The Medieval Peasant House in Midland England

By Nat Alcock and Dan Miles

The aim of this lavishly illustrated book is to provide an in-depth study of the many medieval peasant houses still standing in Midland villages, and of their historical context. In particular, the combination of tree-ring and radiocarbon dating, detailed architectural study and documentary research illuminates both their nature and their status. The results are brought together to provide a new and detailed view of the medieval peasant house, resolving the contradiction between the archaeological and architectural evidence, and illustrating how its social organisation developed in the period before we have extensive documentary evidence for the use of space within the house.

Nat Alcock and Dan Miles' work on Medieval Peasant Houses in Midland England has been nominated for the 2014 Current Archaeology Research Project of the Year.

• Language: English
• Publisher: Oxbow Books
• Release date: Apr 30, 2014
• ISBN: 9781782971177

Nat Alcock

Emeritus Reader in the Department of Chemistry, University of Warwick

Book preview

Published by

Oxbow Books, Oxford, UK

© Nat Alcock and Dan Miles 2013

ISBN 978-1-84217-506-4

eISBN 9781782971177

This book is available direct from

Oxbow Books, Oxford UK

and

The David Brown Book Company

PO Box 511, Oakville, CT 06779, USA

(Phone: [phone]; Fax: [phone])

or via our website

www.oxbowbooks.com

A CIP record for this book is available from the British Library

Library of Congress Cataloging-in-Publication Data

Alcock, N. W. (Nathaniel Warren)

   The medieval peasant house in Midland England/by Nat Alcock and Dan Miles; with contributions by John Chenevix Trench … [et al.]. -- 1st ed.

          p. cm.

     Includes bibliographical references and index.

     ISBN 978-1-84217-506-4

   1. Architecture, Medieval--England--Midlands. 2. Architecture, Domestic--England--Midlands. 3. Dwellings--England--Midlands. 4. Crucks--England--Midlands. 5. Archaeological dating--England--Midlands. I. Miles, D. W. H. II. Title.

    NA963.A445 2013

    728.09424’0902--dc23

2012041704

Front cover right: Phoenix Cottage, 1 Birmingham Road, Stoneleigh, Warwickshire (STO-F); top left: Mill Farm, Mapledurham, Oxfordshire (MDM-A); bottom left: Abel’s Cottage, 43 High Street, Long Crendon, Buckinghamshire (LON-A). Back cover top: Reconstructed elevation for Phoenix Cottage; bottom: Re-creation of the home of Robert Dene of Stoneleigh, Warwickshire.

To J. T. Smith

who compiled the first detailed cruck distribution map, laying the foundation for all our later work

Contents

Preface

This volume presents the results of a study of almost 120 medieval houses in the four Midland counties of Buckinghamshire, Leicestershire, Oxfordshire and Warwickshire (with a few in Gloucestershire and Nottinghamshire), using a combination of tree-ring dating (dendrochronology) and architectural survey, together with the documentary background to another twenty houses. Most of these buildings are of cruck construction – to the extent that the summary title ‘The Cruck-Dating Project’ has often been applied to the project as a whole.

The survey was initiated by Bob Laxton in collaboration with Nat Alcock, in succession to the successful Kent tree-ring dating project carried out by the Nottingham University Tree-ring Dating Laboratory (NUTRDL) from 1986–9. The primary evidence was obtained between 1988 and 1991 by a team consisting of Robert Howard of NUTRDL (dendrochronology) and Dan Miles (architectural survey) under the direction of Nat (N. W.) Alcock (co-ordinating the architectural and documentary study), Bob (R. R.) Laxton and Cliff (C. D.) Litton (tree-ring dating). Most of the plans and sections were measured by Dan Miles during the main phase of the project, but some have come from external sources or from later work and the conventions employed sometimes vary, for example in the inclusion or omission of roof materials. We are extremely grateful to Bob Meeson for redrawing the more than one hundred sections to a uniform scale and style.

Although frustrating for the contributors, the delay in completing this book has had its advantages, through the opportunity both to include additional buildings, and to undertake further research on those already studied. Work after the formal end of the project has included the recording and tree-ring dating by Dan Miles of other medieval houses allowing us to include a dozen extra houses, the updating of sapwood estimates and more detailed documentary research. Very recently, also, it has been possible to apply high-precision radiocarbon dating to some buildings. As a particular example of the positive aspects of this extended period of research, fieldwork by Eric Sewell in Long Crendon, the village already notable for having more cruck buildings than anywhere else in England, has, astonishingly, identified three more cruck-built houses there. Similarly, documentary research on the social background to the medieval houses in Steventon has been transformed by the discovery of extensive hitherto unknown documentary sources.

Apart from sections for which individual authors are named, Nat Alcock has been primarily responsible for chapters 3 and 4 on Planning and Structure, and Dan Miles for chapter 5 on Carpentry. We have worked jointly on the other chapters. We thank the various contributors for their unstinting efforts, and for their patience in waiting for this publication to appear. We are especially grateful to Dr Christopher Currie for making available his research, both published and unpublished, on houses in the Vale of the White Horse, and in discussing the buildings there with us.

The project was funded by the Leverhulme Trust, without whose support it could not have been undertaken. The radiocarbon dating was funded by the Society of Antiquaries (one site), the Vernacular Architecture Group (four sites) and the Oxfordshire Building Record (one site). Generous grants in support of the publication have been made by the Aurelius Charitable Trust. Sadly, during the long period that it has taken for this book to be completed, Bob Laxton, one of the original project directors, has died, as has John Chenevix-Trench, who undertook the initial documentary research on Long Crendon. We hope that they would have approved of the final results.

We thank everyone who has assisted the project, and in particular those who have allowed us free access to record and date their houses. Without their co-operation, the study would have been impossible. Several people provided substantial help with the surveying, notably Catherine Murray (Buckinghamshire), Nick Hill (Leicestershire and Rutland), and members of the Oxfordshire Buildings Record (Oxfordshire, led by David Clark). The archivists at the many repositories where documents have been consulted have been unfailingly helpful.

Despite the long wait between our original survey work and the final appearance of this volume, we hope that everyone interested in the subject of medieval houses will consider the results worth waiting for.

Nat Alcock

Dan Miles

List of Figures and Tables

Part I

Figures

Cross-section drawings of cruck, base-cruck, aisled and box-frame trusses at 1:100 scale

Tables

Part II

Figures

Tables

Appendices: Figures

Appendices: Tables

Picture Credits:

We thank the following organisations and individuals for permission to reproduce figures. Copyright remains with the original owners. Centre for Buckinghamshire Studies (6.3; 8.4.4, 8.5.10); The Master, Fellows and Scholars of Christ’s College, Cambridge (6.15); English Heritage (8.1.1); J.J. Eyston (8.6.11); Neil Finn (8.8.1; 8.8.4); Nick Hill (5.11); The National Archives (8.87); Eric Sewell (8.4.1, 8.5.1); Shakespeare Centre Library and Archive (6.12, 8.11.5); Steventon Parish Council (8.10c.1); John Walker (8.5.4). Acknowledgements for images reproduced in the reports are included in the individual figure captions.

Coastline and boundaries on maps are based on data provided through EDINA UKBORDERS with the support of the ESRC and JISC and uses boundary material which is copyright of the Crown and the ED-LINE Consortium.

Acknowledgements for images reproduced in the reports are included in the individual figure captions.

PART I

Analysis of the Medieval Peasant House

CONVENTIONS

Labelling of features

In descriptions and reports and on plans, bays are labelled as I, II, etc, and trusses as T1, T2, etc.

Dates

Dates cited in the text are given in forms that are intended to indicate their source, their precision and their reliability. A suffix ‘d’ indicating a tree-ring date is occasionally used when this is not self-evident. In the original reports and the dates as published in Vernacular Architecture, central felling dates within the date ranges were given but, following current practice, only the felling date ranges are used.

The following conventions are used:

Sapwood estimates

Felling date ranges have been revised from those originally presented, using the best currently available sapwood estimates. Most of the ranges have also been refined using OxCal. Both the original and revised estimates are given in the individual reports. The following 95% probability estimates are used

The modal dates are used for plotting and statistical analysis but are not cited in the text, since they do not provide precise dates for buildings. When no sapwood survives and only a terminus post quem (tpq) date is known, the modal date is taken to be 25 years after the tpq date.

Dendrochronology Abbreviations

FMR = first measured ring

LHR = last heartwood ring

LMR = last measured (complete) ring

NM = not measured

Dat Cat = Dating Category (see p. xxx)

Units and conversion factors

Dimensions are given in Imperial measure throughout, since these are the units that the carpenters would have used originally. Thus, any standard dimensions or other regularities will be most obvious when expressed in these units. For reasons of space and readability, the metric equivalents of individual dimensions are not included in the text, but can be calculated using the following factors:

The yardland (virgate in Latin) was a conventional measure of the size of an open-field holding, which varied from place to place (and to some extent from holding to holding). At Long Crendon, Buckinghamshire, a yardland contained 30 customary acres, corresponding to 25 statute acres; at Steventon, Oxfordshire, it contained 24 acres; at Stoneleigh, Warwickshire (1597), the yardland was calculated to contain 28 acres on average, and at Diseworth, Leicestershire (1797) each contained 24 acres, although conventionally they were reputed to be of 18 acres.

1

Introduction

T[homas?] Hopkyns is required to build a house of five pairs of forks (edificare super unum toftum unam domum (de v per’ forkes) within two years [at Long Itchington, Warwickshire] (1346–7, Bodleian Library, Ms Trinity 84, p. 65).

When our project on the Medieval Peasant House began, the long-held view of historians and archaeologists was that houses such as that to be built by Thomas Hopkyns at Long Itchington would have lasted for no more than a generation.¹ However, the investigation of documentary sources was undermining these views,² as was evidence that most of the several hundred cruck houses in the Midlands were of medieval date and, simply from their numbers, they could not all be of superior social status. Writing in 2011, the position is reversed. The new orthodoxy is that peasant houses were substantial, and built to last – and that ‘later medieval houses survive in their thousands’.³ However, less attention has been paid to the details of these houses, or to their social status, which may not be as uniform as the label peasant might suggest. Some types of medieval houses, such as the Wealden buildings of Kent, are linked to an emerging group of prosperous yeomen.⁴ In contrast, many of smaller medieval houses in the Midlands, in particular those of cruck construction, can be associated with people of much more modest status. Both these groups, however, fall within the overall definition of peasant within late medieval English society, as the mass of village householders.⁵ They merge at one end into the landless labourers who might occupy their own houses or sometimes live with the masters they served, and at the other into the substantial villagers who had perhaps amassed several originally separate holdings and were farming on a considerable scale. They would normally hold their house and land by customary or copyhold tenure, although a few were free tenants. However, this tenurial distinction was of less importance than the nature and extent of the agricultural and other resources available to them.

Our aim is to provide an in-depth study of the many medieval peasant houses still standing in Midland villages, and of their historical context. In particular, the combination of tree-ring and radiocarbon dating, detailed architectural study and documentary research illuminates both their nature and their status. Within the volume, this introduction examines the strands of evidence used and sets out the detailed strategy adopted in the project, with the study of cruck-built houses at its heart. The four central chapters present the primary results. Chapter 2 examines the dating results in relation to their geographical and chronological range. Chapters 3, 4 and 5 present an examination of the buildings surveyed, correlated with the dating evidence, looking at the developments in planning, structure and carpentry from the thirteenth to the sixteenth centuries. Chapter 6 examines the documentary evidence for medieval peasant houses, both in general terms and in case studies of villages for which exceptionally good documentary and architectural evidence survives. A short additional section examines the social background to the base-cruck houses included in the survey. In Chapter 7, the results are brought together, demonstrating how our work has provided a new and detailed view of the medieval peasant house, resolving the contradiction between the archaeological and architectural evidence. We hope here to show how its social organisation developed in the period before we have extensive documentary evidence for the use of space within the house.

Technical details of the dating methodology are given in Appendix 1, and Appendix 2 provides a complete tabulation of the buildings studied with the codes assigned to each (which are used for reference in the main text). It also includes a list of the buildings rejected on preliminary examination. The dates obtained are listed in Chapter 2.

The heart of an architectural study lies in the individual buildings examined, even more than in the generalisations derived from them. The main text is extensively illustrated with sections and plans of the houses studied; their interpretation, however, requires a close analysis of the buildings, especially in relation to the reconstruction of missing portions or the evidence for such features as smoke louvres. Descriptions of each of the 117 buildings surveyed, accompanied by plans, drawings and photographs, are included on the accompanying CD-ROM. In addition, reports for a further 21 buildings have been compiled, which focus on their documentary history and give only brief details of the buildings (generally based on earlier work). A small selection of these reports are printed in Part II. These include both typical peasant houses and also the most remarkable buildings discovered: LON-G, an aisled hall built in 1205; AST-A, a box-framed chamber block of 1282–6; MDM-A, a virtually complete three-bay cruck house of 1335.

The Evidence for Peasant Houses

Three strands of evidence contribute to our knowledge of medieval peasant housing. Each provides specific insights and each also has its specific weaknesses.

Documents

Documentary sources give good evidence of the dates at which some houses were being constructed. They are concerned to specify the structural character and size, indicating clearly that they were regarded as valuable assets. However, we cannot generally demonstrate that the particular houses referred to might have survived to the present day. The extensive sources, such as probate inventories and surveys, that illuminate post-medieval houses,⁶ are generally non-existent or very sparse until after the Middle Ages. A few exceptions have been discovered, including Stoneleigh, Warwickshire, for which inventories survive from as early as 1537 and regularly include room names; the earliest of these inventories can be seen as describing medieval houses still furnished in medieval style; such other early inventories as have been studied only rarely name the rooms.

Caution is needed in relating the documentary evidence directly to surviving buildings, because it seems clear that by the fifteenth century rural society in some areas was undergoing differentiation, with the emergence below the manorial level of wealthier sub-classes within village society. These groups are identifiable, for example, as the builders of the elegant Wealden houses in Kent.⁷ This trend seems to be associated more with pastoral than arable farming regions and might therefore be relatively unimportant in the Midland counties. Significant late medieval changes have been identified in the sizes of village holdings, but the distribution of land remained generally even (but see the discussion of Oxfordshire in Chapter 6).

The direct identification of the original status for individual standing medieval village houses is exceptionally difficult, because evidence linking post-medieval village holdings with their medieval counterparts is very sparse. However, when the pattern of land-holdings within the village was uniform, it is not necessary to relate a particular house to a particular medieval holding, to demonstrate that it must have belonged to, say, a yardland (30 acre) farm. The documentary case studies focus on places where we can either make links of this type or can find information about the status of individual houses.

Excavation

Excavation can be expected to date house sites reasonably closely and to provide good evidence for their size and room layout, and for the overall pattern of buildings within the croft, but it is unable to identify their associated landholding. The inevitable hiatus caused by later destruction mean that a complete picture is disappointingly rare. More important in relation to the permanence or impermanence of peasant building, is the difficulty of identifying the character of the standing structure from the below-ground evidence. This is exemplified by the reinterpretation of the excavations at Wharram Percy, from single-generation insubstantial structures to long-lived cruck houses, identical in form to those still standing in neighbouring villages.⁸ Thus, archaeological results are of only moderate value in establishing the detailed nature of peasant houses. Rather, the evidence of standing buildings informs the interpretation of excavation results. Partly for this reason and partly because a major review of excavated medieval houses is close to publication,⁹ the evidence of excavation finds only a limited place in the present volume.

Standing Buildings

The third strand of evidence for medieval peasant houses, that of standing buildings, has often only contributed to the discussion in rather general terms. To identify an actual medieval peasant house, three questions must be answered. Is it medieval in date and, if possible, precisely when was it built? Secondly, what was its original form? Thirdly, was it of peasant status? The answers to these three questions for our surveyed houses are given in Chapters 2, 3–5 and 6 respectively.

Research in vernacular architecture has gradually led to the recognition of surviving small medieval houses, by working back from the well-established structural and plan features of sixteenth or seventeenth century houses.¹⁰ The latter can easily be recognised, since from around 1575, many carry inscribed dates.¹¹ In particular, the literary evidence for improvement from the later sixteenth century onwards has been confirmed.¹² The upper floors and chimneys reported as new features in this period are amply exemplified in standing buildings.

Houses that are more primitive than these sixteenth or seventeenth century examples must, by elimination, be medieval. They share in particular the touchstone of the medieval house, the open hall: the main living room originally without an upper floor or a chimney, recognisable by the soot from its central hearth encrusting the roof timbers. Such houses have now been identified in very considerable numbers over the whole of southern England.¹³ In some areas, notably Kent, Sussex and much of East Anglia, the medieval houses have sufficient distinctive details (in such features as their roof structure, crown posts, beam mouldings and bracing patterns), that a reasonably precise typology can be established, linked to buildings of known date. As noted above, the quality of many of the houses in these regions suggests that their builders belonged to a relatively wealthy sub-class within the village community. Smaller and simpler medieval houses have been recorded there, but they are too rare to provide a detailed picture of the houses of more modest peasants.¹⁴

In the Midlands, the medieval houses generally lack such detailing. They are dominated by one structural form, the cruck, in which the roof and walls are supported by paired timbers (cruck blades), reaching from ground to apex in a single sweep. A typical cruck house has three bays (structural units), with four cruck trusses (Fig. 1.1). Most often, the central bay was the open hall, with the roof timbers sooted by the smoke rising from a hearth on the floor; one end bay had an original upper floor, providing sleeping rooms, and the other end was used as service space.

In Fig. 1.2A, the components of a cruck truss are drawn and labelled. Some Midlands box-frame houses (Fig. 1.2B) have also been identified as late medieval as, for example, Cuttle Pool Farm, Knowle, Warwickshire, dated to 1478/9.¹⁵ This structural form becomes dominant in the region after crucks passed out of use, but the present survey and other recent studies have shown that box-framing formed a more significant component of the medieval carpenter’s repertoire than expected, appearing as early as the late thirteenth century in chamber blocks (see Chapters 3 to 5). Surprisingly, perhaps, the earliest surviving houses in the region are not cruck-built, but are aisled (Fig. 1.2C), with free-standing posts set within the walls, or use the more substantial base crucks. These have large curved cruck-like blades that rise only to the tiebeam. They often have aisled end or spere (screen) trusses and use square-set arcade plates rather than the diagonally-set purlins found with crucks (Fig. 1.2D).

Fig. 1.1. A typical Midlands cruck house. Redrawn from Alcock, Meeson et al. (1996) Recording timber-framed buildings: an illustrated glossary.

Key to components (* for more than one label):

Fig. 1.2. Typical timber-framed trusses: (A) a cruck truss; (B) a box-framed truss with clasped purlin roof; (C) an aisled truss; (D) a base-cruck truss with crown-post upper roof and, beyond it a spere truss of aisled form. Drawings by Bob Meeson, based on Alcock, Meeson et al. (1996) Recording Timber-Framed Buildings.

Despite these alternative forms, the dominance of cruck construction in the southern half of Britain is amply demonstrated by the distribution map of cruck houses (Fig. 1.3), and by the absence in the Midlands of similar numbers of box-framed houses. In midland and southern England, the great majority of the crucks are undoubtedly medieval, demonstrated both by their tree-ring dates and their architectural features. Furthermore, the sheer number of cruck houses confirms that they must be peasant houses, rather than being of manorial status or restricted to the wealthiest of villagers. When a village has ten or even twenty such houses, it is a safe deduction that they were the homes of ordinary people including, as our study indicates, the whole hierarchy of rural society, from substantial and middling peasants down to a few smallholders (though not equally likely to survive for all these groups). The status is less obvious for the aisled and the substantial base-cruck houses, for which the evidence is considered in Chapter 6. This map also shows how the cruck distribution has a sharply defined eastern boundary, with crucks being completely absent from the eastern third of England; the cause of this boundary has been an enigma since it was first recognised, but its further examination lies beyond the scope of the present study.¹⁶

Understanding the original form of these houses requires detailed structural analysis, working back from the existing building, mentally stripping away later modernisation and interpreting minute details to re-create the form of the medieval house. This is undertaken, house-by-house, in the individual reports, and an overview of their characteristics is given in chapters 3–5, and summarised in chapter 7.

A particularly significant aspect of the dominance of cruck construction in the region relates to the interpretation of the documentary evidence. Almost invariably, when they give information about the structure itself, they are described as composed of furcae, either individually or in pairs, a term which has long been translated as cruck.¹⁷ Although this meaning has recently been challenged, the suggested alternatives are not convincing. In particular, if this standard term, used in the region where crucks are dominant (and not generally elsewhere), had a different significance, we are left without a plausible word for cruck.¹⁸ Thus, we can use the documentary evidence to support that obtained from standing buildings for the dating and use of cruck construction

Project Strategy

The objective of this project has been to study a considerable sample of what were expected to be medieval peasant houses, applying structural recording and analysis, dendrochronology, and documentary research.¹⁹ Within this overall aim, strategic choices inevitably had to be made. The first was to concentrate on cruck houses. This decision was primarily based on the belief that they dominated the surviving early buildings in the region, and that most such houses from the Midlands southwards were of medieval date. Furthermore, the continuing interest in this form of construction had led to the compilation of a comprehensive list of cruck buildings, providing a ready-made database of examples for study.²⁰ Although some small medieval box-framed buildings (Fig. 1.2B) survive within the region dominated by cruck construction (especially as wings associated with cruck halls), they had not previously been the subject of detailed analysis. However, a number of examples have been included in the study and it has been possible to re-evaluate their significance as components of the regional medieval building tradition.

The project area

The second choice was of the areas to be examined. The distribution map of crucks (Fig. 1.3) shows notable concentrations in north-east and west Yorkshire, along the Severn (Herefordshire, Worcestershire and Shropshire), in south-east and central Wales, and in south Oxfordshire. For the study of medieval peasant houses, the northern areas were rejected because it appeared that almost all the standing buildings were post-medieval, although some evidence suggested that crucks might have been reused from earlier houses.²¹ In the Severn-side counties and Wales, the cruck houses are believed to be medieval, but many of them are very elegant, with strikingly decorative carpentry. It was unclear whether they were the houses of typical medieval peasants or belonged to an elite group of villagers.²²

In the Midlands, the cruck houses in Warwickshire and Leicestershire seemed to be the most ‘ordinary’, often lacking any distinction in their construction. Here, surely, might be found the homes of typical medieval peasants. Detailed study of buildings immediately to the south, in the limestone regions of Oxfordshire and Northamptonshire, seemed less likely to be rewarding. Few crucks had been recorded in this area and these appeared generally to be fragmentary survivals after an extensive sixteenth and seventeenth century rebuilding. However, many cruck houses were known in the southern half of Oxfordshire and the adjoining county of Buckinghamshire, and the area included several ‘cruck villages’ (those containing half-a-dozen or more crucks).²³ Notably early dates had been suggested for some of the Oxfordshire examples, for which confirmation by dendrochronology would be particularly valuable.²⁴ Thus, our survey has concentrated on these four counties (including also that part of West Midlands formerly in Warwickshire) (Fig. 1.4).²⁵ The northern two counties are representative of the grouping of crucks from the Cotswolds northwards, the southern ones of the cruck concentration in the Chilterns and the Vale of White Horse.²⁶ The survey might have been extended to the south, into the rest of Berkshire (from which that part of the study area in south Oxfordshire had been removed in 1974), Wiltshire and Hampshire, but this seemed likely to spread out the buildings that could be recorded too thinly, while proportionately fewer crucks are known there than in the chosen counties.²⁷

Fig. 1.3. Distribution of cruck houses in England and Wales. A total of 3086 examples are plotted. (Updated from the map published in Alcock, ‘Distribution of crucks’, to include newly identified examples.)