcot (see also Rush Court) 3, 6, 18, 26, 33, 38, 50, 52, 55, 59, 70-4, 76-7, 82, 87, 103, 115, 150, 170, 175, 177, 190, 215, 221, 240-1, 246-7 quarry in 99-100 Clapcot Gate: see North Gate Clare Castle (Suff) Nethergate, Redgate and 72 Derneygate Clarendon, Assize of 28-9, 33, 50, Cleeve, William, of Inglesham 140 Clement vII, pope 222 Clement the clerk, reeve 102 Clerks’ Lodging: see Houses of 142-3, 152-3, the Chapel Clergy 155, 158, 163, 176, 179, 186 Cleeve, William 140 Clifton (Notts) 195 Cloere Brien 18, 155 Clophill (Beds), Clopehulla, 217, 220 tithes Cnut, king (1016-35) 203 Cockermouth Castle (Cumbs) 110 Colchester Castle (Essex) 4, 21, 24-6, 401, 44, 47, 124

274

Roman Temple of Claudius College of St Nicholas: see Great Chapel and Houses of the Chapel Clergy Colwerhouse Close constable: see also Edmund Bacun, Miles de Beauchamp, John de Clinton, Miles Crispin, Robert d’oilly, Brien fitzCount, William Marshall,

Constable’s Chamber Constance, duchess of Brittany Conway Castle (Wales) Cookham (Berks) Cooke, William, malster Corfe Castle (Dorset) Cornwall, earldom and duchy earls of, see Edmund, Piers Gaveston, John of Eltham, Reginald, Richard Cotes (Beds), tithes Cottrell, Ambrose Cowley (oxon) Crawford, Thomas, Maj. Gen. Crendon (oxon) Cromwell, oliver Cromwell, Thomas Cross Keys, inn Crouch, Stonor Crowmarsh (oxon): see also Newnham Murren

Crowmarsh Hill Hospital of St Mary Magdalen Cursell, Henry: see also Kersall

Dalling (Middx) David ab Grefeth Day, John Dean’s Lodging: see Houses of the Chapel Clergy

26

Denison, John Derby (Derbs) Derbyshire Dernegate

225 4, 6-7, 26, 2933, 36, 38, 42, 47-52, 54-8, 615, 67-9, 72, 76, 78-9, 82, 84, 867, 92, 95-8, 1023, 106, 109-15, 119, 150, 179, 194-5, 198, 217, 219, 239 86-7, 92 74 111 31 190, 235-6, 2445, 247 45, 61, 98, 110, 257

Despencer, Hugh the Elder, earl of Winchester Hugh the Younger Devizes Castle (Wilts) Devon Domfront, Normandy, great tower donjon: see Great Tower Donnington (Berks) Dorchester (oxon) abbey Dorchester Castle (Dorset) Dover Castle (Kent)

Constable’s Gate keep drawbridge 217, 220 235 218 241 241 34, 108, 138, 140, 151, 242-3, 257 115, 117, 222 231 119, 192 15, 39, 41-2, 50, 77, 143, 148, 150, 163, 172, 192, 215, 234, 236, 242-4, 254 46, 239 49, 53

Draycot Foliat (Wilts) Driver, Edward Drogo de Barentin, sheriff Dublin, vikings of Dudley Castle (Staffs) Dunch, Edmund Durham Castle (Co. Durham)

Eadred, king (948-54), coin of Ealdred, abbot of Abingdon Ealdred, archbishop of York Ealdgyth, daughter of Wigod Earley (Berks) Easole St (in Nonington, Kent) East Garston (Berks), Hesegarestune East Gate: see also Mary Grace

244

49 100 235

275

132, 135 21, 194, 197-8 100, 194-5 62, 72, 100, 139, 171 62 62-4 56 58 44 218 100 101 21, 26 8, 12, 21, 25, 43-4, 49, 59, 60, 61, 65, 66, 79, 92, 97, 110, 121, 124 79, 86 44-5, 90, 124 5, 42, 58-9, 712, 79-82, 90-2, 132, 139, 151-2, 159, 164, 166, 168, 175, 178-9, 188, 191, 237, 239, 243, 257 218 161, 164-5, 186, 193 62, 110 197 257 140 13, 21, 41, 53,

203 108 47 30, 31 94, 99, 218-220 216 218-20 5, 39, 70-1, 77, 79, 99, 103, 105-6, 139, 150, 163, 167, 174,

East Ilsley (Berks) Eddington (Berks), church Edmund, earl of Cornwall

Edmund, earl of Kent Edmund, duke of York Edmund Bacon, constable Edmund de Kendale Edmund Rede Edward, abbot of Reading Edward the Confessor, king (1042-66) Edward the Elder, king (899-924) Edward I, king (1272-1307) Edward II, king (1307-27) Edward III, king (1327-77)

Edward Iv, king (1461-83) Edward v, prince of Wales and king (1483) Edward vI, king (1547-53) Edward, the Black Prince Edward the carpenter Edward the moneyer Edward Norris Eleanor of Aquitaine, queen, wife of Henry II Eleanor de Montfort, daughter of king John Elephant, The, Wallingford inn Elizabeth I, queen (1558-1603)

Elizabeth, sister of Edward Iv Emma de Chausi Emma Stokkere, Richard II’s laundress entrance

176, 178, 185, 234, 238 219 218 4-5, 29, 54, 601, 67, 69, 71, 73, 78, 83-5, 96, 106-7, 110, 134, 151, 172 111-12 65 63, 78, 101, 110 82 221 213 4, 20, 36, 105

see also forebuilding, gateways, posterns Estonia Evelyn, Maj. Arthur Evesham (Worcs), abbey battle Ewelme (oxon) Exeter Castle (Devon) Eythrope (oxon), Aetrope Fabian, king’s forester Falaise Castle, Normandy Fairfax, Thomas Faringdon (oxon, formerly Berks) Faritius, abbot of Abingdon Farnham Castle (Surrey)

31 1, 12, 58-60, 71, 83, 175 4, 60-4, 69, 78, 82, 83, 95-8, 102, 107, 112, 46, 62, 64, 69, 73, 75, 102-3, 105, 107, 11014 4, 65-6, 98, 115 72, 76, 83, 113, 115 54, 117,120 64, 67, 76, 94, 113-4, 134 102 216 67 57, 74

Ferlei, tithe of ferry Fish Street (Fische-stret); now St Mary’s Street fisheries Flanders Flittwick (Beds) Floodgates (sluice system at West Gate) Fludger, Charles Foliot: see Robert, Samson, Walter forebuilding: see Great Tower Fortie, Robert, rector of St Mary’s Fotheringhay Castle (Northants) Francis Lovell, Lord Lovel Freeman, John Freeman, Thomas Freeman, Ann Frobbins Fullerton, James, Sir furnace: see also kiln, oven

60 234 83, 117, 120, 150, 153, 168, 189-90, 196, 229, 232 65 218 105 3, 5, 20, 26, 39, 42, 45, 54, 70-2, 76, 78-80, 86, 88-93, 108, 139, 145, 147, 150, 151, 155-9, 1612, 164-5, 167,

gainery (farm) of Wallingford Castle

276

173, 176, 177, 179, 234, 237, 254, 257 100 140, 153, 190, 243-4, 246 56 60 39, 65, 67, 135, 140, 142, 153, 190-2, 246 3, 21, 22, 24, 26, 39-41, 177 218 90, 99 22 232, 242-3, 248 51, 238, 242-3 56 40-1 n.1, 45, 104 219-20 72, 76-7, 115 227, 229-31, 234 77, 101, 113-5, 141-2, 172, 1745, 219-20, 225 22 218 72-3, 90, 114, 170, 172 193

244 66 65-7 140 140 140 227 190 69, 100

77, 101, 115

Gainsborough (Lincs) gaol: see prisons Gaol or Gayle Barn: see Prison House garden

garderobe: see also Wardrobe Garsington (oxon), gersidona, church of Garston, East (Berks) gatehouse inner (Drawbridge) middle (Bydongate) outer, or main (?Constable’s) gatehouse (not Wallingford) gateways: see also entrances, posterns

Gaveston: see Piers de Gaveston Gawain, Sir his sword Gawynes herber: see Queen’s Arbour Geoffrey, son of Henry II Geoffrey, last prior of Holy Trinity Geoffrey Baynton Geoffrey the Chamberlain Geoffrey Chaucer Geoffrey atte Cuchene Geoffrey Kidwelly, receiver Geoffrey Lecur Geoffrey de Mandeville Geoffrey of Monmouth geophysical survey

73 176 George, The, inn Gerberoy, battle Gervaise de Cornhill Gibbons, Capt., of Henley Gilbert Angevin Gilbert Boghere and brother Thomas Gilbert de Mandeville, aka le Messagier Gilbert de Mordon, sheriff of London Glastonbury (Soms) Gloucester Castle (Gloucs)

6, 12, 18, 54, 70, 72, 76-7, 846, 91, 116, 140, 142-3, 147-8, 151-3, 157, 161, 165-7, 172, 1768, 179, 189-90, 192-3, 222-3, 225, 234 59, 76, 84-5 218, 220, 221 218-20 47, 60, 70-2, 789, 81, 107, 120, 5, 71, 79, 81-2, 86-7, 90, 130, 5, 70-1, 79, 81, 83, 132, 158, 191, 5, 71, 78-9, 86, 106, 1, 26, 40-2, 45, 71, 78-9, 205, 208 5, 70, 72, 13940, 143, 145, 148, 150, 152, 155-9, 161, 164, 166-8, 171-3, 175-8, 185, 1878, 191, 234, 237, 239, 257 60-4, 75, 96-7 6, 73-5, 84, 98 75, 98 69

Gloucester, countess of Godwin Child Godwin Polsore, son of Goodrich Castle (Herefs) goldsmith Goldsmiths Lane Goltho (Lincs) Goring (oxon), prioress of Gough, Richard Great Bridge: see bridge Great Chamber (also known as king’s, queen’s, Lord’s) Great Chapel of St Nicholas

74 222

Great Haseley (oxon) Great Kitchen and related services Great Prison

189 7, 55, 216-18 65 211 115 55 30, 218 74 9, 14-19, 35, 44, 70, 82, 84, 91, 93, 107, 116,

Great Stable Great Tower (also Donjon or Keep) chapel forebuilding

277

130, 140, 150, 161, 172-3, 175, 179-80, 222-5, 250-1, 253-4 234-5, 238 49 109 243 49 61, 110 78, 102, 113 110 29, 73, 117 3, 21, 24-5, 27, 49, 57, 76, 240 78 216 216 159 7, 103 134, 227, 22931, 234 1 101 148, 150, 154-5, 181, 192 58, 79, 82, 84, 87-89, 91, 94, 100, 179 87 3-5, 27, 32, 38, 52-5, 62, 64, 69, 71, 76, 83-4, 8991, 102, 113-14, 127, 134, 138, 146, 151, 179, 225 54 5, 69, 86, 89-91, 100, 179 5, 69, 82, 91, 151 5, 71, 78, 82, 83, 102, 115, 151, 154, 179 91, 139 84, 93-4, 100, 121, 124, 179 2, 5, 44-5, 91-3, 106, 120-1, 124, 130, 132, 179

great chamber kitchen prison privy chamber stair turret Green Chamber Greenwood, Thomas Gregory, John, and son John Groby Castle (Leics) Guildford Castle (Surrey) Guinevere, queen guttering Guy de Beauchamp, earl of Warwick Guy de Montfort Guy of Warwick

Hagbourne (Berks) hall: see Chamber of the Knights’ and Esquires’, Great, Hall of Pleas, Hubert’s, Red Hall of Pleas (aula placitorum) Hampden, John Harold Godwinson, king (1066) Haseley, Great (oxon) Hastings Castle (Sussex) Havell, William Havering Palace (Essex) Hearne, Thomas Hedges, Francis Reade, Maj. John Allnatt John Francis, Sir John Kirby

Sophia, Julia and Ellen Sophie Cassandra Hemsdell, John Hendred, West (Berks), church Henley (oxon) Henrietta Maria, queen, wife of Charles I Henry I, king (1100-35)

94, 121-4 121-4 121-4 121-4 69, 93, 124 69, 84-6 192 192 41 21, 58 74-5 46, 72, 85-6, 88, 92, 100 75

Henry II, king (1154-89), duke of Normandy

Henry III, king (1216-72)

Henry Iv, king (1399-1413) Henry v, king (1413-22) Henry vI, king (1422-61)

60 6, 74-5, 84, 98

Henry vII, king (1485-1509) Henry vIII, king (1509-47)

238

Henry, the Young king (1170-83) 82 240 105 54 3, 21, 28, 36, 40, 47, 116 167-8, 185 58 119, 148, 168-9, 184-5 144 143, 155 144-5, 190 2, 11, 13, 45-7, 67, 72, 77, 88, 91-2, 104, 119, 136-7, 142-4, 147, 153, 155-7, 162, 170-2, 176, 188 144 144 140 55, 215-18, 220, 221 59, 92, 98, 100, 103, 236, 240-3 140, 151, 192, 232-3 4, 5, 7, 29, 41-2,

Henry de Albini Henry of Aldrington, supervisor of works at castle Henry of Almain, son of Richard of Cornwall Henry balistarius (crossbowman) Henry de Bellomonte Henry the Clerk, of St Mary’s Henry of Cornwall, natural son of Earl Richard Henry de Ferrers Henry fitzIvo, mason Henry Lescrop Henry the Larderer Henry of Melburn Henry Merston Henry Norris, constable Henry d’oilly Henry of oxford, sheriff Henry de Scaccario, under-sheriff of Berkshire Henry the Slater Henry Yvele, king’s master mason, brother of Robert Henton, in Chinnor (oxon), and Nethercote in herber: see gardens, Queen’s Arbour Herbert son of Geoffrey the

278

44, 48, 51-2, 556, 73, 81, 105-6, 213, 215-16, 218-19 4, 5, 7, 28-30, 40, 44-5, 48-9, 51-6, 58-9, 74, 79, 82, 103, 105, 107-8, 167, 177, 197-8, 21819, 234 4, 5, 12, 49, 50, 57-60, 85, 87, 90, 101-2, 107, 109, 134, 196-7 64-5, 83, 95, 114 65 65-6, 95, 98, 107-8, 114-15, 221 65-6, 115, 229 4, 5, 64, 67, 69, 83, 107, 120, 215, 222 4-5, 56, 58, 74, 109 217 94, 113-14 60 50 111 102 58 215 50 63, 110 55 69, 112 114 67 51, 75 50, 218 57, 82 211 100 218

216

chamberlain Hereford Castle (Herefs) Hertford Castle (Herts) High Street

Hillingdon (Middx) Hobarte, Henry, Sir Holy Trinity, church, parish, and priory (also Wallingford Priory) Hook Norton (oxon), barony Hopton, Lord Houses of the chapel clergy Clerks’ Lodging

cloister of college clergy buildings Dean’s Lodging Priests’ Lodging

Howbery Park, Crowmarsh (oxon) Hubert (Walter), archbishop of Canterbury Hubert de Burgh, justiciar Hubert’s Hall Hucks, Robert Hucks, William Hucks estate map or plan Huddesbutts (in Shillingford) (oxon) Hugh, priest of Crowmarsh Hugh of Buckland Hugh Daudeley the Elder Hugh Despencer the Elder, earl of Winchester Hugh Despencer the Younger Hugh the Larderer Hugh the Serjeant

Huntingdon (Hunts) castle priory Huntly (Scotland) Hurley Priory (Berks) Hurst, Thomas Hyll, William, dyer

21, 24-5, 47, 60, 62 21, 114 2, 5, 32, 139, 140, 142-4, 1478, 150, 155, 161-3, 166-7, 169, 172, 176-9, 181-2, 190, 199, 222-3, 225, 227, 229-31, 234, 238 56 190 7, 54-5, 72, 83, 198, 213-26, 231, 250 47 241-2

Iffley (oxon) Ince, John, and daughter Frances Ipsden (oxon), chapelry of North Stoke Isabella de Lusignan, queen, wife of king John Isabella of France, queen, wife of Edward II Isabella of France, queen, wife of Richard II Isabella, daughter of Edward III Isabella of Gloucester Ivan Goring, labourer

83, 142, 143, 152-3, 155, 158, 163, 176, 179, 186 5, 83, 106, 153, 189 8, 71, 140-2, 152-3, 162-3, 179, 186 8, 36, 54, 77, 101, 141-3, 1523, 155, 162-4, 176, 178, 179, 186 143

janitor: see porter Jennings, Edward Joan of Navarre, queen, wife of Henry Iv Joan, sister of Henry III Joan of Kent, Princess of Wales Jocelin, bishop of Salisbury Jocelyn Crowmarsh, labourer John, archbishop of Messina John, king (1199-1216)

109 57 84-5 142, 182, 192 141-2, 154, 1912 145, 151, 157, 182 225

John, eldest son of Edward I John, prior of St Frideswide’s John, prior of Wallingford, abbot of St Albans John de Alveton, constable John Anesty, constable John Arden, clerk of the king’s works John Bate, carpenter John Best, carter John de Bousser John Bray, chancery clerk, rector of All Saints John Carpenter

50 216 62, 110 62 62-4, 112 216 102

279

21, 39 53 257 218 120, 133 189

89 192 53 100 61-4, 69, 86, 98, 107, 110-12, 65, 95 105 83-4 100

140 114 58 4, 64, 76, 86, 102-3, 114-15 218-19 100 60 4, 30, 49-50, 55, 56-8, 60, 74, 102, 106-7, 10910, 196 60 114 213, 221 52, 113 33 114 88 102 63, 110 53 69, 112

John de Clinton of Maxstoke, constable John Colles John Coterell, alderman, recorder, MP John de Cotesmore, bailiff John de Coton, sheriff of London John Culham, stonecutter John Derby, MP John Drayton John of Eltham, earl of Cornwall John de Gonwardly John Hubert, reeve John son of Hugh, of Wigenholt or Wikenholt, clerk and sheriff John Humson, bailiff John Hunt, labourer John Juell, constable John Mariot, mayor John Mychell, priest John Perse, surgeon John Plowman of oxford John Plumber, of Berkhamsted John de la Pole, duke of Suffolk John de la Pole, earl of Lincoln John Pope, carter John atte Russhe John Smith of Wallingford John Stacey John de Stonor, justice, keeper of the castle John de Turberville, sheriff John Perse, surgeon John Tyler John Walisse John Warborough, quarryman John Warfield John Warfold John de Wendout, chaplain John ab Zenan jousting: see also tournament

61-2, 70, 95, 98, 102, 110 71 102, 114

Katherine de valois, queen, wife of Henry v Keep: see Great Tower Kenilworth Castle (Warks) Kersall, Henry: see also Cursell Kersall, Richard son of Henry kiln Kinecroft

65

King’s Chamber: see Great Chamber King’s Mead

113 110 84, 102 102, 103 100 64, 106 211 85 57-8, 109-10

kitchen: see also Great Kitchen

Knapp, Henry

labourers

189 102 115 62, 98 115 115 143 94, 100 65 65-6 102 103 97 69, 112 63, 110-11

Lambert Simnel Lambourn (Berks) landgabel Lanfranc, archbishop of Canterbury Lanfred, architect Lancaster, honour of larder Launceston (Cornwall) castle priory Laurence the mason Lawton, Ann Ralph William, of Lawton (Ches) Leaver or Lever, Anthony John Robert Thomas Leeds Castle (Kent) priory Letsham, Daniel Leicester (Leics) abbey castle St Mary in Castro Leland, John Leofwin Lever, Anthony Edward John son of Robert Lewes Castle (Sussex)

51 115 79, 86, 94, 100 100 100 114 100 113 100 70, 97

60 192, 235 192 100, 197, 203 3, 32, 199

280

6, 15, 18, 38, 50, 72-3, 77, 139, 142, 150, 155, 163, 170, 172, 177-8, 2501 9, 12-14, 17, 50, 57, 59-60, 65, 83, 85, 87-9, 91, 109, 121, 153-4, 217, 226 153, 245

7, 67, 69, 98, 100, 102, 114, 189-90, 198, 204, 211 65 219 199, 227 215 47 58 83, 85, 153, 189 40, 205-6 53 211 140 192 192 158, 191-2 192 182 192 111 53 244 49, 194 53 21, 27, 39, 49 53 5, 54, 70-1, 83 218 192 192 192 21, 25, 39, 60

Lincoln (Lincs)

Loader, William London: see also Billingsgate, Queenshithe Holy Sepulchre without Newgate, church of Lundenwic St Bartholomew’s Priory Turnham Green London, Dr, last dean of St Nicholas Louis, prince, of France Lovell, Grigory Luke, Sir Samuel Ludlow Castle (Shrops) castle borough Lydford Castle (Devon) Lydgall. William

3, 57, 194, 1968 21, 26, 32, 3840, 66, 194 235, 244-5 74

William Merwick, Robert Michael of Wolleford Mileham Castle (Norfolk) Miles de Beauchamp, constable Miles Crispin, constable of Wallingford

219-20 197 219 233 83, 134

Miles fitzTurold Mill Brook mills and milling

57 190 234, 236, 239 76 22 21, 24, 41 241

East or King’s Mill North Mill (at North Eyte) South Gate mills moats

Maidstone (Kent) Manasses Arsic, lord of Cogges Manasses de Sanderville Mapledurham (oxon) Margaret of Anjou, queen, wife of Henry vI Margaret de Clare, wife of Earl Edmund of Cornwall Margaret de Clare, niece of Edward II, wife of Piers Gaveston Mariote Litle Marshall, Alfred C. Marshall. John Mitchell Marston, North (Bucks) Mary Grace, the, East Gate chapel Matthew Brand, custodian of the honour of Wallingford Matthew Paris, chronicler Matilda, Empress

Matilda of Wallingford Maurice, Prince Maurice de Berkeley Maurice de London of Kidwelly Maxey, Thomas

23 219 217 218, 220 65, 221

Molines (Moleyns, Mullins), Michael de Montgomery (Wales) Moreton, North and South (oxon)

60, 69 61

Moulsford (oxon) Motehalle : see Hall of Pleas motte

98 119 119, 136 218, 220 167, 174 58 59, 106 4, 29, 49, 52-3, 55-6, 58, 74, 108, 177, 217, 219, 221, 234 4, 29-30, 48, 55, 217, 221 240 62-3, 110 219 191

Muggetour Munt’s Mill Murray, Thomas

281

165, 191 164, 191 211 40 64, 96-7, 112 4, 29-30, 32-3, 38, 48, 53-6, 198, 217-19 218 225 6, 18, 52, 55, 58, 69, 76-7, 834, 92, 98-9, 101, 109, 113, 115, 139, 150, 167, 169-70, 172-5, 179, 234 58, 76-7, 99, 167, 172-3 76, 101, 115, 170, 175 6, 70, 76-7, 120, 174-5, 239 2, 5-6, 18, 46, 59-60, 69-70, 72-3, 90-116, 142, 164-5, 16972, 175, 177-9, 236 153, 240, 246-7 257 39, 72, 170, 192, 218, 220, 225, 128 216, 218, 220 1-2, 5, 9, 11, 14, 20, 25-7, 38-42, 44-7, 52, 59, 70, 72, 86, 90-3, 95, 104-8, 116-17, 120-1, 124, 1278, 130-33, 1367, 140, 143-5, 151, 155-7, 1624, 170, 178-9, 181, 200-1, 2048, 252, 257 91, 95, 108, 139 148, 163, 165-7, 171-2, 175-8, 185 190

Nantes (Brittany) Nedham, William Nettlebed (oxon) Newark (Notts) Newbury (Berks) Newcastle (Tyne and Wear) Newcastle-upon-Lyme (Staffs) Newnham Murren (oxon) Porter’s land in Nicholas, prior of Wallingford Nicholas Harald Nicholas Somerton, escheator Nicholas of Turneston Nigel de Albini of Cainhoe his sons Henry, William and Nigel Noenshynches, midday meal Noricerie, La Norise, nutricium Northampton (Northants) Castle Derngate North Eyte, Northeyte, al. Northacche: see also mills North Gate

Norwich (Norf) castle cathedral Nottingham (Notts) Barker Gate Castle and honour

St Mary’s church Woolpack Lane

odiham Castle (Hants) odo of Bayeux ogbourne St Andrew and ogbourne St George (Wilts), churches old Moor Lane (now St John’s Road)

74 192 86, 94, 99 21, 25, 66 234, 239, 242, 247 21, 45, 243 27 49-50, 53-4, 150 150 213-15 99 114 69, 112 7, 215-17, 221 217

ongar (Boulogne), honor of orchard

orderic vitalis, chronicler oven oxford (oxon)

Beaumont St, palace burh Carmelites Castle

105 82, 88-9 23, 25, 56, 194 21, 38-9, 56, 58, 66, 72 72 76-7, 101, 114, 170-2, 175 5, 9, 11, 18, 389, 55, 70-1, 81, 106, 134, 139, 146, 149-50, 159, 225, 254, 257

Castle St Christ Church, formerly Cardinal College Grandpont Merton College New Road Queen St Rewley Abbey St Frideswide’s Priory St George’s, college of St George’s Tower

21, 25-6, 38, 445, 89, 92, 205 95 4, 30, 32, 56, 194-8 197 4, 8, 21, 27, 302, 39-40, 46, 558, 64, 66, 72, 107, 195-6, 1989 195 197

St John’s Hospital St Michael at the Northgate St Michael at the Southgate St Peter’s College Shoe Lane Tidmarsh Lane oxford Canal Company Wharf, later Town Wharf

96, 111 218, 219 48

palisade

Pamber Forest (Hants) Parry, Thomas, Esq

230-1

282

58 70, 77, 101, 142, 151-4, 179, 189-90, 192-3, 215, 234, 245 21, 47 17, 87, 93, 203, 238 2, 7-8, 60, 74, 95, 98, 100, 114, 140, 159, 189, 192, 197-8, 200, 204, 211, 218, 233-4, 23642, 247, 253 107 200-4, 208 211 4, 6, 21, 24-5, 36, 38, 42, 45, 47, 49-53, 56, 58-9, 66, 70, 74, 92, 100, 105, 108-9, 200-212 203 7, 134, 140, 151, 172, 193, 222, 236 47, 51 242 206 203 67 54, 114 53, 74 1, 23, 39-40, 201-3, 205, 2078 218 23, 201 51 200 203 207 172

2, 5, 20, 39-40, 57, 70, 79, 105, 150, 177, 188, 237 90, 98-100 189

Pasterie, pastry kitchen Pauling, Richard, rector of St Mary’s Payne, Nicholas Thomas, of Wallingford, gent. Pembroke, earl of: see William Marshall pentice perambulation of municipal bounds 1707 and 1806 Peter, parson of St Peter’s Peter Boterel Peter Comester Peter de valognes Peter fitzHerbert Peverel of Nottingham, honour of: see Nottingham Philip II, king of France (11801223) Philip Iv, king of France (12851314) Philip, count of Flanders Philip de Calna Philip de Wygorn Philippa of Hainault, queen, wife of Edward III Piers de Gaveston, earl of Cornwall Pinkney, Thomas, rector of St Mary and St Leonard Pleshey (Essex), castle and borough Plympton (Devon), castle and borough Plowman, John Pollington, Ralph Pont, Richard Pontefract Castle (Yorks) St Clement’s, church portcullis porter Porter’s fee

Porter’s Ditch postern: see also Dernegate

pottery (and ceramics)

89, 91 235

252-4, 257 Priests’ Lodging: see Houses of the Chapel Clergy prison: see also Blindhouse, Great Prison

120, 133, 174, 190 190

5, 85, 91 146-8, 150, 166, 169, 225 189 218 83 216 109

pit prison (puteus) Prison House: also called Gayle House or Gayle Barn Purdon, John

57

Queen Wood (Queneholt) in Watlington (oxon) Queen’s Arbour, Quene(s)herber

61 56 217 109 73, 76, 85 60-4, 75, 96-7

Queen’s Tower

244

Queenshithe (London) quarrying

22

see also Clapcot

22 143 174, 190 140 13, 53, 62 53 47, 71, 78-9, 98, 207 50, 70, 79, 86 50, 73, 139, 141, 148, 150, 167, 176-8, 182, 190 170-1, 190 23, 46-7, 55, 62, 70, 72, 76, 84, 93, 162, 171, 207 11, 13-15, 1719, 197, 200, 203, 208, 225,

Rainald Angevin Ralph, Earl Ralph Arbalaster Ralph Chanu Ralph Culebule Ralph de Porta Ralph Restold or Restwald, under-constable Ralph Saillard Reading (Berks)

abbey Red Hall

283

5-6, 18, 34, 36, 41, 46, 50-1, 578, 62-3, 65, 67, 69, 78, 82-3, 901, 98-9, 102, 105, 107, 10914, 121, 124, 151, 154-5, 179, 200, 203, 207-8, 217. 237-41, 244 83 141, 143, 150-1, 154-5, 176 120, 133

99 6, 9, 15, 17-18, 38, 73, 76-7, 82, 101, 114, 138, 143-4, 148, 151, 163, 170-5, 1778, 183, 188, 250-2, 254 18, 76-7, 87, 91, 139, 252 100 99-100, 113, 203, 206, 208

49 202 50 218 102 50 51, 61, 76, 96, 111 50 2, 7, 11, 51, 75, 99, 117, 118, 198, 234, 23740, 242, 247, 55-6, 73, 89, 117, 213, 218 5-6, 76, 85-7, 89-92, 94, 99, 179

Reginald, earl of Cornwall Reginald Canute Renda, Thomas

Thomas’s wife Sarah Richard I, king (1189-99) Richard II, king (1377-99)

Richard III, king (1483-5) Richard, earl of Cornwall, king of the Romans

Richard d’Albini, abbot of St Albans Richard Bargeman, of Shillingford Richard Blawe, vintner Richard Branche, bailiff of the honour Richard de Briggewater Richard of Chilham, son of king John, keeper of the castle Richard Couper Richard Damory Richard Gratard, MP Richard Grey, Sir, constable Richard Horlock, grocer and MP Richard Kelyng Richard de Louthes Richard Ludlow Richard Mason Richard Puignant Richard Smith, of Benson Richard Stokebrugge Richard de vernon, son of Hugh Richard of Wallingford, master carpenter Richmond (Yorks) ringwork Robert, abbot of St Albans Robert, bishop of Salisbury Robert, earl of Gloucester Robert, earl of Leicester Robert d’Albini

58 218 46, 140-2, 145, 147-8, 151, 1548, 162, 164-6, 176, 182, 191 141 4, 30, 56-7, 74, 106-7, 196 75, 98 64-5, 80, 93-5, 98, 100, 103, 105, 107-9, 114, 172, 229 65, 67 4-6, 11-13, 41, 45, 49, 55, 5860, 67, 75, 78, 83-4, 90, 100, 105-7, 134, 152, 177 215-17, 219

Robert Arsic Robert Baldock, lord chancellor Robert de Basingis Robert Beauchamp Robert de Bellême Robert Bloet, bishop of Lincoln Robert Brompton, glazier Robert Cervesa, of Didcot Robert de Chesney, bishop of Salisbury Robert Corbet Robert Damory Robert fitzWilliam Robert Foliot, archdeacon of oxford Robert Holon Robert Malet of Quainton Robert Mason, of Henley, mason Robert d’oilly

72 Robert d’oilli II Robert de Porta Robert Smith of Wallingford Robert the Watchman Robert Yvele, master mason to Black Prince Rochester Castle (Kent)

102 114 211 57-8 100 63 69, 112 65 102-3 189 69 115 69, 100, 112 217 93-4 189 218 100, 113

Rochford, Lord Rodbourne (Wilts) Roger, bishop of Salisbury Roger Cole Roger de Drayton, dean of St Nicholas Roger d’Ivry Roger de Montgomery, earl of Shrewsbury Roger de Mortimer of Wigmore Roger de Mowbray Roger de Porta Roger Wauton roof, roofing

29, 45 5, 20, 25, 40, 42, 45, 105 213, 217-8 218 55, 108 53 217

Row Barge, beer-house Rupert, Prince Rusden, John

284

219 112 50 100 47 215 94 49 219 49 95 50 219 189 49 92 3-4, 29-30, 32, 47-9, 51, 53, 194, 198, 200, 202, 217-19 51, 53, 59 50 94, 100 50 100 21, 24, 39, 43-4, 61, 92-3, 98, 110 67 218, 220 89, 219 189 71, 84, 110 47, 59 30, 47, 55 46, 62, 64 217 50 62 57-9, 69, 71-2, 78-9, 83-4, 86, 89, 91-2, 94, 98100, 109, 113, 120-1, 128, 211, 218, 257 231 238, 240, 242 140, 147-8, 190,

Rush Court, a manor in Clapcot (Berks) Ryall, Edward

St Albans Abbey (Herts) St Leonard, church and parish St Leonard’s Lane St Lucian, church and parish St John the Baptist, free chapel and hospital St John-super-Aquam, church St John’s Farm St John’s Green St Mary-le-More

St Mary the Less, or de Stallis St Mary’s St St Martin, church and parish St Martin’s Street, formerly St Martyn’s Lane St Michael, church and parish St Nicholas: see Great Chapel St Nicholas, High St St Peter, church and parish

St Peter-in-the-West, gate chapel St Peter’s Street St Ruald or Rumbold, church and parish St valery, honour of Salisbury (Wilts), church of sallyport Sakenate Samson Foliot Samson de la Pomeray Sanchia of Provence, wife of Earl Richard of Cornwall Sandal Castle (Yorks) Sanderville family Sandford (oxon) Sarah de Bidun

244 38, 82, 100, 221, 240 145, 172

Savage, Thomas Scarborough (Yorks) castle priory Schomberg, duke of serjeanties

7, 55, 213-19, 221 37, 55, 234, 244-5 228-31 37, 55 19, 53, 120, 231, 234, 239, 219, 229 220, 227 53, 173, 239 7, 83, 194, 213, 215-16, 219-20, 221, 223, 234-5, 244-5 54, 114 234 7, 215, 219-20, 221, 227 229-30, 234

Shabbington (Bucks), church Shaw (Berks), church Shepherd, George Shillingford (oxon)

Sildenebrugge Shotover Forest, oxon Sewold Graiessune Shrewsbury (Shrops) castle Simon, abbot of St Albans Simon Gurmund Simon de Lagenhull Simon de Montfort, earl of Leicester Simon de Montfort, the younger Skirmer, Richard

54, 114, 229 144 33, 39, 54, 70, 139, 148, 171, 177-8, 188, 192, 197, 229, 234, 244-6 167 228-9 55, 229

Slade, Moses Slotisford Hundred sluice

Smythe, John, mayor Somerton Castle (Lincs) Sonning (Berks) South Gate: see also mills

59, 98, 135 54, 84, 114, 217, 219, 221, 245 46-7, 72, 85, 132 76 218 49 60, 83

Southampton (Hants) stables: see also Great Stable

257 218 234 59

Stafford (Staffs), burh castle stair, staircase, steres

285

235 61 53 53 138 5-6, 41-2, 4951, 55, 57-8, 63, 70, 82, 102, 105, 110, 216, 219 221 218 169, 185 39, 52, 84, 100, 172, 177, 225 38 72, 76-7, 115 172 51, 98, 100, 246 216 3, 18 21, 47, 66 217 102 113 60 60 2, 46, 72, 119, 136, 245, 167, 192 50, 52 11, 18, 46, 70-3. 76, 90, 106, 139, 162, 170-1, 175, 178 246 111 219-20 6, 19, 52, 54, 70-1, 76, 120, 150, 173, 199, 231, 239 3-4, 21, 23-6, 248 61, 69, 151, 153-4, 168, 179, 189, 191, 211, 245 4, 199 21, 23, 25, 27 2, 11, 45-6, 814, 87, 89, 91-3, 100, 120, 124,

stair turret: see Great Tower Stamford (Lincs) Stanford-in-the-vale (Berks) Stanhope, Michael Stanlake (Berks) Stephen, king (1135-54)

Stephen de Segrave, constable Stephen de Tong, chaplain Stephen de Wolleford Stephen’s Mill, Chinnor (oxon) Stigand, archbishop of Canterbury Stoke, North (oxon), church of storage, stores, storehouses

Studie, La Sulgrave (Northants) Swallowfield (Berks), Farley Hill in swannery Sweyn de Mortele Swindon (Wilts) Symonds, Richard

Tantallon (Scotland) Tempsford (Beds) Tewkesbury, battle Thame (oxon) abbey Thames Street Thameside Mansion Therfield Castle (Herts) Thomas, duke of Lancaster

130, 132, 134, 179, 189, 211

Thomas Basset, sheriff Thomas de Beauchamp, earl of Warwick Thomas Beaufort, marquess of Dorset and Somerset Thomas Benshef, mayor Thomas de Berkeley Thomas de Boghere and brother Gilbert Thomas Carpenter, of Warborough Thomas Chaucer Thomas de Cholsey, mayor Thomas Cromwell Thomas de Curtlyngton Thomas Dogett Thomas Haseley Thomas Hatfield, keeper of Nettlebed Thomas Hevyn Thomas Hunte Thomas Mason Thomas More Thomas Reginald of Wallingford Thomas de la Russhe Thomas the Plumber of Abingdon Thomas Polenton Thomas of Walsingham, chronicler Thomas Whythorn Thompson, Dorothy Tickhill Castle (Staffs) timber

3-4, 194, 196-7, 203 4, 21 74 140, 189 219-20 2, 4, 22, 38-9, 41-2, 51-2, 556, 58, 60, 106, 108, 139, 167, 177, 195, 218-9, 234, 62 64 211 218 36-7, 47, 108, 217 39, 53, 219-20, 225 6, 8, 51, 58, 64, 78, 87, 91-3, 957, 100, 111, 119, 121, 153, 158, 190, 226, 236-7, 240, 243, 248 5-6, 69, 83, 112, 151, 179 41 219 6, 18, 76, 173-5, 178 64, 112 218, 220 245-6

257 217, 220 65 240 54 142, 148, 161-2, 219, 227, 22930, 234 14, 145, 155, 159-60, 162-3, 165-6, 176, 178 206 62

Tintagel Castle (Cornwall) Toki, son of Wigod Tostig, Earl tournament: see also jousting towers: see also, Benets, Great, Muggetour, Queen’s trellis: see also garden, vineyard Trevor, Sir Thomas Trokelowe, John, chronicler

286

109 75 65 227 63 61, 110 100 65, 114 227 115, 117, 222 192 189 114 99 189 189 192 189 189 73 74 189 215 189 246 57 1-2, 5, 7, 20, 25, 38, 41-2, 44, 45, 47, 54, 58-9, 69, 71-2, 78-9, 817, 91-2, 94-5, 97-9, 101, 1056, 109, 113, 114, 119-21, 127, 133, 140, 145, 152, 162, 173-4 75 30-1, 49 30, 32, 195 61, 73-4 72, 76-7, 91 190 215

Tuder, owen Tudor, owen, stepfather of Henry vII Tynemouth Priory (Northumb)

165, 191 65

U-shaped feature: see also mills

15, 17-18, 174, 253, 257 49, 89

Upton (Berks)

216-17

vineyard

6, 70, 76-7, 101, 115

Wadard, man of odo of Bayeux Wærferth, bishop of Worcester wage, wages

219 199 50-2, 57, 61, 63, 68-9, 98, 10001, 110, 113, 189, 211 218, 220 72-3, 139, 1701, 173-5, 177 241 74 190 69, 112 109 33, 215 211 115 114

Wainhill, in Chinnor (oxon) Walditch Waller, William, Sir Walter, archdeacon of oxford Walter, Sir John Walter Crispe Walter Foliot Walter Giffard Walter de Fynestocke Walter Stonor, Sir Walter Yonge, Black Prince’s receiver Wardrobe

Wareham Castle (Dorset) Warwick (Warks) castle watch-tower watchmen water, watercourses: see also barge, ferry, mills, moats, sluices, weirs, wharf

Watlington (oxon), bailliwick of

honour of Wallingford park of Wallingford Castle at Watton (Yorks), barn Wells, Edward West Gate (see also Floodgates)

Westbury (Berks) Westminster (London) Wheatley (oxon), Whetelee, quarry White Horse Hill (oxon, formerly Berks) Whitchurch (oxon) Whitelocke, Bulstrode whitewashing Wigod of Wallingford

William I, king (1066-87)

William II, king (1087-1100) William III, king (1685-1702) William, eldest son of Henry II William, sheriff of oxfordshire William de Albini, son of Nigel William de Albini Brito William de Albini pincerna William Aldworth of Clapcot William Arnyet William Baker, of Mongewell William Belson, last steward of priory William de Bereford William Bodington, mayor William Bowe William de Braose William Briewer William Broum William Clerk, supervisor of works William de Clobbho, chaplain William Corbucion William of Cramford

57-8, 60, 64, 69, 75, 79, 85-7, 901, 94, 97, 105, 110 21, 24, 45 233 21, 23, 27, 40, 49, 71, 74-6, 97, 167 79 50-2, 62, 64, 240 2, 5, 18, 20, 389. 46, 58, 70, 72-3, 76-7, 81, 89-90, 93, 99100, 116, 131-2, 134, 139, 148, 150-1, 153, 163, 167-80, 187, 193, 198, 202, 205, 208, 223, 225, 239 59, 219-20

287

18, 59, 67, 94, 99-100 225 142 5, 18, 70-3, 76, 134, 150, 170, 225, 231, 234, 239 63 27, 38, 46, 60, 62, 85, 100, 105, 114 99-100 116 98-9 243 1, 5, 59, 88, 95, 106, 110 4, 29-32, 36-7, 47-9, 55, 75, 116, 194, 198-9, 218 20, 25, 28-30, 33, 36, 47, 49, 66, 105, 108, 116, 177, 195, 198 7, 38, 213, 21517 138 56, 109 51 216-17 217 216-17 115 69, 112 189 222 49-50, 53, 102 227 189 30 49 69, 112 113 113 218 49

William Goldsmith William Hyll, dyer William Longchamp, chancellor William Luvet William Marmion William Marshall, earl of Pembroke and regent William Marshall, constable Willaim Martel William de Merc William of Newburgh, chronicler William Norborow William North, labourer William of Poitiers, chronicler William de la Pole, duke of Suffolk William fitzosbern William fitzPeter William I Peverel William II Peverel William Pret William Shalneston, master carpenter William of Sulham William son of Turold William de Upton William de Warenne William de Welham William Woodward Williams, Thomas Winch Eyte or Eyott

189 189 57 218 114 57

Winchester (Hants)

bishop of Tanner St tower Windsor (Berks) castle

63-4, 69, 96, 102, 110-12 155 57, 109 55 189 100 36 65, 95, 98, 103

Great Forest old Windsor, palace Witney (oxon) Wolsey, Cardinal Thomas Wood Street Woodstock (oxon) Worcester (Worcs), burh castle

47 69, 112 4, 30-1, 195 31, 55, 195 102 86, 94

workhouse Wyatt, Thomas, rector of Ducklington Wycombe, High (Bucks) Wyfold (oxon) Wymondham Priory (Norfolk)

218 218 49 30 82 99 150 148, 163, 169, 172, 184-5, 188

York (Yorks castle

288

3, 21, 24-7, 31, 36, 38-9, 56, 59, 72, 103, 105, 197, 199, 216, 242 37, 189, 192, 217, 231 23 118, 234 7-8, 36, 38, 402, 45, 49-52, 569, 61-2, 64-7, 70, 73-5, 82, 92, 96, 98, 101, 105-8, 110-11, 114, 119-20 101 4, 7 240 215, 222, 225 227, 229, 234 56, 75, 109 3, 199 21, 24, 46, 49, 109 177 240 49 76, 98, 99 215-7 196-8 3, 21, 23, 25, 39, 40, 61, 63