Burton Dassett Southend, Warwickshire: A Medieval Market Village (The Society for Medieval Archaeology Monographs) [1 ed.] 1032430028, 9781032430027

Table of contents :
Cover
Half Title
Copyright
Table of Contents
List of figures
Acknowledgements
Archive and finds location
Summaries
Abbreviations
1 Introduction and background
1.1 Project origins and aims
1.2 Post-excavation to publication
1.3 Archaeological background
1.4 Burton Dassett place-names
1.5 Medieval period investigations in the environs of Burton Dassett
1.6 Summary of geology and soils
1.7 Burton Dassett: its origins and development
1.8 Later documentary evidence
1.9 Fieldwork scope and methodology
1.10 Earthwork survey
1.11 Interpretative considerations of excavation data
1.12 Fieldwalking
2 The archaeological sequence
2.1 Pre-medieval activity
2.2 Beginnings of Southend: 12th to early 13th century
2.3 Occupation of new tenements south of the road
2.4 Planned settlement Phase 2, late 13th century
2.5 The chapel
2.6 The apogee of Chipping Dassett in the early 14th century
2.7 The later 14th century: beginnings of decline
2.8 The early 15th century
2.9 Mid- to late 15th century
2.10 The later 15th century
2.11 The abandonment of Southend and its aftermath (AD 1495 onwards)
3 Spatial organisation and the buildings at Southend
3.1 Spatial organisation
3.2 The buildings at Southend
3.3 Metallurgical analysis of smithing residues
4 Daily life and economy at Southend
4.1 Daily life
4.2 Economy
5 Conclusion
5.1 Burton Dassett Southend: a summary
5.2 A changing research context
5.3 Review of research aims
5.4 Other topics of interest
5.5 Cautionary tales
5.6 Direction and aspirations for future work
Bibliography
Index

Citation preview

BURTON DASSETT SOUTHEND, WARWICKSHIRE A medieval market village by Nicholas Palmer and Jonathan Parkhouse

with contributions from Nat Alcock, Alistair Barclay, Lawrence Barfeld, John Blair, Paul Booth, Roger Brownsword, John Crossling, Christopher Dyer, Blanche Ellis, Alison R Goodall, Ian H Goodall, Julie Hamilton, Rupert Housley, Paul Linford, Susan Lisk, J G McDonnell, Alison Mills, Lisa Moffett, Stuart Palmer, E E H Pitt, Stephanie Rátkai, Wilfred Seaby, A H V Smith, Magdalen Snape, Iain Soden, Brian Spencer and Ann Stirland and illustrations by Nigel Dodds, Andrew Isham, Patricia Mallett, Peter Moore and Candida Stevens with Gavin Lines, Tony McKay, Hazel Martingell and Steve Rigby

THE SOCIETY FOR MEDIEVAL ARCHAEOLOGY MONOGRAPH 44

First published 2023 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2023 The Society for Medieval Archaeology and authors The right of Nicholas Palmer and Jonathan Parkhouse to be identifed as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identifcation and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISSN: 0583-9106 ISBN: 978-1-032-43002-7 (hardback) ISBN: 978-1-032-43001-0 (paperback) ISBN: 978-1-003-36527-3 (ebook) DOI: 10.4324/9781003365273 Publisher’s Note This book has been prepared from camera-ready copy provided by The Society for Medieval Archaeology.

Series editor: Alejandra Gutiérrez The Society for Medieval Archaeology is grateful to Historic England for a grant towards this publication The Society for Medieval Archaeology www.medievalarchaeology.co.uk

Cover: Areas D2 and E under excavation, looking east towards the Burton Dassett hills

CONTENTS List of fgures ............................................................. iv Acknowledgements ................................................... vii Archive and fnds location ........................................ viii Summaries ................................................................. ix Abbreviations ............................................................ xii 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 1.11 1.12

2.4 Planned settlement Phase 2, late 13th century .... 63 2.5 The chapel .......................................................... 71 2.6 The apogee of Chipping Dassett in the early 14th century..................................... 73 2.7 The later 14th century: beginnings of decline .......................................................... 90 2.8 The early 15th century ..................................... 100 2.9 Mid- to late 15th century ................................. 120 2.10 The later 15th century ...................................... 131 2.11 The abandonment of Southend and its aftermath (AD 1495 onwards) .............. 139

Introduction and background Project origins and aims ...................................... 1 Post-excavation to publication ............................ 5 Archaeological background, by S C Palmer ..................................................... 6 Burton Dassett place-names, by C Dyer .......................................................... 10 Medieval period investigations in the environs of Burton Dassett ...................... 11 Summary of geology and soils ............................12 Burton Dassett: its origins and development, by C Dyer ......................................................... 14 Later documentary evidence, by N Alcock ...................................................... 21 Fieldwork scope and methodology ..................... 25 Earthwork survey ............................................. 26 Interpretative considerations of excavation data .............................................. 34 Fieldwalking ......................................................37

3

Spatial organisation and the buildings at Southend 3.1 Spatial organisation ......................................... 143 3.2 The buildings at Southend .............................. 160 3.3 Metallurgical analysis of smithing residues ...... 185 4 Daily life and economy at Southend 4.1 Daily life ......................................................... 189 4.2 Economy ......................................................... 213 5 5.1 5.2 5.3 5.4 5.5 5.6

2 The archaeological sequence 2.1 Pre-medieval activity ........................................ 47 2.2 Beginnings of Southend: 12th to early 13th century ................................ 51 2.3 Occupation of new tenements south of the road ............................................... 56

Conclusion Burton Dassett Southend: a summary ............. A changing research context ........................... Review of research aims ................................. Other topics of interest ................................... Cautionary tales .............................................. Direction and aspirations for future work .......

229 230 232 235 236 236

Bibliography.......................................................... 239 Index ...................................................................... 247

iii

LIST OF FIGURES 2.18 Area D2, Phases D22, late 13th century and D23, 14th century ......................................... 68 2.19 Area E, Phase E2, late 13th century .................. 69 2.20 Plan of the chapel and Priest’s House ................ 72 2.21 Phase plan, c1340 ................................................. 74 2.22 Dassett Southend in c1340 .................................. 75 2.23 Area K, Phase K3, early 14th century ................ 76 2.24 Area H, Phase H3, early 14th century ............... 77 2.25 Area H, house, outbuilding and yard, Phases H2–H5, mid-13th to early 15th century ........... 78 2.26 Area W, water hole and surfaces, from east (Phase W2, 14th–15th century) .......................... 78 2.27 Area I, foundations of possible privy/shed, from north (Phase I3, 14th century) .................. 79 2.28 Area E, Phase E3, early 14th century ................ 81 2.29 Area E, baby burial, Phase E3, early 14th century ................................................................. 82 2.30 Area F, house, from east (Phases F3–F4, 14th to early 15th century) ................................. 84 2.31 Area D1, Phase D13, early to late 14th century ................................................................. 85 2.32 Area A, Phases A3, 14th century and A4, early 15th century ............................................... 87 2.33 Area J, smithy, Phase J3, early to late 14th century ................................................................. 89 2.34 Phase plan, c1390 ................................................. 91 2.35 Area K, Phase K4, mid-14th to early 15th century ................................................................. 92 2.36 Area K, house, from west (Phases K3–K4) ....... 93 2.37 Area K, dung gully, from south (Phase K4) ...... 94 2.38 Area K, possible dog kennel, from west (Phase K4) ............................................................ 94 2.39 Area H, Phases H4 and H5, mid-14th to early 15th century ............................................... 95 2.40 Excavations south of the road, showing maximum extent of buildings at the start of the 15th century .............................................. 96 2.41 Area H, south part, from south (Phases H4–H5) ................................................................ 97 2.42 Area E, Phases E4 and E5, mid-14th to late 15th century ......................................................... 98 2.43 Area E, brewhouse and malting kiln, from south (Phases E3–E4, 14th century) .................. 99 2.44 Area E, malting kiln, from south (Phase E4) .... 99 2.45 Area E, drain in brewhouse, from east (Phase E4) .......................................................... 100 2.46 Area D1, Phase D14, early to mid-15th century ............................................................... 101 2.47 Phase plan, c1425 ............................................... 102 2.48 Area J, smithy, Phase J4, early 15th century ............................................................... 103 2.49 Area J, smithy, from west (Phases J2–J4) ......... 104 2.50 Area J, ditch revetment, from south (Phases J3–J4) .................................................................. 105 2.51 Area H, Phases H4 and H5, mid-14th to early 15th century) ..................................................... 106 2.52 Areas I and W, Phases I4 and W2, early to mid-15th century .............................................. 108 2.53 Area I, house, yard and outbuilding, from east (Phases I3–I4) .................................................... 109

1.1 Location plan ...................................................... 2 1.2 Medieval settlement in the Burton Dassett area ..................................................................... 3 1.3 Burton Dassett Southend general plan ................. 4 1.4 Early settlement in the Burton Dassett area ........ 7 1.5 Geology of the Burton Dassett area .................... 13 1.6 Burton Dassett in c1300 ....................................... 16 1.7 Size of tenant holdings at Great Dassett in 1279–80 ............................................................ 18 1.8 Markets and fairs in the Burton Dassett area ...... 19 1.9 Burton Dassett in 1567 ........................................ 24 1.10 Probate inventory of Roger Heritage entries relating to farming ............................................... 25 1.11 Earthwork survey general .................................... 27 1.12 Earthworks, aerial view from north-east ............ 28 1.13 Earthworks, aerial view from west ..................... 29 1.14 Earthwork survey 1: tenements to north-west of chapel ................................................................ 30 1.15 Earthwork survey 2: excavated tenement ........... 31 1.16 Earthwork survey 3: western area ....................... 32 1.17 Phasing and chronology ....................................... 35 1.18 Fieldwalked areas in relation to earthworks and excavated areas .............................................. 37 1.19 Dovehouse Close feldwalking: rubble distribution ................................................ 38 1.20 Dovehouse Close feldwalking: fnds distributions 1 (total fnds, pottery and animal bone) .................................................. 39 1.21 Dovehouse Close feldwalking: fnds distributions 2 (tile, slate and other fnds).. 40 1.22 Chapel Ground feldwalking: rubble distribution ............................................... 42 1.23 Chapel Ground feldwalking: fnds distributions 1 (pottery and animal bone) ...................................................................... 43 1.24 Chapel Ground feldwalking: fnds distributions 2 (tile and slag)....................... 44 2.1 Tree clearance hollows on north side of road ..... 47 2.2 Pre-medieval (Roman or undated) features on north side of road ............................................ 49 2.3 Excavation within chapel and Priest’s House, 2003 ....................................................................... 52 2.4 Phase plan, c1220 .................................................. 53 2.5 Planning and layout of tenements ....................... 55 2.6 Dimensions of tenements south of the road ....... 56 2.7 Area H, Phase H2, 13th century ......................... 57 2.8 Areas I and W, 12th to 14th centuries ................ 59 2.9 Area I, stone-lined trough in SW corner of house, Phase I2, mid- to late 13th century ........ 60 2.10 Phase K2, mid- to late 13th century house and granary ........................................................... 61 2.11 Area K, granary, from east, Phase K2, midto late 13th century .............................................. 61 2.12 Smithy and environs, Phase J2, 13th century ..... 62 2.13 Dimensions of tenements north of the road ....... 63 2.14 Area B, road, from south, Phase B2 .................... 64 2.15 Phase plan c1290 ................................................... 65 2.16 Area A, Phase A2, later 13th century ................. 66 2.17 Area A, house, from east, Phases A2–A5, late 13th to late 15th centuries ................................... 67 iv

3.5 Distribution of slag across the excavated site ... 153 3.6 Distribution of probable midden deposits and rubbish concentrations ............................... 154 3.7 Quantifcation of probable midden deposits and rubbish concentrations ............................... 155 3.8 Distribution of horse equipment ...................... 158 3.9 Distribution of tools and working waste ......... 159 3.10 Distribution of domestic objects ....................... 161 3.11 Distribution of personal items .......................... 162 3.12 Quantities of selected categories of fnds from areas north and south of the road ............ 163 3.13 Jet and bone rosary beads: context details ....... 163 3.14 Comparative plans of all the houses excavated, by period .......................................... 164 3.15 Dassett Southend houses: dimensions of rooms and foor areas ............................ 166–167 3.16 Model of Area A house ..................................... 168 3.17 Architectural stonework: door jamb inscribed with the name Gormand .................. 169 3.18 Structural stonework and metalwork ............... 170 3.19 Comparative histograms of foor area/plans .... 174 3.20 Floor areas of Dassett Southend house plans compared with peasant houses recorded by Alcock and Miles ............................................... 175 3.21 Access analysis of space per phase and building .............................................................. 176 3.22 Dimensions of outbuildings at Dassett Southend ............................................................ 179 3.23 Comparative outbuilding plans ........................ 180 3.24 Barns and stores, estimated grain storage capacity and possible associated acreages ......... 183 3.25 Hammerscale distribution within smithy ........ 186 3.26 Estimation of smithing slag and hearth bottom generation per year ............................... 187 4.1 Distribution of non-ceramic artefacts by area/tenement .................................................... 190 4.2 Distribution of non-ceramic artefacts by period .......................................................... 192–193 4.3 Numbers of animal species present in the excavated tenements ............................ 194–195 4.4 Iron arrowheads ................................................. 195 4.5 Quantifcation of vessel forms: percentage rim count by period ........................................... 196 4.6 Quantifcation of vessel forms: showing vessel type percentages by area ......................... 196 4.7 Stone mortars ..................................................... 197 4.8 Copper-alloy skimmer and iron knives ........... 198 4.9 Bone knife-handle, pewter spoons, copperalloy thimbles ..................................................... 199 4.10 Copper-alloy buckle ......................................... 200 4.11 Copper-alloy brooch and buckles .................... 201 4.12 Copper-alloy buckles ........................................ 202 4.13 Copper-alloy fnger ring and scabbard chape, iron buckles, belt hooks, rowel spurs and spur buckles ................................................ 203 4.14 Bone tuning pegs and pewter pilgrim badges ................................................................. 205 4.15 Sandstone hone inscribed with ‘Nine Men’s Morris’ board ..................................................... 206

2.54 Area I, hearth, from west (Phase I4, early 15th century) ..................................................... 109 2.55 Areas L, M and N, Phases L1, M1 and N1, medieval ............................................................. 110 2.56 Area L, south part, from west (Phase L1, 14th/15th century) ............................................. 111 2.57 Area F, Phases F3 and F4, early 14th to mid15th century ....................................................... 113 2.58 Area F, house, from east (Phases F4–F5, early/mid-15th century) .................................... 113 2.59 Area D2, house, from east (Phases D23–D24, early 14th to early 15th century) ...................... 114 2.60 Area D2, Phase D24, early to mid-15th century ................................................................ 115 2.61 Area D2, hearth, (Phase D24, early to mid15th century) ...................................................... 116 2.62 Area D2, stone-lined slot, (Phase D24, early to mid-15th century) ......................................... 116 2.63 Area E, house, from east (Phases E4–E5, mid-14th to mid-15th century) ........................ 118 2.64 Area E, stone-lined tank (Phase E5, early/ mid-15th century) .............................................. 119 2.65 Area E, ring hollow, part-excavated (Phase E5, early/mid-15th century) ............................. 120 2.66 Phase plan, c1460 ............................................... 121 2.67 Reconstruction drawing, c1475 ........................ 122 2.68 Area A, Phase A5, mid- to late 15th century ............................................................... 123 2.69 Area A, barn (Phase A5, mid- to late 15th century) .............................................................. 125 2.70 Area D1, Phase D15, mid- to late 15th century ............................................................... 126 2.71 Area D2, house, from east (Phases D24–D26, 15th century) ..................................................... 127 2.72 Area D2, Phase D25, mid- to late 15th century ............................................................... 128 2.73 Area D2, from west (Phases D25–D26, midto late 15th century) .......................................... 129 2.74 Area F, Phases F5 and F6, mid- to late 15th century ............................................................... 130 2.75 Area D2, Phase D26, late 15th century ........... 132 2.76 Area D2, hearth (Phase D26, late 15th century) ............................................................. 133 2.77 Area D2, barn, from east (Phase D26, late 15th century) .................................................... 135 2.78 Area D2, Phase D26, late 15th century, north part of area ........................................................ 135 2.79 Area E, Phase E6, late 15th century ................ 136 2.80 Area F, house, from east (Phases F5–F6, mid- to late 15th century) ............................... 138 2.81 Phase plan, c1495 ............................................... 140 3.1 Pottery distribution across all excavated areas .................................................................... 148 3.2 Animal bone distribution across all excavated areas ................................................... 149 3.3 Broken pottery vessels showing sherd links (1–4) ................................................................... 150 3.4 Broken pottery vessels showing sherd links (5–8) ................................................................... 151

v

4.16 Bone, jet, glass and stone beads, bronze letter ‘S’ .............................................................. 207 4.17 Stone hones by area/phase and date ................ 208 4.18 Stone spindle-whorls and stone mould ............ 209 4.19 Stone spindle-whorls by area/phase and date ... 210 4.20 Iron tools ............................................................ 211 4.21 Pewter and copper-alloy candlesticks and iron candle holders .................................................... 213

4.22 Distribution of medieval coins and jetton ....... 213 4.23 The medieval landscape around Burton Dassett ................................................................. 214 4.24 Iron agricultural tools ....................................... 215 4.25 Main sources of pottery recovered at Burton Dassett, and excavated groups mentioned in the pottery report ............................................. 219

vi

ACKNOWLEDGEMENTS First thanks are due to the landowners, E P Smith and Sons of Northend Farm, and Peter Taylor of Green Farm, Northend, who permitted the excavations to take place and for whom they came as an additional trial on top of the compulsory purchase of their land for the motorway. The willingness of the Smiths to permit access for excavation over two years before construction began was a crucial factor in the viability of the project. In particular we are grateful for the co-operation of Messrs Guy Smith, Edward Smith and Steve Smith. The fnancial resources for the project were provided by the Manpower Services Commission Community Programme, English Heritage, Warwickshire County Council, through the Warwickshire Museum, and Stratford District Council. Central Television kindly sponsored an exhibition about the excavations and the Department of Transport arranged to take early entry to the land on the south side of the road and provided the necessary fencing for the excavation there. The Community Programme schemes were administered by the Warwickshire County Council Managing Agency whose coordinators were Paul Ellis, John Ross and Robin Sudbury. The English Heritage funding for excavation and post-excavation was arranged by the then Ancient Monuments Division inspectors Mike Parker Pearson, Graham Fairclough, Deborah Priddy and Andrew Brown, who were also a source of fruitful discussion and helpful advice. Over 80 people took part in the excavations in various capacities. It is a tribute to their fortitude that work was able to continue despite the appalling freezing mud in winter and the hard-baked conditions in summer. The hard work of keeping the excavation running was carried out by the senior supervisor, Mark Newman, and the supervisors, Stuart Palmer, Paul Booth, Jane Gosling, Nicola King, John Marrow and Andrew Waters. The site planning was organised by Philip Underdown, senior draughtsman, and the

fnds processing by Christina Mezciems, senior fnds assistant. The 2013–2020 project to complete the work was funded by English Heritage (Historic England from April 2015) National Heritage Protection Commissions with the support of the Area Inspector, Ian George and the Regional Science Advisor, Lisa Moffett. The project was managed by Stuart Palmer of Archaeology Warwickshire and monitored for English Heritage/ Historic England by David McOmish. We are grateful to all the specialist contributors for their reports and for their ideas about the site, in particular Nat Alcock, the late Lawrence Barfeld, John Blair, John Crossling, Stephanie Rátkai, Wilfred Seaby and Brian Spencer. The illustrations for the report have been produced by Nigel Dodds, Hazel Martingell, Andrew Isham, Patricia Mallett, Peter Moore, Steve Rigby and Candida Stevens. David Adams and Derek Jenkins helped with survey and salvage recording once motorway construction started, and Roy Emmerson produced the computer plots of the site contour survey. The fnal report has beneftted greatly from comments by the referee appointed by Historic England, who gently pointed out a number of instances where error, verbosity or an absence of clarity had crept in. Alejandra Gutiérrez, editor for the Society for Medieval Archaeology’s monograph series, provided further helpful suggestions. Sophie Blain and Marta Caroscio translated the summaries. Those imperfections that remain are entirely down to the authors. Finally, particular acknowledgement should be made to Helen Maclagan, then Warwickshire County Field Archaeologist, who shouldered much of the burden of organising the excavation and provided constant support and advice, and Professor Christopher Dyer, who throughout the project has been generous with his time, support and encyclopaedic knowledge, and has inspired everyone involved with his interest and enthusiasm. He has also kept the pressure on to ensure the successful completion and publication of the work.

vii

ARCHIVE AND FINDS The fnds from the excavation and the excavation archive are held by the Warwickshire Museum, Market Place, Warwick. The accession number is 13/2020. The detailed reports supporting the narratives and discussion presented in this volume, listed below, have been uploaded to the Archaeology Data Service from which they are freely available (DOI: https://doi.org/10.5284/1083492): 6.1 6.2 6.3 6.4 6.5 7 8.1 8.2 8.3 8.4 8.5

8.6 8.7 8.8

8.9 Domestic stonework, by Iain Soden, John Crossling and Nicholas Palmer 8.10 Architectural stonework, by Iain Soden 8.11 Stone roofng material, by Nicholas Palmer 8.12 Roof tiles and ceramic artefacts, by Susan Lisk 8.13 Archaeometallurgical investigation of the smithy and other evidence, by J G McDonnell and Alison Mills 8.14 Coal, by A H V Smith 8.15 Human remains, by Ann Stirland 8.16 Clay tobacco pipe, by Nicholas Palmer 8.17 Flint, by L H Barfeld 8.18 Late Bronze Age pottery, by Alistair Barclay 8.19 Roman and Saxon pottery, by Paul Booth 8.20 Faunal remains, by Julie Hamilton 8.21 Plant economy, by Lisa Moffett 8.22 Radiocarbon dating of spelt wheat, by Rupert Housley 8.23 Archaeomagnetic dating of hearths, by Paul Linford 9 Miscellaneous data tables

Geology, by John Crossling Soils, by Magdalen Snape Earthwork survey, by Nicholas Palmer Excavation methods, by Nicholas Palmer Fieldwalking, by Nicholas Palmer Stratigraphic sequence, by Nicholas Palmer Medieval pottery, by Stephanie Rátkai Coins and jettons, by Wilfred Seaby Copper-alloy objects, by Alison R Goodall with contribution by John Blair Analyses of copper-alloy objects, by Roger Brownsword and E E H Pitt Pewter objects, by Brian Spencer and Nicholas Palmer, with analyses of pewter spoons by Roger Brownsword and E E H Pitt Lead objects, by Nicholas Palmer Ironwork, by Ian H Goodall, with spurs by Blanche Ellis Bone, jet, glass and miscellaneous objects, by Iain Soden and Nicholas Palmer

viii

SUMMARY

Southend, one of fve medieval settlements in Burton Dassett parish, Warwickshire, was the site of a market promoted by the manorial lord Bartholomew de Sudeley, with a charter being obtained in 1267. The settlement prospered, becoming known as Chipping Dassett, and approached urban status, but then declined throughout the 15th century. It was subjected to depopulation in 1497. The site survived as earthworks in pasture until construction of the M40 motorway necessitated the archaeological programme described here. The only building to survive was the 13th-century chapel of St James, reduced, along with an adjacent post-medieval Priest’s House, to a cow-shed. Open area excavations at Southend investigated parts of ten medieval properties. There was some prehistoric and Romano-British activity, with evidence for woodland regeneration and subsequent clearance in the post-Roman period, despite the Feldon area being one often considered to have little in the way of tree-cover since the Roman period. The main period of occupation lasted from the mid-13th century to the late 15th century, refecting the rise and decline of Chipping Dassett. Over 20 complete plans of houses and outbuildings were recorded, exhibiting a range of building techniques. The remains were well preserved, the surviving stratigraphy protected by demolition rubble. In most houses successive building phases were revealed and many internal features survived. A door jamb inscribed with the name of a tenant family ‘Gormand’ suggests a degree of functional literacy. One of the properties was recognised as a smithy during the excavation and a pioneering sampling and analysis of the ironworking evidence was carried out. The site was also sampled extensively for charred plant remains and, unusually for Warwickshire with its

slightly acid soils, a large assemblage of animal bone was collected. Work on these provides direct evidence of medieval agricultural practice, to be compared with the local historical evidence. The large quantities of fnds recovered, probably the largest assemblage from a medieval rural settlement in the West Midlands, enable the reconstruction of the material culture of a late medieval Warwickshire Feldon village. Although the excavated area lay away from the original settlement nucleus, the investigation revealed the mechanics of 13th-century market development with two separate stages of planned development apparent. After the mid-14th century the tenements show a complex pattern of decline leading up to the depopulation of 1497. The different properties followed varying development paths and the excavations chart a process of general community decline against a background of increasing individual prosperity. The evidence of material culture and settlement morphology, taken together, are relevant to the discussion about differentiation and similarities between urban and rural settlement. The medieval pottery has been crucial to the development of the Warwickshire type series. Identifcation of the pottery sources provides evidence for trade connections between the settlement and the wider market network, with the quantities of material from the Chilvers Coton kilns suggesting that manorial connections with North Warwickshire, where the Sudeley family also held land, were signifcant. The summary narrative and thematic discussions (focused upon material culture, spatial organisation, buildings and economy) in this volume are supplemented by detailed stratigraphic description and specialist reports available online through the Archaeology Data Service.

ix

RÉSUMÉ Southend, un des 5 établissements médiévaux de la paroisse de Burton Dassett, dans le Warwickshire, était le site d’un marché promu par le seigneur féodal Bartholomew de Sudeley, grâce à une charte octroyée en 1267. L’établissement forissant, alors connu sous le nom de Chipping Dassett, approchait le statut de ville, avant de décliner au cours du XVe siècle et de connaître une dépopulation en 1497. Le site est resté enfoui sous une butte de terre dans des pâturages, jusqu’à l’aménagement de l’autoroute M40 qui a nécessité le programme archéologique décrit ici. Seul bâtiment toujours en place, la chapelle Saint-James du XIIIe siècle, avait été convertie, avec la maison paroissiale postmédiévale mitoyenne, en étable. A Southend, l’activité humaine remonte à la Préhistoire, des vestiges préhistoriques et britto-romains y ayant été détectés. Alors qu’on estimait que, depuis la période romaine, la zone de Feldon était peu boisée, des traces de régénérations de forêt, suivies d’éclaircies au cours de la période post-romaine, ont été identifées. Sur le site de Southend, les fouilles archéologiques se sont focalisées sur 10 bâtiments médiévaux. La principale période d’occupation s’est étendue du milieu du XIIIe siècle à la fn du XVe siècle, refétant la montée et le déclin de Chipping Dassett. Les relevés du bâti d’une vingtaine de maisons et de dépendances témoignent d’une grande variété de techniques de construction. La stratigraphie a été préservée grâce aux décombres de démolition qui l’ont protégée et les vestiges sont relativement bien conservés. Dans la plupart des maisons, les phases de construction successives ont été identifées ; de nombreuses structures intérieures sont toujours en place. Un montant de porte gravé du nom d’une famille de locataires : « Gormand », suggère un degré fonctionnel d’alphabétisation. Au cours de la fouille, un des bâtiments a été identifé en tant que forge ; un échantillonnage des traces de ferronnerie et une analyse innovante de celles-ci ont été menés. Un grand nombre de restes de plantes calcinées

a également été échantillonné sur le site, ainsi qu’un large assemblage d’os animaux, dont la préservation, du fait de l’acidité des sols du Warwickshire, s’avère exceptionnelle. L’analyse de ces échantillons révèle une activité agricole médiévale, pratique corroborée par les témoignages historiques locaux. Les grandes quantités de mobilier retrouvées, probablement le plus grand assemblage de l’établissement rural médiéval des West Midlands, ont permis de restituer la culture matérielle d’un village de la fn du Moyen Age du Feldon du Warwickshire. Bien que la zone fouillée soit assez éloignée du centre de l’établissement d’origine, la recherche a révélé des mécaniques du développement du marché au XIIIe siècle, avec 2 phases séparées par un aménagement. Dès la 2nde moitié du XIVe siècle, les bâtiments témoignent d’un processus complexe de déclin menant à la dépopulation de 1497. Les bâtiments se développent de diverses façons et les fouilles retracent un processus de déclin général de la communauté en dépit d’un contexte d’augmentation de la richesse individuelle. La confrontation des témoins de la culture matérielle et de la morphologie du village témoigne de l’étroite frontière entre implantation urbaine et rurale. La céramique médiévale a joué un rôle crucial dans le développement des séries du type Warwickshire. L’identifcation des sources des poteries témoigne de liens commerciaux entre le village et un réseau plus large, avec une grande quantité de matériel provenant des fours de Chilvers Coton, ce qui suggère des liens féodaux importants avec le Warwickshire du nord, où la famille Sudeley détenait aussi des terres. Dans ce volume, le compte-rendu et les discussions thématiques (ciblées sur la culture matérielle, l’organisation spatiale, les bâtiments et l’économie) sont complétés par une description stratigraphique détaillée et des rapports de spécialistes disponibles en ligne via le service de données archéologiques (Archaeology Data Service).

x

RIASSUNTO Southend, uno dei cinque insediamenti medievali della parrocchia di Burton Dassett, nel Warwickshire, fu il sito di un mercato promosso dal signore feudale Bartholomew de Sudeley, attraverso un atto costitutivo emanato nel 1267. L’insediamento prosperò, divenendo noto come Chipping Dassett, fno a raggiungere lo status di città, ma declinò successivamente durante il XV secolo, fno al suo spopolamento nel 1497. Il sito è sopravvissuto interrato in una zona di pascolo fno alla costruzione dell’autostrada M40, quando si rese necessario l’intervento archeologico qui descritto. L’unico edifcio superstite è la duecentesca cappella di St James, ridotta, unitamente all’adiacente canonica postmedievale, a stalla per il bestiame. Gli scavi condotti a Southend hanno indagato alcune parti di dieci proprietà medievali. Si evidenziano attività in epoca preistorica e nel periodo di colonizzazione romana della Britannia, con tracce dello sfruttamento selettivo dei terreni boschivi e del loro successivo abbattimento nella fase post-medievale; ciò sebbene l’area di Feldon sia stata spesso considerata come una zona con scarsa copertura boschiva fn dal periodo medievale. La principale fase di occupazione va dalla metà del XIII fno al XV secolo inoltrato, e mette in luce l’ascesa e il declino di Chipping Dassett. Sono state infatti documentate oltre 20 piante complete di case e annessi, che dimostrano l’impiego di tecniche costruttive diverse. Si tratta di ruderi ben conservati che sono stati preservati dall’accumulo dei detriti sovrastanti. Nella maggior parte delle abitazioni sono emerse successive fasi costruttive, testimoniate da diversi elementi interni. Il nome ‘Gormand’ inciso sullo stipite di una porta, a indicare una famiglia di affttuari, suggerisce un certo livello di alfabetizzazione. Una delle proprietà è stata identifcata come fucina durante gli scavi, ed è stata oggetto di un saggio esplorativo e dell’analisi sui resti metallici. Il sito è stato anche sondato in maniera estensiva alla ricerca di piante carbonizzate e, fatto inusuale per lo Warwickshire, dove il suolo è lievemente acido, un nucleo consistente di

ossa animali è stato rinvenuto. Lo studio di questi materiali mostra una prova diretta dello sfruttamento agricolo nel medioevo, che deve essere confrontato con la documentazione storica locale. La vasta quantità di reperti rinvenuti, probabilmente il nucleo più consistente proveniente da un insediamento nelle West Midlands, permette la ricostruzione della cultura materiale di un villaggio tardomedievale dello Warwickshire quale Feldon. Sebbene l’area oggetto di scavo non coincida con il nucleo dell’insediamento originale, la ricerca ha permesso di comprendere le dinamiche dello sviluppo di un luogo di mercato nel XIII secolo, con l’emergere di due distinte fasi di pianifcazione nel suo sviluppo. A partire dalla metà del XIV secolo i caseggiati mostrano un complesso processo di declino che porterà all’abbandono nel 1497. Le diverse proprietà sono andate incontro a cambiamenti articolati, e gli scavi delineano un generalizzato processo di declino di questa comunità, in un contesto in cui –di contro– andava incrementando la fortuna individuale. I dati relativi alla cultura materiale e alla morfologia dell’insediamento, considerati nel loro complesso, danno un rilevante apporto alla discussione circa le differenze e le somiglianze fra insediamenti urbani e rurali. La ceramica medievale qui rinvenuta è un elemento cruciale nello sviluppo delle tipologie del Warwickshire. L’identifcazione dei luoghi di approvvigionamento rivela i legami commerciali esistenti fra questo insediamento e una più ampia rete di mercato: le quantità non trascurabili di materiali provenienti dalle fornaci di Chilvers Coton, suggerisce delle rilevanti interrelazioni feudali con il Warwickshire settentrionale, dove la famiglia Sudeley vantava ulteriori proprietà. La sintesi esposta e le discussioni tematiche –che si incentrano sulla cultura materiale, l’organizzazione dello spazio, gli edifci, e l’economia– presenti in questo volume, sono integrate da una descrizione stratigrafca dettagliata e da relazioni specialistiche disponibili online grazie all’Archaeology Data Service.

xi

ABBREVIATIONS BAH

Birmingham Archives and Heritage (formerly Birmingham Reference Library) BL British Library BM British Museum CIPM Calendar of Inquisitions Post Mortem HEH Henry E Huntington Library, San Marino, California HER Historic Environment Record Warwickshire NMR National Monuments Record, Historic England, Swindon

NRO PRO SCLA

Northamptonshire Record Offce Public Record Offce Shakespeare Centre Library and Archives, Stratford-upon-Avon (formerly Shakespeare Birthplace Trust Record Offce) SRO Staffordshire County Record Offce TNA The National Archives WRO Warwickshire Record Offce

xii

1 INTRODUCTION AND BACKGROUND

1.1 PROJECT ORIGINS AND AIMS

The parish of Burton Dassett lies midway between Warwick and Banbury in south Warwickshire (Figure 1.1). The parish runs from the Burton Hills, an outlier from the Cotswold escarpment, in the south-east, down into the Lias clay vale, which forms part of the Warwickshire Feldon, to the north-west. Today Burton Dassett contains two villages, Northend and Knightcote, and the smaller hamlet of Burton around the parish church, all of medieval origin (Figure 1.2). There is also the settlement called Temple Herdewyke associated with the army depot, Defence Munitions (formerly MoD) Kineton which is entirely modern, although preserving the memory of an earlier settlement in its name. The pattern of medieval settlement in the parish was more complicated with at least fve centres of settlement in the 14th century. These were Burton, which was presumably the central settlement, Knightcote, Hardwick, Northend and Southend. Southend, the largest of these, was the site of a market for which a charter was obtained in 1267 by the lords of the manor, the de Sudeley family (VCH 1949, 70). The market was successful, and in the early 14th century Burton Dassett had a high tax assessment, presumably refecting the prosperity of its inhabitants, with Southend becoming known as Chipping Dassett and almost approaching the status of a town. Through the 15th century, however, Southend went into a decline almost as rapid as its rise, and in 1497 the remaining inhabitants were evicted, the buildings demolished and the site enclosed for sheep pasture. Burton was depopulated and enclosed at the same time. Hardwick had already been deserted, and both

Knightcote and Northend, the surviving villages, were shrunken. This pattern of late medieval decline and desertion is in many ways typical of the Feldon area. In the parishes around Burton Dassett (Figure 1.2) about half the villages that existed in the Middle Ages had disappeared by the end of the period. Until 1988 the site of Southend (Figure 1.3) was marked by extensive earthworks covering some 30 hectares in the six felds surrounding the 13thcentury St James’s Chapel, the only medieval building to survive, now converted to a house, but until 2003 used as a hovel or animal shelter. A modern road still runs east–west across the site, following the line of a medieval street; a second parallel street is visible as a hollow-way in the earthworks to the north. These are linked by another modern road going to Northend, also following a medieval line. Around the chapel there are two cottages and a farm making the hamlet known as Little Dassett. In 1985 it became clear that the route of the planned extension of the M40 motorway from Oxford to Birmingham would run across the west side of Southend. Consultation between the motorway engineers and the County Field Archaeologist meant that the route was able to avoid most known archaeological sites, but here the proximity of Defence Munitions Kineton, the Burton Dassett Country Park on the hills and the park of Farnborough Hall to the south-east made the placing of the motorway through the site almost inevitable. The motorway itself would cut a swathe some 85m wide across the site. In addition, the construction of a bridge to take the existing east–west road across the site over the motorway and the temporary diversion of the road would destroy further areas to the north and south of the road. The scale of the threat to the site prompted the Warwickshire Museum to organise an excavation

2

Figure 1.1 Location plan of Burton Dassett (outline of Feldon after Warnock 1993)

BU RTON DA S SET T SOU T H E N D

I N T RODUC T ION A N D BAC KGROU N D

3

Figure 1.2 Known medieval settlement in the Burton Dassett area directed by the principal author (NJP) to record as much as possible before motorway construction began. Although Warwickshire had been in the forefront of the modern revival of interest in medieval villages, up to 1985 the contribution of archaeology had lagged behind those of documentary research and feld survey where the pioneering work of Beresford (1945–46), Hilton (1975) and Thorpe (1965) was followed by Bond, Dyer and Hooke (Bond 1974; 1982; Dyer 1981; 1982; Hooke 1985). While there

had been some excavations on deserted village sites in the county, these were all on a small scale (Bond 1982, 161). A large-scale modern excavation on a village site in the county was therefore long overdue. Burton Dassett would have been a suitable site for a research excavation, setting aside the threat of modern development. A number of factors made it possible to regard it as a representative example of a Warwickshire Feldon village. Its multinuclear settlement pattern is repeated in a number of parishes

4

BU RTON DA S SET T SOU T H E N D

Figure 1.3 Burton Dassett Southend plan, showing earthworks, excavated areas and motorway construction corridor

in the immediate vicinity such as Brailes, Tysoe and Chesterton. Its rapid rise and decline refect the general pattern of the area as a whole, albeit in an exaggerated form. Its medieval land-use pattern with a very high proportion of the parish area under arable cultivation is also typical of the area and of the open ‘champion’ landscape of the Feldon. A second area of interest was the presence of the market and the proto-urban character of Southend/Chipping Dassett. It had been said that excavations have concentrated on successful towns and unsuccessful villages and investigation of an unsuccessful town was needed to redress the balance—Chipping Dassett had even been suggested as a possibility (Palliser 1987, 66–67). Finally, there was the fact that there had already been a considerable amount of previous documentary research and feld survey carried out in the parish so that its history was already fairly well understood (Westacott 1924; 1936; Chatwin 1935; Bond 1974, 108; 1982, 157–160; Alcock 1977; 1981). In addition, a small excavation had been carried out at Burton close to the parish church in 1973 (Hunt 1972–73; 1973). In the early 1980s a pessimistic view was taken of the contribution that archaeological excavation would be able to make to the study of medieval rural settlement in the region, because it was felt that meaningful results would only come from large-scale,

long-term excavations involving resources beyond those normally available for rescue archaeology (Bond 1982, 161). At Burton Dassett Southend two crucial factors enabled these conditions to be fulflled: the frst was the willingness of the landowner, E P Smith and Sons of Northend Farm, to allow access to the north part of the site over two years in advance of the start of motorway construction; and the second was the agreement of the Manpower Services Commission to provide a labour force through a series of Community Programme schemes designed to provide work for the long-term unemployed, with the specialist support and direction of the project funded by Warwickshire County Council and then English Heritage. Accordingly, the excavation began in May 1986 and ran continuously until motorway construction work began in September 1988. The excavation, as expressed in the research design, was to pursue general aims of studying the material culture of a Warwickshire Feldon village by area excavation of a number of complete crofts and building complexes and to recover the plans of a series of buildings both domestic and agricultural. Other important objectives were to collect faunal and botanical evidence to provide insight into the agricultural economy and natural environment of the settlement, and to collect substantial pottery and other artefactual assemblages to permit analysis of

I N T RODUC T ION A N D BAC KGROU N D

the circumstances of the inhabitants and of regional patterns of trade. It was also hoped to obtain evidence for the origin and development of the settlement to provide a better understanding of the dynamics of polyfocal settlement and of settlement nucleation, expansion and decline. The excavation covered two areas, to north and south of the modern road across the site which follows the line of a medieval street, probably identifable in documents as ‘Newland’ (Figure 1.3). In the northern feld, known as Chapel Ground (previously Town Field), a detailed survey of the earthworks and a contour survey were carried out before excavation began. Here the main area, excavated from May 1986 to March 1988, covered substantial parts of fve properties in a regular block of six fronting the street (Areas A, D1, D2, E and F) (see Figure 1.15). To the west was part of a north–south side street (Area B) (see Figure 1.16). These parts of the site were totally excavated. Further west (Areas B, C and G) trial trenches were excavated across a number of earthwork platforms. A survey of the fabric of St James’s Chapel and the adjacent 17th-century Priest’s House was also carried out. The second main area, to the south of the road, was in a feld known as Dovehouse Close. In 1987 this feld was ploughed for the frst time in some years and feldwalking (see Figure 1.18, areas Z1–Z6) located further buildings on and off the motorway line. Arrangements were made for more limited excavation, but when the topsoil was removed in 1988 the buildings here were found to be much denser than the feldwalking had suggested. From March to August 1988 parts of fve properties were excavated (Areas H, I, J, K and L), including one used as a smithy (Area J), as well as sections of street and a communal water hole (Area W). These areas were not totally excavated but the buildings were cleared and planned and then selected sample trenches were excavated down to natural substrate. By the end of the excavation therefore, substantial parts of ten properties had been examined, including a good range of houses and outbuildings. Large quantities of fnds and environmental samples had also been collected. The north side of the road was occupied from the later 13th century probably until the evictions of 1497, whereas on the south side occupation began earlier, in the early/mid13th century, and, on the excavated tenements, it also ended earlier, the properties being abandoned by the mid-15th century. The date range of the properties thus spanned the period of Southend’s rise to prosperity and decline, and the two areas to north and south of the road also produced some useful contrasts. Although more work could have been done if extra time had been available, enough was done to cast light on most of the areas identifed in the original research design. By way of postscript, in 1991 Chapel Ground east of the now completed motorway was ploughed up,

5

except for an area around the chapel, for the frst time in living memory. A feldwalking survey (see Figure 1.18, areas Y1–Y4) was carried out to record the surface material disturbed and assess the extent, density and date range of occupation over this part of the settlement. 1.2 POST-EXCAVATION TO PUBLICATION

Following the main excavation in 1989 a full postexcavation programme was launched and by 1992 reports had been prepared on the geology and soils of the area, the archaeological and historical background, coins and jettons, copper alloy, ironwork, pewter, lead, bone, jet and glass objects, domestic and architectural stonework, tile, coal, human remains, clay tobacco pipe, fintwork, Late Bronze Age, Roman and Saxon pottery. Studies of the faunal remains, plant economy, ironworking evidence, radiocarbon dating and remanent magnetic dating of hearths had also been completed. Drafts of the accounts of the feldwork and the medieval pottery report were substantially complete. Only the fnal discussion and conclusions required much further work. However, at this point the principal author was diverted to other responsibilities and work on the project gradually lapsed. Twenty years suddenly went by during which the site began to appear on lists of important unpublished excavations in the region. Nevertheless, it was evident that the Burton Dassett excavations had the potential to make a substantial contribution to medieval settlement studies in the region, illuminating a number of key themes. These include: the evidence for post-Roman woodland regeneration and subsequent clearance; the question of the extent to which a distinction between rural and urban settlement may be made; the physical evidence for setting out a new area of settlement; the unique trajectories of development, modifcation, decline and abandonment exhibited by each tenement; and the structural variety and quality of the buildings. In addition, the fnds assemblage and particularly the pottery, likely to be the largest assemblage from a rural site in the West Midlands, had the potential to illustrate material culture in some detail; whilst the sampling and analysis of the metallurgical debris was unprecedented for its time and remains one of the most important sources of evidence for the workings of a post-Conquest medieval smithy. The project to complete the report was foated in 2012 and began in November 2013. As many of the specialist reports had been completed over 20 years before, contributors were canvassed where possible and offered an opportunity to update their work. Some were prepared for their reports to stand so long as it was made clear that they were written before 1992 and do not consider literature produced after that date. Others agreed to undertake minor revisions. These include the authors of the

6

historical introduction and the reports on geology, plant economy, faunal remains and archaeomagnetic dating. The main sections on which some updating has been carried out are those on medieval pottery, where more is now known about the sources and distribution of the fabrics, and the ironworking evidence which has been reviewed to take account of subsequent progress in the subject. The principal author (NJP) also expanded and revised the discussion sections which had been initiated over 20 years previously, to take account of the considerable quantity of work which had been published in the intervening period. In addition, the opportunity was taken to include an account of archaeological recording carried out in 2003 when the medieval chapel and adjacent 17th-century Priest’s House were converted to domestic use. In 1986 the building had been in danger of collapse, and in the absence of any viable agricultural use that would justify the costs of repair, the conversion of the building, a Scheduled Ancient Monument, was agreed between the landowner and English Heritage. These schemes were approved in 1989 and 1997 but the work only took place in 2003. During this fnal effort to take the project to its conclusion it became evident that the volume and scale of the report as originally proposed was unsuited to modern publication expectations. Not only had medieval settlement studies moved on, but also the character of archaeological publication, with a move away from extensive publication of ‘level 3’ archives (sensu Cunliffe 1983). The project design was revised accordingly, to support a new approach which would draw upon the opportunities afforded by digital archiving and dissemination. This involved summarising the structural sequence from the detailed narratives already compiled and drawing upon the specialist contributions to complete and expand upon the discussion and synthesis of a range of important aspects of the site. These form the basis of this printed monograph. The complementary online reports consist of the full detailed text describing the structural sequences across the site together with the specialist reports in their entirety, available for download from the Archaeology Data Service (DOI: https://doi.org/10.5284/1083492). This process of adaptation and revision was undertaken by the second author ( JP). 1.3 ARCHAEOLOGICAL BACKGROUND

by Stuart C Palmer The medieval villages of Burton Dassett developed in a landscape infuenced by the topography, geology, drainage and soils and also by its human geography. Prior to the excavations in the late 1980s, most of the evidence for early land use in the area had come from chance fnds. The only pre-medieval sites reported had been on the Roman town and villa

BU RTON DA S SET T SOU T H E N D

at Chesterton, on the hillfort at Nadbury Camp, Ratley, and at Ratley Castle. An Iron Age pit group was excavated at Church Hill, Burton Dassett, in 1989 in advance of stone quarrying for the motorway (Booth 1989, 86) although it has yet to be fully analysed. The Saxon activity reported adjacent to Burton Dassett church (Hunt 1973; 1972–73) has since been confrmed as Roman (recently concluded unpublished project by Archaeology Warwickshire) and relates to stone walls and combed tile found in services trenches more recently (Palmer 1998, 85). There had been some systematic feld survey directed specifcally to locating early sites around Chesterton (Bond 1982, fg. 7.1), Kineton (McKay 1985) and Tysoe (by G Miller and K Carrdus), and on some of the few known cropmark sites (Hingley 1986). Compared with the Avon Valley to the north there has been relatively little air reconnaissance in the area south of Chesterton, probably because the subsoil is not generally conducive to the production of cropmarks and thus unrewarding for aerial surveyors, although several undated cropmarks are likely to be later prehistoric. Since 1990, this area has seen relatively little development-led archaeological work compared with other parts of Warwickshire, but in 1999 part of a Roman period farmstead west of Chesterton was excavated in advance of a gas pipeline (Palmer 2010), and in 2012 another in Bishop’s Itchington was located during the evaluation of the site of a proposed wind farm (Kimber and Mayes 2012). Some of the recent work in the area has been carried out by local amateur groups and individual researchers including The Feldon Archaeological Society who have excavated a Roman period villa/ settlement at Grandslet Farm, Burton Dassett, and the Warmington Heritage Group who have carried out excavation and surveys on a similar site at the National Herb Centre, Warmington. Fieldwalking and geophysical survey by the Edge Hill Project has discovered a number of prehistoric and Roman sites, mainly in Tysoe (e.g. Sabin and Donaldson 2010), and there has been further excavation and survey work around Chesterton. The distribution of sites and fnds reported to the Historic Environment Record (Figure 1.4; also table in online archive, Section 9.7, where bibliographic references will be found) does not represent all pre-medieval activity in the area, but refects the minimum of recorded sites. Records of some sites are from an age where far less precision was given to recording the location of fnds, and there was insuffcient detail to enable them to be plotted accurately on Figure 1.4; these are indicated by an asterisk next to their reference numbers in the following account. The lower density of sites towards the east probably refects a lower intensity of feldwork rather than less intensive land use. The categorisation of these sites and fnds is a necessarily obtuse chronology of change across a

I N T RODUC T ION A N D BAC KGROU N D

7

Figure 1.4 Early settlement in the Burton Dassett area (data from Warwickshire Historic Environment Record)

very broad period which barely refects the social and environmental fuctuations of occupation and culture. The paucity of high-quality evidence from which to establish the pre-medieval background inevitably requires that inferences are taken from a much broader area than can be depicted on Figure 1.4; necessarily caveated that models of settlement developed for other regions should not be imported uncritically (Ray 2007, 51–52).

Between c8000 and 2600 BC The small ovate Palaeolithic hand axe found at Chesterton (Figure 1.4 (1)) is likely to have been redeposited during glacial activity and does not refect early human activity within the area. The earliest evidence from the Holocene is Mesolithic fintwork found in small quantities representative of short stays in elevated locations with panoramic

8

views in the Burton Hills, Burton Dassett (2) and at Nadbury (Camp), Ratley (3). These and other sites in the region (cf. Palmer in Alexander et al. 2007) support a model for low-intensity occupation into the 4th millennium BC, as perhaps does the absence of funerary or ceremonial monuments of the earliest Neolithic (Garwood 2007, 199–200). A signifcant concentration of early Neolithic pits and post-holes on the hilltop at Warwick (Gethin et al. 2011) is followed in date by other, disparate river valley groups of which some along the Avon develop as discrete ceremonial complexes by the middle Neolithic (Palmer 2007; 2010). A very low incidence of fintwork at these sites suggests that they were not the focus of permanent settlement but were positioned at a line of communication along the river’s edge where people gathered and ceremonies were enacted. Stone and fint Neolithic axes have been found at Ratley (4*), Tysoe (5), Kineton (6) and Chesterton (7*), with another just outside the mapped area in Harbury (8), of which some probably demonstrate an as yet to be refned long-distance trading network rather than evidence for deforestation. Other fintwork has come from Fenny Compton (9), Compton Verney (10) (where systematic feldwalking yielded a sparse scatter of fint; see Dyer 2000, 54–58), Burton Dassett (11*), Kineton (12), Chesterton (13–16), Avon Dassett (17), Nadbury (18) and Ratley (19). Paucity of raw fint material aside, even though much of this comes from systematic surveys there is no indication of permanency or even repeated visits at these locations. The surge of development-led interventions across the county has so far revealed little data relating to the local environment in the early Holocene and there are no pollen cores from which to adduce the rate of deforestation or the beginnings of agriculture (Greig 2007). Whilst some limited inferences can be made from Neolithic or Early Bronze Age pit groups in the Avon Valley containing feasting remains and incidentally preserved burnt fora and fauna, none have been recorded in the local area from which to extrapolate data. It may therefore be salient that the absence within the region of some typical forms of Neolithic monuments is coincidental with an absence of evidence for agricultural production (Garwood 2007, 196–197), and it would be reasonable to suggest a model for the Burton Dassett area in which gatherer-hunting gave way to domesticated ungulate herding in the early 4th millennium BC, as the forest was restructured piecemeal into pasture (LaFreniere 2008). Between c2600 and 800 BC Evidence for Bronze Age settlement in the region remains equally elusive in the earlier part of the period, albeit with a steady rise in isolated pit fnds in which distinctive Beakers, Collared and other urns were deposited. A profusion of circular cropmarks,

BU RTON DA S SET T SOU T H E N D

assumed to be burial sites but seldom groundtruthed by excavation, is indicative of landscape exploitation but not necessarily of population density and concomitant farming. In the later Bronze Age there is now signifcant evidence for the formalised partition of landscapes associated with isolated farmsteads or agglomerated settlement, the latter so far only found in the east of the county (Palmer 2017). Bronze Age fint arrowheads have come from the Burton Hills (20) and from Nadbury Camp (21), whilst funerary metalwork is recorded from Ratley (22*) and Wormleighton (23*), elevated positions where such burials might be expected. However, a number of earth mounds, in low-lying Bishop’s Itchington (24–26) and Gaydon (27), and ring ditches in Chesterton (28) and Kineton (29), are recorded as possible Bronze Age funerary monuments, although other explanations are perhaps as likely. The only site of this type so far investigated, at Warmington (30), turned out to be of post-medieval origin. A possible burnt mound in Chesterton (31) may also be of Bronze Age date and a Middle Bronze Age cremation cemetery has been recorded at Harbury (Palmer and Hillman 2020; not shown on Figure 1.4). No boundaries, either natural or man-made are known in the Burton Dassett environs despite a high probability of hilltop occupation at, for instance, Nadbury (McArthur 1987–88). Although rare, later Bronze Age pottery has been recovered from across the county where Iron Age sites have been excavated and very often can be said to have represented the earliest evidence for activity on predominantly Iron Age settlement sites (Palmer 2017). Between c800 BC and AD 43 Burton Dassett lies at the northern tip of the Cotswold hillfort zone, with Nadbury Camp, Ratley (32) on the top of the escarpment at Edge Hill and another possible hillfort on Gredenton Hill (33), Fenny Compton, close to the parish boundary with Burton Dassett. The earthworks here suggest a bivallate hillfort of some 2 hectares but it is quite possible that the so-called ramparts are simply cultivation terraces. Little can be inferred from such proximity given that hillforts remain an ambiguous phenomenon and their absence over most of Warwickshire where more settlement data is available excludes them from meaningful hierarchies. Perhaps the most important aspect of their presence is that they demonstrate a requirement for a signifcant investment of physical labour even if the mechanism for its management remains a matter for speculation. The Iron Age activity signifed by the pit group on Church Hill, Burton Dassett (34) and pottery found on Roman sites at Grandslet Farm, Burton Dassett (35) and the National Herb Centre, Warmington (36), attest to the likelihood of a signifcant local population but the data is insuffcient to reveal much about how they lived and what their relationship to hillforts,

I N T RODUC T ION A N D BAC KGROU N D

both putative and confrmed, might have been. In the wider area, excavations at Chesterton Camp (37) found some pits and a burial, a single Iron Age sherd was collected at Barn Hill, Chesterton (38), a Dobunnic stater has been found elsewhere in Chesterton (39) and a Dobunnic half stater of Eisu has been found in Compton Verney (40). Fieldwalking produced a possible Iron Age sherd on one cropmark enclosure in Butlers Marston (41). In the vicinity of Nadbury Camp, Ratley, there are two sets of enclosures (42–43), one very probably of Iron Age date. There are two further enclosures to the north of Warmington (44–45), and others in Bishop’s Itchington (46) and Chesterton (47). Although this ostensibly conforms to a south Warwickshire model of discrete farmstead settlement devoid of proven felds and perhaps indicative of a largely pastoral economy, there is little environmental data and a faunal record too poor (Palmer 2017) to establish the inevitably nuanced farming practices and possible impacts at Burton Dassett. Between cAD 43 and 410 The evidence for activity during the early Roman period recorded on the HER is probably a useful proxy for the later Iron Age landscape, given the availability and subsequent residuality of hard-fred ceramics that are more easily recognised in the ground, and that many excavated sites yield earlier pottery. The abundance of early Roman pottery across south Warwickshire certainly suggests that local populations were relatively early adopters of at least some aspects of Romanised culture and at least partially engaged with the Roman economy, although the Iron Age tradition of circular buildings continues even in urban areas through the 2nd century AD. The relative density of villa sites in the southern part of the county, which may be related to settlement patterns within former tribal areas (Booth 2018, 43), suggests that villas played a signifcant role. In contrast populations in the north of the county seem to have perpetuated Iron Age practices for longer, with the intriguing possibility that this was associated with the continued importance of herding. This may also include horse breeding as implied by the preponderance of horse-related paraphernalia reported to the Portable Antiquities Scheme (Bolton 2010), pit-defned (ranch) boundaries and stock enclosures, and even perhaps the Lunt gyrus. The major settlement in the area was the small town covering around 27 hectares at Chesterton Camp on the Fosse Way (48). At Banbury Road Farm, Kineton (49), there is another settlement with one large stone scatter and nine others possibly fronting a street. This may represent a villagetype site although a villa with principal buildings dispersed over a wide area excavated at Salford Priors may provide a precedent for a regional variation in villa development (Palmer 2000). A concentration

9

of unexcavated or partially excavated villas lies to the south of Chesterton along the Fosse Way with sites known at Ewefelds Farm, Chesterton (50), Hill Farm, Lighthorne (51), Brookhampton, Kineton (52) and Ireland Farm, Gaydon (53), close to the parish boundary with Burton Dassett. Other villas or farmsteads with stone buildings are known in Burton Dassett at Burton (54), just south of the church and at Grandslet Farm (55) east of Northend, and at the National Herb Centre, Warmington (56). A fragment of Roman marble sarcophagus found in Shotteswell (57*) may indicate another villa in that vicinity. Other sites with scatters of pottery and other fnds suggest the presence of smaller farmsteads. These are known in Chesterton (58–61), Harbury (62), Fenny Compton (63), Butlers Marston (64), Oxhill (65), Tysoe (66), Lighthorne (67) and Bishop’s Itchington (68); another one is reported in Burton Dassett, near Knightcote (69*). Two concentrations of pottery in the north of Compton Verney parish seem to represent farmsteads; the sparser distribution of material recovered from feldwalking elsewhere in the parish doubtless results from manuring. Dyer (2000, 57), noting the alignment of later feld boundaries with the Fosse Way, suggests the possibility of Roman feld systems remaining in use to form a template for later landscape elements. This is in accord with the conclusion of the ‘Fields of Britannia’ project that within the Central Province (broadly comparable in extent to that defned by Roberts and Wrathmell 2000a) there is apparently a high (over 70%) correlation between the orientations of Roman feld boundaries and those of medieval feldscapes (Rippon et al. 2014, 207, fg. 5; 2015, 305–342; see also below, 2.1). On Dunsmore it is possible to see 1st millennium BC boundaries fossilised in medieval feldscapes, suggesting a much older establishment and maintenance of estates or land-units (Palmer 2002), which might be repeated where existing parishes are cut by the Fosse. A possible model for medieval parish development in south Warwickshire allows for the presence of at least one villa, which may or may not have emerged from a pre-existing estate, and very likely also including satellite farms, although the evidence to test this model is presently insuffcient. Serendipitous fnds of single coins, small quantities of pottery or other objects at Chesterton (70–72), Harbury (73), Bishop’s Itchington (74*), Ladbroke (75*), Lower Radbourn (76), Watergall (77), Lighthorne (78–80*), Chadshunt (81), Avon Dassett (82), Kineton (83–84), Butlers Marston (85), Ratley and Upton (86–87), Bishop’s Itchington (88), Fenny Compton (89) and Ladbroke (90) do not seem suffcient to indicate actual occupation in a particular location but suggest its presence in the near vicinity. Roman coin hoards from Ladbroke (91) and particularly Warmington (92), may have been deposited as offerings at signifcant places rather than buried in times of strife. South Warwickshire farms, where evidence is available, practised a mixed regime

10

growing mainly spelt and breeding cattle, sheep and less intensively pigs and horses. Their farmscapes then would include access to water, pasture, perhaps wooded pasture and arable. Some felds at least, possibly mostly the infeld, were rectangular and defned by ditches, often associated with trackways (e.g. Salford Priors). One might imagine the undefned outfeld being used for summer pasture and winter foggage. Clay-based farmland such as at that at Southend, far from being intractable to ploughing, might well have been incorporated as fertile agricultural land. Between cAD 410 and 800 Notwithstanding the current debate on the number and intent of continental immigrants in the early postRoman period, it is clear that there was often a shift in settlement foci, a change in building construction methods and a return to handmade ‘bonfre’ pottery in the 5th and 6th centuries, preventing the easy identifcation of such occupation sites. That the majority of evidence for the early part of the period (5th to 7th centuries) within Warwickshire is funerary is due in part to an apparent predilection for siting cemeteries along river valleys; thus the data has been infuenced by discoveries of furnished burials, with their more readily recognisable archaeological signature, during the course of gravel extraction in recent centuries. However, in several locations in south Warwickshire it is possible to suggest continuous occupation of land-units or estates from the later prehistoric period to the present day, with cemeteries at Wasperton and Stretton on Fosse in use across the Roman to Saxon transition. Unhelpfully, it appears that the middle Saxon period in this region was aceramic with pottery only beginning to reappear in the 9th/10th century. Fortunately the widespread adoption of radiocarbon dating human remains can now identify some of the inhabitants who were buried without Germanic grave goods, including re-dating cemeteries such as Tiddington (in Alveston parish) which was previously assumed to be Roman and now shown to be of 7th to 9th century (Palmer 2019). For the early part of the period, the only place in the area where direct settlement evidence has been found is Chesterton (95) where excavations on the Camp found hearths, a comb and what was presumed to be early pottery. Ten burials have also been found just outside the defences of the Camp (96) rather tenuously associated with Anglo-Saxon beads, and early pottery has been found on a Roman farmstead just west of the town (97). In 1774, two or three burials were found at Compton Verney, possibly in a barrow, two of them associated with gold jewellery, a pendant and a bracteate (105*). There are records of three groups of undated early burials in Lighthorne (102–104), one of which produced a set of hanging bowl escutcheons (erroneously described as from

BU RTON DA S SET T SOU T H E N D

Chesterton). At Kineton (100) a cemetery of ten burials, with a ‘javelin’ and a sword, was discovered in a stone pit at Pittern Hill in 1861. In 1891 part of an iron spearhead and two beads were found in sand digging on Oak Hill in Farnborough (101). It seems likely that these were also from a burial (or burials), but the skeleton had decayed. At Camp Lane, Ratley (99), four burials, radiocarbon dated to cal. AD 640–820 exhibited pathologies consistent with poor childhood health (Palmer 2011). The cemetery (93) at Burton Dassett discovered in 1908 during ironstone quarrying on Mount Pleasant Hill overlooking Northend included at least 35 skeletons lying mainly in two lines aligned west–east, with fnds including an iron knife and pottery (Warwick Museum Catalogue A237) and is likely to have been middle Saxon, but the group of undated burials (94) recorded in 1909 from the hill to the south is less certainly Saxon. Across Warwickshire such early Saxon settlements as have been recognised are generally characterised by small groups or even individual sunken-featured buildings (SFBs) suggestive of small, isolated farms, although at Baginton two groups, totalling at least ten SFBs 2km apart, were sited on exposed ridges overlooking the River Sowe (recently concluded unpublished project by Archaeology Warwickshire). Some burials have been found in Roman settlements including a single female burial in the Roman settlement at Grandslet Farm, Burton Dassett (98), and the occurrence of Roman pottery and other fnds in Saxon features alludes to a continuity between the periods in question (e.g. Salford Priors, Baginton). It is likely therefore that the Burton Dassett area was populated in the 5th and 6th centuries, at the least by the descendants of the Roman populace under the cultural infuence of the near continent, if not by some continental immigrants. The cemetery on Mount Pleasant Hill (93) clearly represents a population in the vicinity who it is assumed were agricultural and the forebears of the medieval population. Assertions as to the environment these people inhabited are diffcult to surmise from the archaeological evidence, although given the evidence for place-names in the area (Dyer, below), whilst the Burton Hills may have been wooded, the lower lying fat lands may well have been agriculturally exploited. 1.4 BURTON DASSETT PLACE-NAMES

by Christopher Dyer The 28 settlement names within 8km (5 miles) of Burton Dassett include ten names incorporating the element -tûn and this feature is characteristic of the whole of south-east Warwickshire and adjacent parts of neighbouring counties. The element is thought to belong to an early phase of name formation, though not the earliest. The corresponding numerous leah names in the woodlands to the north and west of

I N T RODUC T ION A N D BAC KGROU N D

Warwickshire are represented in the Burton area by a single name, Ratley, which suggests that woods and woodland clearings were scarce. In addition to tûn names referring to settlements or estates, four cotes and a wic imply minor or outlying places or, in the case of cotes, the presence of smallholdings. Southend and Northend were names applied to outlying settlements at a relatively late date. The earliest names (recorded for the whole country before 730) include those based on topographical features, and in the Burton area three names refer to hills (Farnborough, Gaydon and Hodnell), and four to streams (Avon Dassett, Bishop’s Itchington, Ladbroke and Radbourn). Names which indicate early origin are those using words of preEnglish languages: British or Latin. Chadshunt incorporates the Latin word funta, meaning a well, and Avon Dassett and Bishop’s Itchington were formed from British stream names, Avon and Itchen. Such names could have survived from a phase of contact between the native British population and Germanic incomers, during which the words could be transmitted (for modifcation of this view, Padel 2013). Chesterton includes a loan word from Latin and shows that the English-speaking inhabitants were conscious of Roman antecedents. Dassett, an important district name, has troubled place-name scholars. The two main alternative solutions are to translate it as ‘shelter for deer’ (in Old English) or ‘oak tree wood’ using a British vocabulary. Both interpretations imply a wooded environment, but there is no sign of this in the 11th century. Perhaps the name was formed in the late 1st millennium referring to a limited wooded area, perhaps on the hills above Burton? An alternative is that Dassett refers to the Warwickshire wolds on the western edge of a very large area of woodland on high ground, the wolds of Northamptonshire, Leicestershire and Nottinghamshire, which developed as a pastoral resource, according to a model proposed by Fox (Fox 1989; Watts 2004). A few names were formed when some land was not used for cultivation, such as the marshes at Fenny Compton and Butlers Marston. Ferns grew on the hill at Farnborough, and grazing land must have been a feature at Hardwick. More intense cultivation is recalled by the ‘kitchen garden’, ‘leek garden’ or ‘herb garden’ (-le-actu-n) at Wormleighton. Seven names incorporating personal names, and three using the word ‘king’ and ‘cniht’ suggest the early development of the possession of land, and probably lordship over the places so named (Ekwall 1936; Gover et al. 1936; Gelling 1974; 1982; Watts 2004). The place-names are pre-Conquest in origin, but they did not appear in written sources in most cases until Domesday Book in 1086. Five of the names appear in genuine charters, all but one dated after 950. The formation of the place-names is not closely datable, but most of them are likely to have been in existence by the 8th and 9th centuries. When the settlement of Southend appears in the records around

11

1300, the surrounding countryside had clearly been settled, exploited and named for many centuries. 1.5 MEDIEVAL PERIOD INVESTIGATIONS IN THE ENVIRONS OF BURTON DASSETT

The area around Burton Dassett has fgured large in past regional studies of medieval villages. This is because these were mainly concerned with the phenomenon of late medieval desertion and this area contains one of the highest concentrations of deserted villages in the county (Figure 1.2). The feld study of deserted villages, in this area as elsewhere, began with the task of identifying deserted sites and followed with recording and interpreting their various physical characteristics. Many of the sites in the Burton Dassett area were frst listed by Beresford (1945–46) and have been surveyed by Bond and others. These include Wormleighton, Watergall, Chapel Ascote and Kingston (Bond 1974, fgs 1, 3 and 8), Chesterton, Burton and Southend in Burton Dassett (Bond 1982, fgs 7.1–7.3) and Chadshunt (Dyer 1989b). At the beginning the aim was to understand the chronology of desertion and to produce explanations for it, but the more recent work has tended to show that both the pattern of desertion and its causes were more varied than frst thought (Dyer 1982; 1996, 128–130). More recent work has focused on the question of village origins, either as part of a general account (Dyer 1996) or by adopting a regional approach and studying an area to the south-west of Burton Dassett to exemplify the settlement pattern of the Feldon region (Hooke 1985). Along with the regional studies there have been investigations of individual parishes, of which the classic example remains the pioneering work of Thorpe at Wormleighton (Thorpe 1965; 1975), but work has also been done on Chesterton and Kingston (Hilton 1975), Hodnell, Chesterton, Burton Dassett (Bond 1982), Chadshunt (Dyer 1989b), Nether Itchington (Fowler 1986; Trollope 1987), Farnborough (Wass 2013), Westcote in Tysoe (Dyer 1991b) and Compton Verney (Hilton 1975; Dyer 2000). Most of this research has been documentary or feld survey. Only a handful of medieval excavations have taken place locally, and two of these were on sites other than villages: one on the motte and bailey castle at Ratley (Steane 1991) and the second on part of Warmington Priory which lay in the village just north of the church (excavation by Miss J Morris in 1957–58, Warwick Museum Cat A5361). Since 1990 there has been a number of planning-related evaluations and observations of proposed development sites in villages but very few have produced more than a few sherds of medieval pottery. One exception came in 1991 when trial trenching of the site of a proposed opera house at Compton Verney revealed extensive remains

BU RTON DA S SET T SOU T H E N D

12

of the village of Compton Murdak (Hart 1991; Dyer and Bond 1994; Dyer 2000). Sites in Fenny Compton, Kineton and Warmington have produced more limited remains. There have also been two excavations at Burton in Burton Dassett (see below). In Fenny Compton work in 1994 and 1997 in advance of a housing development on two plots east and west of Manor Cottages revealed medieval buildings fronting Northend Road (Eyre-Morgan 1994; Jones and Palmer 1998a; HER WA 7523). To the east, a stone-built house with a rubble yard to the rear was occupied from the later 13th to the early 15th century. To the west, another house with a possible outbuilding was probably occupied from the 12th/13th century to the early/mid-14th century. In 1995 part of a late medieval structure was excavated fronting Church Street (Oxford Archaeological Unit 1995; Booth 1996; HER WA 7258). In Warmington part of a 14th- to 16thcentury building and a possible fshpond containing 11th- to 13th-century pottery were recorded in 2011 (Yeates 2011; HER WA 13007–8); in Kineton, in Mill Lane in 2004, the remains were revealed of a probably 12th/13th-century building, boundary wall and road surfaces, the building abandoned by the 14th/15th century, and possibly re-occupied from the 15th/16th century until the 17th/18th century (Cook 2005; HER WA 10172–3); and, off St Peter’s Road, medieval yard surfaces and ditches were recorded in 2012 associated with 13th- to early 14th-century pottery (Gethin 2012; HER WA 9016). In Burton Dassett itself the earliest archaeological interest was focused on the parish church and upstanding remains of the chapel at Little Dassett (Cossins 1890; Chatwin 1952, 9). However, the basic multinuclear medieval settlement pattern, the presence of the market, and the story of the parish’s decline and depopulation were all known from quite an early date, but without being related to the surviving remains (Dugdale 1730, I 521–526; Chatwin 1935; VCH 1949, 69–77). The earthworks at Southend were frst identifed as a deserted settlement by Beresford who listed it as ‘the original Dassett’ (Beresford 1945–46, 88), but it was a combination of documentary research and feld survey by Bond that led to the frst detailed consideration of the earthworks in the parish and the defnite identifcation of the site at Little Dassett as Southend/Chipping Dassett (Bond 1974, 94, 108– 109; 1982, 153, 157–160, fgs 7.2 and 7.3). At about the same time, in 1973, an excavation was carried out in Burton in advance of an extension to the graveyard to the south of the parish church (Hunt 1972–73; 1973). The area investigated was 16m by 10.5m and contained a stone wall, 0.7m wide, probably a boundary wall, running north–south. To the west, a rough stone causeway lay across a boundary ditch. On the eastern side, within the property, a spread of ashy soil may well have been

associated with an oven or could have marked the position of a muck heap/midden. The medieval pottery found ranged from the 11th to the 13th century, which suggests that this property may have been abandoned at a relatively early date. The immediate post-medieval and later history of Burton Dassett has also been much studied because of the good survival of documentary evidence, particularly in the papers of the Temple family, formerly of Burton Dassett, later of Stowe, Buckinghamshire (Alcock 1977, 180–184; 1981; Westacott 1936, 96–111). The work on the activities of Peter Temple, the grazier who in the 1540s founded the family fortune and effectively reaped the benefts of the earlier enclosures, also considered the earlier history of the parish (Alcock 1981, 33–37). The result of all this previous research activity meant that the outlines of Burton Dassett’s medieval development were already understood before the excavation project began. This helped considerably both with the defnition of research aims and as a basis for pulling together the documentary evidence to provide a context for the excavated remains. Since 1988 there has been another excavation on the north-western fringe of Burton where the renewal of a water main in 1998 revealed a series of buildings with stone foundations dating to the 13th to 15th centuries extending over c25m of the pipeline (Coutts and Palmer 2000). Historical work on the economic activities of the late medieval inhabitants of Burton Dassett has also continued (Dyer 2012a). 1.6 SUMMARY OF GEOLOGY AND SOILS

The following summary is based upon the full accounts of the geology (by John Crossling) and soils (by Magdalen Snape) of Burton Dassett and its immediate environs, which may be found in the online archive, Sections 6.1 and 6.2. Geology The solid geology in the area around Burton Dassett is restricted in stratigraphic range from Upper Triassic to Middle Jurassic, but despite this a wide variety of rock types can be found (Edmonds et al. 1965; Old et al. 1987) (Figure 1.5). The oldest rocks, the Triassic Mercia Mudstone Group (formerly Keuper Marl), are situated in the extreme north-west of the area and form the lowlying land of the Avon Valley. The Penarth Group (formerly Rhaetic) of the Triassic lies directly upon these mudstones and forms a NE–SW trending escarpment that is closely followed by the route of the Fosse Way. The Langport member of the Penarth Group is a hard porcelaneous limestone which has been locally exploited as a building stone and for roofng material (stone slates) since the Roman period (Besterman nd).

I N T RODUC T ION A N D BAC KGROU N D

13

Figure 1.5 Geology of the Burton Dassett area (contains British Geological Survey © UKRI 2021. All Rights Reserved. https://webapps.bgs.ac.uk/data/maps/maps.cfc?method=viewRecord&mapId=9775 and https://webapps.bgs.ac.uk/data/maps/maps.cfc?method=viewRecord&mapId=9807) All strata to the south and east of this escarpment are of Jurassic age and get progressively younger towards the south-east. The early (lias) part of this sequence includes the Blue Lias Formation; hard limestones which have been used as a local building material since at least Roman times (Besterman nd) and recently have been the primary source of lime for the county’s

cement industry. They are the main building stone in the villages to the north and west of Burton Dassett, from Kineton to Bishop’s Itchington. Although they outcrop on the edge of Burton Dassett parish, they have not been much used there for building. Above the Blue Lias Formation the strata revert back into a continuous sequence of grey mudstones,

14

locally some 100m thick, known as the Charmouth Mudstone Formation. This makes up the bulk of the remaining surface strata in this area. It is typifed by rich gently undulating arable farmland as it weathers to a dark clay which in turn breaks down to give thick fertile soils. The hard Marlstone Rock Formation which forms the main cap-rock to well known local landmarks such as the Burton Dassett Hills and Edge Hill, is an iron-rich sandy limestone which is oolitic in part. This unit is only 4m thick but it extends over a wide area outside of Warwickshire. It is generally called an ironstone and these local outliers form the fringe of the continuous Oxfordshire/Northamptonshire Ironstone. These stones, which are blue green when fresh but weather (rust) to a rich ochreous brown, have been extensively worked as a building stone, forming the main building stone in the Banbury region, including Burton Dassett (Wood-Jones 1963, 3–4, fg. 1). They are known locally as ‘Hornton Stone’ after a village in North Oxfordshire which was the centre of the quarrying industry. There is a series of small disused quarries along the Burton Dassett Hills which presumably provided building stone for the local villages in the medieval period and later. Wood-Jones (1963, 3) implies that Burton Dassett was an important centre for quarrying, but there is no evidence of production for more than local use. The stone contains a high content of iron and on the Burton Dassett Hills was worked for iron for a period during the last century and up until just after the First World War (Tonks 1988, 232–241). Subsequently the price of iron dropped to such an extent that it was no longer viable to work this ore although substantial reserves remain. The extensive remains of these workings are centred at SP 39655202. The uppermost part of the solid geological sequence is the Northampton Sand Formation, a sandy oolitic limestone which represents the lowest unit in the area of the Inferior Oolite which are characterised elsewhere by the honey-coloured Cotswold Limestones, although where they occur here, on the highest points of the Burton Dassett Hills they have far higher concentrations of sand and iron and are only 2–3m thick. Overlying the solid geology is a sequence of drift deposits from the Quaternary period. The vast majority of these deposits are to be found in the north-west quarter of the area, due to the Jurassic escarpment forming a natural barrier frst to the glaciers and later to the rivers. Soils The medieval settlement of Southend and its associated feld system which extended westwards from the settlement, are sited in the lowland area dominated by soils of the Denchworth Association, extending over a wide swathe of land in the area

BU RTON DA S SET T SOU T H E N D

from Ladbroke to Wormleighton to Tysoe. These soils coincide with the Charmouth Mudstone Formation clays, in which the major soil series are the clayey, non-alluvial, seasonally waterlogged, slowly permeable soils of the Denchworth and Lawford series. Under modern conditions the soils will yield moderately good crops of grass or cereals if carefully managed and drained. Topsoils are slow to dry and prone to structural damage when wet and for this reason cultivation is best done in autumn as the land is rarely dry enough to work in spring. Timing of cultivation is vital as, when wet, the soil is vulnerable to smearing and compaction; cloddy seedbeds and ploughpans can result which further impede drainage, lessen seed germination and restrict root growth. When dry, the soils can be hard and diffcult to cultivate. Denchworth soils poach easily especially in the troughs of ridge and furrow. This surface wetness and a weak soil bearing strength limit stocking density and shorten the length of the grazing season (Williamson 2003, 145). In contrast, the Burton Dassett Hills to the southeast of the parish are characterised by soils of the Banbury Association. These soils, formed on steep slopes, are subject to active erosion and soil movement particularly where vegetation is sparse. The Burton Dassett Hills themselves are under permanent pasture due to the limiting factors of steep slopes, stoniness and shallow soils. 1.7 BURTON DASSETT: ITS ORIGINS AND DEVELOPMENT

by Christopher Dyer This survey of Burton Dassett’s medieval development is based on fairly meagre documentary evidence, consisting mainly of Domesday Book, the survey in the Hundred Rolls of 1279–80, and deeds in fve collections, of which the most important relates to the former property of the Catesby family, and is now in the National Archives. The archives of the lords of the main manor of Burton Dassett, the barons of Sudeley in Gloucestershire, have not survived. The results are bound to be tentative because of the thinness of the evidence. Four themes will be examined: the form and distribution of the medieval settlements and especially Southend; the agrarian economy; the commercial development of Southend (called Chipping Dassett for a time); and the 15thcentury decline. Settlements The earliest evidence comes from the placenames. ‘Dassett’ appears to be of British (that is, pre-English) derivation although the meaning is not certain (Gelling 1974, 75). It is a pointer to the existence of a district, comparable with a number of territories in the West Midlands, such

I N T RODUC T ION A N D BAC KGROU N D

as Ismere or the Stoppingas, named in charters of the 8th and 9th centuries. A reasonable hypothesis, based on analogy with better documented places, would be that at some time between AD 600 and 900 ‘Dassett’ was a single large area occupied by a group of people, or forming a unit of government, including at least the present parish of Burton Dassett (4975 acres, 2015 ha), Avon Dassett (1,435 acres, 588 ha), and probably Radway (part of which was linked to Burton Dassett in 1279–80) (VCH 1949, 67, 69, 143; SCLA, CR136/C709a). Other neighbouring parishes, such as Fenny Compton, may well have formed part of ‘Dassett’, and on topographical grounds it would be tempting to include Gaydon and Chadshunt. The district was split up into smaller units in the period before the Norman Conquest, resulting in the pattern of manors in existence at the time of Domesday Book. In 1086 (Burton) Dassett was described separately from (Avon) Dassett, and the statement that 3 knights had 12 villeins and 3 ploughs (Rec Comm 1783, fol 244a) implies that sub-manors had begun to develop, probably the precursors of the holdings of the Knights Templar, the Ardern family and the de Knightcotes, which are more fully recorded in the 12th and 13th centuries. The area supported a population as early as the 6th century, judging from the burials that have been found on the Burton Dassett hills. The cemetery there at Mount Pleasant need not have been immediately adjacent to any settlement. A hint of the former existence of a small farm or hamlet comes from the name of a furlong in the felds of Knightcote called Dunstall, meaning ‘the site of a settlement’, recorded in the 13th century, after its desertion (SCLA, DR98/151). Only extensive archaeological research will tell us when the inhabitants of the manor and parish of Great Dassett (as Burton Dassett was known from 1223 onwards) became grouped into the fve settlements recorded by the 13th century (Figure 1.6); in neighbouring Northamptonshire it has been argued that nucleation took place between the 9th and 12th centuries (Williamson et al. 2013). The likely site of early nucleation (say in the 9th to 10th centuries) would be at Burton, near the church. Here were two points of attraction for settlement: the church itself, which may well have originated well before AD 900 as a minster church serving the scattered inhabitants of the whole Dassett district, and the burh or fortifcation of which the memory is preserved in the place-name Burton. This could refer to a now lost Iron Age hillfort on the church site, or the nearby site on Gredenton Hill in Fenny Compton parish. Alternatively, the burh could have been a defended aristocratic residence or some larger defensive structure of the kind which has been suggested as the origin of the ‘Burtons’ which occupy strategic positions (such as Burton-on-Trent and Bourton-on-the-Water) and may belong to the 9th century, pre-dating the major strongholds

15

such as Stafford, Tamworth and Warwick (Gelling 1981, 16–19). Hardwick’s name implies that it began as an outlying secondary settlement, serving as a base for the management (herding) of animals near the lowlying meadows on the west side of Great Dassett. Knightcote must have been the home of one or more of the Domesday knights, and the main lowland settlement was divided into a Northend and a Southend (on the place-names see Gover et al. 1936, 267–269). It is sometimes said that nucleated settlements were organised by lords, and this is indeed supported by the Knightcote name. This could suggest that a member of the lesser aristocracy was given a holding where he then granted small tenements to peasants and created a village around his house. However, Northend and Southend, both of which belonged to the main manor of Great Dassett, do not seem to represent settlements grouped near manor houses, as the lord’s establishment presumably lay before 1200, as it did later, on the hill-slope at Burton, beside the church. The growth of Northend and Southend must refect the desire of the peasants to be located at a convenient site in relation to their felds and the road system, and the needs of the inhabitants must be taken into account in explaining the origins of the villages. The formation of nucleated villages seems to have been closely linked with the development of open-feld systems. Burton had its own felds, but the other villages were involved in more complex arrangements. They belonged to the ‘bipolar’ type, found elsewhere in the west midland region, in which two villages were located at opposite ends of a single territory, and shared the feld system that lay between them (Dyer 1987, 165–181). So, the northern end of Great Dassett was occupied by a two-feld system, Eastfeld and Westfeld, which was cultivated from Northend and Knightcote, while to the west lay a feld system (Northfeld and Southfeld) used by the inhabitants of Hardwick and Southend (Figure 1.6). Meadows were shared in a similar way. We do not know if this dualism arose from the splitting of existing villages or from the simultaneous growth of the twin settlements (Dodgshon 1980, 108–150). Clearly the economic arrangements of the peasant communities operated independently from the interests of the lords because the sharing of the felds united villages of different lords (Southend belonged to the Sudeley family, the lords of the main manor of Burton, and Hardwick to the Templars in the 12th and 13th centuries, for example). Tenants of adjacent, almost continuous settlements who belonged to the same lord (the Sudeley family) at Northend and Southend were in fact disunited in their agrarian life, with separate attachments to distinct feld systems. Again, we seem to be observing the effects of peasant communities organizing themselves, rather than submitting to seigneurial direction.

16

BU RTON DA S SET T SOU T H E N D

Figure 1.6 Burton Dassett in c1300. The settlements and approximate boundaries of the townships are shown, with the known names of open felds. The ridge and furrow that survived in the 20th century has been added from air photographs. The transcriptions of air photographs were made by staff of Warwickshire Archaeology from photographs in the Warwickshire Historic Environment Record

In the early 14th century the tax lists refect the settlement pattern by listing the fve settlements (and part of Radway) as members of Dassett. In both 1327 and 1332 Southend appears as the largest settlement, with 24 taxpayers, compared with 11 to 15 in the other Dassett villages (Carter 1902, 17; 1926, 25–26; Glasscock 1975, 321). The 14th-century documents tell us a good deal about the layout of Southend. A part of the settlement called Newland (original spelling, Newlond), was frst mentioned in 1321, but it must have originated earlier, perhaps going back to the mid-13th century (TNA, AD E40/ 7014). A deed dated to 1503 refers to a cottage ‘in the street called Newlond’ (SCLA, ER1/66/534). The deeds record messuages and tofts laid out with well defned boundaries in at least one

row. In one case a curtilage was said in 1329 to be surrounded with hedges and ditches (whether a live hedge or fence is not specifed) (TNA, E40/A8114). Some plots of land seem to have been empty at the time of their transfer, while others were occupied with a building or group of buildings. One contained only a barn (TNA, E40/A7657). By an agreement of 1406 a tenant was to build a house of four bays and a chamber annexed of three bays (TNA, E40/ A10790), probably meaning that the main house with its hall was provided with a cross wing. Around the houses were gardens and orchards. Access to the back of the messuages was provided by lanes, one of which was 8ft (c2.4m) wide. One, called Dog Lane, was mentioned in 1503 running behind Newland (TNA, E40/A11116; SCLA, ER1/66/534). Some

I N T RODUC T ION A N D BAC KGROU N D

deeds refer to properties lying near to the market place, but they are insuffciently detailed to indicate whether this consisted of a wide street or of some larger open space. The impression that the documents give of 14th-century Southend is of a compact and densely occupied settlement with a number of streets and lanes. It is assumed (without conclusive proof ) that the street running westward from the chapel in the direction of Kineton that was the focus of the excavation was called Newland in the Middle Ages. The chapel (see Section 2.5) is the only building from the settlement still standing. It is largely undocumented, so that the evidence for its late 13thcentury origins comes from the structure itself. Its survival suggests that it received some maintenance over the centuries; it later became a barn. Permission was given for services to be held in 1380 for three years, but it must have been in use for services both before and after that date (VCH 1949, 69). The name Clerk appears among the tenants in the survey of 1279–80, is found in other documents, and may refer to a clergyman attached to the chapel. By analogy with the many other chapels founded in settlements that lacked their own parish church, the chapel could have been funded by local people who felt a need for religious service and a focus for their sense of community. The beginnings of urban development at Chipping Dassett in the late 13th century may have provided a stimulus for the building of a chapel, which occupied a site at a point where signifcant roads met. The dedication to St James coincides with the feast of St James chosen for Burton’s annual fair in 1267. Perhaps there was an earlier chapel of St James which inspired the choice of the day for the fair, or the fair was held on a day (25 July) which might attract those selling newly sheared wool, and the chapel’s dedication came afterwards? The agrarian economy The Great Dassett villages shared with the rest of the Feldon district of Warwickshire a distinctive agrarian landscape of extensive arable f ields. Domesday records no woodland at Dassett, and the whole district lacked abundant trees. The only identifed woodland in the parish is a grove near the church mentioned in 1545 (Wager 1998, 209, cited from Alcock 1981, 38, but incorrectly dated to 1497). This extended to no more than 20 acres. Some timber and wood probably came to Great Dassett from Chilvers Coton in the Arden, which belonged to the same tenant-in-chief in 1066 and 1086 and continued to be associated with Dassett in the Sudeley barony. Such long-distance connections are well known in Warwickshire but are often assumed to have only been important before the Norman Conquest. However, in the case of the links between Brailes and Tanworth-in-Arden, and between Wasperton and Packwood, timber was still

17

being carried between these manors in the decades around 1400 (BAH, 167904; WRO, CR 2238/1). The largest arable holding in Great Dassett, the demesne of the Sudeley manor, seems to have grown in size from about 100 acres in 1086 (the area likely to have been cultivated by a Domesday plough-team) to about 500 acres in 1279–80, and then to have diminished to about 300 acres in 1367 (though all of these fgures can only be approximations) (Rec Comm 1783, fol 244a; John 1992, 224; SCLA, ER1/66/537). In 1443 a survey records a remarkably large arable demesne of 525 acres (Cal IPM 2010, 26, 62–63). The tenants in 1086 worked with a total of 29 ploughs. They would have each cultivated a smaller amount than a demesne plough, but they must represent an arable area well in excess of a thousand acres. In 1279–80 the tenants held in total about 60½ yardlands ( John 1992, 224–229). The size of this standard unit is not known, but it is just possible that it was as large as 62–64 acres (about 25 ha), because documents mention halfyardlands of 31 and 32 acres, and, in the late 18th century, the Northend and Knightcote yardlands can be reckoned to contain about 59 acres (TNA, E40/6229; SCLA, DR10/2368; WRO, QS75/43) These were very large by the standard of the district: yardlands were as small as 20 acres at neighbouring Chadshunt and Gaydon (SRO, D1734/J2268, fols 8–9), and generally in the Midlands they averaged 30 acres. If the yardland was as large as c60 acres, tenant arable took up a very large area (c3,630 acres), much more than the lords’ demesnes. As a whole, the arable land, including tenant land and the demesnes of the Sudeleys, together with the smaller area cultivated by the other lords, accounted for four-ffths or more of the land in Great Dassett. An agreement for the sale of tithes by Arbury Priory, which held the rectory, suggests that in 1300 about 400 acres (about 160 ha) in the parish were planted with wheat (SCLA, ER1/66/541). In better documented manors in south Warwickshire, wheat usually accounted for about a third of the planted area; the other crops were barley or dredge (mixed barley and oats) and peas, with lesser quantities of oats (e.g. TNA, SC6 1039/14). The amount of grassland in Great Dassett was severely limited, judging from holdings in the 13th century in which less than 4% of the land was described as meadow (Stokes and Wellstood 1932, 179). An important area of permanent pasture lay on the top of the Burton Dassett Hills immediately to the east of Southend, which shows on air photographs as an area without ridge and furrow (see Figure 1.6). Much of the grazing land lay in the open felds after the harvest and during the period of fallow. Pasture was carefully stinted; for example, a 13thcentury deed linked pasture for one aver (draught animal) with a holding of 8 acres of arable (SCLA, DR98/151). Peasant sheep do not appear in the

BU RTON DA S SET T SOU T H E N D

18 Above 1 yard-land

1 yardland

Between 1 yardland and ½ yardland

½ yardland

¼ yardland

Small-holdings

Total

6 (5%)

11 (9 %)

9 (7%)

57 (45%)

33 (26%)

11 (9%)

126 (100%)

Figure 1.7 Size of tenant holdings at Great Dassett in 1279–80

surviving Dassett documents but can be assumed from analogy with other Feldon villages to have grazed the fallow, providing manure for the felds and an important source of cash from their wool. It would be reasonable to suppose that a Dassett half-yardlander would have had two plough-oxen, a horse, a cow or two, two younger beasts, a pig and a dozen sheep (cf. Slavin 2015, 8–9). The Domesday entry for Dassett mentions 75 people, whereas in the Hundred Rolls of 1279–80 126 tenants are listed. The difference in numbers between 1086 and 1279–80 could have been the result of omissions in 1086, but the apparent increase could well refect a growth in population, which encouraged the sub-division of the full yardlands and half-yardlands which most Dassett tenants presumably held at the time of Domesday (there were 58 villeins compared with only nine small-holding bordars). Most tenants held half- or quarter-yardlands in 1279–80 (Figure 1.7) (John 1992, 224–229). In the case of Hardwick, a reduction in the size of holdings had apparently taken place by 1185, when 6 of the 11 tenants had quarter-yardlands (Lees 1935, 28). In the late 13th and early 14th centuries, allowing for omission and evasion in the tax assessments, and assigning a family of fve to each taxpayer or tenant, Great Dassett’s population can be estimated at 700–800, well in excess of the early 20th-century fgures (445–486) and suggesting a density of people of 100 per square mile (39 per km 2 ), which is high by the standard of either medieval or modern rural Europe. It is normally assumed that Feldon villages by about 1300 had expanded to their limits. The land had been extensively colonised, and any further growth in cultivation might have endangered the already delicate balance between arable and pasture. Commercial development Great Dassett was one of the four manors of the Sudeley family in the decades around 1300. They were minor barons, based at Sudeley near Winchcombe in Gloucestershire, and Dassett, which was worth well in excess of £40 per annum, was for them an important source of revenue (Sanders 1960, 85–86; SCLA, ER1/66/537–9; CIPM 2, no. 347; 8, no. 30). They evidently expected it to produce an income in cash, and the customary tenants (who amounted to about 60% of all tenants) were burdened with substantial money rents, such as 15s for a halfyardland, rather than the usual combination of rents

and labour service. Lords such as these, seeking to support the status of a barony on relatively modest landed endowments, must have searched for ways to expand their incomes. Their need is suggested by the indebtedness in the 1290s of John de Sudeley (1280–1336). His father, Bartholomew de Sudeley (1248–1280) had obtained a charter in 1267 by which he could hold a weekly market and an annual fair for three days round St James’s day (25 July) at Dassett (Letters 2003, 2, 359). Bartholomew and his advisers had noted Dassett’s potential for commercial development. The network of markets shows a distinct gap between Kineton (Warwickshire), Chipping Warden (Northamptonshire) and Banbury (Oxfordshire) which the Dassett market would have flled (Figure 1.8). Friday was chosen as the market day so as to avoid clashes with nearby rivals such as Kineton (Tuesday), Southam (Wednesday, later Monday), Bishop’s Itchington (Wednesday), Chipping Warden (Tuesday, later Monday) and Banbury (Thursday) (Barker 1986, 161–175; Postles 1987, 14–26; Goodfellow 1987–88, 305–323; Letters 2003, 1, 260; 2, 280, 358, 361, 363). Indeed, it may have been hoped that the markets would complement one another and traders would move from one market to another on successive days. The market would have been intended to serve an immediate hinterland in Great Dassett and the surrounding villages. It was sited, not near the church and manor house at Burton, but at Southend near the centre of the manor, and near the main road (the Portway, as it was called in the 16th century) from Warwick to Banbury, in the hope of attracting passing trade and perhaps even of developing regular long-distance commerce with these towns. If the Sudeleys followed the example of other lords founding markets, they may have attempted to re-route the main road so as to bring traffc into Southend, or at least to provide an easy diversion. The point chosen for the new market and its associated settlement was also served by an east–west road to Kineton and into the Avon valley; such new commercial ventures often lay at a meeting of thoroughfares. A market charter is a valuable piece of evidence in reconstructing the history of a commercial settlement because it gives a defnite date. Chartered markets were, however, formal legal institutions, connected to changes in commercial life, but not controlling them. Traders and artisans were living at Burton Dassett around the time of the granting of

I N T RODUC T ION A N D BAC KGROU N D

19

Figure 1.8 Markets and fairs in the Burton Dassett area (data from Letters 2003)

the charter, judging from the occupational surnames recorded in 1279–80: these were three people called Marchaund (meaning Merchant) and one called Mercer (a dealer in cloth, especially linen), together with a clerk, lockier (lock-maker), two marshals (smiths) and a tailor. They could have set up their businesses in the 13 years that had elapsed since the grant of the market charter, but they were holders of land, in the case of Thomas le Mercer in large quantity: some of them may well have been living in Burton Dassett before 1267. Marchaund and Mercer are quite unusual names in a rural setting, and their presence points to developing commercial activity. The documents could thus be interpreted as

being compatible with the archaeological dating that suggests that a planned settlement with a commercial dimension may have developed before the market charter was granted. The lords of the manor may not have been the sole initiators of commercial growth but, having observed the ‘merchants’ at work, realised that their manor had a potential for development and invested in a charter. An advantage to them is indicated by the cash rents (15s for a half-yardland) being paid by tenants according to the 1279–80 survey. Tenants with access to a local market, perhaps informal at frst, and then becoming an offcial institution, found it easier to sell their produce and pay their dues to the lord in money.

20

They would also have used markets as consumers, to buy goods such as metalwork and textiles. A further indication of the early chronology of commerce at Burton comes from the place-name Chipping Dassett. The name refers to the market and one suspects that it would take some time for the new name to become established. Its frst known use is in a deed of 1295, which records a purchase of an unspecifed quantity of land for 10 marks (£6 13s 4d) by William Reymund and his wife Emma, who were recorded as holding land in Coventry in 1295 (WRO, N3/560; Willoughby parish deeds). The document was witnessed in Coventry by people who did not live at Burton or its vicinity, apart from William Bele who appears in the 1279–80 survey. A connection with the largest and most rapidly growing city in the region reinforces the idea that the settlement previously known as Southend had become a commercial centre with a wide reputation. Well informed contemporaries saw Dassett as having an urban character, judging from the payment in 1300 by the burgesses of Leicester of cash (perhaps as a bribe) to juries in Warwickshire at Coventry, Warwick, Stratford, Kineton, Alcester, Dassett, Brailes and Henley (Bateson 1899, 233). All the places mentioned except Dassett were boroughs, and they seem to be arranged roughly in ranking order, in which case Dassett was thought to lie between the well established market town of Alcester and the more ambiguous settlement with burgage tenements at Brailes (BAH, 167901). The street name Newland may also be indicative of urban status, because it is found in a number of boroughs, including nearby Southam (Coss and Lancaster Lewis 2013, 577, 579). The Sudeleys attempted to prof it from the Dassett market, not just by collecting the tolls to which the charter of 1267 entitled them, but also by encouraging the growth of the settlement around the market place. Two deeds of the late 13th century, one of which is datable to 1280–90, shows John de Sudeley granting pieces of land that look suspiciously like the tenements found in boroughs (TNA, E40/A10890; SCLA DR 37/2086). They were small plots, unattached to arable in the open felds, and in one case the tenant was to pay the rent of 12d per annum which was so often the sum paid by burgesses. The plots were apparently created after 1279–80, because they are not mentioned in the survey in the Hundred Rolls, when Great Dassett seems strangely devoid of cottagers and smallholders who often indicate the growth of trades and crafts, as people with only an acre or two of land had to fnd additional sources of income (Hilton 1975, 127). The small plots became numerous, judging from the number of deeds, a third of the total in the period 1280–1440, which conveyed messuages, tofts, curtilages or other small parcels, many of them in Southend. Their tenants were freeholders, some with two adjoining plots, and others with as

BU RTON DA S SET T SOU T H E N D

little as a half-messuage. The deeds show a wide range of techniques of conveyancing, including arrangements for mortgages (e.g. TNA, E40/ A6390). Some came into the hands of substantial absentees, like William Clerk of Gaydon, who sublet his plots and houses (e.g. TNA, E40/ A6621). In 1346 another tenant had no less than 11 subtenants (SCLA, DR98/162). Can the settlement around the market at Southend be described as ‘urban’? A major requirement for the defnition of a town is that it should have a dense concentration of people pursuing a variety of non-agricultural activities. In the absence of an occupational census, we can obtain hints of crafts and trades from the surnames which were adopted in the 13th century. At Great Dassett in 1279–80 the tenants called Lokier, Marchant, Mercer and Tailor have been mentioned. Unfortunately, as surnames became hereditary they prove less reliable as a guide to occupations; it is worth noting, though, the appearance in deeds and tax lists of 1280–1350 of a Cobbler, Cook, Couper, Potter, Skinner and Smith. One tenement was in the hands of a miller in the 14th century. He may have used the ‘Stonemill’ which stood on the hills, which has been identifed controversially as a rare tower mill rather than the commonplace post-mill. As the windmill was in a ruinous state by 1369 he might have worked a horsemill at Southend (TNA, E40/A8114, A7657; SCLA, ER1/66/537; Langdon and Watts 2005, 709). In 1360 a tenement was described as a ‘messuage with a bakehouse’, which gives the subsidiary structure such prominence that it may imply bread-making on a commercial scale (SCLA, DR98/179). There is no evidence that the people of Southend were unusually prosperous, and indeed in 1327 and 1332 they paid signifcantly less in taxes than their neighbours at Northend. This does not disprove their urban character, because many well documented small towns contained concentrations of relatively poor traders, especially those dealing in foodstuffs (Hilton 1982, 3–15). The suggestion has been made above that the chapel at Southend was connected to a growth of a community identity, though such chapels are also found in rural communities, such as Hardwick and Knightcote in Great Dassett (Chatwin 1952, 9; SCLA, DR98/150). A new town can often be shown to have developed strong links with its rural hinterland, and to have drawn migrants to build up its population, as is well attested at Stratford-uponAvon between 1196 and 1252 (Carus-Wilson 1965, 46–63). There is a lack of such evidence at Dassett, perhaps because of the late date of its growth, after the fossilisation of surnames. The names formed from place-names tend to refect long-distance moves, for example from Packington (in the Arden) in 1279–80, and from such places as Fillongley, Yardley, Lawford and Ufton later.

I N T RODUC T ION A N D BAC KGROU N D

There is much evidence of connections between Dassett and other manors on the Sudeley estate. The Dastins, a prominent Dassett family, came from the Sudeley area of Gloucestershire, as did a de Guiting who held land at Dassett in 1348 (Royce 1892, vol. 1, 79, 80, 336; SCLA, DR98/169). On 14 June 1348, a deed was witnessed at Dassett by men from Grete and Winchcombe, both adjoining Sudeley (SCLA, DR98/168). Similarly, there were links with the Griff and Chilvers Coton neighbourhood, as revealed by the tenure of land in Northend and Knightcote by a Nuneaton man in 1338 (SCLA, DR98/158, 159, 160). The connections with the Sudeley estate at Griff were reinforced by Arbury Priory’s control of the Dassett rectory, and the Templars’ tenure of manors at Hardwick and Chilvers Coton. Both Arbury Priory and the Templars received benefactions from the Sudeley family. Perhaps an original exchange of goods between Griff and Dassett, such as timber and wood from Arden for Feldon grain, in the long run took on a more commercial character, hence the carriage of pottery, coal and roofng slates from the Nuneaton area to Dassett as shown by the Southend excavations. Perhaps also trade links involving livestock, especially sheep, developed with the Winchcombe district. None of this proves that ‘Chipping Dassett’ was more than a market village with some long-distance connections. As late as 1480–81 the lord of the manor was drawing a small revenue from a ‘booth hall’ and repairing an adjacent ‘shop’ which was probably a seld from which traders would operate. Does this mean that the town was still active? Or was the lord attempting to breathe new life into a moribund trading centre? Chipping Dassett’s tenants (unlike those of Chipping Campden and Chipping Norton) did not hold their messuages and plots by burgage tenure, which prevents us describing the settlement as a borough, though other towns had a long and prosperous life even when the townspeople held their houses on a simple free tenure: in Warwickshire the town of Rugby provides an example. Dassett’s market and commercial community developed relatively late, so that it was not able, like Stratfordupon-Avon, to establish its trading position during the surge of commercial growth around 1200. Small market centres were vulnerable at the end of the Middle Ages to the concentration of commerce in well founded towns. A number of Dassett people joined the Holy Cross Guild at Stratford between 1442 and 1508, and that town may well have been one of the places they preferred for their trading (Macdonald 2007, 177, 197, 213, 259, 263–264, 356–357, 404). Perhaps Dassett can be regarded as a town in some of its phases, notably in the early 14th century. Settlements have been identifed in Warwickshire, some of them boroughs, which experienced episodes of urban activity, at Bidford,

21

Brailes, Bretford, Brinklow and Kineton (Dyer 2012b; 2018; Goddard 2011, 3–31). Decline in the 15th century The rural community as well as the town may have been experiencing some decay in the 15th century. In 1415–16, a court roll for Hardwick (WRO, CR1122/1) includes a reference to a tenant taking over three messuages, which might have led to the removal or decline of two sets of buildings, and another tenant was exempted from repairs to his houses and buildings. It is also noticeable that only three tenants are named in the roll, apart from one nativus (serf ) who had fed the manor. In 1435, an inquisition describes the lands of Richard Verney (SCLA, DR98/188), including a toft in Chipping Dassett, which adjoined another toft. The use of the word ‘toft’ often means that the cluster of buildings which made up a messuage—at least a house and a barn—no longer existed. The document implies that a house site and that of a neighbour were vacant, and this is confrmed by a lease of 1434 for the same holding also describing it as a toft (SCLA, DR98/189). However, the next lease (SCLA, DR98/190), of 1488, describes it as a messuage, as if the house had been rebuilt. There are enough hints here for building plots being abandoned in the early 15th century, as confrmed by the archaeological evidence, to show that there was some justifcation for the argument mounted by the lord of the manor in the early 16th century that before 1485 there had been a shortage of tenants (Dyer 2012a, 29–30). The lord’s explanations ft with the relatively frequent experience, in Warwickshire and in many other parts of England, of inhabitants leaving villages which were afficted by a decline in cultivation and with buildings falling into ruin (Dyer 2010). 1.8 LATER DOCUMENTARY EVIDENCE

by Nat Alcock Later ownership From the 14th century onwards, the main Burton Dassett manor passed through a series of hands (VCH 1949, 69–72; Alcock 1981, 29–30). The last of the Sudeleys died in 1367, when they were succeeded by the Boteler family. On the death without issue in 1473 of Sir Ralph Boteler (created Baron Sudeley in 1441), the ownership was divided between his nephews Sir John Norbury and William Belknap; they partitioned their inheritance in 1496 when William’s son (Sir) Edward Belknap received Burton Dassett (BL Add Ch 5684). Sir Edward died in 1521, leaving the manor to three of his sisters, the wives respectively of Sir Robert Wootton, Sir Philip Cooke and Gerald Dannett. The Dannett family sold their third to Peter Temple in 1557–60, and the undivided

22

manor was partitioned between the co-owners in 1576 (proposed in 1567). The Temple family (by then of Stowe House, Buckinghamshire) acquired the Cooke share of the manor in 1595–1608, and retained it until 1828, when it was bought by Lord Willoughby de Broke of Compton Verney. Enclosure The earliest frm statement that Chipping Dassett had been enclosed comes from the Domesday of Inclosures of 1517 (Leadam 1897, 424–426), which charged Sir Edward Belknap with the enclosure of 360 acres and the destruction of 12 houses in (Chipping) Dassett and Burton; in addition, Sir Edward Raleigh (a minor landowner there) had enclosed 22½ acres, destroying one other house. As is typical of these records, the extent of enclosure is understated. In fact, a 16th-century memorandum confrms what the parish topography suggests, that the enclosure was on a much larger scale, covering 1960 acres and ‘34 ploughs’ (which should correspond to at least that many houses), in Burton, Dassett and Hardwick (see Alcock 1977, 180–184 for both this memorandum and Belknap’s response). Belknap provided two defences to the charges against him; in one of these he suggested both a chronology and an explanation for it. He claimed that it had taken place during the reign of Edward IV, when his predecessor as lord of the manor, Lord Sudeley, was in prison, and when his (Edward’s) lands were held in wardship. During this period, frst the ‘howsing fell in great ruyn and decay’, and secondly the tenant of the demesne, one Roger Heritage, converted much of the manor to pasture. His further claim is interesting (and possibly true), that when the manor was frst enclosed there was a scarcity of pasture, but tenants for arable land were hard to fnd. He also stated that Dassett was still very populous, with 30 households and more than 160 adults, and still retained large areas of arable. In this, he was being misleading, as his statements are correct in relation to the parish of Burton Dassett, where Northend and Knightcote remained in open feld until 1771, with well populated villages then as now. His enclosures involved, however, only that part of the parish comprising the manor and settlements of Burton and Chipping Dassett. Furthermore both Lord Sudeley (Ralph Boteler) and Roger Heritage were conveniently dead and could be blamed without fear of any response. The picture painted by Sir Edward Belknap is that of a classic enclosure following decay and depopulation caused perhaps by epidemics and decreasing land fertility. As such, it appears highly plausible, until the evidence of certain sources contemporary with the enclosure is considered. These allow us to specify its date and they also throw a very different light on its effects on the parish.

BU RTON DA S SET T SOU T H E N D

The moment of enclosure A major phase in the enclosure of Burton Dassett seems to have occurred in 1497. Roger Heritage, the farmer of the demesne, had died in 1495, and his son John inherited his lease and the family farm. In 1496 Edward Belknap took over the lordship of Burton Dassett manor, and the two new arrivals devised a plan for changes that would beneft both of them. On 1 May 1497 Belknap bought John Heritage’s freeholding of a messuage and 1¾ yardlands in ‘Burton and Dassett’, presumably the family’s own holding. Dassett probably referred to Southend (NRO, TS box 4/1a/1; Dyer 2012a, 34). This amounted to at least 52 acres, and perhaps 106 acres, depending on the size of the yardland. This purchase was necessary because the strips of land of which this holding consisted were intermingled with the other parcels of land in the open felds, and a free tenant could insist on his property rights and prevent enclosure. Two days later, Belknap leased back to Heritage for 41 years ‘all the closis of the towneship of Byrton’ with an adjoining feld called Medull Feld, reserving to Belknap ‘the court plasse with all other hosyng of the said towne’ with liberty to dwell in the court place (the manor house) or to pull it down, and ‘lykewyse in all other manner of howsyng’ with access to the housing ‘in building and carrying away’ (HEH, ST Deeds List 2, box 36). This means that John Heritage became the leasehold tenant of a very large area of land, much of it former open-feld arable which had been enclosed with hedges and ditches, while Belknap as lord retained the manor house and all the buildings of the village, which he could demolish. He could remove the materials (timber, stone and slates), most probably for sale. John had moved to Moreton in Marsh (Dyer 2012a, 34), and no longer needed the family house at Burton, while his sheep would have grazed the felds he leased. Other signifcant conditions in this lease were: ‘if it happen any freholder too comen or sowe within the said pastures’ (meaning if a freeman wished to exercise common rights or even cultivate), John would receive a rebate on his rent. This refers to the problems that might arise if a free tenant insisted on retaining rights in the open felds, which Belknap had in the case of Heritage’s holding prevented by buying the land. John Heritage was to bear ‘all reparation of hedgyng and dyking of all manner inclosures’, and at the end of the lease to leave ‘the closys suffciently mounded’ (that is, securely surrounded with ditches, banks and hedges). It is absolutely clear that the township of Burton (by the church) was forcibly depopulated and enclosed in the summer of 1497. The evidence does not cover Chipping Dassett directly (for which 1517 is the terminus post quem), but it is almost certain that its enclosure took place at the same date, following the pattern of simultaneous enclosure and depopulation recorded at Burton. The 1497

I N T RODUC T ION A N D BAC KGROU N D

lease mentions other closes being leased, notably one to John Spencer (probably the Hills), and the deed relates to land in both Burton and Chipping Dassett, which must by then have had a joint feld system. It is hard to see how its open-feld agriculture could have persisted following the enclosure of at least two of its felds, while the agrarian re-organisation would have removed the livelihood of most of its inhabitants. One document does suggest a different picture, a deed dated 10 May 1503, endorsed ‘the church dede’, which is apparently a re-enfeoffment of the church endowment (WRO CR1886/cup 4/ D1; copies in SCLA ER1/6 and DR41/94). The date of this deed is entirely unambiguous in the original, but the presence among the grantees of John Spencer of Hodenhull (Hodnell) is suspicious; he died in 1496/7, and though his nephew then moved to Hodnell, the latter should have been named as Sir John Spencer. The deed describes three cottages and crofts in Chipping Dassett and 16 acres lying dispersed in the felds of Dassett and Hardwick, none of which should have existed at this date if the present analysis is correct. It is most likely that this deed simply repeated an earlier description, and it was probably drawn up in this form to record church rights that were being ignored by the enclosure. Post-enclosure evidence The frst direct mention of Town Field (containing the site of Chipping Dassett) is in a rental of the manor (HEH, ST37), dated 1540, but in several places referring to Sir Edward Belknap (d. 1521) in the present tense, and containing other anachronisms suggesting that it was originally compiled before 1520. Town Field was leased to Henry Makepeace and Robert Broke for £13 6s 8d a year. Henry Makepeace also held a messuage and little close (£2) and another(?) Makepeace held pasture for £23 a year. In a lease of 8 February 1539 (HEH, ST L9E1, register of leases, Temple Land, Bucks), William Brokes, yeoman, son of Robert Brokes, deceased, received those messuages, etc, in Chipping Dassett lately occupied by his father, for 11 years (rent £6 13s 4d); this was probably half of the previous joint holding. The other half may be represented by a lease of 28 April 1540 to John Petiver of the tenement which Henry Makepeace held, for a rent of £12. Both these leases included covenants against tillage. The frst detailed description of the felds comes from a partition of the manor drawn up in 1567 (implemented in 1577). Town Field and the adjoining land was assigned to Peter Temple. Its area was given as 210 acres 18 perches, in the tenure of William Brookes and Thomas Makepeace. A 2-acre close and messuage occupied by William Brookes lay east of this, while Makepeace’s house was in a 3-acre close containing a pond (perhaps the later Little Dassett Farm). Thomas Burbery, who held a large close west of the road (‘Burbery’s Field’) had

23

his messuage, a meadow, and Dovehouse Close (11 acres together) south of Town Field. There was also a ‘cottage or sheppardes house’ in a 2-acre close, used by the shepherds of the tenants of the enormous Traversfeld. The procedure used for enclosing the estate is illustrated by the 1497 lease already discussed; it seems to have been of the simplest: just assigning each great feld to a separate tenant. The same may well have happened in Chipping Dassett, with a few houses retained around the edge of the later Town Field. The small amount of land in the manor held by minor freeholders was laid out on the fringe of the parish. Figure 1.9 provides a reconstruction of how the felds were laid out in 1497, based on evidence from the partition between the three joint owners of the manor in 1567 (see Alcock 1981, 28). Before enclosure Particularly important for its direct evidence for Chipping Dassett in the 15th century is the one surviving medieval account roll, for 1480–81 (NRO, TS box 6/2). This shows a proftable estate, with a rental value of £33 (plus £20 from the demesne). In addition, 17d was received from ‘le Bothehall’, while the only charge (apart from various fees) was 14d for making a new ‘cokyngstole’ (for the punishment of scolds), and 12d for ‘thatching the shop beside the Bothehall’. This gives the clearest possible evidence that 15 years before the enclosure, the place was in good enough shape to have use for a market hall and a shop—as well as a suffcient supply of goodwives and ‘gossips’. It may have been somewhat in decline, but the impression is irresistible that the enclosure destroyed a real community and did not just recognise a prior depopulation. This is in clear contrast to the view of the manor provided by Sir Edward Belknap in his defence against the charge of enclosure, of course a partisan document. How then could this destruction have taken place? In the late 13th century, Chipping Dassett contained a substantial number of freeholdings (though, as noted earlier these appear not to have been held by burgage tenure). By c1520, only about four independent freeholders remained, with very modest holdings in relation to the whole area of the manor. The process of consolidation in the hands of the lord of the manor is not documented, but we can surmise that it was essentially complete by 1497, when we see Belknap making what may well have been the fnal purchase from John Heritage. Certainly, by then he was in almost complete control of the manor, easily able to agree with the remaining freeholders. Most of the latter were local gentry, with minor interests in Burton Dassett, and at least two of them, Verney and Raleigh, were involved in enclosure elsewhere in Warwickshire. It is worth noting that the rewards of enclosure were already substantial by the time of the 1520/40

24

BU RTON DA S SET T SOU T H E N D

Figure 1.9 Burton Dassett in 1567, based on an 1826 map of the parish transcribed to ft the 1883 frst edition OS 6 inch map (Shakespeare Birthplace Trust, Special Collections, Library and Archive DR98/1807, Willoughby de Broke archive). Field names are those given in 1567 or earlier, apart from those in brackets which are from later sources. Dotted lines show uncertain boundaries and internal divisions that seem to be major, original hedge lines. The names of the major sections of the open felds of Knightcote and Northend are taken from the Enclosure Award and earlier deeds. H = house, + = parish church and chapel of Chipping Dassett (from Alcock 1981, 28)

rental. Burton and Dassett produced respectively £84 and £86 in rent, with an additional 12s from chief rents, compared with a total of £53 in 1480. Agriculture before enclosure The evidence for the type of farming carried out in Burton Dassett in the 15th century is extremely limited. Setting to one side the information on feld arrangement (see above), we have a unique survival in the form of Roger Heritage’s probate inventory (TNA, PROB 2/457). This is undated, but Heritage’s will (TNA, PROB 11/10 f 231v) is dated 18 August 1495. The inventory was probably taken soon afterwards, certainly before 12 November 1495, when the will was proved. This inventory was

probably taken in August or September 1495. The items related to farming are listed in Figure 1.10. The extent of Roger Heritage’s arable farming is somewhat uncertain. The inventory shows that he owned 16 oxen, two complete ox teams, capable of ploughing 100 acres or more. A valuation of £3 is given for the fallow that had been ploughed during the summer, which might have been worth 1s an acre, indicating about 60 acres of fallow. Under a two-feld system, he should have had the same area under cultivation. The crops in his barn were, however, only worth £8, which at current prices of 6d or 8d per bushel with each acre producing 8 bushels would have represented the crops of about 30 acres. Furthermore, this ignores the old grain that was still in the barn, crops unsold and unused

. I N T RODUC T ION A N D BAC KGROU N D

25

The crop newly brought into the barne with olde whete and olde peasen by estimacion

£8

Item the Telth [tilth] in the felde falowed [fallow feld]

£3

Item hey in the barne

53s 4d

Item rodder [rother, i.e. cattle] bestes, that is to sey 16 oxen

£8

Item 20 kene [kine]

£8

Item 20 bullockes and heyfors

£4

Summa

£33 13s 4d

Item 4 hovelles with the tymbre in the newe hous

20s

Item 2 cartes and carte geres

40s

Item 3 horses and 4 coltes

53s 4d

Item 3 mares

30s

Item ploughes ploughgere and plough harness

6s 8d

Item shepe 860, price the C [100] oon with another

£8 (£68 16s)*

Summa

£15 10s (correctly £76 6s)

Among debts owed to Roger Heritage Item for the conyes

£6

Item Henry Archer, Richard Makernes, John Smyth and Thomas Hunt owed for tillynge

21s

Figure 1.10 Entries relating to farming in the probate inventory of Roger Heritage. * The inventory has the price per hundred (£8) in the right margin, rather than the correct sum (£68 16s) for all the sheep. This was then erroneously added to the other sums to give the incorrect total for this section

since the harvest of 1494. If the inventory was made in August, however, the 1495 harvest may not have been completed. A further uncertainty is the possibility that some of the grain had been disposed of immediately after the harvest. Even with these uncertainties, adding the harvested acres and fallow acres together indicates a total area of land under cultivation of no more than 120 acres, probably less. This is only a fraction of a demesne that once cultivated 300–500 acres. It is also noteworthy that Roger Heritage had undertaken 26s 8d worth of ploughing for his neighbours, representing another 25–30 acres, suggesting that his plough teams were not being fully utilised on his own land. By contrast, of the scale of his pastoral farming, there can be no doubt. The 20 kine, the bullocks and heifers, imply both substantial dairying and the raising of beef cattle. He also had about eight times as many sheep and cattle as would have been expected on a two-yardland holding in an open-feld village (Dyer 1980, 325; 1991a, 12–14). Roger’s 860 sheep can be compared with the 1,000 sheep his son John kept on his Burton Dassett feld in 1507 and 1510, and with the two or three thousand sheep that Roger’s grandson Peter Temple pastured on 850 acres of the same felds 50 years later (Alcock 1981, 37, 81; Dyer 2012a, 147). Roger might have needed about 200 acres, in addition to whatever pasture his holding provided, which presumably came from his lease of the manorial demesne (recorded in the 1480–81 account, NRO TS box 6/2). We can even suggest where these

860 sheep might have been pastured. One of the great felds frst recorded after the enclosure was Old Leys, described in the 1520s rental as ‘a greate leasowe called the Owld Lease’. This must certainly have existed before 1497, though air photographs show it covered with ridge and furrow. It was perhaps created in the 15th century, an example of the conversion of open feld to pasture that could take place independent of a complete enclosure. Indeed, if Roger Heritage was responsible for its use as pasture, this would accord with Belknap’s claim, though not for the implied suggestion that this had also involved depopulation. Finally, Roger Heritage had the use of the demesne rabbit warren from which he was producing a substantial cash crop: his inventory includes among the debts due to him £6 ‘for the conyes’. Roger’s activities must have resembled those of the other inhabitants of Chipping Dassett in the years before the settlement was destroyed. Although individually they would have been farming on a smaller scale than Roger Heritage’s extensive resources allowed, they too would have had a mixture of arable and grazing land, with the latter having been enlarged by mutual agreement at the expense of the former. 1.9 FIELDWORK SCOPE AND METHODOLOGY

The archaeological programme carried out in connection with the M40 motorway construction

26

at Burton Dassett Southend involved excavation, earthwork survey, feldwalking/surface collection and some building recording. The excavations were focused on areas of the site that were to be affected by the motorway and ancillary works, but the survey work extended outside the threatened area in order to provide context for the excavations. The feldwork began in April 1986 with recording of the earthworks on the motorway line in the feld north of the road known as Chapel Ground (but previously as Town Field). A further survey was carried out of the only standing medieval building in Southend, the Chapel of St James and the adjacent Priest’s House in the south-east corner of Chapel Ground. Although a Scheduled Ancient Monument (Warwickshire no. 68; Heritage List 1005747) this was in a very poor state, in use only as an animal shelter, and near to collapse. Excavation work began in May 1986 with the laying out of trial trenches based on an initial interpretation of the earthworks in areas designated A, B and C (see Figures 1.15 and 1.16). Area A was the westernmost of a block of linear properties defned by ditches with a probable house platform at its southern end. Area B was an area of less prominent earthworks to the west and Area C, to the west again, contained a number of less prominent but possible building platforms. The trenches in Area A and the eastern part of Area B located signifcant medieval building and other remains and these were extended into open areas with the topsoil being removed by machine. The trenches in Area C and the western part of Area B revealed no building remains and work here was not pursued. Further topsoil stripping was carried out to the east of Area A at the south end of the adjacent properties, designated D, E and F. These also contained extensive medieval building remains. In February 1987 the feld south of the road, Dovehouse Close, was ploughed for the frst time in some years, turning up large amounts of building rubble and roof tile. Fieldwalking (see Figure 1.18 and Section 1.11, areas Z1–Z6) confrmed that medieval occupation extended over most of the feld. In May the clearance of Area F began. Work on the north side of the road continued until March/ April 1988 when further trial trenching failed to reveal buildings under prominent earthworks on the western side of the motorway line (Area G). In March 1988 work began south of the road in Dovehouse Close with trial trenching followed by the stripping of areas around the densest areas of building remains revealed. The latest phases on the frontages of Areas H, I and W were cleared and recorded. Area J, the smithy, was investigated and the rear parts of Areas H and K. By this stage ( July 1988) it was clear that the open areas could not be fully excavated in the time remaining and a sampling programme involving trenches and

BU RTON DA S SET T SOU T H E N D

small areas was carried out in Areas H, I and K during July and August. The fnal frantic recording and trenching in Areas L and M was carried out at the end of August before the motorway contractor took possession of the site at the beginning of September 1988. Visits to the site were made while earthmoving for the motorway was going on, but the conditions of work meant that relatively little new information was gathered. A few fnds were recovered, while observation of drainage trenches along the south side of the bridge confrmed that the linear rubble spreads along the northern edge of Dovehouse Close derived from boundary walls rather than buildings. In 1991 the part of Chapel Ground now east of the motorway was ploughed for the frst time in very many years and English Heritage agreed to fund a further feldwalking exercise to plot the extent of medieval remains revealed (see Section 1.11, areas Y1– Y4). Finally, in 2003 a programme of archaeological excavation and recording was conducted in and around the Chapel and Priest’s House as a condition of Scheduled Monument Consent, after negotiations between the landowners and English Heritage to ensure the survival of the building had concluded that the only viable way to fund its repair was to permit its conversion to domestic use (see Section 2.5 below; also online archive, Section 7). 1.10 EARTHWORK SURVEY

The earthwork remains of Burton Dassett Southend cover c30 hectares in the six modern felds (Figures 1.11–1.16) surrounding the former chapel of St James, with the largest area to the north-west in the feld now called Chapel Ground, but which was formerly known as Town Field (Figure 1.11). The earthworks were frst surveyed by Bond in 1973 (Bond 1982, fg. 7.3). This was a sketch survey and, although it shows all the main features and their relationship to each other, it is not metrically accurate. A series of measured surveys of the earthworks was therefore carried out in conjunction with the excavation. When surveyed all the felds were under pasture, with the exception of Dovehouse Close which was surveyed in 1987 after it had been ploughed for the frst time in some years. In Chapel Ground the whole strip affected by the motorway was surveyed in detail, together with the blocks of earthworks along the south and east sides of the feld which seemed to represent the main areas of settlement. The northern part of Chapel Ground and the other felds were surveyed in less detail using a Total Station instrument (further details in online archive, Section 6.3). The earthworks were also studied using air photographs taken between 1969 and 1989. The most useful were those taken by Cambridge University in 1969 and 1978 (Figures 1.12 and 1.13).

I N T RODUC T ION A N D BAC KGROU N D

27

Figure 1.11 Measured survey of surviving earthworks at Burton Dassett immediately prior to excavation

Chapel Ground The earthworks in Chapel Ground varied in character across the feld. The main settlement earthworks lay to the south and east against the modern roads across the site, which seemed to follow the lines of the main medieval streets. These settlement earthworks formed a number of sub-rectangular blocks: one in the south-east corner to the north-west of the chapel, with a second to the north and a third to the west. To the west of the second block there was another rectangular area but this seems to have contained crofts or paddocks rather than building plots. The third block was the most regular and also one where the main excavation in Chapel Ground took place. To the west of this there was an area of more enigmatic earthworks, but which appeared to contain some building platforms. The rest of the feld to the north and west contained a variety of non-settlement earthworks. Area north-west of the chapel The sub-rectangular block of earthworks (Figure 1.14), measuring 100m east–west by 120m north– south, immediately to the north-west of the chapel

probably represented the original core of Southend, given that it lay next to the road junction. Along with the earthworks south-west of the chapel in Barn Close, their character differs distinctly from the neat rectangular confguration further west, being more diffuse and irregular, perhaps indicative of organic growth prior to a more regular planned extension. Although the chapel itself was probably a relatively late feature in the settlement it seems to have been built in the central area, apparently over an earlier structure (see below). To the north this block was bounded by a prominent east–west hollow-way, 11m wide and 0.35m deep, which appeared to have had a drainage ditch along its northern edge. To the west the block was bounded by a deep linear hollow, 7–10m wide by 0.95m deep, at the south end of which was a pond. This pond was post-medieval, the spoil heaps from its construction to east and west clearly overlying the medieval earthworks. It is probable that the north–south hollow was originally a street or lane but that it had been deepened and converted into a drain when the pond was dug. The earthworks within the block had no clear pattern and were diffcult to interpret. The individual plot boundaries were not clear, although the

28

BU RTON DA S SET T SOU T H E N D

Figure 1.12 Aerial view of the earthworks from the north-east. The large central feld is Chapel Ground (reproduced with permission of the Cambridge University Collection of Aerial Photography © copyright reserved; CCA 11, March 1977)

basic arrangement appears to have been roughly rectilinear and aligned on the block boundaries. On air photographs (Figure 1.12) it appears that there were buildings on the east side of the block fronting the north–south road and probably others along the hollow-way. There are also some probable building platforms in the centre of the block behind the frontages. Area north of hollow-way To the north of the hollow-way (Figure 1.11) there was another sub-rectangular block of earthworks, c150m north–south by c100m east–west, running

along the north–south road. Its boundary to the west was formed by a ditch on the same line as that of the previous area. This block was divided into a series of east–west properties, with a continuous row of building platforms to the east fronting onto the road, although the frontage appears to have been set well back from the modern line. The tenement boundaries are unclear, being obscured by (presumably postmedieval) ridge and furrow which gives the block a spuriously regular appearance on the air photographs (Figures 1.12 and 1.13). The block probably contained between eight and ten tenements. In two places there are hollows immediately behind the building platforms that may represent yards. In the south-

I N T RODUC T ION A N D BAC KGROU N D

29

Figure 1.13 Aerial view of the earthworks from the west (reproduced with permission of the Cambridge University Collection of Aerial Photography © copyright reserved; AWV 39, February 1969)

west corner of the block there was a linear hollow parallel to the boundary ditch; this, which probably ran across a number of tenements, was perhaps the result of post-medieval quarrying or earthmoving. To the west of this block of properties there was another sub-rectangular block of earthworks, measuring 175m by 85m. It is possible that this western block contained crofts attached to the tenements to the east, but as only one tenement boundary appeared to run on, and that on a slightly different alignment, it is perhaps more likely that the block was divided into two larger, separate paddocks. The northern of these had east–west ridge and furrow, presumably of later date, across

it. The southern paddock was fatter, apart from a slight mound on its west side. This was probably because in recent times it had served as the Northend village cricket ground. The cricket pavilion, removed by 1986, is visible on the air photograph (Figure 1.13) taken in 1969. This area was also cut by a vehicle track which ran right across the feld and was probably a modern feature. Excavated tenements On the south side of the feld, to the west of the pond, there was another, parallelogram-shaped, block of earthworks measuring c122m east–west by c95m

30

Figure 1.14 Surviving earthworks in tenements to north-west of chapel

BU RTON DA S SET T SOU T H E N D

I N T RODUC T ION A N D BAC KGROU N D

31

Figure 1.15 Earthwork survey and excavated tenements A, D1/D2, E and F, north side of the road

north–south (Figure 1.15). This block contained a well defned series of north–south boundaries delineating properties fronting the east–west road. The main excavated area in Chapel Ground (areas A, D1/D2, E and F) ran along the south side of these properties. On the surface the block initially appeared to be divided into four tenements, two small and two large, but on excavation it emerged that there were probably originally six tenements of near-equal size (20m to 21m wide and 92m to 98m long; presumably intended as units of 4 perches width, at least two of which were only later amalgamated; with this knowledge it is possible to see indications of the original six divisions in the surface). To the north the block was bounded by the continuation of the hollow-way which was here c7.5m wide and c0.35m below the ground level of the properties. Along this stretch there were drainage ditches on both sides of the hollow-way. The western boundary of the block was a deeper boundary ditch, c4.5m wide. The southern frontage was set well back from the modern road, by c16m to the west and c10m to the east. The western tenement (A) was c21m wide by c92m long, with a prominent building platform at its south end. The tenement to the east (D) appeared to be double width, c42m wide by c94m long, although

it subsequently became clear that it had originally been two tenements (D1 and D2). There was a continuous building platform along the southern frontage of both tenements, with a low platform to the north on the west side of D2. The southern third of the tenements was divided off by an east–west ditch, forming a ‘toft and croft’ arrangement. This ditch had a kink in the middle at the point where the original boundary between D1 and D2 would have been. To the east of Tenement D2 the next property (E), c19.5m wide by c97m long, had prominent boundary ditches to east and west. It also had a marked building platform along the south frontage with another projecting northwards from it on the west side. This tenement was also divided by an east–west boundary gully, on a similar line to that in Tenements D1 and D2, but here the evidence was slighter. The easternmost tenement again appeared to be double width, c40m wide by c98m long. It is likely to have been originally two properties, but there is no defnite evidence. Only part of the western half was excavated (Area F). The vestiges of a boundary can in fact be seen at the northern end. There were again building platforms on the southern frontage of both properties, although they were divided by a

32

BU RTON DA S SET T SOU T H E N D

Figure 1.16 Surviving earthworks in the western area, north side of the road. Trenching of the prominent ‘platforms’ in the southwest of this area (G) showed that these features existed entirely in the topsoil, as did the mounds in the south-east (C)

gully, and the one to the east was obscured by a spoil heap from the post-medieval pond. On the western property another platform on the west side projected north from the frontage. On the east side of the east property trampling by animals had enlarged the end of the pond hollow, cutting into the tenement and revealing the masonry foundations of an east–west building. This was well behind the frontage and was presumably an outbuilding. To the north and west there were other prominent banks and hollows that were not easy to interpret, but some of which probably represented other outbuildings. Again, there were traces of an east–west ditch dividing both properties into ‘toft and croft’. Other hollows within the block appear to post-date the medieval confguration and were perhaps the result of quarrying for soil. West of excavated tenements The area to the west of the excavated tenements contained a variety of earthworks that were diffcult to interpret, but some of which seemed likely to represent settlement (Figure 1.16). A number of low platforms may have represented buildings, although they did not seem to sit within defned

properties. A series of trial trenches was excavated across these (Areas B and C). Radiating west and north-westwards from here there were three shallow gullies, of which the two northern ones were probably animal or vehicle tracks and the southern one possibly a drainage ditch. To the west a group of more prominent earthwork platforms was bounded by two deep north–south ditches. Against the south edge of the feld there were two platforms, with one larger one to the north, the group together suggesting a farmstead. To the east there was a less well defned rectangular platform. A single trial trench (G) was placed across these. To the north there were three shallow, parallel gullies running WSW–ENE, probably the south part of a block of late ridge and furrow; and in the north-east corner an oval mound, c10m by c8m, had presumably been caused by some post-medieval agricultural activity. North-western part of feld The north-west part of Chapel Ground also contained earthworks but these do not seem to represent the remains of settlement (Figure 1.11). Some of them

I N T RODUC T ION A N D BAC KGROU N D

were drainage features, others were vestiges of ridge and furrow ploughing. Road to west of Chapel Ground Running northwards along the west side of the western boundary of Chapel Ground, until its removal by the farmer in the late 1980s, there was a substantial rubble road. The line of the modern main road is that of the early 18th-century turnpike and this may have been the original line of the Warwick–Banbury road, although it could have been of only local signifcance. On the 1769 Temple Estate map (HEH, STG Manorial Box 53 [26]) this road is shown more prominently, with hedges on both sides, but by the time of the Ordnance Survey Draft 2-inch survey by Henry Stevens in 1812 it had declined to a track, also shown on the Ordnance Survey 1:2500 1st Edition of 1886, when it had been cut by the railway to the north. Further north it appears to have followed the parish boundary between Gaydon and Burton Dassett before heading towards Kingston and Chesterton. Dovehouse Close The south-western feld, known as Dovehouse Close, was not included on Bond’s 1973 survey (Figure 1.11). This feld had been ploughed in recent times; over most of it the earthworks had been considerably eroded and were much less prominent than in the other felds. However, from within the feld, and on the air photographs, a number of features could be made out. Along the south side of the feld there was a prominent, east–west boundary ditch marking the limit of settlement. The ridge and furrow in the felds to the south ran up to this ditch and stopped. To the east the ditch ran into a presumably post-medieval pond; towards the west a north–south ditch ran up to the road dividing off the western third of the feld. This area was formerly a separate feld known as Bullocks Close. A slight east–west bank possibly representing a property boundary ran across it. In the area to the east two similar banks ran back from the road delineating three large rectangular plots. In the central plot, just north-west of the pond, there was a group of mounds and hollows that were possibly building platforms. These also showed on air photographs (see Figures 1.12 and 1.13) taken when the feld was under cultivation. To the north the photographs also show a line of slight mounds running east–west across the central and east plots, set back from the road, and appearing lighter than the surrounding crop. These may also represent building platforms. Documentary evidence suggests that Dovehouse Close was unoccupied in 1567, but that Bullocks Close was the site of a shepherds’ cottage. It seems that the cottage did not occupy a medieval plot but was newly built; however, it had gone by 1769.

33

To west of Little Dassett Home Farm (Barn Close) Barn Close, the feld to the west of Little Dassett Home Farm, contains very well preserved earthworks (Figure 1.11). Its northern half contains a dense but irregular pattern suggesting an uncertain number of irregular house plots, 50m deep, fronting the modern road. It may be argued that the irregularity of the earthworks here, like those immediately north-west of the chapel, may be indicative of organic development within the core of the original settlement. The feld drains laid in 1986 at the west end of the feld unearthed large quantities of rubble showing that, here at least, these plots were quite densely built up. An irregular but continuous east– west boundary ditch appears to mark the southern limit of the house plots, which thus extended back only half the distance of those in Dovehouse Close further west. To the east this ditch runs into a north– south boundary ditch traversing the feld, which may mean that the plots further east, now occupied by the farm, extended the full width of the feld. To the west the south part of the feld is relatively fat except for the north end of ridge and furrow extending from the felds to the south. This area seems to have retained one house after 1497, as the north part of the feld would seem to correspond to the ‘messuage and orchard’ occupied by Thomas Burbery in 1567, with Dovehouse Close to the west, and the ‘meadow to the south’ covering the south part of the feld. The messuage was, however, probably abandoned by 1658. The site of Home Farm was also occupied in 1567, by Thomas Makepeace, and in 1658. East part of settlement The eastern part of the settlement is covered by two modern felds (Figure 1.11). In the northern one, south of Northend Manor Farm, the earthworks are somewhat eroded by recent ploughing. This makes them diffcult to interpret and the tenement boundaries are unclear. However, the earthworks appear to be generally rectilinear and aligned on the road. The earthworks in the feld to the south are more prominent. On the north side there appear to be two east–west strip properties, the northern one c30m wide, the southern c25m, both c160m long; while to the south there are two larger sub-rectangular plots. The frontage of the northern strip is occupied by the existing Windmill Cottage (although its garden now includes the west end of the southern strip as well). The northern strip property is divided into two parts behind the frontage, while the southern one has a series of cross divisions, probably marking building platforms at the frontage and terracing to the east. The two sub-rectangular plots contain possible building platforms and other earthworks on the same alignment as the boundaries.

34

This part of the site also does not seem to have been completely depopulated in 1497. A house tenanted by Thomas Brooks in 1567 probably occupied the site of Windmill Cottage. A cottage recorded in 1582 may have been located to the north. By 1769 there were buildings on the sites of Windmill Cottage and Little Dassett Home Farm Cottages. 1.11 INTERPRETATIVE CONSIDERATIONS OF EXCAVATION DATA

Building remains The signifcance of the building construction techniques is discussed further below (Section 3.2); however, a note about assumptions made in the interpretation of the buildings may be in order here. Various construction methods were evident, including stone walls, timber-framing on stone footings, timber-framing on post-bases, timberwork supported on earth-fast posts at irregular intervals, and timberwork supported on regularly spaced posts. Some buildings exhibited more than one construction technique, with timber-framed extensions to stone houses, or subsidiary structures such as porches supported on earth-fast posts. Cob walls do not seem to have been a common feature of the area and there was no evidence for their presence at Southend. Buildings supported on earth-fast posts or stone post-bases could generally be recognised relatively easily, with similarities in dimensions, fll or packing being used to assign features to particular structures in cases of stratigraphic complexity. In the case of stone foundations, width appeared to offer the best guide to the character of the walls above (cf. Beresford and Hurst 1971, 94). Those over 0.70m (2ft 4in) wide are thought likely to have supported stone walls to the eaves, while those less than 0.55m (1ft 9in) are likely to have supported timber frames. Widths between these extremes can support stone walls. Buttresses and in situ or loose architectural details such as door or window surrounds were also indicative of stone walls. Regularly spaced pad stones along a wall foundation are likely to be indicative of cruck construction. Phasing and chronology The phasing of the site is primarily based on the observed stratigraphic relationships. Over most of the excavated areas the stratigraphy in and around the buildings, whilst shallow, was relatively well preserved, enabling the groups of contexts belonging to structures and associated activity to be identifed and a sequence of structural phases to be worked out with some confdence (Figure 1.17). However between their initial laying out and fnal abandonment the various tenements

BU RTON DA S SET T SOU T H E N D

had individual development histories which were generally not related stratigraphically. Identifcation of the contemporary phases on different areas has been based mainly on a comparison of the pottery assemblages and seriation of the occurrence and changing proportions of fabrics and wares. Isolated features and groups of features unrelated stratigraphically have also been assigned to particular phases on the basis of their pottery assemblages. For those areas (B, C and G) with little relevant stratigraphy and few fnds, the phase groups are more general. Dating the phases and establishing the chronology of the site has drawn on a number of sources of evidence, but also relies heavily on the pottery assemblages and comparison with dated material from other sites. It should, however, be borne in mind that there will tend to be a degree of residuality in the pottery assemblages, given the relatively shallow stratigraphy, and the conclusions drawn should be treated with a degree of circumspection (see online archive, Section 8.1.4.4). The documentary evidence is of limited help. Dated references to buildings and people in Southend are general evidence of activity at the time but cannot be related to the excavated features. The apparent milestone date of c1497 for depopulation at Southend is not necessarily relevant for the entire settlement; the events of that date were part of a longer process and, as we shall see, some buildings and plots had been abandoned considerably earlier in the 15th century, whilst there are signs that others remained occupied later. The date of 1267 for the acquisition of the market charter for Southend is an abstract privilege; whilst it provides a plausible overall context for expansion of the settlement and the laying out of new properties in the mid-/later 13th century it could well have been granted in anticipation of development or in recognition of marketing activity already begun. The absence of a holding paying 12d per annum in rent from the survey of 1280 and the appearance of such a holding afterwards is evidence of continuing development of the settlement, but again there is no direct link to the excavated areas. Some objective chronology for the latest period is provided by the remanent magnetic dating of hearths in the D2 houses (see report by Paul Linford, online archive, Section 8.23); the data were recalibrated in 2017 using the most up-todate calibration curve Arch_UK1. Date ranges for phase D26, the fnal occupation, of 1449–1539 and 1440–1539 generally ft with the other evidence. However, the recalibrated date range of 1294–1327 for the hearth in the D25 hall was far too early to be compatible with either the overall stratigraphic narrative or the ceramic evidence. Recalibration of data for one of the D24 hearths produced no date later than AD 1000. It is diffcult to account for these discrepancies.

I N T RODUC T ION A N D BAC KGROU N D

35

Figure 1.17 Phasing and chronology. The numbered phases for each individual tenement or excavated area (indicated in the bottom row) are shown against approximate date ranges. Single phases were allocated in respect of each area for pre-medieval features, where these were encountered, and for topsoil (nb in respect of the two-digit phase numbering for the two plots D1 and D2, believed to be a single tenement prior to excavation, the frst digit identifes the plot and the second the phase)

Apart from the pottery, the fnds provided little dating evidence. All the few coins and jettons were residual in demolition contexts or topsoil, except for two Roman coins that were residual in mid-/late 15th-century contexts. Other individual fnds were not closely datable or did not occur in suffcient quantity to be useful for dating. The distribution of the small circular iron buckles which seem to be characteristic of the mid- to late 15th century provides corroboration of the dating of contexts rather than primary evidence (see Figure 3.15). Using these methods, the contexts in each area have been assigned to the dated phases of activity used throughout this report. Figure 1.17 is a schematic representation of the chronology and relationships between the phases across the site, with the phases including buildings shaded (see also online archive, fgs 6.4.2 and 6.4.3). In respect of the phase numbering of the tenements identifed as D1 and D2, which prior to excavation appeared to be a single Tenement D, the frst digit differentiates between the two tenements, the second digit denotes the phase; thus D12 signifes the second phase of activity on Tenement D1.

The nature and quality of the stratigraphic evidence Compared with many medieval village sites the relatively well preserved stratigraphy at Dassett Southend enabled the development of the excavated areas to be traced in some detail. This covers the earliest medieval occupation in the 12th/early 13th century, through the settlement’s rise in the late 13th and early 14th century, to its eventual decline and near total depopulation at the end of the 15th century. Over most of the site, vertical stratigraphy had survived in the form of series of foor layers and hearths within the buildings, and external layers and yard surfaces outside, all protected by extensive spreads of demolition rubble. By the end of the excavation a total of 4597 contexts had been recorded over a total area of 5406m 2 covering parts of ten medieval properties and areas of adjoining public space (online archive, Section 9.1). However, there were limitations to the investigations. The north side of the road was more thoroughly excavated than the south side, with all the 3196m 2 investigated taken down to natural

36

substrate. On the south side only 10% of the 2210m2 opened up was completely excavated. This means that while the demolition layers and the last phase of occupation were recorded, the early phases of these properties were much less thoroughly investigated. The fnal sample trenching was concentrated on the area of the houses and aimed really only to establish the date range of the occupation. The lesser amount of excavation is refected in the smaller number of contexts recorded in the areas on the south side of the road. The other main difference between the north and south was chronological. The medieval occupation to the south started earlier and fnished earlier than that to the north. The total excavation north of the road also meant that large numbers of pre-medieval features, mainly irregular treeclearance hollows, were excavated here, but these were largely absent to the south because the work did not reach these levels. On the north side of the road the area of the houses and their forecourts was fully excavated but the rear parts of the properties which lay outside the area threatened by the motorway were not investigated at all. It is possible that these areas would have contained evidence of more outbuildings, and indicators of activities (such as crafts or horticulture). On the south side of the road the whole tenements were threatened and potentially available for investigation, but their layout was not understood when excavation began and the preliminary feldwalking led to a serious under-estimate of the actual number of properties and buildings in the area. The time available for excavation here was also very limited. Although the full length of the Area J smithy tenement was investigated and the main stripped area extended further into the rear parts of Tenements H and K, the excavation was concentrated on the house areas and their environs, with less attention paid to the rear areas. According to the Geological Survey (1:10560 sheet SP35SE) the natural subsoil over the area of the excavation ought to have consisted of grey Charmouth Mudstone (formerly Lower Lias Clay). In practice, over most of the area, the grey clay was overlaid by localised patches and bands of material, presumably of glacial origin, which varied from yellow-olive, olive brown, dark greyish brown clay /clay loam to yellowish–brown gravel. Some of this material differed very little from the flls of the archaeological features and it was often very diffcult to tell them apart. This diffculty was compounded by the fact that the earliest features were mainly irregular hollows, most probably resulting from tree clearance. The presence of fnds or ironstone rubble in the fll was evidence of archaeological, rather than geological, origin, but many of the features lacked these. The character of the stratigraphy on both sides of the road was generally fairly typical of medieval village sites. However, the largest number of contexts recorded (over 20%) were irregular ‘tree-clearance hollows’ (and their flls), occurring almost exclusively

BU RTON DA S SET T SOU T H E N D

in the pre-medieval phases north of the road (see online archive, Section 9.2). These were fairly unusual but seem to be a feature of the Charmouth Mudstone clay subsoil of the site. The second most numerous category of feature (c12%), was the related ‘hollow’. These are defned as being more irregular than ‘pits’. They occurred in the occupied phases. Many of them were deliberately cut features with various purposes, but a substantial number were probably created as a result of the propensity of the Charmouth Mudstone clay to turn to slurry if trampled when wet and for the continual passage of feet over an area to create a hollow, unless consolidated with rubble. As is typical of village sites, there were relatively few, more regular ‘pit’ contexts among the excavated features: rubbish was not buried in pits but collected into muck heaps or middens for disposal elsewhere. The flls of the excavated features, derived to a greater or lesser degree from the local mudstone clays, generally varied very little in colour and texture. The vast majority consisted of clay loam with a limited colour range from dark greyish-brown-olive to brown-olive- or olive (Munsell 2.5-5Y 4/2–4). Other deposits such as foors and the matrices of rubble and demolition layers also exhibited limited variability. In order to avoid repetition of nearidentical descriptions of texture and colour these have been omitted from the printed narrative except where variation is signifcant. This homogeneity meant that it was very diffcult to recognise the relationship of intercutting features. Many of the recorded relationships were not certain. The synthesis presented here, and the fuller descriptions in Section 7 of the online archive, combine the stratigraphic evidence with that from other sources, notably the fnds, and in some cases the former has been overruled by the latter. Stone walls and foundations consisted invariably of rubble of varying sizes bonded with clay or clay loam; again the detailed descriptions of wall matrices will be found in the online archive but have been omitted from this account, except where variability assists in identifcation of different construction episodes. The excavation produced large quantities of fnds from most areas (see Chapter 4 below, with detailed quantifcation in online archive, Sections 9.3–9.5), although again the reduced scale of excavation south of the road is refected in the area totals. The range of material is wide and comparable to those from other large-scale rural excavations such as Lyveden and West Cotton, Northamptonshire (Steane 1967; Bryant and Steane 1969; 1971; Steane and Bryant 1975; Chapman 2010), and Westbury, Tattenhoe and Great Linford, Buckinghamshire (Ivens et al. 1995; Mynard and Zeepvat 1992, respectively). Like the fnds from these sites, the Dassett Southend assemblage shows the full range of the material culture associated with the late medieval rural population of the English Midlands.

I N T RODUC T ION A N D BAC KGROU N D

The fnds were not, however, evenly distributed across the stratigraphy. As is usual on most sites, a high proportion came from the topsoil and demolition levels. It is also inevitable that the context dating of the fnal phase of occupation (in this case late 15th century on the north side of the road and early 15th century over most of the south side of the road) will have attached not just to the fnds actually deposited in that period, but also to all the residual material that had built up in these layers. The result of this for Dassett Southend is that the fnds were concentrated massively in 15th-century contexts. The assemblages therefore contained substantial amounts of residual material, but it is also plausible that the 15th-century occupants of the properties owned a wider range and greater quantity of possessions than those of previous centuries. Evidence of residuality can be seen in the distribution of the few typologically datable objects, such as spurs (see online archive, Section 8.7 by Blanche Ellis), which all came from mid-/late 15thcentury or later contexts, even those datable to the 14th century. Similarly, the copper-alloy personal and decorative items are distributed over the range of 13th to 15th century despite the tendency to fnd them in late contexts (see online archive, Section 8.3 by Alison R Goodall). The spatial distributions

37

also concentrate very noticeably in those properties (especially D2 and E) that were still occupied in the fnal phases. A similar phenomenon was observed in the ironwork at Caldecote, Hertfordshire, where almost a third of the entire site assemblage came from Croft B, probably the last to be deserted (Beresford 2009, 102). This may suggest that the properties abandoned earliest were thoroughly scavenged and re-usable or recyclable materials removed, but that this did not happen after the more widespread evictions of 1497. Realistically there would have been a limit to what could be salvaged and it is unlikely that any material other than obvious items on the surface would have been removed, but the variation in salvage and recycling further complicates comparison of the assemblages from different tenements. 1.12 FIELDWALKING

Two feldwalking exercises were undertaken as part of the Burton Dassett project: in 1987 on Dovehouse Close and in 1991 on Chapel Ground (Figure 1.18). The methods and results are described in greater detail in the online archive (Section 6.5). The principal conclusions are discussed below.

Figure 1.18 Fieldwalked areas in relation to earthworks and excavated areas. Chapel Ground (Y1–4) was feldwalked in 1991 after excavation and motorway construction; Dovehouse Close (Z1–6) was feldwalked in 1987 prior to excavation in the following year

BU RTON DA S SET T SOU T H E N D

38

Figure 1.19 Rubble distribution as found during feldwalking at Dovehouse Close

Dovehouse Close Six main zones (Z1–Z6) are identified within Dovehouse Close corresponding to the divisions already identifed in the earthworks (Figures 1.19– 1.21). Areas Z1–Z3 covered the original Dovehouse Close, to the west Z4 and Z5 covered the north and south parts of the former Bullocks Close, and to the south-east Z6 covered the north-east corner of the feld to the south. The feldwalking in Z1 revealed two main concentrations of rubble within the area: one in the north-west corner close to the road and a second c40m back towards the west side. There was a scatter of pottery over the whole area dating from the 13th to the 15th century, with a particular concentration c65m back from the road on the eastern side which presumably represented a midden/rubbish disposal area. It was concluded that the two main rubble concentrations represented the only building complexes in the vicinity and excavation started on that basis. In fact, when the topsoil was removed the density of building remains proved to be much greater than expected. In addition to the buildings identifed (Areas K and L) there were others on three more properties (Areas H, I and J). This meant that the excavation strategy had to be adapted accordingly. A second shortcoming of the survey was its failure to predict the presence of a smithy in Area J at the north end of Z1. This was a matter of ill luck. The

smithy lay under the modern entrance to the feld. Quantities of slag were found concentrated in this area but it was initially assumed that they represented modern material dumped around the gateway. In other respects, however, the feldwalking results were more accurate: the date range of occupation did prove to be from the 13th to 15th century, and the identifcation of the midden area behind the Area I house was also probably correct. Z2, west of Z1, covered c100m by c110m, bounded by slight earthwork/rubble banks on both sides. Only a small part of the north-eastern corner of this area was subsequently excavated (Area M). Dense rectangular rubble spreads over its central and western parts clearly represented a number of stone buildings. In the centre the rubble distribution appeared to indicate buildings around a courtyard, but too much reliance should not be placed on this. There were also large quantities of roof tile with a marked concentration over the northernmost rubble spread and high densities to the south of this. To the south there was also a massive concentration of roof slate. Both the tile and slate must represent material from roofs that had decayed in situ. The latest buildings here were clearly substantial, high quality structures. The pottery from Z2 ranged from the 13th to the late 15th/early 16th century with some Cistercian ware, which was virtually absent from Z1 to the east. There was also a scatter of post-medieval pottery, but it seems most likely that this area was

I N T RODUC T ION A N D BAC KGROU N D

Figure 1.20 Numbers of fnds from feldwalking at Dovehouse Close

39

40

BU RTON DA S SET T SOU T H E N D

Figure 1.21 Numbers of tile, slate and other fnds from feldwalking at Dovehouse Close

I N T RODUC T ION A N D BAC KGROU N D

abandoned in the late 15th/early 16th century, slightly later than Z1, and perhaps as part of the 1497 depopulation. Excavation of a dense linear spread of rubble along the north edge of the feld revealed that on the northeast it was derived from a substantial boundary wall rather than buildings. Another spread on the east side also represented a wall. Similar spreads along the east side of the area and the north side of the southern boundary ditch suggests that the whole block was surrounded by boundary walls. Area Z3 was edged on all sides by lines of rubble suggesting further boundary walls and to the south and west by a ditch as well. There was a concentration of rubble on the east side, again c40m back from the road and presumably deriving from a number of buildings. The quantities of tile from the rubble spreads were suffcient to suggest that some tiled roofs had decayed in situ. Again, it is not possible to estimate the number of buildings or to identify the properties with which they were associated. There was insuffcient pottery from the area to gauge the date range of settlement. The west edge of Z3 was the original boundary to Dovehouse Close and seems also to have been the edge of the medieval settlement. Area Z4 corresponded to the north part of Bullocks Close. A spread of rubble on the north side can be identifed with the cottage recorded in 1567 as occupied by the shepherds of the tenants of Travers Field which lay to the south-west. This spread produced a concentration of pottery and some clay pipe suggesting an occupation beginning in the 16th century and lasting until the 18th century. The cottage is not shown on the 1769 Temple Estate map and had presumably been demolished by then. A smaller concentration of tile could suggest that the cottage had a tile roof but that most of it had been salvaged when the building was demolished. The very small quantity of medieval pottery from here suggests that this area lay outside the medieval settlement. The linear rubble spread along the south edge is likely to have marked another boundary wall, but presumably of post-medieval date. Z5, the southern part of Bullocks Close, contained no signifcant concentrations of fnds or rubble, and Z6 lay around a demolished building which was covered by a spread of demolition rubble although its foundations could be discerned. It was probably three-sided, c9m by c4.5m, open to the west, with walls c0.6m wide. It was almost certainly a late 18thor 19th-century hovel similar to the one standing in the north-east corner of Dovehouse Close—neither of these appear on the 1769 map. The f ieldwalking in Dovehouse Close thus produced results in which the rubble and fnds distributions combined to give an apparently convincing and coherent picture. The subsequent excavation of parts of Area Z1 showed, however, that this picture was not necessarily complete, and,

41

in particular, the suggested density of buildings and building plots was very much an under-estimate. This makes it more diffcult to assess the conclusions to be drawn for Areas Z2–Z5, but any assessment of the density of settlement is again likely to err on the side of under-statement. Nonetheless it does seem possible to conclude that the medieval settlement in Dovehouse Close began in the 13th century at the east end of the feld in Areas Z1 and Z2, on the periphery of what is suggested above in the description of the earthworks may well have been the original core of Southend. It then spread westwards but did not extend beyond the ditch marking the western boundary of Area Z3. There were a number of substantial stone buildings in Z2 and Z3 with tile and slate roofs divided between an unknown number of properties. Some of the buildings were on a rough line set c40m back from the road. The occupation lasted into the 15th century in Area Z1 and slightly later, perhaps to the late 15th/ early 16th century, in Area Z2. Chapel Ground The area surveyed in December 1990/January 1991 can be divided into four zones, corresponding to divisions in the earthworks: one area (Y1) to the north-west of the chapel and south of the hollow-way; a second, larger area (Y2) covering the tenements to the north of the hollow-way; the block including the excavated tenements (Y3); and the rest of the feld to the north and west (Y4) (Figures 1.22–1.24). The tile and animal bone assemblages were small and fairly unremarkable. The quantities of rubble and material found were generally very much less than those turned up by the ploughing of Dovehouse Close. Given the other evidence of settlement in this area, this suggests that there was a greater depth of topsoil here and that the plough did not penetrate as deeply into medieval layers. The medieval pottery was generally of a similar date range to that found in the excavations (late 13th to late 15th century). Area Y1 The greatest concentrations of rubble and material were found in this area suggesting that it was densely settled. The main rubble concentrations were on the east side, presumably from the buildings fronting the north–south road. There was also a spread of rubble along the hollow-way deriving from the road metalling. Smaller quantities of rubble, just south of the hollow-way, were not enough to prove the existence of buildings on this frontage, because the rubble could have derived from boundary walls, but suffcient to keep the possibility open. There was also a concentration of tile to the east in the same area as the main rubble spreads and a more extensive concentration of animal bone over the eastern half of the block.

BU RTON DA S SET T SOU T H E N D

42

Figure 1.22 Rubble distribution as found during feldwalking at Chapel Ground

As well as a scatter of 13th- to 15th-century pottery over the whole area there was also a concentration of post-medieval material to the east in the same vicinity as the spreads of building rubble. This raises the possibility of a post-medieval building or buildings in Chapel Ground. The pottery seems to be 17th/18th century rather than 16th century, and is thus unlikely to represent continuity of occupation with the medieval settlement. There is a reference from 1582 to a cottage bounded by Town Field to the west, north and south, but this has previously been tentatively placed east of the road (Figure 1.9). The 1769 map shows no building here. Another possibility is that the material was rubbish dumped from the known building on the site of Windmill Cottage on the east side of the road. Area Y2 Area Y2, to the north of the hollow-way, contained lesser amounts of rubble, but again the concentrations were to the east and will mainly have come from buildings fronting the road. Some of the rubble was east of the probable buildings and may derive from paved forecourts similar to those on the excavated tenements. Densities of pottery, animal bone and tile were also less in this area. The pottery evidence suggests occupation dating mainly from the 13th to the 15th centuries. This area also contained

three scatters of modern material, the two western ones probably relating to the remains of the cricket pavilion (visible on the 1969 air photograph, Figure 1.13). Area Y3 Area Y3 corresponded roughly to the excavated tenements. A fairly marked spread of rubble over its south edge derived from the buildings fronting the east–west street. To the west this spread was rather diffuse, presumably because it derived largely from the excavation backfll and spoil heaps. There were also concentrations of pottery and animal bone and, to a lesser extent, tile, in the same vicinity. As one would expect, the date range of the pottery was similar to that from the excavation with some 13th-century material but most being of 14th- to 15th-century date. The northern parts of the tenements were devoid of rubble, perhaps suggesting that buildings, or stone buildings at any rate, were confned to the south ends of the tenements (and within the excavated area). However, on Dovehouse Close archaeological deposits proved to be far more dense than indicated by surface fnds, and here on Chapel Ground a greater depth of topsoil is suspected, so this inference may be incorrect. A noticeable lack of rubble from the hollow-way north of the tenements may suggest that

I N T RODUC T ION A N D BAC KGROU N D

Figure 1.23 Numbers of pottery sherds and animal bones found during feldwalking at Chapel Ground

43

44

Figure 1.24 Tile fragments and slag found during feldwalking at Chapel Ground

BU RTON DA S SET T SOU T H E N D

I N T RODUC T ION A N D BAC KGROU N D

the metalled street to the east degenerated here to an unpaved back lane, although this could again be caused by the greater depth of topsoil. Area Y4 Allowing for the less intensive collection methods used over the north and west of the feld, the densities here of pottery, animal bone and tile did seem to be less, which supports the suggestion that this area was outside the settlement. One signifcant rubble spread identifed on the north edge of the feld (Figure 1.22) coincided with a slight earthwork mound which was itself immediately south of a small enclosure, perhaps a sheep fold, shown on the 1886 1:2500 OS map. The other noteworthy features in this area were two large concentrations of slag (Figure 1.24). It was originally thought that these might represent medieval ironworking outside the settlement, but the main slag component in the concentrations proved to be a distinctive fowed slag/cinder that may have derived from a boiler or steam engine

45

f irebox. The material thus seems to be postmedieval; it might have derived from a steam engine used in the feld or was perhaps imported from the railway as metalling to provide a hard standing for some other activity. Conclusion Although the small quantities of material recovered mean that the conclusions drawn need to be regarded as tentative, the 1991 feldwalking tended to confrm the extent of settlement and buildings suggested by the earthwork and air photograph evidence, whilst the pottery gave some idea of their date range. It also added a number of useful details to our knowledge of the settlement: the existence of post-medieval activity in Area Y1, the apparent confrmation that there were no major building complexes over the north part of the excavated tenements, and the evidence for the metalling of the hollow-way to the east and its probable degeneration into a unsurfaced track to the north of the excavated tenements.

2 THE ARCHAEOLOGICAL SEQUENCE

2.1 PRE-MEDIEVAL ACTIVITY

Tree-clearance hollows

Where the medieval and later deposits had been removed and the natural substrate exposed it could be seen that features both ‘natural’ and man-made extended over the entire excavated area. On the southern side of the road the natural substrate was exposed in only a few areas and thus very few of the earliest cut features were recorded here, the bulk of the early features being recorded on the northern part of the excavated area.

The natural clay subsoil was cut by a large number of irregular hollows, distributed over the whole area (Figure 2.1). Some 453 of these hollows were identifed, making them by far the most numerous type of feature in the pre-medieval phases. The hollows were of all shapes and sizes up to 5.2m by 1.5m, although the majority were 0.5–2.0m long and 0.5–1.5m wide; their flls ranged from dark greyish-brown to olive brown to olive to olive grey

Figure 2.1 Fully excavated areas on the northern side of the road, showing tree clearance hollows

48

clay loam to clay. Some of the hollows may have been geological in origin, and some were shallow and irregular and may represent variations in the subsoil, or root disturbances, but a relatively large proportion (c25%) contained artefactual material, albeit in small quantities and often abraded, and most were probably of archaeological signifcance. It is suggested that the majority of these hollows were created by tree roots and the grubbing up of the roots when the land was being cleared. As features they are quite unusual, and few excavations have exposed and investigated such features over such a large area. It is likely that they are characteristic of the Charmouth Mudstone clay subsoil at Burton Dassett which is soft enough for roots to penetrate. Tree hollows are less frequent in, for example, gravel subsoils into which roots are less likely to penetrate. The hollows can also be more regular—often kiteshaped—and caused by trees falling over, their roots pulling up an area of the surrounding subsoil, and where wind-throw is suspected the orientation of the depression will often be aligned with the prevailing wind direction. In such cases, humic material was found at the ‘downwind’ end of the hole, having fallen off the root plate. In a few instances, the Burton Dassett hollows contained charcoal and patches of burnt earth/fred clay perhaps caused by burning of roots in situ. The features were diffcult to distinguish and excavate as their flls merged into the natural subsoil and their excavated shapes were often rather arbitrary. Some may not have been real and many will have been overcut on excavation. It seems likely that most of the fnds from these features were already in the topsoil when the hollows were created and became incorporated within their flls. They do not therefore refect the activities that created the hollows and only provide general dating evidence in the form of (possibly quite distant) termini post quem. It was also clear that a number of different periods of tree clearance had taken place. Thus, in Area A a group of hollows was cut by Roman or probably Roman features, and in Area B one hollow was cut by a Roman ditch which was cut in turn by two further hollows. A total of 22 features contained only fint fakes/ blades, 40 contained Roman material, and two contained Saxon pottery sherds. In Area E one contained a fint blade/fake, three contained Roman pottery and, in one case, Saxon pottery. In Area F three features contained worked fint, two contained Roman pottery and one a Saxon sherd. There is, of course, considerable scope for mixing and contamination of deposits in tree hollows/ tree throws through bioturbation, but it seems highly unlikely that all these features represent a single phase. There were also similar features in later phases which contained medieval material, including one group in Area A cut by the original tenement

BU RTON DA S SET T SOU T H E N D

boundary ditches, suggesting that these may have been the result of clearance immediately before the laying out of the area for building. The evidence thus suggests some degree of woodland cover prior to the Roman period and signifcant growth of woodland from or after the Roman period, some or even much of it surviving until the 13th century. This, therefore, is a rare example of evidence for clearance of woodland prior to construction and, moreover, in an area traditionally considered as being relatively lacking in trees. Prehistoric evidence An assemblage of 178 fints was recovered during the course of feldwork (see full report by L H Barfeld in the online archive, Section 8.17). The only distributional trend is a slight increase in quantity towards the north-east of the site which appears to refect the overall depth of archaeological deposit. Most of the f lint was residual within medieval contexts. Mesolithic, Neolithic and Bronze Age material is represented, with the bulk of the assemblage apparently Mesolithic, including a number of items (53% of the complete fakes) being in a lustrous red-brown fint uncommon in West Midlands assemblages, apparently from a single nonlocal source; the absence of this material on other West Midlands sites may refect the social territory of a Mesolithic Group in south Warwickshire, while the quantity of material suggests an encampment in the vicinity. Mesolithic material is relatively rare from the area, represented by only two previous fnds: a few blade fragments from the Burton Hills (Wymer 1977, 327) and a few probable fragments from survey and excavation at Nadbury Camp (French 1981, 15; McArthur 1987–88, 10). Given that the Nadbury material was also collected during systematic work, it is likely that its actual density in the area is greater than the recorded fnds suggest. There is insuffcient Neolithic and later material to indicate any permanent settlement. Neolithic activity is only certainly represented by a fake from a polished fint implement, probably an axe. The reuse of axes in the Late Neolithic and Beaker periods is attested at other West Midlands sites where good quality raw materials were harder to procure. An arrowhead is of (probably late) Bronze Age date. Late Bronze Age (LBA) activity is attested by two small pits in Area G which contained similar and distinctive flls of dark grey clay loam and charcoal. One was a small oval pit, 1.25m by 0.60m by 0.22m deep, which contained Late Bronze Age material consisting of 61 small pottery sherds and crumbs from one or more similar plain vessels in the same fabric and two sherds with fnger-pressed decoration, representing a maximum of fve vessels at most. These are characteristic of Barrett’s decorated ware assemblages (Barrett 1980) which occur from the 8th century BC in the Thames Valley. Little LBA

T H E A RCH A EOL OGIC A L SEQU E NC E

material or ‘post-Deverel Rimbury’ ware had been recorded in Warwickshire, with the exception of the recently excavated material from Whitchurch (Brudenell 2011) and few sites have produced Early Iron Age pottery, so local comparanda are few, although there are affnities with assemblages from Wallingford, Oxfordshire, and assemblages recorded downstream from there (see report on prehistoric pottery by A Barclay, online archive, Section 8.18). The other pit was 0.45m by 0.25m by 0.18m deep and lay about 1m to the east; this contained animal bone but only a crumb of fred clay. The two pits are obviously diffcult to interpret, but the quantity of pottery and animal bone from them suggests activity—possibly occupation—in the immediate vicinity. Both pits were cut by irregular tree clearance hollows. Roman activity A group of Romano-British features containing Roman pottery was excavated (Figure 2.2), all on the north side of the road. Vestiges of a series of ditches, probably feld boundary ditches, can be connected across the area with other undated stretches of ditch on the same alignments to make part of a parallelogramshaped feld system. Running across the north-west corner of Area B were two sections of ditch, separated by a 1.3m gap, aligned NE to SE, and with a U profle. To the east, running WNW to ESE across Area A, there were two parallel groups of ditches some 26m apart. That to the south had two cuts, and its line was continued to the west across Area B and to the east towards Area D1 by further lengths of undated gully.

49

On the north side of Area A an undated gully ran parallel to the Roman ditch. These might have been successive, or possibly contemporary, ditches running either side of a hedge or fence. At the east end of the undated gully was a pit/post-hole containing Roman pottery. About 92m to the east of the NE–SW ditch in Area B, discontinuous undated ditches ran across Area E, also on the same NE–SW alignment. Within the area outlined by the ditches there was also a series of shallow, discontinuous gullies, all running approximately WNW to ESE, some containing Roman pottery, but others undated. These may have been cultivation marks connected with the feld system. Further similar gullies to east and west of the NE–SW ditches suggested that the felds continued to east and west. The site produced over 700 sherds of RomanoBritish pottery, most of which was residual in medieval contexts (see pottery report by P Booth in online archive, Section 8.19). There was noticeably more pottery from the northern side of the road, and it is not inconceivable that the road may mark an earlier boundary. Other residual Roman fnds included two coins (a 2nd-century as, possibly of Faustina II, and a mid-4th century AE2 of Magnentius), and two fragments of vessel glass. This material is likely to represent domestic rubbish incorporated into manure scattered onto the felds, although the coin of Magnentius was pierced for suspension and may have had a second, medieval life as a ‘lucky’ charm (Gilchrist 2008, 141; Kelleher 2018, 79), although of course the coin may have been pierced in the Roman or Saxon period.

Figure 2.2 Plan showing the pre-medieval (Roman or undated) features north of the road

50

The pottery spanned the Roman period with no particular peaks of activity, although there was relatively little material of the 1st century or late 4th century. The composition of the assemblage suggests it might have emanated from a small villa or moderately prosperous farmstead. Many of the charred plant samples from the site, both from premedieval and some medieval contexts, contained remains of spelt or other glume wheat (see report by L Moffett in online archive, Section 8.21). Spelt is not thought to have been grown in England during the medieval period and a radiocarbon date (cal AD 395–650 at 95% confdence level) on some of it (report by R Housley in the online archive, Section 8.22) indicates that it belonged to the late Roman/ early post-Roman period. It must also have been distributed by the manuring of the felds. It is not clear to which settlement the feld system would have belonged. There are substantial Roman sites in the vicinity within 1.5km at Burton, south of the church, and at Grandslet Farm, east of Northend, and within 2.5km a probable farmstead was sited near Knightcote, and a villa at Ireland’s Farm, Gaydon (see Section 1.3 above). Further Romano-British settlements are likely to be found in the area; in 1990 both the Burton and Grandslet Farm sites were known only as small scatters of pottery but recent feldwork has shown that they were substantial settlements with stone buildings. The orientation of the feld system is different to that of the medieval landscape. Such dislocation has been noted at other places in the Midlands and has been taken to point to discontinuity of landscape patterns following a ‘collapse’ at the end of the Roman period (e.g. Williamson 2003, 65–66, fg. 24). Nevertheless, the Fields of Britannia project, a large-scale landscapebased study taking into account a large body of data derived from developer-funded archaeological interventions, has yielded a more nuanced picture with signifcant regional variation. Late RomanoBritish feld systems had the potential to survive in some quantity into the medieval period, especially in the project’s central zone (which approximates to the Central Province defned by Roberts and Wrathmell (2000a), and including the Feldon) where over 70% of late Roman feld systems share a common orientation or alignment with medieval feldscapes (Rippon et al. 2014, 206–207, fg. 5; 2015, 196–199, 305–342; also Dyer 2000, 57, where the possible survival of Roman felds at Compton Verney is suggested). Indeed, the orientation of the NE–SW ditches at Southend is not dissimilar to that of a block of ridge and furrow immediately south of the excavated area. A small number of post-holes in Area A appeared to be aligned on the same general orientation as the feld system, and it is conceivable that they represent a structure of Roman date; there was a slight concentration of Roman material here, but not as much as would be expected if the building

BU RTON DA S SET T SOU T H E N D

had been domestic. On the other hand, the post-holes could have been a chance grouping of different dates. A number of undated and possible post-holes were also scattered across the area, some of which may belong to this phase of activity. Apart from those noted above as sharing the same orientation as the Roman feld system, they were not generally associated with any particular concentration of premedieval material. Some may have been of geological origin, or the results of disturbance from vegetation, while other undated features will have originated during the succeeding phases of activity. Continuity from Romano-British to medieval We do not know the extent of continuity between the Roman and medieval settlement patterns (see also Section 1.3 above). Discussing this question for Oxfordshire villages, Bond (1974, 102–103) identifed various levels of continuity: of population in an area, of habitation on the same site, of ownership of land, and of estate boundaries. At one level the presence of the Anglo-Saxon cemetery on the Burton Hills indicates that there clearly continued to be a population in the area, but there may not have been habitation on the same sites, and we have no evidence for land ownership or estate boundaries. The excavations suggest that there was some degree of dislocation between the alignment of the RomanoBritish feld system and the later medieval landscape pattern and provide some evidence for the regeneration of tree cover after the Roman period over previously cultivated land. Such regeneration may have been relatively local in extent; the Fields of Britannia project examined palaeoenvironmental evidence across England and Wales and noted only small increases in arboreal pollen as a percentage of total land pollen between the Roman and early medieval period that need indicate no more than a reduction in hedge-laying and coppicing (Rippon et al. 2015, 219). There was no evidence for large-scale woodland regeneration in the central zone, unlike the south-east and south-west where increases in tree pollen may refect regeneration of woodland. The Fields of Britannia project has drawn out illuminating regional trends, but some degree of variation is only to be expected, and individual localities will have individual biographies at variance with regional trends (Rippon et al. 2015, 342). Hooke, considering evidence from a different part of the Warwickshire Feldon, the Stour valley, tentatively concluded that it is possible that the later village sites were, even in the Roman period, the most important areas of occupation in the region (Hooke 1985, 131). The evidence from the Burton Dassett area does not support this conclusion. None of the most signifcant Roman sites in the area: the town at Chesterton (Camp); the settlement at Banbury Road, Kineton; the villas at Ewefeld Farm, Chesterton, Hill Farm, Lighthorne, Brookhampton,

T H E A RCH A EOL OGIC A L SEQU E NC E

Kineton, Ireland Farm, Gaydon, Grandslet Farm, Burton Dassett; nor the National Herb Centre, Warmington, developed into villages. Although there was an early Anglo-Saxon presence on the Roman sites at Chesterton Camp and at Grandslet Farm, Burton Dassett, these sites were subsequently abandoned. Similarly, almost all the Romano-British sites where suffcient material has been found to suggest the presence of a lesser, farmstead-type settlement were also abandoned at the end of the Roman period. Chesterton Green is a possible exception, but the extent of the RomanoBritish site is uncertain and Bond (1982, fg. 7.1) shows that the Romano-British occupation does not coincide with the medieval settlement areas at Chesterton. While a number of village sites in the area have produced Roman pottery, including Chadshunt, Watergall, Lower Radbourn and Ratley, only a few sherds have been found at these places, not nearly enough to demonstrate settlement on that exact site. At Dassett Southend, over 700 sherds were not accompanied by any evidence for settlement on the site, so the evidential value of a few sherds is minimal. Nevertheless, the case for continuity is not wholly dependent on medieval villages occupying the same sites as Romano-British farms and it is entirely possible that felds close to settlements were being cultivated in both periods. Della Hooke (1985, 126) suggests a correlation between areas of intensive prehistoric and Roman development and subsequent nucleation, with pre-Conquest charters showing arable extending as far as the boundaries of estates in the Feldon area and the central Avon valley. Whilst there is no evidence for continuity of socio-economic organisation beyond the sub-Roman period, continuity of population is a different matter (Booth 1986, 54), while linguistic continuity may be suggested by the place-name Chadshunt (Ceadeles funtan in a 10th-century charter; Gover et al. 1936, 249), which contains funta, a Latin word for well or spring (see also Section 1.4 above). The whole question of continuity is made more intractable by the dearth of non-funerary archaeological evidence for the entire Anglo-Saxon period in Warwickshire. The one possible, signifcant exception to the overall trend of apparent settlement dislocation is the site at Church Farm, Burton, where there may be some evidence for continuity of settlement. The fnding of a stone wall belonging to a Roman building in a water-main trench on the south side of the medieval settlement (Coutts and Palmer 2000) taken together with the scatter of Roman pottery found in the 1973 excavation by the church (Hunt 1972–73; 1973) hints that the late Anglo-Saxon and medieval settlement may have been placed on the edge of a Roman settlement. On the other hand, it may just be that similar requirements at different periods led to reoccupation of the same site. How this relates to the putative Burton ‘burh’ is unknown.

51

There is no visible trace of this on the ground. It is unlikely to have been at Gredenton Hill which is in another parish and is anyway fairly dubious as a fortifed site. The most likely suggestion seems to be that it was not a massive fortifcation, but a smaller, defended aristocratic residence (see Section 1.4 above). Whatever the evidence from the excavations for regeneration of tree cover after the Roman period and the need for removal of some trees before parts of the excavated site could be developed in the 13th century, in the Middle Ages the Warwickshire Feldon was noted for the absence of woodland and lack of timber which had to be imported, often from the Arden of central and north Warwickshire. The boundary between wooded and open landscapes in Warwickshire is likely to go back at least as far as the Roman period (Roberts and Wrathmell 2000b). Domesday records no woodland at Burton Dassett (see Section 1.7 above), and quantities in the rest of the Feldon are small. For the 24 medieval parishes around Burton Dassett there are no documentary references to woodland at all in 17 of them (Wager 1998, appendix 1). There are single references only in Burton Dassett (to ‘the Grove’ in Burton in 1497), Radway and Wormleighton; and more than one reference only in Chesterton, Lighthorne, Ratley and Tysoe. Sarah Wager, however, points out the diffculties inherent in an uncritical acceptance of Domesday book as a topographic record, with its emphasis on manorial income. She suggests that some categories of woodland such as groves and smaller areas of woodland were omitted from Domesday. Her distribution map of ‘wold’ place-names (Wager 1998, 157, fg. 79) shows, in addition to the group beyond the Dunsmore plateau towards the Northamptonshire boundary, a small cluster around (although not within) Burton Dassett, and follows Fox’s argument (1989) that wald implied the presence of small isolated stands of woodland (see also Section 1.4 above). Such a wooded area may have been deliberately selected as the focus for the planned extension of the settlement in the 13th century. 2.2 BEGINNINGS OF SOUTHEND: 12TH TO EARLY 13TH CENTURY

The pre-1200 rural settlement The excavations provided little in the way of physical evidence for the origins of Southend. On the basis of the pottery, the earliest medieval activity in the excavated areas lay in the south part of the site and dated from the 12th to early 13th century (online archive, Section 8.1.4). It is possible, however, that an agricultural settlement at Southend had already existed for some time before this date, and that it was expanding westwards from an original nucleus to the east around the junction between the main

52

BU RTON DA S SET T SOU T H E N D

Figure 2.3 Excavation in 2003 within the late 13th-century chapel and mid-17th century Priest’s House north–south and east–west roads. The east–west road was probably an early feature linking Burton with the western settlement at Hardwick. If Burton was the original, or at least the central, settlement of Dassett with the parish church and the main manor house and Hardwick and Knightcote had begun to develop by the time of Domesday along with the three separate feld systems, it is not clear when Northend and Southend came into existence. The earliest documentary reference to Herdewic dates to 1123 and that to Kynttecote to 1232. The earliest mention of Suthende Derset dates in fact to c1315, although its existence can perhaps be inferred from the frst mention of Northende which comes in 1285 (Gover et al. 1936, 269), and the settlement may have pre-dated both these mentions, perhaps by a considerable time. It is possible, therefore, that by the late 12th century there were blocks of properties only a short distance east of the excavated areas, around the junction of the east–west and north–south roads, and extending as far as the hollow-way and the northern part of Barn Close Field. This nucleus may then have expanded westwards into the southern part of the excavated area. However, the only archaeological evidence from the road junction area was inconclusive. Fieldwalking in Area Y1 recovered pottery of the 13th, 14th and 15th centuries along with one possible 12th-century sherd from a total assemblage of 117 sherds. Sherds from the area north of the hollow-way (Y2) were also mainly 13th- to 15th-century, with another single 12th-century fragment from a total of 143, but the quantities of material involved were

very small and it is quite possible that the earliest pottery was not adequately represented amongst the assemblage collected. A fragment of rubble-stone wall foundation predating the chapel is also likely to belong to this period, but while its date of construction and nature are uncertain, its construction seems too elaborate for a mere boundary wall (Figure 2.3). It ran just to one side of the central axis of the chapel and was over 16.2m long; the eastern end was squared off, but the ragged western end suggests that it had originally continued further. Along its northern edge at intervals of some 2.6m was a series of fve or six squared protrusions (c0.55m wide by 0.1m deep), suggesting truss bases or internal divisions for a timber-framed building of at least six short bays to the north, or possible external pilasters/buttresses for a stone building to the south, depending on which way the surviving wall faced. It was probably not an earlier chapel—the late 13th-century chapel was shorter than this and it is unlikely that a larger chapel would have been replaced by a smaller one. It could have been a barn; its fragmentary plan is reminiscent of one at Church Enstone, if on a smaller scale (Wood-Jones 1963, 15–19). Its presence shows that this part of the settlement was already built up by the late 13th century and its demolition to allow the construction of the chapel perhaps indicates a desire to locate this important building at a focal point in the settlement. Apart from the building fragment pre-dating the chapel, the earliest medieval activity from the excavated areas comes from south of the road (phases H1, I1, K1).

T H E A RCH A EOL OGIC A L SEQU E NC E

53

Figure 2.4 Plan, phase of c1220, showing features that pre-date the laying out of plots

This included a series of large east–west ditches running across Areas K and H (Figure 2.4). The ditches were only excavated in short sections but ranged in width from 0.96m to 2.4m and in depth from 0.40m to 0.51m and showed signs, such as irregularity of prof ile, of having been recut.

They appeared to turn northwards to the east and presumably represent boundary ditches defning an enclosure, a paddock or small feld to the north, on the south side of the road, and perhaps peripheral to what at this point may still have been a small rural settlement. To the west, one of the ditches

54

butt-ended, suggesting that there may have been an entrance into the enclosure at this point. There was no evidence of buildings or occupation to the north-west within the enclosure. Virtually none of this area was excavated thoroughly but some remains might have been expected in the initial trial trenches or the stripped areas. To the east of the enclosure there was a cluster of smaller features, gullies and hollows, in Areas H and I. Their presence perhaps suggests that by this time this area was a patch of waste ground on the edge of the settlement. If large-scale dumping from beyond the excavated area is ruled out, the amount of residual 12th- and early 13th-century pottery in the later phases on the south side of the road may be indicative of occupation or activity within buildings which is not directly visible in the archaeological record; not only is it entirely possible that such early buildings were timber founded, the excavation strategy on the southern side of the road imposed by the constraints of time would not necessarily have been suited to the identifcation of such structures. It should also be noted, however, that the pottery assemblages on the north side of the road also contained residual 12th- and early 13th-century pottery. Planned settlement phase 1: early 13th century (H2, I2, J2, K2) The next stage of development, dating to the early 13th century, saw the laying out of a series of building plots in the excavated area south of the road (Figure 2.5). The plots formed two rows, one along the road frontage (Areas L, N and J) and a second parallel row to the south (Areas K, H and I). These may have been laid out together as a planned, regular development, although their subsequent occupation was staggered. The evidence for this episode of planning was less extensive than that for the later one north of the road, mainly because less ground was excavated here. Only in Area J was any original boundary feature located exactly. The most likely scheme does not ft the excavated boundaries precisely (Figure 2.6), but there was evidence from Area J that some boundaries had shifted over time, and others had possibly also done so. However, the results are within a reasonable margin of error, and they suggest a more regular layout than the boundaries of a similar development at Kineton (Goddard 2011, 13, map 1). In the northern row, the width of Area L was very similar to that of Area J (c14.5m), and the area in between was approximately twice this (29m). This width approximates to 15.09m or 3 perches of 16½ft, the usual measure of medieval settlement planning in the region (Slater 1982, 195), although the tenements examined north of the road and laid out in a later phase were 4 perches wide. The space between Area J and the western boundary of Dovehouse Close could have accommodated another two property units. Area J was c29m deep, close to 6 perches (30.18m),

BU RTON DA S SET T SOU T H E N D

and the southern boundary of Areas L and N may have followed the same line in the unexcavated strip north of Area K. In the row of properties to the south, the widths of Areas H and I (c16m and 15.5m) were also approximately 3 perches, while Area K, at c42m wide, appeared to cover three units. The boundaries between these tenements were on similar lines to those to the north, although the one between Areas H and K ran too far east at its southern end, and the east boundary of Area I to the south, as represented by the feld boundary, ran too far west. These divergences could, however, be the result of later changes. The southern properties were much longer than the northern ones. Their southern limit is uncertain, although in Area K a short section of rubble foundation was found to the south of the tenement. This lay c60m, or about 12 perches, behind the frontage. Nothing similar, however, was found in the limited investigations at the south end of Area H, and the possible length of 12 perches for these tenements remains speculative, if plausible. Although the evidence is not totally conclusive, this part of the settlement seems to have been deliberately planned and laid out in a single episode, even though the subsequent occupation of the plots was apparently staggered. Four or possibly fve properties 3 perches wide by 6 perches deep were laid out at right-angles to the street frontage, with a gap in the middle, also 3 perches wide, for a road running back from the frontage. The road gave access to the rear of these properties and also to the frontage of a second parallel row of fve larger properties, also 3 perches wide but perhaps 12 perches deep. The smaller size of the northern tenements was balanced by their more favourable position fronting the main street. There was no obvious trace of a back lane giving access to the rear of the southern row, but this area was hardly investigated and such a lane could well have existed. Whilst the rectilinear form of the plots is typical, the plan of this development seems more elaborate than those of most known tenement series. Elaborate schemes are nevertheless diffcult to substantiate using normal methods of topographical analysis and few series have been proved by excavation. The dimensions of the southern plots are similar to those of other planned settlements (to take an urban example, Stratford, at 12 by 3½ perches); the northern plots are rather small but at half size represent a logical division. Where metrological analysis permits, modular planning may be found in a large number of settlements and the use of regular plan measurements extends back well into the early medieval period (Blair 2018, 70–71). Row-plan layouts frequently conformed to standard measurements, with widths in multiples of a perch; whilst there is variation, sometimes regional, in the actual width of a perch, a standard measurement of 16½ft was frequently used. For example, at West Cotton, such a standard

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.5 Planning and layout of medieval tenements, showing 3 and 4 perch plot widths

55

BU RTON DA S SET T SOU T H E N D

56 Tenement Width (m)

Length (m)

actual 3 perches variance actual 6 perches 12 perches variance

L

N

14.5 15.09 -4%

15.09

30.18

30.18

(N + road)

road

(29) 15.09 (-4%)

J 14.5 15.09 -4% 29 30.18 -4%

K ×3 c42 45.26 -7% 60 60.35 -0.6%

H

I

c16 15.09 6%

15.5 15.09 3%

60.35

60.35

Figure 2.6 Dimensions of tenements south of the road

perch was in use from the 10th century (Chapman 2010, 30–32, fg. 4.2); the same length was used in the Whittlewood area along the Buckinghamshire/ Northamptonshire border, although there were instances in that case of a longer perch length being used, often in woodland or forest areas, alongside the standard 16½ft measure ( Jones and Page 2006, 189). There were also indications in Whittlewood of a relationship between a plot’s dimensions and the status of its occupant. The combination of shorter plots next to the main street and longer plots to the rear in this phase of settlement planning at Southend thus hints at some form of differentiation, although this could not be characterised archaeologically. The circumstances of this development are uncertain. It certainly pre-dates the acquisition of the market charter for Southend in 1267 and may just represent a stage in the continuing expansion of the agricultural settlement westwards. Alternatively, it may represent the frst steps towards maximising returns from the estate and promoting a market; as Chris Dyer notes above (Section 1.7) there may well have been commercial activity developing here before the granting of the market charter (see also the discussion below). 2.3 OCCUPATION OF NEW TENEMENTS SOUTH OF THE ROAD

Occupation of the new tenements began over a period in the early/mid-13th century. In the southern row, Area H was the frst to be occupied, while at a slightly later date the three others to the west seem to have been amalgamated and occupied as one unit (Area K), followed by Area I to the east. There was less take up of the shorter properties in the northern row and not all were successfully let. It is suggested that they were intended to appeal particularly to traders or craftsmen and Area J, to the east of the side road, was occupied by a blacksmith. The westernmost property, Area L, was also occupied in later periods, but insuffcient excavation was carried out to say when this occupation began. The other two properties (Area N, and the unnamed property east of Area J) seem not to have been inhabited at all. The frst buildings on the new tenements differed in character, suggesting that, as was normal (Dyer

1986, 29), they were built by the incoming tenants rather than provided by the landlord. Area H house (phase H2) The eastern boundary of Tenement H was marked in its earliest phases by a U-shaped ditch, 0.75m wide and 0.36m deep (Figure 2.7). It is not certain when the ditch was frst dug. Only one cut was visible and no indication was found of any earlier boundary feature. The ditch ended 1.3m from the house; the gap might be explained if there was an access here into the neighbouring property. The house was set in the north-east corner of the tenement and was aligned east–west along the frontage. It was semi-detached, sharing its east wall with the house on Area I. In its original form it was 10m long by 5.2m wide. The surviving wall footings were mainly fairly narrow, suggesting that the building was timber-framed; it was probably three bays long. The north and south walls, of medium–large rubble facing a small–medium core bonded with clay loam, appeared to have been built in sections of varying width (0.9m wide at the eastern end of the north wall, which was only 0.5m wide in the central third widening to 0.9m to the east, and similar variation in the opposite wall), presumably corresponding to the bays of its frame. Perhaps the wider eastern sections of both walls represent an extension to the original building (adding an extra bay c3.5m long). However, there was no trace of an earlier east wall. The west wall, of similar construction to the north and south walls, was 0.85m wide, while the east wall 0.60m was wide and formed a party wall with the neighbouring Area I house. The nature of the bonding suggests that it is more likely that the east wall was originally constructed as part of house H, rather than I. Both the south and east walls contained re-used material including a quern fragment. This building had a foor area of 34m2 (366 sq. ft). No hearths were found within the house nor internal partitions; the building was probably undivided, but the interior was disturbed, and traces of partitions could have disappeared along with any hearth. A series of shallow hollows flled with clay loam may

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.7 Area H, Phase H2, 13th century

57

58

represent the very patchy remnants of foor layers. The house had a south doorway 0.80m wide in the eastern bay, and there was probably another, facing to the north, its presence indicated by a later area of internal metalling. Some care went into the building of this house: the interior was drained by a soakaway drain of large ironstone blocks beneath the foor running northwards into the street under the north wall. To the west of the house the northern boundary of the plot was marked by a narrow rubble wall footing 0.45m wide, exposed in an exploratory trench and aligned with the north-western corner of the house; its full extent is unknown. Behind the house the eastern boundary was a ditch with post-holes, possibly representing a fence, on its east side. To the south-west of the house there were postholes that could have supported part of a small and somewhat irregular timber outbuilding possibly c6m long, although insuffcient was excavated to gain any idea of its plan. In the southern part of the tenement exploratory trenching revealed a series of other early features but the limited extent of excavation makes their interpretation problematic. No clues can be found for the occupation of the Area H tenant in the mid- to late 13th century. The only distinctive characteristic of the fnds assemblage was the very large quantity of residual pottery in the topsoil, of which a high proportion belonged to this period. It is possible that, for a time and contrary to normal practice, domestic waste was not removed but stockpiled on the tenement. Area I house (I2) There was no direct dating evidence for the laying out of the Area I tenement although it seems likely to have occurred in the early/mid-13th century at the same time as the others. The eastern boundary with Area H was formed by the ditch noted above; the walls that marked the western boundary are unlikely to have been original. The Area I house (I2) was built against its neighbour to the west on Area H, sharing the latter’s east wall as a party wall, and parallel with the frontage at the north end of the tenement (Figure 2.8). The north, south and east rubble walls with irregular discontinuous offset foundations varied in width from 0.65m to 0.85m. These dimensions would seem to indicate that the house was stonebuilt, even if it was abutting a timber-framed party wall. It was 13.5m long by 6.0m wide, probably with a three-bay roof structure, and a foor area of 53.13m2 (572 sq. ft). It had a south doorway, on the western side of the central bay. The northern door to the street could have been opposite but may have been 2m further east. Too little of the house was excavated to discover its original internal arrangements; there was no trace of partitions, a hearth or fooring, but against the south wall at the west end there was a

BU RTON DA S SET T SOU T H E N D

rectangular, stone-lined trough, 1.88m long, 0.55m wide and 0.22m deep, the base being lined with fat blocks of ironstone (Figure 2.9). The trough’s function is uncertain, but probably involved some small-scale processing of an animal or agricultural product. There was a similar trough in the D24 house (see below). Post-holes in the sample trench south-west of the house may again have belonged to a contemporary outbuilding, but too little of this area was excavated to be sure even that these represented a building. There was no indication as to the occupation of the inhabitants of the I2 tenement, but among the fnds from south of the house was a musical instrument tuning peg (see Figure 4.14), probably from a harp or fddle. A second similar peg, probably from the same instrument, was found in a later layer to the north. Area K house and granary (K2) As has been seen, Area K covered the equivalent of three plots at the western end of the southern row. None of the original tenement boundary was excavated and the limited nature of the excavation meant that the full building sequence of the house was not defnitively unravelled. However, it seems likely that there were two main phases with the walls of an earlier building running just inside those of the later. The earlier north wall was represented by a 6.8m length of foundation, 0.36m wide and composed of medium–large ironstone rubble bonded with clay loam, except for the central section (2.1m long) which was made of large faced ironstone blocks standing 0.28m high, including one re-used jamb fragment. It was unclear whether this was built against the later wall or had been truncated by it. Running parallel was a 4.15m length of a similar foundation, 0.60m wide, and west of this on the same line was what appeared to be a robber trench. These features are likely to represent the remains of an early building rather than benches built against the walls of the later phase of house K. Whilst the actual ends of the foundations were not located, they did not extend westwards or eastwards beyond later partition walls, and it is plausible that the partition walls may have been on the line of the original end walls of the building. The building will have measured, therefore, 5.8m wide by at least 8.3m long and perhaps, if the position of its end walls were perpetuated by partitions of a later phase of the building, 10.5m long. Internally it would seem to have consisted of a single, two-bay room, 9.1m by 4.5m, with a foor area of 41.4m 2 (446 sq. ft), although further work might have revealed evidence for partitions in this early phase. Too little of its footings survived to say whether the building was timber-framed or stone built, and none of the earlier foor layers or internal features can be associated

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.8 Areas I and W showing phases I1–I3 and W1, 12th to 14th centuries

59

60

Figure 2.9 Area I, stone-lined trough in south-west corner of house, from the south (Phase I2, 13th century)

with the putative early building, although some of them may have belonged to it. Set into the top of the former ditch along the line of the north wall of the house were various layers of small–medium rubble which were probably intended to counteract subsidence of the ditch flls and deposited as part of the house construction. Granary About 7.5m south of the house and aligned approximately parallel to it, there was a small rectangular outbuilding with solidly constructed stone walls, and probably originally a raised foor, which is thus interpreted as a granary (Figure 2.10). This was also built in the mid-/late 13th century. At its full extent the granary measured 9m by 5.0–5.4m and contained two elements: a single cell with solidly constructed, buttressed, stone walls; and a narrower L-shaped stone foundation, on a slightly different alignment, which probably supported a timber staircase on the west end (Figure 2.11). The difference in construction and alignment suggest that the staircase may have been a later addition, although this cannot be proved stratigraphically. While the

BU RTON DA S SET T SOU T H E N D

granary was possibly built from the beginning as a two-storey structure, it seems more likely that it started life as a single storey, stone-walled building, and was then fairly soon afterwards converted to two storeys with the addition of a timber-framed staircase and a possibly timber-framed upper storey. The original stone part of the granary was slightly trapezoid, measuring 7.4m by 5.0–5.4m. Its walls varied in thickness (north 0.6m; east 0.7–0.8m; south 0.8m, west 0.7m), and there were buttresses at the west end of the north wall and the east end of the south wall; their presence shows that these walls were stone, but not whether they were one or two storeys high. However, the relative thinness of the north wall, compared with some of the house walls elsewhere on the site, suggests that the granary was a single storey building in its initial phase. The internal dimensions of the granary were 3.5m by 5.9m, giving a foor area of 20.65m2. In the northeast and south-east corners of the room were two rectangular post-bases of ironstone rubble bonded with clay loam; one 0.45m by 0.45m, the other 0.60m by 0.45m. These post-bases seem likely to have supported a suspended wooden foor, a feature often found in granaries to raise the grain away from damp and vermin. There were also earth and rubble foors, however, and the sequence is not clear. It is perhaps likely that the suspended foor was the original and was replaced when the second storey was added in the subsequent phase. It is possible to offer an estimate of the capacity of the granary at around 550 bushels, the produce of perhaps 1½ yardlands; the calculations and their implications are discussed in Section 3.2 below. Area J: Smithy (Phase J2) One tenant is known to have pursued a nonagricultural occupation: the blacksmith occupying Area J (Figure 2.12). The quantities of ironworking slag from throughout the stratigraphy in this area show that this was the case from the beginning. The tenement appears to have been laid out in the early/mid-part of the 13th century with gullies to west and east. The western gully was replaced by a new boundary ditch, wider and deeper than the original, towards the end of this phase. To the west of the tenement was a north–south road or street c7m wide and surfaced with ironstone rubble. The smith occupied a building ( J2) on the street frontage at the north end of the tenement, which served as both smithy and (possibly) living quarters, although one imagines that the smithing apparatus and debris were not entirely compatible with domestic occupation. The smithy had been heavily robbed after demolition, only the west wall surviving over its entire length with the lines of the other walls marked by robber trenches and discontinuous fragments of masonry. Such masonry as survived was solidly constructed with large roughly faced ironstone blocks

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.10 Phase K2, 13th-century house and granary

Figure 2.11 Area K, granary, from the east (Phase K2, 13th century)

61

62

Figure 2.12 Smithy and environs (Phase J2, 13th century)

BU RTON DA S SET T SOU T H E N D

T H E A RCH A EOL OGIC A L SEQU E NC E

63

facing a smaller rubble core, although the south wall appears to have been narrower (0.7m wide) than the others (0.80–0.95m wide). The building measured 12.2m by 5.2m, with a foor area of 37.5m2 (404 sq. ft). It is not clear whether the building was originally divided internally. It is possible that two post-holes, one 0.35m by 0.37m by 0.23m deep and the other 0.24m by 0.21m by 0.11m deep, may represent the line of an early partition which would have divided the building into two rooms, one to the west 6.75m long, and possibly a two-bay hall workshop, and a single-bay chamber 3.25m long to the east, but the post-holes could have belonged to an early forge structure instead. A southern doorway into the centre of the hall/ workshop may be marked by a pair of external postholes that could have supported a porch, and it is possible there was a northern doorway to the street opposite. A foor consisting of a layer of clay loam covered the western 5m of the building and, although being covered in part by other layers and cut by various features, it appears to have been the basic foor throughout the building’s use. The forge may have been in the north-east corner of the hall/workshop on supports set in a series of post-holes. No hearths were found in this phase, but there may have been a domestic hearth in the centre of the western bay, as in later phases. The only notable fnd from the phase was a carpenter’s reamer made of iron (see Section 3.2, tool no. 13) from the western boundary ditch. 2.4 PLANNED SETTLEMENT PHASE 2, LATE 13TH CENTURY

Not all the tenements on the south side of the road attracted tenants to make their living in them, but Southend was judged suffciently successful by the late 13th century for more tenements to be laid out on the north side of the road. These formed a parallelogram-shaped block of six tenements of approximately equal size and representing another planned development (see Figure 2.5). At this time the area was probably waste ground or paddocks containing a few trees rather than part of the open felds. Given the amount of residual 12th- to early

13th-century pottery throughout the area, rubbish was probably also being dumped here from further east in the settlement. This dumping seems to have been confned to the surface and did not involve below-ground disturbance, which would have been evident on excavation. The more complete excavation in this part of the site meant that the evidence for the original boundaries and measurements was better than south of the road. The original laying out seems to have been done by digging narrow gullies, 0.5– 0.8m wide, along the property boundaries: these survived between Area A and Area D1, and between Area D1 and D2. The original western boundary of Area A had been removed by later ditches, although its line can be estimated. The boundaries between Areas D2 and E, and E and F, had been obliterated by later walls, although the outside edges of the walls probably refect the original line. Even where the original boundary gullies were preserved, it is impossible to be sure from which point original measurements were taken or to gauge the accuracy of medieval measuring cords. This being said, it was, however, possible to take a number of reasonably accurate measurements (Figure 2.13). The excavated tenements varied: 20–22m in width and 92–98m in length, which approximates to a width of 4 perches (20.12m), a length of 20 perches (100.58m), and an area of 0.5 acre. In Area B along the west side of the new block of tenements there was a north–south rubble road up to 5.5m wide (Figure 2.14). The surviving road surfaces were later in date, but the road was probably frst laid out in the late 13th century. The rubble surface was very rough, but cambered, showing that some care had gone into its construction. Although the road surface petered out at its north end, it is likely that it would have linked with the hollow-way visible in the earthworks along the north side of the block to form a back lane giving access to the rear of the tenements. There would also presumably have been a similar side road in the hollow along the eastern side of the block between it and the existing settlement. A spur road leading west from the north–south road in Area B was of similar construction but narrower Tenement

Width (m)

Length (m)

Area (sq. m)

actual 4 perches variance actual 20 perches variance actual 0.5 acres variance

A c22 20.12 9.20% c92 100.58 -8.5% 2024 2023.4 0%

Figure 2.13 Dimensions of tenements north of the road

D1 21.8 20.12 8% c94 100.58 -6.5% 2049.2 2023.4 1.3%

D2 20.8 20.12 3% c94 100.58 -6.5% 1955.2 2023.4 -3.4%

E 20.4 20.12 1% c 97 100.58 -3.5% 1978.8 2023.4 -2.2%

F c20 20.12 -0.6% c98 100.58 -2.6% 1960 2023.4 -3.1%

BU RTON DA S SET T SOU T H E N D

64

Figure 2.14 Area B, rubble road, from the south, probably frst laid out in the late 13th century (Phase B2), although the surviving surfaces are later. The narrower spur road leading west is also visible

(c3.5m wide) and with short lengths of gully to north and south which may have been the remains of roadside drainage gullies; its westwards extent is not known, although it was not picked up in a trial trench 75m to the west. In Areas A, D1 and D2 there were a few tree hollows containing 13th-century pottery which were cut by the original tenement boundary gullies; this suggests that the laying out was accompanied by removal of some trees. In addition, a length of drainage ditch, orientated NNE–SSW diagonally across the plot, was inflled in Area A. This again suggests that the new development was built over an area of pasture or waste adjacent to the existing settlement, rather than over part of the open felds. The slightly unusual parallelogram shape of the block of tenements does not seem, therefore, to be the result of following

existing feld strip boundaries. It actually appears to follow the angle of the existing block of properties to the east, which itself refects the angle of the junction between the main east–west and north–south roads. The buildings constructed on the new plots (Figure 2.15) again varied in size and plan, indicating that they were built by the incoming tenants rather than provided by the landlord. Area A The original eastern boundary of Tenement A was a gully 0.78m wide and 0.32m deep; that along the western side had been obliterated by ditches, much recut, from later phases, whilst the boundary along the frontage had also been destroyed by later features (Figure 2.16).

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.15 Phase plan, c1290

65

66

Figure 2.16 Area A, Phase A2, later 13th century

BU RTON DA S SET T SOU T H E N D

T H E A RCH A EOL OGIC A L SEQU E NC E

The frst house (A2) on Area A was a substantial building, 14m long by 6.4m wide. Only the west and part of the north walls had survived; they were 0.85–0.90m wide and made of large blocks of roughly coursed ironstone rubble facing a medium rubble core. The robber trenches of the other walls were of comparable width, suggesting stone construction to the height of the eaves (Figure 2.17). The house may also have had a roof of Stockingford Shale slates, since three fragments were found within features of this phase, but this could have been added at a later date. There was only the slightest indication of the location of the south door in the form of a patch of later metalling outside the wall 4.0–5.5m from the west end. No trace of internal partitions survived, and it is likely that the house consisted of a single large hall of three bays with a foor area of 57m 2 (614 sq. ft). It had an earth (clay loam) foor, possibly supplemented by rubble, which survived in patches at the east end. Against the north wall in the centre of the eastern bay a hearth was cut by a stakehole which may have supported cooking apparatus, and against the south wall a rectangular rubble base had been provided for some internal fxture, but the earliest internal features of A are diffcult to date, and could have belonged to later phases. To the west of the house a gap 4.5m wide between the house and the western boundary

Figure 2.17 Area A, house, from the east (Phases A2–A5)

67

ditch would have allowed access for carts to the rear of the tenement. No outbuildings belonging to this phase were found. About 11m north of the house a boundary ditch, presumably marking a fence or hedge separated the frontage area from the rest of the tenement. Gaps 2m and 3m wide at either end would have given access between the two areas. The area north of the ditch was almost devoid of features and could have been a garden. Area D2 On the western side, the original tenement boundary survived (Figure 2.18). This was a steep-sided gully c0.45m wide by c0.20m deep. The original eastern and southern boundary gullies had been obliterated by later walls on the same line. Before the laying out it appears that some trees had to be cleared, as the western gully cut through a tree hollow containing 13th-century material. On the D2 tenement the frst house (D22) was a trapezoidal building c11m long by c5.5m wide, aligned along the frontage with a gap of c0.6m between it and the eastern boundary of the tenement. The eastern wall was angled to follow the boundary line. The walls were 0.60–0.65m wide, of ironstone rubble with some facing stones. Some later robbing had taken place along parts of the north and south walls.

BU RTON DA S SET T SOU T H E N D

68

Figure 2.18 Area D2, Phases D22, late 13th century, and D23, 14th century

The internal arrangement of the building was not completely clear. The eastern third was divided off by spere walls (short screens separating crossentry from hall) to the south and, probably, to the north. It is also likely that the western third was similarly divided off, making an arrangement of three roughly equal bays (3m, 3.5m and 3m long by 4.25m wide) with wide doorways (c2m) connecting the bays. The south spere of the eastern partition was supported by a row of four large ironstone blocks. The northern spere was supported by a

stone-packed post-hole, 1.2m out from the north wall. The northern part of the possible western partition would have been supported by the two stone-packed post-holes which were 0.9m and 1.0m out from the wall; there was no evidence for the opposite spere wall at this end of the building. There were one or two clay-lined hearths on the eastern side of the central section, which was presumably the hall; one hearth had a diameter of c0.76m and was 0.06m deep, the diameter of the other was 0.4m with a depth of 0.09m. The location of the external

T H E A RCH A EOL OGIC A L SEQU E NC E

69

Figure 2.19 Area E, Phase E2, late 13th century

doors in this phase is unknown. The total foor area would have been 40.38m 2 (435 sq. ft). Area E No trace survived of the earliest boundaries of Tenement E, which had all been obliterated by later walls, although their lines were clear (Figure 2.19). The earliest house here (E2) was smaller than the others and was set in the south-west corner of the plot, running back from the frontage. It was slightly parallelogram shaped, measuring 5.1m by c7.5m. The north wall and the north ends of the west and

east walls survived, all 0.75m wide, incorporated into later phases of building. The line of the south wall survived as a truncated robber trench. A pit/ post-hole, 0.52m by 0.35m by 0.10m deep, in the north-west corner was probably another construction feature. On the basis of all widths it seems likely that E2 had a single storey and full height stone walls. Internally it was undivided. The foor area was 21.6m 2 (233 sq. ft). A possible concentration of post-holes some 10m east of the house, on the frontage, all undated, outlined a roughly rectangular area c3.5m by 4m; whilst these may have belonged to a timber

70

outbuilding, they were more likely to have been a chance grouping of different dates. Areas D and F No trace of buildings of this period was found on the D1 and F tenements; this may be because these properties were without tenants, or that all traces of the original buildings had been obliterated by later activity. The original tenement boundaries on D1 were represented by U-profle gullies; that on the east was 0.45m wide by 0.20m deep, and that on the west 0.85m wide and 0.30m deep. Later ditches along the southern tenement frontage boundary had removed almost all trace of the original boundary there, but a truncated length of steep-sided hollow at the east end (0.35m wide by 0.06m deep) may have been part of a boundary gully. Adjacent to it a post-hole, 0.30m by 0.26m by 0.12m deep, may also have been a boundary feature. Several pits and irregular hollows in the south-west corner of the tenement, and a truncated irregular gully to their north, also belong to this phase and may well suggest occupation or activity. On Area F there were also a few shallow hollows and post-holes or possible post-holes (F2). The hollows were concentrated around the south and west tenement boundaries and may have been created when the tenement was being laid out, but the original tenement boundaries were obscured by later walls and robber trenches. The post-holes were scattered across the area; they did not make any coherent structure although they could have belonged to an earlier building. South of the road: alterations to Area K granary About 1.2m to the west wall of the granary on Area K (Figures 2.10 and 2.11), and roughly parallel to it, was a discontinuous ragged rubble wall footing c0.50m wide; this overlay a shallow rubble-flled hollow that could have been either the foundation for the wall or remnants of a surface that had subsided into a hollow. The wall had a gap 1.2m wide towards its south end, presumably marking the site of a doorway. Running between this wall and the south-western corner of the original building was another rubble footing. That these walls were later additions is supported by the fact that their alignment differed very slightly from that of the original walls; they were also slighter and probably supported a timber-framed superstructure. It is possible that this was a porch, but more likely that it was a staircase leading up from the north to an upper storey. Within the staircase/porch was a narrow north–south slot, 3.8m long and 0.12m wide although widening out to 0.50m at the north end, which could have been a drain. The doorway in the south part of the added

BU RTON DA S SET T SOU T H E N D

end wall would thus have led into the lower room. The upper storey will have been added at the same time and it is plausible to suggest that it would also have been timber framed. Even discounting the space under the stair, the additional storey would have doubled the storage area to c41.3m 2 (445 sq. ft). The building may have been adapted to a different function at around this time, as the raised foor in the ground foor room seems to have been removed and replaced by earth and rubble fooring. However, the grain store may have been moved to the upper storey and only the ground foor converted to another purpose. A frstfoor granary would have had the same capacity and would have been further from the reach of damp and vermin, although rather less convenient of access. The character of the late 13th-century occupation on the excavated tenements There were relatively few fnds from these phases and nothing that would indicate the occupations of any of the inhabitants other than the smith in Area J. A few remarks can perhaps be made, however, admittedly of a highly speculative nature. The varying sizes of the properties and houses may indicate a social mix among the inhabitants. The large Area K tenement will have belonged to a wealthier tenant, possibly a trader in agricultural produce. The small house on Area E is likely to have belonged to someone of limited means, perhaps starting out on a new career. However, to rent a plot fronting the market place of a new town is the action of someone with ambition, not just someone poor. The larger house on Area A, possibly with an expensive slate roof imported from north Warwickshire, would have belonged to someone already reasonably prosperous. Although it is likely that these properties were among the ones let after 1280 without any holdings in the open felds (see Section 1.7 above), it would have been possible for the inhabitants to acquire plots of land at least as subtenants for their own subsistence from other tenants. In many small towns, such as Kineton and Bretford, it was normal practice for new burgage tenants to be furnished with smallholdings for subsistence (Goddard 2011, 8). Further development south of the road to the west of the excavated areas On the south side of the road on Dovehouse Close there was further development to the west of the excavated tenements, extending c160m westwards along the road. The areas identifed as Z2 and Z3 in the 1987 feldwalking (Section 1.12 above) each seemed to represent properties or blocks of properties bounded by slight earthwork/rubble banks marking the lines of boundary wall (Figure 1.19). Within these were spreads of rubble, roof tile and slate belonging to a number of buildings (Figure 1.21). Given the

T H E A RCH A EOL OGIC A L SEQU E NC E

evidence of the excavated area it is quite possible that other buildings also did not show as surface rubble scatters, although a suffcient area was stripped in the north-east corner of Z2 (Area M) to show that no buildings were fronting the road at this point. The visible rubble spreads in Z2 covered an area measuring c60m east–west by c40m north–south set back 30–40m south of the road, suggesting the presence of multiple properties or a very large farm complex with a number of substantial buildings, at least one with a slate roof and two or more with roofs of ceramic tiles. The pottery evidence suggested that Area Z2 was occupied from the 13th century to the late 15th/early 16th century. In Z3 the rubble spreads were more diffuse, but there was another concentration of roof tile on the eastern side of the area, c40m south of the road and measuring c40m by 40m, indicating the presence of a further building or buildings on a similar line to those on Z2. Too little pottery was recovered from Z3 to suggest a date for its occupation but evidently the settlement had expanded westwards. 2.5 THE CHAPEL

Towards the end of the 13th century the success of the market and the expansion of Southend necessitated the provision of a chapel at which the inhabitants could worship, although inhabitants would still have attended the parish church for Easter communion, burials and other important events. The site chosen for the chapel was at the original core of Southend, away from the new developments but close to the assumed road junction, and its construction involved the demolition of an existing building, possibly a barn. The historical context of the new chapel is discussed above (Section 1.7). It was a rectangular, single cell, stone building measuring 12.6m by 5.75m with diagonal corner buttresses. It was arranged in three bays with doorways, facing north and south, in the centre of the western bays, facing windows in the central and eastern bays, similar to that in the west wall, and probably a larger window in the eastern wall (Figure 2.20). It was probably divided into nave and chancel although no trace of any division survived. The non-matching imposts of the doorways might be dated to the mid-13th century, but these appear to be re-used from another building, and the late 13th-century date (c1290) of the former window in the west wall seems to be the best guide to the date of the new chapel. A desire for economy may be evident in its construction, both as it is the smallest, simplest kind of church possible, and reused stonework seems to have been used for the door imposts and in the walling fabric. The excavation undertaken in 2003 (online archive, Section 7) prior to conversion to domestic use revealed little about the medieval chapel or, for that matter, evidence for medieval activity in what is presumed to be the core of the Southend settlement,

71

in stark contrast to the dense concentrations of building remains, demolition rubble and rubble surfaces packed with fnds evident around the excavated tenements. No contemporary foor surfaces or internal fttings were revealed. The foundations of the west wall were 0.75–0.80m wide with a chamfered plinth 8cm wide on its west side. No new details were revealed of the north wall of the chapel or of the surviving c2m length of the south wall. Nor was there any real trace of the east wall. This must have bridged the early wall foundation and there were ragged gaps in this at the assumed crossing point. The rectangular single-cell plan is the simplest possible form of church. Comparable small rectangular churches have been excavated at Thrislington, County Durham (Austin 1989, 18– 24), Broadfeld, Hertfordshire (Klingelhöfer 1974, 18–19, 27–28, 32) and at St Mary Tanner Street, Winchester, Hampshire (Biddle 1968; 1969; 1970; revised dating: Biddle 1990, 1167). Thrislington, like Southend a rural dependent chapel, was built in the late 11th/12th century. It measured 9.54m by 3.84m, had a single south-west door, and was divided into nave and chancel by an inserted partition. In the chancel there was a step up to the sanctuary. In the 13th/early 14th century wall benches were built around the nave and west end of the chancel, probably in two stages. Broadfeld was a rural chapel that became a parish church. Its original plan, dating to the early 13th century was rectangular, measuring 11.9m by 5.8m; it had facing north and south doors and benches around the walls. In the mid-14th century the church was lengthened and a chancel step inserted. St Mary Tanner Street, Winchester, was a small urban church on a cramped site, dating originally to the 10th century. From the late 11th century to the 14th century, it was rectangular, 12.5m by 5.8m, divided into nave and chancel, with wall benches around the nave, and a sanctuary step across the chancel. These excavated examples, together with the many instances of extant buildings where the initial layout may be adduced from inspection of the plan after subsequent phases of development, suggest that the form of the Southend chapel was not uncommon for the smallest type of church. They also demonstrate that the provision of benches around the walls and a division into nave and chancel were fairly standard arrangements which could also be expected at Southend. An inserted partition dividing off a chancel in the eastern bay could easily have left no trace in the main walls. At Broadfeld, where there were also two facing doors, the southern one, as at Southend, seems to have been the more important, although it is diffcult to say whether this has any general signifcance (Klingelhöfer 1974, 36). The chapel is likely to have remained in repair and in use until at least the end of the 15th century. No signs survive of alterations during the

72

BU RTON DA S SET T SOU T H E N D

Figure 2.20 Plan of the chapel (probably late 13th century) and 17th century Priest’s House, with elevations of the chapel north wall

T H E A RCH A EOL OGIC A L SEQU E NC E

medieval period, although the evidence from the excavated examples already cited is that periodic internal re-arrangements were common. After the abandonment of the buildings at Southend it may have fallen out of use and into decay, although chapels often survived longer than the settlements they had originally served. It was refurbished, with the addition of the adjacent Priest’s House, for a brief period of use during the 17th century by the recusant Lady Wotton (see Section 2.11 below). 2.6 THE APOGEE OF CHIPPING DASSETT IN THE EARLY 14TH CENTURY

The excavations produced abundant evidence of further expansion of the settlement in the early 14th century with new buildings on both sides of the road (Figures 2.21 and 2.22), the provision of a water hole and some re-arrangement of boundaries. This confrms the historical evidence that it was in this period that the settlement reached its peak of prosperity, with the growing use of the name Chipping Dassett, frst recorded in 1295, and references to a land market and a number of occupational names (see above, Section 1.7). Area K, rebuilding of house (K3) In the early 14th century the Area K2 house (above, Section 2.3) was rebuilt on a substantially larger scale (K3) (Figure 2.23). The new house had the largest footprint of those excavated, at 19.8m long by 7.2m wide with a total foor area of 89.49m 2 (963 sq. ft). The walls of the house were constructed of medium–large roughly coursed ironstone rubble facing a core of small–medium rubble bonded with brown/yellowish-brown clay loam. The north wall was 0.70m wide. About 6m from its eastern end there was an imposing doorway to the street, 1.5m wide with an internal splay opening to 1.9m wide. The wall also contained an external buttress, 0.35m wide by 0.15m deep, about 8m from its western end. The south wall varied in width from 0.75m to 1.00m. To the west it had a wide offset footing. There was a southern doorway opposite that in the north wall. This was less well preserved but appeared to be narrower (1.2m externally) with an internal splay (to 1.6m) only on the east side. The east wall and west walls, at c0.90m wide, were slightly wider than the north and south walls, which perhaps means that the building had gabled ends. The original internal arrangement of the house seems to have involved three divisions: a central hall, 9.05m long by 5.4m wide, with opposing doorways at its eastern end, a parlour 3.35m long to the west, and a service end (4.2m long) to the east below the hall. No internal features of this phase were excavated at the west end of the building within the parlour, but some were recorded in the hall and service end.

73

The earliest internal features and foor layers in the hall were only exposed in the western of two 1.5m wide sample trenches excavated across House K, and it is possible that some of them may have belonged to the earlier (K2) phase. Towards the south end of the trench were a post-hole, 0.35m across by 0.10m deep, and a shallow hollow, 0.60m by 0.47m by 0.06m deep, which was sealed by the earliest surviving foor of dark greyish-brown clay loam with brown fecks. Set into the foor was a small circular hearth/burnt patch, in a shallow hollow 0.43m across and 0.06m deep. To the north of this was a series of discrete but probably contemporary patches of rubble foor surface which extended beyond the sample trench both eastwards and westwards. These early foors were overlaid by another, which covered most of the hall area and was probably contemporary with a similar brown clay loam foor around the north door. At the east end of the hall, just to the west of the door passage, there was a hearth, 0.47m by 0.45m, of fat red burnt ironstone blocks, surrounded by a patch of ash on its eastern side. It seems likely that the service end was originally divided from the hall by an ironstone rubble wall which survived for only 2m, but it is likely to have run right across the building, creating a service end 4.2m long. Running eastwards for 1.33m from this was another foundation of similar construction; it was not certain that it belonged to this phase but its similarity to the north–south dividing wall and the fact that it appeared to have been set against an extension of it, make it likely. It presumably subdivided the service end into two rooms, one to the north measuring 2.25m by 4.2m and a second to the south 2.65m by 4.2m. This would correspond to the standard division of the service end of a hall-house into ‘buttery’ and ‘pantry’. The plan of this house is that of the classic medieval hall-house, where bays had specialised functions (Suggett 2018), including hall, parlour/chamber and two service rooms, ‘buttery’ and ‘pantry’. It was the only example excavated at Dassett Southend to match this standard domestic layout exactly, but its plan and size confrm the higher status of the inhabitants of the large Area K tenement. Extension of Area H house with upper room (H3) At some point, probably also in the early 14th century, the Area H house was extended to the west by 1.8m with the addition of an extra bay divided into two small rooms, the northern of which probably contained a stair to an upper room over the western bay of the house (H3) (Figure 2.24). The extension foundations were of medium–large rubble. The south wall was 0.65m wide and the west and north walls were 0.60m wide. There was probably a doorway in the south part of the west wall, although any trace of the door surround had been robbed out.

74

Figure 2.21 Phase plan, c1340

BU RTON DA S SET T SOU T H E N D

T H E A RCH A EOL OGIC A L SEQU E NC E

75

Figure 2.22 Dassett Southend in c1340. The drawing aims to show the excavated area at the peak of its prosperity or just after. The properties along the north side of the road are all shown occupied according to the excavation. Area N and the plot to the west of the smithy are shown as vacant, with the water hole impinging on the latter. There may also have been a house on Area L at this time, but these levels were not excavated. It is likely that the properties further east along the north side of Barn Close were also built up, but the buildings here are vaguely indicated as there is no defnite evidence for this area. Despite the wide market place street and the name of Chipping Dassett the settlement form is that of a village rather than a town. The houses are aligned along the frontages rather than at right angles to them, the frontages are by no means fully built up and the general density of buildings is low. On the other hand the planned rows with plots of equal width and close proximity of some of the buildings are not wholly typical of a village

76

BU RTON DA S SET T SOU T H E N D

Figure 2.23 Area K, Phase K3, early 14th century (compare with Figure 2.10; the house has now been enlarged and sub-divided)

The extension was divided into two small rooms by a short partition wall of clay-bonded coursed rubble. The northern room was 1.2m by 1.7m, with a foor of well laid medium–large ironstone fagstones. The most likely interpretation of this room seems to be that it contained a staircase leading to an upper room over part, at least, of the house. The southern room, 1.3m long by 1.6m wide, was also foored with medium–large ironstone fagstones, and appears to have functioned as a passage or lobby to the door in the west wall (Figure 2.25). Whether the insertion of a foor required the rebuilding of the west end of the house is unknown, but the addition of the stair may suggest that the work was more extensive than just the fooring over of the roof space. The stair, lobby and upper foor would have increased the total foor area of the house to 52.12m 2 (561 sq. ft). The addition of upper rooms to Midlands peasant houses may not have been widespread in the 14th century, although there is some documentary evidence (Dyer 2013, 112), and Alcock and Miles (2013) record one standing example in Steventon, Oxfordshire, dating to 1365/6 (STE-C) and three with upper foors in cross-wings, in Steventon and Sutton Courtney, Oxfordshire, dating to AD 1317/8, 1350/1 and 1355/6 (SUC-A, STE-E and STE-B). A soakaway drain, varying in width between 0.29m and 0.45m, packed with pitched medium–large rubble, ran southwards for 10m from the south-west corner

of the extension and was presumably contemporary. To the west of the house the northern boundary wall was replaced by a wider stone wall of rubble on a footing 0.75–0.85m wide, with a narrower (0.40m) southwards return at its west end. It is possible that these, together with the west wall of the house supported a small lean-to shed or covered area to the west of the house, measuring c4.5m by 2.5m. Construction of water hole (W1) The early 14th century also saw the construction of a communal facility in the form of a water hole in the street north of Areas H and I (Figures 2.8 and 2.26). This involved some re-arrangement of boundaries and the apparent acceptance that this latter property was never going to be occupied. Activity in area W prior to the water hole was evidenced by an early 14th-century pit, over 0.55m wide and 0.50m deep, which contained a bone tuning peg (similar to the peg from House I2); this had been cut by the eastern end of a ditch, 1.79m wide and 0.40m deep, which ran westwards for at least 3.2m, but was then truncated by the later water hole. The inflling of the ditch with clay loam may well have immediately preceded the construction of the water hole. The water hole occupied much of the area to the north of Area I. It consisted of a large sloping-sided hollow, c4m wide by c0.5m deep, running down to the south and east sides of the boundary ditch

T H E A RCH A EOL OGIC A L SEQU E NC E

77

Figure 2.24 Area H, Phase H3, early 14th century; an extension divided into two small rooms has been built against the western end of the H2 house

around the smithy tenement (Area J). The hollow would have enabled animals to drink easily from the ditch, which was c0.45m deeper and presumably went below the water table. The stone revetments and boundary walls now added to the north and west sides of the ditch around Area J would have operated to prevent animals escaping onto the tenement. The hollow was 12m from south to north and tapered northwards; its westward limit was unexcavated but was over 12m from the east edge. Between the water hole and the houses in Areas H and I there was a level street surface c3m wide. Both it and the water hole were presumably paved with ironstone rubble although the original surfaces did not survive. The water hole would have been unsuitable as a source of domestic water. There would have been no fow and it would easily have become stagnant,

quite apart from pollution by animals. It would also have been very silty. Wells would have been a more suitable domestic water source, but none was encountered, although it is possible that there was a communal well in an unexcavated part of the main street. The nearest springs are on the Dassett Hills some 900m away, while the nearest stream is 300m to the south-west (although it is not clear whether this would have been a viable source). The water hole was thus presumably meant for watering animals and would have been particularly useful for those bringing animals to market. It appears to have been a communal feature: it was constructed on too large a scale for the animals of a single household and was also located on public space, although its location tucked away down a side street is a little curious.

78

BU RTON DA S SET T SOU T H E N D

Figure 2.25 Area H, house, outbuilding and yard, from the west (Phases H2–H5). The early 14th-century two-room extension is just beyond the ranging rods

Figure 2.26 Area W, water hole (in use from the early/mid-14th century to around the middle of the 15th century) and adjacent surfaces, from the east (Phase W2)

T H E A RCH A EOL OGIC A L SEQU E NC E

79

Figure 2.27 Area I, foundations of possible privy/shed, from the north (Phase I3, 14th century)

Communal water holes or ponds with stone paving or revetments are not uncommon on village sites. At Hangleton, Sussex, there was one measuring 15.2m by 9.1m by 1.5m deep (50ft by 30ft by 5ft) with a wall along its northern edge (Hurst and Hurst 1964, 115, fg. 1, pl. 3B). This contained mixed 13thto 15th-century pottery. At West Whelpington, Northumberland, a pond was located towards the south edge of the green. It was not investigated but had a straight edge towards the Site 16 buildings which suggests that it was man-made. It is shown on both Period I (late 11th/12th to early 14th century) and Period II (14th to 17th century) plans (Evans et al. 1988, fgs 114 and 115). There was another at Wythemail, Northamptonshire, in the centre of the village, also not investigated, but apparently medieval with a stone lining (Hurst and Hurst 1969, 172). In Warwickshire a probable shallow water hole, 20m by 8.5m across, fed by drainage ditches and surfaced with rubble was found at Goldicote fronting a street towards the western end of a row of tofts on the southern side of the village (Thompson and Palmer 2012, 99, 133, fg. 28). Its fll contained 12th/13thto 15th-century pottery. Enlargement of Area I tenement (I3) Probably at the same time as the construction of the water hole, as part of the re-arrangement of boundaries (although it may have happened later in the century), the eastern half of Area I was extended

northwards with the addition of a square plot c9m by 9m carved partly out of the former street and partly out of the former property to the north (Figure 2.8). Running northwards for 9m from a point 5.5m from the east end of the north wall of the house on Tenement I was a 0.65m-wide boundary wall faced with large blocks, some of which were re-used. About 1.5m from the house there was a gateway through the wall, 0.9m wide. At its north end the wall turned eastwards out of the excavated area. Running parallel, 7.5m to the east, a short section of another north–south wall, 0.78m wide and constructed of large rubble blocks, ran 1m into the excavated area before ending. These walls were probably too far apart to have supported a building and should probably be seen as stone boundary walls enclosing an open area, probably a yard, 7.25m (E–W) by 8.25m (N–S). The eastern of these walls probably marked the eastern boundary of the Area I tenement. Built against the north-east corner of the house was a small, open-sided shed, 1.8m by 1.8m (Figure 2.27); it was presumably of timber, set on stone footings. The east wall rubble footing, 2.1m long by 0.6m wide, overlapped the corner of the house, whilst the north wall footing was 0.37m wide. The shed was f loored with medium rubble bonded with olive-grey clay loam. It was probably a privy, containing a barrel latrine as there was no cess-pit below. It size and location were similar to those of other probable privies on D23 and K4.

80

The foor of the privy was continuous with a rubble surface which extended across the yard up to the gateway in the western boundary wall; to the north it appeared to be petering out close to the edge of the excavated area. While the rubble surface was in use a layer of dark greyish-brown/olive-grey clay loam accumulated over it. It is likely that this was the remains of a muck heap on the surface. It contained moderate amounts of pottery (231 sherds) and animal bone (114 fragments), a carpenter’s iron reamer and a stone spindle-whorl. To the east there was a gap 1m wide between the privy and the eastern yard wall which provided an access into the walled area from the east. The privy was not particularly convenient of access from the house. The quickest route would have been out of the front door into the street and then in through the side gate, which perhaps seems an unlikely arrangement. Two successive small rectangular stone structures, interpreted as garderobes, were recorded on the corner of a mid-14th to early 15th-century house at Great Linford in Buckinghamshire (Mynard and Zeepvat 1992, 71–72, fgs 22–23, Croft F, Building 16). These also lacked convincing cess-pits below, being only 30cm deep and without the characteristic organic fll. Free-standing garderobe pits at Goltho in Lincolnshire and Barton Blount in Derbyshire (Beresford 1975, 44, fg. 21) were about 1.0m square and 0.6m deep with post-holes to support seats (but no evidence of covering structures). These also seem rather shallow. Cess-pits in the settlement at Oversley, Warwickshire, were 1.0m and 1.6m deep (Jones et al. 1997, 44) and possible cess-pits at Upton, Gloucestershire, were 1.8m and 2.1m deep (Rahtz 1969, 91–92, fg. 4). Enlargement of Area K Another similar re-arrangement of boundaries may have taken place at the west end of the street at the same time, involving the encroachment of Area K across the section of street south of Areas L and N. This area lay later within the curtilage of Area K and the later boundary wall running north from the house was on the approximate line of the putative eastern boundary of Area N. This encroachment would have removed the rear access to Area L, but it may have been recognised that, like the plot to the east of Area J, Area N would never be occupied, and a new access to the rear of Area L made across it. Rebuilding of Area E house (E3) The rebuilding of the Area E house involved the addition of a large east–west range onto the east side of the original E2 building (Figure 2.28). The total foor area of the building was now 74.72m 2 (802 sq. ft), which is about three and a half times that of its predecessor. In effect, the original building was

BU RTON DA S SET T SOU T H E N D

converted into a northwards projecting cross-wing at the west end of a larger building along the frontage. A new south wall was built, slightly to the south of the previous wall which was robbed out. The previous north and west walls were retained, but the south part of the original east wall was removed, and a narrow partition inserted dividing off a small room (3.5m by 2.5m) in the north of the cross-wing from the new range which ran through to the south. The partition wall ran at right-angles to the west wall. Its foundation was made of small–medium rubble and clay loam set in a trench 0.30m wide. This would have supported a timber wall. There was a doorway 1.0m wide at the western end. The new range was 17.4m long (including the retained part) by 5.75m wide. Its south and east walls had been set in construction trenches; both walls had been robbed out. A section of the rubble north wall, 0.70m wide, remained in situ. The widths of the surviving walls and their construction and robber trenches suggest that the walls were stone-built to the eaves. In the angle between the new range and its cross-wing there was a timber porch, 1.75m by 1.90m, supported by two post-holes, one 0.45m across by 0.14m deep, the other 0.42m in diameter by 0.10m deep. It is likely that the outward doorway in the porch opened to the east. The new internal arrangements in the house were not completely clear. No defnite trace of a partition separating off the western end of the main range was found and the hall may originally have run through, making it 10m long by 4.5m wide. The latest foor layers in the central and western parts were different, however, suggesting that they may have been divided by an insubstantial partition into a west room (c3.75m by 4.0m) and a hall (c6m by 4.5m). Given the presence of a porch on the north side in the angle of the cross-wing it seems that the north door was just to the west of the main central hearth, and it is possible that there was an opposing southern door although there was no defnite evidence of this. By the end of the phase, an eastern room (5.0m by 4.55m) was divided off from the central hall. It is probable that this was also the original arrangement. It is conceivable that there was an outside doorway in the south-east corner of the eastern room, although the only evidence of this was a spread of rubble outside that may have been a threshold. The evidence is slim and the door remains only a possibility. A separate entrance here might suggest temporary multiple occupancy or possibly use of this room as a shop. The eastern room seems to have been divided from the hall by two opposing spur walls at an angle to the main axis of the hall, with a gap in the middle 1.3m wide. On the south side the footing of medium– large ironstone rubble set in a sub-rectangular pit, presumably for a timber wall, was an original feature. The footing on the north side, 1.86m long by 0.80m wide, was of large, fat, ragged ironstone and Lias

T H E A RCH A EOL OGIC A L SEQU E NC E

81

Figure 2.28 Area E, Phase E3, early 14th century (compare with Figure 2.19). The original house is now the cross-wing of a much larger house, with a probable porch in the angle between the two wings, and the frst of a series of probable brewhouses to the rear rubble blocks packed with dark greyish-brown/olive brown clay loam and small ironstone rubble and set in a shallow hollow. This may have been a replacement for an earlier footing on a similar line. The western room was foored with layers of clay loam and small rubble, surviving in discontinuous patches, which extended along the south edge into the east end of the hall, perhaps indicating the presence of a doorway at the south end of the probable partition. There were patches of similar fooring in the room in the cross-wing to the north. No early foor surfaces survived over the eastern part of the hall, but there

was a central hearth on the west side. This consisted of a fne, rectangular stone platform, 0.75m by 0.70m by 0.19m thick, of seven closely ftting ironstone blocks with a rounded moulding forming a kerb around the edge and set in a shallow construction pit (see online archive, Section 8.10). This was the only hearth excavated with any hint of decoration and it remained in use throughout the phase. Just to the east of the hearth there were three post-holes, all with charcoal and ash in their flls, which may have supported cooking apparatus, although it is possible that they belong to an earlier phase. Otherwise, apart

82

from a couple of shallow hollows, the hall and the western rooms had no observable internal features. The eastern room contained several features although the original arrangement here had been obscured by later activity. Many relationships were not clearly defned by the stratigraphy; most of the features could actually belong to an earlier phase, one or two to a later phase. In the centre of the room there was a shallow hollow, 1.30m by 1.00m by 0.2m deep, with a rectangle of fat stones (0.80m by 0.30m) set in the bottom. The stones were burnt and it is likely that this was the remains of a hearth. To the north-west of the hearth were two pits/hollows of uncertain function. Along the south side of the room was a group of three post-holes, 0.22–0.26m in diameter and 0.10–0.23m deep, and there was another group of four post-holes, all stone-packed, in the south-east corner, 0.22–0.35m diameter and 0.15–0.27m deep. In between these two groups was another group of three, more doubtful, shallow post-holes, 0.17–0.33m diameter and 0.04–0.06m deep, whilst to the north were another genuine post-hole, 0.25m in diameter by 0.14m deep, and another less convincing example 0.25m by 0.21m by 0.04m deep. These features, or some of them, may have supported fxed furniture. Although as suggested by the possible external door, this eastern room may have been a separate unit, its position at the ‘upper’ end of the hall suggests that it may have been a parlour or chamber. Also in the north-west corner of the room, aligned east–west (on the line of the walls), an oval pit, 0.60m by 0.40m by 0.24m deep, contained the near complete, although fragmented, skeleton of a baby (Figure 2.29), probably a neonate (online archive, Section 8.15). These bones, therefore, probably belonged either to a premature newborn or to a baby who died during or just after birth. The skeleton had been laid on its back with its head to the west; its legs were slightly fexed to the right. The pit was flled with brown clay loam, and its top appeared to have been sealed with yellowish-brown clay. The circumstances of this burial must be a matter for speculation. Presumably the child was unbaptised and could not be buried in the churchyard, but whether such careful burial within the house indicates a desire for concealment because of illegitimacy or simply the emotional desire of a mother not to be separated from a child cannot be said. Stillborn, unbaptised babies may have been treated as less than human; in the 17th century a midwife’s oath included an instruction to see that a stillborn was buried in such a secret place ‘as neither any hog, dog nor any other beast may come unto it … [nor] any such child … be cast into the lanes or any other inconvenient place’. The inclusion of such a clause indicates that the decencies of Christian burial were not always observed (Gittings 1984, 83). Other infant burials have been found in and around medieval houses in a variety of circumstances.

BU RTON DA S SET T SOU T H E N D

Figure 2.29 Area E, neonate burial against the north wall of the eastern room, from the east (Phase E3, early 14th century)

It is suggested that this refects the attachment of a peasant family to their toft (Astill 1988, 58). Three burials of neonatal infants were recorded at West Cotton (Chapman 2010, 102, 165, 192, 538): one in a shallow pit in the corner of a barn/malthouse belonging to the 12th-century manor house, one in a barn building that was later converted to a peasant house, although it is not clear to which phase the burial belonged, and a third in a pit cutting rubbish deposits in a yard belonging to a peasant tenement. At Thrislington (Austin 1989, 76, 150, 197) a large part of the skeleton of a baby, perhaps stillborn, came from the fll of a stone-lined storage pit in a building on a peasant toft. In an age of high infant mortality these seem to represent discreet disposal of still births or unbaptised babies. At Tattenhoe, Buckinghamshire, a probable stillborn infant was buried in a shallow depression, aligned east–west with its head to the west and with three animal bones placed over its lower half (Ivens et al. 1995, 31, fgs 17–18). This lay apparently beneath the wall of a building and the animal bones suggested the possibility of a ritual foundation deposit. At Westbury, Buckinghamshire (Ivens et al. 1995, 147,

T H E A RCH A EOL OGIC A L SEQU E NC E

fg. 82) a 5- to 7-month-old foetus was buried in a small pit with its head to the west, also possibly under a 14th-century building. It was suggested that this may have been another possible foundation deposit, or, more likely the disposal of a stillbirth. At the Hamel, Oxford, an infant, new born or a few weeks old, was buried in a small pit cut into the gravel make-up beneath a new foor in a late 13thcentury house (Palmer 1980, 219, fche 1 D06, 2 E03). It was not clear whether the house was occupied or unoccupied at the time, raising the possibility of a desire for concealment of illegitimacy or possibly even a case of infanticide. Other urban examples of infant burials include a neonatal burial at Stert Street, Abingdon, Oxfordshire (Parrington 1979, 3, 16) and a 1-month premature foetus buried in a foor layer at St Aldates, Oxford (Durham 1977, 166). Perhaps the closest parallel to the Burton Dassett burial in the foor of an occupied house is that from Upton, Gloucestershire (Rahtz 1969, 87–88, pl. III, 123–124) where an infant 3 to 6 months old was carefully buried in the corner of a room in a 13th–14th century house, the burial being possibly associated with a spindle-whorl, which may have been a keepsake deposited by the mother (Gilchrist 2008, 133). Perhaps the attachment here was greater because the child had survived for some months. Brewhouse north-west of house E A sub-rectangular pit 1.07m by 1.10m by 0.13m deep and a post-hole, 0.52m by 0.38m by 0.13m deep, cutting the ground surface may belong to the earliest part of this phase. These features were sealed by the construction of an outbuilding, interpreted as a brewhouse (Figure 2.28). The building ran northwards from the north wall of the house; it measured 5.3m by 9.4m and probably had timber-framed walls set on stone footings. There was a wide (2.35m) doorway in the centre of the east wall. The north wall was 0.55m wide and made of small–medium ironstone rubble bonded with clay loam. The east wall was of similar construction although varying in width from 0.45m to 0.55m. The position of the doorway was marked by a stretch of better laid, larger rubble. The north part of the west wall was similar to the north wall, although the south part was composed of larger rubble and it is possible that the south part may have been a rebuilding or, since it survived higher than the other walls, that the upper parts of the wall had a different character to the foundations. Within the building a series of partitions divided off four alcoves along the north and west sides. In the north-west corner two rubble wall footings, set at right-angles, both 0.40–0.48m wide, divided the north end into two parts, one to the west 2.5m by 2.3m, the other to the east 1.6m by 2.6m. In the north-eastern alcove there was an oval hollow, 1.80m by 0.85m by 0.14m deep. This was

83

flled with very dark greyish-brown clay loam and charcoal fecks. In the later phase of the building this area was occupied by a malting kiln and this hollow probably represents the vestiges of an earlier one. In the north-western alcove there was a group of three features, two pits and a post-hole, which could either have pre-dated the building or belonged to its earliest period of use; these were sealed by a foor of brown clay loam. The western side of the building was divided into two further alcoves by a spur wall footing running east from the main west wall. This feature was 2.3m long, 0.45m wide and of similar construction to the two partitions at the northern end. The central alcove also contained a pit (which could have pre-dated the building) 1.45m by 1.04m by 0.22m deep. This was sealed by a layer of clay loam and small rubble which formed a foor in this alcove. On the south side of the southern spur wall was a length of curving gully, 0.26m wide by 0.14m deep, which contained the remains of a rubble stone lining. This ran into a slot in the main west wall, 0.20m deep, which, even though there was no outfall, was presumably a soakaway drain. The drain was initially taken to indicate that this part of the building was used to house animals. However, it now seems more likely that the drain was connected with malting or brewing and served a steeping vat or tank. A similar drain was found in a brewhouse at Great Linford, Buckinghamshire (Mynard and Zeepvat 1992, 60–63, Croft C, Building 13). Brewhouses are common additional outbuildings recorded on peasant farms and brewing was the most widespread of village trades (Dyer 1986, 25; 2002, 170), carried out mainly by women, often as a sideline by peasants’ wives to boost their incomes using grain from the holding. For some poor people, who bought their grain or malt from large-scale producers, brewing could have been their main source of income; whether this was the case for the inhabitants of Tenement E is impossible to say. House F3 The earliest surviving house on Tenement F dated to the early 14th century. It consisted of a slightly trapezoidal house set parallel to the street along the frontage (Figure 2.30) and was 5.6m wide by over 9m long. Its narrow (0.60m) northern wall of claybonded ironstone rubble, the only one surviving, suggested that it had a timber-framed superstructure on a stone footing. The south and west walls both lay on the line of later walls and all trace of them had been removed by later robber trenches. The earliest foor was formed by the top of a clay loam makeup layer. This was cut by a shallow (0.05m) gully, 0.43m wide, running north for c2.5m from the south wall, which may have supported an irregular timber partition with a doorway at its north end dividing off a western room c3.4m long, from a probable hall

84

BU RTON DA S SET T SOU T H E N D

Figure 2.30 Area F, house, from the east (Phases F3–F4). The south and east walls had been entirely robbed out

over 4.5m long, giving the house a foor area of over 32.2m 2 (over 347 sq. ft). No hearths survived in the excavated area. At a later date, a more defnite partition was inserted to make a smaller room, 2m long, at the western end of the building. This room was refoored with a layer of dark greyish-brown clay loam. The refooring and partition seem to have been contemporaneous, as the construction trench for the partition cut the foor layer. The partition itself, which was presumably timber-framed, was supported on a foundation of large ironstone and limestone fagstones. No further foor surfaces survived in the main room of the building and there was no trace of any hearth. On the north side of the room there were two adjacent shallow hollows cutting the foor which could possibly have been post settings. There was another similar feature on the south side of the room. Area D1 (D13): Building and use of house and probable barn/store; midden/muck heap (early to mid-14th century) At the beginning of the phase a new western boundary gully was dug, c0.5m wide by 0.3m deep, which ran up to the north-east corner of the Area A house. This cut off the space between the Area

A house and the original boundary, which appears to mean that the space had been acquired by the Area D1 tenement. The west wall of the new house overlapped the inflled original boundary gully, but there was still room for a pedestrian access between the buildings, 2.2m wide, to the rear of the D1 tenement (Figure 2.31). The house was set parallel to the frontage in the south-western corner of the tenement. The south side of the house had been completely removed by later ditches but if it was built up to the same frontage as the Area A house it would have been c5m wide by 7.3m long, with a foor area that can be estimated at 20.13m 2 (217 sq. ft), making it the smallest of the houses recorded. The north wall was 0.7m wide, the east wall 0.6m, and the west wall 0.6–0.8m wide. It seems likely that these rubble walls were thick enough to have been built of stone to the eaves. Set against the north-east corner of the house there was a roughly rectangular foundation, 2m by 1.8m, and of the same construction as the walls. Its purpose is uncertain. It seems unlikely that a building this small had two storeys and that this foundation supported a staircase. It was too big for a chimney base. A more likely interpretation would be an oven-base, but similar features at West Whelpington, Northumberland, were interpreted as bases for hay ricks (Evans and Jarrett 1987, 297–299, fg. 72).

T H E A RCH A EOL OGIC A L SEQU E NC E

85

Figure 2.31 Area D1, house and vestiges of ancillary building (Phase D13, early to late 14th century)

In the forecourt area south of the house was a hollow, 1.4m by 0.66m by 0.11m. There would probably also have been rubble surfaces leading down from the house to the street but no trace of these survived. To the east of the house, set back from the frontage by c3m, there was a spread of medium rubble which seems to have represented the south-west corner of another building, most of which had been completely removed. The west wall was 0.5m wide, the south wall 0.48m wide, which suggests that they were footings for a timber-framed structure. The rubble within the angle of the walls had a roughly straight north edge; this may have resulted from truncation by later features but, alternatively, it could have marked some internal structure. This building is most likely to have been a barn or store. To the south of the putative barn there were two stone-flled hollows that could have been post-bases supporting

part of a structure against the barn wall; to its north were two further hollows which could have been within the barn but could equally have pre- or postdated it. Running northwards from the north-east corner of the house there were two lengths of a north–south gully, separated by a gap of 1.5m. The gap may have been a deliberate access way into an enclosure against the west side of the property. There was another gap of c1m between the south end of the gully and the corner of the stone foundation by the north-east corner of the D13 building. To the west of the gully, within the enclosure, there was a group of large shallow pit/hollows which may have represented the remains of a midden/muck heap, the hollows being formed by the removal of rubbish to the felds. The pits ranged in size from 0.3m by 0.4m by 0.05m deep to c3m in diameter. These features did not contain large numbers of fnds (a total of only 60

86

BU RTON DA S SET T SOU T H E N D

pottery sherds and 20 fragments of bone from all of them), suggesting that the muck heap here may have been formed of animal and crop processing waste and ash, rather than domestic rubbish.

post of a gate across the access to the rear of the tenement. In the northern part of the tenement the sole evidence of activity recorded consisted of fve pits/hollows.

Area A (A3): Recutting of boundary ditches, kitchen extension to house

Area D2 (D23): Continued occupation of house and construction of outbuildings

By the early 14th century the frst eastern boundary ditch had been flled with clay loam. It was replaced by another ditch, c0.90m wide by 0.46m deep; this was much shorter than its predecessor, extending only 11.5m northwards from the north-east corner of the house (Figure 2.32). Further north, the tenement boundary was now presumably marked by a fence or hedge. The western boundary ditch was also recut slightly to the east. At the south end there was a gully, 0.45m deep by c0.8m wide which was then cut by another ditch, c1.2m wide by 0.5m deep, which may itself have been recut. The north part of this ditch had also been removed by later features. The ditch across the centre of the tenement, perhaps marking the edge of a kitchen garden at the rear of the tenement, was also recut at this time. The new ditch lay c1.3m to the north of its predecessor and could have been dug to the north of an existing fence or hedge. It was up to 1.05m wide by 0.30m deep. Although truncated by a later cut of the west boundary ditch, it is likely that the ditch ran into the continuation of the western boundary ditch. The only access into the northern part of the tenement from the south would have been the 2m gap at the eastern end unless there had been a bridge over the ditch. Just to the north of the house there were two lengths of shallow, east–west gully, 0.70m wide, that may have served for drainage. There was no evident change to building A during the early part of the 14th century, but at some stage in the mid-14th century an extension was added onto the north-east corner of the house, probably to serve as a kitchen. This measured 2.7m by 4m and had timber-framed walls supported on stone footings. The footings of the west and north walls were supported on a roughly L-shaped spread of medium–large rubble. Of these footings themselves only a 1.1m length of the north one survived, 0.35m wide and made of small–medium rubble. The eastern footing also consisted of a ragged spread of medium rubble. No foors survived but, set into the centre of the room on the west side, there was a large heavily burnt hearthstone, 0.35m by 0.30m across. The use of the hearth produced much ash and debris but, since the kitchen seems to have remained in use for the rest of the life of the house, most of this will have been later. A number of shallow pits contained material of this date to the west and north of the house. Cutting one of these, against the west wall of the house, was a post setting, 0.35m by 0.32m, packed with pitched stones which could have supported the

The Area D2 house (Figure 2.32) continued to be occupied in the 14th century, apparently without any major internal re-arrangements (D23) (Figure 2.18). A possible location for the north door c2.5m from the west end of the house is suggested by the presence of an extremely irregular hollow externally, 2.4m by 1.0m, that may have been created by the passage of feet across the threshold. This was cut by a series of features, some of which may have supported a timber porch, c1.5m by c0.6m. There were two post-holes on either side: one to the west (0.52m by 0.43m by 0.17m deep and 0.35m in diameter by 0.15m deep), and one to the east (0.52m by 0.42m by 0.20m deep, and 0.52m by 0.44m by 0.04m deep). A fnal post-hole, 0.43m in diameter by 0.22m deep, with an ashy fll, is more diffcult to explain. This would have been in the middle of the doorway but might have been the result of repair to the threshold. Inside the house, the only trace of fooring of this phase in the hall was a patch of olive brown clay loam with some ash and charcoal which had subsided into a shallow hollow, c1.85m by 1.5m by 0.1m deep. To the south-east there was a group of stakeholes which could have been associated with the hearth, perhaps to support cooking apparatus, which seems to have remained in use from the previous phase. There were seven stakeholes, 2–7cm in diameter and 2–7cm deep, all with high proportions of charcoal in their flls. The fact that the stakeholes crossed the line of the southern spere wall suggests that the wide gap in the partition remained. In the western bay of the building, in the centre, a hearth, 0.85m by 0.40m by 0.06m deep, consisted of a hollow flled with layers of olive brown clay and red/dark red ashy clay loam. At the east end of the building some post-holes may have supported internal fttings: one against the north wall, 0.27m in diameter by 0.14m deep, and two towards the south-east corner, one 0.21m in diameter by 0.07m deep, the other c0.25m in diameter by 0.08m deep, flled with olive brown clay. In addition, there was a square rubble block, 0.20m by 0.20m, in the centre of the foor which could have been a post-base. Set against the north part of the west wall of the house there was a small timber outbuilding or shed, measuring 3.75m by 2.4m, with its structure supported on earth-fast posts. The north wall contained three rubble-packed post-holes (0.24m in diameter by 0.12m deep, 0.31m by 0.20m by 0.16m deep, and 0.25m in diameter by 0.16m deep) and its east end was set in a shallow slot 1.53m long

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.32 Area A, phases A3 and A4 (14th and early 15th centuries)

87

88

by 0.29m wide. The south and west walls were supported by two post-holes at the south-west corner (0.22m by 0.20m by 0.10m deep and 0.20m by 0.19m by 0.07m deep). A patch of rubble may have marked the threshold of a doorway into the centre of its north wall. Just to the north of the outbuilding, midway along its north wall, there was a patch of medium rubble, 0.78m by 0.72m, which might have paved the threshold of a doorway into the outbuilding. Within the outbuilding in the north-western corner were a hollow, 0.4m by 0.36m by 0.04m deep, and a possible post-hole, 0.72m by 0.35m by 0.24m deep. Immediately to the west of the outbuilding there was a large pit 1.4m in diameter by 0.32m deep, cut by an irregular animal disturbance through one side and by a stone-packed post-hole, 0.47m by 0.26m by 0.09m deep, which together with another to the west, 0.57m by 0.26m by 0.07m deep, may have supported a fence running westwards from the outbuilding. To the south and west of the large pit there were other hollows. Area D2 Stable/byre During the life of the tenement the area to the northwest of the house contained a series of outbuildings, mainly of timber. These produced a mass of postholes which proved diffcult to disentangle (Figure 2.18). Using the stratigraphic, spatial and fnds evidence, four successive outbuildings have been identifed with a degree of confdence, although the attribution of all features to particular buildings may not be completely accurate. The earliest of these outbuildings appears to date to this period. It was constructed of timber, 6m long by 4.2m wide, with walls supported on earth-fast posts. The post-holes were at irregular intervals and the building did not seem to have had a regular structure. The 11 postholes identifed as belonging to the external walls ranged in width from 0.15m to 0.66m in width and 0.06m to 0.16m in depth; no foor surfaces survived, but a putative internal partition was represented by two further centrally placed post-holes, one 0.23m diameter by 0.07m deep, the other 0.28m by 0.20m, which would have divided the structure into two slightly unequal compartments 2.75m and 3.25m wide. These sections were of a size to accommodate animals either in stalls, if the partition did not reach right across the building, or in completely separate compartments. The structure seems likely to have been a stable or byre, rather than a small barn, despite the absence of any internal drain. It is likely that any door(s) would have been in the east wall, but no trace was evident. About 3.5m to the east of the possible stable/ byre there was a north–south ditch, U-profled, c1.2m wide by 0.40m deep, and extending from a point 3.5m north of the house to beyond the limit of excavation. This ditch may have divided

BU RTON DA S SET T SOU T H E N D

the rear of the whole tenement into two parts. A number of other shallow hollows and post-holes were located on the tenement to the rear of the house. These were less dense on the eastern part of the plot, east of the north–south ditch, but just outside the north-east corner of the house was a row of three small post-holes (0.25–0.29m wide by 0.06–0.17m deep) with a short slot 0.9m long by 0.3m wide by 0.05m deep. These features probably supported the timbers of a small lean-to shed, c1m square. This could well have been a privy. It would have contained a barrel or bucket latrine as there was no cess-pit below it. Similar structures were found on Areas I3 and K4. Tenement J (J3) The early 14th century saw alterations to the boundaries of Tenement J (Figure 2.33). The construction of the water hole in Area W involved the redigging of the boundary ditch around the south part of Tenement J. The earlier ditch was flled in and replaced with a steep-sided cut revetted on its northern side with small–medium rubble packed with clay. At the south-east corner the stonework of the revetment was absent, but the ditch cut could be made out. This phase of boundary did not survive along the eastern side, but it seems likely that the ditch would have extended northwards along the existing boundary and that the water hole was fed by the boundary ditch. The western boundary ditch was recut at about the same time, presumably as part of the same operation. The new cut was over 1.1m wide by 0.6m deep and was also revetted with coursed large ironstone rubble on its eastern side. This revetment seemed to coincide with the construction of a boundary wall along this side of the property. The wall survived only as fragments but was c0.45m wide and composed of small and large ironstone rubble blocks. The wall extended from the south-west corner of the smithy J3 and was placed immediately on top of the original boundary gully. The remnant of another, similar wall, 0.55m wide, extended along the eastern boundary from the south-eastern corner of the smithy building. It is likely that the wall ran right around the property; its presence along the south edge, set back from the ditch revetment by about 0.60m, is suggested by a shallow gully, perhaps representing a robber trench for a wall. The south part of the tenement seems to have been used in this period for the dumping of rubbish, and thick layers of material built up, presumably representing the remains of muck heaps. Some time after the construction of the smithy building, the northern part of its west wall had a stone foundation of medium ironstone rubble, c0.35m wide, built against it. The foundation did not run along the whole wall but was between 2.6m and 3.2m long. It is not clear whether it was a structural

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.33 Area J, smithy in the early to late 14th century (Phase J3)

89

90

feature to strengthen the end wall or a bench or other internal feature. It may be noted, however, that the width of the robber trench of the east wall suggests that this may also have been widened in a similar fashion. At the east end of the building, the earliest surviving foor was of 14th-century date. This was a patch of olive-grey clay loam and charcoal, found only in a strip where it had been covered by a later partition. A later clay foor covered the whole of the east end. In the western part of the building, a subrectangular stone hearth was set into the foor. This measured 1.00m by 0.74m and was constructed of fat ironstone rubble scorched red and shattered by heat. It was similar to the hearths found in other houses and was presumably domestic. The hearth itself may have been an earlier feature, but it certainly remained in use to the end of the building’s life. 2.7 THE LATER 14TH CENTURY: BEGINNINGS OF DECLINE

The expansion of Dassett Southend is likely to have been halted by the general change of economic circumstances and fall in population which affected the whole country after the early/mid-14th century. It may well be that the Black Death of 1348–50 drastically reduced the number of people in Southend as it did everywhere else, but it does not seem to have caused instant and widespread abandonment of properties. Henceforward signs of decline, however, begin to be evident (Figure 2.34). These are not universal and from the mid-14th century the trajectories of the different excavated tenements started to diverge, presumably refecting the fortunes of their inhabitants. Some went into defnite decline and one was abandoned, others continued as before, while a few continued to expand their buildings. But even for the last group the economic climate had changed and it is noticeable that after this time, with the possible exception of those on Area F (F4 and F6), there were no completely new houses built. Subsequent improvements to the houses were in the form of extensions, or partial rebuilding only. Area K (K4) The fnal period of occupation of Tenement K extended through the later 14th century and just into the 15th century. Most of the fnds from the phase were 14th century, with only a very few contexts containing 15th-century material. In this phase (K4) the house underwent an extensive rearrangement and the granary was refoored; most of the outside surfaces around the house and the use of a midden/muck heap to the south-west of the house also belonged to this period (Figure 2.35). The rearrangement of the house saw the creation of a small lobby (1.7m by 2.7m) opening into the hall, just inside the north door (Figure 2.36). This was done

BU RTON DA S SET T SOU T H E N D

by the insertion of an L-shaped timber partition on narrow stone footings against a shortened length of what had been the eastern cross-wall. This had the effect of reducing the length of the hall to 6.4m, blocking the former through passage and making the south door open not into the hall, but into an enlarged eastern room, now 7.1m long. There was probably a door opening into the hall, 1.0m wide, between the south end of the newly inserted partition and the remains of a section of masonry along the inner edge of the main south wall. This is interpreted above as either a bench or (more likely) the remnant of a wall from the earliest phase of the building, some of which still seems to have been standing. The purpose of the re-arrangement seems to have been to convert the east end into a byre/stable and provide separate entrances for humans and animals. The reduction in domestic accommodation that this involved seems to suggest a drop in status for the house and its occupants and may be interpreted as evidence of decline. The use of the east end as animal housing may also refect the shifting balance in farming from arable to pasture. Running north–south across the eastern room, about 2m from the end wall, was a vertically sided stone-lined slot or gully, 0.25–0.45m wide, set in a wide shallow hollow (Figure 2.37). The sides and base of the slot were lined with medium–large, smooth-faced ironstone blocks. The slot had no outfall and should perhaps be seen as a manure gully, rather than as a drain (cf. Chapelot and Fossier 1985, fg. 72). The gully appeared to have been deliberately inflled to form a rubble surface. The fll included two fragments of worked stone: a drilled block and a small fragment from the side of a stone trough, the latter perhaps part of the furniture of the byre/ stable. The re-arrangement also involved the raising of the foor level at the west end of the eastern room, with an apparent sub-rectangular platform, 2.6m by 2.0m, laid over deposits of rubble and clay loam. To the west of the manure gully the east–west wall which had divided this end of the building during the previous phase seems to have been shortened to make an alcove or stall in the northern part of the room. A sub-rectangular rubble-flled hollow, 0.50m by 0.50m, in the north-eastern corner of the room may have been the base of some further structure. Following the re-arrangement of the building, some of the earlier foors seem to have remained in use in the main central hall, although it is possible that later ones had been worn away. Centrally placed towards the western end of the hall was a substantial hearth, made of fat medium– large burnt red ironstone blocks and incorporating a quern fragment, and measuring 1.1m by 1.1m. A deposit of ash, presumably associated with the hearth, lay to the east. About 3m to the east was another late hearth, the base of which was formed by a spread of small–medium rubble burnt red, c0.60m by 0.80m, and surrounded by another spread

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.34 Phase plan, c1390

91

92

BU RTON DA S SET T SOU T H E N D

Figure 2.35 Area K, mid-14th to early 15th century, Phase K4 (compare with Figure 2.23); the eastern end of the house has been reconfgured with the addition of a drainage gully, perhaps to provide accommodation for livestock

of burnt clay loam and ash. At some stage late in the life of the house the north door was blocked by the insertion of a narrow wall (c0.25m wide) of large ironstone blocks. This may even have been done when the house was abandoned and could suggest that there was some time between the abandonment and the demolition. The partition wall separating the room (?parlour) in the western part of the house may well have been inserted in the previous (K3) phase. Limitations of time meant that foor layers in this room were recorded but not investigated further. A patch of burnt, small–medium ironstone rubble, 0.40m by 0.30m, which probably represented the remains of a hearth, set into layers of clay loam, was overlain in turn by small–medium rubble which appeared to have been metalling rather than demolition rubble. A sub-rectangular area of slightly larger, more densely packed rubble and brown clay loam against the west wall, measuring c1.8m by 0.8m, might have been the remains of a foundation for some fxed furniture or internal structure. The eastern end of the house fronted the street with the north door opening directly onto it. To the east of the house a gateway 4.3m wide will have provided the main access from the street to the rear of the property. To the west, the area to the north of the house formed part of

the tenement and was divided from the street by a boundary wall of ironstone rubble 0.70m wide running NNE–SSW. At its south end there was another narrower gateway (1.5m wide) opening into the street. The street surfaces to the east and north-east of the house were composed of well laid, small, hard-packed rubble. In one place on the east side of the main gateway the surface had been patched with lumps of slag. On the east side of the gateway, up against the boundary wall was a stone foundation measuring 2.8m by 1.3m (Figure 2.38). This consisted of two rectangular blocks of rubble fanking a central alcove. It does not seem too fanciful to suggest that the structure may have been a kennel for a watchdog at one side of the entrance. The foundation seemed oversized just for a kennel and might also have supported a mounting block or loading platform. Running parallel to the east wall of the house, about 2m out from its south end, was a line of medium rubble 1.60m long by 0.40m wide. This may have been the footing for the wall of a timber lean-to shed, 2.3m by 1.6m, against the corner of the house. This was perhaps another privy. Like those on D23 and I3 it would have contained a barrel latrine as there was no sign of a cess-pit below.

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.36 Area K, house, from the west (Phases K3–K4, early 14th to late 14th/early 15th century)

93

94

Around the remainder of the house the external surfaces consisted of rubble spreads of varying size and density, evidently relaid or repaired periodically. Around the south doorway there was a slight mound (c2.5m by 1.2m) of denser, well packed small rubble, forming a ramped surface up to the door from the south-east. To the west and south-west of the house the darkness of the soil marked the position of the midden/muck heap associated with the Area K house. The full extent of this lay outside the trench but it covered an area at least 11m by 10m. The area was only partially investigated, but an east–west trench, c1.2m wide, was excavated across it down to natural substrate, revealing a series of hollows, possible post-holes and gullies beneath the surviving base of the muck heap/midden, up to 0.27m thick. This contained only moderate amounts of pottery (largely 14th century) and animal bone together with a few scraps of metalwork, which may suggest that it was composed largely of organic material. South of building K, the fnal foor in the (?former) granary, surviving as a number of patches, apparently belonged to this phase. Outside, to the north, south and east were rubble surfaces which, although only surviving in patches, had presumably once formed a continuous expanse. The rubble surfaces were not removed so it impossible to say what features lay beneath them. In most cases the rubble appeared to overlie natural substrate, so it is unlikely that any earlier occupation here was very extensive. To the south of the granary the boundary between Tenements H and K was much less well defned. The cart shed outbuilding in the south part of Area H which appeared to be open sided to the west (see below) may have belonged to Area K. The apparent encroachment might have been the result of a joint building operation between neighbours, swapping a small area of land for assistance with building costs. Within Area K the rubble yard surfaces continued southwards along the presumed eastern edge of the property, interrupted only by a large circular feature (unexcavated) which appeared similar to a post-medieval pit a few metres away in Area H (Figure 2.39). On the south side of the cleared area, a stone-lined drain c0.1m wide and lined with small–medium ironstone rubble ran north–south for c4.5m, at the south end of which was a layer of small–medium rubble which could have been either a surface or demolition material. The drain is likely to have been within a building to accommodate animals, a byre or stable, which, in the absence of more extensive demolition layers, can probably be said to have been timber-framed, although its extent is unknown. Further to the south, some 41m south of the granary, a rough rubble foundation running east–west was located in trial trenches (Figure 2.40). This may have represented the southern boundary wall of the property.

BU RTON DA S SET T SOU T H E N D

Figure 2.37 Area K, dung gully in the eastern room of the house, from the south (Phase K4, mid-14th to late 14th/early 15th century)

Figure 2.38 Area K, possible dog kennel adjacent to the eastern plot boundary, from the west (Phase K4, mid-14th to late 14th/early 15th century)

T H E A RCH A EOL OGIC A L SEQU E NC E

95

Figure 2.39 South part of Area H, Phases H4 and H5, mid-14th to early 15th century Area H (H4) On Area H the mid- to late 14th century (H4) saw no changes to the H3 house, but a series of two or three outbuildings were constructed south-west of the house (Figures 2.39, 2.41, 2.51). Running along the west boundary of the tenement was a substantial wall, 19m long with rubble footings 0.70m wide (Figure 2.40). Although two outbuildings were built against it, this wall was probably originally built as a boundary wall. A relative absence of demolition rubble around the wall may suggest that the footings supported a cob wall, but it is more likely that the wall was built of stone and had been thoroughly robbed. Up against the north end of the boundary wall were the remains of a slightly trapezoidal, timber-framed outbuilding measuring 6m (N–S)

by 5.5m (E–W) and divided into two equal sections 2.7m wide (Figure 2.25). The central part of its east wall was made up of a ragged line, 3m long, of medium–large ironstone blocks; to the south was a rough post-base, 0.8m by 0.62m, of medium–large rubble which probably supported the south-east corner. Gaps north and south of the row of ironstone blocks may represent doorways. The line of the north wall was marked by the edge of the foor rather than by any structure. That of the south wall was also marked by the edge of the foor, although here there was also another rough rubble post-base 0.75m by 0.55m halfway along. The building was divided by an east–west partition wall 3.35m long by 0.55m wide running halfway across the building from the west. It ended in one large block that might have supported a post. This, together with the post-base halfway along the southern wall, suggests that the

96

BU RTON DA S SET T SOU T H E N D

Figure 2.40 General plan of excavations south of the road, showing maximum extent of buildings as at the start of the 15th century

T H E A RCH A EOL OGIC A L SEQU E NC E

97

Figure 2.41 South part of Area H, from the south (Phases H4–H5); the wall at centre left is part of an outbuilding, possibly a cart-shed accessed from Area K to the west ridge of the roof of this building was supported on vertical posts, rather than crucks or trusses. Both parts of the building were foored with a densely packed spread of rubble. The division of the building into two suggests that it was not a barn, and the size of the divisions would be suitable for housing animals, despite the absence of an internal drain. Like the D23 stable/byre (see above) this building could have contained two separate stables or, possibly more likely, given that the partition only went part way across, a single space with two stalls. The area immediately south of the house between the eastern tenement boundary ditch and the outbuilding formed an open yard. Running along the south side of the house a spread of small– medium rubble formed a path/surface 1.5–2.0m wide, which merged into another similar surface which ran southwards forming a path 2m wide along the east side of the outbuilding. Further south again, the layers became much darker suggesting the presence of a midden or muck heap: they contained very little pottery and animal bone suggesting that the heap here was largely of organic material, perhaps from the stable/byre. On the east side of the yard area there was a fat-topped, rectangular, stone structure, 1.8m by 1.0m by 0.20m high; its function remains uncertain, but it may simply have been a stand, perhaps for a trough.

In the south part of the area was the remnant of another outbuilding up against the south end of the western tenement boundary wall (Figure 2.41). The south wall of this building was marked by a short section of large rubble foundation and a length of robber trench, 0.65m wide, curving westwards from the boundary wall. The rounded south-west corner can perhaps be taken to suggest that the south wall was of stone to the eaves, although some kind of timber frame involving an end-fork is also possible. The east wall was marked by a robber trench, 0.60m wide, which extended beyond the edge of the excavation some 5m to the north. There was no trace of a north wall and the length of the building is therefore uncertain (but over 6m). It could have extended as far as the northern outbuilding, a length of c12m, but is more likely to have been shorter. Within the building at the south end there were two possible foor layers of clay loam. This building may have been a barn or store. The line of the east wall of the southern outbuilding was continued southwards for 6.3m by another wall, c0.70m wide and constructed of medium–large ironstone rubble. Its function is uncertain. It may well have formed part of another outbuilding to the south (c7m by 5m), but there was no trace of either a west wall or a south wall although

98

BU RTON DA S SET T SOU T H E N D

Figure 2.42 Area E, Phases E4 and E5, mid-14th century to late 15th century (compare with Figure 2.28); the structure is essentially that of the previous phase, but the brewhouse has been re-organised

a later pit had caused much disturbance. The absence of a west wall is more diffcult to explain, especially as it ought to have formed the tenement boundary. It is possible that the building actually belonged to the Area K tenement, forming a structure open-sided to the west, such as a cart shed. Post-bases along its west side could easily have been missed, given the relatively low intensity of excavation here. To the east of the southern outbuilding on Area H was another yard area. This was overlain by the probable remains of an extensive muck heap or midden, measuring c9m by c6m, which contained large quantities of pottery (447 sherds) and animal

bone (56 fragments), suggesting the disposal of much domestic waste. Further south were other rubble areas; time did not permit examination of these, but it was noted that these ended abruptly on an east–west line, south of which the deposits were relatively rubble-free, possibly indicating the presence of a cultivated garden plot (Figure 2.39). Area E (E4) In this period the west end of the house was refoored with another surface of small rubble and clay loam

T H E A RCH A EOL OGIC A L SEQU E NC E

(Figure 2.42). The presence of a separate foor here may mean that the west end was now partitioned off from the hall, although no trace of a partition survived. The north room was also refoored with a similar rubble surface. The main changes in Tenement E in this phase were to the brewhouse, with the north end rebuilt and the south end re-arranged, although each end continued to be used for the same purpose as before, at least for a time. The new north end had wider walls which suggests that they were now built in stone up to the eaves. The new north wall was 1.0m wide and constructed of medium–large ironstone rubble facing a small–medium rubble core bonded with dark greyish-brown clay loam. The northern 2.4m of the west wall was replaced with a wider wall of similar construction, but which curved on the inside, increasing in width from 0.85 to 1.20m. The northern 2.6m of the eastern wall was replaced with a wall 0.8m wide but retaining a right-angled corner. In the rebuilt north-eastern corner of the brewhouse a rectangular malting kiln was constructed, measuring 2.0m long by c1.2m wide and aligned north–south (Figures 2.43 and 2.44). The north wall had a slight batter while the west wall, 0.40m wide, had an offset footing 0.20m wide to the east. The stokehole at the south end of the kiln narrowed to 0.70m wide, and was paved with two large ironstone blocks, around which deposits of very dark grey clay loam and ash had accumulated from use of the kiln. The south end of the building was now divided off by a spur wall at right-angles to the main west wall, fush with the south side of the doorway. This spur wall, presumably of timber, was set in a shallow beam slot, 1.28m long by 0.28m wide by 0.01m deep. The c3m long area divided off was apparently used to house animals, as a new, curving stone-lined drain inserted across its south end indicates. The actual drain was c0.18m wide by 0.12m deep, widening out into a sump 0.45m wide at the east end, and lined with well faced medium ironstone blocks, including a re-used jamb fragment. The drain was set into a steep-sided gully 0.6m wide, widening out to 0.8m wide at the east end (Figure 2.45). The area to the north of the house and east of the brewhouse would have been a yard. The only features here were scattered shallow hollows and a couple of possible post-holes. None of these features was very substantial and it is possible that this area north of the house was occupied by a muck heap, with the hollows being produced by periodic removal of manure. The hollows produced virtually no pottery or animal bone suggesting that any muck heap was composed very largely of organic material. Area D1/D2 The small house and the barn on Area D1 seem to have been demolished in the mid-14th century and

99

Figure 2.43 Area E, brewhouse and malting kiln, from the south (Phases E3–E4; 14th century)

Figure 2.44 Area E, malting kiln, from the south (Phase E4; 14th century)

100

BU RTON DA S SET T SOU T H E N D

dumping and removal of rubbish. Again, these contexts contained little pottery and animal bone (a total of 58 sherds and 38 fragments) suggesting that little domestic waste was deposited onto them. Areas A, I and L in the later 14th century

Figure 2.45 Area E, drain in brewhouse, from the east (Phase E4; 14th century)

the tenement then abandoned. This was the frst property to meet this fate and it may directly refect the aftermath of the Black Death. It is possible that the small size of the existing accommodation was a disincentive to replacement tenants and meant that this tenement was abandoned. By the early to mid-15th century the eastern boundary gully had flled and does not seem to have been replaced, whereas that on the west side separating Tenement D1 from A was renewed (Figure 2.46). This suggests that the tenement had been acquired by the tenants of the D2 tenement and the two properties amalgamated. There was, however, little evidence for major change in the later part of the 14th century to the D2 buildings, suggesting that any enlargement of the landholding was not accompanied by any marked change in the circumstances of the tenants. Pottery cross-joins between D1 and A raise the possibility that D1 may have been managed, at least initially, by the tenants of A. The possible midden/muck heap on D1 remained in use in this period: another group of large irregular pit/hollows was dug in the area and patchy layers deposited over it, possibly again deriving from the

Elsewhere, occupation continued throughout the 14th century, although in some cases it proved diffcult to date building/rebuilding events with precision. Thus, it is quite possible that the kitchen extension on the north-eastern corner of building A3, described above, may date to this phase. On Area I occupation also continued through the 14th century (I3). No structural changes to the house or internal re-organisations were recorded, although the relevant levels were not fully investigated. Outside the house there were rubble surfaces to the east and part of a possible rubble raft probably indicated the presence of a small timber outbuilding to the south-east, c3m wide, set against the eastern tenement boundary, and extending beyond the limit of the excavation. To the south-west and revealed in a sample trench (Figure 2.8), a spread of dark grey clay loam extending beyond the excavated area may have marked the location of another muck heap, possibly set in a ditched enclosure against the western tenement boundary. This contained only a moderate amount of pottery (47 sherds) and little animal bone (8 fragments), and may again have consisted largely of organic material. The very limited amount of excavation on Area L makes it diffcult to say when the buildings there were constructed or frst occupied. The features recorded are likely to relate to the later phases of occupation and are described in more detail in the subsequent section. However, whilst there was no direct dating evidence from occupation deposits, 14th-century material from the tenement is likely to refect occupation in this period. 2.8 THE EARLY 15TH CENTURY

At the beginning of the 15th century, with the exception of Area D1, all the tenements that had been occupied at the start of the previous century were still occupied (Figure 2.47). Over the next half century, however, all but one of the properties on the south side of the road would be abandoned, either because the inhabitants were under pressure or, conversely, because they were leaving for larger or better holdings elsewhere. The smithy tenement, the one defnitely supported to some degree by nonagricultural activity, was among those abandoned. On the north side of the road the picture was somewhat different. Here the occupied tenements would all continue in occupation over this period and the abandoned Tenement D1 would be taken over by the adjacent D2.

T H E A RCH A EOL OGIC A L SEQU E NC E

101

Figure 2.46 Area D1, Phase D14, early to mid-15th century. By this stage the plot had been abandoned and the boundary with D2 to the east removed, suggesting possible acquisition by the D2 tenants. The irregular pits and hollows may derive from dumping and removal of rubbish

Area A (A4) The early 15th century saw much recutting of the western boundary ditch (Phase A4). The frst recut was the last on the original alignment, although at the south end the ditch now ran eastwards right up to the south-western corner of the house, blocking off the former access to the rear of the tenement (Figure 2.32). The new ditch was c1.1m wide by c0.6m deep and in places it too also had been recut, with three cuts being visible in profle although their flls were indistinguishable. After this ditch fell out of use it was replaced by another which ran north-westwards from the north-west corner of the house and then turned to join the original line. Its profle showed that it was also much recut, although it again proved diffcult to separate the various flls along the whole length.

The way the successive ditches moved eastwards may perhaps suggest that they were accompanied by a hedge or fence to the west. The apparent increase in the frequency with which the western boundary ditch was recut in this period may have been simply an accident of survival — earlier recuts having been obliterated by later. However, it may alternatively be evidence of an increasing problem with drainage in this part of the site. A scatter of shallow hollows north of the house also belong to this period. Area J (J4) South of the road the smithy remained in occupation into the early 15th century ( J4) (Figure 2.48). The smithy building was always relatively small, with a foor area of 37.5m 2 (404 sq. ft), and its footprint

102

Figure 2.47 Phase plan, c1425

BU RTON DA S SET T SOU T H E N D

T H E A RCH A EOL OGIC A L SEQU E NC E

103

Figure 2.48 Area J, smithy, Phase J4, early 15th century (compare with Figure 2.33); a possible small outbuilding has been built against the south-eastern corner of the smithy and the forge hearth rebuilt, with two short walls against the main north wall indicating some form of screening and/or a superstructure over the hearth

104

BU RTON DA S SET T SOU T H E N D

Figure 2.49 Area J, smithy building, from the west (Phases J2–J4, mid-/late 13th century to early 15th century)

remained unaltered throughout the occupation of the property, which suggests that its inhabitants, despite their relatively sophisticated craft skills, were never particularly prosperous. During this period there was a fnal internal re-arrangement of the smithy building (Figure 2.49), an outbuilding was constructed against the south-eastern corner of the smithy, the western boundary ditch was again relined and a new ditch was dug to the north of the tenement. The fnal re-arrangement of the smithy involved the construction of two partition walls extending from the north wall to enclose a space 1.3–1.5m wide in the central part of the building. These walls, which may have had a timber superstructure, presumably provided support and formed a screen for a rebuilt forge hearth structure. In the south of the area between the partition walls there were two sub-oval hollows cutting the foor which may have held supports for an anvil (see Figure 3.25, and online archive, Section 8.13). Up against the west side of the western partition wall there was a spread of burnt rubble and ash which probably represented the remains of a hearth belonging to the end of the building’s use. In the eastern room the fnal foor, a patchy clay loam layer, abutted the partition wall next to the forge hearth. A rectangular rubble-flled feature set against the north wall of the building, 0.70m by 0.60m, was probably a foundation for some internal

structure. A pair of large ironstone blocks set into the foor about 1.0m apart between this foundation and the forge hearth partition wall could also have supported some internal ftting. In the previous phase ( J3) a narrow stone foundation had been built against the inner side of the west wall of the smithy, as noted above. Another layer of rubble, 0.5m wide, was now set against the central part of the earlier foundation, widening the wall still further; it was much more ragged than the earlier foundation, but may have fulflled a similar function as a bench or raised work surface. A socketed stone block and two trough fragments from the building may have come from smithy equipment. The contexts sampled within the smithy for hammerscale largely belong to this period; the results of the analysis by Gerry McDonnell (online archive, Section 8.13) are discussed in Chapter 3. A vertically sided pit/post-hole, 0.92m by 0.68m by 0.32m deep, and an apparently worn rubble surface 2.3m wide, in the corner between the rear of the smithy and the eastern tenement boundary wall may represent a lean-to shed or outbuilding There were alterations to the tenement boundary in this phase. The road surface to the west of the smithy was cut by a short length of gully running northwards from the end of the western tenement boundary ditch, which was itself relined with additional stone revetment material in two stages

T H E A RCH A EOL OGIC A L SEQU E NC E

(Figure 2.50). As a result of these alterations the eastern edge of the boundary ditch shifted some 1.2m to the west, so the aim may have been as much to gain ground by encroachment as to strengthen the revetment of the ditch. Running along the front of Area J there was a large deep ditch, over 2m wide by 0.70m deep with sloping sides. In places the profle of the bottom suggested that the ditch had been recut but the flls were indistinguishable. The ditch appears to have been a late feature. It cut across the surface of the road leading south and therefore probably belongs to a period when the buildings on Areas H, I and K were going or had gone out of use. (It is possible that it ended immediately outside the trench, and therefore did not block off the road, but this seems unlikely.) Its greater depth also suggests that it was a late feature. It was certainly open when the smithy building was demolished and could have been so when the building was in use, with the front door being reached by a plank bridge. Tenement J seems to have been abandoned towards the end of the early 15th century. The demolition of the building saw fairly extensive stone robbing from the north, south and east walls. Overlying the last of the spreads of demolition material was an extensive layer of slag which spread from the southern side of the north wall of the smithy, over the remains of the wall and into the ditch in front of the building. It was almost as though this represented a dump of slag in the smithy against the north wall which had collapsed forward when the wall was demolished. Its extent demonstrates that the ditch along the front of the tenement was still open at this time. This ditch was also eventually flled in, although it seems likely to have remained in use for longer than the ditches around the southern, western and eastern tenement boundaries. Area H (H5) Occupation of Area H also continued into the early 15th century (H5) and again no new buildings were constructed (Figure 2.51). Within the house the fnal foor layers were of clay loam, although an area of rubble in the north-east corner may represent metalling around a north doorway (opposite the one in the south wall). The foor was cut by two post-holes, one 0.45m in diameter by 0.04m deep, the other 0.16m by 0.08m by 0.03m deep, which would have supported internal fxtures. To the west of the house the earlier lean-to shed/ covered area remained in use and a layer containing early 15th-century material accumulated over its foor. To the north of the house material of similar date came from the fnal rubble street surfaces. There was also early 15th-century material from a rubble surface within the southernmost possible cart shed outbuilding (which, as suggested above, may have belonged to Area K).

105

Figure 2.50 Area J, ditch revetment along the southern plot boundary, from the south (Phases J3–J4, 14th to early 15th century)

The main demolition of the house (Phase H6), which seems to have taken place in the early 15th century, scattered the usual layers of small–medium rubble and clay loam over the building and its immediate vicinity. The wall foundations were not removed, although robbing of the southern outbuilding took place. The eastern boundary of Tenement H appears to have been flled in at around this period. Area K (K5) The demolition of the Area K buildings seems to have taken place in the early 15th century (K5). As noted above (K4), the north door was blocked by a narrow wall late in the building’s sequence, perhaps indicating that the building had stood vacant for a while prior to demolition. The demolition of the house was less thorough than that of most of the others and, although its walls were razed to the ground, no effort was made to remove the foundation courses. Most of the granary’s foundations were also left, except at the west end. The demolitions scattered rubble over and around both buildings.

106

BU RTON DA S SET T SOU T H E N D

Figure 2.51 Area H, house and outbuildings to the rear, Phases H4 and H5, mid-14th to early 15th century

T H E A RCH A EOL OGIC A L SEQU E NC E

Area I (I4) The occupation of Area I seems to have lasted a little longer than those of its neighbours, into the mid-15th century (Figure 2.52). Early in the century the house was re-arranged internally (I4). The latest arrangement within the house was clearer than the earlier ones. The building was divided into three unequal parts, with a short central hall, and the east end was sub-divided into two small rooms. A north–south partition wall, 0.4m wide, ran across the house just to the east of the doorways, with a gap for a doorway 0.9m wide at the south end. The eastern part of the building was further divided by a rough north–south wall footing, from which ran an east–west wall which divided the east end into two. Both partitions were 0.38m wide and were much more ragged than the one by the external doors and may represent a later sub-division of the eastern part. All these footings would presumably have supported timber-framed partitions (Figure 2.53). The central room measured 3.65m by 4.4m; it must have been the hall as it contained what appeared to be the main hearth, a large irregular arrangement of fat medium–large ironstone blocks stained red by heat (Figure 2.54). It was 1.65m by 1.60m in total but looked as though it may have been constructed in stages. The north-eastern room measured 2.2m by 1.6m whilst the south-eastern room was 2.8m by 2.6m. The similarity between the clay loam foor layers in both eastern rooms and the central room suggests that the partition walls separating them are likely to have been late insertions. The western room measured 5.75m by 4.25m. Against the south wall there was a sub-rectangular spread of rubble and grey clay loam flling a hollow 1.65m long by 1.0m wide. This was presumably a foundation for an internal structure of some kind, perhaps a base for a fxed piece of furniture. In outline the plan of this house was the classic hall-house, with a central hall, fanked by one room at one end and two at the other, like the K3 house. It did not conform entirely to the type, however. It did not have opposing doorways in the hall. The northern doorway to the street did open into the hall, but the southern doorway opened into the western room which was larger than the hall. Running east–west along the centre of the western room an apparent rubble-flled feature 4.4m long by c0.75m wide (unexcavated) may represent the remains of a drain. This might suggest that this end of the building housed animals and the house was therefore a ‘longhouse’. However, this is only a possibility, and if true, would probably represent a late re-arrangement, like that in the Area K4 house. It is more likely that the drain was associated with the ‘processing’ activity represented by the stonelined trough noted in a previous phase (I2, Figure 2.9). Up against the north wall there was an irregular hearth of large fat burnt ironstone blocks. Hearths

107

are often associated with ‘processing’ troughs and its presence would argue against the room being used to house animals. It was not clear, however, whether or not the hearth belonged to the building, as part of it appeared to post-date robbing from the building’s demolition and another hearth to the east was clearly a demolition feature. To the south of the house an extensive rubble surface was laid over the northern part of the rear yard, next to the house. The western part of this appeared to have formed the foor of a new timber outbuilding set against the south-west part of the house. It was supported on narrow rubble footings, of which only remnants had survived; a post-hole at the west end of the southernmost footing may also have been associated. There was no evidence for the western side of this structure, nor any indication as to its function. The building would have been 6.7m long and 4–5m wide, possibly L-shaped and possibly with a porch. Area W (W2) The water hole (W2) north of Area I was also still in use at this time (Figure 2.52). The fnal rubble surfaces over both the water hole and the adjacent street contained early to mid-15th-century material from the fnal period of use, although the surfaces had probably been laid much earlier. Areas L, M and N Investigations in this area were severely constrained by the time available. Most of the contexts recorded were on Tenement L; while no dating material was recovered from occupation contexts the building remnants recorded represent the fnal phase of activity on these tenements and on the basis of the other tenements south of the road are likely to belong to the frst half of the 15th century (Figure 2.55). In addition, a rubble spread to the south of Tenement L, which may have been the foundation for an outbuilding, was of 15th-century date. Areas L, M and N are therefore described at this point in the structural narrative, with the caveat that the dating is less securely based than most of the other elements of this phase; indeed, the material from the topsoil over L was more akin to that from the later phases north of the road, suggesting that this tenement may have remained in occupation longer than the others to the south, and only been demolished in the later 15th century. The earliest wall in the north part of the area formed the boundary between Areas L and M. It belonged to Tenement M and was of one build with the northern boundary wall of Area M. This wall also ran through the south part of Area L and could be traced in the surface rubble distribution, continuing to the south end of the feld, as the boundary between Tenements M and K.

108

BU RTON DA S SET T SOU T H E N D

Figure 2.52 Areas I and W, Phases I4 and W2, early to mid-15th century (compare with Figure 2.8); the house interior has been reconfgured with the insertion of partitions, and an outbuilding constructed against the rear of the house

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.53 Area I, house, yard and outbuilding, from the east (Phases I3–I4, 14th to early 15th century)

Figure 2.54 Area I, hearth in central room of the house, from the west (Phase I4, early 15th century)

109

110

BU RTON DA S SET T SOU T H E N D

Figure 2.55 Areas L, M and N, Phases L1, M1 and N1. Only limited work was undertaken in these areas; most of the fnds were from demolition layers and indicated occupation from at least the 14th century and abandonment and demolition in the 15th century

T H E A RCH A EOL OGIC A L SEQU E NC E

Area L: house and ancillary building The buildings at the north end formed an L-shape, with the house along the street frontage and a barn/ store set at right-angles fronting a yard behind (Figure 2.56). The earliest surviving building on Area L was built against the west boundary wall. It consisted of a north–south range, running back from the frontage. It was 4.4m wide and over 10m long, with an internal partition dividing off a room 4.2m long by 3.8m wide at the north end. Set against this along the frontage to the east was an east–west range, 10m long by 5.2m wide. The sequence of these two buildings seemed fairly clear in plan, although their north walls, albeit much disturbed, seemed to be of one build and different in character to the east wall of the north–south range. The north walls may have been retained from an earlier building parallel to the frontage. (The frontage wall on Area F could also have been retained through different phases.) The north wall was 0.65–0.90m wide. As recorded, it was discontinuous and irregular because of robbing and tree root disturbance. The east wall of the north– south range was built in three sections of different widths. The northern part, 5m long, was 0.60m wide and composed of small–large ironstone rubble; the southern part, of which only 1m lay within the trench, was 0.65m wide, and of medium–large rubble. The 3m length between these was narrower, 0.40m wide, and composed of small–medium rubble; it probably marked a wide doorway opening into the southern part of the building. The narrowness of the northern section of this wall may mean that it supported a timber frame and that the north–south range was of mixed construction, with north and west stone walls originally standing to considerable height, and a timber-framed east wall. The partition wall dividing off the north room was over 0.40m wide, and of medium–large rubble. A small patch of clay loam foor was located at the north end of the north room. The southern room had a foor surface of small to medium rubble and clay loam. The width of the doorway suggests that the southern room may have been a barn/ store building. The east wall of the east–west range was c0.90m wide and made of medium–large ironstone rubble. The south wall was similar in construction, but only 0.60m wide; its west end seemed to have been robbed out. The relative narrowness of the south wall may mean that this range was also of mixed construction, with high stone walls to north and east and timber walls on the south and west. There appeared to be a rubble surface extending over the south wall about halfway along it; this probably marked the position of a doorway, possibly c1.3m wide. The internal dimensions of the range were 8.8m by 3.7m. Its foor was not uncovered; the only internal feature recorded being a burnt patch, 0.60m

111

by 0.40m, against the south wall, and this could have been a demolition feature. The eastern boundary of Area L was marked by another wall which abutted the corner of the east–west range. The boundary wall was of rubble and 0.75–0.80m wide. To the south and west of the buildings a courtyard was surfaced with uneven, but well packed, small–medium rubble, with larger rubble to the south. In the north-east corner of the yard a possible post-base of medium–large rubble, 0.80m by 0.40m, might have supported a lean-to structure against the eastern boundary wall. Overlying the courtyard surface there was a layer of clay loam which might have represented a contemporary accumulation of mud. External features on Tenement L To the north of the Area L buildings there was a surface of small–medium ironstone rubble, which will have been either the street surface or a forecourt to the north of the tenement. The strip excavated across the south part of Area L was really too narrow for much sense to be made of the features uncovered (Figure 2.56). The tenement

Figure 2.56 South part of Area L, from the west (Phase L1, 14th/15th century)

112

boundaries were again marked by walls: to the west the boundary wall between Areas M and L continued through, whilst the eastern boundary wall of Area L continued, here slightly narrower (0.60m) than the section abutting the building. Between these two walls were a number of surfaces and cut features. The fll of a shallow curving east– west gully, 0.52m wide by 0.04m deep, contained 15th-century material; this was cut by a shallow subcircular pit, 1.90m across by 0.06m deep, to the east. Stratigraphically later than these was an extensive spread of rubble, also containing 15th-century material. This contained two elements. A band, about 1m wide and only one course deep, of small–medium rubble faced, like a wall, with medium(–large) rubble, ran across the trench from NE to SW curving slightly westwards. To the west of this and contiguous with it was a sub-rectangular area of small–medium rubble, c3m across. The function of these spreads is uncertain. The one-metre wide rubble band is unlikely to have been a wall foundation given its lack of depth; it is possible that it was a path or section of laid surface, similar to those on the south side of Area D2. The regularity of the rubble to its west suggests that it could have been the rubble foundation for an outbuilding, 4m wide, extending northwards beyond the limit of excavation. Area M The full extent of Area M is uncertain but may have been very large, occupying the full depth of the feld and all the Area Z2 defned in the feldwalking work (see Section 1.12 above). The surface rubble distribution suggested that there was a range of buildings along the north edge of the feld, but when an area here was cleared of topsoil the rubble proved to come only from a substantial boundary wall (the absence of building remains further to the north was confrmed by the salvage recording during motorway construction). It seems that the buildings may actually have been set well back from the frontage to the south and to have corresponded to the group of dense rubble scatters on a line with the Area K, H and I houses. This area lay outside the motorway corridor and was not excavated. The only feature recorded to the west of the boundary with Area L was the northern boundary wall. This was c0.85m wide, with an offset foundation, c0.20m wide, to the south. A length of 44m was uncovered, and it appeared to be built in three sections, on slightly different alignments (from east to west: 22.5m, 9m and over 10m in length). South of the boundary wall the cleared area seemed barren of obvious features and was not further investigated. Area N It seems likely that Area N was originally laid out as a property similar in size to Areas J and L

BU RTON DA S SET T SOU T H E N D

but that it remained unoccupied, as no trace of buildings was found. A patch of small–medium rubble within the east–west trial trench across the southern part of L, M and N may have been the remnant of a surface; it overlay a clay loam layer containing 14th-century material. Area F (F4) On the north side of the road the early/mid-15th century saw the house on the frontage of Tenement F replaced by another on the same alignment (F4). The demolition of the previous house involved the robbing out of the central section of its northern wall, with the resultant robber trench then being flled with rubble and clay (Figure 2.57). The new house was also in the form of a single range parallel to the street, but at 5.2m wide it was slightly narrower than its predecessor. Again, the only surviving structural remains were of the north wall, and the building extended further east beyond the limit of excavation (Figure 2.58). This building was almost certainly divided from the time of its construction, with a separate room at the west end, 2.5m long. This room was bounded by a partition wall with a doorway 0.90m wide at its northern end. The partition wall sat in a shallow construction trench, and consisted of a line of large ironstone blocks, faced only on the west side and packed on the east side with small rubble and clay, which presumably formed the base for a timberframed partition. The western room was foored with clay loam, and the eastern one with clay loam and small rubble. The larger eastern room had a post-hole in the centre of the foor which may have supported another partition, and two hollows containing quantities of ash, although there was no hearth in the area excavated. A single fat-topped stone block set into the foor against the northern wall of the western room may have supported some structure. To the north of the western end of the house there was an irregular gully up to 0.4m wide that ran east–west, parallel with and c1m distant from the house wall; it is likely to have been a drain for water dripping from the eaves of the house. To the east there was another short length of gully, 0.20m wide by 0.05m deep which ran SW–NE. Three large stone-packed post-holes may have been the remnant of a timber outbuilding set against the north-west corner of the house, up against the western boundary of the tenement. Area D (D1/D2 amalgamated) As noted above (Section 2.7), the D1 tenement was the frst within the excavated area to have been abandoned, and by the early to mid-15th century, if not earlier, the boundary between D1 and D2 had been flled so that D1/D2 was by now de facto a single property. The midden/muck heap in the

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.57 Area F, Phases F3 and F4. The 14th century house F3 was slightly wider than its early to mid-15th century replacement F4

Figure 2.58 Area F, house, from the east (Phases F4–F5, early/mid-15th century)

113

114

BU RTON DA S SET T SOU T H E N D

Figure 2.59 Area D2, house, from the east (Phases D23–D24, early 14th to early 15th century) north-west corner of what had been D1 continued in use, with a sequence of patchy layers and shallow pits consistent with dumping and removal of rubbish; the small quantities of pottery and animal bone suggests that little in the way of domestic rubbish was being deposited here (Phase D14). In the south-east part of the former D1 were various somewhat nebulous shallow hollows and an east–west gully c 0.75m wide by 0.08–0.25m deep which was observed for c11.5m before being truncated by later features at its eastern end (Figure 2.46). These features were overlain by spreads of material extending from Area D2, emphasising the disappearance of the boundary and the amalgamation of the two properties. Despite the removal of the D1/D2 boundary, however, the western part of D1 including the midden area may have been sublet to or at least occupied for a period by the tenant of Area A. Such an arrangement would explain why the pattern of pottery crossjoins over Areas A, D1 and D2 is fairly confused, compared with the relatively clear separation between, for example Areas E and F (see Figure 3.3, no. 2).

Alternatively, it is possible that the occupants of both Areas A and D2 had been dumping their rubbish onto D1 while it was abandoned. The muck heap area was also used by the D2 occupants in this period; this is demonstrated by one pottery cross-join link from a foor layer within the D24 house (see Figure 3.4, no. 6). The remaining house on what had been Tenement D2 was now extended westwards (D24) (Figure 2.59), over an area occupied in Phase D23 by an outbuilding. This increased the foor area by 60% to 64.8m 2 (696 sq. ft). At the same time the original part of the house was re-arranged into two rooms and the 7m section of the north wall of the house to the east of the doorway was rebuilt (Figure 2.60). Replacement of north wall and re-arrangement of original D2 house The previous wall may have shown signs of instability, which necessitated its replacement. The other walls seemed unaffected and it is diffcult

T H E A RCH A EOL OGIC A L SEQU E NC E

115

Figure 2.60 Area D2, Phase D24, early to mid-15th century (compare with Figure 2.18); the house has been extended westwards over an earlier outbuilding and the north wall partially rebuilt, with the original house area now divided into two rooms. The outbuilding to the north-west of the house has been rebuilt and the boundary with plot D1 to the west flled in, indicating an amalgamation of the tenements to judge how much of the roof structure would have had to be replaced as well. A trench, 5.8m long by c0.7m wide by 0.25m deep, was dug along the north side of the wall, presumably as part of demolition operations, and then backflled. This was cut by the foundation trench for the new wall; the trench was up to 0.3m wider than the wall to the north and packed with clay loam. The new section of wall was c0.65m wide and made of

roughly coursed small–large rubble. It had extra deep foundations on its northern side and contained re-used material including a broken mortar. The north doorway to the house (0.9m wide) was in the same position as before. It was 0.6m to the west of the position of the south doorway which was probably now located at a point 7.5m from the east end, the latter’s position being marked by a patch of internal threshold metalling.

116

BU RTON DA S SET T SOU T H E N D

Internally this part of the house was now rearranged into two unequal parts: a hall 7m long with a room 2.5m long opening off it at the east end. The division was effected by a rubble footing, 0.25–0.35m wide, which would presumably have supported a timber partition, with a gap for a doorway c0.8m wide at its north end. Parts of two successive clay loam foor layers survived in the east room. In the north-east corner there was a burnt patch, 1.1m by 0.45m, of reddish-yellow and black ash, up against the inner face of the main east wall. Finds from the eastern end of the house included some items indicating, perhaps, at least middling status, notably three beads from a rosary, two of jet and one of bone. More beads occurred in later contexts (see below). There was also a copper-alloy pin, a bone pin, a pewter spoon fragment, an iron knife and a padlock. In the main room (hall) at the beginning of this phase a foor layer of clay loam and charcoal covered most of the central part of the building. Set into this was a central hearth, 1.12m by 0.80m by 0.12m deep, lined with yellowish-brown sandy clay loam and flled with dark reddish-brown/very dark red ashy clay loam, with another smaller patch of dark reddish-brown ash to its north-west. The hall was then foored with clay loam, which survived over its south side. Overlying this, against

the south wall, there was a spread of small–large rubble and tile, 1.3m by 1.0m by c0.08m deep, which was probably metalling around the threshold of the south door. The new foor was accompanied by two new hearths. One, against the centre of the south wall, consisted of a spread of dark reddish-brown and black ash which was bounded to the west by a kerb of medium rubble which may have supported a screen separating the hearth from the doorway. The other hearth, in the centre of the foor, was made of heat-stained, fat, medium rubble blocks, 0.60m by 0.47m, and surrounded to the south by a spread of ash. The central hearth was eventually replaced by another, slightly to the south-east (Figure 2.61). This was sub-rectangular, 1.2m by 0.9m, and made of several ironstone slabs, shattered and reddened by heat. Around the hearth there was a layer of black and red clay loam and ash, presumably deriving from its use. In the north-west corner of the hall, against the north wall, there was another, pear-shaped, hearth, 1.0m by 0.85m, with small–medium rubble lining a shallow hollow and overlaid by ashy silt loam. A small patch of rubble at its south end was probably associated with this hearth. In the south-west corner, against the south wall there was a stone-lined slot or trough (Figure 2.62), 2.1m long by 0.70m wide, running north–south,

Figure 2.61 Area D2, central hearth in main room of the house, from the west (Phase D24, early/mid-15th century)

Figure 2.62 Area D2, stone-lined slot in the main room of the house, from the north (Phase D24, early/mid-15th century)

T H E A RCH A EOL OGIC A L SEQU E NC E

possibly incomplete. As with a similar feature in the I2 house (see below), its function was uncertain, although probably it was some form of ‘processing’ of an animal or agricultural product. Similar features at West Cotton, Northamptonshire, have been linked with fulling or bleaching of linen cloth (Chapman 2010, 237–239; see Section 3.3 below). Finds from the trough included a jug sherd of Midlands Purple, a stone spindle-whorl (no. 92); a copper-alloy annular brooch (no. 1), and a perforated plate (no. 138) which might have been a rose or flter from a watering can, although whether this fnd related to the function of the trough is impossible to tell. Exceptionally it is possible to put a name to the tenant or occupant of the tenement in this period, as it is likely that the door jamb inscribed with the name ‘Gormand’ (see Figure 3.17), found in a layer of rubble of the next phase, was set in the new southern doorway to the street. A family called Gormand, spelled in various ways, often as Goremund, is recorded in Burton Dassett from the 13th to the 15th century, holding land in Northend and Knightcote (Palmer and Dyer 1988, 218). They emerge as prosperous peasants and some of their number were involved in substantial property transactions. A Thomas Gormand had a joint tenancy of 1½ yardlands in 1280 and in 1311 a William Gormand is recorded as a witness to a deed of the lady of the manor of Knightcote. It is possible that a younger son or other relative migrated from Knightcote to Southend to try his hand as a trader in the market or to take up an abandoned agricultural holding, or possibly even both. He certainly seems to have had the fnancial resources both to remodel and extend the D2 house and take on the adjoining plot. Clearly the carving of the name stone was an assertion of status by the Gormands, proclaiming their literacy. It is likely that by this period traders and better-off peasants who might be expected to carry out administrative roles such as churchwarden, or to handle documents such as leases or accounts, would need to have some familiarity with the written word (Palmer and Dyer 1988, 218). And even those among their neighbours who could not have read a deed or letter would certainly have known what the inscription meant. Extension to west end of house The extension added a room, 6.25m long by 5.25m wide, onto the west end to make the house 17.25m long in total (Figure 2.60). A couple of irregular hollows and a shallow pit, stratigraphically pre-dating the extension, seemed to relate to the demolition of the earlier outbuilding and the construction of the extension. On the south-west corner there was a group of four small post-holes, 0.25–0.31m across by 0.11–0.17m deep, that might have supported scaffolding for the building.

117

The north wall of the extension was 0.55–0.60m wide, with a buttress protruding 0.33m beyond the west end of the building. The west wall was 0.55–60m wide with a ragged construction trench on its west side. The south wall was 0.60m wide and set on a footing 0.75m wide; there was a slight indication of another buttress on its west end where the footings bulged outwards. Although the walls of the extension were thus slightly thinner than those of the original house the presence of buttresses does suggest that the walls were also of stone to full height. The new west room measured 5.75m by 4m internally; a foor of clay loam covered the whole room but became very thin in the centre. On the west side, centrally placed, a stone-packed post-hole, 0.33m by 0.25m by 0.12m deep, could have supported a partition or internal fxture. Close to the centre of the south wall, a hearth, 0.60m by 0.50m, consisted of one large fat stone surrounded by small rubble and dark greyish-brown clay loam and charcoal. A cluster of shallow cut features which may have represented the remains of a series of hearths were located in the north-east corner of the room. These were subsequently covered by what appeared to be a replacement foor layer of clay. Rebuilding of outbuilding In the north-western corner of the excavated area, a new and larger outbuilding was built over the previous stable/byre (see Section 2.6 above) with its west wall cutting the flled-in boundary gully (Figure 2.60). Like its predecessor, the new building was constructed of timber supported by earth-fast posts, but although the surviving post-holes were at irregular intervals, two in the centre of the long sides of the structure were paired, suggesting that the building may originally have had two bays of 4.4m. Two centrally placed post-holes in the southern bay could have supported a north–south partition dividing it into two stalls, so perhaps the building retained its function as a stable/byre. There was no drain, however, and other post-holes within the building—one centrally placed in the north bay, one in the south-west corner, and two against the west wall next to a rough oval ashy hearth—hint at other fxtures and activities carried on within what may have been a multi-purpose building. A number of the post-holes had been recut, and it seems clear that their pattern was the result of considerable rebuilding and repair. The location of doors is uncertain, although the density of post-holes along the west wall suggests there was none on that side. North and south of the house Immediately to the north-east of the house an area c7m by c3m contained a very confused series of wide, shallow gully/hollows. They were probably caused

118

BU RTON DA S SET T SOU T H E N D

Figure 2.63 Area E, house, from the east (Phases E4–E5, mid-14th to mid-15th century)

by the passage of feet through water that had drained from the roof of the house. These hollows were succeeded by a more regular drainage gully, c6m long by 1.0m wide, which had the remains of a stone lining along part of its length. Further north and west were various layers and irregular hollows and pits; it is possible that the area had been used, at least in part, for a muck heap, although the generally low levels of domestic rubbish suggest deposition of more organic material, perhaps from the stable/byre adjacent. The only feature attributable to this phase in the forecourt area south of the house was a pit on the south edge of the excavation, 1.68m by 1.06m by 0.18m. Although no defnite evidence of this period survived, it is likely that this area would have been covered by rubble surfaces running from the house down into the street. Area E In the 15th century the house continued in use (Phase E5) with a new series of hearths in the hall

(Figure 2.63). A series of stone-lined gullies and a tank were built in the eastern room. These seem to have been used for some small–scale processing activity. In the brewhouse, the malting kiln also continued in use, and to east and north of the house there were more yard surfaces cut by hollows. By this time the hall area of the building was foored with olive-brown clay. This was different in character to the foor to the west, tending to confrm that, as in the previous phase, the western end of the building was divided off by a partition which left no trace. Just to the south of the north door there was an oval pit, 0.98m by 0.52m by 0.12m deep, across the line of the putative partition. This was flled with brown clay loam and rubble, the rubble perhaps dumped to support the partition across the pit. At the east end of the hall the foor ran through the entrance into the eastern room which might be taken to suggest that there was no door between the two. More hearths were in use in addition to those from the previous phase E4. On the north side was

T H E A RCH A EOL OGIC A L SEQU E NC E

119

a large fat, purple, heat-stained block, fanked by rubble. About 1m to the east was a sub-rectangular patch of fat, red heat-stained, medium rubble, 0.70m by 0.60m. Both these features were overlaid by a spread of dark reddish-brown ash, charcoal and olive clay loam which will have derived from their use. At the east end of the hall, in the angle of the south wall and the spur wall, was a complicated series of hearths cutting the foor. The most substantial of these measured 1.1m by c1.05m and was made of large fat heat-reddened ironstone blocks, bedded on olive yellow sandy clay loam and small rubble, set in a shallow sub-rectangular hollow. The extension of the spur wall to a length of 1.1m was presumably designed to make a proper freback for the adjacent hearth. West of this was a further series of hearths, the earliest of which was associated with a cluster of seven stakeholes and a post-hole. There was nothing to suggest that any of these hearths had any function other than domestic use. Stone-lined tank, gullies and hearths in the east room At some stage in this period, the eastern room was re-arranged with a series of internal stone structures. Up against the north wall there was a rectangular tank, 1.05m by 0.40m by c0.40m deep (Figure 2.64), lined around the sides with squared ironstone blocks. Underneath the tank was a hollow which was probably connected with its construction. Both the western part of the tank lining and an east–west stone-lined gully 4.0m long by 0.36m wide and c0.20m deep running across the centre of the room were packed with small–medium rubble. The gully was set in a shallow hollow c1.2m wide; its bottom was lined with medium–large, fat ironstone blocks, and its sides were lined with medium squared blocks. Running south-westwards from the west end of this was another stone-lined gully which had been more extensively damaged by later features. The two gullies are likely to have functioned as a drain. Up against the south wall of the eastern room there was a rough hearth, 0.83m by 0.48m by 0.09m deep, consisting of greyish-brown clay loam and dark reddish-brown ash, and opposite this, against the north wall there was another, similar hearth, oval, 0.60m by 0.36m by 0.03m deep, of dark reddishbrown ash and charcoal. The gullies and tank were clearly associated, although they did not seem to be physically linked and the bottom of the tank was c0.25m deeper than that of the adjacent gully. Their purpose remains obscure; it seems most likely that some kind of processing of animal or agricultural products was being carried on in the room, especially in view of the presence of the two hearths, and the amount of charcoal in the demolition layers over the tank and gullies. The soil samples from these features gave no helpful indicators. An alternative suggestion,

Figure 2.64 Area E, stone-lined tank at the east end of the house (see Figure 2.42), from the south (Phase E5, early/ mid-15th century)

that the room was used for the accommodation of animals, that the gullies were drains or dung gullies and the tank a watering trough seems unlikely, in the light of the presence of the hearths, the relative depths, the elaborateness of the whole structure and the absence of a separate entrance. The function of these features is discussed further below (Section 3.2). After the tank and gullies fell out of use they were flled in and levelled out. The brewhouse remained in use during this period. The building was refoored periodically with layers of clay, clay loam and rubble. The stokehole of the malting kiln was rebuilt with rather ragged rubble (from which facing stones may have been robbed). In the yard north of the house a rubble path, c1.6m wide, ran from the north porch along the north side and around the east end of the house through the side access to the street. The rubble of the central section of the path on the north side contained large amounts of pottery (521 sherds) and animal bone (175 fragments) suggesting that there had been a muck heap for domestic waste on top of it, perhaps piled up against the house. Another probable path

BU RTON DA S SET T SOU T H E N D

120

composed of less dense rubble ran northwards across the centre of the yard. Cutting into this northern path there was a shallow, ring-shaped hollow, 2.1m in diameter and up to 0.8m wide (Figure 2.65). This may have been produced by the passage of feet along a circular path as a result of the operation of a mill, person or animal powered, and perhaps used for the crushing of malt for brewing. There were various other pits/hollows within the yard area; the quantity of rubble within these features suggests that the rubble surfaces in the yard were originally more extensive. Area I The Area I tenement seems to have remained occupied for slightly longer than the others on the south side of the road and its demolition took place in the mid-15th century (I5). When the house was demolished the entire north wall and the eastern end of the south wall were completely robbed. A small hearth over the former central partition wall, and another which appeared to post-date one of the robber trenches, were presumably related to demolition activity. As usual, layers of small– medium rubble were scattered over and around the house. Relatively large quantities of fnds came from the demolition levels. These included architectural stone fragments including a sill/plinth fragment, a pivot stone, a small stone socket and a possible gutter/downspout fragment with moulded decoration (see online archive, Section 8.10, nos 21, 23, 27 and 45). Area W (water hole) The water hole also seems to have fallen out of use in the mid-15th century (W3), a bit later than the abandonment of the Area J, K and H tenements, but contemporary with that of Area I. Following its disuse, a layer of fairly homogeneous clay loam accumulated over the bottom; this appears to have resulted from silting rather than deliberate inflling. A spread of demolition rubble and dark greyish-brown clay loam extended down from the Area H house and over the silting layer, demonstrating that the water hole was already partly flled when the house on Tenement H was demolished. The water hole itself seems to have survived as a hollow that was eventually flled by further silting. It is possible that rubbish was dumped into the disused water hole, presumably from the adjacent properties. These demolition layers produced 464 pottery sherds and 361 fragments of bone, as well as several metalwork items. Although the demolition layers south of the road produced more metal objects than previous phases, the quantities were far fewer than from equivalent contexts north of the road. This low density of metal fnds from south of the road may refect a poorer material culture/fewer possessions. In addition to the

Figure 2.65 Area E, ring hollow to the north of the house, partexcavated only, from the east (Phase E5, early/mid15th century) fnds from demolition layers there were more from post-medieval and topsoil layers, the large majority of which will have come from the medieval tenements. 2.9 MID- TO LATE 15TH CENTURY

With the exception of Area L, which as noted above appears to have continued in occupation until the mid-15th century, the southern side of the road had been abandoned by this period (Figure 2.66). On the northern side, however, the buildings continued in use, with rebuilding and enlargement of the houses and their ancillary buildings, taking advantage of the ruined buildings to the south as a source of stone (Figure 2.67). Area A The fnal phase of occupation in the mid-/late 15th century saw an extension of the house (A5) with the addition of a trapezoidal room c6.25m long by 5.75m wide onto its west end (Figure 2.68). The room had timber-framed walls set on stone wall footings which were much less substantial than those of the original house. The new room overlaid the western tenement boundary ditch which must have been inflled by this time; the rubble in the top of the ditch and overlying clay loam layers were laid to raise the ground level in advance of construction. A soakaway drain, 0.23m wide by 0.20m deep, was cut through these layers, running westwards from the north-west corner of the original house. The south wall footing of the extension was set in a slight trench; the rubble footing itself was 0.55m wide. The north wall footing was also 0.55m wide

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.66 Phase plan, c1460 (compare with Figure 2.67, which provides a reconstruction of these structures)

121

122

BU RTON DA S SET T SOU T H E N D

Figure 2.67 Dassett Southend c1475 (compare with Figure 2.22). The buildings south of the road had been abandoned and demolished, with the possible exception of those on plot L. The northern area, by contrast, appears to be prosperous, although plot D1 had been abandoned and incorporated into D2. The houses on D2, E and F present a continuous frontage along the road

T H E A RCH A EOL OGIC A L SEQU E NC E

123

Figure 2.68 Area A, Phase A5, mid- to late 15th century (compare with Figure 2.32); the house has been extended to the west and outbuildings constructed to the north

124

but raggedly made of small–medium rubble and did not appear to have been set in a foundation trench. The west wall footing was of a slightly different character, consisting of a line of large rubble blocks, 0.3–0.5m wide. It had been truncated by a later feld drain. The top of the make-up layers formed the foor of the room and there was no evidence of any internal features. The original part of the house remained undivided, its foor still formed by the same clay loam layers. At its east end there were some internal features probably belonging to this period. Centrally placed against the east wall there was a shallow, stone-lined pit, 0.99m by 0.63m. Its lining consisted of six large ironstone blocks, some of which were burnt, as well as smaller stones set in clay. Although the lining contained burnt stones, the absence of ash would tend to suggest that this feature was not a hearth or oven. To the west there was a large oval pit, 2.6m by 1.2m by 0.45m deep, with steep sloping sides and a fattish bottom. The purpose of this pit is uncertain; it was evidently a shortlived feature. A series of patchy layers, protected where they had subsided into the pit, may well have represented the remains of successive foor layers at this end of the building. Forecourt surfaces to south and south-west of house To the south of the house a series of layers were deposited at this time, consisting of a well made surface of small–medium rubble overlying clay loam make-up layers. This served as a forecourt over the whole frontage south of the house. To the west the surface was c2.5m wide, to the east it extended southwards beyond the limit of excavation (over 4.5m), presumably running down into the street. This surface remained in use until the abandonment of the house. Relatively large quantities of pottery (352 sherds) and animal bone (199 fragments) from this surface presumably represented domestic rubbish dumped onto a muck heap outside the house but left behind when the heap was removed to be spread over the felds. Barn This period also saw the construction of a timberframed barn to the north-west of the house (Figure 2.68). The barn was 14.5m long by 4.5m wide and aligned at right-angles to the west tenement boundary. It was probably a four-bay cruck structure: at the west end the bay structure was clear, but to the east some of the post-bases had not survived, and some of the other stone clusters had probably supported the sills of the walls. The western bay was 4m long, the next 3.5m; the other two must also have been c3.5m each. This was the most substantial barn of those excavated but apart from its west wall its remains were relatively ephemeral, consisting of ragged post-bases (Figure 2.69).

BU RTON DA S SET T SOU T H E N D

Some care was taken with the construction of the west wall footing of the barn, which lay over the former boundary ditches, with the wall footing, 6m long by 0.60m wide, laid over a spread of rubble blocks which both consolidated the former boundary ditch fll and acted as a soakaway drain beneath the end of the barn. At either end of the footing there were rubble post-bases, that to the south, 0.70m by 0.75m, and that to the north 0.80m by 0.70m. The other walls were supported on less regular post-bases and clusters of stones. The eastern corner post-base had not survived, nor one of those along the north wall. Along the southern part of the east wall there were two additional adjacent post-bases; the spacing of these is diffcult to explain unless they represented repairs to the wall or supported a door in the end wall of the barn. In fact, the main doors are more likely to have faced each other towards the middle of the long sides although no structural traces of them remained. There were, however, patches of rubble metalling to north and south of the centre-western bay which may have marked the thresholds of the doorways. The space between these doorways could have been used for threshing, although there was no trace of any foor laid within the barn. North of the centre-western bay and opposite the putative door-jamb were two further post settings which were perhaps supports for a lean-to shed or outbuilding (2.3m wide by c4.5m long) in the angle between the barn and the western tenement boundary. To the east a sparse layer of medium rubble, c2.5m by 3.5m, could well have been metalling around the entrance to the barn. Another post-hole 3.1m from the north-east corner of the barn might have been part of another structure or a fence. Yard surfaces between house and barn The area between the house and the barn was surfaced with a number of layers of clay/clay loam with varying proportions of rubble; the density of rubble opposite the west-central bay of the barn perhaps marked the threshold of a doorway into the barn, as suggested above. To the west there was another sparser rubble surface containing a possible post-base set up against the north-west corner of the extension to the house. The post-base contained a socketed stone in the middle, possibly in situ. This could have supported the post of a gate across an entrance, 2.5m wide, to the yard from the west, between the barn and the house extension. Boundary ditches A new western boundary ditch will have been contemporary with the barn. This ran northwards from a point 1.5m north of the barn. It was c1.5m wide by 0.7m deep. This ditch lay to the west of the

T H E A RCH A EOL OGIC A L SEQU E NC E

125

Figure 2.69 Area A, extent of the barn indicated by ranging rods, from the west (Phase A5, mid- to late 15th century)

previous one and may have been dug to the west of an existing hedge or fence. Cut into the rubble surface along the front of the tenement just to the south of the house was another ditch which would have been contemporary with the new western boundary. This was U-shaped, 0.8m wide by 0.4m deep; to the west it probably ended at the south-west corner of the house; to the east it continued along the frontage of Area D1 up to the west end of the Area D2 house. While this ditch was open it would have been necessary to have a bridge across to the door of the Area A house. It may not have been open for long, however; it was noticeable that its fll contained a high proportion of rubble (30%) and it may have been designed to act as a soakaway and deliberately flled with rubble. Some indication of rubble metalling laid over the ditch formed a path up to the door of the house. This ditch may be further evidence of problems with drainage in this part of the site, along with the ditch in front of the smithy ( J4). In the south-west corner of the forecourt area in front of the house

a layer of clay loam gradually accumulated over the existing rubble surface. This probably derived from material washed down from the higher parts of the tenement. The fnal phase of the western boundary ditch was in two sections separated by a gap of c2m opposite the entranceway between the house and barn. To the south the ditch curved to the south-west running into Area B; here it was 1.2m wide by 0.28m deep. To the north it varied between a narrow (0.32m), steep-sided slot to a wider (1.1m) shallow ditch. Its depth, 0.20m, was markedly shallower than that of its predecessors. The south section of the western boundary ditch was cut by a fnal feature when partially flled. This was a stone-packed drain to the west of the west wall of the house and on a slightly divergent line. The drain was c6m long by 0.25m wide and packed with pitched medium–large rubble. The construction of the drain was very late in the site sequence but it is diffcult to imagine the construction of such a feature after the abandonment of the houses.

126

BU RTON DA S SET T SOU T H E N D

Figure 2.70 Area D1; gullies, hollows and drainage ditches of Phase D15, mid- to late 15th century. The large east–west drainage ditch at the front of the plot, which had been recut at least twice, lay just behind the frontage, suggesting that there may have been a fence in front of it. The parallel replacement ditch to the south marks the plot boundary next to the road. The Area A forecourt surfaces extended a short distance in front of D1

Area D (D15) On the west side of what had been Tenement D1 there was a fnal recut of the west boundary gully in this phase (D15). This was 0.55m wide by 0.08m deep and ran northwards from a point 2.5m north of the corner of the Area A house (Figure 2.70). Over the south part of the tenement the latest medieval ground surface was formed by a series of layers of clay loam; the material recovered included 15th-century material but very little from the later part of that century. It therefore seems that little rubbish-generating activity took place here during the latter part of the century and it is possible that the area was cultivated as a garden plot. A number of features were observed although few clues as to their function were gleaned.

A rubble post-base was situated on the west side of the area, which was conceivably part of a leanto structure, 2.5m wide, against the Area A house, although no other trace survived. Some of a series of irregular shallow hollows in the centre of D1 were cut by one of three lengths of discontinuous, east–west gully which appeared to defne a strip c4m wide across the south end of the tenement. The strip was devoid of features and its function is uncertain. A wide shallow gully, 6.5m long, 1.3m wide by 0.10m deep, along the side of the Area D2 barn may have been a drain for water from the barn roof. Running along the southern frontage of the property a series of large east–west drainage ditches ended just on the other side of the former boundary with D2. What was probably the earliest of these,

T H E A RCH A EOL OGIC A L SEQU E NC E

127

Figure 2.71 Area D2, house, from the east (Phases D24–D26, 15th century) 25.5m long by c1.3m wide by 0.8m deep, ran up to the houses on the adjacent properties. The ditch was just behind the frontage suggesting that there may have been a fence along the actual boundary. Its profle suggested that it had been recut on a number of occasions but the flls were diffcult to distinguish fully; three cuts were discernible at its eastern end, two at its western end. This ditch was succeeded by a second, similar ditch up to 1.7m wide by 0.65m deep which ran over the same length but 1.5m to the south, probably along the actual frontage, and possibly to the south of an existing fence or hedge. It also showed signs of having been recut, but again the flls were diffcult to separate. The fnal ditch along the frontage cut both the two earlier ones. This ditch did not end at the boundary of Area A, but continued westwards along the frontage of Area A. Its east end was not located exactly but it was shorter than its predecessors at c17m long, c1.0m wide by 0.52m deep. The fnal phase of the ditches across the south part of the area seems to have flled

or been flled before the fnal abandonment of the property, as there were fve shallow pits/hollows cut into the top of its fll. To the south of the ditches to the west a rubble and clay surface continued the Area A forecourt surface. This extended only c3m into the area. It contained few fnds suggesting the forecourt muck heaps did not extend over this area. Towards the east the surface became much patchier. It seems likely that there would have been a path between the Area D2 and Area A houses, but it was much less well laid than the house forecourts. D2 house (Phase D25) The mid-/late 15th century saw another partial rebuilding of the D2 house (Figure 2.71). The east end was extended up to the neighbouring house on Tenement E, whose west wall now became a party wall; the west end was shortened, but with the addition of an upper storey, reached (in this phase)

128

BU RTON DA S SET T SOU T H E N D

Figure 2.72 Area D2, Phase D25, mid- to late 15th century (compare with Figure 2.60). There has been a further reconfguration of the house, with the addition of an external stair to an upper foor at the rebuilt western end. The eastern end has now been extended to abut the house on plot E, and the timber outbuilding to the north-west has been replaced by an aisled structure of fve bays by an outside lean-to staircase, which would have increased the total foor area. The house, which retained its slightly trapezoidal plan, now measured c16.8m by 5.6m. At the same time the interior was re-arranged into three rooms, with a central hall and a separate passageway to the west of the hall (Figure 2.72). The rebuilding of the west end also involved the partial replacement of the north wall further

east (the part of the original north wall that had not been replaced in the previous phase). Only a short length of the new wall survived, following a line to the north of the earlier wall. The east end of this wall, and the west wall of the extension had been robbed out, but the widths of the robber trenches (0.80m on the west and 0.70 on the north) suggest that the walls had been stone to the eaves. The upper storey, indicated by the presence of a

T H E A RCH A EOL OGIC A L SEQU E NC E

129

Figure 2.73 Area D2, from the west (Phases D25–D26, mid-/late 15th century)

probable staircase, does not seem to have covered the whole area of the western room as there were two hearths in the room below in the subsequent phase which would have needed an outlet for their smoke. Both hearths were, however, located at the east end of the room which perhaps indicates that there was a smoke bay or smoke hood over them. On this basis, the upper room could have measured c3.5m by 4.5m. Set against the west end of the house was a rectangular foundation, 4m by 1.75m, of medium– large rubble. This could have supported a lean-to shed but seems far more likely to have been the base for the external staircase leading to the room over the western end of the house. On the north-west corner of the foundation there was a stone-packed post-hole which may have formed part of a repair to the staircase. The wall dividing off the west room, which had been the west wall of the original building, was rebuilt at this time, but on the original foundations. The new wall was 0.62m wide. The lower western room measured 4.5m by 4.4m and was foored with olive brown clay loam. This foor remained in use into the subsequent phase. To the east of the rebuilt wall there was a north– south passage or lobby 1.65m wide. This was bounded to the east by a wall footing consisting of a single course of medium rubble 0.38m wide, faced on its west

side. This would have supported a timber partition. At the south end of the partition was a doorway 1.3m wide opening into the hall. It is likely (although not certain) that the doorway into the street was now moved to the south end of the passage. Whether the north door was also moved to the north end at this stage is more doubtful. The presence of a layer of threshold metalling in the north-west corner of the hall may well indicate that the north door remained in its previous position in this phase. The passageway was initially foored with a layer of clay loam which may have been continuous with the foor in the hall; it was subsequently overlaid by a well laid surface of pitched, medium rubble, but this only survived over the north end of the passageway. To the east of the passageway was the hall. This, at 4.4m by 4.25m, was a similar size to the other rooms and much smaller than in the previous phase. The hall was divided from the eastern room by a narrow (0.2m) rubble footing, which again would have supported a timber partition. There was a doorway into the eastern room (0.8m wide) at the south end of the partition. At the east end of the house the former east wall was removed and the north and south walls were extended up to the west wall of the Area E house. The effect of this was to lengthen the room by 1.4m, so that it now measured c4.5m by 4.5m. The new area of foor was made up with layers of clay loam similar to that in the hall. In the north-west corner,

130

up against the partition there was a burnt patch, 0.58m by 0.52m, of clay, ash and charcoal. The hall was foored with brownish-yellow clay loam. In the north-west corner was a rectangular spread of medium rubble, 1.4m by 1.2m, around the presumed north door to the house. It contained a re-used door jamb section inscribed with the name ‘Gormand’ (see Section 2.8 above) which may well have been discarded when the south door was moved to the passageway; its disposal may also imply that the Gormand family were no longer the tenants. The hall contained a hearth, 0.8m by 0.5m, of medium–large, heat-reddened slabs surrounded with small rubble against the south wall (Figure 2.73). The reconfguration of the D2 house could have provided self-contained accommodation in the upper room. One may speculate who may have lived there: retired aged parents, or a newly married couple, or it may even have been intended to fulfl a formal maintenance agreement with an unrelated retired former tenant (cf. Dyer 1986, 25), possibly even a member of the Gormand family. Replacement barn/store north-west of house This period saw the replacement of the northwestern outbuilding by another larger barn/store on the same site, encroaching by some 1.5m on to the adjacent D1 tenement (Figure 2.72). Predating construction of the new building were two pits and a gully which appeared to relate to the demolition of the previous one. Immediately before the construction of the new building, rubble was laid over the south and east edges of the barn area.

Figure 2.74 Area F, Phases F5 and F6, mid- to late 15th century

BU RTON DA S SET T SOU T H E N D

This formed a yard surface or path leading from the barn towards the house, which remained in use into the subsequent phase, but it also supported the south wall of the new barn. The new building was probably a timber, aisled barn, of fve short bays, c6m wide by c9.8m long (it might have been longer; its north wall was not defnitely located). The width of the nave was 2.5m and that of the aisles 1.5m. This seems rather narrow for an aisled structure and another possibility is that the building had a normal frame and the aisle posts merely supported partitions dividing the building into fve stalls. The aisled structure seems more likely, however, as the post-holes were large, stone-packed features and appeared designed to carry structural timbers rather than partition studs. Whilst described here as a barn it is possible that it served another purpose. The aisle posts, four on each side, averaged 0.49m across and 0.26m deep, with rubble post-packing. The east wall of the building was marked by a row of post-bases made of medium–large rubble blocks at c1m intervals which probably supported a sill beam. Five post-bases survived along the east side, with a sixth, approximately 1m off-line, probably associated; they ranged in size from 0.36 by 0.28m to 0.95 by 0.20m. All traces of the west and south walls had been removed by a later building and the north wall lay probably just outside the excavated area. The inside edge of the west wall was, however, marked by the edge of a pit dug against it within the barn. This line means that the new building encroached over the original tenement boundary by c1.5m. The south wall presumably lay on top of

T H E A RCH A EOL OGIC A L SEQU E NC E

the edge of the rubble surface. The pit within the building was cut by a post-hole. It was sealed by the main foor of the usual clay loam and cut in turn by two further small post-holes. External features to north and west of house Over the yard between the house and barn was an extensive but patchy spread of clay loam which survived where it had subsided into underlying hollows (Figure 2.72). Running along the north side of the house a shallow east–west gully (0.5m–1.3m wide and 0.16m deep) cut the clay loam spread. By the north door of the house the gully curved outwards, where a second, separate gully, 0.50m wide by 0.20m deep, ran northwards to the edge of the excavation. The east–west gully was probably a drain for rainwater running off the roof of the house; the north–south gully may have served both as a drain and as a boundary dividing up the rear of the tenement. To the west of the house there was one feature belonging to this phase, a north–south gully/depression, 1.98m by 0.88m by 0.14m deep, set against the foundation for the staircase. Area F In the mid-/late 15th century the Area F house was re-arranged internally (F5), the western room being enlarged to 3m with the insertion of a replacement timber partition on stone footings, 0.40m wide with an offset (0.15m wide) on its eastern side (Figure 2.74). The doorway remained at the north end, with the rather narrow (0.7m) threshold probably marked by an area of more rubbly stonework. A new clay loam foor was laid in the western room. To the east this became worn and the top of the previous partition wall began to poke through it. The main room was also refoored at this time. In the south-west corner of the room, overlying the foor, there was a shallow hearth, 0.59m by 0.43m, consisting of an oval patch of ash bounded on one side with burnt ironstone rubble. A cluster of rubble in the north-west corner of the room may possibly have been a post-base against the north wall. Barn/store and yard At about the same time, a new timber-framed outbuilding, 5.7m wide, was built against the north side of the house along the western side of the Area F tenement. Only the southern 2.2m of this was excavated, its north end lying outside the excavated area. The east wall footing of this building was 0.40m wide, formed of coursed medium ironstone rubble bonded with olive clay loam. The northern part of the wall appeared to curve westwards, but insuffcient was excavated to say whether this was a minor irregularity or marked a real shift in line. The west wall, which ran along the tenement boundary,

131

had been removed by a later robber trench, like that of the house itself. The narrowness of its surviving wall-footing strongly suggests that the building was timber-framed or at least formed a timber lean-to against a stone boundary wall. Its size suggests it was probably a barn or store building. It was foored with a layer of small–medium ironstone rubble and olive brown clay loam, on which there was no discernible build-up of occupation material, although it produced 161 pottery sherds. To the east of the building the ground had been made up and another well laid rubble surface put down which will presumably have been an outside yard surface; this too produced a quantity of pottery (101 sherds). Area L Tenement L may have remained occupied for longer than the others in the excavated area south of the road and it is therefore shown on the reconstruction of Southend in around 1475 (Figure 2.67). There was little dating material recovered from its demolition layers (L2), but some of that from the topsoil was akin to that from the fnal phases north of the road, suggesting that it may have been abandoned in the later 15th century. Figure 2.67, a reconstruction of the site c1475, shows the plots to the south side of the road abandoned, apart from Tenement L. This is in striking contrast to the northern side of the road which shows a continuous building frontage along D2, E and F. D1 is now incorporated with D2 whilst the house on A extends the full width of its plot. All the occupied houses have ancillary buildings to the rear. 2.10 THE LATER 15TH CENTURY

The later 15th century saw renewed building activity on Tenements D2 and E, but the process of decline now began to spread north of the road with the abandonment of Area A. This fnal phase of occupation produced large quantities of all categories of fnds from Areas D2 and E. Potsherds from later 15th-century contexts in Area D2, other than demolition contexts, account for 9.8% of the total pottery assemblage from the entire excavation, and likewise 17.6% of the animal bones found in the excavation. Another substantial proportion of the pottery (4.1%) and bone (4.6%) came from Area E. Other fnds refect this trend; excluding nails and horse furniture, 34.7% of non-ceramic artefacts of all phases were derived from D2 and 27.2% from E; 35% of these were from later 15thcentury contexts other than demolition contexts (see online archive, Sections 9.3–9.5). This shows signifcant quantities of fnds from later contexts on these two areas, although these fgures need to be interpreted with caution: it is possible that a

132

BU RTON DA S SET T SOU T H E N D

Figure 2.75 Area D2, Phase D26, late 15th century (compare with Figure 2.72). In the fnal phase of occupation the western external stair was replaced by a small turret, now probably accessed from the inside of the house, whilst the timber outbuilding was replaced by a stone-founded, probably timber-framed, structure. Most of the forecourt surfaces between the house and the road belong to this phase proportion of them had been discarded when the properties were vacated, while others may have been residual material incorporated into imported foor make-up. Area A The demolition of the A5 house in this phase (A6) scattered rubble over all the immediate vicinity, and

the demolition levels produced large quantities of fnds, particularly those over the kitchen area. At the west end of the house there was less demolition rubble, which supports the suggestion that the extension here was timber-framed. The south, east and part of the north walls were robbed at this time, the robber trenches subsequently flling with clay loam and a gradual accumulation of clay.

T H E A RCH A EOL OGIC A L SEQU E NC E

133

Figure 2.76 Area D2, hearth in the south-west corner of the hall, from the east (Phase D26), late 15th century. Archaeomagnetic dating gave a date range at 95% confdence of 1440–1539 Area D2 In contrast, on Tenement D2 the late 15th century saw yet another remodelling of the house (D26) involving the removal of the external timber stair at the west end and its replacement by a new, probably stone-built rectangular stair turret, 2.2m by 3.3m, which would have allowed access to the upper room from inside the house (Figure 2.75). This may indicate that the need for separate accommodation in the upper room at the western end had passed. The earlier stair foundation was replaced by a small rectangular room (2.2m by 3.3m) built against the south-west corner of the house. The south and west walls were 0.65m wide, the north wall 0.52m wide. All were constructed of medium rubble. It seems likely that the room contained a stair which ascended from a door in the south-west corner of the room. In the western room the existing foor remained in use. A stone-packed post-hole, 0.65m by 0.60m by 0.20m deep, was cut into the foor in the centre of the room on the south side. The room contained two hearths. Both lay within 1.2m from the east wall of the room and it is likely that there was a smoke hood or smoke bay above them. One against the wall in the north-east corner measured 0.9m by 0.9m and consisted of one large, fat, burnt slab

surrounded by smaller rubble which may have supported a superstructure. The second hearth in the centre of the room was roughly square, 0.72m by 0.70m, and made of four slabs and two smaller stones, all heat reddened. The magnetic date for the last use of this hearth was between 1449 and 1539 at 95% confdence, almost exactly the same as that for the fnal hearth in the hall (see online archive, Section 8.23). Remnants of possible clay loam foor layers were recorded in the north-west corner of the room. In the entrance passageway to the west of the hall a fnal clay loam foor layer was laid on the south side. To the east the north part of the partition wall was retained but a new central doorway into the hall from the passage (0.9m wide) was created. The south end of the partition was formed by a triangular block of masonry made of large–medium rubble blocks, which also provided a fre-back for a hearth in the south-west corner of the hall (Figure 2.76). The hearth originally consisted of a strip (1.70m by 0.35m) of fat roof tiles, mostly broken in half, set vertically in a matrix of very dark greyish-brown ashy loam. Alongside the tiles there was a rectangular block of burnt ironstone slabs, 0.85m by 0.40m, perhaps representing a later addition to the hearth. A remanent magnetic date suggested a fnal use of the hearth between 1440 and

134

BU RTON DA S SET T SOU T H E N D

1539, consistent with that for the central hearth in the western room (see online archive, Section 8.23). A remnant of a new clay loam foor in the hall survived only on the east side. The eastern room was little changed from the previous phase. A new foor layer of dark greyishbrown clay loam was laid. Cutting this in the southwest corner there was a post-base of medium rubble, 0.60m by 0.40m, set in a shallow hollow.

Within the barn against the south end of the west wall a possible rubble post-base, 0.85m by 0.65m, was located with another similar feature, 0.9m by 0.65m, against the east wall. These could have supported some structure such as a manger (c1.2m wide) across the south end of the building (but this would preclude the existence of a doorway in this area). There was no trace of a foor in the barn (Figures 2.77 and 2.78).

Forecourt surfaces and muck heaps south of the house

Surfaces and muck heaps north of the house

Running westwards from the south-west corner of the house there was a short length of boundary wall, 0.55m wide. This contained a gateway, 1m wide, leading around the end of the house. The gateway involved a step up into the property (a rise of 0.18m), which was paved with one large stone, a re-used section of roughly chamfered plinth. To the south of the house lay a series of rubble forecourt surfaces of uncertain date. This area may well have been paved in previous phases, but the surviving layers contained late material as a result of patching and resurfacing. The surfaces formed a T-shape with a strip 1.5–2.5m wide running along the front of the house and a wider band (6m wide) running up to the door of the house from the street. To the east the surface ran into the similar surface along the front of Area E. To the west it petered out c3m to the west of the gateway to the rear of the property. Within the forecourt surface just outside the house, running north–south, were two lengths of what appeared to be wall foundations, one 0.63m wide, the other 0.85m wide. Both were, however, only one course deep and continuous with the general surface, and it was concluded that they had been deliberately laid as part of the surface. These surfaces all contained very large amounts of pottery and animal bone. This was presumably domestic rubbish dumped outside the house but left behind when the material in the muck heaps was removed to the felds.

Along the southern edge of the rubble around the south-east corner of the barn was a drain, 7cm wide, which widened into a rectangular sump, 0.5m by 0.33m. The area immediately north of the house was paved with a series of small–medium rubble surfaces, which merged into those around the barn. Large concentrations of fnds (more than from any other phase) indicate that there had probably been various muck heaps in the yard; these generally incorporated a higher percentage of domestic waste than those elsewhere in the settlement. The large number of Cistercian ware drinking cups from D26 (210 sherds), particularly when compared with the small quantity from the contemporary E6 phase (9 sherds) raises the possibility that the property was operating as an alehouse in this period. This was a recognised way for households to supplement their income, even for the wealthier ones, and the D2 property was well placed on the market street to attract custom, as well as being adjacent to the brewhouse on Area E. The obvious location for the serving of ale would be in the western ground foor room. The site distribution of Cistercian ware shows a marked concentration around the western end of the house (see online archive, Section 8.14). This, however, refects the location of the contemporary muck heaps to north and south of the building and material from demolition and topsoil layers over them. Relatively little of the Cistercian ware in D26 actually came from layers within the house, and none from the western room. The largest group came from the eastern room, with some from the hall and entrance passage. The general concentration of these drinking vessels, along with a concentration in Areas D2 and E of cisterns (large jars with a bung hole), another type of vessel which might be associated with a hospitality enterprise, offers some support for the alehouse suggestion. The adoption of Cistercian ware seems to have been more prevalent in urban than rural contexts; this may be infuenced by the types of site where archaeological effort has been concentrated, but it is also a refection of changes in lifestyle as the formalisation of food and drink consumption spread down the social scale. Thus the quantity of Cistercian ware may refect the wealth or social pretensions of the inhabitants of D26, or even indicate that the property continued in occupation later than its neighbours, although

Barn/store north-west of the house The latest barn/store building was a rectangular structure 8.4m by 5.4m, with an internal foor area of 27.2m 2 (293 sq. ft). It probably had a timberframed superstructure set on stone footings: its footings were quite narrow (0.55–0.70m on the west, 0.60m on the north and east) and there was relatively little demolition material over its area (Figure 2.75). A fragment of stone window sill from one of the demolition layers could have been re-used material. There was no indication as to the location of a doorway, unless the presence of external rubble surfaces, laid at the start of the previous phase, indicates the presence of one in the south-east corner.

T H E A RCH A EOL OGIC A L SEQU E NC E

Figure 2.77 Area D2, barn, from the east (Phase D26, late 15th century)

Figure 2.78 North part of Area D2, Phase D26, late 15th century

135

136

BU RTON DA S SET T SOU T H E N D

Figure 2.79 Area E, Phase E6, late 15th century (compare with Figure 2.42). The house, extended eastwards up to the gable end of the house on the adjacent plot F, is now part of a continuous building frontage separated from the road by a forecourt area. An upper storey has been added, accessed via a stair at the north end of the former cross-wing. The timber-framed porch in the angle of the west wing has been replaced by a stone-founded structure

this is considered less likely by the excavator. The distribution and signifcance of the Cistercian ware in the fnal phase is discussed further below (Section 3.1 below). Area E In the late 15th century, during its fnal phase of occupation, the house was considerably altered (E6). The west end and north porch were rebuilt, most likely to allow the addition of an upper storey,

reached by a stair inserted in the north end of the former cross-wing (Figure 2.79). In addition, the east end was extended eastwards up to the neighbouring house on Area F, with a room being built across the former side access passage, flling the whole frontage. The central part of the house was also re-arranged internally. Together these improvements gave a total foor area of 107.6m 2 (1159 sq. ft), making E6 the largest of the houses excavated. The arrangement of the upper room with access by the stair from the north porch would make this another self-contained

T H E A RCH A EOL OGIC A L SEQU E NC E

apartment, although one more integrated with the main house than the D25 upper room. Rebuilding of west end of house and north porch The rebuilding of the west end of the house involved the replacement, and presumably the heightening, of the west wall. The most likely explanation of the change was to allow the insertion of an upper room. The new wall, which survived only at the north end, was 1.0m wide. The north wall of the main building was extended across the cross-wing making a narrow room to the north, 1.3m wide, which could have accommodated a staircase rising from the east. This extension to the north wall had been robbed out, but the robber trench was c0.65m wide. At the bottom of the stair there was presumably a doorway leading into the remodelled northern porch in the angle of the two wings. The new porch was designed to give access from outside to the stair and also to the ground foor rooms. It was built of timber as was its predecessor but its north wall was now supported on a rough stone footing, 1.80m long by 0.50m wide, made of small–medium rubble bonded with light brownishgrey clay. The outer doorway now seems to have opened to the east. On the south side of the east door there was a rubble post-base. The surviving foor of the porch was a layer of clay loam and small ironstone rubble. Internal re-arrangement of house The main part of the house was now divided into four sections with a central hall fanked by two rooms to the east and one to the west. The western room was divided off from the hall by a wall footing 0.45m wide and faced only on the east side, which would probably have supported a timber partition. A gap 0.9m wide at the northern end probably marked a doorway. This was opposite the northern porch and the northern doorway would therefore have opened both into the hall and the west room. A south doorway, opening into the hall, may have been located opposite the north one, but the evidence for this is circumstantial. The room to the east of the hall was divided off by a new, but raggedly constructed, rubble wall footing, c0.50m wide, with a gap at the southern end c0.70m wide representing a doorway. A socketed stone in an adjacent layer of demolition rubble may have supported part of a timber door frame. The hall was 6.5m long by 4.4m wide. The earlier main central hearth was replaced, or augmented, by another rougher one to the east, 1.0m by 0.60m, made of seven large fat ironstone rubble blocks. It is also possible that the earlier hearths on the south side of the room remained in use in this phase. The western room was trapezoidal, measuring about 4.2m by 4.4m. Against the west wall a

137

rounded rubble-flled hollow was possibly a postbase supporting some part of the room’s internal structure. To the south along the wall there was a spread of ash and charcoal, 1.3m by 0.65m. This was not substantial enough to be a permanent hearth and will have derived from some temporary activity acceptable in a room with an upper storey. About 0.80m out from the south wall a rectangular cluster of medium rubble, 0.40m by 0.35m, may have been another post-base. The eastern room measured 4.05m by 4.25m. In its north-east corner a length of rubble foundation, 1.2m long by 0.30m wide, against the east wall may have been the base of a step leading through the wall up into the new room built across the former entrance passageway. (The difference in foor levels would have been around 0.32m.) There was a possible hearth in the north-west corner of the room. Extension to east end of house This phase saw the extension of the house eastwards by the addition of a room (3m by 4.5m) across the former access passageway. Before the room was built the ground here and to the north was raised by about 0.18m by dumping a layer of clay loam and small rubble, containing quantities of pottery (460 sherds) and animal bone (414 fragments). Only small sections of the south wall of the room survived; it was perhaps 0.55m wide, of roughly laid medium rubble and, judging by the depth of the robber trench, much less substantial than the walls of the main range. The north wall, 0.5m wide, was also poorly constructed, of small–large rubble. These walls were therefore probably footings for timber framing. Access to the main range of the house seems to have been by a door in the north-western corner of the room. The only internal structure found was a cluster of medium–large rubble, perhaps a post-base, set in a shallow hollow. External features on Area E To the north of the house there were rubble yard surfaces along the north edge of the house, to the east of the brewhouse and across the north-east part of the excavated area. The surface north of the house contained large quantities of pottery (441 sherds) and animal bone (389 fragments), which may have marked the site of a former muck heap/midden. The rubble-free area in the centre of the yard was cut by a shallow hollow with a dark fll which may have marked another midden, although this one contained virtually no pottery, suggesting that any heap here was of organic waste, perhaps from the brewhouse. Three other hollows to the north-east were less distinctive. To the south of the house there was an extensive rubble forecourt surface running down into the

BU RTON DA S SET T SOU T H E N D

138

Figure 2.80 Area F, house, from the east (Phases F5–F6, mid-/late 15th century)

street. Although it is dated to this period, it may well have been laid much earlier and had been continually patched and reworked. It also contained very large quantities of pottery (540 sherds) and animal bone (1,100 fragments) suggesting that further muck or rubbish heaps had existed here. The presence of further large quantities of pottery and bone in the overlying demolition and topsoil layers suggests that the pottery and bone had not simply been incorporated into the forecourt surfaces but had been heaped up and subsequently spread out when the houses were demolished. Area F The late 15th century also saw a further, fnal rebuilding of the house on Tenement F (F6). The new house had an upper room at its western end reached by an internal stair (c0.8m wide) up its north side (Figure 2.74). This gave a total foor area of over 52.36m 2 (564 sq. ft). The new house had the same basic plan as its two predecessors—a range parallel to the street— but at 6.2m it was wider than either. Again, the walls appear to have been timber-framed on stone footings. The new north wall was c0.50m wide. No trace of the south wall survived, having been robbed out, but there were a few fragments of the footing of the west wall at the bottom of its

robber trench (Figure 2.80). It is possible that the west and south walls were not rebuilt but retained from the previous phase. The increased width must mean, however, that there were new trusses and roof structure and it is more probable that the whole frame was replaced. Nevertheless, some components of the previous building were reused in the new one. The western 4.8m of what had previously been the main north wall, as far as its junction with what had been the east wall of the outbuilding constructed in the previous phase, together with the 1.2m section of the eastern outbuilding wall now encompassed within the new structure, survived two or three courses high. The most likely explanation of the retention of these walls appears to be that they were adapted to support a staircase to a room over the western end of the building. Allowing for the timberwork to be slightly narrower than the footings, the stair would have been c0.80m wide. The west bay partition wall also seems to have been retained in the new building. The only trace of foor surviving in the new building was a patch of small ironstone rubble and clay loam forming a surface up against the south side of the new north wall, a metre beyond the foot of the putative stair, which might have formed a patch inside a north doorway. A new timber-framed outbuilding 3.6m wide, narrower than its predecessor, with a 0.60m wide

T H E A RCH A EOL OGIC A L SEQU E NC E

east wall and a foor of rubble and clay loam ran northwards from the north-west corner of the new building. New rubble surfaces were laid over the yard to the north of the house and the forecourt to the south. There were quantities of fnds from this phase, but even allowing for the smaller area excavated the total seemed low compared with those of the D26 and E6 assemblages. 2.11 THE ABANDONMENT OF SOUTHEND AND ITS AFTERMATH (AD 1495 ONWARDS)

The eve of depopulation By c1495 therefore, among the excavated tenements only D2, E and F remained occupied (Figure 2.81). The excavated area on the south side of the road was by now completely abandoned, although it is possible that some part of the Z2 building complex to the west remained occupied. On the north side of the road, the Area D2 house had just been remodelled and a new barn built (D26), the Area E house had just been enlarged to its greatest extent (E6), and the Area F house had also just been rebuilt on an enlarged scale (F6). The buildings on the occupied tenements now flled their whole frontages giving this small part of the settlement an almost urban appearance. This was the position on the eve of the depopulation. While the settlement as a whole had clearly been in decay for most of the century, it is also true that the existing community as represented by the individual inhabitants who remained was really quite prosperous and not ‘struggling to keep their lives together’ (Dyer 2012a, 140). How much warning the inhabitants had of the events of 1497 is uncertain. While it is most unlikely that they would have been aware of John Rous writing in 1486, they would probably have heard the concerns expressed by opponents of enclosure and would certainly have seen the continuing process of depopulation in progress over many of the surrounding parishes. They will have observed the frst enclosures in the southern feld of Southend by Roger Heritage and noticed the consolidation of holdings in the hands of the manorial lord. If the catalyst for the extensive enclosures and depopulation at Burton Dassett was the receipt of full title to the manor by Edward Belknap in 1496, and the succession as lessee of the demesne by John Heritage, following the death of his father Roger in 1495, both these young men would have been known locally as ambitious and acquisitive (Dyer 2012a, v, 34, 222) so their actions may not have come as a complete surprise. It is possible that not all the standing buildings were dismantled or demolished at the precise moment of enclosure, although the presence of an occupied house would have frustrated the aims of laying Town Field down to pasture. The high proportion of the total number of Cistercian ware sherds coming

139

from D2 might indicate that D2 was, perhaps, the building most likely to be in occupation after 1497, although Stephanie Rátkai concludes that ‘it would be unwise to assume that the much higher quantity of Cistercian ware in D2 indicates that this property was necessarily the last one to go out of use or that this was at the very end of the 15th century’ (online archive, Section 8.1.5). The central point of the archaeomagnetic date range for the two main hearths in the D2 building is in the 1490s decade, although the revised calibration now places them within a rather longer time frame (1449–1539 and 1440–1539 at 95% confdence; online archive, Section 8.23). Demolitions in 1497 (D16, D27, E7, F7) As elsewhere, the demolition spread layers of rubble over the former house sites while the better usable building stone was removed for re-use. In fact, the robbing of the main walls of the D2, E and F houses was quite thorough, removing even the foundation courses over long stretches, perhaps suggesting an organised programme. One imagines that the local market in second-hand building material would have been depressed by over a century and a half of abandonment and demolition of buildings and totally saturated by the further wholesale demolitions. Nevertheless, there may have been a steadier market for building stone further afeld where building stone was scarce, and as the depopulating landholders wished to turn the village site into pasture, to achieve a cleared site, they may have been prepared to remove the materials of value and either stockpile them or sell them on. Belknap’s 1497 leaseback of Burton to John Heritage reserves the right to ‘lykewyse in all other manner of howsyng’ with access to the housing ‘in building and carrying away’ (see Section 1.8 above), which indicates that he was concerned to maximise his profts and salvage any saleable materials. Timber, too, would have had potential salvage or resale value within the locality, for example where tenants had been instructed to repair or rebuild houses, or where they did this on their own initiative (Dyer 1982, 27–28; 1986, 22–23). Following the removal of the usable building material the site seems to have been fattened. This involved the spreading out of the muck heaps to the north and south of the houses, judging by the large quantities of pottery and animal bone incorporated into the topsoil. The material does not seem to have been moved any signifcant distance, however, because the concentrations of rubbish remained around the buildings rather than over their interiors. The robbing of building materials was not total, as the roof tile and slate from the buildings of the Z2 complex in Dovehouse Close were abandoned on site. Also, despite the methodical robbing of building materials, judging by the quantities of metal objects found both in the demolition layers over Areas D2, E and F and the post-medieval and

140

BU RTON DA S SET T SOU T H E N D

Figure 2.81 Phase plan, c1495. Occupation within the excavated area immediately prior to enclosure and depopulation was now restricted to a continuous row of buildings on plots D1/D2, E and F, together with their associated outbuildings

T H E A RCH A EOL OGIC A L SEQU E NC E

topsoil layers overlying them, there seems to have been no scavenging of the area for recyclable metalwork, as there may have been to the south of the road. The depopulations did not remove all the houses in Southend. Two farms remained south of the road, to the east of the excavated area. One of these, occupied by William Makepeace in 1567, would have been on the site of the existing Little Dassett Home Farm, the other, described as a ‘messuage and orchard’ occupied by Thomas Burbery in 1567 probably lay to its west in Barn Close. There were also surviving buildings to the east of the road to Northend. These included a house tenanted by Thomas Brooks in 1567, on the site of Windmill Cottage, and probably a cottage to the north recorded in 1582. The chapel also remained standing, probably out of use, and increasingly dilapidated, judging by the work necessary in 1632. Post-medieval activity on site in 1500–1825 After the depopulation and the laying down of the site to pasture, there is very little evidence of activity in the excavated areas. Running east–west through the former smithy building on Area J there was a shallow ditch, probably dating to the 16th century and possibly for drainage. To the south two similar drainage ditches were dug across Area H, probably in the 16th century, one running north-west to south-east through the former house; and a second c24m to the south, running north-east to southwest. Just to the south of the latter there were one or two very large, shallow, circular pits, possibly contemporaneous, of uncertain function but perhaps for the disposal of noxious waste. In the 17th century, two successive drainage ditches were dug across the south-west part of Area A. Elsewhere the f ieldwalking produced some evidence of the new agricultural regime. At the western end of Dovehouse Close, just beyond the limit of the medieval settlement, in the feld later known as Bullock’s Close (Z4), there was a scatter of post-medieval material which would have derived from the ‘cottage or Sheppardes house’ recorded in the 1567 partition of the manor (see Figure 1.9). This, which was occupied by the shepherds in charge of the fock in Traversfeld which lay to the west of the main road, was probably built after the enclosures. The pottery suggests the occupation here began in the 16th century and continued through to the 18th century, although no building is shown here on the 1769 Temple Estate map. It is also possible that the earthwork enclosures in Area G on the north side of the road (Figure 1.16) were sheep folds belonging to this period, although they were not sited immediately opposite the cottage but c100m to the east, and the trial trenching did not produce any features or datable fnds of this period.

141

The distribution of iron ox shoes from the site may also be signifcant (see online archive, Section 8.7, nos 511–521). Although 11 were found, only one came from a medieval context; the others were in the topsoil, north of the road or from feldwalking at Dovehouse Close. Of these two were of medieval type, but the others may possibly have derived from cattle from Wales shod for the journey to market in London who were being fattened up at Burton Dassett on the new pastures. In 1632 the recusant Lady Wotton of Northend Manor brought the chapel back into use as her private chapel and built the two-storey Priest’s House against the east chapel wall for her chaplain (see online archive, Section 7). By the time of the Civil War, she had left Burton Dassett and the chapel had presumably once more fallen out of use. These activities seem to have left little trace other than to the buildings themselves. The feldwalking in 1991 did record a concentration of 17th- to early/ mid-18th-century material on the west side of the road some 100m north of the chapel (Y1; Section 1.12 above). This does not seem to be related to the short-lived re-occupation of the chapel, but it is possible that there may have been a house or cottage here during this period of which no other record survives. Again, no building is shown here on the 1769 Estate map; and the material might alternatively be the result of dumping of rubbish from the house opposite, to the east of the road, on the site of Mill Cottage. After 1500, the excavated area seems to have remained largely as permanent pasture until the late 20th century and the coming of the motorway. In Areas A, D1, D2, E and B, however, attempts had apparently been made to cultivate Town Field/ Chapel Ground over a short period. A series of late plough furrows were found cutting into the medieval demolition layers. The episode seems to have been short lived—the plough furrows remained as discrete and had not merged together into a continuous ploughsoil layer. The furrows produced no dating evidence, although in Area E they were cut by the later feld drains. These contained 3-inch (7.6cm) clay drain pipes which are likely to date to the 19th century; drain pipes and tiles were exempted from tax from 1826 and their price dropped by some 70% following the invention of a machine to extrude pipes in 1845, while government grants became available for drainage from the 1840s (Robinson 1986, 79; Wade Martins 1996, 102–105). This suggests that the episode of ploughing may have taken place just before the drainage pipes were laid, perhaps in the late 18th/early 19th century when a succession of poor harvests and the effects of the Napoleonic wars led to high grain prices and a short-lived expansion of arable acreages.

3 SPATIAL ORGANISATION AND THE BUILDINGS AT SOUTHEND

3.1 SPATIAL ORGANISATION

Communications The position of Chipping Dassett in relation to the contemporary road network is likely to have been critical to its success as a commercial venture. We cannot be certain of the road pattern in the medieval period. The nearest highway of which we can be sure is the Fosse Way, some 7.5km away, although the road from Warwick to Banbury, which now passes just to the west of the settlement (B4100), is likely to have taken something like its present course; the existing road was turnpiked in 1725–26 (Cossens 1946, 85). For the main road to have passed along the road through the excavated area would have involved continuing up the (steep) road to the summit of the Burton Dassett hills and then south-east along the spine of the hills towards Farnborough or turning south-east by the chapel and continuing along the fank of the hills to rejoin the present (M40/ B4100) routes to the south. While it is diffcult to be sure of the route taken through Burton Dassett, the topography of the 13th-century plots fronting on to a wide street is typical of market settlements. They must have placed their houses in that position because they expected interaction with the trade and traffc on the road. Plot layout All the excavated areas showed evidence for deliberate planning, and were apparently laid out shortly before the construction of buildings. As described in Chapter 2, the layouts differed either side of the axial road. To the south were two rows

of plots notionally 3 perches wide; those at the front (L, N and J) only 6 perches deep, those to the rear apparently 12 perches deep (although the southern boundary was only observed in a trial trench at the far end of Area K), with a gap 2 perches wide between the two blocks for access to the rear block via a road occupying one of the front plots. The western rear block, Area K, occupied the width of three of the 3 perch plots. The total depth of the two blocks with the access way was thus 20 perches. West of the excavated area, south of the road, the distribution of rubble recorded during feldwalking was consistent with the planned units having continued in this direction; rubble spreads at the edges of areas Z2 and Z3 were equivalent to 20 perches north–south and 32 perches east–west, making 4 acres which, if plot widths were consistent with those excavated, would have provided space for another eleven plots. To the north of the road the tenements (A, D1, D2, E and F) were 4 perches wide and 20 perches deep (and thus a half acre in area), ending at a back lane. The disparity between north and south refects their having been laid out in two stages. Salvage recording west of the spur road on the western side of Area A showed that there were no buildings here; the plot pattern therefore appears to end at the spur road. Multiphase planning has often been recognised from analysis of settlement morphology, but less frequently in excavation. To the east of the excavated areas less regularity was evident amongst the earthworks. The zone of earthworks north-west of the Chapel (Y1 on Figure 1.18) showed less regularity, although they were contained within boundaries showing some resemblance in terms of shape and dimensions with those around the regular plots of Y3, which

BU RTON DA S SET T SOU T H E N D

144

included the area excavated north of the road; this area, surrounded by roads or lanes on three sides was apparently occupied earlier than the excavated tenements. Layout of buildings on plots All the excavated houses were set on the edge of their plots along the street frontages. While this is characteristic of an urban pattern, it is also typical of villages in the region, continuing so into the postmedieval period (Wood-Jones 1963). Judging by the earthworks, most of the other properties in Southend are likely to have conformed to this pattern, with the possible exceptions of the areas identifed during feldwalking as Z2 and Z3 (Figure 1.18); the rubble within the surface layers here was concentrated some 40–80m back from the frontage, and the part of Z2 that was actually stripped (Area M) contained a boundary wall adjacent to the road but no sign of a building. Insofar as the rubble spread recorded on Z2 can be relied upon, the distribution was not incompatible with a large stone-walled enclosure; this may have been integrated with or imposed upon the 13th-century layout. The quantities of roofng materials in this area suggest that buildings collapsed in situ rather than being demolished. During the early phases, before the street frontage became fully built up, Tenements D1 and possibly E2, had outbuildings on or near the frontage, but the outbuildings on D2, E, F, H, I and L (and J) were for the most part all set out along tenement boundaries running back from the houses. This is also a pattern common in towns but in both cases is really dictated by the narrowness of the tenement plots. There was a preference for the western side of the plot, so that the entrances would have been on the more sheltered eastern side. On the southern side of the road there was also a tendency to place outbuildings next to the tenement boundary, although here a few small structures (I3, J4 and the K4 possible kennel) were constructed against the east tenement boundary. On Plot K, however, wider than the other plots, the granary was set behind the house away from the plot boundaries; this may have been in order to assist the manoeuvring of carts, or may simply refect that there was less constraint on space here. Property boundaries On the northern side of the road, all the identifable plot boundaries consisted of gullies or ditches, which would presumably have been supplemented by fences or hedges. In some cases new boundary ditches were dug parallel to their predecessors leaving a gap that might have been occupied by an existing hedge or fence. More variation was evident south of the road. Area J was delineated by ditches (later replaced by a wall as the water hole on Area W developed),

and the boundary between H and I was also laid out as a ditch. Elsewhere there were stone walls; on the northern (street front) and eastern side of Z3/M, the southern limits of Z2 and Z3, the L/N boundary, and the K/H boundary. It is possible that some of these stone features had replaced earlier ditches or gullies, since the earliest phases were less extensively examined here. The apparent early/ mid-14th-century encroachments of Area K towards the north-west and Area I towards the north-east were also bounded by stone walls. Additionally, as noted above, the distribution of rubble recorded during feldwalking suggests a possible stone-walled enclosure within Area Z2. Dating the boundary walls is diffcult; the plots on this side of the road were earlier than those to the north, and the pottery also points towards early settlement very close to the excavated areas. The walling around K presumably pre-dates the area’s apparent reduction in status in the mid-14th century. We cannot be certain whether the differences in boundary structure were chronological, or related to status. If the wall around K was continuous, as seems likely, then its length would have been in the order of 200m, whilst that around the Z2 enclosure would have been around 390m, a considerable investment in labour, even if shared between neighbours. The use of stone walls as tenement boundaries is relatively unusual; there is little evidence for them at Wharram, Goltho, Barton Blount, or Great Linford, for example, but the availability of suitable material may have been a factor, since some of the boundaries at West Cotton, also on the Jurassic belt, were stone. There was also some evidence for ditches along the street frontage in areas A, D1 and J; these were generally late features. There was less in the way of evidence for boundaries sub-dividing plots, although the gully across most of the width of A presumably served to divide the yard from a garden or paddock area, whilst a discontinuous gully at the southern end of Plot D1, dated to the mid-/late 15th century, some considerable time after its incorporation into Plot D2, marked a narrow strip some 4m or so wide along the street end. Yards Apart from the outbuildings and muck heaps, the most prominent feature of the yards behind the houses were the rubble surfaces which formed paths between the houses and outbuildings, hard standings and bases for muck heaps. The rubble for these surfaces was presumably brought in from a quarry on the Burton Hills and would represent some investment of resources. Given the propensity of the Charmouth Mudstone to remain waterlogged for long periods from autumn into spring and to turn to slurry when walked on wet, these surfaces were, however, really a necessity.

SPAC E A N D BU I L DI NG S

The only other yard features of note were the small ring gully (Figure 2.65) in the E5 yard which may have been worn by the passage of feet operating a mill, perhaps used to crush malt for brewing, and the rectangular stone structure in the H4 yard (Figure 2.51) which may have been the foundation for a trough. Drainage Considering the tendency of the Charmouth Mudstone to waterlogging, there was relatively little evidence for on-site drainage. Relatively short lengths of rubble-flled soakaway drain were recorded on H3, below the putative stair of the K2 granary and (possibly) outside the east wall of the K4 house. A soakaway drain was inserted through the make-up layers laid down prior to construction of the extension to the A5 house. There was also a drain through the rubble surface south of the D26 barn/ store building, and hints of roadside drainage gullies either side of the spur road on Area B. This meagre evidence is in contrast to Goldicote, Warwickshire (Thompson and Palmer 2012, 95) where there were several slab- or pebble-lined drains, at least one of which had been capped. In the later phases at Southend, drainage does seem to have been an issue. There was repeated recutting of the ditches along the western boundary of A during the early 15th century, and a ditch was dug at the front of the smithy on Area J. The deep ditches along the road in front of A and D1, and the drain in front of the D26 barn also belong to this period. It is tempting to see this as a possible response to climatic deterioration (e.g. Lamb 1965; Parry 1978; Campbell, 2016), although diffcult to prove. By the early 15th century the worst of the climatic downturn may have been over, although data are not available at suffcient resolution to be able to identify short-term or local climatic variations with any confdence. Rear plot uses It may be assumed that cultivation of vegetables and garden crops took place at the rear of the plots, and the northern part of Tenement A, accessed via a narrow gap at one end of a ditch at right-angles to the axis of the plot, has been tentatively identifed as a garden. The charred plant remains threw little light on this aspect of the settlement, however. Species such as leeks, cabbages, herbs and fax were absent. There were fragments of Prunus sp. fruit stone (from sloe, bullace, damson or cherry) and one fragment of hazelnut; these may have grown in the hedges which are likely to have grown along the tenement boundaries or been gathered from further afeld. Possibly more data would have been forthcoming had more of the rear areas of the tenements been excavated, most of the external areas which were

145

investigated being the yards and muck heaps close to the buildings, and the reconstruction drawings (Figures 2.22 and 2.67) show token patches of vegetable plot and orchards across the area. Public space About 25% of the excavated area covered space that would have been public. This includes the streets, water hole and the forecourts to the houses. Streets The east–west road across the site was not investigated, having continued in use as a road up to the time of the motorway construction. The most extensively examined section of road was the north–south road in Area B which, although very rough, showed some care in construction as it was cambered, and the spur road which led in a north-westerly direction from its western side was similarly constructed, although narrower (Figure 2.14). The existence of a metalled surface may indicate that the area to which the spur road led was not merely used for grazing; this area of Town Field/Chapel Ground was noticeably devoid of ridge and furrow and was possibly fenced off. It is not inconceivable that this uncultivated area stretching from the edge of the built-up area to the main Banbury to Warwick road was the fairground. A fair on the feast of St James in July would not be able to use arable land with a growing crop, so this area may have been maintained as pasture and reserved for this purpose. The limited amount of metal detecting carried out during motorway construction, however, produced no concentrations of metalwork here, as are sometimes found on fairground sites. Another north–south road giving access to the southernmost block of properties lay between Tenements N and J; only the eastern edge was examined, but its width was recorded as about 7m. In all likelihood it continued around the rear of Tenement J so as to provide access to the water hole as well as linking with the rubble surfaces north of H and K. It was noted that the rubble surfaces over the streets and the forecourts were more extensive and robust than those over private yards, perhaps an indication that maintenance was enforced through community pressure and the manorial court. Water hole The water hole would also have been a community facility. The adjacent forecourt surfaces north of Areas H and I produced fewer fnds, and while this may refect the fact that they were less extensively excavated it may also suggest that no rubbish heaps were allowed to accumulate here where space was restricted. Communal water holes appear to have been the norm, and likely examples from Hangleton (Sussex),

BU RTON DA S SET T SOU T H E N D

146

West Whelpington (Northumberland), Wythenail (Northamptonshire) and Goldicote (Warwickshire) have already been noted in the account of area W1 in Chapter 2. Nevertheless, some village tofts had their own water supply (Astill 1988, 57) in the form of wells or water pits to collect run-off rainwater, as suggested at Goltho and Barton Blount (Beresford 1975, 44). At Caldecote in Hertfordshire, wells on crofts A, B and E were located in corners of the farmyards, 16–17m away from the entrances to the farmhouses (Beresford 2009, 109). At Grenstein in Norfolk, there was both a public pond and three others, presumably private, within individual tofts (Wade-Martins 1980, 107, fg. 49). At Westbury, Buckinghamshire, a pit dug into the boundary ditch of Croft 20, perhaps originally for clay, seems to have been subsequently used as a water collection sump, in the absence of any other water source (Ivens et al. 1995, 180, fg. 100). At Riseholme, Lincolnshire, a stone-lined well was situated within what seems to have been a courtyard at the front of one of the more elaborate tenements (Thompson 1960, 104–105, fgs 30 and 33). At Wharram Percy, North Yorkshire, the village was dependent on the stream and pond for water, as shown by the number of jugs in the pond (Dyer 2012d, 331). Notwithstanding the higher than average concentration of ceramic jugs around the water hole at Southend (online archive, Section 8.1), it would not have been suitable to provide water for domestic purposes, but no wells were found on any of the tenements. It is possible that there was a communal well (or wells) sited in an unexcavated part of one or other of the streets. Wells are less commonly found on medieval village sites than on urban ones. At Great Linford, Buckinghamshire, although there were water holes for animals within the crofts (e.g. Croft F), and a 15th-century pond on Croft L, no wells were recorded, and it is suggested that the inhabitants obtained water for domestic purposes from a spring over 150m east of the excavated crofts (Mynard and Zeepvat 1992, 71, fg. 22 and 91, fg. 33). At Upton, Gloucestershire, the main water supply seems to have been a spring (Rahtz 1969, 98). The nearest springs at Burton Dassett are on the Burton Hills about 900m to the east, an inconvenient but not impossible distance away, especially if a conduit or channel was used. There is a small stream some 200m to the south-west, but it is uncertain whether this would have been a suitable water source. Forecourts Rubble forecourts were recorded in front of all the tenements on the northern side of the road. These were for the most part developed during the late 15th century, and like the drainage ditches cut through them in front of A and D1, may have been a response to wetter conditions. There was also some evidence for a rubble surface south of

the road along the front of area L, although this was only cursorily investigated. The amount of domestic rubbish, pottery and animal bone found in and on the forecourt surfaces south of the houses in Areas A, D2 and E indicates that this notionally public space could be more or less appropriated by householders for their own use, in this case for the siting of muck heaps. The forecourt surfaces leading up to the houses were presumably laid and maintained by the householders, perhaps as a communal co-operative activity, which was probably felt to give them some claim to the use of the space. This was presumably tolerated because it could be done without causing a nuisance to anyone else, given the width of the street compared with the narrow ones found in towns, so that rules restricting rubbish accumulation did not need to be so strictly enforced. The pattern of pottery cross-joins suggests that there may, however, have been episodes of clearance when rubbish was removed from the front of the houses and taken behind the houses. Area B road verge The verge between the Area B road and the western boundary of Area A can also be considered as public space, although the cluster of 14th-century features outside the south-west corner of Area A may indicate that it was informally appropriated by the occupiers of Area A; by the end of phase A5 this area was also being used to provide access to the rear of the tenement, through the gap between the house and the barn. Chapel The construction of the chapel towards the end of the 13th century may be taken as an expression of the success of the new market and the enlarged settlement, now suffciently populous to make such a venture worthwhile, and suffciently prosperous to be able to afford it. The chapel would also have been an expression of the community’s identity, cohesion and sense of belonging. It is worth noting that the main parish church of All Saints also saw enlargement and alterations throughout the occupation of Southend, even though the topographical constraints of the site necessitated several changes in foor level. The main episodes of enlargement at All Saints took place during the later 13th century and are probably indicative of growing population; here there was money to be spent on ornate capitals along the arcade of the northern aisle. In contrast, the simple chapel of St James is indicative of prudent expenditure, with re-used stones being used on door imposts and probably elsewhere in the wall fabric. The Southend community’s pride in possession of a chapel would have been secondary to their belonging to the parish. Major events such as weddings, funerals and Easter worship would have taken place in the parish church,

SPAC E A N D BU I L DI NG S

to the upkeep and furnishing of which they would have been expected to contribute. About 20% of the money to pay for the church fabric came from Southend. Other communal assets There would have been other buildings within the public realm although outside the areas examined. The Booth Hall is known from documentary sources of the later 15th century, although it may have been erected as part of the general expansion in the latter part of the 13th century. It would have been a physical manifestation of Southend’s market function, and would no doubt have been situated in one of the wide streets, perhaps in Newland. By the late 15th century there was also a shop, possibly a seld with individual stalls. There may well have been a communal bakehouse. No baking ovens were identifed in any of the excavated buildings. At West Cotton, baking ovens were identifed on some of the individual tenements (Chapman 2010, 170–172, 188–189), but elsewhere, such as at Cosmeston, Glamorgan, where the bakehouse incorporated a malting oven and was sited in isolation next to the main street (Newman and Parkhouse 1989, 9), they were evidently communal facilities, perhaps in order to make more effcient use of scarce fuel. We may also assume the existence of a mill. A miller is mentioned in the 14th century, as well as a ruinous mill in 1369 (see Section 1.7 above). He may have operated a mill in the obvious location at the top of the hill, where there was a windmill in the post-medieval period a short distance from the extant Beacon Tower (probably built in the 15th century and itself a somewhat unconvincing candidate for a tower windmill), or perhaps a horse mill was working at some other location. Patterns of rubbish disposal and fnds distributions Categories of rubbish As on most village sites, rubbish at Dassett Southend seems to have been accumulated in muck heaps or middens around the properties for later disposal on the felds or elsewhere (Beresford and Hurst 1971, 116), a pattern observed for example at Wharram Percy Area 6 (Wrathmell 1989, fgs 21 and 23). The use of large, deep pits for refuse disposal, common on urban sites, was not practised. It seems possible to divide the waste that would have been generated on site into a number of categories. Primarily there is domestic rubbish emanating from the houses. This would have consisted largely of ash from hearths, soiled foor coverings and other sweepings from the foors, and animal bone waste from food preparation. Broken

147

pottery would be a more occasional addition, but because of its durability it bulks larger in what survives to be excavated. There would also be other broken domestic equipment of wood, leather or stone, of which only the stone and metal fttings from the wood and leatherwork survive. Broken metal objects would presumably be recycled where possible, but some fragments would go into the general rubbish. The ash and organic waste would be suitable as fertiliser for the felds. One might expect food waste to be fed to pigs and other animals, and the bone does exhibit signs of gnawing, mainly by dogs, but the distribution of animal bones correlates closely with that of pottery (Figures 3.1 and 3.2), suggesting that they ended up on the same heaps. Of the 91 contexts containing over 100 fragments of animal bone and the 104 contexts with over 100 sherds of pottery, 58 contexts were common to both groups. The incidence of gnawed bones was consistently around 6% across all species except horse, from which we may infer that a proportion of the bones had been moved from their original location by dogs, and that butchered remains of different species were probably disposed of in a similar fashion, since their exposure to gnawing was apparently the same. The higher incidence of gnawing noted on horse bones, at 11.5%, might mean that horsefesh was being fed on the bone directly to dogs, but the sample is relatively small and the high fgure may be a statistical quirk. There is, however, a contrast with the West Cotton assemblage, where at least 15% of post-cranial bones had been gnawed (Albarella and Davis 2010, 520). As elsewhere, the Dassett Southend houses seem to have been swept regularly so that domestic rubbish did not accumulate inside but was removed by the external doorways to the backyards, or, in the case of the properties on the north side of the road, either to the yards or the front forecourts. The general distributions of both pottery (Figure 3.1) and animal bone (Figure 3.2) show the largest accumulations immediately around the houses rather than within them suggesting that the muck heaps for general domestic waste were in these areas (excluding the large concentrations over the Area A house derived from topsoil material). The evidence from the pottery cross-joins (Figures 3.3 and 3.4) provides some hint as to the trajectory of rubbish, although some caution should be exercised in the interpretation due to the limitations of the methodology applied (online archive, Section 8.1). A large number of cross-joins were identifed (involving some 57 vessels in Area A, 51 in Area D1, many more in D2, but fewer in E and F); some joining sherds came from several contexts, some from different phases, and some from different tenements. Given the number of crossjoins which were identifed, some selection was necessary in order to illustrate the data; Figures 3.3 and 3.4 show selected categories of cross-joins, rather than the entirety of the data. The sherds from

148

Figure 3.1 Pottery distribution across all excavated areas

BU RTON DA S SET T SOU T H E N D

SPAC E A N D BU I L DI NG S

Figure 3.2 Animal bone distribution across all excavated areas

149

150

BU RTON DA S SET T SOU T H E N D

Figure 3.3 Broken pottery vessels showing sherd links across the northern part of the site; selected categories. The lines represent a single vessel with sherds occurring on all fve tenements.

SPAC E A N D BU I L DI NG S

Figure 3.4 Broken pottery vessels showing sherd links across the site

151

152

one vessel occurred in all fve tenements (indicated by the arrowed trace on several of the diagrams); it is diffcult to devise a convincing explanation for such a singular distribution. As the positions of each sherd were not individually plotted, the nodes shown on Figures 3.3 and 3.4 represent the approximate centres of the contexts from which the sherds came. Each sherd or sherd group is shown linked to the nearest neighbour with identifed joins; however, some of the links thus derived may not represent actual trajectories. Even when selectively plotted by particular criteria (such as specifc rubble surfaces or phases of foor), however, it is diffcult to discern clear patterns. It is highly unlikely that some sherds from a recently broken pot would be disposed of to the front of the house and others to the rear; far more likely is that domestic refuse was being moved from the back of the houses to the front, or vice versa. The patterns of cross-joins from the Areas A5 and D26 forecourt rubble surfaces (Figures 3.3, 4 and 3.4, 5) may confate the trajectories of different groups of sherds removed from the houses to the front and the back. However, the nodes with the largest number of intersections on these plans, and on the cross-area and Area A plots (Figure 3.3, 1–2), may indicate the approximate locations of the muck heaps onto which pottery was discarded. A second category of waste would be from both animals and crops, consisting largely of dung and bedding, crop processing waste from threshing or malting, and possibly also human waste from latrines. Agricultural waste would be the largest category by volume and also attractive as fertiliser for the felds. This would also be collected in muck heaps in the yards behind the houses and it might not contain much artefactual material. For those without landholdings to fertilise, this was a saleable commodity and there might therefore have been an incentive to keep the product free from extraneous artefacts. There would also be industrial/craft waste. Some occupations, such as tailoring or shoe-making, would have left relatively little inorganic debris and, if peasants were generally engaged in crafts on a small scale, the detritus of their activity would be widely dispersed. At Dassett Southend this consists almost exclusively of the iron-smithing slag from Area J. Otherwise there were only negligible quantities of copper-alloy working waste (online archive, Section 8.3, nos 193–199), insuffcient to indicate working on any signifcant scale, and occasional objects such as the stone mould (online archive, Section 8.9, no. 87) and a handful of offcuts from bone working (online archive, Section 8.8, nos 19–21), all of which were incorporated into the general domestic waste. Ironworking slag is a waste product whose only attraction would be as infll material or as metalling for surfaces. Although it is found on felds in surface assemblages it will have got there as a background constituent of general rubbish. Its distribution on site (Figure 3.5) therefore concentrates closely on

BU RTON DA S SET T SOU T H E N D

the smithy tenement and its very near vicinity, with only tiny quantities found on the north side of the road, as the smith had little choice but to stockpile it on his own property. The implications of this for understanding the operation of the smithy are discussed below (Section 3.3). Finally there is the waste from building demolition, the material left behind when all the reusable or salvageable elements have been removed. During the life of the settlement small quantities of discarded roof tile and slate from rebuilding and probably also stone rubble were incorporated into external surfaces, but there were no large dumps exclusively of building debris. The largest quantities of building rubble recorded were unsurprisingly from the fnal demolition of the buildings. Even here the larger stones were removed along with most re-usable roofng material, although a number of the demolition layers did contain noticeable quantities of discarded nails. The only place where large quantities of re-usable tile and slate were abandoned was around the demolished Z2 and Z3 buildings in Dovehouse Close. This emphasises the surprising resilience of the local market in reclaimed building materials through the 15th century despite the massive number of abandoned properties there must have been in the area. It also underlines the determination of Belknap’s agents to extract the maximum proft from the depopulations of 1497; the leaseback to John Heritage of 1497 as mentioned above (Section 1.8) reserved the right to all the abandoned building materials and access to remove them. Possible rubbish heaps Three types of excavated feature were identifed during the excavation as probable sites of rubbish heaps/middens (Figures 3.6 and 3.7): areas of distinctively dark grey ashy soil; clusters of shallow pits/hollows with dark flls where muck heaps might have been dug away cutting slightly into the ground surface; and areas of rubble surfaces containing large amounts of pottery and animal bone, also taken to represent the surviving bases of heaps that had been largely removed. The presence of further quantities of pottery and animal bone in demolition layers and topsoil immediately above the rubble surfaces also argues for the existence of distinct muck heaps overlying the surfaces, fattened out and spread after the abandonment of the properties and demolition of the buildings. Finds recovered from the contexts interpreted as middens are, however, unlikely to have retained stratigraphic or chronological integrity. The pottery from such contexts was re-appraised as part of the fnal re-assessment of the pottery report, and the conclusion drawn that while some of the rubbish in a midden is likely to have remained in the hollow formed by removal of the midden, and that successive muck heaps may have occupied the same space, the

SPAC E A N D BU I L DI NG S

Figure 3.5 Distribution of slag across the excavated site

153

154

Figure 3.6 Distribution of probable midden deposits and rubbish concentrations

BU RTON DA S SET T SOU T H E N D

SPAC E A N D BU I L DI NG S Phase A5 D13 D14 D23 D24 D26 D26 D26 D26 D26 E4 E5 E6 E6 E6 H4 H4 I3 I3 J3 K4

Rubble surface (forecourt S of house) Hollows (N of house) Hollows (W of tenement) Hollows (N of house) Hollows (N of house) Rubble surfaces (forecourt S of house) Rubble surface (N of house) Over rubble surface (N of house) Ash/dark layer (N of house) Rubble surface (N of house) Hollows (N of house) Rubble surface (N of house) Rubble surface (N of house) Rubble surface (forecourt S of house) Hollow (N of house) Ash/dark layer (S of house) Ash/dark layer (S of tenement) Rubble surface (yard NE of house) Ash/dark layer (S of house) Ash/dark layer (S of tenement) Ash/dark layer (W of house)

Pottery (sherds) 354

155 Animal bone frags 199

Cu Iron alloy 1

Pewter

Lead

Bone object

Stone object

Architectural stone

Totals 555

60

20

1

81

58

38

96

4

1

5

68

25

1361

3820

6

32

542

898

1

7

124

306

2

5

119

236

2

6

323

565

1

2

5

10

521

175

505

507

540

1100

2 21

4

2 25

447

56

503

268

116

47

8

236

55

2

127

27

1

1

94 2

3

3

5227 2

2

2

1

1450 442 363

1

1

893 15

3

1

3

6

1

5

20

1

1

2

700 1

1023 1

1

1

1668

388 55 293

1

1

2

159

Figure 3.7 Quantifcation of probable midden deposits and rubbish concentrations hollow would subsequently accumulate post-midden material. Furthermore, the rubble surfaces north and south of the plots on the north side of the road differ from the shallow hollow category of midden deposit in containing more bone and pottery as well as a greater quantity of portable fnds including structural ironwork and lock furniture, offering the possibility that some of the material may indicate demolition rather than domestic rubbish disposal. A proportion of the fnds in the contexts are likely to be comingled with the rubble rather than lying on its surface, and while much of the pottery from these contexts is consistent with a late 15th-century date, the admixture of very much earlier pottery indicates an origin in some process other than rubbish disposal. The following account should therefore be approached with some circumspection. The remains of about 21 possible rubbish heaps were identifed across the site, 9 on rubble surfaces, 6 represented by ashy layers and 6 by clusters of shallow

hollows. It is likely that there were others that were not detected. While the heap bases on rubble surfaces were identifed by large quantities of pottery, animal bone and other fnds, those represented by ashy layers and shallow hollows contained varying, generally lesser quantities of these. The group of hollows to the north of the D13 house, the overlying group from D14, and the D23 and D24 hollows in the yard north of the D2 house all produced totals of fewer than 100 pottery and bone fragments combined, as did the hollows north of the E4 house. This suggests that the muck heaps in these locations would have consisted very largely of organic material. Those on D23 and D24 would have comprised dung and animal bedding from the adjacent byre/stables while those on E4 are likely to have been waste from the adjacent brewhouse. In contrast, the muck heap base on the rubble forecourt south of the A5 house and that on the rubble surface north of the E5 house both contained

156

over 500 pottery and bone fragments suggesting that these were largely domestic waste. For phase D26, the fnal late 15th-century occupation of D2, at least fve possible muck heaps were identifed. The major one lay on the rubble forecourt south of the house. The forecourt surface was recorded as a number of contexts and may have contained multiple heaps. It contained massive amounts of pottery and bone and other fnds, a total of over 5100 pottery and bone fragments. It was quite extensive and the only surface recorded here, and it is likely that it also contained reworked rubble from earlier surfaces and that the number of fnds was therefore boosted by residual material. Nevertheless, this would be the most convenient location for a muck heap for disposal of domestic rubbish from the property, because of its closeness to the house and the ease of access from the street for carts for its removal. The heaping of manure on the streets was a recurrent subject of urban byelaws, for example at Atherstone and Nuneaton, Warwickshire (Watkins 1998, 8); the frequency of intervention by manorial courts suggests that the practice was widespread and attempts at regulation were often ineffective. There were also three other heaps on successive rubble surfaces north-west of the house which contained lesser, but still large, amounts of pottery and bone (combined totals of 1440, 888 and 430), while to the east there was an ashy spread heap which contained rather less material (355). Similarly in the same period the rubble forecourt south of the E6 house produced 1640 sherds and fragments (although as noted above there is a possibility that some of the fnds, which included staples, hinge-pivots, a padlock key and hasp, represent demolition) the rubble surfaces just north of the house produced 1012 sherds and fragments, while a shallow hollow east of the brewhouse, admittedly less extensive, contained virtually none. South of the road on phase H4 there were two possible heaps, an ash/dark layer to the south of the house with only 25 sherds and bone fragments, and another, further south on the tenement with over 500. Similarly in phase I3 the rubble surface and overlying layer in the yard to the north-east of the house produced some pottery and bone (384 items), while the ash/dark layer to the south of the house produced very little (55). Only limited amounts of pottery and bone (154) came from another ash/dark layer to the west of the K4 house. These variations suggest that at one time multiple muck heaps were maintained and there was some deliberate separation of rubbish into that which was mainly organic and suitable for spreading on the felds and that with more non-degradable material which was less so. As has already been observed, manure was a saleable commodity for those without their own landholdings to fertilise

BU RTON DA S SET T SOU T H E N D

and there might have been an incentive to keep it free of objects. Disposal of rubbish Manuring of the felds with domestic waste material, including some pottery, was routinely carried out in this period, leaving an archaeological signature in the topsoil which can be used to analyse this important aspect of arable management. Beyond the evidence for tillage provided by low-density pottery scatters, detailed analysis can differentiate between different intensities of manuring, and the expansion and contraction of the open felds ( Jones 2004). There has been no systematic feldwalking survey over the open felds of Burton Dassett away from the settlement, but elsewhere in Warwickshire scatters of medieval pottery are found over the former open felds as a result of manuring. It is usually found in much smaller quantities than Roman pottery. This difference may refect differing patterns of pottery use in the Roman and medieval periods, with more wooden or leather containers or basketry in the latter. Field survey at Compton Verney, Warwickshire, producing broadly similar quantities of Roman and medieval pottery is exceptional, but there was more Roman material than medieval in the felds remote from the village, but signifcantly more medieval than Roman within 1km of the settlement (Dyer 2000, fgs 4 and 7). A similar pattern was evident at Admington, Warwickshire (Chris Dyer pers. comm.). The distinction between Roman and medieval fnds distributions may be because medieval pottery-rich waste was disposed of in dumps elsewhere, but if so, such dumps have proved remarkably resistant to detection by feldwalking. Some deviation from normal rubbish disposal practice may be suspected in the case of Area H, where exceptionally large quantities of largely 13th-century pottery and animal bone (5368 and 1055 respectively) were recovered from the topsoil, suggesting that domestic waste may have been stockpiled at this early period rather than removed. Pottery assemblages from the site contained signifcant residues of late 12th/early 13th-century material, especially on the south side of the road pre-dating the laying out of the house plots (online archive, fgure 8.1.46). On the basis of this, a case could be made for occupation south of the road as early as the 12th century, although, in the excavator’s view, this is less likely, at least within the excavated areas. While pottery-rich waste may have been dumped here prior to actual occupation, it seems unlikely that material would have been carried any great distance for disposal, although Beresford and Hurst (1971, 116) suggest that rubbish with large amounts of pottery and bone might be disposed of on a dump at the settlement edge rather than on the felds. This hypothesis may not actually be supported by archaeological evidence, with little conclusive

SPAC E A N D BU I L DI NG S

evidence for such intact rubbish dumps. Additionally, it is hard to identify where such dumping would have taken place once the plots were laid out; there were no particular concentrations of medieval pottery or animal bone from feldwalking west of the excavated areas, and the low quantity of material from the small areas of ridge and furrow feldwalked south of the excavations (Z6) was consistent with manuring. The proportion of late 12th/early 13th-century pottery amongst the assemblage from Area B as a whole (as well as from Areas A, D1, D2 and E) also seems to indicate that some early rubbish dumping took place here before the tenements and the road were laid out, but here too the quantity of later material does not suggest that this area continued to be a dump for material that was unsuitable for spreading onto the felds, and from the mid-13th century this material must have been disposed of elsewhere. Chance losses In addition to the categories of deliberately discarded bulk material, many excavated fnds will represent chance losses of individual items, such as dress fttings, jewellery, or horse fttings, particularly thrown horseshoes, but also harness fttings and spurs and spur buckles, which might be dropped or become detached without the owner noticing. Some of these might lie where they had fallen, and it is worth noting that a large proportion of the metalwork from around the road on Area B consisted of horse equipment (Figure 3.8), but many items would have been swept up and disposed of with the general rubbish. Small tools and items of domestic equipment such as hones, might also have been accidentally mislaid, but larger ones such as querns or mortars would be less easily lost, and broken fragments would require disposal. These cannot usually be recycled except occasionally as sharpening stones, or as general rubble (quern fragments have been noted incorporated into the fabric of over 60% of medieval Norfolk churches, where distinctive imported Rhineland lava querns are readily identifable; Ashley et al. 2009). Broken metal equipment and utensils could be recycled, but small fragments might be discarded or incorporated into the general rubbish. The spatial distributions of these items (Figures 3.8–3.11) are affected by a number of factors which complicate their interpretation. First, there is the differing intensity of excavation north and south of the road which means that there are more fnds per category from north of the road. Nevertheless, as already observed, more objects came generally from later periods, possibly because of a build-up of residual material, and there is also a marked concentration of fnds on the tenements that were occupied latest. This may refect better access to goods, but perhaps also a higher level of scavenging for recyclable materials by the surviving inhabitants,

157

or their children, leading to fewer fnds from the abandoned properties to the south and west. Comparing the proportions of different fnds groups north and south of the road (Figure 3.12) the effect of more intensive scavenging for metal south of the road can perhaps be seen. Knives, used by everyone, ought to be ubiquitous, but they are virtually absent from south of the road. Similarly, the proportion of horseshoes from these areas (15%) seems inordinately low. Nevertheless, there would have been a limit to the degree of scavenging that was practical; the inhabitants of Southend had no metal detectors. Judging by the concentrations of objects around the houses where the muck heaps for domestic waste were sited, the dominant factor in the distribution of objects lost across each tenement is the pattern of general waste disposal. The majority of objects tended to be found where they were dumped rather than where they were used. This means that caution is necessary if these distributions are used to indicate where activities were carried out. Only objects from foor layers or features within buildings provide perhaps more reliable evidence. The distribution of the jet and bone beads, presumably from the same broken rosary, illustrates the trajectory of lost objects among rubbish outwards from the D2 house, for example (Figure 3.13). The earliest beads, two of jet and one of bone came from the eastern end of the D24 house where the breakage presumably took place. Two jet beads came from later D25 layers in the eastern room, one from the hall, two more from the yard north of the house, and one more from the topsoil north of the house. One bone bead came from the topsoil west of the house, with a possible further bead to the west of the house. Objects connected with sewing, thimbles and pins, and textile working, mainly spindle-whorls, were mainly found in the houses and adjacent rubbish heaps, as one would expect with these essentially domestic activities (Figure 3.9). North of the road other craft tools seemed to concentrate north of the houses, perhaps refecting their use away from the houses or in outbuildings. The distribution of tools does not seem to be a good guide as to where activities were carried out. Only one of the blacksmithing tools found actually came from the smithy tenement; the others came from across the site, from 15th-century layers or topsoil. The wide range of tools found on the D2 and E tenements also demonstrates that they do not necessarily indicate the presence of specialist craftsmen. Woodworking and masonry tools might easily belong to householders as part of a general toolkit, for building maintenance rather than construction. Other tools could be multipurpose: leather-working awls might be used for woodworking and small agricultural tools would be used in gardening. Non-ferrous metalworking tools and waste are perhaps more specialist, but the

158

Figure 3.8 Distribution of horse equipment

BU RTON DA S SET T SOU T H E N D

SPAC E A N D BU I L DI NG S

Figure 3.9 Distribution of tools and working waste

159

160

small quantities found seem more likely to refect occasional visits by itinerant workers than resident bronze or pewter smiths. Unsurprisingly, fnds of domestic equipment (Figure 3.10) were mostly found in the houses and around on the adjacent rubbish heaps. South of the road the distribution is more immediately around buildings, while to the north it is in and around them. Personal items such as clothes fttings and jewellery (Figure 3.11) were also concentrated in and around the buildings, presumably where their former owners spent most of their time. Many of them were again swept up with the general rubbish and incorporated into the rubbish heaps. The marked concentration of small circular iron buckles used on shoes, belts and hose in Areas D2 and E refects their appearance in the later 15th century. Horse equipment (Figure 3.8), mainly horseshoes, was concentrated in the public areas of the site over which horses would have passed, particularly over the house forecourts and along the Area B roads. Spurs and spur buckles came exclusively from Areas D2 and E. The distribution of objects from the medieval tenements at West Cotton, Northamptonshire, exhibited a number of similarities to that at Dassett Southend (Hylton 2010, 341–342), especially the marked variation between tenements. There was also a suggestion that ironwork was generally underrepresented because of recycling. In Areas A, B and E more objects came from the yards (66%) than the houses (33%), but in Area C 25% came from the yards and 75% from the house (although here the yards were not fully excavated). Dress fttings and jewellery were the only category of fnd that came predominantly from the houses, suggesting that they were more frequently lost indoors. Yards with metalled surfaces produced more than unsurfaced areas. It was not possible to defne concentrations that may have indicated midden heaps, but there was a higher density of fnds within 2–3m of the buildings where the soil layers were deeper. 3.2 THE BUILDINGS AT SOUTHEND

Houses The excavations recorded about 25 complete and partial house plans, including rebuildings and extensions, on the ten excavated tenements, dating from the mid-13th to the late 15th century (Figures 3.14 and 3.15). The relatively well preserved stratigraphy meant that more internal features such as hearths survived than on many other sites. Structure Wall thickness may provide a broad guide to superstructure (Beresford and Hurst 1971, 94). Walls over 0.70m will support stone walls to eaves

BU RTON DA S SET T SOU T H E N D

height; those less than 0.55m wide are more likely to support timber superstructures, whilst those in the intermediate range (0.55–0.70m) may support stone walls. It is recorded that average wall thicknesses of the stone houses in the local area is ‘remarkably consistent’ at 2ft 6in (0.75m) or over in medieval structures, but reduced from the end of the 16th century to 1ft 10in (0.56m) (Wood-Jones 1963, 240). On the two surviving structures at Southend the walls of the medieval chapel are c0.7m wide and those of the 17th-century Priest’s House c0.65m wide. On this basis it is likely that the Dassett Southend houses were all constructed with either full-height stone walls or timber-frames on regular or fairly regular stone footings (Figure 3.16), although some minor subsidiary features like porches (E3–5, J2–5) were supported on earth-fast posts. On seven of the ten tenements (A, D1, D2, E, I, J and K) the main parts of the houses are interpreted as having been built with stone walls to the eaves (see Section 1.11 above), and on two (F and H) as having been timber-framed, probably with cruck structures on stone foundations, but possibly as a box-frame. The structures on the tenth (L) may have been a mixture of stone walls and timber-framing; the northern wall of L is less obviously straight than the majority of the Southend walls and appears to be less suited as the support for a sill-beam, but had been subject to signifcant stone-robbing and root disturbance, so it is diffcult to be certain of the form the superstructure took. The prevalence of stone at Southend is somewhat at variance with the current orthodoxy on Midlands medieval peasant housing which sees the peasant house as essentially a timber-framed cruck structure (Dyer 1982; Alcock and Miles 2013). This focus on crucks does seem to underemphasise the local building traditions of stone-built houses in the limestone areas of south Warwickshire and north Oxfordshire (and Gloucestershire and Northamptonshire) which were based on an abundance of good, locally available building stone and, in the case of the Burton Dassett area, a virtual absence of woodland (Hilton 1983, 95). Stonebuilt houses are characteristic of the post-medieval vernacular building tradition of the Banbury area (Wood-Jones 1963) which was based on the use for building of the ferruginous limestone of the Marlstone Rock Formation, which outcrops on the Burton Dassett Hills within 1km of the site. In fact, at Dassett Southend, as elsewhere (Wrathmell 1989, 3), the evidence of the excavations is that the peasant houses of the 13th to late 15th centuries were the direct antecedents of the later, in this case stone-built, vernacular tradition. Other instances in the local area of excavated medieval houses interpretable as originally with full-height stone walls can be cited, including one fronting Northend Road, Fenny Compton, with 0.80m wide foundations, occupied from the later

SPAC E A N D BU I L DI NG S

Figure 3.10 Distribution of domestic objects

161

162

Figure 3.11 Distribution of personal items

BU RTON DA S SET T SOU T H E N D

SPAC E A N D BU I L DI NG S

Copper-alloy belt fttings Iron knives Iron horseshoes Iron tools Structural metalwork: cramps, ties, staples, wedges, studs, clench bolts and roves Structural metalwork: iron hinges, straps, pivots Metal vessels Stone mortars Nails Stone hones Stone querns All items All items except nails

163 North of the road Number % of total 110 81 71 95 56 85 27 75 38 70

South of the road Number % of total 26 19 4 5 10 15 9 25 16 30

Total

78 9 4 1547 35 10 1985 438

15 4 2 903 21 10 1020 117

93 13 6 2450 56 20 3005 555

84 69 66 63 62 50 66 79

16 31 33 37 38 50 34 21

136 75 66 36 54

Figure 3.12 Comparative quantities of selected categories of fnds from areas north and south of road

13th to the early 15th century (Eyre-Morgan 1994; Jones and Palmer 1998a). On the Lias limestone there is House B1 at Goldicote, with c1m wide wall foundations and a stone-mullioned window, occupied from the 13th to the early 15th century (Thompson and Palmer 2012, 81–83, 120–121), and probably at least one 13th/14th-century house at Compton Verney (Hart 1991, 8–10). West Cotton, Northamptonshire, is another extensively excavated site in the limestone area where most of the houses had stone walls to their eaves (Chapman 2010, 221–222). Nevertheless, crucks and stone walls were not mutually exclusive. Wood-Jones (1963, 14–23) identifes a number of early buildings with a localised form of roof structure in which raised crucks were used on stone walls, and it seems likely that these were employed for the roof structures of the Dassett Southend houses. In fact, the closest to Burton Dassett of the houses recorded by Alcock and Miles, the Old Post Offce, Oxhill (Alcock and Miles 2013, OXH-A) has this arrangement. Doorways and windows The evidence of the architectural stonework from the site suggests that the stone houses had some well fnished plain detailing. The door jamb from the rubble of the D25 house, believed to have been incorporated in the structure of the previous phase D24, inscribed with the tenant family’s name ‘Gormand’ (Figure 3.17) was only one of 11 door or window jambs with basic chamfered decoration, coming from phases D22, D25, D27, E4, I4, K2 and K5 (late 13th to late 15th centuries). There were also chamfered blocks from window sills or plinths from phases D26, D27, E6, I3, I4, I5 and K2 (late 15th century), and a stone window mullion from phase A6 (also late 15th century). These details indicate that the houses were likely to have been built by professional masons, albeit to the order of their

Cat. no. J24 J25 B11 J22 J26 J23 J27 J29 B12 J28 B10

Context type Wall matrix Floor Wall matrix Floor Floor Floor Levelling layer Fill of gully Fill of gully Topsoil Topsoil

Context location East end, D24 house

East End, D25 house Hall, D25 house Yard to north of D25 house West of D25 house North of D2 house West of D2 house

Figure 3.13 Jet ( J) and bone (B) rosary beads: context details

tenant-occupiers. This is consistent with a range of documentary evidence which shows peasants hiring specialists for a range of building tasks, not always those requiring high levels of skill, although it is probable that the peasants themselves would have contributed labour for tasks such as site clearance and procurement/carriage of materials (Dyer 2013, 106–108). Similar embellishment is occasionally found elsewhere. At Goldicote there was a single squared door jamb fragment and a moulded coping stone, in addition to a chamfered window mullion (Thompson and Palmer 2012, 120–121, fg. 36, nos 81–83). At Upton the excavator remarked on the quality of the house masonry, but the only decorative fragment was a slightly shouldered, chamfered door lintel possibly from the entrance to house AE, although the buildings are not supposed to have had full-height stone walls (Rahtz 1969, 103, fg. 10, ST78). The medieval peasant tenements of West Cotton were well constructed but without any attempt at general architectural embellishment (Chapman 2010, 217– 223); the only decorative stonework found consisted of a mysterious carved fgure and two fragments of a feur-de-lys fnial (Chapman 2010, fgs 11.11–12,

164

Figure 3.14 Comparative plans of all the houses excavated, by period

BU RTON DA S SET T SOU T H E N D

SPAC E A N D BU I L DI NG S

no. 56; fg. 11.16, nos 29–30). The West Cotton buildings were quite standardised and possibly built by the landlord as a commercial operation without the fnishing touches which one may infer were specifed by the tenant-builders at Dassett Southend. The unstratifed fragment of early 14th-century window tracery from Area D (Figure 3.18, no. 13) could have come in rubble from the parish church or manor house, but it could have also come from an elaborate window in one of the excavated houses. Three of the medieval lesser hall-houses recorded in the local area by Wood-Jones had elaborate traceried hall windows, including Priory Farm and Grange Farm, Balscott, and Leadenporch House, Deddington, Oxfordshire (Wood-Jones 1963, 42–52, fgs 8–10, pl. 2c, 31–36, fg. 6, pl. 2b). This represents a high proportion of the identifed medieval houses, but in the days before dendrochronology it was probably these features that enabled them to be dated as medieval, and such windows may not have been that common. One of the D2 houses, however, may well have been provided with such a window; indeed D2 stands out as being distinctive in terms of overall prosperity. Examples of surviving decorated medieval windows are not common in standing timberframed houses in the region (Alcock and Miles 2013, 99–101), but, of course, later alterations mean that little evidence survives for the position and character of doorways and windows in these houses. The few decorative windows recorded are rare survivals of features that may once have been widespread. Most of the doors at Dassett Southend will have been hung on the heavier of the many iron hinge fttings found, usually with hinge pivots driven into timber jambs (Figure 3.18, nos 170–200) and a few may have been set in lead-lined sockets in masonry (Figure 3.18, nos L12–L13). No examples of the vertical recesses for timber door jambs common at West Cotton (Chapman 2010, 223, fg. 7.69) were noted at Burton Dassett, suggesting that the door frames there were simply ftted fush to the doorways. Three stones with vertical sockets cut into the sides may possibly have been from door cases (Figure 3.18, nos 37 and 39). There were also three pivot stones found (nos 23–25), which represent an alternative method of hanging doors and gates found commonly on village sites from Upton (Hilton and Rahtz 1966, 115, ST 22) to Wharram Percy (Andrews and Milne 1979, 128, fg. 69, nos 21–22; Wrathmell 1989, 56, fgs 20, 37 no. 43). At West Cotton 5 of the 14 were found in situ, on internal doors and external doors of outbuildings (Chapman 2010, 223, fg. 7.71). In fact, they were more common than iron hinge fttings: the West Cotton tenements produced only six hinge pivots and fve hinge fragments in total (Chapman 2010, 361). One of the Burton Dassett pivot stones (no. 24) was found in situ in the centre of the western room of the D25 and D26 house where it must have supported some kind of swivelling apparatus other

165

than a door. Elsewhere other uses for these stones have been suggested, including bearings for lathes or potters’ wheels (Bryant and Steane 1971, 76, fg. 23a and c; Hilton and Rahtz 1966, 115). Most of the windows in the houses would have had wooden frames and been closed with wooden shutters. Many of the more delicate iron hinges, straps and hinge pivots will have supported these shutters, although some will have held doors on furniture. One chamfered stone sill fragment retained a socket, probably for a timber frame, but possibly for an iron grille. Two mid-/late 15th-century contexts in houses A5 and D5 produced fragments of medieval window glass. There was also a fragment of a medieval-type came (lead setting for a glass pane) from the D2 topsoil. A second fragment of came from Area E topsoil will date to post-1600. Medieval village houses are not generally supposed to have had glazed windows (Astill 1988, 57; Dyer 1989a, 165; Beresford 2009, 125), and the evidence from Southend is at best modest. No evidence for medieval glazing was recorded, for example, from the extensive excavations of Midlands peasant houses at Tattenhoe or Westbury, Buckinghamshire (Ivens et al. 1995) or at West Cotton (Chapman 2010, 406). In addition to the fnds at Burton Dassett, fragments of glass and lead cames have, however, been found in medieval contexts on other village sites, such as Wharram Percy (from the very large 15th-century Area 6 house; Andrews and Milne 1979, 115, nos 18– 19, 130; Dyer 2012d, 330); Great Linford (window glass from Crofts A, F and L; Mynard and Zeepvat 1992, 217–220) and Caldecote, Hertfordshire (from Crofts B and D; Beresford 2009, 125). At Goldicote, a stone window mullion with an inserted glazing groove came from House 1 which was abandoned in the early 15th century (Thompson and Palmer 2012, 121, fg. 36, no. 81). All in all there is enough evidence to suggest that occasional glazed windows were to be found by the 15th century, presumably as prestige features, in the halls or parlours of some peasant houses, including one or more at Dassett Southend. Porches The houses on two tenements (E and J) had small timber porches against their rear entrances. The earlier ones (J2, J4, E3, E5) were supported on earthfast posts, while the latest (E6) was set on a stone-wall footing. These presumably allowed wet clothing or footwear to be removed before entry to the house, although the E6 porch offered access both to the hall and to the stair to the upper room. Roof coverings Thatch will have been the main roofng material at Burton Dassett, in the Middle Ages as later (Wood-

BU RTON DA S SET T SOU T H E N D

166 Phase A2

Type S

Bays 3

A3–4

S/TF

3+1

A5

S/TF

4+1

D13

S

1

D22–3

S

3

D24

S

3/4?

D25

S

4?*

D26

S

4?*

E2

S

1

E3–5

S

3+1

E6

S/TF

4+1*

H2

TF

2/3?

H3–5

TF

I2–3

S

2½/ 3½?* 3

I4

S

3?

J2

S

3

J4

S

3

K2

S

2?

K3

S

4

K4–5

S

4

L1

S/TF

3

m ft m ft m ft m ft m ft m ft m ft m ft m ft m ft m ft m ft m ft m ft m ft m ft m ft m ft m ft m ft m ft

H 12×4.75 39.4×15.6 12×4.75 39.4×15.6 12×4.75 39.4×15.6 5.75×c 3.5 18.9×c 11.5 3.5×4.25 11.5×13.9 7.0×4.25 23×13.9 4.4×4.5 14.4×14.8 4.4×4.5 14.4×14.8 6.0×3.6 19.7×11.8 6.0×4.5 19.7×14.8 6.6×4.5 x 14.8 8.5×4.0 27.9×13.1 8.5×4.0 27.9×13.1 12.5×4.25 41.1×13.9 3.75×4.5 12.3×14.8 6.75×3.75 22.1×12.3 4.5×3.75 14.8x 12.3 9×4.6 29.5×15.1 9×5.6 29.5×18.4 6.5×5.6 21.3×18.4 8.8×3.7 28.9×12.1

R1 –

R2 –

R3 –

–

–

–

5.75×4.5 18.9×14.8 –

–

–

–

–

3×4.25 9.8×13.9 5.75×4.25 18.9×13.9 4.5×4.5 14.8×14.8 4.5×4.5 14.8×14.8 –

3×4.25 9.8×13.9 2.5×4.25 8.2×13.9 4.5×4.5 14.8×14.8 4.5×4.5 14.8×14.8 –

–

3.75×4.0 12.3×13.1 4.25×4.5 13.9×14.8 –

2.5×3.5 8.2×11.5 3.5×2.5 11.5×8.2 –

–

–

–

–

–

–

–

5.75×4.25 18.9×13.9 3.25×3.75 10.7×12.3 3.0×3.75 9.8×12.3 –

2.25×1.6 7.4×5.2 –

3.0×2.5 9.8×8.2 –

–

–

–

–

3.3×5.6 10.8×18.4 3.3×5.6 10.8×18.4 4.2×3.8 13.8×12.5

4.25×2.25 13.9×7.4 –

4.25×2.6 13.9×8.5 –

–

–

– – – –

– –

Type: S stone-walled, TF timber-framed, S/TF stone-built with timber-framed extensions; Bays: +1 indicates cross–wing, stair/passage ½ bay, * indicates upper room; H Hall, R1–R3 Rooms/chambers, K Kitchen, W working area, P Passage/Porch, S Stair, UR Upper room. # K4–5 total excludes byre 33.9 sq. m

Figure 3.15 Dassett Southend houses: dimensions of rooms and foor areas

Jones 1963, 244–248). The excavations produced substantial quantities of ceramic and slate roof tiles also, and it is clear that these were used for roofng some of the buildings of the medieval settlement. Most of the 10,629 fragments of roof tile collected during feldwork came from feldwalking, but 3051 fragments (28.7%) were recovered from the excavations and only 12 sherds were from contexts datable to before the 15th century (online archive, Section 8.12). Most of the assemblage (90%) was in a fabric (A) of possible local source. Three fabrics had shelly matrices, generally uncommon in Warwickshire, and may have been from an unknown source on the local Jurassic limestone or from further to the south-east; two of these fabrics were only

present amongst the feldwalked material and may well therefore be later than the excavated buildings. Of those tiles where form could be determined, 1305 fragments were from fat tiles and 110 from curved ridge tiles, all plain and lacking crests or fnials. Flat tiles most often had nibs from hanging, with both square and round peg-holed examples occurring in later 15th-century contexts; only fve tiles had both nibs and peg holes. Peg-tiled roofs tend to be steeply pitched at between 42o and 50o (SPAB 2009, 1). Mortar to fx tiles to the roof was also rare (62 examples), insuffcient to infer anything about how the tiles were hung. Three ridge tiles had mortar on the ends, suggesting that they had been affxed end to end. Glazing was rare, even on the ridge tiles, which is unusual.

SPAC E A N D BU I L DI NG S

167

K –

W –

P –

S –

UR –

2.5×3.75 8.2×12.3 2.5×3.75 8.2×12.3 –

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

1.75×4.5 5.7×14.8 1.75×4.5 5.7×14.8 –

1.5×3.5 4.9×11.5 2.1×3.5 6.9×11.5 –

4.5×3.75 14.8×12.3 4.5×3.75 14.8×12.3 –

–

–

–

–

1.6×1.7 5.2×5.6 1.5×1.5 4.9×4.9 –

–

–

5.0×4.25 16.4×13.9 –

3.6×1.4 11.8×4.6 –

4.25×4.5 13.9×14.8 –

–

–

–

–

1.6×1.3 5.2×4.3 –

1.7×1.2 5.6×3.9 –

?3.5×4.0 ?11.5×13.1 –

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

2.5×3.75 8.2×12.3 –

–

–

–

–

–

–

–

–

–

–

–

–

–

2.5×1.75 8.2×5.7 –

–

–

Floor area 57 sq. m 614 sq. ft 66.4 sq. m 714 sq. ft 92.35 sq. m 993 sq. ft 20.1 sq. m 217 sq. ft 40.4 sq. m 434 sq. ft 64.8 sq. m 696 sq. ft 85.9 sq. m 925 sq. ft 89.1 sq. m 959 sq. ft 21.6 sq. m 233 sq. ft 74.7 sq. m 802 sq. ft 107.6 sq. m 1159 sq. ft 34 sq. m 366 sq. ft 52.1 sq. m 561 sq. ft 53.1 sq. m 572 sq. ft 52.4 sq. m 564 sq. ft 37.5 sq. m 404 sq. ft 37.5 sq. m 404 sq. ft 41.4 sq. m 446 sq. ft 89.5 sq. m 963 sq. ft 59.7 sq. m# 642 sq. ft 48.5 sq. m 522 sq. ft

Figure 3.15 (continued) Dassett Southend houses: dimensions of rooms and foor areas

Where dimensions could be calculated, all but eight of the fat tiles conformed to the 1477 Tiles Statute, which confrmed a standard size of 10½ by 6¼ inches (267 by 159mm) with a minimum thickness of 5/8 inch (16mm), which remained the standard for several centuries (Clifton-Taylor 1972, 269). The ceramic roof tile was concentrated in 15thcentury contexts to the north of the road. It is likely that the D25 and D26 houses had tiled roofs, along with at least parts of the roofs of the later houses on A, E and F. On Area A, it would be the A5 western extension, as the main roof was of slate. South of the road there was very little tile from the Area Z1 feldwalking or from the excavated tenements which lay within this area. This suggests its use only

became common in the mid-15th century after these properties were abandoned. There were also large concentrations in feldwalking Areas Z2 and Z3 to the west, indicating that Z2 contained at least two buildings with tiled roofs that had decayed in situ, and Area Z3 at least one more, perhaps an indication that occupation west of the excavated area south of the road continued for longer (Figure 1.21). In the later period (mid–late 15th century) therefore, tile became important as a roofng material occurring on perhaps 15–25% of the houses. This increase mirrors a general increase in both production centres and use (Lewis 1987, 10). Of the 4569 fragments of roofng slate recovered, about a third (34.3%) were from the excavated tenements (online archive, Section 8.11). The main

168

BU RTON DA S SET T SOU T H E N D

Figure 3.16 Model of Area A house (Phase A5, mid-15th century) (Warwickshire Museum)

material used was Stockingford Shale from north Warwickshire. This could have been quarried at a number of places along the Nuneaton Ridge, including the quarries at Purley or Griff. Its presence at Burton Dassett is further evidence of the connections between the two areas, some 40km apart, reinforced by seigneurial links. Burton Dassett seems to be some way south of the normal distribution of Stockingford Shale: it is common in Coventry (Wright and Stewart 1982, 92, 99–100, fg. 54, 1–7) and present at Kenilworth Castle (Rahtz 1966, 66, n. 1), but absent from Warwick. Only three of the limestone slates used on local post-medieval buildings were recovered from the excavations, so few that they are likely to have arrived in rubble brought in from elsewhere. Given that the ceramic tiles and slate roofs would mostly have been robbed out after abandonment, interpreting the distributions is not straightforward. While massive concentrations of material such as those from Z2 and Z3 indicate the presence of a roof that had decayed in situ, lesser concentrations standing out less prominently from a general background level may indicate debris from construction or demolition of a roof mainly robbed out for re-use. Tile and slate may also have been used in small quantities to line smoke holes or louvres on thatched roofs.

One lesser concentration of slate in Area A suggests that the main part of the house roof was covered with this material. The earliest excavated slate fragments came from phase A2 which saw the original construction of the house in the late 13th century. Although there were only a handful of fragments, the wide walls of the original building would have been able to support the weight and it is possible that it was roofed in slate from the beginning. South of the road there was a massive concentration in Area Z2 of the 1987 feldwalking, suggesting the presence of a slate-roofed building or buildings where the roof had decayed in situ. Other slighter concentrations suggest that one of the buildings in Area L and possibly another in Area I may also have had slate roofs. At a very rough estimate perhaps 10–15% of the houses or 5–7.5% of all buildings had slate roofs, making slate a minor but signifcant component of the roofng on the site. The choice of roof ing material also has implications for the timber supporting it. Slate of 4–6mm thickness is of broadly similar weight to thatch (some of the Southend slates were up to 15mm thick and thus heavier), but ceramic tiles are around twice as heavy as thatch (SVBRG nd) and would have required a robust frame, constructed by a professional carpenter.

SPAC E A N D BU I L DI NG S

Figure 3.17 Architectural stonework: door jamb inscribed with the name Gormand (mid-15th century, from metalling around north door of house D25)

169

170

BU RTON DA S SET T SOU T H E N D

Figure 3.18 Architectural stonework: fragment of early 14th-century window tracery (13); pivot stones (23–25); socket stones, possibly from door cases (37, 39). Structural metalwork: lead socket linings (12, 13); iron hinge pivots (170, 180– 182, 185, 195, 196, 199, 200). Stone: 13: D, unstratifed. 23: I5, mid-15th century demolition. 24: D26, late 15th-century. 25: L3, topsoil. 37: Z1, feldwalking. Lead: L12: A6, late 15th-century demolition. L13: D27, late 15th-century demolition. Iron: 170: H6, early 15th-century demolition. 180–181: E6, late 15th century. 182: D26, late 15th century. 185–196: D27, late 15th-century demolition. 199: A8, topsoil. 200: D17, topsoil

SPAC E A N D BU I L DI NG S

Floors The foors within the excavated houses varied little: most were simply of earth (clay loam), some with an admixture of small rubble. A few foors consisted of small rubble metalling, with patches of larger rubble metalling common around the inside of external doorways, presumably to prevent wet feet from turning the earth foors to mud. There were no tiled or fagstone foors in any of the houses; the entire excavation produced only a single foor tile fragment, with a dark greenish-black glaze, from early to mid-15th century demolition material in House I5, and the local Marlstone Rock Formation limestone does not fracture suitably to make fagstones. Hearths A total of 45 hearths were excavated within the houses; 15 of them were simple patches of burnt soil and ash, with another ten being of ash with more or less rubble in their make-up or surround. The greatest number (19) were in the form of stone platforms made of ironstone slabs set fush with or slightly proud of the foor. These were roughly made of rubble, except for one in the hall of E3 which was neatly fashioned out of seven stones into a rectangle with a roll moulding (online archive, Section 8.10). There was also one elaborate hearth consisting of a bed of pitched roof tiles augmented by a block of stone slabs with a diagonal stone freback in the south-west corner of the D26 hall in one of the latest buildings (Figure 2.76). Although the construction of hearths did not change over the life of the settlement, there did seem to be a chronological variation in their position within rooms. Up to the late 14th century all the hearths were set centrally in the room, with the exception of one in the singleroomed house A2, which was set against a wall. In the 15th century, although there were still central hearths, the majority were placed against walls (18 out of 29). This indicates that central louvres for the removal of smoke were gradually replaced by smoke hoods and chimneys. This, in turn, would have permitted parts of the living space to be lofted over (Wrathmell 2012, 341), although there is no direct evidence of this having taken place at Southend. The shift in position of the hearth is paralleled elsewhere; at Wharram and West Whelpington central hearths were replaced by hearths backing on to a passage partition (Dyer 2012d, 329; Jarrett and Wrathmell 1977, 117, respectively). No traces of clay superstructures or furnaces survived on any of the hearths (cf. Brears 2012, 353). Four earlier hearths, three late 13th/14th-century (in Houses A2, D22–23 and E3), and one early 15thcentury (House E6), appeared to be associated with groups of stakeholes or small post-holes; these perhaps supported some sort of screen or apparatus connected

171

with cooking. A hearth on the northern side of the western room of House I4 contained a socketed stone which may have served a similar purpose. In House D24, a hearth against the south wall of the hall, close to the external door, had a stone kerb on one side which may have supported a projecting screen separating the hearth from the door. Other internal structures within the houses Throughout the houses there were occasional post settings and hollows that might have accommodated internal fxtures of various kinds. Where these were located away from the walls, they tended to concentrate in the rooms off the halls, presumably so that the latter could be kept relatively unencumbered (Figure 3.14, D22–23, E3, D26 and E6). In the hall of the K3 house, along the north side of the western bay and the south side of eastern bay, there were two narrow upstanding stone foundations, measuring 4.3m by 0.4m and 3.4m by 0.6m respectively, apparently sections of wall retained from the previous phase. These may have been the bases of benches along the walls. There was a similar foundation, c3m long by 0.35 wide, set against the western wall of the smithy in phase J3. This may have been another bench, although this one could have been some kind of structural strengthening of the wall. Similar stone foundations interpreted as benches, or possibly cupboard bases, were recorded around the kitchen areas of West Cotton (Chapman 2010, 232–233, fgs 7.80–7.84). A number of other, less substantial, rectangular rubble foundations set against the walls of Dassett Southend houses may have supported other fxed furniture. Against the south wall of the eastern bay of the A2 house, there was a sub-rectangular rubble patch measuring c1.1m by 0.6m; against the west wall of the western room of the K4 house there was one c1.8m by 0.8m; and against the south wall of the western room of the I4 house, there was another measuring 1.65m by 1.0m. Against the north wall of the eastern room of the J4 smithy, there was another foundation measuring 0.7m by 0.6m, but this may have supported part of the bellows apparatus for the forge, possibly in association with the two stone blocks 1m apart set into the foor nearby. In the A5 house, a small, oval stone-lined pit (110) c1.0m by 0.63m across was placed centrally against the eastern wall. The purpose of this is also uncertain. Although some of the stones of the lining were burnt, its fll contained no ash and it does not seem to have been a hearth or oven. Internal post-holes and post-bases were few, and where they were present (e.g. D22–23, eastern room, and E3, eastern room) no observable pattern was evident. It is just possible that such features may have supported bed-frames; references in probate inventories to bedding are more frequent than references to beds, and it has been suggested that

172

where this is the case it may indicate that beds were part of the building structure rather than moveable goods (Brears 2012, 350). Field (1965) notes the absence of references to beds amongst tenants’ principalia in Worcester, suggesting that tenants were responsible for providing their own beds. ‘Processing’ troughs and tank In addition to the possible fxed furniture, a number of structures within the houses were probably related to work or craft processing activities. In the I2 house, running northwards from the south wall of the western bay was a trough cut into the foor, c1.9m by 0.55m by 0.22m deep, lined and foored with rubble blocks (Figure 2.9). The surrounding area was not fully excavated so it is not possible to say whether it was accompanied by other features (although a possible drain and a hearth in this area were recorded as belonging to a later phase). In the western bay of the hall in the D24 house also running northwards from the south wall was another stone-lined trough 2.1m by 0.70m by 0.17m deep, although the northern end of the lining appeared truncated, and rubble in its fll may have been from displaced upper courses of the lining (Figure 2.62). To the north, against the north wall was a hearth which may have been associated. The flls of the trough contained nine pottery sherds (including jug fragments), a spindle-whorl, a copper-alloy brooch and a small perforated disc that might have been a rose or flter from an extremely small watering can. The eastern room of the E5 house seems to have been laid out as a work or ‘processing’ room. Against its north wall there was a small stone-lined tank, 1.05m by 0.40m by c0.40m deep (Figure 2.64). About 0.6m to its south was a stone-lined drain, 4.0m long by 0.36 wide, running at right-angles, and a second drain running south-westwards from the western end of the frst into the south-west corner of the room. To the west of the tank there were hearths against the north and south walls. These features produced only small quantities of pottery, animal bone and some charcoal. Soil samples from these contexts were undistinctive, producing only a few charred plant remains. West Cotton provides a number of parallels for these structures. The Tenement B5/1 ‘processing room’ contained a stone-lined trough, 3.0m by 0.6m by 0.3m deep, associated with a hearth and soakaway (Chapman 2010, 184–186, 237–239, fgs 7.32, 7.87–7.89). The facings of the trough were discoloured by a grey encrustation. Another stonelined trough/gully, 2.30m by 0.6m by 0.23m deep, in the Tenement E13/3 house (Chapman 2010, 197, fgs 7.41–7.42, 7.90) also had a hearth adjacent and was stained grey by chemical action. The trough was later twice reduced in size eventually, the fnal arrangement being stained red/purple. Another stone-lined trough, 1.60m by 0.35m wide by 0.30m

BU RTON DA S SET T SOU T H E N D

deep, was unstained by chemicals in the Tenement D12 kitchen (Chapman 2010, 176, 237, fg. 7.23). The more elaborate arrangement in the Burton Dassett E5 house was similar to that in the West Cotton S17/2 processing room belonging to the late 12th/13th-century manor house. This contained a stone-lined pit with an adjacent larger stone-lined trough set at right-angles to it (Chapman 2010, 106– 107, fgs 5.26–5.28). Both were stained grey or bluegrey by chemical action. Looking just at the Burton Dassett tank there were other, similar stone-lined ‘bin’ features at West Cotton: one in the kitchen of the domestic range of Tenement E13/2, and four others in the Tenement D kitchen/cottage D11. It is suggested that these may have been below ground ‘cool-boxes’ for storage of perishables (Chapman 2010, 178–179, 197, 232, fgs 7.25–7.27, 7.41, 7.83–7.84). The Burton Dassett tank and drains were, however, of one build and almost certainly associated. Another site with an installation similar to the E5 tank and drains was Upton, where the end room of house AE contained a tank or channel, running northwards from the centre of the south wall, with a drainage channel set at right-angles right across the centre of the room and out through the east wall (Rahtz 1969, 87, fgs 5–6). As at Burton Dassett the tank and drain were separate. The function of these structures is uncertain. They seem to have involved the holding and draining away of liquids, possibly heated liquids. The activity, presumably not too noxious, would have been relatively small scale, particularly if the capacity of the tanks and troughs was reduced by the insertion of timber or clay linings to make them liquid-tight. At West Cotton the process caused chemical staining of the structures, but this was not evident at Burton Dassett or Upton. The Upton excavation report does not attempt to identify the process carried out, but elsewhere the excavator suggested that it may have been fulling (Beresford and Hurst 1971, 140) or fulling or tanning (Hilton and Rahtz 1967). Tanning or tawing seems unlikely—the features differ both in character and scale to those recorded in other excavated tanneries (Cherry 1991, 297–298), whilst tawing pits such as those of the late 15th or early 16th century at Glastonbury have lime concretion, as well as diagnostic waste including the foot bones of sheep or goats, which seem to have been invariably attached to the skins when received by the tawyer (Mason 2016, 82). Nevertheless, there would have been a steady supply, if only on a small scale, of skins from dead sheep, so local processing cannot be entirely ruled out, although treatment of skins was also a smelly activity which would have been better undertaken at a distance from dwellings. At West Cotton fulling is also suggested as the most likely option (Chapman 2010, 237–239), the tanks and troughs presumably being used to trample or ‘walk’ the cloth in a bath of water and Fuller’s earth to remove grease and speed the process

SPAC E A N D BU I L DI NG S

of matting (Walton 1991, 330, 332). The fulling suggestion is not totally convincing. While the larger trough features might be suitable for fulling by trampling, some of the others, including the Burton Dassett and Upton tanks, would really be too small since cloth was usually made in 12 or 24 yard lengths (Chorley 1986, 125). While Burton Dassett produced two iron tenter hooks from Area D2, the only fnd relating to the fnishing of cloth from West Cotton was a glass linen smoother and there was a possible example in stone from Upton (Chapman 2010, 387; Hilton and Rahtz 1966, 111, fg. 11, ST4). The bleaching of linen cloth or yarn is offered as another possibility for the use of the features at West Cotton (Chapman 2010, 239) and is a distinct possibility here at Southend; linen was made in smaller lengths than woollen cloth, so the troughs were of a more suitable size for soaking the material in lye before outdoor drying. In any event, some secondary processing of an agricultural product seems more likely. The production of glue from hides, which might be done on a smaller scale, involved prolonged soaking of the hides in water and lime. Alternatively the Burton Dassett E5 tank may have served a function in the malting or brewing carried out on the tenement, such as for steeping the grain to encourage germination in malting. Malting was also a major activity at West Cotton, although there was no evidence for it in the more limited excavated area at Upton. The salting of meat is also a possibility. House size and foor plan evidence The general size and plan complexity of the Burton Dassett houses can be seen to increase markedly over the life of the settlement. Of seven complete plans dating to the 13th century, the average foor area was 437 sq. ft (40.6m 2 ); of seven dating to the 14th century, including one with an upper room (H3), the average foor area was 634 sq. ft (58.9m 2 ); and of eight dating to the 15th century, including three with upper rooms, the average was 814 sq. ft (75.6m 2 ). This kind of increase is a phenomenon observed across Midland and south-eastern England refecting general rising living standards (Astill 1988, 56), possibly here accentuated by the particular development of these houses expanding from small beginnings on new plots. The foor areas of the individual houses serve as a rough index to the relative fortunes of the excavated tenements on either side of the road (Figure 3.19). The frst houses on the newly laid-out tenements were relatively small and most were gradually expanded. On the south side of the road there was no increase in foor space after the early 14th century, and in fact in Area K the amount of living space actually decreased, when the east end of the K4 house was converted to the housing of animals. On the other hand, on the north side of the road, with the

173

exception of Tenement D1 which was abandoned, the houses continued to increase in area through the later 14th and 15th centuries. While the excavated Dassett Southend houses were clearly substantial, they seem to have been somewhat smaller than other comparable houses in the region. Alcock and Miles (2013, 34) describe the ‘characteristic’ Midlands medieval peasant house as a three-bay structure having a foor area of 850 sq. ft (79m 2 ), or 1100 sq. ft (102.1m 2 ) if it had an upper foor in one bay. The 19 Dassett Southend house plans, with total foor areas ranging from 217 sq. ft (20.1m 2 ) (D13) to 1159 sq. ft (107.6m 2 ) (E6), and an average of 644 sq. ft (59.8m2), all fall in the lower part of the distribution of foor areas recorded by Alcock and Miles for examples dated between 1317/18 and c1525 (Figure 3.20). The Dassett Southend average is reduced by the generally smaller 13th-century examples which are absent from the Alcock and Miles group, but this does not affect the general trend. It is possible that the standing houses recorded by these authors are not completely representative of Midlands peasant houses but biased towards the larger examples which are more likely to have survived. The footprint of the fve ‘lesser hall houses’ studied by Wood-Jones (1963) that were suffciently complete to provide comparanda ranged from 414 sq. ft (38.5m 2 ) at the mid-14th century Priory Farm, Balscott, Oxfordshire, to 1102 sq. ft (102.4m 2 ) at Chinner Farm, Chacombe, Northamptonshire, of c1500; again these are generally larger than the Dassett Southend buildings. Field (1965), examining agreements for the construction of peasant houses in Worcestershire, provides an indication of house sizes in terms of bays. These ranged from single-bay to fve-bay structures, with the majority having three bays. Extrapolating from the 18 ft by 15 ft (5.4m by 4.5m) measure for an average bay given by Alcock and Miles (2013, 32) provides an average area of 782 sq. ft (73m 2 ); again, the Dassett Southend houses seem small, although this calculation does not allow for upper rooms, each of which would effectively extend the area by that of an additional bay. The Southend houses do, however, appear to be larger than those 14th- and 15th-century houses recorded by Beresford (1975, fgs 16–17) at Goltho and Barton Blount, where the average areas were 585 sq. ft (54.3m 2 ) and 505 sq. ft (46.9m 2 ) respectively. These houses are perhaps to be considered as belonging to a northern, rather than a midland, tradition and they were generally earlier in date, most having been abandoned by the early 15th century. Thus although they were substantial structures, some incorporating a number of up-market features, the generally smaller than average size of those Dassett Southend houses excavated suggests that a greater proportion of them were occupied by traders, artisans or smallholders of limited means rather than wealthier peasant farmers.

174

Figure 3.19 Comparative histograms of foor area/plans of the excavated houses

BU RTON DA S SET T SOU T H E N D

SPAC E A N D BU I L DI NG S

175

Figure 3.20 Floor area of 18 complete Dassett Southend house phases compared with 48 dated (1317/18–c1525), complete or probably complete, peasant houses recorded by Alcock and Miles (2013, tables 2.2, A2.3 a and c)

House plans Eight of the tenements produced house phase plans of suffcient quality to allow some examination of developments in the use of domestic space (Figures 3.14 and 3.21). Figure 3.21 presents schematic plans with links between rooms and to the outside in order to emphasise the relationship of the rooms to each other. It should be borne in mind that, although the quality of evidence from the excavations was in general good, the identifcation of partitions, hearths and doorways may not always have been complete, particularly in the earlier phases. The range of types shows that even peasant-level houses could exhibit a fair degree of plan variation and even complexity. A number of the houses (A2, D13, E2, H2, I2 and K2) had the most basic of plans, with a single undivided space (such as, for example, S18 at West Cotton, Raunds (Chapman 2010, 93–98), or the ‘typical house’ described at Seacourt, Oxfordshire (Biddle 1962, 120). They would all presumably originally have contained hearths, but traces of these survived only in A2 and I2. These were all among the earlier (13th and early 14th century) and smaller houses, and represented the frst buildings on newly

laid-out properties. The smallest houses (E2 and D13) were both one-bay halls. H2 and K2 had two or three bays, while A2 and I2 had three bays. H2 had opposing doorways, to the street and into the rear of the tenement, in the eastern bay, while I2 had similar doorways, opposing or offset, in the central bay. Even where the plan shows no clear differentiation into rooms, however, it is likely that different activities may have taken place in different parts of the space, and that different activity areas were separated by structures that were temporary or insuffciently substantial to have left any archaeological trace. The next stage up adds an extra room to the hall, reached only by a doorway from the hall. A3 added a kitchen to the north-east of the existing hall. L1 appears to have consisted of a two-bay hall with a separate chamber at the west end. Only one doorway survived, in the centre of the hall leading into the yard. The smithy building (J2) had a two-bay hall/ forge area with a separate chamber to the east. The hall/forge area probably had a central (domestic) hearth in the western bay and facing doorways at the western end of the eastern bay of the hall, the southern doorway to the yard entered through a small porch. The fnal J4 re-arrangement was similar, but with the forge area fully separated off by partitions.

176

BU RTON DA S SET T SOU T H E N D

Figure 3.21 Access analysis of space per phase and building The K4 house would also have consisted just of a hall with a chamber to the west after the conversion of the east end to house animals. In this phase the chamber and both bays of the hall contained central hearths. The H3 house was a variation on the two-room arrangement, with an upper room reached by a stair off the hall. The extension for the stair also contained a lobby/passage providing a second doorway from the rear yard from the west into the hall. The next group had a central hall with single rooms at each end reached from the hall (A5, D22– 23, D24). A5 added an extra room to the west of the existing hall and kitchen. D22–23 had three equal bays, separated by partial partitions, with a central hall. All three rooms contained hearths, although the main hearth was in the hall. The eastern chamber contained post settings for fxed furniture.

D24 re-arranged the previous space into a two-bay central hall with a chamber to the east, and added another larger chamber to the west. The hall had opposing doorways at the east end of its western bay. Both chambers contained hearths against walls, while the hall had central hearths in its eastern bay, hearths against walls in both bays and a stone-lined ‘processing’ trough in its western bay. Two houses at Stoneleigh, Phoenix Cottage and 2 Church Lane, provide Warwickshire examples of three-bay houses with inner room (here foored over) and service end either side of a hall (Alcock and Miles 2013, STO-F and STO-H). Once again, we cannot be certain that all the internal divisions within the Southend buildings will have left an archaeological signature. Another group comprised a central hall with more than two rooms leading off it (K3, I4, E3, E5). The

SPAC E A N D BU I L DI NG S

K3 house had a two-bay hall with central hearths and opposing doorways on the east side of the eastern bay, a chamber to the west and two small service rooms to the east. This represents the classic medieval hall-house, but this was the only example at Dassett Southend exactly conforming to this pattern. The I4 house had a broadly similar plan, but the western room was bigger than the hall and contained the doorway to the yard. It also contained a hearth. The E3 house had a central hall with opposing doorways on its west side and a central hearth, a chamber to the east with a central hearth and post settings for fxed furniture (like the equivalent room in the D22–23 house), and two rooms (without hearths), one leading into the other, to the west. The north doorway from the yard was entered through a small timber porch. The succeeding E5 house kept the same layout, but converted the room to the east of the hall into a workshop or processing room with a stone-lined tank and channels and hearths against the north and south walls. In the hall, as well as the earlier central hearth there were also two hearths towards the north-east corner and a series of fve in the south-east corner. The most complex plans all dated to the mid- to late 15th century and involved multiple suites of rooms (E6, D25 and D26). The E6 house had a two-bay hall, entered from the street and yard by opposing doorways at the west end, with central hearths in the west bay and two hearths against the south wall in the eastern bay. West of the hall, two doorways to north and south opened into a western room, and to the east a south doorway opened into a room containing two hearths by the north and south walls, while another doorway to the north-east led to a further room containing a post-base on its southern side. The north doorway from the hall to the yard opened into a porch, from which another doorway led westwards to a stair to an upper room over the west end of the house. The upper room effectively formed a self-contained ‘apartment’, albeit unheated. The D25 house contained three separate units. It was entered from the street into a passage/lobby, with one doorway into a room to the west and a second doorway to the east into a one-bay hall with a hearth against its south wall. There was an external doorway to the yard in the north-west corner of the hall and a second doorway to the south-east to an eastern room beyond the hall. At the west end of the house an external stair provided a separate entrance from the yard to an upper room over the western room. This would also have been unheated. In the fnal D26 remodelling of the house the entrance passage became a through passage with opposing external doorways separate from the hall, the frst and only time this arrangement was seen. Off the passage, to the west a room contained two hearths, probably in a smoke bay, and an internal stair turret leading to the upper room at its west end. Off the passage to the east was the single-bay hall with a corner hearth

177

to the south-west and a doorway to the south-east leading to an eastern room. The previous three suites had thus been reduced to two, but both contained hearths. Some evidence, notably the large numbers of Cistercian ware drinking vessels found around the house, suggests that the western suite (and possibly the western room in the D25 house) might have been used for the sale of ale. Including these last three, fve houses can be identifed as having an upper room over one bay. The earliest (H3) dated to the early 14th century, the others (D25, D26, E6 and F6) were mid- or late 15th-century, by which time the provision of upper rooms seems to have become standard. A number of trends can perhaps be observed. In the 13th and early 14th centuries the houses were smaller and simpler in plan. The general increase in foor area over time was accompanied by a proliferation of separate rooms and a concomitant increase in plan complexity. The extra rooms offered the opportunity for greater privacy of living. The increased complexity also brought acceptance of a certain amount of wasted space elements (passageways, stairs and porches) whose only function was access or circulation. Where doorways can be identifed, the external access to the houses, both from the public street and from the yards within the tenements, was almost always by opposing doorways directly into the hall. There was only one late example (D26) of a through passage partitioned off from the hall. Other exceptions were the I4 house with a doorway from the street into the hall and a doorway from the yard into the western room, and the D25 house where the street doorway opened into the passage while the yard doorway opened into the hall. The possible hint that there may have been an extra external door at the east end of the E3 house may indicate a temporary separate occupancy of the eastern room or perhaps use for some non-domestic purpose, for example as a shop. The conversion of the east end of the K4 house for animals produced the effect of a baffe-entry arrangement, similar to that of Building 3, phase 1 at Gomeldon, Wiltshire, noted there as anomalous, and perhaps also the result of conversion (Musty and Algar 1986, 137, 149, fg. 5; see also Gardiner 2000, 165–166). Throughout, the hall remained the most important room and the focus for domestic life; but with increasing numbers of rooms leading off it, this importance must have diminished over time. The halls tended to contain the main hearths, but virtually all the multi-room plans had hearths— sometimes more than one—in more than one room, although how many of these were in operation at one time is unknown. The tendency in the later period to place hearths against walls was presumably both to allow for better smoke control by smoke hoods or chimneys, and to provide more unobstructed space in the rooms.

178

The development of the houses largely by extensions or partial rebuilding means that plans were constrained by the previous arrangements and the size and layout of the tenement plots. The one example (K3) of the classic hall-house was one of few complete rebuildings (and on a large plot) where the new building could be laid out to a plan. Although one can see overall patterns in the layout of medieval houses, their builders were not attempting to replicate a standard plan; the variation in the size and arrangement of rooms shows the inhabitants of Southend creating and adapting structures which suited their own needs. Later periods saw the creation of self-contained suites of rooms, possibly to allow multiple occupancy, catering for retired members of a family or previous tenants, or to allow parts of the house to be used for commercial activities such as the sale of ale. Some of the houses also included space for craftwork activities; the Area J smithy is perhaps a special case, but there were the stone-lined troughs in the halls of the I2 and D24 houses, and a whole room was devoted to ‘processing’ at the east end of the E5 house. By virtue of their location, these activities appear to have been on a domestic scale and do not seem to represent large-scale industry, although they may well represent some degree of craft specialisation in keeping with the apparent quasi-urban nature of Southend. This is normal; much medieval industry seems to have been small-scale and home-based and only a few crafts such as tanning and fulling required extensive space for vats or machinery. The adoption of the hall-house plan with its aristocratic origins and hierarchical division of space has been seen as refecting the developing social aspirations of the peasantry (Harris 1978, 31–32; Mason 1973, 30; Mercer 1975, 19–20). Despite the fact that the exact classic plan occurs only once at Dassett Southend, using the position of the main opposing doorways it is possible to identify theoretical ‘high’ and ‘low’ ends in a number of the house plans (K3, E3, D24, E5, D25, D26 and E6), but with little frm evidence about what the rooms were actually used for, either here or at other excavated sites, it is diffcult to say how far the formal elements of ‘high’ and ‘low’ were applied in relation to the householders, their families and servants. All the 15th-century upper rooms (D25, D26 and E6) were over the ‘low’ ends, reducing the height (and status) of the ground foor rooms. Only in D26 was the upper room reached through the ‘low’ end; the other upper rooms were reached from the hall or a separate entrance. In the E5 house the workshop/ processing room with the stone tank and stone-lined gullies occupied the ‘high’ end of the building. How far any hierarchy of rooms was maintained would probably depend as much on practical considerations such as the changing needs and composition of the households, as on the degree of formality in familial relations maintained by individual families.

BU RTON DA S SET T SOU T H E N D

Outbuildings Documentary sources record a number of different types of outbuilding on West Midlands peasant holdings (Dyer 1986, 25). These include barns, granaries, byres, stables, sheep-cotes, pigsties, brewhouses, bakehouses, cart-houses and dovecotes. The commonest was the barn, although these must often have been multi-use buildings, particularly when, as in many cases, they were the only outbuilding on a plot. The excavations produced complete or relatively complete plans for about 12 outbuildings, along with fragments of about 18 more, the evidence in some cases being fairly insubstantial (Figures 3.22 and 3.23). They included the two phases each of the K2 granary and the Area E brewhouse (E3 and E4). Excluding the converted end of the K4 house, there was one byre/stable (K4) indicated by a drain in its foor. Two or three other buildings (D23, D24 and H4), without drains but containing internal divisions that could represent stalls for animals, may also have been byres/stables. One of these (D24) could have been a barn/store partly converted to this use. Only one barn (A5) could be identifed with any confdence, despite relatively meagre structural traces, with wide opposing doorways on a bay that could have been used for threshing. About fve other relatively large outbuildings, undivided internally, seem to represent barns or stores (D25, D26, I4) or probable barns/stores (H4, L1). One probable building, apparently open-sided, may have been a cart-shed (H4/K4), while three smaller ones with foor areas of 16m 2 (172 sq. ft) or less (A5, D23, H3) can be classed as sheds. Another eight probable outbuildings were too fragmentary to categorise (D13, E2, F4, F5, F6, H2, I3, J4). Of these the E2, F4 and H2 buildings were really just clusters of post-holes, the I3 and J4 outbuildings were shed-sized, while the D13, F5 and F6 buildings could have been larger barns/ stores. There were also four other very small outside structures, including three possible privies (D23, I3 and K4) and a dog kennel (K4). The Dassett Southend outbuildings were generally less well constructed than the houses: virtually all were substantially of timber and a signifcant proportion still had structures supported on earthfast posts well into the 15th century. This is in marked contrast to West Cotton, where the houses of the peasant tenements may have lacked the architectural details of those at Dassett Southend but the outbuildings were constructed of stone to the same standard as the houses. Again this may refect the possibility that the West Cotton buildings were specifed by the landlord as a commercial venture rather than by the tenants for their own use. The Dassett Southend outbuildings seemed to be fewer in number and smaller than those recorded at other similar sites, although there may have been

SPAC E A N D BU I L DI NG S

179

Phase A5*

Type Barn

Bays 4

Structure TF, ?crucks on PB/stone footing

A5*

?Shed, lean-to against barn

1

EF posts

D13

?Outbuilding/barn/store

?

TF on stone footing

D23*

?Shed

1

EF posts

D23*

?Byre/stable

?2

EF posts

D23

?Privy/shed

D24*

?Byre/stable/barn/store

?2

EF posts

D25*

Barn/store

?5

EF aisle posts, stone bases

D26*

Barn/store

?2

TF on stone footing

E2

??Outbuilding

?

EF posts

E3*

Brewhouse

3

TF on stone footing

E4*

Brewhouse

3

TF/Stone N end

F4

?Outbuilding

?

EF posts

F5

Outbuilding/barn/store

?

TF on stone footing

F6

Outbuilding/barn/store

?

TF on stone footing

H2

?Outbuilding

?

EF posts

H3

Lean-to/shed

1 bay

H4*

?Byre/stable

2

TF on footing ?open-sided to S TF on stone footing

H4

?Barn/store

?

TF on stone footing

H4/K4

??Barn/store/cart-shed

?

I3

?Privy/shed

TF on footing ?open-sided to W T on stone footing

I3

?Outbuilding/shed

?

?

I4*

Barn/store

?2

TF on stone footing

J4

?Outbuilding/shed

?

EF posts

K2*

Granary

?2

Stone

K2*

Granary/store

?2

Stone/TF, two storey

(K4)

Byre/stable

1½

E end of house

K4

Byre/stable

?

TF? stone-lined drain

K4

?Kennel

Stone

K4

?Privy/shed

T on stone footing

L1

?Barn/store

EF posts

?

TF on stone footings

* illustrated on Figure 3.23. EF earth-fast, PB post-base, T timber, TF timber-framed

Figure 3.22 Dimensions of outbuildings at Dassett Southend as excavated

Dimensions m 14.5×4.5 ft 47.5×14.8 m ?5.4×2.3 ft ?17.7×7.5 m ? ft m 3.75×2.4 ft 12.3×7.9 m 6×4.2 ft 19.7×13.8 m c1×1 ft c3.3×3.3 m 8.8×6 ft 28.9×19.7 m ?9.8×6 ft ?32.1×19.7 m 8.4×5.4 ft 27.6×17.7 m ??4×3.5 ft ??13.1×11.5 m 9.4×5.3 ft 30.8×17.4 m 9.4×5.3 ft 30.8×17.4 m ? ft m 2.2+×5.7 ft 7.2+×18.7 m 1+×3.5 ft 3.3+×11.5 m ? ft m 4.2×3.0 ft 13.8×9.8 m 6×5.5 ft 19.7ft×18 m 6m+×4.8 ft 19.7ft+×15.7 m ?6+×4.5 ft ?19.7+×14.8 m 1.8×1.8 ft 5.9×5.9 m ?×c3 ft ?×c9.8 m 6.7×4–5.5 ft 22×13.1–18 m 3.2+×3.6 ft 10.5+×11.8 m 7.2×5 ft 23.6×16.4 m 9.2×5 ft 30.2 x16.4 m 5.7×4.2–7.1 ft 18.7×13.8–23.3 m ? ft m 2.8×1.2 ft 9.2×3.9 m 2.3×1.6 ft 7.5×5.2 m 4.3+×4.3 ft 14.1×14.1

Floor area 65.25 sq. m 702 sq. ft ?12.4 sq. m ?133 sq. ft ? 9 sq. m 97 sq. ft 25.2 sq. m 271 sq. ft c1 sq. m c11 sq. ft 52.8 sq. m 568 sq. ft 58.8 sq. m 633 sq. ft 27.2 sq. m 293 sq. ft ??14 sq. m ??151 sq. ft 49.8 sq. m 536 sq. ft 49.8 sq. m 536 sq. ft ? ? ? ? 12.6 sq. m 135 sq. ft 33 sq. m 355 sq. ft ? ? 3.24 sq. m 35 sq. ft ? 33.8 sq. m 364 sq. ft ? 36 sq. m 387 sq. ft c82 sq. m c883 sq. ft 33.9 sq. m 365 sq. ft ?

3.7 sq. m 39 sq. ft ?

180

BU RTON DA S SET T SOU T H E N D

Figure 3.23 Comparative outbuilding plans by phase

others further back on the tenements outside the excavated areas. Unlike the houses the outbuildings did not increase in size over the life of the excavated tenements. They tended to be rebuilt rather than extended. More outbuildings were identifed for the later periods, which may indicate a propensity to erect another building when additional space was required rather than adapt existing structures, and may refect a more general trend after c1350 (Gardiner 2000, 162); alternatively this may refect the better survival of structural components such as post-holes and post-bases amongst the later deposits. Developments in the sizes and use of outbuildings are more diffcult to trace than those of the houses. The remains of the structure on A5 identifed as a four-bayed barn—a length of wall at one end and incomplete opposing pairs of ragged post-bases along the main longitudinal walls—were really very slight, to the extent that the loss of even one of the surviving structural elements would have made the identifcation of this building even more problematic (Figure 2.68). Similar structures elsewhere on the site could have left no trace. Not only were the outbuildings less substantial but they were not always situated in close proximity to the dwelling. The considerable variety of functions and structural methods does, however, seem to be a special characteristic of the Burton Dassett outbuildings.

Area K granary The earliest outbuilding recorded was the mid-/ late 13th-century Area K granary (K2). This was the only outbuilding constructed with full stone walls. It is interpreted as a granary because of its relatively small size, the solidity of its construction, with buttressed corners, and the evidence that it had a suspended foor supported on corner post-bases. The internal foor area of the granary was 20.65m2, which suggests that its maximum capacity might be estimated at 20m3 or the equivalent of 549 bushels. At a yield of 12 bushels of grain per acre, 549 bushels would be the product of 45 acres, or a holding of 90 acres worked on a two-course rotation. On the basis of this calculation, the granary might be expected to hold the produce of 1½ of the large 60 acre Great Dassett yardlands (see Section 1.7 above). This would put the tenant of Area K in the top 5% of Great Dassett landholders who held more than a single yardland. Grain was normally stored in sheaf, however, and only threshed gradually as required for use or sale. At planting time threshing would be stepped up to produce the necessary quantities of seed corn. A cultivator would therefore never need to store such a quantity of threshed grain. One person who might would be a grain dealer—and it is possible that the tenant of the Area K tenement pursued this activity—supplying grain to the inhabitants of the

SPAC E A N D BU I L DI NG S

new developments and to local smallholders, or even further afeld to Banbury or Coventry, although, if the granary was his only store, the scale of his trading was relatively modest. A total of 549 bushels, would represent seed corn needed to plant only 270 acres of wheat or 137 acres of barley, when the Sudeley demesne cultivated 500 acres in 1280, and tenant arable covered c3630 acres (see Section 1.7 above). It is possible that the plot contained other outbuildings, including a general purpose barn, in the unexcavated area to the south-west. A grain dealer’s turnover was, however, likely to be well in excess of his storage capacity, and an annual turnover of 5000 bushels might only require 500 bushels to be stored at any one time. Grain dealers seem rarely to have been specialists and the trade might well have been one of a portfolio of activities. A stonemason in York was a part-time dealer in grain, for example (Chris Dyer pers. comm.). Few granaries have been identifed on village sites because generally peasant farmers had no need to store large quantities of threshed grain. For this reason, a small square building with a wide stone foundation on Croft B at Caldecote, Hertfordshire, dated to c1460–1630, was interpreted, after some debate, as a dovecote rather than a granary (Beresford 2009, 109, 131). The remodelling of the granary which took place in the late 13th century (K2) added a probably timber framed upper storey reached by an external stair. This would have doubled the storage capacity of the building, but it is possible that the original suspended foor on the ground foor was removed at this time and replaced by an earth-and-rubble foor, the lower space now becoming a simple store. The upper foor may have remained as a granary, or been given over to some other function. Area E brewhouse The outbuilding to the north-west of the Area E house which went through two phases of development (E3 and E4–5) is interpreted as a brewhouse, primarily on the basis of successive malting kilns in its north-east corner and successive drains in its foor. First constructed in the early 14th century (E3), the building measured 9.4m by 5.3m, probably with a timber-frame supported on narrow stone footings (Figure 2.28). It had a wide doorway in its eastern wall. The interior was divided into four alcoves by partitions on narrow stone footings, perhaps representing raised grain storage areas, or a raised foor on which to spread the grain to germinate. The malting kiln, of which only vestiges of the earliest phase survived, lay in the north-eastern alcove. The adjacent north-western alcove was almost fully divided off from the rest of the room and it is possible that its foor level was raised to provide access to the top of the kiln. At the south end, south of a spur wall, a curving gully may have been the drain

181

from a steeping vat although there was no outfall and it would have functioned, perhaps unsatisfactorily, as a soakaway. In the following phase (E4, Figure 2.42) the north end of the building was rebuilt and the partitions removed, although the malting kiln remained in the north-east corner and the drain at the southern end, suggesting that the overall working arrangements were retained, steeping the barley at the southern end, spreading it out to germinate in the centre of the room, then drying to prevent further sprouting in the oven at the north-eastern corner. This later layout is broadly similar to that of a mid-14th- to mid-15th-century brewhouse at Great Linford (Mynard and Zeepvat 1992, 60, fgs 17–18, pls 6–7, Croft C, building 13). This measured 7.5m by 5.5m, had a malting kiln in one corner, a circular oven/vat base in another, and a drain across the f loor, and there were similar arrangements in two further structures there (Croft E, building 11, dated to the 14th century, and another of probably similar date on Croft G below building 20; Mynard and Zeepvat 1992, 65–67, fg. 20 and 74–76, fg. 26). The word ‘kiln’ often appears in documents. Medieval malting kilns are found either as freestanding structures, with round or rectangular kiln chambers and probably originally timber or clay superstructures, or within buildings. Those within buildings divide into those with a small foor area where the building is effectively an adjunct to the kiln (as at West Cotton)—these are probably ‘malthouses’—and those in larger buildings, often attached to manorial complexes and often containing other structures such as vat bases and ovens, brewhouses or combined brew/bakehouses. The manor house at Great Linford had a substantial brewhouse adjacent from the late 14th to late 15th century which also incorporated a probable breadoven (Mynard and Zeepvat 1992, 99–101, fg. 41), in addition to the examples cited above, of which only that on Croft G included evidence for the full set of brewhouse components: oven, malt kiln and base for a lead vessel or vat. Another category has kilns at one end of a larger building. These include those at Houndtor (Barns 1–3) and Hutholes (House 6) (both in Devon) (Beresford 1979, 140–142, fg. 24), described as corndrying ovens, where there were two kilns side by side, in each case the larger one on the right described as an oven and the smaller one on the left as a kiln; the consistency of the arrangement indicates a difference in function, and later all the ovens were flled in and replaced by higher level ovens. These are a less satisfactory parallel for the Burton Dassett example than those instances where a kiln occupies one corner of a building. At Caldecote, Hertfordshire, a kiln and a circular oven/vat base were recorded at one end of the mid-15th-century Croft B barn, part of which had been converted into a bakehouse/brewhouse (Beresford 2009, 108–109, fg. 9.14).

182

The West Cotton series of malthouses with their associated structures suggests activity on a much larger scale than at Burton Dassett was being carried out as a commercial operation over and above what was required for domestic consumption; clearly that locality specialised in malt as there is evidence for its being traded from Higham Ferrers down to London (Chapman 2010, 225–229). A probable local parallel for the Building E arrangement at Burton Dassett is the 15th-century Structure III at Bascote (Warwickshire) (Litherland et al. 2008, 88–90, fg. 5). Although this is not discussed as a malting kiln/brewhouse in the report and the samples from hearth 114 contained no distinctive plant remains suggestive of malting, the layout of the various features is very similar. The circular structure F157 interpreted as an oven might have been the base of a steeping vat similar to those at West Cotton. Analysis of the charred plant remains from phase E5 of the Burton Dassett kiln (online archive, Section 8.21) showed that barley, oats and wheat were being processed, along with a relative abundance (c25%) of weed seeds, chiefy Brassica spp. The barley and oats may have been grown together as dredge. Well over half the barley grains had germinated, but the wheat grains were apparently ungerminated. Wheat was sometimes malted to make a superior ale, but this was uncommon. The evidence thus suggests possible use of the kiln for drying or parching grain as well as for malting barley. Similar ‘dual use’ is also likely at West Cotton, where the presence of Brassica in quantity is also attested (Campbell 2010, 475 and 496); it is also noted that the addition of mustard seed to ale increased thirst and was therefore used in ales intended for sale, although it may also have been a contaminant of the cereal crops. The weeds in the Southend malting kiln may even have been used as fuel for the kiln. Byre/stables Four possible byres or stables were identifed, one incorporated into a house and three free-standing. During the last phase of House K4 (mid-14th to early 15th century) a stone-lined dung gully was inserted across the eastern room of the house, now divided from the rest of the house by a dog-legged timber partition on a narrow stone wall which provided entry to the house via the original north door. The north door now led into a small lobby and entry to the stable via what had originally been the opposing south door (Figure 2.35). A short length of narrow wall foundation from the previous phase at right-angles to the drain was incorporated into the new arrangement to divide off a stall some 2.2m wide at the northern end of the room, but the fact that the dung gully extended across the full width of the building suggests that at least one other animal could have been accommodated.

BU RTON DA S SET T SOU T H E N D

Another drain at the southern end of Area K (originally recorded as part of Area H; Figure 2.39), c4.5m long and c0.1m wide, may originally have been within a building, which, in the absence of any other structural evidence, is likely to have been timber framed and constructed to accommodate animals. The earliest outbuilding to the north-west of the house on the D2 tenement dated to the 14th century (D23, Figure 2.18). It measured 6m by 4.5m giving a foor area of 25.2m 2 (271 sq. ft) with a relatively primitive structure supported on irregularly spaced lines of earth-fast posts. It was possibly divided into two areas of unequal size, suggesting use for the accommodation of animals rather than for storage, although there was no internal drain. The complexity of the post-holes on this part of the plot, deriving from more than one phase of outbuilding, however, makes any interpretation somewhat tentative. Finally, on Tenement H (Phase H4, mid- to late 14th century) a slightly trapezoidal timberframed structure, 6m by 5.5m, was built against the boundary wall with Area K to the west, divided into two sections 2.7m wide. The west wall was the boundary wall with K, the east wall was simply a line of ironstone blocks with 1.5m gaps at either end, perhaps doorways. The north and south walls were represented only by the edges of the rubble foor, apart from two post-bases on the south side; timber sill-beams cannot be ruled out. A stone partition extended halfway across the building from the western wall, terminating in a large block that may have been a post support. The size and proportions of these arrangements suggest animal accommodation, despite the lack of an internal dung gully. Byres at one end of a house are not uncommon (for example Building 5 on Croft B and Building 16 on Croft F at Great Linford have rooms at the western end identifed as byres, or possibly a dairy in one instance, on the basis of drains across the foor; Mynard and Zeepvat 1992, fgs 15 and 23); they are reminiscent of the traditional longhouse with accommodation for animals at one end, although even here ‘in many cases it is hard to prove that animals were stalled in the low end’ (Grenville 1997, 137). There are fewer excavated examples of free-standing stables or byres, however; none was conclusively identifed at West Cotton or Great Linford. It is probable that byres were built and demolished as needed; there is no reason to expect all peasant houses to have had byres. Barns and store buildings One defning characteristic of a barn is that it is a large undivided space. Some of the structures excavated at Burton Dassett were probably divided and might be better considered as store buildings; unfortunately evidence for internal partitions in what in some cases are somewhat insubstantial buildings may not always be archaeologically recoverable.

SPAC E A N D BU I L DI NG S

183

The best evidence comes from the 15th-century phases, such as the sequence of structures north-west of house D24 against what had been the boundary with Tenement D1. The earliest structure in the sequence was the putative byre/stable described above. In the early to mid-15th century this was replaced by a larger timber structure, 8.8m by 6.0m, supported on earth-fast posts (Figure 2.60). These survived at irregular intervals and there was some recutting indicating repair and rebuilding, but there was some pairing of posts evident suggesting that the building might been divided into two bays 4.4m wide. The total area was c53m 2. In the following phase D25 (mid- to late 15th century) this was replaced by an aisled building of fve narrow bays with stone-packed earth-fast aisle posts and external walls supported by sill beams set on post-bases (Figure 2.72); this was separated stratigraphically from its predecessor, demonstrating that the rebuild was complete. At 6m wide and at least 9.8m long (c59m 2 ), it was larger than its predecessor. Finally this structure was replaced in phase D26 by a smaller building 8.4m by 5.4m (c44m 2 ) consisting of what must have been a timber superstructure set on narrow stone footings (Figure 2.77). The evidence for a possible building on the fnal occupation phase of Area I (early to mid-15th century) was insubstantial, consisting of traces of rubble footings and a post-hole (Figure 2.52); it incorporated part of the rubble yard surface extending up to the house as its foor. It would have been 6.7m long and about 4m to 5m wide. The largest of all the free-standing outbuildings was that constructed north-west of house A in the mid-/late 15th century; it was 14.5m long by 4.5m wide (Figure 2.68). This was probably of four bays. The western end wall was founded on stone in order to consolidate the flls of an earlier boundary ditch, but the remainder of the structure was apparently founded upon post-bases, not all of which were present. It appears to have had a foor area of c65m 2. Other possible barns or storage buildings were recorded on D1 (phase 3), consisting of one corner of a structure, presumably timber-framed, resting

barn/store

date

total foor area

A5

mid-/late 15th century early/mid-15th century mid-/late 15th century late 15th century early/mid-15th century

65.25 m2 702 sq ft 52.8 m2 568 sq ft 58.8 m2 633 sq ft 45.4 m2 489 sq ft 33.8 m2 364 sq ft

D24 D25 D26 I4

estimated storage area (8ft stack) 35 m2 377 sq ft 36 m2 388 sq ft 37.8 m2 407 sq ft 19.5 m2 210 sq ft 20.52 m2 221 sq ft

on narrow stone footings and with a rubble foor (Figure 2.31); on H4, consisting of short lengths of rubble foundation, partly robbed (Figures 2.39 and 2.41) and on L, where the width of the door to the southern room of the wing of the partly excavated building may indicate a storage function, whilst at the southern end of plot K, but apparently facing on to plot H, was a possible open-fronted cart shed (Figure 2.51). Some idea of the typical size of medieval barns in the region can be gained from Field’s study of Worcestershire building agreements (Field 1965). These include 35 relating to barns, dating to between 1384 and 1507, of which 28 (or 80%) were specifed to be three-bay, 5 four-bay, 1 fve-bay and 1 two-bay. If a typical bay in a barn was 15ft by 15ft (4.6m by 4.6m), then the foor area of a one-bay barn would be 225 sq. ft (21m 2 ), that of a three-bay barn would be 675 sq. ft (63m 2 ) and a fve-bay barn 1125 sq. ft (105m 2 ). The average for Field’s barns would be 713 sq. ft (66.6m 2 ). Including the D24 barn/store/byre/stable, the fve complete Dassett Southend barn/store buildings ranged from 364 sq. ft (33.8m 2 ) (I4) to 702 sq. ft (65.25m 2 ) (A5) with an average of 551 sq. ft (51.21m 2 ). Again, these are at the low end of Field’s Worcestershire range, with even the largest A5 barn smaller than his average. The seven barns recorded at Caldecote, Hertfordshire, (including one which also served as a malt/brewhouse) were all bigger than the Dassett Southend average. They ranged from 576 sq. ft (54m 2 ) to 1222 sq. ft (114m 2 ) with an average of 800 sq. ft (74m2 ) (Beresford 2009, 130, fgs 9.39, 9.12 and 9.14). Similarly at Goltho and Barton Blount (Beresford 1975, 45) the three recorded Period 3 barns had foor areas of 544 sq. ft (51m 2 ) (Barton Blount Croft G), 598 sq. ft (56m 2 ) (Goltho Croft C) and 704 sq. ft (65m 2 ) (Barton Blount Croft E). Taking a 13th-century fgure that a stack of grain in sheaf measuring 30ft by 15ft by 10ft in a large seigneurial barn contains 40 quarters or 320 bushels (Dyer 2012c, 321), allowing for stacks of only 8ft rather than 10ft in the smaller Dassett Southend

estimated capacity (bushels) 214

estimated acreage at 8 bushels/acre 27 acres

estimated acreage at 12 bushels/acre 18 acres

estimated acreage at 16 bushels/acre 14 acres

220

28 acres

18 acres

15 acres

231

29 acres

19 acres

15 acres

119

15 acres

10 acres

8 acres

126

16 acres

10 acres

8 acres

Figure 3.24 Barns and stores, estimated grain storage capacity and possible associated acreages

184

buildings, one bushel would occupy a notional 1.758 sq. ft. Taking the fve Dassett Southend barn/stores, allowing for some access space within the buildings and a threshing area in the through passage of the A5 barn, it is possible to produce a speculative estimate of the effective storage area and therefore the capacity in bushels of each building. From this, it is also possible to calculate a very rough estimate of the planted area that would have been necessary to produce this quantity of grain, allowing for a range of grain yields per acre (Figure 3.24). Assuming a twocourse rotation, the estimated acreage fgures need to be doubled to give an idea of the landholdings involved. Thus for A5 we may suggest an area of 28 to 54 acres, for D24 30 to 56 acres, for D25 30 to 58 acres, for D26 16 to 30 acres, and for I4 16 to 32 acres. All these buildings dated to after 1350 when the average size of landholdings was larger (Dyer 1989a, 141–143, table 11). The calculation is tenuous, but it gives an approximate order of magnitude. By the standards of the day, an acreage of 30 acres was reasonably large; most peasants had holdings of 15–30 acres (representing around 7–20 sown acres), depending on rotation, and 15 acres may have been enough for self-suffciency, without an absolute need to supplement income with additional craft activities. It is unlikely that all barns were used exclusively for grain storage; in all probability they would have been multi-purpose buildings. All of these buildings, apart from those on plot I, are on the north side of the road where, in spite of their size, the buildings seem to have acquired something of a quasi-urban character by the late 15th century. Sheds A small number of buildings were recognised which on the basis of their size (less than the single bay of a barn) and generally insubstantial structure were classed as sheds: a lean-to building against the west end of house D23, supported on earth-fast posts; another lean-to (c4.5m by 2.5m) on stone foundations west of House H3 (Figures 2.24 and 2.25); and a lean-to (c2.3m by 4.5m) represented by two post-holes in the angle between the north side of the barn and the western tenement boundary. These could have served any purpose from keeping poultry to storing tools, although the structure on H5 yielded pottery which may be suggestive of a dairying function. Other structural traces were also recorded which may well have represented outbuildings, although they were largely so insubstantial as to defy any convincing interpretation: a group of undated postholes east of the E2 house, post-holes and walling along the western boundary of Tenement F, a cluster of post-holes south-west of house H2, an area of possible rubble raft foundation of foor against the eastern tenement boundary of I3, and a stone-

BU RTON DA S SET T SOU T H E N D

packed post-hole and possible rubble foor surface in the angle of the J4 house and its eastern tenement boundary. Some of the post-holes may alternatively have belonged to animal pens or drying racks. Possible privies There were three possible privies, all represented by small rectangular timber structures set against the corners of houses. That on the north-east corner of the late 13th-century D23 house was c1m square with timberwork supported on a line of three postholes with a slot at right-angles (Figure 2.18). The one on the north-east corner of the 14th-century I3 house had narrow stone footings on two sides (Figure 2.52), while that on the south-east corner of the early 14th-century K4 house was supported on a single narrow stone footing (Figure 2.35). None of the structures had cess-pits below, which would mean that they would have contained barrel or bucket latrines. These would be more easily emptied than cess-pits, but this would need to be done more frequently. Two successive small rectangular stone structures, interpreted as garderobes, were recorded on the corner of a mid-14th/early 15th house at Great Linford (Mynard and Zeepvat 1992, fgs 22–23, Croft F, Building 16). These also lacked convincing cess-pits below, being only 30cm deep and without the characteristic organic fll. There was a garderobe, 0.91m wide, on the eastern corner of the 15th-century farmhouse on Croft B at Caldecote, interpreted as serving a frst-foor room (Beresford 2009, 106, fgs 9.12, 9.15). Freestanding garderobe pits at Goltho and Barton Blount (Beresford 1975, 44, fg. 21) were also small and rather shallow (1m by 1m and 0.6m deep), with post-holes to support seats, but with no evidence of covering structures. Deeper cess-pits were found in the settlement at Oversley, Warwickshire (1.0m and 1.6m deep) ( Jones et al. 1997, 44) and similarly, perhaps, at Upton (1.8m and 2.1m deep) (Rahtz 1969, 91–92, fg. 4). Possible dog kennel The fnal outbuilding/structure recorded was the possible dog kennel on the east side of the entrance to the yard of the Area K property, a very suitable location for a watchdog (Figures 2.35 and 2.38). Remains of dogs were scattered throughout the deposits at Southend; the few instances where measurement was possible indicate a withers height in the range 40–50cm, but there is little to indicate breed or stature (online archive, Section 8.20). The kennel was an alcove (1.2m by 0.5m) in the base of a stone foundation (2.8m by 1.3m). The foundation seemed oversized just for the kennel and may also have supported something like a mounting block or a platform with steps for the loading of carts.

SPAC E A N D BU I L DI NG S

Smithy building The smithy on Area J apparently saw relatively little in the way of structural change between construction in the mid-13th century and abandonment around the early 15th century. Extensive robbing after demolition means, however, that few details had survived. A pair of post-holes may indicate a porch to the south, which may have served to limit the amount of light coming into the building to enable the smith to control the heat of the forge, as the colours of fame and metals served to gauge temperature. An opposing doorway on the north side seems less likely in view of the probable internal layout but there would have been space to the left of the likely position of the forge. The building was never enlarged, which may refect the overall prosperity, or lack of it, of the smiths. By comparison, the building identifed as a smithy at Goltho was around 31m 2 (336 sq. ft), smaller than that at Burton Dassett (37.5m 2, 404 sq. ft), but there it is suggested that the smith lived in an adjacent house, providing additional (and probably less unpleasant) domestic space (Beresford 1975, 46). At some point a stone foundation was built against the northern part of the west wall, which may have been the base of a bench or to strengthen the wall. The building does not appear to have been merely a workshop, as there is evidence of domestic occupation. It is not inconceivable that the smith had another forge elsewhere, perhaps in Kineton, with the Southend smithy only occupied periodically, or by a servant. Certainly, by the second half of the 15th century smithing had ceased not only within Area J but also within the areas excavated and feldwalked, so the remaining tenants north of the road must have looked further afeld for their ironworking needs. 3.3 METALLURGICAL ANALYSIS OF SMITHING RESIDUES

The smithy was the subject of detailed metallurgical analysis, which is reported in detail by Gerry McDonnell and Alison Mills (online archive, Section 8.13). This work included the most systematic sampling of ironworking residues that had been undertaken on a medieval smithy and the site is unique in the British and European archaeological record. The conclusions of the analysis are worth summarising here. Over 96% of the slag recovered at Burton Dassett came from Area J and the areas immediately adjacent (H, I, W and K), with 65% of the slag (226.8 kg) from J itself. The greatest quantity of stratifed material on Area J came from early 15th-century contexts of the fnal phase. There was a slag dump just south of the building, with more from the foor of the outbuilding to the east. It is also likely that there was another slag dump inside the smithy against

185

the north wall, some of which had slumped off the remains of the wall when it was demolished. Little slag was recovered from the tenement ditches and it seems likely that much of the slag was either dumped outside the building or removed elsewhere for use as hardcore. Slags from the adjacent areas were mostly derived either from deliberate spreads of slag used as surfacing or from within demolition layers. A background level of very small levels of slag from early 13th-century contexts may hint at an earlier smithy within the early settlement core. The dense concentration of slag, from both excavation and feldwalking, in the vicinity of the smithy establishes a model for the distribution of slag across rural settlements. The pattern is similar to that observed at Wharram (McDonnell et al. 2012) and at a 14th/15th-century site at London Road, Crawley (Andrews 2001). The lack of any further concentration of smithing debris at Burton Dassett makes it unlikely that smithing continued at Southend after the abandonment of the building on Area J. The weight of hearth bottoms as a proportion of the total weight of all diagnostic smithing slag was 10%, similar to Wharram Percy, which is the only comparable rural site. The range of hearth bottom weights and sizes is typical, suggesting use of a large hearth that allowed the bottoms to develop unconstrained. Analysis of the smithing slags indicates that the Burton Dassett smiths were skilled craftsmen. The fuxing of the iron with sand was well controlled, with little metal being lost and just suffcient fux being added to absorb iron oxide scale, indicating effcient use of resources. Burton Dassett was the frst medieval site which was systematically sampled for hammerscale in Europe. Unlike slags, hammerscales are not taken away for use as hardcore and remain close to the smithy and primary slag dumps. The purpose of the sampling was to confrm the identity of the building as a smithy and to examine its distribution in order to identify domestic areas as well as the forge and working area, in addition to investigating the relationship between magnetic susceptibility of soil samples and hammerscale concentrations. Most of the hammerscale was fake hammerscale (dislodged from the surface of the iron worked in the forge) although spheroidal hammerscale (spheroidised liquid slag expelled from the hammering together of two pieces of iron when fre-welded) was present in most samples. This implies that the smiths undertook the full range of iron-smithing techniques, certainly hot forging and by assumption cold work too, together with f ire-welding which implies the manufacture of steel-edged tools. Proportions of hammerscale in the magnetic fraction were signifcantly higher from contexts within the smithy than from those outside it, which showed no obvious patterning. Contour maps of magnetic susceptibility and magnetic weight within

BU RTON DA S SET T SOU T H E N D

186

Figure 3.25 Hammerscale distribution within the smithy

the smithy are shown in Figure 3.25. They show accumulations of residue along wall lines and within depressions, which are not surprising, together with uniformly higher readings in the central northern part of the building and low readings within the central structure. This differentiation suggests that the surface from which the samples were taken represented the smithy foor. The generally central distribution of hammerscale would be consistent with a central hearth, probably constructed at waist level, which would have protected the foor area immediately below it from falling hammerscale. This hypothesis

is supported by the comparable results obtained from analysis of samples from a smithy foor at the Ryedale Folk Museum, Hutton-le-Hole, North Yorkshire, using the same methodology as at Burton Dassett. It is possible to suggest, therefore, a forge hearth in the centre of the northern side of the building directly over the area of low susceptibility, with a hearth to the left of this and the anvil positioned in the middle of the room in the area of highest susceptibility. Given that the overall distribution of slag, and thus the fall-off patterns, are understood, it is likely that the quantity of slag recovered (344kg) approaches

SPAC E A N D BU I L DI NG S

187

complete recovery. Since the forge was in operation for somewhere in the order of two centuries, it is possible to draw some tentative conclusions about the extent of activity. Some 90% of the slag consisted of smithing slag lumps, as opposed to hearth bottoms. This suggests annual generation of only c1.5 kg of slag. This might indicate that the smithy was a standard tenement building which was re-organised and used temporarily as a smithy for a relatively short period prior to abandonment of the tenements south of the road. The distribution of slag throughout the stratigraphy makes this unlikely. The conclusion may therefore be drawn that the smith occupied the site over a long period, but that the forge was only used part-time or intermittently, perhaps even seasonally. If this is the case, then the quantity of slag and the number of hearth bottoms seems to suggest that the hearth bottoms developed during one specifc operation. We cannot be entirely certain, however, that the quantity of slag recovered represents the greater part of the entire output of waste during the period of use of the smithy. Figure 3.26 shows estimates of smithing slag and hearth bottom weight per year against different periods of operation and different slag recovery rates. If the recovery of slag is much lower than seems to be the case and the smithy only operated for a single generation, then about 1kg of smithing slag would have been generated per week, with a hearth bottom being formed every six weeks. These data are unique in being the frst to provide estimates of slag production for a rural settlement and provide valuable comparanda for future investigations. Three models for rural smithing operations have been proposed on the basis of the Wharram Percy evidence (McDonnell et al. 2012). The frst is that of a self-suffcient smith producing iron and steel off-site and manufacturing artefacts for the local community in the village forge. The second imports iron and steel in bar form; while the third is essentially a recycling operation based on scrap metal. There are also differing levels of skill, from simple production of items such as nails and hooks, through the manufacture of edged tools, up to highly skilled

actual recovered SS HB kg kg 310 35 Smithy life SS (years) kg/year 25 12 50 6 100 3 150 2 200 2 250 1

HB

HB no. 59

if 75% recovered SS HB kg kg 413 47

HB

SS

kg/year

no./year

kg/year

1 1