The Archaeology of Medieval Villages Currently Inhabited in Europe 1789693004, 9781789693003

The study of deserted villages abandoned during the last millennium in Europe has been the primary focus of archaeologic

301 81 20MB

English Pages 132 [131] Year 2019

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

The Archaeology of Medieval Villages Currently Inhabited in Europe
 1789693004, 9781789693003

Table of contents :
Cover
Title Page
Copyright Information
Author Biographies
Preface
List of Figures and Tables
Introduction
Test pit excavation as a method for reconstructing the development of currently occupied rural settlements: evidence from England
Carenza Lewis
Village Archaeology in France. A twenty-five year retrospective
Edith Peytremann
Investigating medieval village formation in the Netherlands
J.P.W. Verspay, H. Renes, B. Groenewoudt, J. van Doesburg
Is this a village? Approaching nucleated settlements in Scandinavian contexts
Ingvild Øye
The Archaeology of Currently Inhabited Villages in Spain:The Case of Asturias
Margarita Fernández Mier
Jesús Fernández Fernández
Conclusion
Chris Wickham

Citation preview

The Archaeology of Medieval Villages Currently Inhabited in Europe edited by

Jesús Fernández Fernández Margarita Fernández Mier

The Archaeology of Medieval Villages Currently Inhabited in Europe edited by

Jesús Fernández Fernández Margarita Fernández Mier

Archaeopress Archaeology

Archaeopress Publishing Ltd Summertown Pavilion 18-24 Middle Way Summertown Oxford OX2 7LG www.archaeopress.com

ISBN 978-1-78969-300-3 ISBN 978-1-78969-301-0 (e-Pdf)

© the individual authors and Archaeopress 2019 Cover images: Village of Villanueva de Santu Adrianu, mentioned in IXct. documents. View from the north. Drawing and reconstruction of the buildings in the nucleated tun at the farm Seim, situated close to the fjord in Vik in the Sognefjord, as it appeared around 1870. Uncertain buildings drawn in dotted lines. (Drawing: A. Berg 1952).

All rights reserved. No part of this book may be reproduced, or transmitted, in any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior written permission of the copyright owners.

This book is available direct from Archaeopress or from our website www.archaeopress.com

Contents List of Figures and Tables����������������������������������������������������������������������������������������������������������������iii Author Biographies��������������������������������������������������������������������������������������������������������������������������� v Preface�������������������������������������������������������������������������������������������������������������������������������������������� vii Introduction�������������������������������������������������������������������������������������������������������������������������������������� 1 Test pit excavation as a method for reconstructing the development of currently-occupied rural settlements: Evidence from England ��������������������������������������������������������������������������������� 7 Carenza Lewis Village Archaeology in France. A twenty-five year retrospective��������������������������������������������������35 Edith Peytremann Investigating medieval village formation in the Netherlands��������������������������������������������������������53 J.P.W. Verspay, H. Renes, B. Groenewoudt, J. van Doesburg Is this a village? Approaching nucleated settlements in Scandinavian contexts���������������������������73 Ingvild Øye The Archaeology of Currently Inhabited Villages in Spain: The Case of Asturias��������������������������91 Margarita Fernández Mier, Jesús Fernández Fernández Conclusion��������������������������������������������������������������������������������������������������������������������������������������117 Chris Wickham

i

ii

List of Figures and Tables Test pit excavation as a method for reconstructing the development of currently occupied rural settlements: evidence from England Figure 1. A 1m square test pit under excavation in Pirton (Hertfordshire) in 2010.�������������������������������������������������������11 Figure 2. Eastern England, showing the location of parishes included in the East Anglia CORS project between 2005 and 2017. ����������������������������������������������������������������������������������������������������������������������������������������������������������������������13 Figure 3. Pirton as depicted on the 6” to 1 mile Ordnance Survey map c. 1880, with the three main zones of settlement highlighted (Ordnance Survey map copyright University of Edinburgh/Digimap, reproduced with permission).������������������������������������������������������������������������������������������������������������������������������������������������������������������14 Figure 4. Test pits excavated at Pirton showing test pits producing Romano-British pottery (black/grey circles), test pits which did not produce Romano-British pottery (white squares) and names of streets and farms mentioned in text.����������������������������������������������������������������������������������������������������������������������������������������������������������������15 Figure 5. Test pits excavated at Pirton showing test pits producing early/middle Anglo-Saxon pottery (black/grey circles) and test pits which did not produce early/middle Anglo-Saxon pottery pottery (white squares). �������������17 Figure 6. Test pits excavated at Pirton showing test pits producing late Anglo-Saxon pottery (black/grey circles) and test pits which did not produce late Anglo-Saxon pottery pottery (white squares). ���������������������������������������17 Figure 7. Test pits excavated at Pirton showing test pits producing high medieval pottery (black/grey circles) and test pits which did not produce high medieval pottery (white squares). �����������������������������������������������������������19 Figure 8. Test pits excavated at Pirton showing test pits producing late medieval pottery (black/grey circles) and test pits which did not produce late medieval pottery (white squares). ������������������������������������������������������������19 Figure 9. Test pits excavated at Pirton showing test pits producing post-medieval pottery (black/grey circles) and test pits which did not produce post-medieval pottery (white squares). �����������������������������������������������������������20 Figure 10. CORS discussed in text showing percentage of habitative pits per historic period against average for all East Anglian CORS included on Table 1.����������������������������������������������������������������������������������������������������������������������24 Table 1. Test pits excavated within CORS in East Anglia by May 2017, listed alphabetically by parish name, showing the number and percentage of excavated pits per parish producing possibly habitative numbers of pottery sherds of Roman, Anglo-Saxon, medieval and post-medieval date���������������������������������������������������������������26

Village Archaeology in France. A twenty-five year retrospective Figure 1. Location of sites mentioned in the article��������������������������������������������������������������������������������������������������������������36 Figure 2. French functioning preventive archaeology���������������������������������������������������������������������������������������������������������39 Figure 3. Location of test pits areas on the 19th cadastre of Bonnée (Loiret), © S. Joly��������������������������������������������������41 Figure 4. Vic (Indre): location of excavation areas since 1998 on the current cadastre plan����������������������������������������42 Figure 5. Location of test pits and excavations on the 19th Cadastre of Tremblay-en-France (Seine-Saint-Denis)��44 Figure 6. Vauchrétien (Maine-et-Loire): location of the test pits on the current cadastre plan�����������������������������������45

Investigating medieval village formation in the Netherlands Figure 1. Cycle of archaeological heritage management (simplified).�������������������������������������������������������������������������������54 Figure 2. Overview of all the initial archaeological sites (white = urban; black = non-urban).�������������������������������������58 Figure 3. Overview of all the archaeological sites that are considered relevant by the set criteria.����������������������������58 Table 1. Overview of archaeological periods according to the Archeologisch Basisregister (ABR) used in this article.�������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������59 Figure 4. Overview of relevant sites by site type. Where multiple types are present the least generic is displayed.59 Figure 5. The number of sites per archaeological period during the time frame 750-1650 from a total number of 190 sites.���������������������������������������������������������������������������������������������������������������������������������������������������������������������������������60 Figure 6. The habitation period of multi-phased sites measured in the number of ABR-periods they cover�������������60 Figure 7. The archaeological period in which the multi-phased sites started.�����������������������������������������������������������������60 Figure 8. The period in which the settlements started that continued to the nineteenth century (red) or the present-day (blue). ��������������������������������������������������������������������������������������������������������������������������������������������������������������61 Figure 9. Overview of regions for which synthesising models on the formation of villages were available.��������������63 Figure 10. Overview of the combined outcome of the inventories of archaeological excavations, historical geographical case studies and (supra)regional models.�������������������������������������������������������������������������������������������������65

iii

Is this a village? Approaching nucleated settlements in Scandinavian contexts Figure 1. Prospect of Vik, c. 1930 with the medieval Hove church in front on the terrace behind an expanding settlement facing the Sognefjord. The new Vik church from 1877 can be discerned to the left, while Hopperstad stave church further west is out of the picture (Photo: Fylkesarkivet i Sogn og Fjordane, with permission).���������������������������������������������������������������������������������������������������������������������������������������������������������������������������73 Figure 2. Section of the cadastre map from Vik, 1864: Above to the left, the clustered tun at Hopperstad, with the stave church in its outskirts. On the eastern side of the river Hopra, the farm Hove appears with its three clusters with the church in the centre (Sunnfjord og Ytre Sogn jordskifterett, with permission).�����������������������74 Figure 3. Drawing and reconstruction of the buildings in the nucleated tun at the farm Seim, situated close to the fjord in Vik in the Sognefjord, as it appeared around 1870. Uncertain buildings drawn in dotted lines. (Drawing: A. Berg 1952).������������������������������������������������������������������������������������������������������������������������������������������������������77 Figure 4. Map showing nucleated farms from western Norway mentioned in the text (Drawing: P. Bækken).���������80

The Archaeology of Currently Inhabited Villages in Spain: The Case of Asturias Figure 1. Example of traditional mosaic landscape from the Northwest of the Iberian Peninsula. Mountain pasture area in Vigaña, Asturias.���������������������������������������������������������������������������������������������������������������������������������������93 Figure 2. Location map of study cases in Asturias.����������������������������������������������������������������������������������������������������������������96 Figure 3. Example of geoarchaeological survey.��������������������������������������������������������������������������������������������������������������������97 Figure 4. Territory of Vigaña (Belmonte de Miranda), excavated sites and chronologies obtained: 1. El Castru: fortified hillfort dated from the 6th century BC to the 1st century AD. 2. Las Corvas: Neolithic settlement with crop spaces superimposed until today. 3. Arrichere: agrarian terraces from the nineteenth century. 4. La Granda: agrarian terrace from the 16th century. 5: Las Murias: fields, not dated. 6: L’Eirón: agrarian terraces from the 19th century. 7: La Sienra, agrarian space from the Neolithic to the present. 8: L’Hortal: possible neolithic storage silos. 9: La Escuela: negative structures from medieval period. 10: Late medieval Necrópolis. 11: Medieval Necropolis: occupation from the seventh century AD to the present. 12: Braña d’Estoupiel. lo: traces of activities linked to the megalithic structure of La Chalga. Livestock and artisan activities of modern times. 13. Braña Folgueras: fences related to modern agricultural activities. 14. Hermitage of Linares: settlement occupation from the ancient bronze to the final bronze with reuse as a necropolis from the X.C century and subsequent use as chapel. 15: The Cuernu: megalithic structure.��������������������������������������������������������99 Figure 5. Las Corvas: Neolithic settlement. Fireplace within domestic structures.�������������������������������������������������������100 Figure 6. Hermitage of L.linares: bronze age structures.��������������������������������������������������������������������������������������������������101 Table 1. Vigaña radiocarbon data.������������������������������������������������������������������������������������������������������������������������������������������102 Figure 7. San Pedru de Vigaña. Early medieval post holes under Late medieval necropolis.��������������������������������������103 Figure 8. Village of Villanueva de Santo Adriano, mentioned in IXct. documents. View from the north.����������������105 Table 2. Results of the surface archaeological survey.�������������������������������������������������������������������������������������������������������105 Figure 9. Prospected areas and surveys with codes.�����������������������������������������������������������������������������������������������������������106 Figure 10. Different structures excavated (S. Romano). 1. Paved floor of medieval origin. A fireplace above was dated between XIIIth-XIVth centuries (CDR, UE7). 2. Post holes of the XIIth ct. (MUR, EU 12). 3. Detail of an excavated agrarian terrace dated from XV-XVIIth ct. 4. Post hole and structures filled up by sediments dated between the X-XI centuries (CDR, EU 11 -post hole excavated in EU 13).�����������������������������������������������������������������107 Table 3. Stratigraphic sequences noted in the soil and chronological correlation between SUs. The century is given where there is absolute dating. ����������������������������������������������������������������������������������������������������������������������������109 Table 4. Villanueva radiocarbon data.�����������������������������������������������������������������������������������������������������������������������������������110 Table 5. Vigaña and Villanueva archaeological results comparison.�������������������������������������������������������������������������������111

iv

Author Biographies Jan van Doesburg (1963) studied Medieval Archaeology at the University of Amsterdam, The Netherlands. Since 1990 he is working for the Cultural Heritage Agency (Ministry of Education, Culture and Science) as Senior Researcher Archaeology Middle Ages and Early Modern period. Jan is currently finishing his thesis on land use, development of elite groups and castle building in the Kromme Rijn area (central part of the Netherlands) during the Middle Ages. His main interest lies with medieval settlements, castles/ moated sites, post-medieval archaeology and material culture from the Early and High Middle Ages. Jesús Fernández Fernández’s lines of research and interests focus on Medieval Archaeology, Historical Archaeology, Landscape Archaeology, GIS and Heritage studies. His recent research interests and publications focus on the landscape and settlement transformations in the Asturian area, particularly in the early medieval period. Jesús has been the director of several archaeological fieldwork projects and a member of research projects in various universities. Currently he is co-director with Gabriel Moshenska and Margarita Fernández Mier of the Villanueva de Santu Adrianu medieval settlement excavation project. Currently he is teaching and researching at Oviedo University within the programme Marie-Curie COFUND. Fernández is also a social entrepreneur and director of La Ponte-Ecomuseum, an archaeological-museological community project in Asturias, founded in 2012 and an award winner in 2016 (Leading Culture Destination Awards) and 2019 (Hispania Nostra Awards for Good Practices in Cultural Heritage). Margarita Fernández Mier is Professor of Medieval History at the University of Oviedo (Spain). Her principal interest is the study of early medieval societies in the north-western Iberian Peninsula, as well as the settlement analysis and the organisation of agrarian landscapes research. Margarita’s work is based on written and archaeological records and a long term analysis, from Roman times to Middle Ages. She is PI of the ‘Local spaces and social complexity: the medieval roots of a twentieth-century debate (ELCOS)’ project: a interdisciplinary research group funded by the Spanish Government which aims to situate the present-day rural communities of Southern Europe as inheritors of a centuries-long experience of collective organisation from medieval times. Margarita is the lead investigator of he LLABOR research group working on Agrarian and Public Archaeology in Spain and Latin America. Bert Groenewoudt wrote his PhD on the applicability of non-destructive methods of prospection and assessment of archaeological sites (1994). He studied Prehistory at the University of Amsterdam and works as a senior landscape archaeologist with the Cultural Heritage Agency of the Netherlands (RCE). Project manager of the Dutch Valetta Harvest Project (until 2016) and of the Archaeological Research Agenda of the Netherlands. His research focusses on man-landscape interaction, long-term processes of change, land-use dynamics and the archaeology of ‘marginal’ landscapes. Carenza Lewis is Professor of Public Understanding of Research at the University of Lincoln and an archaeologist with research interests in rural settlement and childhood. Formerly an investigator for RHCME, presenter on Channel 4s television series Time Team and founding director of Access Cambridge Archaeology, she has published widely while leading initiatives engaging wider publics with heritage including the Higher Education Field Academy, Cambridge Community Heritage and Unearthing Middlefields Utopia. Director of Our Lincolnshire, from 2019-22 she is leading Community Archaeology in Rural Environments Meeting Societal Challenges (CARE-MSoC), a European Commission project exploring the social benefits of involving residents of rural communities in the Netherlands, Czech Republic and Poland in local archaeological excavations.

v

Dr Edith Peytremann is currently research engineer at National Institut for preventive archaeological research (Inrap) to Nantes and lecturer at the University of Nantes. She’s permanent Researcher at the Centre de Recherches Archéologiques et Historiques Anciennes et Médiévales (CRAHM-UMR 6273 Caen). She holds a HDR (French post-doctoral degree allowing its holder to supervise PhD students). She is also been Co-President of the French Association of Merovingian Archaeology since 2014. Her main research topics are rural settlements and site of early Médieval Ages, burial practices of early Medieval Ages and the furniture of early Medieval Ages except the pottery. She led and leads many archaeological excavations in Lorraine, Franche-Comté, Alsace and the Pays de la Loire. She is the author of numerous archaeological publications, the last of which is En marge du village. La zone d’activités spécifiques et les groupes funéraires de Sermersheim (Bas-Rhin) du VIe au XIIe siècle. Hans Renes (1954) is historical geographer at Utrecht University (Faculty of Geosciences) and professor of heritage studies at Vrije Universiteit Amsterdam (Faculty of Humanities), the Netherlands. He published on many different aspects of landscape history of the Netherlands and Europe as well as on the relation between landscape heritage and planning. Archaeologist Johan Verspay (1982) is a Phd candidate at the University of West Bohemia where he studies the transformation of the countryside of Brabant (NL & BE) during the Late Middle Ages and Early Modern period. As a research staff member at the University of Amsterdam he recently completed the Valetta Harvest project on medieval village formation. He is currently working on a follow-up project (CARE-MSoc), in the Groene Woud region (NL) in which archaeology, historical geography and medieval history are integrated to reconstruct the formation an development of villages through community archaeology. Ingvild Øye, Dr.philos. Professor emeritus of Medieval Archaeology, University of Bergen. Former posts at the Medieval Collections of the University Museum of Bergen and the Bryggens Museum; Director of Bryggens Museum and the medieval monuments Håkonshallen/Rosenkrantz Tower. She has published books, parts of books and articles related to both rural and urban archaeology of the Middle Ages, including landscape archaeology, settlement history within broader timeframes. Her publications also comprise works on material culture, especially tools and technology, many related to agricultural and textile production. She has been member in various national and international networks, a.o. as national representative in the international network Ruralia conferences on medieval rural archaeology (2004-2015). She is also member of the international Advisory Board of the journal Landscape History (2008-) and the editorial board for the academic book series The Northern World, North Europe and the Baltic 400-1400 AD, Economies and Cultures, Brill Academic Publishers. (2000-2016).

vi

Preface This book was developed in the context of research looking at the network of Early Medieval settlements in the north-west of the Iberian Peninsula. This has been one of the priorities of medieval archaeology in Spain over the past decade and has mainly focused on the study of abandoned villages. However, research into inhabited villages on the Iberian Peninsula has only taken place over the past five years or so, compared with the longer tradition of historical research in other European countries where this topic has been considered in greater depth. These countries have also addressed the theoretical challenges of such work, as well as those relating to archaeological practice. The resulting ability to compare the different experiences and issues arising from the geographical and cultural diversity across Europe is of particular interest when carrying out research in this field. Our experience over the past few years of research in Spain has led us to try to deepen our understanding of three fundamental aspects of the work. The first looks at the origins of inhabited villages with the aim of establishing a chronology for their occupation while analysing their relationship with rural Roman settlements. This type of settlement is fairly elusive in some of the areas covered by the research. A second aspect — in which little progress has been made to date — is to understand the reasons why some settlements, which are still in existence today, have survived, while others were abandoned. Thirdly, given the need to work with the populations who still live in these spaces, this type of archaeology has become an appropriate setting in which to engage more deeply with a true social archaeology which actively involves the local communities living in the areas being researched. To begin addressing these issues, we organised an event entitled ‘International colloquium on the archaeology of medieval villages currently inhabited in Europe’, held on 7 May 2016 at the University of Oxford and attended by several European researchers, with the aim of reflecting on the archaeological issues represented by medieval villages that are still inhabited today. As Chris Wickham argues, a comparative analysis may help us transcend the limits imposed by national archaeological traditions, which limit our ability to interpret phenomena produced, at least on a European level. Our aim was therefore to discuss different research agendas and their results and, ultimately, to gain a better understanding of the ways in which medieval rural communities — which have been resilient and have survived while also being transformed over the longue durée — were organised. All of the above took place within a general discussion framework, leading to a greater understanding of the situation in Europe. In the course of this colloquium, the diversity of situations and issues faced across the continent became apparent, as did the many obvious commonalities, and this is reflected in this collection of papers. The book is a compilation of some of the papers given at that event and other contributions that have broadened the context of the discussion at a European level. The colloquium could not have taken place without the participation of a number of people and institutions to whom we would like to extend special thanks. Firstly, we wish to thank Chris Wickham for his logistical support, and the Faculty of History and the University of Oxford for providing us with financial support and a space in the Radcliffe Humanities Building in which to hold the 2016 event. Secondly, thanks are due to the Instituto Humanismo y Tradición Clásica at the University of León, which contributed logistical and financial support. Andrew Reynolds (UCL Institute of Archaeology), Stephen Mileson (University of Oxford) and Wim De Clercq (University of Ghent) made some interesting contributions to the colloquium. Although they were not, in the end, involved in this publication, we are very grateful for their participation. Special thanks are due to Wendy Davies, who chaired the discussion at the end of the event. We would also like to extend our gratitude to all those who came to the Oxford colloquium and participated actively in the final discussion with their ideas and suggestions, all of which we have taken into account while gathering material and writing this summary. We are also grateful to the external reviewers, whose suggestions and comments greatly helped in improving the contributors’ texts. Finally, thanks are due to Archaeopress, and in particular to David Davison, for his confidence in our project from the very beginning. This book grew out of the ELCOS research project, Espacios locales y complejidad social. Las raíces medievales de un problema del siglo XXI [Local spaces and social complexity: the medieval roots of a 21st century issue], HAR2016-76094-C4-1-R, funded by the Spanish Ministry of Science and Innovation. vii

viii

Introduction Rural areas in present-day Europe have experienced an uneven process of abandonment, affecting a wide range of places across the continent to varying degrees. Rural communities have been transformed in different ways over the past few centuries, especially from the first industrial revolution onwards. From the 18th and 19th centuries onward, traditional communities in northern and central Europe were broken up, leading to the current pattern of small and medium-sized private farming enterprises and the growth of the service economy in rural areas. However, in southern European countries, some rural communities have continued to exist, having undergone varying degrees of transformation and using different methods for managing the land with resilient collective practices. These communities have adapted and redefined themselves over the centuries, and are still active today. Before the beginning of this process, traditional communities — despite the differences between them — followed a standard model, whose roots originated in the Early Middle Ages. However, in the process they also witnessed different turning points which led to the survival of some settlements, while acting to the detriment of others. Thus, over the last millennium, a large number of villages were abandoned, in a wide range of circumstances and over a long period of time. Over the past few decades, these villages have been the main focus of archaeological interventions. Historical research into the socioeconomic processes and the organisation of land in the Middle Ages has concentrated on these abandoned medieval villages. This has encouraged a significant renewal of studies, mainly from the 1960s, into rural medieval settlements and landscapes all over Europe, which have placed great importance on the inhabited areas of villages (for example, Hamerow 2002; 2012; Francovich & Hodges 2003; Peytremann 2003; Valenti 2004; Quirós Castillo, 2009, Valais, 2012). This important leap — which is both qualitative and quantitative — has been encouraged by a number of different, converging factors, in which the role played by so-called preventive archaeology is of particular importance. It has had a key role in the excavation of medieval sites (see, for example, the papers by Vespay et al and Peytremann in this publication). In some regions, such as Île-de-France, a large number of early medieval villages have been excavated in the course of the construction of large public works, which have had a significant impact on studies of French medieval history (Peytremann, 2003). These interventions have sometimes formed part of larger research projects around monuments or heritage sites which, in some cases, have provided critical information on the first early medieval settlements (e.g. Azkarate GaraiOlaun and Quirós Castillo, 2001). Together, they have contributed to a marked increase in the number of records available on the medieval period across the whole of Europe, while at the same time facilitating a methodological renewal as a result of area excavations and the reflection on the different factors and processes of village abandonment. In the same way, they have contributed to analysis of how each archaeological site has been generated, what its subsequent evolution has been, and what tools allow us to understand the meaning of the material culture associated with them (Schiffer, 1987, Lamotta, Schiffer, 1999). In parallel with this positive development, an imbalance in the type of deposit studied was strengthened, related to the evolution of European rural history itself. The progressive development of the early medieval archaeology was focused on studies of sites that allowed progress in the characterization of settlements and questioned the origins of the feudalisation processes. On the contrary, less attention was paid to settlements from other periods better documented by written sources, which slowed the development of an archeology focused on the problems related to rural communities after the twelfth century, always with great asymmetry between different historiographies. The large volume of information generated about the early medieval settlements has stimulated the debates over the existing historical paradigms, resulting in the categories and concepts used in the studies of early medieval settlements being called into question and opening up new perspectives and changes in focus. The volume of archaeological data has undoubtedly stimulated the discussion on 1

The Archaeology of Medieval Villages Currently Inhabited in Europe

the types of settlement, given the variety of situations that have been documented, as well as on the internal organisation of those communities and their relationships with centres of power operating at a supralocal level. Thus, the gradual incorporation of archaeological information on the Early Middle Ages has facilitated the discussion on the historiographical categories that were widespread during the 1990s, through studies of the land based on written sources. Concepts such as ‘proto-villages’, prévillageois (Guadagnin, 1988) or non-villages (Lewis, Mitchell-Fox & Dier, 1997) are used to define early medieval settlements, with the concept of a ‘village’ being reserved for inhabited areas linked to feudal structures (Zadora-Rio, 1995; Peytremann, 2003; Francovich and Hodges, 2003; Valenti, 2004; Reynolds, 2005; Wickham, 2005). Gradually, archaeological practice has marked out a landscape dotted with farms, hamlets and villages, with the differences between the categories being defined according to different parameters on which researchers have not always agreed. Of course, these forms of settlement differ greatly across Europe, meaning that there is a need to define the terminology used in the categories analysed, as exemplified in some of the works contained in this book (see Oye and Vesya et al) and which is further complicated by linguistic differences. More recently, also with different rhythms according to respective European historiographic traditions, working areas surrounding the villages have been incorporated into the investigation integrating the study of territoriality and agricultural practices through the analysis of different landscape elements like the forms of the plots, their excavation and the bio- and geoarchaeological studies. This has deeper knowledge of peasant work practices, their knowledge of territorial management, governance forms, and local communities’ indicators of identity. The excavation of cultivation fields, the investigation of mountain areas and pastures, and the archaeological study of forests has opened a new lines of work that have advanced a holistic approach to the deserted villages and their complementary surrounding productive spaces, especially in the Anglo-Saxon sphere (Ripon, 2008, Williamson, 2012, Quirós Castillo, 2014). In summary, the complexity of this historical record has led to a need for a renewal of methodological approaches. This has taken place through the setting up of interdisciplinary teams, as well as the systematic incorporation of archaeobiological and micromorphological samples, converging with the interests of environmental studies that have led the way in recent decades. Given this increase, and the fact that abandoned areas have been studied in a greater level of detail, interventions in small and medium-sized settlements, which have continued to be occupied until the present day, are more exceptional. It is true that spaces relating to centres of power (churches and castles) have been subject to archaeological surveys, but this has not happened in the rest of the inhabited areas and even less so in the production areas of villages. The result is a bias in the research agenda. Only those settlements which, for some reason, have been abandoned or, conversely, those which have been successful, ceasing to be villages and becoming towns, have been studied. It should not be forgotten that in recent decades, urban archaeology has seen a spectacular level of growth, linked to the activity of preventive archaeology. Today’s villages, which are largely medieval in origin, if not older, have remained on the margins of these programmes, although their situations vary depending on the historiographic traditions in different European countries. Cartographic records of early medieval settlements, which have served as the basis for hypotheses about territorial networks, are partial and will not become meaningful until the information from villages still inhabited today is incorporated. The deserted villages were only one part of the medieval rural network of settlements along with the villages that still survive today: they operate in unison and are part of a complex network integrated into different territorial frameworks that cannot be understood as separate phenomena. Due to all of these reasons we need to complement the information from deserted villages with the chronology of the villages still inhabited today. We should ask why some settlements survive while others disappear, and also question the role played by the different social groups that have taken part in these processes. The promotion of one settlement at the expense of the other has broad notable repercussions: there are processes of re-territorialization in which all local communities are involved, which imply relevant changes to local identities and also new conflicts. To understand these processes 2

Introduction

it is necessary to incorporate social complexity into the research agenda of medieval rural societies in the coming decades, analysing factors both external and internal to the community, and the role played by different social actors, both the peasantry, fleeing from a homogeneous reading as a group, and the lordships, with special attention to the church and the monarchy. The five papers in this book reflect the different trajectories of this type of work in Europe. In the UK (Lewis) and the Netherlands (Vespay et al), complex projects have been undertaken which have already generated a large volume of information, and which have shown the partial nature of the conclusions reached when based only on data from abandoned villages. England has the longest history of this type of intervention (Lewis, 2007; 2017). In her paper, Carenza Lewis outlines a methodological approach which focuses on the use of small surveys (test pits) to document currently occupied rural settlements (CORS). Eighty percent of medieval habitats are likely to have survived to the present day; however, we scarcely have any data on their historic development over the longue durée or on the distinctive features that these villages had at different times throughout their history. Work is pending in Britain to combine the data of these projects mainly focused in social aspects with a complementary and more complex reading of the abundant data provided by deserted villages. Equally relevant, due to the volume of data it has provided, is the project undertaken in the Netherlands (Malta-archaeology). One of the main objectives of Dutch archaeological research is to examine the formation and development of medieval and early modern villages. The project makes use of data from the study of abandoned villages, as well as that obtained from occupied villages. This has led to the collaboration of various different administrative entities in the management of this information. The experience of these projects impacts on a theme that is also addressed in these papers: information over the longue durée that can be gleaned through interventions on inhabited villages. On the one hand, the pre-existence of prehistoric settlements has been documented, in some cases dating back to the Neolithic era. This creates the possibility of identifying archaeological sites in territories with scarce documentation of open prehistoric settlements. On the other hand, on many occasions it is possible to verify the origin of these settlements in Roman times, with different characteristics and typologies; the coincidence of prehistoric and Roman sites with the medieval villages implies that we might question not only their origins, but also their continuity of use, breaks and ways in which they have been reused. The study of changes, permanent features, adaptations and modifications that have been features of these settlements from the Middle Ages to the present day it is also necessary, open up a variety of themes that will help us to gain a more complex understanding of these communities. The lack of attention given to the archaeology of inhabited villages could lead to a situation that is paradoxical, to say the least. For example, it could give us a better understanding of the features of early medieval settlements - particularly deserted villages. However, we do not have information adout the internal processes of transformation which survival villages have undergone, and of their responses to the different historical conditions throughout the Middle Ages and the Early Modern era. This brings us directly to the concept of resilience, which we view as a robust conceptual tool with which to consider the complex development of these communities. The concept of a ‘resilient settlement’ has been used recently to analyse the evolution of pre-industrial settlements, evaluating the ability of these communities to overcome crises caused by external agents. This has involved an analysis of their ability to recover from demographic crises, the decline of the areas that they used for economic purposes and the destruction of their dwelling areas (Curtis 2014). It is our view that some of the papers in this publication pave the way for a more complex understanding of the concept. We should not only be studying these communities’ capacity for survival and recovery up to the time of industrial revolution, but also the ways in which they were able to adapt and redefine themselves, what factors provided stability in contrast to deserted villages, and the forms of governance that guided their actions. We should also seek to understand the active role played by farming communities in these processes. At the same time, archaeological practice must allow for the study of the reuse and redefinitions of the elements comprising the agrarian landscape. Many of these elements are medieval in origin and continued to exist across the whole of Europe until the agrarian reforms of the 18th and 19th centuries, and in southern Europe until the present day. This can be seen in the research carried out 3

The Archaeology of Medieval Villages Currently Inhabited in Europe

in northern Spain (see Fernández Mier, Fernández Fernández) and in some of the examples from France (see Peytreman). The resilience of an inhabited space in the face of external processes and the resilience of the forms of the landscape, take on new functions and new meanings with the introduction of new crops, technological developments and new forms of land management and ownership. Another element that stands out in all the papers is the need for methodological reflection. The contributors are unanimous in their view that interdisciplinary studies should be a priority, with data from intensive archaeological interventions being considered alongside data from written, cartographic and linguistic sources. This is an indispensable first step in planning intensive archaeology, as argued in the paper on archaeological sites in Norway. Such information is essential for interpreting archaeological data, which tends to be fragmentary. The impossibility of carrying out area excavations yields different types of data which are difficult to interpret, making it necessary to create corpora of information. These should take into account the data gleaned from both inhabited and abandoned villages, as suggested in the project carried out in the Netherlands. Furthermore, there is a need for a general research framework as the basis for historical questions and reflections in a more general sense, leading to better coordination of the research agenda (Vespay et al.). The need to explore new approaches to archaeological intervention is also noted, such as archaeological and geotechnical surveys that may yield data and facilitate the planning of archaeological interventions, as has happened in some projects in France and Spain (see Peytreman and Fernández Mier, Fernández Fernández). A final important aspect is the involvement of local communities in the entire archaeological process; this is argued in the papers looking at the UK and Spain. The concept of Public Archaeology has been developed in the English-speaking world since the 1970s (McGimsey, 1972). It calls for a greater level of participation by local communities in the management of archaeology. This practice has been given a significant boost by post-processual theories, which demand the integration of the whole of society into the archaeological debate. The papers in this publication all touch on this aspect, from two different points of view. Firstly, the involvement of the educational community in the entire excavation process, which in England has led to the development of a broad programme led by Carenza Lewis’ projects. Secondly, the creation of different types of tool for the involvement of local communities in the generation of knowledge which, at the same time, is of use to them in the processes of change and transformation in which they are currently immersed. This is particularly the case in southern Europe, as seen in the work carried out on northern Spain. Taken as a whole, the range of papers in this publication allow us to take a closer look at an archaeology with enormous interpretive potential, which requires its own research agendas. These need to take place on at least three different levels: local, regional and European. At a local level, there are areas where not even preventive archaeology is undertaken in medieval settlements of this type (for example, in Spain archaeological surveys are not carried out in rural settlements at the same rate as they are in historic centres of towns and cities, meaning that there is a marked bias in the research). The raising of awareness and involvement of local government is key here, as Vespay et al have argued. On the other hand, where regular and systematic interventions are carried out, an appropriate degree of contextualisation of the results is required in regional interpretive frameworks, which is complemented by historical research. The papers by Peytremann and Lewis show the potential of this type of research, through which it is possible to question established historical paradigms. Finally, it is important to integrate the variety of experiences at a European level, as an initial step towards a better understanding of the informative significance and the interpretive potential of these inhabited places, which tend to be small and rural. This provides us with a better understanding of our diversity on a continental level, as well as those elements that we share and have in common. Our aim with this publication is to take an initial, modest look at this level, as a necessary step towards a future summary in which more European countries are represented. This collection of papers proposes methodologies, reflections and ways of approaching all these issues and scales of work. We hope that it will be of interest to readers who are researchers, those who are technicians or who work on an area of cultural heritage that is essentially rural, and as yet not well-known.

4

Introduction

Bibliography AA. VV. 1965. Villages désertés et histoire économique XI-XVIII siècle, París, S.E.V.P.E.N. Azkarate Garai-Olaun, A., Quirós Castillo, J. A. 2001. Arquitectura doméstica altomedieval en la Península Ibérica. Reflexiones a partir de las excavaciones arqueológicas de la catedral de Santa María de VitoriaGasteiz, País Vasco, Archeologia Medievale XXVII: 25-60. Curtis, D. R. 2014. Coping with crisis. The resilience and Vulnerability of the Pre-industrial Settlements. Farnham: Ashgate. Francovich, R., Hodges, R. 2003. Villa to Village. The Transformation of the Roman Countryside in Italy, c. 4001000. London: Bloomsbury Academic. Guadagnin, R. (ed.) 1988. Un village au temps de Charlemagne. Moines et paysans de l’abbaye de Saint-Denis du VII siècle a l’An Mil. Paris: RMN Hamerow, E. 2002. Early Medieval Settlements. The Archaeology of Rural Communities in Northest Europe, AD 400900. Oxford: Oxford University Press. Hamerow, H., 2012. Rural Settlement and Society in Anglo-Saxon England. Oxford University Press. Oxford: Oxford University Press. Lamotta, V., Schiffer, M.B. 1999. Formation processes of house floor assemblages. In P.M. Allison (ed.), The Archaeology of Household Activities: 19-29. Routledge. Lewis, C., Mitchell-Fox, P., Dyer, C. 1997. Village, Hamlet and Field. Changing medieval settlements in central England. Manchester-Nueva York: Windgather Press Lewis, C., 2007. New Avenues for the Investigation of Currently Occupied Medieval Rural Settlement: Preliminary Observations from the Higher Education Field Academy. Medieval Archaeology 51: 133-163. Lewis, C., 2017. Test pit excavation within currently occupied rural settlements: results of the English CORS project in 2016. Medieval Settlement Research 32: 70-78. McGimsey, C. 1972. Public archaeology. New York: Seminar Press. Peytremann, É., 2003. Archéologie de l’habitat rural dans le nord de la France du IVe au XIIe siècle. SaintGermain-en-Laye: AFAM Quirós Castillo, J. A. 2009. Early Medieval Villages in Spain in the light of European experience. New approaches in peasant archaeology. In J. A. Quirós Castillo (ed.), The archaeology of villages in the early middle ages: 13-26. Bilbao: University of the Basque Country. Quirós Castillo, J. A. 2014. Oltre la fragmentazione posprocesualista. Archeologia agraria nel Norvest della Spagna. Archeologia Medievale 41: 23-38. Reynolds, A. 2005. On farmers, traders and kings: archaeological reflections of social complexity in early medieval north-western Europe. Early Medieval Europe 13 (1): 97-118. Rippon S.J. 2008. Beyond the Medieval Village: The Diversification of Landscape Character in Southern Britain. Oxford: Oxford University Press. Schiffer, M.B. 1987. Formation Process of the Archaeological Record. Pennsylvania State University: University of Utah Press. Valenti, M. 2004. Insediamento altomedievale nelle campagne toscane. Paesaggi, popolamento e villaggi tra VI e X secolo. Florencia: All’insegna del Giglio. Wickham, C. 2005. Framing the Early Middle Ages: Europe and the Mediterranean 400–800. Oxford: Oxford University Press. Williamson, T. 2012. Environment, Society and Landscape in Early Medieval England. Woodbridge: The Boydell Press.

5

6

Test pit excavation as a method for reconstructing the development of currently-occupied rural settlements: Evidence from England Carenza Lewis Introduction – The challenges of archaeology in inhabited medieval settlements Across Europe, rural settlement is an important area of research for medieval archaeologists, as the successive volumes from the Ruralia International Association for the Archaeology of Medieval and Post-Medieval Settlement attest (Bis-Worch and Theune (eds.) 2017; Brady (ed.) 2019; Fridrich et al. (eds.) 1996; Fridrich et al.(eds.) 1998; Klápštĕ (ed.) 2000; Klápštĕ (ed.) 2002; Klápštĕ (ed.) 2005; Klápštĕ (ed.) 2013; Klápštĕ (ed.) 2016; Klápštĕ and Sommer (eds.) 2007; Klápštĕ and Sommer (eds.) 2009; Klápštĕ and Sommer (eds.) 2011). However, much of this attention has focussed on deserted sites, while still-inhabited places have been relatively neglected. This paper presents and contextualises a novel approach - test pit excavation - which has recently been used very effectively within inhabited rural settlements in the United Kingdom (UK), exploring its contribution to knowledge of settlement development, and to wider historical enquiry, and its potential for use in inhabited rural settlements beyond the UK. Rural settlement has been extensively studied in the UK (Dyer and Everson 2012; Gerrard 2003), inspired by the pioneering work of W. G. Hoskins, Maurice Beresford and John Hurst in the 1940s and 1950s (Beresford 1954; Beresford and Hurst (eds.) 1989; Beresford and Hurst 1990; Hoskins 1955). In most parts of the country, the visible remains of many deserted and extensively shrunken rural settlements have been the subject of archaeological investigation using a range of techniques. Excavations have taken place on hundreds of settlements, advancing knowledge and understanding of the origins, development, adaptation and contraction of rural settlements, and illuminating many aspects of life in the medieval countryside (e.g. Austin 1989; Dixon 1997; Horning 2004; Ivens et al. 1995; Kelly 1982; Musty and Algar 1986; Thompson 1960). However, the time-consuming nature of excavation - 30 years of excavation by Beresford and Hurst at Wharrram Percy in Yorkshire ultimately covered only 6.5% of the site (Wrathmell (ed.) 2012: 365) – has led to other methods being used as well. Topographical surveys of earthworks of abandoned settlement (often complemented by aerial photography) in England, Northern Ireland, Scotland and Wales have enabled the layout and extent of thousands of former settlements to be recorded across large areas (e.g. Chart 1940; Cushion and Davison 2003; Everson et al. 1991; Hartley 1989; RCHAMS 1993) with some surveys aiming to be comprehensive across entire counties (e.g. RCHME 1975; RCHME 1979; RCHME 1981; RCHME 1985). Another widely used technique has been field-walking, which has identified and dated large numbers of lost sites visible only as pottery scatters, with scores of the most useful surveys covering entire parishes or contiguous groups of parishes (e.g. Cambridge Archaeological Field Group 2015; Davison 1990; Hall 1996; Parry 2006) and transformed understanding of processes such as the origins of nucleated settlements (e.g. Foard 1978). Work such as this has advanced knowledge and nourished intensive academic debate regarding the origins, development and character of medieval settlement across the UK (e.g. Aston, Austin and Dyer (eds.) 1989; Blair 2018; Dyer and Jones (ed.) 2010; Hooke (ed.) 1985; Jones and Page 2006; Lewis, Mitchell Fox and Dyer 1997; RCAHMS 2002; Rippon 2018; Roberts 2006; Roberts and Wrathmell 2002; Taylor 1983; Yeoman 1997). As periodic reviews have shown, the study of medieval settlement in the UK could be said with some justification to be well-advanced (Christie and Stamper (eds.) 2012; Govan (ed.) 2003). However, much of this work has focussed on deserted sites, with medieval settlements which are inhabited today being somewhat overlooked. The main reasons for this are easy to see: explaining the circumstances surrounding the abandonment of deserted sites presents an immediately appealing challenge, while the 7

The Archaeology of Medieval Villages Currently Inhabited in Europe

lack of later site-use means that medieval remains are more likely to be accessible and well-preserved. Conversely, in inhabited settlements, medieval remains are more difficult for the archaeologist both to identify and to access, while the presence of medieval remains is much more difficult to predict. This neglect of inhabited medieval settlements is problematic because abandoned sites represent only a small minority of the total number of all medieval settlement. In the UK, there are around 3,000 deserted settlements of known or suspected medieval (11th to 16th century) date (Beresford and Hurst 1989), compared with around 50,000 or so places in existence today as villages, hamlets, towns, suburban areas and other settlements (TownsList.co.uk (accessed 14 May 2017)). The latter can usefully be defined as ‘currently occupied rural settlements’ (CORS), avoiding any implicit assumptions regarding whether or not the sites which are occupied today were inhabited in the medieval period, and if so, whether or not their use since then has been continuous. Quantifying how many of today’s CORS are likely to occupy the sites of medieval settlements is difficult as there are no medieval written sources listing all inhabited places in the UK (Pugh 1978). Place-name scholars covering all the UK list around 17,000 ‘major place names’ of pre-modern origin (Mills 1991: xii), but these studies omit many commonly duplicated names such as Weston or Thorpe as well as discrete places distinguished by suffixes or prefixes such as Upper/Lower, Great/Little, North/South and so on. In England, Domesday Book in 1086 lists around 13,000 place names (Welldon Finn 1973: xi), and approximately the same number (13,089) are recorded 250 years later in the Lay Subsidy of 1334 (Glasscock 1975: xxx). However, both these sources considerably underestimate the total number of settlements of medieval date, partly because some regions are omitted (Cumbria, Northumberland and County Durham are entirely excluded from Domesday Book), but also because data from some places are included anonymously within the returns of larger settlements or estate centres. In Norfolk, for example, 749 named places were recorded in Domesday Book (http://opendomesday.org/county/norfolk/), but around 1,500 manors are recorded in early 14th century records (Campbell 2000: 30). In addition, the record of a single named manor may encompass a large number of discrete hamlets or farmsteads if the conventional threshold of dispersion of 150m is applied (Roberts 1987: Figure 2.3a). This is especially likely in in areas of dispersed settlement, which extended across much of the UK in the medieval period (Rippon 2008; Roberts and Wrathmell 2000). Taking this into account suggests a total of perhaps as many as 22,000 or more discrete settlements in 11th – 14th century England (Roberts and Wrathmell 2000: 11-12). After adding Wales, Scotland and Northern Ireland, an estimate of 30,000 or so for the number of medieval settlements across the UK may be reasonable. Even allowing for a considerable margin of error, the significance of this figure for the study of rural settlement is that it shows that permanent desertion was an unusual fate for a medieval settlement in the UK, probably affecting little more than 10% of the total. The simple fact alone that deserted sites are a small minority would suggest that more archaeological attention should be given to CORS than has previously been the case, but this is reinforced by the fact that deserted sites may also be atypical. In the UK deserted sites are considerably more common in central England than elsewhere, and very uncommon in the west and north of the UK (Beresford and Hurst 1989; Roberts and Wrathmell 2000), so their distribution is clearly not typical of CORS. Furthermore, research has shown that in many cases deserted settlements were smaller than non-deserted settlements, appeared later in documentary records, paid less tax and were sited in less favourable locations (Lewis, Mitchell Fox and Dyer 1997: 143-155 and op. cit.). The conclusion to be drawn from this must be that our ability to understand the development of rural settlement in the medieval period risks being compromised if our evidence base is dominated by studies of deserted sites. The argument in favour of archaeological investigation of CORS is further strengthened by the fact that CORS cover significant extents of land about whose archaeology little is known: as noted above, on archaeological distribution maps, of all periods, most of the land currently covered by habitation is blank. Finally, land within or around CORS can be particularly prone to development, a threat which can be more effectively managed if the extent and character of the underlying archaeological resource is better known and understood, which is difficult to establish without excavation. 8

Carenza Lewis - Test pit excavations in currently-occupied rural settlements

However, in spite of this, until recently CORS were rarely included on most UK local authority records as medieval settlements: while most known upstanding medieval buildings were included on such records, the settlement itself was not. By contrast, most deserted settlements were included on such official records – deserted sites being identified as settlements of medieval date in a way that CORS were not. The omission was usually due to a lack of corroborating archaeological evidence. Very few regions in the UK had seen any systematic research-driven primary investigation aimed at CORS, and most of that which had taken place was of a top-down nature not involving original fieldwork (e.g. Roberts 1987; Lewis, Mitchell Fox and Dyer 1997; Roberts and Wrathmell 2000). This was due in no small part to the difficulty of conducting such fieldwork: the built-up environment renders techniques such as excavation problematic and fieldwalking and aerial photography impossible, while geophysical survey rarely yields useful results due to interference from modern features. Archaeological excavation which does take place within CORS is limited in most cases to observations or investigations carried out in advance of construction. These may be noted summarily in the proceedings of national groups such as the Society for Medieval Archaeology (http://www.medievalarchaeology.co.uk/index.php/publications-2/medieval-britain-and-ireland/) or the Medieval Settlement Research Group (https://medieval-settlement.com/publications/journal), or published in more detail in county archaeological journals (e.g. Cessford and Dickens 2005; Clelland and Mepham 2015). These provide useful insights, especially when large areas are excavated (e.g. Auduoy and Chapman 2009) or outcomes from many sites are synthesised (e.g. Rippon, Wainwright and Smart 2014; Wright 2015; Blair 2018). However, such excavations are usually only able to explore a very small part of any settlement, and the difficulties of using such data to reconstruct settlement development have been demonstrated with characteristic candour by Chris Taylor, who showed how his theories regarding the development of Whittlesford (Cambridgeshire) had an unsettling capacity to be overturned by excavations in new areas (Taylor 1989). What is needed is an approach which enables archaeologists to gain a more representative picture of medieval activity within CORS. This paper explores the capacity of test pit excavation to provide this. The development of test pit excavation in CORS in UK The use of ‘test pits’ – small hand-dug archaeological trenches typically just 1m square – to identify sites and establish the extent and/or depth of archaeological deposits has been an established archaeological technique since the 1970s (Coles 1972: 138-40; Barker 1986: 69-70). On open land (Hodges 1991), or on lightly vegetated terrain such as open woodland (Hayes 1985), the technique has been used in the UK and beyond (Gerrard and Aston 2007: 246), particularly effectively on prehistoric sites. Given that on open sites any number of test pits can potentially be placed anywhere, methodological debate in the later 20th century on the use of test pits largely focussed on identifying statistically valid strategies for determining how many pits should be excavated, and how these should be sited in order to deliver an unbiased sample (Orton 2000). When contemplating a programme of test pit excavation within built-up areas, however, these theoretical issues are over-ridden by the more practical difficulty that modern development renders much of the area completely inaccessible for excavation: site selection must instead be pragmatic and contingent. Simultaneously, localised variations in recent and current land-use which, in CORS, may change every few metres from one garden, yard, drive, playground, verge or green to the next, may affect preservation and so might be presumed to invalidate observations based on plotting the distribution of finds from different pits across the site. Given such circumstances, test pitting within an inhabited settlement was not seriously considered until it was pioneered by Chris Gerrard, Mick Aston, Michael Costen and Teresa Hall at Shapwick in Somerset (Gerrard and Aston 2007: 244-261). A ten-year programme of interdisciplinary fieldwork across that parish had started in 1990, and by the middle of the decade Gerrard et al. were well aware that the un-built-up areas within the inhabited village of Shapwick were too small to accommodate the open excavations they were carrying out elsewhere, and that such excavations would in any case take too long to be acceptable to most garden owners. However, having established good community relations in the earlier years of the project, when casting around for a new approach with which to investigate the village, they judged 9

The Archaeology of Medieval Villages Currently Inhabited in Europe

that small test pits sited within gardens, which could be completed in a day or so and cause very little disturbance to plants or people, might be acceptable. The experiment proved remarkably successful, and a total of 81 one-metre square test pits were ultimately dug in different gardens over six years. These produced substantial amounts of pottery and other finds which achieved the intended aims: firstly of establishing whether there was a pre-existing settlement underlying Shapwick; and secondly of showing that the resulting data could be used to reconstruct the historic development of the present village (Gerrard and Aston 2007: 248-253). Since the proof of concept has been established, test pit excavation has been used in other inhabited villages, notably between 2000-05 by Richard Jones in the Whittlewood Forest area of the BuckinghamshireNorthamptonshire border, investigating settlement development across a dozen parishes including Whittlebury, Leckhampstead, Lillingstone Lovell, Silverstone and Wicken (Jones and Page 2001; Page and Jones 2003a; Page and Jones 2003b; Jones and Page 2006). In 2003 the technique came to the attention of millions of viewers of the UK archaeological television series Time Team, when 42 pits were excavated on live television within the CORS of Great Easton in Leicestershire (Cooper and Priest 2003) and many more within hundreds of other gardens across the UK (Lewis 2003). In 2005 the use of test pit excavation to investigate CORS was scaled up still further when this author began a long-term programme in East Anglia (Lewis 2007a; Lewis 2014a). This has to date completed more than 2000 test pits within around 60 CORS in six counties (Table 1) (Lewis 2005; Lewis 2006; Lewis 2007b; Lewis 2008; Lewis 2009; Lewis 2011; Lewis 2012; Lewis 2013; Lewis 2014b; Lewis 2015a; Lewis 2016a; Lewis 2017a; Lewis 2018a). Most of these test pits have been excavated by thousands of members of the public participating in educational programmes (Lewis 2014c) and community projects (Lewis 2015b). With such a large number of pits excavated, the East Anglian project allows the potential of test pit excavation within CORS to be assessed across a large number of different settlements. Aims of the East Anglian CORS project The initial aim of the East Anglian CORS project in 2005 was simply to use test pit excavation to reconstruct the long-term development of a selection of CORS, by quantifying and mapping pottery finds to show which parts of those settlements were used by humans at different dates (Lewis 2007a), adding to the corpus of CORS whose medieval development had been archaeologically explored in order to help counter the bias favouring deserted sites. As the programme expanded, it became apparent that the outcomes of work on the individual settlements had the potential to be aggregated, synthesised and compared in meta-analyses which could explore wider historical questions. East Anglia was an appropriate area to study for a number of reasons. It has continuous sequences of pottery throughout the medieval period and displays considerable variation in landscape type, patterns of settlement form and levels of desertion (Martin 2012; Rippon 2008; Roberts & Wrathmell 2000: 22–3 & 40–44; Williamson 2003). Furthermore, when the project started in 2005, medieval settlement in the region had seen less fieldwork than other parts of the UK such as the English Midlands (Christie & Stamper 2012; Martin 2012: 228–9) and was thus a prime candidate for an intensive archaeological research programme. Pragmatically, East Anglia was also easily accessible from the University of Cambridge which hosted the CORS project from 2005-15. Methodology of the East Anglian CORS project Site selection A basic premise of East Anglian CORS project was that any settlement could potentially be included. Funding was mostly sourced from non-academic bodies interested in providing educational and community projects, which usefully enabled an open-ended, evolutionary, exploratory long-term programme of investigation not restricted by narrow research questions. The choice of settlements to be included in the project was primarily driven by local interest or educational need (Lewis 2007a; Lewis 2014c) rather 10

Carenza Lewis - Test pit excavations in currently-occupied rural settlements

than any conventional statistical method: the random element introduced by this co-produced selection strategy eliminated other selection biases which might have been generated had settlement selection been made in pursuance of a specific research question. As the programme developed, it succeeded in including CORS of different sizes (from small towns to hamlets) and different forms (including nucleated and dispersed settlements), sited in different pre-modern landscape zones (river valley, watershed, fenland, coastal, wood-pasture) and on differing geologies. Once each CORS was selected, the choice of sites for individual test pits within the CORS was made in collaboration with local residents, often supporters of local museums or members of local historical or archaeological societies, of which there are a considerable number in England (Thomas 2010). Local coordinators promoted the project in their community and identified sites where excavation could take place. The aim was for pits to be distributed across the settlement so that all areas are sampled, including areas of apparently recent origin such as 20th century new-build estates. In the UK, test pits can be sited wherever the landowner permits: any unbuilt-up land within CORS can thus be used as long as the site is safe and not one of the relatively small number of legally protected scheduled ancient monuments. In practice most test pits were excavated in private gardens. Excavation methods The excavations were mostly carried out by volunteer members of the public with little or no previous experience of archaeology. The excavation methodology was therefore devised to minimize the risk of information of archaeological significance being overlooked. Volunteers were briefed by a professional archaeologist and provided with written handbook and a 16-page pro-forma Test Pit Record booklet to record their test pit excavation. The recording booklet included designated spaces, prompts and 1:10 planning grids in order to ensure that all required observations were made and recorded, and a sufficient number of roving professional archaeologists were present to monitor the excavations, identify finds and provide advice. This system ensured that while the personnel and conditions for each pit were different, the process followed was always the same and was carried out to the same standards. This allowed the data from different pits to be compared. Each test pit measured one metre square (Figure 1) and was marked out with string and nails before its location was recorded using taped survey combined with GPS to draw a sketch plan (in the record booklet) showing the test pit in relation to nearby permanent mapped features such as buildings and property boundaries. Any turf was removed and excavation then proceeded in a series of 10cm horizontal spits. Before each spit was excavated its surface was photographed, drawn at 1:10 scale in the record booklet and its colour recorded with reference to a standardised colour chart included in the written handbook. All spoil was screened for artefacts using a 10mm mesh, unless the spoil could not pass through the sieve in which case it was hand-searched. Volunteers were asked to err on the side of caution by retaining everything they thought might even possibly be of interest: finds were then reviewed by a suitable specialist

Figure 1. A 1m square test pit under excavation in Pirton (Hertfordshire) in 2010.

11

The Archaeology of Medieval Villages Currently Inhabited in Europe

to determine what should ultimately be retained according to the CORS project discard policy. Finds from each spit were kept separate so that different layers could be identified retrospectively if changes were not noted during excavation, and each spit was given a separate context number. Features were rare, encountered in approximately 10% of pits, and up to 90% of these were relatively recent, dating to the 19th – 21st centuries. Cut features such as ditches or pits, if encountered, were excavated sequentially in the normal way under guidance from a professional archaeologist. Built features such as walls were carefully cleaned, planned and left in situ. In situ human remains (which were very rare, encountered in only c. 0.2% of pits) were cleaned, recorded and left in place. Excavation continued until impeded by a feature (ancient or modern) whose removal was inappropriate or impossible, or until volunteers reached natural, or ran out of time, or reached a depth of 1.2m (below which UK guidelines require excavations to be shored, which is not practical in a 1m square test pit). When excavation ceased, the sections were drawn at 1:10 scale in the record booklet and the pit was then backfilled and any turf replaced. After each excavation was completed, the finds and record booklet data were analysed and a summary submitted to local authority Historic Environment Record (HER) curators. Lists of pottery finds and maps showing the distribution of pits yielding pottery of different dates were made available online (http://www.access.arch.cam.ac.uk/reports) and synthesising reports were drafted and uploaded to the same site after test pit excavation in each CORS ceased (e.g. Lewis and Baillie 2014; Lewis and Pryor 2014a; Lewis and Pryor 2014b; Lewis, Ranson and GBAHG 2014; Ranson 2013). Finds were retained by the supervising archaeological organisation unless site owners wished to keep them. Experience showed that the most financially cost-effective strategy, and the most socially impactful, was for a number of test pits to be excavated simultaneously. This allowed a supervising archaeologist to perform a roving role overseeing several pits, and volunteers to enjoy the social aspect of being simultaneously part of a small team cooperating to complete their own test pit and part of a large team collaborating to advance understanding of the whole settlement by synthesising the outcomes from different pits. Educational projects for teenagers typically completed 10-12 pits over two days (Lewis 2007a), while community projects with more adults, if organised efficiently with sufficient archaeological support, could complete 30 or more pits over a single weekend (Lewis 2015b). Test pit excavation within each CORS generally continued until it either became impossible to find additional new sites to excavate or it was deemed that sufficient pits had been excavated to provide a representative picture, and little new understanding would be gained from further test pit excavation. Identifying when the latter point has been reached is not an exact science, but experience has shown that while any pit is potentially able to produce useful data about the particular spot it occupies, for more general inferences about settlement development, coverage of perhaps 20% of properties may be needed in smaller places with fewer than 30 or so properties present today; but in larger places such as Pirton, excavations in 5%-10% of properties may be sufficient, especially given a reasonably evenly spread distribution. A pragmatic attitude to compromise is often required as it will not always be possible to achieve even coverage, but experience has indicated that overall, 30-60 pits in an average-sized CORS is likely to provide a useful overview of the past development of that settlement. Principles of data analysis Analysis pertaining to medieval settlement development using test pit data from CORS has focused on pottery because this was extensively used in eastern England and can thus act as a proxy for human presence (Lewis 2007a; Lewis 2016b). Medieval ceramics were easily broken and are taphonomically durable in most archaeological contexts, while sherds are relatively easy to see during excavation and can be dated visually without incurring prohibitive costs. Medieval potters produced a wide range of wares, most industries remaining in production for a century or more (McCarthy & Brooks 1988): given this long productive lifespan, short-term perturbations in settlement and demography are unlikely to register in pottery assemblages, but sustained long-term change will be apparent. 12

Carenza Lewis - Test pit excavations in currently-occupied rural settlements

Making inferences using pottery from test pits requires care, of course, as these are inevitably based on small samples potentially affected by a range of biases (Lewis 2007a: 139–40). Experience and comparison with data from excavation and field-walking (Haselgrove et al. 1985; Davison 1990; Parker Pearson and Schadla-Hall 1994), suggests that for the Roman, late Anglo-Saxon, medieval and postmedieval periods, the recovery of five or more sherds from any one test pit is very likely to indicate contemporary settlement in the immediate vicinity; two to four sherds is more than would be expected from non-habitative activity such as manuring of arable land and therefore may indicate settlement nearby; while a single sherd, especially if small and heavily abraded, is much less likely to indicate settlement nearby (Jones 2004; Lewis 2005: 11–12). Inferences based on pits from several different modern properties are considered more reliable than those based on single pits, while negative evidence (i.e. the absence of pottery) must be used with considerable caution, with inferences based on several pits in different properties more reliable than those based on single pits. Outcomes of the East Anglian CORS project Some of the outcomes of test pit excavation in CORS in East Anglia are summarised below in order to enable the value of this technique for reconstructing the development of CORS to be assessed. This encompasses 2,022 test pit excavations carried out in 59 CORS in East Anglia (Table 1; Figure 2). The average number of pits per CORS is 34, but in fact the numbers vary from seven (Bures and Mount Bures, Essex) to 114 (Pirton, Hertfordshire). Twelve CORS have had fewer than 20 pits excavated, and 31 have had more than 30.

Figure 2. Eastern England, showing the location of parishes included in the East Anglia CORS project between 2005 and 2017.

13

The Archaeology of Medieval Villages Currently Inhabited in Europe

Figure 3. Pirton as depicted on the 6” to 1 mile Ordnance Survey map c. 1880, with the three main zones of settlement highlighted (Ordnance Survey map copyright University of Edinburgh/Digimap, reproduced with permission).

Pirton Pirton in Hertfordshire is reviewed here in detail because more test pits have been excavated in this village than in any other CORS to date (Lewis 2007: 52-3; Lewis 2008: 64; Lewis 2009: 51; Lewis 2011: 53-4; Lewis 2012: 48; https://www.access.arch.cam.ac.uk/reports/hertfordshire/pirton), and so it provides a useful demonstration of what this technique can reveal. Figure 3 shows the footprint of the 19th century settlement and some of the street names and farms mentioned in the text; Figures 4-9 show the distribution of test pits containing pottery from different historic periods, with the pits which did produce pottery for that period shown as circles, and those which did not shown as white squares. The parish of Pirton had a population of 1,274 recorded in the most recent (2011) census, most of whom live in the nucleated village of c. 526 houses which today clusters east, west and (mainly) north of the 12th century church (HHER 4315) (see Figure 7), which itself lies adjacent to earthwork remains of a 12th century motte and bailey castle (SAM Herts 13612). Earthworks south and east of the motte (HHER 746) are likely to represent the remains of manorial features, possibly contained within a large outer castle bailey (although lack of excavation in this scheduled (protected) area means these features are not well-understood). In the 19th century, before extensive 20th century expansion on the northeast and south-west, maps show that the settlement comprised three discrete zones (Figure 3). The first of these clustered south-west of the church around Great Green and Bury End; the larger second zone lay in the centre of the present settlement, arranged mainly along the High Street and Royal Oak Lane; the third zone comprised two smaller rows of houses somewhat intermittently spaced along the north end of High Street and along Shillington Road and Burge End Lane. Additionally, three detached farms lay c. 200m to the north and west of Zone 3, two of which have moats around them: fragmentary remains of the moat at Burge End Farm lie c. 0.3km to the north of the present village, while a second, better-preserved moat lies 500m to the south-west at Parsonage/Rectory Farm. Walnut Tree Farm, 14

Carenza Lewis - Test pit excavations in currently-occupied rural settlements

beyond the southern fringes of the settlement lies c. 300m south of Zone 2. Excavation of 114 test pits in Pirton across most of its present footprint (shaded on Figures 4-9) has allowed the use of this area over time, period by period, to be reconstructed in some detail (Figures 4-9). (NB Prehistoric material is excluded from this overview of material more pertinent to medieval settlement development.) Romano-British Pirton (late 1st – late 4th centuries AD) Analysis of material of Roman date (late 1st – late 4th century AD) (Figure 4) shows that 14 of the 30 pits in the area between Rectory Farm and Burge End Farm produced habitative amounts of pottery of this date, as did five more widely spaced pits on the south-eastern edge of the present village south of Little Green. Very little pottery of Roman date was found in the intervening area (between Great Green, High Street and Royal Oak Road), now the centre of the village. The test pit data thus indicate that habitation at Pirton in the Romano-British period comprised two discrete settlements, both located on the fringes of the present CORS and separated by c. 800m. The northern settlement appears to be the larger, and to have been arranged as a linear row at least 450m long, aligned along the southern side of a small stream flowing north-east. The southerly settlement may have constituted two or more separate small sites located on land which today drains to the east. This pattern is typical of RomanoBritish lowland England, with field-walking often revealing settlements separated by less than 1km (e.g. Rogerson et al. 1997: 13-17). Early/middle Anglo-Saxon Pirton (early 5th – early 9th centuries AD) The test pit data for the early and middle Anglo-Saxon period (early 5th – early 9th centuries) (Figure 5) indicate the northerly Romano-British settlement at Pirton was entirely depopulated, as not a single sherd of pottery dating to any point between the early 5th and mid-9th century was found

Figure 4. Test pits excavated at Pirton showing test pits producing Romano-British pottery (black/grey circles), test pits which did not produce Romano-British pottery (white squares) and names of streets and farms mentioned in text.

15

The Archaeology of Medieval Villages Currently Inhabited in Europe

in any of the 30 or so pits excavated between Rectory Farm and Burge End Farm. In fact, the only evidence from test pits of any activity in this period comes from one pit on the southern edge of the present village (south-east of Little Green), which produced a single sherd of hand-made organictempered pottery of 5th-to-8th century date. It is interesting to note that this pit also produced pottery of Roman date, possibly suggesting a degree of post-Roman continuity of site-use here. The scarcity of pottery suggests that this settlement (if this is what the pottery represents) was small in extent and relatively short-lived. None of the test pits at Pirton produced any pottery of middle Anglo-Saxon date (mid-7th – early 9th centuries), but in 2016 radiocarbon dating and reassessment of the pottery from larger excavations in 1993 behind the Fox Inn on the High Street (HHER 9676; Blair 2018, 323-4) showed some of this material to date to the 7th to 9th centuries (pers. comm. G Burleigh email 3/2/2017). This suggests that the early Anglo-Saxon activity south-east of Little Green had shifted location to the area near the Fox Inn by the middle Anglo-Saxon period. However, the lack of pottery of this date from nearby test pits suggest that activity may have been limited in extent or duration. This pattern is reflected from excavation elsewhere in England which shows the settlements of this date were typically small and mobile within the landscape (Hamerow 2012: 65-73; Wright 2015). Late medieval and post-medieval Pirton (late 14th to late 18th centuries AD) In the later Anglo-Saxon (late 9th - late 11th century) (Figure 6) the number of test pits producing habitative amounts of pottery at Pirton soars to 33. These 33 pits cluster in three distinct zones: in the north-west, centre and south-east of the present village. The north-western zone, between Rectory Farm and Burge End Farm, is represented by just six intermittently scattered pits, none producing more than four sherds. This material is difficult to interpret but it is plausible that it indicates three or more separate farmsteads, suggesting that some of the farms now present in this area may be of Anglo-Saxon or Saxo-Norman origin. The form of the settlement (if this is what the later Anglo-Saxon pottery represents) in this area is very different to that indicated by the dense and continuous spread of Roman-British material in the same area, and it must be emphasised that there is no evidence of continuity of occupation over the 500 years between the Roman and late Anglo-Saxon periods. In contrast, in the central part of today’s settlement (along and between High Street and Royal Oak Lane) a continuous spread of test pits produced late Anglo-Saxon pottery, suggests that a nucleated village had come into existence in this area by the 11th century. Pottery of 9th to 11th century date comes from all test pits around the Fox Inn, where the 1993 excavation (noted above) revealed a late Anglo-Saxon cemetery and east–west orientated building plausibly interpreted as a church (HHER 9677: note 2016; Fenton 1993). The test pit data suggests a close spatial link between church and community at his time, and the rectilinear distribution of pits yielding pottery of this date provides a more tentative hint that this village may possibly have been planned on a gridded layout, as has been noted elsewhere (e.g. Blair 2013; Mortimer 2000; Ravensdale 1974: 121–44; Blair 2018). Whatever its precise layout, the late Anglo-Saxon nucleated village in the centre of Pirton seems to represent a new incarnation of settlement on this site, located on land which was uninhabited in the Romano-British and early Anglo-Saxon periods and which may well have developed from a pre-village nucleus of 7th to 9th century date, following a developmental sequence which has been identified elsewhere in central England (Jones and Page 2006: 87-92; Wright 2015: 179-181). A third discrete area of late Anglo-Saxon settlement at Pirton is represented by a cluster of a dozen test pits south of Little Green along Walnut Tree Lane, on the southern edge of the present settlement. These pits all lie immediately adjacent to the field east of Walnut Tree Lane which contains the earthworks of features thought to be part of the 12th century castle bailey (noted above). This raises the interesting possibility that the motte and bailey castle were deliberately built near an Anglo-Saxon site, possibly also of high status. 16

Carenza Lewis - Test pit excavations in currently-occupied rural settlements

Figure 5. Test pits excavated at Pirton showing test pits producing early/middle Anglo-Saxon pottery (black/grey circles) and test pits which did not produce early/middle Anglo-Saxon pottery pottery (white squares).

Figure 6. Test pits excavated at Pirton showing test pits producing late Anglo-Saxon pottery (black/grey circles) and test pits which did not produce late Anglo-Saxon pottery pottery (white squares).

17

The Archaeology of Medieval Villages Currently Inhabited in Europe

High medieval Pirton (early 12th – early 14th centuries AD) The test pits at Pirton show the volume of pottery to soar again in the high medieval period (early 12th– early 14th century) (Figure 7), with the number of pits producing more than a single sherd increasing three-fold from 33 to 91. While there was a general increase in pottery manufacture and use in the 12th and 13th centuries (McCarthy and Brooks 1988: 68-73; Courtney 1997: 97), analysis of the pottery from the Pirton test pits, which focusses on changes in the number of sites producing pottery, rather than changes in the number of sherds per se, suggests that the population of Pirton may have tripled in size between the 11th and 14th century. The distribution of test pits producing high medieval pottery shows some distinct patterns. Ribbon development along High Street, along with the addition of a new linear row settlement along Burge End Lane and Shillington Road, served to connect the formerly separate farmsteads between Rectory Farm and Burge End Farm to the nucleated village along High Street, considerably increasing the overall footprint of the medieval settlement. The test pit data also indicate that a new settlement at Great Green (HHER 12427) came into existence at this date. Great Green is close to the motte-and-bailey castle (HHER 32) and the present church (HHER 4315), both of which are likely to be of 12th century date, and the test pit data suggest that in this area, castle, church and settlement may all be the result of seigneurial construction and replanning in the late 11th and 12th century (probably occurring simultaneously to the cessation of use of the cemetery and church near the Fox Inn). Overall, the test pit data indicates that by the 13th century, Pirton had developed into a large, densely inhabited nucleated village measuring c. 800m by 600m, which had acquired a compact but nonetheless polyfocal form as the new and existing nodes of settlement grew in size and agglomerated. The only areas which did not produce pottery of this date are around Little Green and on the east side of Royal Oak Lane itself. As this absence is reflected in several pits excavated in different plots, it is likely to represent a real pattern, and hints at that possibility that the green near the village pond may then have been larger than it is now, and that Royal Oak Lane demarcated the eastern limit of the high medieval village. Late medieval Pirton (late 14th – mid-16th centuries AD) In the later medieval period (late 14th – mid-16th century) (Figure 8), a very different pattern is evident in the test pit data. Overall the number of pits producing habitative volumes of pottery plummets from 91 to 32: Pirton was reduced to barely a third of its former size. The obvious causes of this depopulation, given the date of the pottery, are the successive climatological and epidemiological disasters of the 14th century, culminating in the 1348-9 outbreak of plague known as the Black Death (Lewis 2016b). No part of Pirton was unaffected, but some distinct patterns are apparent. Royal Oak Lane and Burge End Lane were particularly badly impacted: virtually no pottery was found in these areas, while the areas along Shillington Road and east of Walnut Tree Road fared little better. The formerly densely populated area along, and east of, High Street was reduced to an intermittent scatter of inhabited plots which appear to have lain between others which had either been entirely vacated, or had been combined with those of former neighbours to create a smaller number of properties expansively sited within larger plots. The area around Great Green fared better, although its fringes appear to have been abandoned. Likewise, in the north of the village the locations producing pottery appear to have contracted from a continuous spread to cover only the areas immediately adjacent to Rectory Farm, Wrights Farm, Hammonds Farm and Burge End Farm. The test pit data for the post-medieval period (late 16th – later 18th century) (Figure 9) indicate that recovery at Pirton was slow, and that when it did occur the margins of the settlement (including the west end of Shillington Lane and the areas south of West Lane and north-west of Little Green) remained largely unoccupied, with habitative volumes of pottery of late 16th to late 18th century date coming predominantly from the three zones shown as inhabited on the 19th century Ordnance Survey map (Figure 3). 18

Carenza Lewis - Test pit excavations in currently-occupied rural settlements

Figure 7. Test pits excavated at Pirton showing test pits producing high medieval pottery (black/grey circles) and test pits which did not produce high medieval pottery (white squares).

Figure 8. Test pits excavated at Pirton showing test pits producing late medieval pottery (black/grey circles) and test pits which did not produce late medieval pottery (white squares).

19

The Archaeology of Medieval Villages Currently Inhabited in Europe

Figure 9. Test pits excavated at Pirton showing test pits producing post-medieval pottery (black/grey circles) and test pits which did not produce post-medieval pottery (white squares).

Conclusion from Pirton The pottery data from the test pits thus shows how the layout of Pirton as depicted on the earliest reliable 19th century maps (Figure 3) developed over nearly 2,000 years of remarkably dynamic change in the size and layout of the settlement, including two periods of apparently catastrophic decline. While each period left its mark, the form of the settlement in the 19th century has been shown to be essentially an artefact of the post-medieval period. Other settlements in eastern England Similar work at dozens of other CORS (Table 1) is building up a substantial corpus of case studies of places whose spatial development and demographic fluctuations can, as at Pirton, be reconstructed as never before. To give some idea of the diversity of trajectories these reveal, shorter summaries of data from a selection of these settlements are presented below, drawing on reviews published in Medieval Settlement Research (Lewis 2005; Lewis 2006; Lewis 2007b; Lewis 2008; Lewis 2009; Lewis 2011; Lewis 2012; Lewis 2013; Lewis 2014b; Lewis 2015a; Lewis 2016a; Lewis 2017a; Lewis 2018a) and available online (https:// www.access.arch.cam.ac.uk/reports/). (See Figure 2 for locations.) Acle Acle (Norfolk) is now a small town near the major medieval city of Norwich, overlooking the Norfolk Broads adjacent to upper tidal reaches of the River Bure. Here the excavation of 45 pits (Ranson 2017a; Lewis 2009: 53; Lewis 2011: 54-5; Lewis 2012: 48-9; Lewis 2014: 76; Lewis 2015: 44-5) indicated that settlement in the Roman period comprised a dispersed scatter of perhaps five small foci. This was very different in form to the next known settlement, dating to the late Anglo-Saxon period, which took the form of a small cluster. This expanded in the high medieval period to include at least two outlying suburbs to the south 20

Carenza Lewis - Test pit excavations in currently-occupied rural settlements

and east of today’s settlement as well as farms or possibly fields to the north-west, before contracting after the Black Death into a tightly clustered nucleation overlying the earlier Anglo-Saxon core at the centre of the present settlement. Ashwell A total of 50 test pits excavated in the now-large nucleated village of Ashwell (Hertfordshire) (Lewis 2012: 47-8; Lewis 2013: 84; Ranson 2013) which grew up around a spring near a Roman temple site (Jackson and Burleigh 2018) showed the CORS to have overlain perhaps four small Romano-British settlements, each represented by just a single pit, only one of which produced more than four sherds of pottery. Despite 14 burgesses being recorded in Domesday Book (indicating a substantial, implicitly urban, holding) (Williams and Martin (ed. and trans.) 2003: 375), no test pits produced any pottery pre-dating the Norman Conquest, an enigma whose resolution will require further investigation. The rectangular gridded plan evident in the present settlement may originate in the high medieval period, as pottery of this date was found in pits across this area, but not all pits produced such material, suggesting that the settlement was not very densely populated. In contrast, after the 14th century Ashwell collapsed into a densely inhabited but very much smaller nucleated settlement tightly clustered around the church, so the present plan may in fact post-date the medieval period. Binham Excavation of 60 pits showed Binham (Norfolk) (Lewis 2009: 53-5; Lewis 2011: 55; Lewis 2012: 49; Lewis 2013: 85; Lewis 2014: 76-7; Ranson 2017b), located on sandy soils near the major medieval shrine of Walsingham, to lie adjacent to a Roman settlement, probably a villa, and indicated that subsequent activity in the early and middle Anglo-Saxon period shifted westwards to occupy two discrete areas. Binham has been identified as a possible ‘central place’ in the 5th to 6th century on the basis of exceptional finds from metal-detecting (Behr et. al. 2014) and the test pit excavations thus provide some context for these, showing where contemporary settlement may have been. Although the Benedictine priory is not documented until 1091, pottery of 5th to 9th century date was found in test pits along its precinct boundary, including sherds of Ipswich Ware found in an undisturbed beam slot a metre below the present ground surface close to the priory entrance. It is tempting to speculate that this may relate to earlier, undocumented religious use of the priory site, and it certainly hints at the possibility that community memory of this spot as an important middle Anglo-Saxon site may explain the choice of site for the later priory (Lewis 2009: 53-5; Lewis 2011: 55; Ranson 2017b: 73-4). In the later Anglo-Saxon period, the focus of activity shifts east and west, creating two settlements separated by c. 300m. The high medieval period sees settlement intensify south of the priory, causing the two settlements here to become conjoined, while a new hamlet extension appears east of the priory precinct boundary. The post-Black Death period sees severe contraction, when the number of pits producing more than a single sherd of pottery falls from 24 to seven, with the area closest to the priory worst affected. The centre of the present CORS, where an inn and shops now cluster around two small greens, does not appear to come into habitative use until the 16th or 17th century, although insufficient pits have been excavated here to be entirely confident about the development of this part of the settlement. Clavering 40 test pits excavated in the dispersed wood-pasture settlement landscape around at Clavering (Essex) (Lewis 2013: 83; https://www.access.arch.cam.ac.uk/reports/essex/clavering) produced small amounts of Romano-British pottery from three discrete sites, hinting that the presently very dispersed settlement pattern was equally so in the Roman period. No pottery was found dating to between the end of the Roman period and the early 12th century, but in contrast, the high medieval period saw an explosion of settlement across the landscape, as the number of pottery-producing sites climbed from zero in the 21

The Archaeology of Medieval Villages Currently Inhabited in Europe

Anglo-Saxon period to above average for the region in the high medieval period. Settlement then included a small nucleated village clustered around a church and manorial site (EHER 115: EHER 113) along with numerous even smaller dispersed hamlets scattered along lanes throughout the parish, many with ‘green’ or ‘end’ names and some possibly comprising little more than single homesteads. A significant number of these sites are furnished with moats. This expansion ceased in the later medieval period: most outlying sites produced no later medieval pottery at all, although the nucleated settlement around the church seems to have fared better than more remote sites. All but one of the test pits produced pottery of 16th to 18th century date, showing that when recovery took place, the dispersed character of the settlement pattern established in the high medieval period was re-established. Great Shelford Excavation of 41 pits in the river valley village of Great Shelford (Cambridgeshire) (Collins 2013; Lewis 2006: 40; Lewis 2008: 61; Lewis 2011: 51; Lewis 2012: 44-5) showed Romano-British pottery to be restricted to just four pits near to the river in the south of the present settlement, with only one producing more than a single sherd and none producing more than four. Early and middle Anglo-Saxon activity was not detected, but pottery of late Anglo-Saxon date indicates a shift northwards, with activity focussing in three discrete zones: near the church, along the High Street and on the edge of High Green, although the latter is represented by just a single pit producing fewer than five sherds. Growth in the extent and density of habitation is demonstrated by the large number of pits (22, or 54% of the total excavated) producing habitative levels of high medieval pottery, but like many CORS in the East Anglian study, this was reversed in the post-Black Death period when the number fell to just nine. High Green, the area which saw the most growth in the high medieval period, saw the most severe contraction. Long Melford A total of 74 test pits excavated within the large Stour Valley village of Long Melford (Suffolk), Lewis 2015: 46; Lewis 2016: 61; Lewis 2017: 76; https://www.access.arch.cam.ac.uk/reports/suffolk/longmelford) today more than 3.5km in length, showed this to be the only CORS in the East Anglian study area to be located directly on top of Roman settlement of any size. 28 test pits (nearly 40% of the total number excavated by 2017), produced more than a single sherd of Romano-British pottery, most of these concentrated in the centre of the present settlement. No pottery of early or middle Anglo-Saxon date was found in any of the pits, and when the settlement next becomes archaeologically visible, in the later Anglo-Saxon period, the main focus of activity had shifted more the 500m to the north around a large green near the church. This settlement extended northwards in the high medieval period, while to the south the area of the former Roman settlement was re-occupied, as was another zone c. 300m further to the south of this. For the post-Black Death period all these areas see an increase in the number of sites producing pottery, with growth in the southernmost two zones sufficient to cause these discrete nodes to coalesce into a single very long linear settlement extending for more than 1km along the High Street. Nayland Excavation of 50 pits in the valley-bottom settlement of Nayland (Suffolk) (Collins, Lewis and Pryor 2015; Lewis 2013: 87-8; Lewis 2015: 46) indicated that a single small Romano-British rural settlement on the eastern margins of the existing settlement was subsequently abandoned and the low-lying site of the CORS appears to have remained uninhabited until a small late Anglo-Saxon pre-village nucleus came into existence near the church in Nayland. This grew into a large and densely packed nucleated settlement in the high medieval period, with the number of pits producing habitative amounts of pottery soaring from two to 29. The settlement continued to grow in the post-Black Death period, but in density rather than area, probably due to lack of space in the valley bottom site: despite this apparent handicap, the number of pottery-producing sites within the village rose again in this period to 34, or 68% of the total excavated. 22

Carenza Lewis - Test pit excavations in currently-occupied rural settlements

Sharnbrook A total of 68 test pits in the large stream-side nucleated village of Sharnbrook (Bedfordshire) (Collins and Pryor 2017; Lewis 2007: 49; Lewis 2008: 60-1; Lewis 2009: 44; Lewis 2011: 48-50; Lewis 2012: 42-3; Lewis 2013: 79) showed Romano-British activity be restricted to the edge of an area of 20th century housing on the extreme southernmost periphery of the present village and a single sherd from one pit, some 600m to the north-west and also on the edge of the present village. Early Anglo-Saxon pottery was found from two pits along the High Street, one of which contained a single tiny sherd of Romano-British pottery weighing less than 5g. Neither of these pits produced any late Anglo-Saxon pottery, with settlement in the 9th to 11th century apparently dispersed in four discrete small nodes, three of them along the High Street, perhaps arranged as an interrupted row as each appear to be separated from the next by at least 200m and none extend to more than one test pit. The high medieval period saw settlement along the High Street increase dramatically in extent and density, with the number of pottery-producing pits rising from one to 12. Settlement along the northern end of the High Street also expanded, and six other discrete areas also produced pottery from two or more pits, hinting at the likelihood of other nodes of settlement in the vicinity. These all appear to be new foundations, as none produced any Anglo-Saxon pottery. Post-Black Death contraction is clearly evident, with the number of habitative pits dropping from 27 to 16. Although this decline is less severe than in many other CORS, it appears to affect all parts of the settlement. Walberswick 42 pits excavated within the now-small coastal village of Walberswick (Suffolk) (Collins 2017; Lewis 2014: 79-80; Lewis 2015: 48; Lewis 2016: 61; Lewis 2017: 77) showed Romano-British pottery to be restricted to three small widely dispersed locations, a complete lack of 5th to 9th century pottery, and late AngloSaxon pottery to be found on four discrete sites each separated from the other by at least 200m. Pottery of high medieval date was found widely across all the present settlement, including in areas to the north and south of the main street where maps show no settlement to have been present in the 19th century. Walberswick shows no sign of late medieval contraction, but there is a noticeable shift westwards in the distribution of pits producing habitative levels of pottery, and streets around the church appear to be newly established for habitation at this time. Synthesis: Test pit data illuminating wider research themes The examples summarised above show how detailed settlement narratives can be reconstructed using data from test pit excavations, in a wide range of CORS, of different forms, located in different landscapes. Beyond this, however, comparative analysis can contextualise individual trajectories (Lewis 2014). It can be noted from the statistics in Table 1, for example, that on average, while c. 9% of test pits have produced two or more sherds of Romano-British pottery, only 1.6% produced even a single sherd of early or middle Anglo-Saxon pottery, but 10.5% produced two or more sherds of late Anglo-Saxon pottery, rising to nearly 40% in the High Medieval period, plunging to 20% after the 14th century before surging again to c. 60% in the post-medieval period (Figure 10). There is much to ponder in data such as this, but it can briefly be observed here that Pirton, Acle, Binham and Long Melford all have substantially more pits producing Roman pottery than average (17%, 18%, 20% and 38% respectively), Nayland (8%) is close to average and Great Shelford and Walberswick (both 2%) are well below average. The chart also highlights differences in trajectories, with Pirton contracting particularly severely in the post-Black Death period, while Long Melford, Nayland and Walberswick all bucked a general trend of severe post-14th century decline evident in all the other settlements discussed in this paper. Long Melford, Nayland and Walberswick are all in Suffolk, and all were also strongly reliant on non-agricultural economic activity, which may provide a clue as to their resilience. Detailed discussion of these observations is beyond the scope of this paper, but the significance of the contextualisation provided by the ‘bigger picture’ of the test pit data is evident. The way in which test pit data can contribute to exploring major longue dureé themes in and across different historic periods can be illustrated here (briefly) with reference to some examples. 23

The Archaeology of Medieval Villages Currently Inhabited in Europe

Figure 10. CORS discussed in text showing percentage of habitative pits per historic period against average for all East Anglian CORS included on Table 1.

Early medieval settlement continuity and discontinuity The degree of continuity (or otherwise) between Roman and later settlement and land-use is a widely debated issue (Dark & Dark 1997; Hingley 2007; Rippon 2018) which remains incompletely understood (Oake et al. 2007: 12–15; Wade 2000: 23). It is pertinent to this issue to observe that the East Anglian CORS generally yield notably little Romano-British material, with fewer than 10% producing more than a single sherd. This contrasts starkly with evidence from field-walking in the region which shows a dense pattern of settlement in the Romano-British period (e.g. Davison 1990; Parry 2006; Rogerson et al. 1997). Furthermore, the CORS data shows that when pottery of Roman date is found, this is almost always on the edges of the later settlements, as is the case at Pirton. In fact, in all bar two cases, Romano-British material is almost entirely absent from the locations in which pottery of late Anglo-Saxon or high medieval date is found (Cooper 2013:37). It is well-known that pottery use was less in the early and middle Anglo-Saxon periods (early 5th – early 9th centuries) than in the periods before and after, and this makes activity of this period difficult to explore from test pit excavation. This is frustrating, as there would appear to be a significant shift in settlement location at some point in these centuries, but it is difficult to pin down precisely when this occurs from the test pit data. There appears to be some degree of continuity of site-use from the late Roman into early Anglo-Saxon period, with test pits producing material of the latter date often the same or close to those with Romano-British material. But there is very little spatial correlation between sites yielding

24

Carenza Lewis - Test pit excavations in currently-occupied rural settlements

Romano-British pottery and those yielding later Anglo-Saxon or high medieval pottery. This prompts the inference that settlements which developed from perhaps the 7th – 9th centuries (and certainly from the 9th – 11th centuries) may have deliberately avoided locations which had been inhabited in the Roman period (Lewis 2018b and in prep). The origins of the nucleated village and the medieval development of dispersed settlements Test pit data from CORS can also inform debate around the vexed question of the origins of the nucleated village, which has been extensively researched in the UK (Taylor 1983; Hooke (ed.) 1985; Lewis et al. 1997; Roberts 1987; Roberts and Wrathmell 2002; Jones and Page 2006). With the possible exception of places such as Binham and Gaywood, the East Anglian test pit data show no evidence for large densely inhabited settlements (which might merit the term nucleated) underlying CORS before the late Anglo-Saxon period (Cooper 2013; Lewis 2010) and, more surprisingly, less evidence than might be expected for compact nucleations even in this later period. In most of the East Anglian CORS, the areas from which late Anglo-Saxon pottery has been recovered are limited in extent, suggesting the settlements were either very small at this time (e.g. Coddenham, Great Shelford, Nayland), or remained essentially dispersed (e.g. Sharnbrook, Carleton Rode). Pirton, where there is a large, compact pre-Norman nucleated settlement, appears to be in a minority, at least in this region. It seems to be the case that while the sites of most later nucleated villages carried some form of habitation in the pre-Norman era, it was only from the 12th century that most of these developed into large, densely-packed settlements. The test-pit data also show where, when and how dispersed settlements developed, a subject of considerable recent interest ( e.g. Rippon 2008), with the East Anglian CORS data showing some elements of dispersed settlement patterns to appear in the late Anglo-Saxon period (e.g. Carleton Rode and Chediston), a few earlier ( e.g. Little Hallingbury) and others later (e.g. Clavering). One theme to emerge from this is the potential of the test pit data to inform our understanding of the impact of the 12th and early 13th centuries, when the pottery indicates that many new subsidiary settlements were founded and most existing settlements grew considerably in size and density, at a time when many may also have been re-planned (Blair 2018; Creighton and Rippon 2017). The impact of the Black Death Another long-standing and challenging historical debate which the test pit data can inform is the extent and impact of the demographic downturn of the 14th century, which was caused by a range of factors culminating in the Black Death pandemic of plague in 1346–50 (Campbell 2016; Lagerås 2016). Estimates of the scale of population decline have ranged widely from 23% (Ziegler 1969: 185) to as high as 62.5% (Benedictow 2004: 377). Resolving these substantial differences of opinion has been hampered by the lack of standardized ‘before and after’ demographic data spanning the longue durée, but pottery from test pits can act as a proxy for this. The aggregated data from the East Anglian CORS has shown that the number of pits producing potentially habitative levels of pottery to fall across the region from nearly 40% before the mid-14th century to barely 20% afterwards (Lewis 2016b), thus the test pit data suggest that the population of eastern England region was c. 45% lower in the centuries after the Black Death than it had been before. This figure falls approximately midway between the historically-estimated figures noted above. As well as providing a new source of data for assessing overall demographic change, the test pit data allow the long-term impact of this demographic turbulence on contemporary communities to be mapped at a range of scales, and the effect on each settlement measured and compared with others (Lewis forthcoming 2019). Analysing this date reveals a number of specific responses, including an apparent preference for locations close to churches (especially notable in settlements which contacted most severely after the 14th century) and the post-14th century abandonment of the most recently inhabited parts of settlements (Lewis forthcoming 2018).

25

Table 1. Test pits excavated within CORS in East Anglia by May 2017, listed alphabetically by parish name, showing the number and percentage of excavated pits per parish producing possibly habitative numbers of pottery sherds of Roman, Anglo-Saxon, medieval and post-medieval date. No. pits with 1+ sherds

% of all dug pits with 1+ sherds

Change since previous period in no. of pits with 1+ sherds

2022

8 4 0 12 1 0 2 0 2 2 1 2 2 4 0 5 1 1 1 1 0 0 0 3 1 3 28 0 6 0 4 2 9 19 0 5 0 0 0 1 0 5 0 0 2 3 0 0 0 0 1 2 1 5 0 3 5 5 4 3.5 166

Change since previous period in no. of pits with 1+ sherds

45 50 7 60 14 9 32 57 23 47 33 40 59 34 24 44 39 10 35 41 36 26 33 37 24 54 74 40 32 7 50 24 36 114 26 31 10 31 35 68 19 23 15 16 32 27 31 24 24 33 18 23 42 58 18 34 14 23 57

% of all dug pits with 1+ sherds

Norfolk Herts Suffolk Norfolk Norfolk Suffolk Norfolk Norfolk Cambs Suffolk Suffolk Essex Suffolk Cambs Essex Norfolk Norfolk Cambs Herts Cambs Suffolk Norfolk Norfolk Cambs Cambs Essex Suffolk Essex Cambs Essex Suffolk Norfolk Cambs Herts Beds Cambs Cambs Beds Cambs Beds Beds Beds Suffolk Suffolk Essex Cambs Suffolk Cambs Norfolk Cambs Essex Cambs Suffolk Essex Cambs Cambs Cambs Norfolk Essex

Middle Anglo-Saxon

No. pits with 1+ sherds

Acle Ashwell Bures Binham Blo’ Norton Bramford Brundall Carleton Rode Castor Chediston Clare Clavering Coddenham Cottenham Daw’s Heath Garboldisham Gaywood Girton Great Amwell Great Shelford Hessett Hillington Hindringham Houghton Isleham Little Hallingbury Long Melford Manuden Meldreth Mount Bures Nayland Paston Peakirk Pirton Potton Rampton Ramsey Riseley Sawtry Sharnbrook Shefford Shillington Snape Southwold/Reydon Southminster Stapleford Sudbury Swaffham Bulbeck Terrington St Clement Thorney Thorrington Ufford Walberswick West Mersea West Wickham Willingham Wisbech St Mary Wiveton Writtle Average Total

Early Anglo-Saxon % of all dug pits producing 2-4+ sherds

county

No. pits with 2-4+ sherds

Parish name

Total no. pits dug (to May 2017)

Roman

18 8 0 20 7 0 6 0 9 4 3 5 3 12 0 11 3 10 3 2 0 0 0 8 4 6 38 0 19 0 8 8 25 17 0 16 0 0 0 1 0 22 0 0 6 11 0 0 0 0 6 9 2 9 0 9 36 22 7

0 1 0 2 0 0 1 1 1 0 0 0 6 4 0 1 0 0 0 0 0 1 1 1 0 0 0 1 0 0 0 0 0 1 0 0 0 1 0 2 0 0 0 0 0 0 5 0 0 0 0 0 0 0 0 2 0 0 0 0.7 32

0 2 0 3 0 0 3 2 4 0 0 0 10 12 0 2 0 0 0 0 0 4 3 3 0 0 0 3 0 0 0 0 0 1 0 0 0 3 0 3 0 0 0 0 0 0 16 0 0 0 0 0 0 0 0 6 0 0 0 0 1.6

8 3 0 10 1 0 1 1 1 2 1 2 4 0 0 4 1 1 1 1 0 1 1 2 1 3 28 1 6 0 4 2 9 18 0 5 0 1 0 1 0 5 0 0 2 3 5 0 0 0 1 2 1 5 0 1 5 5 4

0 0 0 6 0 0 0 1 0 0 1 0 3 2 0 1 8 0 0 0 0 2 2 1 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 2 0 0 0 0 0 0.7 33

0 0 0 10 0 0 0 2 0 0 3 0 5 6 0 2 21 0 0 0 0 8 6 3 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 0 0 0 0 0 0 3 0 0 0 0 0

0 1 0 4 0 0 1 0 1 0 1 0 3 2 0 0 8 0 0 0 0 1 1 0 0 2 0 1 0 0 0 0 0 1 0 0 0 1 0 2 0 0 0 0 0 0 3 0 0 0 0 0 0 2 0 2 0 0 0

8.2

-134

1.6

1

of excavated pits per parish producing possibly habitative numbers of pottery sherds of Roman, Anglo-Saxon, medieval and post-medieval date.

-9 -6 -2 -17 0 -1 -3 -13 -7 6 -7 -6 -12 -11 0 -5 -18 -3 -1 -13 -8 -12 -12 -14 -4 -4 11 -4 -8 -4 5 -6 -5 -59 -2 -10 -3 4 -4 -11 -3 -10 -3 0 0 -3 -5 -4 -6 14 -2 -1 7 -6 -7 -6 1 -7 -2

22.7

-321

37 31 4 26 10 7 20 38 9 34 25 30 39 24 9 27 17 2 12 30 22 6 21 21 19 31 44 26 15 5 42 11 11 67 6 19 4 18 19 32 12 14 3 11 26 11 28 19 16 25 12 10 34 21 12 14 6 19 44 25.1 1207

Change since previous period in no. of pits with 2-4+ sherds

569

27 22 0 12 21 22 0 14 17 47 21 20 14 9 4 7 8 0 9 22 19 0 9 24 13 6 31 13 47 29 68 4 17 28 4 16 40 65 14 24 5 17 20 38 22 19 48 33 29 61 11 4 69 10 11 21 21 17 39

% of all dug pits producing 2-4+ sherds

38.6

12 11 0 7 3 2 0 8 4 22 7 8 8 3 1 3 3 0 3 9 7 0 3 9 3 3 23 5 15 2 34 1 6 32 1 5 4 20 5 16 1 4 3 6 7 5 15 8 7 20 2 1 29 6 2 7 3 4 22 9.6 460

No. pits with 2-4+ sherds

17 17 2 17 3 2 2 14 4 14 6 14 10 9 1 3 14 1 4 17 10 3 6 18 7 7 7 5 13 4 27 5 1 58 3 14 7 15 6 23 4 10 6 6 7 6 14 6 9 5 4 0 18 12 9 13 2 10 24

Change since previous period in no. of pits with 2-4+ sherds

47 34 29 40 21 33 9 37 48 34 42 35 34 41 4 18 54 30 11 54 42 46 45 62 29 13 16 23 72 86 58 29 31 80 12 48 70 52 26 40 21 61 40 38 22 30 65 50 54 18 22 9 52 21 50 38 14 48 42

Post-medieval

% of all dug pits producing 2-4+ sherds

179

21 17 2 24 3 3 3 21 11 16 14 14 20 14 1 8 21 3 4 22 15 12 15 23 7 7 12 9 23 6 29 7 11 91 3 15 7 16 9 27 4 14 6 6 7 8 20 12 13 6 4 2 22 12 9 13 2 11 24 16.3 781

No. pits with 2-4+ sherds

4 0 0 1 0 1 1 6 7 2 7 0 7 3 0 4 1 2 0 5 5 13 7 4 0 2 5 4 10 2 2 2 10 33 0 1 0 1 3 4 0 4 0 0 0 2 4 6 4 1 0 2 4 2 0 0 0 1 0

Change since previous period in no. of pits with 2-4+ sherds

9 0 0 12 0 11 3 12 30 4 24 0 17 15 0 11 18 20 0 12 14 58 27 14 0 0 7 10 31 29 4 8 28 29 0 3 0 3 9 6 0 17 0 0 0 7 19 25 17 3 0 9 10 0 0 0 0 4 0 0 10.5

Late medieval

% of all dug pits producing 2-4+ sherds

Change since previous period in no. of pits with 2-4+ sherds

4 0 0 7 0 1 1 7 7 2 8 0 10 5 0 5 7 2 0 5 5 15 9 5 0 0 5 4 10 2 2 2 10 33 0 1 0 1 3 4 0 4 0 0 0 2 6 6 4 1 0 2 4 0 0 0 0 1 0 4.4 212

No. pits with 2-4+ sherds

% of all dug pits producing 2-4+ sherds

High Medieval

No. pits with 2-4+ sherds

Late Anglo-Saxon

82 62 57 43 71 78 63 67 39 72 76 75 66 71 38 61 44 20 34 73 61 23 64 57 79 57 59 65 47 71 84 46 31 59 23 61 40 58 54 47 63 61 20 69 81 41 90 79 67 76 67 43 81 36 67 41 43 83 77

25 20 4 19 7 5 20 30 5 12 18 22 31 21 8 24 14 2 9 21 15 6 18 12 16 28 21 21 0 3 8 10 5 35 5 14 0 -2 14 16 11 10 0 5 19 6 13 11 9 5 10 9 5 15 10 7 3 15 22

59.7

747

The Archaeology of Medieval Villages Currently Inhabited in Europe

Conclusion The above review of the aims and outcomes of test pit excavation in eastern England shows that if appropriate methods are used, pottery from test pits can be dated and mapped to reconstruct the changing spatial extent of habitation in CORS over centuries, in ways that no other data can. Importantly, this enables research-driven archaeological excavation to explore not just the possibly atypical minority of deserted (or very shrunken) settlements, but to include also the majority of settlements which did not become permanently deserted. Changes in the number of sites producing pottery can be used as proxies to reconstruct variation in population levels over long periods of time, allowing the technique to inform broader questions pertaining to settlement, demography and landscape, such as late antique/postclassical continuity and change, the origins of the nucleated village and the impact of the Black Death. Such questions are international in scope: much archaeological fieldwork around the Mediterranean, for example, explores similar issues ( e.g. Hodges et al. 1995; Barker (ed.) 1995; Christie (ed.) 2000; Arthur 2000; Leone & Mattingley 2000; Sanders 2000; Bowden & Hodges 2000) but typically excludes built-up areas, leaving these as black holes in our maps of the past about which little can be said. Although beyond the scope of this paper, it should be noted that test pit excavation is also valuable for management purposes because it can explore areas which are otherwise archaeologically inaccessible. It should also be noted that test pit excavation within CORS has proved to be an effective means of delivering wider social benefits, such as raising academic aspirations and developing transferrable skills and increasing community cohesion (Lewis 2014c; Lewis 2017b) as well enriching the lives of members of the public who participate in the excavations (Alonzo Gonzalez and Fernandez Fernandez 2013; Lewis 2015b). This review of work in eastern England must have implications for other European countries. The archaeological neglect of CORS by medieval settlement researchers is widespread across our continent, but the questions which test pit excavation can address (at scales from local to trans-national) are (as noted above) pertinent across Europe and (given an absence of legal constraints or deeply stratified deposits), the test pit methodology is potentially capable of being carried out in any rural settlement in any country. Test pit excavation has already produced useful results from CORS elsewhere in England (Denison-Edson and Mills 2014; Hurst 2014; Johnson and Howard-Davis 2016) and beyond, including France (Turner & Webster 2012; Turner et al. 2013) and Spain (Fernandez Mier et al. 2014; Narbarte Hernández et. al. 2018). As more work is carried out, data from each additional excavated test pit can be contextualised increasingly meaningfully through comparison with the existing dataset, and also simultaneously, of course, serves to expand that dataset, whether by adding detail to settlements where some test pit excavation has already been carried out, or by expanding coverage into new areas. Furthermore, by involving members of the wider public including resident communities, test pit excavation within CORS can also help to protect resources and build for the future. While heritage protection legislation may currently impede test pit excavation in some countries, the data from eastern England show that the more widely test pit excavation is carried out within today’s rural settlements, the more useful the outcomes are likely to become. Abbreviations CORS: Currently Occupied Rural Settlement DMV: Deserted Medieval Village HHER: Hertfordshire Historic Environment Record (a list of recorded archaeological sites and finds maintained by the local authority, searchable online at https://www.heritagegateway.org.uk/gateway/) SAM: Scheduled Ancient Monument. References Alonzo Gonzalez, P. & Fernandez Fernandez, P. 2013. Rural development and heritage commons management in Asturias (Spain): The Ecomuseum of Santo Adriano. Journal of Settlements and Spatial Planning Special Issue 2: 245–53. 28

Carenza Lewis - Test pit excavations in currently-occupied rural settlements

Arthur, P. 2000. From vicus to village: Italian landscapes, AD 400–1000. In N. Christie (ed.) Landscapes of Change: Rural Evolutions in Late Antiquity and the Early Middle Ages: 103–34. Aldershot, Ashgate. Audouy, M. and Chapman, A. 2009. Raunds: The origins and growth of a midland village AD 450-1500. Oxford, Oxbow Books. Austin, D. 1989. The deserted medieval village of Thrislington, County Durham: Excavations 1973-1974. Society for Medieval Archaeology Monograph Series 12. Aston, M., Austin, D. and Dyer, C. C. (eds.) 1989. The Rural Settlements of Medieval England. Oxford, Blackwell. Barker, P. 1986. Understanding Archaeological Excavation. London, Batsford. Barker, G. (ed.) 1995. A Mediterranean Valley: Landscape Archaeology and Annales History in the Biferno Valley. Leicester, Leicester University Press. Behr, C., Pestell, T. with Hines, J. 2014. ‘The Bracteate Hoard from Binham — An Early Anglo-Saxon Central Place?’ Medieval Archaeology 58: 44-77. Benedictow, O. 2004. The Black Death: the Complete History. Woodbridge, Boydell & Brewer. Beresford, M. W. 1954. The Lost Villages of England. London, Lutterworth Press. Beresford, M. and Hurst, J. (eds.) 1989. Deserted Medieval Villages (second edition). Stroud, Alan Sutton Beresford, M. and Hurst, J. G. 1990. Wharram Percy: Deserted Medieval Village. London, Batsford. Bis-Worch, C. and Theune, C. (eds.) 2017. Religion, cults and rituals in the medieval rural environment. Ruralia (Ruralia 11), Sidestone Press.  Blair, J. 2013. Grid Planning in Anglo-Saxon Settlements: the short perch and the four-perch module. Anglo-Saxon Studies in Archaeology and History 18: 18-61. Blair, J. 2018. Building Anglo-Saxon England. Princeton and Oxford, Princeton University Press. Bowden, W. & D. Hodges, 2000. Balkan ghosts? Nationalism and the question of rural continuity in Albania. In N. Christie (ed.) Landscapes of Change: Rural Evolutions in Late Antiquity and the Early Middle Ages: 195– 222. Aldershot, Ashgate, Brady, N. (ed.). 2019. Transitions and Transformations in the Medieval and Early Modern Countryside. Ruralia (Ruralia 12), Sidestone Press.  Cambridge Archaeological Field Group. 2015. Change in a South Cambridgeshire parish: Understanding the Bronze Age to late medieval settlement within Haslingfield. Proceedings of the Cambridge Antiquarian Society 104: 71-87. Campbell, B. M. S. 2000. English Seigniorial Agriculture 1250-1450. Cambridge, Cambridge University Press. Campbell, B. M. S. 2016. The Great Transition: Climate, Disease and society in the late-medieval world. Cambridge, Cambridge University Press. Christie, N. (ed.), 2000. Landscapes of Change: Rural Evolutions in Late Antiquity and the Early Middle Ages. Aldershot, Ashgate. Christie, N. and Stamper, P. (eds.) 2012. Medieval rural settlement: Britain and Ireland, AD 800–1600, Oxford, Oxbow Books. Chart, D.A. 1940. A preliminary survey of the ancient monuments of Northern Ireland (PSAMNI). Belfast, HMSO. Cessford, C. and Dickens, A. 2005. The Manor of Hintona: the origins and development of Church End, Cherry Hinton. Proceedings of the Cambridge Antiquarian Society 94: 51-72. Christie, N. and Stamper, P. (eds.) 2012. Medieval Rural Settlement: Britain and Ireland, AD 800-1600. Oxford, Oxbow Books. Clelland, S. and Mepham, L. 2014. An Anglo-Saxon Site at Thrapston Road, Spaldwick. Proceedings of the Cambridge Antiquarian Society 103: 117-131. Coles, J. 1972. Field Archaeology in Britain. London, Methuen. Collins, C. 2013. Archaeological Test Pit Excavations in Great Shelford, Cambridgeshire 2006, 2007, 2008, 2010, 2011, 2012 and 2013. Cambridge, Cambridge University Archaeology Department (available online at http:// www.access.arch.cam.ac.uk/reports/cambridgeshire/great-shelford). Collins, C. 2017. Archaeological Test Pit Excavations in Walberswick, Suffolk 2013, 2014, 2015 and 2016. Cambridge, Cambridge University Archaeology Department (available online at http://www.access.arch.cam. ac.uk/reports/suffolk/walberswick). Collins, C., Lewis, C. and Pryor, A. 2017. Archaeological Test Pit Excavations in Nayland. Suffolk 2012 and 2014. Cambridge, Cambridge University Archaeology Department (available online at http://www.access. arch.cam.ac.uk/reports/suffolk/nayland). 29

The Archaeology of Medieval Villages Currently Inhabited in Europe

Collins, C. and Pryor, A. 2017. Archaeological Test Pit Excavations in Sharnbrook, Bedfordshire 2007-2012. Cambridge, Cambridge University Archaeology Department (available online at http://www.access. arch.cam.ac.uk/reports/bedfordshire/sharnbrook). Cooper, C. 2013. Continuity and change in context: evidence for the Roman to Medieval transition from the HEFA CORS Project. Unpublished MSc. thesis, University of Oxford. Cooper, N. and Priest, V. 2003. Sampling a medieval village in a day. Medieval Settlement Research Group Annual Report 18: 53-56. Courtney, P. 1997. Ceramics and the history of consumption: Pitfalls and prospects. Medieval Ceramics 21: 95-108. Creighton, O. and Rippon, S. 2017. Conquest, colonisation and the countryside: Archaeology and the mid11th to mid-12th century rural landscape. In D. Hadley and C. Dyer (eds.) The Archaeology of the 11th century: continuities and transformations. Society for Medieval Archaeology Monograph 38: 57-87. London, Routledge. Cushion, B. and Davison, A. 2003. Earthworks of Norfolk. Dereham, East Anglian Archaeology Report 104. Dark, K. & Dark, P. 1997. The Landscape of Roman Britain. Stroud, Sutton Davison, A. 1990. The Evolution of Settlement in Three Parishes in South-East Norfolk. East Anglian Archaeology 49. Denison-Edson, P. and Mills, A. 2014. ‘The Swaledale Big Dig’. The Archaeological Forum Journal CBA Yorkshire (new series) 3: 65-76. Dixon, P. 1997. A rural medieval settlement in Roxburghshire: excavations at Springwood Park, Kelso, 1985-6. Proceedings of the Society of Antiquaries for Scotland 128: 671-751. Dyer, C. and Everson, P. 2012. The Development of the Study of Medieval Settlements, In N. Christie and P Stamper (eds) Medieval Rural Settlement in Britain and Ireland AD 800-1600. Oxford, Oxbow Books. Dyer, C. and Jones, R. (eds.) 2010. Deserted Villages Revisited. Hatfield, University of Hertfordshire Press. Everson, P. L., Taylor, C. C. and Dunn, C. J. 1991. Change and continuity - Rural settlement in north-west Lincolnshire. London, HMSO. Fenton, P. 1993. An archaeological evaluation of land to the rear of The Fox Public House, Pirton, Hertfordshire. Unpublished excavation report RNO 109. SHT3576 Fernández Mier, M., Fernández Fernández, M. J., Alonso González, P., López Sáez, J. A., Pérez Díaz, S. and Hernández Beloqui, B. 2014. The investigation of currently inhabited villages of medieval origin: Agrarian archaeology in Asturias (Spain). Quaternary International 346: 41-55. Foard, G. 1978. ‘Systematic Fieldwalking and the investigation of Saxon settlement in Northamptonshire’ World Archaeology 9: 357-74. Fridrich, J., Klápštĕ, J., Smetanka, Z. and Sommer, P. (eds.). 1996. Ruralia 1: Conference Ruralia 1_ Prague, 8th – 14th September 1995. Prague, Institute of Archaeology. Fridrich, J., Klápštĕ, J., Smetanka, Z. and Sommer, P. (eds). 1998. Ruralia 2: Conference Ruralia 2_ Spa, 1st – 7th September 1997. Prague, Institute of Archaeology. Gerrard, C. 2003. Medieval Archaeology: understanding traditions and contemporary approaches. London, Routledge. Gerrard, C. and Aston, M. 2007. The Shapwick Project, Somerset: A Rural Landscape Explored. Leeds, Society for Medieval Archaeology. Glasscock, R, (ed.) 1975. The Lay Subsidy of 1334. London, Oxford University Press. Govan, S. (ed.). 2003. Medieval or Later Settlement in Scotland: 10 years on. Edinburgh, Historic Scotland. Hall, D. 1996. The Fenland Project, Number 10: Cambridgeshire Survey, The Isle of Ely and Wisbech. Cambridge, East Anglian Archaeology Report 79. Hartley, F. 1989. The Medieval Earthworks of Central Leicestershire. Leicester, Leicestershire Museums, Arts and Records Service. Haselgrove, C. Millett, M. and Smith, I. (eds.) 1985. Archaeology from the Ploughsoil: Studies in the Interpretation of Field Survey Data. Sheffield, University of Sheffield Department of Archaeology. Hamerow, H. 2012. Rural Settlements and Society in Anglo-Saxon England. Oxford, Oxford University Press. Hayes, P. 1985. ‘The San Vincenzo Survey’ in S. Macready and F. H. Thompson (eds.) Archaeological field Survey in Britain and Abroad. London, Society of Antiquaries. 30

Carenza Lewis - Test pit excavations in currently-occupied rural settlements

Hingley, R., 2007. The Roman landscape of Britain: from Hoskins to today. In A. Fleming and R. Hingley (eds.) Landscape History after Hoskins: Prehistoric and Roman Landscapes 1: 101–13. Macclesfield, Windgather Press Hodges, R. & C. Wickham with G. Barker, 1995. The evolution of hilltop villages (ad 600–1500). In G. Barker (ed.) A Mediterranean Valley: Landscape Archaeology and Annales History in the Biferno Valley: 254–85. Leicester, Leicester University Press. Hodges, R. 1991. Wall-to-wall history: the story of Roystone Grange. London, Duckworth. Hooke, D. (ed.). 1985. Medieval Villages. Oxford, Oxford University Committee for Archaeology (Monograph 5). Horning, A. 2004. Archaeological explorations of cultural identify and rural economy in the north of Ireland: Goodland, Co. Antrim. International Journal of Historical Archaeology 8 (3): 199-215. Hoskins, W. G. 1955. The Making of the English Landscape. London, Hodder and Stoughton. Hurst, D. 2014. Test pitting in Worcestershire in Currently Occupied Rural Settlements. Transactions of the Worcestershire Archaeological Society 24: 279-281. Ivens, R., Busby, P. and Shepherd, N. 1995. Tattenhoe and Westbury: Two deserted medieval settlements in Milton Keynes. Buckingham, Buckinghamshire Archaeological Society Monograph. Jackson, R. and Burleigh, G. 2018. Dea Senuna: Treasure, Cult and Ritual at Ashwell, Hertfordshire. British Museum Research Publication Vol 194. London, British Museum. Johnson, D. and Howard Davis, C. 2016. Long Preston Bid Dig: Report on an archaeological test pitting project in Long Preston, North Yorkshire. Yorkshire Dales National Park Authority Report SYD 14771. Jones, R. 2004. ‘Signatures in the soil: the use of ceramic manure scatters in the identification of medieval arable farming regimes.’ Archaeological Journal 161: 159-88 Jones, R. and Page, M. 2001. Medieval settlements and landscapes in the Whittlewood Area: Interim Report 2001-2. Medieval Settlement Research Group Annual Report 16: 15-25. Jones, R. and Page, M. 2006. Medieval Villages in an English landscape: Beginnings and Ends. Macclesfield, Windgather Press. Kelly, R. S. 1982. The excavation of a medieval farmstead at Cefn Graenog, Clynnog, Gwynedd. Bulletin of the Board Celtic Studies 29: 859-908. Klápštĕ, J. (ed.). 2000. Ruralia 3: Conference Ruralia 2_ Maynooth, 3rd – 9th September 1999. Prague, Institute of Archaeology. Klápštĕ, J. (ed.). 2002. The Rural House, from the Migration Period to the Oldest Still Standing Buildings. Ruralia (Ruralia 4). Turnhout, Brepols Publishers.  Klápštĕ, J. (ed.). 2005. Water Management in Medieval Rural Economy. Ruralia (Ruralia 5). Turnhout, Brepols Publishers.  Klápštĕ, J. (ed.). 2013. Hierarchies in rural settlements. Ruralia (Ruralia 9). Turnhout, Brepols Publishers.  Klápštĕ, J. (ed.). 2016. Agrarian Technology in the Medieval Landscape. Ruralia (Ruralia 10). Turnhout, Brepols Publishers.  Klápštĕ, J. and Sommer, P. (eds.). 2007. Arts and Crafts in Medieval Rural Environment. Ruralia (Ruralia 6). Turnhout, Brepols Publishers.  Klápštĕ, J. and Sommer, P. (eds.). 2009. Medieval Rural Settlement in Marginal Landscapes. Ruralia (Ruralia 7). Turnhout, Brepols Publishers.  Klápštĕ, J. and Sommer, P. (eds.). 2011. Food in the Medieval Rural Environment. Ruralia (Ruralia 8). Turnhout, Brepols Publishers.  Lagerås, P. 2015. Environment, Society and the Black Death: an interdisciplinary approach to the late-medieval crisis in Sweden. Oxford, Oxbow Books. Leone, A. & D. Mattingley, 2000. Vandal, Byzantine and Arab rural landscapes in North Africa. In N. Christie (ed.) Landscapes of Change: Rural Evolutions in Late Antiquity and the Early Middle Ages: 135–62. Aldershot, Ashgate. Lewis, C. 2003. Test Pitting medieval settlements – Big Dig 2003. Medieval Settlement Research Group Annual Report 18: 46-7. Lewis, C. 2005. Test pit excavation within currently occupied settlements in East Anglia in 2005. Medieval Settlement Research Group Annual Report 20: 9-16. Lewis, C. 2006. Test pit excavation within currently occupied settlements in East Anglia in 2006. Medieval Settlement Research Group Annual Report 21: 37-44. 31

The Archaeology of Medieval Villages Currently Inhabited in Europe

Lewis, C. 2007a. New avenues in the investigation of currently-occupied rural settlements – preliminary results from the Higher Education Field Academy. Medieval Archaeology 51: 133-164 Lewis, C. 2007b. Test pit excavation within currently occupied settlements in East Anglia in 2007. Medieval Settlement Research Group Annual Report 22: 48-56 Lewis, C. 2008. Test pit excavation within currently occupied settlements in East Anglia in 2008. Medieval Settlement Research Group Annual Report 23: 60-68. Lewis, C. 2009. Test pit excavation within currently occupied settlements in East Anglia in 2009. Medieval Settlement Research Group Annual Report 24: 43-58. Lewis, C. 2010. Exploring black holes: Recent investigations in currently occupied rural settlements in Eastern England. In N. Higham and M Ryan (eds.) The Landscape Archaeology of Anglo-Saxon England. (Manchester Centre for Anglo-Saxon Studies Series): 83-105. Woodbridge, Boydell and Brewer. Lewis, C. 2011. Test pit excavation within occupied settlements in East Anglia in 2010. Medieval Settlement Research 26: 48-59. Lewis, C. 2012. Test pit excavation within occupied settlements in East Anglia in 2011. Medieval Settlement Research 27: 42-56. Lewis, C. 2013. Test pit excavation within occupied settlements in East Anglia in 2012. Medieval Settlement Research 28: 77-89. Lewis, C. 2014a. The Power of Pits: Archaeology, outreach and research in living landscapes. In K. Boyle, R. Rabett and C. Hunt (eds.) Living in the Landscape: 321-338. Cambridge: McDonald Institute for Archaeological Research Monograph. Lewis, C. 2014b. ‘Test pit excavation within occupied settlements in East Anglia in 2013.’ Medieval Settlement Research 29: 66-81. Lewis, C. 2014c. ‘Cooler than a trip to Alton Towers’: Assessing the impact of the Higher Education Field Academy 2005-2011 in Public Archaeology 13 (issue 4): 295-322. Lewis, C. 2015a. Test pit excavation within occupied settlements in East Anglia in 2014. Medieval Settlement Research 30: 39-49. Lewis, C. 2015b. Archaeological Excavation and Deep Mapping in Historic Rural Communities. Humanities 2015 (4): 393–417. Lewis, C. 2016a. Test pit excavation within occupied settlements in East Anglia in 2015. Medieval Settlement Research 31: 53-62. Lewis, C. 2016b. Disaster Recovery? – New archaeological evidence from eastern England for the impact of the ‘calamitous’ 14th century. Antiquity 90 (Issue 351): 777-797. Lewis, C. 2017. Evidencing the impact of widening participation access programmes: Assessment within the Higher Education Field Academy. Widening Participation and Lifelong Learning 19 (number 2): 87-112. Lewis, C. 2017. Test pit excavation within occupied settlements in East Anglia in 2016. Medieval Settlement Research 32: 70-78. Lewis, C. 2018a. Test pit excavation within occupied settlements in East Anglia in 2017. Medieval Settlement Research 33: 77-88. Lewis, C. 2018b. Missing the Antique? Reconstructing the Roman-Medieval transition in rural settlements in England. Paper presented 6 Sept 2018. Barcelona, 24th European Archaeological Association Annual Conference: Barcelona. Lewis, C. (forthcoming 2018). ‘Reconstructing the impact of 14th century demographic disasters on late medieval rural communities in England’ in Peter Brown, Paolo Forlin and Chris Gerrard (eds.) Waiting for the End of the World: Perceptions of Disaster and Risk in Medieval Europe. Society for Medieval Archaeology Monograph Series, London: Routledge. Lewis, C., Mitchell Fox, P. and Dyer, C. C. 1997. Village, Hamlet and Field. Manchester, Manchester University Press. Lewis, C. and Baillie, B. 2014. Archaeological test pit excavations in West Wickham, Cambridgeshire, 2013. Cambridge, Cambridge University Archaeology Department (available online at http://www.access. arch.cam.ac.uk/). Lewis, C. and Pryor, A. 2014a. Archaeological test pit excavations in Meldreth, Cambridgeshire, 2013. Cambridge, Cambridge University Archaeology Department (available online at http://www.access.arch.cam. ac.uk/). 32

Carenza Lewis - Test pit excavations in currently-occupied rural settlements

Lewis, C. and Pryor, A. 2014b. Archaeological test pit excavations in Shillington, Bedfordshire, 2013. Cambridge, Cambridge University Archaeology Department (available online at http://www.access.arch.cam. ac.uk/). Lewis, C., Ranson, C. and GBAHG 2014. Archaeological test pit excavations in Great Bowden, Leicestershire, 2014. Cambridge, Cambridge University Archaeology Department (available online at http://www.access. arch.cam.ac.uk/). Martin, E. 2012. Norfolk, Suffolk and Essex: Medieval Rural Settlement in ‘Greater East Anglia’. In Christie and Stamper (eds.) Medieval Rural Settlement: 225-248. Macclesfield, Windgather Press. McCarthy, M. and Brooks, C. 1988. Medieval Pottery in Britain AD 900-1600. Leicester, Leicester University Press. Mills, A. D. 1991. Oxford Dictionary of British place-names. Oxford, Oxford University Press. Mortimer, R. 2000. Village Development and Ceramic Sequence: the Middle to Late Saxon Village at Lordship Lane, Cottenham, Cambridgeshire. Proceedings of the Cambridge Antiquarian Society 89: 5-34. Musty, J. and Algar, D. 1986. Excavations at the Deserted Medieval Village of Gomeldon, near Salisbury. The Wiltshire Archaeological and Natural History Magazine 80: 27-169. Narbarte Hernández, J., Rodríguez Lejarza, A., Santeramo, R., Quirós Castillo, J. and Iriarte Avilés, E. 2018. Evidence for ancient occupation in currently inhabited rural settlements: the archaeological project of Aizarna (Gipuzkoa). Munibe Anthropology Archaeology 69 (online) accessed 28/9/2018 at doi. org/10.21630/maa.2018.69.03 Oake, M., Luke, M., Dawson, M., Edgeworth M. and P. Murphy, 2007. Bedfordshire Archaeology. Research and Archaeology: Resource Assessment, Research Agenda and Strategy (Bedfordshire Archaeology Monograph 9). Bedford, Bedfordshire County Council/The Bedfordshire Archaeological Council. Orton, C. 2000. Sampling in Archaeology. Cambridge, Cambridge University Press. Page, M. and Jones, R. 2003a. Medieval Settlements and Landscapes in the Whittlewood Area: Interim Report 2002-3. Medieval Settlement Research Group Annual Report 18, 27-36. Page, M. and Jones, R. 2003b. Medieval Settlements and Landscapes in the Whittlewood Area: Interim Report 2003-4. Medieval Settlement Research Group Annual Report 18, 37-45. Parker Pearson, M. and Schadla-Hall, T. (eds.) 1994. Looking at the land: Archaeological Landscapes in Eastern England. Leicester, Leicestershire Museums, Arts and Records Service. Parry, S. 2006. The Raunds Area Survey. Oxford, Oxbow Books. Pugh, R. B. 1978. England’s Earliest Gazeteer? Historical Research 51 (issue 124): 113–123. Ranson, C. 2013. Archaeological Test Pit Excavations in Ashwell, Hertfordshire, 2011 & 2012. Cambridge. Cambridge University Archaeology Department (available online at http://www.access.arch.cam.ac.uk/reports/ hertfordshire/ashwell). Ranson, C. 2017a. Archaeological Test Pit Excavations in Acle, Norfolk, 2009, 2010, 2011, 2013 and 2014. Cambridge, Cambridge University Archaeology Department (available online at http://www.access.arch.cam. ac.uk/reports/norfolk/acle). Ranson, C. 2017b. Archaeological Test Pit Excavations in Binham, Norfolk, 2009-2013. Cambridge, Cambridge University Archaeology Department (available online at http://www.access.arch.cam.ac.uk/reports/ norfolk/binham). Ravensdale, J. 1974. Liable to floods: Village landscape on the edge of the Fens, AD 450-1850. Cambridge, Cambridge University Press. RCAHMS. 1993. Waternish, Skye and Lochalsh District, Highland Region: An Archaeological Survey. Edinburgh, RCAHMS. RCAHMS. 2002. But the walls remained. Edinburgh, RCAHMS. RCHME. 1975. Archaeological sites in North-East Northamptonshire. London, HMSO. RCHME. 1979. Archaeological sites in Central Northamptonshire. London, HMSO. RCHME. 1981. Archaeological sites in North-West Northamptonshire. London, HMSO. RCHME. 1985. Archaeological sites in South-West Northamptonshire. London, HMSO. Rippon, S. 2008. Beyond the Medieval Village: the Diversification of Landscape Character in Southern Britain. Oxford, Oxford University Press. Rippon, S. 2018. Kingdom, Civitas, and County. The evolution of territorial identity in the English landscape. Oxford, Oxford University Press. 33

The Archaeology of Medieval Villages Currently Inhabited in Europe

Rippon, S., Wainwirght, A. and Smart, C. 2014. Farming Regions in Medieval England: The Archaeobotanical and Zooarchaeological Evidence. Medieval Archaeology 58: 195-255. Roberts, B.K. 1987. The Making of the English Village. London, Longman. Roberts, B. K. and Wrathmell, S. 2000. An Atlas of Rural Settlement in England. London, HMSO. Roberts, B. K. and Wrathmell, S. 2002. Region and Place: A study of English rural settlement. London, English Heritage. Roberts, K. (ed.) 2006. Lost Farmsteads: Deserted Settlements in Wales. York, CBA Research Report 148. Rogerson, A., Davison, A., Pritchard, D. and Silvester, R. 1997. Barton Bendish and Caldecote: Fieldwork in south-west Norfolk. Dereham, East Anglian Archaeology Report 80. Sanders, G.D.R. 2000. Problems in interpreting rural and urban settlement in southern Greece, AD 365– 700. In N. Christie (ed.) Landscapes of Change: Rural Evolutions in Late Antiquity and the Early Middle Ages: 163–94. Aldershot, Ashgate. Taylor, C. C. 1983. Village and Farmstead. London, George Philip. Taylor, C. C. 1989. ‘Whittlesford: Study of a River-edge Village. In M.A. Aston, D. Austin C.C. Dyer (eds.) The Rural Settlements of Medieval England: 231-246. Oxford, Oxbow Books. Thomas, S. 2010. Thomas, L. 2010. Community Archaeology in the UK: Recent Findings. Unpublished Report by the Council for British Archaeology, available at http://www.sccjr.ac.uk/publications/ community-archaeology-in-the-uk-recent-findings/ (accessed 19 May 2017) Thompson, F. H. 1960. The deserted village of Riseholme near Lincoln. Medieval Archaeology IV: 95-108. Turner, S. and Webster, J. 2012. Medieval and later settlement around Chessenon (Charente), France fieldwork in 2011. Medieval Settlement Research 28: 60–66. Turner, S., Webster, J. and Duggan, M. 2013. Medieval and later settlement around Chessenon (Charente), France: fieldwork in 2012. Medieval Settlement Research 28: 93–5 Wade, K., 2000. Anglo-Saxon and Medieval (Rural). In N. Brown & J. Glazebrook (eds.) Research and Archaeology: a Framework for the Eastern Counties 2 - Research Agenda and Strategy: 23–26. Norwich, Scole Archaeological Committee for East Anglia (EAA Occasional Paper 8). Welldon Finn, R. 1973. Domesday Book: a guide. Chichester, Phillimore Press. Williams, A. and Martin, G. (trans. and ed.). 2003. Domesday Book: a complete translation. London, Folio Society. Williamson, T. 2003. Shaping Medieval Landscapes: Settlement, Society, Environment. Macclesfield, Windgather Press. Wrathmell, S. (ed.). 2012. A history of Wharram Percy and its neighbours. York, York University Archaeological Publications. Wright, D. 2015. Middle Saxon Settlement and Society. Oxford, Archaeopress. Yeoman, P. 1997. Medieval Scotland. Edinburgh, Historic Scotland. Ziegler, P, 1969. The Black Death. Stroud, Sutton Publishing.

34

Village Archaeology in France. A twenty-five year retrospective Edith Peytremann Introduction The archaeology of the medieval village, in France and indeed throughout Europe, has focused mainly, from the 1970s until the present, on deserted villages, regardless of their date of abandonment (École pratique des hautes études. Section Sciences économiques et sociales 1965). It is from these results, with the addition of textual analyses and the memory of the village as they were before the Second World War, that archetypes of the medieval village have been constructed, and, more recently, those of the early Middle Ages (Peytremann 2014). The gradual expansion to a larger time scale and to a wider spatial scale has allowed researchers to revise their interpretative framework and consider the village not as a fixed entity responding to an unchanging definition but as an entity in constant development progressively adapting to the socio-economic and ecological needs of the populations it supports. It remains for us to understand how these developments occurred and why they sometimes failed. An historiography of archaeology within the village will be our first concern, then we will discuss briefly how French archaeology functions; some examples of investigation within a village setting follow. Finally, we will offer a synthesis from two perpectives: the methodology implemented and the results obtained, followed by an overview of the projects in progress or to come, showing in particular the impact on village archaeology on the various laws concerned with environmental impact and town planning. Brief historiography of village archaeology in France Some isolated and pioneering experiments took place in the 1950s in France. The work of Arthur Stieber in Alsace can be cited in example. A researcher at the CNRS from 1952 to 1961 (Schnitzler et al. 2015), Stieber opened a number of trenches in several Alsatian villages such as Furdenheim (Schnitzler et al. 2015) (Figure 1). Although the results were noted in the archaeological chronicles of Gallia or in short articles appearing in the Cahiers d’Archéologie et d’Histoire d’Alsace (Stieber 1954), Stieber, has never published a synthesis of his excavations in the villages. Better known is the work of CNRS archaeologist Claude Raynaud who, in the 1980s, conducted a series of excavations within and around the Occitanian village of Lunel-Viel (Gard), resulting in several publications (Raynaud 1990; Raynaud 2007). In the specific case of Lunel-Viel, a multi-year coordinated research programme studied the impact on the habitat, the settlement and the development of land-use. The principal results indicate that a settlement, which remains occupied to this day, first emerged in the 1st century BC. It experienced several phases of desertion, particularly during the ancient period. Fifteen years of excavations have allowed the discovery of several residential areas and three necropoleis outside and within the currently-inhabited village. Many questions remain unresolved, in particular concerning the location of the Carolingian centre though the Saint-Vincent church, with a surrounding ditch, known in the centre of the current village, offers a clue. To the example of Lunel-Viel may be added the village of Vieux (Calvados), which since 1987 has been the subject of intensified surveillance. This watching brief has led to numerous trial trenches and excavations, sometimes concerning very small areas. A first synthesis article appeared in 2013 (Hincker 2013), which shows the evolution of the village from its origins as an ancient settlement and then its bipolarization into two settlement zones associating funerary areas and churches, both located on the periphery of the old centre. During the 1990s, the process of archaeological investigation in currently-inhabited villages in France has developed. Nevertheless, they remain few in number compared with excavations of deserted village sites, even though researchers like P. Périn and Cl. Lorren have since 1993 advocated such research (Lorren et al. 1995). A first experiment was conducted in the North of France, following the creation of 35

The Archaeology of Medieval Villages Currently Inhabited in Europe

1 10 9 11 8 2

Hauts-de-France 24 6 4 5 13 14 Ile10de- 1 9 11 France 2 12

18

Normandie B retagne

3

Grand-Est

8

25

20

Pays de la Loire

Hauts-de-France 18

19 B retagne

15

16

Centre-Val-de-Loire 24 12

17

4 13

21 Normandie

15

6 5 14 IledeFrance

Bourgogne-Franche-3 Grand-Est Comté

22

25

20

Pays de la Loire 19

16

Centre-Val-de-Loire

17

Bourgogne-FrancheComté

21

Nouvelle-Aquitaine

Auvergne-Rhône-Alpes

22

Auvergne-Rhône-Alpes

Nouvelle-Aquitaine

Occitanie

7

Occitanie

7

ProvenceCote-d’Azur ProvenceCote-d’Azur

23 23

© E. Peytremann

© E. Peytremann

1: Dechy (Nord)(Nord) , 2:, 2: Guesnain Furdenheim (Bas-Rhin), 4: 1: Dechy Guesnain (Nord), (Nord), 3:3:Furdenheim (Bas-Rhin), 4: Roissy-en-France (Val-d’Oise),5:5:Tremblay-en-France Tremblay-en-France (Seine-Saint-Denis), 6: Roissy-en-France (Val-d’Oise), (Seine-Saint-Denis), 6: Louvres (Val-d’Oise), 7: Lunel-Viel (Hérault), 8: Lauwin-Planque (Nord), 9: Louvres (Val-d’Oise), 7: Lunel-Viel (Hérault), 8: Lauwin-Planque (Nord), 9: Flers-en-escrebieux (Nord), 10: Cantin (Nord), 11: Auby (Nord), 12: Villiers-le-Bel Flers-en-escrebieux (Nord), Cantin (Nord),14:11:Gagny Auby(Seine-Saint-Denis), (Nord), 12: Villiers-le-Bel (Val-d’Oise), 13: Drancy10: (Seine-Saint-Denis), 15: (Val-d’Oise), 13: Drancy (Seine-Saint-Denis), 14: Gagny (Seine-Saint-Denis), 15: Yutz (Moselle), 16: Bourogne (Territoire de Belfort), 17: Joué-Lès-Tours (Indre-et-Loire), Vieux (Calvados), 19: Vauchrétien (Maine-et-Loire), 20: Yutz (Moselle), 16:18:Bourogne (Territoire de Belfort), 17: Joué-Lès-Tours Bonnée (Loiret), 21: Esvres (Indre-et-Loire), 22: Nohant-Vic (Indre), 23: (Indre-et-Loire), 18: Vieux (Calvados), 19: Vauchrétien (Maine-et-Loire), 20: Villongue-de-la-Salanque (Pyrénées-Orientales), 24: Couloisy (Oise), 25: Bonnée Bréal-sous-Vitré (Loiret), 21:(Ille-et-Vilaine). Esvres (Indre-et-Loire), 22: Nohant-Vic (Indre), 23: Villongue-de-la-Salanque (Pyrénées-Orientales), 24: Couloisy (Oise), 25: Fig. 1. Location of sites mentioned in the article Figure 1. Location of sites mentioned in the article Bréal-sous-Vitré (Ille-et-Vilaine). Fig. 1. Location of sites mentioned in the article 36

Edith Peytremann - Village Archaeology in France. A twenty-five year retrospective

an archaeology service attached to the city of Douai. Several villages (Lauwin-Planque, Dechy, Guesnain, Flers-en-Escrebieux and Auby; Figure1) were selected for archaeological survey in the form of energy network monitoring, test pits or area excavations (Willot et al. 2013). The first results for the villages of Guesnain and Dechy were published in 2002 (Compagnon et al. 2002). The benefit here is the show two unique evolutions occuring in villages with different grouping phases. This experiment is still on-going. Experiments in the Paris region are different in that most of the excavations within villages were not integrated into a research programme; they were instead due to chance deriving from construction work in the villages, mainly during the building of new infrastructures. Several villages have thus been the object of this sort of archaeological excavation. Examples include Roissy-en-France, Tremblay-enFrance, Villiers-le-Bel or Louvres (Figure 1). The creation of departmental archaeological services in the 1990s, sometimes taking over from important local associations, also contributed to the archaeological monitoring of certain villages, such as in Seine-Saint-Denis with the villages of Drancy, Tremblay-enFrance and Gagny (Gonçalves-Buissart et al. 2012). This work resulted in excavation reports and notices published in the Collectif Research Project on Rural Settlement in the Île-de-France (Gentili et al. 2003; Gentili et al. 2009). There have been, however, some more substantial articles published recently, in particular on Roissy-en-France (Dufour 2013) or Villiers-le Bel (Gentili 2014). In Lorraine, archaeological monitoring of the commune of Yutz (Figure 1) began in 1989 within the village and in the surrounding area, in order to trace the evolution of the village in its territory. An article of synthesis published in 1999 focuses on the long-term evolution and land-use of this terroir (from the 10th BC to the 20th century AD) (Blaising 2000). This innovative publication is to first to approach the study of the village from a long-term perspective with extensive use of palaeo-environmental data, in particular the carpology and the palynology. Such a comprehensive synthesis was possible because most of the archaeological work was directed by the same archaeologist. In Alsace, some excavation, trial-trenching and energy network monitoring have also taken place in inhabited villages, mainly since the 1990s. Each village shows a different evolution, some with origins dating back to the Neolithic, others apparently being created at the beginning of the Merovingian period (Peytremann 2013a). Interest in the history of the village intensified at the beginning of the 21st century, as indicated by the 2003 publication of a research program on the morphogenesis of medieval agglomerations (Gauthiez et al. 2003). Archaeology, however, is more rarely called upon for this kind of study of villages than of towns and the morphological approach, using planimetric documentation to understand the formation of agglomerations, is mostly highlighted. Archaeological investigations continue today in villages in more and more regions, but theoretical discussion of the process of village formation and synthetic publications are lacking. Since 2004, excavations in the village have become commonplace, as can be seen in Franche-Comté in the village of Bourogne (Figure 1) (Billoin et al. 2006) and in the Centre region. These usually consist of trial-trenching, more rarely of area excavations, and these investigations are usually the result of chance, without real follow-up work in particular villages (archaeological zoning; see below). A round table held in 2007 in Saint-Germain-en-Laye on the theme ‘Archaeology of the village, Archaeology in the village’ marks an important milestone in France since it draws attention towards archaeological investigations previously judged unimportant because they usually concern only small areas. This round table brought together archaeologists working for the state services, for local authorities (departments, regions, municipalities), or for universities as well as others employed by the French National Institute for Preventive Archaeology (INRAP). This resulted in a collective article, indeed a manifesto, advocating the recognition of village archaeology, in all its aspects, as a field in its own right. This was published in 2009 in the journal Les nouvelles de l’archéologie (Carré et al. 2009). The round table proceedings themselves were later published in 2013 (Mahé-Hourlier et al. 2013).

37

The Archaeology of Medieval Villages Currently Inhabited in Europe

Since this round table, there has been an upsurge in the number of archaeological excavations within villages everywhere in France; for the most part these are not integrated into a specific research programme, with the exception of a programme in the Centre region and another in Languedoc-Roussillon in southern France which will be presented below. To illustrate this upsurge, mention may be made of the research programme1 initiated in the Hauts-de-France region in the departments of Aisne and Oise (Hugonnier et al. 2014, 58). Although this program is not exclusively devoted to the results of excavations in villages, it is noteworthy that in a corpus of 31 excavated medieval rural sites more than 25% of them are located in the centres of the villages (Hugonnier et al. 2014, 58). The surface area of excavations in villages varies from 80 to 8000m². Among publications reporting this dynamic is a book published in 2012 entitled Église, cimetière et paroissiens. This follows up on a preventive excavation that took place in the village of Bréal-sous-Vitré in Brittany (Colleter et al. 2012). Concerning a small area of (250 m²), this excavation disclosed successive modifications in the use of this space: from the 7th to the 11th century the use was funerary and probably religious; from the end of the 11th to the end of the 13th century the use was indeterminate; then funerary again until the second half of the seventeenth century. During the 18th century, the space was briefly used for casting bells and then transformed into a garden which lasted until the beginning of the 21st century. Anthropological and archival studies allow the specific interpretation the origin of the parish centre. This book is one of the few monographs now available summarising research regarding a currently-inhabited village. Two years later an issue of the journal Archeopages was entirely dedicated to the village. The articles here compiled and the discussions represent a real advance in the study of villages over their long term, for they take into account occupation from the Neolithic era, Protohistory or Antiquity as well as medieval and post-medieval times. At least three articles in this issue report on investigations in other present-day villages. The following year, an important article on the village of Joué-Lès-Tours (Figure 1), which benefited from two archaeological excavations, was published (Papin et al. 2015). Finally, a two-day seminar on ‘Evaluating the Village’ was organized by Jean-Philippe Chimier, Nicolas Fouillet and Stéphane Joly at the University of Tours early in 2016, (http://citeres.univtours.fr/spip. php?article2353). Interest in this research field and discussion of how best to frame it were further advanced in a symposium held in 2017 in Perpignan on the theme of ‘Archaeology in the village: the village and its transformations, from the Middle Ages to the modern cadastre’. How archaeology functions in France and how this affects village archaeology To understand why archaeology in currently-inhabited villages is weak, it is necessary to mention quickly the operating method of French archaeology and its various actors. In France archaeology is subject to two types of operation: preventive archaeology and planned archaeology. Preventive archaeology is dependent on spatial planning works (construction of roads, allotment, railways tracks, etc.). It is predominant in France. Different operators can be involved: public bodies such as the French national Institut for Preventive Archaeological Research (Inrap), archaeological services owned by authorities (city, departments and regions) or private bodies, since 2003, such as the Eveha, Archéodunum, Archéopole private archaeological companies. Planned archaeology depends generally on a research program led by university archaeologists or those working for the CNRS or the state service. Preventive archaeology is governed by a 2001 law that was amended in 2003 (Anon 2016a). In this law, it is specified that it is the State (Anon 2016a) that decides when archaeological surveys must be carried out ahead of infrastructure work, and following the results of these tests, if there must be an excavation. The This program entitled ‘Habitats and territories, medieval occupations, Aisne and Oise’ corresponds to a collective action of research carried out within the framework of Inrap with researchers of all institutional origins. 1 

38

Edith Peytremann - Village Archaeology in France. A twenty-five year retrospective

Regional archaeology service (RAS): - receives construction permits automaticly if the area is > 3 hectares - decides the realisation of preliminary evaluation: - depending on different criteria - depending archaeological zoning if the area is < 3 hectares

French national Institut for Preventive Archaeological Research or archaeological services of regional authorities carry out the test pits (approximatly 10% of the concerned area). This intervention is financed though a royalty (0,53€ about m²)

Regional archaeology service : - tests are negative building can begin - tests are positive: RAS decides the realisation or not of a excavation after consulting the commission of external experts. If there is a excavation:

Planning authoritie: chooses the operator to carry out the excavation according the guidelines of the RAS. Fig. 2. French functioning of preventive archaeology

Figure 2. French functioning preventive archaeology

preliminary evaluations cover approximately 10% of the area covered by the developments. For all works on an area greater than 30,000m², all requests are sent to the Regional Archaeology Service, which may or may not require surveys according the presence of a known archaeological site, and/or the sedimentary, topographic and geographical context. In the case of works on a smaller area, this procedure depends on the region. The regional archaeological services (state) can indeed decide on an archaeological zoning (Anon 2016a) in a city, a village or around an important archaeological site. There may be several levels, between 0 and 30,000m², within this zoning, which govern the conducting of archaeological test pits. The cost of the surveys is financed by a charge2 payable by the developers (Figure 2). The villages which are subject to zoning are not very many in the country and often only the area around the church or castle is concerned. This provision refers to the assumption that the church or the castle, are located at the source of the village and they are the centre. It can be seen from the legislation that it is difficult to require access to excavate test pits in currently-inhabited villages, particularly if the area subject to the forthcoming construction is small. However, the situation is now changing, thanks to the implementation of a new law. Indeed, since a change to national planning law was instituted to protect agricultural land within a broader environmental policy framework,3 construction work has moved back towards the centre of the villages. The charge is calculated according the area covered by works from 3000 m² in application of the article L.524-7-II of the heritage Code. 3  Grenelle de l’environnement 2007-2010 : LOI n° 2009-967 du 3 août 2009 de programmation relative à la mise en œuvre du 2 

39

The Archaeology of Medieval Villages Currently Inhabited in Europe

Moreover, the 2005 Disability law,4 requiring the accessibility of all public buildings, has resulted in construction work associated with churches that affect the underground levels. These innovations have led the State Regional Archaeological Services to adjust their zoning policy. In the next few years, there should be more zoning, leading to the conducting of surveys on small areas and in the long term the excavation of villages. This new feature also poses the problem of financial resources required to pay for more extensive surveys and which exceed the price set by the development charge, and it also poses the problem of the methods to be implemented: French village archaeology is booming! Some recent examples of archaeology in currently-inhabited villages Since the 2000s, experimental programmes have been carried out in several villages in the Centre region. Some involve deliberate excavation and several years of follow-up in the same village, others have been opportunistic interventions. Since 2010, in any event, the Regional Archaeology Service has had a more prescriptive policy in regard to the medieval churches within the villages as part of a funerary topography project. Around forty interventions (energy network monitoring, test pits and excavations) were counted in 2016; this led to the discovery of 26 sites which provided data on funerary topography as well as on the age of the boroughs. We comment below on two examples from this regional program, and then on others from further afield. The village of Bonnée (Loiret)5 Bonnée is a small village with about 700 inhabitants, located on a slight relief framed by former channels of the Loire 46km from Orléans (Figure 1). The conservation of the archaeological layers is good. The only previously-known archaeological vestiges in the village are the remains of an ancient amphitheatre discovered in the 19th century. Since 2001, the regional service established a ‘watching brief ’ zone in the village when the municipality announced development plans. Currently, no old building stands in the village. Since 2001, seven preliminary evaluations have been carried out on the village centre (Figure 3), corresponding to 8.4ha of evaluated land or 8834m² where the topsoil was stripped (Joly 2009 ; Joly 2014). They were all positive. No excavations, however have been undertaken. In the majority of cases, the Regional Archaeology Service estimates that the data collected were sufficient, in one case the developer did not follow up on his development intention. The results after fifteen years of follow-up indicate that the data collected vary in quality and that post-excavation studies of finds are lacking; palaeoenvironmental studies were rarely carried out and topographic data are unreliable in the absence of systematic geo-referencing. On the other hand, knowledge about the development of the village has increased and it is possible to restore some of it. The village dates back to the Bronze Age, but its topographical evolution remains difficult to understand in the light of premature data analysis. It does seem that the centres and peripheries of the different settlements have shifted over time, to judge from the location of the ancient agglomeration. Various types of remains were discovered providing evidence for domestic structures, funerary activities and public use. Because it was impossible to synthesize all the data, in particular topographical data, a project for the creation of GIS documentary studies and artefacts studies was set up. It is important to emphasize that this project is primarily linked to the personal involvement of an Inrap researcher, Stéphane Joly, and that it does not depend on or respond to a research policy defined by the Culture or Research Ministry. Grenelle de l’environnement NOR: DEVX0811607L, II, section 1, art. 7 et LOI n° 2010-788 du 12 juillet 2010 portant engagement national pour l’environnement (1) NOR: DEVX0822225L  4  Loi n° 2005-102 du 11 février 2005 pour l’égalité des droits et des chances, la participation et la citoyenneté des personnes handicapées  NOR: SANX0300217L  5  All the data and illustrations concerning Bonnée were provided to us by Stéphane Joly, Inrap whom we thank warmly.

40

Edith Peytremann - Village Archaeology in France. A twenty-five year retrospective

Figure 3. Location of test pits areas on the 19th cadastre of Bonnée (Loiret), © S. Joly Figure 3. Location of test pits areas on the 19th cadastre of Bonnée (Loiret), © S. Joly

The village of Nohant-Vic (Indre) Nohant-Vic is a small village located about thirty kilometres south-east of Chateauroux (Figure 1). It currently has 500 inhabitants. The current village results from the grouping of the villages of Nohant and Vic in the nineteenth century. The archaeological context here is particularly rich for both former boroughs and the territory separating them. This rich context, and in particular the presence of a church attested from the eleventh century in the village of Vic, determined the decision by the Regional Archaeology Service to establish a watching-brief in this borough. Recent research (surveys, energy network monitoring and excavations) has focused on the former borough of Vic (Figure 4). Occupation in ancient times (1st and 2nd centuries AD) is attested by remains of walls, a well, pits and levels of embankments. These remains are insufficient to characterize the occupation. A second antique occupation, dated this time from the Late Empire, was also identified. Here again the vestiges discovered (level of embankments and walls) do not make it possible to understand its nature. Southeast of the church, a Merovingian funerary area was identified but cannot be precisely located. Numerous burials from the 11th-12th centuries, from the 13th-15th centuries and the 19th century show that an extensive 41

The Archaeology of Medieval Villages Currently Inhabited in Europe

excavation areas

0

1/1000

20 m

Figure 4. Vic (Indre): location of excavation areas since 1998 on the current cadastre plan

Figure 4. Vic (Indre): location of excavation areas since 1998 on the current cadastre plan

cemetery developed around the church at least 80m to the southeast. A funerary area more specifically dated from the 13th-15th centuries was observed more than 100m to the south-east of the church. The textual records indicate that in 1821 a new cemetery was put into service east of the road crossing the village and went out of use in 1861 (Pichon 2011).

42

Edith Peytremann - Village Archaeology in France. A twenty-five year retrospective

The village of Tremblay-en-France (Seine-Saint-Denis)6 (Figure 1) Tremblay’s case is unique in the sense that there have been archaeological excavations on the territory of the village since the 1970s, first conducted by an association of amateur volunteers, and then from 1991, continued by the archaeological service of the department (cf. above). After 20 years of research (Figure 5), it is possible to offer a synthesis, both documentary and archaeological (Lafarge et al. 2009). The archival studies establish that Tremblay was a lordship of the abbey of Saint-Denis. Occupation here, which began at the end of the 5th century, developed around a road junction. A settlement was composed of small agricultural units on elongated plots separated from each other by small ditches. These plots form a loose group amid which was installed, at a yet unknown date, a Merovingian church and a cemetery; later by the 9th century at least, a manor house was built. From the 13th century onwards, the number of land parcels diminished and buildings were added along roadside. The Tremblay excavations have led to several instances of follow-up: a documentary analysis,7 a study of building techniques, artefact studies (ceramics, small finds), and a study of fauna. So far no overall synthesis has been published,8 but site records are accessible (Gonçalves 2003c; Gonçalves 2003a; Gonçalves 2003b; Lafarge 2003). The village of Vauchrétien (Maine-et-Loire) The example of Vauchrétien is characteristic of the fate of most surveys in a village environment. A small village, Vauchrétien is located about 16km south-west of the town of Angers (Figure1), in a region characterized by dispersed settlement. Work on the development of the village centre, near the church, was the reason for the archaeological trial trenches. The oldest preserved part of the church is attributed to the 10th century. Surveys conducted in 2013 constitute the first archaeological intervention in the centre of the village. All the trenches (Figure 6) have disclosed ancient structural features such as ditches, a destruction level, pits and various medieval vestiges. The latter include part of a cemetery, pit-features and a well filled up during the 13th century (Peytremann 2013b). When the regional Archaeological Service made development contingent on excavation, the project was cancelled as too costly. Summary of achievements and ongoing projects What methodologies are best suited to study the history of villages ? The various examples considered above highlight several methodological issues which affect both the fieldwork and data analysis. From the standpoint of archaeological interventions: We have considered instances when archaeological excavation concerned mainly small areas and consequently it took many years before a more general interpretation could be attempted. (see Tremblayen-France, Yutz, etc.). This approach, dependent on the medium and long term, is actually similar to what often happens in urban archaeology, where only the accumulation of data in the long term makes it possible to interpret the historical development of the city. The narrow dimensions of the land-parcels under study sometimes makes the use of mechanical excavators impossible and thus requires other technical aproaches. Drilling cores will sometime overcome this difficulty. This method of geotechnical prospection, also used in urban areas, was helpfully employed in the village of Villelongue-les-Salanques (see below). Another non-destructive geotechnical method, already tried in urban areas, might also useful in villages. It consists of measuring the compactness of the subsoil using a light dynamic penetrometer PANDA®. These density measurements make it possible to estimate the thickness of the archaeological layers and to propose hypotheses about their nature (Galinié et al. 2003). So far, this technique has not yet been used in currently-occupied villages. The methodological similarity of excavations with complex and compacted stratigraphies from the urban 6  7  8 

All data and illustrations concerning Tremblay-en-France were provided to us by Cristina Gonçalves and Ivan Lafarge. Including that of a land register from 14th century A project of synthesis by C. Gonçalves is ongoing.

43

The Archaeology of Medieval Villages Currently Inhabited in Europe

PETIT TREMBLAY

21

Terre Rouge

23

De

ne

Ru

e

ne

16

uv

e

9

11

7

Les Petites Vignes

Juiv

ie

PETIT TREMBLAY 22

20

19

rue

12

des

Vign

ay bl em Tr

ˆ

n ti ar mm Da

es

C

Gr a n d e R u e

Le Grand Pré h

e

m

13

i

12

de

in em Ch

er

St

Go

lay

la

de

de

mb

de

in

ou

re àT

e

C

m he

s ni

sse

ru

La Croix Pigeot ou le Gandenoux

n d

e la

GRAND TREMBLAY P

i s

18 s

o

t t e

La Rue de Boulangerie

5

6

15 17

Pl ace

Ru

e du za

rt

T

re

m

b

la

y

8

ˆ

500 m

d

P



Organization Year(s)

23

CG93

2010

Type of Appelation archaeological work test pits La Croix Pigeot

22

CG93

2010

test pits

2007/2008

test pits excavation test pits

a

ha

e

ss

au

4

Ru

et

0

19&20 CG93

2

3

de

S Le

e

an

N

Ru

24

et

Gr

Building Test pits Excavation

pp

s

25

1

u Cr

it

14

pu

10

e G ra n d e R u

La Pissotte

v

e

V

le il

p

in

te

é

Site Settlement

Protohistory Ancient 6th-10th 11th-13th 14th-15th 16th-18th times

Indivision Popot, chemin des Settlement Saints-Pères & La Maladrerie, terrains du Figaro Settlement

5

CG93

1992

21

CG93

2009/2010

1

JPGF

1988

test pits excavation excavation

Chemin de la Pissotte

2

JPGF

1991/1992

excavation

Ferme Cuypers

Settlement

8

CG93

1994

test pits

Route de la Raperie - IUT

Plotting

4

JPGF

1992

excavation

Le Clos Saint-Charles

Settlement

11

CG93

1998

test pits

Ruelle de la Détourne

Settlement

16

CG93

2005

test pits

Ruelle de la Détourne

Cimetery

17

CG93

2008

3

JPGF

1992

energy network Place de l’église monitoring excavation Fontaine Gancien

6

CG93

1993

excavation

10

CG93

1998&2007 excavation

& Rue Jules Cimetière Rue Cruppet

Guesde,

Plotting rue

du Settlement Settlement

burials Settlement

Chemin de la Pissotte

Settlement

Allée des Tilleuls

Settlement Settlement Settlement Settlement

12&13 CG93

1997/1998

14

CG93

1999

Test Château Bleu pits&evaluation excavation Rue des Fossés

15

CG93

2000

excavation

Rue Louis-Eschard

24

CG93

2012/2013

& Rue de Chalmassy

Settlement ?

25

CG93

2013

test pits excavation test pits

CG93

1996

test pits

7

CG93

1994

18

CG93

2009

8-10 rue des Fossés, allée du Genévrier Ruelles du Cimetière et de la Boulangerie energy network Rue du Cimetière - impasse de la monitoring Poste test pits Rue Louis-Eschard, rue Gosse

Settlement ?

9

Colluvial deposits Plotting Farm

Fig. 5. Location of test pits and excavations on the 19th Cadastre of Tremblay-en-France (Seine-Saint-Denis)

Figure 5. Location of test pits and excavations on the 19th Cadastre of Tremblay-en-France (Seine-Saint-Denis)

44

Edith Peytremann - Village Archaeology in France. A twenty-five year retrospective

Figure 6. Vauchrétien (Maine-et-Loire): location of the test pits on the current cadastre plan

Figure 6. Vauchrétien (Maine-et-Loire): location of the test pits on the current cadastre plan

context with long-occupied sites within villages suggest that this approach might be productive. Indeed, villages sometimes contain strata with organic levels analogous to the ‘dark earth’ of urban contexts. Close study is called for here. These stratigraphic sequences within core areas of inhabited villages are often different to those in rural sites outside them, which are less dense and complex.

45

The Archaeology of Medieval Villages Currently Inhabited in Europe

The collection of soil samples for the purpose of palaeo-environmental studies, though still insufficiently practiced, is essential to understand the topographical development and the uses of the land of the village: buildings, crops, grazing, funerary, storage, etc. Unlike the Spanish experiments (Fernández Mier et al. 2014), there have been few studies in France on the cultivated areas associated with the village. Palynological and physicochemical analyses of the soils are virtually non-existent, and radiocarbon dating remains rare. The work of Nicolas Poirier (Poirier 2013) in Berry and that of Jean-Marie Blaising (Blaising 2003) in Lorraine nevertheless deserves mention. Another methodological point worth stressing concerns the geo-location of the data from a GIS perspective. Geo-referencing is essential for medium- and long-term research, as is the archiving of excavation data. Studies of all of the archaeological material, whether from diagnostics or excavations, must be carried out systematically not only in order to obtain dates but also to understand how the different parts of the village developed and were used. Such studies also provide information needed to help us understand the role of the village in the larger territorial network. The archaeological study of buildings, now greatly developed in urban environments, is too little practised in regard to villages. It is needed to complete the data on the development of the built environment within the village. A final remark: the optimal use of all these methods to maximise the collected data regarding the history of the village supposes that the diagnostic data (obtained from ground survey and test-pits) diagnoses are treated in the same way as those obtained from fuller excavations. This is not or has not always been the case in France. Optimal analysis assumes that critical studies can be undertaken as trial-trenching is completed, in particular palaeo-environmental and anthropological analyses. Respect for such protocols would require a flexibility of implementation not always compatible with the finances available and with administrative functioning, in particular as Inrap currently carries out the archaeological surveys. From the perspective of data analysis: The most successful examples (Tremblay-en-France, Dechy, Yutz, etc.) show that studies of written sources and old maps and plans are essential components of research regarding village evolution. These studies must include the geo-referencing of documents, within the limits of what is possible, so that they can subsequently feed into the GIS, which, while serving as a working tool, also facilitates the archiving of the data. As in the case of urban archaeology, the analysis of village development should no longer focus on specific time periods; it should embrace the analysis of the long-term village development which in turn will involve interdisciplinary analysis. The establishment of a methodological protocol for the implementation of field interventions appears essential to collect and organize all data. This is the only way to compare village evolutions at various local, regional and other levels and to understand what the factors which drive these evolutions and who actors are. First results Looking back over some twenty years, let us consider what results have been achieved. They are not all of equal value. Some are important for our general understanding; others pertain more to regional development. In this article, we will consider only results of more general scope. The data deriving from excavations in the currently-occupied villages do not support the arguments developed by Robert Fossier and other French researchers (Fossier 1982, 191–192) proposing that the 46

Edith Peytremann - Village Archaeology in France. A twenty-five year retrospective

birth of the village in France took place around the year 1000. These data suggest that the chronology of the emergence of the village is more complex and far exceeds the frameworks so far proposed. Indeed, the various examples cited or discussed suggest that the villages may have different chronological origins. In the current state of research, the earliest period of origin that has been recorded is the Neolithic in Alsace (Peytremann 2013a, 227). Some villages provide evidence of occupation during the Bronze or Iron Ages (Bonnée, Couloisy (Oise), etc.) but the most numerous examples date to Antiquity (Yutz, Vieux, LunelViel (Herault), etc.). More village creations then take place throughout the Middle Ages9 and even in the modern and contemporary periods. The determining factors in the development of the village also indicate great variability, which depend largely on the date of creation. They correspond to natural features or anthropic interests such as road crossings, rivers, etc, which in some cases have long been identified by geographers, historians and archaeologists. These villages do share one point in common: the establishment of a church, at a variable date, in a part of the village that may also vary, even though the development of the building over the centuries ends up, in the modern period at the latest, in a central location within the village. The establishment of an elite dwelling in the village is on the other hand much less systematic. Another particularly interesting result emerges from these examples, for example from the case of the site of Vic (Indre). This is the plurality of funerary spaces within the village. These did not necessarily remain static around the church. Although the transferring of cemeteries to a site or sites outside the centre of the village is a well-known phenomenon from the mid-19th century onwards, the discovery that such an event could occur at earlier times, or the creation of funerary spaces which operated simultaneously along with the continues use of the churchyard introduces new factors to consider for the topographical study of the funerary spaces within the villages. It is already clear that the residence-church and cemetery relationship was more complex than had been assumed some thirty years ago, but now it seems that the village dynamics that lead to significant changes in spatial relationships (agricultural, funerary and built) has been underestimated. The example of the village of Bonnée demonstrates clearly that the core of the settlement was displaced between the ancient and the medieval periods. At the scale of the currentlyoccupied village, the same dynamics seem to be at work on settlement sites located outside the village, producing similar evidence of displacement. However, the we can no longer simply assume that these sites were deserted in favour of the currently-occupied village site. Although this may have sometimes been the case, other forces may have been at work. As pointed out by N. Mahé-Hourlier and Jean-Michel Willot, the importance of land ownership and the ways in which it is passed on are all factors that must be taken into account in order to understand the abandonment of the sites of rural habitats at the early and high Middle Ages and the formation of some villages (Mahé-Hourlier 2013, 217) (Willot et al. 2013, 157). Archaeology in village: a developing archaeology Let us illustrate the dynamism of village research in France today by looking in some detail at two research programs.10 They are different from the habitation and territory programme previously mentioned and from other fieldwork described above insofar as these projects do not derive solely from opportunistic discoveries and preventive archaeological investigations. Instead, they derive from particular programmed investigations (excavations or field-walking) in and around villages. In both cases, the archaeology is also associated with documentary and archival research. The first is the EVENA project, a research programme managed by the CITERES Laboratory and the Inrap since 2011 (Chimier et al. 2012). It aimed to carry out an archaeological assessment of the village of Esvres (dpt Indre-et-Loire) and its rural environment in the long term. The archaeological data came from different origins (field-walking, energy network monitoring, test pits, excavations). The field-walking 9  10 

For village creations between the 5th and 12th centuries, see Peytremann 2003. We thank warmly Jean-Philippe Chimier and Olivier Passarius for all data were provided to us.

47

The Archaeology of Medieval Villages Currently Inhabited in Europe

surveys covered 380 ha, and trial-trenching was carried out over 125 hectares. Ten sites, in addition to the remains located in the centre of the village, have been uncovered. They include agricultural establishments, from the early Iron Age to the Roman periods, or from the beginning of the Middle Ages, and funerary sites from the end of late Iron Age and the Gallo-Roman period (Chimier et al. 2014). A GIS has been set up to manage different types of data at different levels of interpretation. It integrates all available sources, studies of archival documentation, studies of built heritage, reviews of old archaeological data, the results of preventive excavations and surveys, of diagnostic test pits and field-walking. At the end of this five-year program and in advance of the preliminary publication currently under way, the researchers carried out a methodological assessment at a seminar. The paradox pointed out on this occasion is that the diversity of fields involved has led to too complex and unwiedly collection of data to manage under the program. Moreover, the vectoring of the documentation proved extremely time-consuming. The second project, entitled Villages of yesterday, today’s villages in Roussillon, is a collective project, begun in 2010, bringing together preventive archaeologists, academic historians, and other researchers. Its goals are to clarify the origins, formation, evolution and transformations of group of currently-inhabited villages on the Eastern Pyreneean plain. The impetus for this research project came as a result of a planned archaeological excavation at the village church of Villelongue-de-la-Salanque (Pyrénées orientales). Three trenches were dug to a depth of 1.30 m; beyond that coring was carried out. The excavation results allow us to completely reconsider the interpretation of the village’s development, which was based on written sources and analysis of the 19th century cadastre (Catafau 2014). The results indicate that the topography of the village was completely remodelled after significant flooding and that a ditch previously interpreted as an enclosure, in fact is the vestige of a partly-canalised former stream channel. The overall project allows for archaeological research in a several currently-occupied villages (test pits, energy networks monitoring, building archaeology studies, etc.) in order to understand the dynamics of their morphogenesis, the role of landscape patterns from ancient times in their organization and the reasons for their persistence. The study of landscape forms as fixed in the cadastral plans of the 19th century is also one of the goals of this project. The approach here is oriented towards comparing written sources (texts), planimetric documents (the Napoleonic cadastre cadastres and 18th century predecessors) and archaeological data in order to develop a critical interprétation in a regressive perspective. At the heart of this methodology is integrating the combined results of excavation and study of the built environment in currently-occupied villages, with information provided by written sources and the Napoleonic cadastre. It requires constant dialogue and back and forth interrogation, calling into question previous interpretations of texts, the translation of specific phrases, the chronology or the too easily-regressive reading of the cadastres. Conclusion Looking back over 20 years of experience, our review of archaeological research concerning currentlyoccupied villages in France offers mixed results. As we have seen, much has been done, but generally as a result of opportunities linked to development, without real preparation and the few data collected, accumulated sometimes over several years, have often fallen into oblivion in the absence of the means to analyze them. Nevertheless, there is a will now to advance this field of research. Indeed, it is included in the revision of the national programming conducted by the National Council for Archaeological Research,, a branch of the Ministry of Culture, which defines and oversees archaeological policy in France (Peytremann et al. 2013; Anon 2016b: 150). However, the financial and technical problems confronting the implementation of a true archaeological policy within villages must not be denied. Methodological reflection is essential to overcome the technical difficulties associated with this research, which often must be conducted over the long term before tangible results can be obtained. Preliminary results are, however encouraging and stimulate reflection. They offer a vision of the village that is totally renewed and much more diversified than what was previously thought. Village origins 48

Edith Peytremann - Village Archaeology in France. A twenty-five year retrospective

extend backwards over a long period of time, from the Neolithic to the modern era; the factors governing implantations are as variable as the subsequent evolutions. We must now speak of village in the plural. Church and cemetery, the common features in so many European villages, must be considered in the light of their own development. These early results encourage us to consider villages as association of spaces whose uses have been modified (or might have been modified) over time in accordance with economic factors (greater or lesser importance of crops, livestock, crafts, etc.), sociological factors (presence or absence of an elite, relative power of such as reflected in topography), religious factors (presence or absence of a parish church or of a chapel) but also climatic factors. We should also note that in France, studies of agrarian spaces remain undeveloped, thus depriving us of a key element of the village, its territory (terroir). The sharing of the different European experiences, in the context of this colloquium, is, in my opinion, a good starting point for the elaboration of a common methodology which will not only make singularities as both at national and local level stand out, but will also bring arguments before political and economic decision-makers the importance of this research. After the development of rural archaeology in the 1970s and urban archaeology in the 1980s, village archaeology is now developing with its own particular features: stratification, destruction, interdisciplinary, multidisciplinary, and emphasis on study of the overall rural environment from the perspective of the long term. It is now up to us to make this field of village archaeology as much a success as its two ‘sisters’, if I may thus express it, have known. Acknowledgment May Bonnie Effros and Bailey Yong be thanked for their attentive proofreading of my English translation. Bibliography Anon. 2016a. Livre 5 : Archéologie. In Code du patrimoine, Available at: https://www.legifrance.gouv.fr/ affichCode.do?cidTexte=LEGITEXT000006074236. Anon. 2016b. Programmation nationale de la recherche archéologique. Paris: Ministère de la Culture et de la Communication. Billoin, D., Humbert, S., and Putelat, O. 2006. Bourogne (Territoire de Belfort), ‘cimetière communal’. Rapport final d’opération. Dijon: INRAP Grand-Est Sud. Blaising, J.-M. 2000. Évolution de l’habitat sur le terroir de Yutz (Moselle-France). In J. Klápště (ed) Ruralia III, conferences Ruralia III, Maynooth, 3rd-9th September 1999: 120–155. Pragues: Institute of Archaeology Blaising, J.-M. 2003. Parcellaire laniéré, billons et crêtes de labours, les structures agraires lorraines du XIIe s. à hier. In Sols et structures agraires, 45–52. Sarrebourg: ADRAL, Inrap, DRAC lorraine Carré, F. et al. 2009. Histoire(s) de(s) village(s) L’archéologie en contexte villageois, un enjeu pour la compréhension de la dynamique des habitats médiévaux. Les Nouvelles de l’archéologie 116: 51–59. Catafau, A. 2014. Les Celleres et la naissance du village en Roussillon. Perpignan, France: Presses universitaires de Perpignan. Chimier, J.-P., Dubois, J., Fouillet, N., and Pouyet, N. 2014. Esvres-sur-Indre, de la Protohistoire récente au début du Moyen Âge. In É. Zadora-Rio (ed.), Atlas Archéologique de Touraine.Supplément à la Revue Archéologique du Centre de la France, http://a2t.univ-tours.fr/notice.php?id=65. Tours: FERACF Available at: http://a2t.univ-tours.fr/notice.php?id=65. Chimier, P., and Fouillet, N. 2012. Evena. Présentation du programme de recherche archéologique sur le village d’Esvres et son territoire communal et premiers résultats (2011-2012). Bulletin de la Société archéologique de Touraine 58: 101–108. Colleter, R., Le Boulanger, F., and Pichot, D. 2012. Église, cimetière et paroissiens. Bréal-sous-Vitré (Ille-etVilaine), étude historique, archéologique et anthropologique (VIIe-XVIIIe siècle). Paris: Errance. Compagnon, É., Louis, É., and Willot, J.-M. 2002. Au coeur des villages médiévaux du Nord de la France: les fouilles de Guesnain et de Déchy (Nord). Douai, France: Arkeos : Service archéologique de Douai. 49

The Archaeology of Medieval Villages Currently Inhabited in Europe

Dufour, J.-Y. 2013. Le village de Roissy-en-France au haut Moyen Âge. In N. Mahé-Hourlier and S. Poignant (eds.), Archéologie du village, Archéologie dans le village dans le nord de la France (Ve-XIIIe siècles). Actes de la table ronde, 22-24 novembre 2007, MAN, Saint-Germain-en-Laye: 23–36. Saint-Germainen-Laye: AFAM École pratique des hautes études. Section Sciences économiques et sociales ed. 1965. Villages désertés et histoire économique: XIe-XVIIIe siècles. Paris: S.E.V.P.E.N, 1965. Fernández Mier, M. et al. 2014. The investigation of currently inhabited villages of medieval origin:Agrarian archaeology in Asturias (Spain). Quaternary International 346: 41–55. Fossier, R. 1982. Enfance de l’Europe. Aspects économiques et sociaux. 1/ L’homme et son espace 1. Paris: PUF. Galinié, H. et al. 2003. Utilisation du pénétrométre dynamique de type Panda en milieu urbain pour l’évaluation et la caractérisation du dépôt archéologique. Revue d’Archéométrie 27: 15–26. Gauthiez, B., Zadora-Rio, É., and Galinié, H. 2003. Village et ville au Moyen Âge : les dynamiques morphologiques. Tours: Presses universitaires François Rabelais. Gentili, F. 2014. Archéologie d’un village du Moyen Âge à la période moderne. L’exemple de Villiers-le-Bel. Archéopages 40: 132–139. Gentili, F., Lefèvre, A. and Mahé, N. eds. 2003. L’habitat rural du haut Moyen Âge en Île-de-France. Guiry-enVexin (Val-d’Oise), France: C.R.A.V.F. Gentili, F., Lefèvre, A. and Mahé, N. 2009. L’habitat rural du haut Moyen Âge en Île-de-France. Guiry-en-Vexin (Val-d’Oise): C.R.A.V.F. Gonçalves, C. 2003a. Notice de site : Tremblay-en-France (Seine-Saint-Denis) Allée des Tilleuls, route de Roissy. In F. Gentili, A. Lefèvre and N. Mahé (eds.), L’habitat rural du haut Moyen Âge en Île-de-France. 1er supplément au Bulletin archéologique du Vexin français et du Val-d’Oise: 75–78. Guiry-en-Vexin: C.R.A.V.F. Gonçalves, C. 2003b. Notice de site: Tremblay-en-France (Seine-Saint-Denis) Château Bleu. In F. Gentili, A. Lefèvre and N. Mahé (eds.), L’habitat rural du haut Moyen Âge en Île-de-France. 1er supplément au Bulletin archéologique du Vexin français et du Val-d’Oise: 78–81. Guiry -en-Vexin: C.R.A.V.F. Gonçalves, C. 2003c. Notice de site : Tremblay-en-France (Seine-Saint-Denis) rue Louis Eschard, allée de la mairie. In F. Gentili, A. Lefèvre and N. Mahé (eds.), L’habitat rural du haut Moyen Âge en Île-de-France. 1er supplément du Bulletin archéologique du Vexin français et du Val-d’Oise: 56–63. Guiry-en-Vexin Gonçalves-Buissart, C., Lafarge, I., and Le Forestier, C. 2012. Les habitats ruraux du haut Moyen Âge en Seine-Saint-Denis. Etats des Lieux. Archéopages 34: 48–57. Hincker, V. 2013. Origine et constitution du village de Vieux (Calvados), restitution à partir des sources archéologiques. In N. Mahé-Hourlier and S. Poignant (eds.), Archéologie du village, Archéologie dans le village dans le nord de la France (Ve-XIIIe siècles). Actes de la table ronde, 22-24 novembre 2007, MAN, SaintGermain-en-Laye: 175–197. Saint-Germain-en-Laye: AFAM Hugonnier, L. et al. 2014. État de la recherche sur le ‘fait villageois’ en Picardie méridionale. Nouvelles données et nouvelles perspectives. Archéopages 40: 56–63. Joly, S. 2009. Bonnée (Loiret), Rue des Sentes : des occupations milieu 1er s. ap. J-C. et fin IIIe-Ve s. en périphérie de l’agglomération antique : rapport de diagnostic. Pantin: Inrap CIF. Joly, S. 2014. Bonnée, Loiret, rue Creuse : Un nouveau quartier de l’agglomération antique : projet de lotissement communal : rapport de diagnostic. Pantin: Inrap CIF. Lafarge, I. 2003. Notice de site : Tremblay-en-France (Seine-Saint-Denis) 70ter, route de Roissy. In F. Gentili, A. Lefèvre and N. Mahé (eds.), L’habitat rural du haut Moyen Âge en Île-de-France. 1er supplément au Bulletin archéologique du Vexin français et du Val-d’Oise: 69–71. Guiry -en-Vexin: C.R.A.V.F. Lafarge, I., & Gonçalves-Buissart, C. 2009. Tremblay-en-France (93), évocation de la formation du village à partir de l’habitat dispersé du haut Moyen Âge. In F. Gentili, A. Lefèvre and N. Mahé (eds.), L’habitat rural du haut Moyen Âge en Île-de-France. 2e supplément au Bulletin archéologique du Vexin français et du Val-d’Oise: 271–275. Guiry -en-Vexin: C.R.A.V.F. Lorren, C., and Périn, P. eds. 1995. L’habitat rural du haut Moyen Âge (France, Pays-Bas, Danemark et GrandeBretagne). Rouen, France: Association française d’archéologie mérovingienne. Mahé-Hourlier, N. 2013. De la fouille archéologique au domaine du haut Moyen Âge... In N. Mahé-Hourlier and S. Poignant (eds.), Archéologie du village, archéologie dans le village dans le nord de la France (Ve-XIIIe siècles). Actes de la table ronde, 22-24 novembre 2007, M.A.N, Saint-Germain-en-Laye, 207–220. Saint-Germainen-Laye: AFAM

50

Edith Peytremann - Village Archaeology in France. A twenty-five year retrospective

Mahé-Hourlier, N., and Poignant, S. eds. 2013. Archéologie du village, archéologie dans le village dans le nord de la France (Ve - XIIIe siècles): actes de la table ronde des 22-24 novembre 2007, M.A.N., SaintGermain-en-Laye. Saint-Germain-en-Laye: Association française d’Archéologie mérovingienne. Papin, P., Livet, J., Vanhove, C., and Yvernaul, F. 2015. Contribution à la connaissance de la topographie d’un centre paroissial en milieu rural : le cas de Joué-lès-Tours (Indre-et-Loire). Revue archéologique du Centre de la France [En ligne], 54. Available at: http://racf.revues.org/2284 [Accessed september 30, 2016]. Peytremann, É. 2003. Archéologie de l’habitat rural dans le nord de la France du IVe au XIIe siècle. SaintGermain-en-Laye: AFAM. Peytremann, É. 2013a. Un prétexte pour s’interroger sur la formation des villages : l’exemple des sites alsaciens du haut Moyen Âge. In N. Mahé-Hourlier and S. Poignant (eds.), Archéologie du village, Archéologie dans le village dans le nord de la France (Ve-XIIIe siècles). Actes de la table ronde, 22-24 novembre 2007, MAN, Saint-Germain-en-Laye: 221–236. Saint-Germain-en-Laye: AFAM Peytremann, É. 2013b. Vauchrétien, Maine-et-Loire., route de Notre Dame d’Allençon, allées des Platanes, rue Principale. Rapport de diagnostic. Cesson-Sévigné: Inrap Grand Ouest. Peytremann, É. 2014. La notion de village en France au premier Moyen Âge. Retour sur un débat. Archéopages (40): p.84–91. Peytremann, É., & Boucharlat, É. 2013. Programme 20 « Espace rural, peuplement et productions agricoles aux époques gallo-romaine, médiévale et moderne ». Ager. Available at: https://f.hypotheses.org/wpcontent/blogs.dir/1082/files/2013/04/Annexe_3.pdf [Accessed May 19, 2016]. Pichon, I. 2011. Nohant-Vic, ‘Place de l’église’. Nouvelles données sur le bourg de Vic. Rapport final d’opération de diagnostic et de suivi de travaux. Tours: Inrap Centre-Île-de-France. Poirier, N. 2013. Un espace rural à la loupe: paysage, peuplement et territoires en Berry de la préhistoire à nos jours. Tours: Presses universitaires François-Rabelais. Raynaud, C. 1990. Le village gallo-romain et médiéval de Lunel Viel, Hérault: la fouille du quartier ouest, 1981-1983. Besançon, France: Les Belles Lettres. Raynaud, C. 2007. Archéologie d’un village languedocien. Lunel Viel (Hérault) du 1er au XVIIIe siècle. Lattes: Éd. de l’Association pour le développement de l’archéologie en Languedoc-Rousillon. Schnitzler, B., Stahl, M., and Triantafilidis, G. 2015. Archives scientifiques de l’archéologie : fonds Arthur Stieber. Strasbourg: SRA Alsace. Stieber, A. 1954. Découverte d’un fond de cabane mérovingien à Furdenheim. Cahiers d’Archéologie et d’Histoire d’Alsace 134: 97–100. Willot, J.-M., Bernez, S. and Séverin, C. 2013. Du domaine monastique carolingien au village du bas Moyen Âge : approche archéologique d’une transformation du paysage rural en Ostrevent (Nord). In N. Mahé-Hourlier and S. Poignant (eds.), Archéologie du village, Archéologie dans le village dans le nord de la France (Ve-XIIIe siècles). Actes de la table ronde, 22-24 novembre 2007, MAN, Saint-Germain-en-Laye.T XXIX des Mémoires de l’AFAM: 141–159. Saint-Germain-en-Laye: AFAM

51

52

Investigating medieval village formation in the Netherlands1 An assessment of the output of 10 years of development-led archaeology J.P.W. Verspay, H. Renes, B. Groenewoudt, J. van Doesburg Introduction The formation and development of villages in the Middle Ages and Early Modern period is one of the current priorities of the Dutch Archaeological Research Agenda (NOaA) (Groenewoudt et al. 2017). It is one of ten research topics (kenniskansen) selected by the Cultural Heritage Agency of the Netherlands (Rijksdienst voor het Cultureel Erfgoed; RCE), that presently form major gaps in our knowledge and, at the same time, have a high potential to be bridged with the data of recent archaeological excavations (De Groot and Groenewoudt 2014: 7; Eerden et al. 2017; Groenewoudt 2015). To this purpose the RCE initiated the Valetta-Harvest programme, a stimulus programme aimed at synthesizing findings from development led archaeological excavations into new insights on the subject (Groenewoudt 2015). At the beginning of 2017 a study on village formation in the Netherlands was completed by a research team of the University of Amsterdam (Verspay et al. 2017). This article presents its main outcome: the potential and limits of the data from Malta driven research on the subject of village formation and recommendations for future advancement. The Dutch Archaeological system and the stagnation of the empirical cycle The signing and implementation of the Valletta Treaty (‘Malta’) had a profound impact on archaeological heritage policies in the Netherlands (Willems 2014). Prior to the embedding of the Treaty in Dutch legislation archaeological research was mainly conducted by universities, municipal archaeologists and the State Service for Archaeological Investigations (than: Rijksdienst voor Oudheidkundig Bodemonderzoek; ROB). Although the treaty was only put into law in the Netherlands in 2007, 15 years after its first signing, it had already been gradually implemented in the preceding years. In the new situation, archaeology became part of spatial planning and was to be financed according to the polluter-pays principle.2 The fieldwork was delegated to the market. As a result new archaeological firms emerged, some as spin-offs from the universities or the state department, others as private enterprises. The state department itself moved away from undertaking excavations and became a governing body, aimed at performing a connective role between politics, science and the executive parties. By 2007 nearly all archaeological fieldwork was conducted as part of spatial development (rescue archaeology) by contractors and, to a far lesser extent, municipal archaeological services. This development certainly had its merits. It brought a major increase in funding for archaeological research and resulted in a rise of jobs and the professionalization of the field. Archaeological research expanded to the entire country, whereas earlier university and state department research projects focussed on specific regions only. Also, unlike many previous rescue excavations, the results now got published as a rule. This development, however, came at a price (Willems 2014). The focus of the archaeological research shifted to the sub-local level of individual sites. The increasing number and scope of private contractors led to a fragmentation of research and knowledge. This situation worsened when universities with their regional research traditions, gradually retracted from the market. Although an overarching research agenda existed on the national level from 2005 onwards and in some provinces and towns provincial and local research agendas were developed,3 in most cases it proved to be too abstract to provide sufficient guidance This article is based on the Valetta Harvest report: Village Formation in the Netherlands during the Middle Ages (AD 800 – 1600): An assessment of recent excavations and a path to progress by J.P.W. Verspay, A.M.J.H. Huijbers, H. van Londen, J. Renes and J. Symonds 2017. 2  According to this principle the initiator(s) of activities that threaten to damage or destroy scientifically valuable archaeological remains are obliged to bear the cost of their protection or excavation. 3  Of the 12 provinces only 3 have their own research agenda as do 18 of the 400 municipalities (measurement 17 Januari 2017) (www.erfgoedmonitor.nl). 1 

53

The Archaeology of Medieval Villages Currently Inhabited in Europe

and unity for the individual excavation (De Groot and Groenewoudt 2014; Groenewoudt 2017). As a result a large number of single-site reports have been produced by contracting organisations, but local and regional syntheses remained rare (Figure 1). Dutch universities, which were regarded as the main actors to collate these studies into new knowledge, have been unable to keep up with the massive flow of new data due to lack of funding, staff or interest. So, compared to the large amount of new data that has been gathered since 2007, the progress in our understanding of the past has been limited.

Figure 1. Cycle of archaeological heritage management (simplified).

Transforming data into knowledge The issue of archaeological inertia was one of the main problems identified in the evaluation of the Monuments Act in 2012 (Van der Reijden, Keers and Van Rossum 2011).4 To counter this, the RCE initiated the programme ‘Kenniskaart Archeologie’ (Mapping Archaeological Knowledge) (Lauwerier et al. 2017, 2018), which aims to provide a better infrastructure for the assessment of interests within archaeological heritage management. Key principle to this process is that new insights generated by Malta research improve our understanding of the past. This provides the basis for increasingly accurate forms of predictive modelling and in time will have corresponding benefits for (the implementation of) local and national archaeological policies. Within this agenda, the RCE initiated the joint project ‘Nieuwe NOaA - Oogst voor Malta’ (New National Research Agenda - Valletta Harvest) to synthesise archaeological research and to update and improve the Dutch Archaeological Research Agenda (NOaA 2.0). At the start, major gaps in our knowledge were identified (kennislacunes) based on the current research agenda and on nationwide assessments, such as the Archeologiebalans 2002 (Lauwerier and Lotte 2002) and the Erfgoedbalans 2009 (Beukers 2009)(Eerden and Lobbes 2014). Next, an inventory was made of archaeological studies that were conducted during 2000-2014 to determine which of these gaps have the potential to be filled (kenniskansen) (De Groot and Groenewoudt 2014; Eerden et al. 2017). Topics that have been recently addressed in synthesising studies or articles were struck from the list. The remaining topics were then prioritised according to so-called ‘archaeological regions’5, period and theme. From these, ten topics were selected, among which was the theme of the present paper: ‘Village formation in the Middle Ages (ca. 800-1600)’. Defining the problem Despite the long tradition of settlement archaeology in the Netherlands, many questions remain unanswered. For the Middle Ages, one of the main questions is when the settlement became fixed and took the form of (present-day) villages. The current state of knowledge concerning settlement development (in general) and village formation (in particular) shows large regional differences. Hence, while an overview of the different settlement types in time and space has yet to be compiled for the north-western half of the Netherlands, particularly for the peat regions, models on the development Letter from the State Secretary for Education, Culture and Science (OCW) Zijlstra to the House of Representatives regarding the evaluation of The Archaeological Heritage Management Act, dated 7 February 2012. 5  An archaeological region (archeoregio) is an area with a shared history of landscape and habitation with similar formation processes of the soil archive (both physical geography and archaeological sites) (Beukers 2009: 112-113; Groenewoudt 1994; Lauwerier and Lotte 2002). 4 

54

Verspay et al - Investigating medieval village formation in the Netherlands

of settlement have already been devised for the south, southeast and east of the country (e.g. Heidinga 1987; Theuws 1989, 2008, 2011; Spek 2004; Spek et al. 2010; Van Beek 2009; Van der Velde 2011).6 In most cases, knowledge was largely confined to the level of the individual settlement (structure, nature, size and function), with comparative research focusing predominantly on classifying these settlements into different types.7 Rather less attention has been placed on examining the relationship between settlements, the possible meaning of different settlement types and how networks of settlement developed over time. In general, there is a structural lack of information about each region from the Late Middle Ages and Early Modern periods. This can be attributed to changes in house construction which has rendered them archaeologically invisible8; to the location of sites beneath current buildings, making them harder to access, and to a lack of interest among archaeologists and policymakers. The picture of late medieval and early modern settlements in the Netherlands therefore relies primarily on abandoned settlements that were later incorporated into the arable field systems. This has created a problem, inasmuch as we have some understanding of aspects of habitation in the early and high Middle Ages, but know very little about any links with modern villages and hamlets. In recent years the development of rural settlements during the Late Middle Ages and Early Modern period is gaining interest, partly because of the evaluation of our current knowledge (Beukers 2009) and the research priorities set by the RCE. Aims of our study Our research project was not about finding the origin of villages, but rather about understanding the processes behind, and the material expression of, village formation. Its aims were threefold. First, to establish to what extent archaeological research in the Netherlands - conducted in a commercial setting over the past decade (Malta archaeology)- produced relevant data that can be used to answer the key questions posed in the National Archaeological Research Agenda (NOaA) regarding village formation. Second, to provide answers to these questions based on the evidence from these recent excavations. Finally, to generate input for the upcoming update of the research agenda. These three aims have been used to guide the research project through the following phases: Phase 1: Collation and assessment of site-based archaeological data; Phase 2: Synthesis of village formation based on relevant sites in relation to the research questions; Phase 3: The drafting of ready-for-use input for the new NOaA. Defining the village It is not easy to define a village and in many ways the term is often used in a relational way. The village is a layered concept which can refer to different elements, i.e. an administrative organisation, a social community or a geographical entity. It is important, therefore, to clearly define this term prior to our analysis. Following the aims our study this definition needs to do justice to their historical interpretation, but at the same time reflect an archaeologically visible entity. From a historical point of view, a lot of definitions are possible for villages and village communities, because of their overlap with other local socio-spatial entities, territories, jurisdictions and communities There are nuances to the general trend as detailed studies have been conducted for some of the peat areas (e.g. De Bont 2008). These, however, focus primarily on the reclamation of these lands and only to a lesser degree address the development of settlement. Also, not every area within the more intensively studied sandy parts of the Netherlands has been covered equally thorough. 7  In particular, this is the case with the monodisciplinary, typological studies up until about 1990. After that the studies shift to an interdisciplinary approach focusing more on dynamics. 8  The timber frame was no longer set into the ground but placed on stone or brick footings. This meant that far fewer soil features were left behind, making the house plan difficult or impossible to identify. (e.g. Zimmermann 1998; Verspay 2007; Van Doesburg and Groenewoudt 2014; Schabbink 2015). 6 

55

The Archaeology of Medieval Villages Currently Inhabited in Europe

like parishes, alderman court territories and manorial complexes (Genicot 1990; Reynolds 1997; Troßbach and Zimmermann 2006). For practical reasons we looked primarily at the relationship between physical villages and societal processes influencing the physical dimension. The concept of village in a wide sense can be defined as a local socio-spatial unit on the level of the local community with varying societal and physical dimensions, but with at least one concentrated settlement, small or large. In a narrow sense a village is a concentrated settlement. A smaller or larger share of the inhabitants might identify themselves with the village community, the name of the village and the village territory, suggesting the existence of a notion of territoriality (Sack 1986). For archaeological research, the definition of a village as a concentrated settlement is relevant. Although rows of farms and even groups of dispersed farms could also be seen as a village, for the purposes of archaeological research they are better seen as groups of individual farms. In this study we understand ‘village’ to mean a large, stable, permanent (as against, for example, seasonal settlements), non-urban habitation agglomeration (or its archaeological equivalent) involving secondary and tertiary economic and social amenities. This definition only covers the physical settlement in a narrow sense, i.e. the village itself. The village community and territory therefore fall outside of our definition, although relations with surrounding areas, certainly for predominantly agrarian villages, did of course have an impact on the village. Combining this morphological definition with historical processes, we define village formation to be the clustering of farms and other houses into larger and stable settlement agglomerations in which secondary and tertiary economic and social amenities are developed. The complex, layered and many-sided nature of villages requires an interdisciplinary approach in which archaeological data and models are combined with historical, historical geographical and anthropological models, theories and insights. Archaeological manifestations In order to study a complex process such as village formation and to compile an inventory of relevant excavations, it was necessary to describe how the key elements and processes manifest themselves archaeologically and what indicators we needed to be looking for. Following our definition a village is first and foremost a collection of houses with a certain spatial cohesion or structure. This involves a degree of proximity or clustering and a location on a road or waterway or around a green, square or field. A green or square can be identified by a central open space in the settlement, possibly featuring a drinking hole or fence. Until about the fourteenth century9 the load-bearing structure of these houses was set into the ground, which meant the houses left behind the imprint of a house plan in the soil. Later, buildings on footings (poeren) made their appearance, making house-plans much more difficult to interpret (partial or complete). Other toft elements such as wells, pits, ditches and, in sandy regions from the fifteenth century onwards, deep-litter byres continued to be identifiable soil features. First and foremost, the amenities comprised of a church or chapel and graveyard. In the early phase the church was still a wooden building but at the latest, from the Late Middle Ages onwards, it was built of brick or stone. Graveyards usually lay in the direct vicinity of churches.

There are some regional differences on the exact moment and pace in which the use of padstones or brick footings became common practice in the construction of houses (Schabbink 2015: 209-221). 9 

56

Verspay et al - Investigating medieval village formation in the Netherlands

Other amenities are mainly manifested as traces of mills, artisanal activity or trade in ground features (ovens, pits, etc.) and in find material (loom weights, cloth seals, coin weights, the use of barrels in well structures, and so on). A certain level of stability and continuity can be assumed for villages. This is expressed in successive overlapping floorplans with a long duration of use and may have increased with the advent of more durable construction on brick footings and walls (Zimmermann 1998: 172). Village formation can manifest itself archaeologically as a development whereby habitation clustered in successive phases from small, dispersed settlements to a larger settlement with the appearance of the above-mentioned amenities (artisanal industry, trade and church) that were not yet present in the smaller settlements from the previous period.10 This process could also involve the emergence and growth of a settlement core with the above-mentioned characteristics around a new focal point such as an aristocratic residence or manorial court. New foundations are harder to identify because they entail an absence of occupation in a previous period. These characteristics made up part of the criteria that were used to inventory and assess relevant research data in Phase 1. Phase 1: Survey of relevant archaeological data The first part of our research aimed at collating archaeological excavations that could (potentially) provide relevant information for a synthesis on village formation and subsequently assess their usefulness for doing so. It was designed as a dual approach in which we would use data from both deserted settlements and existing villages. The archaeological data from the rural settlements was to be used from a progressive perspective (bottom-up) to see how, when and where the process of nucleation took place. In parallel, existing villages with a known medieval origin could be studied from a regressive perspective (top-down) to see when, how and why these places came to be. For these villages a wider range of sources was available such as historical studies, topographical maps and other archival data. The survey of archaeological data from Malta excavations started with a thorough querying of ARCHIS, the national archaeological database. Following the criteria we derived from archaeological, historicalgeographical and anthropological theory, as well as the given parameters, our survey produced 725 excavations published in the past decade.11 Of these, 410 were located outside urban areas (Figure 2).12 For 377 of these sites (92%) a (digital) report could be collected. All of these reports were assessed. Of these, 190 (50%) qualified as being relevant, 46% of the total number of the rural sites (Figure 3). The spatial distribution of excavated sites varies significantly between the various archaeological regions (AR). Relevant excavations were most abundant in the sandy areas of Brabant (AR 4) and OverijsselGuelders (AR 3), as well as the riverine area of Utrecht-Guelders (AR 13) whereas hardly any data was produced for the Limburg sand and loess area (AR 5 & 6), the clay area of Zeeland (AR 14) and Flevoland (AR 10) and the peat area of Friesland (AR 9). This pattern is of little historical significance and primarily reflects the areas where modern constructional developments are concentrated. The overwhelming majority of sites (181) consist of settlement remains in the form of one or more (farm)yards, including artificial dwelling mounds (terpen or wierden) (figure 5). Churches and graveyards were only present in a limited number of excavations (6). Mills were even fewer in number (2). In a substantial number of these sites (30), remnants of artisanal production were found, mostly pointing Other modes of development are possible too, such as the formation of a settlement nucleus without a contraction or relocation of the surrounding dispersed settlement, or the foundation of a completely new settlement cluster. 11  Sites selected were based on site type, site date, research type and publication date. 12  Cities selected were based on towns with borough rights during the given period. Since sites were documented by presentday place names, medieval villages that were incorporated at a later stage could not be excluded from this selection. 10 

57

The Archaeology of Medieval Villages Currently Inhabited in Europe

to iron production or processing. Because these remains are not always explicitly discussed in the site reports, the real number may be higher. To establish chronological trends and relations, the datings of the sites were included in our survey. Rather than using absolute dates, we chose to classify the individual sites by archaeological period (ABR) as is used in the ARCHIS database (Table 1).13 Looking at the dates of individual sites, we see a relatively large increase in the number of sites from the Early Middle Ages C (VMEC) to the High Middle Ages (LMEA) (Figure 5). For the subsequent periods, the numbers are much smaller.

Figure 2 Overview of all the initial archaeological sites (white = urban; black = non-urban).

There are, however, regional differences. These can be related to the landscape types involved. The curve works particularly well for the sandy regions (AR 1-5), with their long settlement history. In other archaeological regions, the different histories are shown in regionally different curves. In the wetland landscapes of Holland (AR 8-10) and Flevoland (AR12), for example, large-scale settlement only developed during the High Middle Ages when the large reclamations were undertaken. But the curve not only reflects the settlement history, but also shows the effects of archaeological formation processes. The decrease in the Late Middle Ages and the Early Modern period is not in line with the known demographics. The lack of sites from these periods can probably be attributed to developments resulting in a reduction of archaeological visibility, problematic accessibility and a lack of professional interest. Next, the settlement sites were categorised in single-phase and multiphase settlements and we looked at the duration of habitation at a given location. From this we could distinguish sites that

Figure 3 Overview of all the archaeological sites that are considered relevant by the set criteria.

58

Archeologisch Basisregister Plus at: http://abr. erfgoedthesaurus.nl/ 13 

Verspay et al - Investigating medieval village formation in the Netherlands

were inhabited for either a short or a longer period. Most of our sites (126) were multi-phased and were inhabited over a longer period of time. Measured in ABRperiods, the actual duration per site varied from one to nine (Figure 6). This shows that most of these sites were in use for up to four periods. Only one site, Hallum Hellema 2007, was occupied continuously since the Roman period. If we look at the starting period of these multi-phased settlements, regardless of their duration, it is very clear that most began, again, in the High Middle Ages (Figure 7).14 Between the different archaeological regions, some additional trends could be distinguished. To get an idea of the history of the presentday villages we also looked at the starting period of settlements that continued to the present (NTC). Finding a continuous series of houses or farmyards often proved problematic. Nonetheless, the majority of the present settlements turned out to

Figure 4 Overview of relevant sites by site type. Where multiple types are present the least generic is displayed.

Table 1. Overview of archaeological periods according to the Archeologisch Basisregister (ABR) used in this article. Period

Abbreviation (NL)

Date

Modern Period

NT

AD 1500 – present

Modern Period C

NTC

AD 1850 – present

Modern Period B

NTB

AD 1650 – 1849

Modern Period A

NTA

AD 1500 – 1649

ME

AD 450 – 1499

LME

AD 1050 – 1499

Late Middle Ages A

LMEB

AD 1250 – 1499

Late Middle Ages A

LMEA

AD 1050 – 1249

VME

AD 450 – 1049

Early Middle Ages D

VMED

AD 900 – 1049

Early Middle Ages C

VMEC

AD 725 – 899

Early Middle Ages B

VMEB

AD 525 – 724

Early Middle Ages A

VMEA

AD 450 – 524

ROM

12 BC – AD 449

Late Roman Period

ROML

AD 270 – 449

Middle Roman Period

ROMM

AD 70 – 269

Early Roman Period

ROMV

12 BC – AD 69

Middle Ages Late Middle Ages

Early Middle Ages

Roman period

14 

For this overview, only uninterrupted series of habituation have been included.

59

The Archaeology of Medieval Villages Currently Inhabited in Europe

Figure 5 The number of sites per archaeological period during the time frame 750-1650 from a total number of 190 sites.

Figure 6 The habitation period of multiphased sites measured in the number of ABR-periods they cover.

date back (at least) to the High or Late Middle Ages (LMEA/LMEB), although some have a Carolingian origin (Figure 8).

Figure 7 The archaeological period in which the multi-phased sites started.

Assessment The analysis of the survey results revealed that a substantial number of relevant excavations were undertaken, yet that the data - given the current approach - are of limited use for an inductive synthesis on village formation. This can be attributed to a number of factors: 1. The quality of the data in ARCHIS varies considerably This holds true for both the general information in the records, as well as the content of the individual projects. We encountered incomplete records, inconsistencies in descriptions and (multiple) duplications. The descriptions of individual excavations displayed a large variation in extent, detail and care. However, in contrast to ARCHIS, the summaries of excavation reports provided in the digital repository DANS were very useful. To a large extent, the quality of the digital data seems to be related to the particular 60

Verspay et al - Investigating medieval village formation in the Netherlands

Figure 8 The period in which the settlements started that continued to the nineteenth century (red) or the present-day (blue).

contractors and the time of the data input. Overall, the quality appears to have improved significantly in more recent projects. Nevertheless it was found that although ARCHIS is well suited for recording basic project information, its current database structure makes it of limited use for synthesis. 2. Most small scale research does not permit conclusions on settlement level Studies of rural settlement on the level of the individual village proved to be rare and often consisted of older research, e.g. Kootwijk (Heidinga 1987) and Gasselte (Waterbolk and Harsema 1979). This can in part be attributed to the scale of the excavation, which is a determining factor in its potential to produce relevant information on settlement topography (in a bottom-up approach). The extent of the excavation must be sufficient to get a clear view on the level of the toft and its surroundings and also large enough to cover a significant part of the settlement. More important, however, is the fact that the excavations in general focussed primarily on building plans and, to a far lesser extent, on tofts. Only rarely is this done from a perspective of the wider settlement. This severely limits the potential of individual sites to contribute to the topic of village formation process. 3. Few village areas with multiple excavations Since no excavation covers an entire village, data on the formation and development of these settlements needs to be compiled from several sites. Ideally we would be looking at multiple observations within one village territory. This, however, proved to be challenging as for only six villages three or four relevant excavations were available. Although this number improved on the municipal level, still only fifteen municipalities had three to five published observations. Upon closer inspection these excavation proved to be of various size and usefulness and often lay dispersed within the municipal territory. 4. Background information often locally specific and a historical narrative. Most of the settlements that have been studied, were founded in an already inhabited landscape. To explain the settlement dynamics, physical-geographical analysis by itself does not suffice. Historical or 61

The Archaeology of Medieval Villages Currently Inhabited in Europe

historical-geographical information is usually scarce, however. When this information is present, it often seems to be locally specific and follow a historical narrative, making it hard to distil general information out of this. Nonetheless, exceptions are present, e.g. in Zwaag (Schrikx 2008; 2009; 2013) and Warnsveld (Fermin 2009; 2011; 2012; 2013), and offer a valuable context and wider region. 5. Historical village centres Perhaps the most significant issue our survey exposed, was the difficulty to obtain data from currently inhabited historical village centres. This problem is the result of a combination of factors. First, while plenty of (local) historical data is available, most is not yet digitized, limiting the possibilities for synthesis at this stage. Historical topographical maps or land register plans have only been vectorised for limited parts of the Netherlands.15 Therefore, it was not possible to compile an integrated overview of archaeological observations in historical village centres and present an overview of the age of existing villages at this stage. Subsequently, an overview of medieval church locations, important indicators for historical village centres, to act as a reference point for spatial queries, turned out to be too ambitious. Next, very few excavations have been conducted in these village centres. A verification of the individual sites that were considered to be relevant, showed that only 25 (13%) were situated in the residential cores as mapped in the early nineteenth century. The great majority of the archaeological excavations was conducted in the surrounding country-side, as these are the areas that are developed when existing villages are expanded. As a result the archaeological dataset proved to be highly biased towards deserted settlements. In addition, the few excavations that were carried out in village centres are usually fairly small as most interventions in these existing village centres are often of limited size. Moreover, dependent upon the requisite set by a municipality on the size and depth of a (planned) disturbance in their zoning plan (vrijstellingsgrens), archaeological research is frequently not required. Even when excavation is required, often, only a part of a site can be uncovered. The fragmentation is worsened by disturbances caused by subsequent occupation phases, making it difficult to interpret the data adequately or to assess its value.16 Testing existing settlement models Following the outcome of our survey we had to conclude that there were too few relevant sites for a reliable synthesis of village formation based on qualitative analysis. Moreover, these sites differ substantially, both in their nature and the quality of the available data, and are unevenly distributed within the archaeological regions. Therefore we decided to adjust our research strategy and turn to a deductive approach in which the data is used to test existing hypotheses from historical geography, archaeology and anthropology on the topic. Recent settlement models could be collected for nine of the fourteen archaeological regions (Figure 9). As it turned out, the level of synthesis varied considerably between the archaeological regions. While a lot of work had been done in the cover sand areas, we were hard pressed to find anything for the clay areas, the riverine areas and the loess areas. As for other regions, no synthesis has yet been made. Some of these models are (out)dated, as their theoretical underpinnings are generally considered obsolete, yet quite a number of syntheses were developed recently as part of dissertations or other Malta Harvest projects. Here we face a problem that these are pretty much up-to-date with the major excavations. Testing these models would not so much entail confronting them with new data, but rather with different theories. Another difficulty we found was that most of our settlement models were not very specific on village formation. This could be because the time frame of a study only partially overlapped and did not Scans of these maps, however, are freely available for the whole county. Because of the almost inherent fragmented nature of the sites in village centres, they will frequently score low on their Physical Quality, one of the three main values in the Dutch Valuation system (Deeben et al. 1999; Willems and Brandt 2004). The valuation will therefore rely heavily on the Intrinsic Quality. This, however, requires a good understanding of the research potential of the remains and their value in correlation with other sites, both of which depend on the local and regional context. 15  16 

62

Verspay et al - Investigating medieval village formation in the Netherlands

cover the main part of this development (Dijkstra 2011), or because the focus lay on the reclamations rather than the developments of the settlements (De Bont 2008). The main difficulty of the models, however, was that most of them describe the trends in the settlement pattern in general terms, like a clustering around churches or on manorial estates. Similarly, the causes or motives for these developments are explained by abstract processes like ‘manorialisation’ and ‘institutionalisation’. These motives are fairly similar for all regions because of a bias towards written sources.17 The problem is not the validity of these explanations. On the contrary, these processes were most likely an important factor in the forming of stable agglomerations. The difficulty, however, lies with the possibilities to observe these processes archaeologically. Moreover, Figure 9 Overview of regions for which synthesising models on the these models only cover the outlines of formation of villages were available. the processes involved (e.g. the creation of manorial estates, the establishment of churches, demographic growth, the emergence of an early market economy, labour differentiation and specialisation). They do not provide much clarity on the variety found in the composition, layout and development of both the archaeological settlements and the historical villages. This means that although the current models are correct, they are not complete and lack the level of detail required to address this diversity. But it is also the question as to whether these spatial developments can be captured in general explanations at all. Comprehensive case studies by Spek show the importance of local factors in formation processes like the geographical location and natural opportunities for agriculture, the ownership of land, either privately held or organised in estates, access to trade networks and the proximity of main infrastructure (Spek 2004: 978). Leenders adds the importance of territorial and economic politics to this list, in which villages were given privileges or were even established from scratch by competing lords in order to outmanoeuvre their opponents (Leenders 2011). All of this illustrates the complexity of the topic of village formation and the difficulty to address this by using regional models. In order to understand this process one needs find out to how generals trends played out on a local level and in what form. General factors in Village formation To get a better understanding of the main processes of nucleation and village formation reference literature from related disciplines was studied. These disciplines, historical geography, history and anthropology, offer different but complementary windows upon the topic, offering insights into both the spatial and temporal development of the village form.

17 

It is stressed by Reynolds that historical texts offer a bias towards vertical relationships (Reynolds 1997).

63

The Archaeology of Medieval Villages Currently Inhabited in Europe

On a general level these main processes were found to be fairly consistent and well understood. In a recent review article the medieval historian Curtis surveyed the emergence of concentrated settlements across medieval Western Europe, and identified four thematic frameworks which have been regularly deployed by medieval historians to explain village formation (Curtis 2013: 226-227). These are: 1. Power, coercion, and lordship; 2. Communalism and territorial formalisation; 3. Field systems and resource management; 4. Urbanisation and market integration. These frameworks are interchangeable and overlap and should be regarded in their chronological, geographical and social context. It is possible that all four of the explanatory frameworks outlined here could apply to a given settlement at different points in its development (Curtis 2013: 251). The ‘power and coercion’ and ‘open field’ explanations for village formation have been dominant, but are in need of evaluation due to new evidence arising from across Western Europe (Curtis 2013: 250). Anthropological and Development Studies have the capacity to identify the relationship between social processes and rural settlements within living agricultural societies to uncover the circumstances in which concentrated settlements arise. Netting (1993: 157-188) and Stone (1996: 2856) found population density, intensity of farming, pressure on land, property concepts and modes of inheritance to be key variables in this process. Netting (1993: 164) views rural habitation as a situation within a continuum; a model which on one side hand sees a low population density, a low land pressure, an absence of a notion of private property, an absence of autonomous household farms, extensive farming and concentrated settlement; and at the other end of the spectrum has a high population density, high land pressure, presence of private property, presence of autonomous household farms, intensive farming and dispersed settlement (Huybers, Renes and Symonds 2017). Silberfein (1998) identified a number of social and economic aspects of society as the main reasons to live in villages. These social and economic reasons are typified as horizontal and vertical relations (Reynolds 1997). It is stressed by Reynolds that historical texts offer a bias towards vertical relationships (top-down, concurrent with Curtis’ first framework of power, coercion and lordship). Dutch archaeological models, such as those presented by Theuws, indeed put a focus on manorial lords and ecclesiastical institutions. Equal attention should then be given to five types of horizontal relationships visible in the Middle Ages of Western Europe: 1. Social relationships: solidarity, kinship, identification with community and community territory; 2. Social-defensive relationships: communal defence at times of conflict; 3. Social-economic relationships: cooperation in the commons; 4. Social-juridical relationships: common duties (paying taxes like tributes and tithes and common rights and duties with respect to the commons, common alderman court); 5. Social-religious relationships: confession of faith, joining common mass, worship of common saint, identification with a common saint and church territory (parish or not). All of these frameworks are thought to be relevant to the Dutch situation, albeit in different ways. Each factor can have its own network and sphere of influence, which does not necessarily correlate with a geographical area. In addition, it is apparent that similar patterns may appear in the landscape for different reasons and vice versa. On a theoretical level, the models and concepts present the dynamic complexity of processes behind village formation. And it is clear from this overview that no single model will fit all occurrences of village formation. It does however introduce human agency as a determining factor for variety and similarity. And that raises the question whether the projection of explanatory models on particular historical situations will ever lead to a satisfactory result. The aim of local and regional research then is to understand, rather than explain. 64

Verspay et al - Investigating medieval village formation in the Netherlands

Historical-geography: A bridge between site and synthesis The study of the regional models revealed that the level of synthesis on settlement development varied substantially between the regions. More importantly, the current models generally proved to be too abstract to test the data from the excavations. The levels of observation were too far apart. In order to connect these information levels in a meaningful manner we turned to historical geography. Compared to the traditional morphogenetic deconstruction, based mainly on nineteenth century maps and secondary literature, an in-depth retrospective analysis of village territories using a wider range of primary sources has proven to be more successful, especially if conducted as part of an interdisciplinary study. In its most elaborate form this approach was described by Spek and applied in the case studies of his dissertation (Spek 2004: 1004), but others have also produced useful results in lighter studies along these lines (e.g. Fermin 2015; Keunen 2006; Vangheluwe and Spek 2008). One of the main advantages of historical geography is that its scope is more in harmony with the scale of objects we study, as it tends to look at villages as a whole in relation to their surrounding territory. Archaeological observations hardly ever cover more than some fragments of a settlement. Further, by combining maps and fieldwork with the analysis of written records it can give an insight into processes like land ownership and common law that played a key part in village formation, but are virtually invisible archaeologically. This provides the framework in which archaeological observations can contribute by providing a dating range to known elements, add information on elements that are not known from the written record and test the geographical reconstructions in the field. Phase 2a: Analysis of four case studies Following these conclusions we made an inventory of historical-geographical case studies of villages or village territories that produced (or enabled) a reconstruction of its forming and development. We combined the results of the earlier inventories to get an overview of the available data on the different levels of observation. This showed that only for a very limited number of village territories, a combination of regional models, historical geographical studies and relevant archaeological observations were available (Figure 10). For further analysis we selected four case studies for which relevant information was available at all three levels: a (supra) regional model, (multiple) relevant and highly rated archaeological excavations, and a historical geographical analysis of the village (territory): Warnsveld (Gld.), Kerk-Avezaath and Kapel-Avezaath (Gld.), Limmen (N-H.) and Someren (N-Br.). Interestingly, for one of our best case studies, the village of Warnsveld, the archaeological data consisted predominantly of small-scale observations such as trial trenches and watching briefs. Because the researchers worked closely from and within a historical geographical framework, these observations gained a high scientific yield (e.g. Fermin and Groothedde 2005; Fermin 2009; 2011; 2012; 2013).

Figure 10 Overview of the combined outcome of the inventories of archaeological excavations, historical geographical case studies and (supra)regional models.

65

The Archaeology of Medieval Villages Currently Inhabited in Europe

The individual cases were described and analysed according to the following format: 1. Presentation of the available dataset; 2. Presentation of the current regional settlement model(s); 3. Presentation of a historical-geographical study; 4. Presentation of archaeological observations; 5. Description of the settlement history (500) 800-1600; 6. Explanation of the settlement history (500) 800-1600; 7. Extrapolation to the (archaeological) region; 8. Conclusion. First, an overview was provided of all the data we collected for the case from the various disciplines. Next, the current regional settlement model was presented and specified to the development of villages. Then, if available, the historical geographical data was discussed to provide an overview of the village and a model for its development on a local level, to identify key factors in this development and deduce archaeological indicators to test these. Next, the archaeological observations were presented and used to describe the characteristics of the settlement remains according to their composition, lay-out and date. This provided the information for the reconstruction of the development of the village between AD 500 and 1600.18 Subsequently, we tried to identify the major processes that led directly or indirectly to the formation of the village and check what the general social factors contributed to this. Finally, we extrapolated our findings to the (archaeological) region to see whether these factors also determined the development of other villages in this area, and to establish at what level these processes took place. In our case studies we found that a combination of most of the factors identified above played a role in the formation of each of the villages on a general level. However, on a local level the precise form of these factors could vary considerably and it was often difficult to establish how these influenced the development of the settlement exactly. The founding of a church could trigger nucleation in some villages, for example, while in other settlements dispersed hamlets continued to be the norm. In addition, like Spek (2004: 978), we found that similar processes could lead to different layouts whereas villages with a similar layout could have different origins. Each of the villages had its own distinct formation in which local factors and human agency played an important role. Moreover, a survey of settlements in the surrounding area clearly showed that the development of the case studies was not ‘typical’ for the region. Phase 2b: Synthesis of village formation On a general level the factors involved in village formation are consistent with the models presented by Curtis (2013) and Silberfein (1998). The diversity we found in our case studies can be attributed to the major influence of local factors and human agency in the process of village formation. This could be the geographical location and natural opportunities for agriculture, the ownership of land, access to trade networks and the proximity of main infrastructure, as well as territorial and economic politics. From this diversity we can conclude that the formation of villages is not so much the logical result of a specific factor or decision, but rather the outcome of a context in which a set of conditions were met that made the development of a village possible or expedient. For example, when demographic growth reached the limits of the available arable land, this could give rise to cottagers. The emergence of cities and an early market economy presented a demand and market, where opportunities for (part-time) specialisation and trade occurred. Churches and manorial courts acted as religious and administrative centres and formed a focal point around which inhabitants with a less land-bound livelihood clustered, sometimes stimulated by the establishment of a market place. Changes in the social-judicial positioning of people, in which they were no longer bound to the land, meant that they could now move to a different location, and cohabitation in a clustered settlement could be more favourable for cooperation or to 18 

We expanded the period to the Early Middle Ages to provide a better context for the development of the settlement.

66

Verspay et al - Investigating medieval village formation in the Netherlands

maintain social relations. Others might choose a more autonomous position. Other factors like territorial or economic politics or the geography might dictate and channel these developments, for example stream banks in tidal marshes heavily favour linear settlement, as does a location along a major road. The importance of local factors means that in order to compile a comprehensive synthesis on village formation, one has to unravel and understand the process on a local level first. Only then is it possible to deduce the general trends and construct a regional model. Although we found some excellent case studies, their number is currently far too small to endeavour such a synthesis. For now, the knowledge gap remains. The good news is that despite the complexity of village formation and the fact that some of the major processes involved are not directly observable archaeologically, it is still very much possible to gain good insight into the process. This can be seen in some of our case studies. The key is interdisciplinary research, specifically the use of historical geography, and a detailed reconstruction of (late) medieval landownership and land use. An historical-geographical and historical overview of a village territory not only enables better interpretation and cohesion of the (often small-scale) archaeological observations, it also provides a tool for more accurate predictions, more purposeful questions and selection. As such, it would be beneficial for municipal planning and heritage policy, in which it could be integrated. Phase 3: How to proceed The peculiarities of research in built-up areas mean that it is quite different from the studies that are conducted in the countryside, which determine the norm in current rural archaeology. In essence, village archaeology resembles urban archaeology. This is reflected, in part, in the scale of observational opportunities which are mostly limited to small plots and keyholes. Our case studies, however, illustrate that these snippets of information obtained from small-scale excavations and observations in test trenches and watching briefs are invaluable in reconstructing the development of a village. It is therefore important to acknowledge the potential of small-scale interventions in municipal policy and to embed these observations on a local level in municipal policy. To do so effectively requires: •• Municipal research agenda: Since the National Archaeological Research Agenda stands at some distance from the decision-making process at the local level, it is important to also incorporate the topic of village formation in the local research agenda and translate it into the predictive map and regulations. •• Historical geographical framework: The local research agenda would benefit greatly from a historical-geographical study of the village area as it would act as a framework for interpreting (small-scale) archaeological observations and lead to a better understanding of village formation as part of the wider settlement dynamics. Moreover, it would enable to further concentrate on the predictive maps. A good example of the integration of historical information and policy can be seen in the heritage maps of Southeast Brabant.19 •• Continuity of knowledge: Because local factors play a major role in village formation and the available data is often fragmented, it is important to maintain an overview to adequately interpret and assess the potential of the data of these small scale observations. This enables continuous synthesis of local information and up-to-date information for policy and efficient decision-making. Within local policy the extent of the exemption area before archaeological research is required should be minimised initially. To minimise the burden on small (private) developments, yet benefit from the observational opportunities, one could make more use of watching briefs, combined with a general design brief (PvE) for this purpose, based on the historical-geographical framework.

19 

Heritage map for Southeast Brabant at http://atlas.odzob.nl/erfgoed.

67

The Archaeology of Medieval Villages Currently Inhabited in Europe

Village archaeology greatly benefits from a stable focal point of knowledge for a coherent interpretation of (small-scale) observations and assessment of the relevance of archaeological involvement in various locations. It is no coincidence that the best case studies currently available are connected to municipal archaeological services; possessing the additional advantage of being closely situated to the subject, being able to respond quickly to opportunities and threats that arise, and having the overview to do so efficiently. However, there are no compelling reasons for village archaeology to be restricted to this organisational form. To conclude, the study of village formation is best served by a local research agenda based on comprehensive historical-geographical study of the area and active coordination by people who know the area well. A case like Warnsveld shows that this really pays off. The ensuing bias of the current system In our current archaeological heritage management system in the Netherlands, excavations are all but development driven. As a result, the data on medieval rural settlement primarily stems from the countryside surrounding the historical villages, as these are the areas in which development is most frequently undertaken. Very few excavations are conducted in the residential cores. Because of this, our understanding of the development of rural settlement is mainly based on deserted settlements. This imbalance is problematic, as the relation with the development of the historical village centres often remains unclear. To counter this inherent bias, the research on village formation would greatly benefit from purposeful studies of historical village centres. This would provide a more representative range of sites, which enables a more comprehensive understanding of rural settlement in general and villages in particular. In addition, the outcome could provide valuable input for municipal planning and heritage policy. Such a study should be a combination of integrated landscape archaeological and historical geographical research (e.g. Spek 2004: 1003) and an archaeological prospection of the historical village and the surrounding area (e.g. Lewis 2005; 2013; 2014) with additional designated observations (e.g. Nováček and Vařeka 1996; Vařeka et al. 2010). Since these villages are inhabited, a Community Archaeology approach such as in the CORS programme (Lewis 2014) is invaluable to obtain archaeological data and make use of local knowledge and to generate support. A project along these lines has recently been initiated in a joint effort by Lincoln University (UK), the University of West Bohemia (CZ), the Adam Mickiewicz University (PL) and the University of Amsterdam (NL): Community Archaeology in Rural Environments.20 (Verspay in prep.). Bibliography Beukers, E. (ed.) 2009. Erfgoedbalans 2009: Archeologie, monumenten en cultuurlandschap in Nederland. Amersfoort. Curtis, D. R., 2013. The Emergence of Concentrated Settlements in Medieval Western Europe: Explanatory Frameworks in the Historiography. Canadian Journal of History/Annales. De Bont, C. H. M. 2008. Vergeten land: Ontginning, bewoning en waterbeheer in de westnederlandse veengebieden. Wageningen. De Groot, T. and B. Groenewoudt 2014. Programma Kenniskaart Archeologie: ‘Kenniskansen’ voor AMZ-relevant synthetiserend onderzoek op basis van Malta-rapportages 2. Project Oogst voor Malta. Amersfoort. Deeben, J.H.C., B.J. Groenewoudt, D.P. Hallewas and W.J.H. Willems 1999. Proposals for a practical system of significance evaluation in archaeological heritage management, European Journal of Archaeology 2, 177-199. Dijkstra, M. F. P., 2011. Rondom de mondingen van Rijn & Maas: landschap en bewoning tussen de 3e en 9e eeuw in Zuid-Holland, in het bijzonder de Oude Rijnstreek. Amsterdam (dissertation University of Amsterdam). Eerden, M. and M. Lobbes 2014. Rapporten Malta-opgravingen 2007-2013 beter ontsloten. Amersfoort. 20 

Information on the Dutch part or this research can be found at: www.gemeenschapsarcheologie.nl

68

Verspay et al - Investigating medieval village formation in the Netherlands

Eerden, M. C., B. J. Groenewoudt, T. de Groot, E. M. Theunissen and R. Feiken 2017. Synthesising data from development-led archaeological research. In: R. C. G. M. Lauwerier, M. C. Eerden, B. J. Groenewoudt, M. A. Lascaris, E. Rensink, B. I. Smit, B. P. Speleers and J. van Doesburg (eds), Knowledge for Informed Choices. Tools for more effective and efficient selection of valuable archaeology in the Netherlands. Amersfoort (Nederlandse Archeologische Rapporten 55): 195-209. Fermin, H. A. C., 2009. Het verleden van Warnsveld verkend (2): Archeologische begeleiding van de rioolvervanging in de Molenstraat, het Bongerspad, de Tuinstraat en de Schoolstraat te Warnsveld (gem. Zutphen). Zutphen (Zutphense Archeologische Publicaties 47). Fermin, H. A. C., 2011. Het verleden van Warnsveld verkend (3): Archeologische begeleiding van twee rioolvervangingen in Warnsveld (Laan 1940-1945 en Landweg). Zutphen (Zutphense Archeologische Publicaties 61). Fermin, H. A. C., 2012. Rhienderink, Brake en Scheurkamp: Archeologisch onderzoek en historisch onderzoek naar de ontstaansgeschiedenis van drie historische erven naar aanleiding van de rioolvervanging in de Rhienderinklaan, Kozakkenlaan, Veldesebosweg, Abersonlaan en Peppelenweg in Warnsveld (gem. Zutphen). Zutphen (Zutphense Archeologische Publicaties 71). Fermin, H. A. C., 2013. De Weme van Warnsveld: Archeologisch en historisch onderzoek rond de oude pastorie op Rijksstraatweg 57 in Warnsveld (gem. Zutphen). Zutphen (Zutphense Archeologische Publicaties 76). Fermin, H. A. C., 2015. Rhienderink revisited: Een middeleeuwse greppel op het perceel Veldweidelaan 4 te Warnsveld. Zutphen (Zutphense Archeologische Publicaties 102). Fermin, H. A. C. and M. Groothedde 2005. Het verleden van Warnsveld verkend: Archeologische waarnemingen bij de cunetbegeleiding rond de Bonendaal en het proefsleuvenonderzoek bij Den Bouw (Abersonplein 9) in 20042005 te Warnsveld. Zutphen (Zutphense Archeologische Publicaties 18). Genicot, L. 1990. Rural communities in the medieval west, Baltimore. Groenewoudt, B. J. 1994. Prospectie, waardering en selectie van archeologische vindplaatsen: Een beleidsgerichte verkenning van middelen en mogelijkheden, Amsterdam (dissertation University of Amsterdam). Groenewoudt, B. J. 2015. Valletta Harvest: value for money. Dutch initiatives to make ‘Malta’ excavation results relevant to heritage management, science and society. Contribution EAC Occasional Paper No. 10. Proceedings of the International Conference Amersfoort, The Netherlands, 20-22 March 2014. Groenewoudt, B. J., M. C. Eerden, T. de Groot and E. M. Theunissen 2017. Answers to questions: The new National Archaeological Research Agenda of the Netherlands. In R. C. G. M. Lauwerier, M. C. Eerden, B. J. Groenewoudt, M. A. Lascaris, E. Rensink, B. I. Smit, B. P. Speleers and J. van Doesburg (eds), Knowledge for informed choices. Tools for a more effective and efficient selection of valuable archaeology in the Netherlands. Amersfoort (Nederlandse Archeologische Rapporten): 179-194. Heidinga, H. A., 1987. Medieval settlement and economy north of the lower Rhine: Archaeology and history of Kootwijk and the Veluwe (the Netherlands). Assen (Cingula 9). Huijbers, A. M. J. H., J. Renes and J. Symonds 2017. Conceptual framework. In J. P. W. Verspay, A. M. J. H. Huijbers, H. van Londen, J. Renes and J. Symonds, Village formation in the Netherlands (AD 800-1600): An assessment of recent excavations and a path to progress. Amersfoort (Nederlandse Archeologische Rapporten 52). Keunen, L. J. , 2006. ‟…der grosse und kleine hoff undt darauf bestehendes dorpff Neede…”: De ontwikkeling van de hof en het dorp Neede. Borculo (Kleine Reeks: Bijdragen tot de geschiedenis van Stadt en Heerlijkheid Borculo 12). Lauwerier, R. C. G. M. and R. M. Lotte (eds) 2002. Archeologiebalans 2002, Amersfoort. Lauwerier, R.C.G.M., M.C. Eerden, B.J. Groenewoudt, M.A. Lascaris, E. Rensink, B.I. Smit, B.P. Speleers & J. van Doesburg (eds.) 2017. Knowledge for informed choices. Tools for a more effective and efficient selection of valuable archaeology in the Netherlands, Amersfoort (Nederlandse Archeologische Rapporten 55). Lauwerier, R.C.G.M. et al. 2018. Toolbox for Archaeological Heritage Management. Maps, Methods and More for Effective and Efficient Selection of Valuable Archaeology, Internet Archaeology 49. https:// doi.org/10.11141/ia.49.8 Leenders, K. A. H. W., 2011. Late dorpsvorming in het Brabantse zand. Historisch Geografisch Tijdschrift 29: 73-80. Lewis, C. 2005. Test pit excavation within currently occupied rural settlements in East Anglia in 2005. Annual Report of the Medieval Settlement Research Group 20: 9-16. Lewis, C., 2013. Test pit excavation within currently occupied rural settlements – results of the University of Cambridge CORS project in 2012. Annual Report of the Medieval Settlement Research Group 28: 77-89.

69

The Archaeology of Medieval Villages Currently Inhabited in Europe

Lewis, C., 2014. The Power of Pits: Archaeology, Outreach and Research in Living Landscapes. In K. Boyle, R. J. Rabett and C. O. Hunt (eds.), Living in the Landscape: Essays in Honour of Graeme Barker. Cambridge: 321-338. Netting, R., 1993. Smallholders, householders. The Environment in Anthropology: A Reader in Ecology, Culture, and Sustainable Living: 10-14. Nováček, K. and P. Vařeka, 1996. Archaeological research of present-day villages of a medieval origin in Bohemia. In J-M. Pesez (ed.), Ruralia I: Conference Ruralia I (Prague, 8th-14th september 1995). Prague: 314-320. Reynolds, S., 1997 (1984). Kingdoms and Communities in Western Europe 900-1300, Oxford. Sack, R. D., 1986. Human territoriality: Its theory and history. Cambridge. Schabbink, M. (ed.), 2015. Vier eeuwen boeren: Synthese Oogst voor Malta onderzoek: Archeologische sporen van boerderijen en erven 1250-1650. Amersfoort (Nederlandse Archeologische Rapporten 49). Scholte Lubberink, H.B.G., L.J. Keunen & N.W. Willemse, 2015. Op het kruispunt van de vier windstreken. Synthese Oogst voor Malta onderzoek de Gelderse Vallei (Utrechts-Gelders zandgebied), Amersfoort (Nederlandse Archeologische Rapporten 48). Schrikx, C. P., 2008. Bewoningssporen vanaf de 12e eeuw aan de Dorpsstraat van Zwaag (gemeente Hoorn). Hoorn (Verslagen van de Archeologische Dienst Hoorn 5). Schrikx, C. P., 2009. Archeologisch onderzoek op het perceel Dorpsstraat 186 in Zwaag, Hoorn, (Hoornse Archeologische Rapporten 13). Schrikx, C. P., 2013. Sporen van 12de-eeuws Zwaag: Archeologisch onderzoek op het perceel Dorpsstraat 176 in Zwaag, gemeente Hoorn. Hoorn (West-Friese Archeologische Rapporten 59). Silberfein, M. (ed.), 1998. Rural Settlement Structure and African Development. Boulder. Spek, T., 2004. Het Drentse esdorpenlandschap: Een historisch-geografische studie, Utrecht. Spek, T., H. van der Velde, H. Hannink and B. Terlouw 2010. Mens en land in het hart van Salland: bewoningsen landschapsgeschiedenis van het kerspel Raalte. Utrecht. Stone, G. D., 1996. Settlement ecology: The social and spatial organization of Kofyar agriculture, Tucson. Theuws, F. C., 1989: Middeleeuwse parochiecentra in de Kempen 1000-1350. In F. Theuws and A. Verhoeven (eds), Het Kempenproject 3: De Middeleeuwen centraal. Waalre: 97-216. Theuws F. C. W. J., 2008. Settlement research and the process of manorialization in Northern Austrasia. In S. Gasparri (ed.), 774: ipotesi su una transizione: atti del seminario di Poggibonsi, 16-18 febbraio 2006. Turnhout: 199-220. Theuws, F., 2011: Nederzettingsontwikkeling in de Middeleeuwen: een model en enige thema’s voor toekomstig onderzoek. In F. Theuws, M. van der Heiden and J. Verspay, De archeologie van de Brabantse akkers. Amsterdam (Themata 4): 60-77. Troßbach, W. and C. Zimmermann 2006. Die Geschichte des Dorfes: Von den Anfängen im Frankenreich zur bundesdeutschen Gegenwart. Stuttgart. Van Beek, R., 2009. Reliëf in Tijd en Ruimte: Interdisciplinair onderzoek naar bewoning en landschap van OostNederland tussen vroege prehistorie en middeleeuwen. Wageningen (dissertation Wageningen University). Van der Reijden, H., G. Keers and H. van Rossums 2011. Ruimte voor archeologie: Synthese van de themaveldrapportages. Amsterdam (RIGO rapport P18090). Van der Velde, H. 2011. Wonen in een grensgebied: Een langetermijngeschiedenis van het Oost-Nederlandse cultuurlandschap (500 v. Chr.-1300 na Chr.). Amersfoort (Nederlandse Archeologische Rapporten 40). Van Doesburg, J. 2013. Manors (curtes): New archaeological evidence from the Netherlands, In J. Klápště and P. Sommer (eds), Hierarchies in rural settlements, Turnhout (Ruralia IX): 212-235. Van Doesburg, J. and B. Groenewoudt 2014. In search of the invisible farm: Looking for archaeological evidence of late medieval rural settlement in the sandy landscapes of the Netherlands (1250–1650 A.D.), In I. Bahacova and P. Sommer (eds), Medieval Europe in Motion: In honour of Jan Klápště. Prague: 51-68. Vangheluwe, D. and T. Spek 2008. De laatmiddeleeuwse transitie van landbouw en landschap in de NoordBrabantse Kempen. Historisch-Geografisch Tijdschrift 26: 1-23. Vařeka, P., F. Kostrouch, P. Kočár, Z. Sůvová, 2010. A contribution to research on deserted villages of Medieval origin. The remains of Late Medieval buildings on house plot no. 121 in Mikulčice, Přehled výzkumů 51: 249-265. Verspay, J. P. W., 2007. Onzichtbare Erven: Het Brabantse platteland in de Late Middeleeuwen. Amsterdam (master thesis University of Amsterdam). Verspay, J.P.W., in prep. Expedition Liempde: Archaeological research in a present-day village of medieval origin in North-Brabant (NL), Plzeň. 70

Verspay et al - Investigating medieval village formation in the Netherlands

Verspay, J. P. W., A. M. J. H. Huijbers, H. van Londen, J. Renes and J. Symonds, 2017. Village Formation in the Netherlands during the Middle Ages (AD 800 – 1600): An assessment of recent excavations and a path to progress. Amersfoort (Nederlandse Archeologische Rapporten 52). Waterbolk, H. T., 2009. Getimmerd verleden: Sporen van voor- en vroeghistorische houtbouw op de zanden kleigronden tussen Eems en IJssel. Groningen. Waterbolk, H. T. and O. H. Harsema, 1979. Medieval farmsteads at Gasselte. Palaeohistoria 21, 227-265. Willems, W. J. H. 2014. Malta and its consequences: a mixed blessing. In V. M. van der Haas and P. A. C. Schut, The Valletta Convention: twenty years after – benefits, problems, challenges. Brussels (EAC Occasional Paper 9), 151-156. Willems, W. J. H., & Brandt, R. W. 2004. Dutch archaeology quality standard. Zimmermann, W. H., 1998. Pfosten, Ständer und Schwelle und der Übergang vom Pfosten- zum Ständerbau - Eine Studie zu Innovation und Beharrung im Hausbau: Zu Konstruktion und Haltbarkeit prähistorischer bis neuzeitlicher Holzbauten von den Nord- und Ostseeländern bis zu den Alpen. Probleme der Küstenforschung im südlichen Nordseegebiet 25: 9-241.

71

72

Is this a village? Approaching nucleated settlements in Scandinavian contexts Ingvild Øye In 1859, following the second Land Consolidation Act in Norway two years earlier, the Ministry of the Interior appointed a land commission to restructure the farmland in the community of Vik in western Norway. This community is located mid-fjord on the southern side of the 200km long Sognefjord, and is one of the richer agricultural communities in this part of the country. The mission was to carry out the first large-scale reallocation of the joint ownership of all infield farmland in the two parishes of Vik; Hopperstad and Hove, both with their early medieval churches at farms with the same names, both also with monumental Viking Age grave mounds and the largest concentration at Hove (Figure 1). The reallocation also partly involved outfields right up to the mountains. All land of individual parts of altogether 1500 arable fields and meadows was measured for size and assessed for productivity to achieve an equal and just deal for the owners. It affected 18 more or less agglomerated farms of altogether 61 holdings. The farmland comprised 1739 decares and included 1539 borderlines (Hoprekstad 1956). However, at this stage it did not affect the old structures of largely nucleated settlements. This huge enterprise took five years, leaving a document of 160 closely written pages and maps that gives a unique insight into older settlement and field structures of seemingly old age (Figure 2). Was this a village, then? The general framework and the conceptual and analytical tool in Norwegian agrarian studies has been the farm denoted by a name of its own (Norw. navnegård) – equivalent with the later land-assessed farm, as a separate fiscal object. Still, it could cover a varied number of holdings; units of production run by a household. Farms could be clustered or more dispersed, regardless of structure, size and degree of

Figure 1. Prospect of Vik, c. 1930 with the medieval Hove church in front on the terrace behind an expanding settlement facing the Sognefjord. The new Vik church from 1877 can be discerned to the left, while Hopperstad stave church further west is out of the picture (Photo: Fylkesarkivet i Sogn og Fjordane, with permission).

73

The Archaeology of Medieval Villages Currently Inhabited in Europe

Figure 2. Section of the cadastre map from Vik, 1864: Above to the left, the clustered tun at Hopperstad, with the stave church in its outskirts. On the eastern side of the river Hopra, the farm Hove appears with its three clusters with the church in the centre (Sunnfjord og Ytre Sogn jordskifterett, with permission).

subdivision into co-located holdings (Sandnes 1979). Co-location was the case for many of the farms in Vik, such as the farm Hopperstad, with around fifty buildings in the 1850s including outhouses (cf. Figure 2). The suffix -staðir in the name also indicates Viking Age origins at the latest. Farm names therefore also reflect settlement hierarchies as well as chronological layers. 74

Ingvild Øye - Is this a village? Approaching nucleated settlements in Scandinavian contexts

What, then, is the difference between clustered farms in the Norwegian sense, and hamlets and villages as the usual concepts in other parts of Scandinavia? The criteria are not so clear. In Sweden the concept, ‘by’, has been seen as the equivalent to the English hamlet although the area need not comprise more than two farmsteads (Sporrong 1990: 468). Villages are generally larger than hamlets, and often classified according to both structure and function (Adams 1977: 64). However, all Scandinavian languages lack the linguistic distinction between village and hamlet. Hence, quantitative as well as structural criteria largely distinguish between the two – which is the case for farms in Norwegian contexts too. Such criteria, however, vary from one country to another. In Scandinavian, as well as European research, the lower limit of a village is set at around 10–15 farms (Hybel 1989: 188). In recent Scandinavian research, both archaeological and historical, the line between hamlets and villages has been rather unclear, when in Denmark the minimum size of a village counts only three farms (e.g. Liebgott 1989: 26). The terminological dividing lines between farms, hamlets and villages are thus rather vague and the factual differences are exaggerated (Widgren 1997; Øye 2000). There are reasons, then, to look closer into the matter. This article focuses on structural, functional and chronological aspects of agglomerated rural settlements in Norway related to other Scandinavian countries. The aim is twofold – to trace nucleation processes in time and space within regions and across national borders. The second concerns varied research traditions and perceptions related to both deserted medieval settlements and more sustainable nucleated settlements that have lasted into our recent history. The overall aim is to achieve a broader understanding of nucleation processes in different landscapes in time and space, related to various environments and other factors that promoted nucleation. What’s in a name? The linguistic distinction between farms, holdings, hamlets and villages in Scandinavian research of the Middle Ages, largely disappears within the medieval terminology.The common noun for the inhabited and cultivated area was ON bær or býr, deriving from búa, to dwell, reside. The synonyms ból and bólstaðr were other common terms (KLNM II: 282). The nucleus of the habitation, called bœli or tún (etymogically parallel to OE tūn), refers to the habituated area and the space between the farmhouses – the farmyard. The noun ON tuft, and the derivations, toft, topt, tupt and tompt, denoting sites or plots within the settlements – built, unbuilt, owned, rented, or abandoned – connect to both urban and rural nucleated settlements. Furthermore, ON hofudból, denoted central farms and related to social hierarchies, perhaps similar to a manor. The ON term garðr, etymologically identical with English ‘yard’ and the Scottish garth (Winchester 1987: 60), originally denoted the fenced-in settled and arable land. Its secondary extended meaning as a farm and a settled, economic and agrarian unit, as used in Norway today, first seems to originate from the 13th century. At that time, it appears as more frequent in the eastern part of the country and the Trøndelag region in mid-Norway than in the western parts. These areas, south-western, western and northern parts of the country, were to a larger degree characterised by nucleated and more composite clustered settlements (Holmsen et al. 1956: 29). Today they are referred to as ‘klyngetun,’ a common tún where buildings belonging to different households were centred. Originally, Norse people also lacked a special term for holding (Norw. bruk), as a unit of production occupied by a household, normally a family (Sandnes 1979: 166). In Denmark and Sweden, this is still the case, and the notion ‘gård’ there denotes settlements of one household, equivalent to holding in Norway. This may explain the later terminological divide as for nucleated settlements in the Scandinavian countries. However, in Norway, the specific name of a so-called navnegård covered it all, by giving various agricultural units, whether clustered or single, a specific common name. A farm name covers, then, several layers of meaning, as for status, structure, size and time depth. In fact, it corresponds to the naming of villages also with their own names and without specifying holdings/farmsteads therein. The terminology used, and linguistic differences between regions and countries may, then, conceal similarities, while medieval terminology reveals them. The English term ‘village’ of Latin vill seems to equate OE tūn and as in ON tún, meaning the habitation of the land of common and open fields – a core land, either as a basic unit in whole, or in parts. This again has developed into ‘town’ and township, denoting 75

The Archaeology of Medieval Villages Currently Inhabited in Europe

the agricultural landscape and core unit of landscape organisation of the corporate field system. This may vary in size and shape of township boundaries, and with an interwoven structure of field systems. In many cases, such townships have survived into the recent past. In northern regions of England, they often coincide with the medieval ecclesiastical parish, and larger parishes concur with multiple townships. They show a variety of forms within the broad structures of extensive communal fields and are associated with a wide range of dispersed settlements and intermediate types too (Moorehouse and Bond 2016). Can this also be the case in Vik? When approaching more sustainable units than deserted settlements from a specific period and rather as integral processes of social interaction in long-term perspectives, the aim is to widen the outlooks and understanding of nucleation processes in the past. State of research – outlooks and perspectives Although there are many historical parallels between Scandinavian countries, in research and scholarly traditions when approaching Viking Age and medieval settlements, there are also differences as for source material, methodologies and approaches. History has generally provided important frameworks and chronologies on a macro level and of influence on approaches and interpretations within other disciplines too. Archaeology on its side is increasingly able to provide new data, and revise chronology. Today, new evidence on both micro and macro levels is gained from using a whole range of refined methodologies; localising structures from the air from aerial photos and LIDAR, and at ground level from total stations and GIS, and on micro levels from excavations on different scales. Knowledge related to rural and agrarian development in the medieval period also relies on evidence and approaches achieved over time. Results from multi-disciplinary studies within history, economic history, toponymy, geography, philology as well as archaeology, form a scholarly basis, although under continuous discussion and revision. In Norway, as in Scandinavia generally, the study of medieval rural settlements dates back to the early years of the 20th century, both as a specialisation and as subtopics within political and economic history. Still, it builds on earlier achievements related to data collection, registrations of different source categories – narratives, law codes and diplomas, letters and documents of juridical bearing. Since the 1930s to the present, a retrogressive method has gained a strong position in Norwegian agrarian studies, in history, geography, and partly ethnology and archaeology. The overall theoretical and methodological framework has shown a high degree of continuity, within a settlement pattern of smaller and dispersed farmsteads and a system of self-sufficiency compared to the more nucleated and regulated settlements further south. Norway’s general unfavourable conditions for agriculture and egalitarian conditions in peasant community compared to its neighbouring countries to the south have been emphasised. Because of a high degree of structural continuity in the Norwegian agrarian society, and its physical environments, the method is regarded as especially suitable (Holmsen 1981 [1940]). Accounts, tax lists, record evidence from the post-mediaeval periods and historical and cadastral maps, are used for documenting old farm structures and field systems. When used in combination with the more fragmentary medieval sources, and not least archaeological sources, it is possible to reveal and substantiate archaic features. Both historians and archaeologists have largely comprehended the development of farms as an organic evolution, from larger to smaller units, responding to demographic and economic variables. Old farms divided into separate farms with new names, or as nucleated holdings within the same farm, reflect such processes. A general theory has been that the old kin society of extended families with collective property rights was broken up in the early medieval ages, starting as a process in the Viking period and leading to a more individual and private orientation in settlement and organisation with separate farms (e.g. Holmsen 1966 [1935]: 27). In a project initiated by the Institute for Comparative Research in Human Culture in Oslo in the 1940s, the relationship between separate farms and different types of nucleated settlements appeared on the agenda for the first time and several case studies were implemented to throw light on their structure 76

Ingvild Øye - Is this a village? Approaching nucleated settlements in Scandinavian contexts

(Holmsen et al. 1956). Farms and settlements were also assessed in light of medieval jurisdiction, where old principles could still be traced in the landscape. The provincial Gulathing law for western Norway representative for the mid-12th century and the national Landlaw of 1274, refer to two principles of field division: (i) a division into two compact parts, and (ii) a division into strips of equal length and breadth (Bjørkvik 1956). The latter principle has led to a complicated field system that prevails in the hamlet-like farms in the west and north. The form and size of the fields could vary from long narrow strips to open fields or plots without systematic arrangements (cf. Figure 2). This system was partly based on common ownership and partly on an intricate division of territory, which gave each farmer a large number of scattered and intermingled strips and plots. The size and productivity of the infield decided how far it could be divided. The form and size of the fields could vary from long narrow strips to open fields that were not systematically arranged. However, they could also be allocated so that one group of holdings would have its strips and plots in one unit (Bjørkvik 1956). Such division of fields according to ‘just and equal’ principles both in quality and in extent is also a generally well-known system in North Western Europe (Dodgshon 1980: 36). Population growth was the initiator for splitting up in even smaller units, especially where arable land was limited, such as in western and northern Norway. At larger farms with wider areas of cultivated land, the core area could expand and the farmland shared communally between the households. When such nucleation processes started, has been debated but less studied. Attempts to reconstruct nucleated farmsteads as they appeared before the large division of farms from the middle of the 19th century onwards were also made (Figure 3), but restricted in time to the situation before the effects of the Land

Figure 3. Drawing and reconstruction of the buildings in the nucleated tun at the farm Seim, situated close to the fjord in Vik in the Sognefjord, as it appeared around 1870. Uncertain buildings drawn in dotted lines. (Drawing: A. Berg 1952).

77

The Archaeology of Medieval Villages Currently Inhabited in Europe

Commission Acts had given full effect (Berg 1968). However, with the exception of a comprehensive study from southwestern Norway (Rønneseth 2001 [1974]), they have seldom been studied in long-term perspectives and without greater impact on later research. Since the large-scale restructuration of clustered farms in the 19th century, the old farm structures have more or less disappeared in the rural landscape in Norway. Only a few have been preserved more or less intact. Over time, the opportunities for small-scale farming have also declined, thus creating a new wave of deserted farms that provide new archaeological prospects for assessing old land use. In Norway, the archaeological research of medieval rural settlements has mainly concentrated on deserted and rather marginal, dispersed and small-scale single farms in the south-west and western parts of the country and farm mounds in the northern parts of the country (Bertelsen 1979; V. Martens 2016; Petersen 1933; 1936). Until the 1970s, archaeological investigations were mainly engaged in building constructions and the layout of buildings. However, new interest in the ecological processes in this field called for closer co-operation between archaeologists and natural scientists, resulting in investigations of usage and the sustainable capacity of the natural resources (e.g. Kaland 1987; Pedersen 1990; Randers 1981). Generally, it has been difficult to trace more sustainable settlements from the Viking period and the Middle Ages (J. Martens et al. 2009), probably due to the permanency in their location. Danish and Swedish archaeological research on medieval nucleated settlements has generally concentrated on buildings related to later known plot structures within smaller and larger villages, but generally less on the surrounding environments. These studies cover both prehistoric moving and medieval stationary villages, based on both small-scale and larger open field excavations. As in Norway, information from later documents and maps has been important for understanding the more fragmented archaeological remains as parts within larger structural frameworks, and the connection between plots, farms/holdings and the common farmland as known from later periods (Steensberg 1956). The main outline of villages within the borders of the medieval Danish realm (including Skåne, Blekinge, Halland and Västermanland, presently in Sweden) consisted of regulated plots and buildings for the individual households. These plots, denoted as toft, were located along a common road and an open common space, and surrounded by common fields, regulated within a rotational arable farming system. This regular and seemingly planned pattern has been seen as results of a controlling influence from political authorities or property owners from the turn of the first millennium to the 13th century (e.g. Porsmose 1988). Today, many of the old conceptions of society, structurally, socially and chronologically, have been questioned. The most influential frameworks for explaining patterns of medieval settlement concentration and their relation to social and economic change in Norway now rather focus on power, coercion, lordship and communalism. These factors have impact on resource-management and rural structures too. They indicate a more hierarchical structure, rooted in landownership over larger territories, more like multiple estates further south and to the west, although on a smaller scale (F. Iversen 1999; 2004; T. Iversen and Myking 2005; D. Skre 1998). Medieval landownership in Scandinavia reflects considerable regional differences. Around 1300, a large percentage of Scandinavian peasants were tenants, having farmed some 70 per cent of the land in Norway (Øye 2004: 8). In Denmark, closely integrated in estates, tenanted land amounted to as much as 85–90 per cent of the land (Hybel and Poulsen 2007: 209–10). In Sweden, freeholders, especially in the northern areas held more land; around 50 per cent (Söderberg and Myrdal 2002: 89). In Norway, tenanted land was mainly concentrated to the coastal districts, connected to estates around the major towns, while freehold was more common in inland and upland areas. Tenancy has largely been related to the advance of religious institutions; the king and local magnates expanding as main landowners in the early and high Middle Ages but may go further back based on archaeology in light of medieval sources. When medieval law refers to the buried in gravemounds, ON haugóðalsmaðr, 78

Ingvild Øye - Is this a village? Approaching nucleated settlements in Scandinavian contexts

for attesting status and lineages in conflicts of allodial heritage rights, it supports an interpretation as visual manifestations of property rights and built by the heirs to demonstrate their hereditary rights. Studies of distribution and spatial patterns of prehistoric monumental graves at borders and in close contact with historical settlements have shown to coincide significantly with later patterns of ownership of freehold and estates. Such references substantiate, then, patterns of early landownership and their absence indirectly reflects inequalities as for use rights (F. Iversen 1999; 2008; D. Skre 1998). Differences in landownership and hierarchies within rural societies may have contributed to regional differences in settlements, land use and social landscapes. Today, the older view of medieval Norway as primarily an egalitarian society, based on relatively homogenous farms and holdings is no longer obvious, although with some truth when regarded in broader geographical perspectives. Around AD 1300, clear regional patterns in settlement and land use appear on macro levels. By this time, nucleated settlements, whether called farms, villages or hamlets, were located in the areas with the best farmland – primarily in present-day southern Sweden and Denmark, and more scattered in western Norway and further north. Still, both nucleation and expansion of settlements are generally connected with upward demographic trends and are seen as a response to population pressure. However, the physical expansion, tenure and reorganisation of settled and farmed land are processes also linked with agricultural advances, new methods of tillage, improved technology, and collective management of fields. Assessing agglomerated settlements across national borders and from varied perspectives as long-term processes may give a broader understanding than when seen in shorter and closed periods and within limited areas. Structural, functional and chronological patterns of nucleation A landscape has varying qualities and options but is never static and depends on use and societal conditions. Nevertheless, the physical geography and its potential for long-term settlements able to feed its inhabitants, plays a crucial role. Available resources other than cultivable land, and not least in Norway, outlying areas, such as for hunting, fishing, iron extraction and quarrying, were central for producing a surplus. Hence, a combination of physical and geographical elements, climate, terrain and soil, as well as outlying resources, were significant factors for settlements and livelihood. These environmental conditions vary significantly in Norway, stretching across 13 degrees of latitude, as for climatic conditions, and between north and south, coast and inland, and lowland and highland with vast mountainous areas. Up to the present, no more than 3 per cent of the land has been under cultivation (Øye 2004; 2011). The most favourable conditions for agriculture are located in the eastern part of the country and in the region of Trøndelag. Western Norway, on the other hand, is characterised by its broken topography with uneven and limited pockets of glacial and post-glacial deposits suitable for cultivation – along the coast, by the fjords and in the valleys, stretching from sea to mountains. In western Norway, with its characteristic topography and differences in physical environment from coast to inland, the medieval settlement pattern was rather dispersed, often defined topographically by large stretches of uncultivated land bounding the farm areas. Therefore, the region comprised a considerable range of settlement types, including clustered settlements with multiple farms as well as truly dispersed and isolated farms. There is, however, a striking concurrence between nucleation of settlements on the most productive soils in mid- and inner fjord areas of western Norway, on open land of self-drained moraine terraces or moderate slopes. Neither Denmark nor Sweden are geographically homogenous, although the physical landscape is altogether more open and less hilly, but still varied in productivity. Generally, moraine soils facilitated a more intensive arable cultivation than more sandy and wet soils (Jacobsen 2013). The nucleation of settlements and the causes is not so clear, but hardly due to nature alone. Fragmentary source material and partial excavations make it difficult to get a full view of such processes, and retrospective analyses based on later information, may not reflect earlier development. Archaeological excavations in different contexts, combined with other source material are crucial, then, for establishing a more precise chronology and fuller understanding of such processes. 79

The Archaeology of Medieval Villages Currently Inhabited in Europe

Five cases from western Norway In a multidisciplinary research project, The traditional farm in western Norway (1995–1998), the opportunities for approaching farms and agrarian landscapes that were still in use but to a lesser degree affected by modern machine-driven farming, were tested out with small-scale excavations as for time depth and structural patterns (Austad and Øye 2001). One of the inherent theoretical and methodological conditions was that different time-scales and structures of different age are still preserved and can be discerned in the landscape, both above and under the ground. This is often the case in the somewhat vertical landscapes of western Norway, often with ‘pockets’ of fossilized, preserved agrarian structures, such as lynchets, clearance cairns, fences, and in some cases house sites too. By studying four historic farms in various zones from coast to inland, the aim was to gain a broader and more holistic understanding of the development of rural agricultural landscapes in the region. Today, the number of case studies has been extended to 11, of which five are historically nucleated farms (Øye 2011). In this context, the aim is to explore both variations and general traits in nucleation processes, and assess factors that may have contributed or restricted nucleation over time. The five hamlet-like farms represent various types of settlements of different scale and social status. They are all recorded as well established farms in written sources from the first part of the 14th century. All connect to fjord districts but at various gradients from coast to inland (Figure 4). The settlement clusters,

KVÅLE

ORNES

GRINDE VIK LEE

HAVRÅ

INDRE MATRE

Figure 4. Map showing nucleated farms from western Norway mentioned in the text (Drawing: P. Bækken).

80

Ingvild Øye - Is this a village? Approaching nucleated settlements in Scandinavian contexts

‘tun’, and main arable land, all locate to moraine terraces and productive soils at various altitudes, such as Cambro-Silurian minerals, phyllite, and others. On two ‘fjord farms’, both within topographically delimited areas, Indre Matre and Havrå in the county of Hordaland, the clustered settlements are situated on terraces, c. 40 and 75m a.s.l. respectively. The farm territories, however, stretch from sea level up into valleys of outlying grazing land, at c. 700 and c. 350m a.s.l. The farms in Sogn, Grinde, Lee and Ornes, also within topographically delimited areas, have their settlement areas at 100, 300 and 120m a.s.l, respectively. One of them, Ornes (also known as Urnes) situated in the inner parts of the long Sognefjord, stands out with its stave church from the early 12th century but with one or two forerunners dated to the early 11th century; the earliest dated stave church in Norway, and today on UNESCO’s world heritage list. It is located close to the old clustered tún, as documented in a cadastral map from 1865 (Hamre et al. 2007). During the 19th century, three of these farms were subject to land reallocation, and the first in 1830, at Indre Matre. Here, the common infield was largely divided in two, between freeholders living on the farm and the other owner, a larger manor outside the area (Zehetner 2007: 78–84). Similar but more comprehensive and complex divisions were carried out between freeholders at Ornes in 1865 and at Grinde in 1874, after both farms had expanded considerably in numbers of households during the preceding centuries. At Grinde, the old settlement cluster was largely broken up into separate units, and with a new relocated unit. Lee and Havrå, on the other hand, smaller than the others, seem only to have followed old medieval norms, without the same complex procedures. At Lee, an out-of-court agreement in 1865 between the then four holders, divided the common fields in two parts with a line through the middle of the largest arable field (Valvik 1998: 33). It was in accordance with the two partite system and ‘fair and just’ principles, based on the same extent and value, and same advantages as they had earlier derived from their fragmented shares. At Havrå, however, there seems to have been a more or less continuous shift of fields, as according to old norms, tenants could demand new assessment when a new tenant took over. Here, the settlement area was not affected other than to the extent caused by demographic changes. By the middle of the 19th century, the number of buildings – dwellings, barns, cow-houses and other outhouses – ranged from about 20 to more than 50 buildings in their common clusters at these five farms. The highest number was at Ornes with around 50 buildings in the common settlement nucleus on the terrace, located on both sides of the old road that still leads from the fjord and further eastwards to neighbouring farms and with some 20 more scattered outhouses (Lia 2005). The old road also connects to the early medieval church located on the north-eastern side of the tun. Altogether, the settlement covered an area of 0.9ha (Hamre et al. 2007). Ornes was clearly the largest of these farms; about four times larger than the average in its district according to tax lists from the first nationwide land register of 1667 (Iversen 1999: 49–50). Grinde was 1.7 times larger than farms in its district, while Lee ranged 0.7 below the average. Havrå, however, scored 1.75 above the average in its district (Austad and Øye 2001: 154), which was generally at a lower level than the farms in inner Sogn. Like Lee, Indre Matre was below the average, 0.8, in 1647 (Zehetner 2007: 241). However, the number of households and the area they disposed fluctuated over time. By the time of the land allocation, Ornes was run by six landholders, and Grinde and Lee by four each. At Havrå, the number expanded from eight to ten tenanted households from AD 1700 to 1900 (B. Skre 1994, 41–42), and at Indre Matre, four and additionally two crofters by the end of the 1800s (Zehetner 2007: 84–86). At this time, the farms had reached a peak in number of households, people, and buildings. Because of the land consolidation, we have accurate information about location and number of buildings, road systems, arable fields and grassland, productivity class, and the relation between them. At Ornes, for example, the cadastre map exposes a complex system of strip farming, creating a mosaic of grassland. Here, the arable land comprised 13.34ha (c. 31%) and grassland and pastures (c. 61%) of the farmland, and divided in altogether 459 patches of fields of varying sizes and with defined ownership. The best cultivation areas were in front and at the sides of the tun (Hamre et al. 2007). Grinde reveals a similar structure. Here, systematic name documentation of infield, outfield and shieling areas has given a yield of more than 600 individual names, including the outlying areas that reflect area use and ways 81

The Archaeology of Medieval Villages Currently Inhabited in Europe

of structuring them when approaching them archaeologically. The name flora within the main fence indicates intensive farming and on the terrace of more or less continuous cultivation (Øye 2002: 35–36). The three other farms, without similar maps, have not left the same detailed information. Still, ‘small’ names registered as part of various campaigns and projects from the 1930s onwards, have provided information about farm structures, farming and farming systems at the other farms too. The registered names vary in number from around 100 to 400 names (Julshamn 1998: 28–38; Larsen 2016: 188; B. Skre 1994; Valvik 1998; Zehtener 2007). At all the farms, the micro names were in fact useful tools in the field surveys of man-shaped structures and formed points of departure when choosing areas for small-scale excavations, making it possible to explore these last reminiscences of old farming systems. To illuminate the question of nucleation, the dating of main structural elements such as settlement sites, agricultural structures, and ways of farming within fields, fences, and main communication lines has been crucial to establishing how they connect or disconnect to later known patterns. These smallscale archaeological excavations have thus given information about the time depth of various land use, cultivation, shifting farming methods of extensive and intensive tillage, expansion and decrease of farmed land. Indirectly, intensification and expansion reflect periods of increased activity. Three of the farms, Ornes, Grinde and Indre Matre, have prehistoric grave mounds, substantiating their social status at an early stage. It is therefore interesting to compare them with archaeological traces from various structures within various zones; the closest infield, the outer infield and more distant areas related to the settlement core. At Ornes, eight larger cairns and mounds surrounded and demarcated the whole territory of what became the farm. Of these, four larger cairns marked the extreme point of the promontory close to the sea at the mouth of Lustrafjord, seemingly demonstrating territorial rights to the seagoing traffic into this sidearm of the main Sognefjord. One of the cairns, dated to late Bronze Age (1100–500 BC), contained graves from the Roman and Migration periods too, attesting continuity over centuries. The other four mounds seem to a have had a similar demarcating role related to the terrace of the best agricultural area and the settlement area. Two of them, dated to the Migration period (AD 400–550), were located in front of and at each end of the terrace, one of them rather monumental; 25m across and 2m high. The two other mounds, both dated to the Viking period (AD 800–1000), stood at the rear of the settlement area – one behind the church and the other further east close to a path used for leading cattle to the outfields, and marked in the cadastre map. The monuments from the Viking period bridging the upper terrace were less monumental and connected to later borders. Another visible monument, a standing stone, is located on the terrace brink and relates to an old road down to the sea, and in between two larger arable fields and close to an old infield border too, as marked in the cadastre map. Just at this point, the settlement becomes visible when coming up the sea road. The stone is of a type generally dated to the early Iron Age in Norway but was also productive into the late Iron Age (Lia 2005: 49–51). The Ornes archaeological traces from the various zones of close and distant cultivation are best represented in the zones beyond the main terrace. Here, in the deeper ploughed fields with less preserved fossil remains, the oldest dated layer stems from the early Viking period. Outside this area, a series of 14C-datings from various lynchets and structures give a fuller picture of long-term use; from the late Bronze Age through all following periods into the early Modern period. The outer zones near the outfields, were also well represented with fossil structures, and provided even longer time depth of use stretching from about 3500 BC, the Neolithic as the earliest indication, through the late Bronze Age and the Pre-Roman period. Except for a lacuna between 320 BC–AD 100, they showed time representations from all later periods into the Early Modern period, and marked by intensive and expanding cultivation in the Middle Ages (Lia 2005: 70, 80, 88). In main outlines, these representations seem, then, to match with the dating of the earliest grave cairn by the fjord and shows that the area had been taken into use before it was erected, and more intensively used from the late Bronze Age and early Iron Age onwards. The spatial pattern of these prehistoric monuments, then, coincides amazingly with the main structures documented in the cadastre map. An extensive and intensive use of areas outside the best cultivation area signifies large-scale extensive farming at an early stage. 82

Ingvild Øye - Is this a village? Approaching nucleated settlements in Scandinavian contexts

Although it has been difficult to date the main arable terrace earlier than to AD 800 because of heavier cultivation, the other dated structures both indirectly and directly, demonstrate Ornes’ position as a productive and demarcated area at an early stage. Grinde is also marked by early grave monuments, but none of them excavated. Based on construction, shape and size, however, they appear as belonging to the early Iron Age, or probably late Bronze Age. One of them, a large cairn 25m across and c. 2–3m high, is situated at a crag close to a main arable terrace and two other cairns further east on the same terrace. Earlier there were also four larger cairns near the sea front, like at Ornes, and the largest furthest west on a promontory close to the farm’s present border. Here, a chest of flagstones was found, probably from the late Bronze Age (Øye 2002: 27–8). In this case, too, the monuments seem to concur with the earliest dating of land clearing and cultivation in the early Bronze Age (1875–1680 BC). Traces of more intensive farming are dated to the Pre-Roman period, c. 510–200 BC, and with further expansion of the late known infields in the late Roman period, AD 200–410 (Austad and Øye 2001; Øye 2002: 42, 50). The prehistoric graves at Indre Matre show another pattern. Some 20 grave mounds and graves are located in grave fields on each side of an old pathway leading from the settlement area to the outfields on the main terrace north and northeast of the tun. In this transition zone between infield and outfield areas and close to arable fields, the largest field (c. 140 x 40m) has two cairns and 10 mounds of varying size between 4 and 16m across and from ground level to 2m in height. Being broadly dated to the early and late Iron Age, they cover a period of nearly a millennium. The smaller grave field on the other side with six mounds is close to another pathway leading to an area of now deserted fossil arable fields, which are rather similar in size and structure and probably of the same time span. Additionally, three single grave cairns have been located, one in the outfields along the same pathway as the largest grave field, the two others in separate arable fields at some distance from the tun area (Zehetner 2007: 196–207). As they are closely connected to arable land of old age and to the settlement site, the graves probably relate to people connected to the farm. Being built as marked visual monuments and, as mentioned earlier, probably reflect property rights, they may possibly indicate an early two-partite division and demarcation of landownership. Through documentary sources, this can first be evidenced by the early 1600s, but test pitting in this border zone may indicate an older origin (Zehetner 2007: 210–11). At Havrå, with no physically marked grave monuments, trenches through the main infield fence and the pathway leading from the tun to the outfields have also corroborated an early dating of the later known farm structure. Here, the pathway for leading the cattle out of the infield area seems to date to the late Roman period, several centuries after the first clearing of land in the late Neolithic. The oldest traces of cultivation date to the Bronze Age, and through pollen analysis, cultivation of barley is documented to the late Bronze Age (Julshamn 1998). This is somewhat later than documented at Lee, attested by macrofossils. However, the large-scale stone-built wall around the infield, by then based on intensive cultivation, seems to be much later, dated to the late 13th/14th centuries and seems to represent an extension of the infield at this stage (Julshamn 1998: 78). A social marking in the landscape is moreover reflected in the early establishment of private churches on central farms, as referred to as a special category in the Gulathing law. The early church at Ornes clearly belongs to this category, based on the proximity to the common tun, and its small size, 16x7m. By the 12th century, written sources substantiate Ornes’ role as a magnate’s seat, when Sverris saga, written around 1200, connects high-ranking people in the King’s hirð, so-called lendirmennir and to Ornes as their residential farm (Magerøy 1991).1 Men of this status were prominent local magnates, attached to the king and owed him fealty and service. In this position, they received crown land as a personal fief but at the same time enjoyed a substantial income from their own land (Helle 2003: 200). This also must have been the case at Ornes, appearing as one of the largest farms in Sogn in the Middle Ages, owning parts of farms in neighbouring areas (Holmsen 1937: 54). Around AD 1300, this lineage was broken but the new owners still belonged to This has earlier been interpreted as another farm, Ænæs in Kvinnherad. The philologist, Professor Hallvard Magerøy has reassessed the sources and convincingly argued that it is a misconception and should be Ornes. 1 

83

The Archaeology of Medieval Villages Currently Inhabited in Europe

the upper echelon of society, and with their own huge estates in other parts of the country. The farm thus remained in the ownership of magnates and nobility for at least a period of 500 years, and also later related to the upper social strata (Magerøy 1991). Judging by the gravemounds and intensive farming over wide areas, its importance had roots of great time depth. The spatial demarcation, the location of the settlement area with its panoramic view over the inlet to the Lustrafjord underlines Ornes’s strategic position too, able to control seagoing traffic and access to areas further inland. The five nucleated farms, then, represent farms of different scales and social topography but all have documented structures of long durability and show long-lasting use within larger areas, and a high degree of stability in the settlement core and surrounding infields. The traces of intensification and expansion of land evidently respond to demographic patterns, which are more difficult to trace in the landscape based on limited archaeological opportunities. Monumental manifestation also coincides with periods of intensification in both the initial phase and later expansion and intensification in the early Iron Age, and later in the Viking period and early Middle Ages. At all the investigated farms, the infields expanded during the late Iron Age and early Middle Ages, that also saw an expansion of activities in outlying and mountainous areas, related to animal farming, and other resources (Øye 2002, 2011). The connection to prehistoric grave mounds, medieval churches or chapels, also has structural, chronological and social impact and largely concurs and corroborates the findings in a temporal perspective. From the Middle Ages onwards, written sources add information about shifting numbers of holdings and changing ownership, and general demographic trends over the centuries. The archaeological observations seem generally to concur with a general increasing demographic trend in the Viking period and early and high Middle Ages. However, the following period of demographic crisis and desertion is not reflected in the archaeological material, although Lee may be doubted. Although Grinde does not appear in a general census in 1520, there are no signs of desertion in the archaeologically dated layers in profiles from the farm (Austad and Øye 2001). However, larger and more sustainable farms like these were generally less affected by decline than the more marginal farms were. Still, it must be admitted that the methods used may be too coarse for assessing demographic trends other than as indicative, and in broad outlines. Agglomeration and central places In rare cases, medieval and thus contemporary sources give detailed glimpses of the built environment of nucleated settlements. This is the case at Kvåle in Sogndal. This is also one of the larger farms in Sogn and western Norway and, like Ornes, housed lendirmenn and their households from the late 12th century onwards. In 1314, Kvåle was divided between three heirs, and therefore recorded in a legal document. The arable land and farmland is not mentioned, only the buildings and unbuilt plots clustered together in the same tun. It comprised altogether 20−30 buildings: several dwelling houses, both old and new, and additionally, outhouses, barns, byres, a horse stable, woodsheds, an oust, a sauna, a house for cooking, a porch and a weaving house, and also a church on the outskirts of the cluster, and several buildings belonging to the priest (Øye 1986: 410–12). As a residential farm for the elite, and, judging by the mention of premises for both male and female servants it was worked by subordinates. The church, built in stone probably in the late 12th century, was seemingly privately founded and later became the centre in a small parish, probably originating from an early Kvåle estate, thus adding income of land rent and tithes to sustain the church and priest. Medieval documents underline the size and wealth of the farm, of 120 so-called mánaðarmataból representing one of the highest values in the country (Øye 1986: 356). This is a measure used for taxation of land and land rent, originally representing the value of the daily food ration for one man in one month (Steinnes 1936: 84). This is exactly the same value as recorded at Ornes (Magerøy 1991: 252). In the late Middle Ages and in the 16th and 17th centuries, 6–7 tenants with their households worked the farm (Øye 2000: 17). The church was by then abandoned; deserted in the early 1500s according to documentary sources (Øye 1986: 266), and Kvåle seemingly had lost its earlier status. A study of medieval church sites in the inner parts of Sogn shows that as many as 12 of 22 medieval churches were located at larger nucleated farms and in close connection to the settlement areas. They are, like Ornes, Kvåle and the churches in Vik, interpreted as early privately founded churches, which 84

Ingvild Øye - Is this a village? Approaching nucleated settlements in Scandinavian contexts

later became churches within smaller or larger parishes (Brattekværne 2006). A study of abandoned and deserted medieval churches and chapels in two counties in western Norway shows a similar pattern, where the great majority of altogether 34 deserted medieval churches were located close to old clustered settlements of high rank, and as originally private churches (Buckholm 1998). They indicate, then, an expanding period of high population density, nucleation at larger farms, hierarchical and manor-like structures, strengthened by new institutions, and their rise in the early Middle Ages and decline by the end of the medieval period. Only within a limited area of Sogndal three of four medieval churches disappeared by the late Middle Ages. However, in Vik only one of three was deserted, but to be rebuilt in the 19th century, as the medieval churches Hove and Hopperstad no longer fulfilled the requirements of parish churches and again fell into private ownership. Like at Ornes, they are now protected heritage. Farm structures, composite or single farms, and location, should therefore not only be seen as selfsustaining units but also as closely integrated within hierarchical structures and related to larger socioeconomic and socio-political systems. High quality land, convergence of communication lines, and centrality of these settlements were other central factors for nucleation and social and economic status. Once established, the settlements had great durability. Many could have been of the size of the clustered settlement at Kvåle, stimulated by both sociopolitical and demographic factors with an escalating population from round the turn of the millennium shift with a peak around AD 1300 and heavily reduced during the late medieval period to expand again in the Early Modern period. However, as late as by the late 18th century, the population does not seem to have regained the levels of the early 14th century (Lunden 2004: 159). However, not all clustered farms were large or had central functions. The investigations show that even moderate sized farms such as Havrå, Lee and Indre Matre went through similar nucleation processes over time but on a smaller scale. Regional differences in processes of agglomeration? Are there then parallels between the agglomerated settlements appearing in western Norway with archaeologically documented villages further south? Based on different methodologies and opportunities as for investigations, such comparisons are rather difficult. While Danish archaeologists have been able to detect and investigate larger areas of nucleated settlements from the Viking period into the Middle Ages, this has not been possible in Norway. Another major difference is the physical landscape itself and degree of open land, which has obvious consequences for ways and space for cultivation, and consequently for access to larger areas of land, and then again expansion, mobility and lastly stability. A striking contrast to the place bound settlements in western Norway is therefore the settlements further south, that moved cyclically within larger resource areas based on extensive farming, as revealed at the large-scale excavation at Vorbasse in Jutland (Hvass 1986) and at other sites, close to later stationary villages from the Middle Ages. These settlements based on extensive farming, were relocated cyclically within larger resource areas, until they became stationary in the early Middle Ages, and some also became church centres. The chronological frames for this change vary from c. AD 1000–1200 and AD 950–1100 (Grøngaard Jeppesen 1981; Porsmose 1981; 1988). However, later studies have shown that the formation and regulations of nucleated villages happened to a large degree in the 12th century but developed over a longer time span into the 14th century (Carelli 2001: 32–38). Excavations of the church village Kyrkheddinge in Skåne have been able to trace the development of a main farm from the late Viking Age as it expanded as a nucleated village from the beginning of the 12th century, but still not within regular plots. By the end of the century, a stone church was under construction, possibly replacing a wooden church from the preceding century. Judging by later landownership in the late Middle Ages when larger parts of the land were owned by the church and the local elite, local magnates or the church seem to have taken a founding role in Kyrkheddinge, similar to what has been observed in other rural nucleated settlements with churches too (Schmidt Sabo 2001). This process also coincides in time when nucleated farms in western Norway became church centres and strengthened their position as central loci within close-by estates, and later as centres within their own parishes. 85

The Archaeology of Medieval Villages Currently Inhabited in Europe

A large-scale excavation in Tårnby in Zealand of 7500 m2 made it possible to follow the development of one of the farmsteads within its plot structure of the medieval village of 12 farmsteads from around AD 1100 into the early Modern period. Although the number was stable, complex restructuring of buildings went on over time (Svart Kristiansen 2005). Another analysis of structures and chronology in 35 partially excavated village sites has further substantiated that they were established from the turn of the millennium into the 13th and 14th centuries. What is interesting compared to Norwegian clustered settlements is that the plots and settlement patterns were generally more flexible and varied than shown in later documentary evidence. They also show regional variations at least until the beginning of the 13th and 14th centuries, and then seemingly based on a general regulation (Svart Kristiansen 2006). Such a development has also opened for new interpretations, caused by changes in social patterns – based on smaller households within smaller economic units (Carelli 2001; Svart Kristiansen 2006; 2016). This may well be the case in Norway too. In Norway, the nucleated settlements cannot be exposed at this level based on contemporary sources, but only indirectly through investigations of the farming land and so far based on small-scale operations. By analysing the stratigraphy in the profiles and radiocarbon dating, phases of later structures, as evidenced though cadastre maps, show some common characteristics: a common road going through the settlement cluster, which connects to the best agricultural land. Larger nucleated settlements centrally located on larger areas of cultivable land, also here appear as centres for magnates and as church centres in the early and high Middle Ages, private at first instance to become centres in smaller or larger parishes by the 14th century at the latest. Together, archaeology and landscape studies combined with written evidence have been able to build a more solid empirical platform, which brings nuances into the broader picture of settlement patterns in time and space in both Norway and in southern areas of Scandinavia. Recent research also modifies and questions earlier interpretations about structures and chronology. It also gives a more varied pattern of nucleated settlements and changes over a longer time span than forwarded in earlier interpretations. Altogether, nucleated farms as well as villages may have developed over longer drawn-out periods, and the more sustainable of them avoided desertion. Farms, hamlets or villages? Returning now to Vik and the initial question posed – what is a village? When the Bishop in Bergen visited his churches in his diocese in 1823, and churches in Vik, he described the place as a small village. Besides clustered farms on the lower terrace, of which three were close-by early medieval churches, some 240 leaseholders living off their various crafts and services occupied the seafront (Hopperstad 1956). Similar composite agglomerations had also developed in other areas along the fjord with denser populated settlements, such as Sogndal, although here three of the four medieval churches no longer existed. Could these medieval nucleated settlements also be called villages? Terminologically, we have seen overlapping medieval terms across national borders in larger parts of Scandinavia. The number of buildings within nucleated settlements, whether called farm or village, also overlap. Demographic patterns and socio-political changes also follow parallel trajectories, although not necessarily altogether synchronic. The elite, powerful lords or ecclesiastical institutions – also played a crucial role as agents for forming central places in Scandinavia in the early and high Middle Ages, like in Britain too (Lewis, Mitchell- Fox and Dyer 2001). However, when studying nucleation in long-term perspectives, a fixed and static terminology – farm, hamlet or village – may seem less relevant as an analytical tool – as places change in time, functionally, structurally and in importance but have a longer prehistory explaining the development. In western Norway, such long-term development can be observed at places such as Ornes and Kvåle, probably qualifying for all of these concepts over time, from farm, hamlet to village, and back to farm again, based 86

Ingvild Øye - Is this a village? Approaching nucleated settlements in Scandinavian contexts

on degree of nucleation, size, structure and central functions over time. In Scandinavia, moreover, the difference between hamlets, farms, and villages is rather blurred by the minimum definition of a village. Also in Britain, it can be difficult to draw the line between these categories (Fowler 2002: 118–120). Still, the physical environments promote and limit opportunities, and nucleated settlements cannot always be compared meaningfully disregarding physical environments and wider societal conditions. In Scandinavia, with the most fertile agricultural areas to the south, changes in arable farming seem to appear earlier than further north. In the western and northern areas of Scandinavia, innovations were largely related to the various resources in the outlying areas; fishing, hunting, quarrying and iron production, although intensification by using ard and plough besides spades and hoes was also important. The development, then, does not appear synchronically and in all parts of Scandinavia (Øye 2016). Regional studies serve, then, as favourable frameworks for comparative studies, being able to reveal various patterns and processes in time and space. They make it, then, easier to detect both general conditions and variations, continuity and changes. These opportunities of regional studies are also experiences made in England (Lewis, Mitchell-Fox and Dyer 2001: 25). When assessing nucleation processes in long-term perspectives it may also be rewarding to assess them at various levels spatially, at micro, meso and macro levels, and in time, within Braudel’s wellknown interconnected tripartite division of time – conditions that favour processes of longue durée, more cyclical processes of shorter periods, and events of far reaching and long lasting effects. All these elements interact and can be recognised when studying nucleation processes in long-term perspectives. Physical environments shape opportunities for sustaining larger concentrations of people and regional variations of long durability related to resources, regardless of shifting technologies and methods of taking them into use. Climatic conditions, demographic fluctuations, new ways of organisation and exploitation of land and resources may lead to slow-moving changes of shorter and longer durability. More rapid changes, caused by specific events, such as plagues, wars and socio-political factors of shifting power and new institutions could also have long lasting effects. All these influences contributed in the shaping of nucleated settlements but of varying impact over time. The environmental conditions for food production was the longest lasting, and common for all the areas studied here. Central location and communication lines also stand out as elements of long lasting importance. Demographic patterns interacting with societal development are more complex and archeologically detected only indirectly as broader trends, as periods of intensification and expansion. Such processes appear far back in the late Bronze Age, in the Roman period and most visibly in the Viking period and the early Middle Ages. Some had clear consequences in the rural landscapes, with rising power centres in areas able to feed larger populations and provide surplus, and within hierarchical structures, organised within estates and with a subordinate work force and tenants. Hence, shifting socio-political conditions had great impact on what places became nucleated that later could also cause their decline. Such trends can be recognised over larger territories in Scandinavia too. However, nucleated settlements of long-lasting durability seem largely to be due to their environmental opportunities, related to landed resources, methods of production and ways of organisation. Ornes and similar places, give then insight into social relationships and management, not unlike settlements denoted as villages in other Scandinavian areas too. Extending the scope in time and across regions and countries seems then rewarding for how to recognise and understand how both specific and general mechanisms within various types of landscapes and societal conditions interacted in time and space. Today, new technologies, other livelihoods and communications have largely broken these lines. However, a few of the places under discussion still appear as truly nucleated villages, such as Vik, with central functions of today and as a meeting point between the fjord and inland areas and with still-standing churches as visual symbolic manifestations of old structures and heritage. 87

The Archaeology of Medieval Villages Currently Inhabited in Europe

References Adams, I. H. 1976. Agrarian Landscape forms. London. Austad, I. and Øye, I. 2001. Den tradisjonelle vestlandsgården som kulturbiologisk system. In B. Skar (ed.), Kulturminner og miljø. Forskning i grenseland mellom natur og kultur: 135–205, Oslo. Norsk institutt for kulturminneforskning. Berg, A. 1968. Norske Gardstun. Instituttet for sammenlignende kulturforskning, Skrifter serie B, vol. 55, Oslo. Bertelsen, R. 1979. Farm mounds in north Norway. A review of recent research, Norwegian Archaeological Review 12: 48–56. Bjørkvik, H. 1956. Farm Territories. Habitation and Field Systems, Boundaries and Common Ownership. In A. Holmsen, H. Bjørkvik and R. Frimannslund (eds.), The old Norwegian peasant community. Investigations undertaken by the Institute for Comparative Research in Human Culture. Scandinavian Economic History Review: 33–61. Brattekværne, C. 2006. Kirkesteder i Indre Sogn: en arkeologisk analyse av religionsskiftet, med utgangspunkt i kirkesteder i Indre Sogn, Unpublished master thesis in archaeology, University of Bergen. Buckholm. M. B. 1998. Nedlagte kirker og kirkesteder fra middelalderen i Hordaland og Sogn og Fjordane: Hvilke kirkesteder ble nedlagt, når, og hvorfor? Arkeologiske avhandlinger og rapporter fra Universitetet i Bergen 1. Bergen. Carelli, P. 2001. En kapitalistisk anda. Kulturälla förändringar i 1100-talets Danmark. Lund Studies in Medieval Archaeology 26, Stockholm. Dodgshon, R. 1980. The origin of the British field systems. An interpretation. London. Erixon, S. 1960. Svenska byar utan systematisk reglering, En jämförande undersökning, Stockholm. Fowler, P. 2002. Farming in the First Millenium AD. British agriculture between Julius Caesar and William the Conqueror. Cambridge. Grøngaard Jeppesen, T. 1981. Middelalderbyens oppståen. Kontinuitet og brud i den fynske agrarbebyggelsen mellom yngre jernalder og tidlig middelalder. Hamre, L. N., Domaas, S. T, Austad, I. 2007. Land-cover and structural changes in western Norwegian cultural landscape since 1865, based on old cadastral map and field survey. Landscape Ecology Vol. 22, Issue 10, 1563–1574. Helle, K. 1995. Down to 1536. In R. Danielsen, S. Dyrvik, T. Grønlie, K. Helle and E. Hovland (eds.), Norway: a history from the Vikings to our own time: 3–119. Norwegian University Press, Oslo. Helle, K. 2003. The Norwegian kingdom: Succession disputes and consolidation. In K. Helle (ed.), The Cambridge History of Scandinavia, Vol. I. Prehistory to 1520: 369–91. Cambridge, Cambridge University Press. Holmsen, A. 1937. Økonomisk og administrativ historie, Sogn. Norske bygder b. IV, Bergen. Holmsen, A. 1961. Norges historie. Fra de eldste tider til 1660. Oslo. Holmsen, A. 1966 [1935]. Gard – bygd – rike. Festskrift til Andreas Holmsens 60-års dag 5. juni 1966. Oslo-BergenTrondheim, Norwegian University Press. Holmsen, A. 1981 [1940]. Nye metoder innen en særskilt gren av norsk historieforskning, Historisk Tidsskrift, 32: 27–45. Holmsen, A., Bjørkvik, H. and Frimannslund, R. (eds.) 1956. The old Norwegian peasant community. Investigations undertaken by the Institute for Comparative Research in Human Culture (Oslo), Scandinavian Economic History Review 4: 17–81. Hoprekstad, O. 1956. Storskiftet i Vik 1859–1864, Bygdabok for Vik i Sogn Bd. II: 299–313, Bergen. Hvass, S. 1983. Vorbasse. The development of a settlement through the first millennium A.D. Journal of Danish Archaeology 2: 126–136. Hybel, N. 1989. Crisis and Change. The Concept of Crisis in the Light of Agrarian Reorganization in Late Medieval England, Århus. Hybel, N. and Poulsen, B. 2007. The Danish Resources c. 1000–1550. Growth and Recession. Leiden, Brill. Iversen, F. 1999. Var middelalderens lendmannsgårder kjerner i eldre godssamlinger? En analyse av romlig organisering av graver og eiendomsstruktur i Hordaland og Sogn og Fjordane. Arkeologiske avhandlinger og rapporter fra Universitetet i Bergen. Bind 4. Bergen. 88

Ingvild Øye - Is this a village? Approaching nucleated settlements in Scandinavian contexts

Iversen, F. 2005. Royal manors and estates in Western Norway in the late Iron Age and Middle Ages, In T. Iversen and J. R. Myking (eds.) 2005. Land, lords and peasants. Peasants’ right to control land in the Middle Ages and the early modern period – Norway, Scandinavia and the Alpine region. Report from a seminar in Trondheim, November 2004, Skriftserie fra Institutt for historie og klassiske fag, NTNU: 133–144, Trondheim. Iversen, T. 1996. Trelldommen. Norsk slaveri i middelalderen. Historisk institutt, Universitetet i Bergen, Skrifter I, Bergen. Jacobsen, J.G.G. 2013. Viking Landscape Regionality. An Interdisciplinary Model for Regional Classification of Cultural Geography in medieval Scandinavia c. 800–1300. In R. Efe, I. Atalay and I .Cürebal (eds.), International Geography Symposium – GEOMED 2013: 89–100. Julshamn, L. 1998. Havrå – fra fjorden til fjellet. En arkeologisk punktundersøkelse av gården Havrå, Osterøy, Hordaland. Unpublished master thesis in archaeology, University of Bergen. Kaland, S. H. H. 1987. Viking/Medieval Settlement in the Heathland Area of Nordhordland, Universitetets Oldsaksamlings skrifter. Nye rekke nr. 9, In J. Knirk (ed.), Proceedings of the Tenth Viking Congress: 171– 90, Oslo. KLNM= Kulturhistorisk leksikon for nordisk middelalder, 1–22, Oslo 1956–78. Larsen, J. 2016. Archaeological investigations of an agrarian landscape in western Norway – The fjord farm Indre Matre. In J. Klápsté (ed.), Agrarian Technology in the Medieval Landscape, Ruralia X: 185–194, Turnhout, Brepols. Lewis, C, Mitchell-Fox P. and Dyer, C, 2001. Village, Hamlet and Field. Changing Medieval Settlements in Central England, London. Lia, V. 2005. Ornes i Luster – en arkeologisk landskapsanalyse med punktundersøkelser i innmark. Unpublished master thesis in archaeology, University of Bergen. Liebgott, N.-K. 1989. Dansk middelaldearkæologi, Copenhagen. Lunden, K. 2004. Recession and new expansion 1350–1814. In R. Almås (ed.), Norwegian Agricultural History: 144 –232, Trondheim, Tapir academic press. Magerøy, H. 1991. Urnes stavkyrkje, Ornes-ætta og Ornes-godset. Norröna et islandica. Utvalde artiklar av Hallvard Magerøy: 233–257, Øvre Ervik. Martens, J., Martens, V, and Stene, K. 2009 (eds.). Den tapte middelalder? Middelalderens sentrale landbebyggelse. Varia 7. Oslo. Martens, V. 2016. North Norwegian farm mounds – economic resources and landscape conditions, In J. Klápsté (ed.), Agrarian Technology in the Medieval Landscape, Ruralia X: 173 –183, Turnhout, Brepols. Moorehouse, S. and Bond, J. 2016. An approach to understanding medieval field systems across Europe: the structure and anatomy of township field systems. In J. Klápsté (ed.), Agrarian Technology in the Medieval Landscape, Ruralia X: 1–48, Turnhout, Brepols. Pedersen, E.A. 1990. Rydningsrøysfelt og gravminner. Spor av eldre bosetningsstruktur på Østlandet, Viking 53: 50–66. Petersen J. 1933. Gamle gårdsanlegg i Rogaland. Instituttet for sammenlignende kulturforskning. Serie B XXXI, Oslo. Petersen J. 1936. Gamle gårdsanlegg i Rogaland. Fortsettelse. Instituttet for sammenlignende kulturforskning. Serie B XXIII, Oslo. Porsmose, E.1981. Den regulerede landsby I–II. Studier over bebyggelsesudviklingen på Fyn fra ca 1700 til ca 1000 e. Kr. fødsel, Odense. Porsmose, E. 1988. Middelalder o.1000–1536. In C. Bjørn (ed.), Det danske landbrugs historie i Oldtid og middelalder: 205-417, Odense. Randers, K. 1981. Høybøen – en ødegård på Sotra. En undersøkelse av bruksperioder og erverv basert på bosetningsspor fra eldre jernalder og middelalder, Unpublished master thesis in archaeology, University of Bergen. Rønneseth, O. 2001 [1974]. Gard og gjerde – faser i utviklingen av Jærens kulturlandskap. Oversettelse og forord av F-A. Stylegar. Stavanger. Sandnes, J. 1979. Ødegårsdprosjektet og tallet på gårdsbruk i Norge i høgmellomalderen. Historisk Tidsskrift bd. 48 1979: 114–119. 89

The Archaeology of Medieval Villages Currently Inhabited in Europe

Schmidt Sabo, K. 2005. Den medeltida byns sociale dimensioner, Riksantikvarieämbetet Arkeologiska undersökingar Skrifter 67. Lund Studies in Historian Archaeology 1. Stockholm. Skre, B. G.1994. Havråboka. Soga om ein gamal gard på Osterøy. Osterøy. Skre, D. 1998. Herredømmet. Bosetning og besittelse på Romerike 200–1350 e. Kr., Acta Humaniora, University of Oslo. Sporrong, U. 1990. By. Nationalencyklopedian. Bd. 3. Steensberg, A. 1956. Bebyggelsen på landet i Danmark i historisk tid. Nordisk Kultur XIII. Landbruk ock bebyggelse, Stockholm. Steinnes, A. 1936. Mål, vekt og verderekning i Noreg i millomalderen og ei tid etter. Nordisk kultur XXX, Stockholm. Svart Kristiansen, M. 2011. Studies of tofts, peasants and hierarchies in medieval rural Denmark – some comments from an archaeological perspective. In J. Klápsté (ed.), Hierarchies in rural settlements, Ruralia IX: 313–328, Turnhout, Brepols. Söderberg J. and Myrdal, J. 2002. The Agrarian Economy of Sixteenth Century Sweden. Acta Universitatis Stockholmiensis. Stockholm Studies in Economic History 35, Stockholm, Almqvist and Wiksell International. Valvik, K. A. 1998. Lee – en tradisjonell vestlandsgård? En arkeologisk punktundersøkelse av gården Lee, Vik, Sogn og Fjordane. Unpublished master thesis in archaeology, University of Bergen. Widgren, M. 1997. Bysamfällighet og tegskifte i Bohuslän 1300–1750. Skrifter utgävna av Bohuslän. Winchester, A. J. L. 1987. Landscape and Society in Medieval Cumbria. Edinburgh). Zehetner, J. L. 2007. Fra sted til gård. En agrararkeologisk analyse av Indre Matre i Kvinnherad, Hordaland. Unpublished doctoral thesis, University of Bergen. Øye, I. 1986. Sogndal i mellomalderen ca. 1050−1537. In Sogndal bygdebok b. I: 239–460, Bergen. Øye, I. 2000. Norway in the Middle Ages. Farms or hamlets – and villages too? In: Ruralia III, Conference Ruralia III, Maynooth 3rd–9th Sept. 1999: 12–23, Prague. Øye, I. (ed.), 2002. Vestlandsgården – fire arkeologiske undersøkelser: Havrå – Grinde – Lee – Ormelid ved L. Julshamn, R. L. Bade, K. A.Valvik og J. Larsen. Arkeologiske avhandlinger og rapporter fra Universitetet i Bergen 8. Bergen. Øye, I. 2004. Agricultural conditions and rural societies c. 800−1350. In R. Almås (ed.), Norwegian Agricultural History: 79–140, Trondheim, Tapir academic press. Øye, I. 2011. Agrarian Landscapes – Chronological Issues and Archaeological Challenges. In S. Sigmundsson (ed.), Viking Settlements and Viking Society. Papers from the Proceedings of the Sixteenth Viking Congress: 494– 506, Reykjavík. Øye I. 2016. Agrarian Technology and land use in Scandinavian landscapes c. 800–1300 AD. In J. Klápsté (ed.), Agrarian Technology in the Medieval Landscape, Ruralia X: 157–172, Turnhout, Brepols

90

The Archaeology of Currently Inhabited Villages in Spain: The Case of Asturias1 Margarita Fernández Mier Jesús Fernández Fernández Introduction: The study of abandoned villages and the renewal of medieval archaeology The recent publication of the book Treinta años de Arqueología Medieval en España [Thirty Years of Medieval Archaeology in Spain] (Quirós Castillo, 2018) is a good reflection of the process of consolidation that medieval archaeology has undergone in Spain. For decades, the archaeology of Al-Andalus was unquestionably the focus of theoretical reflection and empirical data (García Porras, 2018). However, in recent years an updated research agenda has renewed interest in northern Spanish medieval societies. This process has seen a convergence of different factors: the questions asked within medieval rural history in the late twentieth century (Fernández Mier, 2018b); the demonumentalisation of research in medieval archaeology, which has moved from churches, cemeteries and castles to the landscape (Martín Civantos, 2018); the spectacular development of preventive archaeology, especially in regard to cities and urban growth (Vigil-Escalera, 2018), and the impact of the broad experience of European research, initially Italian and French, and later British. All these strands have led to the generation of a programme of work, historical questions and the implementation of projects looking at abandoned villages. Their influence across the different regions of northern Spain has been uneven. In this context, we must analyse the redefinition of the archaeology of local communities and of rural societies through a wide range of methodological approaches. Such approaches have become more complex over the past 30 years, ranging from the identification of inhabited areas to the application of archaeobiological and geoarchaeological studies (Quirós Castillo, 2009; Vigil Escalera, 2015). Similarly, research carried out in other European historical contexts is shown in other works in this monograph. A large part of Spain’s landscape is now devoted to agricultural production. This is true of the Meseta Central, a wide, flat plateau in central Spain. Several abandoned medieval villages have been identified — not without difficulty — in these agricultural areas, dating from different periods and associated with different historical processes (Quirós Castillo, 2014). Within the framework of landscape archaeology, some efforts have contributed to reflections on the methodology of the surveying and analysis of these large, ploughed, agricultural areas, which have led to the recovery of abundant surface archaeological material. Although systematic surveys have been infrequent, recent full-coverage surveys in Spain have looked at forms of settlement in Late Antiquity and the Early Middle Ages (Tejerizo García 2017). However, only a few areas have been studied and there has been very little reflective appraisal of the information collected, especially in relation to the designation of agricultural and domestic spaces within the village. Given the scarcity of systematic surveys, most of the information on abandoned medieval villages comes from the widespread activities of commercial archaeology. This has produced a significant body of information that has been disseminated to varying degrees across Spain’s regions. The importance of this preventative archaeology has led some commentators to argue that it has been the real driving force behind the revival of the discipline, in particular with regard to methodology (Quirós Castillo, 2012), since it has paved the way for more widespread area excavations, the gradual incorporation of archaeobiological studies and the sequencing of common ceramics across wide areas. This has encouraged reflection on the concept of an archaeological site, especially with regard to ways of analysing and researching spaces used for agriculture and livestock that are linked to domestic areas. It has also led to the incorporation of a This study forms part of the ELCOS research project, Espacios locales y complejidad social. Las raíces medievales de un problema del siglo XXI [Local spaces and social complexity: the medieval roots of a 21st century issue], HAR2016-76094-C4-1-R, funded by the Ministry of Science and Innovation. 1 

91

The Archaeology of Medieval Villages Currently Inhabited in Europe

large number of actors into archaeological practice and to a reflection on rural medieval communities, against a background of a marginal interest in these matters by a large part of Hispanic medieval studies. Despite this historiographical renewal, monographs looking at rural archaeological sites are scarce. Aldaieta (Azkarate Garai-Olaun, 1999), Zaballa (Quirós Castillo, 2012), Gasteiz (Azkarate Garai-Olaun, Solaun Bustinza, 2013), Zornoztegui (Quirós Castillo, 2018) and Aistra (Reynolds, Quirós Castillo, 2007) are the main reference works in this field. One significant piece of information is the geographical diversity shown by the research: some areas have been widely documented, such as Madrid (Vigil-Escalera, 2012), the Basque Country (Quirós Castillo, 2012) and parts of Catalonia (López Mullor, 2011, Roig, 2013); while neither research projects nor preventive archaeology have advanced the definition of the network of medieval settlements in other areas (Tejerizo García, Quirós Castillo, 2018). The study of abandoned villages, carried out through archaeological projects by research groups as well as commercial archaeology, has, over the past two decades, generated a large body of data. This research has led to a greater understanding of the network of settlements of the post-Roman period in rural areas (Vigil-Escalera, 2015); of the changes that took place as a result of the imposition of Islamic rule from the eighth century onwards (Vigil Escalera, Quirós, 2013); and of the impact of the late-medieval crisis of the fourteenth century (Reglero de la Fuente and Sáez Sáiz, 1999). Thus, in areas devoted to agricultural activity, a large number of sites were surveyed during the twentieth century, many of which had been significantly altered by agricultural works relating to the mechanisation of the countryside over the course of the last century. The resilient landscape of mountainous areas The renewal of our understanding of the communities of northern Spain has been supported by the study of abandoned villages in the Meseta Central and Catalonia. However, it must be taken into account that a large part of Spain — almost half of its expanse — is mountainous, and the features of populated areas in these regions are very different. These population centres were less mobile and very few abandoned villages have been documented, with the survival of medieval settlements to the present day being fairly common. We believe that a discussion of the network of medieval settlements cannot take place solely through the study of abandoned villages, since this necessarily involves a partial view. This complex situation requires the inclusion of research into inhabited villages. In order to do this, interdisciplinary studies are required which also examine the causes of the uneven existence of abandoned medieval villages across the different areas. Spain’s mountainous regions have landscapes that were only partially affected by the liberal reforms of the nineteenth century and the land consolidation of the twentieth century.2 They are exceptional witnesses to the evolution of forms of ownership and exploitation of the land over the centuries, which have preserved the morphology of the landscape as well as the identities of the local communities who settled there during the Middle Ages (Fig. 1). These mountainous regions are characterised by low levels of mobility in their population centres and by landscapes with marked subdivisions of property, organised around smallholdings in a mainly mosaic landscape. Each family unit had a different number of small plots scattered across the different agricultural units that made up the communities’ crop and livestock farming areas. The predominance of these methods of distributing smallholdings and property encouraged the survival of collective and semi-collective practices in the use and management of the land until the mid-twentieth century (Fernández Mier, 2010). It is still possible to trace the forms of governance associated with this type of ownership, relating to the management of pasture and forest areas, which are still collectively owned under a wide variety of legal arrangements. Their survival makes In 1952, the first Spanish regulation on land consolidation was enacted. Up to the 1980s, projects were undertaken across the entire country to alleviate the problems of small-scale farming with one objective: to create smallholdings large enough to allow the mechanisation of the countryside. Land consolidation slowed down after the 1980s and its impact across Spain’s autonomous communities has been uneven, and marginal in mountainous areas. This is in contrast to the long history of land consolidation in other European countries and the role that it has played in the redefinition of ownership structures (Crecente Maseda, Álvarez López, 1999). 2 

92

Fernández Mier, Fernández Fernández - The Archaeology of Currently Inhabited Villages in Spain

Figure 1. Example of traditional mosaic landscape from the Northwest of the Iberian Peninsula. Mountain pasture area in Vigaña, Asturias.

the management of the land more complex, as well as hampering the success of the small amount of land consolidation that continues to take place under the governments of today’s autonomous communities in Spain. Associated with this landscape, a population has survived that maintains significant links with the farming communities that occupied these lands in the past; or at least more than with the commercial farmers who were the product of the nineteenth-century agricultural revolution in Europe. We can view and analyse these farming communities through an anthropological lens, which would allow us to expand the historical questions that for decades have centred research on these communities, generally viewed from the perspective of the data provided by documentary sources from the large estates. We must move towards the design of a research programme that will analyse farming communities through their own reality, as producers who established particular relationships with the natural conditions This was based on an exhaustive knowledge of the environment in which they lived, and on the use of particular strategies with regard to their own reproduction and permanence. We must evaluate their ability to make production-related decisions and to participate actively in all processes (Wolf, 1971). These family units possessed a knowledge of the farming ecosystems which allowed them to adapt to adverse conditions and to pressures from outside their communities. This was the foundation of the multifunctionality3 of these agricultural holdings, based on criteria of sustainability and efficiency that led them to develop multiple strategies (Rosset, 1999) involving not only the family unit but the entire community; a process which, evidently, was not without internal conflict. Understanding this complexity from the point of view of mid-twentieth century ethnographic practice enables a closer look at the functioning of these local communities and the family units that lay at their heart in the past. It should be the essential starting point from which to consider any research into the communities using the regressive method and over the longue durée. It has been difficult to locate abandoned villages through archaeological surveys in these areas. There is no doubt that they exist, and some have been excavated, such as the sites of Els Altimiris (Sancho i Planas, 3 

About the discussion on multifunctionality see Fernández Mier 2010: 44.

93

The Archaeology of Medieval Villages Currently Inhabited in Europe

2009) in Catalonia, and As Pereiras (Aboal-Fernández, Cobas, 1999) and A Pousada (Blanco et al, 2009), in Galicia. Given the small number of abandoned villages, early investigations in the 1990s (Fernández Mier, 1999, Bolós, 2004) considered the hypothesis of the survival of the early medieval settlement until the present day. This is associated with a landscape whose agricultural organisation is based around the continual reuse of old shapes and morphologies, which take on new meanings and functions over the course of the centuries. We should remember that the survival of inhabited spaces and the reuse of plots of land does not imply immobility in the land’s use, management or ownership. These were undoubtedly shaped by the social formations which comprised family units as well as local communities (Wolf, 1971). The objective, therefore, is to focus attention on the study of settlements that are still occupied today. This should be based on the archaeological record and include an analysis of all their features, from domestic spaces to forest areas, taking into consideration the way in which they operated just before the breaking up of pre-industrial communities. This will lead to an understanding of their origins, as well as of their resilience and their ability to overcome crises at different moments in time, enabling their capacity to survive up to the present day. It was therefore necessary to define a research strategy that would facilitate an understanding, based on the study of these settlements, not only of their origins but also of their internal structure, the management of their production spaces, the strategies and institutions governing the management of their resources, their internal hierarchy, the dynamism or persistence of forms of property and the extent to which they coordinated with other supralocal entities. All these elements allow us to analyse the reasons for their resilience, in which geographical factors and the ways in which they coordinated the management of resources doubtless played a key role (Curtis, 2014). The journey towards implementing the study of inhabited villages has been slow and few working groups have looked at the subject in Spain, for various reasons. The first is the absence of a preventive archaeology in rural areas that considers the heritage value of population centres and the shapes of the landscape, beyond the monuments or unique ethnographic buildings that are included in inventories of heritage. In fact, the absence of systematic archaeological research linked to the land consolidation which continues to take place today, means the loss of an important body of information. Land consolidation involves the reorganisation of the network of pathways, the removal of the different types of plot boundaries, changes to irrigation systems, the disappearance of buildings for agricultural or livestock-related use, the destruction of the micromorphology of the land, changes to the forms of ownership and management of the land, and the loss of biodiversity. It is a complete redesign of the landscape, carried out without prior study of the forms of the land boundaries, archaeological interventions on the elements that will disappear as a result of the consolidation, or anthropological studies of the old forms of land management (Fernández Mier, 2018). The documentation of these elements has been dependent on the personal initiative of specific working groups or the awareness of professional archaeologists. The most significant example of this has been the work carried out within INCIPIT4 [Institute of Heritage Sciences] (Santiago de Compostela) which, in the context of monitoring large-scale public works involving the creation of large trenches (roads, motorways, gas pipelines), has involved archaeological interventions on a variety of agricultural and livestock-related elements from all historical periods. It has paid particular attention to areas that have been terraced, plot boundaries, agricultural fields, livestock-related buildings and irrigation channels, all of which form part of Galicia’s traditional landscape (Criado et al, 2017). But this practice is far from widespread in Spain. A second factor is the difficulty of carrying out works in parts of villages that are not protected by the various heritage laws (either national or regional) since they are not part of the archaeological inventories. This entails carrying out archaeological interventions on privately-owned land where it is difficult to work within the guidelines for archaeological practice in Spain, and which demands a The Instituto de Ciencias del Patrimonio (CSIC) [Spanish National Research Council] is the successor to the old Laboratorio de Patrimonio [Heritage Laboratory], under whose remit this monitoring was carried out. The results appeared in various issues of the journals CAPA and TAPA, published by CSIC (http://www.incipit.csic.es/es/Publicaciones.aspx). 4 

94

Fernández Mier, Fernández Fernández - The Archaeology of Currently Inhabited Villages in Spain

process of negotiation between archaeologists and owners, separate from official communications with administrative bodies. This issue leads us to consider another relevant matter: the need to involve the communities that we are studying in the social construction of knowledge, and not just as mere recipients of the academic output generated through the study. Various strands of work from the perspective of agricultural archaeology have emphasised the need to reformulate the concept of the rural landscape and of the relationship that archaeological practice should have with the local communities that are being researched (Ayán, Gago, 2012; Martín Civantos, Ruiz, Martín, 2017; Alonso, 2017). There must be an involvement with these communities, which have shown resilience, possess extensive knowledge and are now undergoing a profound process of transformation involving various policies (European, regional and national). We believe that our research should deepen the connection between past and present, allowing us to produce alternative discourses that question the dominant narratives on land management emanating from agricultural policies relating to productivity (Alonso et al, 2018), and encouraging the promotion of alternative types of heritage management in practice at a local level (Alonso et al., 2018; Fernández et al., 2015). Three research groups in northern Spain have focused their attention on inhabited villages, looking at different elements of the landscape and with varying degrees of involvement in this type of research: the work undertaken in Asturias by the research group LLABOR, whose results we shall see below; and the research group GIPYPAC5 which, based on a thorough examination of abandoned medieval villages, has begun to undertake works in inhabited areas in Gipuzkoa, with interesting results that have led to the documentation of the Roman origins of some settlements (Narbarte et al, 2018). In addition to these studies, which are clearly oriented towards a holistic study of inhabited villages, some research has been undertaken from the perspective of preventive archaeology. This has considered elements of the agricultural spaces of present-day villages affected by large infrastructures, although the research in these villages has not been carried out in a holistic manner (Criado et al, 2017). The archaeology of inhabited villages in Asturias In the 1990s, a research project was undertaken based on ethnohistory,6 with the intention of understanding the formation of the rural landscape of the Asturian mountains. It combined historical information with an anthropological approach. The studies of some marginalised rural social groups (García Martínez, 1988) and the approach from the perspective of historical geography (García Fernández, 1988, Rodríguez Gutiérrez, 1984, 1989) were the basis of research focusing on the production cycle of the farming community through a study of the landscape. At the same time, a deeper analysis of the toponymy was undertaken (García Arias, 1995), leading to the creation of an important corpus of data. This gave historians access to an extremely valuable record when interpreted alongside the historical and archaeological information. Thus, in the early 1990s, various different strands of work converged on the subject of the rural landscape, from anthropology, historical geography, rural history and linguistics, clearly influenced by French historiography. The historical studies carried out in this context focused on giving historical depth to a landscape that has been shaped by different, multifunctional elements, and been resilient up to the present day (Fernández Mier, 2010). The ethnographic and anthropological approach pointed the research methodology towards a regressive reading of the landscape, going back to the assumptions of the classic work of M. Bloch (1931) on French rural history and of some works that, at the time, reflected on the study of the forms of land boundaries (Verhulst 1995, Chouquer, 1996a and b; 1997) and of all the elements comprising the landscape (Barceló 1988, Guilaine, 1991). The methodology was an interdisciplinary approach that used written documentation, both medieval and early modern. It included the systematic compilation Grupo de investigación en Patrimonio y Paisajes Culturales. Universidad del País Vasco [Research Group on Heritage and Cultural Landscapes, University of the Basque Country] led by Juan Antonio Quirós Castillo. 6  Project led by Professor F. Javier Fernández Conde. 5 

95

The Archaeology of Medieval Villages Currently Inhabited in Europe

of all place names, focusing on minor place names and taking into consideration the fact that all the toponyms are part of a system that relates to the inhabitants’ own concept of their land (Fernández Mier, 2006). It also involved the study of the archaeological information revealed by a surface survey of both archaeological sites and elements associated with production processes (terracing, emerging irrigation systems, methods of closing off plot boundaries, networks of pathways). It placed particular importance on the ethnographic work that has shown us how these villages functioned before they were broken up, from the 1970s onwards (Fernández Mier, 1999; Míguez Mariñas, 1999; Fernández Fernández, 2014; Muñiz López, 2014). The anthropological approach led to an understanding of the systematic nature of the way in which the land was used in the villages: the link immutability between domestic areas and those devoted to production; the centrality of the so-called marginal spaces (pasture and forest) in the communities’ economies; the farming communities’ deep, complex knowledge of the areas that they cultivated; the rationales for the complementarity of uses, enabling the conditions within the different ecological niches to the exploited; the forms of governance and the creation of identity-based processes; and the actions of external agents and the communities’ defence mechanisms when faced with these actors from outside their community; in short, the cumulative and social nature of the landscape. All this was accompanied by a belief in the need to cross the formally established borders to address relevant issues, making use of the wide range of resources, methods and techniques from different disciplines. The research methodology: from a surface survey to archaeological and geo-archaeological studies These theoretical premises formed the basis of two research projects that, in the last ten years, have focused archaeological interventions on inhabited villages: Vigaña (Miranda) and Villanueva (Santu Adrianu) and initial works in Ambas (Grau), all in Asturias (Fig. 2). These have been combined, theoretically and methodologically, with the archaeology undertaken in abandoned medieval villages. Thus, the land of medieval villages has been viewed as a unique site on which thorough interventions have been carried out in different places. The result has been the recovery of some astonishing information on historical periods about which little was known at a regional level and over the longue durée, simply through observation and intervention on the landscape with a ‘demonumentalised’ gaze. In addition, there has been a focus on the sustainability of the communities’ ways of using the land in the past and the ability of the rural population to articulate that use (Fernández Mier, 2018; Fernández Fernández, 2014, 2017).

Figure 2. Location map of study cases in Asturias.

96

Fernández Mier, Fernández Fernández - The Archaeology of Currently Inhabited Villages in Spain

The information generated by surface surveys has guided the planning of archaeological interventions, both in population centres and in areas devoted to agricultural and livestock-related activity. Priority has been given to small pits (varying from 1x1 to 3x4 and increasing in line with the data obtained), carrying out stratigraphic readings and collecting all the excavated material (Fernández Mier, Alonso González, 2016; Fernández Fernández, 2014, 2017). In all cases, chemical and micromorphological analyses of the soils, studies of the pollen and radiocarbon dating have been carried out. The investment of the large number of working hours required by test-pits on farmland has led to a reorientation in the most recent interventions towards mechanical geo-archaeological works that permit the evaluation of the site’s potential. A mechanical auger is used, creating balks that are 2 m deep. These have been used in spaces excavated in early interventions, to contrast the information that both intervention methodologies provide, as well as in new projects embarked on as part of more recent research (Fig. 3). Working on inhabited villages has significant implications for the historical information obtained. Data from inhabited areas is extremely fragmentary due to the impossibility of carrying out area excavations. This makes it difficult to interpret and it requires a significant amount of work to produce the archaeological data, since a considerable number of surveys must be carried out to obtain synchronic data. Faced with this issue, diachronic information improves the possibility of studying the land over the longue durée. Surveys on farmland and pasture have led to the documentation of archaeological sites, in some cases under sediment packages that are 3-4 m deep, dating to the early days of the occupation of the land in late prehistory. This shows the continual use of the area under different types of habitat and models of exploitation, at least from the Neolithic period to the present day. In the same way, our understanding of the communities’ relationships with the land, their ability to transform the landscape, and the survival of or changes to agricultural and livestockrelated techniques, are deepened. This information allows us to look in greater depth at the way in which the communities connected with their land.

Figure 3. Example of geoarchaeological survey.

97

The Archaeology of Medieval Villages Currently Inhabited in Europe

Vigaña: a mountain village over the longue durée Our first case study is a village referred to in medieval sources as Vigaña Basel or Vigaña Arceu, with the latter still used today. It is a semi-mountainous village located in the valley of the River Pigüeña in the Cantabrian Mountains, at an altitude of 650 m, occupying half the side of the valley. Its climate has long, cold winters with temperatures below freezing and gales, snow and frost. Its summers are cool and short. Its location allows for the exploitation of the land in the area around the river, at an altitude of 150 m, as well as of the important pasture areas located in the mid-mountain areas, at a height of 1100 m (Fig. 4). Vigaña has a mosaic landscape, and currently has around a dozen houses that are in use throughout the year; a tiny fraction of the 80 houses that existed in the early twentieth century. Only five of the current inhabitants work in the primary sector, in extensive farming and breeding of a native breed of cattle that is suited to the mountainous terrain. This activity centres on the trade in meat, which has increasingly prevailed as the sole form of work since the mid-twentieth century, replacing the traditional multifunctionality of the land, which balanced agricultural production and livestock farming. The work still retains a family feel, although on a part-time basis since it is fitted around other tasks relating to the cattle farming sector. CAP subsidies are fundamental in ensuring their profitability and they are underpinned by support for an activity that sustains biodiversity and encourages the development of work relating to the tourism sector. Although this extensive cattle farming is completely market-oriented, the forms of ownership and use of the land maintain the features of the first half of the twentieth century: a staggered use of the land throughout the year, so as to exploit the conditions within different ecological niches; and the ownership and collective use of a large part of the areas of pasture and forest under a system of privately owned property, acquired by the community in the nineteenth century following confiscations by the liberal governments. Thus the plots of land, landscape and forms of management and ownership of the land have been preserved, having been only partially modified by the reforms of the nineteenth and twentieth centuries. However, the communities have re-centred their economic activity, in several cases through migration, while in others they have moved towards extensive livestock farming, dependent on subsidies. This has led them into agricultural post-production where their activity is more closely linked to the preservation of biodiversity and of a landscape centred around its enjoyment by the urban classes, than with their value as producers. The starting point for the archaeological interventions in Vigaña was a sound understanding of the reality of this village; as well as an analysis of the landscape through the lens of anthropology, toponymy and documentary evidence, and through the plots of land that were established here in the Early Middle Ages (Fernández Mier, 1999). Working on the basis of the territorial boundary of the village as it was described in the twelfth century, and which still exists today as a single site (Fernández Mier, 2018), eight archaeological excavations were planned, involving sixteen interventions that varied in size and focus: - The archaeological excavation of the exterior of the Church of San Pedro de Vigaña (Fig. 4: 10, 11), in spaces not occupied by agricultural units or by what is now the cemetery (Fernández Mier, 2015). - Small surveys inside the village (Fig.4: 9); in this case only a small amount of information was produced due to the superimposition of houses on the site. - Surveys of varying sizes in places used for agricultural activity, both in the vicinity of populated areas and in agricultural areas located further away but closer to archaeological sites from the prehistoric period (Fig. 4: 2,4,5,6,7,8). The size of the surveys varied between 2 x 1 m and 10 x 10 m, depending on the results obtained in smaller interventions (Fernández Mier, Alonso González, 2016). - Intensive archaeological interventions in the Iron Age settlement (Fig. 4: 1, 3) located in the immediate vicinity of the village, with the aim of ascertaining when it was abandoned (González Álvarez et al, 2018). - Archaeological trial trenches of 1 x 1m in the main pasture areas (Fig. 4: 12,13) as a surveying method, given the absence of archaeological material on the surface (Fernández Mier, González Álvarez, 2013). - Area excavation at one end of Vigaña’s boundary, where there is now a small chapel serving the valley communities, built on the site of a medieval cemetery (Fernández Mier et al, 2018). (Fig. 4: 14). - Area excavation of a megalithic structure located at the land’s highest point, at an altitude of 1100 m. (Fig. 4: 15) 98

Figure 4. Territory of Vigaña (Belmonte de Miranda), excavated sites and chronologies obtained: 1. El Castru: fortified hillfort dated from the 6th century BC to the 1st century AD. 2. Las Corvas: Neolithic settlement with crop spaces superimposed until today. 3. Arrichere: agrarian terraces from the nineteenth century. 4. La Granda: agrarian terrace from the 16th century. 5: Las Murias: fields, not dated. 6: L’Eirón: agrarian terraces from the 19th century. 7: La Sienra, agrarian space from the Neolithic to the present. 8: L’Hortal: possible neolithic storage silos. 9: La Escuela: negative structures from medieval period. 10: Late medieval Necrópolis. 11: Medieval Necropolis: occupation from the seventh century AD to the present. 12: Braña d’Estoupiel.lo: traces of activities linked to the megalithic structure of La Chalga. Livestock and artisan activities of modern times. 13. Braña Folgueras: fences related to modern agricultural activities. 14. Hermitage of Linares: settlement occupation from the ancient bronze to the final bronze with reuse as a necropolis from the X.C century and subsequent use as chapel. 15: The Cuernu: megalithic structure.

Fernández Mier, Fernández Fernández - The Archaeology of Currently Inhabited Villages in Spain

99

The Archaeology of Medieval Villages Currently Inhabited in Europe

These interventions have resulted in the processes of occupation and transformation of the landscape of the village being traced, from the earliest uses of the land in the Neolithic era. Farmland such as the site of Las Corvas conceals levels of open habitat from the Neolithic period (between 5000 and 6000 years BP), sealed by later post-depositional processes, in turn bolstered by the irrigation practices associated with corn production from the seventeenth century onwards (Fig. 4: 2; Fig. 5). Other surveys on farmland such as La Sienra or L’Hortal (Fig. 4: 7, 8) show traces of different uses, either agricultural or domestic, also in the Neolithic era. The lower levels of the stratigraphy show that they have been greatly altered by the erosive actions of both natural processes and later agricultural practices. There is evidence of significant occupation during the Bronze Age (between 4500 and 3000 years BP) in the location of Fontevigaña/L.linares over a long period of time, associated with monumental fire-burning structures and terraced areas which are unprecedented in the Cantabrian Bronze Age (Fig. 4: 14; Fig. 6). This complex occupation makes it more difficult to gain an understanding of this period in north eastern Spain, since it leads us to question the itinerant character of the population at that time. This monumental site became an important place that was often visited in later eras, as evidenced by archaeological material from the Roman and Late Antique periods. Later, the site was occupied by a medieval cemetery, from the late tenth- to the early-eleventh century. The archaeological context is associated with a significant set of pottery finds, which we believe relates to the existence of an abandoned medieval village in the area. The archaeological material from the early modern era again indicates that the site was visited until the small hermitage was built, which still exists today. This dates back to the seventeenth century, as do the coins found around the chapel. The site is a good example of the symbolic importance of the area in the Bronze Age for later societies. It reflects the way in which a prehistoric monument was incorporated into daily life and played a key social role, probably related to its original ritual function. This role continued until it became a cemetery in the Middle Ages, and a place of worship from the seventeenth century onwards. The excavation of the Iron Age fortified settlement of El Castru has produced information from the eighth century BCE to the change in era, coinciding with the arrival of the Romans in the north of

Figure 5. Las Corvas: Neolithic settlement. Fireplace within domestic structures.

100

Fernández Mier, Fernández Fernández - The Archaeology of Currently Inhabited Villages in Spain

Spain (Fig. 4: 1, 3). The archaeobiological materials analysed on the site show a complementarity of agricultural and livestock-related activities (González Álvarez et al, 2018). The survey of Las Corvas, located in the immediate vicinity of the settlement, outside of the site’s defensive structure, seems to indicate the location of farmland next to the walls. This would have existed at the same time as the hillfort (Fig. 4: 2). As shown in Table 1, radiocarbon dating allows us to establish a complex sequence of use of a segment of the valley from the Neolithic age until the change in era. This included changes in the places and forms of inhabitation, although the land was not abandoned and movement took place within a very limited radius. Two megalithic monuments are associated with these settlements: one is a tumulustype structure, located 900 m away, and the other is a cave-like monument at an altitude of 1100 m. They are located in the main pasture area but as yet we do not have more precise information about their age. This data on late prehistory shows the significant transformations that the landscape underwent during this period, and the continual occupation of the land from neolithisation onwards, due to the large areas of pasture.

Figure 6. Hermitage of L.linares: bronze age structures.

It is striking that there is very little material there that can be linked to the late Roman imperial period. Some minuscule fragments of terra sigillata hispanica, which are very worn and come from the excavation of two areas of the cemetery (the surroundings of the Church of San Pedru de Vigaña and L.linares; Fig. 4: 10,11,14), indicate the presence of some sort of settlement in the area. However, this pottery is not sufficiently significant to establish the existence of a Roman imperial occupation. The proximity of the gold mines in the neighbouring valley of the La Uxa river, exploitation of which was confined to the first and second centuries, could explain the absence of a clear Roman imperial occupation in the area around Vigaña (Fernández Mier, 1999, Villa Valdés 2007). The excavations of the medieval cemeteries offer limited and fragmentary information on the Roman Imperial era, but a complete sequence of information from the Early Middle Ages to the present day. The oldest burials in the cemetery of San Pedru de Vigaña (one of which contains significant grave goods, which is very unusual in this area) are suggestive of the first half of the seventh century AD (Fig. 4: 10, 11. Fig. 7). A series of negative stratigraphic units (post-holes of different sizes located in the two parts of the cemetery that were excavated) allow us to hypothesise on the existence of post-hole structures that saw the end of this early use of the area as a burial place, although we are unable to identify the nature of its subsequent use. The area once again became a cemetery in the twelfth century, with a second layer of tombs being documented, lying from east to west and slightly turned away from the axis of the church that exists there today. They may be associated with an earlier religious building that would have 101

The Archaeology of Medieval Villages Currently Inhabited in Europe

Table 1. Vigaña radiocarbon data.

been smaller than the fifteenth century church that stands there today. They are well-made tombs with anthropomorphous headstones and, in some cases, large slabs that have been well-worked, and the coffins and lids sealed with red clay. Practically all the graves of this phase have been reused. The third phase of interments was superimposed over these ones, also lying from west to east, parallel to the church. The tombs have slabs with medium-sized, straight headstones which have been barely worked. They are mainly located around the Church’s chancel. In the most recent phase, the burials are characterised by simple graves without a covering, with the different graves often overlapping and reinterments being common. They lie from west to east, parallel with the church that stands there today. All the burial phases have been cut into by various works associated with church repairs. Finally, the cemetery that exists today is located to the south of the church. A small vegetable plot now exists on part of the cemetery to the north. A 1.5 m stone wall has been built, attached to the rock and filled with earth, creating a small plot for growing crops. There are several similar vegetable plots in the village. Although we do not have sufficient 102

Fernández Mier, Fernández Fernández - The Archaeology of Currently Inhabited Villages in Spain

historical data that would enable us to date the vegetable plot, some oral traditions indicate that it is unlikely to have existed before the nineteenth century. Next to the vegetable plot is a granary, which also sits over part of the cemetery. Information has also been generated from a diachronic perspective by some surveys carried out in farmland close to dwelling areas, such as in the site known as La Sienra (Fig. 4: 7), with a simple stratigraphic sequence. Under the humic layer, several units associated with medieval, early modern and contemporary cultivation layers were identified, based on the scant quantities of ceramics found. These units superimpose another layer with an abundance of carbons and ceramic material dated to the early medieval era, covering other layers that suggest prehistoric origins. The pollen record in this early medieval layer suggests a highly deforested landscape. The arboreal stratum comprises chestnut, savin juniper, hazel, yew, birch, black poplar, poplar and pine (which may have come from further afield), with chestnut being the most abundant. The Figure 7. San Pedru de Vigaña. Early medieval post holes under Late herbaceous vegetation is dominated medieval necropolis. by grasses, indicating the presence of pastures. There are significant quantities of other herbaceous plants that grow in humid conditions, as well as ferns, indicating the existence of waterlogged areas or areas with high levels of moisture in the soil. The anthropization of the area is evident throughout the entire sequence, based on the scant presence of an arboreal stratum as well as on the strong presence in the palynological record of ruderal plants that are typical of humanised environments. On the other hand, there is no evidence that would suggest an agricultural use of the area, indicating a livestock-related specialisation from the early Middle Ages onwards. Chemical analyses of the soils show levels of nitrogen, phosphorous and calcium, indicating a progressive enriching of organic material in this period, while the acidity of the soil was increasing. The only nutrient to have diminished in quantity is potassium, associated with lower levels of ash. Higher levels of this element are found in the stratigraphic unit dating from the late Neolithic era, together with a lower incidence of phosphorous, nitrogen and calcium. This data may indicate the use of slash-and-burn techniques as a method of fertilising farmland in shifting cultivation in this prehistoric phase, distinguishing this agricultural phase from medieval and contemporary farming practices based on organic fertilisation by livestock (Fernandez Mier et al, 2014). Other surveys on agricultural land (Fig. 4: 4 and 6) indicate a dramatic restructuring from the sixteenth century onwards, with the creation of terraced spaces, and widespread modification of the landscape in the nineteenth century. This involved the creation of small terraces with stone walls and the increasing 103

The Archaeology of Medieval Villages Currently Inhabited in Europe

use of drystone walls to section off plots (Fig. 4: 2,3). This process dramatically transformed the landscape and was the result of the influence of the liberal policies of the nineteenth century (Fernández Mier, Alonso González, 2016). In conclusion, various different occupations since the seventh century have been documented beneath the present-day village of Vigaña. Burial places have alternated with structures for domestic use around the church that stands there today. They existed alongside spaces used for intensive cattle farming in one of the village’s main production areas. In the twelfth century, a period for which there is abundant documentary information about the village, including a definition of its boundaries, a new stage of burials has been documented. These were related to the building of a church and reflect an important restructuring of the land. This coincided with the consolidation of the nearby monastery of Lapedo and the disappearance of some centres such as Fontevigaña, as well as the consolidation of others around a church, such as Vigaña (Fernández Mier, 2019). The survival of the location of the prehistoric sites in the collective memory of the village is remarkable, whether through legend or references to buried treasure (as in La Chalga and El Castru), or the reoccupation and later Christianisation of the Bronze Age site in Fontevigaña/ L.linares. The prehistoric monuments played an active role in the land and the life of the community, reflecting the mythical world view of space and time which was reinterpreted over the millennia, and appropriated by Christianity from the early days of the Middle Ages. Villanueva The second case study is Villanueva de Santu Adrianu, an extremely small population centre (currently approximately 50 inhabitants) located in the valley on the banks of the River Trubia, in a semi-mountainous area in central Asturias. The area has an oceanic climate as a result of its proximity to the sea, with abundant rainfall all year round and an extremely rugged landscape with steep slopes and inclines. Villanueva sits on the alluvial plain at the bottom of the valley and is made up of eight areas (S. Romanu, El Puente/Villanueva, Traslaponte, El Carme, L’Arcellada, El Sabil, La Villa Fondera and Las Xanas) that give the village its characteristic polynuclear structure (Fig. 8). The limits of the village and its farmland are also those of the parish (San Romano, approximately 6 km2) and currently there are no other significant dwelling areas. The landscape of today can also be characterised as mosaic. Its traditional arrangement of collective and semi-collective land uses continued until the twentieth century. The analysis of the medieval documentary evidence was essential in assigning a chronology to some of the village’s areas and agricultural spaces. The first documentary reference to the village is found in a bequest by King Alfonso III of Asturias to a monastery located in nearby Tuñón, approximately 3 km from Villanueva, dating from the ninth century (891 AD) but copied in the twelfth century.7 The nature of this bequest is clear: it was a monastery founded on a legacy of the king and his wife, Jimena, to which they donated several assets, liturgical objects, serfs, fortifications and a series of areas of land and spaces to be used for agriculture and livestock. This centre of power was the seed of a new area called Santu Adrianu (Fernández Conde and Pedregal Montes, 1998). As part of the regal endowment, a pre-Romanesque temple was built and devoted to the king and his wife, representing the status of the monarchy of the kingdom of Asturias. It was a foundation intended to serve as a governing centre of the large legacy left as part of its founding charter. The list of the areas is clear and allows us to take a look at the early settlement during these centuries. Of the thirteen areas that could be identified in the parts of the document considered to be authentic, eleven continued to be inhabited, either as villages or hamlets, until the beginning of the second half of the twentieth century. There are doubts over the location of the two areas that were not continually inhabited, and they may have been villages that were in fact inhabited. The reference to a nearby town It is an interpolated document. Documentary studies of both indicate the veracity of some parts and show that the original documents included a bequest (Fernández Conde and Pedregal Montes, 1995-1996, 1998). In any case, for our study the aim is to be able to identify the degree of continuity between the medieval settlement and the current one, in order to consider the historical issues. 7 

104

Fernández Mier, Fernández Fernández - The Archaeology of Currently Inhabited Villages in Spain

Figure 8. Village of Villanueva de Santo Adriano, mentioned in IXct. documents. View from the north.

close to the River Trubia, called San Romano (a name by which it continued to be known until at least the high Middle Ages), appears in the original part of the document. The first reference to a ‘villa nueva’ [‘new town’] appeared in the fourteenth century, in the parochial census carried out by the bishop of Oviedo, which would later replace San Romano. The ‘villa vieja’ [‘old town’] (San Romano) was absorbed into the village as another neighbourhood, and still bears its old name. The archaeological interventions — both the surface surveys of the surrounding agricultural spaces and the various archaeological surveys — have focused on this population centre (Fig. 9). A summary of these results has already been published (Fernández Fernández, 2014; Fernández Fernández et al, 2018; Fernández Mier et al, 2014). Only a brief outline of the information will be given here. Documentary studies, the morphological analysis of the plots of land, toponymy and archaeological surveys (Table 2) led us to establish an initial hypothesis: that the area of San Romano is the oldest, corroborated by the presence of medieval pottery in its surroundings. This pottery is less abundant around the new town, where early modern and contemporary ceramics are more commonly found. Based on this information, new surveys were planned for plots of land around the area of San Romano (Fig. 9). Below is a summary of the results obtained (Tables 2 and 3). Table 2. Results of the surface archaeological survey. Toponymyagricultural space

Total number of fragments collected

Medieval pottery and percentage of the total

Early modern and contemporary pottery

Güerta de S. Romano

17

yes (47%)

yes

La Veiga

30

yes (3%)

yes

Buyera

11

yes (100%)

no

Güertu l’Arcellada

14

no

yes

Cericéu

0

no

no

105

The Archaeology of Medieval Villages Currently Inhabited in Europe

Figure 9. Prospected areas and surveys with codes.

The excavation of two areas of early medieval farmland show early activity relating to the presence of a farming community in the area, whose origins we can place between the eighth and tenth centuries as a result of radiocarbon dating from two surveys: MNS (SU 7) and PMR (SU 9). This has been ascertained using chemical analysis of the soil, a study of the material culture, and pollen analysis, all of which indicate the properties of cultivated soil (Fernández Fernández, 2014). These studies have confirmed the presence of cerealia pollen in a reforested landscape, in relation to the underlying archaeological layers (Roman era. PMR, table 23). We do not have any more information on this early village other than the presence of ancient historical negative structures, later filled in during the Early Middle Ages (CDR, SU 11). This filling, which appears to correspond to a period of intensified occupation, is called dark earth (PMR, 10th century; CDR, 10th-11th centuries, table 3). It contains large quantities of remains of fauna, pottery and iron (mainly nails) in a matrix that has been darkened by abundant remains of carbonised plant material. This type of medieval soil is sometimes known as dark earth-like, to differentiate it from the dark earth of urban environments, where it was first defined (Macphail and Linderholm, 2004). Soil with these properties has been identified and called dark earth in other rural European sites (Loveluck, 2004). Dark earth has sometimes been understood to be the result of a combination of different activities on a single space over a long period of time, for example dwellings and agriculture or vice versa (Courty, Golberg and Macphail, 1989). Pollen data indicates a significant change in the use of this space, with the disappearance of cereal crops and the existence of taxons associated with populated areas. The material culture is composed of remains of mainly black and grey pottery, suggestive of early medieval periods (Requejo Pagés, 2003-2004). The fauna consumed consisted of goat, pig and cow in similar proportions. In summary, the data suggests the presence of a village with a diverse economy following a similar pattern to that described in other abandoned villages of the same period in north eastern Spain (Vigil-Escalera, 2003; Quirós Castillo and Vigil-Escalera, 2003).

106

Fernández Mier, Fernández Fernández - The Archaeology of Currently Inhabited Villages in Spain

Figure 10. Different structures excavated (S. Romano). 1. Paved floor of medieval origin. A fireplace above was dated between XIIIth-XIVth centuries (CDR, UE7). 2. Post holes of the XIIth ct. (MUR, EU 12). 3. Detail of an excavated agrarian terrace dated from XV-XVIIth ct. 4. Post hole and structures filled up by sediments dated between the X-XI centuries (CDR, EU 11 -post hole excavated in EU 13).

The village’s next evolutionary phase relates to a set of domestic structures dating from around the twelfth century, identified in the MUR survey (slightly further away from the current population centre, Fig 10). These relate to a dwelling space built from post-holes in which several domestic areas have been identified, although it has not been completely excavated (Fig 10, 2). The material culture is composed of similar elements to those found in previous phases, without notable changes to the ceramic tradition. These are fragments of cooking items such as pots, pans, jugs, etc, with incised decorations, made using a potter’s wheel or by hand, generally fired either in reduction or through double stage firing (Fernández Fernández et al, 2018). Another interesting piece of information is that the post-holes in this domestic unit were excavated from Iron Age earth, indicating, in the same way as the findings of the PMR and CDR surveys, a lack of precise archaeological information from between the early Iron Age and the eighth century AD.

107

The Archaeology of Medieval Villages Currently Inhabited in Europe

What is interesting about the presence of this structure is that it is located on the lower levels of one of the historic open-field areas of the village, traditionally dedicated to the extensive farming of corn, beans and potatoes. Its excavation has allowed us to view the formative sequence of this type of agricultural unit, present in many villages in Asturias (where they are known by different names such erías, veigas and cortinales) and in a large number of regions in Europe (such as open fields in England, zelgensystem in Germany and trest in Brittany, France), from an archaeological perspective. Historical and archaeological research has suggested that open field systems were developed throughout the Middle Ages (Zadora-Rio, 1991; Oosthuizen, 2010, Williamson, 2012). However, there is still no precise archaeological information on this matter in Asturias. Our data indicates that, at least until the fourteenth century, these spaces still showed signs of multifunctionality, with agricultural and habitational uses both forming part of the archaeological sequence, just as observed in the previous period (the ninth to the eleventh century; Fernández Fernández, 2014a), proving that the duration of these phenomena was longer. Taphonomic data indicates that the soils that filled the twelfth century structures were clearly agricultural and probably part of an open field system. From the fourteenth century onwards, the area continued to be used for agriculture, apart from a period of two centuries during which time it was inundated with sterile deposits resulting from a torrential flood (Fernández and Moshenska, 2017). The next phase of inhabitation to be identified is suggestive of the thirteenth to the fourteenth century, with a domestic space built on the dark earth of the tenth to the eleventh century (CDR). It is a levelled kitchen floor with a hearth. The associated fauna indicates a greater presence of pigs than of goats and cows, suggesting a degree of agricultural specialisation and the loss of the faming community’s autonomy in the early Middle Ages, when beef consumption appeared to become more common (Fernández Fernández, 2014b). Radiocarbon dating points to the thirteenth or early fourteenth century. These dates coincide with the point at which the post-holes structures ceased to exist, dated to around the twelfth century by the MUR survey. This new dwelling structure was linked to the current village. It was around this time that a clear and permanent division between the dwelling and productive spaces occurred, which still exists today but was not seen in earlier periods. Another piece of data supporting this view is the fact that several tiles were found in the destruction layer of the late medieval village (SU 3 MUR), indicating the frequent use of coverings of this type, rather than those of perishable materials. The latter would have deteriorated, while houses were made of stone and were more permanent. In conclusion, a greater degree of mobility and signs of multifunctionality could be seen in the phases prior to the thirteenth and fourteenth centuries, when dwelling areas appeared to become permanently separate from working areas, with the formation of an open field system and the establishment of the village. Why did this happen at this time? Why did these changes take place? We have evidence of this through twelfth-century written documents from the Episcopal estate (both proprietary and jurisdictional) in the area. It was at this point that the farming community may have lost control of the communal pastures and ownership of the livestock, giving rise to a greater level of manorial pressure. This resulted in agricultural specialisation by the village’s farming communities, obliging them to intensify farming and to create spaces for specialised agriculture such as open fields and, indirectly, forcing them to make their dwelling places permanent. The soil that covered over the twelfth century structures in MUR (SU 504) shows a process of intensified agrarian activity and a huge increase in domestic residues in the upper part (just before the point at which the area was flooded, SU 4). The impression gained from this data is that pressure was growing on agricultural spaces which, unable to regenerate naturally, required the use of fertilisers. However, since cattle manure does not appear to have been an available resource, these communities were obliged to use domestic waste as fertiliser. All this points towards an intensification in production, the consolidation of permanent, regulated farmland and a degree of reduction in livestock. Interventions on different types of farmland such as agricultural terraces have allowed us to gain a view of the combination of changes that occurred to the landscape over this historical period. This began with the construction of agricultural terraces between the tenth and the eleventh centuries on earlier 108

Fernández Mier, Fernández Fernández - The Archaeology of Currently Inhabited Villages in Spain

Table 3. Stratigraphic sequences noted in the soil and chronological correlation between SUs. The century is given where there is absolute dating. Chronology

Test pit PMR

Test pit CDR

Test pit MUR

Test pit MNS

Late Modern

SU 1 (agrarian soil)

SU 1 (agrarian soil)

SU 1 (agrarian soil)

SU 1 (agrarian soil)

Early Modern

SU 2-3-4-5 (agrarian soil and structures)

SU 2-3 (agrarian soil, 16th c. AD)

SU 1 (agrarian soil)

SU 4 (agrarian soil)

Central-Late Medieval

SU 6-7 (flood deposits)

SU 4-5 (flood deposits) SU 7 (domestic structure, 13th-14th c. AD)

SU 2-3 (flood deposits, after 13th-14th c. AD) SU 5 and 4 (agrarian soil, after 13th-14th c. AD)

SU 6 (stony soil, 13th14th c. AD)

Central Middle Ages

SU 12 and post holes (domestic structure, 11th-12th c. AD)

Early Medieval, 10th-11th c.

SU8 (dark earth, 10th c. AD)

SU10 (dark earth, 10th-11th c. AD).

SU 5 (agrarian terrace, 10th-11th c. AD)

Early Medieval 8th-9th c.

SU 9 (agrarian soil, before 10th c. AD).

SU 11, post hole, before 10th-11th c. AD).

SU 7 (agrarian soil, 8th c. AD)

SU 10 (pasture soil?, roman pottery)

SU 13 (pasture soil?, roman pottery)

Late Antiquity Roman period Iron Age

SU 50, pebbly soil structure (Iron Age, 3rd-4th c. BC)

farmland (eighth century) in an area linked to the Church (MNS). Another terrace was built over the medieval one in the early modern era, which continued to be used until it was last widened between the nineteenth and twentieth centuries. We can see how the church played a fundamental role in the organisation of the settlements in these flat areas at the base of the valleys, with the dating of the oldest medieval one having been identified through this survey (ca. eighth century). The terracing (SU 5, MNS) coincides with the point at which the population began to grow (between the tenth and the eleventh centuries) and predates the village of the thirteenth to the fourteenth century. Through other interventions, we have also been able to discern a moment of significant reorganisation in the land, in around the sixteenth century. This was linked to growing population levels and the emergence of new crops from the Americas, which obliged farming communities to restructure their working spaces and systems of agricultural rotation. A total of four agricultural terraces were excavated, two of which were created in the early modern age and the remaining two built in around the sixteenth century (Fernández Fernández, 2014). Discussion The archaeological excavation of these two villages has allowed a close look at the genealogy of the landscape over the longue durée (Table 5). Interventions on a wide variety of places have provided us with information on the relationship that prehistoric communities had with the environment and their ability to monumentalise and conceptualise the land. Sometimes these prehistoric settlements occupy sites close to the location of villages that exist today and are concealed below large areas of sedimentation resulting from medieval and early modern farming. Archaeological interventions in these farming areas with complex depositional processes — encouraged by agricultural activity — have proved to be the ideal methodology for the documentation of periods about which little is known at a regional level. This has shed light on previously unknown ways in which the land was occupied in the late prehistory of the Cantabrian region, showing a great degree of complexity and variety, but also a significant level of resilience. These were communities who moved around a defined area over millennia, transforming the 109

The Archaeology of Medieval Villages Currently Inhabited in Europe

Table 4. Villanueva radiocarbon data.

environment and building monuments that would go on to play an active role in the communities of subsequent eras. These communities integrated them as key elements that formed part of their world view of the land with which they identified (Díaz-Guardamino et al, 2015), as we have seen in Vigaña. 110

Fernández Mier, Fernández Fernández - The Archaeology of Currently Inhabited Villages in Spain

Understanding the occupation of the land in the Roman period is more complex. In both case studies, fragmentary and diffuse information has been identified, indicating the existence of some type of Roman occupation. The current data makes it impossible to define the type of settlement and its date, and we can only confirm that it relates directly to the layers of medieval occupation. It is not until the seventh and eighth centuries that we are able to document the existence of dwelling places (represented by post-holes and domestic remains), farmland and burial places. These elements represent an important degree of mobility which, nonetheless, appeared to be restricted to a specific area. This was associated with a territoriality that can be glimpsed in the documentation of subsequent centuries. Despite the fragmentary information, data from the cemeteries allows us to suggest the hypothesis of a degree of hierarchical social organisation, with one of the seventh-century burials in Vigaña appearing to indicate the pre-eminence of some individuals at the heart of these communities. These individuals were able to connect with powerful groups acting at a supralocal level, giving them access to a degree of prestige. Studies of farmland, although they provide only fragmentary information, begin to show communities displaying a certain degree of specialisation relating to the geographical conditions of their environment. Although in Villanueva there is evidence of mixed economic activity based on a complementarity between agriculture and livestock farming, in Vigaña the data obtained to date suggests a specialisation in intensive livestock farming. From the tenth century onwards, and at different times in the two sites, important transformations in the landscape are seen, involving the construction of terraced areas, the reorganisation of burial areas, references in the documentation to precise territorial delimitations of village boundaries, and the Table 5. Vigaña and Villanueva archaeological results comparison. Chronology

Vigaña

Villanueva

Neolithic/Bronze age

Domestic areas with fireplaces

Iron Age

Hillfort: VIII-I. B.C.

Indeterminate structures in fluvial context, outside of hillforts

Roman-Late Antiquity (I-VI centuries)

Limited information indicating the presence of population. It is not possible to describe the type of occupation.

Limited information indicating the presence of population. It is not possible to describe the type of occupation.

Early Middle Ages (VII-IX centuries)

Necropolis. Evidences of domestic uses in secondary position. Agricultural soils in the surroundings of the village

Post-holes and dark earths. Agricultural soils in the surroundings of the village.

X-XI centuries

Evidences of domestic uses in secondary position Agricultural soils in the surroundings of the village

Dark earths. Settlement intensification, construction of agrarian terraces associated with the territorial reorganization driven by a monastic foundation.

XII century

Territorial reorganization; desserted village in the new territorial boundaries and construction of agrarian spaces associated with a monastic foundation.

Timber building

XIII-XIV centuries

Domestic and agrarian areas disengagement and stabilisation process. Open-fields creation.

XVI-XVII centuries

Construction of new agrarian terraces and agrarian spaces

Construction of new agrarian terraces and agrarian spaces

XIX century

Enclosure of some village common lands and pasturelands. Emergence of small intensive agrarian areas inside the villages

Enclosure of some village common lands and pasturelands. Emergence of small intensive agrarian areas inside the villages

111

The Archaeology of Medieval Villages Currently Inhabited in Europe

abandonment of some settlements in favour of the consolidation of others. This was all linked to the presence of ecclesiastical entities that took on a central role in the organisation of the land, influencing economic practices and establishing themselves as one of the defining elements in the identities of village communities. The founding of the monastery of Santu Adrianu de Tuñón in the tenth century, just outside Villanueva, and the consolidation of the Monastery of Lapedo in the twelfth century in the Pigüeña valley, eight kilometres from Vigaña, with the likely construction of a church in the village at the same time, show the process of territorial redefinition associated with ecclesiastical institutions. The documentation of domestic structures up to the thirteenth century below the agricultural layers of the village of Villanueva is of particular interest. It indicates that it was not until that point that the locations of dwelling areas and farmland were consolidated in a way that continued throughout the preindustrial period. This sheds light on the point at which open field systems and dwelling areas (which up until then seem to have been characterised by a great deal of mobility) were established. The interpretive approach described here shows a sequence similar to that documented in other areas that have been researched in north western Spain, based on abandoned medieval villages, such as in the Basque Country (Quirós Castillo, 2012, 2018). It consolidates the information on farming communities in the Cantabrian region as well as showing, within some more general features, the distinctive characteristics of each territory. Conclusion The archaeology of villages that exist today has proved to be a powerful tool for uncovering complex diachronic information. It has enabled us to consider new theoretical and methodological questions to improve our understanding of farming communities over the longue durée. Firstly, it has qualitatively and quantitatively improved our ability to obtain information on periods about which there was little information. This allows us to fill in historical gaps, as well as to question interpretations which argue that there was mobility on land without stable population centres or societies characterised by primitive economies. It was simply a question of asking different questions of the land and considering the study of the landscape, taking into account the vision of the landscape held by the actors who inhabit the land today. The repeated occupation of the same places over the centuries leads to a deeper understanding of the relationships between the communities and the land, the way in which they adapted to their environment and the internal dynamics that led to their redefinition, and their ability to respond to external influences. These exist, but we must interpret them based on the communities’ ability to integrate them into their own social formations and to establish strategies to enable their survival. On the other hand, systematic surveys of the land and intensive interventions begin to shed light on abandoned medieval villages about which very little is known in most of north western Spain. They provide limited archaeological information on the surface, but continue to play an active role within the imagination of the communities, whether through oral traditions or through the enshrinement of their location. This shows the need to deepen the definition of research strategies which allow the study of populations from both types of settlement (those which have been abandoned and those which are currently inhabited) and that allow us to reflect on the dynamics that have led to the survival of some settlements when others have disappeared, and the different territorial responses. Finally, we must not forget the real protagonists of this research: today’s rural communities. These communities have shaped the landscape, preserved traditional forms of production and land management, and integrated old sites into their world view. We must ensure that we work actively with them, involving them in the research of their own historical reality, and allowing them to participate in the scientific knowledge that has been accumulated and communicated via different types of activities and approaches. The vehicle through which the research and its dissemination have been articulated is a community organisation called La Ponte-Ecomuséu, involving both local academics and members of the 112

Fernández Mier, Fernández Fernández - The Archaeology of Currently Inhabited Villages in Spain

community with no connections to archaeology. The Ecomuseum organises different types of activities such as discussions, workshops for children, projects with local schools, and educational or interpretive tours, all of which lead to an understanding of the historical evolution of the village and its complexity. These activities are always well-received by the local community and help to involve it in the necessarily reflective process that must accompany academic and scientific practice. Thus, bridges are built between expert and academic knowledge and its practices, and the local population, who often feel excluded or isolated from this type of project (Alonso et al, 2018). References Aboal Fernández, R, Parcero Oubiña, C. 1999. Primeros resultados de documentación arqueológica del yacimiento romano medieval de As Pereiras (Amoeiuro, Ourense). Gallaecia 18: 301-312. Alonso González, P. 2017. El Antipatrimonio. Fetichismo y dominación en Maragatería. Madrid, CSIC. Alonso González, P., Macías Vázquez, A., Fernández Fernández, J. 2016. Governance Structures for the Heritage Commons: La Ponte-Ecomuséu-Ecomuseum of Santo Adriano, Spain”. In Gould, Peter, Pyburn, K. Anne (Eds.), Collision or Collaboration. Archaeology Encounters Economic Development: 153 - 170. Switzerland, Springer. Alonso González, P., Fernández Mier, M. and Fernández Fernández, J. 2018. La ambivalencia del paisaje. De la genealogía a la arqueología agraria. Munibe Anthropology Archaeology 69: 283-296. Ariño Gil, E. 2013. El hábitat rural en la península Ibérica entre finales del siglo IV y principios del VIII: un ensayo interpretativo. Antiquité Tardive 21: 93-123. Ayán Vila, X.M., Gago Mariño, M. 2012. Herdeiros pola forza: patrimonio cultural, poder e sociedade na Galicia do século XXI. A Coruña, 2.0 Editora. Azkarate Garai-Olaun, A. 1999. Necrópolis tardoantigua de Aldaieta (Nanclares de Gamboa, Álava). Vitoria, Diputación Foral of Álava. Azkarate Garai-Olaun, A., Solaun Bustinza, J.L. 2013. Arqueología e historia de una cidad: los orígenes de VitoriaGasteiz. Bilbao, University of the Basque Country. Barcelo, M. 1988. Arqueología Medieval: en las afueras del medievalismo. Barcelona, Crítica. Ballesteros Arias, P., Blanco Rotea, R. 2009. Aldeas y espacios agrarios altomedievales en Galicia. In J. A. Quirós Castillo (ed.), The arqueology of early medieval villages in Europe. Documentos de Arqueología e Historia 1: 115-136. Bilbao, University of the Basque Country. Blanco Rotea, R., Prieto- Martínez, P., Ballesteros Arias, P., López González, F. 2009. El despoblado de A Pousada: La formación de una aldea rural en la Alta Edad Media. Tapa 41: 111-120 Bloch, M. 1931. Les caractères originaux de l’histoire rurale française. París, Société d’édition «Les Belles Lettres». Bolós, J. 2004. Els orígens medievals del paisatge catalá. Barcelona, Abadía de Monserrat. Courty, M.A.; Goldberg, P. y Macphail,R.I. 1989. Soils and micromorphologyin archaeology. Cambridge, Cambridge University Press. Crescente Maseda, R., Álvarez López, C.J. 1999. Una revisión de la concentración parcelaria en Europa. Estudios Agrosociales y pesqueros 137: 35-47. Criado Boado, F., Parcero Oubiña, C., Otero Vilariño, C. and Cabrejas, E. (eds). 2017. Atlas arqueolóxico da paisaxe galega. Santiago, Xerais. Curtis, D.R. 2014. Coping with Crisis. The resilience and vulnerability of Pre-Industrial Settlements. Dorchester, Routledge. Chouquer, G. 1996a. Les formes du paysage, tome 1- Études sur les parcellaires. París, Errance. Chouquer, G., 1996b. Les formes du paysage, tome 2- Archéologie des parcellaires. París, Errance Chouquer, G., 1997. Les formes du paysage, tome 3-L’Analyse des systems spatiaux, París, Errance. Díaz-Guardamino, M., García Sanjuán, L., Wheatley, D. 2015. The lives of Prehistoric Monuments. Iron Ages, Roman and Medieval Europe. Oxford, Oxford University Press. Escalona Monge, J. 2009. The Early Castilian Peasantry: An Archaeological Turn? Journal of Medieval Iberian Sudies 1 (2): 119-145. 113

The Archaeology of Medieval Villages Currently Inhabited in Europe

Fernández Fernández, J. 2014, Estudios multiescalares sobre el Valle del Trubia (Asturias, España), Oviedo, Universidad de Oviedo. Fernández Fernández, J. 2017. Arqueología de una aldea medieval y su espacio agrario: Villanueva de Santo Adriano (Asturias). Historia Agraria 72: 79-107. Fernández Fernández, J., Alonso González, P. and y Navajas Corral, O. 2015. La Ponte-Ecomuséu: una herramienta de desarrollo rural basada en la socialización del patrimonio cultural. La Descommunal 1 (2): 117-130. Fernández Mier, M. 1999. Génesis del territorio en la Edad Media: arqueología del paisaje y evolución histórica en la montaña asturiana: el valle del río Pigüeña. Oviedo: University of Oviedo. Fernández Mier, M. 2006. La toponimia como fuente para la historia rural: la territorialidad de la aldea feudal. Territorio, Sociedad y Poder 1: 35-52. Fernández Mier, M. 2010. Campos de cultivo en la Cordillera Cantábrica. La agricultura en zonas de montaña. In H. Kirchner (ed), Por una arqueología agraria. Perspectivas de investigación sobre los espacios de cultivo en las sociedades medievales hispánicas: 41-59. Oxford, Archaeopress. Fernández Mier, M. 2015. La articulación social de la Alta Edad Media asturiana. In J.A. Quirós Castillo and S. Castellanos García (eds), Identidad y etnicidad en Hispania. Propuestas teóricas y cultura material entre los siglos V al VIII. DAM 8: 181-200. Vitoria: University of the Basque Country. Fernández Mier, M. 2018a. El paisaje rural medieval: de los dominios monásticos a la arqueología agraria. In D. Soto Fernández and J.M. Lana Berasain (eds), Del pasado al futuro como problema. La historia agraria española en el siglo XXI: 1-32. Zaragoza, SEHA. Fernández Mier, M. 2018b. De la Arqueología del Paisaje a la Arqueología Agraria. In J.A. Quirós Castillo, Treinta años de Arqueología medieval en España: 225-270. Oxford, Archaeopress. Fernández Mier, M. 2019. Peasant Communities and Landscape design in the North West of the Iberian. In J. Escalona, Policy and neighbourhood, in press. Fernández Mier, M., Fernández Fernández, J., Alonso González, P., López Saez, J.A., Pérez Díaz, S. and Hernández Beloqui, B. 2014. The investigation of currently inhabited villages of medieval origin: Agrarian archaeology in Asturias (Spain). Quaternary International 346: 41-55. Fernández Mier, M. and Alonso González, P. 2016. Medieval north-west Spain: What can agrarian archaeology tell us about living rural landscapes? In J. Klapste, (ed), Agrarian Technology in the Medieval Landscape, Ruralia X: 291-308. Turnhout, Brepols. Fernández Mier, M. and González Álvarez, D. 2013. Más allá de la aldea: estudios diacrónicos del paisaje en el entorno de Vigaña (Belmonte de Miranda). In Excavaciones Arqueológicas en Asturias 7 (20172012): 353-365. Oviedo, Principado de Asturias. Fernández Mier, M., González Álvarez, D., Martínez Gallardo, C., López Gómez, P., Martínez Barrios, C. 2018. Nes llendes de l’aldea: paisaxe y territorio en Vigaña (Miranda). In Excavaciones Arqueológicas en Asturias 8 (2013-2017): 359-370. Oviedo, Principado de Asturias. Fernández Conde, F.J. and Pedregal Montes, M.A. 1998. Evolución histórica del territoriode Santo Adriano y génesis delpoblamiento medieval. Studia Historica 16: 129-172. García Arias, X. Ll. 1995. Toponimia: teoría y actuación. Oviedo, Academia de la Llíngua Asturiana. García Fernández, J. 1988. Sociedad y organización tradicional del espacio en Asturias. Gijón, Silverio Cañada Editor. García Martínez, A. 1988. Los vaqueiros de alzada de Asturias. Oviedo, RIDEA. García Porras, A. 2018. Treinta años de una nueva arqueología de al-Andalus. in J.A. Quirós Castillo, Treinta años de Arqueología medieval en España: 95-122. Oxford, Archaeopress. González Álvarez, D., Marín Suárez, C., Farci, C., López Gómez, P., López Sáez, J.A., Martínez Barrios, C., Martinón-Torres, M., Menéndez Blanco, A., Moreno-García, M., Núñez de la Fuente, S., PeñaChocarro, L., Pérez-Jordá, G., Rodríguez Fernández, J., Tejerizo García, C., Fernández Mier, M. 2018. El Castru (Vigaña, Balmonte de Miranda, Asturias): un pequeño poblado fortificado de las montañas occidentales cantábricas durante la Edad del Hierro. Munibe Anthropology Archaeology 69: 211-237. González-Ruibal, A., Alonso González, P. and Criado-Boado, F. 2017. Against reactionary populism: Towards a new public archaeology. Antiquity 92: 507-515. Guilaine, J. 1991. Pour une archéologie agraire. París, A. Colin. 114

Fernández Mier, Fernández Fernández - The Archaeology of Currently Inhabited Villages in Spain

Kinnaird, T., Bolós, J. and Turner, S. 2017. Optically-Stimulated Luminescence Profiling and Dating of Historic Agricultural Terraces in Catalonia (Spain). Journal of Archaeological Science 78: 66–77. López Mullor, A. 2011. La arqueología medieval en Cataluña, 1980-2010. Boletín de Arqueología Medieval 15: 265-398. MacPhail R. I. and Linderholm, J. 2004. “Dark Earth”: Recent Studies of “Dark Earth” and “Dark-Earthlike”. Microstratigraphy in England, UK. In L. Verslype and R. Brulet (Eds.), Terres Noires-Dark Earth. Actes de la table ronde internationale tenueà Louvain-la-Neuve, les 09 et 10 novembre 2001: 35-42. Louvainla-Neuve, Université Catholique de Louvain. Martín Civantos, J.M. 2018. La Arqueología del paisaje como lugar donde hacer realmente compleja nuestra disciplina. In J.A. Quirós Castillo, Treinta años de Arqueología medieval en España: 205-224. Oxford, Archaeopress. Míguez Mariñas, M.I. 2010. Espacio y propiedad en Asturias: Arqueología del paisaje y evolución social en el concejo de Castrillón. Gijón, Trea. Muñiz López, I. 2014. El pueblo dormido, Banduxu (Proaza). Historia, Paisaje y arqueología de la aldea asturiana, Oviedo, Tesis doctoral: http://hdl.handle.net/10651/29286. Narbarte Hernández, J., Rodriguez Lejarza, A., Santeramo, R., Quirós Castillo, J.A., Iriarte Avilés, E. 2018. Evidencias de ocupación antigua en núcleos rurales actualmente ocupados: el proyecto arqueológico de Aizarna (Guipúzcoa). Munibe Anthropology Archaeology 69: 239-256. Quirós Castillo, J. A. 2009. The Archaeology of Medieval Villages en Europa. Bilbao, University of the Basque Country. Quirós Castillo, J. A. 2012. Arqueología del campesinado medieval: La aldea de Zaballa. Bilbao, University of the Basque Country. Quirós Castillo, J. A. 2013. El poblamiento rural de época visigoda en Hispania. Arqueología del campesinado en el interior peninsular. Bilbao, University of the Basque Caountry. Quirós Castillo, J. A. (ed) 2014. Agrarian Archaeology in early Medieval Europe, Quaternary International 346: 1-162. Quirós Castillo, J.A. (ed) 2018. Treinta años de Arqueología Medieval en España, Oxford, Archaeopress. Reglero de la Fuente, C., Sáez Sáiz, I. 1999. El despoblado medieval de Fuenteungrillo (Valladolid): análisis de las estructuras del castillo. In Actas del V Congreso de Arqueología Medieval Española Vol. 1: 77-83. Valladolid, Junta de Castilla y León. Reynolds, A. and Quirós Castillo, J.A.. 2007. Aistra (Zalduondo). Arkeoikuska 06: 94-100. Rodríguez Gutiérrez, F.1984. Transformación y crisis de un espacio de montaña: el Concejo de Lena. Oviedo, Ayuntamiento de Oviedo. Rodríguez Gutiérrez, F. 1989. La organización agraria de la montaña central asturiana. Oviedo, RIDEA. Roig Buxó, J. 2009. Asentamientos rurales y poblados tardoantiguos y altomedievales en Cataluña (siglos VI al X). In J.A. Quirós Castillo (ed), Archaeology of Early Medieval Villages in Europe: 2017-251. Bilbao, University of the Basque Country. Roig Buxó, J. 2013. Trabajo, mercancía y conocimiento. La Arqueología como profesión, J.A. Quirós Castillo (ed), La materialidadde la Historia. La Arqueologia en los inicios el siglo XXI: 177‐200. Madrid, Akal. Rosset, P. 1999. The multiple functions and benefits of small farms agriculture. Oakland, Food First/The Institute for Food and Development Policy. Ruiz-Ruiz, J.F., Martín Civantos, J., 2017. La gestión comunitaria del agua en la cara Norte de Sierra Nevada: Acción colectiva y saberes etnoecológicos en los sistemas de riego de origen andalusí. E-rph: Revista Electrónica de Patrimonio Histórico: 76-103. Sánchez Hernández, A. 2006. La concentración parcelaria y sus fases. Redur 4: 75-85. Sancho i Planas, M. 2009. Aldeas tardoantiguas y aldeas altomedievales en la sierra del Montsec (Prepirineo leridano): hábitat y territorio. In J.A. Quirós Castillo, The Early medieval villages in Europe: 275-288. Bilbao, University of the Basque Country. Tejerizo García, C. 2017. Arqueología de las sociedades campesinas en la Cuenca del Duero durante la primera Alta Edad Media. Bilbao, University of the Basque Country. Tejerizo García, C. and Quirós Castillo, J.A. 2018. Treinta años de arqueología en el norte de la Península Ibérica. La “otra” Arqueología Medieval. In J.A. Quirós Castillo (ed), Treinta años de Arqueología medieval en España: 95-122. Oxford, Archaeopress. 115

The Archaeology of Medieval Villages Currently Inhabited in Europe

Vigil-Escalera Guirado, A. 2003, Excavaciones arqueológicas en el yacimiento “El Pelícano” (Arroyomolinos, Madrid). Informe depositado en la Dirección General de Patrimonio Histórico de laComunidad de Madrid. Vigil-Escalera Guirado, A. 2015. Los primeros paisajes altomedievales en el interior de Hispania. Registros campesinos del siglo V d.C. Bilbao, University of the Basque Country. Vigil-Escalera Guirado, A. 2018. Los últimos treinta años de la arqueología de época visigoda y altomedieval. In J.A. Quirós Castillo, Treinta años de Arqueología medieval en España: 271-294. Oxford, Archaeopress. Vigil-Escalera Guirado, A. and Quirós Castillo, J.A. 2013. Un ensayo de interpretación del registro arqueológico, en Quirós Castillo, J.A. (dir), El poblamiento rural de época visigoda en Hispania. Arqueología del campesinado en el interior peninsular. Bilbao, University of the Basque Country: 357-399 Villa Valdés, A. 2007. Explotaciones auríferas en la sierra de Begega (Belmonte de Miranda): principales resultados de la intervención arqueológica. In Excavaciones Arquelógicas en Asturias 1999-2002: 295-303. Oviedo, Principado of Asturias. Wolf, E. R. 1971. Los campesinos. Barcelona, Labor.

116

Conclusion Chris Wickham How to deal with the archaeology of still-inhabited villages (CORS, ‘currently occupied rural settlements’, as Carenza Lewis here calls them) is one of the major current next-step problems in medieval archaeology. Anyone who has tried to do meta-analyses of current archaeological data knows how skewed the bulk of our material is. If early medieval settlements are fragmented, elusive, temporary, maybe this is simply because the more stable and permanent ones are under existing villages and towns? It is an easy assumption to make, but it is one which cries out for testing. And, until very recently, that testing was rarely carried out, as all the contributors to this small but significant volume make clear. Nor is it easy to do it, of course; and it would have been forgivable if these contributors had restricted themselves to methodological points about how it could be done, in targeted research programmes, in the future. But for the most part they have gone much further than that, and have offered preliminary (indeed, in some cases quite full) results of attempts to get past the problem that open-area excavations in existing settlements are almost never possible; it is that concrete contribution which makes the book significant. The problem here is a particularly medieval one. It is easily arguable that the settlement pattern of the western Roman empire, which underlay three or four out of our five empirical articles here, was not like that of later centuries in its basic elements; it was structured in large part around isolated estate centres, even if larger settlements were prominent in some areas, and it has never been very difficult to find its key elements in open landscapes in most of Europe. But the middle ages is a different matter; at some point, varying from place to place, settlement became more stable, and it becomes harder to believe that a deserted settlement is at all typical of those which show greater continuity – as Lewis shows, deserted settlements are only around 10% of the total for medieval England after 1000, and they are in general poorer and weaker sites. In Denmark one can study settlements until they ceased to move around, in the eleventh or twelfth centuries; Ingvild Øye cites the best-known instance of this, Vorbasse. In parts of southern Europe, one can do so until settlements became nucleated on the tops of hills, as was the case for much of central and southern Italy at different moments between 700 and 1200 (not that this makes earlier settlement easier to find there, unfortunately). After these dates, however, one can either resort to guesswork, as people have hitherto, or else – and more satisfactorily – one must confront the difficult task of figuring out exactly how to work archaeologically with villages where people actually live. As all our contributors stress, this can only be done in a multidisciplinary way. As I am a historian, this seems obvious to me; I know that there are plenty of schools of archaeology which prefer to avoid the written record, but when settlements are still lived in there are so many kinds of documentary evidence – and, of course, also of environmental, geographical, anthropological evidence – that it would be pointless to imagine not using them. This points us in the direction of post-processual archaeological methods; they were not, by and large, stressed here, but they were discussed at the conference, for Stephen Mileson, who has pioneered such methods in Oxfordshire, also spoke at it.1 Otherwise, however, catastal maps, rentals, post-mortem inventories of possessions, judicial documents, all contribute to the construction of late material and early modern material culture – for nearly all our contributors agreed that this was the most problematic period for us – along with the study of standing buildings, which has been the most common recourse of archaeologists of the period hitherto. But, actually, the reason why post-processual methods have not been stressed here is also because our contributors have found other ways to confront the problem. These are in each case different ways, but, put together, they all contribute to a handbook of methods, and successful ones, for getting around the problem that people still live on the sites one wishes to study. The new syntheses possible on the basis of large-scale test-pitting, which Lewis develops most of all here, are a particularly striking example, one which I would expect to see tried across Europe 1 

See in particular S. Mileson, ‘Openness and closure in the later medieval village’, Past and present, 234 (2017), pp. 3–37.

117

The Archaeology of Medieval Villages Currently Inhabited in Europe

from now on (although in the very heavily built-up villages of southern Europe, where gardens are fewer, it could be harder), but every paper presents others. There were strong signs at the conference that people were very keen to learn from each other, and publishing these articles together will, I hope, push that mutual learning process further along. What do we learn here in concrete terms, however? First, a terminological point. Some of our contributors are unhappy with too wide a use of the word ‘village’; Johan Verspay and his team, for example, wish to restrict it to stable concentrated settlements, and Øye sees villages and clusters of farms as partially separate concepts – even if she stresses both that Scandinavian languages do not help much in setting out a clear terminological difference here, and that, in reality, individual settlements moved between settlement types quite often, with isolated farms perhaps clustering and then becoming villages, but then perhaps turning back into hamlets thereafter as well. (Norway is a good region for realising that, in fact, given its relatively dispersed and fluid settlement typology.) I have myself never been concerned about using the word ‘village’ for a settlement which heavily dispersed in its spatial pattern, or else quite small, ‘hamlet-like’. As long as the people living in a settlement had a conception of a territory around that settlement which was defined by their occupation of the terrain, how the settlement was actually configured spatially has never seemed to me to matter – although, of course, a nucleated village and a heavily dispersed one will normally have importantly different social structures.2 That ‘conception of a territory’ is a historian’s definition, perhaps, but the more multidisciplinary one is, the less that might matter. Anyway, in this brief conclusion I refer to villages and settlements interchangeably. This initial point is important to make because one of the most significant conclusions which our contributors come to is that, when one actually sets out to study existing villages in detail, one discovers that they change in structure all the time – in their size, as Øye stresses, or in their internal spatial relationships, as Lewis, Edith Peytremann and Margarita Fernández Mier/Jesús Fernández all stress, or in their basic economic functions and/or relations with the agrarian landscape, as is implicit in the arguments of all the contributors but is stressed most by Øye and Fernández Mier/Fernández. We cannot naïvely assume that the nineteenth-century village plans which are available across Europe reflect the settlement patternings even of 1500, let alone 1100. Too many interventions can occur in between. This discovery, while it might be upsetting to adherents of traditional typologies of village forms, such as the Gewanndorf or the Waldhufendorf, is liberating for the rest of us, for it gives back historical process to a millennium of otherwise continuous settlement stability. The empirical conclusions that one can otherwise come to on the basis of reading this book are more contingent; it is not easy to compare case studies from regions as different as Norway, the Netherlands, France, England and the northern Spanish mountains, in any more systematic way. But two points can at least be made. First, actually-existing villages could be very old. In the Netherlands, perhaps not; they were there only sometimes Carolingian and hardly ever Roman (Verspay et al.). In Lewis’s part of eastern England, too, there was, even in still-occupied villages, a clear break from Roman-British settlement, and not necessarily a firm early medieval basis to the clearer settlement structures of the centuries after 900/1000. But in Norway, the Asturias, parts of France, settlement occupation could go back to the Bronze age, even the Neolithic. (In France, the Roman period was a frequent point of reference, as Peytremann shows; in the Asturias, by contrast, that period was less visible than was the seventh century and onwards, and also the pre-Roman period.) One would, of course, have to look at each site in detail to be sure what those early dates actually meant: are they simply signs that older earthworks and mounds were numinous points of reference to much later settlers? Do they simply show that the land was worked continuously, rather than that settlement was continuous? – for there were certainly settlement breaks in stable agrarian landscapes in the case of several of the individual villages discussed here. And so on. And, as I have already said, the internal structuring of villages will often have changed dramatically across these long tracts of time. But this is still an important corrective to the assumptions 2 

See for example C. Wickham, ‘La cristalización de la aldea’, in XXXIV semana de estudios medievales (Pamplona, 2008), pp. 33-51.

118

Chris Wickham - Conclusion

of many scholars that ‘the village’ is a product of this last millennium only; we have to deal with a long back history in many places. The second point is less surprising to experts in this field, and even to followers of quite traditional scholarship, but it was striking all the same to find stress on it converging across discussions of such different regions: one major moment for the structuring of the lived-in landscape everywhere was the foundation of village churches, which was often a development of the twelfth century. This is partly a matter of power: élites built churches, and used them to impose their power over local societies which had been, in some cases, less dominated before; collective religious life was also an important element of village solidarity, and in this period it became crystallised in and around buildings which in nearly every part of Europe were much more elaborate, solid, expensive, than traditional, even élite, secular constructions. Churches did not necessarily mark the ‘birth’ of the village (least of all this), or the move to nucleated settlement, or the end of a more fluid history for settlement; but they were an important structuring element for all that, and such churches (which are often still there, in their early form or as later reconstructions) remained a point of reference for whatever changes would come later to the lived spaces of parishioners. Every village is and was different – a point constantly stressed during the conference, particularly by Peytremann, who offers here an impressive synthesis of very disparate examples – but there were all the same some common (although, doubtless, not universal) developments too, and it is worth signalling them when we see them. Finally, I think it is worth urging that a new focus on studying still-existing settlements, as is so well set out here, should not lead to the separation of this form of study from that, more fully-established, of studying deserted settlements. Jesús Fernández Fernández and Margarita Fernández Mier in their introduction stress that one major unanswered question, when looking at the ‘resilience’ of villages, is what were the reasons for some settlements surviving while others failed; we cannot think of answering this unless we study both. We will have to take all kinds of settlement into consideration, if we want, through comparison, to arrive at lasting syntheses.

119