Early Medieval Britain: The Rebirth of Towns in the Post-Roman West 0521885949, 9780521885942

The growth and development of towns and urbanism in the pre-modern world has been of interest to archaeologists since th

950 221 12MB

English Pages XVIII+228 [250] Year 2018

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Early Medieval Britain: The Rebirth of Towns in the Post-Roman West
 0521885949,  9780521885942

Table of contents :
List of Figures ix
List of Tables xii
Preface xiii
Introduction 1
Who Were the Anglo-Saxons? 1
An Outsider's View of Anglo-Saxon Urban Origins 3
Archaeology and Urban Origins 5
Theoretical Perspectives 6
A Processual Plus Perspective 9
Models of Medieval Urbanism 12
The Organization of This Volume 15
1. The End of Urbanism in Roman Britain 18
Introduction 18
The Archaeological Evidence 27
Winchester 27
London 30
Wroxeter 32
Other Roman Towns 36
Recent Approaches to Late Roman Urbanism in Britain 43
Box: Oppida 44
Conclusions 46
2. Early Anglo-Saxon England: Settlement, Society, and Culture 50
Historical Background 50
Chronological Issues 55
The Archaeology of Early Anglo-Saxon Settlement 57
Settlement Changes in the Late 6th and 7th Centuries 71
Western Britain in the 5th–7th Centuries 83
3. Middle Saxon Settlement and the Rise of the Emporia: The Archaeology of the Wics and Contemporary Sites 86
Introduction 86
Background 88
Middle Saxon Changes in Settlement Patterns 89
The "7th-Century Shuffle" and the Establishment of New Estate Centers 90
Archaeology of the Wic Sites or Emporia 97
Ipswich 98
Hamwic (Middle Saxon Southampton) 106
History of the Excavations 106
The Archaeological Evidence 108
Eorforwic (Anglian York) 114
History of the Excavations 114
Archaeology at Anglian York 116
Lundenwic 121
History of the Excavations 121
Anglo-Saxon Lundenwic 123
Other Possible Emporia 128
Conclusions Regarding the Wic Sites 128
The End of the Emporia 134
Other Possible Middle Anglo-Saxon Towns: 8th-Century Mercian Burhs 136
4. Towns in Late Anglo-Saxon England 138
Background and Introduction 138
Late Anglo-Saxon Winchester and a Model for Planned Towns in Wessex 141
Box: Geophysical Survey in Archaeology 150
London: From Wic to Burh 157
Yorvik: Anglo-Scandinavian York 160
Viking Dublin 166
Late Saxon Ipswich 168
Norwich 169
Thetford 171
The Pattern of Urban Development in Late Saxon England 171
Conclusions 178
The Rebirth of Towns in Anglo-Saxon England 178
Examining the Urban Revolution in Early Medieval Britain 182
Urbanism in Later Prehistoric Europe 185
Comparisons to the Hawaiian Case: State Formation without Urbanism 188
Final Thoughts 191
Bibliography 193
Index 217

Citation preview

Early Medieval Britain

The growth and development of towns and urbanism in the premodern world has been of interest to archaeologists since the 19th century. Much of the early archaeological research on urban origins focused on regions such as Mesopotamia, Egypt, and Mesoamerica. Intensive archaeological research that has been conducted since the 1960s, much of it as a result of urban redevelopment, has shed new light on the development of towns in Anglo-Saxon England. In this book, Pam J. Crabtree uses up-to-date archaeological data to explore urban origins in early medieval Britain. She argues that many Roman towns remained important places on the landscape, despite losing most of their urban character by the 5th century. Beginning with the decline of towns in the 4th and 5th centuries, Crabtree then details the origins and development of towns in Britain from the 7th century through the Norman Conquest in the mid-11th century. She also sets the development of early medieval urbanism in Britain within a broader, comparative framework. Pam J. Crabtree is a Professor of Anthropology at New York University. She received her BA from Barnard College and her PhD from the University of Pennsylvania. Crabtree has been actively involved in medieval archaeology since 1971, and her research has been funded by Fulbright grants, the National Science Foundation, and the National Endowment of the Humanities. She is co-editor (with Peter Bogucki) of European Archaeology as Anthropology (2017). In addition to her work on Anglo-Saxon England, Crabtree has taken part in archaeological projects in Ireland, Belgium, France, Germany, Israel, Turkey, Ukraine, Armenia, Egypt, India, and historic sites in the US.

Case Studies in Early Societies Series Editor Rita P. Wright, New York University This series aims to introduce students to early societies that have been the subject of sustained archaeological research. Each study is also designed to demonstrate a contemporary method of archaeological analysis in action, and the authors are all specialists currently engaged in field research. The books have been planned to cover many of the same fundamental issues. Tracing long-term developments, and describing and analyzing a discrete segment in the prehistory or history of a region, they represent an invaluable tool for comparative analysis. Clear, well organized, authoritative, and succinct, the case studies are an important resource for students, and for scholars in related fields, such as anthropology, ethnohistory, history, and political science. They also offer the general reader accessible introductions to important archaeological sites.

Other titles in the series include: Ancient Mesopotamia Susan Pollock Ancient Oaxaca Richard E. Blanton, Gary M. Feinman, Stephen A. Kowalewski, Linda M. Nicholas Ancient Maya Arthur Demarest Ancient Jomon of Japan Junko Habu Ancient Puebloan Southwest John Kantner Ancient Cahokia and the Mississippians Timothy R. Pauketat Ancient Middle Niger Rod McIntosh Ancient Egyptian Civilization Robert Wenke Ancient Tiwanaku John Janusek The Ancient Indus Rita P. Wright Ancient Inca Alan L. Kolata Ancient Central China Rowan K. Flad, Pochan Chen The Colonial Caribbean James A. Delle Ancient Teotihuacan George L. Cowgill

Early Medieval Britain The Rebirth of Towns in the Post-Roman West Pam J. Crabtree New York University

University Printing House, Cambridge CB2 8BS, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 79 Anson Road, #06-04/06, Singapore 079906 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9780521885942 DOI: 10.1017/9781139017121 © Cambridge University Press 2018 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2018 Printed in the United States of America by Sheridan Books, Inc. A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data Names: Crabtree, Pam J., author. Title: Early medieval Britain: the rebirth of towns in the post-Roman West / Pam J. Crabtree, New York University. Description: New York: Cambridge University Press, [2018] | Series: Case studies in early societies | Includes bibliographical references and index. Identifiers: LCCN 2018003786 | ISBN 9780521885942 (hardback) Subjects: LCSH: Cities and towns – Great Britain – History – Medieval period, 1066–1485. Classification: LCC HT133.C713 2018 | DDC 307.760941/02–dc23 LC record available at https://lccn.loc.gov/2018003786 ISBN 978-0-521-88594-2 Hardback ISBN 978-0-521-71370-2 Paperback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

For Howard Dalton Winters

Contents

List of Figures List of Tables Preface

page ix xii xiii

Introduction

1 1 3 5 6 9 12 15

Who Were the Anglo-Saxons? An Outsider’s View of Anglo-Saxon Urban Origins Archaeology and Urban Origins Theoretical Perspectives A Processual Plus Perspective Models of Medieval Urbanism The Organization of This Volume

1

The End of Urbanism in Roman Britain Introduction The Archaeological Evidence Winchester London Wroxeter Other Roman Towns

Recent Approaches to Late Roman Urbanism in Britain Box: Oppida

Conclusions

2

Early Anglo-Saxon England: Settlement, Society, and Culture Historical Background Chronological Issues The Archaeology of Early Anglo-Saxon Settlement Settlement Changes in the Late 6th and 7th Centuries Western Britain in the 5th–7th Centuries

3

Middle Saxon Settlement and the Rise of the Emporia: The Archaeology of the Wics and Contemporary Sites Introduction

vii

18 18 27 27 30 32 36 43 44 46 50 50 55 57 71 83 86 86

viii

Contents Background Middle Saxon Changes in Settlement Patterns The “7th-Century Shuffle” and the Establishment of New Estate Centers Archaeology of the Wic Sites or Emporia Ipswich Hamwic (Middle Saxon Southampton) History of the Excavations The Archaeological Evidence Eorforwic (Anglian York) History of the Excavations Archaeology at Anglian York Lundenwic History of the Excavations Anglo-Saxon Lundenwic Other Possible Emporia Conclusions Regarding the Wic Sites

The End of the Emporia Other Possible Middle Anglo-Saxon Towns: 8th-Century Mercian Burhs

4

Towns in Late Anglo-Saxon England Background and Introduction Late Anglo-Saxon Winchester and a Model for Planned Towns in Wessex Box: Geophysical Survey in Archaeology

London: From Wic to Burh Yorvik: Anglo-Scandinavian York Viking Dublin Late Saxon Ipswich Norwich Thetford The Pattern of Urban Development in Late Saxon England

Conclusions The Rebirth of Towns in Anglo-Saxon England Examining the Urban Revolution in Early Medieval Britain Urbanism in Later Prehistoric Europe Comparisons to the Hawaiian Case: State Formation without Urbanism Final Thoughts

Bibliography Index

88 89 90 97 98 106 106 108 114 114 116 121 121 123 128 128 134 136 138 138 141 150 157 160 166 168 169 171 171 178 178 182 185 188 191 193 217

Figures

1.1

Map of Britain showing the locations of the Romano-British sites discussed in this chapter. page 19 1.2 The Packard Automobile plant in Detroit. The last car left its assembly plant in 1956. 24 1.3 Many of the unoccupied parts of Detroit have returned to prairie. 26 1.4 Photograph showing a Roman street at the Brook Street site in Winchester in 1971. The Late Saxon and modern streets are located to the left (east) of the Roman one. 28 1.5 Wroxeter “old work.” Portions of the support for the hypocaust of the Roman baths can be seen in the foreground. 33 1.6 Reconstruction of the mid- to late 6th-century timber buildings at Wroxeter. 34 1.7 Map showing the location of the Pagan cremation cemeteries surrounding sub-Roman Lincoln. 37 1.8 Photograph of the ruins of St. Martin’s Church at the deserted medieval village of Wharram Percy in North Yorkshire. 41 1.9 Map showing the distribution of the Late Iron Age oppida in temperate Europe. 45 1.10 Reconstruction of sub-Roman Canterbury (from a painting at the Canterbury Roman Museum). 47 1.11 A reconstruction of the South Gate of Roman Silchester. Portions of this gate survive to this day. 49 2.1 Map of southern Britain showing the locations of the Anglo-Saxon sites discussed in this chapter. 58 2.2 Plan of West Stow, redrawn. 59 2.3 Reconstructed sunken-featured building at West Stow County Park. The program of reconstruction began in 1974. 60 2.4 Plan of the site of Mucking, Essex. 63 ix

x

2.5 2.6 2.7 2.8 2.9 2.10

2.11 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12

3.13 3.14 4.1 4.2 4.3 4.4 4.5

Figures

Anglo-Saxon polities at the time of the Tribal Hidage, probably in the late 7th century. Plan of Yeavering in the 7th century. Reconstruction of structure C12 at Cowdery’s Down. Image of Sutton Hoo Mound 1 as it exists today. Photograph of the reconstruction of the grave from Mound 1, Sutton Hoo (from the Sutton Hoo Museum) Photograph of artifacts recovered from Rendlesham, a royal estate center that is probably associated with the cemetery at Sutton Hoo. Reconstruction of the Prittlewell, Essex chamber burial. Map of Britain showing the location of the sites discussed in this chapter. Plan of Staunch Meadow, Brandon, Suffolk. Plan of Middle Saxon Wicken Bonhunt, Essex. Plan of Ipswich excavations showing the location of sites within the town. Image of Ipswich Ware pottery jar. Map of England showing locations of Ipswich Ware finds. Map of Southampton showing location of the Hamwic in relation to Roman and medieval Southampton. Plan of the excavations at St. Mary’s Stadium, Hamwic, Southampton. Map of Roman York. Photograph of the Multangular tower, York. Map of York showing location of Eorforwic and the Fishergate excavations. Map of London showing location of the Royal Opera House (2), Julibee Hall (3), and St. Martin’s courtyard (1) sites. Reconstruction of Middle Saxon Lundenwic. Image of a sword hilt from the Staffordshire Hoard. Map showing the location of the sites discussed in this chapter. Street plan of Winchester showing the Roman and Late Saxon streets. Plan of Wareham. Plan of Cricklade. A magnetometer in use. Readings are taken at 50 cm intervals.

70 72 74 76 77

79 80 87 92 94 100 101 103 107 109 116 117 119

124 125 130 140 143 147 149 150

Figures

4.6 An example of a resistivity survey. A current is passed through the outer two probes, and the inner probes measure the degree of resistivity. 4.7 Ground-penetrating radar in use. 4.8 Plan of Anglo-Scandinavian York showing the location of the Coppergate site. 4.9 Reconstruction of Viking Dublin.

xi

151 152 161 167

Tables

1 Basic chronology of Roman and Anglo-Saxon England 3.1 Major sites discussed in Chapter 3 3.2 Ceramic chronology for Anglo-Saxon and early medieval Ipswich

xii

page 2 91 99

Preface

My interest in Anglo-Saxon England goes back to my undergraduate days. I was an art history major at Barnard with a special interest in early medieval art, but when I was a junior I had the opportunity to work at Winchester, and it changed my life. I  worked at the Brook Street site back in the summer of 1971. This was the final season at Brook Street, and the final season of the ten-year program of excavation at Winchester under the leadership of Martin Biddle. I was fortunate to be part of the excavations when we were excavating the earliest of the Anglo-Saxon features and coming down into the Roman features, including a Roman road. I spent part of my summer excavating an Anglo-Saxon wicker-lined well and a fence line. Friends of mine excavated the old Roman road surface, which produced boot nails and low-denomination base-metal Roman coins. I was hooked on archaeology for life. Winchester was the training ground for an entire generation of archaeologists in the UK and North America, and I will be eternally grateful for the opportunity to be part of it. Thank you, Professor Emeritus Martin Biddle, CBE. I returned to Barnard in the fall of 1971 and began taking courses in anthropology and anthropological archaeology, since most US archaeology programs are housed in departments of anthropology. I also looked around for PhD programs in anthropological archaeology that included medieval archaeology. At that time, the only person teaching medieval archaeology within an anthropology department was Professor Bernard Wailes at the University of Pennsylvania. I will be eternally grateful to Bernard Wailes for taking a chance on a medieval art historian with some basic statistics and Latin skills. Bernard also offered me the opportunity to excavate at the Iron Age royal site of Dún Ailinne, Co. Kildare, Ireland. I have been involved with Dún Ailinne throughout my entire career, and we just completed an excavation season at Dún Ailinne in the summer of 2016. I wish that Bernard were still with us, and I want to thank him for being my PhD adviser and my friend. When I  entered the University of Pennsylvania, I  was interested in Anglo-Saxon settlement patterns (or, as we might talk about them today, xiii

xiv

Preface

landscape studies). I was specifically interested in the question of possible Roman-to-Saxon continuities on the rural level. Unfortunately, I was about 35 years too early for the kinds of GIS tools that are available to archaeologists today. I became a zooarchaeologist by accident. As part of the first year core program in anthropological archaeology, I took the human osteology lab with Professor Alan Mann. As someone who trained in art history, the whole process terrified me, but as a result of my art history training, I have a pretty good visual memory. I did surprisingly well in the osteology lab, and I went on to take another course in human osteology and to take faunal analysis with Dr. Dexter Perkins, Jr. Dexter is best known for his work on possible early sheep domestication at Zawi Chemi Shandiar in Iraqi Kurdistan (Perkins 1964), and he and his partner, Pat Daly, regularly taught zooarchaeology at both Penn and Columbia University. As a result of the zooarchaeology course, I realized that zooarchaeology provided a way to look at early medieval landscape use in England. Dr. Perkins was a member of my PhD committee, and he passed away shortly after I received my PhD. I will always be grateful to him for giving me my start in zooarchaeology. While I was an early PhD student in archaeology, I spent two summers working at the excavations at York under the direction of P. V. Addyman and the York Archaeological Trust. I  spent the first season excavating, sorting medieval pottery, and drawing a few human skeletons. (Thank you, Alan Mann, for teaching me human osteology.) In 1974, I  spent part of the summer in the bone lab. I  am so impressed with what has happened in York in the past 40 years, and I am delighted to see that the Historic Towns Atlas for York has recently been published (Addyman 2015). When I began looking for a PhD research topic, I was told that the animal bones from the Early Anglo-Saxon settlement at West Stow were being studied by another researcher. I did a lot of reading on Late Iron Age and Roman sites in Britain, and returned to the UK in 1976 to see whether there might be a collection or series of collections available for study. I met Roger Jones of the old Ancient Monuments Laboratory in London and Jennie Coy, who had recently been hired as a research fellow in zooarchaeology at the University of Southampton. They told me that there were a number of Late Iron Age collections available, and they also had an enormous collection from West Stow. As it turned out, the person who had initially agreed to work on the faunal material from West Stow could not carry out the project, and the excavator, Stanley West, was anxious to have the material analyzed. I met with Stanley West in the Suffolk Archaeology Unit, and he was agreeable, so I  returned to the United States to write grant proposals.

Preface

xv

I returned to the UK a year later. I  was fortunate to receive a Fulbright-Hayes grant, which allowed me to be an overseas visiting student at Southampton University’s Archaeology Department and to make use of the comparative faunal collection that Jennie Coy had developed. My Fulbright was extended until 1979, and I also received grants from the United States National Science Foundation (BNS 7708141) and the Wenner-Gren Foundation for Anthropological Research (Grant No. 3267). I am so grateful to the late Jennie Coy, who helped me at all stages of the project, and to the Archaeology Department at Southampton, then chaired by Lord Colin Renfrew, which provided me with a home away from home. I am also grateful to Dr. Mark Maltby, now a reader in archaeology at Bournemouth University, who was then a junior research fellow in Southampton’s Faunal Remains Unit. Mark has been one of the pioneers in medieval urban zooarchaeology, and his work on Roman and medieval Exeter (Maltby 1979) remains a model for all of us to emulate. Although I returned to the UK in the 1980s for several conferences and short projects, I did not undertake another big Anglo-Saxon project until 1990. Shortly after the completion of the excavations at the Middle Anglo-Saxon estate center of Brandon in northwest Suffolk, I  had the opportunity to work on the huge collection of faunal remains from that site. The work was supported by a summer grant from the US National Endowment for the Humanities and by English Heritage. I am grateful to Andrew Tester and Bob Carr for giving me the opportunity to work on the material and to Sebastian Payne for his constant support throughout the project. The Brandon project allowed me to explore changes in animal husbandry practices, hunting patterns, and diet between the Early Anglo-Saxon period at West Stow and the Middle Anglo-Saxon period. Although it took quite a while for the final Brandon site report to appear, the final product (Tester et al. 2014) was certainly worth the wait. I am also grateful to my husband, Doug Campana, who worked alongside me on the Brandon project, and who spent more hours than he would like to admit measuring chicken bones. Our measurement archive has been sent to the Chicken Coop, as a study of the Scientific and Cultural Perceptions of Human–Chicken Interactions (http://scicultchickens.org/), and I hope that our data will lead to a greater understanding of the role of chickens in Anglo-Saxon society. After we had completed the initial report on the Brandon faunal material, I  was offered the opportunity to analyze the animal bone material from two other Middle Anglo-Saxon sites, Ipswich and Wicken Bonhunt. The original identifications were made by Patricia Stevens, and I  will be forever grateful for her help, her careful identifications,

xvi

Preface

and her hard work. I was asked to analyze the material and to prepare short reports on the animal bones that would appear as part of the site reports and longer reports that would be filed as archive reports for English Heritage. Unfortunately, the final site reports on Ipswich and Wicken Bonhunt were never published, and I was given the opportunity to publish the Middle Saxon faunal remains from these two sites in an East Anglian Archaeology monograph entitled Middle Saxon Animal Husbandry in East Anglia (Crabtree 2012). Both sites have important implications for our understanding of the beginnings of urbanization in the Middle Anglo-Saxon period. Wicken Bonhunt is a Middle Saxon estate center in Essex near the Cambridgeshire border with evidence for specialized pork production, possibly as early as the late Early Saxon period (late 6th–7th century). It was excavated by Keith Wade of the Suffolk Archaeological Unit. Ipswich is one of the emporium or wic sites, centers of regional and international trade in the Middle Saxon period. It is the only one of the wics that survived as an urban center in the Late Saxon and early medieval period. The animal bone data from Ipswich will be published as a future monograph in the East Anglian Archaeology series (Crabtree under review). I am grateful to Keith Wade, former county archaeologist for Suffolk in the UK, for the opportunity to work on these important archaeological materials, and I am delighted that many of the reports on Ipswich have been made available through the Archaeological Data Service. I am also grateful to Sebastian Payne who invited me to work on these data. I began work on the faunal remains from the Late Roman small town of Icklingham, located about 6 miles (10 km) from West Stow, in 1989. I initially examined the material that was excavated by Stanley West and Jude Plouviez back in the 1970s. I  had hoped to include some of this material in my PhD thesis, but I simply did not have time to do the work in the late 1970s. In 2005, I had the opportunity to study the Icklingham material that was collected by Drs. Catherine Hills and Jess Tipper as part of the Lark Valley Archaeological Project. I  was able to finish the West–Plouviez faunal collection in 2008. I am very grateful to both Jude Plouviez and Catherine Hills for allowing me to study these materials. Catherine arranged for me to stay at Cambridge while I  was working on her materials, and Jude allowed me to stay in her house while I was finishing the Icklingham collection. I  owe a debt of gratitude to both of them. As a student of Bernard Wailes, I have always felt that the data from later prehistoric and early medieval Europe have an important role to play in our broader understanding of social, political, and economic change in premodern societies. I  was honoured when my colleague

Preface

xvii

Professor Rita Wright invited me to submit a proposal for her series Case Studies in Early Societies, and I regret that it has taken me so long to complete this project. I  faced several years when I  had to combine a heavy teaching load during the academic year and field work in several countries during the summer. I was able finally to finish this book thanks to a visiting research scholarship at New York University’s Institute for the Study of the Ancient World (ISAW) during the fall semester of 2015. I am profoundly grateful to Professor Roger Bagnall, the recently retired director of the institute, and to Shelby White, the institute’s founder, for this opportunity. I also benefited enormously from the seminar on the ancient economy that was co-taught by my colleagues Lorenzo d’Alfonso and Elizabeth Murphy. Thank you to everyone at ISAW. Finally, I  also would like to thank my husband, Douglas Campana, who read, edited, and commented on many drafts of this book. I  am profoundly grateful to my colleague Professor Rita Wright, the editor of this series, who encouraged me to write this book. I  appreciate her patience and her support. Dr.  Zenobie Susanne Garrett also read the book, and I am most grateful for her input. I would also like to thank the two anonymous reviewers whose comments have made this a much better volume. The errors, of course, are all mine. This book is dedicated to the memory of my late colleague Professor Howard Dalton Winters. Howard was one of the pioneers and founders of processual archaeology in North America, and his work on the Riverton Culture (Winters 1969) remains a classic in archaeological analysis. Howard was a wonderful supporter of mine when I  was a very junior faculty member at New York University. One of the greatest pleasures in my teaching career was co-teaching environmental archaeology with Howard back in the early 1990s. Howard was looking forward to a long and productive retirement when he passed away from a stroke in 1994. In the two decades since his passing, I  have taken over some of his courses, including Prehistoric Hunters and Gatherers, and Archaeology and the Environment. I  hope that he would approve of the job I  have done teaching his courses. Howard, thank you for always being in my corner. You are missed.

Introduction

This is a book about the origins and development of towns and cities in Anglo-Saxon England from about 400 CE to the time of the Norman Conquest in 1066. It is a topic that has fascinated me since I excavated at the Brook Street site in Winchester over 45 years ago, and it is a topic that has engaged scholars in both history and archaeology throughout the 20th and 21st centuries. In this introductory chapter, I want to introduce Anglo-Saxon England and address some of the ways that archaeologists have approached the study of urbanism in the past. This chapter will provide a brief introduction to the chronology of Anglo-Saxon archaeology. It will then address some of the theoretical approaches that have been used to answer questions about the origins and development of towns in medieval Europe and elsewhere. I will also discuss my theoretical perspective and introduce the rest of the chapters in this volume. Who Were the Anglo-Saxons? While the Anglo-Saxons are part of every schoolchild’s education in the United Kingdom, the medieval world in general and the Anglo-Saxons in particular are often missing from both secondary and higher education in the United States. Courses on world history and western civilization generally begin with Egypt and Mesopotamia, move quickly to Classical Greece and Rome, briefly mention the medieval period between 400 and 1400 CE, and then move on to the Renaissance, the Reformation, the Enlightenment, and the modern world. In popular culture the medieval period is often seen as the “Dark Ages” or the “Age of Faith” (see, for example, Durant 1980) without recognizing that this was a period of substantial social, political, and economic changes that began with the decline of the western Roman Empire and ended with the culture contact that resulted from the European voyages of discovery. The earlier parts of the medieval period have always represented a substantial challenge for traditional historians, since there are relatively few documents that survive from this period. For example, Sawyer’s classic, From Roman Britain 1

2

Introduction

Table 1. Basic chronology of Roman and Anglo-Saxon England Period

Dates

Roman Sub-Roman Early Anglo-Saxon Middle Anglo-Saxon Late Anglo-Saxon

43–ca. 410 CE ca. 410–ca. 450 CE ca. 450–ca. 650 CE ca. 650–ca. 850 CE ca. 850–1066 CE

Note: The Late Anglo-Saxon period is often referred to as the Anglo-Scandinavian period in eastern England.

to Norman England (1998), has relatively little to say about the 5th and 6th centuries, since the textual evidence is so limited. The first chapter is entitled “The Seventh Century and Before.” Agricultural historians Banham and Faith (2014, 15)  have argued that this period should be seen as “a proto-historic, rather than a truly historic period” due to the lack of written source material. Fortunately for us, the development of medieval archaeology, especially since the end of World War II, has helped to shed new light on the “Dark Ages.” Britain became a province of the Roman Empire in the 1st century CE, and the Roman army was finally removed from Britain in 407 CE, almost 300 years later. By the end of the first decade of the 5th century the inhabitants of Britain were on their own politically. Historical sources suggest that Germanic immigrants from the regions that today include the northern portions of the Netherlands and Germany and eastern Denmark began entering southern and eastern England sometime in the mid-5th century, although the nature and extent of this migration has been a matter of serious debate (see Chapter 2). The term “AngloSaxon” has been applied to the period from the mid-5th through the mid11th centuries CE in England. It is based on historical sources, such as the 8th-century Venerable Bede, who wrote that these migrants included Angles, Saxons, and Jutes. Efforts to identify specifically Anglian, Saxon, and Jutish material culture in the archaeological record have met with, at best, mixed success, and the culture we describe as Anglo-Saxon is clearly an amalgam of native Romano-British and Germanic practices (see Hills 2003 for a more detailed discussion of the origins of the English). In dividing up the period between the end of Roman rule in Britain and the Norman Conquest, I will adopt the conventional chronology as a heuristic device (see Table 1). I have used the term “sub-Roman” to describe the first half of the 5th century CE, a period that is poorly known both archaeologically and historically (see Chapter  1). It is important

An Outsider’s View of Anglo-Saxon Urban Origins

3

to recognize that (1) recent archaeological research has challenged this traditional chronology in several ways, and (2) important changes take place both within and across these time periods. We cannot and should not view them as a series of static time slices. As noted above, there is very limited historical evidence for the Early Anglo-Saxon period. Hills (2003, 110) notes that: The historical sources tell us reliably only two things. First, that during the 5th century AD Britain ceased to be part of the Roman Empire and was subject to attack from a variety of peoples, including Saxons. Secondly, that the rulers of the peoples living in eastern and southern Britain by the 8th century, and perhaps before that, believed they could trace their ancestry back to heroic Germanic leaders from the continent.

As a result, most of our data on settlement patterns and lifeways for the Early Anglo-Saxon period come from archaeology. The Middle Anglo-Saxon period, conventionally dated between 650 and 850 CE, is the period that saw the development of the emporia or wic sites, arguably the first towns in post-Roman Britain. These sites appear to have served as centers of craft production and of regional and international trade. The period ends with the appearance of the Great Viking Army in eastern England. During much of the Late Anglo-Saxon period, traditionally dated between 850 and 1066 CE, Scandinavians controlled large parts of eastern England. The late 9th and 10th centuries were probably some of the most politically and militarily tumultuous in all of English history, as the Vikings and the kings of Mercia and Wessex vied for political and military control of the region. The period ends with the Norman Conquest of England in 1066 CE. An Outsider’s View of Anglo-Saxon Urban Origins This volume will focus on the development of urbanism in later AngloSaxon England (ca. 700–1050 CE) and the background to urban development (ca. 400–700 CE). It will highlight the archaeological evidence for the rebirth of towns in the post-Roman world, since historical data for the earlier part of Anglo-Saxon England are strictly limited. Although I  have been active in the archaeology and zooarchaeology of AngloSaxon England since my postgraduate days in the 1970s, I come to this project as a bit of an outsider. When I first arrived at the University of Southampton in the 1970s, my late friend and colleague Jennie Coy asked me why an American like me would be interested in the Anglo-Saxons. It is a good question. As noted above, while the Anglo-Saxon period is

4

Introduction

part of the national dialogue in the United Kingdom, the Anglo-Saxons are among the missing in secondary and tertiary education in the United States. The few Anglo-Saxonists in the US are to be found primarily in university departments of English and history. I am one of the very few university archaeologists in the United States whose primary interest is in Anglo-Saxon England. The number of medieval archaeologists in the United States can probably be counted on the fingers of two hands. As a result, the treatment of the Middle Ages in the United States focuses almost entirely on literary and other written sources. There is almost no role for material culture outside of art history. On the other hand, I do have some advantages as an outsider. Modern post-processual archaeologists have challenged the earlier notion that archaeological research is entirely objective. We always view the past through the lens of the present, and Anglo-Saxon studies are no exception (see, for example, Harke 1998). For example, in the 19th and early 20th centuries, the scholarly view of Early Anglo-Saxon England was based primarily on the documentary sources, such as Gildas, a 6thcentury British cleric writing in what today is Wales, and Bede, an 8thcentury ecclesiastic writing A History of the English Church and People in Northumbria (Bede, tr. Sherley-Price 1968). The general view based on the limited historical record was that the arrival of the Anglo-Saxons in post-Roman Britain in the 5th century led to the defeat and disappearance of the Romanized Britons, who were killed in battle, enslaved, or driven into western Britain and Wales. This viewpoint can be seen as both Germanist and migrationist (see Oosthuizen 2016 for a modern critique of this approach). Ideas about the relationship between Roman Britain and Anglo-Saxon England began to change in the post-war period. The role of Germanic migration was downplayed, and archaeologists began to look for possible continuities between Roman Britain and AngloSaxon England at both rural and urban sites. The search for continuities between Late Roman/early post-Roman Britain and Early Anglo-Saxon England is certainly a valid one, and it was one of the questions that formed part of my PhD research on the faunal remains from the AngloSaxon village of West Stow. It is certainly no coincidence, however, that these questions began to be asked in the aftermath of the two world wars that pitted Germany against the United Kingdom. In addition, much of the research on Anglo-Saxon England has been written by British people for a British audience. In some cases it has been both literally and figuratively insular. My goal in this volume is to bring Anglo-Saxon archaeology to a wider audience and to involve Anglo-Saxon archaeology in the broader comparative discussions about early urban origins. Since the focus of this volume is specifically on the origins of urbanism in Anglo-Saxon England, I will begin with a brief review of some of the

Archaeology and Urban Origins

5

general archaeological literature on urban origins. I  will then address some of the specific theories that have been put forth to explain the origins of early medieval towns in Britain and northwest Europe. Archaeology and Urban Origins I will begin with a very simple definition of urbanism, drawing on the criteria that seem to be most universal for archaeologists studying urban origins. First and foremost, towns and cities should have relatively large, dense residential populations. This allows us to distinguish cities and towns from ceremonial centers, which may have monumental architecture, including religious monuments, but may be home only to a small permanent population. Second, cities should be home to some people who are engaged in activities other than food procurement (animal husbandry, hunting, fishing, agriculture, and plant collecting) on a full-time basis. These may include people working as craft specialists, laborers, judges and administrators, traders and moneyers, and religious specialists. Third, cities should show some evidence for organization and planning, which can be taken as evidence for complex political organizations. These need not necessarily include a king or emperor. Sodalities and complex kinship systems may also form part of urban governments, as they did in places such as Jenne Jeno in Mali in the 1st millennium CE (R. McIntosh 2005). Urban planning can take a variety of forms ranging from street plans to defenses to the zoning of residential and commercial areas. This is essentially a functional definition of urbanism, and I recognize that urbanism is a process rather than an event and that we can talk about a wide spectrum of urban societies. As Carballo and Fortenberry (2015, 542) note: Consideration of the spectrum of urbanism illustrates that early urban ‘revolutions’ do not constitute an end point in archaeological study and reveals the ways in which certain settlements were more urban than others, as remains the case among modern cities and towns.

I will return to this discussion in Chapter  3 when I  consider whether the Middle Anglo-Saxon settlements known as wics or emporia should be seen as urban. I will consider whether these settlements are stepping stones on the way to the development of towns in the Late Saxon period, and/or whether they are reflective of a particular moment in the development of trade around the North Sea. Urbanism, however, is more than just the presence of a substantial population and evidence for occupational specialization and some degree of urban planning; urbanism is a process that transforms the countryside as well as the emerging urban center. For example, in describing the process of city-state formation in Mesopotamia, Yoffee (1995, 284)  noted

6

Introduction

that “the countryside was created as a hinterland of city-states.” More recently, Monroe (2011) has adopted a landscape approach to the question of urban development along the Slave Coast in West Africa. He suggests, “Cities are thus settlements that provide specialized services to a broader hinterland. The key issue, therefore, is not what a city is, but what a city does for rural communities within its sphere of influence” (Monroe 2011, 400). This approach is not unlike the definition of urbanism adopted by Russo (1998, 22; following Reynolds 1977), who suggests the following working definition of a medieval town: [A] permanent human settlement with a significant portion of the population living off a variety of non-agricultural occupations and forming a social unit more or less distinct from the surrounding countryside.

In this volume, I plan to draw on the landscape approach. I will examine not only the archaeology of emerging urban sites, but also changes in rural settlements that may reflect changes in relationships between emerging urban centers and the surrounding countryside. However, this is not strictly a book about the landscape archaeology of Anglo-Saxon England. I plan to draw on other theoretical perspectives from contemporary archaeological thought. Theoretical Perspectives The rise of complex societies, including the beginnings and development of urbanism, is one of the most interesting and important questions faced by archaeologists and anthropologists. Unfortunately, much of our understanding of the process of urban development in the archaeological record has been based on the study of a limited number of ancient societies. Childe, who was perhaps the most important archaeological theorist of the first half of the 20th century, focused primarily on Egypt, Mesopotamia, and the Indus civilization in his classic studies such as Man Makes Himself (1936). At that time, relatively little was known about the archaeology of early China, and Childe was largely unaware of the archaeological record for the emergence of complex societies in the Americas. Medieval Europe only played a minor role in Childe’s theoretical framework. In What Happened in History Childe (1942, 25) argued, for example, that feudalism tied medieval cultivators to the soil and emancipated them from Roman slavery. However, he had nothing to say about the development of towns and cities in medieval Europe. In a subsequent publication, Childe (1950) introduced his concept of the “urban revolution,” based on an analogy with the Industrial Revolution of the late 19th and earlier 20th centuries. It was seen as

Theoretical Perspectives

7

a change in technology that had ramifications for many other aspects of society. For Childe, the urban revolution was the second of two prehistoric revolutions. The first, the agricultural revolution, including the domestication of plants and animals, was a necessary precursor for the second, since farming provided the food surpluses that could be used to support non-farmers, including everyone from rulers to laborers, and to support everything from regional and long-distance trade to the construction of monumental architecture. Childe identified ten criteria that he saw as central to what he termed the “urban revolution.” They included urbanism, craft-specialization, the concentration of economic surplus through taxation or tribute, class stratification, state formation, monumental public works, long-distance trade, standardized art, writing, and advances in the sciences, including mathematics and astronomy. Redman (1978, 218)  subsequently subdivided these into primary and secondary characteristics. Redman saw the first five criteria – urbanism, craft-specialization, the concentration of economic surplus through taxation or tribute, class stratification, and state formation – as primary. He saw the secondary criteria as consequences of the first five. Today, most archaeologists would reject a “Chinese-menu” approach to the study of the origins of complex societies. We know that the Inca, the largest empire in the Americas prior to European contact, had no formal writing system, although they did have an elaborate system of record-keeping based on knotted strings known as the quipu (Asher and Asher 1997). Long-distance trade is not a unique feature of complex, urban societies. For example, the Natufians were late Pleistocene hunter-gatherers who lived in the southern Levantine regions of the Middle East. Natufian sites in the Jordan Valley have yielded beads made of shells from both the Mediterranean and the Red Sea (Crabtree et al. 1991). Neolithic monuments in Europe, including megalithic tombs like Newgrange in Ireland that reflect complex astronomical observations, were constructed millennia before the beginnings of urbanism in that part of the world (Harbison 1988, 76). Similar lists of traits have been used to define urbanism in the medieval world. Medieval towns were seen as having one or more of the following:  defenses, a planned street system, a market, a mint, legal autonomy, a role as a central place, a large and dense population, a diverse economic base, houses and plots of urban type, social differentiation, complex religious organization, and a judicial center (Biddle 1976, 100, following Heighway 1972). This list presents many of the same problems as Childe’s ten criteria do. Some complex early medieval defense systems such as the Danewerke in Denmark (Hellmuth Andersen et al. 1976) and Offa’s Dyke near the

8

Introduction

Welsh border in Britain (Hill and Worthington 2003) are not directly associated with urban centers. Judicial centers and execution sites, such as Sutton Hoo (Carver 2005), also exist outside urban centers. Other criteria, such as legal autonomy, seem to be based on our knowledge of later medieval cities and may not be appropriate for the initial development of urbanism in the earlier medieval world. In summary, trait lists are not necessarily reliable ways to define early medieval urbanism. Despite these reservations, however, it is clear that questions about the origins of urbanism and state formation lie at the heart of most studies of the emergence of complex societies. In the 1960s and 1970s, both archaeologists (e.g., Adams 1966) and anthropologists (e.g., Service 1975) who were interested in urban development focused primarily, although not exclusively, on a limited number of ancient societies that were seen as examples of independent or primary urban development and state formation. These included regions such as ancient Egypt, Mesoamerica, China, the Indus region, Peru, and Mesopotamia where states and cities were supposed to have developed without much in the way of outside influence. However, this perspective began to change in the 1980s. There are several reasons for this change. First, cultural anthropologists abandoned their interest in cultural evolution, leaving these questions to the archaeologists. This has led to an approach that is based increasingly on archaeological data rather than on ethnographic parallels to the 19th- and 20th-century societies studied by earlier generations of social anthropologists. This is, in many ways, a welcome development. For example, for too much of the later 20th century, archaeologists viewed ancient hunter-gatherer societies through the lens of the !Kung San (Lee 1968, 1979). We saw these societies as unchanged and unchanging, failing to acknowledge that many non-western societies studied by early anthropologists were undergoing rapid social, political, and economic changes as a result of culture contact and colonialism. The archaeological record allows us to look for greater variability among both hunter-gatherer and early urban societies without viewing them through the prism of a small number of well-known ethnographic case studies. Second, as a truly world archaeology has developed in the past 40 years, studies of the development of urbanism and state formation have been carried out in regions outside the Middle East and Latin America. An excellent example is the research conducted by S. K. McIntosh (1999) and R. McIntosh (2005) on early urbanism in the Middle Niger region in Mali. While the development of urbanism in this region was traditionally linked to contacts with Islamic traders in the Middle Ages, their research has shown that urbanism developed independently in Mali well before the beginnings of the trans-Saharan trade. Similarly, Kirch (2010,

A Processual Plus Perspective

9

2012) has argued that states developed independently in Hawaiʽi in the late pre-contact period. While the archaic Hawaiian states were nonurban, the process of state formation presents some interesting parallels to the beginnings of state formation in Anglo-Saxon England (Crabtree 2017). Several decades of rescue archaeology have shed new light on the Mississippian urban center at Cahokia near St. Louis, MO, and some researchers (J. Z. Holt 2009) have suggested that Cahokia may be the center of a performance state, similar to the arguments that have been made for the Maya (Demarest 2005). In short, we now have a broad range of data on the emergence of complex societies in a number of different regions of the world. These case studies can provide useful comparative data for the study of the emergence of Anglo-Saxon urbanism. Third, many archaeologists and anthropologists (Service 1975; Kirch 2010) have focused on the differences between primary state formation and secondary states. In early medieval Europe, Hodges (1982) suggested that state formation (and concomitant urbanism) in Carolingian France was primary, while broadly contemporary state formation in Anglo-Saxon England was secondary. I would argue that this is a distinction without a real difference. There is no question that the inhabitants of Hawaiʽi were relatively isolated from other complex state-level societies prior to the late 18th century. However, this does not appear to be the case in many other parts of the world. For example, detailed archaeological research in the Middle East and Mesoamerica has cast some doubt on the alleged independence or primacy of urban development and state formation in these so-called primary centers. Research in Mesoamerica has shown that the Olmecs are less of a mother culture (cultura madre) and more of an older sister culture to the complex urban societies that developed in Oaxaca, the Valley of Mexico, and the Maya regions (see, for example, Lesure 2004). Archaeological research that has been carried out in the geographical region that stretches between Egypt and the Indus has identified a number of additional early urban societies, such as Elamites in Iran, and these early complex societies appear to have been in contact with one another. In short, the evidence for contacts between early complex societies in the so-called primary centers of urbanism has made the distinction between primary urban societies and other early urban societies less meaningful, and the study of urban origins has encompassed a wider range of societies in the past 30 years. A Processual Plus Perspective The theoretical model that I use is one that Michelle Hegmon (2003) first described as “processual plus.” It grows out of the new or processual

10

Introduction

archaeology movement that dominated archaeological thought from the 1960s through the early 1980s (see, for example, Binford 1962; Binford and Binford 1968; Watson et al. 1971). The goal of processual archaeology was to focus research on the processes of cultural change in the past, including questions such as the origins of plant and animal domestication and the origins and growth of cities and states, Childe’s (1936) two revolutions. While processual archaeology never managed to come up with law-like generalizations about the processes of social change similar to the laws of physics (contra Watson et  al. 1971), the processual movement did lead to significant improvements in archaeological methods, including improvements in archaeological surveying techniques, development of systematic methods for the collection and analysis of plant and animal remains from archaeological sites, and advances in the stratigraphic excavation and interpretation of archaeological sites, among other methodological advances. The use of these techniques has radically transformed the ways in which Anglo-Saxon archaeology is practiced today. By the mid-1980s, however, archaeologists were also beginning to recognize some of the inherent weakness in the processual approach (see, for example, Hodder 1986). A  relentless focus on general models left little room for human agency or historical contingency. In addition, the processual model paid little attention to issues of personal identity at a time when issues surrounding race, class, and gender were becoming increasingly important in many of the other social sciences (Conkey and Spector 1984). Archaeologically, men and women were simply reduced to “faceless blobs” (Tringham 1991, 94). The processual plus approach acknowledges these criticisms and tries to combine the methodological advances of the processual era with an awareness of the importance of social and historical issues such as agency and identity. It also recognizes that social institutions are the products of everyday decisions made by people living their lives (Bourdieu 1977; Giddens 1984). In other words, day-to-day lives of common people matter, and their decisions can give us insight into how societies develop and change. I am a zooarchaeologist, an archaeologist who analyzes animal bone remains in an attempt to reconstruct past animal husbandry practices, hunting patterns, and diets. I  have spent a good portion of my academic career studying animal bone assemblages from Anglo-Saxon sites in eastern England. As such, I  have a strong interest in the medieval economy. As a student who was trained during the processual era, I was intrigued by an approach known as economic prehistory (see Higgs 1972; Higgs 1975). This methodological and theoretical approach focused on the use of data such as animal bones and seed remains to reconstruct

A Processual Plus Perspective

11

the economies of prehistoric and early historic sites. Unfortunately, the extreme version of economic prehistory argued that “economy is everything, and culture is noise.” This approach, like the broader processual movement, overlooks the importance of human agency, individual decision-making, and the ways in which economic behavior is integrated with other aspects of society. New models of economic analysis allow us to incorporate the study of ancient economies into a broader framework. The New Institutional Economics (NIE) is an economic perspective that incorporates institutions, transaction costs, and political economy into the study of both ancient and modern economies (North 2005, 21; see also Coase 2005; North 1993). For example, New Institutional Economists examine the relationship between trade and the legal and judicial systems. Are there institutions such as laws and courts that are designed to protect traders against theft and fraud in the market place? Does the political system encourage or discourage trade and risk-taking? How is the security of foreign and non-local traders guaranteed? How heavily is trade taxed, and what kinds of exemptions are available to traders? In short, how do the political, social, and legal systems work to encourage or discourage trade and other similar economic activities? The New Institutional Economics also attempts to address a long-term debate within economic anthropology. For several decades, economic anthropology has been dominated by a debate between the substantivists and the formalists (see Moreland 2000a for a discussion of these issues in medieval archaeology). The substantivists argued that ancient and non-western economies were embedded in social, political, and religious systems (Polanyi 1957). Many substantivists adopted a neo-evolutionary model, arguing that certain forms of trade and exchange were associated with specific forms of socio-political organization. Formalists, on the other hand, argued that ancient economies could be studied using the same market principles that economists use to study modern economies. New Institutional economists note that no markets operate on strictly neoclassical principles (Coase 2005). Neoclassical models fail to account for transaction costs, the costs of making an exchange and participating in a market. For example, corruption, such as payola, is a transaction cost, while coinage can reduce transaction costs by easing the process of exchange. The important feature of the New Institutional Economics is that it shows how both ancient and modern western economies are embedded in broader socio-political contexts. In this case, culture is not noise. By understanding the social and political institutions that may encourage or discourage trade or exchange, we can shed new light on past economic behavior. Moreover, the New Institutional Economics

12

Introduction

allows us to avoid the evolutionary assumptions that underlie many of the substantivist models. Market exchange, gift exchange, and reciprocal exchange are not necessarily tied to particular socio-political forms. Multiple forms of exchange can exist at the same time. For example, in highland New Guinea, cash wages have not replaced the traditional exchange systems. Instead cash is sometimes included along with other valuables in the traditional exchange networks (Rena Lederman, personal communication; see Lederman 1986). In addition, as Wright (2013, 54)  notes, the division of exchange into gift-giving vs. market exchange may be oversimplified. Developments in exchange theory show that objects gain value through their history. Even objects that enter the system through the marketplace may gain value as a result of subsequent non-market exchanges. To give a modern example, my greatgrandmother purchased her tea set from a market, but its value to me is that it was also owned by my grandmother, my mother, and now by me. One of the advantages of being an archaeologist in the 21st century is that there are a range of interpretive models that can be used to understand the origins of urbanism. This volume will take a processual plus approach in studying the rebirth of towns in post-Roman Europe. The question I am asking is essentially a processual one – can we use archaeological data to understand the origins and growth of towns in AngloSaxon England? I  will draw on a wide range of archaeological data in an attempt to answer that question. However, I realize that a traditional processual approach failed to address many of the issues that make the Anglo-Saxon case distinctive. I will draw on a range of recent theoretical approaches including, for example, landscape archaeology and the New Institutional Economics as a way of understanding the process of urban growth and development. Models of Medieval Urbanism This book explores the rise of urbanism in early medieval northwest Europe as part of a broader study of urban origins. The origin of medieval urbanism was of great interest to historians throughout the 20th century. While imprisoned during World War I, the great Belgian historian Henri Pirenne wrote much of his classic work, Medieval Cities: Their Origins and the Revival of Trade (Pirenne 1925 [1969]). Pirenne’s ideas were further explored in Mohammed and Charlemagne, which was published posthumously in 1937. Pirenne suggested that Roman patterns of urbanism and trade were not completely disrupted until the spread of Islam in the 8th century. It was only then that northwest Europe was isolated from the Mediterranean, leading to new patterns of trade and new institutions

Models of Medieval Urbanism

13

of government. He further argued that the disappearance of trade put an end to the last vestiges of city life in western Europe in the turbulent 9th century (Pirenne 1925 [1969], 65). Towns and castles were merely fortified places and administrative centers. However, these places provided the foundations for the revival of commerce in the 10th century (Pirenne 1925 [1969], 76). Pirenne’s model was based almost exclusively on historical sources, since urban medieval archaeology did not develop until after World War II (Crabtree 2004, 2009). While many aspects of the Pirenne thesis can be tested using archaeological data (see, for example, Hodges and Whitehouse 1983), archaeological research on early medieval northwest Europe only began in earnest in the mid-20th century. Since modern cities are located on top of most medieval towns and cities, the devastation that was caused by World War II allowed archaeologists to have access to large areas of many cities such as London for the first time. Since that time, rescue or preventive archaeology, carried out in advance of urban renewal, has produced a wealth of data on urban origins in medieval Europe. Archaeological approaches to medieval urbanism developed as a response to urban redevelopment in the post-war years. Archaeologists had been interested in the urban revolution since the days of V. Gordon Childe, but the beginnings of the processual movement in archaeology in the 1960s and 1970s encouraged excavators to use archaeology to examine the processes of cultural change. Urban origins and development were important processes that could be examined archaeologically; this led to important studies of the development of ancient cities in the Near East, Mesoamerica, and elsewhere, including medieval Europe. In the 1980s, Richard Hodges (1982, 1989, 2012) developed a processual model for the origins and growth of medieval cities, based in part on his own research on pottery assemblages from Hamwic (Hodges 1980). Drawing on economic anthropology, geography, and processual archaeology, Hodges developed a model for state formation, urban origins, and trade in early medieval northwest Europe. In Dark Age Economics, he argued that there was a systematic relationship between the shift from chiefdoms to states, the change from a redistributive system to a market economy, and the nature of the markets themselves. Chiefs and kings had a central role in the process, initially by controlling access to foreign luxury goods and subsequently by monitoring market exchange and benefiting from taxation. The agents of change were the kings, rather than the peasants, even though these kings were dependent on an economic surplus to support those people who were engaged in noneconomic pursuits such as trade and craft production (Hodges 1989,

14

Introduction

130). Hodges concludes his book by noting, “At each developmental stage the central person is gaining greater mastery of the traditional levelling mechanisms.” He further notes that “the archaeology of trade … demonstrates the complexity that great men had to tackle” (Hodges 1989, 197–8). Hodges’s book can be seen as a substantivist approach to the study of urban origins, in that, following Polanyi (1957), he saw economic behavior as embedded in a broader social and political context. This, however, was not the only approach adopted by archaeologists studying early medieval origins in the late 20th century. Peter Wells (1984) adopted a more formalist approach to the study of early medieval urban origins in Farms, Villages, and Cities. While most of the book focuses on the origins of towns in later Iron Age Europe, the final chapter addresses the growth of towns in post-Roman Europe. Wells was not concerned with the role of the emerging political and social elites in the formation of early medieval towns. Instead, he was concerned with the roles played by entrepreneurial individuals. Wells argued that the critical factors in the development of urbanism were “individual initiative and motivation to acquire specific luxury goods, and the mobilization of communities to produce materials that could be exchanged for desired luxuries” (Wells 1984, 204). He argued that the people who were buried in rich graves from the 6th century onward “can be interpreted principally as those who played entrepreneurial roles in coordinating trade and reaped the benefits from that trade” (Wells 1984, 198). Wells’s model failed to stimulate the kinds of interest that Hodges’s did, but the two models illustrate the diversity of theoretical approaches to archaeological research into early medieval urbanism in the 1980s. The value of Hodges’s book is that it stimulated a great deal of research on early medieval trade, urban origins, and state formation. At the time that the first edition of the book was published (Hodges 1982), the locations of Middle Anglo-Saxon London and York were unknown, and there were few detailed published accounts of early medieval rural sites in Britain. Much of Hodges’s analysis was based on historic sources and numismatics. In the 30 years or more since the publication of Dark Age Economics, new archaeological discoveries and changing interpretive frameworks in archaeology have challenged Hodges’s initial model. He revisited it in Dark Age Economics:  A New Audit (Hodges 2012). He concludes by noting that “the Dark Age economy, as it now appears, was more varied, more regional and more complex than DAE envisioned” (Hodges 2012, 138). One of the goals of this volume is to examine some of this complexity and variety. Another goal is to provide a corrective for the top-down approach that was taken in the initial Dark Age Economics

The Organization of This Volume

15

publication. We now have many more rural sites that have been excavated using modern methods, and we can begin to examine the Dark Age economy both from the top down and from the bottom up. There is now a rich literature on the rise of towns and the relationship between urbanism and state formation in later Anglo-Saxon England. Much of it has been written since Hodges’s (1982) pioneering study Dark Age Economics 30  years ago. However, there is a real need to revisit these issues in light of the new archaeological scholarship that has emerged in the past dozen years. The new developments include the redating of the beginnings of Ipswich Ware pottery. The start of this industry, which included the first industrially produced ceramics in postRoman Britain, was traditionally dated to ca. 650 CE. It was one of the features of material culture that was used to define the Middle AngloSaxon period. However, research by Blinkhorn (1999, 2012) suggests that the industry did not start until the beginning of the 8th century. This means that sites like Early Saxon West Stow, where 381 sherds of Ipswich Ware were recovered from the latest features, must have continued into the early 8th century, rather than having been abandoned around 650 CE. In addition, the publication of the Boss Hall and Buttermarket sites from Ipswich (Scull and Wade 2009) has led to a serious revision of the early history and development of Ipswich. Ongoing research by the Museum of London (Cowie 2008) has shown that, contrary to earlier expectations (see, for example, Scull 1997, 273), there is almost no evidence for continued occupation of Roman London during the 5th and 6th centuries CE. In addition, the publication of a number of new Early and Middle Anglo-Saxon rural sites, including the Middle Saxon estate center of Flixborough in Lincolnshire (Loveluck 2007), provides new data on the changes in rural settlement and subsistence that may have accompanied the origins and growth of urban settlements. The Organization of This Volume This volume will examine the archaeological evidence for the beginnings of urbanism in Anglo-Saxon England. It is not designed to be just another archaeology of Anglo-Saxon England. There are many of those available (see, for example, Arnold 1997; Kharkov 1999; Hinton et al. 2011; Higham and Ryan 2013), and there are many people who are far more qualified than I am to write such a volume. Instead, I want to focus on those factors – including economic, political, and social factors – that led to the development of towns in the later Saxon period, towns that are still occupied today, over 1000 years later. In Chapter 1, I will begin in a somewhat unusual place, by examining the state of urbanism in 4th- and

16

Introduction

5th-century Roman Britain. I will argue that urbanism in Roman Britain had largely failed by the middle part of the 5th century. While Roman towns may have continued to exist as parts of the landscape, most ceased functioning as urban centers by the early decades of the 5th century. I will focus on three case studies – London, Winchester, and Wroxeter. All three Roman towns have been subject to long-term excavation programs using modern archaeological methods, and these three cities show different patterns of decline in the Late Roman and early post-Roman period. Chapter 2 will examine the archaeological evidence for settlement in the 5th, 6th, and 7th centuries. One problem that archaeologists face is that while that between 30,000 and 40,000 burials have been identified from the 5th through the early 7th centuries, relatively few settlement sites from this period are known, and even fewer have been fully published. What can these settlements tell us about the social, political, and economic organization of Early Anglo-Saxon England? How different is Early Anglo-Saxon settlement in eastern England from contemporary settlements in other parts of Britain? Along with these questions, I will address the questions of what happened to the Romano-British population of eastern England in light of recent genetic and archaeological studies. And finally, can we see changes in settlement between the 5th and the 7th centuries that may reflect social changes that were taking places within the Early Anglo-Saxon period? In particular, I will examine the archaeological evidence for the beginnings of settlement differentiation in the late 6th and 7th century and the so-called “7th-century shuffle” (see, for example, Arnold and Wardle 1981). Do these settlement shifts reflect changes in social and political organization that laid the foundations for the emergence of trading centers or wics? Chapter  3 focuses on the foundation and growth of the emporia or wics. How, when, and why did these settlements develop? What is their relationship to the rise of the powerful Middle Anglo-Saxon kingdoms of Wessex, Mercia, East Anglia, and Northumbria and the foundation of the wics? The question of whether these sites should be considered as urban will be addressed, along with the question of whether these settlements should be described as “proto-urban.” An even more important question that needs to be examined is the relationship between the foundation of these settlements and the emergence of a small number of powerful Anglo-Saxon kingdoms known as the heptarchy. Hodges (1982) adopted a top-down model to explain the rise of the wics, suggesting that they were places where early kings could control the trade of high-status items, especially those that were being imported from the European continent. This “prestige goods” model has been challenged from a number of different perspectives in recent years. Moreland (2001) has

The Organization of This Volume

17

emphasized the role that these centers played in regional as well as international trade in the Middle Anglo-Saxon period. In addition, the identification of a number of productive sites (Pestell and Ulmschneider 2003), which have produced large quantities of metalwork including coins, has led archaeologists to conclude that local and regional markets may have been widespread in Middle Saxon England, indicating that these wics may have been a small part of a much wider network of trade and exchange in the 8th and 9th centuries. Others have suggested that the primary role of the wics may have been for the collection of duties, rather than for the control of prestige imports. I hope to show that the rise of the wics was part of a series of bottom-up social and economic changes whose origins can be seen in the later 6th and 7th centuries CE. Finally, I  will examine the relationship between these centers of trade and craft production and the towns that were founded in the 10th and 11th centuries CE. Chapter 4 will focus on the development of a widespread network of towns during the Late Anglo-Saxon period. Since only one of the wics, Ipswich, remained a town in the later Anglo-Saxon period, I will look for other possible antecedents for the Late Saxon towns. In particular, I will examine whether the construction of a series of fortified places known as burhs played a role in the development of Late Anglo-Saxon urbanism and the growth of Anglo-Scandinavian York and its relationship to other emerging towns in the Viking world. I will also consider the political, economic, and military roles of these towns in Late Anglo-Saxon England. In the concluding chapter, I  will try to set the archaeological evidence for the development of urbanism in the Anglo-Saxon period into a broader archaeological context. What can the Anglo-Saxon data tell us about urban origins in the ancient and medieval worlds? In discussing the development of small towns in Anglo-Saxon England, Christie and Creighton (2013a, 9) note that “we need to see these sites in their broader context  – not only in the context of their surrounding landscapes, but also escaping the paradigms of Anglo-Saxon studies to appreciate their place in the longer-term history of urban evolution and the development of royal authority.” This is my goal for the final chapter and the conclusion. I will also briefly compare the development of urbanism in AngloSaxon England with its development in other parts of the world.

1

The End of Urbanism in Roman Britain

Introduction This is a volume that explores the origins and growth of urbanism in early medieval Britain. However, the early medieval towns were not the first towns in Britain. A small number of the largest Iron Age settlements may have functioned as towns (see box on the oppida below), but widespread urbanism was introduced to Britain after the Roman conquest of the region in the mid-1st century CE. The archaeological traces of Roman imperialism in Britain were catalogued by Collingwood and Richmond (1969) nearly 50 years ago. They include forts and fortresses, roads, literacy, and many items of material culture. However, among the most important Roman foundations were the towns, since urbanism was seen as an essential part of Roman life. Urban centers included large cities such as London, coloniae such as York that were home to retired members of the Roman military, and the civitas capitals, such as Winchester, that served as administrative and market towns for local tribal regions (Figure 1.1). At the bottom end of the scale were the small towns, many located on or near major roadways and trade routes. Larger Roman towns were generally laid out on a grid plan. Public buildings, including fora, amphitheaters, and baths, were often constructed in the earlier part of the Roman period, generally in the 1st and 2nd centuries CE. In the later part of the Roman period, large landowners also built elaborately constructed town houses within the Roman towns and cities. Towns and cities played a variety of different roles in Romano-British life between the 1st and the 4th centuries CE. They served as ports, markets, administrative centers, centers for a variety of different religious cults including Christianity, and locations for a number of specialized crafts. In addition, Esmonde Cleary (1989) has argued that one of the main functions of the urban centers was the collection of taxes in money and in kind. He further suggested that one main function of coinage in the Roman Empire was the conversion of taxes in kind into taxes in coin. 18

0

100

200

300

400 km

Figure  1.1. Map of Britain showing the locations of the RomanoBritish sites discussed in this chapter. Drawn by Douglas V. Campana using open-source software. Map numbers: London – Londinium 1, Queenford Farm 2, Berinsfield 3, Wroxeter – Viroconium 4, Winchester – Venta Belgarum 5, Lincoln – Lindum Colonia 6, St. Albans – Verulamium 7, Silchester – Atrebatum 8, Bath – Aquae Sulis 9, Cirenchester – Corinium 10, Cadbury 11, Canterbury – Durovernum Cantiacorum 12

20

The End of Urbanism in Roman Britain

While many of the larger-scale Roman industries, most notably the pottery and metal extraction industries, were located in the countryside, towns were home to a range of craft specialists. One of the specialized crafts that has recently been identified on the basis of archaeological data is butchery. Seetah (2006) has shown that Roman butchery in London was a specialized industry involving a standardized tool kit and consistent methods for dismembering a carcass. Maltby (2010) has provided similar evidence for the civitas capital of Winchester. Even small towns, such as the Late Roman town of Icklingham in northwest Suffolk, had specialist butchers who dismembered cattle in standardized ways (Crabtree 2010a, 193). There are three questions that must be answered before we can assess the influence, if any, that Roman urbanism in Britain had on the rise of towns in the later Anglo-Saxon period. First we must determine whether occupation in Roman towns in Britain continued beyond the first few decades of the 5th century. Second, we must try to determine whether any such occupation was truly urban in character. Finally, we need to consider the relationship between these data and the broader questions of the survival of Romano-British culture in the post-Roman period. From a political and military perspective, Britain was facing a number of problems in the later Roman period. Jones (1998, 246)  identifies some of the issues that plagued Britain in the 4th century, including the fact that the northern frontier was never permanently pacified and that the Roman garrison in Britain was inadequate. There were also problems of internal security, and, in a broader sense, Britain was an overreach for the Romans. He argues that Late Roman Britain was destabilized both by frontier problems and by the actions of a series of ultimately unsuccessful usurpers who were supported by the people of Britain. Jones (1998, 255)  suggests that ancient empires, including the Romans, reach a point where increased expansion provides diminishing return and maintenance costs lead to increasing stress and demand for resources. He further suggests that Britain was never effectively Romanized (Jones 1998, 257). This last point is controversial, given the epistemological problems surrounding the concept of “Romanization.” The notion of Romanization implies that the relationship between the colonizer and the colonized is a one-way street. Recent research has provided examples from around the world showing that the relationship between the colonizers and the colonized is complex and multifaceted (see, for example, Stein 2005).

Introduction

21

Nevertheless, it is clear that the Roman Empire gave up on Britain about the same time as the Britons gave up on Rome. While historical sources for the early 5th century are few in number, they do suggest that Roman rule in Britain was effectively over by the end of the first decade of the 5th century. In 407 Constantine III apparently left Britain with most of the Roman army that had been stationed there. The British were forced to take up arms to defend their cities from barbarian attacks. In 410, the 6th-century Byzantine historian Zozimus reports that the emperor Honorius responded to a plea from the Britons with letters telling the civitates of Britain to see to their own defenses. Todd (1999, 208–9) sees these letters as an acknowledgment that the ties between Britain and the Roman Empire had been broken. Not all scholars are confident that the letters, in fact, refer to Britain (see, for example, Bartholomew 1982), but it is clear that the Romans made no attempt to reconquer Britain after 410. The study of 5th-century Britain provides some formidable challenges to archaeologists as well. The first and most basic problem is chronology. Roman sites are often dated using material culture, such as coins and pottery. Most of the Roman pottery industries in Britain appear to have ceased functioning early in the 5th century, so mass-produced Roman pottery cannot be used to date 5th-century sites in Britain. Coinage presents similar problems. The last small denomination coins that are widespread in Britain are those of the house of Theodosius, dating to 395– 402 CE. Many of these coins show extensive wear, but whether this is a result of long-term use or simply of poor manufacture remains an open question (Esmonde Clearly 2011, 18). There is another possibility as well. Gardner (2007, 147) notes that small denomination coins are often found in Late Roman rubbish deposits. Does this represent an active rejection of the Roman monetary economy by the inhabitants of Britain in the early 5th century? Did these coins become as worthless in the period after 410 CE as confederate dollars did in the United States after the Civil War? The alternative to the use of coins and pottery for dating purposes is the adoption of chronological techniques that have been used in prehistoric sites, including radiocarbon dating and dendrochronology. Traditional radiocarbon dating required about 25 grams of organic material, and the dates that resulted often had standard deviations that were too large to be useful in a historical context. The advent of accelerator mass spectrometry (AMS) radiocarbon dating (Hedges and Gowlett 1986) allowed archaeologists to date smaller samples, including items that lived and died in a single year such as single cereal grains or single tree rings, potentially reducing errors. AMS dates

22

The End of Urbanism in Roman Britain

have been used in the analysis of the Late Roman and Early Anglo-Saxon cemeteries at Queenford Farm and Berinsfield in Oxfordshire (Hills and O’Connell 2009) to suggest that the cemeteries were used sequentially rather than concurrently. Dendrochronology has been used less frequently in Late Roman and early medieval contexts. Many sites lack the well-preserved timbers that are required for tree-ring dates. Another problem is related to the nature of the archaeological evidence itself. As we will see below and in the next chapter, there is clear evidence for a shift from construction in stone to timber-building by the early 5th century in Britain. Archaeologically, the remnants of timber buildings are far more ephemeral than those of masonry structures, making 5th-century architecture far more difficult to identify in the material record. In many older excavations, traces of Late Roman timber buildings may simply have been missed if archaeologists were not looking for postholes, beam slots, and other evidence for timber architecture. However, excavations since about 1970 have generally been conducted using modern methods of excavation, so we should be able to distinguish between the absence of evidence for these timber buildings and a true absence of Late Roman and post-Roman timber construction. Clearly, the excavations at Wroxeter, described in detail below, have shown that timber structures can be identified when careful, open-area excavation is carried out. A review of the entire history of Roman urbanism is well beyond the scope of this volume. However, a major question for students of early medieval urbanism is whether one can identify a degree of continuity between Roman and early medieval urbanism. Is there, for example, evidence for continued activity in the Roman towns in the 5th and 6th centuries CE that might have served as the basis for the development of early medieval urbanism in the 7th and 8th centuries CE? In addition, we need to determine whether this activity was, in fact, truly urban in character, or whether the Roman towns continued as parts of the built landscape, but lost most or all of their urban functions after Britain ceased to be a part of the Roman Empire. The views about the fate of the Roman towns in Britain are varied. On the one hand, Henig (2011, 516) has argued that “not a single one of them died; all of them retained a real presence and vitality in the landscape until the seventh century at the least.” If Henig is correct, then the rebirth of towns in Britain owes much to the continued presence of Romano-British towns. On the other hand, Esmonde Cleary (1989, 161)  argued, “The generality of the archaeological evidence points to a sudden and total collapse of the Romanised way of life in the island in the generation or so after 411. In that time, the towns, the villas, the

Introduction

23

industries and the other material evidence diagnostic of Roman Britain disappeared.” An early review of the archaeological evidence for possible urban continuity into the 5th and 6th centuries reached similar conclusions (Brooks 1986). Brooks (1986, 99) argued that “the principal towns of Roman Britain were deserted by the mid fifth century, and remained so for at least a hundred years.” While Esmonde Cleary (2011) now agrees that the evidence for occupation of some of the RomanoBritish towns may extend further into the 5th century than was originally thought, he sees this as a change in chronology, rather than a change in interpretation. He now sees Roman Britain as an example of a failed state or a collapsed state (Esmonde Cleary 2011, 20). It is difficult to reconcile these two views of Late Roman towns. Either they continued as vital parts of the landscape until the 7th century, or they disappeared long before the appearance of the first post-Roman towns in the mid-7th century CE. One possible sign of urban abandonment has been the presence of the so-called “dark earth” in Roman towns in Britain beginning in the Late Roman period. Detailed studies of these deposits, both in the UK and on the continent, have shown that they can be caused by a variety of different factors (Macphail et al. 2003). In addition to abandonment, these deposits can be formed by middening and by low-intensity activities such as gardening and grazing. Most of these activities would indicate a decline in urban activities, but more of these deposits need to be examined using modern micromorphological techniques to determine whether they were formed as a result of the total abandonment of the site or whether they result from some less intensive activities. The results may differ from town to town, and even from area to area within a single town. What is missing from the discussions of late 4th-century and 5thcentury Roman towns in Britain is a model for urban decline. With certain caveats, I  would like to suggest that the history of the city of Detroit, Michigan can provide us with such a model. What was to become the city of Detroit was founded as a French fort and missionary outpost in 1701. By the early 20th century, it had grown to become a center of commerce and industry and one of America’s largest cities. In particular, it was home to the US automobile industry; the headquarters for the major American automotive manufacturers were located in the Detroit area. The US automobile industry prospered in the postWorld War II period due to pent-up demand and limited competition from foreign car companies. The major car companies and their subsidiaries provided many high-paying blue collar jobs in Detroit and the surrounding area.

24

The End of Urbanism in Roman Britain

Demographic data tell the tale of Detroit’s rise and fall. In 1950, Detroit was home to 1.85  million people. It was the fifth largest city in the United States; only New York City, Chicago, Philadelphia, and Los Angeles were larger. At that time, Detroit had the highest median income and highest percentage of home ownership of any major city in the US. However, Detroit’s prosperity would not last for long. The oil crises of the 1970s and increasing competition from Japanese, German, and other foreign auto manufacturers led to a decline in US market share at home and around the world. US automakers relocated their plants away from Michigan to states (mostly in the South) that had lower labor costs and to plants abroad. The decline in the US auto industry, along with “white flight” after the racial tensions of the 1960s and 1970s, had a major impact on Detroit’s population numbers. The 2010 census showed that Detroit’s population had been reduced to 714,000 and that the population had declined a full 25 percent between 2000 and 2010. So what lessons can be learned from Detroit’s decline? One of the most important is that standing buildings alone do not make an urban environment. The Packard plant still stands in Detroit, and the last car left its assembly line in 1956 (Figure 1.2). A plan for the redevelopment

Figure 1.2. The Packard Automobile plant in Detroit. The last car left its assembly plant in 1956. Photo credit: Frank Wulfers, reproduced with permission.

Introduction

25

of the site began in May of 2017, more than 60 years later. The Michigan Central Station remains a prominent landmark, even though it was abandoned as a railroad station in 1988. There are approximately 70,000 abandoned buildings in Detroit. While most of these structures are residential, they include 48 office buildings. Although these abandoned buildings can attract vermin and criminal activity, the city simply does not have the money to pull down these structures. Abandoned structures, especially stone structures, may have survived in Roman towns many decades and even centuries, but this alone is no evidence for a continued urban presence within these towns. Second, even though portions of Detroit’s urban core and a number of neighborhoods remain in use, many sections of the city are no longer viable. Today only about half of the 139 square miles (260 sq. km) that make up Detroit are actually in use. Areas where most of the houses have been abandoned and subsequently burned out or bulldozed have returned to prairie, and the transition has taken place quite rapidly (Figure  1.3). Other areas have been turned into urban gardens. Trees sprout from the tops of abandoned office buildings. The city of Detroit is even considering cutting off urban services from some parts of the city, although the decision to abandon certain neighborhoods is a politically difficult one. We need to bear these data in mind when we examine archaeological evidence from urban sites. Abandonment of one part of the city may not mean that the entire town was abandoned at that time. However, these data do show that abandoned residential areas can return to nature in much less than half a century. A third lesson is that the decline of a major industry can lead to urban decline even when the municipal, state, and federal governments continue to function. Many of those who remain in Detroit rely on government transfer payments, including Medicare, Medicaid, Social Security, unemployment insurance, food stamps (now known as SNAP or the Supplementary Nutrition Assistance Program), and AFDC (aid to families with dependent children), for all or part of their income. For example, in Wayne County, which includes Detroit, over 500,000 people were receiving food stamps in 2010.1 The situation in the British towns was undoubtedly even direr since one of the major functions of the Romano-British towns was the collection of taxes in coin and in kind (Esmonde Cleary 1989). With the withdrawal of Roman military and political power, a major raison d’être for many of these Roman towns was removed. It is reasonable to suggest that many of these towns may have experienced an even more rapid population decline than Detroit has 1

http://frac.org/wp-content/uploads/2011/01/urbansnapreport_jan2011.pdf

26

The End of Urbanism in Roman Britain

Figure 1.3. Many of the unoccupied parts of Detroit have returned to prairie. Photo credit: Wendy A. Carter, reproduced with permission.

experienced, since there was no longer a central government supporting these urban residents. In addition, the Detroit example shows that local economic changes can lead to substantial population decline. We do not need to call on external forces such as warfare or pestilence. By analogy, it is reasonable to suggest that many of the Romano-British towns may have failed on their own; we do not need to blame Anglo-Saxon immigrants, barbarian threats, or epidemic diseases. Of course, Detroit is not a perfect analogy for Romano-British towns and cities. Detroit is much larger, both in terms of population and in terms of area, than any town or city in Roman Britain. Imperial Rome itself may have been home to only half a million people (Storey 1997), although estimates of about a million residents are common (see for example Hodges and Whitehouse 1983, 48–52). In addition, Detroit continues to exist, albeit in a much reduced form. However, given the high unemployment rate, which some have estimated to be as high as 50 percent of the

The Archaeological Evidence

27

workforce,2 and the fact that nearly half the adult population of Detroit is functionally illiterate,3 it is questionable whether the city would continue to exist without government transfer payments such as unemployment insurance and Social Security. Despite these differences, Detroit may provide some insight into the decline of urbanism in Late Roman Britain. The Archaeological Evidence In order to explore the possibilities for urban continuity between Late Roman and post-Roman Britain, we will begin by examining three Romano-British cities that have had substantial programs of excavation since the 1960s – Winchester, London, and Wroxeter (Figure 1.1). Winchester (Venta Belgarum) was chosen because it was home to the first large-scale urban excavations in Britain in the post-World War II period, although the results of some of these excavations remain unpublished. London (Londinium) was chosen because it was the largest city in Roman Britain, and rescue excavations carried out by the Museum of London and others have revealed much of its Roman and postRoman history. Wroxeter (Viroconium), the fourth largest city in Roman Britain, was chosen because careful open-area excavations and gradiometer surveys have revealed evidence for 5th–6th-century occupation of portions of the city. We will then briefly examine the possible evidence for continued occupation at a number of other Romano-British urban sites. The chapter will conclude with an evaluation of the nature of the occupation in Roman towns during the 5th and 6th centuries to determine whether this occupation should be considered urban and the extent to which these settlements may have affected the development of urban centers during the Middle and Late Anglo-Saxon periods. Winchester Winchester or Venta Belgarum was the fifth largest town in Roman Britain. This civitas capital served much of central southern England. The defenses were laid out in 70 CE, and the street grid was established in the 1st and 2nd centuries. A major program of excavation began in Winchester in the early 1960s under the direction of Martin Biddle (1990). At the beginning of the program, most scholars thought that the modern street grid in the city preserved its Roman layout. The excavations that were carried out within the town between 1961 and 2 3

www.huffingtonpost.com/2009/12/16/detroits-unemployment-rat_n_394559.html www.huffingtonpost.com/ 2011/ 05/ 07/ detroit- illiteracy- nearly- half- education_ n_ 858307.html

28

The End of Urbanism in Roman Britain

Figure  1.4. Photograph showing a Roman street at the Brook Street site in Winchester in 1971. The Late Saxon and modern streets are located to the left (east) of the Roman one. Photo credit:  Pam J.  Crabtree, reproduced with permission of the Winchester Excavation Committee.

1971 showed that Winchester was extensively replanned in the 9th century, probably under Alfred the Great (see Chapter 4), and that the Late Saxon city layout differs significantly from the Roman grid plan that underlies it (Figure 1.4). The Early Roman buildings in the town were timber, but they were replaced by stone construction in the 2nd century. The town included town houses with mosaic floors and hypocausts. With the exception of the forum and the temple, which were located on the tufa island in the former channel of the Itchen River, little is known of the location of the public buildings. Extensive suburbs developed to the north of the town along the roads to Cirencester and Silchester. In

The Archaeological Evidence

29

addition, there were extensive extramural cemeteries which have been examined archaeologically since 1972 (see, for example Booth et  al. 2010; Stuckert 2016). While Winchester was initially unfortified in the 1st century CE, construction of earth and timber defenses began in the Flavian period (69–96 CE). The earthwork defenses were augmented in the late 2nd century, and a stone wall surrounding the town was constructed in the 3rd century CE. Archaeological data from the extensive excavations at Winchester show that the character of the town began to change around 350 CE. There is no evidence for the construction of any new town houses after about 350 CE, and the existing housing stock gradually fell into disrepair (Qualmann et  al. 2010, 8). By the mid-4th century, dark earth had begun to form, and the street grid was in poor shape. However, the extramural cemeteries continued in use throughout the 4th century, and evidence for metal-working actually increases during the 4th century. The 4th-century Notitia Dignitatum4 records a gynaecium (state textile mill) at Venta, probably, but not certainly, Venta Belgarum, and the cemetery data from Lankhills suggest that Winchester was home to a diverse population in the 4th century, including some individuals whose isotopic profiles are non-local (Booth et al. 2010, 528). In addition, the presence of crossbow brooches amongst the grave goods at the Lankhills cemetery suggests that military or other official personnel were present in the town. There is no obvious parallel for this combination of military and industrial activity in Roman lowland Britain (Booth et al. 2010, 528). The data suggest that Winchester changed from an administrative center to an industrial center during the later 4th century CE. One possibility is that the gynaecium and the metal shops may have been producing uniforms and weapons for the Roman army. Alternatively, late 4th-century Winchester may have been a fortified center for collection and storage of taxes in kind (Booth et al. 2010, 5). Despite the evidence for significant occupation in the late 4th century, Winchester’s urban character did not last much past 400 CE. There were attempts at metalling the streets south of the forum after the 360s– 370s (Biddle 1970, 312–13), but the use of the extramural cemeteries ends abruptly around 400 CE (Booth et  al. 2010, 5–6). By the early 5th century, the town houses were reduced to shanties (Qualmann et al. 2010, 8). While some authors have suggested that Winchester may have continued to represent a center of authority in the post-Roman period 4

Literally a “list of offices,” the Notitia Dignitatum is a Late Roman administrative document.

30

The End of Urbanism in Roman Britain

(Biddle and Kjolbye-Biddle 2007; see also Todd 1999, 211) there is no evidence that it continued to exist as an urban entity beyond the beginning of the 5th century. Although there is some possible 5th- and 6th-century settlement within the walls of Roman Winchester, the character of that settlement is radically different from the earlier Roman town. For example, the ditch of the Roman road on the Brook Street site appears to have been used for some form of horticulture (Biddle and Kjolbye-Biddle 2007, 197). Only one of the five Late Roman east–west streets remained in use, and only two of the seven north–south streets were in use between the 5th and the 8th century. While Winchester may have been the center of an agricultural estate by the Middle Saxon period, there is nothing in the 5th- to 8th-century record to suggest that it retained its urban character. London London was the largest city in Roman Britain and served as the capital of the province. The question of Roman/post-Roman continuity is a critical one, since London was not only the largest city in Roman Britain, but it was also the site of a wic (see Chapter 3) in the Middle Saxon period, and it eventually developed into the largest city in medieval and modern England. Twentieth-century scholars, including Mortimer Wheeler (1934, 290) and Martin Biddle (Biddle et al. 1973, 18), thought that the walled city of London was never completely abandoned in the post-Roman period (5th–6th centuries). However, detailed excavations that have been carried out within the Roman city now suggest otherwise. Serious archaeological work began in London in the aftermath of World War II (Biddle et al. 1973) when sites that had previously been covered with buildings were exposed for the first time. Work has continued for the past 60 years in the form of rescue archaeology in advance of construction. Much of this work has been conducted under the auspices of the Museum of London. The history of Roman London differs somewhat from other large and smaller towns in Roman Britain, in that London appears to have reached its maximum period of prosperity in the 2nd century CE (Esmonde Cleary 1989, 82). At that time it served as the principal port of Britain, and archaeological data show that massive docks and waterfront structures were constructed then (Milne 1985). Imports included terra sigillata and amphorae. However, trade from the other western provinces began to dry up by the end of the 2nd century. Archaeologically, this decline is reflected in the sharply decreased numbers of buildings, pits,

The Archaeological Evidence

31

wells, and animal bones that can be dated to 150–400 CE in Roman London (Marsden and West 1992; Cowie 2008, 50). Urban contraction can also be seen in the form of “dark earth,” which covers the most recent Romano-British occupation levels. While dark earth formation may result from a number of different causes, environmental analyses indicate the dark earth within the city of London was caused by weathering, root action, burrowing animals, and worms (Macphail 1998; Macphail et  al. 2003). Pollen grains and phytoliths recovered from the dark earth are consistent with those of a grassland or urban wasteland environment, indicating abandonment rather than land used for cultivation. Macphail (2015) recently suggested that the Late Roman roads in London show a rural signature, characterized by soils, sands, and dung. This process has obvious parallels to the process of urban abandonment seen in Detroit today (Figure  1.3), where grassland and prairie have overtaken former residential areas in less than 50 years. Cowie (2008, 50) suggests that London must have been abandoned around the time that Roman rule in Britain ceased (ca. 410 CE) or shortly thereafter. Artifactual evidence from the early post-Roman period is equally sparse. The only well-stratified and well-dated 5th-century artifact that has been recovered from within the walls of Londinium is a quoit brooch. The brooch was found lying among fallen roof tiles in the former Roman bathhouse at Billingsgate. This mid-5th-century brooch is not necessarily evidence for Anglo-Saxon settlement within the former Roman town. It is more likely that it was dropped by someone looking for salvageable tiles or other reusable Roman material. As Plouviez (1985, 85)  has noted, the Late Roman pottery found at the Early Anglo-Saxon site of West Stow appears to have been scavenged, based on the unusually high number of bases found in the Early Saxon contexts. This scavenging continued through the late 6th century, long after the nearby Late Roman town of Icklingham was abandoned. It is not unreasonable to suggest that the inhabitants of post-Roman Britain, whether British or Saxon, may have scoured the former Roman city of London for usable materials. In Detroit and elsewhere, abandoned homes are often stripped of their copper piping and of other valuable building materials and appliances. Cowie (2008, 51)  also suggests that various factors could have led to a precipitous decline in the population of the London region in the 5th and 6th centuries, including plague, warfare, and climatic deterioration. However, it may not be necessary to resort to such exogenous factors to explain much of London’s decline. London’s role as a port city began to decline early in the Late Roman period, and its role as

32

The End of Urbanism in Roman Britain

an administrative center would have largely disappeared with the withdrawal of Roman military and political power in the early 5th century. Residents of premodern cities usually experienced higher mortality than their rural counterparts (Scheidel 2001, 15), and ancient cities relied on immigration to sustain their populations. As London’s role in trade and administration declined, there would have been far less to attract rural residents to the city. While disease, warfare, and environmental changes may have played a role in London’s decline, much of its desertion may have resulted from people “voting with their feet.” Rural residents no longer wanted to relocate to London, and the residents of London died out or moved elsewhere. This could easily take place within a generation or two. The latest known Roman activity in central London comes from the site of St. Martin in the Fields, a mile (2 km) west of the Roman city. The site lies on the western edge of what was to become the AngloSaxon town of Lundenwic. Recent excavations have uncovered a heavy Roman sarcophagus, dating from the late 4th or early 5th century CE. The find is interesting because it is contemporary with St. Martin himself, who was a Roman soldier who converted to Christianity and died in 397 CE. A Roman tile kiln, dated using archaeomagnetics to the first half of the 5th century, has also been recovered from the site. It is the latest Roman structure discovered in London to date, and its presence suggests that there may be a large building nearby. Seventhcentury Anglo-Saxon burials have also been found at the site. The presence of the brick kiln suggests that some aspects of Roman life survived into the 5th century in the London region. While this site may well have served as a cult site throughout the post-Roman period, it cannot be used to make the case for continuous urban occupation of Roman London.5 Wroxeter The Late Roman and post-Roman history of Wroxeter differs markedly from that of either Winchester or London. Wroxeter is located in Shropshire, near the Welsh/English border. The Roman settlement there was the fourth largest town in Roman Britain, and the “old work” is the largest remaining example of Roman civilian architecture in Britain (Figure  1.5). The town was initially established as a Roman legionary fortress, but it was transformed into a civilian town in the late 1st century CE. The nucleus of the city lay within the former fortress and included 5

www.museumoflondon.org.uk/Explore-online/Past/MissingLink/Excavations/

The Archaeological Evidence

33

Figure 1.5. Wroxeter “old work.” Portions of the support for the hypocaust of the Roman baths can be seen in the foreground. Copyright: Historic England, reproduced with permission.

public buildings such as a forum, baths, and a mansio (a building associated with the imperial postal system). The forum and the baths were constructed in the early 2nd century. The town covered an area of about 78 ha. A complete geophysical survey of the town has identified over 260 Roman-period buildings, including houses of the lower-class inhabitants and possible industrial areas. The town walls were constructed in the late 2nd century and refurbished in the 4th century CE. During the past 40 years, there has been increasing interest in exploring the possibility that urban life in some Romano-British towns continued into the 5th century and possibly even longer. Based on the large-scale excavations at Wroxeter, White and Barker (1998) have suggested that the town was occupied into the post-Roman period, as late as the 6th or possibly 7th century CE. The archaeological evidence indicates that the baths basilica was maintained into the 5th century CE, although the structures were quite dilapidated (White and Barker 1998, 119). The basilica was subsequently closed to the public, and a number of small, flimsy wattle-and-daub structures were constructed inside the building. The structure appears to have been used for industrial purposes at this time. The roof of the basilica was subsequently taken down, and the shell of the building appears to have been used as an open-air market, along

34

The End of Urbanism in Roman Britain

Figure  1.6. Reconstruction of the mid- to late 6th-century timber buildings at Wroxeter. Copyright: Historic England, reproduced with permission.

with a municipal office or tax collection point, in the early 6th century (Barker et al. 1997, 230). In the mid- to late 6th century, much of the remaining shell of the basilica was demolished, and a series of rubble platforms was constructed. White and Barker (1998, 121) estimate that up to 250 cubic meters of rubble were used to construct these platforms, and they suggest that these platforms served as foundations for a large timber-framed building (Building 10), and a series of smaller timber buildings (Figure 1.6). The building platforms appear to have been laid out in Roman feet (White and Barker 1998, 124), which has often been interpreted as evidence for continuity between Late Roman and Dark Age Britain. It should be noted, however, that the English measurement system is still in use in the US (even though it was abandoned in Britain over 40  years ago), and we still measure nails by pennyweight (d). The US, of course, has not been part of the British Empire since the late 1700s. The faunal remains from Wroxeter (Hammon 2011) do suggest some degree of continuity between Late Roman and sub-Roman Wroxeter. Throughout the sequence, the faunal assemblages are dominated by the remains of cattle and pig, and there is no clear evidence for size changes

The Archaeological Evidence

35

between Late Roman and sub-Roman Wroxeter. It should be noted, however, that the Wroxeter cattle never showed the size improvements that are commonly seen in Roman cattle from southeastern England (Hammon 2011, 16; see also Rizzetto et  al. 2017). One of the most striking finds is the proximal phalanx of a Barbary ape (Macaca sylvanus) dated to the 6th or 7th century CE, which may indicate continuing contact with the Mediterranean world (Hammon 2011, 16). The artifactual evidence also includes a Palestinian oil or wine jar from the same time period (White and Barker 1998, 128). The evidence for continued occupation of Wroxeter into the postRoman period raises several interesting issues. The first, of course, is who was occupying this settlement after the end of Roman rule. White and Barker (1998) suggest at least two possibilities. The first is that it may have been home to a British tribal leader, someone who had the leadership ability to oversee the construction of the timber platforms. This leader might have seized power in the aftermath of the Roman withdrawal and may have been one of the tyranni (tyrants or leaders) mentioned by Gildas, a 6th-century British cleric writing in Wales. Alternately, Wroxeter might have been the seat of a bishop. White and Barker (1998, 125) note that bishoprics were established in most Roman towns in the 4th century and that many may have become self-perpetuating. If this were the case, then post-Roman Wroxeter can be seen as a parallel to many of the Gaulish towns that survived as walled centers of power for bishops. There is one major difference between the Gaulish towns and Wroxeter, however. Most of the Gaulish bishoprics survived into the Middle Ages and modern times. Wroxeter, however, appears to have been abandoned sometime in the 7th century. There is no evidence for a catastrophic end to the town. Many of the buildings were systematically dismantled, but Building 31 was left to collapse (White and Barker 1998, 136). Whether this abandonment is related to the increasing Anglo-Saxon presence in western Britain in the 7th century remains an open question. However, the important point here is that Wroxeter did not survive to become an early medieval town, so it is unlikely to have had any influence on the redevelopment of towns after 700 CE. Recent reanalysis of the stratigraphy, chronology, and artifacts recovered from Wroxeter has cast some doubt on the survival of Wroxeter as an urban center into the 6th and 7th centuries (Lane 2014). Lane (2014, 505–6) notes that most of the timber buildings at Wroxeter were based on recognizing patterns in the rubble spreads, rather than by evidence for postholes and post pads. In addition, none of the radiocarbon dates can be used to support the chronological sequence proposed by White and Barker (Lane 2014, 509). Moreover, Wroxeter lacks any evidence for

36

The End of Urbanism in Roman Britain

the imported pottery and glass that are typically found on 5th- to 7thcentury sites in western Britain (Lane 2014, 510, Fig. 4). While Lane’s (2014) analysis does not preclude some 5th-century activity in Wroxeter, it does cast doubt on Wroxeter’s survival as an urban center into the 6th and 7th centuries CE. Other Roman Towns In addition to the large scale-excavations that have been carried out at Wroxeter described above, a number of other towns have provided evidence for some degree of activity continuing into the 5th century. The evidence includes data from Lincoln, Silchester, Bath, and Verulamium (see Henig 2011 and Speed 2014 for an up-to-date review of the evidence), as well as the evidence from St. Martin in the Fields, London, discussed above. The evidence from Lincoln is difficult to interpret. Excavations conducted at the former church of St. Paul in the Bail (M. J.  Jones 1994) revealed an archaeological sequence that extended well before the Norman Conquest. At the base of this sequence were two structures that were probably built of timber and located in the courtyard of the Roman forum and that may have been churches. Coin evidence suggests that these structures are no earlier than the 4th century, but radiocarbon dates are consistent with a Late Roman, sub-Roman, or even a post-Roman date for these buildings. It is possible that these buildings are part of a bishopric established in the late 4th century. The church may have been moved to this location in the early 5th century, when the bishop became “head of a dwindling population eking out a different sort of life within the former city” (M. J. Jones 1994, 339). Alternatively, M. J. Jones (1994, 340) has suggested that this may be “the creation of a cult and political centre by a British ruler controlling at least the area of the Roman city, and modelling himself to some extent on Roman lines.” Gilmour (2007) has noted that only two of the associated burials appear to date to the 5th century; the rest of the burials are 7th–9th century in date. Regardless of whether these possible churches are Late Roman or sub-Roman in date, there is no evidence from these excavations to suggest that urban life in Lincoln continued beyond the first decades of the 5th century. The data suggest that we are looking at a cult center instead. Recently, Green (2012) has used historical, linguistic, and archaeological evidence to suggest that Lincoln may have been the center of a small British polity in the 5th and 6th centuries. She argues that the cremation cemeteries that surround the town of Lincoln may represent Germanic mercenary troops who were invited to Britain to provide security for the Lincoln polity and region (Figure  1.7). While Green

The Archaeological Evidence

37

Figure 1.7. Map showing the location of the Pagan cremation cemeteries surrounding sub-Roman Lincoln. Redrawn from Green (2012, Fig. 11, p. 63) with permission of the author.

38

The End of Urbanism in Roman Britain

makes a strong case for Lincoln as the center of a post-Roman polity, she does not see the town of Lincoln as maintaining its urban character into the 5th and 6th centuries. The data from Verulamium Insula XXVII have been the subject of debate for many years. Frere’s (1983) 1955–61 excavations there revealed a 22-room town house, Building 2, which included three with polychrome mosaics. The rooms opened onto a palisaded corridor and ranged around a garden or courtyard. The mosaics were initially dated to the late 4th century, based on their association with late 4th-century coins. One of the mosaics was subsequently cut by what was interpreted as a corn-drying oven, and a barn was built over the remains of the town house. A Roman-style water pipe was subsequently installed at the site. These data were interpreted to suggest that town life in Verulamium must have continued into the mid-5th century CE. Neal (2003) has re-examined the evidence. He argues that the mosaics are much earlier than the late 4th century. While he admits that “activity in the area possibly did continue into the late fourth, early 5th century” (Neal 2003, 201), he argues that the data from Insula XXVII cannot be used to argue for the continuation of urban life in Verulamium into the mid-5th century. Frere (2011) has recently responded to the critiques of the dating of House 2. He argues that the coin finds provide a terminus post quem of 380 CE for the construction of this large, high-status house, but he notes that this is the only part of the Insula XXVII/2 sequence for which direct dating evidence exists. He defends the long chronology for this part of the site by noting that the Second Period mosaic in Room 15/15 lasted long enough to require repair, that the stoke hole of the corn dryer that was inserted into the mosaic lasted long enough to require repair, and that this was all followed by the collapse of the house, the construction of a barn, and the subsequent construction of a water pipe. That there may have been occupation in Verulamium that extended into the first half of the 5th century should not surprise us. The question that must be raised is whether the water pipe line, dated by Frere to ca. 455–70, “implies the survival of some civic authority or organization” (Frere 2011, 8), and whether that in turn means that Roman urban life continued in Verulamium into the mid-5th century. Silchester, the Roman town of Calleva Atrebatum, was another cantonal capital. It is located in Hampshire, midway between the modern towns of Basingstoke and Reading. Like Wroxeter, it was not home to a medieval or a modern town. Although the Roman walls remain to a height of over 4 meters in some places, most of the interior of the former Roman town is in fields today. Excavations at the former Roman

The Archaeological Evidence

39

town were initially carried out in Victorian times; a modern series of excavations has focused on Insula IX, in the center of the Roman city. The modern excavations have been carried out by Reading University since 1997 and have focused on the history of Roman life within the town (Fulford et al. 2006). Recent excavations have shown that Insula IX was not abandoned before the 5th century. A Roman column which was reused as a property marker in the 5th century was recovered from the Reading University excavations. The marker has an inscription in ogham, a form of writing that developed in Ireland in the late 4th century. Evidence suggesting that the abandonment of Silchester may have been deliberate includes the infilling of some of the 4th-century wells. The excavators have argued that all three properties on Insula IX were occupied into the early 5th century (Fulford et al. 2006, 273), although the intensity of occupation clearly declines by the mid-4th century. However, they note that “Calleva cannot be regarded as a town in a Roman sense much after the middle decades of the fifth century.”6 Another Roman town with possible evidence for post-Roman activity is Bath (Aquae Sulis) in Somerset. The town is best known historically for its hot mineral springs and for its Roman-period temple to Sulis Minerva. The baths and temple precinct were extensively excavated by Cunliffe and Davenport (1985) between 1978 and 1984. Their excavations showed that the surface of the inner temple precinct was resurfaced around 300 CE (Period 4). This was followed by a series of repaving levels (Levels 5a–f), and the deliberate demolition of the temple in Period 6. The excavators suggested that Period 5 began in the late 4th century, but could have continued into the late 5th or possibly the early 6th century (Cunliffe and Davenport 1985, 75). The chronology of the temple precinct was recently re-examined by Gerrard (2007), using stratigraphy, artifactual evidence, and new radiocarbon dates on animal bone remains from Period 5. Gerrard suggests that Period 5 begins around 350 CE and ends around 100 years later. He suggests that the destruction of the temple in Period 6 took place in the later 5th century, probably around 450 but before 500 (Garrard 2007, 160). This destruction is unrelated to the end of Roman Britain in the early 5th century and to the foundation of a nunnery on the site in the 7th century CE. Garrard (2007, 161) suggests that Phase 5 represents “episodic activity taking place with reference to the ever more dilapidated monumental structures.”

6

www.silchester.rdg.ac.uk/guide/end

40

The End of Urbanism in Roman Britain

While this evidence certainly suggests that there was continued activity in some Romano-British towns into the first half of the 5th century and perhaps later, what archaeologists must determine is whether this activity represents a true continuation of Roman urbanism or whether this simply represents low-level activity in towns that had lost their urban character by the first few decades of the 5th century CE. The question is whether we are seeing some life in the Roman towns or whether we are actually seeing town life continue into the post-Roman period (Wacher 1995, 408; Gerrard 2007, 149). The archaeological data suggest that we are seeing several different patterns.The first is represented by cities like London that appear to have lost their urban character by the early 5th century. They accord with the model proposed by Esmonde Cleary (1989, 146)  over 25 years ago, when he suggested that the large Roman towns in Britain “were abandoned not long after the latest Roman coins to reach Britain became current.” At that time, he suggested a date in the first third of the 5th century for the abandonment of the Roman towns and cities in Britain. Using the Detroit analogy, if we assume that one of the primary functions of the Roman towns and cities in Britain was the collection of taxes to support the Roman state and military, then the disappearance of this industry should have led to a contraction of these sites. However, depopulation would not have happened overnight, since the towns also served as markets, centers for specialized production, and local administrative centers. We might expect the depopulation to take a generation or two if it were not forced by warfare or similar circumstances. A second pattern is apparent in centers like Lincoln and Bath. In those centers, religious buildings appear to have continued in use into the subRoman period. For example, Gilmour (2007, 254) suggests that the later of the two possible churches underlying St. Paul in the Bail may not have been demolished until about 650 CE. However, these towns lack evidence for traditional urban life. They lacked sizeable, dense populations that included substantial numbers of people engaged in non-subsistence pursuits. It is possible that they continued as cult centers or even centers of authority, but they seem to have lost their urban character by the middle of the 5th century, if not earlier. At a more pedestrian level, they may resemble the churches that survived in otherwise deserted or shrunken medieval villages (Figure 1.8). Finally, there is the case of Wroxeter, where White and Barker (1998) have argued for continuing settlement in the sub-Roman period. Certainly, the character of settlement is different from what was seen at Wroxeter during the first four centuries CE. However, Wroxeter retains many characteristics of earlier Roman towns and cities. Its

The Archaeological Evidence

41

Figure  1.8. Photograph of the ruins of St. Martin’s Church at the deserted medieval village of Wharram Percy in North Yorkshire. Copyright: Historic England, reproduced with permission.

market appears to survive into the post-Roman period, and the “Great Rebuilding” in the 6th century suggests that it may still have been home to specialized craftspeople (White and Barker 1998, 128). This activity may have been supported by anona or taxes in kind drawn from the town and its immediate surrounds. As Hammon (2011, 24) notes, “Perhaps we should accept that no single model is adequate to explain the myriad of responses to the demise of Roman Britain.” On the other hand, Lane (2014) has challenged the evidence for continued urban occupation at Wroxeter in the 6th and 7th centuries. Even if the original interpretation of Wroxeter is correct, I would argue that Wroxeter may well be the exception that proves the rule of urban decline in the post-Roman period. While much of the evidence for catastrophic decline comes from sites in southeastern Britain, the case of Cirencester in Gloucestershire is illustrative as well. Cirencester was the second largest city in Roman Britain, home to perhaps 15,000–20,000

42

The End of Urbanism in Roman Britain

people at its height.7 Like Wroxeter, it is located in western England, and at the beginning of the 4th century CE it was made the administrative area of Britannia Prima, which included most of southwestern England and Wales. It was also home to a thriving mosaic industry; at least 80 mosaics have been recovered from the remains of the city (p. 9). The 5th-century history of Cirencester is very different from that of Wroxeter. By the mid-5th century, all the great buildings of Cirencester were in ruins, and there is no evidence for any new construction within the city walls for at least 300 years (p. 11). The evidence that does exist for 5th-century activity comes primarily from the Roman amphitheater, which was located outside the city walls. By the 5th century, it was no longer used for performances and spectacles. Instead the entrances to the structure appear to have been narrowed, and a ditch was dug along the southern side, turning the amphitheater into a stronghold. Timber buildings of 5th-century date have also been recovered from the amphitheater.8 In many ways, what we see at Wroxeter and Cirencester has parallels to some of the Iron Age hillforts in southwest Britain that were reoccupied in the 5th century. One of the best known is Congresbury–Cadbury, which was excavated between 1968 and 1973 (Rahtz et al. 1992). Excavations revealed that the Iron Age defenses of the site were refurbished in the 5th century. The sub-Roman settlement itself includes circular and rectangular structures, as well as evidence for metal-working and exotic imports such as glass from the Rhineland and Mediterranean olives. The settlement activity dates between 450 and 550 CE, and it is paralleled by the evidence for the reoccupation of a number of other hillforts in Somerset (Burrow 1981). These hillforts, like Wroxeter, may well have been home to local British chiefs. The sites are located in the west of Britain, where the Anglo-Saxon presence is limited before the 7th century CE. Most show some links to the Roman past. In the case of Wroxeter, it is the use of the Roman foot in the layout of the structures; at Congresbury–Cadbury, it is in the continuing contacts with the Roman world. However, with the exception of Wroxeter, none of these settlements was truly urban in character, and more importantly, none led to the development of a town or city after 700 CE. If, on the other hand, Wroxeter represents the seat of a bishop, then it is one British example of a pattern that was widespread in Gaul. In postRoman Gaul the towns remained as fortified centers for bishoprics and thereby maintained at least some of their urban functions. Whether these 7

8

Cirencester Archaeological Strategy 2006, p.  10:  www.cmis.cotswold.gov.uk/CMIS WebPublic/Binary.ashx?Document=11426 www.english-heritage.org.uk/daysout/properties/cirencester-amphitheatre/history-andresearch/

Recent Approaches to Late Roman Urbanism in Britain

43

sites can truly be seen as urban or whether they should be seen as ceremonial centers is certainly an open question. Where Wroxeter and the Gaulish/Frankish towns differ is that the Frankish centers developed into towns once again in the Carolingian period, while Wroxeter was deserted by the end of the 7th century CE. Recent Approaches to Late Roman Urbanism in Britain In 2011, Adam Rogers published a study of Late Roman towns in Britain, focusing on the construction of new structures and industrial activities carried out within public buildings in Roman towns in the 3rd and 4th centuries CE (Rogers 2011). He prefers the term “town” to city (Rogers 2011, 3), and he draws his data from three classes of Roman towns (Latin oppida) – coloniae, municipia, and civitas capitals. Rogers suggests that meaningful new activities were carried out in the forum–basilica complexes of Roman towns during the 3rd and 4th centuries CE, including both new constructions and industrial activities (Rogers 2011, 179). He notes, for example, the evidence for metal-working within the Roman basilica at Silchester (Fulford and Timby 2000). Rogers’s main goals, however, are more broadly theoretical. He attempts to challenge the notions of decline in Late Roman towns, arguing that “Rather than viewing urbanism in terms of greatness and decline, the Roman towns, as they changed over time, were successful expressions of the sites in their landscapes” (Rogers 2011, 182). As such, he rejects an economic or political view of urbanism, favoring instead the long-term importance of places in the landscape. He argues that the status of a place is more important than the physical conditions of the settlements themselves (Rogers 2011, 3, 38–41). The problem that I see with this approach is that the Late Roman cities may have remained important places on the landscape while at the same time losing many of their urban characteristics, such as a large population or evidence for non-agricultural specialists. For example, Winchester may have served as the center of an agricultural estate from the 5th through the 8th centuries CE, but this is not evidence for its continued function as a town. Speed (2014) approaches some of the same questions from an empirical perspective. Drawing on a broader range of archaeological data, he asks 1) whether Roman towns survive into the 5th and 6th centuries CE, and 2) whether the decay seen in Roman towns begins before the 5th century CE (Speed 2014, 2). As such, his study examines the history of nine Roman towns from 300 to 600 CE. This study is important because he draws on much of the archaeological “grey literature,” data that exist as reports or archive material but have not been published in books, monographs, and journal articles.

44

The End of Urbanism in Roman Britain

Box: Oppida The term oppidum (plural oppida) is a familiar one to students of later European prehistory. It was used by Julius Caesar to describe the large, fortified sites that served as political capitals of various tribes that he encountered in Gaul in the mid-1st century BCE. Archaeologically, the oppida (Figure 1.9) are described as large fortified sites up to 600 ha in area that served a variety of functions in late pre-Roman Europe. In addition to serving as political capitals for a number of Iron Age tribes, these sites also served as centers of craft production, industrial activity, and trade. Evidence from the site of Manching in Bavaria, Germany, indicated that these pre-Roman towns had planned layouts and substantial residential populations. They may even have served as residences for the Iron Age leaders, although the archaeological evidence for this is more limited (Bintliff 1984, 195–200, n. 67; see also Collis 2004). These Iron Age oppida are often seen as the first evidence for urbanism in Europe north of the Alps (see, for example, Wells 1984). Many of the oppida are located on the European continent, but the term has also been used to describe some of the largest settlements in Britain during the late pre-Roman and Early Roman period, e.g., Camulodunum (modern Colchester). For an up-to-datereview of the British oppida, see Pitts (2010). The term oppidum is also used to describe three classes of Roman towns (Rogers 2011, 4).The first was a colonia or colony.The residents of these towns were Roman citizens, and the town populations also included substantial numbers of retired members of the Roman military. Both York and Lincoln were classified as coloniae. The second category is the municipium. The residents of these towns were not automatically Roman citizens. Verulamium (Roman St. Albans) is the one recognized municipium in Roman Britain. The third class of towns were the so-called civitas capitals (Wacher 1966), including towns such as Winchester, which served as administrative and market towns for the various tribal groups in Roman Britain. The Roman name for Winchester, Venta Belgarum, probably means market of the Belgae.

45

Figure 1.9. Map showing the distribution of the Late Iron Age oppida in temperate Europe. Redrawn after Wells (1999, Fig. 6). Copyright: Princeton University Press, reproduced with permission.

46

The End of Urbanism in Roman Britain

Although Speed is a strong advocate for activity in Roman towns during the 5th and 6th centuries, he concludes that while some of these towns served as power bases for smaller regional polities, sub-Roman urbanism was “not ultimately successful” (Speed 2014, 125). While Late Roman towns may have remained meaningful places in the 5th and 6th centuries, the nature of these settlements is ruralized, more on the scale of the kinds of settlements that are known from 5th- and 6th-century rural sites (see Chapter 2). This is part of a process that Brogiolo and Chavarria (2015) have described as the “suburbanization” of Roman cities in the Mediterranean in the 4th through 8th centuries. They note that Roman towns in the Mediterranean show evidence for craft production within public buildings and evidence for agriculture during this period. Of the nine Late Roman towns surveyed by Speed, the one with the greatest evidence for 5th- and 6th-century activity is Canterbury. In 4thcentury Canterbury, timber structures occupy larger, earlier masonry buildings and are also built over earlier buildings (Speed 2014, 39). Speed identified just over 40 5th- to 8th-century buildings within the Roman town of Canterbury. Some are sunken-featured buildings (see Chapter  2) within decaying Roman buildings, and most are located close to the Riding gate. Interestingly, the only place where the Roman street plan survives in Canterbury is around the town gates (Speed 2014, 40). One of the biggest problems in the interpretation of the data from Canterbury is that it is not clear whether the town was deserted and then repopulated in the mid-5th century. A  reconstruction of Canterbury in the 5th century is shown in Figure 1.10. Conclusions So how do we envisage Britain in the first half of the 5th century? Until the 20th century, scholars relied primarily on historical accounts such as that of the 6th-century Welsh cleric Gildas who suggested that Romanized Britons were driven from the cities of Britannia by the arrival of the Anglo-Saxons (see Chapter 2). Clearly this is not the case. The archaeological data point to the gradual decline of most Roman towns and cities, rather than any form of military conquest. This decline is clearly related to the overall decline in the economy of the Roman Empire in the mid1st millennium CE (see McCormick 2002).9 In the years following World War II, a number of scholars suggested that Roman towns may have continued in some form well beyond the Roman military withdrawal 9

McCormick (2002:  115) has argued that the “economy which sustained the Roman Empire collapsed” between 250 and 650 CE, but he notes that the process of decline was not necessarily uniform or even unidirectional in the short term.

Conclusions

47

Figure  1.10. Reconstruction of sub-Roman Canterbury (from a painting at the Canterbury Roman Museum). Copyright: Canterbury Museum, reproduced with permission.

from Britain in the beginning of the 5th century. This view certainly colored Frere’s interpretation of Insula XXVII at Verulamium, and it lay behind Biddle’s excavation program at Winchester since Winchester was then thought to retain its Roman street plan. While recent research has indicated that activity in some of the Roman towns continued into the 5th century, evidence for continuity in urban functioning is still lacking with the exception of Wroxeter. Many Roman towns in Britain had clearly lost most of their population by the first decades of the 5th century, and with the withdrawal of Roman imperial power, they lost their role as centers for the collection of taxes. This decline should not be blamed solely on catastrophic factors such as political instability, epidemic disease, and climatic change, although all these may have played a role. Human agency played an important role in the decline. There was no longer anything to attract people to these former cities, and, in a period of political instability, life in the countryside, with more direct access to cereals and livestock, may have seemed much more attractive. Following Dietler (2010) I  would argue that the early 5th century in Britain should be seen as a post-colonial period when a number of aspects of Roman economy and material culture appear to be actively rejected, especially in southeastern England. One example would be the abandonment of coinage in the early years of the 5th century. As noted

48

The End of Urbanism in Roman Britain

above, many small denomination coins appear to have been discarded in the late 4th or early 5th century. As Esmonde Cleary (1989) points out, when the coinage supply was cut off in the mid-4th century there is evidence for counterfeiting, but there is no such evidence for the early 5th century. While about 90 hoards of gold and silver coins can be dated between 350 and 420, none can be dated after about 425 CE (Todd 1999, 224). If coinage was seen as a symbol of Roman taxation, it may well have been actively rejected in the immediate post-Roman period. Other aspects of elite Roman material culture may have been rejected as well. For example, a possible corn drying oven was constructed by cutting through the mosaic floor in Insula XXVII in Verulamium, and this may indicate that such mosaics were no longer valued as symbols of wealth and power. We do not need to posit a peasant revolt in the early 5th century (Faulkner 2001) to explain the disappearance of the symbols of elite Roman culture. It may simply reflect a choice by the British population to reject the outward symbols of Roman colonialism. The theme of this book is the rise of urbanism in early medieval Britain. The critical issue here is that by the mid-5th century, Britain was no longer an urban society. While small numbers of individuals may have remained in some towns and cities, they were essentially urban homesteaders. With the possible exception of Wroxeter, the towns themselves no longer served as centers of trade, craft specialization, and administration. They were no longer populated by substantial numbers of people who were engaged in non-subsistence pursuits. By the mid5th century, Britain, like contemporary Ireland, Scandinavia, and Free Germany, was a largely rural society. Ottaway (1996, 122) sums up the end of Roman urbanism as follows: The evidence as it appears to us today is not so much a violent bang, but rather for a feeble whimper as the urban economy expired and townsfolk drifted away to the countryside to become subsistence farmers.

Binchy (1954) famously described early medieval Irish society as rural, tribal, and hierarchical. The same description is probably appropriate for Britain in the 5th century. Britain was no longer an urban society, and most people were probably engaged in some form of mixed agriculture and animal husbandry. As Esmonde Cleary (2011, 23) notes, the landscape must have been peopled and worked or it would have rapidly reverted to woodland. Farming is likely to have been less intensive, since farmers were no longer producing to support the Roman towns and the military garrisons. Smaller tribal polities appear to have re-emerged after the end of Roman rule; some of these may have been centered in the hillforts, such as Congresbury–Cadbury, that were reoccupied in the 5th

Conclusions

49

Figure 1.11. A reconstruction of the South Gate of Roman Silchester. Portions of this gate remain as part of the landscape to this day. Copyright: English Heritage, reproduced with permission.

century. Others may have existed at Wroxeter and in the remains of the amphitheater at Cirencester. The leaders of these polities clearly had the ability to mobilize labor to reconstruct the defenses at Congresbury– Cadbury and to construct the rubble platforms that underlay the subRoman timber buildings at Wroxeter. Even though Britain was no longer an urban society, the Roman towns and cities themselves did not disappear. They remained part of the built landscape, and, as is the case with the amphitheater and city walls at Silchester, they continue to be a part of the built landscape to this day (Figure 1.11). It is also likely that political leaders exercised control over the towns and the areas surrounding them. As K. Dark (2000, 58) has noted, these towns may have had some lingering political importance, even if their resident populations were quite small. For example, the Anglo-Saxon Chronicle records that the Saxons took Cirencester, Gloucester, and Bath after the Battle of Dyrham near Bath in 577. It was as part of this politically fragmented, post-colonial sub-Roman world that we see the first appearance of the Anglo-Saxons, probably in the second quarter of the 5th century, and that is the subject of the following chapter.

2

Early Anglo-Saxon England: Settlement, Society, and Culture

This chapter will explore the archaeology of Early Anglo-Saxon settlement in England from the 5th through the 7th centuries CE. As we saw in Chapter  1, urbanism in Britain effectively ended sometime in the early decades of the 5th century. Here we will examine the nature of the settlements that were established in Britain in the 5th century and what they can tell us about social organization in early post-Roman Britain. Before we review the archaeological evidence, we will briefly examine the historical background to the Early Anglo-Saxon period. Historical Background The 5th century represents a major period of change in British archaeology and history. As we saw in Chapter  1, the diocese of Britain was no longer part of the Roman Empire after the early years of the 400s. Whether the emperor Honorius’s letter was addressed to the Britons or not, the removal of Roman troops in 407 essentially left the diocese on its own. It should not surprise us that many of the institutions that were put in place to support the Roman administration – including coinage, taxation, and urban centers  – largely disappear along with the end of Roman political hegemony. The nature of the early post-Roman world has been a subject of debate for centuries (see Wood 2013 for a comprehensive review of this topic). As modern medieval archaeology has developed in the past 60 years, most of the early models were based on historical sources. The problem is that there are very few contemporary written sources for 5thand 6th-century Britain. Of the major British sources – Gildas, Bede, and the Anglo-Saxon Chronicle  – only Gildas can be seen as broadly contemporary with the events he describes. Gildas was a British cleric who was probably born in northwest Britain, was educated in a classical tradition, and was writing in Latin in Wales, probably around 540 50

Historical Background

51

CE, although some authors would date this work as early as the late 5th century (see, for example, Higham 1994, 141). Gildas’s De excidio et conquestu Britanniae (On the Ruin and Conquest of Britain) was meant as a sermon, condemning his contemporaries for the state of affairs in Britain at that time. The first part, however, provides a brief history of Britain from the time of the Roman conquest up until Gildas’s time. The Venerable Bede, who wrote his Historia ecclesiastica gentis Angelorum (An Ecclesiastical History of the English People) in the early 8th century, drew heavily on Gildas for his early history. The Anglo-Saxon Chronicle, which was compiled in Wessex in the 9th century and survives in nine copies, probably incorporates written records from the 7th century onward but is not describing truly contemporary events until the reign of King Alfred. In short, the British historical record for the 5th and 6th centuries is very limited. Moreover, as Barbara Yorke (1990, 1993) has suggested, many of the documents that do survive were constructed to legitimate the kings who emerged in the late 6th and early 7th centuries CE. The documents provided genealogies for these early kings, usually going back to a pair of eponymous ancestors, whose names, like Hengist and Horsa, began with the same letter. They are not a reliable reflection of the history, in the modern sense of the word, of the 5th and earlier 6th centuries. If we use these documents as historical sources, we run the risk of treating these documents the way that Archbishop Ussher used the “begats” in the Old Testament to reconstruct the age of the world. The British sources are supplemented by even more limited sources from the European continent. Prosper Tiro, writing in southern Gaul, reports that Germanus, the Bishop of Auxerre, visited Britain in 429 to combat the Pelagian heresy. Germanus also visited the shrine of St. Alban at Verulamium at that time. A Gaulish chronicler suggests that the provinces of Britain were subjugated by the Saxons in 441–2. The British and continental sources were used to develop a picture of post-Roman Britain that remained the dominant narrative through the 1950s. The historical account can be summarized as follows. With the withdrawal of Roman military forces, the leaders of the diocese of Britain were forced to see to their own defenses. In order to protect Britain from external attack, a British leader, described by Gildas as a “proud tyrant,” invited the Saxons to settle in Britain. The Saxons turned on their hosts and overran the eastern part of the country. The British defeated the Saxons at the battle of Mons Badonicus, leading to a period of peace that seems to have prevailed in Gildas’s time, but eventually the Saxons conquered most of eastern England. The indigenous British population was slaughtered, forced west, or subjugated by the

52

Society and Settlement in Early Anglo-Saxon England

Anglo-Saxons. Bede (1.15) provides a particularly colorful description of the process: [T]hese heathen conquerors devastated the surrounding cities and countryside, extended their conflagrations from the eastern to the western shores without opposition and established a stronghold over nearly all the doomed island. Public and private buildings were razed; priests were slain at the altar; bishops and people alike, regardless of rank, were destroyed with fire and sword, and none remained to bury those who had suffered a cruel death. A few wretched survivors captured in the hills were butchered wholesale, and others, desperate with hunger, came out and surrendered to the enemy for food, although they were doomed to lifelong slavery even if they escaped instant massacre. Some fled overseas in their misery; others, clinging to their homeland, eked out a wretched and fearful existence among the mountains, forests, and crags, ever on the alert for danger.  (Bede 1.15, tr. Sherley-Price 1968, 57)

During the past 50 years, this traditional model has been challenged by archaeologists, historians, and, most recently, geneticists. The concerns raised include questions about the role of migration in the process, and even whether there was an Anglo-Saxon migration at all. Others have also questioned whether there are possible continuities between Late Roman Britain and Early Anglo-Saxon England, on the rural if not the urban level. Finally, geneticists have begun to examine the relative contributions of British and Anglo-Saxon males to the modern English gene pool and their implications for Anglo-Saxon settlement and social organization. The question of the nature, scale, and existence of the Anglo-Saxon migrations to Britain is a particularly contentious one. The traditional historical model was a migrationist one, suggesting that the cultural and linguistic changes that appear in Anglo-Saxon England resulted from substantial population movement. While migration was seen as one of the classic causes of cultural change in the first half of the 20th century, migration was de-emphasized as a means of change in the 1960s as the theoretical focus in archaeology shifted away from culture history and toward more processual, ecological, and economic explanations for social change. In addition, as a result of two world wars, there was certainly a trend toward focusing on the British, rather than Saxon, antecedents of contemporary English society (Harke 1998). In recent years, this debate has been characterized as one between the “movers” and the “shakers” (Halsall 1999, 132; see also Wood 2013, 310–12). The movers are those scholars who see the migrating peoples as a catalyst for social, political, and economic changes in the Late Roman and post-Roman world. The shakers, on the other hand, argue

Historical Background

53

that tensions within the Late Roman Empire shook it to its foundations, and the barbarian invasions are simply symptomatic of larger internal problems. Critiques of the 5th-century migrationist model take several forms. At a most basic level, historian Michael E. Jones (1998) argued that a close but literal reading of the historical texts would suggest that the Anglo-Saxons arrived in a relatively small number of boats. Jones (1998, 47)  notes that Gildas tells us that the Anglo-Saxons arrived in three boats, followed by reinforcements. He sees Gildas’s account as a classic example of a mercenary revolt, and warns that “the small number of invaders described by Gildas must be treated seriously by the historian” (Jones 1998, 53). Bede (1.15), following Gildas, tells us that the Angles or Saxons initially came to Britain in three long ships at the invitation of King Vortigern. Given what we know about pre-Viking watercraft, this would not involve a large number of immigrants (see also Higham 1994, 41). Other estimates of the number of immigrants vary, but most range between the low tens of thousands and the low hundreds of thousands (Thomas et al. 2006, 2651). Other scholars have argued that what we may see is a classic example of elite dominance (Renfrew 1987; Ward-Perkins 2000; Anthony 2007, 117–19; see Hills 2003 for a comprehensive review of this issue). In this model, a small number of militarily successful Saxons may have led much of the rest of the population to emulate their success by adopting their language and material culture. The recently published cemetery at Wasperton in Warwickshire (Carver et al. 2009) provides the kind of data that can be used to support a model of acculturation. The cemetery, which was in continuous use from the 4th through the 7th centuries, shows burial rites that seem to reflect political, rather than demographic, changes. The cemetery begins with traditional 4th-century Roman burials. These are followed by what appear to be Christian burials in the 5th-century sub-Roman period. These are followed by Anglo-Saxon burials in the later 5th and 6th centuries. The earlier cremations from the later 5th and early 6th centuries show parallels with the Rhineland, while the later inhumations show links with East Anglia and Wessex. The burial rites in this frontier area seem to reflect the changing political dominance in the region. The inhabitants of the site, who were bread-makers during the Roman period, cast their lot with the Christians of western Britain in the first part of the 5th century, but shifted their allegiance to the dominant Anglo-Saxon polities by the 6th century. K. Dark (2000) has argued for continuity of British power and influence throughout the 5th into the 6th century in both eastern and western Britain. While the case for British continuity in the west can be made

54

Society and Settlement in Early Anglo-Saxon England

on the basis of a range of historical, archaeological, and linguistic criteria (see previous chapter and below), the case for the same degree of continuity in eastern England is somewhat harder to make. Dark has argued that the areas that show an absence of early Pagan Anglo-Saxon cemeteries in the east are areas that were controlled politically and militarily until the English kingship begins to emerge in the late 6th century. Given the absence of settlement continuity in many urban and rural settlements in eastern England, Dark’s case remains unproved. However, the possibilities for long-term continuities in the use of the landscape (see, for example, Oosthuizen 2016; see also Oostheuzen 2017) would lend support to Dark’s model. Recently, population geneticists have entered the fray. Using Ychromosome data drawn from a sample of modern British males from central England and North Wales, Weale et  al. (2002) argued that the modern English population was indistinguishable from modern Frisians, but significantly different from the modern population of North Wales. They suggest that these data are best explained by a substantial migration of Anglo-Saxon males into England, but not Wales, in the immediate post-Roman period. While the data certainly show that modern English males are genetically different from modern males in North Wales, the assumptions used in this model and the conclusions drawn from it are more problematic. First, it is not clear that the modern population of rural North Wales can be taken as similar to the population of eastern England during the Roman period. North Wales has experienced substantial depopulation in modern times, and Wales also experienced substantial contact with Ireland in the early Middle Ages. The authors do not address the substantial genetic variation that is seen within the modern North Wales population. Second, the modern genetic data from central England were drawn from people living in small towns whose paternal grandfathers had been born in the same region. This was done as a way of minimizing the effects of migration to large towns and cities in the post-medieval period. However, as we shall see below, Early AngloSaxon England was non-urban, so this may not address issues of population movement in the early medieval period. The third problem is that massive migration may not be the only explanation for genetic difference. Other explanations for these data are possible. For example, Thomas et al. (2006) suggest that an apartheid-like social structure that limited the reproductive opportunities for British males could also explain the genetic patterns seen by Weale et al. (2002). Recently, Pattison (2008) has suggested that a steady, small-scale flow of immigrants from northwest Europe could also explain the Y-chromosome data seen by Weale et al. (2002) and Capelli et al. (2003). In short, the genetic data can be

Chronological Issues

55

interpreted to support either a migrationist model or an elite dominance model for the arrival of the Anglo-Saxons in southern and eastern Britain. A recent study by Leslie et  al. (2015) has attempted to clarify the genetic issues by studying genome-wide SNP (single nucleotide polymorphism) data from 2039 individuals from the United Kingdom and comparing those data to genetic data obtained from over 6000 individuals from across western Europe. Their data provide clear evidence for an Anglo-Saxon migration, but they suggest that the proportion of Anglo-Saxon ancestry in the modern central and southern English population is likely to be between 10 and 40 percent (Leslie et al. 2015, 313). In addition, their data indicate that there was very little input from the Danish Vikings, despite their political control of parts of eastern England in the Late Anglo-Saxon period (Leslie et al. 2015, 313). Chronological Issues One of the critical issues archaeologists face in addressing questions of Early Anglo-Saxon settlement and the possibilities of continuities in settlement patterns from the Roman/sub-Roman period to the AngloSaxon era is the question of dating. As noted in the previous chapter, two of the materials that are commonly used to date Roman sites in Britain and elsewhere are pottery and coins. Both present serious problems when trying to date early 5th-century sites. The last Roman bronze coins to have reached the diocese of Britain were those of the House of Theodosius (388–402). While it is certainly possible that some of these coins stayed in circulation after the first decade of the 5th century, there are good reasons to think otherwise. Low-value bronze coins would have served two principal functions in Late Roman Britain. First, they served an administrative function. They were used to convert between taxes in kind and taxes in coin and to pay for services. Second, they also served a commercial function in 4th-century Roman Britain. As Esmonde Cleary (1989, 96) has shown, when there was an acute shortage of bronze coinage during the 350s, there was extensive counterfeiting. Most of the copied coins appear to have been used for commercial transactions. No such copying took place after 402, indicating that coins were no longer needed for commercial purposes. Data from 140 Roman period sites in Britain show that coin loss increased substantially in the Late Roman period (Reece 1995, 183; Gardner 2007, Fig. 3.8). While this increasing loss rate may reflect the more widespread use of coinage in commercial transactions during the 4th century, Gardner’s (2007) study of Roman military sites shows that many Late

56

Society and Settlement in Early Anglo-Saxon England

Roman coins ended up in refuse contexts in the late 4th century. As Gardner (2007, 79)  notes, the presence of these coins in Late Roman refuse deposits may “suggest a tension between coins being more widely used than they were previously, and at the same time less worthwhile to keep – to the extent that many were thrown away.” These data suggest that early 5th-century Britain no longer had a monetary economy. Coins, therefore, cannot be used to date most 5th-century sites. Pottery presents many of the same problems.Terra sigillata or “Samian” fine wares were imported into Britain from central and southern Gaul in the 1st and 2nd centuries. The chronology of this industry is well established, and many of these ceramics can be dated to within a decade or two. In the Late Roman period, a number of centers of pottery production were established in southern and eastern Britain. These kilns were generally located in rural locations, but were situated in places where they could provide fine wares to the civitas capitals (Esmonde Cleary 1989, 91). Unfortunately, these production centers seem to have gone out of use at the beginning of the 5th century. While some Roman pottery sherds were found at Early Anglo-Saxon sites such as West Stow in Suffolk, the disproportionately high number of bases and their continued use into the 6th century suggest that these pots may have been selectively chosen from a rubbish deposit and that this scavenging continued for an extended period (Plouviez 1985, 85). Given the problems with using Late Roman coins and pottery to date the 5th-century sites in Britain, most chronologies have been based on artifact typologies, and particularly on comparisons to better-dated material from the European continent. Dendrochronology has played a role in areas where waterlogging has led to the preservation of wood, but its role to date has been relatively limited. Until recently, radiocarbon dating played a relatively minor role in dating Early Anglo-Saxon sites. Traditional methods of radiocarbon dating require large samples (generally at least 25 grams of material) and often produce standard deviations that are too large to be useful in what is a historical time scale. As noted above, the advent of AMS dating in the 1980s, which allows archaeologists to date smaller samples, and with the development of better methods for preparing and cleaning samples, radiocarbon dates are playing an increasingly important role in Anglo-Saxon archaeology. For example, a series of carefully selected radiocarbon samples from the ostensibly Late Roman cemetery of Queenford Farm and the nearby Early Saxon cemetery Berinsfield in Oxfordshire allowed Hills and O’Connell (2009) to show that the cemeteries had been used sequentially rather than concurrently. The radiocarbon dates indicated that the Queenford Farm cemetery was primarily in use during the 4th century,

The Archaeology of Early Anglo-Saxon Settlement

57

while the Berinsfield cemetery was in use from the 5th to the early 6th century. If the two cemeteries had been contemporary, they might have provided some support for the apartheid-like model for Anglo-Saxon social organization. Instead, these data suggest that the native population in the region adopted Anglo-Saxon burial customs in the 5th century. Despite the chronological problems that plague the archaeology of the 5th century, the archaeological data suggest that much of the diocese of Britain went through a process of de-Romanization in the late 4th and early 5th century. As we saw in Chapter 1, this is much more than a political change. Not only were the Roman troops withdrawn from Britannia, never to return, but the Roman towns and villas were depopulated as well. Smaller British polities survived in the west, in places such as Wroxeter and Congresbury–Cadbury, but by the end of the first half of the 5th century, Britain had largely ceased to be Roman. This chapter will explore the nature of the settlements that replaced the towns, villas, and small rural settlements in eastern England during the 5th century. The Archaeology of Early Anglo-Saxon Settlement One of the problems facing archaeologists who study Early Anglo-Saxon settlement in Britain is that relatively few Early Saxon settlements have been excavated to a modern standard, and even fewer have been completely published. While somewhere between 30,000 and 40,000 Pagan Anglo-Saxon burials have been discovered in Britain, there are probably only a few dozen 5th- and 6th-century settlement sites that have been excavated in any detail, and only a handful of those have been comprehensively published. The earliest large-scale excavation of an Early Anglo-Saxon settlement was undertaken by E. T. Leeds at the site of Sutton Courtenay in Berkshire (Leeds 1923, 1926–7, 1947) (see Figure  2.1). In the 1920s, Leeds salvaged about two dozen sunken-featured buildings (SFBs) from gravel quarrying. These structures, known as grubenhäuser on the European continent, are generally rectangular with rounded corners. They are sunk about a half meter into the earth and are marked by one, or sometimes three, postholes at either short end. While Leeds identified a small number of postholes, he did not identify any posthole structures other than the grubenhäuser, and he assumed that these sunken-featured buildings served as the dwellings for the Early Anglo-Saxons. The first Early Anglo-Saxon settlement to have been extensively excavated and published was West Stow in Suffolk (Figure  2.2) (West 1985). The West Stow site is located in the Lark River valley in the Breckland region of northwest Suffolk, near Bury St. Edmunds. The

0

100

200

300

400 km

Figure 2.1. Map of southern Britain showing the locations of the AngloSaxon sites discussed in this chapter. Drawn by Douglas V. Campana using open-source software. Map numbers: Sutton Courtenay 1, West Stow 2, West Heslerton 3, Mucking 4, Quarrington 5, Kilham 6, Yeavering 7, Bloodmoor Hill 8, Cowdery’s Down 9, Lyminge 10, Sutton Hoo / Tranmer House 11, Rendlesham 12, Prittlewell 13, Taplow 14, Dinas Powys 15, Congresbury–Cadbury 16

The Archaeology of Early Anglo-Saxon Settlement

59

Figure 2.2. Plan of West Stow, redrawn. Copyright: Suffolk County Council.

site was associated with a cemetery which was discovered in the course of gravel digging in the 19th century. More than 100 graves, including both inhumations and a few cremations, were excavated in the 1840s and 1850s. The excavations recovered 151 objects, plus beads, but the quality of the recording is so poor that the individual grave groups cannot be reconstructed. The West Stow settlement site was discovered by Basil Brown in the 1940s, and preliminary excavations took place there under the direction of Vera Evison in the late 1950s. Major excavations were carried out at the West Stow settlement by Stanley West between 1965 and 1972 (West 1969, 1985). West had attended Cambridge University at a time when the theoretical focus of archaeology was shifting toward questions of economy, technology, and environment, so the focus of the excavation was on the recovery of material that could be used to reconstruct Early Anglo-Saxon settlement patterns and subsistence technology. For example, flotation techniques, which were being developed while the excavation was ongoing, were used at the site in 1972 (Murphy 1994). Excavations at West Stow uncovered 69 sunken-featured buildings clustered around seven small “halls,” spread across an area of about

60

Society and Settlement in Early Anglo-Saxon England

Figure  2.3. Reconstructed sunken-featured building at West Stow County Park. The program of reconstruction began in 1974. Photo credit: Pam J. Crabtree.

1.8 ha. Unlike Leeds, West (1985, 116–18) argued that the bottoms of the SFBs were not living floors. Instead, the excavations at West Stow suggested that the pits were covered by larger, suspended timber floors. The reconstructions of these dwellings can be seen at the West Stow Anglo-Saxon village today (Figure 2.3). The halls are rectangular postbuilt structures that are generally less than 10 m in length, far smaller than similar structures on the continent (West 1985, 19). Detailed chronological analyses indicate that only about three of these settlement clusters would have been in use at any one time. The halls were interpreted as the main dwellings for each cluster, while the sunken-featured buildings served as outbuildings such as workshops and barns. West (1985, 168–9) suggested that the site was probably occupied by three extended families throughout most of its history. Initial analysis suggested that the site was first occupied in the early 5th century, and that it was abandoned around the middle of the 7th century CE. A small number of sherds of Ipswich ware were recovered from the fills of some of the latest of the SFBs. Other fragments of Ipswich Ware were recovered from the ditches and from the general cultural layer, known as “Layer 2.” Reanalysis of the chronology of Ipswich Ware indicates that it was not produced until about 700–20 CE

The Archaeology of Early Anglo-Saxon Settlement

61

(Blinkhorn 1999, 2012). This means that the West Stow settlement must have survived into the early 8th century, adding at least 50 years to the settlement’s history. West (2001), however, notes that the longer chronology is inconsistent with the experimental data on the lifespans of the timber buildings. West (1985, 167) believed that he had excavated the total Early AngloSaxon settlement at West Stow. However, recent discoveries have also challenged West’s view of West Stow as a small hamlet that was probably occupied by no more than three extended families at any one time. Rescue excavations were carried out in 2007 in advance of the construction of a new gift shop for the visitors’ center at the West Stow site (Gill et  al. 2015). These excavations, known as West Stow West, identified six additional sunken-featured buildings and a posthole structure, indicating that the site may have been much more extensive than was originally thought.1 With these new discoveries West Stow looks more like the other sprawling Anglo-Saxon settlements that were established in the 5th century. The long-term programs of excavation at West Stow allow us to say something about the way the Early Anglo-Saxons made use of the Lark Valley environment. The settlement itself was located on the river terrace. This terrace area also would have provided fields for agriculture and some limited woodland with pannage for pigs. The rich grazing areas closer to the river were probably used by cattle. The Lark River would also have supplied freshwater fish, and it was home to aquatic birds such as ducks. The upland areas of the Breckland region would have been ideal for grazing sheep and a few goats. The distribution of known Anglo-Saxon finds in the Lark Valley suggests that these settlements were strung out along both sides of the river. This settlement pattern would have given each farmer access to the rich pastures of the Lark Valley, the farmland and woodland along the river terraces, and the upland grazing areas. The most extensively excavated and fully published of the Early Anglo-Saxon settlements is the site of Mucking in Essex (Hamerow 1993; Hirst and Clark 2009). Crop marks seen from the air initially revealed the archaeological potential of the Mucking site, which is located on the 100-meter gravel terrace overlooking the Thames estuary (Hamerow 1991, 3). The site was excavated by the late M.  U.  and M.  T. Jones, but the results have been published by other scholars. 1

The excavation has not yet been published, but the excavators are planning to publish the results in Medieval Archaeology. Gill et al. (2015) is part of the grey literature. Some information on the faunal remains recovered from West Stow West has been published by Crabtree and Campana (2015).

62

Society and Settlement in Early Anglo-Saxon England

Excavations revealed two Pagan cemeteries, 53 posthole buildings, and 203 grubenhäuser spread over 18 ha. The site was continuously occupied from the first half of the 5th century through the late 7th or early 8th century. Two models for the settlement history have been proposed. The first by Hamerow (1993, 86–9) suggests that the Mucking settlement is made up of a series of farmsteads whose locations shifted around the two cemetery sites (Figure 2.4). Ancestral ties to the cemeteries explain why the Early Anglo-Saxons continued to inhabit this relatively inhospitable site, although it may have been initially chosen for its strategic location overlooking the Thames. Tipper (2000, 2004) suggested an alternative model based on the restudy of the ceramic and the settlement data. He suggested that there was a single nucleus of settlement in the 5th century located in the southwest area of the site. This was augmented by a secondary settlement at the north end of the site which was first established in the 6th century. While Hamerow saw the settlement as shifting throughout the Saxon period, Tipper suggested that two separate and stable communities, separated by the cemeteries, occupied the terrace area during the 6th and 7th centuries. Tipper also noted the three ditched enclosures in the central area of the settlement, which probably date to the late 7th to early 8th century, and suggested that they may have been associated with the contraction of the settlement at the end of the Early Anglo-Saxon period (Tipper 2004, 37–9). Similar ditches have been identified from the final phase at West Stow (West 1985, 54), some of which contained Ipswich Ware, indicating that they were filled in the early 8th century. These data hint at the changes in land tenure and/or organization that seem to have taken place at the end of the Early Anglo-Saxon period. Hirst and Clark (2009, 445)  note that after Mucking was abandoned in the late 7th or early 8th century, the site reverted to agriculture and that the subsequent field system developed out of these 7th–8th-century ditched enclosures. A third large Early Anglo-Saxon site has been excavated at West Heslerton in Yorkshire. The site has been extensively surveyed and excavated under the direction of Dominic Powlesland, but it remains incompletely published (Powlesland et al. 1986; Powlesland et al. 1998; Powlesland 1999, 2000). The site is located on the south side of the Vale of Pickering and at the foot of the Yorkshire Wolds (Powlesland 1999). Excavations were carried out at the settlement between 1986 and 1995. The site shows some 4th-century Late Roman activity, principally associated with the ritual use of a spring. The Early and Middle AngloSaxon features are distributed over 13 ha and include 130 grubenhäuser

2700 N N

2600 N

10

0

2500 N 2400 N 2300 N 2200 N 105

2100 N 2000 N 1900 N 1800 N 1200 E 1700 N 1600 N

1100 E

1500 N 95

90

100

105

1400 N 1300 N 1200 N

Anglo-Saxon Cemetery I

1100 N

1000 E

1000 N Anglo-Saxon Cemetery II

900 N 800 N

900 E 700 N Grubenhaus posthole building grave Anglo-Saxon ditch

90

95

500 N

100

105

600 N

400 N 0 300 N

800 E

0

100 m 250 ft

200 N 100 N

100 W 0

700 E

100 E

600 E 200 E 300 E 400 E 500 E

Figure 2.4. Plan of the site of Mucking, Essex. Reproduced with permission of Sue Hirst who holds the copyright.

64

Society and Settlement in Early Anglo-Saxon England

and 90 posthole structures, plus a cemetery that was excavated earlier (Powlesland et al. 1986; Powlesland 1999). The Anglo-Saxon settlement was discovered in 1982 in an area that was experiencing damage from plowing. A spring located in the center of the south half of the settlement was the setting for the construction of two shrines in the 4th century. The presence of some very Late Roman or sub-Roman pottery suggests that the shrine remained an important ritual location into the early 5th century. The stream, which is fed by the natural spring, formed the focus of the Early Anglo-Saxon settlement. The layout of West Heslerton shows some zoning that is not apparent at either West Stow or Mucking. There appear to be distinct areas of the site dedicated to housing, crafts, and agricultural processing. No boundary lines or ditches are apparent in the northern half of the settlement. The northeast appears to have served primarily as a dwelling zone, and posthole structures or “halls” dominate in this area. A high concentration of grubenhäuser are located to the west of the stream channel. As is the case at West Stow, these structures seem to have been covered by a wooden floor. Powlesland (1999) and his colleagues interpret this as an area devoted to crafts and agricultural processing. In addition to crop storage, crafts carried out in this area include spinning, weaving, metal-working, animal processing and butchery, and possibly bone-, leather-, and horn-working. There is no evidence to suggest that these crafts were carried out on a large scale. Rather, they seem to represent local production for local consumption. The southern part of the settlement has a very different layout. Extensive enclosures are visible on both sides of the stream. These enclosures appear to have been built in the Late Roman period, and their layout was maintained throughout the Early Anglo-Saxon period. They are marked by ditches and slots for fencing. During the Middle Saxon period, when settlement contracted to the Late Roman core of the site, new enclosures and boundary ditches were constructed. Powlesland (1999) has suggested that the Middle Saxon settlement may represent an early manor. Coin evidence suggests that the site was abandoned around 850 CE. The excavations at West Heslerton raise two interesting questions for the study of early medieval urbanism. The first, of course, is the question of Roman-to-Saxon continuity at the site. The site was clearly used in the late 4th century, and probably into the early 5th century, as a shrine. Its location at a spring is not surprising, since many Roman and pre-Roman ritual sites are located at springs and other aquatic features. Powlesland (1999) dates the Early Anglo-Saxon settlement to ca. 475–675 CE.

The Archaeology of Early Anglo-Saxon Settlement

65

Whether the late 5th-century inhabitants represent Romanized Britons, immigrant Anglo-Saxons, or some combination of the two, what we see at West Heslerton is clearly not continuity in Roman urban settlement. Other questions of Roman–Saxon continuity will have to await the full publication of the excavation. A second and more critical question is whether the ordered layout of the site represents a type of “proto-urban” settlement in the Early Anglo-Saxon period. Powlesland (1999) asks whether the zoning and planning seen at West Heslerton represent a failed Early Anglo-Saxon attempt to establish sites of a more urban nature. He further wonders whether this was part of a broader trend toward Early Saxon urbanism which could not be sustained. While we await the final publication of the site, I  think that the answer to these questions should be an emphatic no. While there is certainly evidence for settlement planning and zoning at West Heslerton, there is no clear evidence for substantial craft specialization. As Powlesland (1999, 62)  notes, there is no evidence for large-scale textile production at West Heslerton. Similar evidence for small-scale textile production has been recovered from other Early Saxon sites, including West Stow and Kilham in East Yorkshire (Hunter-Mann 2000, 2001). In many ways, the zoning we see at West Heslerton presages the kinds of zoning we see in some later Early Saxon and Middle Saxon rural settlements. For example, at Brandon in Suffolk (Tester et al. 2014, 372–4; see Chapter 3) there is clear evidence for a waterfront industrial zone where activities such as textile production took place. Moreover, there is no evidence that the site of West Heslerton was providing services to a broader hinterland. While the lava querns provide some evidence for long-distance trade (see below), there is no evidence to suggest that West Heslerton was a focal point for broader trade in these imported items or that it served as a market for the surrounding countryside. The planning clearly points to some degree of leadership, but that should not surprise us. Historical and archaeological evidence both indicate that leadership was a feature of both native British and Anglo-Saxon societies. What conclusions can be drawn about Early Anglo-Saxon settlement from the large-scale excavations at West Stow, Mucking, and West Heslerton? All three appear to be rural settlements of substantial size that include both post-built structures and grubenhaüser. The small “halls” appear to have served as dwellings, while the sunkenfeatured buildings were used as workshops and storage facilities. While these sites appear to have made use of the Roman landscape, there is no clear continuity in residential settlement. West Heslerton

66

Society and Settlement in Early Anglo-Saxon England

is built around what was a Late Roman ritual site. The West Stow settlement is located on an unoccupied site, although it is close to the Late Roman settlement of Icklingham. There was also a Late Roman settlement in the area of Mucking, although the Roman material from the site has not yet been fully published. In addition, Cemetery II at Mucking was bounded by Roman enclosure ditches on both the north and the south (Hirst and Clark 2009, 759). This may suggest continuity of land use or land tenure, even if there was no obvious continuity of settlement. The economic data from these sites point to a degree of autarky or economic self-sufficiency (Crabtree 2010b). While the faunal data from Mucking are limited due to poor preservation, and the data from West Heslerton have not been fully published, the extensive faunal collection from West Stow indicates that animal husbandry was not geared toward specialized products for exchange (Crabtree 1990). Faunal collections from a range of smaller Early Anglo-Saxon sites, including Quarrington in Lincolnshire (Rackham 2003) and Kilham in East Yorkshire (Archer 2003), reveal an Early Anglo-Saxon economy that is unfocused and designed to produce a range of products, such as meat, milk, wool, and traction, for local consumption. Gerrard (2013, 13) has recently argued that Late Roman farming communities were overproducing crops to meet the demands of the Roman state. When these demands disappeared in the 5th century, the agropastoral system changed to one that was focused on less intensive forms of agricultural production such as cattle and sheep pastoralism. A recent study of Anglo-Saxon deer hunting (Sykes 2010, 178) shows that Early Anglo-Saxon sites, including West Stow, tended to include a high proportion of meaty body parts. This is consistent with animals that were butchered at the kill site and whose low-utility parts were discarded there. Only the meat-bearing bones were carried back to the settlements. Hunting was not a common practice, and it was carried out as a means of risk reduction, rather than as a means of social differentiation. In short, hunting served as a dietary supplement at times when other food sources may have been running low. While subsistence practices were focused primarily on local needs, this does not mean that these sites were isolated and cut off from local, regional, and international trade networks.The presence of Neidermendig lava for quern stones at West Heslerton (Powlesland 1998) in both Early and Middle Saxon contexts points to trade with the continent, although whether this trade was direct or indirect is not known. At a regional level, the presence of a marine flatfish in 6th-century contexts from West Stow indicates trade and contact with the coast (Crabtree 1990, 27),

The Archaeology of Early Anglo-Saxon Settlement

67

and the presence of a beaver incisor may indicate contact with the fenland region. By the mid- to late 6th century, Illington-Lackford pottery appears at a series of settlement and cemetery sites including West Stow in a 900 square km region in East Anglia. The pottery is marked by distinctive stamps and designs, and it is the first standardized pottery to appear in post-Roman Britain. Its presence at West Stow and other contemporary East Anglian sites also points to local and regional exchange, although this may have taken one of a number of forms, including trade, gift exchange, and redistribution by a local leader (Russel 1984, 528; Hamerow 2006, 279). If the latter is the case, then craft specialists may have been attached to local leaders, rather than itinerant or independent crafts workers. Another striking feature of the settlement evidence from the 5th and early 6th century is that the architecture and settlement layouts show little evidence for differences in status or wealth. All the dwelling units appear to be made up of a small, central hall surrounded by a series of sunken-featured buildings that served as outbuildings. There is no clear evidence for high-status or low-status dwellings before the late 6th century. As Hamerow (2004, 307) notes, status does not appears to be expressed by the ability to support large numbers of people in a single household. Many of these Early Anglo-Saxon settlements have little in the way of boundaries. While some possible animal pens are present, there are no enclosures around buildings or sets of buildings before the late 6th century at the earliest (Hamerow 2004, 307). For example, the ditches that are visible at West Stow can be assigned to the final phase of the settlement. That phase was initially dated to the late 6th–7th centuries CE. However, a number of these ditches include Ipswich Ware, which suggests that they may be as late as the early 8th century CE. The settlements are also somewhat mobile; dwellings are not rebuilt in the same place, although at West Heslerton there are examples of replacement buildings that are adjacent to earlier buildings (Powlesland 2000, 24). The individual dwellings are generally not surrounded by fences, ditches, or other territorial markers. The question of the nature of the society that inhabited these sites is an interesting one. The excavators of both West Heslerton (Powlesland 1998) and West Stow (West 1985) do not see these settlements as high-status communities. Higham (2004, 15–16), however, suggests that the quantity of artifacts recovered from West Stow, and probably West Heslerton, is too large to have been produced by an economy based primarily on subsistence farming. He notes that the excavation at West Stow produced, among other things, 66 iron knives, a large

68

Society and Settlement in Early Anglo-Saxon England

number of spindle whorls and loom weights, and a range of other metal items including arrowheads, shield bosses, and brooches. Powlesland (1998) reports that the larger West Heslerton site produced over 200 iron knives, plus additional fragments. Higham seems to imply that the denizens of West Stow and probably West Heslerton must have relied heavily on tribute to support their lifestyles. There are three problems with this argument. First, knives would have served a variety of functions, from eating utensils to butchery tools to weapons, in Early Anglo-Saxon society. It is not unreasonable to assume that the community of West Stow discarded 66 knives over a period of almost 300 years. The second is that smaller Early Anglo-Saxon sites, such as Quarrington and Kilham, seem to show the same economic patterns as those seen at West Stow. The third is that there are good continental models for Migration Period sites that rely primarily on tribute. At sites such as Gennep (Heidinga 1994, 2001) in the Netherlands, the inhabitants appear to have subsisted primarily on tribute. The site provides extensive evidence for hunting and metal-working, but the evidence for animal husbandry and agriculture is limited to horse breeding. At West Stow, the evidence for hunting is limited, and the presence of neonatal and young juvenile cattle, sheep, and pigs indicates that animal husbandry was taking place on site. No one, however, would argue that Early Anglo-Saxon sites and communities were egalitarian. The variations seen in the quantities and quality of grave goods in the large inhumations and cremation cemeteries point to a society that had significant differences in social status, political power, and material wealth. The problem is that we have almost no historical evidence about the nature of 5th- and 6th-century Anglo-Saxon society. A  number of authors have noted that many of the 7th-century Anglo-Saxon kingdoms are similar in area and geographic extent to the pre-Roman Iron Age tribal territories in Britain. Higham (2004, 8) notes that the territory of the Trinovantes is remarkably similar to the territory of the 7th-century East Saxons. However, the question of whether there was a degree of continuity in political administration at the local level is far more difficult to answer. For example, we do not know the structure and extent of a single Roman estate, so it is difficult to determine whether any of these estates, or any smaller land units, may have passed into Saxon hands. Moreover, there has been a tendency to project the social and political structure known from 7th- and 8th-century England back onto the 5th and 6th century, even though it is likely that Early Saxon England, as well as sub-Roman Britain, was made up of a series of smaller and more loosely organized polities.

The Archaeology of Early Anglo-Saxon Settlement

69

Archaeological and landscape data from West Sussex provide some evidence to support the presence of these smaller 5th- and 6th-century polities (Semple 2008). Graves and cemeteries in this region were located along crests of escarpments and on edges of rough pasture overlooking the coast, the estuary, and the rivers (Semple 2008, 417). Both primary and secondary barrows were located so as to dominate entry points into west and central Sussex (Semple 2008, 420). Semple (2008, 423) has argued that these funerary monuments define a series of small, river-centered polities that emerged in the political chaos of the post-Roman period. That these small polities are centered around river valleys is particularly significant, since McCormick (2010) has argued that one of the most significant changes in transportation during the 5th century was the change from land-based transport during the Roman period to water-based transport in the post-Roman era. If we imagine that much of eastern England was made up of small, river-based polities in the 5th and 6th centuries, then there is no need for the substantial food rents that characterized the 7th and 8th centuries CE. In the Middle and Late Anglo-Saxon periods, the king and his followers “ate” their way through their kingdoms. A possible Middle Saxon food-rent collection site has been identified as Higham Ferrers, Northamptonshire (Hardy et  al. 2007). In the 5th and 6th centuries, it is likely that chiefs, war leaders, and retainers could move through their entire kingdoms in a matter of days, and the burden of taxes and tribute is likely to have been much less. As Hamerow (2004, 307–9) has suggested, “the seventh and eighth centuries seem truly to be the period when, at least in southern England, settlements became more firmly inscribed onto the landscape as both secular and religious landlords extended growing influence over their estates.” If the west Sussex model can be extended to other parts of eastern England, it suggests that we are looking at a region made up of a patchwork of small polities. Some of these polities appear to have survived into the Middle Anglo-Saxon period (7th–9th centuries CE), since they appear in the Tribal Hidage, a document that seems to be a tribute list for the emerging Mercian kingdom in the Midlands (Figure 2.5). It is also likely that these small polities could unite for purposes such as warfare and raiding. For example, Yorke (2003) has suggested that the Angles and Saxons may have been confederations or gentes in 5th- and 6thcentury Britain. Small British polities also existed in the regions outside the initial Anglo-Saxon settlements. Many of the known Early Anglo-Saxon settlements are located on comparatively marginal land, such as breckland and gravel. While those

70

Society and Settlement in Early Anglo-Saxon England

Figure 2.5. Anglo-Saxon polities at the time of the Tribal Hidage, probably in the late 7th century. Redrawn after Yorke (1990, Map 1, p. 12), with permission of the author.

who would see substantial continuity in post-Roman rural settlement might argue that the better lands were already taken by native British farmers, it is equally possible that these lands were chosen for nonagricultural reasons. Moreover, while these areas may not have been ideal for the intensive cereal and livestock farming that was practiced during the Roman period to meet the needs of the towns and the

Settlement Changes in the Late 6th and 7th Centuries

71

military, they would have been appropriate for more extensive subsistence practices that were designed to meet local needs. This is consistent with the pollen data from East Anglia, which provide no evidence for woodland regeneration in the post-Roman period, but which suggest a decline in cereal agriculture in favor of stock-raising. The archaeological data suggest that for most of the 5th and 6th centuries CE Britain was a rural society. The towns and cities of Roman Britain were depopulated, and Anglo-Saxon settlements that were established between 450 and 550 were strikingly non-urban in character. While towns and cities were seen as essential elements of Roman culture and worldview, it might be argued that cemeteries played a more prominent role in Early Anglo-Saxon society. Lucy (2000, 152)  notes that cemeteries are located on high ground while settlements are located in more sheltered valley regions. While Early Anglo-Saxon rural settlements like West Stow and West Heslerton were occupied throughout the entire Early Saxon period, by the end of the 6th century we see changes in the character of Anglo-Saxon settlements that may mirror other substantial changes in Anglo-Saxon society. Settlement Changes in the Late 6th and 7th Centuries In this section, I  will argue that the archaeological evidence points to changes in settlement patterns that first appear in the late 6th and 7th centuries CE. These include the appearance of rural settlements with more elaborate architecture than the typical Early Saxon villages such as West Stow, the appearance of possible royal sites, and the establishment of the earliest of the rural estate centers. These data suggest that a true settlement hierarchy is emerging in Anglo-Saxon England by the 7th century CE. I will further suggest that these changes mirror broader changes in Anglo-Saxon society and that they laid the foundation for the political consolidation and urbanism that become apparent in the Middle and Late Anglo-Saxon periods. At the top of the settlement hierarchy is the royal palace at Yeavering in Northumberland (Hope-Taylor 1977). The site was discovered through aerial photography in 1949 and subsequently excavated by Brian HopeTaylor between 1953 and 1962 since the site was threatened by quarrying. Yeavering is the site of prehistoric activity, including a henge monument and burials. The Anglo-Saxon settlement at the site was established in the second half of the 6th century as one or two small homesteads (Scull 1991), and a number of Early Anglo-Saxon burials were placed near the prehistoric ones. In the early 7th century (Phase IIIc) it developed into an elaborate settlement and ceremonial complex (Figure 2.6). The

72

Society and Settlement in Early Anglo-Saxon England

Figure 2.6. Plan of  Yeavering in the 7th century. Copyright: Historic England, reproduced with permission.

complex included a large timber hall, Structure A4, as well as a number of other carefully laid-out timber buildings. The hall is one of the largest buildings of this date that is known from England. Other structures included a large enclosure whose function is not clear, possible Pagan monuments including postholes that may have served as settings for carved poles, and a grandstand-like structure with tiers of steps leading down to a throne. The site does not appear to have been occupied year round. It is probable that the early Northumbrian kings were eating their way through their territories in the 7th century. While the king was in residence at Yeavering, he may have collected tribute, hosted feasts, and settled disputes. A second level in the settlement hierarchy is represented by the settlement at Bloodmoor Hill, Carlton Colville, Suffolk (Lucy et  al. 2009). The site was occupied from the 6th to the earlier 8th century CE, and it was subject to a recent program of excavation. The excavated area included over 30,000 square meters, and the excavations uncovered 38 sunken-featured buildings, 9 post-built structures, and over 270 pits. The site yielded extensive evidence for metal-working, including crucibles, molds, and substantial collections of scrap metal. These data suggest that

Settlement Changes in the Late 6th and 7th Centuries

73

specialized craft production was taking place at the site, probably carried out by specialists who were attached to the lord of the estate. The mid- to late 7th-century graveyard also included high-status female burials. The excavators have suggested that Bloodmoor Hill may represent an early estate center. Archaeozoological data also support this interpretation. While most Early Anglo-Saxon sites have a pastoral economy based on autarky or economic self-sufficiency (Crabtree 2010b), the pattern of animal management seen at Bloodmoor Hill is more complex. While pigs seem to have been raised for local consumption and cattle were kept for dairying and for traction, the sheep seem to have been reared for more specialized wool production (Higbee 2009). This kind of specialized production is seen much more commonly at later Middle Saxon estate centers including Brandon in Suffolk, Wicken Bonhunt in Essex, and Flixborough in Lincolnshire (Dobney et al. 2007; Crabtree 2010b, 2012; Crabtree and Campana 2015). Additional intriguing data come from the 7th-century site of Cowdery’s Down in Hampshire (Millett and James 1983). Excavations revealed 18 structures that were dated to the 6th–7th centuries CE on the basis of radiocarbon. They included 16 posthole or trench-built constructions and only two sunken-featured buildings. The buildings were assigned to three sequential phases, and the settlement was composed of three, six, and ten major structures along with two fenced enclosures in each phase. A  striking feature of Cowdery’s Down is the presence of these fenced enclosures, which are not seen on earlier sites. The size and techniques used to construct these buildings set them apart from most of what is known of Anglo-Saxon domestic architecture. Marshall and Marshall (1991, 1993) surveyed 267 Anglo-Saxon domestic buildings from the 6th through the 9th centuries CE. Several of the buildings from Cowdery’s Down are among the largest and widest structures included in their survey. The largest structures at Cowdery’s Down are also distinguished by the presence of possible cruck architecture (Figure 2.7). Using this method, the outer door frames extend into the roof, and the roof is supported by curved timbers known as crucks. The presence of this type of architecture suggests the presence of skilled carpenters who may have been attached to local lords. Millett and James (1983, 247) suggest that the large buildings reflect high social status. In each phase, a major structure is located inside, but close to, the enclosures. The excavators suggest that this may represent a chief’s house. There is also evidence for careful planning in Phase C.  The largest building, structure C12, is centrally located on the crest of a ridge. Very few artifacts were recovered from the site, and the excavators (Millett and James 1983, 249) suggest that the

74

Society and Settlement in Early Anglo-Saxon England

Figure 2.7. Reconstruction of structure C12 at Cowdery’s Down. Copyright: Royal Archaeological Society, reproduced with permission.

site might have been occupied seasonally as part of a peripatetic kingship, a feature that is well known from the later Middle Saxon period. Cowdery’s Down, along with Yeavering, provides some of the earliest evidence for repair, remodeling, and in situ replacement of Anglo-Saxon buildings. Two of the buildings from Phase B of the Anglo-Saxon settlement were rebuilt in the same location in Phase C (Millett and James 1983, 213; Hamerow 2011, 135).2 The choice to rebuild on the same footprint, along with the presence of fenced enclosures around the structures, suggests that attitudes toward space, and possibly land and property ownership, were beginning to change in the 7th century. In this context it is interesting to note that some of the very latest structures at West Stow are a series of ditches, some of which cut through earlier Anglo-Saxon features, which may represent property boundaries. As noted above, these ditches were originally dated to the 7th century on

2

See Hamerow 2012 for an up-to-date review of Anglo-Saxon rural settlements.

Settlement Changes in the Late 6th and 7th Centuries

75

the basis of Ipswich Ware (West 1985, 54). With the redating of Ipswich Ware, they are now almost certainly 8th century. The process of settlement transformation in the late 6th and 7th centuries can be seen most clearly at the Anglo-Saxon site of Lyminge in Kent (Thomas 2013). The Early Saxon settlement at Lyminge was established in the 5th century. Up until the late 6th century, the settlement is composed of small, earth-fast timber buildings and sunkenfeatured buildings, but unlike the settlement evidence known from West Stow and Mucking, artifactual and burial evidence indicate that Lyminge was home to one or more rich and powerful households (Thomas 2013, 120). The small finds from the site include a diversity of glass vessel forms, and the Early Anglo-Saxon faunal assemblage is dominated by the remains of pigs, which are often associated with high-status settlements (Thomas 2013, 125). The Lyminge cemeteries included evidence for a horse burial and a grave of a 6th-century woman who was wearing a Scandinavian gold bracteate. A  bracteate is a small, circular gold pendant that is worn around the neck; it is often associated with women of the highest rank. The character of the settlement changes in the late 6th century with the construction of a very large timber hall at the headwaters of the Nailbourne River. The hall measures 21 m by 8.5 m and includes internal partitions at the east end. A smaller post-in-trench building is located to the east of the great hall, suggesting that these buildings are part of a planned cluster of high-status buildings. The great hall is closely comparable to the buildings seen at Cowdery’s Down and Yeavering (Thomas 2013, 126–7). Place-name evidence suggests that Lyminge was a royal vill at this time.3 The site subsequently emerges as a monastic center in the late 7th century. While there is no clear archaeological evidence for the development of urban centers in the later 6th and 7th centuries CE, the changes that we see in settlement patterns are reflected in other aspects of the archaeological record and in the beginnings of the historical record as well. Thirty years ago, Arnold (1984, 280) pointed out that we see a clear change in Anglo-Saxon burial practices in the early 7th century. In the 6th century and earlier, the richest Pagan Saxon graves are found in communal cemeteries. By the 7th century, the richest graves contain even greater wealth and are often found isolated from the rest of the population. 3

A royal vill or villa regia was part of the peripatetic settlement system that characterized the Anglo-Saxon kings from the 7th century onward. Food rents were collected at the royal vills, and the king and his retinue ate their way through the kingdom by stopping at various vills throughout the year. A similar system has been documented for pre-contact and early historic Hawaii (see Kirch 2012).

76

Society and Settlement in Early Anglo-Saxon England

Figure 2.8. Image of Sutton Hoo Mound 1 as it exists today. Photo credit:  Pam J.  Crabtree, reproduced with permission of the National Trust.

The classic example of a wealthy 7th-century grave is the burial in Mound 1 at Sutton Hoo (Figure 2.8). The site was initially discovered in 1939 on the eve of World War II and was published by Bruce-Mitford (1972). Mound 1 contained the remains of a ship burial, although the body of the deceased was never recovered. A  richly furnished burial chamber was constructed in the center of the 27-m-long ship. The grave was equipped with a helmet, shield, and sword, as well as large quantities of decorated metalwork, including a belt buckle, shoulder clasps, and a purse lid. The purse contained 37 gold tremises, each from a different mint in Francia, as well as three coin blanks and two ingots. The body was accompanied by Byzantine silver bowls and two silver spoons engraved with the names Paulos and Saulos (Paul and Saul). Other grave goods include a mail coat, leather shoes, textiles, and drinking horns made of aurochs (wild cattle) horn (Figure 2.9). Between 1983 and 2001 Martin Carver (2005) led an ambitious program of survey and excavation that was designed to place the Sutton Hoo grave mounds into a broader landscape context (see also Williamson 2009). It included extensive survey of the cemetery and the surrounding area and re-excavation of a number of the burial mounds. Carver’s

Settlement Changes in the Late 6th and 7th Centuries

77

Figure 2.9. Photograph of the reconstruction of the grave from Mound 1, Sutton Hoo (from the Sutton Hoo Museum). Photo credit:  Pam J.  Crabtree, reproduced with permission of the National Trust.

research showed that the initial 6th-century cemetery at Tranmer House was established in the northern part of the Sutton Hoo area. It contained cremations and inhumations and continued in use until the close of the 6th century (Fern 2015). It began as a folk cemetery, but wealthy individuals were clearly present by the later part of the 6th century. The successor cemetery at Sutton Hoo that included Mound 1 was opened at the end of the 6th century “to commemorate people with enhanced social pretensions” (Carver 2005, 490). This cemetery includes nearly 20 mounds that were constructed in the late 6th and 7th centuries. These burials may be associated with the kings that appear in Anglo-Saxon king lists for East Anglia, since the site is close to the royal center at Rendlesham. During the 8th to 11th century, the site was transformed into a place of execution. What kind of settlement might have been associated with Sutton Hoo? While Bede reports that Swithelm, king of the East Saxons, was baptized at Rendlesham, the exact location of this Anglo-Saxon royal settlement was unknown until recently. In 2007 a local landowner in Rendlesham reported that illegal metal detectorists were looting his fields at night. His

78

Society and Settlement in Early Anglo-Saxon England

complaint eventually led to an extensive program of archaeological survey that covered 160 ha. Systematic metal detecting was used to identify and record finds from the plow soil, and magnetometry (see Chapter 4) was used to identify buried remains in the areas where the surface finds were densest (Scull et al. 2016). Magnetometers detect slight changes in the earth’s magnetic field, which can be caused by sub-surface features such as pits and ditches and iron objects. The archaeologists also examined aerial photographs of the region. The survey revealed a concentration of small finds and features in an area covering about 50 ha (Scull et al. 2016, 1597). This area is located about 4 miles (7 km) northwest of Sutton Hoo along the River Deben. The Anglo-Saxon finds range in date from the 5th through the 8th century CE, but the site reached its zenith between the early to mid-6th century and the second quarter of the 8th century (Scull et al. 2016, 1601; Minter et al. 2014, 52). The finds provide evidence for metal-working, as well as dress fittings, jewelry, silver coins, and mounts for hanging bowls, all from northern and western Britain (Figure  2.10). Imported items include Coptic bowls from the eastern Mediterranean, in addition to Byzantine copper coins and Merovingian gold coins (Plouviez 2014). The presence of coins and weights indicates that Rendlesham may have been the location of periodic fairs or markets. The evidence for trade and commerce decreases in the early 8th century, just as Ipswich was expanding (Minter et al. 2014, 53). The role of Ipswich as a major center for trade and exchange during the 8th and 9th centuries will be examined in Chapter 3. The rich and extensive settlement at Rendlesham indicates that the emerging East Anglian royal house was not just collecting food rents from the surrounding countryside, but it was also engaged in craft production, trade, and exchange. Scull et al. (2016, 1605) suggest that it served as “a permanent centre for agrarian or economic administration, a periodic residence for a peripatetic elite, and a periodic meeting places for military and jurisdictional assemblies.” While Sutton Hoo Mound 1 is probably the best known of the wealthy early 7th-century Anglo-Saxon graves, it is certainly not the only one. One of the most spectacular finds of the 21st century is the so-called prince’s grave at Prittlewell in Essex (Blair, Barnham, and Blackwell 2004). The burial was discovered in a known Anglo-Saxon cemetery by the Museum of London Archaeological Service in 2003 in advance of a road improvement project. It included an underground chamber that was probably originally covered with a mound. The chamber had timber-lined walls and a plank roof and was 4 m square and about 1.5 m deep. When the roof collapsed, the contents of the burial were sealed (Figure 2.11). Although the acidic soils destroyed all traces of the body, the burial was furnished

Settlement Changes in the Late 6th and 7th Centuries

79

Figure  2.10. Photograph of artifacts recovered from Rendlesham, a royal estate center that is probably associated with the cemetery at Sutton Hoo. Copyright: Suffolk County Council, reproduced with permission.

with weapons and feasting equipment, as well as two gold foil crosses, two Merovingian gold coins, and a lyre. Other imported items include a flagon and bowl from the eastern Mediterranean, Coptic bowls, and a Byzantine silver spoon similar to the ones that were recovered from Sutton Hoo. A number of other rich late 6th–earlier 7th-century princely burials are known from other parts of southeast England, although most were excavated before the advent of modern archaeological techniques. These burials are generally covered with barrows and usually command long views. A  well-known example is the 7th-century burial at Taplow in Buckinghamshire. The barrow, which dominates the surrounding landscape, was excavated in 1883 under less than optimal conditions. The deceased was buried facing west in an oak chamber. He was laid out on a bier and accompanied by three sets of weapons and 19 feasting and drinking vessels, including four glass claw beakers and gilt silver

80

Society and Settlement in Early Anglo-Saxon England

Figure 2.11. Reconstruction of the Prittlewell, Essex chamber burial. Copyright: Museum of London Archaeology, reproduced with permission.

mounts for drinking horns that were made from aurochs (wild cattle) horn. Since aurochs had been extinct in Britain for many centuries, these horns must have been imported from somewhere in east-central Europe. Other imported items include a Coptic bowl and stand. Luxury items such as the gold and garnet buckle are now housed in the British Museum. Many of the items in the burial appear to be Kentish in origin, and Webster (2001) notes that this region represented the furthest west extent of Kentish power at about 600 CE; the area subsequently came under Mercian domination. Webster suggests that the interred may represent a Kentish sub-king who ruled under Mercian authority. The archaeological data clearly show that by the late 6th to earlier 7th-century housing and burial rites were being used to express differences in social status, political power, and material wealth. While Anglo-Saxon

Settlement Changes in the Late 6th and 7th Centuries

81

England of the 5th and earlier 6th century was by no means an egalitarian society, these archaeological data can be interpreted to suggest the emergence of a small, powerful class of chiefs or kings by the late 6th century. Drawing on the more limited historical data, Yorke (1989, 1993, 2003) has suggested that the 5th- and 6th-century confederations of Angles and Saxons broke up and were transformed into a series of smaller kingdoms around 600 CE. She further suggests that it was the Franks trying to control cross-channel trade who may have provided the impetus that caused powerful military leaders to transform themselves into ruling dynasties (Yorke 2003). The question of how the transformation took place has been poorly theorized, and only a few explanations for this process have been proposed. Many historians have simply avoided this issue because the documentary record for the period is so limited. Unlike prehistoric archaeologists, many Anglo-Saxon archaeologists have been reluctant to draw parallels to similar processes of socio-political change outside Europe. However, Scull (1993), Yorke (2003), and Oosthuizen (2011, 2016) have addressed this issue from three different, but not necessarily mutually exclusive, perspectives. Scull (1993) was one of the first archaeologists to address this issue from a processual perspective. Like Hodges (1982) before him, Scull (1993, 67) suggests that the control of trade played a major role in the emergence of Anglo-Saxon kingship. Scull sees Early Anglo-Saxon society as made up of internally ranked, patrilineal descent groups farming their own territories. He suggests that there was competition between these lineages, and, by the later 6th–7th century, there was competition between elites at both the local and regional levels and an increase in social ranking. This can be seen in the emergence of the settlement hierarchy and the princely graves. Scull (1993, 76) also suggests that external links may have played a role in this transformation. Scull (1993, 77)  sees this transformation as a multi-stage process beginning with ranked lineages competing with one another. Some initially achieve local hegemony; then competition between local chiefs leads to temporary regional or interregional hegemonies. These, in turn, are transformed into a more permanent political hierarchy with a more formal administrative organization. Scull (1993, 77) sees control of trade as a critical factor in this process because it allowed for the acquisition and redistribution of exotic imported items. He also suggests that population growth may have been a motivating factor, as it put pressure on land as a social resource (Scull 1993, 77–8). Yorke (2003) has taken a different approach to the formation of the Anglo-Saxon kingdoms. As noted above, she suggests that in the late 6th

82

Society and Settlement in Early Anglo-Saxon England

to 7th century the older confederations of Angles, Saxons, and Jutes were broken up. At that time, we see the beginnings of a process of secondary state formation. While Yorke does not provide an explicit model for this process, she suggests that the ability to create and exploit both agricultural surpluses and other sources of wealth played an important role in this transformation (Yorke 2003, 402). She also suggests that the conversion to Christianity may have been an important underpinning of these Early Anglo-Saxon kingdoms (Yorke 2003, 406). Oosthuizen (2011, 2013) has recently proposed an alternative model for kingdom building in the 6th–7th centuries. She notes that Early Anglo-Saxon societies were pastorally oriented and suggests that there was a long-term continuity in grazing rights and rights to the common from prehistoric to early historic times. Oosthuizen suggests that these flexible, adaptive, and absorptive rights may have provided the foundation for the political, economic, and social changes of the Middle AngloSaxon period. While Scull’s model is based on competition, Oosthuizen’s focuses on consensus, legitimacy, and collective assent. Oosthuizen (2016) has developed these ideas further in a recently published paper. She suggests that long-term continuities in agricultural land use and forms of collective government are related to substantial continuity in population between Roman Britain and Early Anglo-Saxon England. She argues against “migration from northwest Europe as being the triggering factor in the emergence of new forms of political governance that led to the emergence of the early ‘Anglo-Saxon’ kingdoms” (Oosthuizen 2016, 209). However, she admits that this argument does not mean that there is no merit in an elite replacement or dominance model. Long-term continuities in property rights, land use, and population are not necessarily inconsistent with the emergence of a dominant military elite, which may have included both indigenous inhabitants and continental migrants or their descendants. One point that deserves further discussion is Yorke’s contention that the ability to create and exploit agricultural surpluses (and other sources of wealth) was crucial to state formation in Anglo-Saxon England. Here Yorke is essentially talking about what Childe (1950) referred to as a social surplus. It is the creation and extraction of the agricultural surplus that is critical to its use in political economy. Patron–client relationships, structured relationships between people of unequal rank, may have played a critical role in the creation and extraction of social surplus in Anglo-Saxon England. These relationships are features of traditional Roman, Celtic, and Germanic societies, and they would have entailed rights and responsibilities on behalf of both the patron and the client.

Western Britain in the 5th–7th Centuries

83

Archaeological data and literary evidence can be used to try to reconstruct this process. One of the obligations that a client would have had to his chief or king would have been food rent. The chief could have used the food rent as a form of social capital to support activities such as feasting. Feasting plays an important role in Anglo-Saxon literature such as Beowulf, and it is significant that the princes’ graves contain a number of items that seem to be associated with feasting. Feasting encourages loyalty, which is crucial in warfare and in competition between rival elites. The princes’ graves all include substantial sets of weaponry. Control of exotic luxury items is important not only because they are symbols of status and wealth, but also because they can be used to reward loyalty in followers. I would argue that those leaders who were most successful patrons were likely to have emerged as kings in the early 7th century. It could even be argued that some of the continental elites were patrons of the emerging Anglo-Saxon royal houses. Formalizing many of the traditional patron–client ties may have served as the basis for the administrative structures of the Middle Anglo-Saxon period. A focus on patronage is compatible with the models proposed by Scull, Yorke, and Oosthuizen. Following Scull (1993), successful patrons will be better able to compete with other elites at the local and regional levels. Successful patrons may have styled themselves as the “heirs of Rome,” following Yorke (2003, 402; see also Oosthuizen 2016). As Hunter (1974) has shown, by the beginning of the Middle Anglo-Saxon period, the Roman and Germanic traditions had been successfully fused in the minds of the Anglo-Saxons. In addition, successful patrons may have been part of large groups who shared common grazing rights, following Oosthuizen (2011). Elite patrons who were best able to control economic surpluses and use them to their own benefit may have emerged as kings by the early 7th century. In conclusion, it is worth emphasizing that despite the changes in settlement patterns, social organization, burial practices, and rural economy that took place in the later part of the Early Anglo-Saxon period, these changes did not directly lead to or necessitate the re-emergence of towns at this time. However, these changes provide the social and economic background for the re-emergence of towns in the Middle Saxon period, which will be discussed in the following chapter. Western Britain in the 5th–7th Centuries The patterns of settlement, subsistence, and technology that existed in western Britain during the 5th through the 7th centuries are strikingly

84

Society and Settlement in Early Anglo-Saxon England

different from those that we have seen in eastern England. At the top of the settlement hierarchy are a series of fortified sites, including a number of reoccupied and refortified Iron Age hillforts (Burrow 1981). These includes sites such as Dinas Powys in Glamorgan, Wales (Alcock 1963, 2001) and Congresbury–Cadbury in Somerset (Rahtz et al. 1992), which may have served as royal centers for sub-Roman British polities. These polities were ruled by kings, described by Gildas as tyrants, but who are probably military strongmen and their retinues (Costen 2011, 16). The various fortified sites may have been used on a peripatetic basis by these kings (Thomas 1988, 429). One of the most striking features of these western British sites is that they have also yielded evidence for Mediterranean and continental imports. A comprehensive study of imported pottery and glass from sites in southwestern England, Wales, Scotland, and Ireland has pointed to two successive patterns of trade between Atlantic Britain and Ireland and the Mediterranean and continental Europe (Campbell 2007). The first period features Roman amphorae and fine red-slipped tablewares from the eastern Mediterranean and North Africa and can be dated to between the late 5th and the mid-6th century. Distributional data show that most of these imports are concentrated in southwestern Britain at sites such as Tintagel in Cornwall (Thomas 1988), and the Byzantine Empire’s need for tin may have motivated this trade (Campbell 2007, 132–3 and Fig. 83). The second period, dating from the later 6th and 7th centuries, is marked by imports from western Merovingian France, including glass and coarse- and fine-ware pottery. Some of this pottery, including E-Ware,4 may have been used to import commodities such as dyestuffs, and it was then reused for lighting, industrial purposes, and possibly as cooking pots (Campbell 2007, 51). Campbell (1996, 1999) has suggested that surplus wealth, including Mediterranean imports, was used by these tyrants or petty kings to develop more complex political hierarchies. Many of the royal sites show evidence for craft specialization such as jewelry-making. Unlike later Anglo-Saxon England, however, western Britain did not develop “a true market economy with coinage, a merchant class and towns” (Campbell 2007, 140). Why these failed to develop is not clear. It is in the Middle Anglo-Saxon period in eastern England that we see the beginnings of these developments. The period is traditionally dated between 650 and 850 CE, and it was an era of major social, economic, and political change in eastern England. Some of the most striking changes 4

The original research on E-Ware was carried out by B. Wailes (1964; see also Crabtree 2014). The major problem is that the kilns where this pottery was produced have not yet been identified.

Western Britain in the 5th–7th Centuries

85

are seen in settlement patterns, including the widespread appearance of both secular and ecclesiastical estate centers, the appearance of “productive sites” with large quantities of metalwork finds (Ulmschneider 2000; Ulmschneider and Pestell 2003; see also Pestell 2011), and the rise of a group of craft production and trading centers known as the emporia or wics that may represent the beginnings of urbanism in post-Roman Britain. We will examine these changes in the following chapter.

3

Middle Saxon Settlement and the Rise of the Emporia: The Archaeology of the Wics and Contemporary Sites

Introduction Archaeologists have long recognized that substantial changes in settlement patterns began in the 7th century in many parts of Anglo-Saxon England. The so-called “7th-century shuffle” or “Middle Saxon shift” (Arnold and Wardle 1981; Hamerow 1991) was marked by the abandonment of many Early Anglo-Saxon habitations and the establishment of new types of settlements including high-status estate centers and the emporia, centers of craft production and regional and international trade (Hill and Cowie 2001). As we saw in Chapter 2, the process of settlement shift is actually more protracted than was traditionally thought. High-status settlements, including royal residences such as Yeavering and possible estate centers, first make their appearance in the late 6th and early 7th centuries. The final phases of sites such as West Stow were traditionally dated to the mid-7th century based on the presence of small quantities of Ipswich Ware pottery. Blinkhorn’s (1999, 2012) research, however, has shown that Ipswich Ware pottery was not produced until about 700 CE, and it was not produced in substantial quantities until at least a quarter of a century later. The new dating evidence is important because it suggests that we are not looking at a short-term episode of settlement shift in response to changing political or environmental factors. Instead, we may be looking at a more protracted period of settlement change. This appears to be a process that took place somewhat more gradually throughout the “long eighth century” (see Hansen and Wickham 2000). This chapter will focus on the archaeology of the emporia or wic sites – including Ipswich, Hamwic (Saxon Southampton), Lundenwic (Middle Saxon London), and Eorforwic (Anglo-Saxon York)  – in the Middle Saxon period (Figure 3.1). In the early medieval historical sources, the 86

0

100

200

300

400 km

Figure 3.1. Map of Britain showing the location of the sites discussed in this chapter. Drawn by Douglas V.  Campana using open-source software. Map numbers: Ipswich 1, Hamwic 2, Lundenwic 3, Eoforwic 4, Ramsbury 5, Brandon 6, Flixborough 7, Wicken Bonhunt 8, Fordwich 9, Sandwich 10, Sandtun 11

88

Middle Saxon Settlement and the Rise of the Emporia

Latin words emporium (trading place or market), mercimonium (merchandise or goods), and portus (harbor or port) are sometimes used to refer to these sites. For example, Alcuin describes 8th-century York as “emporium terrae commune marisque,” which has been translated as a seat of commerce by land and sea (Palliser 2014, 35–6). However, the sites are commonly referred to as wics since many include the Old English suffix wic, which here refers to trading settlement or harbor (Rumble 2001; Malcolm and Bowsher 2003, 2). It is derived from the Latin word vicus, which can refer to a village, a neighborhood, or part of a town. O’Connor (2001, 54) has provided a useful working definition of the wics. He suggests that they are “undefended sites of nucleated occupation of the later seventh to mid-ninth centuries, located around the North Sea basin and its tributary rivers, having no clear ecclesiastical function and showing evidence of, and trade in, a range of commodities, typically high value and low volume.” He further suggests that wics were not “sites of agrarian production, that is, they are not ‘farms’ ” (O’Connor 2001, 54). Whether this definition is appropriate for all the wic sites is a question that we will attempt to answer in this chapter. Ipswich, Southampton, London, and York have all been the focus of substantial programs of urban archaeology since the early 1970s, and our understanding of their layout and character has changed substantially in the past 40 years. We will consider whether these sites represent the earliest urban settlements in post-Roman Britain. We will also consider whether the term “proto-urban,” which is often applied to these sites, is appropriate. If these sites themselves are not fully functioning towns, then did they pave the way for the development of true towns in the Late Anglo-Saxon period? However, we will begin by examining changing settlement patterns in the Anglo-Saxon countryside, since, as Falconer (1995) has argued, the rise of urbanism affects not only the urban centers themselves, but also the surrounding countryside. Background The Middle Anglo-Saxon period in England was an era of significant social, political, economic, and religious change. It is important to review some of these changes as they are all directly or indirectly related to the appearance of the first towns in post-Roman Britain. The reintroduction of Christianity to eastern England began with Augustine’s mission to Canterbury in 597. Historical sources indicate a degree of ambivalence toward Christianity in the 7th century. Some kings converted to Christianity only to have their sons revert to

Middle Saxon Changes in Settlement Patterns

89

Paganism. This ambivalence can be seen in the Sutton Hoo burial itself. Although the ship burial with its weapons and feasting equipment is laid out in a traditional Pagan style, the presence of two silver spoons engraved with the names Paulos and Saulos provides at least a nod to Christianity. By the late 7th and 8th centuries, however, the nascent English church was playing an increasingly important role in Anglo-Saxon politics (see Blair 2005 for a detailed discussion of the role of the church in Anglo-Saxon England). Historic charters reflect grants of large estates to monasteries and other religious leaders by the Anglo-Saxon kings. The emerging Anglo-Saxon kings became patrons of the church. In return, the church provided a degree of legitimacy for these Anglo-Saxon monarchs (Yorke 2003, 405). Politically, the Middle Saxon period is characterized by the emergence of the heptarchy, seven Anglo-Saxon kingdoms including Northumbria, Wessex, East Anglia, Mercia, Essex, Sussex, and Kent. The Tribal Hidage (Figure  2.5), a late 7th-century document which probably shows an assessment for tribute to the Mercian king Wolfhere (658–75), reveals the existence of a number of smaller polities in areas such as the Fenland. Some of these smaller polities may well have been clients of the larger kingdoms of the heptarchy. By the end of the Middle Anglo-Saxon period, four kingdoms remain dominant:  Mercia, East Anglia, Northumbria, and Wessex. It is worth noting that the four major wic sites are located in these kingdoms – Hamwic in Wessex, Ipswich in East Anglia, Lundenwic in Mercia, and Eorforwic in Northumbria. The Middle Anglo-Saxon period has traditionally been dated between about 650 and 850 CE. This dating was based, in part, on the production and distribution of Ipswich Ware, the first mass-produced pottery in postRoman Britain (see below). Ipswich Ware was distributed throughout East Anglia and at major ecclesiastical and secular sites throughout eastern England. As noted above, recent research (Blinkhorn 1999, 2012), however, has shown that Ipswich Ware was not produced until the early 8th century (ca. 700–720 CE). This change should encourage scholars to rethink the chronology of the Middle Saxon period (see the discussion of the “7th-century shuffle” below). The end of the Middle Anglo-Saxon period is more unambiguously dated by the appearance of the Viking army in eastern England. Middle Saxon Changes in Settlement Patterns A number of changes are apparent in the patterns of settlement in Middle Anglo-Saxon England beginning in the 7th century. As noted

90

Middle Saxon Settlement and the Rise of the Emporia

above, the first, sometimes termed the “7th-century shuffle” or “Middle Saxon shift” (Arnold and Wardle 1981; Hamerow 1991), is marked by the abandonment of several important Early Anglo-Saxon sites and the establishment of new sites in the early Middle Anglo-Saxon period. These include a number of both ecclesiastical and secular estate centers. A second change is the appearance of “productive sites,” sites that produced substantial quantities of coins and copper-alloy and other non-ferrous metal artifacts (Ulmschneider 2000; Ulmschneider and Pestell 2003). A  third change is the appearance of small-scale trading places along both sides of the North Sea (Loveluck and Tys 2006). Finally, the Middle Saxon period is marked by the appearance of the emporia, centers of trade and manufacture, probably operating under royal auspices. We will examine each of these developments separately. The “7th-Century Shuffle” and the Establishment of New Estate Centers As noted above, one striking feature of the Middle Saxon period is the abandonment of a number of substantial Early Anglo-Saxon settlements and the establishment of new settlements in areas that were not intensively occupied in the Early Anglo-Saxon period. In terms of the AngloSaxon villages we examined in the previous chapter, both West Stow and Mucking appear to have been abandoned around the beginning of the Middle Saxon period, while West Heslerton survived into the Middle Saxon period, but in a much reduced form. The latest phase of West Stow is particularly instructive here. The final features at West Stow contain sherds of Ipswich Ware, indicating that they survived into the beginning of the 8th century. These features include a number of ditches, probably boundary ditches that might reflect changes in land ownership and/or land use practices at the beginning of the Middle Saxon period. The site of Bloodmoor Hill, which was mentioned in Chapter 2, may be the site of an early estate center. During the Middle Anglo-Saxon period, a number of new estate centers were established; many of these were in locations that had no substantial Early Anglo-Saxon activity. While the church played a major role in the establishment of some of these settlements, both ecclesiastical and non-ecclesiastical centers have been identified, and several have been the subject of major programs of excavation since the 1970s. One of these Middle Anglo-Saxon estate centers that was initially excavated in the 1970s was the site of Ramsbury in Wiltshire (Haslam

The “7th-Century Shuffle”

91

Table 3.1. Major sites discussed in Chapter 3 Site Type

Name

Location

Date

Estate center Estate center Estate center Estate center Wic Wic

Ramsbury Brandon Flixborough Wicken Bonhunt Gippeswic Hamwic

Middle Saxon Middle Saxon Middle to Late Saxon Middle Saxon Middle Saxon and later Middle Saxon

Wic Wic ?Wic ?Wic ?Wic

Eorforwic Lundenwic Fordwich Sandwich Sandtun

Wiltshire Suffolk Lincolnshire Essex Ipswich, Suffolk Southampton, Hampshire York London Kent Kent Kent

Middle Saxon and later Middle Saxon Middle Saxon and later ?Middle and Late Saxon ca. 700–ca. 875

1980). (The sites discussed in this chapter are listed in Table 3.1; see Figure 3.1 for site locations.) Although it was originally identified as a specialized metal-working site, it now appears to have been an estate center in the Middle Saxon period. The faunal remains from the site are striking because they include a high proportion of deer remains (Coy 1980). As Sykes (2010) has shown, deer hunting appears to have become a prestige activity in Middle Saxon times, while in earlier Anglo-Saxon times deer hunting served primarily as a supplement to the diet. Another important Middle Anglo-Saxon estate center is the site of Staunch Meadow Brandon in northwestern Suffolk. The Suffolk County Archaeological Unit, under the direction of Andrew Tester and Bob Carr (Carr et al. 1988; Tester et al. 2014) carried out eight seasons of excavation at the site between 1980 and 1988. Approximately one-third of the site, an area of 13,000 square meters, was excavated in advance of the construction of playing fields. Since the site was never plowed, the Middle Anglo-Saxon remains were exceptionally well preserved. This site is located on a branch of the Fens which follows the Little Ouse River. It sits on a small rise or “island” formed of windblown sands located in the floodplain of the river (Tester et al. 2014, 1), and it was approached by a small causeway (Figure 3.2). The site was initially occupied in the mid-7th century, and from that time until the early 8th century it was characterized by a dispersed settlement pattern. Buildings were erected on the northern and southern part of the settlement during this time. During the first half of the 8th

92

Middle Saxon Settlement and the Rise of the Emporia

waterfront

medieval enclosure

medieval church

high-status area

church

N

0

50 m

causeway

Figure 3.2. Plan of Staunch Meadow, Brandon, Suffolk. Copyright: Suffolk County Council, reproduced with permission.

century, settlement expanded in the center of the site, including a church and cemetery. Although no written evidence exists for Middle Saxon Brandon, Tester et al. (2014, 393) suggest that this represents the construction of a minster church, possibly associated with a parent monastery at Ely. A palisade-type fence and gate separated the church and cemetery from two farming households that were established in the western part of the site. In the later 8th and early 9th centuries, the area along the waterfront was fenced off, and three large halls were constructed there. Textile processing,

The “7th-Century Shuffle”

93

including bleaching and dyeing, was carried out in this part of the site. Artifactual remains, such as silver pins, indicate that Brandon was a wealthy community, and the presence of a stylus and ink wells demonstrates that at least some members of the community were literate. Brandon was clearly a Christian community, and it may have served as a monastic double house for part, if not all, of its existence. The evidence for wealth and conspicuous consumption suggests that this may have been a monastic foundation with a royal or aristocratic patron (Tester et al. 2014, 390). Overall, the excavation revealed 35 post-built timber buildings, plus fence lines, pits, ditches, hearths, and two churches and two cemeteries. In addition to the evidence from the riverside industrial area for wool and linen processing, faunal remains indicate that the site was engaged in specialized wool production (Crabtree 2007; Crabtree and Campana 2014, 2015). Recent analyses of the ceramics and other datable artifacts indicate that the site was abandoned in the later 9th century, probably as a result of the Danish invasions of East Anglia (Tester et al. 2014, 393). While Brandon may have been the center of an ecclesiastical estate for all or part of its history, Wicken Bonhunt in northwest Essex appears to have been the center of a secular estate. Wicken Bonhunt is a multiperiod settlement that was recovered as a result of ploughing in 1967. Rescue excavations were carried out between 1971 and 1973 (Wade 1980) on the portions of the site that were directly threatened by farming operations. Ceramics indicate a late 6th–7th-century presence at the site, but no Early Anglo-Saxon structures were recovered during excavation. The Middle Saxon features include boundary ditches, at least 28 structures, and two wells (Figure 3.3). A large channel to the south of the site may be the leat of a watermill. Although there is no direct evidence for the function of any of the Middle Saxon structures, some of the buildings are likely to have served as workshops, barns, and byres (Wade 1980, 96). There appear to be at least two major phases of building activity at the Bonhunt site (Wade 1980, 96). The location of the settlement boundaries suggests that one-third to one-half of the site has been excavated. The building alignments indicate that their layout was deliberately planned, suggesting a high degree of organization (Wade 1980, 98). The faunal remains from Wicken Bonhunt point to specialized pork production (Crabtree 1996, 2012). Flixborough in Lincolnshire is a Middle Anglo-Saxon estate center that has been both extensively excavated and thoroughly published (Dobney et al. 2007; Loveluck 2007; Loveluck and Atkinson 2007). The Flixborough estate center seems to have had two very different functions during the Middle and Early Late Saxon periods. In the earlier part of

94

Middle Saxon Settlement and the Rise of the Emporia

Figure 3.3. Plan of Middle Saxon Wicken Bonhunt, Essex. Copyright: Suffolk County Council, reproduced with permission.

the Middle Anglo-Saxon period, Flixborough appears to have been the center of a high-status secular estate. The artifactual and faunal data point to extensive and elaborate feasting activities. By the mid-9th century, Flixborough appears to have fallen into ecclesiastical hands. The feasting was replaced by an intensified focus on wool and textile production. The faunal remains show an increase in the proportion of sheep, and the artifactual data indicate that Flixborough was a center of specialized textile manufacture.

The “7th-Century Shuffle”

95

Alternative explanations of the Flixborough data are possible. Blair (2011) has noted that only a small part of the Flixborough settlement was excavated (see Blair 2011, 103, Fig. 3), and much of the material that was recovered came from reworked midden material rather than from areas in and around the buildings. Based on the large number of styli recovered from Flixborough, as well as the tendency for the buildings to be replaced on the same footprint and the maintenance of the same layout for centuries, Blair (2011) suggests that the site may have been monastic throughout its existence. Monasteries had access to cheap labor in the form of lay brothers, which may have made them particularly well positioned to engage in manufacture and trade (Blair 2005, 257). A striking feature of many Middle Anglo-Saxon estate centers is that they appear to be engaged in the specialized production of commodities and finished goods. Some are agropastoral, including wool at Brandon and pork at Wicken Bonhunt, while others are finished products, including textiles at Flixborough and metal tools at Ramsbury. These data indicate that the Middle Anglo-Saxon inhabitants of southern and eastern England were engaged in substantial trade and exchange activities. Agropastoral goods were certainly moving through the system as part of the feorm or food-rent system, but the identification of a number of “productive sites” points to more widespread patterns of local and regional trade and exchange (Ulmschneider 2000; Ulmschneider and Pestell 2003). The so-called productive sites are defined by the presence of substantial quantities of metal finds, including coins, copper-alloy objects, and other nonferrous metal items. The identification of these sites has been aided by the establishment of the Portable Antiquities Scheme (PAS) in England and Wales in 1997 (see, for example, Richards et al. 2009). This is a voluntary effort to encourage collaboration between private metal detectorists and archaeologists. It allows the location of metal finds that do not qualify as treasure to be recorded at county councils and local museums. The category of productive sites includes both sites such as Brandon and sites that are known only from metal detecting. In addition to the presence of coins and other metal artifacts, productive sites were located near major land or water routes. These data suggest that the productive sites were involved in some form of commerce, such as local markets or fairs. The church seems to have played a role in many, but not all, of the productive sites. As a result of gifts from kings, the church became a large landowner in Middle Anglo-Saxon times. Monastic foundations had access to substantial amounts of cheap labor. In addition to agropastoral production, many monastic houses engaged in such activities as craft and

96

Middle Saxon Settlement and the Rise of the Emporia

salt production. Since church estates were important centers of landholding, production, and consumption, as well as trade and exchange, many of the productive sites may be church-related (Ulmschneider 2000, 77). A final important feature of the Middle Saxon period is the evidence for agricultural specialization and intensification. Much of this information comes from detailed studies of animal bones and plant remains that have been recovered from archaeological sites. Faunal remains from a number of Middle Saxon sites in southern and eastern England provide evidence for specialization in the production of commodities such as wool and pork, and this shift appears to have begun early as the later 6th or early 7th centuries (Crabtree 2010b). This can be seen most clearly in the high proportion of pig remains from the 6th–7th-century contexts at the rural estate center of Wicken Bonhunt in Essex (Crabtree 2012, 57) and the high numbers of older sheep seen at the possible early rural estate center of Bloodmoor Hill (Higbee 2009, 58).1 Recent studies of cereal crop production based on charred plant remains point to a similar pattern of intensification (McKerracher 2016). While earlier studies suggested that wheat replaced hulled barley as the dominant cereal crop in Middle Saxon times (Banham 2010), recent research based on archaeobotanical remains that have been systematically collected during the past 25 years indicate that both barley and wheat were common cereal crops throughout the Early and Middle Saxon periods. The critical change that appears in the 7th and especially in the 8th century is the increasingly widespread use of oats and rye. McKerracher (2016) interprets these data to suggest that at a time when increasing amounts of land were being brought under cultivation Middle Saxon farmers made strategic choices of crops that were most appropriate to the local environmental conditions. It is worth noting that these changes took place at climatic amelioration, when conditions were both warmer and dryer than they were in the earlier Saxon period (P. Dark 2000, 171; Williamson 2013, 45). We will explore the archaeology of the wic sites below in the context of broader changes in Middle Saxon settlement and subsistence. A final important change in our understanding of the Middle Saxon period is a re-evaluation of the nature of coastal settlement on both sides 1

While the suggestion that the archaeological evidence for specialization in animal husbandry began as early as the late 6th–7th centuries has recently been challenged by Holmes (2013), the faunal assemblages she examined do not include any assemblages that can be dated to the late Early Anglo-Saxon period (Holmes 2013, 266–7). Her analysis did not include the available published data from both Bloodmoor Hill and late Early Anglo-Saxon Wicken Bonhunt.

Archaeology of the Wic Sites or Emporia

97

of the North Sea from about 600 to 1000 CE (Loveluck and Tys 2006). Recent surveys and excavations, combined with the reanalysis of older data, have shown that the coastal regions on both sides of the North Sea were permanently occupied throughout this period. They provide evidence for smaller trading centers, coastal settlements, and beach/dune sites that have access to imported items and that must have been part of an active coastal trade network during the early medieval period. They note that the inhabitants of these areas engaged in production of commodities like salt, fish, livestock, and wool for trade (Loveluck and Tys 2006, 161) and that they also have greater access to imported goods than their inland contemporaries (Loveluck and Tys 2006, 162). They were part of a broader network of trade and exchange that included everything from small beach and dune sites to the emporia or wics, the largest of the early medieval trading communities. A similar range of harbors and trading places has been identified on the Danish island of Gotland dating to the same period, ca. 600–1000 CE. Carlsson (1992) has suggested that the Danish trading places may have served as an important point of connection between the trading places, including the wic sites, of the North Sea and those of the Baltic regions. One possible trading center has recently been excavated in the Lincolnshire fenlands. The site, located near the village of Little Carlton, was initially discovered by a metal detectorist who reported the discovery of an 8th-century silver writing stylus to the Portable Antiquities Scheme in England. The site, which would have been an island in the fenlands during the Anglo-Saxon period, was subsequently excavated by archaeologists from Sheffield University (Townend et al. 2016). The Little Carlton site is clearly a high-status Middle AngloSaxon occupation dated between ca. 650 and 780 CE. Finds recovered from the site include 21 styli, a lead tablet bearing the name Cudberg, about 300 dress pins, and a large number of coins from the 7th and 8th centuries. The site, which was previously unknown, appears to be a center of extensive trading and industrial activity. The substantial evidence for literacy indicates that this site might be monastic, which would make it quite similar to the productive sites that were discussed above. Archaeology of the Wic Sites or Emporia Of the well-known emporium sites  – Ipswich, London, Saxon Southampton, and York – Ipswich and Hamwic have been excavated over the longest period of time. We will examine the archaeological evidence for these two towns first, and then we will explore the archaeology of

98

Middle Saxon Settlement and the Rise of the Emporia

Middle Saxon York and London. We will conclude with some comments on other possible emporium sites in eastern England. Ipswich In the 1950s, John Hurst and Stanley West examined the medieval pottery that was housed in the Ipswich Museum. They determined that most of it was Anglo-Saxon and that much of it had been made within the town of Ipswich. The earliest of the pottery was what they termed Ipswich Ware, and they dated it to the Middle Saxon period (ca. 650–850 CE). Stanley West also conducted the first government-funded excavations within the town of Ipswich (see Wade n.d. for a detailed history of the excavations in Ipswich). West’s excavations at Cox Lane in 1958 and Shire Hall Yard in 1959 revealed that Middle Saxon Ipswich (Gippeswic in Anglo-Saxon) covered an area of at least 30 ha and functioned as an international port (West 1963). By the early 1970s archaeologists recognized that Ipswich was one of a small number of northwest European emporia, substantial early medieval trading settlements that display some urban characteristics, including evidence for craft specialization, a substantial population, and a planned layout. Archaeological research in Ipswich expanded in 1974 with the creation of the Suffolk Archaeological Unit, initially under the auspices of the Scole Committee for East Anglian Archaeology and subsequently as a unit of the Suffolk County Council. Keith Wade was appointed as the urban archaeologist that year, and the Origins of Ipswich Project was begun in 1974 in order to study the growth and development of Ipswich in the early medieval period. Ipswich was a de novo settlement in the early medieval period. The nearest large Roman town is located at Felixstowe. The chronology for Ipswich, based on ceramics, is shown in Table 3.2. Recent archaeological research has identified a major cemetery and settlement in Ipswich on the north bank of the River Orwell that dates to the 7th century CE (Scull 2009). The Buttermarket cemetery included the graves of at least 71 men, women, and children, and it has been radiocarbon dated to between 610/635 CE and 680/690 CE (Wade n.d.). Figure  3.4 shows the location of the sites that were excavated within the town as part of the Origins of Ipswich project. Only a portion of the cemetery was excavated, but Scull (2013, 223) estimates that the cemetery originally included between 400 and 450 individuals. Many of the burials were placed in wood-lined coffins or containers, and two of the containers appear to be small boats. These types of graves are rare in England, but

Archaeology of the Wic Sites or Emporia

99

Table 3.2. Ceramic chronology for Anglo-Saxon and early medieval Ipswich Period

Dates

Pottery

Early Middle Saxon Middle Saxon

600–700 CE 700–850/880 CE

Early Late Saxon

850/880–900 CE

Middle Late Saxon Early Medieval

900–1000 CE 1000–1150

Handmade pottery Ipswich Ware and continental imports Thetford Ware; decrease in Ipswich Ware Thetford Ware and St. Neot’s Ware Early Medieval Ware

Data source: Wade (2012).

common in the Merovingian regions of the European continent (Scull 2013, 224). Some of the burials were surrounded by ditches, suggesting that they may have been covered with a mound, similar to the mounded graves known from the Sutton Hoo cemetery located 9 miles (15 km) northeast of Ipswich (Carver 2005). Evidence for the 7th-century settlement in Ipswich was discovered at the St. Peter’s Street/Greyfriars Road site located near the modern waterfront. Excavations revealed two grubenhaüser and nine pits, which contained handmade pottery which predates the manufacture of Ipswich Ware. The 7th-century settlement in Ipswich covered between 6 and 30 ha on the north bank of the Orwell River (Scull 2013, 220). The pottery from the site included a high proportion of imported wares, suggesting that this site was designed as a special-purpose settlement to channel exchange contacts with the continent (Scull 2009, 314). Wade (1988, 1993) has suggested that the foundation of Ipswich may have been associated with the East Anglian royal house, although Kentish or continental authorities may also have played a role. As Loveluck and Tys (2006) have demonstrated, small-scale trading sites are found along both sides of the North Sea beginning as early as 600 CE. In his pioneering work Dark Age Economics, Hodges (1982) initially suggested that the role of these sites was to control trade in high-status imported goods, positing what would today be termed a prestige goods model (Friedman and Rowlands 1977). However, more recent research has suggested that these sites may have served to channel and tax trade (see, for example, Loveluck and Tys 2006). Taxes and duties would have provided a significant source of revenue, and the ability to waive taxes would have been a source of power. Moreover, Moreland (2001) has emphasized the importance of regional trade in the economies of the emporia.

100

Middle Saxon Settlement and the Rise of the Emporia

Figure  3.4. Plan of Ipswich excavations showing the location of sites within the town. Copyright: Suffolk County Council, reproduced with permission.

Archaeology of the Wic Sites or Emporia

101

Figure 3.5 Image of Ipswich Ware pottery jar. Copyright: Suffolk County Council Archaeological Service, reproduced with permission.

While the presence of imported pottery indicates the 7th-century settlement at Ipswich was engaged in cross-channel trade, this early settlement was set within a rural agrarian landscape (Scull 2009, 314). Evidence for urban planning and denser settlement does not appear until around 700 CE. Ipswich expanded rapidly in the 8th century, and this expansion may have begun as early as 680–700 CE. By the 8th century, the town had grown to about 50 ha, expanding northward into previously unoccupied heathland. At its height in the Middle Saxon period, Ipswich was home to about 2000–3000 people (Hamerow 2007, 223). A new grid pattern of streets was laid out at this time. It was centered on the marketplace and included a probable royal chapel dedicated to St. Mildred. In the town center, houses were built up against the edges of the streets, and craft activities were carried out in the back yards. Craft activities included widespread bone- and antler-working, spinning and weaving, and metal-working. Evidence for leather-working has been recovered from waterlogged deposits. However, the most striking feature of Middle Saxon Ipswich is the large-scale production of Ipswich Ware pottery (Figure 3.5).

102

Middle Saxon Settlement and the Rise of the Emporia

While earlier Anglo-Saxon pottery was hand-built and fired in bonfires, Ipswich Ware pottery was coil-built, finished on a slow wheel, and kilnfired. It is the first industrially produced pottery in Britain since the end of the Roman period. Three kilns for the production of Ipswich Ware have been identified. Two were recovered from the Cox Lane area of Ipswich, and a third was discovered during excavations at the Buttermarket site (Blinkhorn 1989). The Buttermarket kiln operated during the latter part of the Middle Saxon period from about 800–850 CE, but there is some limited evidence for other earlier kilns in the same area (Blinkhorn 1989, 13). As noted above, recent research (Blinkhorn 1999, 2012) has indicated that Ipswich Ware was first produced around 720 CE, substantially later than the traditional date of 650. Paul Blinkhorn (2012) has carried out a detailed study of Ipswich Ware. His research shows that in East Anglia and the adjacent areas of eastern Cambridgeshire and possibly northern Essex, Ipswich Ware was the primary form of pottery used on Middle Saxon sites, and handbuilt pottery is rare. Blinkhorn termed this area the Primary Zone. The Primary Zone corresponds to the areas of East Anglian hegemony. Outside the Primary Zone, Ipswich Ware pottery is found at the wic sites, at royal vills and other lordly settlements, at ecclesiastical settlements, and at sites with evidence of specialized agricultural or craft production (Blinkhorn 2012, 99). These sites are generally located near route ways including rivers and Roman roads. A  map of the distribution of Ipswich Ware is shown in Figure 3.6. One of the striking features of this distribution is the paucity of finds of Ipswich Ware in Wessex. Blinkhorn (2012, 87)  notes that there are few East Anglian coins in Wessex, suggesting little economic contact between the two regions in the 8th century. Most of the Ipswich pottery that appears outside the Primary Zone consists of large jars. When residues are present on these pots, most are animal fats. It is unlikely that preserved meat and dairy products would have been shipped from Ipswich to the rural areas of eastern England, so these residues are probably the result of reusing these pots as cooking vessels. Blinkhorn (2012, 98) has suggested that these vessels may have been used initially to transport salt. Salt was a crucial commodity for the preservation of meat, fish, and dairy products. In addition, salt was used in industrial processes, including dyeing and tanning. Blinkhorn (2012, 99)  suggested that the Middle Saxon “elite, secular and ecclesiastical, may have controlled this trade and profited greatly from it in the form of tolls, taxations and food-rents.” Rather than seeing the wic sites as engaged primarily in the controlled exchange of foreign luxury items (Hodges 1982), Blinkhorn’s model shows the importance of both

Archaeology of the Wic Sites or Emporia

103

Figure 3.6. Map of England showing locations of Ipswich Ware finds. Redrawn with permission of Paul Blinkhorn, after Blinkhorn (2012, p. 70, Fig. 36).

intensified rural production and regional trade in the emergence of the wic sites (see also Moreland 2000b; Crabtree 2010b). Archaeological evidence for trade between Ipswich and the European continent includes hone stones from Norway and lava querns and pottery from the Rhineland, Flanders, and northern France (Scull 2009, 318).

104

Middle Saxon Settlement and the Rise of the Emporia

Much of this imported pottery was used by the foreign traders who were residing in Ipswich, but some may have made its way into AngloSaxon households. Wooden barrels provide evidence for trade in wine. Some of these barrels were reused as linings for shallow wells, and an example was recovered from Brook Street in Ipswich. Goods that moved outward from Ipswich probably included woven cloth and wool (Wade n.d.). Certainly spinning and weaving are among the crafts documented at Middle Saxon Ipswich (Wade n.d.), and zooarchaeological data suggest a shift toward specialized wool production at rural sites and estate centers that may have begun as early as the late 6th/early 7th century (Crabtree 2010b). By the mid-8th century, the enlarged emporium of Ipswich was linked to the surrounding countryside by markets and a substantial element of monetary exchange (Scull 2013, 237). Coins began to be minted in East Anglia sometime in the 7th century (Williams 2013, 136). By the late 7th and early 8th centuries, large numbers of sceattas (silver pennies) were being minted in East Anglia, and coinage from other parts of England and the continent was in circulation as well (Naismith 2013, 144–5). The widespread use of coinage would have facilitated marketing by reducing transaction costs and would also have helped to link the surrounding rural areas of Suffolk to the markets in Ipswich. The edges of Ipswich were less densely settled than the town center. At the Foundation Street site near the eastern margin of the town, there are fewer buildings, and the houses are set back from the street (Wade n.d.). This part of the town provides more evidence for agricultural activities such as cereal processing. The question of how much of Ipswich’s food was raised in and around the town in the Middle Saxon period is not an easy one to answer. The faunal assemblages recovered from 11 Middle Saxon sites in Ipswich indicate that the diet was dominated by beef, followed by pork, with relatively small amounts of lamb and mutton (Crabtree 2012, 2016). These sites also yielded sizeable numbers of chicken and goose bones, but wild bird and mammal bones are quite rare (Crabtree 2012, 21–3). The age profiles for the cattle indicate that most of these animals were market-age and older adult animals, including oxen that may have been slaughtered after their working lives were over. Following on the pioneering work of Jennifer Bourdillon at Hamwic (Middle Saxon Southampton), I initially thought that Ipswich may have been provisioned with food renders (see below and Bourdillon 1988; see also Hamerow 2007; Crabtree and Stevens 1994). Food renders are taxes in kind paid to the king. The corollary of this is that the residents

Archaeology of the Wic Sites or Emporia

105

of Ipswich and other emporia may have been prevented from trading directly with food producers (Hamerow 2007, 221). The argument for this was based on the presence of large numbers of older animals at the wic sites. These older animals may have been relatively tough and unpalatable. Recent data from Ipswich, however, challenge this model. As will be discussed below, Ipswich survived as a regional market town into the later Middle Ages and modern times. Comparison of the Middle Saxon faunal assemblage from Ipswich with the Late Saxon and early medieval (11th–12th  century) assemblages show no appreciable changes through time (Crabtree 2016, Crabtree under review).2 Historical and archaeological evidence indicates that markets were providing food in the later periods, and it is likely that they were operating in the 8th and 9th centuries as well. When combined with the evidence for productive sites, the data from Ipswich suggest that markets and fairs were increasingly common in the Middle Saxon period. As noted in the introductory chapter, markets are one of the important features of medieval urbanism. Ipswich continued to serve as a center for craft production and exchange during the Late Saxon period. The town was initially fortified during the 10th century, probably under the Danish occupation. By the Late Saxon period, the economy of Ipswich was focused on regional rather than international trade. The town continued as a regional market town into the early medieval period (Wade 1988, 1993, 2000). The Origins of Ipswich Project (1974–90) led to the excavation of 36 different archaeological sites, including 27 sites within the Anglo-Saxon and medieval fortifications. Since 1991 all excavations in Ipswich have been developer-funded, although relatively little excavation has been carried out within the town over the past quarter century. The one exception was a planned redevelopment of the docks, but this project was postponed as a result of the financial crisis of 2008. Unfortunately, the excavations at Ipswich have not been fully published, and complete publication is not likely since the Suffolk Archaeological Unit was recently broken up. Specialist reports are available for the pottery (Blinkhorn 2012) and the faunal remains (Crabtree 2012, 2016), and a publication on the worked bone and antler is also in preparation (Riddler in prep.). However, the data that are available allow us to compare Ipswich to the other, better published wic sites.

2

For a comprehensive and up-to-date review of the animal bone remains from AngloSaxon sites see Holmes (2014).

106

Middle Saxon Settlement and the Rise of the Emporia

Hamwic (Middle Saxon Southampton) History of the Excavations Hamwic, also known as Hamtun, is located on the west bank of the River Itchen, just north of the point where the rivers Test and Itchen converge and spill into the Solent (Figure 3.7). Both of these rivers were navigable; the bend in the river near Hamwic provided a measure of protection for ships. In addition, boats could be beached on the Itchen flats, reducing the need for an extensive system of wharves and docks. Like Ipswich, Hamwic was a de novo settlement. The Roman settlement at Bitterne Manor was located a mile (1.5 km) northeast of Hamwic on a promontory on the east side of the River Itchen. It is often identified as the site of Clausentum, which appears in the Antonine Itinerary,3 but the identification is not entirely secure (Morton 1992, 2). The site was a small Roman town in the 1st and 2nd centuries CE. A wall was erected around the site in the later 3rd century. There is no evidence to suggest that there is any relationship between the Roman site at Bitterne and the later foundation of the Anglo-Saxon site of Hamwic. As we will see below, the main Anglo-Saxon occupation of Hamwic begins in the later 7th century CE and ends around 900. By about 1000 CE most of Hamwic had reverted to fields, which were used for lowintensity farming activities such as grazing (Brisbane 1994). The Hamwic area was resettled about 1830, and small antiquarian excavations carried out between 1825 and 1945 uncovered about 2–3 percent of the Middle Saxon site (Brisbane 1994, 28).4 Digging for brick earth in the 19th and 20th centuries damaged much of the Anglo-Saxon material in central Hamwic, but the lower levels of Anglo-Saxon pits and wells survived the brick earth collection. Traces of timber buildings survive only in those areas where digging for brick earth did not take place. Since Southampton was a major British port, parts of the city were heavily damaged during World War II, making some previously inaccessible areas available for excavation. In the late 1960s, Peter Addyman and David Hill (1968, 1969) summarized what was then known about the archaeology of Southampton. They argued that the only way to learn more would be through large-scale, open-area excavation within the Middle Saxon town.

3

4

The Antonine Itinerary is a list of Roman roads and stations that was probably recorded in the 3rd century CE. Morton (1992) provides a useful history of the excavations that have taken place in Saxon Southampton.

Archaeology of the Wic Sites or Emporia

107

CUT-THORN CROSS

PORT SW OO D

ST RE ET

COMMON LAND

ST. DENYS’ PRIORY

HILL LANE

Roman CLAUSENTUM

REE

S ST

ARY

ST M

PADWELL CROSS

T

MILLBR OOK LANE

EAST MARLANDS

Saxon HAMWIC

ch

en

ve

r

It

WEST MARLANDS

Ri

HOUNDWELL

HOGLANDS

MO

ST. MARYS CHURCH

DE

RN

SH

OR

EL

INE Medieval SOUTHAMPTON

0

300 m

Figure 3.7. Map of Southampton showing location of the Hamwic in relation to Roman and medieval Southampton. Reproduced with permission of Professor Mark Brisbane.

108

Middle Saxon Settlement and the Rise of the Emporia

In 1971, a program of excavation began on a series of sites in Melbourne Street (Holdsworth 1980) under the auspices of the Southampton Archaeological Research Committee. The sites were all disturbed by brick earth digging or by later housing developments and their utility trenches. The 1971–6 excavations uncovered Middle Saxon pits, wells, and postholes that represented the remains of structures. The excavations also yielded a sizable and well-preserved faunal collection that provided evidence for Middle Saxon diet and clear evidence for bone- and antler-working. The subsequent large-scale, open-area excavations at the Six Dials site were carried out under the aegis of the Southampton Archaeological Research Committee and later the Southampton City Museum (Andrews 1997; see also Hamerow 2002). The excavations were carried out between 1977 and 1986. Since this area was not heavily damaged by brick earth digging, it provided evidence for Middle Saxon houses and the street plan and layout of the town during the 8th and 9th centuries CE. I was an archaeology student at the University of Southampton while the excavations were taking place, and I was able to visit the Six Dials site while excavations were in progress. Between 1946 and 1990, about 4 percent of Hamwic was excavated using modern methods (Brisbane 1994, 28). While the large, open-area excavations at Melbourne Street and Six Dials provided important information on the nature of the Middle Saxon settlement at Hamwic, its origins remained relatively unknown until about 20 years ago. In 1998 excavations began in advance of the construction of a new football stadium in the St. Mary’s area of Southampton (ancient Hamwic). These excavations revealed a high-status cemetery dated to the second half of the 7th century and the first quarter of the 8th century. This high-status cemetery has shed important light on the initial founding of Hamwic (Stoodley 2002; Birbeck 2005). The Archaeological Evidence The early history of Hamwic has been a subject of debate for decades, but the recent excavations at St. Mary’s Stadium may shed new light on the debate. The excavations were developer-funded and only explored the areas that would be directly affected by the construction of the football stands. The areas that would serve as the football pitch and parking lots remain unexplored (Figure  3.8). Excavations revealed 24 inhumations and 18 cremation burials. An additional two inhumations that were almost certainly part of this cemetery were discovered as part of the Melbourne Street project. Many of the graves were in poor condition, having been disturbed by later 8th- and 9th-century settlement

Archaeology of the Wic Sites or Emporia

109

Figure  3.8. Plan of the excavations at St. Mary’s Stadium, Hamwic, Southampton. Copyright: Wessex Archaeology, reproduced with permission.

110

Middle Saxon Settlement and the Rise of the Emporia

features (Stoodley 2002, 321). A  high proportion of the graves were accompanied by grave goods, including 12 individuals (46  percent of the inhumations) who were accompanied by weapons. This certainly indicates that the people buried at the stadium were high-status individuals. Stoodley (2002, 326) has suggested that the high proportion of weapons burials may indicate that Hamwic began as a “royal foundation controlled by royal representatives, i.e., an estate.” Birbeck (2005, 190) has carried this idea further, suggesting that the 7th century saw a rise in agricultural surpluses, allowing inhabitants of the Hamwic region to participate in trade and marketing. He suggested that royal agents may have regulated trade in the 7th century from a royal estate, and that a royal estate in Hamwic would be well positioned to regulate trade in farming surpluses. Birbeck (2005, 194)  further suggested that the royal estate may have developed into a center of trade and manufacture in the 8th century. By that time, the role of the royal agents would have been to tax trade rather than to control it (Birbeck 2005, 190). Twenty-five years earlier Holdsworth (1980, 1)  raised the question of whether the establishment of Hamwic was a response to the increasing political and economic security brought about during the reign of King Ine of Wessex (688–726) or whether the town was established by direct royal prerogative. The data from the St. Mary’s Stadium lend some support to the second hypothesis. Additional support may be found in the use of both Hamwic and Hamtun to describe the settlement. While Hamwic describes a trading function, Hamtun implies that the site may also have had an administrative role (Stoodley 2002, 317). This model, however, presents several problems. The first, and perhaps the most critical, is that the excavations in Hamwic have recovered no evidence for 7th-century settlement whatsoever. Ipswich, in contrast, has produced evidence for pre-urban settlement in addition to the 7th-century Buttermarket cemetery. At present, the evidence for a royal estate at Hamwic in the 7th century remains conjectural. However, it is consistent with the more widespread evidence for agricultural surpluses and trade in the 7th century. While Hodges (1982) argued that AngloSaxon kings established the emporia to control trade in imported highstatus materials, archaeological and zooarchaeological evidence points to the intensification of agriculture and local trade before the establishment of the emporia (see, for example, Crabtree 2010b). As a large-scale, open-area excavation in an area where there was no brick earth quarrying, the Six Dials excavations revealed the remains of over 60 buildings, plus fence lines and back yards. Most of the buildings were 4–5 m wide and up to 10 m long, and some may have served as combined homes and workshops (Brisbane 1994, 31). There is almost

Archaeology of the Wic Sites or Emporia

111

no evidence for occupation before these houses were constructed around 700 CE. The extensive Six Dials excavations also revealed details of the layout of the wic. A  boundary ditch was dug around 700 CE. The ditch was not defensive in nature, but instead seems to have been designed to separate the town from the surrounding countryside, suggesting established property rights. In terms of the street plan, the major north–south streets in Hamwic were laid out first, and the east–west streets were laid out a few years later to form a nearly rectilinear pattern. The streets were paved with uniformly sized gravel (Brisbane 1994, 31). Since the Six Dials area is at the opposite end of the town from the river landings, the town plan appears to have been conceived as a unified entity from the start (Bourdillon 1988, 179). At its height, Hamwic was probably home to between 2000 and 3000 people (Andrews 1997, 253). The excavations at Hamwic yielded evidence for a wide variety of crafts. Metal-working activities included gold-, copper-alloy-, and ironworking. A wide range of crafts were based on primary and secondary animal products,5 including bone- and antler-working (Driver 1984), leather-working, butchery, and wool processing and textile production. Artifactual evidence for wool processing includes spindle whorls, bone needles, wool combs, and thread pickers (Brisbane 1994, 32). Evidence for glass-working and wood-working was also recovered from Six Dials. Unlike later Anglo-Saxon and medieval towns, some of the craft activities were not spatially segregated (Brisbane 1994, 31). However, Riddler (2001) has argued that bone- and antler-working was carried out in three specific areas of Hamwic. Two were on the periphery of the town, and the third was located in a possible foreign enclave. The imported pottery from Hamwic indicates trade with several continental ports during the 8th and 9th centuries. Most of the imported pottery comes from Carolingian Francia, with smaller quantities of pottery from the Rhineland also present. Hodges (1980) suggested that most of these pottery vessels were used by foreign traders who were present in Hamwic. Coins also provide evidence for international trade, including coins minted in the Low Countries, Frisia, and Denmark (Andrews 1997, 254). In addition to the long-distance trade, much of the trade was probably local as well (see Moreland 2000). However, there is no evidence for a clear marketplace where goods were bought, sold, and exchanged (Andrews 1997, 255). 5

Primary products are those that are obtained from a dead animal such as meat, hide, and bone. Secondary products are those that can be taken from the animal while it is still alive, such as milk and wool.

112

Middle Saxon Settlement and the Rise of the Emporia

In terms of regional trade, the large numbers of craft activities based on animal products point to trade and exchanges with farms in the surrounding river valleys. The rough and ready character of the butchery at Hamwic, described by Bourdillon (1988, 180) as “bold and slapdash chopping,” suggests that the town lacked the specialized butchers that are documented from Roman and later medieval towns (Seetah 2006, 2007), as well as from Middle Anglo-Saxon London. In addition the finds of sceattas (Anglo-Saxon silver coins) are concentrated primarily within the town of Hamwic, indicating that the coins were used primarily for marketing within the town rather than throughout the Wessex countryside. Some of the most important evidence for Hamwic’s economy comes from the study of the enormous collections of animal bone remains recovered from the Melbourne Street and Six Dials sites. Prior to the 1970s, studies of animal bone remains tended to be more zoological than archaeological or historical. For example, Clutton-Brock’s (1976) review of the animal bone evidence from Anglo-Saxon archaeological sites includes only five sites, and two of them are not even Anglo-Saxon. Older zooarchaeological studies often ignore archaeological context entirely. For example, the study of the enormous faunal assemblage from the Iron Age oppidum of Manching in Germany examined the bones in terms of the years in which they were excavated rather than the levels or features they were recovered from (Boessneck et  al. 1971; see also Crabtree 2018). New zooarchaeological methods6 developed in the 1970s allowed zooarchaeologists to focus more closely on archaeological questions such as urban provisioning and the relationships between town and countryside. The zooarchaeological work at Hamwic is one of the best examples of the new zooarchaeology. Jennifer Bourdillon and the late Jennie Coy examined about 75,000 animal bone fragments that were recovered from five Melbourne Street sites (Bourdillon and Coy 1980), and Bourdillon (1988, 1994) used these data plus the animal bone evidence from Six Dials to develop a model of the way in which Hamwic was provisioned with meat and other animal products such as wool, traction, hides, milk, and horns. At the contemporary continental emporium of Dorestad in the Netherlands, farm buildings were identified within the town itself (Prummel 1983). This is not the case at Hamwic, whose residents must have procured their meat from farms located outside the town. The near 6

These include methods for determining the ages of animals when they died (Silver 1969; Grant 1976, 1982; Payne 1993) and standardized measurements (von den Driesch 1976) that can be used to determine animal sizes (Von den Driesch and Boessneck 1974).

Archaeology of the Wic Sites or Emporia

113

absence of very young animals in the Hamwic pits and other features point to a real separation between the animal producers in the countryside and the meat consumers in the town. Animal bone remains from Hamwic indicate that the Middle Saxon diet was based on domestic animals, primarily cattle and sheep, plus some pigs, and chickens and geese. A  few goat bones were also identified, but most of these were male horn cores that may have provided the raw materials for horn-working. Wild animals and birds were quite rare. While fine sieving at the Six Dials site produced somewhat larger numbers of fish bones, most of these were locally available estuarine fish (Bourdillon 1994, 122). Most pigs, which are raised for primary products such as meat and fat, were killed shortly after they reached maturity, but about half the cattle and sheep bones came from older individuals with well-worn permanent teeth. These animals would have yielded very tough meat, requiring extensive stewing to make it palatable. The age and sex profiles for the sheep from the Melbourne Street sites suggest that these animals were drawn from flocks of wethers (castrated males) that were raised primarily for their wool. Wool production peaks at about five to seven years in sheep (O’Connor 2010, 12), and these animals may have been sent to market soon after that. Similarly, many of the cattle were older oxen that may have pulled carts and plows before ending up in the stew pot. One of the cattle bones from the Melbourne Street sites (Bourdillon and Coy 1980, 94) appears to show traction pathology (see Bartosiewicz et al. 1997). However, the vast majority of the bones recovered from the Hamwic sites appear to have come from large, sturdy animals. One of the most remarkable features of the faunal assemblages from Hamwic is their uniformity. There are no clear differences in diet across the settlement, and only limited evidence for changes through time. The uniformity in diet is paralleled by the lack of variation in the housing at Hamwic. It stands in sharp contrast to the historical evidence, such as King Ine’s 7th-century law code, which points to significant differences in social status and political power in Middle Saxon Wessex. Bourdillon (1988, 1994) has crafted an elegant model to explain how the residents of Hamwic may have obtained their meat. She notes that the faunal remains from Hamwic reflect a low diversity in food resources and animal age structure, and she suggests that the population of the town was supported by a controlled food supply. In other words, there was little direct contact between producers and consumers. Bourdillon suggests that the inhabitants of Hamwic were provisioned with food renders which were paid to the king. These renders were normally paid to a royal vill, and, as noted above, historic sources point to

114

Middle Saxon Settlement and the Rise of the Emporia

the possible presence of a royal estate at Hamwic, although there is as yet no archaeological evidence to support this contention. Certainly, the absence of farms within the town of Hamwic provides indirect support for provisioning based on food renders, but the new data on marketing in Middle Saxon Ipswich shows that not all the wics were provisioned in that way. Unlike Ipswich, Hamwic ceased functioning as a town sometime between 850 and 900 CE (Andrews 1997, 255). While Morton (1992) argued that Hamwic was abandoned in the late 9th century, Andrews suggests that a few elements of dispersed settlement may have survived after 900, although Six Dials was abandoned by that time (Andrews 1997, 255). Viking raids and disputes among Charlemagne’s heirs have been suggested as possible causes for Hamwic’s decline (Hodges 1982, 156–7), but there is no evidence that Viking raids led to Hamwic’s demise. Winchester re-emerges as an important town center about 900 CE, and it is possible that Winchester took over Hamwic’s functions, including that of a royal vill (Yorke 1984, 66). Even without the Vikings, it is likely that Hamwic’s indefensible position along the Itchen played a role in its demise. As we will see in Chapter 4, Winchester was redeveloped within the former Roman city walls in the late 9th or early 10th century. One of the most important questions we need to ask is whether we can consider Hamwic a town. Some scholars have used the term “protourban” to describe Hamwic and the other emporium sites. In many ways this is a problematic term since it implies an almost teleological process of urban development. As we have seen, Ipswich did develop as a market town for southeastern Suffolk, while Hamwic had largely disappeared by 900 CE. Hamwic is distinct from contemporary rural settlements in terms of its size. At its height, it was probably home to 2000–3000 residents. I  would argue that the size of Hamwic and Ipswich, when combined with the evidence for craft specialization, planned street layouts, urban plots, and evidence for both local and international trade, qualifies these sites as urban. Andrews (1997, 255) has come to a similar conclusion, suggesting that these sites should still be referred to as towns, even though they lack some of the attributes that we see in later medieval urban centers. Eorforwic (Anglian York) History of the Excavations York is a picturesque city with a long and important history as a Roman colonia, an early medieval trading center, a Viking town, and an important

Archaeology of the Wic Sites or Emporia

115

city in later medieval and modern English history.7 Like many European towns in the second half of the 20th century, York’s rich archaeological heritage was threatened by modern urban redevelopment. In 1969, an amateur archaeological society, the York Excavation Group, was formed to help preserve York’s archaeological resources. The group did some important work at York Minster, but it was not equipped to address the challenges presented by modern urban renewal. In 1972, the York Archaeological Trust was founded as an independent charity which sought to preserve York’s past for future generations. The Trust sponsored a number of large and small excavations within the city, beginning in 1972. The original director of the trust was P. V. Addyman, a British archaeologist who was educated at Cambridge and who had substantial previous experience in medieval archaeology. Addyman served as the director from 1972 until 2002. Initially, the York Archaeological Trust was modeled on the Winchester Excavations Committee that had sponsored in 1961–71 excavations at Winchester (see Chapter  4). However, the York Archaeological Trust recognized that one of the problems facing the Winchester group was its slow rate of publication. Instead of publishing large site report volumes, the York Archaeological Trust decided to publish each site as a series of fascicles, so that in many cases the archaeology, artifactual studies, and environmental evidence for a single site appear in different small volumes. This is the case for the Fishergate site that provided much of what we know about Eorforwic. The York Archaeological Trust is best known for its sponsorship of the Coppergate excavations (1976–81) which produced about 40,000 Viking period artifacts, leading to the opening of the Yorvik Viking Centre in 1984. The Trust subsequently sponsored the Fishergate excavations (1985–6) which began to fill in the archaeological evidence for York between the Roman and the Viking periods. A recent developed-funded project on the campus of the University of York has added on 6th- and 7th-century settlement in the area (Spall and Toop 2008). The York Archaeological Trust has also been committed to public outreach. In 1990 the Trust opened an Archaeological Resource Centre (ARC) committed to public education about archaeology (Jones 1990). I was fortunate to visit ARC shortly after it opened in 1990, and I was impressed with its commitment to the public. In 2006 ARC was transformed into DIG, which provides hands-on learning for school children about archaeology. 7

The excavators at York have chosen to use the term Anglian, rather than Middle Saxon or Middle Anglo-Saxon, for the pre-Viking settlement at York.

116

Middle Saxon Settlement and the Rise of the Emporia

Archaeology at Anglian York York, located at the confluence of the Ouse and Foss rivers, was an important military and civilian site during the Roman period in Britain. The site was attractive to the Romans because of its position on a major river system with a varied surrounding countryside and because of its defensibility (Palliser 2014, 5–6). The Roman 9th legion built a fortress along the River Ouse in 71 CE. The fortress covered 26 ha and was initially surrounded by a ditch, earthen ramparts, and gates, towers, and a palisade made of timber (Figure 3.9). In the early 3rd century CE the timber fortifications were replaced by walls, gates, and towers made of stone, including an elaborate polygonal tower known as the Multangular

Sycamore Terrace

Porta decumana R

LA TE

R

R

A

VIA

E

RIA TO AE PR

Ou

e

Pr ae to n rium

PR

u r

Pr inc ipia

a

IN

T

CIP AL

P

IS

N

T

U

R

VIA

er

t

t

E

Riv

e

A

Anglian Tower P

VIA

IN

DE

TE

CU

M

RV AL

AN

A

LU

M

Porta principalis dextra dextra

se

R

A

E

TO

R

II

A

Ba

th

s

Porta principalis principalis sinistra sinistra

Porta praetoria

Baths

? Forum Baths

Riv

os er F

s

16–22 Coppergate

Baths

Sites discussed in the text Mosaic finds Roads and streets, known and conjectured Cemeteries Principal settled areas 0 0

100 100

200 200

300 metres 300 yards

Contains Ordnance Survey data © Crown copyright and database right 2010

Figure 3.9. Map of Roman York. Copyright: Historic Towns Trust, reproduced with permission.

Archaeology of the Wic Sites or Emporia

117

Figure 3.10. Photograph of the Multangular tower, York. Photo credit: Pam J. Crabtree.

tower on the southwest corner of the fort, which survives to this day (Figure 3.10). Civilian settlement outside the fortress began in the late 1st century CE. Although the initial civilian settlement may have had the status of a vicus or canaba,8 the settlement had achieved the status of a colonia by 237 CE. In the 3rd and 4th centuries York also served as the capital of Britannia Inferior and then Britannia Secunda.9 It was also the seat of a bishopric in the late 4th century (Ottaway 1999, 147). Until recently, little was known about York in the 5th and 6th centuries CE (Tweddle et al. 1999, 207). There are no historical records for York between the early 4th and the early 7th centuries CE (Spall and Toop 2008, 2). While mid- to late 5th-century Anglo-Saxon cemeteries have been identified in the York region, there is only limited occupation within the former Roman city, and no clear evidence for continuity 8

9

A vicus is a small town while a canaba is a settlement near a Roman fort that housed military dependents and civilian contractors. In the 3rd century under Severus, Britannia was divided into two provinces – Britannia Superior and Britannia Inferior. In the 4th century under Constantine, it was divided into several provinces.

118

Middle Saxon Settlement and the Rise of the Emporia

of urban settlement (Spall and Toop 2008, 18). There is some limited activity around the principia (the center of the Roman fort) in the 5th century, and a contemporary timber building was erected in the colonia (Palliser 2014, 17). However, the final use of the principia was as a small farm based on pig-rearing (Palliser 2014, 29). Environmental evidence from York in the form of dark earth indicates dramatically reduced occupation and soil that has been reworked by agriculture (Palliser 2014, 18–19). As noted above, recent rescue excavations on the University of York campus have begun to fill in some of the gaps. Archaeological monitoring at Heslington Hill on the north campus of York University revealed two intact archaeological deposits that yielded pottery, small finds, and animal bones. Glass beads from the Netherlands provide some evidence for trade (Spall and Toop 2008, 12). The deposits appear to be refuse associated with a settlement dated between 550 and 650 CE. The settlement is located on a glacial moraine, on land that may have been part of a Roman villa estate. The settlement was abandoned in the 7th century as part of the “Middle Saxon Shift” (Spall and Toop 2008, 18), at the same time that the new Middle Anglo-Saxon settlement of Eorforwic, best known from the excavations at the Fishergate site, was established just to the northeast of the confluence of the Ouse and the Foss and a mile (1.5 km) to the west of Heslington Hill (O’Connor 1991; Maiman 1993; Rogers 1993; Kemp 1996). The excavations at Fishergate were carried out by the York Archaeological Trust between 1985 and 1986. The map (Figure  3.11) shows that the site is located to the south and well outside of the old Roman fortress. The excavations at Fishergate revealed a settlement that was established in the very late 7th or early 8th century, and occupation continued until the 860s or 870s. Based on archaeological, historical, and place-name evidence, Kemp (1996, 64–6) identified the Fishergate site as a wic. The emporium is estimated to have covered between 26 and 65 ha and may have been home to between 1000 and 1500 people (Hamerow 2007, 223). The Middle Anglo-Saxon features uncovered included boundary ditches, postholes, stake holes, and pits which were interpreted as the remains of a planned settlement that included rectangular post-built structures, property boundaries, and a road. The pits were rich in artifactual material including evidence for a variety of craft activities  – iron- and other metal-working, the manufacture of bone and antler combs, glass-working, textile production, wood-working, and fur preparation (Kemp 1991, 211; Spall and Toop 2008, 13; Palliser 2014, 37). Other artifacts recovered included personal items such as dress pins

Archaeology of the Wic Sites or Emporia

FISHERGATE TOWER

CITY WALL

TOW E R S T R E E T

ILL DM L EA

LANE

REET GEORGE ST

PICCADILLY

CASTLE

119

E BAR

FISHERGAT

FAWCET STREET

ATE

ERG FISH

RIVER FOSS

FEWELL WAR

HOTEL

R

VE

RI SE

OU GE LANE

BLUE BRID

0 0

50 100

200

100 300

150 Metres 400

500 Feet

Figure  3.11. Map of York showing location of Eorforwic and the Fishergate excavations. Reproduced with permission of the York Archaeological Trust.

120

Middle Saxon Settlement and the Rise of the Emporia

and brooches. The artifactual evidence for craft production is complemented by evidence for trade in the form of coins, glass vessels, and lava querns from the Rhineland and imported pottery from France and the Low Countries (Kemp 1991, 213–16). The animal bone remains from the Fishergate site were analyzed by O’Connor (1991). The large Middle Anglo-Saxon faunal assemblage was dominated by the remains of cattle and sheep, with smaller numbers of pigs, supplemented by chickens, geese, and fish. Hunted mammals and birds were very rare. The kill patterns for cattle and sheep show a wide age distribution, including young juvenile through adult animals. The kill pattern for pigs is more focused, with an emphasis on younger pigs that were just over a year old and older pigs that were just over two years old when they were slaughtered (O’Connor 1991, 248–51). O’Connor (1991, 282; see also O’Connor 2001) concludes that it is likely that Fishergate was provisioned indirectly from regional food rents. However, his data are not a particularly good fit for the model of indirect provisioning where consumers have no contact with producers. In most cases of indirect provisioning, the consumers are supplied with a limited number of species from a limited range of age classes (see, for example, Zeder 1988, 1991). While the range of species at Middle Anglo-Saxon York is certainly limited, the residents consumed cattle and sheep from quite a wide range of age classes. These data could be equally consistent with an early market economy, especially since the Fishergate data are quite similar to the later 9th-century data from Late Anglo-Saxon Coppergate site (O’Connor 1991, 278). O’Connor (2001, 60) notes that the 10th–11th-century Coppergate material shows an increasing diversity in bird remains, as well as increased numbers of birds, fish, and pigs, and he attributes this change to increased consumer choice and direct procurement of meat by consumers. We will return to this issue when we discuss the Coppergate material in Chapter 4. Excavations carried out at the Bridge Lane site in 2000–1 revealed additional Middle Anglo-Saxon material from part of the same settlement as the Fishergate site. Blue Bridge Lane is located just south of Fishergate, and excavations there revealed a series of aligned pits, providing additional evidence for property demarcations (Spalland Toop 2008, 13). The pits were originally cess pits, but were subsequently used for waste disposal. Spall and Toop (2008, 13) note that the use of these pits for disposal of midden waste points to the spatial constraints that are typical of an increasingly urban environment. Similar practices were seen in New York City during its period of rapid growth in the 19th century (see, for example, Yamin 2001). The contents of these pits included both personal items and the remains of craft production. The principal crafts

Archaeology of the Wic Sites or Emporia

121

represented are bone-, antler-, and horn-working and textile production, crafts that are based on primary and secondary animal products. Interestingly, the animal bone remains also include a higher proportion of sheep than was seen at Fishergate (Spall and Toop 2008, 15). However, like Fishergate, the bone assemblage was dominated by the remains of domestic animals. These data point to increasing links between the wic and the surrounding countryside. The prevailing interpretation of Middle Anglo-Saxon Fishergate is that it was a wic site that was part of a polyfocal settlement that included a possible royal center, an ecclesiastical center that was located within the former Roman fortress, and a trade and craft production zone (Palliser 2014, 36–7). As Tweddle (1999, 212) suggests, “In neither the 8th nor the 9th centuries, however, was York a densely settled urban area. Instead, scattered among the still impressive remains of the Roman city, were various foci of settlement, serving different purposes:  royal, ecclesiastical, manufacturing and trading.” Palliser (2014, 38) attributes York’s rebirth to the reintroduction of Christianity to Britain and to the Mediterranean package of values that came along with it. He also sees the wic sites as the apex of a hierarchy of productive sites, an issue that we will return to at the end of this chapter (Palliser 2014, 37). There are some problems with the polyfocal model. While it is clear that a bishopric was established in York in the 7th century, the historical and archaeological evidence for a royal center is more limited (see, for example, Rollason 2003, 178). Instead of viewing the Fishergate settlement as a de novo settlement established as a royally controlled center, Spall and Toop (2008, 20) see the establishment of the wic as part of the broader social and economic changes that have their origins in the economic intensification and social changes that are seen in the 7th century and that were described in the first section of this chapter. Spall and Toop (2008, 21) also suggest that changing attitudes toward Romanitas and the ancient Roman world may have attracted people back to York. While the Early Anglo-Saxons may have consciously avoided some Roman sites and rejected aspects of Roman culture, Hunter (1974), in a seminal paper, suggested that by the Middle and Late Anglo-Saxon periods, the inhabitants of Britain often conflated Germanic and Roman traditions to create a single imagined historical past. Lundenwic History of the Excavations As we saw in Chapter 1, Roman London was shrinking in size during the later Roman period, and there is almost no evidence for 5th- and

122

Middle Saxon Settlement and the Rise of the Emporia

6th-century occupation within the walls of the old Roman city (Cowie and Blackmore 2008, 127). Although Bede makes it clear that a major market town for all nations existed in London in the Middle Anglo-Saxon period, the location of Lundenwic remained unknown until the 1980s. In 1984, Martin Biddle (1984) and Alan Vince (1984, see also Vince 1990, 13–25), working independently, identified the location of Lundenwic by plotting the location of known finds of Middle Anglo-Saxon pottery, coins, and other artifacts and combining these data with place-name and topographic evidence. The distribution centered on an area about 1.2 miles (2 km) long and 0.3 miles (0.5 km) wide that was located about half a mile (1 km) upstream from the old walled Roman city. The area was on the line of an old Roman road which is now followed by Fleet Street and the Strand (Ottaway 1996, 121). The identification of the location of Middle Anglo-Saxon London is one of the most important discoveries in modern medieval archaeology, and most of the archaeological work in the area has been carried out by archaeologists associated with the Museum of London. The Department of Urban Archaeology (DUA) was created in the 1970s, in part as a response to the publication of The Future of London’s Past (Biddle et al. 1973). In 1991 the DUA joined with the Museum’s Department of Greater London Archaeology (DGLA) to form the Museum of London Archaeological Service (MoLAS), now known as the Museum of London Archaeology (MoLA). MoLAS was set up as an independent archaeological contractor in response to changing antiquities legislation. Today MoLA is an independent chartable company conducting archaeological research in London and other parts of the UK. While Biddle’s and Vince’s research suggested a possible location for Lundenwic, it was the DGLA’s excavations at Jubilee Hall in 1985 and Maiden Lane in the following year that provided the first substantial archaeological evidence for Middle Anglo-Saxon London (Cowie et al. 1988). The excavations at Jubilee Hall revealed structures, wells, pits, and a single inhumation, while those at Maiden Lane uncovered ditches, pits, and a midden. These excavations yielded evidence for craft activities, including metal-working, bone-working, and textile manufacture, as well as for long distance and regional trade. Kentish hone stones, quern fragments, and a coin minted in Kent provide evidence for regional trade. Regional trade also supplied raw materials such as wool, metal, and antler. International trade provided Rhenish quern stones and pottery from France, the Low Countries, and the Rhineland. The presence of continental tablewares may also provide indirect evidence for the importation of wine (Cowie et al. 1988, 81). The evidence for craft production and regional and long-distance trade

Archaeology of the Wic Sites or Emporia

123

is similar to the kinds of materials that were recovered from Ipswich and Hamwic. While many sites in London have yielded evidence for Middle AngloSaxon settlement since the mid-1980s, some of the most significant work was carried out in 1996 when a 2500-square-meter open-area excavation was carried out at the site of the Royal Opera House, representing about 0.4  percent of the estimated area of Lundenwic (Malcolm and Bowsher 2003, 2). Additional developer-funded excavations were carried out in the area of St. Martin’s Courtyard between 2006 and 2009. These excavations shed light on the northwest corner of Middle Anglo-Saxon London (Fowler and Taylor 2013). The data from these excavations, when combined with the evidence from other smaller Middle AngloSaxon sites, provide a detailed picture of urban life in 7th–9th-century London (see Cowie et al. 2012 for an overview). Anglo-Saxon Lundenwic The excavations carried out at the Royal Opera House in London at the end of the last century revealed important new information on the layout of the Middle Anglo-Saxon town, the craft activities that were carried out there, and the ways in which the residents of Lundenwic were provisioned with food and raw materials. A small number of late 6th- to 7th-century finds suggest that a small and possibly seasonal settlement developed by about 600 CE and that this site developed into the port of Lundenwic (Malcolm and Bowsher 2003, 2). The Thames river provided access to continental ports, and the riverfront along the Strand would have allowed traders to beach their boats and conduct trade there (Milne and Goodburn 1990, 629; Malcolm and Bowsher 2003, 187). The Royal Opera House is located to the north of the Strand and provided evidence for the community of craft producers who resided in the town and may have supplied goods for trade. Occupation of the Royal Opera House site began in the 7th century (Figure 3.12). A  main road, Road 1, was constructed through the settlement sometime in the 7th century, probably postdating the earliest buildings on the site. The road runs in a north–south direction, and, if extended northward, would intersect the main east–west Roman road (modern New Oxford Street), which was still in use in Middle AngloSaxon times (Malcolm and Bowsher 2003, 145). Alleyways connect the buildings and plots to the road. While the main road was made of layers of compacted gravel and regularly maintained, the composition of the alleys was more varied and generally included more sand. The main road was regularly cleaned, but rubbish built up on the alleyways that were sometimes used for market stalls or shops.

124

Middle Saxon Settlement and the Rise of the Emporia

N ew

rd Oxfo

t Stree

t

iles St G

tree hS Hig

Lu

nd

Dr ur

yL an

en wi

c

e t tree et S Fle

ych Aldw

1

2

arti Upper St M

3

a Str

nd

loo ter Wa

n’s Lane

s ame

r Th Rive

Br e idg

Figure  3.12. Map of London showing location of the Royal Opera House (2), Julibee Hall (3), and St. Martin’s courtyard (1)  sites. Redrawn after Fowler and Taylor (2013, Fig.  7) with permission of Museum of London Archaeology.

Over 60 Middle Anglo-Saxon buildings were discovered during the Royal Opera House excavations. The typical house was constructed of posts that supported a timber frame, but sometimes the posts were placed on a sill beam that was either set on the ground surface or set in a trench. The walls were made of wattle and daub – wickerwork made of wooden rods 1–2 cm in diameter and covered with mud plaster (Malcolm and Bowsher 2003, 150–2). The typical house was about 12 m long and about 5.5 m wide with a pitched roof and gable ends. Doorways were often positioned in the middle of the long side of the house, although other variations exist (Figure 3.13). The Royal Opera House excavations yielded evidence for a range of crafts and activities that were carried out in the area. Evidence for textile

Archaeology of the Wic Sites or Emporia

125

Figure 3.13. Reconstruction of Middle Saxon Lundenwic. Copyright: Museum of London Archaeology, reproduced with permission.

production includes loom weights, spindle whorls, and bone thread pickers and pin beaters used to separate the threads on a loom. Textile production increased in the 8th century at the time when Ipswich Ware pottery reached the site, and Blackmore (1999) has suggested that this activity may have been stimulated by Frisian traders who appear to be the most active foreign traders in the pre-Viking period. Other crafts and industries that are represented include bone- and antler-working, iron smithing, non-ferrous metal-working including copper-alloy-, gold-, silver-, and lead-working, and tanning. There is no clear zoning and clustering of single crafts and trades. Instead, the organization of these trades was based on a model of “cooperative diversity” (Malcolm and Bowsher 2003, xv), where related trades were sometimes located close to one another so that, for example, iron-working for the production of knives was located next to bone-working for handles (Malcolm and Bowsher 2003, 196). Artifacts from the Royal Opera House excavations have provided substantial evidence for both regional and international trade. Imported items include lava querns from Carolingia and pottery. Approximately 12 percent of the pottery from the site was produced on the European continent (Malcolm and Bowsher 2003, 187), including pottery from

126

Middle Saxon Settlement and the Rise of the Emporia

northern France and Belgium, the Meuse Valley, and the Rhineland (Blackmore 2003). The presence of Ipswich Ware also points to the importance of regional trade. In the case of both the imported pottery and the Ipswich Ware, it is not clear whether the pots themselves or their contents were more important. Oysters were also imported from the East Anglian coast (Winder and Gerber-Parfitt 2003). Excavations at rural settlements in the London region have provided some evidence for how the emporium may have been supplied with food (Cowie and Blackmore 2008). For example, the Middle Saxon Site U in Whitehall, Middlesex appears to be a processing and distribution center for beef and mutton. The faunal assemblage from the site includes very few sub-adult animals, and the assemblage is dominated by primary butchery waste. Cowie and Blackmore (2008, 162) suggest that this settlement is a farm and redistribution center serving the needs of Lundenwic. They further suggest that during the Early Anglo-Saxon period there is a shift in rural agricultural production from producing goods for local consumption to producing a surplus to support the royal palaces and Lundenwic (Cowie and Blackmore 2008, 158). Population sizes are notoriously difficult to estimate for archaeological sites. However, based on the currently available evidence, Lundenwic is likely to have been home to a larger population than either Hamwic or Ipswich. During the mid-8th century, London may have been home to between 8500 and 13,700 people, making it three or four times larger than the other two sites (Malcolm and Bowsher 2003, 193). Cowie and Blackmore (2012, 116)  subsequently argued for a smaller population of about 6000–7000 people. The conditions in which these early Londoners lived were often less than hygienic. Water wells were located in close proximity to cess pits, and the residents may have also been exposed to toxic substances used in tanning and metallurgy. The Royal Opera House area appears to have experienced a decline in both population size and settlement density in the 9th century. The recovery of locks and keys points to a concern for security, and trash disposal appears to have been poorly regulated. It was at this time that a large defensive ditch was excavated across the northern portion of the site, possibly as a result of Viking attacks. The Vikings took the city in 871 CE, and Lundenwic was abandoned in favor of the City of London area to the east. Recent excavations at St. Martin’s Courtyard have helped to identify the northwest corner of Ludenwic (Fowler and Taylor 2013). Two linear features may mark the western extent of the settlement. The earlier one is dated to the late 7th or earlier 8th century. The second ditch marks the westward expansion of the site in the 8th century. While there are

Archaeology of the Wic Sites or Emporia

127

hints of earlier occupation, extensive settlement in this area of London begins in the late 7th or early 8th century CE. The material remains provide evidence for iron smithing, non-ferrous metal-working, the stabling of animals, and possibly the processing of crops. These are the kind of activities that one might expect to find on the outskirts of town. The site was covered with a layer of rubbish and industrial waste dating to ca. 770–850 CE. However, the site may not have been completely abandoned at this time, since two buildings appear to have been erected in the area after 770 CE, and the excavations uncovered a male burial that has been radiocarbon dated to 720–940 CE (Nicholls 2013, 97). The area appears to have experienced a revival in the late 8th through mid-9th century before its final abandonment. The site is significant for its excellent organic preservation, including the remains of a number of wooden artifacts and leather shoes. The emporia are the largest sites in a network of towns and trading places that range from small beaching sites along both sides of the North Sea to periodic fairs and markets at productive sites and estate centers to Lundenwic (see Loveluck and Tys 2006). While Ipswich, London, York, and Hamwic have received the most attention from archaeologists and historians, a number of smaller sites may also have served as wics or trading centers. Two possible wic sites that deserve further attention are Fordwich (Hill 2001) and Sandwich in Kent (Holman 2001). Fordwich was part of the Early Anglo-Saxon royal estate at Stury. Fordwich stands on the opposite side of the Great Stour River from Stury. Its location would have provided deeper water and a more convenient location than Stury, which is located at an awkward bend in the river. Fordwich first appears in the historical records in the late 7th century. Charters from 747 and 761 describe tolls on ships visiting Forduuic, documenting its status as a port in the 8th century (Kent Historic Towns Survey 2004a, 5). The Anglo-Saxon settlement may have stood on higher ground than the modern town. Some features are visible in aerial photographs, but almost no archaeological research has been undertaken at the site. Fordwich became the principal port for Canterbury after 1075. Sandwich appears to have been founded on the south bank of the River Stour during Anglo-Saxon times. The Life of St. Wilfrid indicates that the saint arrived at “a port of safety in Sandwich” in 666 CE (Kent Historic Towns Survey 2004b). However, there is no Middle or Late Anglo-Saxon archaeological evidence recovered from within the modern town. It is therefore likely that the Anglo-Saxon settlement is not in the same place as the medieval town. Given the environmental conditions in the Anglo-Saxon period, the site is likely to have been on high ground.

128

Middle Saxon Settlement and the Rise of the Emporia

Other Possible Emporia Hill et al. (2001, 95–103) provide a list of sites that previous authors had identified as possible emporia. While many have turned up appropriate finds, none is the size of the four large wics described in detail in this book. In addition to Fordwich and Sandwich in Kent, some of the other possible wic sites include Dover, Santun, and Sarre in Kent, Norwich in Norfolk, and Dunwich in Suffolk. Much more archaeological research is needed to determine the dating and character of these smaller sites, although the existing evidence for both Norwich and Dunwich is unconvincing (Cowie 2001, 15). Sandtun in Kent, however, appears to be a smaller site in a hierarchy of Middle Anglo-Saxon trading places. Initial excavations were carried out at Sandtun in the 1940s. This work was reassessed and new rescue and research excavations were carried out at the site between 1993 and 1998 (Gardiner et al. 2001). The site is located in a sandy, coastal area in the northeast corner of the Romney March. Sandtun appears to have been occupied between about 700 and 850–75 CE. The ceramic assemblage includes a very high percentage of continental pottery, 29 percent by weight (Gardiner et al. 2001, 271). The faunal assemblage from the site includes large numbers of wetland and coastal wild birds, as well as over 4000 bones of marine and estuarine fish. Craft activities include bone-working and the manufacture of spindle whorls. The site itself appears to be much smaller and less densely occupied than Hamwic (Gardiner et  al. 2001, 270), but the presence of craft activities indicates that it is more than a temporary camp. It appears to be a coastal site that is involved in fishing and also both continental and coastal trade. Gardiner et al. (2001, 276–7) suggest that trade may have taken place at a number of different ports in Kent, but it may have been administered through a smaller number of centers like Canterbury. Conclusions Regarding the Wic Sites To what extent should the major wic sites – Ipswich, Hamwic, Eorforwic, and Lundenwic  – be treated as a group of similar sites, and what do these sites tell us about the rebirth of urbanism in Anglo-Saxon England? To answer these questions, the origins of these sites, their layout and functions, and their abandonment or continuation after the 9th century will be compared. All four sites appear to have been established in the 7th century CE. Although Anglo-Saxon York and London are relatively close to their

Archaeology of the Wic Sites or Emporia

129

Roman predecessors, none of these sites is sitting on a Roman foundation. The sites are de novo settlements of the early Middle Anglo-Saxon period. Both Hamwic and Ipswich were preceded by high-status 7thcentury cemetery sites, at St. Mary’s Stadium and Buttermarket, respectively. While Hamwic may have been established on or near a royal vill, the archaeological evidence for this is limited at best. There is no evidence for a royal vill at Ipswich; the royal settlement is located at Rendlesham near Sutton Hoo. Although Middle Anglo-Saxon York has often been seen as a polyfocal central place, including Eorforwic as a center of trade and manufacture, an ecclesiastical center within the old Roman fort, and a royal center, clear evidence for a 7th–9th-century royal center within Middle Anglo-Saxon York is still lacking. There is no clear archaeological evidence for a royal settlement in or near Lundenwic, but the bishop and his church were located just inside the walls of the otherwise largely deserted old Roman city. The absence of a clear royal settlement within these wics does not mean that the Anglo-Saxon kings were not actively involved in these trading towns. It is no surprise that these wics are located within the territories of four of the most powerful kingdoms of Middle Anglo-Saxon England. Ipswich is located in the southeastern part of East Anglia; York is located in Northumbria; Hamwic is located in the southeastern part of Wessex; and Mercia, the most powerful of the 8th-century kingdoms, controlled London from the 660s through the 8th century, with a possible gap of less than 15 years in the late 7th century (Maddicott 2005). London’s location along the great artery of the Thames provided it with a much larger hinterland than the other wics had (Maddicott 2005, 14). While Hodges (1982) initially argued that the role of the wics was to control trade, especially in imported luxury goods, research that has been conducted since the 1980s has shown that trade, including trade with the European continent, was widespread in the later 7th and 8th centuries. Instead, the wic sites provided an opportunity for these Anglo-Saxon kings to concentrate and tax trade. Not only were the tolls an important source of revenue, but the ability to remit tolls was a source of power for the king. This can be seen clearly in Æthelbald of Mercia’s (r. 716– 57) seven toll-remitting charters (Maddicott 2005, 25), mostly issued to powerful religious houses. The Middle Anglo-Saxon period was an era of intense competition between the developing Anglo-Saxon kingdoms. Recent archaeological studies support the hypothesis that “Intense interstate competition tends to coincide with high levels of tax extraction in the form of taxation of goods or taxation of (military) labor, or both” (Monson and Scheidel 2015, 20). The ability to tax goods at the port of London, when combined with control over the salt works at Droitwich

130

Middle Saxon Settlement and the Rise of the Emporia

Figure 3.14. Image of a sword hilt from the Staffordshire Hoard. Copyright: Birmingham City Museums, reproduced with permission.

(Maddicott 2005), would have given the Mercian polity a distinct advantage in its competition with the other Anglo-Saxon polities. The recent discovery of the Staffordshire hoard, the largest collection of Anglo-Saxon gold- and silver-work ever found, should be understood in the context of Middle Anglo-Saxon warfare and competition. The hoard was discovered near Litchfield in 2009, and it includes 3500 items, almost entirely war gear, including over 5 kg of gold and 1.4 kg of silver (Figure 3.14). The find was recovered in what was then Mercian territory, and it has been dated to the 7th/8th centuries CE. It may represent a collection of war trophies (Leahy and Bland 2009, 11–12). The plans of the wic sites themselves show some important similarities and differences. Despite the difficulties of estimating population sizes, it is clear that London is by far the largest of the four sites. Hamwic and Ipswich seem to be broadly similar in size, and York is clearly the smallest of the four, home to only about 1000–1500 people (Hamerow 2007, 223). Some of this may be a product of location. London clearly had the largest hinterland, and the Thames estuary would have provided easy access to the continental ports. Hamwic would have had easy access to northern Francia, while Ipswich was ideally suited for trade with the Low

Archaeology of the Wic Sites or Emporia

131

Countries. York’s inland, riverine location would have been less favorable to long-distance trade with the continent. Both Lundenwic and Hamwic had street plans that were laid out sometime around 700 CE. The houses in both towns were clearly laid out in relation to the streets. The open-area excavations at the Royal Opera House in London and Six Dials in Hamwic made it possible to reconstruct the urban plans for these two towns. Our data for Ipswich are poorer, since most of the areas excavated as part of the Origins of Ipswich were much smaller in scale, but there is evidence for a grid plan that was laid out around 700 CE and centered on the market. Anglian York appears to have been much less densely occupied than the other three wic sites. All four sites provide evidence for significant trade activity. Imported materials, such as continental pottery and lava quern stones, are easy to recognize, but the widespread distribution of Ipswich Ware in greater East Anglia underlines the importance of the emporia as centers of regional trade as well. The presence of sceattas, silver coins minted in Britain and on the continent, would have facilitated trade by reducing transaction costs. In a recent review of 8th- and 9th-century coinage in southern England, Naismith (2012) has argued that these coins were used in markets, rather than as gift exchange (contra Hodges 1982). Naismith (2012) further suggests that the Anglo-Saxon economy was partially monetized in the 8th and 9th centuries CE. It is important to note that many of the goods and services that were traded at the wic sites are likely to be archaeologically invisible. Maddicott (2005, 10)  notes that Bede described a Northumbrian noble named Imma who was given to a Mercia war leader after the Battle of Trent and was subsequently sold to a Frisian slaver in London. Other invisible items include agricultural products such as wool. Zooarchaeological studies carried out in both Anglo-Saxon England and Carolingian Francia point to increasing wool production in the 8th and 9th centuries (Crabtree 2010b). All four emporia have yielded substantial evidence for craft production, including such crafts as bone- and antler-working, metalsmithing, and textile production. Most of these industries are relatively small scale, and in London related crafts tended to be located close together, although there is no clear segregation of individual crafts as was common in later medieval cities. Many of these crafts are based on primary and secondary animal products, including wool, horn, bone, and leather, pointing to the interrelationship between these early emporia and the countryside. Not only did the surrounding farms provide food for craft workers and other non-agricultural laborers, they also provided many of the raw

132

Middle Saxon Settlement and the Rise of the Emporia

materials that formed the bases of these crafts. The emporia emerge in the 7th and 8th centuries at a time of increasing agricultural productivity in southern and eastern England. Estate centers, both ecclesiastical and secular, served as centers of “consumption, exchange, agrarian improvement, technological innovation and trade” (Maddicott 2005, 14). For example, a complete iron coulter10 was found at the base of SFB 1 at Lyminge (Thomas 2013, 123). This 7th-century find is the earliest known evidence for the heavy plow in post-Roman Britain. Heavy plows allow farmers to cultivate the heavy but fertile clay soils that were often avoided by the Early Anglo-Saxons. As Moreland (2000b) has eloquently argued, we should see the development of the emporia as a consequence of intensified rural production rather than as the engine that drove rural economic changes. While there is evidence for a number of smaller-scale craft activities at Ipswich, including bone- and antler-working and metal-working, the production of Ipswich Ware pottery was carried out on an industrial scale not seen in Britain since the end of the Roman Empire. As noted above, the Ipswich Ware pottery that appears outside the primary zone of East Anglia may have been used to transport salt, another invisible but vital commodity. Given the archaeological evidence for trade, craft production, and urban planning, should we see the emporium sites as towns? My answer is an unqualified yes. First, these sites show a level of settlement density and economic diversity that is not seen in even the largest high-status rural settlements. While the term “proto-urban” has sometimes been used to describe these settlements, it is not clear why the term “proto” is necessary or warranted. The emporia are not the same as the towns of late medieval Europe, but they are not like Roman towns either. Protourban is poorly defined, but it has been used to refer to very early towns, e.g., sites like 7th-millennium Çatal Höyük in Anatolia, or to refer to settlements that share both urban and rural characteristics. While we can identify some possible farm sites on the edges of some of the wics, neither use of the term proto-urban seems particularly appropriate here. I  would prefer to refer to the wic sites as early towns or early urban centers. They qualify as towns based on their size, planned layouts, and evidence for trade, craft specialization, and markets. In a recent keynote address, the medieval historian Chris Dyer (2015) provided a social and economic definition of a medieval town as a permanent settlement with 10

A coulter is part of a heavy plow. The coulter cuts an edge ahead of the plowshare. The moldboard then turns over the soil to form a ridge.

Archaeology of the Wic Sites or Emporia

133

non-agricultural occupations (specifically crafts and trade), a market, and a population of more than 300. Using these criteria, the wics definitely qualify as towns. The emporia, however, are very different kinds of towns from the late medieval towns that are more familiar to us from historical records. They lack the charters and legal autonomy that characterized many late medieval urban centers. They are not judicial centers either, and many lack the complex religious organizations that are a common feature of later medieval population (cf. Biddle 1976, 100). Perhaps the biggest difference is that some of these emporia appear to be agricultural towns or what the German scholars refer to an Ackerburgerstädt. While many of the residents of the wics are clearly engaged in trade and craft activities, others, particularly those who lived in the peripheral areas of some of the wics, do appear to be engaged in animal husbandry and agriculture. The Peabody/National Gallery site on the edges of Lundenwic may well be a farm (West 1989; Whytehead et al. 1989), and some of the buildings on the outskirts of Ipswich may have served agricultural functions. The presence of some farmers who are resident within early medieval towns is not unique to the English wic sites; the continental emporium of Dorestad also appears to have been home to three large farms and their fenced compounds from the late 7th through the later 8th centuries (van Es 1980a, b). These farms were subsequently replaced by seven houses, most of which were occupied by craft producers (Hodges 2012).11 During the earliest phases of medieval Antwerp (radiocarbon dated to the 8th–10th centuries CE) animals were also stabled within the town (Crabtree et al. 2017). The archaeological evidence for sites like Sandtun show that the picture of continental trade being conducted through a small number of sites like Hamwic and Lundenwic is far too simple (cf. Hodges 1982). We are certainly looking at a hierarchy of trading places of varying scales that were active during the Middle Anglo-Saxon period. As Gardiner et  al. (2001, 278)  conclude, “The discovery and examination of the site of Sandtun throws into relief the paucity of our understanding about settlement and trade during the mid-Saxon period.” While I think that our knowledge of Middle Saxon settlement and trade has advanced substantially since the late 1990s, there still is much to learn. 11

See also www.world-archaeology.com/world/europe/holland/richard-hodges-travels-todorestad-netherlands-3.htm

134

Middle Saxon Settlement and the Rise of the Emporia

The End of the Emporia What is the legacy of the Middle Saxon emporia? Were they a failed experiment in early medieval urbanism, or did they set the stage for the development of towns in the Late Anglo-Saxon period? A discussion of this topic needs to begin with Ipswich, since it was transformed from a wic to a market town in the Late Anglo-Saxon and medieval periods. Ipswich is located in one of the parts of Britain that came under Viking control in the late 9th century. Viking raids are recorded in Britain beginning in the late 8th century. The raids were sporadic until the 840s, and many concentrated on monasteries, which were often wealthy and unfortified. In the 860s, the Vikings began to assemble larger armies with the goal of conquering English territory. The Viking army arrived in East Anglia in 865, and Ipswich was under Danish control from about 880 until 920 CE. During this period, Ipswich was fortified with a town wall for the first time. The construction of the wall closed and diverted some streets. Sunken-featured structures were constructed within the town for the first time. They also appear for the first time in Viking York (see Chapter 4). There is a marked increase in craft activities during the Early Late Saxon Period in Ipswich. Archaeological excavations have revealed evidence for the smelting and smithing of iron as well as copper-alloy metallurgy. Molds recovered through excavation provide evidence for the production of brooches. The production of Thetford Ware replaces Ipswich Ware at this time, but the pottery kilns remain in the northeast part of the town. In short, craft activities continue in Ipswich under Viking control. Ipswich was reconquered by Wessex in 920, and the story of Ipswich in the later parts of the Late Anglo-Saxon period will be continued in Chapter 4. Hamwic’s history is very different. The town was largely abandoned sometime between 850 and 900 CE, although a few dispersed settlements may have survived into the early 10th century. As noted above, Hodges (1982) suggested that Viking raids may have played a role in Hamwic’s demise, although we have no clear archaeological or historical evidence to support this contention. Nevertheless, security may have played a role in the abandonment of Hamwic, whether the threats were from Viking raiders or elsewhere. The site of Hamwic is low-lying and indefensible. Winchester, just 10 miles (16 km) to the north, was established as a bishopric in the 7th century, and Winchester may have taken over from Hamwic as a center for trade and craft production in the period around 900. As we will see in Chapter 4, the town of Winchester was replanned in the late 9th or early 10th century CE.

The End of the Emporia

135

Lundenwic and Eorforwic represent intermediate cases, since urbanism continued in both London and York in the Late Anglo-Saxon period, but the locations of the major urban centers shifted in the late 9th or early 10th centuries. The dating of the final phases of Lundenwic has been challenging. The Royal Opera House site clearly reveals evidence for urban decline between 770 and 850 CE (Malcolm and Bowsher 2003, 122). Lundenwic was probably no longer a large and nucleated settlement by the early 9th century (Cowie and Blackmore 2012, 209), and it was abandoned sometime in the mid-9th century (Cowie and Blackmore 2012, 111). The abandonment of the Royal Opera House site is marked by extensive dark earth deposits. Shortly thereafter, a new urban settlement was established within the walls of the old Roman town. Whether this shift in settlement was the result of Viking raids or the series of fires that devastated Lundenwic, the result was to shift the Anglo-Saxon town about half a mile (1 km) to the east. Similarly, in York, the focus of settlement shifted away from Fishergate and toward the upstream Coppergate area at the end of the 9th century. The question of whether these wics should be seen as failed settlements and products of the social, economic, and political circumstances of the long 8th century was addressed by Cowie and Blackmore (2012, 209) for Lundenwic. They argue that: [Lundenwic’s] development in the late 7th century marked London’s rebirth as an urban settlement. From this point on, and despite the troubles and dislocations of the 9th century, London’s role as a central place and important point of contact for travellers and traders from the Continent has never really ceased.

The role of historical contingency in the expansion and contraction of the wics also needs to be emphasized. The wic sites expanded in the later 7th and 8th centuries at a time of increasing agricultural productivity and prosperity. It is not surprising that they contracted in the 9th century at a time of increasing Viking raids and the eventual occupation of many parts of eastern England by the Viking army, as well as the break-up of the Carolingian Empire. Under these turbulent circumstances, towns based on craft production and trade with little in the way of defenses would not have been particularly attractive places to reside. The settlement shift from Lundenwic to the London burh, the old Roman town, is entirely understandable under these circumstances. We will explore this shift in the following chapter. The archaeology of the wic sites shows that the rebirth of towns in post-Roman Britain had its ups and downs as a result of specific regional and historical circumstances. Finally, the wic sites can contribute to our understanding of social change in Middle Anglo-Saxon England. A  close examination of the

136

Middle Saxon Settlement and the Rise of the Emporia

archaeological evidence from both urban and rural sites in Middle Anglo-Saxon England shows that there is no clear correlation between the means of trade and exchange and the socio-political evolution. In his original formulation, Hodges (1989, Table 6) following Polanyi (1957) saw complex chiefdoms as characterized by systems of redistributive exchange, such as food rents, where goods flow toward the chief and are then redistributed to his followers through mechanisms such as feasting. States, on the other hand, are supposed to be characterized by market exchange. The wics and the contemporary rural sites provide evidence for multiple systems of governance and exchange operating at the same time. Feasting was clearly taking place at high-status sites such as Flixborough, while coastal, “productive,” and wic sites provide evidence for marketing at different levels of intensity. Rural agricultural resources, such as livestock pastures, may have been managed collectively by farmers based on practices that can be traced back to pre-Anglo-Saxon times (Oosthuizen 2011). As Hodges (2012, 138) noted, simple processual models of socioeconomic change can mask the diversity that is apparent in the archaeological and historical records. Other Possible Middle Anglo-Saxon Towns: 8th-Century Mercian Burhs While the wic sites are the most well-known of the large Middle AngloSaxon settlements, Haslam (1987) has suggested that fortified places may also have played a role in the development of Middle Anglo-Saxon urbanism. A  burh is a fortified place. The Old English term refers to a fortification or fortified settlement. A  10th-century document, now known as the Burghal Hidage, lists 30 burhs in Wessex and 31 in Mercia. In 1987, Haslam (1987) suggested that there was a relationship between the construction of fortresses or burhs and the development of “urban” markets in Mercia. In the 8th century, Mercia emerged as the dominant Anglo-Saxon polity in Britain. Haslam suggested that King Offa, who ruled Mercia from 757 to 796, developed a series of fortified places that also developed as market centers. Citing the historian Nicholas Brooks (1971, 72), Haslam (1987, 77) noted that the 8th–9th-century military obligations to the Anglo-Saxon kings included “army-service, bridge-work and fortress-work” and that the fortress and bridge worked together as a single unit to secure the river crossing. These defensive measures took on great importance in the late 8th and 9th centuries due to the increasing threat from the Vikings. Haslam (1987) identified a series of possible Mercian burhs. Some were located within older Roman defenses, while others were newly

8th-Century Mercian Burhs

137

fortified in the late 9th century. These fortifications are associated with important river crossings, so that they could prevent land armies from crossing the river and ships from sailing up the river. Haslam (1987, 87– 8) further suggested that the defended area within the fortress was only a small part of the settlement system. The focus of the settlement would have been on a large extramural market, while the fortified area itself would have served ecclesiastical functions as homes to minster churches, administrative functions as homes to the king’s reeve who collected taxes and tribute, and places of refuge. The problem with this model, as Haslam (1987, 88)  notes, is that “It is not possible from the archaeological evidence alone to determine either the nature or the date of the extramural activity in these places.” While the possibility of markets associated with these burhs is an intriguing one, a market alone does not make a town. Haslam (1987) attempted to link these developments to a wider interest in the relationship between the control of trade, urbanism, and state formation as set forward by Hodges (1982) in Dark Age Economics. Lacking clear evidence for Middle Anglo-Saxon settlement and marketing in the extramural areas, the development of this system of burhs seems more closely related to state formation in Middle Anglo-Saxon England than to the origins of urbanism. However, the fortified bridge and burh combinations may have laid the basis for later medieval urbanism at these sites. We will revisit the question of the burhs in Chapter  4, since fortified places may have served as the foundations for towns beginning in the late 9th century CE.

4

Towns in Late Anglo-Saxon England

Background and Introduction The 9th century was marked by a number of disruptive political changes, both in England and on the European continent. The unity of the Carolingian Empire did not last long past Charlemagne’s death. Charlemagne was succeeded by his son, Louis the Pious, in 814, but by the 830s the empire was wracked by civil war. The wars ended with the Treaty of Verdun in 843 that divided the empire into three parts. One part was to become France, a second was to become Germany, and the third was a region in between that stretched from northern Italy to the Low Countries. As noted in Chapter  3, the nature of Viking involvement in Britain also changed in the mid-9th century. While the Vikings engaged in raids on soft targets, primarily but not exclusively monasteries, until about 840, by the middle of the century, the Danes were assembling larger armies with the aim of conquering Britain. The Great Army arrived in East Anglia in 865 and captured York in 866. The Vikings eventually conquered East Anglia and killed its king in 870 and subsequently conquered eastern Mercia in 877, leaving Wessex the only Anglo-Saxon kingdom that was not under Viking control. The Great Army attempted to invade Wessex but was defeated by King Alfred the Great at the Battle of Eddington in 878. Reynolds (2009, 77) has described the later 9th and 10th centuries as “some of the most turbulent periods in English history.” King Alfred was succeeded by his son, Edward the Elder (r. 899–924), and his grandson, Æthelstan (r. 924–39). Edward reconquered southern England from the Danes and incorporated Mercia into his kingdom. Aethelstan defeated the Viking army that invaded Northumbria in 937, but the final Viking ruler in York was not expelled until 954. This expulsion led to a generation of peace under King Edgar (r. 959–75), but this was followed by a second Viking age. The late 10th and early 11th

138

Background and Introduction

139

centuries were characterized by Viking raids and extortion, including rapidly increasing payments of Danegeld by the English. The Danes succeeded in placing a Scandinavian on the English throne in 1016, and King Cnut reigned until 1042. Cnut was succeeded by the English king Edward the Confessor, who reigned until 1066. His death without issue led to a battle for succession that ultimately resulted in the Norman Conquest under William. This political instability was accompanied by changes in AngloSaxon society, and specifically in Anglo-Saxon kingship. As Reynolds (2009) points out, by the late 9th, 10th, and 11th centuries the AngloSaxon kings had developed substantial economic, military, and judicial authority. The Late Saxon kings maintained the network of royal palaces that had initially been established during the Middle Anglo-Saxon period. These palaces were part of the peripatetic pattern that allowed the king and his retinue to move through the kingdom on a regular basis and make use of the food rent or feorm. Archaeologically, the best known of the Late Saxon palaces is Cheddar in Somerset (Rahtz 1962–3; Rahtz et al. 1979). The principal feature of the palace was a large, bow-shaped hall, initially constructed some time before 930. The hall, with the addition of a masonry chapel at one end, was rebuilt twice in the Late Saxon period. Anglo-Saxon kings also had the ability to raise large armies, to build fortifications based on a system of military obligations, and to enforce a complex judicial organization (Reynolds 2009, 69). Clear evidence for the judicial power of the Anglo-Saxon kings is seen in the execution places, such as the one that was discovered at Sutton Hoo by Carver (2005). It is against this backdrop of political, economic, and social change that we will examine the development of later Anglo-Saxon urbanism. I will begin by examining the archaeological evidence for the development of urbanism in the Late Saxon period before discussing the models that have been proposed to explain this development. As we noted in Chapter 3, many of the wic sites seem to have declined in the 9th century, while other new towns rose to take their place. I will begin with the study of Winchester. Winchester has been a focus of intensive open-area and rescue excavation since the 1960s, and its Late Saxon layout has been taken as a possible model for other towns in Wessex. My next case study will be London, since we see a clear shift in population from the older wic back to the old Roman town in the late 9th century. I will then take a look at Viking York (Yorvik), which has been seen as a model for Anglo-Scandinavian urbanism in northern England and elsewhere. Finally, I will examine some of the smaller market towns of the Late Anglo-Saxon period, beginning with Ipswich, which transitioned

0

100

200

300

400 km

Figure  4.1. Map showing the location of the sites discussed in this chapter. Drawn by Douglas V. Campana using open-source software. Map numbers: London 1, Winchester 2, Wareham 3, Cricklade 4, Wallingford 5, Exeter 6, Chichester 7, Bath 8, Worchester 9, Yorvik 10, Viking Dublin 11, Ipswich 12, Norwich 13, Thetford 14

Late Anglo-Saxon Winchester

141

from a wic to a regional market town in the late 9th and 10th centuries (Figure 4.1). Late Anglo-Saxon Winchester and a Model for Planned Towns in Wessex As noted at the end of Chapter  3, Haslam (1987) suggested that a number of Anglo-Saxon towns in Mercia might have been based on burhs or fortified places. Haslam’s model draws on an earlier model first proposed by Biddle and Hill (1971). Using data from the longterm program of excavation at Winchester in Hampshire, Biddle and Hill suggested that a number of fortified places in Wessex were planned or replanned no later than the mid-10th century and that these sites developed as both defensive and commercial centers in the Late AngloSaxon period. As noted in Chapter 1, a long-term program of research and rescue excavation was undertaken in Winchester between 1961 and 1971 under the direction of the Winchester Excavation Committee and Martin Biddle (Biddle 1990). The excavations centered around four large-scale projects: the excavation of the old and new minster churches which predate the construction of the Norman cathedral, the excavation of the medieval bishop’s palace of Wolvsey, the excavation of the north bailey of the royal castle at Castle Yard, and the excavation of 12 Anglo-Saxon and medieval houses along the west side of Lower Brook Street, which was known as Tanner Street in the Middle Ages. The choice to focus on the castle, the bishop’s palace, and the pre-Norman churches reflects a mid-20th-century approach to medieval archaeology. The interest in these areas grew out of history, art history, and architectural history, and these disciplines tend to focus on the activities of the political, economic, and social elite classes. The Brook Street excavations, on the other hand, provided a much fuller picture of the day-to-day lives of the inhabitants of Late Anglo-Saxon Winchester. In addition to the four major excavation projects, a number of smaller-scale excavations were carried out in and around the town. Unfortunately, the final publication of a number of these sites, including the Lower Brook Street site, has not been completed, and researchers must rely on the yearly interim reports on the excavations that were published in Antiquaries Journal over 40 years ago. At the time that theWinchester excavations began, many archaeologists and lay people thought that Winchester was unique among British towns because it was supposed to have retained its Roman street plan.

142

Towns in Late Anglo-Saxon England

However, excavations soon showed that the Roman grid plan was very different from the modern one (Figure  4.2). Winchester’s medieval street plan had four characteristic elements (Biddle and Hill 1971). The first is a main east–west High Street. The line of this street is close to, but not exactly the same as, the line of the Roman street. Biddle and Hill (1971, 73) argue that this Roman east–west street was not entirely abandoned at the end of the Roman period and that it may have survived as a thoroughfare into the post-Roman period. Historical sources indicate that the High Street was in existence in its current form by about 900 CE (Biddle and Hill 1971, 74). The second element of the plan was two back streets that ran parallel to the High Street on both the north (St. Georges Street/Silver Hill) and south (St. Clement Street). The third element of the plan was a series of north–south streets that ran perpendicular to the High Street. One of these was Lower Brook Street, which was the site of a major open-area excavation between 1965 and 1971. The fourth element was a set of intramural streets that ran along the inside of the old Roman walls. These include North Walls on the north side of town and St. Swithin Street on the south. Charters show that the High Street, back streets, and north–south streets are all present by the late 10th century, but that the intramural streets do not appear in the documents until later. Subsequent archaeological evidence indicates that Winchester was replanned between 880 and 886 CE (Biddle 1975, 310), during the reign of King Alfred the Great. These Late Saxon streets formed a rectilinear grid plan, but one that was very different from the earlier Roman grid (Figure  4.2, see also Figure  1.4). Additional Anglo-Saxon streets were discovered underneath the Norman castle. These streets show evidence for extended use and repair before they were covered by the castle in 1067 (Biddle and Hill 1971, 78). Thus, the historical and archaeological data indicate that Winchester’s street layout was extensively replanned in the late 9th century CE. As noted above, large-scale, open-area excavations were carried out at the Brook Street site from 1965 to 1971, but it was only in the final season that the 8th–9th-century occupation of the site was reached.1 During the medieval period, the Brook Street area was home to two of the 57 medieval parish churches that are known to have been built in Winchester by the late 13th century, many of which were Anglo-Saxon 1

As I noted in the preface, I was lucky enough to be a student volunteer on the Brook Street site in the summer of 1971, and the experience led me to apply to graduate school to study medieval archaeology.

Late Anglo-Saxon Winchester

143

Roman streets

e

R

Itc

h

n

Anglo-Saxon streets

.

Site of later castle New Minster Royal Palace? Nunnaminster

Old Minster

0 0

100 100

200 200

250 metres 250 yards

Figure  4.2. Street plan of Winchester showing the Roman and Late Saxon streets. Copyright: Winchester Excavation Committee, produced by Giles Darkes, reproduced with permission.

in date (Barlow et al. 1976, 459). St. Pancras was located on the west side of the site, while St. Mary’s developed out of a 9th-century nonecclesiastical building located in the southern part of the excavated area. The latest Roman occupation underneath the St. Pancras church was followed by a long period of farming activity when the area appears to have been plowed. The first church was built over the last cultivated surface. The first post-Roman occupation of the Brook Street site is a small 7th-century cemetery, dating to the conversion period, i.e.,

144

Towns in Late Anglo-Saxon England

the period of the 7th and 8th centuries when there is evidence for a large-scale conversion from Paganism to Christianity.2 This cemetery may not have been associated with a church, since regular burial in churchyards only became common in the 9th and 10th centuries CE (see Cherryson 2010). The other building from the pre-urban period was the 9th-century masonry building that was located in the southern part of the site and that was initially non-ecclesiastical in nature. It is one of the very few preNorman non-ecclesiastical masonry buildings in England. It is possible that the building was the center of an agricultural estate that was located within the walls of the former Roman town. As noted above, in the later Middle Ages this building was transformed into St. Mary’s church. During the 7th to 9th centuries, there were large open spaces within Winchester’s Roman walls, and, as noted above, a number of these were under cultivation. The area within the walls was home to a bishopric and a small number of private dwellings. Barlow et al. (1976, 45) stress the decidedly non-urban character of Middle Anglo-Saxon Winchester, especially when compared to Hamwic. However, archaeological evidence suggests that Winchester’s urban redevelopment began in the late 9th century. Biddle (1975, 311) notes that in the late 9th century timber buildings were constructed along the western margin of Lower Brook Street. During the course of the excavations twelve timber houses were uncovered on the Brook Street site. House IX, which was initially built as a single 9.5 by 5 m room, revealed a sequence of 17 chalk floors (Biddle 1975, 311). The house plots were separated by fence lines, and the pits in the back yards of the houses probably served as tanning pits. The name, Tanner Street, is first recorded in 990 CE (Biddle 1975, 317). As Biddle (1975, 320) notes, although the late 9th century was a 2

There is little evidence to suggest that Christianity survived beyond the 5th century in much of eastern England, but it did survive in Wales and parts of western Britain. The archaeological evidence from Wasperton in Warwickshire (Carver et  al. 2009, see also the Archaeological Data Service (http://archaeologydataservice.ac.uk/archives/view/ wasperton_eh_2008/overview.cfm)) illustrates these religious changes. The Late Roman and earlier 5th-century graves at Wasperton are all oriented east–west and include a high proportion of coffins. These are typical of Christian Roman burials in western Britain. At the Wasperton cemetery there is a switch to Anglo-Saxon style graves in the later 5th century, including 22 cremations with an average date of 480 CE and 53 late 5th- to 7th-century inhumation burials. These later burials appear to be Pagan. The reconversion from Paganism to Christianity began with St. Augustine’s visit to Canterbury in 597 CE. However, the reconversion to Christianity took nearly two centuries to complete. For a fuller discussion of Anglo-Saxon Paganism, see Carver, Sanmark, and Semple (2010). For an overview of the archaeology of the conversion period in East Anglia, see Hoggett (2010).

Late Anglo-Saxon Winchester

145

formative period for the history of Winchester, this period has “no surviving documentary evidence of any kind.” This is why archaeology has played such a crucial role in our understanding of the rebirth of towns in Anglo-Saxon England. The replanned town of Winchester was established within the old Roman walls, enclosing an area of 58.2 ha, making it one of the largest urban centers in Britain at this time (Barlow et  al. 1976, 451). The areas occupied by the ecclesiastical and other enclosures were small enough to be incorporated into the new late 9th-century street plan. Archaeological and historical data suggest that the initial street frontages were parcelled out into large blocks, which would allow them to be subdivided into tenements. As Barlow et al. (1976, 452) note, the challenge that the planners of Winchester faced was to encourage settlement within the burh, since the initial function of this replanning was for defense in a time of political and military unrest. It is not clear how rapidly the area within the Roman walls was developed, and it is likely that the development took place over many decades (Barlow et al. 1976, 453–4). This should not surprise us, since the development of towns and cities is a process, not a single event. Street names, like Tanner Street, indicate that trades were localized in different parts of the town by the late 10th century (Barlow et al. 1976, 455). The trades that can be identified by late 10th-century street names include butchery, tanning, and shield-making, and the archaeological evidence from the Brook Street site confirms that tanning was established by the 11th century (Barlow et al. 1976, 459). One of the most important features of Winchester that may distinguish it from other contemporary settlements is its importance as an ecclesiastical center (Barlow et al. 1976, 464). The bishopric of Winchester was established in 660 CE. The minster church itself, the “Old Minster,” was initially constructed in 648. Shortly after Winchester was replanned in the late 9th century, a “New Minster” was constructed between 901 and 903. The nearby Old Minster became a priory church for a monastic community, and the Nunnaminster, later known as St. Mary’s Abbey, was also founded in the first decade of the 10th century. The presence of these religious communities may have encouraged the development of crafts and trades within the town of Winchester. In addition, both the Old Minster and the New Minster held vast estates in the surrounding countryside. Banham and Faith (2014, 238)  suggest that the rise of Winchester as a commercial center may have been accompanied by an intensification of production on these estates. Winchester could have provided an important market for wool and other agricultural products from these large land holdings.

146

Towns in Late Anglo-Saxon England

While the archaeological data from Winchester demonstrated that the town was substantially replanned in the late 9th century CE, the model proposed by Biddle and Hill (1971) has been extended beyond Winchester to other parts of the kingdom of Wessex. Biddle and Hill note that the Burghal Hidage, a document compiled between 911 and 914 CE,3 represents a strategic plan for Wessex based on a system of fortified places or burhs, each one no further than 40 miles (65 km) from the next. However, not all the burhs developed into towns. Some are very small sites, including forts and fortified bridgeheads. These are similar to the 8th-century burhs identified by Haslam (1987) for Middle Saxon Mercia. However, Biddle and Hill focus on six other towns whose layouts are similar to Winchester’s. Three were originally walled Roman towns, and three appear to be new Late Anglo-Saxon foundations/ fortifications. The three towns that included new earthen fortifications were Wareham in Dorset, Cricklade in Wiltshire, and Wallingford in Oxfordshire. These three towns have been subject to very different amounts of archaeological survey and excavation, and the extent to which these sites could be considered urban in the late 9th and 10th centuries remains a matter of considerable debate. The archaeological record and historical data from Wareham illustrate some of these problems. Wareham is strategically located on a dry point between the rivers Frome and Piddle and at the head of the Wareham Channel of the Poole harbor (Figure 4.3). By the end of the Anglo-Saxon period Wareham had emerged as one of the most important towns in the county, but its earlier history is less clear. Wareham is the site of a possible British Christian monastery that predates the Saxon conquest of the region in the late 7th century. It subsequently became the site of a minster church and a large royal estate (Dorset Historic Towns Survey:  Wareham 2011). The area was overrun by the Great Viking Army in 876, and shortly thereafter it was established as one of King Alfred’s burhs. Archaeological work carried out on the west walls of the town by the Royal Commission on Historical Monuments in the 1950s showed that the initial phase of the town walls included an earthen bank and ditch along with a timber palisade (Royal Commission on Historical Monuments 1959). These fortifications appear to date from around the time of Alfred the Great. The walls surround the north, east, and west, sides of the town. The south side of the town fronts onto the river Frome,

3

Haslam would date this document to 878–9 (see Haslam 2005, 2009, 111–13).

Late Anglo-Saxon Winchester

147

Figure 4.3. Plan of Wareham. Redrawn with permission of the editors of Medieval Archaeology.

and the nature of the defenses along this side is unknown. A  second phase of construction of a stone-faced rampart appears to have taken place sometime before the Norman Conquest. In summary, Wareham appears to have been initially fortified in the late 9th century, and these fortifications were maintained and strengthened during the Late AngloSaxon period. The nature of the occupation within the walls is poorly known. It has been assumed that the grid-plan layout of the streets was established at the same time as the walls, but it is possible that the plan developed in a piecemeal fashion (Dorset Historic Towns Survey: Wareham 2011,

148

Towns in Late Anglo-Saxon England

26). Although it has also been assumed that markets were held along the main streets, the nature and extent of settlement within the town during the Late Saxon period is also unknown. The Dorset Historic Towns Survey concludes that two of the critical research questions for Wareham are:  (1) is there evidence to show that the street plan is contemporary with the 9th-century construction of the town walls, and (2)  what is the extent of occupation within the walls during the Late Anglo-Saxon period (Dorset Historic Towns Survey:  Wareham 2011, 111)? The archaeological evidence for Cricklade in Wiltshire presents similar problems. The town is located in the upper River Thames valley at an important river crossing (Figure 4.4). The town appears to have been fortified in the late 9th century (Haslam 1976: 17). Based on the presence of a number of Middle Saxon potsherds, Cricklade appears to have been a pre-existing settlement of some sort (Haslam 1984, 107). The original defenses of the burh included a clay bank that was 6 m wide surrounded by a triple ditch arrangement that included two smaller ditches and a wide outer ditch (Haslam 1984, 107; Wiltshire County Archaeological Services 2004, 9). The defenses also included an intramural walkway, and it is possible that this walkway was constructed before the erection of the town walls (Haslam 1984, 109–10). The type of stockade that stood on top of the bank is unknown due to weathering, but the defenses were enhanced by the construction of a stone wall in the 10th or 11th century (Wiltshire County Archaeological Services 2004, 9). Unfortunately, the extent of Late Anglo-Saxon settlement within the burh is unknown. The archaeology of Wallingford in Oxfordshire is far better known as a result of the Burh to Borough research project that was conducted in Wallingford between 2008 and 2010 (Christie and Creighton 2013a; Christie et  al. 2009). The survey and excavation project was a collaboration of scholars from the universities of Exeter, Leicester, and Oxford and also involved substantial participation by local community members. The town was chosen for research because Wallingford was often seen as a classic example of combination of a fortified burh and incipient Late Anglo-Saxon town (see Biddle and Hill 1971, 81). In addition, since Wallingford did not develop as a thriving town in the post-medieval period, there were many open spaces within the town that could be explored archaeologically. The investigators combined large-scale geophysical surveys in these open areas with smaller, targeted excavations that were designed to answer specific archaeological questions (see Box:  Geophysical Survey in Archaeology). In

Late Anglo-Saxon Winchester

149

Figure  4.4. Plan of Cricklade. Redrawn with permission of Jeremy Haslam and the editors of Medieval Archaeology.

addition, the excavation placed small test trenches within garden plots, a method that has been used with success in the archaeological study of occupied medieval villages (see, for example, Whitlock 2014). The results of this research shed important new light on Wallingford’s history.

150

Towns in Late Anglo-Saxon England

Box: Geophysical Survey in Archaeology Archaeological excavation is expensive and time-consuming. In many cases, archaeologists do not know what they are going to find until they put a shovel or a trowel in the ground. New methods of geophysical survey allow archaeologists to “see” below the ground without excavations. In some cases, these surveys can eliminate the need for further excavation. In others, they have helped archaeologists identify areas of a site or town that warrant additional smaller-scale excavations. Three methods of geophysical survey are commonly used in archaeological research:  magnetometry, resistivity, and ground-penetrating radar. Magnetometers are sensitive instruments that can measure slight deviations in the earth’s magnetic field. These deviations can be caused by underground pits, ditches, and other sub-surface features. The deviations result from the differing magnetic intensity of features such as backfilled ditches (Figure  4.5).

Figure  4.5. A magnetometer in use. Readings are taken at 50  cm intervals. Photo credit:  Douglas V.  Campana. I  have permission to reproduce this image from both Doug Campana and Roseanne Schot, who appears in the photo.

Late Anglo-Saxon Winchester

151

Resistivity survey measures the electrical resistance to a current that is passed through the earth between two probes (Figure  4.6). Features such as stone walls show a high resistance to the passage of the current, while features that tend to hold moisture, such as refilled ditches, have a very low resistance to the passage of a current. At Wallingford, resistivity and magnetometry were carried out over more than 11 ha as part of the Burh to Borough project. The project also made use of ground-penetrating radar (GPR) to examine two parts of Wallingford’s castle (Figure  4.7). GPR uses radar to provide a profile of the sub-surface layers at an archaeological site. It can be used to determine the depth of buried archaeological features (Christie, Creighton, and Edgeworth 2013).

Figure  4.6. An example of a resistivity survey. A  current is passed through the outer two probes, and the inner probes measure the degree of resistivity. Photo credit:  Douglas V.  Campana. I  have permission to reproduce this image from both Doug Campana and Ger Dowling, who appears in the photo.

152

Towns in Late Anglo-Saxon England

Figure 4.7. Ground-penetrating radar in use. Copyright: GSSI, reproduced with permission.

Archaeological research showed that Wallingford was not an entirely de novo site. Stray Roman-period finds and a 5th–6th-century cemetery near the southwest corner of the defenses point to some pre-9th-century occupation (Hamerow and Westlake 2013). Wallingford is located on a crossing point of the River Thames, and the Late Saxon earthen fortifications include a large rampart that rises 7–8 m above the surrounding ditch. The ramparts themselves were built over plow soil, indicating that the area was under cultivation in the Anglo-Saxon period. As part of the Late Saxon system of defense, a fortified bridgehead was constructed on the opposite site of the River Thames (Christie et al. 2009). These imposing defenses surrounded a town that was not completely occupied in the Late Saxon period. The area within the walls included many open areas that may have hosted periodic fairs and/or were used for gardens and grazing. The Late Saxon town had a strongly ecclesiastical focus, with many churches and chapels set up in the gates. These may also have doubled as aristocratic residences (Christie and Creighton 2013b, 144). It is possible that the open areas inside the burh were part

Late Anglo-Saxon Winchester

153

of an ecclesiastical estate. Rather than seeing Late Saxon Wallingford as a densely populated town, it appears to have served as periodic place of refuge or a location for mustering armies. It is far less densely populated than the earlier Middle Saxon wic sites. In summary, the available archaeological data show that Wareham, Wallingford, and Cricklade were fortified around 900 CE. All three have grid-plan layouts that are broadly similar to the town plan that has been established for Late Anglo-Saxon Winchester. It is not entirely clear, however, that these street plans were laid out at the same time as the defenses were constructed. Moreover, even if these street grids were established in the late 9th or early 10th century, it is not clear that these fortified burhs actually served as towns at that time. These burhs lack the dense occupation and the evidence for non-agricultural activities that we have used to define urbanism. They seem, instead, to have served as places of refuge, ecclesiastical centers, and places for farming activities, and probable sites of periodic fairs or markets. They have more in common with large rural estates than they do with the wics or other medieval urban sites. Biddle and Hill (1971, 81)  also identified three additional former Roman towns  – Exeter, Chichester, and Bath  – that appeared to have been replanned in the late 9th or early 10th century. All three towns have seen substantial histories of archaeological excavation and research, although much of the recent work at Exeter has been developer-funded and has not been published (Mark Maltby, personal communication). The critical question for archaeology is: to what extent do these towns show urban development in the 10th and 11th centuries CE? Exeter is located in the southwestern part of Britain, and it was initially established as a legionary fortress in the mid-1st century CE. The Roman legion departed in 75 CE, and the Roman town (civitas) of Isca Dumnorum was established within the walls of the former fortress a few years later (Allan et al. 1984, 385). There is no clear archaeological evidence for extensive sub-Roman occupation in Exeter, but a small Christian community may have survived in or close to the Roman town after the loss of most urban institutions (Allan et al. 1984, 409). Written sources indicate that Exeter was one of the burhs established by Alfred in the late 9th century, but none of the surviving wall masonry dates to the Late Saxon period (Allan et al. 1984, 396). While Exeter was clearly a thriving town with a large population at the time of the Norman Conquest, there is little evidence for urban life in Exeter before the late 10th or early 11th century, and the problematic term “proto-urban” has been used to describe Exeter in the early 10th century (Allen et al. 1984, 410). In summary, there is clearly a time lag between the refortification

154

Towns in Late Anglo-Saxon England

of Exeter as a burh in the late 9th century and its development as a town in the 11th century. The archaeological evidence from Chichester in West Sussex paints a slightly more complicated picture. Like Exeter, Roman Chichester began its life as a legionary fort. After the abandonment of the fort by the Second Roman Legion, the area developed into a thriving Roman town, Noviomagus Reginorum. There is little evidence for sub-Roman activity in Chichester other than the discovery of a gold solidus of Valentinian III (425–53), and no evidence for continuity of urban occupation between the Roman town and the subsequent Saxon occupation (Munby 1984, 322–3). Anglo-Saxon occupation resumed in the 7th century CE, and charter evidence suggests that an increasing amount of land within the town came under ecclesiastical control in the 7th and 8th centuries CE (Jervis 2009, 63). Archaeological data indicate that Chichester developed as a town from the 9th through the 11th centuries CE, and there is at least some evidence for limited cross-Channel trading in the earlier part of the 9th century in the form of a coin of Charles the Bald, one of Charlemagne’s grandsons (Jervis 2009, 71). Therefore, Chichester appears to have been an established Anglo-Saxon settlement at the time it was incorporated into the burghal system in the late 9th century. The early 10th-century Burghal Hidage indicates that it was one of the largest of the Late Saxon burhs. Ceramic evidence can be used to trace the development of craft production within Chichester. The earliest 7th–9th-century pottery was shell-tempered and fired in a reducing atmosphere. The clays came from a variety of different sources outside the town, which may indicate that the town was populated by people from a variety of rural settlements outside the town in the 9th and 10th centuries. During the Late Saxon period, pottery production shifts from a household-level of production to workshops. Five clamp kilns were recovered from excavations at Chapel Street, and these kilns seem to have supplied the whole town during the Late Saxon period. These data point to a gradual development of the town from the 7th century through the Late Saxon period. Bath was the subject of a long-term program of excavation from the 1950s to the 1980s (Cunliffe 1979), but the historic character of the town means that the potential for future large-scale programs of survey and excavation is slight (O’Leary 1981). The town was best known in both Roman and Saxon times for its baths. However, its strategic location at a river crossing on the Foss Way made it an excellent candidate for the burghal system. In 1980, a timber outwork was identified outside and parallel to the old Roman walls. The timber outwork was set on the lip of a recut ditch that was originally dug in the 4th century CE (O’Leary 1981, 1).

Late Anglo-Saxon Winchester

155

A 7th-century charter records the establishment of a convent of the Holy Virgin, possibly a double monastery, in Bath, although the location of this monastery remains a matter of debate (Cunliffe 1984, 347– 8). Historic documents also record the presence of the Mercian kings Ecgfrith and Burhred in Bath in 796 and 864 respectively, but unfortunately there is little archaeological evidence for the nature of 7th–9thcentury settlement within the town (Cunliffe 1984, 349). There was a market at Bath’s High Street in the later medieval period, and it is probable that it dates back to Anglo-Saxon times. In the early 10th century, Alfred the Great’s son Edward established a mint at Bath, and mints are usually associated with ports and market towns. Edward also ordered that all buying and selling should be conducted at a port with a reeve as a witness. In Anglo-Saxon times a reeve was an administrative agent of an Anglo-Saxon king whose duties included tax collection. The reason for this was to hinder the sale of stolen property. The first known reeve from Bath was a man named Alfred whose death is recorded in the Anglo-Saxon Chronicle for 906 (Manco 1998, 42). While the archaeological data are still lacking, the historical evidence points to Bath as a market center in Late Saxon times. By the late 11th century, Bath was the largest town in Somerset (Manco 1998, 43). Should Winchester be seen as a model for planned towns in Late Anglo-Saxon Wessex? From the archaeological data presented here it is clear that the extent of urban development in these burghal towns varied widely. Wareham, Cricklade, and Wallingford were not densely populated, and they lack the presence of non-agricultural specialists, such as potters and metalworkers. The evidence from Exeter provides little evidence for urban settlement until the late 10th or 11th centuries. Chichester seems to have developed gradually as a town from the 7th century onward, and Bath appears to have been a market center in Late Saxon times. These data have led recent scholars to suggest that the initial function of these burhs should be seen as primarily military (Holt 2009). Christie and Creighton (2013b, 405) warn us about seeing these late 9th-century burhs as a “visionary exercise in town planning,” even though many of these places did develop into sizeable towns in subsequent centuries (see also Holt 2009, 61). Christie and Creighton (2013b, 405) note, however, that the scale of these places and the investments that were made in them suggest that roles in addition to defense may have been planned for them in the future. One of the major roles that these fortified places may have played is as market centers. The political chaos of the late 9th and 10th centuries would have made unprotected markets vulnerable to attacks by Vikings and others. The construction and reuse of these fortified places

156

Towns in Late Anglo-Saxon England

would have reduced the costs of trade by reducing the threat of loss. In terms of the New Institutional Economics, well-fortified market places reduce transaction costs and therefore encourage trade. The presence of mints in some of these towns also serves to reduce transaction costs. These changes would have had additional advantages for the kings of Wessex, since more trade would have yielded more tolls, which would have strengthened the kings’ financial positions, as the need for revenues increases substantially in wartime. The burhs’ roles as both markets and places of refuge would also have provided significant benefits for the agricultural populations in the surrounding countryside. As Holt (2009, 57) notes, urbanization is a process that creates both the town and the countryside. The charter of the borough of Worchester in Mercia, written between 884 and 901, illustrates the dual roles that these fortified places played. The charter indicates that the ealdorman Æthelred (King Alfred’s trusted sub-regent in western Mercia) ordered the burh of Worcester to be built for the protection of all the people. It also notes that half of the market rights, i.e., revenues, were granted to the Church of St. Peter to repair the borough walls (Manco 1998, 40). The other half remained with the king. While the market mentioned in the charter may not necessarily be the same market that was located on the High Street in the later Middle Ages (Holt 2009, 66), it is clear that the charter has implications both for defense and for trade. The other point that should be made is that, based on present evidence, Winchester is the only one of these towns that appears to have been substantially replanned in the late 9th or early 10th century CE. Holt (2009) argues that the central zone of Worcester was not replanned in the late 9th century and that a later 10th- or early 11th-century date is more likely. Most of these other sites do not provide enough detailed stratigraphic evidence to determine whether their street grids were laid out at the time of Alfred the Great and his immediate successors. Back in 1990, Vince (1990, 154) suggested that many of the burhs in Mercia and Wessex “may have remained sparsely populated fortresses until well into the tenth century.” More than twenty-five years later, we do not have the data to challenge this interpretation. Given the paucity of other data, it is worth trying to identify the unique features of Winchester that led to its precocious urban development. The first is that Winchester included a substantial ecclesiastical community from the 7th century onward. While there is a sizeable ecclesiastical presence in many of these burghal towns, it is not necessarily true of all of them. In addition, Winchester may have taken over some of the administrative functions that were previously based in Hamwic.

London: From Wic to Burh

157

Winchester is located only about 10 miles (16 km) north of Hamwic, and the two towns are connected by a Roman road. Winchester’s protected location within the old Roman walls may have made it a more favorable location for marketing and trade. These features may make Winchester somewhat unique among the burghal towns in Wessex. London: From Wic to Burh London suffered from Viking attacks from the 830s onward, and the Viking army occupied London in 871–2 CE (Haslam 2010a, 111). After King Alfred’s defeat of the Viking army at the Battle of Edington, the Vikings retreated to a newly created kingdom in East Anglia, and the Viking army that had been based in Fulham to the west of London left England for the continent. The Anglo-Saxon Chronicle reports that London was restored by Alfred in 886, and this date has traditionally been used in the examination of the changes in London’s settlement that can be dated to the late 9th century CE. Haslam (2010a, 2010b) has recently suggested an alternative model for the redevelopment of London. He suggests that the rebuilding of London by Alfred took place soon after he regained control of London from the Vikings, probably beginning in late 879 or 880. Haslam has argued that in order to create both a fortress and a sustainable community within it, including trade and revenues for the king, London’s streets, wharves, and markets were replanned as part of a single system. He argues for the early date based on the recovery of three monogram coins from the Bull Wharf along the Thames. These coins were only issued for a very brief period of time after Alfred’s reconquest of London (Haslam 2010b, 209). Moreover, Haslam suggests that these coins indicate that trade was being carried out along the foreshore south of the London burh in the early 880s (Haslam 2010a, 113–14). In addition, the presence of earlier 9th-century objects in the area suggests that trading had been taking place in this region for some time (Haslam 2010b, 209). Haslam (2010a, 2010b) has suggested that Alfred’s plan for London would have included several closely interrelated elements. Since trading was being carried out along the Thames foreshore south of the old Roman town, roads were built that connected the hithes4 along the Thames with the gateways in the old Roman riverside wall. The plan would have involved repair of the Roman fortifications and gates, including the riverside wall. New gates in the riverside wall would have linked the hithes along the Thames to the new street plan within the 4

A hithe (hythe) is a small port or harbor, especially on a river.

158

Towns in Late Anglo-Saxon England

walls. Within the walls, a realigned street at Cheapside and Poultry Street may have served as an internal market. In addition, Haslam argues that a wooden bridge over the Thames, built on top of the Roman stone foundations, would have connected the London burh to the Southwark area on the far side of the river, although there is no archaeological evidence for this. Haslam (2010a, 137)  sees this as a systemic replanning of London that was designed to combine military effectiveness with “commercial, social and indeed religious viability and sustainability.” He argues that many of the features that are known from later historical and topographic studies can be traced to the Late Saxon period (Haslam 2010a, 138). While Haslam has presented us with an elegant historical model, it is problematic from an archaeological perspective for several reasons. Archaeological data are almost never precise enough to distinguish dates that are only six years apart. The presence of the monogram coins at Bull Wharf only indicates that the area was used as a port in the early 880s. It cannot help us determine whether London was replanned in 879 or 886, as the Anglo-Saxon Chronicle reports. While it is possible to make a circumstantial case for the presence of a bridge connecting London and Southwark in the late 9th century, we have no documentary or archaeological evidence to support this claim. Cowie (2014, 24) has noted that the first evidence for the bridge is later and based on two timbers that are dated to ca. 987–1032 based on dendrochronology. The strongest evidence that Haslam presents is the data that suggest that the roads that connect the hithes along the Thames and the gaps/gates in the old Roman riverside wall were developed in the late 9th century, since there is clear evidence for riverside commerce at that time (see below). In order to test Haslam’s model, we need more Late Saxon data from within the old Roman city. Even if Haslam is correct and London was replanned as a system in the late 9th century, the question still remains:  when did Late Saxon London develop a true urban character with a dense population including people engaged in non-agrarian pursuits? As archaeologists, we need to be interested in urban development and not just in urban origins. Vince (1990, 154) argued that London remained sparsely populated until “well into the tenth century.” Are there any new archaeological data that would allow us to challenge Vince’s view? Archaeological excavations and watching briefs5 at Æthelred’s Hithe/ Bull Wharf have helped to answer this question (Ayre and Wroe-Brown 5

In North America this is known as archaeological monitoring. Archaeological monitoring involves watching the construction process to make sure that no archaeological remains are uncovered or disturbed.

London: From Wic to Burh

159

2015, 159). Two open-area excavations along the Thames foreshore south of the Roman sea wall have revealed little activity in the area during the Early and Middle Anglo-Saxon periods, with the exception of two Middle Anglo-Saxon burials. The presence of two Northumbrian strycas (coins from northern England), along with some historical evidence, suggest that trade may have been taking place along the western portion of this foreshore as early as the mid-9th century. There is no evidence for structures from this period, so trade may have taken place from boats that were beached on the foreshore of the Thames (Ayre and Wroe-Brown 2015, 162–3). The first structures built in the area were trestles for a gangplank that were erected during the late 9th or early 10th century. These were followed by the construction of an embankment and the first building in the area, which was probably built in the third quarter of the 10th century (Ayre and Wroe-Brown 2015, 138). In the late 10th or early 11th century, a series of embankments was constructed along the waterfront at a time when London’s prosperity was growing (Ayre and Wroe-Brown 2015, 139). The embankments were marked by rows of piles, sometimes with timbers behind them, and the enclosed areas were filled with a variety of materials (Ayre and Wroe-Brown 2015, 140). The archaeological evidence shows that by the late 10th or early 11th century the riverside area was being divided into plots, possibly with different owners (Ayre and Wroe-Brown 2015, 165). While activity declined at Middle AngloSaxon period Lundenwic in the late 8th and early 9th century, the evidence from the Bull Wharf site shows the gradual redevelopment of the Lundenburh waterfront from the late 9th through the 11th centuries as a center of local, regional, and international trade (Ayre and Wroe-Brown 2015, 185–6). The data from the areas north of the foreshore are somewhat less conclusive. There is almost no evidence for occupation in the old Roman city during the 5th and 6th centuries (Cowie 2008). The reoccupation of the Londinium area began with the establishment of a cathedral within the walls, most likely near the site of the present-day St. Paul’s Cathedral, although there is no clear evidence for a Saxon cathedral church on or near the site (Cowie 2014, 20). More than a quarter-century after Vince’s (1990) classic book on Saxon London, there is still limited evidence for occupation within the old Roman city during the later 9th and early 10th century CE. There are few datable artifacts from this period. Cowie (2014, 21) notes that charter evidence points to some maritime trade in the 9th century, based on a charter granting a property within the walls to the Bishop of

160

Towns in Late Anglo-Saxon England

Worchester in 857. This property was probably located in the western part of the walled city. As noted above, Haslam (2009, 2010a, 2010b) argued that the entire area within the old city walls was replanned by Alfred the Great in the late 9th century. Very little of the Roman street plan is apparent in the Late Saxon layout of the town. Other scholars, however, have suggested that Alfred’s replanning was limited to a core area within the Roman city, extending south from Cheapside to the Late Anglo-Saxon waterfront (Cowie 2014, 22). Other streets would have been added gradually during the late 10th century. By that time, the occupied area would have expanded to about 30 ha with a population of about 5000 people (Cowie 2014, 22). While the density of housing in some areas of the city increased during the 11th century, parts of the intramural area “still lay empty or thinly populated at the end of the 11th century” (Cowie 2014, 23). These data suggest that Late Saxon London developed gradually between the time of Alfred the Great and the Norman Conquest. Yorvik: Anglo-Scandinavian York The city of York was captured by the Great Viking Army in 866–7. The Vikings were probably attracted to York because of its wealth, which was a result of its role in manufacturing and trading activities (Hall et  al. 2014, 707). As described in the previous chapter, York was one of the most important 8th–9th-century wic sites, and it served as a center of both regional and international trade. Unfortunately, the archaeological record for these later 9th-century events is quite limited. However, the deposit of an 8th-century helmet sometime around 890 CE may be a reflection of the political turmoil of the later 9th century. The helmet was manufactured between about 750 and 775 CE, and it was deposited in a small wood-lined pit at the Coppergate site in York (Hall 1984, 34–6; Hall et al. 2014, 707). Much of what we know about the development of urbanism in York from the late 9th through the 11th century comes from the open-area excavation carried out at 16–22 Coppergate. The site is located on a moraine on a spur of land near the confluence of the Foss and the Ouse rivers (Figure 4.8). It is located about 150 m south of the old Roman fortress (Hall et al. 2014). The Coppergate site was excavated between 1976 and 1981. In 1966 and 1970, York City Council had acquired two adjacent derelict properties on Coppergate – the old White Horse pub and the former Craven’s sweet factory. Although this area was scheduled for redevelopment, the economic conditions of the time led to its postponement, offering the

High

Ouse

gate

Coppergate

Piccadilly

Castlegate

Watching Brief

o rF

ss

ve

Ri 0

100 Meters

Figure 4.8. Plan of Anglo-ScandinavianYork showing the location (dark shading) of the Coppergate site. The area that is labeled as a watching brief was monitored, but it was not subject to large-scale excavation. Redrawn with permission of the York Archaeological Trust.

162

Towns in Late Anglo-Saxon England

York Archaeological Trust the opportunity to carry out a large-scale excavation, covering 1000 square meters, without delaying a development project (Hall 1984, 9). The large scale of the excavations and the excellent state of preservation, including many organic remains such as timbers, make the Coppergate site one of the most important sites for the study of later Anglo-Saxon urbanism. As Hall et al. (2014, 701) note, the archaeology of the Coppergate area is a key to our understanding of the rebirth of urbanism in York after decline and abandonment of this area in the post-Roman period. In addition, the York Archaeological Trust has made a commitment to publishing its results in a timely manner. These include a popular presentation of the excavation results (Hall 1984), a more detailed report on the archaeology of Viking York published by English Heritage (Hall 1994), and a final comprehensive site report (Hall et al. 2014), which was published after the death of Richard Hall who directed the excavations at the Coppergate site. The final report includes new information on dendrochronology, radiocarbon dating, and the settlement archaeology of the Coppergate site. In addition, separate specialist reports have been published on aspects of Coppergate material culture, including the pottery (Mainman 1990), textiles (Rogers 1997), and the faunal remains (O’Connor 1989). As noted above, the Coppergate area of York appears to have been abandoned in the post-Roman period. There is no evidence for 5th- and 6th-century activity at the site, although there are antiquarian accounts of Pagan Saxon burials from the nearby Pavement Street area (Hall et al. 2014, 701). There is little evidence for 7th-century activity either, although a number of disturbed burials from the Coppergate site may be as early as the late 7th or early 8th century (Hall et al. 2014, 701). These conversion-period burials might be associated with All Saints Church, Pavement or St. Mary’s Church, both of which appear to be early foundations. It is equally possible that these burials were not associated with any church, since non-churchyard burial was quite common in the Middle Anglo-Saxon period (Buckberry 2010). The presence of 8th-century pottery, along with the burials, points to some settlement activity in the Coppergate area during that period, although no structures of that date have been found at the Coppergate site itself. This may be part of a broad expansion of settlement and trade in York in the 8th century (see Chapter 3). The second half of the 9th century was a period of political turmoil in York. As noted above, the Viking army conquered York in 866/7, and King Alfred made peace with the invaders 30 years later in 894. It is in the early 10th century that we see clear evidence for the urban development of the Coppergate area. Property boundaries marked by fence

Yorvik: Anglo-Scandinavian York

163

lines were laid out in the early 10th century, followed by the construction of four tenement houses around 930–5 (Hall et al. 2014, 708). Palliser (2014, 66)  notes that this is probably the point when York becomes a single urban entity, as opposed to a series of shifting locations associated with churches, elite residences, and loci of manufacture and trade. The earliest houses were post-and-wattle6 dwellings with their short ends fronting on Coppergate. These buildings were continuously repaired and occupied in the first half of the 10th century. Palliser (2014, 66) notes that Viking York, Yorvik, was a denser and more compact city than its Anglian (Middle Anglo-Saxon) predecessor, Eorforwic. Its population may have reached 10,000–15,000 people before the Norman Conquest in 1066 (Palliser 2014, 67). Artifactual evidence shows that both domestic and craft activities took place within and around these tenements. Domestic activities included small-scale animal husbandry, horticulture, textile manufacture, and building maintenance. The tools and dyes used in textile processing are typically Anglo-Saxon rather than Scandinavian (Rogers 1997). Commercial activities included a range of crafts – metal-, bone-, antler-, amber-, and jet-working, as well as wood-turning (Hall et al. 2014, 710; Palliser 2014, 69). The name, Coppergate, meaning wooden “cup-turner street,” is first recorded in the 12th century, but the archaeological evidence shows that this craft was established in the area by the first half of the 10th century (Hall et al. 2014, 709). Many of the raw materials used in these crafts would have been available from the surrounding countryside, but other raw materials, such as Baltic amber, were imported from other parts of the Viking world (Hall et al. 2014, 710; Palliser 2014, 69). The archaeological evidence for metal-working includes evidence for both gold- and silver-working, as well as coin production. The evidence for coin production includes lead trial pieces and an iron coin die. One of the trial pieces is an impression of a coin of King Athelstan of York (929–39 CE), while the die is for a coin that was inscribed to Saint Peter, probably referring to York Cathedral which was dedicated to St. Peter (Hall 1984, 60–3). The die would have been used to make silver pennies, the principal coinage of 10th-century England. These pennies are the successors to the sceattas that are commonly found in Middle Anglo-Saxon sites. Since the production of coinage was under strict royal control, finds of dies and trial pieces are exceptionally rare, and the Coppergate finds have shed new light on the ways that these pennies were produced. From the perspective of the New Institutional 6

Wattle is made up of thick branches that are woven together between the posts. The wattle structures are then plastered with clay.

164

Towns in Late Anglo-Saxon England

Economics, coinage would have facilitated trade and exchange. Coins were also used as part of the tribute payments paid by the Anglo-Saxons to the Vikings. The post-and-wattle dwellings went out of use after about 25 years. Dendrochronology suggests that they were demolished just before 955–6 (Hall et  al. 2014, 711). The demolition may be related to a change in ownership of the tract that resulted from the political turmoil of the 940s and 950s. Erik Bloodaxe served as King of Northumbria twice, once in the 940s and a second time in the early 950s. After his departure in 954, York became part of the English kingdom. After a brief period of dumping and pit construction, a new series of post-and-plank structures were constructed on Coppergate around 960 (Hall et al. 2014, 711–12). These buildings had partial basements, up to 1.5 m deep, and the walls of the buildings were constructed of oak posts and planks (Hall 1984, 67). While these dwellings differed in style and construction from their early 10th-century predecessors, their locations continued to respect the early 10th-century property boundaries. Houses continued to be built on the same plots until the 19th century (Palliser 2014, 73). While domestic activities continued to be carried out in the Coppergate tenements in the late 10th century, craft activities appear to decline. Much higher levels of gold and silver production, as well as moneying and international trade, were seen in the post-and-wattle buildings that can be dated to the second quarter of the 10th century. Glassmaking is one of the few crafts that is well documented for the late 10th century. In the late 10th century, York became more of a local and regional market and less of an international one (Hall et al. 2014, 715). However, York was part of a ceramic province that extended from East Anglia northward along the east coast of England (Hall et  al. 2014, 717). Terry O’Connor (1989) carried out a pioneering study of the animal bone remains recovered from the Coppergate site that shed light on how these Anglo-Scandinavian tenement dwellers were provisioned with food and other animal products. O’Connor’s data show that the animal assemblages from all periods are dominated by the remains of cattle. This is likely a result of the available pasturage around the town. The Vale of Pickering is a low-lying, flat area that is especially suited to grass production, and beef and dairy herds are an important part of the agricultural economy to this day (O’Connor 1989, 199). The faunal material from the early 10th century is generally quite similar to the earlier faunal assemblages from Anglian Eorforwic. The material from the later 10th and 11th centuries is more diverse, with increasing numbers of pigs and

Yorvik: Anglo-Scandinavian York

165

wild birds (O’Connor 1989, 199–201). The presence of large numbers of cattle and pigs in the faunal assemblage points to important interactions between the town of Yorvik and its surrounding countryside. While cattle were clearly raised outside the city, it is possible that the pigs may have been raised within the town itself. However, recent isotopic studies (Hammond and O’Connor 2013) show that nearly all the pigs from the Coppergate and Swingate sites in York were raised in rural or woodland locations outside the city. There is also a marked change in the patterns of fish exploitation at the Coppergate site. The early 10th-century material includes more freshwater and migratory fish, e.g., salmon and carp, while the later faunal assemblages include more cod and other cod-family bones, reflecting “a shift in exploitation in the early 11th century from local rivers to the open sea” (O’Connor 1989, 197). O’Connor’s pioneering work recognized what is now known as the “Fish Event Horizon,” a marked increase in the consumption of marine fish that is seen around 1000 CE, especially in medieval town assemblages (Barrett et al. 2004). Subsequent isotopic analyses (Barrett et  al. 2011) have shown that this fish event initially involved fishing in local waters during the 11th and 12th centuries CE, while cod was obtained from more remote areas including northern Scotland, northern Norway, and Iceland from the 13th century onward. A recent meta-analysis of fish recovered from medieval London shows that cod only became common within the city around 1000 CE (Orton et al. 2014, 524). The London cod was initially caught locally and eaten fresh (Orton et  al. 2014, 526). The expansion of sea fishing may have been driven by some combination of urban demand and technological innovation in fishing practices. The data on fish utilization from London and York are paralleled by contemporary data from Flanders, indicating that this change is part of a widespread alteration in early medieval subsistence strategies (Barrett et al. 2011). While the changes in fish utilization at York reflect broad changes in medieval fishing strategies and diet, the other evidence for settlement patterns, subsistence practices, and technology point to more specifically local, regional, and national patterns in urban development. The 1988–9 excavations at the Queen’s Hotel site in York showed that the earlier Anglian buildings respected the Roman street pattern, while the 10th-century structures follow the modern street lines (Palliser 2014, 70). York’s defenses were also refurbished at the same time. These data show that early 10th-century urban replanning was not limited to the areas in Wessex that were under Anglo-Saxon control. While much of the material culture seen at Yorvik, such as textiles, is Anglo-Saxon

166

Towns in Late Anglo-Saxon England

rather than specifically Viking in style (Rogers 1997; Palliser 2014, 55), imported material including Baltic amber and Irish clothing pins indicate that Yorvik in the late 9th and 10th century was part of a broader Viking trading province. By the late 10th and 11th century, York seems to have become primarily a local and regional market. Much of the bulk trade was local and overland (Palliser 2014, 74). From a landscape perspective, a critical part of the urban revolution is the creation of a town and its rural hinterland. Central and eastern Yorkshire are some of the most agriculturally productive parts of eastern England, and increasing rural production may also have stimulated York’s development (Palliser 2014, 67). The faunal remains from Coppergate certainly point to a close relationship between the urban consumers in Yorvik and the rural producers in the surrounding areas. Viking Dublin While Ireland’s early medieval monasteries may have served some of the functions of towns (see, for example, Dougherty 1985),7 Wallace (2004, 466) has argued that “the concept of mainstream urbanism was introduced to Ireland, possibly from ninth-century England, with the Scandinavians acting as catalysts who transferred the idea.” Pat Wallace, the former director of the Irish National Museum, excavated the Fishamble Street– Wood Quay site, which provided important information about the settlement patterns of Viking Age Dublin (Wallace 2004, 2015). Since Yorvik and Dublin were founded by Danish Vikings at roughly the same time, it is useful to briefly compare the structure and layout of these two urban centers, especially since both York and Dublin were ruled by the 7

Dougherty (1985) argued that 9th-century monastic communities in Ireland carried out some of the same functions as cities in the ancient world. Valante (1998) challenged this interpretation, based on documentary sources that suggest that monasteries were self-sufficient communities and that the pre-900 monastic communities were engaged primarily in farming and herding. Moreover, she argued (Valante 1998, 12) that 9th-century monasteries could not be seen as towns because they were not true centers of trade and exchange. She also noted that while monasteries had craftsmen, they were not part of an industrial economy (Valante 1998, 17). John Soderberg (2001) has approached this problem from a different perspective. He examined the faunal remains from the large monastic community at Clonmacnoise in County Offaly. The Clonmacnoise assemblage is dated to around 700  CE, and, like the faunal assemblages from early urban Dublin, was dominated by the remains of cattle. The age profiles for the Clonmacnoise and the Dublin cattle were quite similar, including large numbers of older animals. Soderberg suggested that Clonmacnoise, lik Dublin, was provisioned with cattle from the surrounding countryside. If Soderberg is correct, this would indicate that these early monasteries were not self-sufficient and may have been engaged in activities other than agriculture and animal husbandry.

Viking Dublin

167

Figure 4.9. Reconstruction of Viking Dublin. Copyright: National Museum of Ireland, reproduced with permission.

descendants of Ivar the Boneless for most of the 80 years following 867 (Palliser 2014, 56). Excavations that have been carried out in Dublin since 1960, and particularly those conducted at the Fishamble Street–Wood Quay site (Wallace 2015), have revealed the character, layout, and material culture of early medieval Dublin. Viking raids on Ireland began in the late 8th century, and historical records report that the first Viking settlements in the form of naval camps or longphoirt were established near the mouth of the River Liffey in 841. However, their exact location is unknown (Edwards 1990, 180). The earliest urban settlement in Dublin was established at the confluence of the River Liffey and its southern tributary, the Poddle (Wallace 2004, 466). The settlement was located on high ground overlooking the rivers. The 10th–11th-century streets do not form a grid pattern; instead, they follow the natural contours of the land (Edwards 1990, 182). Lanes radiated from these streets, and each house plot had a pathway leading from the street or lane (Figure 4.9). The plot boundaries were marked by post-and-wattle fences. Within each plot, a post-and-wattle building was

168

Towns in Late Anglo-Saxon England

located with one end facing the street or lane. These buildings were generally sub-rectangular in shape and internally divided into three parts. The similarities to Yorvik include the long-term survival of the plot lines themselves and the use of post-and-wattle building methods. The internal organization of the houses is quite different, and the Coppergate area lacks the lanes that are seen in Dublin. As is the case in York, cattle were not kept in the urban plots but were reared outside the town (Wallace 2004, 467). Specialized crafts carried out in 10th-century Dublin included nonferrous metal-working, the production of antler combs, and wood-carving. Since the Vikings did not make their own pottery, wooden vessels took on a special importance (Edwards 1990, 185). Raw materials, such as Baltic amber, were imported to Dublin, while finished goods, including pins, combs, brooches, and woollens, were exported. Wallace (2004, 467) notes that Dublin prospered in the 11th century when the Irish Sea became a “Viking Lake.” In contrast, York prospered in the 11th century by becoming a local and regional market center, strengthening its ties to the surrounding countryside. Late Saxon Ipswich As noted in Chapter 3, Ipswich is the one of the wics that survived as an urban center into the Late Saxon period and later. Ipswich was occupied by the Vikings in 879/880, and it was ruled by them for 40 years until the West Saxons regained control of the area in 920. The period of Viking occupation was a period of substantial change (Wade n.d.). During this period, Ipswich was surrounded by defenses for the first time. Sunkenfeatured buildings, similar to the ones seen in Anglo-Scandinavian York, also appear for the first time.8 There is a marked increase in craft activities including iron- and copper-alloy metal-working. The presence of molds indicates that brooches were being manufactured in Ipswich, and the manufacture of a type of pottery known as Thetford Ware began to replace the production of Ipswich Ware. Evidence for the manufacture of this pottery was first discovered at Thetford (see below), but it was probably first developed by the Ipswich Ware potters in the late 9th century in Ipswich. It is a medium sandy grey ware that was manufactured on a potter’s wheel.9 The pottery kilns were still located in the northeast part of Ipswich at this time (Wade n.d.). 8

9

Note that these houses are different in form and construction from the sunken-featured buildings or grubenhaüser known from Early Anglo-Saxon sites such as West Stow. For a brief illustrated introduction to Middle and Late Saxon pottery from East Anglia, see www.spoilheap.co.uk/pot_gall1.htm.

Norwich

169

The town of Ipswich grew relatively little after its West Saxon reconquest in the early 10th century, but it remained one of the ten largest English towns at the time of the Norman Conquest. Its economy continued to be based on craft production and trade, and a mint was established in Ipswich in the reign of King Edgar (r. 959–75). Local and regional trade, especially focused on the East Midlands, dominated Ipswich in the 10th century, but international trade became more important once again in the 11th century (Wade n.d.). Species ratios and age profiles for the cattle from Ipswich are largely unchanged from the Middle Saxon period to the Norman Conquest, indicating that cattle were driven to market in Ipswich from the surrounding countryside (Crabtree 2016). In the 10th and 11th centuries, the Middle Saxon street plan remained largely intact, except for the changes caused by the construction of the defenses. The Late Saxon townscape is more uniform than its Middle Saxon predecessor, with buildings generally set back between 10 and 15 m from the street frontages. Some of the sunken-featured buildings increase in size and become two-storied, with the sunken portion transformed into a cellar or half-cellar. In summary, the political changes of the 9th and 10th century do not seem to have had a substantial effect on Ipswich’s role as a center of craft specialization and trade. The major change is the construction of the town’s defenses. These would have reduced transaction costs by making marketing and other forms of exchange safer. The presence of a mint and the increasing use of coinage would also have reduced transaction costs in the marketplace. While Ipswich was the only major town in Middle Saxon East Anglia, two other urban centers developed in Late Saxon/Anglo-Scandinavian East Anglia  – Norwich and Thetford. I  will briefly examine the archaeological evidence for these two East Anglian urban centers before we examine the models for the development of urbanism in England during the Late Saxon period. Norwich Norwich was one of the largest towns in England at the time of the Norman Conquest, but its roots date back to the Middle Anglo-Saxon period. It is located on the River Wensum, and today it serves as the county town for Norfolk. While earlier models suggested that Norwich developed out of a polyfocal settlement, including settlements along the river and up the hill to the south, recent archaeological research indicates that the early settlement was concentrated on the gravel terraces on either side of the Wensum (Ayers 2011, 71). The settlement was not dense on either the north or the south side of the river (Ayres 2011, 72), but

170

Towns in Late Anglo-Saxon England

evidence for coins and some imported Carolingian pottery point to some level of trade and exchange. Ayres (2011, 76) has argued that “Middle Saxon Norwich, therefore, perhaps consisted, at best, of modest mercantile settlements on both sides of the River Wensum.” These settlements appear to have been undefended. A defensive earthwork, including a bank and ditch, was constructed on the north side of the Wensum in the late 9th or early 10th century while Norwich was under Viking rule. The earthwork surrounds the possible mercantile area that was originally founded in the 8th and 9th centuries. Its structure parallels Danish fortifications in Britain and on the continent, including the fortification surrounding Ipswich that was constructed at about the same time (Ayres 2011, 77). Norwich became part of the Anglo-Saxon kingdom in 917 under the rule of Edward the Elder. Archaeological evidence, in the form of coin finds, suggest that Norwich had developed as a center of regional importance by the 930s, and it is possible that a mint was operating there as early as the 880s when Norwich was part of the Danelaw (Ayres 2011, 79). The area south of the River Wensum had a strong Scandinavian presence; all the diagnostically Scandinavian articles recovered from Norwich were recovered from the southern part of the settlement (Ayres 2011, 81). Urban occupation in this area had developed by the late 10th century (Ayres 2011, 80). While there is some debate about the details, the area to the south of the river appears to have been laid out in a rectilinear fashion (Ayres 2011, 81). It is not clear, however, whether this was a new foundation or whether it was structured around earlier Middle Saxon settlement. However, the presence of this rectilinear layout points to urban planning and some form of central authority (Ayres 2011, 83). In addition there is evidence for a 2-meter-deep defensive ditch on the south side of the town, but it appears to be somewhat later in date than the D-shaped ditch surrounding the northern portion of Norwich. By the 11th century, occupation extended beyond the enclosed area, and by the time of the Norman Conquest, Norwich was home to more than 5000 people (Blair 2013, 341). Evidence for craft activities from Late Anglo-Saxon Norwich includes pottery manufacture, antler-working, metal-working (including goldworking), textile manufacture, and horn-working, and possible warehouses have been identified along the river (Ayres 2011, 84–5). Sunkenfeatured buildings from the Greyfriars site have been identified as workshops (Emery 2007, 40). While some international trade may have been conducted in Norwich, most of the trade appears to have been local in nature (Ayres 2011, 85). In summary, Norwich had developed as

The Pattern of Urban Development in Late Saxon England

171

a center of craft production and trade serving the surrounding countryside by the late 11th century. Thetford Thetford is located on the south bank of the Little Ouse river at its confluence with the Thet in the Breckland region of southwest Norfolk. It is located only about 7 miles (11 km) from the Middle Saxon estate center of Brandon (see Chapter 3). Archaeological research has revealed substantial Middle Saxon settlement around the edges of what would become the Anglo-Scandinavian town. Excavations along the Brandon Road (Atkins and Connor 2010) uncovered a part of what appears to be a large Middle Saxon village. The excavations also revealed evidence for ferrous metal-working and possible steel production, and the settlement may have served as a periodic market. The Anglo-Saxon Chronicle notes that the Vikings established their winter quarters in Thetford in 870 CE. The Anglo-Scandinavian settlement was surrounded by a large and curving bank and ditch. Archaeological excavations within the Late Saxon town show that Thetford developed as a thriving urban community in the 10th and 11th centuries. Crafts and industries were carried out within the defended area south of the Little Ouse, including pottery manufacture, wood-working, textile manufacture, and leather-working (Rogerson and Dallas 1984). The crowded south bank of the river was home to both industry and commerce, including the manufacture of the eponymous Thetford Ware, a widespread form of Late Saxon pottery that, as noted above, was initially developed at Ipswich. The pottery is a sandy grey wheel-thrown ware. The commerce and trade carried out at Thetford appears to be mostly regional. There is little evidence for overseas trade (Dallas 1993). Like Norwich and Ipswich, Thetford seems to have been a center for regional trade and craft production in the Late Saxon period. The Pattern of Urban Development in Late Saxon England With the exception of Ipswich, the Middle Saxon wic sites seem to have lost their urban character in the early to middle part of the 9th century. In the period between the late 9th and the late 11th century, AngloSaxon England saw the development of a number of towns that served as centers of regional commerce and trade. Understanding this process of urban growth and development is critical to our understanding of how Anglo-Saxon society changed during the later Anglo-Saxon period.

172

Towns in Late Anglo-Saxon England

When we take a closer look at the data, both from the Anglo-Saxon burhs and from the Anglo-Scandinavian towns within the Danelaw, the general pattern seems to be that towns developed gradually from the later 9th through the 11th century. Nearly 30  years ago, Astill (1991) suggested that urban development took place somewhat earlier north, while many of the towns in the south appear to have developed more rapidly in the later 10th and 11th centuries, and recent archaeological research seems to support his conclusions. While a number of larger towns, including London, York, and Norwich, have yielded evidence for international trade with the European continent and Ireland, many of the Late Saxon towns appear to be based on local and regional manufacturing and trade, indicating growing links between the towns and their surrounding countrysides. S. R. H. Jones (1993) used a New Institutional Economics approach to explain the rise of towns in England during the two centuries before the Norman Conquest. He argued that the Middle Saxon economy was focused on subsistence production and that trade took place primarily through gift exchange and institutional redistribution, a system where royal and monastic estates supplied themselves with goods and services (Jones 1993, 664). He further suggested that there was limited market exchange in the Middle Saxon period and that Middle Saxon coinage may have been used “to serve as a store of wealth rather than a medium of exchange” (Jones 1993, 662). Jones (1993, 660) further suggested that the Viking raids forced an “emergency conversion” from gift exchange and institutional redistribution to market exchange. The concept of an emergency conversion is drawn from the ethnographic literature on markets in Africa (Bohannan and Dalton 1962). Jones (1993) argued that the Viking invasions forced the conversion to a market economy so that the Anglo-Saxons could raise money for defense and for tribute. In addition, he suggested that structural changes in Late Anglo-Saxon society reduced transaction costs which, in turn, encouraged market exchange. These changes included the creation of a system of fortified burhs, as well as improvements in coinage and new legal codes. Jones (1993, 675) concludes by suggesting that: The main impetus to trade, however, was provided by the Viking invaders. The need to generate and monetize surpluses, either to buy off  Vikings or wage war against them, forced the Anglo-Saxons to trade whether they wanted to or not.

It has been over 20 years since Jones’s article was written, and new archaeological data recovered from both Middle and Late Anglo-Saxon sites may allow us to reconsider some of his arguments. Although Jones (1993, 659) challenges Hodges’s view that the wics and burhs were developed

The Pattern of Urban Development in Late Saxon England

173

by kings to monopolize the control of prestige goods, Jones nevertheless adopts a substantivist and neo-evolutionary view of the Middle AngloSaxon economy. I would argue that gift exchange should not be seen as a primitive precursor to market exchange; instead, the two forms can and do coexist, even in modern industrial societies. For example, giftgiving plays a major role in the US economy. In 2012, Christmas holiday shopping made up 19.3 percent of all US retail sales.10 In addition, recent archaeological research has provided more evidence for Middle Saxon marketing than was known in the 1990s. As noted in Chapter 3, the PAS (Portable Antiquities Scheme) has allowed archaeologists to identify a large number of Middle Saxon “productive sites,” and many of these sites appear to have been home to periodic markets or fairs. Moreover, the Middle Saxon economy appears to have been more monetized than was traditionally thought. The finds of coins in and around the wic sites point to a partially monetized and market-based economy. While the Viking threat certainly caused Alfred and his successors to raise both cash and labor through traditional obligations – such as military service and bridgework  – and through taxes and tolls, it is likely that the political competition between Middle Anglo-Saxon kings had a similar effect in the 8th century. In addition, the argument that the 8th-century economy was primarily engaged in subsistence production has been challenged by archaeological research at sites such as Flixborough, Brandon, and Wicken Bonhunt (see Chapter  3; see also Crabtree 2010b). Moreover, Jones (1993, 672; see also Astill 1991) notes the Danish boroughs grew more rapidly than the English ones during the 10th century, an argument that is inconsistent with the emergency conversion model. Where Jones is correct is that the construction of the burhs and similar fortified places in the Danelaw reduced both the costs and the risks of transacting business. None of the Middle Saxon wic sites was fortified, while all the Late Saxon market towns were. The archaeological data show that the development of most of these towns was gradual, and it is clear that the primary role of these fortified locations was initially for defense. As Jones (1993, 668) himself notes, “The most far reaching of Alfred’s measures, however, was to set in motion the systematic fortification and garrisoning of a series of burhs.” The archaeological data from towns like Lundenburgh suggest that urbanization was a gradual process that extended over several centuries. Even at the time of the Norman Conquest, some areas of 10

www.statista.com/statistics/243452/holiday-sales-as-a-percentage-of-industry-sales-inthe-us/.

174

Towns in Late Anglo-Saxon England

Lundenburgh were sparsely settled at best. I  would argue that the fortifications served to reduce transaction costs, but that the development of the Late Saxon towns was dependent on forging links with the surrounding countryside. The countryside provided raw materials like timber, hides, and food, and the towns provided markets and finished goods for the surrounding rural areas. The great medieval historian Sir Frank Stenton famously defined a town as a wall, a mint, and a market (Stenton 1971, 537). While this is certainly not a universal definition of urbanism, it is probably a fair definition for Anglo-Saxon England on the eve of the Norman Conquest. Both fortifications and coinage would have reduced transaction costs and encouraged market-based exchange. However, the gradual nature of this process from the later 9th through the 11th century points out the importance of the truly processual approach. It is also worth noting that most, but not all, the burhs developed into medieval towns. The process of urban development was a long-term one, and its origins date back to the early part of the Middle Saxon era. When we examine the long-term process of urban development in Anglo-Saxon England, several important conclusions can be drawn from the archaeological data. First, while most Roman urban centers did not survive as towns beyond the 5th century, they remained important places on the landscape. Churches and cathedrals were established within the old Roman town walls of London and Winchester as early as the 7th century CE. Substantial amounts of land within the old Roman town of Chichester appear to have come under ecclesiastical control as early as the 7th and 8th centuries. In Winchester, the Brook Street area itself may have been part of an agricultural estate in the Middle Saxon period, and although the Roman street plan did not survive in Winchester, the main east–west street may have survived as a thoroughfare. It is therefore not surprising that when Alfred was seeking to develop a system of defense for the kingdom of Wessex he made use of a number of older Roman fortifications. As a number of other scholars have noted, we should not see King Alfred, great though he was, as a visionary urban planner. However, the reuse and refurbishing of these Roman fortifications may have created the conditions that led to their gradual redevelopment as urban centers. It is useful to emphasize that some, but not all, of these burhs developed as thriving urban settlements by the end of the 11th century. The locations of these towns  – whether along rivers or near major roads and trade routes – would have made them attractive places to both Romans and Saxons alike. The second point I  would like to emphasize is the issue of historical contingency. I  argued in Chapter  3 that the largest of the wics

The Pattern of Urban Development in Late Saxon England

175

should be seen as towns. Archaeological data show that they were planned settlements with substantial populations including merchants, craftworkers, and others who were not engaged in subsistence activities such as agriculture and animal husbandry. Evidence recovered in the past 30  years points to a network of trading places in Middle AngloSaxon England, and the wic sites appear to be at the apex of this trading network. The wics were also closely engaged with the surrounding countryside. These centers provided goods such as Ipswich Ware pottery that were used in the surrounding rural settlements, and local farming communities provided the denizens of the wics with food and raw materials for craft production. Archaeological data show that these 8th–9thcentury urban communities were not fortified. The 9th century was the beginning of a period of intense political turmoil in England and on the continent. The political chaos in post-Charlemagne Carolingia may have affected the availability of goods from the continent. In addition, these large unfortified settlements would have been vulnerable to Viking raids and other depredations. In this uncertain environment it is not surprising that some of these trading places are gradually deserted in the 9th century. This should not cause us to describe these wics as failed or proto-urban. Ipswich was continuously occupied throughout this period, and London and York both developed as urban settlements once again in the 10th and 11th century. By way of comparison, even New York City, the largest city in the US and the center of the world financial industry, lost nearly 1 million people between 1970 and 1980. The city was saved from bankruptcy by a federal loan and debt restructuring. The causes for this decline include but are not limited to the switch from an industrial to a service economy, increasing crime rates, and white flight to the suburbs. Today, New York City has recovered, and it is home to more than 8 million people. These data should remind us that urban growth is not always linear, and it can be affected by social, political, and economic factors. The same was certainly true in the Anglo-Saxon period. The third point that I would make is that there does seem to be a link between Anglo-Saxon state formation and urban development, but it is not the simple neo-evolutionary link that the original Dark Age Economics posited. A  wide variety of different types of exchange  – gift exchange, barter, redistribution, and market exchange  – existed throughout the Middle and later Anglo-Saxon period in England. Beginning in the 7th century, local, regional, and international trade took place at a variety of venues, from small shoreline trading sites to the wics. Even the most powerful Anglo-Saxon kings did not have the power to control this trade, but they did have the power to tax it. Taxes on trade were just one of several sources of revenue for the Anglo-Saxon kings. Others included food

176

Towns in Late Anglo-Saxon England

rents and labor taxes in the form of military service and the maintenance of bridges and other defenses. Tribute and war booty also played a role. The need to tax trade would have given the Anglo-Saxon kings a strong interest in the development and maintenance of towns, and this may explain Alfred’s and his successors’ investment in the burhs. While the burghal system was initially a system of defense, the presence of these fortified sites and the heavy investments made in them would certainly have facilitated the localization and taxation of commerce. Finally, it is not surprising that we cannot identify a single cause or prime mover for the development of towns in later Anglo-Saxon England. Several factors certainly played a role, and one is certainly the degree of intensification and specialization in farming practices. During the Roman period, agriculture in Britain was intensive. Rural residents needed to produce a surplus which could be sold at market for cash to pay Roman taxes and rents. Animal husbandry was focused on cattle, including large cattle which were raised for meat and for traction (see Rizzetto et al. 2017). The end of Roman Britain led to changes in livestock husbandry and farming. As Banham and Faith (2014) have emphasized, the 5th and 6th centuries appear to have been a period of abatement in Anglo-Saxon farming. Some of the areas with the heaviest clay soils were abandoned, and unspecialized pastoralism grew at the expense of cereal agriculture. Archaeological evidence, however, points to increasing intensification and specialization in agriculture and animal husbandry beginning as early as the 7th century (Crabtree 2010b, 2014). A 7th-century iron plow share has been identified at the site of Lyminge, and the organic remains from sites such as Brandon, Wicken Bonhunt, and Bloodmoor Hill point to increasing specialization in animal husbandry. Agricultural intensification would have provided the economic background for the emergence of towns in the Middle and Late AngloSaxon periods. Intensive farming was needed to support craftworkers and other non-agricultural specialists, and more specialized farming practices were needed to provide raw materials such as wool. A second factor is the changing nature of trade networks. In the introduction, we noted that the great Belgian historian Henri Pirenne argued that Roman trading patterns in northwest Europe continued until the spread of Islam in the 8th century. Archaeological evidence shows that the non-Anglo-Saxon parts of western Britain maintained trade contacts with the Mediterranean during the 5th and 6th centuries. Late Roman amphorae and red-slipped tablewares were imported into southwestern Britain between the late 5th and mid-6th centuries. This trading system shows links with Byzantium and may be related to the acquisition of tin, lead, and silver from the southwestern regions of Britain (Campbell

The Pattern of Urban Development in Late Saxon England

177

2007). Evidence for long-distance trade in Anglo-Saxon England between 450 and 600 is relatively limited. Beginning in the 7th century, however, eastern and southern England became part of a separate North Sea trading province that included Merovingian Gaul, the Rhineland, and the Low Countries. This trade provided opportunities both for entrepreneurial individuals and for emerging Anglo-Saxon kings for whom this trade represented a new source of revenue. This is the final point that I  would like to emphasize here. The formalist–substantivist divide that dominated economic anthropology in the 1980s gave us only a partial picture of the relationship between trade and urbanism in early medieval Europe. The original Dark Age Economics model was too heavily focused on long-distance trade and the role of kings in the control of the trade in imported prestige goods. This model overemphasized the power of the emerging Saxon kings. Their role in trade appears to be one of revenue extraction rather than overall control. In addition, archaeological research conducted since the 1980s has emphasized the importance of regional trade in the emergence of towns in the Middle and Late Saxon periods. On the other hand, Wells’s focus on entrepreneurs may have given us an incomplete picture as well. People from many different levels of Anglo-Saxon society appear to have been engaged in trade. The small-scale trading places identified on both sides of the North Sea and along the Danish coast were probably used by local farmers, fishermen, and middlemen. Similarly, Henning (2007, 31) describes the traders and craftsmen who lived in the emporia and the wics in contemporary Francia as the “true keepers of the light of the urban economy” because these settlements provided niches for selfdetermined action such as craft production for unknown customers. It would be a stretch to see these individuals as the kinds of entrepreneurs that Wells envisions in Farms,Villages, and Cities. In addition, monasteries and wealthy estate centers were also involved in trade. Documentary evidence describes trade fairs at major monasteries in Carolingia, and the evidence from the so-called productive sites in Britain suggest that these sites were also home to periodic markets and fairs. This would have represented a source of revenue for the monasteries. In short, many different people within Anglo-Saxon society would have benefited from increased trade. A focus on kings and entrepreneurial elites provides only part of the picture. We will return to some of these issues in the concluding chapter.

Conclusions

In the previous four chapters we explored the changes in settlement patterns in England from the end of the Roman period through the 11th century CE. The goal of this final chapter is twofold. First I will explore the patterns of settlement change seen in the archaeological record from the 4th through the 11th centuries CE in Britain. Then I will try to place this pattern of change in a broader comparative context. What can the development of urbanism in Anglo-Saxon England tell us about the broader theoretical issues of urban origins and development, and how do the patterns of urban origins that we see in early medieval Britain compare to the growth and development of urbanism in other parts of the ancient world? The Rebirth of Towns in Anglo-Saxon England The archaeological data from most of the larger Roman towns in Britain suggest that they lost their urban character sometime in the 5th century. This does not necessarily mean that there was not some occupation in the towns, but it does mean that they were not densely populated and that they were no longer centers of trade, craft production, and exchange. Religious institutions, such as St. Paul in the Bail in Lincoln, may well have survived into the 5th and even the 6th centuries CE, but a church alone does not make a town. Similarly, there is clear evidence for some settlement around the city gates in Canterbury in the 5th century, but many parts of the Roman city were unoccupied, and it is not clear whether or not the 5th-century settlement was continuous with the earlier Roman occupation. In many ways, the archaeological evidence for Wroxeter shows that what was happening in western Britain differed from what was happening in many other parts of the former Roman province. The painstaking archaeological work at Wroxeter (White and Barker 1998) has shown that post-Roman urban occupation may have survived through the 5th and 6th centuries, probably as a center of a small British polity led by a 178

The Rebirth of Towns in Anglo-Saxon England

179

king or bishop. While the northern part of the city appears to have been abandoned in the post-Roman period, 33 buildings were constructed in the area of the baths basilica during the 5th and 6th centuries (White and Barker 1998, 127). It is also possible that the former basilica was being used as a market, although this cannot be proved conclusively (White and Barker 1998, 121). Wroxeter was finally abandoned sometime in the 7th century, and unlike many other former Roman towns, it was never reoccupied as an urban center. Wroxeter eventually came under Mercian control, and its successor was a small village located at the crossing of the River Severn. In many ways, Wroxeter has more in common with the Iron Age hill forts of southwestern Britain that were reoccupied in the post-Roman period and served as centers of small British polities (Burrow 1981) than with the earlier Roman towns such as Winchester and London. Both Wroxeter and the hillforts seem to have been centers of British polities that developed in the immediate post-Roman period. These polities in areas like Somerset seem to have been absorbed into the expanding Anglo-Saxon kingdom of Wessex, while the territory of Wroxeter eventually became part of Mercia. By the time of King Offa’s reign (757–96 CE), the Mercian kingdom had expanded as far as the Welsh border, and Wessex had expanded into Devon by the early 9th century and had gained control of part of Cornwall by the time of Alfred the Great. Even though archaeological evidence suggests that most of the Late Roman towns in Britain did not survive as urban entities beyond the early 5th century, these places remained as part of the post-Roman landscape. As described in Chapter 4, the one part of the late 9th-century street plan from Winchester that is relatively closely aligned with the Roman plan was the major east–west street (Figure 4.2). This evidence suggests that the street may have survived as a Roman thoroughfare, even though Winchester had lost its urban character by the early part of the 5th century CE. One of the roles that towns played in Roman Britain was as markets. These markets allowed farmers to exchange surplus agricultural produce for coins which could be used to pay taxes to the Roman state. The need for such a sizable agricultural and pastoral surplus would have diminished with the withdrawal of Roman military and political power in the first decade of the 5th century. Pollen evidence provides comparatively little evidence for forest regeneration in the immediate post-Roman period (P. Dark 2000). Those areas that do show forest regeneration are concentrated in northern and northwestern parts of England and in southeast Scotland (P. Dark 2000, 134). In other regions, pasture may have expanded at the expense of arable (P. Dark 2000, 142–3). These

180

Conclusions

data suggest that sub-Roman/Early Anglo-Saxon Britain remained home to substantial populations in the 5th and 6th centuries, a view that is certainly contrary to the kinds of slaughter and mayhem that is described by Gildas. The lack of forest regeneration also indicates that much of the British landscape continued to be used as cultivated fields and/or pastureland throughout the Dark Ages. It is reasonable to assume that post-Roman agriculture in eastern Britain was not as intensive as it had been during the Roman period, since Roman institutions including the military no longer had to be supported. Banham and Faith (2014) have described the early post-Roman period as a period of “abatement” or when the more intensive farming practices of the Roman period were replaced by small-scale cereal production and an increase in pastoralism. The limited archaeofaunal data that we have for this period points to a significant degree of autarky in subsistence practices. As Oosthuizen (2016) and others have argued, traditional land units and local subsistence practices may have survived, even if the towns and villas did not. It is fair to say, however, that during the 250  years from the early to mid-5th through the later 7th centuries, there was no settlement in eastern England that could reasonably be identified as a town. Most of the settlements appear to be small rural communities like West Stow and Mucking. While West Heslerton may show some internal zoning in terms of its activities, it is not a town. The spatial segregation that we see in the activities there is much closer to the kind of zoning of activities that we see in Middle Saxon rural settlements like Brandon. West Heslerton lacks the evidence for craft specialization that we would expect to see in a true town. Archaeological evidence shows that the 7th century was a period of substantial political, social, and economic changes, and these changes are reflected in new and different forms of settlement. The new forms of settlement include estate centers. Some, like Brandon and Flixborough, appear to have been monastic communities for all or at least part of their existence, while others were royal vills that formed part of the peripatetic kingship that characterized the Middle and Late Anglo-Saxon period. The recently excavated food-rent collection center at Higham Ferrers in Northampton (Hardy et al. 2007) shows how the king and his retinue may have eaten their way through their kingdom. A hierarchy of trading places first makes its appearance in the 7th century. At the bottom of the hierarchy are the small coastal sites that were identified by Loveluck and Tys (2006) on both sides of the North Sea. At a second level are the so-called “productive sites” that have produced large numbers of coins and non-ferrous metal artifacts. Sites such as Brandon are included among the productive sites. Many appear to be

The Rebirth of Towns in Anglo-Saxon England

181

associated with churches and other religious foundations, and they appear to be the sites of periodic markets or fairs. At the top of the hierarchy are the coastal and riverine wic sites that were founded in the 7th century and expand dramatically in the 8th century CE. While Lundenwic, Hamwic, Ipswich, and Eorforwic are the largest and best known archaeologically, a number of other possible wic sites have also been identified. One intriguing possibility is Dunwich in northern Suffolk. Dunwich was the site of a Roman fort and also home to the first bishop of East Anglia. Established in the 7th century, the bishopric was the home of St. Felix, who Christianized the East Anglian polity. Much of Dunwich has been lost to coastal erosion beginning as early as the 11th century, but at the time of Domesday, Dunwich was the tenth largest town in England. Much of the town center was lost to coastal erosion during the 13th and 14th centuries. Ongoing archaeological research, however, may allow some of the Anglo-Saxon and medieval town to be recovered (see, for example, Sear et al. 2013). What the available archaeological data do show is that eastern England was home to a range of different types of markets during the Middle Saxon period, ranging from small beachheads to large towns with populations in the thousands. In addition, the wic sites were home to a range of craft specialists, many working with material such as bone, antler, wool, and horn that would have been obtained from the surrounding countryside. In addition, these craftworkers were supplied with food, either directly through markets or indirectly through food rents or renders. Recent numismatic studies indicate that the Middle Saxon economy was partially monetized (Naismith 2012), and the use of coins would have facilitated market transactions. In addition, taxes on trade would have supported the Anglo-Saxon kings, and this revenue would have played a critical role at a time of fairly regular interpolity and intrapolity warfare. Since many of the crafts that were manufactured in the wics were based on animal products such as bone, horn, and wool, these emporia also served to establish links between urban dwellers and the surrounding countryside. The data from the wics and other contemporary sites indicate that Middle Saxon England was an early urban society. There is no question that nearly all of the large wics declined in the 9th century, and the reasons are likely related to the general political unrest in the 9th century. The combination of Viking raids and the political break-up of the Carolingian empire under Charlemagne’s grandsons would have made trading at large, unfortified coastal sites less attractive. It also would have represented a loss of revenue for the Anglo-Saxon kings. It is not surprising, therefore, that we see the re-establishment of fortified towns, both in the Danelaw and in Wessex, beginning in the late

182

Conclusions

9th century and continuing under Alfred the Great’s successors. These towns were fortified by walls and ditches in both regions. From the perspective of the New Institutional Economics, fortified towns would have facilitated trade and other forms of exchange by protecting the safety of the traders, craftworkers, and marketplaces. The final point to be made is that the development of the Late Saxon and Anglo-Scandinavian towns in the 10th and 11th century was a gradual process. While most of Alfred’s burhs developed into towns, the process did not happen overnight. Examining the Urban Revolution in Early Medieval Britain What do these data tell us about the causes and process of urbanization in Anglo-Saxon England? Two main conclusions can be drawn from the available data. The first is that there is no single factor or “prime mover” that can explain the rebirth of towns in Anglo-Saxon England. I will focus instead on several interconnected factors, including but not limited to political competition, increased regional and long-distance trade, and agricultural intensification, all of which played a role in the process. The second point is that we can see urbanization in AngloSaxon England as part of a broader pattern of social and economic change that was taking place in several regions of northwest Europe at about the same time. It is a pattern that has clear parallels to the situation in the greater Middle East in the 3rd millennium BCE, and one that I will discuss in greater detail after addressing the process in Anglo-Saxon England itself. Anglo-Saxon society was never an egalitarian one. Like most societies in later prehistoric and early medieval Europe, it was characterized by substantial differences in social status, political power, and material wealth. However, the archaeological and the limited historical records suggest that these differences increased substantially in the 7th century, as evidenced by changes in both settlement patterns and burial practices. The appearance of wealthy burials like Sutton Hoo Mound 1 and the Yeavering palace, along with the construction of royal genealogies, point to an increase in social stratification in the 7th century. What is the source of this increased wealth, power, and status in the Middle Saxon period? In Anglo-Saxon England, as in the rest of medieval Europe, land was a major source of wealth. Zooarchaeological (Crabtree 2010b, 2012; Crabtree and Campana 2015) and archaeobotanical (McKerracher 2016) data point to increasing surplus production from the 7th century onward. The discovery of a 7th-century iron coulter at Lyminge in Kent indicates at least the occasional use of the heavy

The Urban Revolution in Early Medieval Britain

183

plow, which would have allowed farmers to cultivate the heavier clay soils, also pointing to agricultural intensification at this time (see Thomas et al. 2016). Seventh- and 8th-century records also point to increasing competition among the Anglo-Saxon kings, and this competition and warfare must have required new or increased revenue sources, what Childe (1950) described as mobilizing the social surplus. Tribute from smaller and less powerful polities would have provided one source of revenue. The Tribal Hidage, a 7thcentury document probably reflecting the assessment of the tribute by King Wulfhere (658–75) is an example (York 1990, 10). Food rents would have been used to support the peripatetic king and his retinue, but the taxation of trade may have become an increasingly important revenue stream in the Middle Saxon period. As we have already seen, regional and international trade increased in the late 7th and 8th centuries with the establishment of a hierarchy of trading places, including the establishment of the emporia. The importance of taxing trade as a revenue stream was not limited to the Anglo-Saxon kingdoms. Davis (2016) has recently made a similar argument for Carolingian Francia. She noted that the Carolingian Empire controlled its money supply tightly, and trade was channeled to points on the frontier. The collection of taxes and duties on trade provided revenue for Charlemagne and his agents. The Middle Saxon kingdoms’ abilities to mobilize the social surplus, through food rents, labor service, and levies on trade, allowed the kingdoms of Wessex, Mercia, East Anglia, and Northumbria to expand at the expense of many smaller Anglo-Saxon and British polities. It also made eastern England a target for Viking raids and ultimately for the settlement and political control of much of eastern England. The importance of towns to the political economy of early medieval Britain is seen in the construction of towns, not only by Alfred of Wessex and his successors, but also in the Anglo-Scandinavian regions of eastern England and in the areas of Ireland that were also settled by the Vikings. This point leads to the second conclusion that can be drawn about the development of urbanism and state formation in Anglo-Saxon England. In the original edition of Dark Age Economics, Hodges (1982, 1989) argued for a close relationship between urbanism and state formation. He further argued that state formation in Carolingian Francia was an example of primary state formation, while Anglo-Saxon state formation should be seen as secondary. In many ways, this formulation obscures some of the interesting variation that we see in state formation in different parts of northwestern Europe in the 1st millennium CE. In a recent publication, the historian Jennifer Davis (2015) has emphasized

184

Conclusions

that the Carolingian Empire was, in fact, an empire.1 In scale, it was far larger than the kingdoms of the Anglo-Saxon heptarchy; it was larger than anything known in British history until the post-medieval period with the establishment of colonies in North America. State formation also developed in Iron Age/Viking Denmark in the 1st millennium (Randsburg 1980; Thurston 2001). Thurston (2001, 7) argued that state formation in southern Denmark involved the transition from a corporate form of rule to an exclusionary or network form of rule. The characteristics of exclusionary rule included a focus on personal prestige, power, and wealth, along with elite aggrandizement, princely burials, and an interest in long-distance trade and status-related craft goods. While some of these features are also seen in 7th-century Anglo-Saxon England, the Anglo-Saxon case also entails agricultural intensification and the mobilization of the resulting agricultural surplus. Rather than reducing these three cases to examples of primary and secondary state formation, I think it is more useful to look at them as examples of differently organized early states that engaged in trade, giftexchange, and other forms of interaction, including raiding. It would be more useful to see the relationships between the peoples of the North Sea region as an interaction sphere. The wic sites and the later AngloSaxon towns would have served as points of contact between AngloSaxon traders and foreign merchants and between town residents and inhabitants of the surrounding countryside. These urban trading places are sites where commodities and gifts are exchanged, but they are also sites where ideologies and identities are negotiated. A possible parallel can be seen in the 3rd millennium in the Near East. From the time of Childe (1935, 1936) through the 1970s (Service 1975), archaeologists and anthropologists focused on Egypt, Mesopotamia, and the Indus Valley as examples of primary state formation and early urbanism. Less attention was paid to the other 3rd-millennium complex societies that developed in adjacent areas such as the southern Levant and Iran. We now know that a wide range of complex societies developed in the Middle East during the Bronze Age and that these complex urban societies, although differently organized, interacted with one another in interesting ways. In particular, recent archaeological attention has focused on the Arabian Peninsula and the Gulf as a possible area of interaction between the Harappans and Mesopotamia (see, for example, Thornton et  al. 2016). Since trade has played a role in many models of urban origins going back to the early 20th century, understanding 1

Davis (2015) presents an innovative analysis of the structure of power and governance in the Carolingian Empire. She suggests that Charlemagne was a flexible and pragmatic ruler who learned from the peoples he conquered.

Urbanism in Later Prehistoric Europe

185

how these early urban societies interacted with their neighbors, whether through trade or other means, should be an important part of our understanding of urban process. Compared to the ancient and medieval cites in other parts of the world, Anglo-Saxon towns and early medieval towns in general are relatively small. As we noted in Chapter  3, the wics were home to several thousand people. The highest population estimates for Lundenwic are less than 15,000, and York, Hamwic, and Ipswich are considerably smaller. In the 10th century, Winchester may have been home to about 8000 people (Lambert, n.d.), and London was home to about 15,000 people at the time of the Norman Conquest. These early medieval towns are dwarfed by such cities as Mohenjo-daro and Harappa in the Indus, Teotihuacán in the Valley of Mexico, or Anyang in the Late Bronze Age of western China (see Wright 2010; Campbell in press; Cowgill 2015). Anyang was home to at least 100,000 people in the Late Bronze Age (Campbell in press), while Teotihuacán was home to  around 85,000 at its height (Cowgill 2015, 143). Wright (2010, 107–10) estimates the population of Mohenjo-daro at between 20,000 and 40,000 and the population of Harappa at 25,000– 30,000. In fact, one of the striking features of later European prehistory and European medieval archaeology is how late and limited the evidence for urbanism is in temperate Europe. As part of an attempt to understand the origins of medieval European urbanism, I want to start with a brief discussion of the evidence for towns and cities in later prehistoric Europe and follow with a comparison of state formation in Anglo-Saxon England to early state formation in pre-contact Hawaiʽi, a region where state formation took place without concomitant urbanism. Urbanism in Later Prehistoric Europe While towns and cities were fundamental aspects of both the ancient Greek and the ancient Roman economy, political organization, and society, evidence for urbanism north of the Alps in the pre-Roman world is both late and limited. In the early part of the 20th century, the great prehistorian V. Gordon Childe took an essentially diffusionist approach in order to develop a chronology for European prehistory before the development of radiocarbon age determination. Since elements of civilization, including writing and agriculture, had spread from the Near East and the eastern Mediterranean, he suggested that, in the absence of evidence to the contrary, other elements of social and technological complexity may also have diffused from the east (Childe 1925, 1957). With the advent of radiocarbon dating, it became clear that many innovations, including the development of copper metallurgy and the construction of megalithic

186

Conclusions

stone monuments, developed independently in Europe (Renfrew 1973). Subsequent research has shown that the oldest of the megalithic tombs are over 1000 years older than the Egyptian pyramids. The development of modern dating methods, however, has done little to alter the view that urbanism in temperate Europe north of the Alps was both late and limited. While Neolithic Europeans established trade networks in commodities such as salt and flint and Bronze Age Europeans developed complex metallurgical technologies, there is little evidence for anything that can be called a town in temperate Europe before the later part of the Hallstatt Iron Age (ca. 600–480 BCE). At this time, a small number of fortified hilltop settlements appear in west-central Europe that appear to be involved, directly or indirectly, with trade with the Greek trading colony of Massilia (modern Marseilles). Massilia was established by the east Greeks (living in what today is modern Turkey) around 600 BCE. Hillforts such as the Heuneberg (Arnold and Murray 2002) in Baden-Württemberg, southwest Germany, have yielded evidence for metal-working and textile production. The Heuneberg hillfort is surrounded by wealthy burials accompanied by rich grave goods including imports. It has also yielded imported Greek pottery and a fortification system of mud bricks with a limestone foundation that is based on a Greek model. The Heuneberg, at its height, may have been home to several thousand people. However, the Heuneberg and the other similar sites in west-central Europe seem to have gone out of use when Greek trade routes changed around 450 BCE. The oppida of the last two centuries BCE also appear to have functioned as towns (see Chapter 1). These are large fortified sites that appear to have served as political capitals for different Iron Age tribes in temperate Europe. Most are located along major rivers or trade routes, and the largest of these sites is over 600 ha in extent. One of the best known is the site of Manching in southern Germany, which has been subject to a series of excavation campaigns since the 1950s. This 380-ha site sits on a river terrace along the Danube, and excavations have revealed a planned layout and evidence for pottery-, glass-, and metalworking (Malin-Boyce 2004). The nearby oppidum of Kelheim, located at the confluence of the Danube and the Altmühl, encloses an area of 600 ha including important iron ore deposits that were used from the Iron Age through the Middle Ages (Wells 1993). Excavations carried out at Kelheim between 1987 and 1993 have produced evidence for intensive iron smelting and forging designed for trade. Tentative estimates for the population of Kelheim range between 500 and 2000 people (Wells 2004, 249). Some oppida, including Manching, were gradually abandoned in the 1st century BCE; others were destroyed by the Romans and still

Urbanism in Later Prehistoric Europe

187

continued to be occupied for several generations or more after the conquest of Gaul by the Romans. When we examine the archaeological record for later European prehistory as a whole, we see a patchwork of socially complex and technologically sophisticated societies in which urbanism played a relatively minor role. Urbanism in prehistoric Europe seems to emerge in response to a very specific set of historical circumstances. In the case of the late Hallstatt towns, urbanism seems to be related to trade with the Greek colony at Marseilles. The French oppida, many of which were founded around 120 BCE, appear to have been a response to the Roman conquest of southern Gaul a few years earlier (Collis 2004, 156). The British oppida were established a bit later, in the later 1st century BCE and the 1st century CE, and these sites provide extensive evidence for trade with the Roman world. It is probably fair to suggest that the oppida can be seen as a response to Roman expansion which both led to a need for defence and provided opportunities to trade and exchange. It is possible to argue that trade provided similar opportunities for Anglo-Saxon communities beginning in the 7th century. Unlike the western parts of Britain, which continued to have trade relations with the eastern Mediterranean in the 5th and 6th centuries, the eastern parts of England were largely isolated from the Mediterranean world after the withdrawal of the Roman legions from Britain in 407 and the abandonment of the provinces of Britain around 410 CE. The beginnings of a North Sea trading province in the 7th century may have provided new opportunities for the emerging Anglo-Saxon kings and, equally importantly, for other parts of Anglo-Saxon society. Given the number of small trading places and productive sites that have been identified along with the emporia, it would have been difficult for Anglo-Saxon kings to control trade. However, taxing trade would have been an important source of revenue for these emerging royal households. The elites would not have been the only elements of Anglo-Saxon society who benefited from the establishment of the emporia and other smaller trading sites. Craftworkers would have had the opportunity to sell their wares to a broader market. The data from Lundenwic show that closely related crafts were located near one another, one example of how craftworkers may have benefited from the establishment of the wics. Traders, merchants, and middlemen would also have benefited. In addition, the emporia would have provided a market for commodities such as salt and for surplus agricultural products, including wool, horn, and hides. The faunal data from Ipswich, for example, show that the inhabitants of the Middle Saxon wic were provisioned with older male oxen that were no longer needed for traction and transport (Crabtree 2016). In short, we

188

Conclusions

can no longer see the wic sites as representing a top-down model for the control of imported prestige goods. Part of their success may stem from the fact that multiple levels of Anglo-Saxon society benefited from their presence. The decline of the emporia in the early 9th century may also be related to more limited opportunities in the face of increasing raids and political instability. To sum up, for most of later European prehistory Europe was home to technologically and socially complex societies. Crumley (1974) drew on both literary and archaeological data to suggest that some of the Late Iron Age polities were, in fact, states. However, urbanism developed quite late in European prehistory, and when viewed in light of the spectrum of urban development, Iron Age towns are relatively small in terms of population size. Towns developed in the Iron Age when specific opportunities for trade and interaction with foreign communities arose. Urbanism was not a fundamental part of prehistoric European society, and it developed comparatively slowly in the post-Roman period in northwest Europe as well. Given the relatively late development of urbanism in Europe, it is worthwhile to take a closer look at Hawaiʽi, a place where we have archaeological evidence for state formation without any evidence for urbanism prior to European contact. Comparisons to the Hawaiian Case: State Formation without Urbanism In many parts of the world, including Late Bronze Age China (Campbell, in press), the Indus (Wright 2010), Sumer (Pollock 1999), and Teotihuacán (Cowgill 2015), urbanism and state formation go hand in hand. In some cases, the states were small city states that included the urban center and the surrounding countryside, as was the case in early Sumer. In other cases, the states were based on larger territories. Both urbanism and state formation were two crucial parts of Childe’s urban revolution. As I have argued elsewhere (Crabtree 2017), the patterns of state formation in pre-contact Hawaiʽi and Middle Anglo-Saxon England show some interesting similarities. Here we will examine both the similarities and the differences between the Hawaiian and the Anglo-Saxon case studies in an attempt to determine why, despite the similarities in the patterns of state formation, urbanism developed in Anglo-Saxon England, but it never developed in pre-contact Hawaiʽi. Kirch (2010, 2012) has recently argued that Hawaiʽi represents an important new case of primary state formation. While Polynesia has traditionally been seen as the home of chiefdoms (Kirch 1989), Kirch argues that a number of changes took place in Hawaiʽi in pre-contact times that

Comparisons to Social Complexity in Hawai‘i

189

represent a unique case of state formation in pre-colonial Polynesia. Kirch (2012) identified a number of social, political, and economic changes that took place in the centuries after the initial settlement of Hawaiʽi that mark the transition from chiefdom to state. They include the construction of genealogies that were exclusively for the aliʽi (nobility), which served to distinguish them from the commoners (makaʽainana). The aliʽi ate their way through their kingdoms in the same way as the Middle and Late Anglo-Saxon kings made use of food rents by moving through their retinues from one royal vill to the next. The Hawaiians described the process as “a shark going inland is my chief” (Kirch 2012). After Hawaiʽi was settled from the Marquesas Islands about 1000 years ago, it rapidly filled up with farmers who combined agriculture based on taro and other “canoe plants”2 with animal husbandry, fishing, and collecting. Population growth led to agricultural intensification. In the well-watered areas, canals were constructed to allow for the irrigation and expansion of wetland taro. In the drier areas of East Maui and the Big Island, permanent dryland taro fields were established. The fields were bordered by rows of sugar cane, which served as windbreaks and which may have helped maintain the fertility of the soil (Patrick Kirch, personal communication). Just as the Middle Anglo-Saxon kings maintained strong relationships with the Church by donating land for churches and monasteries, the Hawaiian chiefs/kings also had close ties to the priesthood. Major temples, known as heiau, were located next to chiefly residences. Examples include the Piʽlanihale heiau in East Maui. Built of basalt, it rises to a height of 15 m, from a base of 104 by 126 m. By the 16th century CE, Hawaiʽi faced a problem of environmental circumscription  – there were no more lands available for dry farming on the island of Hawaiʽi and East Maui (Kirch 2012, 213). Following a model first suggested by Carneiro (1970), Kirch (2012) has suggested that environmental circumscription led to competition between the Hawaiian chiefs and ultimately state formation, although the Hawaiian Islands were not unified as a single kingdom until the early 19th century under Kamehameha I, nearly a quarter-century after the initial European contact in 1788. The Hawaiian case study has several interesting parallels to AngloSaxon England in the 7th and 8th centuries CE. In both areas, the construction of genealogies helped to distinguish the emerging chiefs/kings 2

The Hawaiians introduced about two dozen species of plants to Hawaiʽi. In addition to the staple crop, taro (kalo), they also cultivated bananas, breadfruit, sweet potatoes, sugar cane, and a range of other plants used for food and fiber.

190

Conclusions

from the commoners. As noted above, in pre-contact Hawaiʽi only chiefs were allowed to have genealogies. In Anglo-Saxon England, all the extant royal genealogies claim descent from one of the Germanic gods, usually Wodin, the god of war (Yorke 1990, 14–15). Both areas show evidence for close relations between religion and the state, and both the Anglo-Saxon kings and the Hawaiian chiefs relied on food rents to support themselves and their retainers/retinues. Both regions show evidence for agricultural intensification, and both 16th–18th-century Hawaiʽi and 7th–8thcentury England were characterized by competition and warfare. The Hawaiian and Anglo-Saxon cases are not entirely similar, however. The Hawaiians were facing environmental circumscription by the 16th century. On the other hand, the agricultural economy of AngloSaxon England was exceptionally productive and continued to experience intensification throughout the Late Anglo-Saxon and medieval periods. I  have argued elsewhere (Crabtree 2017) that the emerging Anglo-Saxon kingdoms may have faced a degree of political circumscription. While Wessex expanded into Somerset and Devon in the 7th and 8th centuries and Mercia had expanded to the Welsh border by the time of King Offa in the late 8th century, the Anglo-Saxons were unable to expand into the regions of Wales and Scotland that were home to well-established polities. What we see instead is competition among the powerful Middle Saxon polities such as Mercia and East Anglia to control other areas within Anglo-Saxon England. The Staffordshire hoard is a material manifestation of this inter-polity competition. The other striking difference, of course, is that the pre-contact Hawaiians never developed towns. Hawaiʽi remained a rural farming society until the 19th century, while towns were established in Middle Saxon England by the 8th century CE. There are several possible explanations for this difference. The first, of course, is the relative and absolute isolation of the Hawaiian Islands. Eastern England, on the other hand is a relatively short boat ride away from continental Europe. While the channel crossing is famously choppy, marathon swimmers have made the crossing many times. The second is the nature of the pre-contact Hawaiian land divisions. The traditional large landholding unit was the ahupuaʽa, a land division that extended from the highest mountains to the sea coast. Nearly all the necessary resources for daily life would have been available within a single ahupuaʽa, which made regional trade much less necessary. However, exchange at the local level was critical to the Hawaiians’ success, since very different resources were available at different elevations within the land divisions. The critical difference appears to be pre-contact Hawaiʽi’s isolation. This can be seen clearly in the history of Honolulu after European

Final Thoughts

191

contact. The first European sailor to sail into Honolulu Harbor was Captain William Brown of Great Britain in 1794. Many other ships followed, and during the late 18th and early 19th centuries Hawaiʽi became “a provisioning stop on the Northwest Coast-Canton fur trade route” (Sahlins 1992, 3). Hawaiʽi subsequently supplied sandalwood to the Chinese market, and when the supplies of this commodity disappeared, the Hawaiians turned to provisioning the Pacific whaling fleets. By the 1840s, Honolulu’s population was between 10,000 and 12,000 people (Sahlins 1992, 4). The Hawaiian case shows the importance of taking the long view in studying the history of urbanism. Drawing on a series of case studies from pre-Hispanic Mexico and early historic North America, Carballo and Fortenberry (2015) have argued that the artificial division between prehistoric and historic societies has hindered the study of urban growth and development. The Hawaiian case shows that contact with European, North American, and Chinese traders fundamentally changed the settlement patterns of pre-contact Hawaiʽi to one in which Honolulu became an important urban port. Final Thoughts Compared to many of the other urban societies in the ancient and the medieval world, Anglo-Saxon urbanism is relatively small-scale. The wics and Late Saxon towns lack many of the more complex administrative functions of later medieval cities. The wics lack the monumental architecture seen at sites like Teotihuacán, and their populations are much smaller than those of other cities in antiquity. So why should these towns be of interest to students of urban history? From a landscape perspective, Anglo-Saxon towns transformed the relationship between these early urban centers and the countryside, beginning with the Middle Saxon wics. This transformation can be seen in the distribution of Ipswich Ware pottery. In what Blinkhorn (2012, 99) has termed the “Primary Zone,” more or less equal to the Middle Saxon kingdom of East Anglia (modern Norfolk and Suffolk), Ipswich Ware pottery was used as everyday domestic pottery. Outside this zone, in other parts of eastern England, this pottery was an item of trade, and some forms may have been used to transport other commodities such as salt. Ipswich Ware replaced local handmade forms of pottery beginning in the 8th century. In return, Ipswich and the other wic sites consumed foodstuffs from the surrounding countryside and provided markets for surplus farm products. In terms of landscape archaeology, symbiotic relationships between the towns and the surrounding countryside were created for the first time since the end of the Roman period. The archaeological record, from small trading

192

Conclusions

places to productive sites to the wics and towns, points to an increase in trade and exchange in Anglo-Saxon England from the 7th century onward. The perspective of the New Institutional Economics suggests that the establishment of the wics and the later Saxon towns would have facilitated trade and exchange in a number of different ways. In Lundenwic, the zoning that brought together workers in related trades would have expedited the production of manufactured goods. The use of coinage would also have facilitated trade. The fortification of the Late Saxon and Anglo-Scandinavian towns in the late 9th and 10th century would also have facilitated trade and other forms of exchange by providing security for both traders and town residents. Even though the populations of the wics were relatively low and many of the Late Saxon towns grew slowly in the 10th century, by creating new relationships between the towns and the surrounding countryside and by facilitating trade these early towns laid the basis for the larger, multifunctional urban centers of the later Middle Ages and the early modern world. What can the study of Anglo-Saxon urbanism tell us about the bigpicture questions surrounding what Childe termed the urban revolution? Childe was one of the most important theoreticians in the history of archaeology, and his models of the urban revolution and the Neolithic revolution that preceded it helped to transform archaeological thought in the first half of the 20th century. The interest in neo-evolutionary models of social change, along with the rise of the processual movement in archaeology in the 1960s and 1970s, led to a search for the commonalities that are shared by complex societies in different parts of the world. It is only in the past 30 years that we have also come to appreciate the diversity that characterizes those ancient and medieval societies that we identify as cities and states. Medieval archaeology has been ignored by scholars in the United States because it does not fit our narrative for the development of “western civilization.” We look to the ancient Near East to understand the background for the glory that was Greece and the grandeur that was Rome. The early medieval world, with its limited textual and literary sources, is often dismissed as just the “Dark Ages.” However, it was these Dark Age societies that formed the background for the later medieval and modern worlds. Archaeology can shed new light on these Dark Ages, and the study of early medieval urbanism can help us understand how trade, early state formation, and historical contingencies played important roles in the rebirth of towns in the post-Roman West.

Bibliography

Adams, R. McC., 1966. The Evolution of Urban Society: Early Mesopotamia and Prehispanic Mexico. Chicago: Aldine. Addyman, P. V., 2015. Historic Towns Atlas: York. Oxford:  Oxbow Books for the Historic Towns and York Archaeological Trust. Addyman, P. V. and Hill, D. H., 1968. Saxon Southampton: a review of the evidence part I: history, location, date and character of the town. Proceedings of the Hampshire Field Club 25, 61–93. 1969. Saxon Southampton: a review of the evidence part II: industry, trade and everyday life. Proceedings of the Hampshire Field Club 26, 61–96. Alcock, L., 1963. Dinas Powys. Cardiff: University of Wales Press. 2001. Dinas Powys. In Medieval Archaeology:  An Encyclopedia, edited by P. Crabtree, pp. 80–1. New York: Garland. Allan, J., Henderson, C., and Higham, R., 1984. Saxon Exeter. In AngloSaxon Towns in Southern England, edited by J. Haslam, pp. 385–411. Chichester: Phillimore. Andrews, P., 1997. Excavations at Hamwic, Volume 2: Excavations at Six Dials. London: Council for British Archaeology Research Report No. 109. Anthony, D., 2007. The Horse, the Wheel, and Language: How Bronze-Age Riders from the Eurasian Steppes Shaped the Modern World. Princeton:  Princeton University Press. Archer, S., 2003. The zooarchaeology of an Anglo-Saxon village:  the faunal assemblage from Kilham, East Yorkshire. Unpublished MA thesis, University of York. Arnold, B. and Murray, M., 2002. A landscape of ancestors in southwest Germany. Antiquity 76 (292), 321–2. Arnold, C., 1984. Social evolution in Post-Roman western Europe. In European Social Evolution: Archaeological Perspectives, edited by J. Bintliff , pp. 277–83. Bradford: University of Bradford. 1997. The Archaeology of the Early Anglo-Saxon Kingdoms, 2nd edition. London: Routledge. Arnold, C. and Wardle, P., 1981. Early medieval settlement patterns in England. Medieval Archaeology 25, 145–9. Asher, M. and Asher, R., 1997. Mathematics of the Incas: The Code of the Quipu. Mineola, NY: Dover Publications. Astill, G. G., 1991. Towns and town hierarchies in Saxon England. Oxford Journal of Archaeology 10, 95–117. 193

194

Bibliography

Atkins, R. and Connor, A., 2010. Farmers and Ironsmiths: Prehistoric, Roman and Anglo-Saxon Occupation beside Brandon Road, Thetford, Norfolk. East Anglian Archaeology 134. Bar Hill: Oxford Archaeology East. Ayre, J. and Wroe-Brown, R., 2015. The post-Roman foreshore and the origins of the late Anglo-Saxon waterfront and the dock of Æthelred’s Hithe:  excavations at Bull Wharf, City of London. Archaeological Journal 172 (1), 121–94. Ayres, B., 2011. The growth of an urban landscape:  recent research in early medieval Norwich. Early Medieval Europe 19 (1), 62–90. Banham, D., 2010. ‘In the sweat of thy brow shalt thou eat bread’:  Cereals and cereal production in the Anglo-Saxon landscape. In The Landscape Archaeology of Anglo-Saxon England, edited by N. J. Higham and M. J. Ryan, pp. 175–92. Woodbridge: The Boydell Press. Banham, D. and Faith, R., 2014. Anglo-Saxon Farms and Farming. Oxford: Oxford University Press. Barker, P., White, R., Pretty, K., Bird, H., and Corbishley, M., 1997. The Baths Basilica Wroxeter Excavations, 1966–90. London:  English Heritage Archaeological Report 8. Barlow, F., Biddle, M., von Feilitzen, O., and Keene, D. J., 1976. Winchester in the Early Middle Ages:  An Edition and Discussion of the Winton Domesday. Oxford: Clarendon. Barrett, J., Locker, A. M., and Roberts, C. M., 2004. ‘Dark Age Economics’ revisited: the English fish bone evidence. Antiquity 78, 618–36. Barrett, J. H., Orton, D. C., Johnstone, C., Harland, J., Van Neer, W., Ervynck, A., Roberts, C. M., Locker, A. M., Amundsen, C., Enghoff , F., HamiltonDyer, S., Heinrich, D., Hufthammer, A. K., Jones, A. K.  G., Jonsson, L., Makowiecki, D., Pope, P., O’Connell, T. C., DeRoo, T., and Richards, M., 2011. Interpreting the expansion of sea fishing in medieval Europe using stable carbon isotope analysis of archaeological cod bones. Journal of Archaeological Science 38, 1516–24. Bartholomew, P., 1982. Fifth-century facts. Britannia 13, 261–70. Bartosiewicz, L., Van Neer, W., and Lentacker, A., 1997. Draught Cattle:  Their Osteological Identification and History. Tervuren: Musée Royale de l’Afrique Centrale. Biddle, M., 1970. Excavations at Winchester 1969. Eighth Interim Report. Antiquaries Journal 50, 277–326. 1975. Excavations at Winchester, 1971: Tenth and Final Interim Report Part II. Antiquaries Journal 55 (2), 295–337. 1976. Towns. In The Archaeology of Anglo-Saxon England, edited by D. M. Wilson, pp. 99–150. Cambridge: Cambridge University Press. 1984. London on the Strand. Popular Archaeology 6 (1), 23–7. 1990. The study of Winchester:  archaeology and history in a British town, 1961–1983. Albert Ricket Archaeological Trust Lecture. In British Academy Papers on Anglo-Saxon England, edited by E. G. Stanley, pp. 299–341. Oxford: Oxford University Press. Biddle, M. and Hill, D., 1971. Late Saxon planned towns. Antiquaries Journal 71, 70–85.

Bibliography

195

Biddle, M., Hudson, D. M., and Heighway, C. M., 1973. The Future of London’s Past. London: Rescue Publication No. 4. Biddle, M. and Kjølbye-Biddle, B., 2007. Winchester: from Venta to Wintancæstir. In Pagans and Christians: From Antiquity to the Middle Ages: Papers in Honour of Martin Henig Presented on the Occasion of His 65th Birthday, edited by Lauren Gilmour, pp. 189–214. Oxford: Archaeopress. Binchy, D. A., 1954. Secular institutions. In Early Irish Society, edited by B. Miles Dillon, pp. 52–65. Dublin: Mercier Press. Binford, L. R., 1962. Archaeology as anthropology. American Antiquity 28, 217–25. Binford, S. R. and Binford, L. R., eds., 1968. New Perspectives in Archaeology. Chicago: Aldine. Bintliff , J., 1984. Iron Age Europe in the context of social evolution from the Bronze Age through historic times. In European Social Evolution:  Archaeological Perspectives, edited by J. Bintliff , pp. 157–215. Bradford:  University of Bradford. Birbeck, V., with Smith, R. J. C., Andrews, P., and Stoodley, N., 2005. The Origins of Mid-Saxon Southampton:  Excavations at the Friends Provident St. Mary’s Stadium, 1998–2000. Salisbury: Wessex Archaeology. Blackmore, L., 1999. Aspects of trade and exchange evidenced by recent work on Saxon and medieval pottery from London. Transactions of the London and Middlesex Archaeological Society 50, 38–54. 2003.The pottery. In Middle Saxon London: Excavations at the Royal Opera House 1989–1999, edited by G. Malcolm, D. Bowsher, with R. Cowie, pp. 225–41. London: Museum of London Archaeology Service Monograph 15. Blair, I., Barham, E., and Blackmore, L., 2004. My Lord Essex. British Archaeology 76, 10–17. Blair, J., 2005. The Church in Anglo-Saxon Society. Oxford: Oxford University Press. 2011. Flixborough revisited. Anglo-Saxon Studies in Archaeology and History 17, 101–7. 2013. Norwich. In The Blackwell Encyclopedia of Anglo-Saxon England, edited by M. Lapidge, J. Blair, S. Keynes, and D. Scragg, p. 341. Oxford: Blackwell. Blinkhorn, P., 1989. Middle Saxon pottery from the Buttermarket kiln, Ipswich, Suffolk. Medieval Ceramics 13, 12–16. 1999. Of cabbages and kings: production, trade and consumption in Middle Saxon England. In Anglo-Saxon Trading Centres and Their Hinterlands: Beyond the Emporia, edited by M. Anderton, pp. 4–23. Glasgow: Cruithne Press. 2012. The Ipswich Ware Project:  Ceramics, Trade and Society in Middle Saxon England. London: Medieval Pottery Research Group Occasional Paper 7. Boessneck, J., von den Driesch, A., Meyer-Lemppenau, U., and Weschler-von Ohlen, E., 1971. Die Tierknockenfunde aus dem Oppidum von Manching. Wiesbaden: Die Ausgrabungen von Manching 6. Bohannan, P. and Dalton, G., 1962. Markets in Africa. Evanston, IL: Northwestern University Press. Booth, P., Simmonds, A., Boyle, A., Clough, S., Cool, H. E.  M., and Poor, D., 2010. The Late Roman Cemetery at Lankhills:  Excavations 2000–2005. Oxford: Oxford Archaeology.

196

Bibliography

Bourdieu, P., 1977. Outline of a Theory of Practice, translated by R.  Nice. Cambridge: Cambridge University Press. Bourdillon, J., 1988. Countryside and town:  the animal resources of Saxon Southampton. In Anglo-Saxon Settlements, edited by Della Hooke, pp. 177– 95. Oxford: Blackwell. 1994. The animal provisioning of Saxon Southampton. In Environment and Economy in Anglo-Saxon England, edited by J. Rackham, pp. 120–5. London: Council for British Archaeology Research Report 89. Bourdillon, J. and Coy, J. 1980. The animal bones. In Excavations at Melbourne Street, Southampton, 1971–76, by P. Holdsworth, pp. 79–121. London: Council for British Archaeology Research Report 33. Brisbane, M., 1994. Hamwic (Saxon Southampton):  the origin and development of an eighth century port and production centre. Actes des congrès de la Société d’Archéologie Médiévale 4, 27–34. Brogiolo, G. P. and Chavarria, A., 2015. The archaeology of the early medieval city: state of research and future perspectives. Paper presented at the Conference on the Environmental Archaeology of European Cities, Brussels, Belgium, 27–29 May, 2015. Brooks, D. A., 1986. A review of the evidence for continuity in British towns between the 5th and 6th centuries. Oxford Journal of Archaeology 5, 77–102. Brooks, N. P., 1971. The development of military obligations in eighth- and ninth-century England. In England before the Conquest, edited by P. Clemoes and K. Hughes, pp. 69–84. Cambridge: Cambridge University Press. Bruce-Mitford, R., 1972. Sutton Hoo Ship Burial, Volume 1. London:  British Museum. Buckberry, J., 2010. Cemetery diversity in the mid to late Anglo-Saxon period in Lincolnshire and Yorkshire. In Burial in Later Anglo-Saxon England: c. 650– 1100 AD, edited by J. Buckberry and A. Cherryson, pp. 1–25. Oxford: Oxbow. Burrow, I., 1981. Hillfort and Hill-Top Settlement in Somerset in the First to Eighth Centuries A.D. Oxford: British Archaeological Reports, British Series, 91. Campbell, E., 1996. Trade in the Dark Age West: peripheral activity? In Scotland in Dark Age Britain, edited by B. Crawford, pp. 79–91. St. Andrews: University of St. Andrews. 1999. Saints and Sea-Kings: The First Kingdom of the Scots. Edinburgh: Historic Scotland/Cannongate. 2007. Continental and Mediterranean Imports to Atlantic Britain and Ireland, AD 400–800. CBA Research Report 157. York: Council for British Archaeology. Campbell, R., in press. Violence, Kinship and the State: The Great Settlement Shang and Its World. Cambridge: Cambridge University Press. Carballo, D. A. and Fortenberry, B., 2015. Bridging prehistory and history in the archaeology of cities. Journal of Field Archaeology 40 (5), 542–59. Carlsson, D., 1992. Harbours and trading places on Gotland AD 600–1000. In Aspects of Maritime Scandinavia AD 200–1200: Proceedings of the Nordic Seminar on Maritime Aspects of Archaeology, edited by I. O. CrumlinPederson, pp. 145–58. Roskilde: Vikingeskipshallen. Carneiro, R. L., 1970. A theory of the origin of the state. Science 169, 733–8. Carver, M. O. H., 2005. Sutton Hoo: A Seventh-Century Princely Burial Ground and Its Context. Report of the Research Committee of the Society of Antiquaries of London 69. London: British Museum Press.

Bibliography

197

Carver, M., Hills, C., and Scheschkewitz, J., 2009. Wasperton: A Roman, British and Anglo-Saxon Community in Central England. Woodbridge: Boydell. Carver, M., Sanmark, A., and Semple, S., 2010. Signals of Belief in Early England: Anglo-Saxon Paganism Revisited. Oxford: Oxbow. Cherryson, A. K., 2010. ‘Such a resting place as is necessary for us in God’s sight and fitting in the eyes of the world’: Saxon Southampton and the development of churchyard burial. In J. Buckberry and A. K. Cherryson, eds., Burial in Later Anglo-Saxon England, 650–1100 AD, pp. 54–72. Oxford: Oxbow. Childe, V. G., 1925. The Dawn of European Civilization. New York: Knopf . 1935. New Light on the Most Ancient East:  The Oriental Prelude to European Prehistory. London: Kegan Paul. 1936. Man Makes Himself. London: Watts and Co. 1942. What Happened in History. Harmondsworth: Penguin Books. 1950. The Urban Revolution. Town Planning Review 21, 3–17. 1957. The Dawn of European Civilization, 6th edition. New York: Knopf . Christie, N. and Creighton, O., 2013a. Transforming Townscapes from Burh to Borough: The Archaeology of Wallingford, AD 800–1400. London:  Society for Medieval Archaeology Monograph 35. 2013b. Wallingford and early medieval urbanism. In Transforming Townscapes from Burh to Borough:  The Archaeology of Wallingford, AD 800–1400 by N. Christie and O. Creighton, pp. 405–7. London:  Society for Medieval Archaeology Monograph 35. Christie, N., Creighton, O., and Edgeworth, M., 2013. Investigating the townscape and hinterland: methods and sources. In Transforming Townscapes from Burh to Borough: The Archaeology of Wallingford, AD 800–1400, by N. Christie and O. Creighton, pp. 15–44. London:  Society for Medieval Archaeology Monograph 35. Christie, N., Edgeworth, M., and Hamerow, H., 2009. The big dig: Wallingford. British Archaeology 106 (May/June), available on line at www.archaeologyuk. org/ba/ba106/feat3.shtml. Clutton-Brock, J., 1976. The animal resources. In The Archaeology of Anglo-Saxon England, edited by D. M. Wilson, pp. 373–92. London: Methuen. Coase, R., 2005. The institutional structure of production. In Handbook of New Institutional Economics, edited by C. Ménard and M. M. Sherley, pp. 31–9. New York: Springer. Collingwood, R., and Richmond, I., 1969. The Archaeology of Roman Britain. London: Methuen. Collis, J., 2004. Oppida. In Ancient Europe 8000 BC–AD 1000: Encyclopedia of the Barbarian World, Volume 2, edited by P. Bogucki and P. Crabtree, pp. 154–7. New York: Charles Scribner’s Sons. Conkey, M. W. and Spector, J. D., 1984. Archaeology and the study of gender. In Advances in Archaeological Method and Theory, Volume 7, edited by M. B. Schiffer, pp. 1–38. New York: Academic Press. Costen, M., 2011. Anglo-Saxon Somerset. Oxford: Oxbow. Cowgill, G., 2015. Ancient Teotihuacán:  Early Urbanism in Central Mexico. Cambridge: Cambridge University Press. Cowie, R., 2001. English wics: problems with discovery and interpretation. In Wics: The Early Medieval Trading Centres of Northern Europe, edited by D. Hill and D. Cowie, pp. 14–21. Sheffield: Sheffield Academic Press.

198

Bibliography

2008. Descent into darkness: London in the 5th and 6th centuries. In Londinium and Beyond: Essays on Roman London and Its Hinterland for Harvey Sheldon, edited by J. Clark, J. Cotton, J. Hall, and H. Swain, pp. 49–53. York: Council for British Archaeology Research Report 156. 2014. Lundenburgh:  the archaeology of Late Saxon London. In Hidden Histories and Records of Antiquity: Essays on Saxon and Medieval London for John Clark, Curator Emeritus, Museum of London, edited by J. Cotton, J. Hall, J.  Keily, R. Sherris, and R. Stephenson, pp. 17–26. London and Middlesex Archaeological Society Special Paper 17. Cowie, R. and Blackmore, L., 2008. Early and Middle Saxon Rural Settlement in the London Region. London: Museum of London Archaeological Service Monograph Series 41. Cowie, R., Blackmore, L., Davis, A., Keily, J., and Rielly, K., 2012. Lundenwic: Excavations in Middle Saxon London, 1987–2000. London: Museum of London Archaeology 63. Cowie, R., Whytehead, R. L., and Blackmore, L., 1988. Two Middle Saxon occupation sites: excavations at Jubilee Hall and 21–22 Maiden Lane. Transactions of the London and Middlesex Archaeological Society 39, 47–163. Coy, J., 1980. Section 3: the animal bones. In J. Haslam, A Middle Saxon iron smelting site at Ramsbury, Wiltshire. Medieval Archaeology 24, 41–51. Crabtree, P. J., 1990. West Stow:  Early Anglo-Saxon Animal Husbandry. East Anglian Archaeology 47. Ipswich: Suffolk County Planning Department. 1996. Production and consumption in an early complex society: animal use in Middle Saxon East Anglia. World Archaeology 28, 58–75. 2004. Archaeology. In Dictionary of the Middle Ages, Supplement 1, edited by W. C. Jordan, pp. 25–30. New York: Charles Scribner’s Sons. 2007. Animals as material culture in Middle Saxon England: the zooarchaeological evidence for wool production at Brandon. In Breaking and Shaping Beastly Bodies: Animals as Material Culture in the Middle Ages, edited by A. Plushkowski, pp. 161–9. Oxford: Oxbow. 2009. The archaeology of medieval Europe. History Compass 7 (3), 385–402. 2010a. Zooarchaeology and colonialism in Roman Britain:  evidence from Icklingham. In Anthropological Approaches to Zooarchaeology:  Colonialism, Complexity and Animal Transformations, edited by D. Campana, P. Crabtree, S. DeFrance, J. Lev-Tov, and A. M. Choyke, pp. 190–4. Oxford: Oxbow. 2010b. Agricultural innovation and socio-economic change in early medieval Europe: evidence from Britain and France. World Archaeology 42, 122–36. 2012. Middle Anglo-Saxon Animal Husbandry in East Anglia. East Anglian Archaeology 143. Bury St. Edmunds: Suffolk County Council Archaeological Service. 2014. 2013 Farrell Lecture  – Remembering Bernard Wailes:  archaeological approaches to Late Iron Age and early medieval Ireland. Eolas 7, 92–102. 2016. Zooarchaeology at medieval Ipswich: from wic to regional market town. In Objects, Environment and Everyday Life in Medieval Europe, edited by Ben Jervis, Lee Broderick, and Idoia Grau-Sologestoa, pp. 19–39. Turnhout, Belgium: Brepols. 2017. State formation in Anglo-Saxon England. In European Archaeology as Anthropology: Papers in Memory of BernardWailes, edited by P. J. Crabtree and

Bibliography

199

P. Bogucki, pp. 245–59. Philadelphia:  University of Pennsylvania Museum of Anthropology and Archaeology. 2018. The value of studying large faunal collections using traditional zooarchaeological methods:  a case study from Anglo-Saxon England. In Zooarchaeology in Practice: Case Studies in Methodology and Interpretation in Archaeofaunal Analysis, edited by C. Giovas and M. Le Febvre, pp. 173–88. New York: Springer. under review. Provisioning Ipswich. Under review for East Anglian Archaeology. Crabtree, P. and Campana, D., 2014. Animal bone. In Brandon, Staunch Meadow, Suffolk: A High Status Middle Saxon Settlement on the Fen Edge, by Andrew Tester, Sue Anderson, Ian Riddler, and Bob Carr. East Anglian Archaeology 151, pp. 296–312. Bury St. Edmunds: Suffolk County Council Archaeological Service. 2015. Wool production, wealth, and trade in Middle Saxon England. In Animals and Inequality in the Ancient World, edited by B. S. Arbuckle and S. A. McCarty, pp. 337–53. Boulder: University of Colorado Press. Crabtree, P. J., Campana, D. V., Belfer-Cohen, A., and Bar-Yosef , D., 1991. First results of the excavations at Salibiya I, Lower Jordan Valley. In The Natufian Culture in the Levant, edited by O. Bar-Yosef and F. Valla, pp. 161–72. Ann Arbor, MI: International Monographs in Prehistory. Crabtree, P. and Stevens, P. 1994. The animal bone remains from Ipswich. Unpublished animal bone report submitted to English Heritage in August, 1994. Cristian Capelli, C., Redhead, N., Abernethy, J. K., Gratrix, F., Wilson, J. F., Moen, T., Hervig, T., Richards, M., Stumpf , M. P.  H., Underhill, P. A., Bradshaw, P., Shaha, A., Thomas, M. G., Bradman, N., and Goldstein, D. B., 2003. A Y chromosome census of the British Isles. Current Biology 13, 979–84. Crumley, C., 1974. Celtic Social Structure:  The Generation of Archaeologically Testable Hypotheses from Literary Evidence. Ann Arbor: University of Michigan Museum of Anthropology, Anthropological Papers Number 54. Cunliffe, B., ed., 1979. Excavations in Bath 1950–1975. Bristol:  Committee on Rescue Archaeology in Avon, Gloucestershire and Somerset Excavation Report 1. 1984, Saxon Bath. In Anglo-Saxon Towns in Southern England, edited by J. Haslam, pp. 345–58. Chichester: Phillimore. Cunliffe, B. and Davenport, P., 1985. The Temple of Sulis Minerva at Bath, Volume 1: The Site. Oxford: Oxford University School of Archaeology. Dallas, C., 1993. Excavations at Thetford by B.  K. Davidson between 1964 and 1970. East Anglian Archaeology 64. Dereham: Field Archaeology Division, Norfolk Museums Service. Dark, K., 2000. Britain and the End of the Roman Empire. Stroud: Tempus. Dark, P., 2000. The Environment of Britain in the First Millennium AD. London: Duckworth. Davis, J., 2015. Charlemagne’s Practice of Empire. Cambridge:  Cambridge University Press. 2016. The political economy of the Carolingian Empire. Paper presented at the conference on The Global Eighth Century: A Comparative Assessment

200

Bibliography

of Socio-Economic and Political Developments in the Eighth Century CE across Eurasia, Institute for the Study of the Ancient World, New  York University, April 14–15, 2016. Demarest, R., 2005. Ancient Maya: The Rise and Fall of a Rain Forest Civilization. Cambridge: Cambridge University Press. Dietler, M., 2010. Archaeologies of Colonialism:  Consumption, Entanglement, and Violence in Ancient Mediterranean France. Berkeley:  University of California Press. Dobney, K., Jaques, D., Barrett, J., and Johnstone, C., 2007. Farmers, Monks and Aristocrats:  The Environmental Archaeology of Anglo-Saxon Flixborough. Oxford: Oxbow. Dorset Historic Towns Survey, 2011. Dorset Historic Towns Survey: Wareham, available online at www.dorsetforyou.com/article/398181/Wareham--historic-towns-survey. Dougherty, C., 1985. The Monastic Town in Early Medieval Ireland. Oxford: British Archaeological Reports, International Series 225. Driver, J. 1984. Zooarchaeological analysis of raw material selection by a Saxon artisan. Journal of Field Archaeology 11, 397–403. Durant, W., 1980. The Age of Faith. The Story of Civilization, Book 4. New York: Simon and Shuster. Dyer, C., 2015. Historical and archaeological evidence for the rural hinterlands of English medieval towns. Paper presented the Conference on the Environmental Archaeology of European Cities, Brussels, Belgium, 27–29 May 2015. Edwards, N., 1990. The Archaeology of Early Medieval Ireland. Philadelphia: University of Pennsylvania Press. Emery, P. A., 2007. Norwich Greyfriars: Pro-Conquest Town and Medieval Friary. East Anglian Archaeology 120. Dereham:  Historic Environment, Norfolk Museums and Archaeology Service. Esmonde Cleary, A. S., 1989. The Ending of Roman Britain. London: Batsford. 2011. The endings(s) of Roman Britain. In The Oxford Handbook of AngloSaxon Archaeology, edited by H. Hamerow, D. A. Hinton, and S. Crawford, pp. 13–29. Oxford: Oxford University Press. Falconer, S., 1995. Rural responses to early urbanism:  village and household economy at Tell el-Hayyat, Jordan. Journal of Field Archaeology 22 (4), 399–419. Faulkner, N., 2001. The Decline and Fall of Roman Britain. Stroud: Tempus. Fern C., 2015. Before Sutton Hoo: The Prehistoric Remains and Early Anglo-Saxon Cemetery at Tranmer House, Bromeswell, Suffolk. East Anglian Archaeology 155. Bury St. Edmunds: Suffolk County Council Archaeological Service. Fowler, L. and Taylor, R., 2013. At the Limits of Lundenwic:  Excavations in the North-West of Middle Saxon London. London:  Museum of London Archaeology, Archaeology Studies 27. Frankenstein, S. and Rowlands, M., 1978. The internal structure and regional context of early Iron Age society in south-western Germany. Bulletin of the Institute of Archaeology, University of London 15, 73–112. Frere, S. S., 1983. Verulamium Excavations II. London: Reports of the Research Committee of the Society of Antiquaries of London 41.

Bibliography

201

2011. The saga of Verulamium building XXVII 2. Britannia 42, 263–74. Fulford, M. and Timby, J., 2000. Late Iron Age and Roman Silchester: Excavations on the Site of the Forum-Basilica 1977, 1980–86, Britannia Monograph 15. London: Society for the Promotion of Roman Studies. Fulford, M., Clarke, A., and Eckhardt, H., 2006. Life and Labour in Late Roman Silchester:  Excavations in Insula IX since 1997. London:  Society for the Promotion of Roman Studies. Gardiner, M., Cross, R., Macpherson-Grant, N., Riddler, I., Blackmore, L., Chick, D., Hamilton-Dyer, S., Murray, E., and Weir, D., 2001. Continental trade and non-urban ports in mid-Anglo-Saxon England:  excavations at Sandtun, West Hythe, Kent. Archaeological Journal 158, 161–290. Gardner, A., 2007. An Archaeology of Identity: Soldiers and Society in Later Roman Britain. Walnut Creek, CA: Left Coast Press. Gerrard, J., 2007. The temple of Sulis Minerva at Bath and the end of Roman Britain. Antiquaries Journal 87, 148–64. 2013. The Ruin of Roman Britain:  An Archaeological Perspective. Cambridge: Cambridge University Press. Giddens, A., 1984. The Constitution of Society. Berkeley: University of California Press. Gill, D., Goffin, R., and Carruth, J., 2015. New Museum Building, West Stow Anglo-Saxon Village WSW076, SACIC Archaeological Excavation Report No. 2015/083. Available online at http://grey-lit-suffolkarchaeology. s3.amazonaws.com/2015_083.pdf . Gilmour, B., 2007. Sub-Roman or Saxon. Pagan or Christian: who was buried in the early cemetery at St. Paul-in-the-Bail, Lincoln? In Pagans and Christians: From Antiquity to the Middle Ages: Papers in Honour of Martin Henig, Presented on the Occasion of his 65th Birthday, edited by Lauren Gilmour, pp. 229–56. Oxford: British Archaeological Reports, International Series 1610. Grant, A., 1976. The use of tooth wear as a guide to the age of domestic animals. In B. Cunliffe, Excavations at Portchester Castle, Volume 2: Saxon, pp. 245–79. London: Society of Antiquaries. 1982. The use of tooth wear as a guide to the age of domestic ungulates. In Ageing and Sexing Animal Bones from Archaeological Sites, edited by B. Wilson, C. Grigson, and S. Payne, pp. 91–108. Oxford:  British Archaeological Reports, British Series 109. Green, C. R., 2012. Britons and Anglo-Saxons: Lincolnshire AD 400–650. Studies in the History of Lincolnshire 3.  Lincoln:  Society for Lincolnshire History and Archaeology. Hall, R., 1984. The Excavations atYork: TheViking Dig. London: The Bodley Head. 1994. Viking Age York. London: Batsford. Hall, R. A., with Evans, D. T., Hunter-Mann, K., and Mainman, A. J., 2014. Anglo-Scandinavian Occupation at 16–22 Coppergate. Archaeology of York 8/ 5. York: Council for British Archaeology for the York Archaeological Trust. Halsall, G., 1999. Movers and shakers: the barbarians and the fall of Rome. Early Medieval Europe 8 (1), 131–45. Hamerow, H., 1991. Settlement mobility and the ‘Middle Saxon shift’:  rural settlements and settlement patterns in Anglo-Saxon England. Anglo-Saxon England 20, 1–17.

202

Bibliography

1993. Excavations at Mucking, Volume 2:  The Anglo-Saxon Settlement. London: English Heritage Archaeological Report 21. 2002. Early Medieval Settlements:  The Archaeology of Rural Communities in Northwest Europe, 400–900. Oxford: Oxford University Press. 2004. The archaeology of Early Anglo-Saxon settlements:  past, present and future. In Landscapes of Change:  Rural Evolutions in Late Antiquity and the Early Middle Ages, edited by N. Christie, pp. 301–16. Aldershot: Ashgate. 2006. The earliest Anglo-Saxon kingdoms. In The New Cambridge Medieval History, c. 500–700, edited by P. Fouracre, pp. 263–89. Cambridge: Cambridge University Press. 2007. Agrarian production and the emporia of mid Saxon England, ca. 650– 850. In Post-Roman Towns, Trade and Settlement in Europe and Byzantum, Volume 1:  The Heirs of the Roman West, edited by J. Henning, pp. 219–32. Berlin: Walter de Gruyter. 2011. Anglo-Saxon timber buildings and their social context. In The Oxford Handbook of Anglo-Saxon Archaeology, edited by H. Hamerow, D. Hinton, and S. Crawford, pp. 128–55. Oxford: Oxford University Press. 2012. Rural Settlements and Society in Anglo-Saxon England. Oxford:  Oxford University Press. Hamerow, H. with Westlake, S., 2013. Saxons before the burh: the early AngloSaxon cemetery at Wallingford. In Transforming Townscapes from Burh to Borough:  The Archaeology of Wallingford, AD 800–1400, by N. Christie and O. Creighton, pp. 57–63. London:  Society for Medieval Archaeology Monograph 35. Hammon, A., 2011. Understanding the Romano-British–early medieval transition: a zooarchaeological perspective from Wroxeter (Viroconium Cornoviorum). Britannia 42, 275–305. Hammond, C. and O’Connor, T., 2013. Pig diet in medieval York: carbon and nitrogen stable isotopes. Archaeological and Anthropological Sciences 5, 123–7. Hansen, I. L. and Wickham, C., 2000. The Long Eighth Century:  Production, Distribution and Demand. Leiden: Brill. Harbison, P., 1988. Pre-Christian Ireland: From the First Settlers to the Early Celts. London: Thames and Hudson. Hardy, A., Charles, B. M., andWilliams, R. J., 2007. Death andTaxes: The Archaeology of a Middle Saxon Estate Centre at Higham Ferrers, Northhamptonshire. Oxford: Oxbow. Härke, H., 1998. Archaeologists and migrations: a problem of attitude? Current Anthropology 39, 19–45. Haslam, J., 1976. Wiltshire Towns: The Archaeological Potential. Devizes: Wiltshire Archaeological and Natural History Society. 1980. A  Middle Saxon iron smelting site at Ramsbury, Wiltshire. Medieval Archaeology 24, 1–68. 1984. The towns of Wiltshire. In Anglo-Saxon Towns in Southern England, edited by J. Haslam, pp. 87–147. Chichester: Phillimore. 1987. Market and fortress in England in the reign of Offa. World Archaeology 18, 76–93. 2005. King Alfred and the Vikings: strategies and tactics, 876–886 AD. AngloSaxon Studies in Archaeology and History 13, 121–53.

Bibliography

203

2009. The development of Late Saxon Christchurch, Dorset, and the Burghal Hideage. Medieval Archaeology 53, 95–118. 2010a. The development of London by King Alfred:  a reassessment. Transactions of the London and Middlesex Archaeological Society 61, 109–44. 2010b. King Alfred and the development of London. London Archaeologist (Spring), 208–12. Hedges, R. E. M. and Gowlett, J. A. J., 1986. Radiocarbon dating by accelerator mass spectrometry. Scientific American 254 (1), 100–7. Hegmon, M., 2003. Setting theoretical egos aside: issues and theory in North American archaeology. American Antiquity 68 (2), 213–43. Heidinga, H. A., 1994. Frankish settlement at Gennep: a Migration Period center in the Dutch Meuse area. In The Archaeology of Gudme and Lundeborg: Papers Presented at a Conference at Svendborg, October 1991, edited by P. O. Nielson, K. Randsborg, and H. Thrane, pp. 202–8. Copenhagen:  Akademisk Forlag Universitetsforlaget i Kobenhaven. 2001. Gennep. In Medieval Archaeology: An Encyclopedia, edited by P. Crabtree, pp. 132–4. New York: Garland. Heighway, C. M., ed., 1972. The Erosion of History: Archaeology and Planning in Towns: A Study of Historic Towns Affected by Modern Development in England, Wales and Scotland. London: Council for British Archaeology. Helmuth Andersen, H., Madsen, H. J., and Voss, O., 1976. Danevirke. Copenhagen: Jysk Arkælogisk Selskab. Henig, M., 2011. The fate of Late Roman towns. In The Oxford Handbook of Anglo-Saxon Archaeology, edited by H. Hamerow, D. A. Hinton, and S. Crawford, pp. 515–33. Oxford: Oxford University Press. Hennig, J., 2007. Early European towns: the development of the economy in the Frankish realm between dynamism and deceleration AD 500–1100. In Trade and Settlement in Europe and Byzantium, Volume 1:  The Heirs of the West, edited by J. Hennig, pp. 3–40. Berlin: deGruyter. Higbee, L., 2009. Mammal and bird bone. In The Anglo-Saxon Settlement at Bloodmoor Hill Carlton Colville, Suffolk, by S. Lucy, J. Tipper, and A. Dickens, pp. 279–304. East Anglian Archaeology 131. Cambridge:  Cambridge Archaeological Unit. Higgs, E. S., ed., 1972. Papers in Economic Prehistory. Studies by members and associates of the British Academy Major Research Project in the early history of agriculture. Cambridge: Cambridge University Press. ed., 1975. Paleoeconomy. Cambridge: Cambridge University Press. Higham, N., 1994. The English Conquest: Gildas and Britain in the Fifth Century. Manchester: Manchester University Press. 2004. From sub-Roman Britain to Anglo-Saxon England:  Debating the Insular Dark Ages. History Compass 2, 1–29. Higham, N. and Ryan, M., 2013. The Anglo-Saxon World. New Haven, CT: Yale University Press. Hill, D., 2001. Fordwich, Kent. In Wics:  The Early Medieval Trading Centres of Northern Europe, edited by D. Hill and D. Cowie, p. 98. Sheffield: Sheffield Academic Press. Hill, D., Bailey, M., Gardiner, M., Griffiths, D. W., Holman, D., Parfitt, K., Perkins, D., and Riddler, I., 2001. Gazetteer of possible Anglo-Saxon wics.

204

Bibliography

In Wics:  The Early Medieval Trading Centres of Northern Europe, edited by D. Hill and D. Cowie, pp. 95–103. Sheffield: Sheffield Academic Press. Hill, D. and Worthington, M., 2003. Offa’s Dyke:  History and Guide. Stroud: Tempus Publishing. Hills, C. M., 2003. Origins of the English. London: Duckworth. Hills, C. M. and O’Connell, T. C., 2009. New light on the Anglo-Saxon succession: two cemeteries and their dates. Antiquity 83 (233), 1096–108. Hinton, D., Crawford, S., and Hamerow, H., 2011. The Oxford Handbook of Anglo-Saxon Archaeology. Oxford: Oxford University Press. Hirst, D. and Clark, D., 2009. Excavations at Mucking, Volume 3:  The AngloSaxon Cemeteries. London: Museum of London Archaeology. Hirst, S., 2004. The Prittlewell Prince: The Discovery of a Rich Anglo-Saxon Burial in Essex. London: Museum of London Archaeology. Hodder, I., 1986. Reading the Past:  Current Approaches to Interpretation in Archaeology. Cambridge: Cambridge University Press. Hodges, R. 1980. The pottery. In Excavations at Melbourne Street, Southampton, 1971–76, by P. Holdsworth, pp. 40–58. London:  Council for British Archaeology Research Report 33. 1982. Dark Age Economics:  The Origins of Towns and Trade AD 600–1000. London: Duckworth. 1989. Dark Age Economics: The Origins of Towns and Trade AD 600–1000, 2nd edition. London: Duckworth. 2012. Dark Age Economics: A New Audit. London: Duckworth. Hodges, R. and Whitehouse, D., 1983. Mohammed, Charlemagne and the Origins of Europe. Ithaca, NY: Cornell University Press. Hoggett, R., 2010. The Archaeology of East Anglian Conversion. Rochester, NY: Boydell. Holdsworth, P. E., 1980. Excavations at Melbourne Street, Southampton, 1971–76. London: Council for British Archaeology Research Report 33. Holman, D., 2001. Sandwich, Kent. In Wics:  The Early Medieval Trading Centres of Northern Europe, edited by D. Hill and D. Cowie, pp. 101–2. Sheffield: Sheffield Academic Press. Holmes, M., 2013. Entrepreneurs and traditional farmers:  the effects of an emerging market in Middle Saxon England. In Barely Surviving or More than Enough? The Environmental Archaeology of Subsistence, Specialization and Surplus Food Production, edited by M. Groot, D. Lentjes, and J. Zeiler, pp. 247–78. Leiden: Sidestone Press. 2014. Animals in Saxon and Scandinavian England: Backbones of Economy and Society. Leiden: Sidestone Press. Holt, H., 2009. The urban transformation in England. Anglo-Norman Studies 32, 57–78. Holt, J. Z., 2009. Rethinking the Ramey State: was Cahokia the center of a theater state? American Antiquity 74 (2), 231–54. Hope-Taylor, B., 1977. Yeavering: An Anglo-British Centre of Early Northumbria. DoE Archaeological Report 7. London: HMSO. Hunter, M., 1974. Germanic and Roman antiquity and the sense of the past in Anglo-Saxon England. Anglo-Saxon England 3, 29–50. Hunter-Mann, K., 2000. Lowthorpe Beck, Kilham, East Yorkshire:  Report on the Training Excavation, 2000. York: York Archaeological Trust.

Bibliography

205

2001. Lowthorpe Beck, Kilham, East Yorkshire: Report on the Training Excavation (Second Season), 2001. York: York Archaeological Trust. Hurst, J. J. and West, S. E., 1957. Saxo-Norman pottery in East Anglia. Part II: Thetford Ware with an account of Middle Saxon Ipswich Ware. Proceedings of the Cambridge Antiquarian Society 50, 29–60. Jervis, B., 2009. Pottery from Late Saxon Chichester: a reassessment of the evidence. Sussex Archaeological Collections 147, 61–76. Jones, A., 1990. Archaeological reconstruction and education at the Yorvik Viking Centre and Archaeological Research Centre, UK. In Constructed Past:  Archaeology, Education and the Public, edited by P. Planet and P. G. Stone, pp. 258–68. London: Routledge. Jones, Michael E., 1998. The End of Roman Britain. Ithaca, NY:  Cornell University Press. Jones, M. J., 1994. St. Paul in the Bail, Lincoln: Britain in Europe? In Churches Built in Ancient Times: Recent Studies in Early Christian Archaeology, edited by K. S. Painter, pp. 325–47. London: Society of Antiquaries. Jones, S. R. H., 1993. Transaction costs, institutional change, and the emergence of a market economy in later Anglo-Saxon England. The Economic History Review, New Series 46 (4), 658–78. Kemp, B., 1996. Anglian Settlement at 46–52 Fishergate. The Archaeology of York 7/1. London: Council for British Archaeology. Kemp, R. L., 1991. The archaeology of 46–54 Fishergate. In Bones from 46–54 Fishergate, by T.  P. O’Connor, pp. 211–20. The Archaeology of York 15/4. London: Council for British Archaeology. Kent Historic Towns Survey, 2004a. Fordwich: Archaeological Assessment Document. Maidstone: English Heritage and Kent County Council. 2004b. Sandwich:  Archaeological Assessment Document. Maidstone:  English Heritage and Kent County Council. Kharkov, C., ed., 1999. The Archaeology of Anglo-Saxon England: Basic Readings. New York: Garland. Kirch, P., 1989. The Evolution of the Polynesian Chiefdoms. Cambridge: Cambridge University Press. 2010. How Chiefs Became Kings: Divine Kingship and the Rise of Archaic States in Ancient Hawaiʽi. Berkeley: University of California Press. 2012. A Shark Going Inland Is My Chief. Berkeley: University of California Press. Lambert, T., n.d. A brief history of Winchester, Hampshire. Available online at www.localhistories.org/winchester.html. Lane, A., 2014. Wroxeter and the end of Roman Britain. Antiquity 88 (340), 501–15. Leahy, K. and Bland, R., 2009. The Staffordshire Hoard. London:  British Museum Press. Lederman, R., 1986. What Gifts Engender: Social Relations and Politics in Mendi, Highland Papua New Guinea. Cambridge: Cambridge University Press. Lee, R. B., 1968. What hunters do for a living, or, how to make out on scarce resources. In Man the Hunter, edited by I. DeVore and R. B. Lee, pp. 30–48. Chicago: Aldine. 1979. The!Kung San: Men,Women and Work in a Foraging Society. Cambridge: Cambridge University Press.

206

Bibliography

Leeds, E. T., 1923. A Saxon village at Sutton Courtenay, Berkshire. Archaeologia 73, 147–92. 1926–7. A Saxon village at Sutton Courtenay, Berkshire:  second report. Archaeologia 76, 59–80. 1947. A Saxon village at Sutton Courtenay, Berkshire: third report. Archaeologia 112, 79–94. Leslie, S., Winney, B., Hellenthal, G., Davison, D., Boumertit, A., Day, T., Hutnik, K., Royrvik, E., Cunliffe, B., Welcome Trust Case Control Consortium, International Multiple Sclerosis Consortium, Lawson, D. J., Falush, D., Freeman, C., Pirinen, M., Myers, S., Robinson, M., Donnelly, P., and Bodmer, W., 2015. The fine-scale genetic structure of the British population. Nature 519, 309–14. Lesure, R. G., 2004. Shared art styles and long-distance contact in early Mesoamerica. In Mesoamerican Archaeology, edited by Julia A. Hendon and Rosemary A. Joyce, pp. 73–96. Oxford: Blackwell. Loveluck, C., 2007. Rural Settlement, Lifestyles and Social Change in the Later First Millennium AD: Anglo-Saxon Flixborough in ItsWider Context. Oxford: Oxbow. Loveluck, C. and Tys, D., 2006. Coastal societies, exchange and identity along the Channel and southern North Sea shores of Europe, AD 600–1000. Journal of Maritime Archaeology 1, 140–69. Loveluck, C. and Atkinson, D., 2007. The Early Medieval Settlement Remains from Flixborough, Lincolnshire:  The Occupation Sequence, c.  AD 600–1000. Oxford: Oxbow. Lucy, S., 2000. The Anglo-Saxon Way of Death:  Burial Rites in Early England. Stroud: Sutton. Lucy, S., Tipper, J., and Dickens, A., 2009. The Anglo-Saxon Settlement at Bloodmoor Hill Carlton Colville, Suffolk. East Anglian Archaeology 131. Cambridge: Cambridge Archaeological Unit. Macphail, R. I., 1998. A reply to Carter and Davidson’s “An evaluation of the contribution of soil micromorphology to the study of ancient arable agriculture.” Geoarchaeology 13, 549–64. 2015. Modelling European ancient settlements  – their composition and morphology:  the state of the science employing contributions from soil micromorphology and associated geoarchaeological techniques. Paper presented the Conference on the Environmental Archaeology of European Cities, Brussels, Belgium, 27–29 May 2015. Macphail, R. I., Galinié, H., and Verhaeghe, F., 2003. A future for dark earth? Antiquity 77(296), 349–58. Maddicott, J. R., 2005. London and Droitwich, c. 650–750: trade, industry and the rise of Mercia. Anglo-Saxon England 34, 7–58. Mainman, A. J., 1990. Anglo-Scandinavian Pottery from Coppergate. The Archaeology of York 16/5. London:  Society of Antiquaries for the York Archaeological Trust. 1993. Pottery from 46–54 Fishergate. The Archaeology of  York 16/6. York: Society of Antiquaries for the York Archaeological Trust. Malcolm, G. and Bowsher, D., with Cowie R., 2003. Middle Saxon London: Excavations at the Royal Opera House 1989–99. London: Museum of London Archaeology Service Monograph 15.

Bibliography

207

Malin-Boyce, S., 2004. Manching. In Ancient Europe 8000 BC–AD 1000: Encyclopedia of the Barbarian World, edited by P. Bogucki and P. Crabtree, pp. 158–9. New York: Charles Scribner’s Sons. Maltby, M., 1979. Faunal Studies on Urban Sites: The Animal Bones from Exeter 1971–1975. Exeter Archaeological Reports,Volume 2. Sheffield: Department of Prehistory and Archaeology, University of Sheffield. 2010. Feeding a Roman Town:  Environmental Evidence from Excavations in Winchester, 1972–1985. Winchester: Winchester Museums. Manco, J., 1998. Saxon Bath: the legacy of Rome and the Saxon rebirth. Bath History 7, 27–55. Marsden, P. and West, B., 1992. Population change in Roman London. Britannia 23, 133–40. Marshall, A. and Marshall, G., 1991. A survey and analysis of the buildings of Early and Middle Anglo-Saxon England. Medieval Archaeology 35, 29–43. 1993. Differentiation, change and continuity in Anglo-Saxon buildings. Archaeological Journal 150, 366–402. McCormick, M., 2002. Origins of the European Economy:  Communications and Commerce AD 300–900. Cambridge: Cambridge University Press. 2010. A small town experiences the fall of the Roman Empire: (2) Thinking about Tarquimpol’s new answers to an old question. Paper presented at the 75th meeting of the Society for American Archaeology, St. Louis, MO, April 14–18. McIntosh, R. J., 2005. Ancient Middle Niger:  Urbanism and the Self-Organizing Past. Cambridge: Cambridge University Press. McIntosh, S. K., 1999. Modeling political organization in large-scale settlement clusters: a case study from the inland Niger Delta. In Beyond Chiefdoms: Pathways to Complexity in Africa, edited by S. K. McIntosh, pp.  66–79. Cambridge: Cambridge University Press. Millett, J. with James, S., 1983. Excavations at Cowdery’s Down, Basingstoke, Hampshire, 1978–81. Archaeological Journal 140, 151–279. Milne, G., 1985. The Port of Roman London. London: Batsford. Milne, G. and Goodburn, D., 1990. The early medieval port of London AD 700– 1200. Antiquity 64, 629–36. Minter, F., Plouviez, J., and Scull, C., 2014. Where kings lived:  Rendlesham rediscovered. British Archaeology 137, 50–5. Monroe, J. C., 2011. Urbanism on West Africa’s slave coast. American Scientist 99, 400–19. Monson, A. and Scheidel, W., 2015. Studying fiscal regimes. In Fiscal Regimes and the Political Economy of Premodern States, edited by A. Monson and W. Scheidel, pp. 3–27. Cambridge: Cambridge University Press. Moreland, J., 2000a. Concepts of early medieval economy. In The Long Eighth Century:  Production, Distribution and Demand, edited by I. L. Hansen and C. Wickham, pp. 1–34. Brill: Leiden. 2000b. The significance of production in eighth-century England. In The Long Eighth Century: Production, Distribution and Demand, edited by I. L. Hansen and C. Wickham, pp. 69–104. Brill: Leiden. 2001. Emporia. In Medieval Archaeology:  An Encyclopedia, edited by P. J. Crabtree, pp. 92–7. New York: Garland.

208

Bibliography

Morton, A. D., ed., 1992. Excavations at Hamwic, Volume 1: Excavations 1946– 83, Excluding Six Dials and Melbourne Street. London:  Council for British Archaeology Research Report No. 84. Munby, J., 1984. Saxon Chichester and its predecessors. In Anglo-Saxon Towns in Southern England, edited by J. Haslam, pp. 315–33. Chichester: Phillimore. Murphy, P. 1994. The Anglo-Saxon landscape and rural economy: some results from sites in East Anglia and Essex. In Environment and Economy in AngloSaxon England, edited by J. Rackham, pp. 23–39. London:  Council for British Archaeology Research Report 89. Naismith, R., 2012. Money and Power in Anglo-Saxon England:  The Southern English Kingdoms 757–865. Cambridge: Cambridge University Press. 2013. Coinage in pre-Viking East Anglia. In East Anglia and Its North Sea World in the Middle Ages, edited by D. Bates and R. Liddiard, pp. 137–51. Woodbridge: Boydell. Neal, D. S., 2003. Building 2, Insula XXVII from Verulamium: a reinterpretation of the evidence. In The Archaeology of Roman Towns: Studies in Honour of John S. Wacher, edited by P. Wilson, pp. 195–202. Oxford: Oxbow. Nicholls, M., 2013. Radiocarbon dating. In At the Limits of Lundenwic: Excavations in the North-West of Middle Saxon London at St Martin’s Courtyard, 2006–9, by L. Fowler and R. Taylor, p. 97. London:  Museum of London Archaeology, Archaeological Studies 27. North, D. C., 1993. Nobel Prize Lecture by Douglas C.  North, December 9, 1993. 2005. Institutions and the performance of economies over time. In Handbook of New Institutional Economics, edited by C. Ménard and M. M. Sherley, pp. 21–30. New York: Springer. O’Connor, T. P., 1989. Bones from the Anglo-Scandinavian Levels at 16–22 Coppergate. The Archaeology of York 15/3. York:  Council for British Archaeology for the York Archaeological Trust. 1991. Bones from 46–54 Fishergate. The Archaeology of York 15/4.York: Council for British Archaeology for the York Archaeological Trust. 2001. On the interpretation of animal bone assemblages from wics. In Wics: The Early Medieval Trading Centres of Northern Europe, edited by D. Hill and D. Cowie, pp. 54–60. Sheffield: Sheffield Academic Press. 2010. Livestock and deadstock in early medieval Europe from the North Sea to the Baltic. Environmental Archaeology 15, 1–15. O’Leary, T. J., 1981. Excavations at Upper Borough Walls, Bath, 1980. Medieval Archaeology 25, 1–31. Oosthuizen, S., 2011. Archaeology, common rights and the origins of AngloSaxon identity. Early Medieval Europe 19 (2), 153–81. 2013. Tradition and Transformation in Anglo-Saxon England:  Archaeology, Common Rights and Landscape. London: Bloomsbury Academic. 2016. Recognizing and moving on from a failed paradigm: the case of agricultural landscapes in Anglo-Saxon England, c.  AD 400–800. Journal of Archaeological Research 24 (2), 179–227. 2017. The Anglo-Saxon Fenland. Oxford: Windgather. Orton, D. C., Morris, J., Locker, A., and Barrett, J. H., 2014. Fish for the city: meta-analysis of archaeological cod remains and the growth of London’s northern trade. Antiquity 88, 516–30.

Bibliography

209

Ottaway, P., 1996. Archaeology in British Towns from the Emperor Claudius to the Black Death. London: Routledge. 1999. York:  the study of a Late Roman colonia. In The Coloniae of Roman Britain: New Studies and a Review: Papers of the Conference Held at Gloucester on 5–6 July, 1997, edited by H. Hurst, pp. 136–50. Portsmouth, RI: Journal of Roman Archaeology Supplementary Series 36. Palliser, D. M., 2014. Medieval York 600–1540. Oxford: Oxford University Press. Pattison, J. E., 2008. Is it necessary to assume an apartheid-like social structure in Early Anglo-Saxon England? Proceedings of Royal Society B: Biological Sciences 275, 2423–9. Payne, S., 1973. Kill-off patterns in sheep and goats: the mandibles from Asçvan Kale. Anatolian Studies 23, 281–303. Perkins, D., Jr., 1964. Prehistoric fauna from Shanidar, Iraq. Science 144, 1565–6. Pestell, T., 2011. Markets, emporia, wics, and ‘productive’ sites: pre-Viking trade centres in Anglo-Saxon England. In the Oxford Handbook of Anglo-Saxon England, edited by H. Hamerow, D. A. Hinton, and S. Crawford, pp. 556– 79. Oxford: Oxford University Press. Pestell, T. and Ulmschneider, K., 2003. Markets in Early Medieval Europe: Trading and “Productive” Sites, 650–850. Macclesfield, UK: Windgather Press. Pitts, M., 2010. Re-thinking the southern British oppida: networks, kingdoms, and material culture. European Journal of Archaeology 13 (1), 32–63. Pirenne, H., 1925 (1969). Medieval Cities: Their Origins and the Revival of Trade. Princeton, NJ: Princeton University Press. 1937 (1939). Mohammed and Charlemagne, translated from the French by Bernard Miall. New York: Barnes and Noble. Plouviez, J., 1985. The late Romano-British pottery. In West Stow:  The AngloSaxon Village, Volume 1, by S. E. West, pp. 82–5. East Anglian Archaeology 24. Ipswich: Suffolk County Planning Department. 2014. The Rendlesham Survey: investigating a royal Anglo-Saxon landscape. Paper presented at the Castle Museum, Ipswich, December 6, 2014. Polanyi, K., 1957. The economy as instituted process. In Trade and Market in Early Empires, edited by K. Polanyi, C. M. Arensburg, and H. W. Pearson, pp. 243–69. Glencoe, IL: The Free Press. Pollock, S., 1999. Ancient Mesopotamia. Cambridge: Cambridge University Press. Powlesland, D., ed., 1998. The West Heslerton Assessment. Internet Archaeology 5. http://dx.doi.org/10.11141/ia.5.4. 1999. The Anglo-Saxon settlement at West Heslerton, North Yorkshire. In Northumbria’s Golden Age, edited by J. Hawkes and S. Mills, pp. 55–65. Stroud: Sutton. 2000. West Heslerton settlement mobility:  a case study in static development. In Early Diera: Archaeological Studies of the East Riding in the Fourth to Ninth Centuries AD, edited by H. Geake and J. Kenny, pp. 19–26. Oxford: Oxbow. Powlesland, D., Haughton, C. A., and Hanson, J, H., 1986. Excavations at Heslerton, North Yorkshire, 1978–1982. Archaeological Journal 143, 53–173. Prummel, W., 1983. Excavations at Dorestad 2:  Early Medieval Dorestad, an Archaeozoological Study. Amersfoort: ROB. Qualmann, K. E., Rees, H., and Scobie, G. D., 2010. Introduction to Roman Winchester. In Feeding a RomanTown: Environmental Evidence from Excavations

210

Bibliography

in Winchester, 1972–1985, by M. Maltby, pp. 4–9. Winchester:  Winchester Museums. Rackham, J., 2003. Animal bones. In An Early to Middle Saxon settlement at Quarrington, Lincolnshire (ed. G. Taylor). Antiquaries Journal 83, 258–73. Rahtz, P., 1962–3. The Saxon and medieval palaces at Cheddar, Somerset: an interim report of the excavations in 1960–62. Medieval Archaeology 6–7, 53–66. Rahtz, P., Anderson, F. W., and Hirst, S. M., 1979. The Saxon and Medieval Palaces at Cheddar:  Excavations 1960–62. Oxford:  British Archaeological Reports, British Series 65. Rahtz, P., Woodward, A., Burrow, I., Everton, A., Watts, L., Leach, P., Hirst, S., Fowler, P., and Gardner, K., 1992. Cadbury Congresbury 1968–73. Oxford: British Archaeological Reports, British Series 223. Randsborg, K., 1980. The Viking Age in Denmark:  The Formation of a State. London: St. Martin’s Press. Redman, C. L., 1978. The Rise of Civilization: From Early Farmers to Urban Society in the Ancient Near East. San Francisco: W. H. Freeman and Company. Reece, R., 1995. Site-finds in Roman Britain. Britannia 26, 179–206. Renfrew, C., 1973. Before Civilization: The Radiocarbon Revolution and Prehistoric Europe. London: Cape. 1987. Archaeology and Language:  The Puzzle of Indo-European Origins. Cambridge: Cambridge University Press. Reynolds, A., 2005. On farmers, traders and kings: archaeological reflections of social complexity in early medieval north western Europe. Early Medieval Europe 13, 97–118. 2009. The landscape archaeology of secular power in 8th–11th century England. Territorio, Sociedad y Poder 2, 67–88. Reynolds, S., 1977. An Introduction to the History of English Medieval Towns. Oxford: Clarendon. Richards, J. D., Naylor, J., and Holas-Clark, C., 2009. Anglo-Saxon landscape and economy: using portable antiquities to study Anglo-Saxon and Viking Age England. Internet Archaeology 25, http://intarch.ac.uk/journal/issue25/ richards_index.html. Riddler, I., 2001. The spatial organization of bone working at Hamwic. In Wics:  The Early Medieval Trading Centres of Northern Europe, edited by D. Hill and D. Cowie, pp. 61–6. Sheffield: Sheffield Academic Press. in preparation. Ipswich, Bone and Antler. East Anglian Archaeology. Rizzetto, M., Crabtree, P., and Albarella, U., 2017. Livestock changes at the beginning and end of the Roman period in Britain: issues of acculturation, adaptation and ‘improvement’. European Journal of Archaeology 20 (3), 535–56. Rogers, A., 2011. Late Roman Towns in Britain: Rethinking Change and Decline. Cambridge: Cambridge University Press. Rogers, N. S.  H., 1993. Anglian and Other Finds from 46–54 Fishergate. The Archaeology of York 17/9. York:  Society of Antiquaries for the York Archaeological Trust. Rogers, P. W., 1997. Textile Production at 16–33 Coppergate. The Archaeology of York 17. York: Society of Antiquaries for the York Archaeological Trust.

Bibliography

211

Rogerson, A. and Dallas, C., 1984. Excavations at Thetford 1948–59 and 1973– 80. East Anglian Archaeology 22. Dereham: Norfolk Archaeological Unit, Norfolk Museums Service. Rollason, P., 2003. Northumbria, 500–1100: Creation and Destruction of a Kingdom. Cambridge: Cambridge University Press. Royal Commission on Historic Monuments, 1959. Wareham West Walls. Medieval Archaeology 3, 120–38. Rumble, A., 2001. Notes on the linguistic and onomastic characteristics of Old English wic. In Wics: The Early Medieval Trading Centres of Northern Europe, edited by D. Hill and R. Cowey, pp. 1–2. Sheffield: Sheffield Academic Press. Russel, A., 1984. Early Anglo-Saxon ceramics from East Anglia:  a microprovenience study. PhD thesis, University of Southampton. Russo, D. G., 1998. Town Origins and Development in Early England, c. 400–950 AD. Westport, CT: Greenwood Press. Sahlins, M., 1992. Anahulu: The Anthropology of History in the Kingdom of Hawaii, Volume 1: Historical Ethnography. Chicago: University of Chicago Press. Sawyer, P. H., 1998. From Roman Britain to Norman England, 2nd edition. London: Routledge. Scheidel, W., 2001. Roman age structure: evidence and models. Journal of Roman Studies 91, 1–26. Scull, C., 1991. Post-Roman Phase I  at Yeavering:  a reconsideration. Medieval Archaeology 35, 51–63. 1993. Archaeology, early Anglo-Saxon society and the origins of the Anglo-Saxon kingdoms. Anglo-Saxon Studies in Archaeology and History 6, 65–82. 1997. Urban centres in pre-Viking England? In The Anglo-Saxons from the Migration Period to the Eighth Century: An Ethnographic Perspective, edited by J. Hines, pp. 269–310. Woodbridge: Boydell. 2009. Early Medieval (Late 5th–Early 8th Centuries AD) Cemeteries at Boss Hall and Buttermarket, Ipswich, Suffolk. Leeds:  Society for Medieval Archaeology Monograph 27. 2013. Ipswich: contexts of funerary evidence from an urban precursor of the seventh century AD. In East Anglia and Its North Sea World in the Middle Ages, edited by D. Bates and R. Liddiard, pp. 218–29. Woodbridge: Boydell. Scull, C., Minter, F., and Plouviez, J., 2016. Social and economic complexity in early medieval England:  a central place complex of the East Anglian kingdom at Rendlesham, Suffolk. Antiquity 90 (354), 1594–612. Scull, C. and Wade, K., 2009. Buttermarket and the development of Ipswich. In Early Medieval (Late 5th–Early 8th Centuries AD) Cemeteries at Boss Hall and Buttermarket, Ipswich, Suffolk, by C. Scull, pp. 313–19. Leeds:  Society for Medieval Archaeology Monograph 27. Sear, D., Murdock, A., LeBas, T., Baggeley, P., and Gubbins, G., 2013. Dunwich, Suffolk:  mapping and assessing the inundated medieval town. Dunwich Project Final Report, available at www.dunwich.org.uk/resources/ documents/dunwich_12_report.pdf . Seetah, K., 2006. A multidisciplinary approach to Romano-British cattle butchery. In Integrating Zooarchaeology, edited by M. Maltby, pp. 109–16. Oxford: Oxbow.

212

Bibliography

2007. Butchery as an analytical tool:  a comparative study of the RomanoBritish and medieval periods. Unpublished PhD thesis, Cambridge University, Cambridge, UK. Semple, S., 2008. Politics and princes. Oxford Journal of Archaeology 27 (4), 407–29. Service, E. R., 1975. Origins of the State and Civilization: The Process of Cultural Evolution. New York: W. W. Norton and Co. Sherley-Price, L., tr., 1968. Bede’s History of the English Church and People. Baltimore, MD: Penguin. Silver, I. A., 1969. The ageing of the domestic mammals. In Science in Archaeology, edited by D. Brothwell and E. S. Higgs, pp. 283–302. London: Thames and Hudson. Soderberg, J., 2001. Feeding communities: monasteries and urban development in early medieval Ireland. In Shaping Community: The Art and Archaeology of Monasticism, edited by S. McNally, pp. 67–73. Oxford: British Archaeological Reports, International Series 941. Spall, C. A. and Toop, N. J., 2008. Before Eorforwic: new light on York in the 6th–7th centuries. Medieval Archaeology 52, 1–25. Speed, G., 2014. Towns in the Dark:  Urban Transformations from Late Roman Britain to Anglo-Saxon England. Oxford: Archaeopress. Stein, G., ed., 2005. The Archaeology of Colonial Encounters:  Comparative Perspectives. Santa Fe, NM: School of American Research Press. Stenton, F. M., 1971. Anglo-Saxon England, 3rd edition. Oxford:  Oxford University Press. Stoodley, N., 2002. The origins of Hamwic and its central role in the seventh century as revealed by recent archaeological discoveries. In Central Places in the Migration and Merovingian Periods, edited by B. Hårdh and L. Larsen, pp. 317–31. Uppsala: Almqvist and Wiksell International. 2011. Burial practice in seventh-century Hampshire:  St. Mary’s Stadium in context. In Burial in Later Anglo-Saxon England, 650–1100 AD, edited by J. Buckberry and A. K. Cherryson, pp. 38–53. Oxford: Oxbow. Storey, G. R., 1997. The population of ancient Rome. Antiquity 71, 966–78. Stuckert, C., ed., 2016. The People of Early Winchester. Winchester:  Winchester Excavation Committee. Sykes, N., 2010. Deer, land, knives and halls:  social change in early medieval England. Antiquaries Journal 90, 175–93. Tester, A., Anderson, S., Riddler, I., and Carr, R., 2014. Staunch Meadow, Brandon, Suffolk:  A High Status Middle Saxon Settlement on the Fen Edge. East Anglian Archaeology 151. Bury St. Edmunds: Suffolk County Council Archaeological Service.. Thomas, C., 1988. Tintagel Castle. Antiquity 62, 421–34. Thomas, G., 2013. Life before the minster:  the social dynamics of monastic foundation at Anglo-Saxon Lyminge, Kent. Antiquaries Journal 93, 109–45. Thomas, G., McDonnell, G., Merkel, J., and Marshall, P., 2016. Technology, ritual and Anglo-Saxon agrarian production:  the biography of a seventhcentury century plough coulter from Lyminge, Kent. Antiquity 90 (351), 742–58.

Bibliography

213

Thomas, M. G., Stumpf , M. P.  H., and Härke, H., 2006. Evidence for an apartheid-like social structure in early Anglo-Saxon England. Proceedings of Royal Society B: Biological Sciences 273, 2651–7. Thornton, C. P., Cable, C. M., and Possehl, G. L., eds., 2016. The Bronze Age Towers at Bat, Sultanate of Oman: Research by the Bat Archaeological Project 2007–12. Philadelphia: University Museum. Thurston, T. L., 2001. Landscapes of Power, Landscapes of Conflict: State Formation in the South Scandinavian Iron Age. New York: Springer Scientific. Tipper, J., 2000. Grubenhäuser:  pitfills and pitfalls. PhD thesis, Cambridge University. 2004. The Grubenhaus in Anglo-Saxon England: An Analysis and Interpretation of the Evidence from a Most Distinctive Building Type. Yeddingham: Landscape Research Centre Archaeological Monograph 2.1. Todd, M., 1999. Roman Britain, 3rd edition. Oxford: Blackwell. Townend, P., Willmott, H., Daubney, A., and Vickers, G., 2016. The mystery in the marsh: exploring the Anglo-Saxon island at Little Carleton. Current Archaeology 313, available online at www.archaeology.co.uk/articles/features/ the- mystery- in- the- marsh- exploring- an- anglo- saxon- island- at- littlecarlton.htm. Tringham, R., 1991. Households with faces: the challenge of gender in prehistoric architectural remains. In Engendering Archaeology:  Women and Prehistory, edited by J. M. Gero and M. W. Conkey, pp. 93–131. Oxford: Blackwell. Tweddle, D., Moulden, J., and Logan E., 1999. Anglian York:  A Survey of the Evidence. York: Archaeology of York 7/2, pp. 115–313. Ulmschneider, K., 2000. Settlement, economy and the ‘productive’ site: Middle Anglo-Saxon Lincolnshire, AD 650–780. Medieval Archaeology 54, 53–77. Ulmschneider, K. and Pestell, T., eds., 2003. Markets in Early Medieval Europe: Trading and “Productive” Sites. Macclesfield, UK: Windgather Press. Valante, M. A., 1998. Reassessing the Irish monastic town. Irish Historical Studies 31 (121), 1–18. Van Es, W. and Verwers, W. J. H., 1980a. Excavations at Dorestad 1, the Harbour: Hoogstraat. Amersfoort: ROB/RCE. 1980b. Excavations at Dorestad 3: Hoogstraat 0, II–IV. Amersfoort: ROB/RCE. Vince, A., 1984. The Aldwych:  Mid-Saxon London discovered. Current Archaeology 8 (4), 310–12. 1990. Saxon London: An Archaeological Investigation. London: Seaby. Von den Driesch, A., 1976. A Guide to the Measurement of Animal Bones from Archaeological Sites. Peabody Museum Bulletin 1. Harvard University. Von den Driesch, A., and Boessneck, J., 1974. Kritische Anmerkungen zur Widerristhöhenberechnung aus Längenmassen vor- und frühgeschichtlicher Tierknocken. Säugetierkunliche Mitteilungen 22, 325–48. Wacher, J., ed., 1966. Civitas Capitals of Roman Britain. Leicester:  Leicester University Press. 1995. The Towns of Roman Britain. London: Batsford. Wade, K., 1980. A settlement at Bonhunt farm, Wicken Bonhunt, Essex. In Archaeology in Essex to AD 1500, edited by D. G. Buckley, pp. 96–102. London: Council for British Archaeology Research Report 34.

214

Bibliography

1988. Ipswich. In The Rebirth of Towns in the West, AD 700–1050, edited by R.  Hodges and B. Hobley, pp. 93–100. London:  Council for British Archaeology Research Report 68. 1993. The urbanisation of East Anglia: the Ipswich perspective. In Flatlands and Wetlands: Current Themes in East Anglian Archaeology, edited by J. Gardiner, pp. 141–51. East Anglian Archaeology 50. Norwich:  Scole Archaeological Committee. 2000. Ipswich. In Medieval Archaeology: An Encyclopedia, edited by P. Crabtree, pp. 173–5. New York: Garland. 2012. The chronological framework [for Ipswich]. Available from the Archaeological Data Service at http://archaeologydataservice.ac.uk/ archiveDS/archiveDownload?t=arch-1644-1/dissemination/pdf/Metadata/ The_Chronological_Framework_by_Keith_Wade.pdf n.d. A history of archaeology in Ipswich and its Anglo-Saxon origins. Available online at http://ipswichat.org.uk/AboutUS/History.aspk. Wailes, B., 1964. Some imported pottery in western Britain, AD 400–800:  its connection with Frankish and Visigothic Gaul. Unpublished PhD thesis, Cambridge University. Wallace, P., 2004.Viking Dublin. In Ancient Europe 8000 BC–1000 AD: Encyclopedia of the Barbarian World, edited by P. Bogucki and P. J. Crabtree, pp. 466–8. New York: Charles Scribner’s Sons. 2015. Viking Dublin: The Wood Quay Excavations. Sallins: Irish Academic Press. Ward-Perkins, B., 2000. Why did the Anglo-Saxons not become more British? The English History Review 115, 513–33. Watson, P. J., LeBlanc, S. A., and Redman, C. L., 1971. Explanation in Archaeology: An Explicitly Scientific Approach. New York: Columbia University Press. Weale, M. E., Weiss, D. A., Jager, R. F., Bradman, N., and Thomas, M. G., 2002. Y chromosome evidence for Anglo-Saxon mass migration. Molecular Biology and Evolution 19, 1008–21. Webster, L., 2001. The rise, fall and resuscitation of the Taplow Burial. Resumé of a lecture given to the Sutton Hoo Society, October 28, 2001. Available online at www.suttonhoo.org/Saxon/Saxon_pdf/w_Saxon35.pdf . Wells, P. S., 1984. Farms,Villages and Cities: Commerce and Urban Origins in Late Prehistoric Europe. Ithaca, NY: Cornell University Press. ed., 1993. Settlement, Economy, and Cultural Change at the End of the European Iron Age:  Excavations at Kelheim in Bavaria, 1987–1991. Ann Arbor, MI: International Monographs in Prehistory. 2004. Kelheim. In Ancient Europe 8000 BC–AD 1000:  Encyclopedia of the Barbarian World, edited by P. Bogucki and P. Crabtree, pp. 247–9. New York: Charles Scribner’s Sons. West, B., 1989. Birds and mammals from the Peabody Site. Transactions of the London and Middlesex Archaeological Society 40, 151–67. West, S. E., 1963. Excavations at Cox Lane (1958) and the town defenses, Shire Hall Yard (1959). Proceedings of the Suffolk Institute of Archaeology 29, 233–303. 1969. The Anglo-Saxon village of West Stow:  an interim report on the excavations. Medieval Archaeology 23, 1–20.

Bibliography

215

1985. West Stow:  The Anglo-Saxon Village. East Anglian Archaeology 24. Ipswich: Suffolk County Planning Department. 2001. West Stow Revisited. Bury St. Edmunds:  West Stow County Park and Anglo-Saxon Village. Wheeler, R. E.  M., 1934. The topography of Saxon London. Antiquity 31, 290–302. White, R. and Barker, B., 1998. Wroxeter:  Life and Death of a Roman City. Stroud: Tempus. Whitlock, A., 2014. The archaeology of a medieval village:  Willingham. Paper presented at the Medieval Archaeology Conference, SUNY Cortland, October 8, 2014. Whytehead, R. I. and Cowie, R. L., with Blackmore, L., 1989. Excavations at the Peabody Site, Chandos Place and the National Gallery. Transactions of the London and Middlesex Archaeological Society 40, 35–176. Williams, G., 2013. The circulation, minting, and use of coins in East Anglia, c. AD 580–675. In East Anglia and Its North Sea World in the Middle Ages, edited by D. Bates and R. Liddiard, pp. 120–36. Woodbridge: Boydell. Williamson, T., 2009. Sutton Hoo and Its Landscape: The Context of Monuments. Oxford: Windgather Press. 2013. Environment, Society and Landscape in Early Medieval England: Time and Topography. Woodbridge: The Boydell Press. Wiltshire County Archaeological Services, 2004. The archaeology of Wiltshire’s towns: an extensive urban survey – Cricklade. Archaeological Data Service, available online at http://archaeologydataservice.ac.uk/archiveDS/archive Download?t=arch-906-1/dissemination/pdf/EUS_Texts/Cricklade.pdf . Winder, J. and Gerber-Parfitt, S. 2003. The oyster shells. In Middle Saxon London: Excavations at the Royal Opera House 1989–99, edited by G. Malcolm, D.  Bowsher, and R.  Cowie, pp.  325–32. London:  Museum of London Archaeological Studies 15. Winters, H. D., 1969. The Riverton Culture:  A Second Millennium Culture in the Central Wabash Valley. Illinois Archaeological Survey, Monograph No. 1. Springfield:  Illinois Archaeological Survey and the Illinois State Museum. Wood, I., 2013. The Modern Origins of the Early Middle Ages. Oxford:  Oxford University Press. Wright, R., 2010. The Ancient Indus: Urbanism, Economy, and Society. Cambridge: Cambridge University Press. 2013. Commodities and things:  the Kulli in context. In Connections and Complexity: New Approaches to the Archaeology of South Asia, edited by A.  S.  Abraham, P. Gullapalli, T. P. Rakzek, and U. Z. Rizvi, pp. 47–62. Walnut Creek, CA: Left Coast Press. Yamin, R., 2001. From tanning to tea: the evolution of a neighborhood. Historical Archaeology 35 (3), 5–15. Yoffee, N., 1995. Political economy of early Mesopotamian states. Annual Review of Anthropology 24, 281–311. Yorke, B., 1984. The bishops of Winchester, the kings of Wessex and the development of Winchester in the 9th and 10th centuries. Proceedings of the Hampshire Field Club and Archaeological Society 40, 61–70.

216

Bibliography

1989. The Jutes of Hampshire and Wight and the origins of Wessex. In The Origins of Anglo-Saxon Kingdoms:  Studies in the Early History of Britain, edited by S. Bassett, pp. 84–96. Leicester: Leicester University Press. 1990. Kings and Kingdoms in Early Anglo-Saxon England. London: Routledge. 1993. Fact or fiction? The written evidence for the fifth and sixth centuries AD. Anglo-Saxon Studies in Archaeology and History 6, 45–50. 2003. Anglo-Saxon gentes and regna. In Regna and Gentes:  The Relationship between Late Antique and Early Medieval Peoples and Kingdoms in the Transformation of the Roman World, edited by H.-W. Goetz, J. Jarnut, and W. Pohl, pp. 381–407. Leiden: Brill. Zeder, M., 1988. Understanding urban process through the study of specialized subsistence economy in the Near East. Journal of Anthropological Archaeology 7, 1–55. 1991. Feeding Cities:  Specialized Animal Economy in the Ancient Near East. Washington, DC: Smithsonian Institution Press.

Index

abandonment, 23, 25, 31, 35, 39, 40, 86, 90, 134–135 urban, 23, 31 acculturation, 53 acidic soils, 78 Ackerburgerstädt, 133 administrative centers, 13, 18, 29, 32, 40 agricultural intensification, 96, 110, 145, 176, 182–183, 184, 189–190 agricultural surpluses, 82, 110 agriculture, 5, 46, 61–62, 68, 110, 118, 133, 175, 176 Alfred, King, 138, 146, 153, 155–157, 160, 173, 174, 176, 182 amber, Baltic, 163, 166, 168 amphitheatres, 18, 42, 48–49 Anglian York, see Eorforwic Anglo-Saxon burial practices, 57, 75 Anglo-Saxon burials, 53, 57, 71 Anglo-Saxon Chronicle, 49–51, 155, 157–158, 171 Anglo-Saxon kingdoms, 16, 68, 81–82, 89, 129, 138, 179, 183, 190 Anglo-Saxon kings, 77, 89, 129, 136, 139, 175–177, 181–183, 187, 190 Anglo-Saxon settlement, 31, 52, 57–61, 64, 69–71, 74, 90, 118, 127 Anglo-Saxon society, 65, 68, 71, 81–82, 139, 171, 177, 182, 187–188 Anglo-Saxon towns, 32, 135, 136, 141, 148, 184–185, 191 Anglo-Saxon urbanism, 9, 139, 162, 191–192

Anglo-Saxons, 1–4, 52–53, 73, 98–99, 111–112, 163–164, 172, 188, 190 definition, 1–3 urban origins, 3–5 Anglo-Scandinavian towns, 171–172, 182, 192 animal bones, 10, 31, 39, 96, 104, 112–113, 118, 120–121, 164 animal husbandry, 5, 48, 66, 68, 133, 175, 176, 189 animals, 7, 10, 66, 104, 113, 127, 133 anthropologists, 6, 8–9, 184 antlers, 105, 121, 122, 163, 181 architecture, 22, 67, 71, 73 monumental, 5, 7, 191 art history, 4, 141 artifacts, 35, 67, 73, 79f. 2.10., 118, 122, 125 artifactual evidence, 31, 35, 39, 111, 120, 163 Æthelstan, 138 Baltic amber, 163, 166, 168 Bath, 19, 36, 39, 40, 49, 140, 153, 154–155 baths, 18, 33, 39, 154, 179 battles, 4, 49, 51, 131, 138–139, 157 Bede, 4, 50, 52, 53, 122, 131 Berinsfield, 19, 22, 56 Biddle, Martin, 7, 27, 29–30, 122, 133, 141–142, 144, 146, 148 bishoprics, 35, 36, 42, 117, 121, 134, 144–145, 181

217

218

Index

bishops, 35, 36, 42, 51–52, 129, 159, 179 Blinkhorn, Paul, 15, 61, 86, 89, 102, 105, 191 Bloodmoor Hill, 58, 72–73, 90, 96 boats, 53, 106, 123, 159 bones, 105, 108, 111–112, 113, 118, 121, 131–132, 163, 181 animal, 10, 31, 39, 96, 104, 112–113, 118, 120–121, 164 boundary ditches, 64, 90, 93, 111, 118 Brandon, 65, 73, 87, 92f. 3.2., 171, 173, 176, 180 Breckland region, 57, 61, 171 brooches, 31, 68, 120, 134, 168 Brook Street site, 1, 28f. 1.4., 141–145 buildings, 30, 33–35, 36–38, 73–75, 91–95, 110, 144, 164, 168–169 post-and-wattle, 163, 164, 167 public, 18, 28, 32, 43, 46 timber, 22, 34, 35, 42, 61, 72, 75, 93, 106 Bull Wharf, 157–158 burghal system, 154–155, 156–157, 176 burhs, 136–137, 141, 145, 146, 148–152, 153–154, 155–157, 172–174, 176 fortified, 148, 153, 172 burial practices, 83, 182 Anglo-Saxon, 57, 75 burials, 16, 36, 71, 76–77, 78–80, 98–99, 162, 184 Anglo-Saxon, 53, 57, 71 Essex chamber, 80f. 2.11. graves, 16 ship, 76, 89 wealthy, 182, 186 butchery, 20, 64, 111–112, 145 Buttermarket, 15, 98, 102, 129 Cadbury, 19, 58 Canterbury, 19, 46, 88, 127–128, 178

capitals, civitas, 18–20, 27, 43–44, 56 Carolingian Empire, 111, 135, 138, 181, 183–184 cattle, 34, 61, 73, 104, 113, 120, 164–165, 168–169, 176 cemeteries, 53, 56–57, 59, 62–64, 69, 71, 76–79, 92–93, 98 cremation, 36–37, 68 extra-mural, 29 ceramics, see pottery charters, 89, 127, 133, 142, 156, 159 Cheapside, 158, 160 Chichester, 140, 153–155, 174 chiefdoms, 13, 188–189 Childe, Gordon, 6–7, 10, 13, 82, 183, 184, 185, 188, 192 Christianity, 18, 32, 82, 88–89, 121, 144 chronology, 1, 2, 21, 23, 35, 39, 56, 60, 89 churches, 36, 89, 90, 92–93, 95, 162–163, 178, 181, 189 Cirencester, 19, 28, 41–42, 48–49 cities, 5–6, 13–14, 25–27, 32, 40–43, 48–49, 160, 165, 185, see also towns medieval, 8, 12–13, 131, 191 Roman, 30–32, 36, 39, 44–46, 117, 121–122, 129, 158, 159–160 walled, 30, 160 city walls, 42, 49, 114, 160 civitas capitals, 18–20, 27, 43–44, 56 Cleary, Esmonde, 18, 22–23, 25, 30, 40, 47–48, 55–56 coinage, 18, 21, 55–56, 104, 111–112, 163–164, 169–170, 172–174, 179–181 coloniae, 18, 43–44, 117–118 commerce, 13, 23, 78, 88, 95, 158, 171, 176, see also trade commodities, 88, 95, 96–97, 102, 132, 184, 186, 187, 191 competition, 81–83, 130, 183, 189–190

Index complex societies, 6, 7–8, 9, 184, 188, 192 Congresbury, 42, 48, 57–58, 84 construction, 29, 30, 38, 91–92, 108, 147–148, 169, 182–183, 189 consumers, 113, 120 consumption, 93, 96, 132, 165 local, 64, 66, 73, 126 continent, 56, 57–60, 99, 103–104, 129–131, 135, 138, 170, 175 continuity, 30, 34, 47, 53–54, 55, 64–65, 66, 68, 82 control, 16–17, 83, 110, 173, 175, 177, 179, 188, 190 of trade, 81, 99, 110, 129, 137, 187 conversion, emergency, 172–173 Coppergate, 115, 120, 160–166, 168 Coptic bowls, 78–80 costs, transaction, 11, 156, 169, 172, 174 countryside, 5–6, 20, 47–48, 52, 112–113, 131, 156, 174, 191 surrounding, 6, 116, 163, 165, 168–169, 171–172, 174–175, 181, 191–192 Cowdery’s Down, 58, 73–74 craft activities, 101, 111–112, 118, 122–123, 128, 132–134, 163–164, 168, 170 craft production, 44, 46, 85–86, 120–121, 131–132, 134–135, 171, 177–178 craft specialization, 5, 18–20, 48, 65, 67, 168, 169, 180, 181 crafts, 64, 95, 104, 111, 120–121, 124–125, 131–133, 163–164, 171 cremation cemeteries, 36–37, 68 Cricklade, 140, 146–149, 153, 155 cult centers, 32, 40 culture, 2, 11, 50 material, 4, 15, 18, 21, 47–48, 53, 165–167 Roman, 48, 71, 121 Romano-British, 18–20

219 Danelaw, 170, 172, 173, 181, see also Vikings Dark Age Economics, 13, 14–15, 99, 137, 175, 177, 183 dark earth, 23, 29, 31, 118 dating, radiocarbon, 21, 35–36, 39, 56, 162, 185 de novo settlements, 98, 106, 121, 129 defenses, 27–29, 135, 136, 147–148, 152–153, 155–156, 168–169, 172–173, 176, see also ditches; walls; forts/fortifications defensive ditches, 126, 170 dendrochronology, 21–22, 56, 158, 162, 164 dense populations, 5, 7, 40, 158 depopulation, 40, 54 Detroit analogy, 23–27, 31, 40 development, 1–2, 12–13, 17, 135–137, 139, 154, 171, 173–174, 185–186 of urbanism, 3, 6, 8, 14, 17, 139, 145, 160, 178 diet, 10, 91, 104, 113, 165 Dinas Powys, 58, 84 dioceses, 50, 51, 55, 57 ditches, 60, 62–64, 67, 74, 78, 90, 146–150, 170, 171 backfilled, 150 boundary, 64, 90, 93, 111, 118 defensive, 126, 170 Dorestad, 112, 133 drinking horns, 76, 80 Dublin, 166–168 Dunwich, 128, 181 dwellings, 57–60, 65, 67, 164 post-and-wattle, 163, 164 Early Anglo-Saxon period, 24t. 1., 16, 50–86 archaeology of settlement, 57–71 chronological issues, 55–57 historical background, 50–55 settlement changes in late 6th and 7th centuries, 71–83

220

Index

Early Anglo-Saxon period (cont.) western Britain in 5th–7th centuries, 83–85 Early Anglo-Saxon sites, 56, 66, 68, 73 early medieval urbanism, 8, 14, 22, 64, 134, 192 earthworks, 29, 170 east Anglia, 89, 99, 102–104, 129, 131, 132, 134, 181, 190 Eastern Mediterranean, 78–79, 84, 185, 187 east–west streets, 30, 111, 142, 174, 179 ecclesiastical centers, 85, 90, 102, 121, 129, 132, 145, 153 economic changes, 1, 8, 17, 52, 121, 180, 182, 189 economy, 11, 46, 59, 67, 99, 105, 169, 173, 185 market, 13, 84, 120, 172 Edward the Elder, 138, 155, 170 Egypt, 1, 6, 9, 184 elites, 81, 83, 102, 187 emergency conversion, 172–173 emporia, 85–137, 187–188 archaeology, 97–133 end, 134–136 enclosures, 64, 67, 73, 145 ditched, 62 fenced, 73–74 Eorforwic, 86–87, 89, 91t. 3.1., 114–121, 128–129, 131, 135 Essex, 61–63, 73, 78, 89, 91t. 3.1., 96 estate centers, 86, 90–91, 96, 104, 127, 132, 180 rural, 71, 96 estates, 68–69, 73, 110, 145 royal, 79f. 2.10., 110, 114, 127 European continent, see continent evidence, 27–30, 34–40, 41–46, 65, 99–108, 110–111, 120–123, 131–132, 185–186 artifactual, 31, 35, 39, 111, 120, 163 historical, 3, 68, 113, 134, 155, 159

exchange, 11–12, 78, 95–96, 97, 136, 169–170, 172, 190, 192 gifts, 12, 67, 131, 172–173, 175, 184 market, 12, 13, 136, 172–173, 175 theory, 12 Exeter, 140, 148, 153–154, 155 extra-mural cemeteries, 29 fairs, 95, 105, 173, 177, 181 farmers, 48, 92, 96, 132, 133, 136, 179, 183, 189 farms, 14, 88, 112, 114, 126, 133, 177 faunal assemblages, 34, 91, 93–94, 104–105, 112, 113, 126, 128, 165–166 feasting, 79, 83, 94, 136 fish, 97, 102, 113, 120, 165 Fishergate, 115, 118–121, 135 Flixborough, 15, 73, 87, 91t. 3.1., 93–95, 136, 173, 180 food, 52, 104–105, 113–114, 123, 126, 131, 164, 174–175, 181 rents, 78, 83, 120, 136, 139, 181, 183, 189–190 Fordwich, 87, 91t. 3.1., 127–128 foreign traders, 8, 104, 111, 125, 191 foreshore, 157, 159 fortifications, 136–137, 139, 146–147, 152, 170, 174, 192 fortified burhs, 148, 153, 172 fortified places, 13, 136–137, 141, 146, 155–156, 173 fortified sites, 44, 84, 176, 186 fortresses, 18, 116–118, 121, 136–137, 157, 160 forts/fortifications, Roman, 118, 121, 129, 157, 160, 174, 181 France, 111, 120, 122, 138, 177, 183 gates, 46, 49f. 1.11., 92, 116, 152, 157 Gaul, 35, 42–44, 187 genealogies, 51, 189–190 geophysical survey, 148–150

Index gift exchange, 12, 67, 131, 172–173, 175, 184 Gildas, 4, 35, 46, 50–53, 84, 180 Gippeswic, 91t. 3.1., 98 glass, 36, 42, 84, 186 gold, 48, 80, 125, 130, 163–164 goods, 88, 104, 111, 129–131, 172, 175 GPR, 59, 69, 75–77, 81, 83, 98, 108–110, see ground penetrating radar graves; see also burials Great Viking Army, 3, 146, 160 grid plans, Roman, 28, 142 ground penetrating radar (GPR), 150–152 Grubenhäuser, 57, 62 halls, 59–60, 64, 65, 92 Hamwic, 86–87, 89, 91t. 3.1., 104–114, 126–127, 128–131, 133, 134, 156–157, see also Southampton handmade pottery, 99t. 3.2. harbors, 88, 97, see also ports Haslam, Jeremy, 90, 136–137, 141, 146–149, 157–158, 160 Hawaiʽi, 188–191 heptarchy, 16, 89 Heslington Hill, 118 hierarchy, 121, 128, 133, 180–181, 183 hillforts, 42, 48, 179, 186 historical contingency, 10, 135, 174, 192 historical evidence, 3, 68, 113, 134, 155, 159 historical sources, 2–3, 13, 21, 50–51, 86–88, 142 Hodges, Richard, 13–15, 16, 99, 102, 110, 111, 114, 133, 134–137 horns, 76, 80, 112, 131, 181, 187 house plots, 144, 167 houses, 21, 25, 38, 99–101, 104, 124, 131, 133, 164 town, 18, 28–29, 38

221 human agency, 10–11, 47 hunting, 5, 66, 68 husbandry, animal, 5, 48, 66, 68, 133, 175, 176, 189 imported pottery, 36, 84, 101, 104, 111, 120, 126 Indus, 8, 9, 185, 188 industrial activities, 29, 43, 97 industries, 15, 20, 23, 40, 56, 125, 131, 171 inhumations, 53, 59, 77, 108–110, see also burials instability, political, 47, 139, 160, 162, 164, 188 intensification, agricultural, 96, 110, 145, 176, 182–183, 184, 189–190 international trade, 111, 114, 122, 125, 159, 160, 164, 169, 170–172 Ipswich, 15, 86–88, 98–105, 114, 126–127, 128–131, 132–134, 168–169, 171 Ipswich Ware, 15, 60, 75, 86, 89–90, 99–102t. 3.2., 125–126, 132, 191 Late Anglo-Saxon, 168–169 markets, 104, 169 Ireland, 7, 39, 48, 54, 84, 166–167, 172, 183 Iron Age, 42–45, 68, 112, 179, 184, 186, 188 iron smithing, 125, 127 Itchen, River, 28, 106 Kelheim, 186 Kent, 75, 89, 91t. 3.1., 122, 127–128, 182 Kilham, 58, 65–66, 68 kilns, pottery, 134, 168 kingdoms, Anglo-Saxon, 16, 68, 81–82, 89, 129, 138, 179, 183, 190 kings, 3, 5, 13, 77, 83–84, 138–139, 156, 157, 177–179

222

Index

kings (cont.) Anglo-Saxon, 77, 89, 129, 136, 139, 175–177, 181–183, 187, 190 kingship peripatetic, 74, 180, 183 landscape, 16, 22–23, 43, 48–49, 54, 69, 174 archaeology, 6, 12, 191 Lankhills, 29 Late Anglo-Saxon period, 23t. 1., 17, 134–135, 138–177 background, 138–143 Ipswich, 168–169 London, 157–160 Norwich, 169–171 pattern of urban development, 171–177 Viking Dublin, 166–168 Winchester, 141–146 Yorvik, 160–166 Late Roman period, 20, 21–23, 30, 31, 36, 43–46, 55–56, 64–66 layout, 42, 64, 88, 93–95, 108, 111, 123, 128, 166–167 Lincoln, 19, 36–38, 40, 44, 178 Lincolnshire, 15, 66, 73, 91t. 3.1., 93 London, 15–16, 18–20, 27, 122–124, 126–127, 128, 129–131, 139–140, 157–158 archaeological evidence, 30–32 burh, 135, 157–158 Late Anglo-Saxon, 157–160 Middle Anglo-Saxon, 112, 122–123 riverside wall, 157–158 Roman, 15, 30–31, 32, 121 Royal Opera House, 123–126 St. Martin in the Fields, 32, 36, 124f. 3.12. Strand, 122–123 long-distance trade, 7, 65, 111, 122, 131, 177, 182, 184 Low Countries, 111, 120, 122, 130, 138, 177

Lundenwic, 86–87, 89, 91t. 3.1., 121–129, 131, 133, 135, 185 Lyminge, 58, 75, 132, 176, 182 magnetometry, 78, 150 market centers, 136, 155, 168 market economy, 13, 84, 120, 172 market exchange, 12, 13, 136, 172–173, 175 market towns, 18, 44, 105, 114, 122, 134, 139–141, 155 marketing, 110, 112, 114, 136–137, 157, 169 marketplaces, 12, 101, 111, 169, 182 markets, 11–12, 40–41, 104, 105, 131, 132–133, 155–156, 174, 179 periodic, 171, 173, 177, 181 regional, 17, 164, 166 material culture, 4, 15, 18, 21, 47–48, 53, 165–167 meat, 66, 102, 112–113, 120, 176 medieval towns, 5, 6, 7, 13–14, 18, 111–112, 132–133, 181, 185 medieval urbanism, 12–13, 105, 137 models, 12–15 Mediterranean, 7, 12, 46, 84, 176 Eastern, 78–79, 84, 185, 187 Melbourne Street, 108, 112–113 Mercia, 3, 89, 129, 136, 141, 156, 179, 183, 190 Mercian burhs, 136–137 Mesoamerica, 8–9, 13 Mesopotamia, 1, 5–6, 8, 184 metal, 95, 122, 163, 186 metalwork, 17, 76, 85 Middle Anglo-Saxon period, 64–65, 82–137, 169–174 background, 88–89 changes in settlement patterns, 89–97 dating, 89 migration, 2, 52, 54, 82 migrationist model, 53, 55 military service, 173, 176

Index minster churches, 92, 137, 141, 145–146 mints, 7, 76, 155–156, 169–170, 174 monasteries, 89, 95, 134, 138, 155, 177, 189 monumental architecture, 5, 7, 191 mosaics, 38, 42, 48 mounds, 76–77, 78, 99 Mucking, 58, 61–64, 65–66, 75, 90, 180 neighborhoods, 25, 88 networks, 17, 97, 127, 139, 175, 184 New Institutional Economics (NIE), 11–12, 156, 163, 182, 192 Norman Conquest, 1, 2–3, 147, 153, 160, 163, 169–170, 172, 173–174 North Sea, 5, 88, 90, 97, 99, 127, 177, 180, 184 Northumbria, 4, 16, 72, 89, 129, 164, 183 northwest Suffolk, 20, 57, 91 Norwich, 128, 140, 169–171, 172 occupation, 15, 20, 23, 27, 29, 35, 38–39, 147–148, 159 O’Connor, Terry, 88, 113, 118–120, 162, 164–165 open area excavations, 22, 27, 106–108, 110, 123, 131, 142, 159, 160 Opera House, 123–125, 131, 135 oppida, 18, 44, 186–187 opportunity, 54, 129, 162, 177, 187–188 organization, social, 50, 52, 57, 83 Ouse, River, 91, 116, 118, 160, 171 Oxfordshire, 22, 56, 146, 148 patrons, 82–83, 89 periodic markets, 171, 173, 177, 181 peripatetic kingship, 74, 180, 183 pigs, 34, 61, 68, 73, 75, 96, 113, 120, 164–165 Pirenne, Henri, 12–13

223 pits, 30, 72, 78, 93, 99, 108, 118, 120, 122 planned towns/settlements, 118, 141, 155, 175 plants, 5, 7, 10, 24, 96 political instability/turmoil, 47, 139, 160, 162, 164, 188 political power, 25, 32, 68, 80, 113, 179, 182 polyfocal settlement, 121, 169 population, 24, 26, 29, 31–32, 53–54, 82, 185, 186, 191–192 dense, 5, 7, 40, 158 sizes, 126, 130, 188 substantial, 5, 26, 44, 98, 175, 180 pork, 95, 96, 104 ports, 18, 30, 88, 123, 127–128, 129, 155, 158 post-and-wattle buildings, 163, 164, 167 post-holes, 22, 57, 61–62, 64, 73 post-Roman period, 3–4, 29, 31–33, 35, 41, 69–71, 88–89, 162, 179 pottery, 20, 21, 55–56, 93, 98–99, 102–103, 105, 125, 168 continental, 128, 131 first industrially produced, 89, 102 handmade, 99t. 3.2. imported, 36, 84, 101, 104, 111, 120, 126 Ipswich Ware, see Ipswich, Ware kilns, 134, 168 Thetford Ware, 99t. 3.2., 134, 168 power, 35, 48, 99, 129, 175, 177, 182, 184 political, 25, 32, 68, 80, 113, 179, 182 Powlesland, Dominic, 62–65, 66–68 prairie, 25–26, 31 pre-contact Hawaiʽi, 188, 190, 191 primary state formation, 9, 183–184, 188 Prittlewell, 58, 78–80 processual approach, 9–10, 12–13, 81, 136, 174, 192 processual-plus perspective, 9–12

224

Index

productive sites, 85, 90, 95–96, 97, 173, 177, 180, 187, 192 public buildings, 18, 28, 32, 43, 46 Quarrington, 58, 66, 68 Queenford Farm, 19, 22, 56 Queen’s Hotel, 165 radiocarbon dating, 21, 35–36, 39, 56, 162, 185 raids, Viking, 114, 134–135, 139, 167, 172, 175, 181, 183 Ramsbury, 87, 90–91, 95 raw materials, 113, 123, 131, 163, 168, 174–175, 176 rebirth of towns in Anglo-Saxon England, 128, 145, 162, 178–182 reconstruction, 34f. 1.6., 46, 49f. 1.11., 60, 77f. 2.9., 80f. 2.11. redevelopment, 24, 35, 157, 160 gradual, 159, 174 planned, 105 redistribution center, 126 institutional, 172 regional market towns, 105, 141 regional markets, 17, 164, 166 Rendlesham, 58, 77–79, 129 rents, food, 78, 83, 120, 136, 139, 181, 183, 189–190 repair, 38, 74, 142, 156, 157 rescue excavations, 9, 27, 30, 61, 93, 118, 139–141 residents, 26, 32, 44, 104, 112–114, 120, 123, 126, 133 rural, 32, 176 resistivity, 150–151 retinues, 84, 139, 180, 183, 189 revenue, 156, 157, 177, 181, 183 sources, 99, 175, 177, 183 Rhineland, 53, 103, 111, 120, 122, 126, 177 river terraces, 61, 186 rivers, 61, 102, 106, 136–137, 146, 154, 158, 167, 169–171

riverside wall, 157–158 roads, 18, 28, 118, 122, 123, 157–158 Roman, 30, 31, 102, 123, 157 Roman cities, 30–32, 36, 39, 44–46, 117, 121–122, 129, 158, 159–160 Roman culture, 48, 71, 121 Roman forts/fortifications, 118, 121, 129, 157, 160, 174, 181 Roman grid plans, 28, 142 Roman London, 15, 30–31, 32, 121 Roman period, 18–20, 26–27, 30, 41, 44, 53–54, 69–71, 176, 179–180 Early, 44 Late, 20, 21–23, 30, 31, 36, 43–46, 55–56, 64–66 Roman roads, 30, 31, 102, 123, 157 Roman street plans, 46–47, 141, 160, 174 Roman towns, 18–20, 22–23, 25, 31, 38–40, 43–47, 135, 153–154, 178–179 Roman urbanism, 16, 20, 22, 40, 48 end in Britain, 18–49 recent approaches, 43–46 Roman walls, 38, 142, 145, 154, 157, 159 Romano-British culture, 18–20 Romano-British towns, 22–23, 25–27, 33, 40 royal centers, 77, 84, 121, 129 royal estates, 79f. 2.10., 110, 114, 127 royal vills, 75, 102, 113–114, 129, 180, 189 rural settlements, 6, 54, 57, 65, 70–71, 126, 132, 175, 180 rural sites, 14–15, 46, 104, 136 salt, 96, 97, 102, 186, 187, 191 Sandtun, 87, 91t. 3.1., 128, 133 Sandwich, 87, 91t. 3.1., 127–128 Scandinavians, 3, 139, 163, 166 Scull, C., 15, 71, 78, 81–83, 98–101, 103–104 secondary state formation, 82, 184

Index settlement changes, 71–83, 86, 89–97, 178 density, 126, 132 hierarchy, 71–72, 81, 84 patterns, 55, 59, 61, 71, 75, 83–86, 88–89, 91, 165–166 settlements, 5–6, 16, 40–42, 61–64, 66–67, 69–73, 117–118, 169–171, 180 polyfocal, 121, 169 rural, 6, 54, 57, 65, 70–71, 126, 132, 175, 180 urban, 15, 65, 88, 118, 135, 155, 167, 174–175, see also cities; towns SFBs, see sunken-featured buildings, 46 sheep, 68, 73, 113, 120 shrines, 51, 64 Silchester, 19, 28, 36, 38–39, 43, 49 Six Dials, 108, 110–113, 114, 131 social changes, 10, 16, 52, 82, 121, 135, 139, 192 social organization, 50, 52, 57, 83 societies, 7, 8, 9–11, 67–68, 182, 185, 187 Anglo-Saxon, 65, 68, 71, 81–82, 139, 171, 177, 182, 187–188 complex, 6, 7–8, 9, 184, 188, 192 urban, 5, 7–8, 9, 48–49, 181, 185, 191 Somerset, 39, 42, 84, 139, 155, 179, 190 Southampton, 3, 88, 91t. 3.1., 107f. 3.7., see also Hamwic specialization, 96, 176 craft, 5, 18–20, 48, 65, 67, 168, 169, 180, 181 specialized production, 40, 73, 93–95, 104 St. Albans, 19, 44, 51 St. Martin in the Fields, 32, 36, 124f. 3.12. Staffordshire Hoard, 130, 190 state formation, 7–9, 13, 14–15, 82, 137, 183–184, 185, 188–189, 192

225 primary, 9, 183–184, 188 secondary, 82, 184 status, 43, 67, 83, 117, 127, 182, 184 stone walls, 29, 148–151 Strand, 122–123 stratigraphy, 10, 35, 39, 156 street plans, 5, 27–29, 108, 111, 131, 145, 148, 153, 156 Roman, 46–47, 141, 160, 174 sub-Roman period, 2, 34–35, 36, 40–42, 44–47, 49, 53, 64, 68 subsistence production, 172, 173 Suffolk, 56, 57–59, 65, 72–73, 79f. 2.10., 91–94, 98–100, 104, 105 sunken-featured buildings (SFBs), 46, 57–61, 65, 67, 72–73, 75, 132, 168–169, 170 surpluses, agricultural, 82, 110 surrounding countryside, 6, 116, 163, 165, 168–169, 171–172, 174–175, 181, 191–192 Sutton Courtenay, 57–58 Sutton Hoo, 8, 58, 76–79, 89, 99, 129, 139, 182 symbols, 48, 82–83 tanning, 102, 125–126, 145 Taplow, 58, 79 taxes, 7, 18, 55, 99, 102–104, 129, 175–176, 179, 183 temples, 28, 39 Teotihuacán, 185, 188, 191 textile production, 65, 111, 118, 121, 122, 124–125, 131, 163, 170–171 Thames, 62, 123, 129, 148, 152, 157–159 theoretical perspectives, 1, 6–9 Thetford, 140, 168–169, 171 Thetford Ware, 99t. 3.2., 134, 168 timber buildings, 22, 34, 35, 42, 61, 72, 75, 93, 106 tolls, 102, 127, 129, 156, 173 town houses, 18, 28–29, 38 town walls, 33, 134, 146–148, 174

226

Index

towns, 17–20, 22–29, 39–46, 104–105, 110–114, 132–135, 145–157, 164–168, 173–175 Anglo-Saxon, 32, 135, 136, 141, 148, 184–185, 191 Late, 138–177 Anglo-Scandinavian, 171–172, 182, 192 market, 18, 44, 105, 114, 122, 134, 139–141, 155 medieval, 5, 6, 7, 13–14, 18, 111–112, 132–133, 181, 185 planned, 118, 141, 155, 175 Roman, 18–20, 22–23, 25, 31, 38–40, 43–47, 135, 153–154, 178–179 Romano-British, 22–23, 25–27, 33, 40 trade, 11–14, 128–130, 132–133, 156–157, 170–172, 177–178, 183–185, 186–187, 191–192 centers of, 16, 48, 85, 90, 97, 110, 127, 129 control of, 81, 99, 110, 129, 137, 187 and exchange, 11, 17, 78, 96, 97, 136, 170, 187, 192 international, 111, 114, 122, 125, 159, 160, 164, 169, 170–172 long-distance, 7, 65, 111, 122, 131, 177, 182, 184 routes, 18, 174, 186, 191 taxation, 99, 110, 129, 175–176, 181, 183, 187 traders, 5, 11, 123, 135, 177, 182, 187, 192 foreign, 8, 104, 111, 125, 191 trading places, 97, 127, 133, 175, 180, 183 Tranmer House, 58, 77 transaction costs, 11, 156, 169, 172, 174 Tribal Hidage, 69–70, 89, 183 tribute, 7, 68, 69, 89, 137, 172, 183 tyrants, 35, 84

urban centers, 8, 9, 16, 18, 35–36, 166–168, 169, 174, 179 urban development, 3, 6, 8–9, 155, 156, 158, 171–172, 174, 175 urban life, 33, 36–38, 123, 153 urban origins, 4, 5, 9, 12–14, 17, 158, 178, 184 Anglo-Saxons, 3–5 and archaeology, 5–6 urban revolution, 6–7, 13, 166, 182–185, 188, 192 urban settlements, 15, 65, 88, 118, 135, 155, 167, 174–175 urban societies, 5, 7–8, 9, 48–49, 181, 185, 191 urbanism, 5, 7–9, 15–16, 18, 43, 183, 185, 187–188 Anglo-Saxon, 9, 139, 162, 191–192 development, 3, 6, 8, 14, 17, 139, 145, 160, 178 early medieval, 8, 14, 22, 64, 134, 192 later prehistoric Europe, 185–188 medieval, 12–13, 105, 137 Roman, see Roman urbanism Venta Belgarum, 19, 27–29, 44 Verulamium, 19, 36, 38, 44, 46–48, 51 Viking armies, 89, 134, 135, 138, 146, 157, 160, 162 Viking attacks, 126, 157 Viking Dublin, 140, 166–168 Viking raids, 114, 134–135, 139, 167, 172, 175, 181, 183 Viking York, see Yorvik Vikings, 134, 136–138, 155, 157, 160, 164, 166–168, 171, 172 villas, 22, 57, 180 vills, royal, 75, 102, 113–114, 129, 180, 189 Vince, Alan, 122, 156, 158, 159 Wade, Keith, 15, 93, 98–99, 104–105, 168–169 Wales, 4, 35, 42, 50, 54, 84, 95, 190

Index Wallingford, 140, 146, 148–153, 155 walls, 30, 31, 122, 134–135, 144, 146–148, 152–153, 158, 159 city, 42, 49, 114, 160 Roman, 38, 142, 145, 154, 157, 159 town, 33, 134, 146–148, 174 Wareham, 140, 146–148, 153, 155 warfare, 26, 31–32, 40, 69, 83, 183, 190 wealth, 13, 48, 67, 75, 82–83, 93, 160, 182, 184 weapons, 29, 68, 79, 89, 110 weaving, 64, 101, 104 wells, 14, 39, 44–45, 93, 104, 106–108, 122, 126, 186 Wells, Peter, 14, 44–45, 177, 186 Wensum, River, 169–170 Wessex, 89, 102, 110, 138, 139–141, 146, 156–157, 179, 183 West Heslerton, 58, 62–68, 71, 90, 180 West Stow, 4, 15, 31, 58–64, 65–68, 71, 74–75, 86, 90 West Sussex, 69, 154 Wicken Bonhunt, 73, 87, 91t. 3.1., 93–95, 96, 173, 176 wics, 16–17, 85–88, 91, 127–131, 132–134, 135–136, 174–175, 181, 191–192, see also emporia Wiltshire, 90–91, 146–148

227 Winchester, 18–20, 27–30, 44–43, 114–115, 141–146, 156–157, 174, 179 Brook Street site, 1, 28f. 1.4., 141–145 wool, 66, 93–95, 96–97, 104, 111, 112–113, 131, 176, 181 Worchester, 140, 156, 160 workshops, 60, 65, 93, 110, 154, 170 Wroxeter, 16, 19, 22, 27, 32–36, 38, 40–43, 47–49, 178–179 Yeavering, 58, 71–72, 74–75, 86 York, 97–98, 114–119, 121, 127, 129–130, 135, 160–168, 183, 185 Anglian, see Eorforwic Anglo-Scandinavian, see Yorvik Coppergate, 115, 120, 160–166, 168 Fishergate, 115, 118–121, 135 Queen’s Hotel site, 165 York Archaeological Trust, 115, 118–119, 161f. 4.8. Yorke, Barbara, 51, 69–70, 81–83, 89, 114, 190 Yorvik, 17, 134, 139–140 Late Anglo-Saxon period, 160–166 zoning, 5, 64–65, 180, 192 zooarchaeology, 3, 10, 104, 110, 112, 131, 182